Enhanced β-carotene production in Yarrowia lipolytica through the metabolic and fermentation engineering Open Access

Abstract

 

β-Carotene is a kind of high-value tetraterpene compound, which shows various applications in medical, agricultural, and industrial areas owing to its antioxidant, antitumor, and anti-inflammatory activities. In this study, Yarrowia lipolytica was successfully metabolically modified through the construction and optimization of β-carotene biosynthetic pathway for β-carotene production. The β-carotene titer in the engineered strain Yli-C with the introduction of the carotenogenesis genes crtI, crtE, and crtYB can reach 34.5 mg/L. With the overexpression of key gene in the mevalonate pathway and the enhanced expression of the fatty acid synthesis pathway, the β-carotene titer of the engineered strain Yli-CAH reached 87 mg/L, which was 152% higher than that of the strain Yli-C. Through the further expression of the rate-limiting enzyme tHMGR and the copy number of β-carotene synthesis related genes, the β-carotene production of Yli-C2AH2 strain reached 117.5 mg/L. The final strain Yli-C2AH2 produced 2.7 g/L β-carotene titer by fed-batch fermentation in a 5.0-L fermenter. This research will greatly speed up the process of developing microbial cell factories for the commercial production of β-carotene.

One-Sentence Summary

In this study, the β-carotene synthesis pathway in engineered Yarrowia lipolytica was enhanced, and the fermentation conditions were optimized for high β-carotene production.

Graphical Abstract

Synthesis pathways of β-carotene and fatty acids in the engineered Yarrowia lipolytica.

Open in new tabDownload slide

Introduction

β-Carotene (C₄₀H₅₆), a precursor of vitamin A, plays vital roles in anticancer, antioxidant, and anti-cardiovascular diseases (Eggersdorfer & Wyss, 2018; Wang et al., 2021). β-Carotene can be widely found in plants, fungi, and algae; however, its extraction efficiency from these natural resources is extremely low (Liu et al., 2021a; Maoka, 2020). Currently, chemical synthesis is still one of the main methods for large-scale β-carotene production; however, it also faces several challenges, such as complexed processes and toxic by-products. Recently, microbial fermentation has emerged as a promising alternative for β-carotene production owing to its biological safety and environmental friendliness as well as sustainability (Choudhari et al., 2008; Sevgili & Erkmen, 2019; Wang et al., 2019).

With the rapid development of synthetic biology, the metabolic construction of model microbes such as Escherichia coli and Saccharomyces cerevisiae for natural product synthesis has emerged as an attractive method over the conventional β-carotene production owing to their mature manipulation tools and easy cultivation (Bu et al., 2020; Jing et al., 2021; Ma et al., 2019b; Verwaal et al., 2007). For instance, the recombinant β-carotene producing E. coli has been successfully constructed after the expression and optimization of β-carotene metabolism modules, which can produce 2.1 g/L of β-carotene (Zhao et al., 2013). The exogenous addition of oleic acid and down-regulation of squalene synthase (ERG9) enabled the genetically engineered S. cerevisiae to produce 142 mg/L of β-carotene (Bu et al., 2022). Different from these two well-known model microbes, Yarrowia lipolytica is considered as a potential host for the synthesis of terpenoids, as it has powerful acetyl-CoA synthesis and energy supply systems, which has also been Genetically Recognized as Safe for the commercial production of food-grade chemicals (Ma et al., 2019a; Muhammad et al., 2020). Furthermore, Y. lipolytica is a naturally oil-producing yeast, making it more suitable for the production of hydrophobic β-carotene, as β-carotene is a highly hydrophobic compound, which has been reported to be accumulated in the lipid droplets of microorganisms (Larroude et al., 2018; Liu et al., 2021b; Sun et al., 2018; Suwaleerat et al., 2018; Zieniuk & Fabiszewska, 2018). In addition, Y. lipolytica possesses a wide substrate spectrum, including hydrophobic and hydrophilic carbon sources, making it more suitable for industrial β-carotene applications (Wierzchowska et al., 2021).

Two natural pathways, including the 2-C-methyl-D-erythritol-4-phosphate pathway (MEP) and the mevalonate pathway (MVA), are mainly responsible for the production of β-carotene precursors, including isopentenyl pyrophosphate (IPP) and dimethylallyl diphosphate (DMAPP) (Wang et al., 2019). The MVA pathway naturally exists in Y. lipolytica, in which two molecules of acetyl-CoA can be condensed into one molecule of acetoacetyl-CoA under the catalysis of acetyl-CoA thiolase (ATTC). Then, acetoacetyl-CoA can be further converted to 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) under the catalysis of hydroxymethylglutaryl-CoA synthase (HMGS) followed by the removal of one molecule of coenzyme A from HMG-CoA to form MVA by HMG-CoA reductase (HMGR) with the electrons provided by NADPH. Finally, IPP can be synthesized under the action of mevalonate kinase (MVK), phosphomevalonate kinase (PMK), and methylpropionate pyrophosphate decarboxylase (MVD1), which can be further isomerized to DMAPP through the catalysis of isopentenyl diphosphate isomerase (Ma et al., 2016; Matthaus et al., 2014; Wang et al., 2019, 2021). The isoprene precursors will undergo several steps of condensation reactions catalyzed by the relevant terpene synthases to produce the corresponding terpenoids (Fig. 1).

In this study, we first expressed the β-carotene synthesis module originated from Xanthophyllomyces dendrorhous into the genome of Y. lipolytica. Then, the rate-limiting step in MVA pathway was eliminated through the metabolic engineering optimization, while the lipid synthesis capacity of the engineered strain was improved. On this basis, key genes in multiple copies were expressed to enhance the β-carotene synthesis flux. Our results indicated that the intracellular lipids can promote β-carotene synthesis, and the multi-copy expression of key genes is an important means to improve the metabolic flux. The combination of metabolic and fermentation engineering can greatly improve the product synthesis capacity of industrially produced strains.

Materials and Methods

Strains, Media, and Reagents

Escherichia coli strain DH5α, which was used for constructing and propagating recombinant plasmids was grown in Luria–Bertani (LB) medium (10 g/L NaCl, 10 g/L tryptone, and 5 g/L yeast extract) solidified with 20 g/L agar. Ampicillin (50 mg/L) was added to the LB medium when necessary. Escherichia coli strains were cultured at 37°C with shaking (220 rpm). Yarrowia lipolytica strains used in this study were grown in YPD medium (20 g/L glucose, 20 g/L tryptone, and 10 g/L yeast extract), YPG medium (20 g/L glycerol, 20 g/L tryptone, and 10 g/L yeast extract), or synthetic dextrose medium (6.7 g/L YNB, 0.01μmol (NH₄)₂Fe(SO₄)₂, 2 g/L glucose, and appropriate amount of amino acid mixture). Hygromycin B (500 mg/L) and agar (20 g/L) were added to the YPD media when necessary. Yarrowia lipolytica strains were cultured at 30°C with shaking (220 rpm).

Construction of Plasmids

The primers (Table 1) used for constructing plasmids were synthesized by Genscript (Nanjing, China). Target DNA fragments were amplified by a polymerase chain reaction (PCR) using PrimeSTAR Max (Vazyme). The restriction enzymes used in this study were purchased from New England Biolabs (Beijing, China). The heterologous genes were optimized according to the codon preferences of Y. lipolytica and synthesized by General Biosystems (Anhui, China). All plasmids were constructed using the Gibson assembly method. The initial plasmids 113-GPD-TEF, Pki-TEF used in this study were kindly donated by Prof. Qingsheng Qi, Shandong University. The 113-GPD-TEF is an integrative vector with the URA3 marker. The PAN7-1 plasmid is a modified fungal expression vector with hygromycin B (hph) and URA3 markers. The Pki-TEF vector is a modified fungal expression vector with the Leu2 marker. The crtE, crtI, and crtYB fragments were amplified by PCR with crtE-F, crtE-R, crtI-F, crtI-R, crtYB-F, and crtYB-R primers, respectively. The above amplified fragments were integrated with plasmid vectors to obtain 113-GPD-TEF-crtI, 113-GPD-TEF-crtIE, and Pki-TEF-crtYB plasmids, respectively. P_TEF1-crtYB-T_cyc1 gene cassette was amplified with PcrtYB-F, TcrtYB-R primers and later integrated on 113-GPD-TEF-crtIE in the same way to generate plasmid 113-GPD-TEF-crtIEYB. P_GPD-hph-T_xpr2 was amplified from the PAN7-1 plasmid with hph-F and hph-R primers, and acc1 was amplified with acc1-F and acc1-R primers. 113-GPD-TEF-hph and Pki-TEF-acc1 plasmids were subsequently constructed, respectively. Yarrowia lipolytica endogenous tHMGR was amplified with tHMGR-F and tHMGR-R primers and integrated into 113-GPD-TEF-hph to obtain 113-GPD-TEF-hph-tHMGR. Next, P_TEFin-tHMGR-Tcyc1 was amplified from 113-GPD-TEF-hph-tHMGR with 2tHMGR-F and 2tHMGR-R primers, and 113-GPD-TEF-hph-2tHMGR plasmid was constructed. P_GPD-hph-T_xpr2 was amplified with 2tHMGR-F and 2tHMGR-R primers, and 113-GPD-TEF-hph-2crtIEYB was obtained after inserting hph gene cassette into 113-GPD-TEF-crtIEYB. The plasmids used in this study are listed in Table 2.

Construction of Yeast Strains

Yeast transformation was performed using a Frozen EZ Yeast Transformation II Kit™ (Zymo Research, USA), and transformants were selected on SD-LEU-URA and hygromycin plates. Since the methods of gene integration and strain verification for strains are all the same, strain Yli-C was taken as an example to describe the specific operations. The strain Yli-C was constructed by transforming plasmid 113-GPD-TEF-crtIEYB. Briefly, the plasmid was first linearized by enzymatic digestion, and then 200–1000 ng (no more than 10 μL) of the gene fragment was treated with 50 μL of po1f competent cells by the Frozen-EZ Yeast Transformation II Kit™ and incubated at 30°C for 1 hr. A total volume of 100 μL of the transformation mixture was plated onto the screening plate and incubated at 30°C for 2–4 days. Transformants were verified by colony PCR. All yeast strains used in this study were listed in Table 3. The genetic modifications of all corresponding strain phases were shown in Fig. 4c.

Analysis of β-Carotene

To analyze the β-carotene and lycopene contents, 2 mL aliquot of the fermentation broth was centrifuged to collect cells. Cells were accordingly resuspended in 3 mL of dimethyl sulfoxide and incubated for 10 min at 60°C and then for 15 min at 50°C after adding an equal volume of acetone. The samples were then centrifuged (13000 g for 5 min). The supernatants containing β-carotene, lycopene, or other carotenoids were analyzed using the high-performance liquid chromatography (HPLC; Agilent Technologies 1200 Infinity Series system, Agilent, USA) with a variable-wavelength detector (450 nm) and an Acclaim™ 120 C18 column (5 µm, 4.6 × 250 mm). The mobile phase (at 30°C) consisted of 1 mL/min methanol, acetonitrile, and dichloromethane (42:42:16).

Fermentation in Shake Flasks and a 5.0 L Bioreactor

For the shake-flask fermentations, 50 mL of YPD media with 0.5 mL of precultured yeasts in a 250-mL flask. Cells were cultured at 30°C and 220 rpm for 7 days. Fed-batch cultivations were performed in a 5.0 L bioreactor with an initial working volume of 2.0 L containing 30 g/L glucose, 20 g/L peptone, and 10 g/L yeast extract. The medium was inoculated with a 500-mL aliquot of a 24-hr preculture, which was prepared in a shake flask containing YPD medium. The initial OD₆₀₀ for the fermentations was ∼1.2. The pO₂ level was set at 20% air saturation using the cascade controls for the agitation speed (666 rpm) with a constant air input flow rate of 1 v/v/min. The temperature was maintained at 25°C throughout the operation. The pH was maintained at 5.6 by adding 4.0 M NaOH or 4.0 M HCl. Foam was eliminated by adding antifoam 204 (Sigma-Aldrich, St. Louis). After the fermentation, broth OD₆₀₀ was stabilized (24–48 hr), 2.0 mM of H₂O₂ was added to induce oxidative stress on cells.

Results

Metabolic Construction of β-Carotene Producing Yarrowia lipolytica

The biosynthesis of β-carotene via acetyl-CoA in Y. lipolytica requires a long metabolic pathway involved in multiple enzymes (Fig. 1). In order to achieve the heterologous synthesis of β-carotene in Y. lipolytica, which cannot indigenously produce β-carotene, we first overexpressed structural genes responsible for β-carotene synthesis in Y. lipolytica. It has been reported that Y. lipolytica has a specific codon preference for heterologous genes. Accordingly, we first codon-optimized genes of crtI (encoded phytoene desaturase), crtE (encoded geranylgeranyl diphosphate synthase), and crtYB (encoded phytoene synthase and lycopene cyclase) originate from X. dendrorhous, and then integrated them into the chromosome of Y. lipolytica using the linearized 113-GPD-TEF-crtIEYB (including crtI, crtE, and crtYB) vector inserted with strong promoters GPD and TEF, generating the genetically engineered strain Yli-C. After the plate screening of the uracil-deficient type, positive transformants with orange–red colors were chosen, which were cultivated in shake flasks with YPD as medium for 192 hr. HPLC analysis showed that the β-carotene titer could reach 34.5 mg/L with OD₆₀₀ of 26 by the engineered strain Yli-C, indicating that the heterologous biosynthesis of β-carotene has been successfully achieved in engineered Y. lipolytica (Fig. 2b). In contrast, no β-carotene production was detected in the original strain Y. lipolytica po1f (Fig. 2a). The β-carotene yield of strain Yli-C was basically equal to those of other studies (Gao et al., 2017; Ma et al., 2022).

Yarrowia lipolytica has been reported to show a wide substrate spectrum, and the carbon metabolic flow of Y. lipolytica can only be concentrated in the glycolytic reactions when glucose or glycerol was used as the sole carbon source (Guo et al., 2022). When glucose was used as the carbon source, OD₆₀₀ of strain Yli-C reached the maximum of 26.5, while the OD₆₀₀ was 22.0 when glycerol was used as the carbon source. The β-carotene production from glucose was increased by 18% compared to glycerol as the carbon source (Fig. 2c). Then, the effect of temperatures ranging from 20°C to 30°C on β-carotene production was further investigated. As shown in Fig. 2d, the maximum β-carotene production could reach 40 mg/L at 25°C, which was 60% and 14.3% higher than that of 20°C and 30°C, respectively. This result is in accordance with that obtained by Zhou et al. (2018, 2019).

Enhancement of Gene Expression Levels in Fatty Acid Synthesis and MVA Pathways

3-Hydroxy-3-methylglutaryl coenzyme A reductase (HMGR) can irreversibly catalyze the reaction of HMG-CoA to mevalonate, making HMGR a recognized rate-limiting enzyme in the MVA pathway (Polakowski et al., 1998). HMGR is a membrane protein resident in the endoplasmic reticulum, which consists of an N-terminal membrane anchor and a catalytic structural domain extending to the cytoplasmic C-terminus (Burg & Espenshade, 2011; Li et al., 2019; Muhammad et al., 2020). It was found that truncating 500 amino acids at the N-terminal end of HMGR can effectively prevent the structural domain from self-degradation, making the protein structure more stable in the cytoplasm of Y. lipolytica (Friesen & Rodwell, 2004; Gao et al., 2017; Ma et al., 2019b). Strain Yli-CH was accordingly genetically constructed through further overexpressing the endogenous tHMGR from Y. lipolytica in strain Yli-C. After 168 hr of shake flask fermentation, the β-carotene titer of strain Yli-CH can reach 50.53 mg/L, which was 17% higher than the parent strain Yli-C (Fig. 3a). This further proves that unlocking the rate-limiting step in MVA pathway can increase the metabolic flux and facilitate the β-carotene synthesis.

Generally, the lipid-soluble β-carotene is mainly stored in the cell membrane, which greatly reduces the flexibility of cell membranes, causing certain cytotoxicity to cells (Gao et al., 2017; Jing et al., 2022). The intracellular lipids of oleaginous yeasts can provide ideal intracellular storage space for β-carotene, which can effectively reduce the toxic effects of β-carotene on cells (Sun et al., 2016). Acetyl-CoA carboxylase, a rate-limiting enzyme in the process of intracellular fatty acid synthesis, is mainly responsible for the conversion of acetyl-CoA to malonyl-CoA. Malonyl-CoA can further bind to acyl carrier proteins under the action of fatty acid synthase complex and uses itself as a C₂ donor to synthesize long-chain fatty acids, thereby enhancing the metabolism of intracellular fatty acid pathway (Fig. 1; Wang et al., 2020). Accordingly, the recombinant strain Yli-C was used as the chassis cell to obtain an engineered strain Yli-CA by overexpressing acc1 from the oleaginous yeast of Cryptococcus podzolicus to enhance its fatty acid synthesis pathway. The engineered strain Yli-CA showed excellent lipid synthesis ability based on β-carotene synthesis, whose β-carotene titer can reach 53.24 mg/L in shake flasks (Fig. 3b).

The above fermentation data show that the independent overexpression of tHMGR and acc1 in Yli-C promoted β-carotene synthesis without affecting cell growth, indicating that the enhanced metabolic expression of MVA pathway and fatty acid synthesis pathway is an effective strategy to improve the β-carotene production. To further improve β-carotene production, the genetic expression of these two metabolic pathways was jointly enhanced. Accordingly, tHMGR and acc1 were simultaneously overexpressed in strain Yli-C to obtain the engineered strain Yli-CAH. Under similar fermentation conditions, the β-carotene titer of the engineered strain Yli-CAH in shake flasks finally reached 87 mg/L, which was 1.72, 1.63, and 1.85 times of those in strain Yli-CH, Yli-CA, and Yli-C, respectively. Notably, the β-carotene yield of recombinant strain Yli-CAH was 85% higher than that of the starting strain Yli-C. Meanwhile, the engineered strain Yli-CAH possessed better growth profiles than strains Yli-CH, Yli-CA, and Yli-C (Fig. 3c).

Push–Pull Strategy to Drive the Metabolic Flux Toward β-Carotene Synthesis

As known, multi-copy expression of key genes in β-carotene synthetic pathway can increase the expression of corresponding encoded enzymes and enhance the final β-carotene production (Bourgeois et al., 2018; Zhang et al., 2020b). For example, 46–60 mg/g DCW of lycopene was obtained after the adjustment of copy number of lycopene synthesis-related genes in Y. lipolytica (Zhang et al., 2019). A total of 159 mg/g DCW of β-carotene can also be obtained after the multi-copy expression of genes in MVA and downstream product synthesis pathways in Y. lipolytica (Liu et al., 2021b). Therefore, the multi-copy expression of tHMGR and crtIEYB was carried out in the following studies to further drive the carbon flux from acetyl-CoA towards β-carotene.

The recombinant strain Yli-C2AH with two copies of crtIEYB was obtained after the integration of 113-GPD-TEF-hph-2crtIEYB plasmid, which carries hph screening marker into the Yli-CAH genome. Strain Yli-C2AH2 was obtained through the integration of the linearized 113-GPD-TEF-hph-2tHMGR plasmid with the genome of Yli-C2AH. Yli-C2AH2, which has three copies of tHMGR including its own endogenous hmgr in MVA pathway. From the transformed plates, it can be clearly observed that the engineered strain Yli-C2AH2 with multiple copies of tHMGR and crtIEYB genes showed darker colors than the engineered strains Yli-CAH and Yli-C2AH, suggesting that the enhanced metabolic expression in both MVA pathway and β-carotene synthesis modules makes strain Yli-C2AH2 more efficiently synthesize β-carotene. To further confirm this result, strains constructed in this section were subjected to shake flask fermentation. Strain Yli-C2AH eventually accumulated 95.3 mg/L of β-carotene after 7 days, which was 11.5% higher than that of strain Yli-CAH and 176% higher than that of the original engineered strain Yli-C (Fig. 4a). Strain Yli-C2AH2 was also able to accumulate up to 117.5 mg/L of β-carotene after 7 days, which was 24.3% higher than that of strain Yli-C2AH and 240.6% higher than that of strain Yli-C (Fig. 4b).

Optimization of Fermentation Conditions

It was known that there was a positive correlation within intracellular carbon-rich metabolites, including those of TCA cycle, gluconeogenesis, and carotenoid synthesis (Park et al., 2017). For example, citric acid cleavage in cells can produce oxaloacetate and acetyl-CoA, which was able to increase the intracellular acetyl-CoA pool. Indeed, when the engineered strain Yli-C2AH2 was subjected to glucose–sodium citrate mixed media with sodium citrate added in gradients (0, 5, 10, and 15 g/L), the β-carotene production of strain Yli-C2AH2 could reach 120.3 mg/L when the glucose to sodium citrate ratio was set at 4:1 (5 g/L sodium citrate), which was 2.5% higher than that in YPD media (Fig. 5a). The microbial growth improved with the increase of sodium citrate addition, and strain Yli-C2AH2 showed the optimal growth trend at 15 g/L sodium citrate (glucose: sodium citrate 4:3) with the lowest β-carotene production.

It has been found that exogenous addition of oxidants can promote the synthesis of β-carotene with antioxidant activity in β-carotene-producing strains, thereby reducing the effect of oxidants on cells (Yan et al., 2011). H₂O₂ is generally used as oxidant in fermentation system to carry out oxidative stress in the cells. Under the optimal conditions of the above experiments, H₂O₂ with varying concentrations of 0, 1, 2, and 2.5 mM was added to the fermentation system to monitor the β-carotene production by strain Yli-C2AH2. Since H₂O₂ is toxic to cell, it was added after 24 hr, when cells have reached a certain amount of growth. As seen in Fig. 5b, it was found that the β-carotene titer was increased when H₂O₂ was added compared to that without the addition of H₂O₂. A maximum of 154 mg/L of β-carotene was obtained when H₂O₂ concentration was 2 mM increased by 60% compared with the control. Significantly, the addition of 1 mM H₂O₂ produced 139 mg/L of β-carotene, which may be due to the insufficient stimulation to cells, while 2.5 mM of H₂O₂ may produce excessive free radicals such as OH·, resulting in lower β-carotene concentrations (121 mg/L). Based on the above, 2 mM of H₂O₂ and 5 g/L of sodium citrate were added to the shake flask fermentation system of Yli-C2AH2 to finally obtain 160 mg/L of β-carotene.

5.0 L Fermenter Amplification Fermentation

In order to evaluate the β-carotene production capacity of the final constructed engineered strain Yli-C2AH2, the fed-batch fermentation at 25°C with constant pH of 5.6 in a 5.0-L fermenter was carried out. As seen in Fig. 6b, 2 mM of H₂O₂ was added at 48 hr of fermentation for the oxidative stress when the strain reached the stabilization stage. The fermentation broth reached a maximum OD₆₀₀ of 164.4 at 60 hr, and finally, 2.7 g/L (51.34 mg/g DCW) of β-carotene was produced after 168 hr of fermentation. As shown in Fig. 6a, the lipid droplets with different sizes can be produced in the cell, which can be clearly observed under an ordinary light microscope, and β-carotene was mainly stored in these lipid droplets. This also confirms our hypothesis that the overexpression of acc1 can provide more storage space for β-carotene after increasing the lipid synthesis capacity of cells. In the work of Ma et al., they also enhanced the cellular storage of lycopene by the overexpression of ole1 and deletion of Seipin (FLD1) to regulate lipid droplet size in S. cerevisiae (Ma et al., 2019a). After 168 hr of fermentation, the color of the fermentation broth changed from light to dark, indicating the increase of β-carotene content.

Discussion

By introducing an exogenous β-carotene synthesis module, the engineered Yli-C successfully achieved the heterologous synthesis of β-carotene. The strain accumulated 40 mg/L of β-carotene at 25°C using glucose as the carbon source. This phenomenon may be due to the fact that the consumption of glucose is able to produce more cofactor of ATP than glycerol during the glycolytic pathway, and the accumulated ATP is more favorable for β-carotene synthesis. Hence, glucose will be adopted as the carbon source for the following studies. On the selection of fermentation temperature, many studies have indicated that low-temperature fermentations are more favorable for carotenoid biosynthesis by some microorganisms, such as the natural carotenoid-producing strains Rhodotorula/Rhodosporidium and X. dendrorhous. The variation of carotenoid content in yeast with temperature may be related to the effect of temperature on the activity of enzymes related to carotenoid synthesis and the expression level of their coding genes (Li et al., 2022). Although we can conclude that low temperatures favor carotenoid synthesis, precise temperature control in industrialization requires more expensive costs, so mining high-temperature-tolerant components in strains and improving their biological activity at high temperatures through enzyme engineering modifications are promising research directions for β-carotene biosynthesis in the future.

The unlocking rate-limiting step resulted in a 17% increase in β-carotene titer in Yli-CH, and the enhancement of the fatty acid synthesis pathway resulted in a 23% increase in β-carotene titer in Yli-CA. The strain Yli-CAH, which optimized both pathways simultaneously, had the highest β-carotene titer of 87 mg/L. The results may be due to the increased lipid content making the cell membrane more flexible, thus reducing the damage of β-carotene to the cell membrane. This work is also similar to the results of another study in which the toxicity of lycopene to S. cerevisiae was mitigated by increasing the unsaturated fatty acid content in cell membranes through the overexpression of ole1 encoding stearoyl-CoA 9-unsaturase (Hong et al., 2019). As lipid synthesis and β-carotene synthesis utilize the same precursor of acetyl-CoA in microbes to further improve the β-carotene production, it is worthwhile to study how to balance the carbon flux distribution between the two modules to achieve an optimal ratio in the future.

Simultaneous multi-copy expression of tHMGR and crtIEYB in Yli-CAH resulted in a β-carotene titer of 117.5 mg/L in Yli-C2AH2. Thus, the increase of key gene copy number can effectively enhance the metabolic flux in MVA and β-carotene synthesis pathways to convert more acetyl-CoA into β-carotene. However, the expression of multiple copies of exogenous genes in the cell will inevitably cause a certain degree of metabolic burden on the host, and therefore gene copy number amplification needs to be strictly controlled. The expression of key enzymes can be enhanced by increasing the copy number of the corresponding genes without affecting the cell growth as much as possible.

In the optimization of fermentation conditions, the addition of citrate had a slight effect on β-carotene production, while the addition of H₂O₂ had a significant increase in β-carotene production. This may be due to the increase in TCA cycle intermediates by cleavage of sodium citrate can enhance cellular respiration, resulting in enhanced growth and decreased product synthesis. As reported in the previous study, the down-regulation of TCA cycle flux resulted in more acetyl-CoA to be diverted to carotenogenesis (Li et al., 2022). The supplementation of sodium citrate mainly increased the microbial growth and did not effectively convert acetyl-CoA to β-carotene. Since cell density is directly related to product yield, efficient expression of key enzymes in the intracellular MVA pathway as well as in the β-carotene synthesis pathway and improvement of the catalytic activity of the enzymes may be critical in future studies to influence the yield of products based on high cell density. As known, β-carotene has antioxidant properties and is effective in reducing cellular and tissue damage induced by reactive oxygen species. It has been found that the addition of exogenous oxidants can promote β-carotene synthesis, as cells stimulate β-carotene synthesis in the absence of antioxidant enzymes to eliminate free radical damage to cells (Xue & Ahring, 2011; Yan et al., 2011).

Finally, based on the phenomenon that the secondary metabolite β-carotene started to be accumulated rapidly after strain Yli-C2AH2 entered the stationary phase, the continuous fermentation strategy in the subsequent study could effectively extend the stationary phase to further increase the β-carotene production. Moreover, in the future study, the two-stage fermentation strategy can be adopted to further improve the β-carotene production, in which high density of cells can be obtained in the first state, and fermentation conditions can be adjusted including temperature, pH, and light to promote the efficient synthesis of β-carotene in the late fermentation stage. In the past studies on β-carotene production by Y. lipolytica, researchers have mostly focused on optimization of metabolic pathways and overexpression of key enzymes (Table 4). In this study, we attempted to increase β-carotene production by enhancing the fatty acid metabolic pathway to improve β-carotene storage space and exogenously adding H₂O₂ for oxidative stress on the basis of increasing the copy number of key genes. The combination of metabolic engineering and fermentation strategy provides a way to synthesize fat-soluble terpenoids and natural products with antioxidant activity by using Y. lipolytica in the future.

Conclusions

Natural oleaginous yeast is an excellent chassis cell for the production of hydrophobic terpenoids. In this study, metabolic and fermentation engineering strategies were used to construct a Y. lipolytica cell factory, targeting the high value-added chemical β-carotene. We first overexpressed exogenous genes in Y. lipolytica to construct a β-carotene synthesis module, enhanced the product and fatty acid synthesis pathways, and then optimized the copy number of key genes to obtain an engineered Yli-C2AH2 with high heterologous β-carotene production. To evaluate the product synthesis capacity of this strain, fed-batch fermentation was performed in a 5.0-L fermenter after process conditions and medium were optimized, resulting in a final yield of 2.7 g/L (51 mg/g DCW), which lays the foundation for the industrial production of β-carotene using Y. lipolytica.

Acknowledgements

Not applicable.

Author contributions

Z.W.M. and X.F.X. designed the experiment. J.Y.W. operated the whole experiment, analyzed the data, and wrote the manuscript. W.J.N., G.H.Y., J.Y.J., and J.W.K. provided the experimental consideration and analyzed the partial data. J.M., X.F.X., and Z.W.M. supervised the whole research and revised the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the National Key R & D Program of China (2019YFA0905700), National Natural Science Foundation of China (22078151, 22178169), Jiangsu Agricultural Science and Technology Independent Innovation Fund Project (CX(21)3120), Jiangsu Synergetic Innovation Center for Advanced Bio-Manufacture (XTD2215), and Young Elite Scientist Sponsorship Program by CAST (YESS20200174).

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References