Construction of Eicosatetraenoic Acid Producing Cell Factory by Genetic Engineering of Mucor circinelloides

: Eicosatetraenoic acid


Introduction
Polyunsaturated fatty acids (PUFAs), particularly very-long-chain polyunsaturated fatty acids (VLCPUFAs) with chain lengths varying from C20 to C22 and containing more than two double bonds, play an important role in regulating various physiological processes, such as hemostasis (blood clotting), reproduction, immune and inflammatory responses [1]. ω-3 PUFAs, one kind of PUFA, are essential to human nutrition and health [2]. Since ω-3 PUFAs cannot be de novo synthesized by mammals, they must be obtained through diet. Due to the vital role of eicosapentaenoic acid (EPA, 20:5 ω-3) and docosahexaenoic acid (DHA, 22:6 ω-3) in human health, they received growing attention and were well-studied in recent years [3,4]. In addition, stearidonic acid (SDA, 18:4 ω-3) and docosapentaenoic acid (DPA, 22:5 ω-3) have been developed and utilized [5][6][7][8]. However, efficient genetic transformation methods [22,23]. In recent years, considerable research work has been carried out to modify this fungus to produce high-value PUFAs. Yang et al. [5] overexpressed the D15D gene to construct an SDA-producing cell factory, which was reported to convert LA to ALA [24] and also GLA to SDA in M. circinelloides WJ11. Meanwhile, the other strain, M. circinelloides CBS 277.49, was also used to construct a DGLA-producing cell factory by overexpressing D6E (GLELO) gene that converts GLA to DGLA [25]. Therefore, SDA and DGLA-producing strains were constructed by overexpressing delta-15 desaturase and D6E(GLELO) genes with a yield of 340 mg/L and 74.61 mg/L, respectively. Furthermore, SDA can be catalyzed by D6E(GLELO) gene to produce ETA, and DGLA can be catalyzed by delta-17 desaturase to produce ETA [26]. Therefore, these mutants have great potential for the production of VLPUFAs.
Oomycetes such as Pythium and Phytophtora are known to produce VLCPUFAs (e.g., ARA and EPA [27]). Fatty acid profiles of these organisms suggested that they produce these VLCPUFAs through the desaturase/elongase pathway rather than the polyketide pathway used by marine bacteria and thraustochytrids [28]. Therefore, in this study, we determined the fatty acid composition of the recombinant strains by overexpressing two different heterologous delta-17 desaturase genes (one from P. aphanidermatum and the other from P. parasitica) in M. circinelloides. To the best of our knowledge, this is the first study on producing ETA by overexpressing the delta-17 desaturase gene in M. circinelloides, which provided alternative routes for the production of very long-chain PUFAs.

Strains, Transformation, and Fermentation Conditions
The delta-17 desaturase genes from P. aphanidermatum and P. parasitica were synthesized by Shanghai Biological Engineering Co., Ltd(Shanghai, China). The DGLA-producing leucine-auxotrophic strain SD14, which contained the delta-15 desaturase gene and delta-6 elongase (D6E) gene of M. alpina and was derived from M. circinelloides WJ11 in our lab (Junhuan Yang, unpublished results), was used as the recipient strain in transformation experiments to overexpress the delta17-desaturase genes. Escherichia coli Top10 was used for plasmid construction, preservation, and propagation and grown in Luria-Bertani (LB) medium contained (10 g/L tryptone, 5 g/L yeast extract, 10 g/L NaCl, and 20 g/L agar) at 37 °C, shaking at 200 rpm, and supplemented with kanamycin or ampicillin (100 mg/mL) when necessary [17,29,30]. Plasmid pCRC176 and pCRC177 were used as PaD17 and PpD17 overexpression host vectors generated from pMAT2076, respectively. ω-3 desaturases (e.g., delta-17 desaturase) are necessary for the desaturation of ω-6 PUFAs into ω-3 counterparts and play an important role in ETA synthesis from DGLA [9] ( Figure 1). Since these enzymes play an essential role in the synthesis of long-chain ω-3 PUFAs, genetic engineering of micro-organisms to produce ω-3 PUFAs through the ω-3 pathway requires the ω-3 desaturases identified and characterized from numerous sources. Researchers demonstrated that the ω-3 desaturases from plants preferentially desaturated C18 (ω-6 PUFAs) compared with the C20 substrates [12,13]. Some ω-3 desaturases have recently been found to have high activity as delta-17 desaturases and are favored for C20 substrates. The ω-3 desaturase (SDD17) from Saprolegnia diclina was found to exclusively desaturate C20 PUFAs with a remarkable preference for arachidonic acid (ARA, 20:4 ω-6) [14]. Another ω-3 desaturase (PID17) from Phytophthora infestans could also convert 31% of ARA to EPA [15]. In addition, Xue et al. [2] found that three ω-3 desaturases (PaD17 from Pythium aphanidermatum, PsD17 from Phytophthora sojae, PrD17 from Phytophthora ramorum) had strong delta-17 desaturase activity and PaD17 had the highest conversion rate of ARA to EPA. When the PaD17 gene was transformed into Mortierella alpina the conversion rate of ARA increased by 49.7% in the previous study [16], but the conversion rate of DGLA was only 14%. Similarly, the PpD17 gene from P. parasitica was transformed into M. alpina and converted the ARA to EPA with a substrate conversion rate of 70% [17], and the conversion rate of DGLA was almost 25%. Thus, overexpression of delta-17 desaturase could upregulate the biochemical pathway of EPA production in fungi.
The oleaginous filamentous fungus, Mucor circinelloides, is a model organism to study the mechanism of lipid accumulation. It has an innate ability to accumulate more than 36% intracellular lipids and attracts considerable interest. Meanwhile, it has the enzyme system to synthesize high content of γ-linolenic acid (GLA, C18:3) [18], which has many beneficial effects for humans [19,20], making this strain a good model micro-organism for GLA production. Furthermore, this fungus has explicit genetic background [21] and efficient genetic transformation methods [22,23]. In recent years, considerable research work has been carried out to modify this fungus to produce high-value PUFAs. Yang et al. [5] overexpressed the D15D gene to construct an SDA-producing cell factory, which was reported to convert LA to ALA [24] and also GLA to SDA in M. circinelloides WJ11. Meanwhile, the other strain, M. circinelloides CBS 277.49, was also used to construct a DGLA-producing cell factory by overexpressing D6E (GLELO) gene that converts GLA to DGLA [25]. Therefore, SDA and DGLA-producing strains were constructed by overexpressing delta-15 desaturase and D6E(GLELO) genes with a yield of 340 mg/L and 74.61 mg/L, respectively. Furthermore, SDA can be catalyzed by D6E(GLELO) gene to produce ETA, and DGLA can be catalyzed by delta-17 desaturase to produce ETA [26]. Therefore, these mutants have great potential for the production of VLPUFAs.
Oomycetes such as Pythium and Phytophtora are known to produce VLCPUFAs (e.g., ARA and EPA [27]). Fatty acid profiles of these organisms suggested that they produce these VLCPUFAs through the desaturase/elongase pathway rather than the polyketide pathway used by marine bacteria and thraustochytrids [28]. Therefore, in this study, we determined the fatty acid composition of the recombinant strains by overexpressing two different heterologous delta-17 desaturase genes (one from P. aphanidermatum and the other from P. parasitica) in M. circinelloides. To the best of our knowledge, this is the first study on producing ETA by overexpressing the delta-17 desaturase gene in M. circinelloides, which provided alternative routes for the production of very long-chain PUFAs.

Strains, Transformation, and Fermentation Conditions
The delta-17 desaturase genes from P. aphanidermatum and P. parasitica were synthesized by Shanghai Biological Engineering Co., Ltd(Shanghai, China). The DGLA-producing leucine-auxotrophic strain SD14, which contained the delta-15 desaturase gene and delta-6 elongase (D6E) gene of M. alpina and was derived from M. circinelloides WJ11 in our lab (Junhuan Yang, unpublished results), was used as the recipient strain in transformation experiments to overexpress the delta17-desaturase genes. Escherichia coli Top10 was used for plasmid construction, preservation, and propagation and grown in Luria-Bertani (LB) medium contained (10 g/L tryptone, 5 g/L yeast extract, 10 g/L NaCl, and 20 g/L agar) at 37 • C, shaking at 200 rpm, and supplemented with kanamycin or ampicillin (100 mg/mL) when necessary [17,29,30]. Plasmid pCRC176 and pCRC177 were used as PaD17 and PpD17 overexpression host vectors generated from pMAT2076, respectively. Subsequently, the recombinant strains Mc-PaD17 and Mc-PpD17 were generated by overexpressing the PaD17 and PpD17 genes from SD14. The fungus was grown at 28 • C in YPG medium (20 g/L glucose, 10 g/L peptone, 3 g/L yeast extract, and 20 g/L agar) [31] or in minimal leucine media YNB contained (10 g/L glucose, 1.5 g/L ammonium sulfate, 1.5 g/L L-glutamic acid, 0.5 g/L yeast nitrogen base without amino acids and ammonium sulfate, and 20 g/L agar) [32,33], which were adjusted to pH 4.5 or 3.2 for mycelial or colonial growth separately, and supplemented with leucine (0.4 g/L) when required. Transformation and selection procedures were carried out as previously described [34,35].

Plasmids Construction
Plasmid pMAT2076 (Supplementary Figure S1), which was used to overexpress the target genes in M. circinelloides, was constructed previously by our laboratory [37,38]. It contained a selectable marker (LeuA) and a strong promoter (Pzrt1) flanked by up-and down-stream sodit-a sequences. The DNA fragments of the delta-17 desaturase gene from P. aphanidermatum and P. parasitica were obtained by PCR amplification using the primers D17Pa-F/R and D17Pp-F/R (Table S1), respectively, which contained 25 bp homologous sequences on both sides of linearized pMAT2076. Subsequently, these PCR fragments were inserted into the linearized vector pMAT2076 digested with XhoI restriction endonuclease to generate plasmid pCRC176 (contained a PaD17 gene) and pCRC177 (contained a PpD17 gene) using the One Step cloning kit (Takara) ( Figure S1).

Cell Dry Weight, Lipid Extraction and Analysis
The fungal cell biomass was collected after filtration through a Buchner funnel, washed three times with distilled water, and then frozen overnight at −80 • C and then freezedried for 48 h. The cell dry weight (CDW) of each sample was calculated gravimetrically. Subsequently, the CDW of every sample was measured. Lipid was extracted using the Folch [39] method with some modifications. About 10~15 mg of dry mycelia was digested with the 4 M HCl, and then chloroform/methanol (2:1, v/v) was added. The internal standard was pentadecanoic acid (15:0). Transfer the lower layer to a new lipid extraction bottle and blow dry with nitrogen, and then 10% H 2 SO 4 /methanol (w/w) was added for methylation. The fatty acid methyl esters (FAMEs) were isolated with n-hexane (HPLC grade), and then the fatty acid profile was analyzed by gas chromatography (GC) following the manufacturer's instructions. The GC reaction condition was as follows: ramp from 80 • C to 160 • C at 10 • C/min, ramp to 200 • C at 5 • C/min, then ramp to 230 • C at 5 • C/min, and hold for 12 min. Finally, the identification and quantification of individual chromatographic peaks were carried out by comparison to the external FAME standard mixture (SupelcoR 37 Component FAME Mix).

Determination of Glucose and Ammonium Concentrations during the Fermentation
The concentrations of glucose and ammonium in the fermentation medium were determined by using a glucose oxidase electrode biosensor (SBA-40E, Institute of Biology, Shandong Academy of Sciences, Jinan, China) and the indophenol method as described [40,41], respectively.

Genomic DNA Extraction, RNA Isolation and RT-qPCR Analysis
The genomic DNA from reconstitution strains, which were incubated for 48 h in YPG medium at 28 • C, were extracted using the DNA Quick Plant System kit (Tiangen Biotech Co., Ltd., Beijing, China). For RNA isolation and real-time quantitative PCR (RT-qPCR) analysis, the fungus was cultivated for 96 h with a modified K&R medium at 28 • C, and then the fresh mycelium was harvested at 6, 12, 24, and 72 h. The total RNA of each sample was extracted with Trizol after crumbling under liquid nitrogen. Afterward, the total RNA was reverse-transcribed to cDNA by Evo M-MLV RT Mix Kit with gDNA Clean for qPCR (Accurate Biotech Co., Ltd., Changsha, China) as described in the instructions. The relative transcription levels of desaturase genes in these strains were detected using SYBR Green Premix Pro Taq HS qPCR Kit in LightCycler 96 (Rotkreuz, Switzerland) according to the manufacturer's introduction. The specific primers for RT-qPCR were listed in Supplementary Table S1, and the actin gene of M. circinelloides was seen as the housekeeping gene. All data were employed by the 2 −∆∆Ct method [42]. Three independent biological replicated experiments are used for data analysis.

Statistical Analysis
All experimental data were analyzed by the Student's t-test of IBM SPSS 21.0. Each experiment contained three replicates, and the results were presented as Means ± SD. p < 0.05 was considered a significant difference.

Generation of Delta-17 Overexpressing Strains of M. circinelloides by Genetic Engineering
A delta-17 desaturase (GenBank accession No. FW362186.1) from P. aphanidermatum, and a delta-17 desaturase (GenBank accession No. KT372001) from P. parasitica were investigated in this study. To identify the effect of these exogenous genes on fatty acid synthesis, single overexpression recombinant strains were constructed, and the target genes were identified by PCR analysis. For each overexpression plasmid, two independent transformants carrying delta-17 desaturase gene were selected, named Mc-PaD17-1, Mc-PaD17-2, Mc-PpD17-1, and Mc-PpD17-2, respectively. Amplification was carried out using a primer pair (sodit-a-F/R) listed in Table S1. Bands approximately 7.

Genomic DNA Extraction, RNA Isolation and RT-qPCR Analysis
The genomic DNA from reconstitution strains, which were incubated for 48 h in YPG medium at 28 °C, were extracted using the DNA Quick Plant System kit (Tiangen Biotech Co., Ltd., Beijing, China). For RNA isolation and real-time quantitative PCR (RT-qPCR) analysis, the fungus was cultivated for 96 h with a modified K&R medium at 28 °C, and then the fresh mycelium was harvested at 6, 12, 24, and 72 h. The total RNA of each sample was extracted with Trizol after crumbling under liquid nitrogen. Afterward, the total RNA was reverse-transcribed to cDNA by Evo M-MLV RT Mix Kit with gDNA Clean for qPCR (Accurate Biotech Co., Ltd., Changsha, China) as described in the instructions. The relative transcription levels of desaturase genes in these strains were detected using SYBR Green Premix Pro Taq HS qPCR Kit in LightCycler 96 (Rotkreuz, Switzerland) according to the manufacturer's introduction. The specific primers for RT-qPCR were listed in Supplementary Table S1, and the actin gene of M. circinelloides was seen as the housekeeping gene. All data were employed by the 2 −∆∆Ct method [42]. Three independent biological replicated experiments are used for data analysis.

Statistical Analysis
All experimental data were analyzed by the Student's t-test of IBM SPSS 21.0. Each experiment contained three replicates, and the results were presented as Means ± SD. p < 0.05 was considered a significant difference.

Generation of Delta-17 Overexpressing Strains of M. circinelloides by Genetic Engineering
A delta-17 desaturase (GenBank accession No. FW362186.1) from P. aphanidermatum, and a delta-17 desaturase (GenBank accession No. KT372001) from P. parasitica were investigated in this study. To identify the effect of these exogenous genes on fatty acid synthesis, single overexpression recombinant strains were constructed, and the target genes were identified by PCR analysis. For each overexpression plasmid, two independent transformants carrying delta-17 desaturase gene were selected, named Mc-PaD17-1, Mc-PaD17-2, Mc-PpD17-1, and Mc-PpD17-2, respectively. Amplification was carried out using a primer pair (sodit-a-F/R) listed in Table S1. Bands approximately 7.

Expression Levels of Delta-17 Desaturase Gene in Strains
The relative mRNA levels of the delta-17 desaturase gene from P. aphanidermatum and P. parasitica were analyzed in the selected recombination strains grown in a 1.5 L fermenter incubated with K&R medium by RT-qPCR. As shown in Figure 3, it was obvious that delta-17 desaturase gene expression levels in recombinant strains increased rapidly from 6 to 12 h and then gradually declined during the fermentation process. The expression levels of the delta-17 desaturase genes increased to 18.9-and 72.7-folds at 12

Expression Levels of Delta-17 Desaturase Gene in Strains
The relative mRNA levels of the delta-17 desaturase gene from P. aphaniderm and P. parasitica were analyzed in the selected recombination strains grown in a 1.5 L menter incubated with K&R medium by RT-qPCR. As shown in Figure 3, it was obv that delta-17 desaturase gene expression levels in recombinant strains increased rap from 6 to 12 h and then gradually declined during the fermentation process. The exp sion levels of the delta-17 desaturase genes increased to 18.9-and 72.7-folds at 12 h in PaD17 and Mc-PpD17, respectively. The RT-qPCR results suggested that the PaD17 and PpD17 gene were transcribed in Mc-PaD17 and Mc-PpD17, respectively.

Cell Growth and Lipid Accumulation in Delta-17 Desaturase Gene Overexpressing Stra
To explore the effect of the delta-17 desaturase genes on lipid biosynthesis in M cinelloides, cell growth and lipid accumulation in Mc-PaD17 and Mc-PpD17 were m ured, the fresh mycelium was harvested in a 1.5 L fermenter cultured with modified medium at 96 h of cultivation. From Figure 4A, the results showed that the cell gro patterns of these three strains were impacted by the delta-17 desaturase genes. Su quently, the cell proliferation of these overexpressing strains exhibited a continuous gro trend in Figure 4A, and the cell lipid content achieved the peak at 96 h in Figure 4D.
The cell growth curve showed that the cell proliferation was slow before 24 h in PaD17, while the Mc-PpD17 stain did not show a significant difference compared wit control strain (Mc-2076) ( Figure 4A). After that, the cell proliferation of the PaD17 ov pressing strain was increased by 1.4-fold (up to 25.98 g/L) compared to the control. T the lipid content was increased by about 23.52% (from 23.38% to 28.88%) compared t control. Simultaneously, the overexpression of the PpD17 gene promoted cell repro tion, and the lipid-free CDW improved by 33.43% (Figure S2), and the lipid content increased by 32.38% at 96 h, compared with the control strain. For the three strains consumption trends of nitrogen were similar ( Figure 4C). After 12 h fermentation, am

Cell Growth and Lipid Accumulation in Delta-17 Desaturase Gene Overexpressing Strains
To explore the effect of the delta-17 desaturase genes on lipid biosynthesis in M. circinelloides, cell growth and lipid accumulation in Mc-PaD17 and Mc-PpD17 were measured, the fresh mycelium was harvested in a 1.5 L fermenter cultured with modified K&R medium at 96 h of cultivation. From Figure 4A, the results showed that the cell growth patterns of these three strains were impacted by the delta-17 desaturase genes. Subsequently, the cell proliferation of these overexpressing strains exhibited a continuous growth trend in Figure 4A, and the cell lipid content achieved the peak at 96 h in Figure 4D.
The cell growth curve showed that the cell proliferation was slow before 24 h in Mc-PaD17, while the Mc-PpD17 stain did not show a significant difference compared with the control strain (Mc-2076) ( Figure 4A). After that, the cell proliferation of the PaD17 overexpressing strain was increased by 1.4-fold (up to 25.98 g/L) compared to the control. Thus, the lipid content was increased by about 23.52% (from 23.38% to 28.88%) compared to the control. Simultaneously, the overexpression of the PpD17 gene promoted cell reproduction, and the lipid-free CDW improved by 33.43% (Figure S2), and the lipid content was increased by 32.38% at 96 h, compared with the control strain. For the three strains, the consumption trends of nitrogen were similar ( Figure 4C). After 12 h fermentation, ammonium was depleted in these three strains. Since lipid accumulation is correlated with the amount of nitrogen [43], these strains started to accumulate lipids when the nitrogen was depleted (at 12 h) from the growth media. Before 24 h of incubation, overexpression of delta-17 desaturase genes enhanced the glucose homogenization, and the lipid accumulation of the Mc-PaD17 was increased by about 48.90 (from 7.71% to 11.48%), while the lipid accumulation of the Mc-PpD17 was increased to 10.43%, an increase of 35.28% compared with Mc-2076 ( Figure 4D). In conclusion, the overexpression of the heterologous genes had a significant effect on cell growth and lipid content.  Figure 4D). In conclusion, the overexpression of the heterologous genes had a significant effect on cell growth and lipid content.

ETA Accumulation in Delta-17 Desaturase Gene Overexpressing Strains
FAMEs analysis suggested that fatty acid composition was similar in the delta-17 overexpressing strain and the control strain (Mc-PpD17 and Mc-2076, respectively), except in the case of ETA had no significant accumulation in the control strain. In Mc-PpD17, the ETA percentage in the total lipid content was higher at the beginning of fermentation, and there was a decreasing trend with the progress of time. However, the total ETA yield started to increase over time, and the total ETA yield was calculated based on the total fatty acids (TFAs) content. Considering that the maximum ETA yields of 114.69 mg/L were detected, and the total fatty acids were raised to 30.  Table 1. Taken together, the findings of fatty acid composition demonstrated that overexpression of delta-17 desaturase genes induced significant changes in other fatty acid contents compared with the control strain.

ETA Accumulation in Delta-17 Desaturase Gene Overexpressing Strains
FAMEs analysis suggested that fatty acid composition was similar in the delta-17 overexpressing strain and the control strain (Mc-PpD17 and Mc-2076, respectively), except in the case of ETA had no significant accumulation in the control strain. In Mc-PpD17, the ETA percentage in the total lipid content was higher at the beginning of fermentation, and there was a decreasing trend with the progress of time. However, the total ETA yield started to increase over time, and the total ETA yield was calculated based on the total fatty acids (TFAs) content. Considering that the maximum ETA yields of 114.69 mg/L were detected, and the total fatty acids were raised to 30 The values represent the mean ± SD of three independent experiments.

Effect of Overexpression of Delta-17 Desaturase Genes on the Transcription Level of Key Genes for Fatty Acid Biosynthesis
To further understand the effects of overexpressed delta-17 desaturase genes in M. circinelloides, the relative expression levels of the key enzymes associated with lipid biosynthesis were determined at 24 h by RT-qPCR when the fatty acids were accumulated rapidly. There were two genes (cme1, gene ID: scaffold00036.12; cme2, and gene ID: scaffold00049.37) that encoded the malic enzyme in cytosol (cme), and three genes (mme1, gene ID: scaffold000188.29; mme2, and gene ID: scaffold000295.10, mme3, and gene ID: scaffold00014.40) located in the mitochondria for encoding the malic enzyme (ME), which converted malate to pyruvate and were assumed to provide reducing power NADPH for the synthesis of fatty acids [29,44,45]. The results demonstrated that the cme1, cme2, mme2, and mme3 expression levels were significantly up-regulated in Mc-PpD17, whereas the transcription level of cme2 was increased in Mc-PaD17 ( Figure 5A-E). The genes (fas1, gene ID: scaffold0002.57; fas2, gene ID: scaffold000111.12) encoded the cytosolic fatty acid synthase (fas), which were the key enzymes in the fatty acid synthesis pathway and probably catalyzed de novo fatty acid synthesis [29]. As can be observed from Figure 5G, the expression level of the fas2 was significantly increased in overexpressing delta-17 desaturase strains (Mc-PaD17 and Mc-PpD17) compared with the control strain. In addition, the genes (acc1, gene ID: scaffold00021.30; acc2, gene ID: scaffold00023.50) encoding acetyl-CoA carboxylase (ACC), which could catalyze the carboxylation of acetyl-CoA to produce malonic acid monoacyl-CoA, and was an intermediate metabolite that played an important role in the fatty acid synthesis and metabolism [46]. The results show that the transcription levels of acc1 were significantly up-regulated in Mc-PpD17 in Figure 5H, while acc1 and acc2 had no significant difference in Mc-PaD17 in Figure 5H,I. Simultaneously, the delta-6 elongase enzyme (D6E, GLELO), which catalyzes GLA elongation to produce DGLA [47], which was the precursor of ETA, was analyzed. As shown in Figure 5J, the expression level of the D6E (GLELO) was up-regulated significantly in Mc-PaD17 and Mc-PpD17, compared with the control strain (Mc-2076). Therefore, the RT-qPCR results demonstrated that the overexpression of delta-17 desaturase genes could manipulate the expression of cme2, fas2, and D6E and thereby contribute to altering lipid biosynthesis in M. circinelloides.

Discussion
Various micro-organisms, fungi in particular, can effectively synthesize and accumulate VLCPUFAs through desaturation and elongation pathways [1]. Although the biosynthesis of ω-3 VLCPUFAs has been successfully reconstituted in some oleaginous fungi using desaturases and elongases, the desirable level and composition of certain fatty acids have not been fully achieved [48]. Production of specialty fatty acids in the storage lipids involves many coordinated biochemical processes, and each process involves many intricate enzymatic reactions [49]. The oleaginous fungus, M. circinelloides, has been used as the model to investigate lipid biosynthesis for more than 30 years [50]. Meanwhile, in M. circinelloides, SDA and DGLA were synthesized by overexpressing delta-15 desaturase and D6E(GLELO) gene [5,25]. On the other hand, P. aphandermatum was reported to convert DGLA to ARA through delta-5 desaturase; in addition, the conversion of ARA to EPA and DGLA to ETA is an important step process catalyzed by delta-17desaturase enzyme [2]. Similarly, the P. parasitica strain can also produce ARA, ETA, and EPA.
Genetic engineering of micro-organisms to produce EPA, DHA, and ETA through the desaturase/elongase pathway requires the overexpression of diverse pathway enzymes [9]. Nevertheless, whether the delta-6 or delta-8 pathway is used, a ω-3 desaturase is essential for the conversion of ω-6 PUFAs into their ω-3 counterparts [2]. Because of the fundamental role delta-17 desaturase enzymes play in the synthesis of long-chain ω-3 PUFAs, there has been a significant effort to determine and characterize these enzymes from different sources [17]. The DGLA-and SDA-producing cell factories have been

Discussion
Various micro-organisms, fungi in particular, can effectively synthesize and accumulate VLCPUFAs through desaturation and elongation pathways [1]. Although the biosynthesis of ω-3 VLCPUFAs has been successfully reconstituted in some oleaginous fungi using desaturases and elongases, the desirable level and composition of certain fatty acids have not been fully achieved [48]. Production of specialty fatty acids in the storage lipids involves many coordinated biochemical processes, and each process involves many intricate enzymatic reactions [49]. The oleaginous fungus, M. circinelloides, has been used as the model to investigate lipid biosynthesis for more than 30 years [50]. Meanwhile, in M. circinelloides, SDA and DGLA were synthesized by overexpressing delta-15 desaturase and D6E(GLELO) gene [5,25]. On the other hand, P. aphandermatum was reported to convert DGLA to ARA through delta-5 desaturase; in addition, the conversion of ARA to EPA and DGLA to ETA is an important step process catalyzed by delta-17desaturase enzyme [2]. Similarly, the P. parasitica strain can also produce ARA, ETA, and EPA.
Genetic engineering of micro-organisms to produce EPA, DHA, and ETA through the desaturase/elongase pathway requires the overexpression of diverse pathway enzymes [9]. Nevertheless, whether the delta-6 or delta-8 pathway is used, a ω-3 desaturase is essential for the conversion of ω-6 PUFAs into their ω-3 counterparts [2]. Because of the fundamental role delta-17 desaturase enzymes play in the synthesis of long-chain ω-3 PUFAs, there has been a significant effort to determine and characterize these enzymes from different sources [17]. The DGLA-and SDA-producing cell factories have been constructed by overexpressing D6E (GLELO) gene and D15D desaturase gene in M. circinelloides, respectively [6,25]. To explore the effect of the delta-17 desaturase gene in governing fatty acid synthesis in oleaginous fungi, therefore, in this study, we constructed the delta-17 desaturase gene overexpression strains by genetic engineering strategies (Figure 2). Our findings suggested that overexpressing delta-17 desaturase genes significantly enhanced the efficiency of carbon utilization in the nitrogen-limited broth in the early fermentation stage (24 h) and caused an increase in lipid accumulation significantly increased by 23.52% and 32.38% in Mc-PaD17 and Mc-PpD17, respectively, compared to that of the control strain ( Figure 4B,D). These results are in agreement with the previous study, which demonstrated the overexpression of the ω-3 desaturases gene in M. alpina resulted in a significant increase in the total lipid content (up to 6.46 g/L) over the parent strain [16].
Moreover, the heterologous overexpressing PpD17 desaturase gene in M. circinelloides increased the ETA content up to 114.69 mg/L and improved lipid accumulation. On the other hand, the DGLA, which is considered the precursor for the synthesis of ETA, was significantly decreased in Mc-PpD17. Notably, the conversion rate of DGLA to ETA reached up to 27.93% in Mc-PpD17, which is similar to the conversion rate of DGLA in the transformant of Saccharomyces cerevisiae (overexpressing PpD17 gene) grown on the media supplemented with DGLA [17]. Li et al. increased the copy number of the genes encoding fatty acid desaturase and elongase in Pichia pastoris, resulting in improved arachidonic acid and eicosapentaenoic acid production [51]. Therefore, we can use the same method to increase the content of ETA. Nevertheless, the reviewed data showed that no ETA accumulated by heterologous overexpressing the PaD17 gene in M. circinelloides (Table 1). It was found that the PaD17 desaturase had a stronger substrate preference for ARA rather than DGLA, which might explain the ineffectiveness of PaD17 desaturase in ETA production in M. circinelloides [2]. A previous study demonstrated that the fatty acid composition of S. cerevisiae transformant strains was analyzed and showed that 18-carbon and 20-carbon n-6 PUFAs were accumulated and ω-3 desaturase converted DGLA to ETA when grown in a complete medium supplemented with 0.1% of exogenous (DGLA) fatty acid [52]. Moreover, Rong et al. reported that oRiFADS17 and oObFADS17 possessed strong delta-17 desaturase activity, which exhibited a remarkable increase in desaturation activity on C20 fatty acids compared to C18 fatty acids and finally converted n-6 PUFAs into n-3 PUFAs [53]. Therefore, we speculated that the heterologous overexpressing delta-17 desaturase genes might work under certain unknown conditions in M. circinelloides.
To investigate the impact of delta-17 desaturase overexpression on M. circinelloides, the transcriptional levels of the key genes for fatty acid biosynthesis in the early fermentation stage (24 h) were analyzed by RT-qPCR ( Figure 5). In oleaginous micro-organisms, malic enzyme (ME) regulates fatty acid biosynthesis by providing carbon and NADPH and organic acid synthesis, participating in membrane lipid regeneration, and promoting fruit development in plants [54]. The ME genes from M. circinelloides were overexpressed in M. circinelloides and Rhodotorula glutinis and led to a 2.5-and 2.1-fold increase in lipid accumulation, respectively [55,56]. Our results indicated that the lipid content in the recombinant strain Mc-PpD17 was significantly increased compared to the control strain, which may be caused by the enhanced levels of NADPH generated by the increased expression levels of ME genes ( Figure 5A,B,D,E). Another study reported the transcriptional levels of the acc1 gene in Mc-PpD17 were noticeably higher than that in Mc-2076 and Mc-PaD17 (Figure 5 H). Similarly, the previous study showed that the increased expression of the acc genes in transformants triggered the accumulation of acetyl-CoA, which was the substrate for fatty acid synthesis [57]. Fatty acid synthase (FAS) is considered a key lipogenic enzyme [58] in the endogenous lipogenesis pathway that primarily catalyzes the synthesis of the long-chain saturated fatty acid palmitate from acetyl-CoA and malonyl-CoA and NADPH [37] and the expression of FAS in Saccharomyces cerevisiae triggered short-chain fatty acid production [59]. In this study, it was found that the ex-pression levels of fas2 in the overexpressing strains had significantly increased compared to the control strain ( Figure 5G). To the best of our knowledge, DGLA, which was the elongated product of GLA catalyzed by the enzyme delta-6 elongase (D6E(GLELO)) [50], was the precursor substance for the synthesis of ETA. By overexpressing D6E(GLELO) gene in M. circinelloides CBS277, the DGLA percentage was up to 5.72% [25]. The transcriptional level of D6E(GLELO) gene was remarkably up regulated in the mutant strains ( Figure 5J). Thus, the sufficient accumulation of DGLA in the mutants might provide sufficient substrate for the synthesis of ETA. These above results indicated that the overexpression of the delta-17 desaturase gene induced an increase in NADPH and acetyl-CoA and provided abundant substrate to synthetic ETA. Considering the significant increase in ETA for the TFAs and the lipid contents in the transformant strains, we can hypothesize that the overexpression of the delta-17 desaturase could improve the activities of key enzymes in the NADPH synthesis pathway and further increase ETA production and lipid accumulation.

Conclusions
In conclusion, we successfully constructed the ETA biosynthesis pathway in M. circinelloildes by overexpression of the delta-17 desaturase gene from P. parasitica. As far as we know, this is the first time to report the construction of an ETA-producing cell factory in M. circinelloildes. The results indicated that the introduction of exogenous D17 genes affected lipid profiles, lipid content, and enzymes related to fatty acid synthesis. This work established a new context for further research for improved production of ETA in M. circinelloides, which may apply in industrial ETA production. Simultaneously, ETA is a precursor substance of EPA. This study laid the foundation for further implementation of the synthesis of ω-3 PUFAs by oleaginous fungi. Further, this work has covered the way for further development of strains with varying LCPUFAs compositions tailored for specific applications and for developing a versatile platform for the production of other high-value lipid products.

Conflicts of Interest:
The authors declare that there is no competing financial interest with the publication of this paper.