Removal of lycopene substrate inhibition enables high carotenoid productivity in Yarrowia lipolytica

Culture conditions and media

Escherichia coli DH5α cells harboring plasmid were cultured in Luria-Bertani (LB) media (BD bioscience) supplemented with corresponding antibiotics (100 μg/mL ampicillin and 50 μg/mL kanamycin) for plasmid propagation at 37 °C for 16 h. All Yarrowia lipolytica strains used in this study were cultivated at 30 °C with shaking at 230 rpm. The media used for culturing Y. lipolytica strains was prepared as follows. YPD media containing 10 g/L yeast extract (BD bioscience), 20 g/L peptone (BD bioscience), and 20 g/L glucose (Sigma-Aldrich) was used for carotenoid fermentation. YNB media composed of 1.7 g/L yeast nitrogen base without amino acids and ammonium sulfate (YNB, VWR Life Science), 20 g/L glucose, 5 g/L ammonium sulfate (VWR Life Science), 15 g/L agar (BD bioscience), and 0.77 g/L appropriate complete supplement mixture minus uracil, leucine, or tryptophan (Sunrise science products) was used for selecting transformed Y. lipolytica strains. The media with varying Carbon-to-Nitrogen (C/N) ratios provided in Supplementary Table 2 were used to regulate the carbon flux distribution between isoprenoid and lipid synthesis for optimal carotenoid fermentation.

Construction of plasmids and strains

E. coli strain DH5α was selected for plasmid propagation. Y. lipolytica po1f strain served as the base strain, and its derivatives and plasmids used in this study are listed in Supplementary Table 4. The primers (synthesized in Sigma-Aldrich) used for plasmid construction are provided in Supplementary Table 5. The restriction enzymes Not1 and Dpn1 were purchased from New England Biolabs (NEB). KAPA HiFi DNA Polymerase with high-fidelity purchased from KapaBiosystems was used for gene amplification for plasmid construction. GoTaq DNA polymerase (Promega) was used for colony PCR identification. PCR fragments were purified using the ZYMO Fragment Recovery Kit (ZYMO research). Plasmids were constructed using Gibson Assembly kit (NEB) followed by transformation into DH5α cells by heat shock. The successfully constructed plasmids were extracted by the QIAprep Spin Miniprep Kit (Qiagen) and then sequenced at Quintara Biosciences. All engineered Y. lipolytica strains were constructed by transforming linearized plasmids (Not1 digestion) using the lithium-acetate method. Recombinants were verified by PCR amplification from genomic DNA. The carotenoid biosynthetic genes evaluated in this study were codon-optimized towards Y. lipolytica and synthesized by GeneArt (Thermo Fisher Scientific).

TRP1 disruption in po1f strain using CRISPR-Cas9

For TRP1 disruption, the CRISPR-Cas9 plasmid³⁹ containing gRNA (ACGCCGAGGAGTGGTACCGG) targeting the TRP1 (YALI0B07667g) gene of Y. lipolytica was transformed into strain po1f using Ura3 as the auxotrophic marker. The strain with tryptophan auxotrophy was obtained by selecting on YNB-Ura and YNB-Ura-Trp plates. After that, the positive clones were inoculated onto YPD plates and sub-cultured three times to lose the CRISPR-Cas9 plasmid, resulting in the po1f-T strain (ura3^–, leu2^–, trp1^–).

Auxotrophic markers curation by the Cre-loxP system

In order to rescue URA3, LEU2 and TRP1 auxotrophic markers, plasmid pYLMA-Cre harboring Cre recombinase was transformed into target Y. lipolytica strains. Transformants were cultured on YPD agar plate supplemented with a final concentration of 250 mg/L hygromycin B (Sigma-Aldrich) for antibiotic selection. After 2~3 days of cultivation at 30 °C, colonies were randomly transferred onto a new YPD agar plate containing hygromycin B, and cultured for 1 more day to allow for more successful marker deletions. Markers curation was then confirmed by transferring the colonies onto YNB-Ura, YNB-Leu, and YNB-Trp agar plates, respectively. Successful deletion of all three markers in strains will show a phenotype with uracil, leucine and tryptophan deficiency. Plasmid pYLMA-Cre in cells was then removed by incubating engineered strains on YPD agar plates at 30 °C for 24 h, with 2~3 repeats.

Shake flask fermentations

Single colonies of recombinant strains were picked from plate, inoculated into 2 mL YPD media, and cultivated overnight (16~18 h) at 30 °C and 230 rpm. The culture was then transferred to a 50 mL shake flask containing 10 mL YPD media (initial OD₆₀₀ = 0.1), and cultivated at 30 °C with shaking at 230 rpm for 3~5 days. When applicable, 30 mM isoprenol (Sigma-Aldrich) was added into YPD media when glucose of culture was nearly consumed.

Bioreactor fermentations

Fed-batch fermentations were performed in a 3 L bioreactor (New Brunswick Bioflo115 system). The initial fermentation was completed with 1 L medium containing 100 g/L glucose, 100 g/L peptone, and 50 g/L yeast extract. The temperature was maintained at 30 °C. The dissolved oxygen was controlled at 20% with an agitation cascade of 250~800 rpm. Air was sparged into fermenter at 2 vvm. The pH was maintained at 6.8 by feeding 5 M NaOH or 5 M HCL. Foam was prevented by the addition of antifoam 204 (Sigma-Aldrich). The fed-batch process was initiated after 48 h of cultivation with the 10 × Y₁₀P₁₀D₅₀ media consisting of 100 g/L yeast extract, 100 g/L peptone and 500 g/L glucose. Once the media feeding starting, the agitation and aeration was changed and held constantly at 600 rpm and 0.3 vvm, respectively. Samples were taken every 24 h to measure OD₆₀₀, glucose concentration, and carotenoid titer.

Quantification of residual glucose in media

Glucose concentration in media was determined by High-Performance Liquid Chromatography (HPLC, Agilent technologies 1260) equipped with a refractive index detector and a HPX-87H column (Bio-Rad). The 500 μL sample was extracted and centrifuged at 12,000 g for 5 min, and then the supernatant was filtered through 0.2 μm syringe filters prior to injection. The mobile phase consisted of 14 mM sulfuric acid (Sigma-Aldrich) with a flow rate of 0.7 mL/min at 50 °C. The injection volume was 10 μL.

Lipid extraction and quantification

The fatty acids synthesized by Y. lipolytica including palmitate (C16:0), palmitoleate (C16:1), stearate (C18:0), oleate (C18:1) and linoleate (C18:2) were quantified using a Gas Chromatography coupled to a Flame Ionization Detector (GC-FID). 0.1~1 mL cell culture was extracted from each bioreactor such that the sample contained approximately 1 mg biomass. A centrifugation step at 16,000 g for 10 min was performed and the supernatant discarded. 0.5 mL of a 0.5 M sodium hydroxide-methanol solution (20 g/L sodium hydroxide in anhydrous methanol) was mixed with the cell pellets, followed by the addition of 100 μL internal standards containing 2 mg/mL methyl tridecanoate (Sigma-Aldrich) and 2 mg/mL glyceryl triheptadecanoate (Sigma-Aldrich) dissolved in hexane. Methyl tridecanoate was used for volume loss correction during sample preparation and glyceryl triheptadecanoate was used for transesterification efficiency correction. The samples were vortexed for 1 h to allow for the transesterification of lipids to fatty acid methyl esters (FAMEs). Afterwards, 40 μL of 98% sulphuric acid (Sigma-Aldrich) was added to neutralize the pH. The FAMEs were then extracted through the addition of 0.5 mL hexane followed by vortexing for 30 min. Centrifugation at 12,000 g for 1 min was then performed to remove cellular debris and the top hexane layer was extracted for analysis. Separation of the FAME species was achieved on an Agilent HP-INNOWax capillary column. The injection volume was 1 μL, split ratio was 10, and the injection temperature was 260 °C. The column was held at a constant temperature of 200 °C and helium was used as the carrier gas with a flow rate of 1.5 mL/min. The FID was set at a temperature of 260 °C with the flow rates of helium make up gas, hydrogen, and air at 25 mL/min, 30 mL/min, and 300 mL/min, respectively.

Intracellular metabolites extraction and quantification

To extract intracellular metabolites (e.g., IPP/DMAPP, and GGPP), 1 mL culture was filtered through a 25-mm 0.2 μm nylon filter using vacuum filtration. The cells were washed immediately with 2 mL of water preheated to 30 °C, and the filter was submerged in ice-cold extraction buffer (40% methanol + 40% acetonitrile + 20% water). After incubation at −20 °C for 20 min, the extract solution was centrifuged at 16,000 g for 10 min, and the supernatant was transferred to a new tube and dried. The sample was resuspended with 50 μL water, and then centrifuged at 16,000 g for 10 min. Metabolites in the supernatant were quantified by liquid chromatography-tandem mass spectrometry (LC-MS/MS) comprised of an Agilent 1100 series LC system and an AB Sciex API-4000 MS. 10 μL sample was injected and separation was achieved on a Waters XBridge C-18 column with a mobile phase consisting of solution A (0.1% tributylamine, 0.12% acetic acid, 0.5% 5 M NH₄OH in water, v/v) and solution B (100% acetonitrile). The flow rate was 0.3 mL/min and the following gradients were used: 0–5 min, 0% B; 5–20 min, 0~65% B; 20–25 min, 65% B; 25–30 min, 100% B; 30–35 min, 100% B; 35–36 min 100~0% B, 0% B until 45 min. The analytes were then analyzed using Analyst 1.6.2 and MAVEN 707, and compared to standard curves generated using chemical standards purchased from Sigma-Aldrich and Cayman Chemicals.

Labeling experiments

Strains used in labeling studies were revived in YNB media with [U-¹³C]glucose as the sole carbon source. They were then subcultured in the same media and grown until early stationary phase at 30 °C. Samples were taken before the start of the pulse addition of an extra carbon source using the same intercellular metabolite extraction method. Afterwards, 10 mM stearic acid was added to the corresponding cultures, and measurements of metabolite isotopic enrichments were taken at different time points. The optical densities associated with each sample were also recorded. IPP/DMAPP, and GGPP were quantified by LC-MS/MS. All MS data from labeling experiments were corrected for natural abundance using IsoCor⁴⁰.

Extraction of carotenoids

Carotenoid extraction was performed as described⁴¹ with the following modification. Briefly, 100 μL culture was centrifuged for 1 min at 16,000 g, and cell pellets were suspended in 900 μL dimethyl sulfoxide (DMSO, Sigma-Aldrich) prior to heating at 50 °C for 1 h until the cells bleached in a water bath. The DMSO extracts were briefly mixed with 450 μL of methanol and centrifuged at 16,000 g for 5 min. The resultant supernatants were transferred into 96-well assay plates or glass vials for carotenoid analysis and quantification.

Analysis and quantification of carotenoids

The production of carotenoids was expressed as grams per liter of fermentation broth (g/L) and milligrams per gram of dry cell weight (mg/g DCW). Optical densities were measured at 600 nm with Thermo Spectronic Genesys 20 (Thermo Scientific) and used to calculate cell mass (DCW = 0.35 × OD₆₀₀ for β-carotene and DCW = 0.30 × OD₆₀₀ for lycopene, Supplementary Fig. 18). The analysis and quantification of β-carotene was performed by HPLC (SHIMADZU LC-20 AT) equipped with a Kromasil C18 column (4.6 mm × 250 mm) and UV/VIS detection at 450 nm. The mobile phase consisted of acetonitrile-methanol-isopropanol (5:3:2 v/v) with a flow rate of 1 mL/min at 40 °C. The analysis and quantification of lycopene were performed with Spectramax M2e Microplate Reader (Molecular devices) or HPLC at 470 nm. The standard curves of β-carotene and lycopene (Sigma-Aldrich) were prepared by running the same extraction process as the samples.

Quantitative real-time PCR

Real-time PCR (RT-PCR) was used to estimate the relative gene expression. mRNA extracted by MasterPure^TM Yeast RNA purification kit (Lucigen, Wisconsin, USA) was used as the template. RT-PCR was carried out on an iCycler (Bio-Rad, USA) using iScript^TM one-step RT-PCR kit with SYBR Green Supermix (Bio-Rad, USA) according to the manufacturer’s instructions. ACT1 was used as an internal control gene for normalization. The relative gene expression was calculated using the comparative 2^−△△CT or 2^−△CT method.

In vitro enzymatic assays

Yeast microsomes for in vitro enzymatic assays were prepared as described previously⁴². Briefly, strains harboring wild type or mutated CarRP were grown overnight in YNB media at 30 °C and then inoculated into 200 mL YNB media to an initial OD₆₀₀ of 0.1. After 24 h cultivation, cells were collected by centrifugation at 4,000 g for 10 min. Resuspension of the cells in TEK buffer (50 mM Tris-HCl, pH 6.8, 1 mM EDTA, 0.1 M KCl) followed, and the solution was kept at room temperature for 5 min. Afterwards, the cells were recovered by centrifugation, washed in TES buffer (50 mM Tris-HCl, pH 6.8, 1 mM EDTA, 0.6 M sorbitol), resuspended in TESM buffer (50 mM Tris-HCl, pH 6.8, 1 mM EDTA, 0.6 M sorbitol, 14 mM 2-mercaptoethanol), and left at room temperature for 10 min. Then, the cells were recovered once again by centrifugation, washed in extraction buffer (50 mM Tris-HCl, pH 6.8, 1 mM EDTA, 1 mM PMSF), and resuspended in extraction buffer. Glass beads were added to each sample, which were intermittently vortexed for 30 s and placed on ice for 30 s for a total of 15 repeats. The cell pellets were then discarded by centrifugation at 4,000 g, 4 °C for 10 min, and the supernatant was transferred to a 50 mL tube. The crude yeast microsomal fraction collected above was used for in vitro assays of lycopene cyclase. Standard enzyme assays were performed in a total volume of 200 μL containing 50 mM Tris-HCl (pH 6.8), 1 mM phenylmethylsulfonyl fluoride (PMSF) and 1 mg of microsomal proteins. Serial concentrations of lycopene (50~350 μmol/L) dissolved in dimethyl sulfoxide (DMSO) were used as the substrate. Reactions were initiated by substrate addition, incubated at 30 °C with gentle shaking for 16 h, and then terminated by adding 200 μL ethyl acetate. The solution was vortex for 10 min, and the organic phase was collected by centrifugation and analyzed by HPLC.

Generation of protein variants

The variants were generated by analyzing the amino acid conserved and co-evolutionary information of this protein family from Position Specific Scoring Matrix (PSSM). Here we generate the matrix using psiblast from ncbi-blast-2.7.1 + ⁴³ with uniref90⁴⁴ as the database (https://www.uniprot.org/help/uniref) and an E-value of 0.01 running 3 iterations. For all positions in the protein, the PSSM score which represents conservatism of amino acids was calculated for both lycopene cyclase and other homologous proteins of this family. The higher score indicates the more conservative of the amino acid in this position. We screened the substitutions that could be replaced with more conserved amino acids based on the PSSM scores. The different value between the potential substitutions and wild-type amino acids was calculated and the scores were sorted. All glycine substitutions were removed from the scoring. Top 25 scoring substitutions were combined into double substitutions randomly. A distance matrix was computed using the PAM30 substitution matrix and clustered using Agglomerative in sklearn⁴⁵ into 25 clusters to minimize the number for test. The variants were chosen randomly within each cluster. The ranked 26–50 scoring substitutions were ordered as single mutational variants.

Homology model of lycopene cyclase

A homology model was generated of the lycopene cyclase using TrRosetta server²¹ submitted the sequence of the R domain (1–239 amino acids) of CarRP.

Calculation of the C/N ratio in media

The C/N ratio was calculated by referring to the composition of Yeast extract (Bacto^TM) and Peptone (Bacto^TM) in the BD Bionutrients^TM technical manual (Supplementary Table 6) (https://legacy.bd.com/ds/technicalCenter/misc/lcn01558-bionutrients-manual.pdf). The carbon in Yeast extract and Peptone was ignored because of its extremely lower concentration relative to that of glucose. The total nitrogen in Yeast extract and Peptone is 10.9% and 15.4%, respectively. The C/N ratio was generated with the following formula 1. X, Y, and Z represent the concentration of glucose, yeast extract and peptone respectively.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.