3.2 Augmenting NAPDH and acetyl-CoA precursor pathways to improve squalene production
NADPH as the primary biological reducing equivalent protects cell from oxidative stress and extend carbon-carbon backbones, which was also reported as the major rate-limiting precursor in fatty acids synthesis in oleaginous species (Qiao et al., 2017; Wasylenko, Ahn, & Stephanopoulos, 2015). HMG-CoA reductase (HMGR) is the first rate-limiting enzyme in the mevalonate pathway and plays critical role in regulating squalene biosynthesis (Ma et al., 2019). 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) is reductively hydrolyzed to mevalonate by releasing coenzyme A with NADPH as reducing equivalent (Cao, Wei, Lin, & Hua, 2017). Based on previous work, source of cytosolic NADPH in the Baker’s yeast may originate from various alternative routes depending on the carbon source and genetic background of the yeast strain (Huan Liu, Marsafari, Deng, & Xu, 2019; Minard, Jennings, Loftus, Xuan, & McAlister-Henn, 1998; Minard & McAlister-Henn, 2005). With glucose as carbon sources, cytosolic NADPH primarily relies on the pentose phosphate pathway. Other cytosolic NADPH pathways include NADP-specific isocitrate dehydrogenase (IDP2), malic enzyme (ylMAE), mannitol dehydrogenase (ylMnDH1, ylMnDH2), 6-phosphogluconate dehydrogenase (ylGND2) and succinate semialdehyde dehydrogenase (ylUGA2) (Huan Liu, Monireh Marsafari, Li Deng, et al., 2019) (Fig. 1). In this work, we tested a collection of auxiliary cytosolic NADPH pathways and investigated how these pathways may enhance squalene production and cellular fitness on the basis of co-expressionSQS-ylHMG (Fig. 3A). Among these chosen NADPHs, mannitol dehydrogenase (ylMnDH2, encoded by YALI0D18964g) presented the best results to improve squalene production. Mannitol, a more reduced sugar alcohol compared to glucose, played an essential role in modulating cytosolic NADPHs through the mannitol cycle. This could partially explain why mannitol was the major byproduct during lipid accumulation phase in Y. lipolytica (P. Xu, Qiao, & Stephanopoulos, 2017). When ylMnDH2 was overexpressed with SQS and ylHMG (strain HLYaliS02 , Supplymentary Table S2), the engineered strain produced 11% more squalene with volumetric production titer increased to 135.22 mg/L, despite relatively decreased yield of 32.33 mg/g DCW (Fig. 3A). This is possibly ascribed to the increased cell fitness and lipid content after enhancing the supplement of NADPH.
Apart from NADPH, acetyl-CoA, is an essential metabolic intermediate connecting glycolysis, Krebs cycle, and glyoxylate shunt pathways. Acetyl-CoA is also the intermediate metabolite participated in lipid synthesis, peroxisomal lipid oxidation and amino acid degradation pathways. It links both anabolism and catabolism, is the starting molecule in MVA pathway. Cytosolic acetyl-CoA was found as a critical precursor to boost secondary metabolite production (Huan Liu, Marsafari, Wang, Deng, & Xu, 2019). For example, engineering alternative cytosolic acetyl-CoA pathways were proven to be efficient strategies to improve fatty acids and isoprenoid production in both Bakers’ yeast and Y. lipolytica (Hu Liu, Fan, Wang, Li, & Zhou, 2019; Meadows et al., 2016; van Rossum, Kozak, Pronk, & van Maris, 2016). Therefore, we next investigated whether endogenous and various heterologous acetyl-CoA pathways could improve squalene production. First, we investigated the pyruvate decarboxylase (PDC), acetylaldehyde dehydrogenase (ALD) and acetyl-CoA synthase (ACS) bypass (Fig. 1) and compared the efficiency of this route from Y. lipolytica , S. cerevisiae andE.coli (Fig. 3B). By overexpression of pyruvate decarboxylase (ScPDC) from S. cerevisiae and acetylaldehyde dehydrogenase (EcPuuc) from E.coli , we obtained only 106.54 mg/L of squalene (Fig. 3B). We observed that the cell growth fitness was negatively impacted due to the expression of heterologous genes, possibly due to the accumulation of the toxic aldehyde intermediate. We next attempted the endogenous ATP citrate lyase, which is the primary acetyl-CoA route to Y. lipolytica metabolism. ATP citrate lyase (ACL) was mainly used for supply of the cytosolic acetyl-CoA, which was proven to have two isoforms encoded by two separate genes in Y. lipolytica (ACL1 and ACL2) (Nowrousian, Kück, Loser, & Weltring, 2000). Endogenous ylACL1 (YALI0E34793g) and ylACL2 (YALI0D24431g) genes were subsequently tested. A 19.5% increase in squalene synthesis was obtained in the resulting strains HLYaliS03 with ylACL2 overexpressed along with SQS and ylHMG, leading to the titer of squalene 144.96 mg/L (Fig. 3B). The increase was probably a result of the pushing strategies for acetyl-CoA enrichment by expressing ACL2 so that adequate cytosolic acetyl-CoA could be pushed into the MVA pathway for the synthesis of squalene. Surprisingly, the specific yield reduced to 25.27 mg/g DCW which was caused by the enhancement of cell growth due to the increased lipid content. This increased lipid content may also serve as the storage space to sequestrate squalene in our engineered cell.