Improving the Synthesis of Odd-Chain Fatty Acids in the Oleaginous Yeast Yarrowia lipolytica

Abstract

(1) Background: Odd-chain fatty acids (OCFAs) have garnered attention for their potential health benefits and unique roles in various biochemical pathways. Yarrowia lipolytica, a versatile yeast species, is increasingly studied for its capability to produce OCFAs under controlled genetic and environmental conditions. However, optimizing the synthesis of specific OCFAs, such as cis-9-heptadecenoic acid (C17:1), remains a challenge. (2) Methods: The gene coding for the Δ9 fatty acid desaturase, YlOLE1, and the gene coding the diacylglycerol O-acyltransferase 2, YlDGA2, were overexpressed in Y. lipolytica. With the engineered strain, the main goal was to fine-tune the production of OCFA-enriched lipids by optimizing the concentrations of sodium propionate and sodium acetate used as precursors for synthesizing odd- and even-chain fatty acids, respectively. (3) Results: In the strain overexpressing only YlDGA2, no significant changes in fatty acid composition or lipid content were observed compared to the control strain. However, in the strain overexpressing both genes, while no significant changes in lipid content were noted, a significant increase was observed in OCFA content. The optimal conditions for maximizing the cell density and the C17:1 content in lipids were found to be 2.23 g/L of sodium propionate and 17.48 g/L of sodium acetate. These conditions resulted in a cell density (optical density at 600 nm) of 19.5 ± 0.46 and a C17:1 content of 45.56% ± 1.29 in the culture medium after 168 h of fermentation. (4) Conclusions: By overexpressing the YlOLE1 gene and optimizing the concentrations of fatty acid precursors, it was possible to increase the content of OCFAs, mainly C17:1, in lipids synthesized by Y. lipolytica.

1. Introduction

Finding alternatives to conventional petroleum-derived fuels has become imperative in a world where the call for sustainable and environmentally friendly practices is increasingly urgent [1]. The growing interest in a greener and cleaner ecosystem has influenced a revolution in the lipid-derived oleochemical field. Among the multiple possibilities, microbial oils have risen as a promising substitution, presenting an exceptional solution to the global quest for sustainable energy sources [2,3,4].

Microbial oils are lipids and fatty acid-derived products synthesized by microorganisms such as bacteria, yeasts, and algae through fermentation processes [5]. Microbial oil production often requires fine-tuning culture conditions, nutrient availability, and various other factors to boost the accumulation of lipids [6,7,8,9]. Among the several microorganisms studied, Yarrowia lipolytica, an oleaginous yeast, has emerged as one of the most promising for lipid production [10,11]. Through diverse metabolic engineering approaches, it can accumulate high lipid content in its biomass, offering a potential alternative to fossil resources [12]. This model organism is considered non-pathogenic and is Generally Recognized As Safe (GRAS) by the American Food and Drug Administration (FDA). It requires strictly aerobic conditions, a temperature around 30 °C, and is adaptable to a mildly acidic pH range of 5.5–6.5. In addition, Y. lipolytica demonstrates remarkable versatility by assimilating a diverse range of substrates. It is reported to valorize a broad range of cheap and renewable feedstocks including sugars, volatile fatty acids, alkanes, and municipal organic wastes [13,14,15]. Moreover, mono-, di-, and polysaccharides that Y. lipolytica cannot assimilate naturally can be processed by either introducing heterologous enzymes (e.g., SUC2 from Saccharomyces cerevisiae for sucrose metabolism or glucoamylase from Aspergillus niger for starch metabolism) or awakening native enzymes (e.g., XK for xylose metabolism) in Y. lipolytica [16,17,18]. This substrate flexibility offers an environmentally friendly approach to valorizing raw materials to high-value chemicals with a reduced fossil carbon footprint and improved process benefits [19].

The metabolic engineering of Y. lipolytica also extends to the production of high-value-added lipids, specifically odd-chain fatty acids (OCFAs), which are not commonly found in the yeast’s natural products [20,21]. OCFAs emerge as a distinctive solution in the field of lipid-derived oleochemicals, promising sustainable and economically feasible production. They are of great interest due to their unique structural and functional properties, which differ significantly from even-chain fatty acids. They serve as valuable precursors in the synthesis of high-value chemicals, such as flavors, fragrances, and pharmaceuticals, offering a pathway to products that are otherwise difficult to obtain from traditional sources. Additionally, OCFAs have been shown to possess anti-inflammatory and antimicrobial properties, making them promising for therapeutic applications [22].

OCFA synthesis is initiated by supplementing the media with a three-carbon atom substrate (e.g., sodium propionate), which is converted into propionyl-CoA that condensates with a malonyl-CoA to yield 3-oxovaleryl-ACP, a five-carbon (C5) compound serving as the starting point for OCFA synthesis. Several studies reported that propionate supplementation has led to an increase in lipid yields [20,23]. However, inhibitory effects of this substrate were reported for cell growth, particularly when the concentration exceeded 5 g/L [24,25]. Thus, it is important to balance between acetyl-CoA and propionyl-CoA precursor pools to find the optimal culture conditions and avoid growth inhibition. In addition, it was shown from previous studies that desaturases located in the endoplasmic reticulum play a crucial role in lipid biosynthesis [26,27]. For instance, OLE1 is a Δ9 fatty acid desaturase gene that kickstarts the synthesis of palmitoleic (C16:1) or oleic acid (C18:1), whereas FAD2 is a Δ12 desaturase that introduces a second double bond, leading usually to the formation of linoleic acid (C18:2). Remarkably, the overexpression of OLE1 unexpectedly led to an increased lipid yield in previously engineered strains [28].

In light of these observations, this study aimed to further enhance OCFA synthesis by genetically engineering the Y. lipolytica OCFA-producing strain JMY7877, previously studied [19], and optimizing the concentrations of two key substrates for fatty acid biosynthesis: sodium propionate and sodium acetate.

2. Materials and Methods

2.1. Strains, Plasmids, and Media

The strain JMY7877 previously studied [19] was transformed to integrate a second copy of DGA2 and OLE1 genes from Y. lipolytica (YlDGA2 and YlOLE1, respectively). All strains and plasmids used in this project are described in

Table 1. Plasmids and strains used in this study.

Strain	Description	Plasmid	Auxotrophy
JMY7877	Po1d Δphd1 Δmfe1 Δtgl4 + pTEF-YlDGA2 + pTEF-YlGPD1 + hp4d-YlLDP1 + pTEF-RePCT + pTEF-ScSUC2-LEU2 ex + pTEF-YlHXK1-URA3 ex	JME2103 JME2347	U+L+
JMY9031	Y7877 − URA3 − LEU2	JME547	U−L−
JMY9112	Y9031 + pTEF-YlDGA2-URA3 ex	JME1132	U+L−
JMY9178	Y9112 + pTEF-YlOLE1-LEU2 ex	JME5112	U+L+

and illustrated in Figure 1. Yeast extract peptone dextrose (YPD)-rich medium was prepared containing yeast extract (10 g/L), peptone (20 g/L), and glucose (20 g/L). Yeast nitrogen base minimal media were prepared containing YNB (1.7 g/L) without amino acids and ammonium sulfate, NH₄Cl (5 g/L), and 50 mM KH₂PO₄-Na₂HPO₄. To complement strain auxotrophy, 0.1 g/L of uracil and/or leucine was supplemented in the media. Solid media were prepared by adding 15 g/L agar.

Standard molecular biology techniques were performed as previously described [29]. For gene overexpression, plasmids were prepared by NotI digestion and transformed into Y. lipolytica strains using the lithium acetate method, as described previously [30]. Restriction enzymes were obtained from New England Biolabs, PCR amplification was performed with Promega GoTaq DNA polymerases, and plasmids were purified with a Macherey-Nagel Plasmid Miniprep Kit. YlOLE1 and YlDGA2 gene integration were confirmed via colony PCR using the primers 5′-CTGTCGAGGGCTCCATCCGATG-3′ and 5′-CTAAGCAGCCATGCCAGACATAC-3′, and 5′-TCTGGAATCTACGCTTGTTCAG-3′ and 5′-ATAAGATTGAGACCGTTCTGG-3′, respectively.

2.2. Fermentation Process

2.2.1. Preparation of Stock Cell Culture

To prepare a stock cell culture for the JMY9178 strain, culture was performed in a 500 mL Erlenmeyer flask containing 100 mL of a medium composed of sucrose (20 g/L), yeast extract (10 g/L), and peptone (20 g/L), along with Na₂HPO4 (0.05 M) and KH₂PO₄ (0.05 M) used as pH buffers to maintain a pH of ≈6. All components were previously prepared and autoclaved in an HMC HV-110L autoclave (HMC Europe GmbH, Tüßling, Germany) for 20 min at 121 °C to ensure sterility.

The medium was inoculated with 2 mL of a previously prepared stock solution of the strain JMY9178. Subsequently, the flask was placed in a shaker (New Brunswick Scientific, Edison, NJ, USA) set at 28 °C, with agitation maintained at 170 rpm for 24 h. Following this incubation period, 1 mL of cell culture was combined with 1 mL of a 50% sterile solution of pure glycerol (w/v) and preserved in cryotubes at −80 °C.

2.2.2. Culture Media Composition

The fermentation media were composed of all components listed in

Table 2. Growth media composition.

Component	Concentrations
Sugar beet molasses *	20 gsucrose/L
Crude glycerol *	30 g/L
Yeast extract	1 g/L
NH4Cl	0.5 g/L
Sodium acetate	[0; 5; 10; 15; 20 g/L]
Sodium propionate	[0; 0.5; 1; 2; 3.5; 5 g/L]
Na2HPO4	0.05 M
KH2PO4	0.05 M
Saline solution **	1 mL/L
Distilled water	-

* The concentration of sugar beet molasses was adjusted according to its sucrose content (≈800 g_sucrose/L_molasses). Crude glycerol, with a purity of approximately 80%, was used directly to prepare the culture media. The concentrations of sucrose in sugar beet molasses and glycerol in crude glycerol were determined by HPLC, as previously described [19]. ** The composition of the 1000× concentration saline solution with a final volume of 1 L was as follows: H₃PO₄ 85% liquid (107 g), KCl (20 g), NaCl (20 g), MgSO₄·7H₂O (27 g), ZnSO₄·7H₂O (7.7 g), MnSO₄·H₂O (0.47 g), CoCl₂·6H₂O (0.3 g), CuSO₄·5H₂O (0.6 g), Na₂MoO₄·2H₂O (0.094 g), H₃BO₃ (0.3 g), and water up to 1 L.

. Sugar beet molasses (≈800 g/L sucrose, Tereos Company, Origny-Sainte-Benoite, France) and crude glycerol (≈80% purity, SAS PIVERT, Venette, France) were used in this study as low-cost substrates for fermentation. Based on the previously published paper, the concentrations of sugar beet molasses and crude glycerol were set at 20 g/L and 30 g/L, respectively [19]. The medium was supplemented with yeast extract and ammonium chloride as organic and inorganic nitrogen sources, respectively (Table 2). Both components allowed a controlled carbon-to-nitrogen (C/N) ratio. Excessive nitrogen concentration could divert nutritional resources toward biomass growth rather than lipid accumulation [31]. Sodium acetate and sodium propionate used as precursors of acetyl-CoA and propionyl-CoA, respectively, were added at different concentrations to determine the experimental domain for the design of experiments approach. Na₂HPO₄ and KH₂PO₄ served as pH buffers, maintaining the pH ≈ 6. Distilled water was then added to fulfill the volume requirement of the media.

2.2.3. Fermentation Conditions

The fermentation process was performed in 500 mL Erlenmeyer flasks placed in a shaking incubator at 28 °C, 170 rpm agitation, for 168 h. Each flask contained 100 mL of culture medium. The yeast cell growth was monitored by periodically sampling 1 mL from each flask during fermentation. The samples were diluted 20 times with a non-inoculated medium used as blank. The optical density (OD) was then measured over the blank at a wavelength of 600 nm using a spectrophotometer (Thermo BioMate 3, Waltham, MA, USA). At the end of fermentation, 50 mL of the culture medium was used to quantify the dry cell weight (DCW). Accumulated biomass was harvested by centrifugation for 10 min at 4600 rpm. It was then freeze-dried for three days using an Alpha 1-4 LD Plus lyophilizer (Martin Christ, Osterode am Harz, Germany) set at a temperature of −70 °C and a pressure of 0.52 mbar. Dry biomass was subsequently used for lipid extraction and fatty acid quantification.

2.2.4. Optimization of the Concentrations of Fatty Acid Precursors

A central composite experimental design was applied to optimize the concentrations of two variables: sodium acetate (X₁, ranging from 15 to 20 g/L) and sodium propionate (X₂, ranging from 2 to 3.5 g/L). These experimental domains were selected based on results obtained by varying the concentrations of the precursors. The responses investigated in this experimental design were the maximization of the cell growth (measured by optical density at 600 nm, Y₁) and the percentage of C17:1 in lipids (Y₂), selected as the primary OCFA synthesized. This design consisted of 22 experiments (4 points from the factorial design and 4 star points, all repeated twice (4 × 2 + 4 × 2 = 16 points)), along with 3 center points (repetitions for statistical analysis) and 3 test points for the validation of the predicted models.

With two variables and 5 levels, the experimental data were fitted to a second-degree regression model represented by Equation (1):

Y= β₀ + β₁X₁ +β₂X₂ +β₁₂X₁X₂ +β₁₁X₁² +β₂₂X₂²

(1)

In this equation, Y denotes the predicted response, X₁ and X₂ represent, respectively, the variables 1 and 2, β₀ is the constant coefficient, β₁ and β₂ are the linear coefficients, β₁₂ denotes the interaction term coefficient, and β₁₁ and β₂₂ denote the quadratic effect term coefficients.

Fermentation of the 22 Erlenmeyer flasks corresponding to the runs of the experimental design was performed as mentioned in Section 2.2.3. The two responses (i.e., cell density (OD_600nm) and C17:1 content (%)) were recorded for each experiment and entered into NemrodW software (v.2017) for analysis.

2.3. Lipid Extraction and Fatty Acid Analysis

2.3.1. Lipid Extraction from Y. lipolytica Biomass

Lipid quantification was performed as previously demonstrated [32], with slight modifications. Lyophilized biomass obtained at the end of fermentation was ground in a bead miller (Retsch MM400, Haan, Germany) for 3 × 5 min at 30 Hz and then washed by Soxhlet with 250 mL n-hexane for 24 h. After evaporation of n-hexane at 60 °C and reduced pressure using a rotary evaporation system, the lipid content (%) was determined, according to Equation (2).

$$Lipid content (\%)={\displaystyle \frac{mass of recovered lipids }{mass of the sample }}\times 100$$

(2)

For fatty acid analysis, lipid extraction from Y. lipolytica freeze-dried biomass was performed as previously outlined [19]. Approximately 30 mg of dry biomass were placed into a 2 mL tube with two stainless steel beads (5 mm diameter). Subsequently, 1 mL of a cyclohexane–isopropanol mixture (2:1, v:v) was added to each tube, and the contents were milled for 5 min at 30 Hz using the bead miller Retsch MM400. Afterward, the mixtures were centrifuged for 10 min at 13,500 rpm using a Micro Star 12 centrifuge (VWR, Rosny-Sous-Bois, France), and the resulting supernatants were transferred each to a glass tube. The extraction process was repeated three times by adding 1 mL of cyclohexane–isopropanol, bead-milling the samples, and centrifuging until a total of 3 mL of supernatant was gathered in the glass tube. The resulting volume of supernatants was evaporated under a nitrogen flow at 60 °C, and the collected lipids were subjected to a transesterification reaction.

2.3.2. Preparation and Analysis of Fatty Acid Methyl Esters

Each lipid sample was resuspended in 0.3 mL of toluene (Carlo Erba, Val de Reuil, France), and a 1 mL methanolic HCl (3 M) (Sigma Aldrich, Saint-Quentin-Fallavier, France) was added to each mixture. Following this, the tubes were purged with nitrogen to prevent oxidation and were placed in an 80 °C oven for 2 h. The samples were mixed by vortexing every 30 min. To stop the transesterification reaction, 0.5 mL of sodium bisulfite (5% w:v prepared in water, Acros Organics, Fisher Scientific, Illkirch, France) was introduced, and the tube underwent vortexing for 10 s. The extraction of fatty acid methyl esters (FAMEs) was accomplished by adding 1.7 mL of cyclohexane (Carlo Erba, Val-de-Reuil, France) and vortexing for 1 min. After centrifugation at 2500 rpm for 2 min, 200 μL was withdrawn from the upper organic phase containing FAMEs and transferred to an injection vial. The vial content was then evaporated under a nitrogen flow and reconstituted in 500 μL of cyclohexane for gas chromatography (GC) analysis.

GC analysis of FAMEs was performed using an Agilent Technologies GC apparatus equipped with a capillary column (AIT-50, 50% cyanopropyl-50% methyl polysiloxane, 60 m × 320 μm × 0.25 μm) and coupled to a flame ionization detector (GC-FID Agilent Technologies 7890A, Les Ulis, France). FAMEs were identified compared to a commercial standard mixture (FAME37, Merck, Fontenay Sous Bois, France). The GC oven was initially set at 150 °C for 2 min, then gradually increased to 220 °C at a rate of 1.5 °C/min, and maintained at this temperature for 30 min. Hydrogen was employed to ignite the flame, and helium served as the carrier gas with a flow rate of 30 mL/min. The injector temperature was set at 250 °C and 1 µL of the FAME sample was injected. After a 30 min run, FAME peaks were identified, integrated, and quantified.

2.4. Statistical Analysis

One-way analysis of variance (ANOVA) was used to determine the significant differences using StatPlus V6 software for Macintosh systems.

3. Results and Discussion

3.1. YlOLE1-Overexpressing Strain JMY9178

Previous studies demonstrated that the potential of Y. lipolytica to produce and accumulate OCFAs can be improved by a combinatorial strategy to simultaneously overexpress and delete multiple key genes associated with lipid biosynthesis and degradation, respectively [20,33]. In this work, the gene coding for the Δ9 fatty acid desaturase, YlOLE1, and the gene coding the diacylglycerol O-acyltransferase 2, YlDGA2, were overexpressed. The OLE1 enzyme catalyzes the desaturation of saturated fatty acids by introducing a double bond between the ninth and tenth carbon atoms from the carboxyl end of the fatty acid chain [34], whereas the DGA2 enzyme is recognized as a key component of the lipid pathway by performing the final step of TAG biosynthesis (the incorporation of the third acyl-CoA onto the diacylglycerol backbone) and its storage [35]. In the strain JMY9112 overexpressing only YlDGA2, no significant changes in the fatty acid composition or lipid content were observed, compared to the control strain JMY7877. However, in the strain JMY9178 overexpressing both genes (i.e., YlOLE1 and YlDGA2), significant changes were observed in the OCFA content. Although there was no significant increase in the total lipid content, the C17:1 (%) and the ratio of unsaturated fatty acids to total fatty acids were improved in the strain JMY9178, compared to the control JMY7877, studied previously [19].

For this comparison, microbial cultures were performed under similar conditions. The concentrations of the fatty acid precursors, sodium acetate and sodium propionate, were set at 20 g/L and 5 g/L, respectively. The results showed that in strains JMY7877 and JMY9178, the cell densities (OD_600nm) after 168 h of fermentation were 18.90 ± 0.23 and 14.8 ± 1.89, respectively. OCFAs represented 58% ± 4.08 and 67% ± 1.17 of the total fatty acids, while the C17:1 content was 36.45% ± 1.2 and 50.4% ± 0.18, respectively. Overexpression of the YlOLE1 gene significantly increased the content in OCFAs and C17:1 and decreased the cell density. This decrease in cell density was also observed in a previous work where the authors overexpressed OLE3 into Schizochytrium sp. S31 [36]. In addition, the OCFA content was among the highest reported in the literature [37], while the C17:1 content obtained in this study was the highest reported to date.

3.2. Impact of Fatty Acid Precursors on Yeast Cell Growth and OCFA Synthesis

3.2.1. Impact of Sodium Propionate

OCFAs are synthesized using various three-carbon precursors. Propionate, for example, has been widely used in microbial production systems. For instance, Candida sp. DBM 2163, Trichosporon cutaneum, and Y. lipolytica have demonstrated the ability to incorporate propionic acid to produce significant amounts of C17:1 [33,38,39]. Other microorganisms, such as Rhodococcus opacus PD630, utilize 1-propanol as a precursor to produce OCFAs [40]. Although required for OCFA synthesis, the literature has reported that propionate may negatively impact Y. lipolytica growth and lipid biosynthesis [24,37,41]. Therefore, the salt form, sodium propionate, has been used in some studies to reduce the toxicity of propionic acid, especially in Y. lipolytica [19,32]. To further reduce the inhibitory effect of this precursor, multiple fermentation conditions were conducted in this work with the YlDGA2/YlOLE1 overexpressing JMY9178 strain. Various concentrations of sodium propionate (0–5 g/L) were tested while maintaining a constant sodium acetate concentration of 20 g/L. The results of these fermentations are shown in Figure 2.

The results indicate that cell density (OD_600nm) decreases as the concentration of sodium propionate increases. The lowest cell density (OD_600nm = 14.8 ± 1.89) was observed at the highest sodium propionate concentration (5 g/L). An inverse correlation was observed between the sodium propionate concentration and cell density. Additionally, the control medium, which did not contain sodium propionate, achieved the highest cell density after 168 h of fermentation (Figure 2). These observations suggest that sodium propionate exerted an inhibitory effect on cell growth, with higher concentrations inhibiting cell proliferation. However, as previously mentioned, propionate plays a crucial role in OCFA synthesis. These fermentations were conducted to determine the concentration range of propionate that supports both cell growth (biomass accumulation) and the highest content of OCFAs. Therefore, the fatty acid composition was analyzed after 168 h of fermentation. The results obtained are summarized in

Table 3. Dry cell weight (DCW) and FAME (fatty acid methyl ester) profile of each medium containing different concentrations of sodium propionate. FAs denotes fatty acids.

Sodium Propionate (g/L)	0	0.5	1	2	3.5	5
DCW (g/L)	19.42 ± 2.9	18.31 ± 2.4	17.0 ± 0.07	15.6 ± 0.71	11.3 ± 0.14	10.5 ± 1.13
FAMEs (%)
C15:0	0.46 ± 0.07	1.67 ± 0.05	2.75 ± 0.15	4.07 ± 0.2	2.69 ± 0.34	2.07 ± 0.19
C16:0	25.25 ± 2.22	24.36 ± 1.39	19.77 ± 0.08	7.21 ± 0.44	3.2 ± 0.29	1.49 ± 0.46
C16:1	11.24 ± 0.13	10.95 ± 0.91	7.87 ± 0.47	4.37 ± 0.04	3.05 ± 0.17	2.22 ± 0.11
C17:0	0.5 ± 0.07	2.27 ± 0.03	4.44 ± 0.52	6.01 ± 0.25	3.12 ± 0.04	1.63 ± 0.18
C17:1	3.67 ± 0.25	11.83 ± 0.66	20.31 ± 0.05	46.17 ± 1.47	51.78 ± 0.08	50.4 ± 0.18
C18:0	6.15 ± 1.73	4.87 ± 0.94	4.05 ± 0.47	1.37 ± 0.32	0.69 ± 0.01	0.22 ± 0.31
C18:1	48.54 ± 3.34	40.8 ± 0.68	36.8 ± 0.49	21.59 ± 0.31	19.58 ± 0.02	18.65 ± 1.36
C18:2	0.09 ± 0.06	0.42 ± 0.05	1.17 ± 0.01	3.77 ± 0/04	6.47 ± 0.78	6.9 ± 0.9
C19:0	4.08 ± 0.28	3.02 ± 0.39	2.79 ± 0.04	2.65 ± 0.01	3.37 ± 0.11	4.76 ± 0.13
C19:1	0	0	0	2.73 ± 0.01	5.96 ± 0.13	8.51 ± 0.49
Saturated FAs	36.44 ± 4.36	36.21 ± 2.8	33.8 ± 1.27	21.31 ± 6.07	13.08 ± 0.69	10.19 ± 1.27
Unsaturated FAs	63.51 ± 3.81	64.01 ± 2.3	66.16 ± 1.03	78.64 ± 6.05	86.86 ± 0.69	86.69 ± 3.04
OCFAs	8.71 ± 0.67	18.8 ± 1.12	30.3 ± 0.76	61.63 ± 4.45	66.94 ± 0.35	67.39 ± 1.17

.

The GC-FID analysis of the extracted lipids revealed that the engineered strain JMY9178, cultivated in the presence of 2, 3.5, and 5 g/L of sodium propionate, accumulated the highest OCFA content (up to 67.39% ± 1.17 with 5 g/L sodium propionate) relative to the total fatty acids. In contrast, media with lower concentrations of sodium propionate accumulated lower OCFA levels (8.71% ± 0.67, 18.8% ± 1.12, and 30.3% ± 0.76 for 0, 0.5, and 1 g/L sodium propionate, respectively). These results confirm that sodium propionate is essential for OCFA synthesis, as highlighted in recent studies [19,37,42]. Additionally, cis-9-heptadecenoic acid (C17:1) was the major OCFA synthesized by the yeast, with its accumulation increasing as the sodium propionate concentration increased from 0.5 to 5 g/L. Other OCFAs, such as pentadecanoic acid (C15:0), heptadecanoic acid (C17:0), nonadecanoic acid (C19:0), and nonadecenoic acid (C19:1), were produced in smaller amounts.

These findings suggest that while the engineered Y. lipolytica strain JMY9178 can proliferate in media with low sodium propionate concentrations, it requires higher concentrations of sodium propionate to synthesize efficiently OCFAs. To achieve both high cell growth and significant C17:1 production, the concentration range of sodium propionate was identified as 2–3.5 g/L for further optimization with response surface methodology.

3.2.2. Impact of Sodium Acetate

Acetate plays an essential role in yeast cell proliferation and is a precursor of even-chain fatty acids (ECFAs). Its concentration in the culture medium ranged from 0 to 20 g/L, with a fixed concentration of sodium propionate at 5 g/L. Figure 3 shows the cell density measurements (OD_600nm) obtained for each experimental condition.

The results indicate that the highest cell growth (OD_600nm = 14.8 ± 1.89) was achieved at the highest sodium acetate concentration (20 g/L). In contrast, lower concentrations of sodium acetate (5 g/L) led to a significantly lower cell density (OD_600nm = 3.83 ± 0.11). When fermentation was conducted without sodium acetate, the adaptation phase lasted for 24 h before yeast proliferation began, ultimately reaching a cell density that was not significantly different from that obtained with 15 g/L of sodium acetate. Moreover, a significant increase (p < 0.05) in cell density (OD_600nm) was observed as the sodium acetate concentration increased from 5 to 20 g/L. These findings demonstrate that the highest cell densities were observed within the sodium acetate concentration range of 15–20 g/L.

Table 4. Dry cell weight (DCW) and FAME profile of each media containing different concentrations of sodium acetate.

Sodium Acetate (g/L)	0	5	10	15	20
DCW (g/L)	6.30 ± 0.48	3.94 ± 0.17	4.92 ± 0.96	6.85 ± 0.03	10.5 ± 1.13
FAMES (%)
C15:0	7.59 ± 0.33	6.93 ± 0.25	3.47 ± 0.10	1.81 ± 0.19	2.07 ± 0.19
C16:0	5.32 ± 0.37	3.83 ± 0.38	1.89 ± 0.14	1.24 ± 0.15	1.49 ± 0.46
C16:1	2.93 ± 0.06	3.52 ± 0.28	2.48 ± 0.06	1.94 ± 0.10	2.22 ± 0.11
C17:0	7.93 ± 0.90	2.32 ± 0.23	1.42 ± 0.27	1.29 ± 0.08	1.63 ± 0.18
C17:1	53.42 ± 0.69	50.49 ± 0.88	50.40 ± 1.73	49.11 ± 0.06	50.4 ± 0.18
C18:0	1.26 ± 0.01	0.71 ± 0.09	0.37 ± 0.00	0.44 ± 0.09	0.22 ± 0.31
C18:1	14.83 ± 0.18	18.71 ± 0.59	20.17 ± 1.22	18.09 ± 1.34	18.65 ± 1.36
C18:2	2.32 ± 0.07	2.20 ± 0.40	6.15 ± 0.49	9.93 ± 0.70	6.9 ± 0.9
C19:0	2.07 ± 0.27	5.83 ± 0.59	5.65 ± 0.01	4.83 ± 0.13	4.76 ± 0.13
C19:1	2.32 ± 0.37	5.44 ± 0.54	7.97 ± 0.61	11.66 ± 1.76	8.51 ± 0.49
Saturated FAs	24.15	19.61	12.80	9.59	10.19
Unsaturated FAs	75.81	80.35	87.16	90.72	86.69
OCFAs	73.31	71.00	68.91	68.69	67.39
ECFAs	26.69	29	31.09	31.31	32.61

shows the effect of increasing sodium acetate concentrations on dry cell weight (DCW) and FAME profiles in the engineered strain JMY9178. As the sodium acetate concentration increased from 0 to 20 g/L, both cell growth and fatty acid composition were significantly affected. At 20 g/L sodium acetate concentration, the DCW reached its highest value of 10.5 ± 1.13 g/L, much higher than at lower concentrations, such as with 5 g/L (3.94 ± 0.17 g_DCW/L) and 15 g/L (6.85 ± 0.03 g_DCW/L) sodium acetate. This indicates that higher sodium acetate concentrations strongly favor cell growth. This observation concurs with the results reported in a previous work, where the dry cell weight increased by increasing the concentration of acetic acid up to a certain level [43].

Regarding the FAME profile, C17:1 remained the major OCFA for all tested sodium acetate concentrations, with percentages ranging from 49.11% ± 0.06 to 53.42% ± 0.69, compared to the total fatty acids. However, as the sodium acetate concentration increased, the content of other OCFAs (i.e., C15:0, C17:0, and C19:0) decreased. In contrast, C19:1 increased from 2.32% ± 0.37 in the absence of sodium acetate to 11.66% ± 1.76 at 15 g/L before decreasing to 8.51% ± 0.49 at 20 g/L.

It could be also observed from Table 4 that as the sodium acetate concentration increases, the synthesis of ECFAs is favored, as evidenced by the higher proportions of C16:0, C18:0, and C18:1 at elevated sodium acetate concentrations. The proportion of saturated fatty acids decreased from 24.15% at 0 g/L to 10.19% at 20 g/L, while the percentage of unsaturated fatty acids increased from 75.81% to 86.69%. This indicates that higher sodium acetate concentrations drive the synthesis of ECFAs, which likely competes with OCFA synthesis.

In terms of total OCFA content, the highest accumulation was observed in the control (73.31%) and at 5 g/L sodium acetate (71%). However, as the sodium acetate concentration increased, the OCFA content gradually decreased to 67.39% at 20 g/L. This reduction in OCFA content with increasing sodium acetate concentration in the culture medium can be attributed to the promotion of ECFA synthesis, over the synthesis of OCFAs.

Taking together the results of cell density (OD_600nm), DCW, and OCFA content in lipids, it seems that the concentration of sodium acetate should be optimized in the range of 15–20 g/L, which will be further studied using response surface methodology.

3.3. Optimization of Concentration of FA Precursors

3.3.1. Model’s Validation and Fitting

A central composite design was applied to optimize the culture conditions of Y. lipolytica, focusing on maximizing two responses: the cell density (OD_600nm) and the C17:1 content (%) in lipids. The experimental domains of the factors were defined according to the results obtained in Section 3.2.1 and Section 3.2.2.

Table 5. Experimental conditions and the measured responses of the design of experiments. Coded variables are provided in parentheses.

	Factors		Measured Responses			Predicted Responses
Experiment	Sodium Acetate (g/L)	Sodium Propionate (g/L)	Cell Density (OD600nm)	DCW (g/L)	C17:1 (%)	Cell Density (OD600nm)	C17:1 (%)
1	15 (−1)	2 (−1)	21.88	10.44	41.02	22.070	39.260
2	15 (−1)	2 (−1)	20.99	9.92	40.71	22.070	39.260
3	20 (+1)	2 (−1)	28.12	11.14	44.74	24.040	40.220
4	20 (+1)	2 (−1)	23.17	12.04	41.68	24.040	40.220
5	15 (−1)	3.5 (+1)	11.58	7.47	41.16	13.925	47.011
6	15 (−1)	3.5 (+1)	17.42	8.12	46.06	13.925	47.011
7	20 (+1)	3.5 (+1)	23.17	9.52	43.44	19.599	46.906
8	20 (+1)	3.5 (+1)	21.88	9.69	46.23	19.599	46.906
9	13.96 (−α)	2.75 (0)	15.74	9.37	47.35	15.393	46.798
10	13.96 (−α)	2.75 (0)	16.43	8.97	48.18	15.393	46.798
11	21.04 (+α)	2.75 (0)	17.82	9.85	45.56	20.798	47.403
12	21.04 (+α)	2.75 (0)	18.22	10.27	47.56	20.798	47.403
13	17.5 (0)	1.69 (−α)	24.75	13.78	30.19	26.172	34.494
14	17.5 (0)	1.69 (−α)	27.42	13.51	31.48	26.172	34.494
15	17.5 (0)	3.81 (+α)	15.25	9.09	48.34	17.272	44.702
16	17.5 (0)	3.81 (+α)	15.05	9.21	49.28	17.272	44.702
17	17.5 (0)	2.75 (0)	20.59	9.47	49.19	17.269	48.811
18	17.5 (0)	2.75 (0)	16.34	9.74	47.96	17.269	48.811
19	17.5 (0)	2.75 (0)	15.94	9.52	47.46	17.269	48.811
20	15.97 (−0.612)	2.48 (−0.36)	16.53	9.84	48.38	17.845	46.450
21	19.03 (+0.612)	2.48 (−0.36)	19.5	9.70	48.25	19.784	46.828
22	17.5 (0)	3.28 (+0.707)	16.34	8.99	48.14	16.157	49.060

displays the experimental conditions along with the corresponding measured responses obtained from the experiments.

The experimental design was validated using the test points corresponding to experiments 20–22. The analysis of the residues showed non-significant differences (p > 0.05) between the measured responses and the predicted ones, indicating the validation of both models (Table 5). The measured responses of the test points were then included in the models provided using NemrodW software as follows (Equations (3) and (4)) (* means significant coefficient):

Y_{cell density (OD600nm)} = 17.269* + 1.911* X₁ + 0.926 X₁X₂ − 3.147* X₂ + 0.413 (X₁)² + 2.226* (X₂)²

(3)

Y_{% C17:1} = 48.811* + 0.214 X₁ + 3.609* X₂ − 0.266 X₁X₂ − 0.855 (X₁)² − 4.607* (X₂)²

(4)

In addition, the results of the ANOVA for cell density (

Table 6. Analysis of variances for the responses (a) Y₁: cell density (OD_600nm) and (b) Y₂: C17:1 (%). * indicates significant differences.

(a)
Source of Variation	Sum of Squares	Degrees of Freedom	Mean Squares	F-Ratio	p-Value
Regression	280.508	5	56.101	9.059	0.0307 *
Residue	99.088	16	6.193
Lack of fit	51.372	6	8.562	1.794	19.7
Pure error	47.716	10	4.771
Total	379.596	21
(b)
Source of Variation	Sum of Squares	Degrees of Freedom	Mean Squares	F-Ratio	p-Value
Regression	428.727	5	85.745	33.195	<0.01 *
Residue	152.087	16	9.505
Lack of fit	126.257	6	21.042	8.147	0.219 *
Pure error	25.830	10	2.583
Total	5880.815	21

) indicate that there are significant differences between the regression model and the residual error, as evidenced by the p-value of 0.0307 (p < 0.05). However, no significant difference is observed between the lack of fit and the pure error (p = 19.7, p > 0.05). This indicates that the model adequately fits the data. For the response C17:1(%), the ANOVA results in Table 6 show a highly significant difference between the regression model and the residual error, with a p-value of less than 0.01. However, unlike the cell density model, there is no significant difference between the lack of fit and the pure error (p = 0.219, p > 0.05), indicating that the model might still need further refinement to better capture the response variability. It should be noted that although the p-value is higher than 5%, this is due to the fact that the pure error for the C17:1 content is very low (indicating a small difference between repetitions). Therefore, it is expected that the model error does not fall below the pure error. In other words, considering this, we can consider the model for C17:1 content as adequate.

3.3.2. Optimal Culture Condition Analyses

The optimal path refers to the trajectory within the experimental design space that leads to the best combination of factor levels to achieve the desired objectives, here maximizing the responses of cell density (OD_600nm) and C17:1 content (%). These paths are derived from the mathematical models (Equations (3) and (4)) built from the experimental data and are used to guide decision-making toward the optimal conditions. Figure 4 illustrates the optimal path for each response provided by NemrodW software as well as the desirability functions, allowing us to find a compromise between the optimal conditions for each model.

Figure 4a illustrates that maximizing cell density requires an increase in the concentration of sodium acetate (factor 1). As more biomass accumulates, the overall lipid content also increases. Additionally, increasing the concentration of sodium propionate up to a certain point near the center value contributes to maximizing the cell density. Beyond a certain concentration of sodium propionate, a slight decrease in the cell density occurs. Conversely, to optimize the C17:1 content (%), an increase in the sodium propionate concentration (factor 2) and a decrease in the sodium acetate concentration (factor 1) is necessary (Figure 4b). As previously mentioned, sodium propionate and sodium acetate are precursors for the synthesis of propionyl-CoA and acetyl-CoA, respectively. Increasing the sodium acetate concentration reduces the OCFA content, despite its importance in constructing the fatty acid backbone and lipid synthesis in cell membranes. On the other hand, increasing the sodium propionate concentration enhances the OCFA content but impairs cell growth due to the toxicity of sodium propionate at higher concentrations. The results of the optimal paths for both responses concur with the conclusions made in Section 3.2.1 and Section 3.2.2.

Based on the above-mentioned observations, a compromise must be made to determine the optimal concentrations of the various compounds in the culture medium. For this purpose, desirability functions were used in NemrodW software (Figure 4c,d) to find the optimal conditions that allow for maximizing both responses. According to NemrodW software, the optimal concentrations predicted for sodium acetate and sodium propionate were 21.02 g/L and 2.86 g/L, respectively. However, analyzing the isoresponse curves (contour plots) proved interesting, as they sometimes offer better insights than calculations alone. The results of the isoresponse curves are provided in Figure 5.

By analyzing Figure 5, it was more advantageous from economic considerations to select optimal concentrations better than those obtained through the desirability functions. The predicted concentrations for sodium acetate and sodium propionate from Figure 5 were 17.48 g/L (coded variable X₁ = −0.01) and 2.23 g/L (coded variable X₂ = −0.699), respectively. Under these conditions, the predicted responses were a cell density (OD_600nm) of 20.54 ± 5.69 and a C17:1 content of 44.06% ± 3.64. These concentrations were then tested in Erlenmeyer flasks, yielding results of 19.5 ± 0.46 for cell density and 45.56% ± 1.29 for C17:1 content, which were not significantly different from the predicted values (p > 0.05).

Compared to our results reported previously [19], the optimization of sodium acetate and sodium propionate concentrations in the present study resulted in similar cell growth (19.5 ± 0.46 vs. 18.90 ± 0.23 in [19]) and no significant changes in the lipid content. Notably, the concentrations of sodium acetate and sodium propionate were previously found to be 20 g/L and 5 g/L [19], respectively, whereas they were optimized to 17.48 g/L and 2.23 g/L in the current work.

Furthermore, the overexpression of the YlOLE1 gene increased the C17:1 content from 36.45% ± 1.2 using the strain JMY7877 in [19] to 45.56% ± 1.29 in the present study with the strain JMY9178. This concentration of C17:1 is, to our knowledge, the highest reported so far in the literature using agro-industrial by-products (i.e., sugar beet molasses and crude glycerol) as low-cost substrates. These results not only led to improving cell growth and lipid production but also minimizing substrate consumption, thus contributing to a more cost-efficient process. A further improvement for better efficiency could involve using acetate and propionate—the main volatile fatty acids produced during the acidification step of the methanization process—as substrates for ECFA and OCFA synthesis, as recently reviewed [44].

4. Conclusions

In this study, the Y. lipolytica engineered strain JMY9178 led to an increase in OCFA accumulation. Additionally, the optimization of cell growth and the production of cis-9-heptadecenoic acid (C17:1) was achieved using a central composite design. This optimization focused on varying the concentrations of sodium acetate and sodium propionate, which were used as precursors for fatty acid synthesis. According to the results of the response methodology approach, cell growth corresponding to an OD_600nm of 19.5 ± 0.46 and approximately 45.56% ± 1.29 of C17:1 was obtained, while the inhibitory effect of sodium propionate was reduced by lowering its concentration to 2.23 g/L, compared to 5 g/L used in the previous study. Furthermore, the sodium acetate concentration was reduced to 17.48 g/L, compared to 20 g/L in the earlier study. These findings provide valuable insights and support the industrial application of OCFAs, and especially C17:1, paving the way for its potential implementation in larger-scale production processes.