3.4 Shake flask cultivation of engineered strain with pH and carbon/nitrogen ratio optimization
Despite that glucose is the preferred carbon source for cell growth, a similar level of squalene production was detected in our engineered strains (HLYaliS01 and HLYaliS02 , HLYaliS03 ).Y. lipolytica is a natural lipid producer, accumulating up to 30%~60% cell weight as lipid, which leads to a strong competition for the precursor acetyl-CoA (P. Xu et al., 2017). Meanwhile, cultivation pH and media C/N ratio were two critical factors that affect cellular morphology and growth in Y. lipolytica(Szabo, 1999).
In our previous work, we observed a quick declining of cultivation pH from 6 to 3.5 in polyketide synthesis, due to the accumulation of citric acid when glucose was utilized. The pH variations negatively affect strain physiology, alter cell membrane permeability and limit nutrients transportation due to the loss of proton driving force. A significant improvement of polyketide titer was observed by combining PBS buffer with 1 mg/L cerulenin supplementations. Cerulenin is known to irreversibly form a covalent adduct with the active site (cysteine residue) of β-ketoacyl-ACP synthase, inhibiting the elongation of the fatty acid backbone (Huan Liu, Monireh Marsafari, Fang Wang, et al., 2019). Thus, a similar strategy was applied to promote squalene production by strain HLYaliS02 (shown in Fig. 4 A and Supplementary Fig. S3). When the engineered strain was cultivated in the minimal YNB media with 0.2 M phosphoric buffer solution (PBS, pH 6.0), squalene production was increased to 354.44 mg/L at 168 h (Supplementary Fig. S3), which was an increase of 88.4%, compared with the results from pH uncontrolled (Supplementary Fig. S1 C) experiment. The improvement is ascribed to the better growth fitness under pH control and the biomass of strain HLYaliS02 reached 13.98 g/L DCW with the squalene specific yield at 25.35 mg/g DCW (Supplementary Table S3). The major byproduct mannitol accumulated up to 2.2 g/L at 48 hour and citric acid reached 7.81 g/L at 96 h; both mannitol and citrate were subsequently reincorporated into cell metabolism (Supplementary Fig. S3). But only ~50% of glucose was utilized in this process which was consistent with our previous work, indicating the supplementation of PO43- buffer may negatively impact the glucose uptake rate. To further enhance squalene synthesis, 1 mg/L cerulenin was supplemented to the minimal YNB-PBS media at 48 h and the squalene production increased to 384.13 mg/L at 188 h, an 8.4% increase compared with the result without cerulenin (Fig. 4 A). A similar fermentation profile of glucose consumption, mannitol, and citric acid accumulation was found: half of glucose was utilized while 14.9 g/L DCW was obtained with the squalene specific yield at 25.78 mg/g DCW (Supplementary Table S3). Byproduct mannitol reached 1.9 g/L at 48 h, but citric acid increased to 9.7 g/L, which was higher than that in the YNB-PBS media without cerulenin supplemented, possibly due to the fact that inhibition of the endogenous fatty acid synthase may prevent citrate from being converted to acetyl-CoA and oxaloacetate by ATP-citrate lyase (encoded by ACL).
We next investigated the effect of C/N ratio on squalene production in YNB-PBS media supplemented with cerulenin (Fig. 4 B). Various C/N ratios including 10:1, 20:1, 40:1, 60:1 and 80:1 was studied (Supplementary Fig. S4). When the C/N ratio was set at 60:1, a similar fermentation profile of glucose consumption, mannitol, dry cell weight, citric acid was obtained, compared to the metabolic profile for C/N 80:1. Squalene titer at C/N 60:1 reached 396 mg/L at 120 h with increased productivity. The highest squalene titer was achieved in the media with C/N ratio 40:1, reaching 502.75 mg/L at 120 h with the yield to 32.6 mg/g DCW (Fig. 4 B, supplementary Table S3), which was 30.8% higher than the squalene production form C/N ratio 80:1 media. We speculated that acetyl-CoA flux was enlarged and flowed to MVA pathway, since we observed less citric acid accumulation (1.9 g/L) at the end of fermentation. However, when the C/N ratio was further reduced to 20:1 or 10:1, adverse effect was obtained with decreasing squalene production (Supplementary Fig. S4). We speculated that the superfluous nitrogen provision may lead more carbons flowing to cell growth. These results illustrated that C/N ratio plays an important role in the redistribution of carbon flux and strongly influenced the accumulation of squalene. Further downregulation of acetyl-CoA carboxylase (ACC) may be required to improve squalene production. ACCase, as the malonyl-CoA source pathway and the acetyl-CoA sink pathway during lipogenesis, was primarily controlled through the phosphorylation of serine residues by Snf1-mediated AMP kinase. Inhibition of fatty acid synthase pathway and nitrogen starvation was proven to be effective strategies to activate Snf1 kinase and slows down ACC1 activity (Seip, Jackson, He, Zhu, & Hong, 2013; Zhang, Galdieri, & Vancura, 2013). It was consistent with our findings that medium C/N ratio was beneficial for squalene synthesis. By applying these engineering strategies, we have obtained an oleaginous yeast strain with a similar squalene level to the strainS. cerevisiae (Han, Seo, Song, Lee, & Choi, 2018; Huang et al., 2018). This work highlights the potential of engineering Y. lipolytica as a promising microbial platform for efficient synthesis of squalene and terpene-related compounds.