Thraustochytrid strains
T. aureum ATCC 34304 was obtained from the American Type Culture Collection (USA). Parietichytrium sp. SEK358, SEK36447, I65-24A, and A. limacinum mh018648 were isolated from seawater from the Ishigaki Islands, Iriomote Islands, Ishigaki Islands, and Yaeyama Islands, Okinawa, Japan, respectively. Parietichytrium sp. SEK364 was identified as Parietichytrium sarkarianum by analyses of 18S rRNA gene, morphology and cultural properties47. The Parietichytrium strains used in each experiment are indicated in the legend of each Figure and Table.
Culture of thraustochytrids
Thraustochytrid strains were grown in 20 mL of GY medium [3% glucose and 1% yeast extract in 50% artificial sea water(3% NaCl, 0.07% KCl, 1.08% MgCl2 6H2O, 0.54% MgSO4 7H2O, 0.1% CaCl2 2H2O)] with 0.1% vitamin mixture (0.2% vitamin B1, 0.001% vitamin B2, and 0.001% vitamin B12) and 0.2% trace elements (3% EDTA di-sodium, 0.15% FeCl3·6H2O, 3.4% H3BO4, 0.43% MnCl2·4H2O, 0.13% ZnSO4·7H2O, 0.026% CoCl2·6H2O, 0.013% NiSO4·6H2O, 0.001% CuSO4·5H2O, and 0.0025% Na2MoO4·2H2O) in a 100 mL flask at 25 °C with rotation at 150 rpm. Potato dextrose agar (PDA) plates (0.8% potato dextrose agar, 50% artificial sea water, 2% agar) containing appropriate antibiotics were used to select Parietichytrium sp. and T. aureum mutants. The growth of these strains was monitored by measuring the optical density at 600 nm (OD 600), dry cell weight (DCW), and consumption of glucose in the medium. The glucose concentration was measured by the Glucose CII test (FUJIFILM Wako Pure Chemical Co., Japan) and the concentration of dissolved oxygen (DO) was measured by a DO meter.
GC analysis to determine the fatty acid composition
Cells were cultured in a 100 mL flask containing 40 mL of GY medium at 25 °C with shaking at 150 rpm. Two mL of culture fluid was harvested, and cells were collected by centrifugation (2500 × g, 3 min), dried by lyophilization, and their DCW was measured. The fatty acids were extracted as fatty acid methyl esters (FAMEs) from dried cells as described in ref. 49. FAMEs were analyzed by GC-2014 (Shimadzu) equipped with an ULBON HR SS-10 column (Shinwa Chemical Industries). The column temperature was programmed to increase from 160 to 200 °C by 2 °C/min, and then from 200 to 220 °C by 5 °C/min.
Genome sequencing and assembly
Genomic DNA from thraustochytrids was extracted by the standard phenol–chloroform method. The draft genome sequences of T. aureum ATCC 34304 and Parietichytrium sp. I65-24A were obtained using a 454 GS FLX system (Roche Diagnostics) and assembled using Newbler assembler version 2.7 software (Roche Diagnostics). The genomic DNA from T. aureum was sequenced using a paired-end sequencing strategy, and a total of 369 Mbp was obtained from 1,046,218 reads (approximately 9-fold coverage). Assembly of all reads resulted in 14,205 contigs (>500 bp) and 463 scaffolds. The genomic DNA from Parietichytrium sp. was sequenced using a whole-genome shotgun and a paired-end sequencing strategy, and a total of 1556 Mbp was obtained from 4,511,709 reads (approximately 29-fold coverage). Assembly of all reads resulted in 11,544 contigs (>500 bp) and 360 scaffolds. The draft genome sequences of Parietichytrium sp. and T. aureum have been deposited into DDBJ under accession numbers BLSF01000001 to BLSF01000360 and BLSG01000001 to BLSG01000463, respectively.
Expression of DES and ELO genes in S. cerevisiae
The candidate genes for ELO and DES were extracted from the Parietichytrium draft genome database by local BLAST using previously known ELO/DES genes as query sequences. The ORFs of the putative ELO (ELO-1, ELO-2, and ELO-3) and DES (DES-1, DES-2, DES-3, DES-4, DES-5, and DES-6) were amplified by PCR using cDNA or genomic DNA of Parietichytrium, and inserted into the MCS of pYES2/CT (Invitrogen). The ω3DES of S. diclina33 and M. alpina50 were obtained from genomic DNA and cDNA, respectively, by PCR, and inserted into pYES2/CT. The expression vectors were introduced into S. cerevisiae INVSc1 (Invitrogen) using the lithium acetate method51. The transformants were selected by plating on synthetic agar plates lacking uracil (SC-ura). S. cerevisiae transformants harboring ELO and DES were cultured in SC-ura medium containing 2% glucose at 25 °C for 1 day, and then cultured for an additional 1 day in SC-ura medium containing 2% galactose and 50 μM fatty acids (substrate), as described in Supplementary Fig. S1a. The cells were collected by centrifugation at 2000 × g for 10 min and lyophilized. The fatty acid profiles were obtained by GC analysis. ELO or DES activity was expressed as follows: activity (%) = product area × 100/(substrate area + product area).
Quantitative real-time PCR
Real-time PCR was performed using an Mx3000P qPCR System (Agilent Technologies) with TB Green Premix Ex Taq II (Tli RNaseH Plus) (Takara-Bio) using cDNA from Parietichytrium (2 days culture) as a template. The copy numbers of three ELO and six DES were measured using plasmids containing the ORF of each gene as a standard.
Disruption of genes in Parietichytrium sp. and T. aureum
The genes of Δ4DES and C20ELO in Parietichytrium sp., and Δ4DES and PUFA-S in T. aureum were disrupted by homologous recombination using targeting constructs (Supplementary Fig. S13) as described in the previous report35. The linearized knockout targeting constructs containing antibiotic resistance gene were obtained by PCR, introduced into thraustochytrids with a PDS-1000/He (BioRad) using 1550 psi (for Parietichytrium sp.) or 1100 psi (for T. aureum) rupture disks, and then cultured on PDA plate containing appropriate antibiotics. After transfection, ∆4DES and C20ELO KO mutants of Parietichytrium sp. SEK358 (Supplementary Fig. S13a, b) were selected on PDA medium containing 0.5 mg/mL of G418. Two different selection markers were used for disruption of the target genes of Parietichytrium sp. SEK364 and T. aureum (Supplementary Fig. S13c–e). After transfection of C20ELO knockout construct containing NeoR (Supplementary Fig. S13c), Parietichytrium sp. SEK364 was cultured on 2 mg/mL G418, then first allele knockout strain was selected by PCR. For the disruption of second allele of C20ELO of Parietichytrium sp. SEK364, hygromycin resistance gene (HygR) containing knockout construct (Supplementary Fig. S13c) was introduced into first allele knockout strain, then cultured on 2 mg/mL hygromycin containing PDA. The first allele of Δ4DES in T. aureum was replaced with the knockout construct containing the Blasticidin S resistance gene (BlaR) expression cassette (first allele KO construct) and selected on PDA medium containing 0.2 mg/mL of Blasticidin S (Supplementary Fig. S13d). The second allele was replaced with the construct containing the GFP-fused zeocin resistance gene (ZeoR) expression cassette (second allele KO construct) and selected on PDA medium containing 20 mg/mL of zeocin (Supplementary Fig. S13d). In addition to zeocin resistance, fluorescence derived from GFP-ZeoR was used as indicator to select transfectants because of low antibiotic efficiency of zeocin. The first allele of PUFA-S in T. aureum was replaced with the disruption construct containing NeoR expression cassette (first allele KO construct, Supplementary Fig. S13e) and selected on PDA medium containing 2 mg/mL of G418. The second allele was replaced with the construct containing the hygromycin resistance gene (HygR) expression cassette (second allele KO construct, Supplementary Fig. S13e) and selected on PDA medium containing 2 mg/mL of hygromycin, generating the PUFA-S KO strain. The PUFA-S-disrupted and Δ4DES-disrupted mutant (PUFA-S/D4DES KO) was obtained by introducing the Δ4DES knockout construct into the PUFA-S KO strain. The mutants were selected by PCR, and the gene disruption of mutants was confirmed by Southern blot analysis using WT DNA as a control.
Development of a high-expression promoter for Parietichytrium sp
To overexpress ω3DES, which is a bottleneck for EPA or n-3DPA production by Parietichytrium, we developed high-expression promoters in this study. SmDNAV is a double-stranded DNA virus infecting S. minutum34. To identify the immediate early genes of SmDNAV, DNA microarray was performed in SmDNAV-infected S. minutum treated with or without cycloheximide. The 1 kbp upstream sequence of immediate early genes was isolated from the genomic DNA of SmDNAV and inserted upstream of NeoR. Then, the promoter activity was assessed by the transformation efficiency or mRNA expression level of NeoR, which was quantified by real-time PCR. The promoter region demonstrating the highest transcriptional activity was used as the SmDNAV-derived promoter (Smp1P) in this study.
Generation of ω3DES-overexpressing mutants
The codon optimized ω3DES gene of S. diclina was chemically synthesized (Biomatik), and inserted between the Smp1 or EF1α promoter and ubiquitin terminator (Supplementary Fig. S6d). The linearized expression construct was prepared by PCR using the plasmid as a template, and then introduced into Δ4DES KO and C20ELO KO strains of Parietichytrium sp. by biolistic transformation with a PDS-1000/He (BioRad). C20ELO KO/ω3DES OE was selected by 1 mg/mL of Blasticidin-containing PDA. ∆4DES KO/ω3DES OE was selected by 1 mg/mL hygromycin-containing PDA.
Tracing the metabolism of stable isotope-labeled fatty acids
Deuterium-labeled palmitic acid (d31-C16:0) and 13C18-labeled oleic acid (13C18-C18:1) were purchased from Cambridge Isotope Laboratories, Inc. 13C18-C18:1 was added to the medium at a final concentration of 0.25 mM. Cells were collected 4, 8, 24, and 48 h after adding stable isotope-labeled fatty acids. The cell pellet and supernatant were separated by centrifugation at 5,000 × g for 3 min. The cell pellet was dissolved in 200 µL of distilled water. Then, the pellet was crushed at 3000 rpm for 60 s using a bead beater (μT-12, TAITEC) with glass beads (diameter: 0.4 mm, AS ONE Corp.) and kept on ice for 60 s. This procedure was repeated three times to prepare the cell lysate. Cellular lipids were extracted from 90 µL of cell lysate by adding 375 μL of chloroform/methanol (1:2, v/v) containing 50 µM internal standard (C12:0). For saponification, 10 µL of 4 M KOH was added. After incubation at 37 °C for 3 h, 10 µL of 2 M acetic acid, 125 µL of chloroform, and 125 µL of water were added, and the organic phase was separated by centrifugation at 17,500 × g for 3 min. Then, 150 μL of the organic phase was transferred to vials, and the stable isotope-labeled fatty acids were measured using LC-ESI MS/MS (3200 QTRAP, SCIEX) equipped with an InertSustain C18 column (2.1 × 150 mm, 5 μm, GL Sciences) using a gradient starting with 10% solvent B (methanol with 2.5 mM ammonium acetate) in solvent A (distilled water with 2.5 mM ammonium acetate) and reaching 90% solvent B for 1 min, followed by 95% solvent B for 15 min. The column was equilibrated for 5 min before the next run. Isotope-labeled fatty acids were detected by MRM or pseudo-MRM, for which the conditions are described in Supplementary data set 1.
LC-ESI MS/MS to determine the lipid composition
The cell pellet and supernatant were separated by centrifugation at 5000 × g for 3 min, and the DCW was measured after the pellet was lyophilized. The 2 mg of dried cells was dissolved in 300 μL of chloroform/methanol (2:1, v/v) containing 10 μM PC22:0 (11:0/11:0), 10 μM LPC13:0, and 20 μM TG36:0 (12:0/12:0/12:0) as internal standards, and then crushed by sonication for 60 s and kept on ice for 60 s. This procedure was repeated three times. After incubation at 25 °C for 30 min, the supernatant and cell debris were separated by centrifugation at 17,500 × g for 3 min. The 240 μL of supernatant was mixed with 60 μL of water, and then centrifuged at 17,500 × g for 3 min. Forty microliters of the organic phase was transferred into autoinjector vials containing 960 μL of 2-propanol, and then the cellular lipids were measured by LC-ESI MS/MS using a high-performance liquid chromatography system (1200 series, Agilent Technologies) equipped with a mass spectrometer (3200 QTRAP LC-MS/MS system). A binary solvent gradient with a flow rate of 200 μL/min was used to separate phospholipids and neutral lipids by reversed-phase chromatography using an InertSustain C18 column (2.1 × 150 mm, 5 μm)39,52. The gradient was started with 0% solvent B1 (2-propanol with 0.1% formic acid and 0.028% ammonia) in solvent A1 (acetonitrile/methanol/water, 19:19:2, v/v/v containing 0.1% formic acid and 0.028% ammonia) and was maintained for 3 min. The gradient reached 40% B for 24 min, then 50% B for 26 min, and was maintained for 3 min. The gradient was returned to the starting conditions for 1 min and the column was equilibrated for 7 min before the next run. LPC, PC, LPE, PE, diacylglycerol (DAG), and TAG containing C16:0, C18:0, C18:1, C18:2, C18:3, C20:3, C20:4, C20:5, C22:4, C22:5, and C22:6 fatty acids were detected using MRM. The peak intensity of each glycerolipid species was analyzed using MultiQuant 3.0.1 (SCIEX). MRM conditions are described in Supplementary data set 1.
Fed-batch culture of Parietichytrium sp
Mutant strains of Parietichytrium sp. SEK364 (C20ELO KO and C20ELO KO/ω3DES OE) Parietichytrium sp. SEK358 (Δ4DES KO and Δ4DES KO/ω3DES OE) were precultured in 500-mL flasks containing 100 mL of GY medium at 28 °C with shaking at 120 rpm. After glucose consumption, the cells were transferred to a Small Scale Bioreactor BMP-Z (ABLE Co. & Biott Co., Ltd) equipped with a 1 L vessel containing 400 mL of GY medium, the temperature and pH of which were maintained at 28 °C and 6.8 ± 0.2, respectively. The DO level in the medium was maintained above 10% by adjusting the agitation speed (from 600 to 1200 rpm) and air flow rate (from 0.25 to 0.5 L/min). The glucose concentration was continuously monitored and maintained below 60 g/L throughout cultivation by adding feed solution (50% glucose). Samples were collected at the time points indicated in Fig. 3, and the DCW, TFA, EPA, and n-3DPA levels were measured by the method described in “Methods” section.
Statistics and reproducibility
The results are expressed as mean ± SD.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
