Skip to main content

Altering the fatty acid profile of Yarrowia lipolytica to mimic cocoa butter by genetic engineering of desaturases

Abstract

Background

Demand for Cocoa butter is steadily increasing, but the supply of cocoa beans is naturally limited and under threat from global warming. One route to meeting the future demand for cocoa butter equivalent (CBE) could be to utilize microbial cell factories such as the oleaginous yeast Yarrowia lipolytica.

Results

The main goal was to achieve triacyl-glycerol (TAG) storage lipids in Y. lipolytica mimicking cocoa butter. This was accomplished by replacing the native Δ9 fatty acid desaturase (Ole1p) with homologs from other species and changing the expression of both Ole1p and the Δ12 fatty acid desaturase (Fad2p). We thereby abolished the palmitoleic acid and reduced the linoleic acid content in TAG, while the oleic acid content was reduced to approximately 40 percent of the total fatty acids. The proportion of fatty acids in TAG changed dramatically over time during growth, and the fatty acid composition of TAG, free fatty acids and phospholipids was found to be very different.

Conclusions

We show that the fatty acid profile in the TAG of Y. lipolytica can be altered to mimic cocoa butter. We also demonstrate that a wide range of fatty acid profiles can be achieved while maintaining good growth and high lipid accumulation, which, together with the ability of Y. lipolytica to utilize a wide variety of carbon sources, opens up the path toward sustainable production of CBE and other food oils.

Introduction

Cocoa butter is one of the main ingredients in chocolate and is additionally used in other products such as cosmetics. Cocoa butter mainly consists of triacylglycerol (TAG, three fatty acids attached to a glycerol backbone) and is produced from cocoa tree seeds (Theobroma cacao). Originally native to Central America, the growing area has spread to almost all tropical countries of the world [1]. The cocoa tree needs 5 years to bear fruits and about ten years to reach its maximum yield. Additionally, it only grows in the tropical cocoa belt between 10° and 20° north to south [1]. There is a natural fluctuation in production, which, combined with rising pressure from pests, disease [2], and global warming, is projected to impact the cocoa production areas [3], which might cause shortages in the future.

A solution to counterbalance fluctuation and sustainably prevent supply shortfalls is producing cocoa butter equivalents (CBE) in microbial cell factories, such as yeast. CBE are lipids with similar properties as cocoa butter and can be blended with natural cocoa butter [4]. The fatty acid profile of cocoa butter mainly consists of palmitic acid (C16:0, 26.2%), stearic acid (C18:0, 35.8%) and oleic acid (C18:1, 33.6%), and a small amount of linoleic acid (C18:2, 2.7%) [5]. Multiple studies have demonstrated cocoa butter equivalent production in the model organism Saccharomyces cerevisiae [4, 6, 7]. However, S. cerevisiae does not have a high native lipid accumulation capacity, limiting its productivity and the economic feasibility of the process [4].

A more promising type of cell factory to produce lipids and lipid derivates are oleaginous yeasts, which are characterized by having at least 20% of their cell dry weight as lipids, mainly constituted of TAGs [8]. Multiple oleaginous yeasts (e.g. Yarrowia lipolytica, Rhodotorula toruloides, Trichosporon oleaginosus) have potential as platform strains for CBE production and have recently been reviewed [9].

Y. lipolytica is the most studied oleaginous yeast and is generally regarded as safe (GRAS) [10]. Additionally, Y. lipolytica has a high tolerance to salt and acidic pH [11] and can be engineered to consume various carbon sources (e.g. pentoses such as xylose or lipids) [12]. The primary fatty acids found in Y. lipolytica are palmitic acid (C16:0), palmitoleic acid (C16:1), stearic acid (C18:0), oleic acid (C18:1), and linoleic acid (C18:2) [13]. De novo fatty acid synthesis produces palmitic acid (C16:0) and stearic acid (C18:0) that are bound to Coenzyme A and the Δ9 desaturase Ole1p can desaturate them to form palmitoleic acid (C16:1) and oleic acid (C18:1) [14]. The latter can be converted to phosphatidylcholine and further desaturated by the Δ12 desaturase Fad2p (Fig. 1A) [15]. Both desaturases are membrane-bound and receive electrons from NADH via an NADH-dependent cytochrome b5 reductase [16,17,18].

Fig. 1
figure 1

Overview of fatty acid synthesis in Yarrowia lipolytica and engineering strategy of the study. A The three main fatty acid species found in cocoa butter are marked with a cocoa pod. Names in blue display the proteins catalyzing the reaction. Fas fatty acid synthetase, Elo2 fatty acid elongase, Ole1 Δ9 fatty acid desaturase, Fad2 Δ12 fatty acid desaturase. B The three main experimental strategies of this study: first, the native OLE1 was exchanged with homologs to abolish palmitoleic acid (C16:1). Second, the native OLE1 promoter was exchanged to balance palmitic, stearic and oleic acid (C16:0, C18:0 and C18:1). Finally, the FAD2 promoter was exchanged to reduce the linoleic acid content (C18:2)

The primary aim of this study was to mimic the fatty acid profile of cocoa butter in the oleaginous yeast Y. lipolytica by genetic engineering of its desaturases. A secondary goal was to develop a palette of high lipid content strains with different fatty acid profiles to produce a selection of microbial oils with different properties. A previous study compared different oleaginous yeasts for their potential as a platform to produce cocoa butter-like lipids and found that the main drawback of Y. lipolytica was its high palmitoleic acid (C16:1) and linoleic acid (C18:2) content [19]. To overcome these drawbacks, we first exchanged its Ole1p with homologs from other organisms with low affinity for C16:0 fatty acids to reduce the cellular palmitoleic acid (C16:1) content (Fig. 1B). Then, we exchanged the native OLE1 promoter to vary the expression level of the OLE1 homologs and analyzed the fatty acid profiles to identify a strain with a similar FA profile as cocoa butter while maintaining a high lipid production. To further study the impact of different fatty acid profiles on the growth and lipid content, we also tested different expression levels of FAD2.

Results

Exchange of OLE1 with homologs from Arxula adeninivorans (AaOLE1), Gloeophyllum trabeum (GtOLE1), and Rhodotorula toruloides (RtOLE1)

To establish a food oil production strain, it is crucial to alter its fatty acid profile while maintaining high lipid titers. We first compared the fatty acid composition of a non-engineered strain (OKYL029, mhy1Δ to avoid undesired pseudohyphal growth) [20] with a derived strain engineered for lipid production (OKYL049—deletion of sterol ester production (are1Δ) and increased TAG production (DGA1 overexpression)) [21], leading to a 2.5-fold increase in lipid content. We noted that OKYL049 displayed a decrease in linoleic acid (C18:2) and an increase in stearic acid (C18:0), both of which properties are desirable for CBE-production (Fig. 2). Therefore, all subsequent work was performed in the OKYL049 background.

Fig. 2
figure 2

Replacement of native OLE1 with homologs in FAD2 wild type and deletion strain. Strains were expressing homologs of Ole1p under the control of the native OLE1 promoter. Strains were cultivated for 96 h in LPU media, and the fatty acid profile was determined by FAME extraction. Lipid content represents g FAME per g cell dry weight. Displayed is the mean and standard error of n ≥ 4 replicates. Data for OKYL029 from [20]. Yl Y. lipolytica, Gt G. trabeum, Rt R. toruloides, Aa A. adeninivorans, WT wild type, Δ deletion

Cocoa butter is largely devoid of palmitoleic acid (C16:1), and our first objective was to reduce the C16:1 content while maintaining good growth and lipid production. We chose to modulate the activity of the ∆9 desaturase Ole1p. As previously reported, the deletion of OLE1 was not viable unless supplemented with unsaturated fatty acids [22] (Additional file 1: Fig. S1). During these experiments, we noted that even supplementation with heptadecenoic acid (C17:1), a fatty acid not naturally produced by Y. lipolytica, allowed growth (Additional file 1: Fig. S1). Instead of deleting OLE1, we exchanged the open reading frame of the native YlOLE1 with codon-optimized homologs of OLE1 from Arxula adeninivorans (AaOLE1), Gloeophyllum trabeum (GtOLE), and Rhodotorula toruloides (RtOLE1), all of which have been reported to have low affinity for C16 fatty acids [22]. The growth of these strains was largely unaffected (Fig. 2). As expected, we found that the palmitoleic acid content (C16:1) was highly reduced in all strains expressing any of the three OLE1-homologs (Fig. 2), supporting that the activity of all three homologs with palmitic acid (C16:0) as substrate is lower compared to the native YlOLE1. However, the strain expressing AaOLE1 (OKYL059) showed clear, albeit low, levels of C16:1. The strain expressing RtOLE1 (OKYL056) showed the most cocoa butter-like fatty acid profile (26% C16:0, 0% C16:1, 24% C18:0, 42% C18:1, 8% C18:2) (Fig. 2).

Cocoa butter is very low in linoleic acid (C18:2) content. In order to abolish this fatty acid, we deleted the gene encoding the Δ12 desaturase (FAD2) [23] in the above detailed strains expressing different OLE1-homologs. As expected, the FAD2-deletion abolished C18:2 from the fatty acid profiles (Fig. 2). The deletion of FAD2 in the wild-type OLE1 strain did not reduce growth or lipid production (AOYL02; Fig. 2). Among the FAD2 deletion strains, KSYL14 (expressing GtOLE1) showed the fatty acid profile most similar to cocoa butter (20.5% C16:0, 1.1% C16:1, 38% C18:0, 40.4% C18:1). However, the biomass and lipid content of this strain were strongly reduced compared to the parental strain (AOYL02) (18 to 10 g/L biomass and 45 to 20% lipid content, respectively). Supplementing unsaturated fatty acids to the media recovered the growth defect of KSYL14 (Additional file 1: Fig. S3). In the strains expressing RtOLE1 or AaOLE1, oleic acid (C18:1) proportion in the fatty acid profile was increased.

Promoter exchange of OLE1

All our engineered strains displayed a level of C18:1 that was too high compared to cocoa butter. Therefore, as a second step towards a tailored fatty acid profile, we decided to alter the expression level of the two OLE1 homologs that showed the best results (i.e. highly reduced C16:1 content and no reduction in growth or lipid production) in the previous experiment (Fig. 2), RtOLE1 and AaOLE1. This set of experiments aimed to have enough expression of the OLE1 homologs to supply the cells with unsaturated fatty acids during exponential phase to maintain good growth, while having a minimal expression during the lipid production phase to reduce the amount of oleic acid (C18:1) in the final product. From published transcriptomics analyses [24], we identified several promoters that were hypothesized to result in time-dependent expression. The expression from the promoter of the ribosomal protein-encoding gene RP30 (pRP30) was expected to be very strong during the early growth phase and highly downregulated during the lipid production phase. The expression of pGAPDH was expected to be high during the whole growth phase and only slightly downregulated during lipid production (Fig. 3A). The native OLE1 promoter was expected to be downregulated during the growth phase and upregulated during lipid production. We initially decided to exchange the native promoter of OLE1 for pRP30 or pGAPDH, in strains expressing either RtOLE1 or AaOLE1 with and without FAD2-deletion (promoters were arbitrarily defined as 1000 bp upstream of the start codon of the respective genes).

Fig. 3
figure 3

Promoter exchange of OLE1 homologs. The native OLE1 gene was exchanged for either RtOLE1 or AaOLE1, and the native promoter was exchanged for either pGAPDH or pRP30. A Strains were cultivated for 96 h in LPU media, and the fatty acid profile was determined by FAME extraction. Lipid content represents g FAME per g cell dry weight. Displayed is the mean and standard error of n ≥ 4 replicates. B Scheme of expected promoter activity of pRP30 and pGAPDH based on data from [24]. Dashed line marks the onset of lipid production due to nitrogen limitation. C Growth curves of strain KSYL07. Cells were cultivated in 96-well plates in LPU media with or without the addition of palmitoleic acid (C16:1) or oleic acid (C18:1) (500 mg/L in 1% Tween 20), and OD600 was measured every 30 min. The lines and shadows represent the average and standard deviation of quadruplicates, respectively

All strains expressing the Ole1p version after pRP30 showed a substantial decrease in growth and lipid content (Fig. 3B), which could be recovered by supplementing the media with unsaturated fatty acids (Fig. 3C; Additional file 1: Fig. S4). The expression of OLE1-homologs with pRP30 is not sufficient to sustain a normal lipid production phase.

In contrast to pRP30, the pGAPDH-regulated strains showed similar undesired fatty acid profiles and growth as with the native OLE1 promoter (cf. OKYL061 and OKYL063, Fig. 3B, OKYL056 and OKYL059, Fig. 2). Combining pGAPDH-regulated OLE1 expression with the absence of FAD2 also yielded lower growth and lipid content, albeit not as severe as observed when under the control of pRP30 (cf. KSYL09 and KSYL12, Fig. 3, KSYL15 and KSYL16 Fig. 2).

From our first experiments, the strains OKYL056 (pOLE1—RtOLE1) and OKYL061 (pGAPDH—RtOLE1) were the most promising cocoa butter production strains because of their good growth performances and desirable fatty acid profiles. Since deleting FAD2 had a negative impact on the growth and lipid production when combined with promoter exchange of RtOLE1, we decided to continue with OKYL056 for our following experiments.

Promoter exchange of FAD2

Our first experiments focused only on the Δ9 desaturase Ole1p, and we observed a negative impact on growth and lipid production if the Δ12 desaturase Fad2p was deleted in combination with an exchange of Ole1p or its promoter (Figs. 2, 3). To further understand the importance of Fad2p, we exchanged the native FAD2 promoter for testing a range of expression levels. Primarily, we aimed to reduce the linoleic acid (C18:2) content, but we additionally tested strong promoters to explore the range of possible fatty acid profiles which might be of interest for other applications. The native OLE1 promoter is downregulated in the growth phase and upregulated in the lipid production phase. ICL1, RP30 and TEF1 promoter were expected to be very active during the growth phase and downregulated in the lipid production phase. Whereas pICL1 has a lower expression difference between growth and lipid phase. pGAPDH was reported to be active longer during growth (compared to pICL1, pRP30 and pTEF1) but still downregulated during lipid production (pOLE1—Cluster 5c; pICL1—Cluster 1b; pRP30 and pTEF1—Cluster 1a; pGAPDH—Cluster 3 [24]). We first conducted this experiment with strains expressing RtOLE1. FAD2 expression under pICL1 yielded similar results as the native promoter, while pTEF1 and pOLE1 regulated FAD2 expression yielded similar results as with pGAPDH. Therefore, for the strain expressing native YlOLE1 we discarded the pICL1, pTEF1 and pOLE1 regulated FAD2.

We found that the linoleic acid (C18:2) content in the fatty acid profile can vary without a substantial influence on the growth or lipid content (Fig. 4), ranging from no C18:2 (KSYL015) to 16% C18:2 (YMYL03). Additionally, oleic acid (C18:1) and linoleic acid (C18:2) were linked in an inverse correlated manner—similar to stearic acid (C18:0) and oleic acid (C18:1) in our first experiment (Fig. 3). In this case, however, the stearic acid (C18:0) content was largely unaffected by the changes of oleic and linoleic acid. We observed this correlation for both the native OLE1 and RtOLE1.

Fig. 4
figure 4

Promoter exchange of native FAD2 gene. Strains were expressing the native FAD2 gene under the control of different promoters. The strains were either expressing the native YlOLE1 or RtOLE1 under the control of the native OLE1 promoter, respectively. Strains were cultivated for 96 h in LPU media, and the fatty acid profile was determined by FAME extraction. Lipid content represents g FAME per g cell dry weight. Displayed is the mean and standard error of n ≥ 4 replicates

Interestingly, the expression with the strong TEF1 promoter [25] (YMYL04) resulted in a similar fatty acid profile as the expression with the GAPDH or OLE1 promoter (YMYL03 and YMYL05) (Fig. 4), suggesting that high expression levels of desaturases are not the sole factor to influence the fatty acid profile, but that additional regulation exists.

Based on the fatty acid profile, we concluded that the strain OKYL056 (FAD2 wt, RtOLE1) was the most promising strain for CBE production. The strain KSYL15 (fad2Δ, RtOLE1) also showed a promising fatty acid profile, however, the palmitic acid (C16:0) and oleic acid (C18:1) content of OKYL056 was closer to that of cocoa butter.

Characterization of the most promising cocoa butter strain and related strains

To better understand and characterize the most promising CBE production strain (OKYL056), we monitored its growth, lipid production, and fatty acid profile over time in nitrogen-limited cultures. We also included the parental strain (OKYL049) and a strain with additional FAD2 deletion (KSYL15) to understand the effect of the individual genetic modifications.

During the first 24 h after inoculation, lipid content remained stable while fatty acid profiles changed. We observed a shift to higher saturation in the C18 fatty acid species over time in all strains, while the ratio of C16 to C18 remained similar (~ 1:3). In the strains OKYL049 and OKYL056, the relative linoleic acid content (C18:2) decreased between 10 h (30%) and 24 h (10%). Simultaneously, the relative oleic acid content (C18:1) increased by approximately the same percentage (Fig. 5A). Except for the absence of palmitoleic acid (C16:1) in OKYL056, the fatty acid profiles of OKYL049 and OKYL056 were very similar during the first 24 h, while we previously observed that fatty acid profiles at 96 h were different (Fig. 2).

Fig. 5
figure 5

Fatty acid profile during growth and lipid class composition after 96 h of cultivation. Strains were cultivated in LPU media for up to 96 h. Displayed is the mean and standard error of n ≥ 4 replicates. A The fatty acid profile was determined by FAME extraction. Lipid content represents g FAME per g cell dry weight. B Cells were cultured for 96 h, and solid-phase extraction was performed, followed by FAME extraction of each fraction: neutral lipids (NL—including cholesterol, cholesterol ester, triacylglycerol, diacylglycerol, monoacylglycerol), free fatty acids (FFA) and phospholipids (PL). Bars show the percentage of the individual fatty acid of the total fatty acids from each fraction. The colour of the bars is the same as in A. The table shows the contribution of each lipid fraction to the sum of all fractions. Av. average, SE standard error

The FAD2 deleted strain (KSYL15) lacks the Δ12 desaturase to produce linoleic acid (C18:2). Here the oleic acid content (C18:1) stayed constant over the first 18 h, and the amount of C18 fatty acids was similar to the other two strains despite the lack of linoleic acid (C18:2).

After 24 h, all strains started to accumulate lipids, and their fatty acid profiles changed between 24 and 36 h. Although the lipid content kept increasing after 36 h, the fatty acid profile remained constant (Fig. 5A).

These results showed that the fatty acid profile varied depending on the growth phase of the cultures. During the exponential phase, the change was mainly visible in the ratio of saturated to unsaturated fatty acids, while with the beginning of lipid accumulation, the shift occurred in the fatty acid chain length.

We performed solid-phase extraction (SPE) to further understand the growth limitations and the distribution of fatty acids in different lipid classes. We separated the neutral lipids (NL) that mainly consist of TAGs, free fatty acids (FFA) and phospholipids (PL) and analyzed the fatty acid distribution of each fraction.

The fatty acid composition of the neutral lipid (NL) fraction was very close to that of the total FAME fatty acid composition (Figs. 4, 5B) which most likely reflects the dominance of the NL (> 90% of total lipid). The composition of the free fatty acid pool was very different from the NL fraction, being devoid of palmitoleic acid (C16:1) and linoleic acid (C18:2). The absence of C18:2 can be explained by the functionality of the ∆12 desaturase, which only acts on fatty acids bound to membranes [15]. The absence of C16:1 in FFA is potentially of interest, but elucidating the reasons for this would require substantial further work. The phospholipid (PL) fatty acid composition was, in its turn, different from that of the NL and free fatty acids. We observed all dominant fatty acid species except for stearic acid (C18:0) in the PL fraction of the strain expressing both native desaturases (OKYL049). Oleic acid (C18:1) was the most dominant fatty acid with almost 75%. The oleic acid (C18:1) content increased to nearly 98% in KSYL15 (RtOLE1 expression and FAD2 deletion), and PL content increased from 1.83% (OKYL056) to 3.72% (KSYL15) with the deletion of FAD2 (Fig. 5B). This result indicated that a higher amount of PL could compensate for the lack of polyunsaturated fatty acids (C18:2) in the PL fraction. Interestingly, overexpression of FAD2 (YMYL03-05, Fig. 4) led to an overall increase of linoleic acid (C18:2) but did not decrease the total PL amount (Additional file 1: Fig. S5) compared to the wild type expression of FAD2. Different FAD2 expression levels increased both linoleic acid (C18:2) and palmitic acid (C16:0) in the PL (Additional file 1: Fig. S5), indicating a regulation of the cell to maintain membrane fluidity [26].

Additionally, the dominance of oleic acid (C18:1) in the PL fraction can explain why growth-deficient strains did show a higher recovery of growth when supplemented with oleic acid (C18:1) compared to palmitoleic acid (C16:1) (Fig. 2B; Additional file 1: Fig. S3).

Discussion

This study shows that the fatty acid profile of Y. lipolytica can be altered to mimic cocoa butter by exchanging the Δ9 desaturase Ole1p. We found that growth and lipid production are depending on the expression of both desaturases, Ole1p and Fad2p. Taking this into account, the most promising strain for mimicking cocoa butter, in terms of fatty acid profile, growth, and lipid production, was obtained by exchanging the native YlOLE1 with OLE1 from R. toruloides (RtOLE1, strain OKYL056).

Our results showed a broader range of expression levels for FAD2 without a negative impact on lipid production or growth than OLE1. For FAD2, high expression (TEF1 promoter), low expression (ICL1 promoter) and deletion yielded strains with high lipid content and different fatty acid profiles (Fig. 4). However, the expression of OLE1 with different promoters yielded some strains with reduced growth and lipid production or complete growth deficiency (e.g. RP30 promoter) (Fig. 3). In fact, some of the most promising fatty acid profiles correlate with poor growth, e.g. GtOLE1, and OLE1 homologs under control of the RP30 promoter. Adaptive laboratory evolution (ALE) experiments or targetted mutagenesis are approaches that could be used to further increase the growth and lipid content of such strains with favourable fatty acid profiles.

We found a link between the C18 fatty acids, which might be bottlenecks for future engineering of Y. lipolytica for tailored fatty acid profiles. When exchanging the native OLE1 with other homologs, we observed that stearic and oleic acid (C18:0 and C18:1) are inversely correlated, while linoleic acid (C18:2) remained relatively unaffected (Fig. 2). During the promoter exchange of FAD2, we saw an inverse correlation between oleic and linoleic acid (C18:1 and C18:2), while stearic acid (C18:0) was unaffected (Fig. 4). However, these observations are based on total lipid extraction analysis. Because of the high contribution of the neutral lipid fraction (mainly TAG) to the total lipids (Fig. 5), a total lipid extraction is sufficient to estimate the TAG composition of strains. Therefore, the observed inverse correlation applies to TAGs but not necessarily to other lipid classes.

The separation of lipid classes with SPE allowed us to analyze the phospholipids (PL), the main building blocks of cell membranes. Within the PL, an increase of linoleic acid (C18:2) is accompanied by an increase of palmitic acid (C16:0) (Fig. 5; Additional file 1: Fig. S5). Polyunsaturated fatty acids (e.g. C18:2) increase the fluidity of biological membranes, while unsaturated fatty acids (e.g. C16:0) reduce their fluidity [26]. The cell might increase saturated fatty acids (C16:0) to maintain the membrane fluidity for optimal growth. Additionally, we found that oleic acid (C18:1) is the dominant fatty acid in the PL fraction, confirming previous findings [27]. In our engineered strain (KSYL15), oleic acid (C18:1) can make up 98% of the PLs without affecting growth or lipid production. We saw an increase in the total PL fraction in that strain, which could be used for metabolic engineering applications that use PLs as a precursor, e.g. branched fatty acids production [28].

In this study, we mainly focused on mimicking the fatty acid profile of cocoa butter. However, to create proper cocoa butter equivalents (CBE), the TAG structure of natural cocoa butter has to be imitated to ensure similar physical and chemical behaviour of the product [29]. A similar TAG structure can be achieved by introducing GPAT, LPAT and DGAT genes from T. cacao, as previously demonstrated in S. cerevisiae [6]. Most relevant for Y. lipolytica will be the Sn3 position (introduced by DGAT and PDAT), which in Y. lipolytica is naturally occupied by a monounsaturated fatty acid (C16:1 or C18:1), but in cocoa butter is occupied by a saturated fatty acid (C16:0 or C18:0). The Sn1 position in Y. lipolytica is mainly occupied by a saturated and Sn2 by an unsaturated fatty acid similar to cocoa butter. [19]

Other studies in Y. lipolytica to mimic cocoa butter have focused on the growth and fatty acid profile of wild-type strains when fed with different lipidic substrates (e.g. stearin, hydrolyzed oleic rapeseed oil, saturated free fatty acids) [30, 31]. Combining optimized media composition with genetically engineered strains can potentially allow for better tailored fatty acid profiles. This approach would also have the advantage of exploiting Y. lipolytica's ability to consume a wide range of substrates (e.g. lipids or glycerol) to use waste streams as a low-cost carbon source [32, 33].

This study does yield promising results concerning CBE production and demonstrates the possibility of creating a pallet of strains expressing a wide range of fatty acid profiles while still maintaining a high lipid content that can be used for other applications apart from food/feed supplementation [34], e.g. biodiesel with different melting temperatures [35] or biopolymers [36].

Conclusion

In this study, we used genetic engineering of the desaturases in Y. lipolytica and demonstrated the feasibility of producing fatty acid profiles to mimic cocoa butter and other oils. Our study demonstrates the wide range of fatty acid profiles that Y. lipolytica can produce while maintaining good growth and high lipid production. In combination with further genetic engineering and process optimization, e.g., alternative carbon sources, our results can contribute to establishing microbial cell factories for the sustainable production of food oils.

Method

Strains and strain construction

All Yarrowia lipolytica strains in this study are derived from the W29 background strain (Y-63746 from the ARS Culture Collection, Peoria, USA; a.k.a. ATCC20460/CBS7504). The initial starting strain of this study was OKYL029 which carries the MHY1 deletion to prevent filamentous growth [20].

The obese strain (OKYL049) was constructed by additionally deleting the gene ARE1 and overexpressing DGA1 (Tef1in promoter, PEX20 terminator) [21] and is referred to as the wild type (WT) in this study. The genotype and parental relations of all strains are summarized in the Additional file 1: Table.

For the construction of gRNA plasmids for the CRISPR-Cas9 driven gene deletion and marker-free integration, the EasycloneYALI toolbox was used [37]. The deletion repair fragments were constructed from equal amounts of two single-stranded oligonucleotides (around 100 bp; 100 pmol/µL), incubated for 5 min at 95 °C and allowed to cool down to room temperature. 3 µL of this repair fragment was used for deletion. Integration cassettes for marker-free integration were obtained by digesting the corresponding plasmids (see Additional file 1). After a gel cleanup, 1 µg of repair fragment was used for transformation. For each transformation 500 ng of gRNA plasmid was used.

Transformation of Y. lipolytica was performed using a lithium-acetate based heat shock method previously described [38].

Media and growth conditions

Lipid production media (LPU) consisted of 1.5 g/L yeast extract (Merck), 0.85 g/L, casamino acids (Formedium), 1.7 g/L Yeast Nitrogen Base without amino acids and ammonium sulfate (Formedium), 5.1 g/L potassium hydrogen phthalate (Merck) buffer adjusted to pH 5.5, 100 g/L glucose (Sigma-Aldrich), and 0.5 g/L urea (Sigma-Aldrich) [22]. For fatty acid supplementation, 1% (v/v) of Tween 20 (Merck) was added together with the indicated amount of fatty acid (palmitoleic, oleic, heptadecenoic acid, Sigma-Aldrich).

YPD plates contained 20 g/L peptone from meat (Merck), 10 g/L yeast extract, 20 g/L glucose, and 20 g/L agar (VWR). For selection, YPD plates were supplemented with 250 mg/L Nourseothricin (Jena Bioscience). LB plates contained 10 g/L peptone from casein (Merck), 10 g/L NaCl (Merck), 5 g/L yeast extract, 16 g/L agar and were set to pH 7.0 with 5 M NaOH (Merck).

All Y. lipolytica strains were cultivated in suspension at 30 °C in 10 mL media in 100 mL Erlenmeyer flasks at 200 rpm shaking unless differently stated.

Escherichia coli DH5 alpha was used for plasmid construction and purification and was cultivated in LB broth or on agar plates supplemented with 100 µg/mL ampicillin at 37 °C.

Lipid extraction and quantification

Y. lipolytica strains were cultivated in LPU media for 96 h before fatty acid methyl ester (FAME) extraction to measure cellular lipid content. The protocol used was previously described [39]. In short, 100 µL of cell culture was spun down, the supernatant was discarded, and the cells were washed with 1 mL water. The suspension was spun down again, and the supernatant was removed. The cell pellet was dried in a vacuum dry freezer for one day. Then glyceryl triheptadecanoate (TAG(17:0/17:0/17:0)) (Sigma-Aldrich) was added to the cell pellet as an internal standard. 500 µL of a methanol solution containing 1 M NaOH was added, and the samples were vortexed at 1200 rpm at room temperature for 1 h. The solution was neutralized by carefully adding 80 µL of 50% sulfuric acid (Merck). The FAMEs were extracted by adding 500 µL n-hexane (Merck). Phases were separated by centrifugation for 1 min at 10,000×g. 200 µL of the upper hexane phase was mixed with 800 µL hexane, and 1 µL of this sample was analyzed on GC–MS (Thermo Scientific Trace 1310 coupled to a Thermo Scientific ISQ LT with a ZBFAME column (Phenomenex, length: 20 m; Inner Diameter: 0.18 mm; Film Thickness: 0.15 um)) with a method concisting of 2 min hold at 80 °C, followed by a ramp of 40 °C/min until 160 °C, a ramp of 5 °C/min until 185 °C, a ramp of 40 °C/min until 260 and a final hold of 260 °C for 30 s. The GLC reference standard GLC 403 was used as a standard curve.

The biomass/cell dry weight of each culture was calculated by filtration and allowed the calculation of lipid content as g FAME per g CDW [20].

Solid-phase extraction (SPE)

Cell pellets of 100 µL culture were washed with 1 mL water and freeze-dried for one day. A mix of internal standards for each lipid fraction (Glyceryl triheptadecanoate, Pentadecanoic acid, 19:0 PC, Sigma-Aldrich) and ca. 50 µL of acid-washed glass beads were added. 1 ml of chloroform:methanol 2:1 (v:v) (Merck) was added, and the sample was vortexed at 1200 rpm at room temperature for 1 h. SPE cartridges (Supelclean™ LC-NH2 SPE Tube, Sigma-Aldrich) were placed on Vac Elut apparatus, activated by adding 4 mL of hexane, and Samples were loaded on the cartridges. Eluation of the different lipid fractions was achieved by adding 4 mL of chloroform:isopropanol 2:1 (v:v)(Merck) (neutral lipid fraction, including cholesterol, cholesterol ester, triacylglycerol (TAG), diacylglycerol (DAG), monoacylglycerol (MAG)), 2% acetic acid (Sigma-Aldrich) in diethyl ether (Sigma-Aldrich) (fatty acid fraction) and methanol (phospholipid fraction). The solvent of collected lipid fractions was evaporated and resuspended in 1 mL chloroform:methanol 2:1 (v:v), transferred to 1.5 mL reaction tubes and subject to FAME extraction before analysis on GC–MS. [40] For SPE samples the GLC reference standard GLC 426 was used, and the method concisting of 2 min hold at 80 °C, followed by a ramp of 40 °C/min until 120 °C, a ramp of 3.5 °C/min until 165 °C, a ramp of 40 °C/min until 260 and a final hold of 260 °C for 30 s.

Growth profiler

Y. lipolytica strains were grown in LPU media overnight and inoculated in 150 µL of LPU media to a starting OD600 of 0.05 in 96-well plates. The cells were cultivated in quadruplicates at 30 °C and 200 rpm. The growth was measured every 30 min by the Growth Profiler 960 (Enzyscreen B.V., Heemstede, The Netherlands).

Availability of data and materials

The datasets supporting the conclusions of this article are included within the article and its Additional files.

Abbreviations

C16:0:

Palmitic acid

C16:1:

Palmitoleic acid

C18:0:

Stearic acid

C18:1:

Oleic acid

C18:2:

Linoleic acid

ALE:

Adaptive laboratory evolution

CBE:

Cocoa butter equivalents

DGA:

Diacylglycerol

FAME:

Fatty acid methyl ester

FFA:

Free fatty acids

GC–MS:

Gas chromatography–mass spectrometry

HPLC:

High-performance liquid chromatography

MAG:

Monoacylglycerol

NL:

Neutral lipids

ORF:

Open reading frame

PL:

Phospholipids

SPE:

Solid-phase extraction

TAG:

Triacylglyceride

References

  1. Verna R. The history and science of chocolate. Malays J Pathol. 2013;35:111.

    CAS  PubMed  Google Scholar 

  2. Clough Y, Faust H, Tscharntke T. Cacao boom and bust: sustainability of agroforests and opportunities for biodiversity conservation. Conserv Lett. 2009;2:197–205.

    Article  Google Scholar 

  3. Schroth G, Läderach P, Martinez-Valle AI, Bunn C, Jassogne L. Vulnerability to climate change of cocoa in West Africa: patterns, opportunities and limits to adaptation. Sci Total Environ. 2016;556:231–41.

    Article  CAS  Google Scholar 

  4. Bergenholm D, Gossing M, Wei Y, Siewers V, Nielsen J. Modulation of saturation and chain length of fatty acids in Saccharomyces cerevisiae for production of cocoa butter-like lipids. Biotechnol Bioeng. 2018;115:932–42.

    Article  CAS  Google Scholar 

  5. Lipp M, Simoneau C, Ulberth F, Anklam E, Crews C, Brereton P, et al. Composition of genuine cocoa butter and cocoa butter equivalents. J Food Compos Anal. 2001;14:399–408.

    Article  CAS  Google Scholar 

  6. Wei Y, Gossing M, Bergenholm D, Siewers V, Nielsen J. Increasing cocoa butter-like lipid production of Saccharomyces cerevisiae by expression of selected cocoa genes. AMB Express. 2017;7:1–11.

    Article  Google Scholar 

  7. Wei Y, Bergenholm D, Gossing M, Siewers V, Nielsen J. Expression of cocoa genes in Saccharomyces cerevisiae improves cocoa butter production. Microb Cell Fact. 2018;17:11.

    Article  Google Scholar 

  8. Ratledge C. Regulation of lipid accumulation in oleaginous micro-organisms. Biochem Soc Trans. 2002;30:4.

    Article  Google Scholar 

  9. Szczepańska P, Hapeta P, Lazar Z. Advances in production of high-value lipids by oleaginous yeasts. Crit Rev Biotechnol. 2021;42:1–22.

    Article  Google Scholar 

  10. EFSA Panel on Nutrition NF, Allergens F, Turck D, Castenmiller J, de Henauw S, Hirsch-Ernst K, et al. Safety of Yarrowia lipolytica yeast biomass as a novel food pursuant to Regulation (EU) 2015/2283. EFSA J. 2019;17:e05594.

    Google Scholar 

  11. Miller KK, Alper HS. Yarrowia lipolytica: more than an oleaginous workhorse. Appl Microbiol Biotechnol. 2019; 1–12.

  12. Spagnuolo M, Hussain MS, Gambill L, Blenner M. Alternative substrate metabolism in Yarrowia lipolytica. Front Microbiol. 2018;9.

  13. Beopoulos A, Cescut J, Haddouche R, Uribelarrea J-L, Molina-Jouve C, Nicaud J-M. Yarrowia lipolytica as a model for bio-oil production. Prog Lipid Res. 2009;48:375–87.

    Article  CAS  Google Scholar 

  14. Xu P, Qiao K, Ahn WS, Stephanopoulos G. Engineering Yarrowia lipolytica as a platform for synthesis of drop-in transportation fuels and oleochemicals. Proc Natl Acad Sci. 2016;113:10848–53.

    Article  CAS  Google Scholar 

  15. Ferrante G, Kates M. Pathways for desaturation of oleoyl chains in Candida lipolytica. Can J Biochem Cell Biol. 1983;61:1191–6.

    Article  CAS  Google Scholar 

  16. Mitchell AG, Martin CE. A novel cytochrome b5-like domain is linked to the carboxyl terminus of the Saccharomyces cerevisiae Δ-9 fatty acid desaturase. J Biol Chem. 1995;270:29766–72.

    Article  CAS  Google Scholar 

  17. Nagao K, Murakami A, Umeda M. Structure and function of Δ9-fatty acid desaturase. Chem Pharm Bull. 2019;67:327–32.

    Article  CAS  Google Scholar 

  18. Nachtschatt M, Okada S, Speight R. Integral membrane fatty acid desaturases: a review of biochemical, structural, and biotechnological advances. Eur J Lipid Sci Technol. 2020;122:2000181.

    Article  CAS  Google Scholar 

  19. Wei Y, Siewers V, Nielsen J. Cocoa butter-like lipid production ability of non-oleaginous and oleaginous yeasts under nitrogen-limited culture conditions. Appl Microbiol Biotechnol. 2017;101:3577–85.

    Article  CAS  Google Scholar 

  20. Konzock O, Norbeck J. Deletion of MHY1 abolishes hyphae formation in Yarrowia lipolytica without negative effects on stress tolerance. PLoS ONE. 2020;15:e0231161.

    Article  CAS  Google Scholar 

  21. Konzock O, Zaghen S, Norbeck J. Tolerance of Yarrowia lipolytica to inhibitors commonly found in lignocellulosic hydrolysates. BMC Microbiol. 2021;21:1–10.

    Article  Google Scholar 

  22. Tsakraklides V, Kamineni A, Consiglio AL, MacEwen K, Friedlander J, Blitzblau HG, et al. High-oleate yeast oil without polyunsaturated fatty acids. Biotechnol Biofuels. 2018;11:131.

    Article  Google Scholar 

  23. Tezaki S, Iwama R, Kobayashi S, Shiwa Y, Yoshikawa H, Ohta A, et al. Δ12-fatty acid desaturase is involved in growth at low temperature in yeast Yarrowia lipolytica. Biochem Biophys Res Commun. 2017;488:165–70.

    Article  CAS  Google Scholar 

  24. Morin N, Cescut J, Beopoulos A, Lelandais G, Le Berre V, Uribelarrea J-L, et al. Transcriptomic analyses during the transition from biomass production to lipid accumulation in the oleaginous yeast Yarrowia lipolytica. PLoS ONE. 2011;6:e27966.

    Article  CAS  Google Scholar 

  25. Shabbir Hussain M, Gambill L, Smith S, Blenner MA. Engineering promoter architecture in oleaginous yeast Yarrowia lipolytica. ACS Synth Biol. 2015;5:213–23.

    Article  Google Scholar 

  26. Cronan JE, Gelmann EP. Physical properties of membrane lipids: biological relevance and regulation. Bacteriol Rev. 1975;39:25.

    Article  Google Scholar 

  27. Sekova VY, Dergacheva DI, Isakova EP, Gessler NN, Tereshina VM, Deryabina YI. Soluble sugar and lipid readjustments in the Yarrowia lipolytica yeast at various temperatures and pH. Metabolites. 2019;9:307.

    Article  CAS  Google Scholar 

  28. Blitzblau HG, Consiglio AL, Teixeira P, Crabtree DV, Chen S, Konzock O, et al. Production of 10-methyl branched fatty acids in yeast. Biotechnol Biofuels. 2021;14:12.

    Article  CAS  Google Scholar 

  29. Lippe M, Anklam E. Review of cocoa butter and alternative fats for use in chocolate—part A. Composit Data Food Chem. 1998;62:73–97.

    Article  Google Scholar 

  30. Papanikolaou S, Chevalot I, Komaitis M, Aggelis G, Marc I. Kinetic profile of the cellular lipid composition in an oleaginous Yarrowia lipolytica capable of producing a cocoa-butter substitute from industrial fats. Antonie Van Leeuwenhoek. 2001;80:215–24.

    Article  CAS  Google Scholar 

  31. Papanikolaou S, Muniglia L, Chevalot I, Aggelis G, Marc I. Accumulation of a cocoa-butter-like lipid by Yarrowia lipolytica cultivated on agro-industrial residues. Curr Microbiol. 2003;46:0124–30.

    Article  CAS  Google Scholar 

  32. Papanikolaou S, Aggelis G. Yarrowia lipolytica: a model microorganism used for the production of tailor‐made lipids. Eur J Lipid Sci Technol. 2010;16.

  33. Dobrowolski A, Drzymała K, Mituła P, Mirończuk AM. Production of tailor-made fatty acids from crude glycerol at low pH by Yarrowia lipolytica. Biores Technol. 2020;314:123746.

    Article  CAS  Google Scholar 

  34. Vasconcelos B, Teixeira JC, Dragone G, Teixeira JA. Oleaginous yeasts for sustainable lipid production—from biodiesel to surf boards, a wide range of “green” applications. Appl Microbiol Biotechnol. 2019. https://doi.org/10.1007/s00253-019-09742-x.

    Article  PubMed  Google Scholar 

  35. Magalhães AMS, Pereira E, Meirelles AJA, Sampaio KA, Maximo GJ. Proposing blends for improving the cold flow properties of ethylic biodiesel. Fuel. 2019;253:50–9.

    Article  Google Scholar 

  36. Cordova LT, Walton H, Biswas E, Monroe K, Quirino R, Alper HS. Data-driven approach to decipher the role of triglyceride composition on the thermomechanical properties of thermosetting polymers using vegetable and microbial oils. ACS Appl Polym Mater. 2021;3:4485–94.

    Article  CAS  Google Scholar 

  37. Holkenbrink C, Dam MI, Kildegaard KR, Beder J, Dahlin J, Belda DD, et al. EasyCloneYALI: CRISPR/Cas9-based synthetic toolbox for engineering of the yeast Yarrowia lipolytica. Biotechnol J. 2018;13:1700543.

    Article  Google Scholar 

  38. Chen D-C, Beckerich J-M, Gaillardin C. One-step transformation of the dimorphic yeast Yarrowia lipolytica. Appl Microbiol Biotechnol. 1997;48:232–5.

    Article  CAS  Google Scholar 

  39. Qiao K, Abidi SHI, Liu H, Zhang H, Chakraborty S, Watson N, et al. Engineering lipid overproduction in the oleaginous yeast Yarrowia lipolytica. Metab Eng. 2015;29:56–65.

    Article  CAS  Google Scholar 

  40. Kaluzny MA, Duncan LA, Merritt MV, Epps DE. Rapid separation of lipid classes in high yield and purity using bonded phase columns. J Lipid Res. 1985;26:135–40.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

We would also like to thank Eduard Kerkhoven for his valuable comments on the manuscript.

Funding

Open access funding provided by Chalmers University of Technology. This work was supported by the FORMAS (formas.se) Grant Number 2017-01281.

Author information

Authors and Affiliations

Authors

Contributions

OK and JN planned the study design and supervised the experimental work. OK, YM, SZ and AS performed the experimental work. OK, and YM analyzed the data. OK wrote the original manuscript. OK and JN revised the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Joakim Norbeck.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1.

Supplementary figures and list of Strains, Primers, and DNA sequences used in this study.

Additional file 2.

Raw data.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Konzock, O., Matsushita, Y., Zaghen, S. et al. Altering the fatty acid profile of Yarrowia lipolytica to mimic cocoa butter by genetic engineering of desaturases. Microb Cell Fact 21, 25 (2022). https://doi.org/10.1186/s12934-022-01748-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12934-022-01748-x

Keywords