Assessment of Fatty Acids Profile and Omega-3 Polyunsaturated Fatty Acid Production by the Oleaginous Marine Thraustochytrid Aurantiochytrium sp. T66 Cultivated on Volatile Fatty Acids

Thraustochytrids are considered natural producers of omega-3 fatty acids as they can synthesize up to 70% docosahexaenoic acids (DHA) of total lipids. However, commercial and sustainable production of microbial DHA is limited by elevated cost of carbon substrates for thraustochytrids cultivation. This problem can be addressed by utilizing low-cost renewable substrates. In the present study, growth, lipid accumulation and fatty acid profiles of the marine thraustochytrid Aurantiochytrium sp. T66 (ATCC-PRA-276) cultivated on volatile fatty acids (C1, formic acid; C2, acetic acid; C3, propionic acid; C4, butyric acid; C5, valeric acid and C6, caproic acid) and glucose as control were evaluated for the first time. This strain showed an inability to utilize C3, C5 and C6 as a substrate when provided at >2 g/L, while efficiently utilizing C2 and C4 up to 40 g/L. The highest cell dry weight (12.35 g/L) and total lipid concentration (6.59 g/L) were attained when this strain was cultivated on 40 g/L of butyric acid, followed by cultivation on glucose (11.87 g/L and 5.34 g/L, respectively) and acetic acid (8.70 g/L and 3.43 g/L, respectively). With 40 g/L butyric acid, the maximum docosahexaenoic acid content was 2.81 g/L, corresponding to 42.63% w/w of total lipids and a yield of 0.23 g/gcell dry weight (CDW). This marine oleaginous microorganism showed an elevated potential for polyunsaturated fatty acids production at higher acetic and butyric acid concentrations than previously reported. Moreover, fluorescence microscopy revealed that growth on butyric acid caused cell size to increase to 45 µm, one of the largest values reported for oleaginous microorganisms, as well as the presence of numerous tiny lipid droplets.


Introduction
Omega-3 and omega-6 fatty acids are considered essential fatty acids in the human diet because humans are unable to synthesize them in sufficient amounts [1]. Biosynthesis of eicosapentaenoic acid (C20:5n−3; EPA), docosapentaenoic acid (C22:5n−3, DPA) and docosahexaenoic acid (C22:6n−3, DHA) occurs from the parent omega-3 fatty acid α-linolenic acid (C18:3n−3), however, the conversion rates are too low to meet daily requirements [2,3]. Therefore, it is recommended to consume EPA and DHA through the diet. DHA plays an important role in the function and responsiveness of cell membranes, tissue metabolism and hormonal and other signaling pathways [4,5]. It also constitutes the major fatty acid in the brain and retina [6]. Significant quantities of DHA need to be supplemented by the diet during DHA content was also studied. By understanding which specific VFAs promote high growth and DHA production, it will be possible to fine-tune the dark fermentation process and deliver a mixture of VFAs suitable for DHA production. Ultimately, the current work aims to give additional value to the traditional bio-hydrogen production process, by providing a nutraceutical compound (DHA) alongside bioenergy (hydrogen), while simultaneously treating organic waste streams.

Microorganism and Cultivation
The marine thraustochytrid Aurantiochytrium sp. T66 (ATCC-PRA-276) was procured from the ATCC. It was initially cultured on ATCC ® 790 By + medium containing yeast extract (1 g/L), peptone (1 g/L), glucose (5 g/L) and seawater (1000 mL) at 25 • C with 180 rpm for 48 h. After that, the thraustochytrid was cultivated at 25 • C with 180 rpm on seed culture medium containing glucose (20 g/L), yeast extract and 50% artificial seawater for 24 h. Artificial seawater composition was similar to that reported previously [36].

Batch Cultivation on Different Volatile Fatty Acids
The commercial VFAs formic acid (C1), acetic acid (C2), propionic acid (C3), butyric acid (C4), caproic acid (C5) and valeric acid (C6) were purchased from Sigma Aldrich (St. Louis, MO, USA). All VFAs were used as sole carbon source without any further sugar addition. Initially, 2 g/L of each VFA were mixed with an appropriate amount of yeast extract to maintain a C/N ratio of 10, followed by addition of 50% v/v artificial seawater. Tris (hydroxymethyl) aminomethane was used as buffer at a concentration of 100 mM. The volume was adjusted to 90% of the final volume and the pH was set to 6.8 with 3 M NaOH and 3 M HCl. Cultivation experiments were carried out in 250-mL Erlenmeyer flasks with 100 mL of working solution. Seed culture (10% v/v) was used to inoculate the medium and flasks were incubated in an orbital shaker with 180 rpm at 25 • C until stationary phase was reached. Growth was monitored by taking optical density readings at 680 nm (OD680) with a UV-Vis spectrophotometer (Spectra Max M2; Molecular Devices, San Jose, CA, USA). For VFAs that Aurantiochytrium sp. T66 could grow on, the concentration was gradually increased up to the point where the strain was not able to consume it. The C/N ratio for each trial was adjusted to 10 with yeast extract.

Morphologic Analysis and Estimation of Lipid Accumulation by Fluorescence Microscopy
Aliquots of 1 mL were harvested from the culture at early stationary phase, centrifuged and resuspended in 100 µL of 0.9% saline solution. 4,4-difluoro-1,3,5,7,8-pentamethyl-4-bora-3a,4adiaza-s-indacene (BODIPY 493/503 ) stock solution was prepared in DMSO (0.1 mg/mL) and 2 µL of it was added to the 100-µL samples. The samples were incubated for 5 min in the dark and imaging was carried out on a digital inverted fluorescence microscope equipped with a GFP light cube (EVOS-FL, (Thermo Fisher Scientific, Waltham, MA, USA). Cell size was determined by ImageJ 1.8.0 software (NIH, Bethesda, MD, USA).

Cell Dry Weight (CDW) Determination
At the end of each cultivation the cells were harvested to determine the cell dry weight. The culture was harvested into 50-mL tubes as soon as it reached stationary phase, as determined by daily OD680 measurements and centrifuged at 8000 rpm (7881× g) for 10 min in an Eppendorf 5804 R (Eppendorf Nordic A/S, Sverige Filial, Gulbackegatan, Malmö, Sweden) centrifuge with an F-34-6-38 rotor. Next, 2 mL of the supernatant was used to determine the residual carbon source (see Section 2.5). The cell pellet was washed three times with deionized (DI)-water, resuspended in DI-water, and finally pipetted into a preweighed metallic pan. The pans were dried in an oven at 50 • C until constant weight was attained. Cell dry weight was then determined gravimetrically.

Total Lipid Determination
To determine the amount of lipids in the cells, the dried biomass was milled with a mortar and pestle into a fine powder according to our previous report [37]. The powder was blended with chloroform:methanol (2:1, v/v) and incubated for 2 h under shaking. Subsequently, DI-water was added to the slurry at a volume equal to the volume of the slurry. The mixture was mixed thoroughly and afterwards centrifuged at 8000 rpm (7881× g) for 10 min. The bottom clear phase was aspirated into a preweighed glass plate and placed into a hot-air oven at 50 • C to evaporate the solvent. Lipid weight was then determined gravimetrically, and lipids were stored in a freezer at −20 • C until further analysis.

Residual Carbon Estimation by High-Performance Liquid Chromatography (HPLC)
Residual carbon source concentration was determined at the end of the cultivation by high-performance liquid chromatography (HPLC) analysis of the supernatant. Samples were prepared by filtration through a 0.2-µm syringe filter (Sartorius TM Minisart TM RC) into HPLC vials. A PerkinElmer (Waltham, Massachusetts, USA) Series 200 HPLC instrument equipped with a Bio-Rad Aminex HPX-87H Column (#1250140) was used to analyze C1, C2, C3, C4 and glucose. The analysis was programmed for 30 min and the column temperature was set to 65 • C with 5-mM H 2 SO 4 as mobile solvent. A refractive index (RI) detector was used. Determination of residual C5 and C6 concentrations was performed as described above, with the difference that analysis time was extended to 50 min and the mobile phase was 90% v/v 5-mM H 2 SO 4 and 10% v/v acetonitrile. Detection took place with a UV-Vis detector PerkinElmer (Waltham, Massachusetts, USA) set to 210 nm.

Analysis of Fatty Acid Profile by Gas Chromatography-Mass Spectrometry (GC-MS)
Lipids were transesterified as described by Wychen et al. (2013) [38] with some modifications. Specifically, a 50 to 100 mg lipid sample was dissolved in 2 mL of chloroform:methanol solution (2:1, v/v) in a screw-capped glass vial and 3 mL of 0.6 M HCl:methanol was added to it. The liquid was mixed and placed in a preheated water bath at 85 • C for 1 h. After that, it was cooled at room temperature to avoid extensive evaporation. The cooled liquid was transferred into 50-mL tubes and 3 mL HPLC-grade hexane was added. The tubes were mixed well and centrifuged at 8000 rpm (7881× g) for 10 min. The upper hexane layer was transferred to new glass vials. The fatty acid profile was determined by GC (Clarus 690; PerkinElmer) with MS (Clarus SQ8; PerkinElmer) using a capillary column (Elite 5MS; 30 m, 0.25 mm ID, 0.25 µm df, Cat. # N9316282; PerkinElmer). The oven was programmed to 50 • C for 0.5 min; temperature was then ramped to 194 • C at 30 • C/min for 3.5 min and then to 240 • C at 5 • C/min, where it was held for 10 min. The injection port temperature was adjusted to 250 • C and 1 µL of sample was injected with He as carrier gas in split mode (10:1). Solvent delay time was 3 min. The MS transfer line temperature was adjusted to 250 • C with 170 • C source temperature. Mass spectra (mass range 50 to 400 m/z) were recorded at 3 scans/s with electron ionization at 70 eV. Individual fatty acid methyl ester (FAME) peaks were identified by their mass spectra and further confirmed by a standard (Supelco 37 Component FAME Mix: 47885-U; Sigma-Aldrich (St. Louis, MO, USA). DHA concentration (g/L) and yield (mg/g CDW ) were estimated by the following equations:

Statistical Analysis
All experiments in this study were conducted in triplicate. All data are expressed as a mean ± standard deviation. The experimental data were subjected to one-way analysis of variance (ANOVA) as implemented in the Graph Pad Prism 8.4.2, 676 (GraphPad Software, San Diego, CA, USA) statistics platform. Tukey simultaneous tests were conducted to determine the statistical differences between the results obtained with various carbon sources. To ascertain that the observed variations were statistically significant, the probability (p) values were determined. A 95% confidence level (p < 0.05) was applied for all analyses.

Batch Cultivation of Aurantiochytrium sp. T66 on Different Volatile Fatty Acids
In the present study, the marine oleaginous thraustochytrid Aurantiochytrium sp. T66 was grown on artificial medium containing various concentrations of VFAs. When Aurantiochytrium sp. T66 was cultivated in 2 g/L of C1, C2, C3, C4, C5 and glucose ( Figure 1A), cell dry weight was 0.35 g/L, 0.35 g/L, 0.34 g/L, 0.52 g/L, 0.45 g/L, 0.34 g/L and 0.33 g/L, respectively. The corresponding total lipid concentration was 0.07 g/L, 0.10 g/L, 0.06 g/L, 0.12 g/L, 0.08 g/L, 0.04 g/L and 0.08 g/L ( Figure 1). Total carbon source consumption was complete only when cultivations were carried out in C2, C4 and glucose. The value of 2 g/L was chosen because it had been reported as the upper limit to avoid inhibition in other oleaginous microorganisms. One of them, the heterotrophic microalga Chlorella protothecoides (UTEX 25), exhibited optimal growth at 2 g/L VFAs (acetic acid:propionic acid:butyric acid; 6:1:3), resulting in 0.58 g/L cell dry weight and 0.278 g/L lipids [32]. Ramping the concentration of VFAs from 2 g/L to 4 g/L, severely inhibited growth and lipid accumulation by this microalga, which dropped to 0.33 g/L and 0.136 g/L, respectively [32]. In an effort to identify whether Aurantiochytrium sp. T66 could tolerate higher VFA concentrations, 10 g/L of carbon source was used ( Figure 1B). Cell dry weight was now 0.25 g/L, 2.10 g/L, 0.38 g/L, 3.23 g/L, 1.46 g/L, 0.51 g/L and 2.68 g/L in C1, C2, C3, C4, C5 and glucose, respectively. The respective lipid content was 10.57%, 30.32%, 12.90%, 43.11%, 12.60%, 15.45% and 41.13% ( Figure 1B). The above findings showed that C2 and C4 were suitable substrates, with C4 in particular yielding better results than glucose. Almost complete consumption of C2 and C4 was observed at the end of cultivation. In contrast, thraustochytrids cultivated on C1, C3 and C6 exhibited growth inhibition and final cell dry weights below 1 g/L at both 2 g/L and 10 g/L of the substrate (Figure 1). Growth on C5 was partially inhibited and C5 consumption was 31.85%.
Based on the results obtained so far, the most suitable VFAs appeared to be acetic acid (C2) and butyric acid (C4); hence, their concentration was increased to 20 g/L (Figure 2A). This resulted in a The above findings showed that C2 and C4 were suitable substrates, with C4 in particular yielding better results than glucose. Almost complete consumption of C2 and C4 was observed at the end of cultivation. In contrast, thraustochytrids cultivated on C1, C3 and C6 exhibited growth inhibition and final cell dry weights below 1 g/L at both 2 g/L and 10 g/L of the substrate (Figure 1). Growth on C5 was partially inhibited and C5 consumption was 31.85%.
Based on the results obtained so far, the most suitable VFAs appeared to be acetic acid (C2) and butyric acid (C4); hence, their concentration was increased to 20 g/L (Figure 2A). This resulted in a cell dry weight of 5.68 g/L, 8.00 g/L and 6.87 g/L when the microorganism was cultivated on C2, C4 and glucose, respectively, and a corresponding lipid concentration of 2.44 g/L, 4.05 g/L and 3.34 g/L. Because the cells showed no growth inhibition under these conditions, the concentration of VFAs was further increased to 40 g/L ( Figure 2B). The highest cell dry weight (12.35 g/L) and total lipid concentration (6.59 g/L) was obtained with C4, followed by glucose (11.87 g/L and 5.34 g/L, respectively), whereas C2 resulted in lower biomass and lipid accumulation. Further increasing the concentration of VFAs (C2 and C4) to 60 g/L resulted in growth inhibition, which was not observed with 60 g/L of glucose (data not shown). Notably, 40 g/L of VFAs is higher than what was reported as tolerable by other oleaginous microorganisms. However, no study was reported to utilize VFAs as a sole carbon source by thraustochytrids. Among various types of oleaginous microorganisms only the oleaginous microalgae Crypthecodinium cohnii could utilize a high concentration of acetic acid and butyric acid as sole carbon sources for PUFA production. When Chalima et al. (2019) cultivated C. cohnii on various VFAs at an initial concentration of 5 to 50 g/L, the highest amount of biomass was obtained on 30 g/L of acetate (~6 g/L), 10 g/L of propionate (~3.8 g/L) and 15 g/L of butyrate (~3.9 g/L), while further increasing the concentration of substrates inhibited microbial growth [35]. Likewise, Ratledge et al. (2001) suggested that C. cohnii could grow on up to 1-16 g/L sodium acetate as a sole carbon source in a pH auxostat with acetate used to control the pH of the medium [39]. In another study, the freshwater microalga C. protothecoides was cultivated heterotrophically in waste activated sludge whose total chemical oxygen demand was 3840 mg/L, including acetic acid (1.2 g/L), propionic acid (0.45 g/L), butyric acid (0.23 g/L), isobutyric acid (0.24 g/L), valeric acid (0.36 g/L) and isovaleric acid (0.14 g/L) [40]. Waste activated sludge yielded a cell dry weight of 0.34 g/L; which was only slightly lower than that obtained on 15 g/L of pure glucose (0.63 g/L) or waste activated sludge enriched with selenite (0.50 g/L) [40]. Another freshwater microalga, Chlorella sorokiniana, was also heterotrophically cultivated on acetate, butyrate and a mixture of butyrate and acetate, but the concentration of these substrates was considerably lower (500-mM acetate and butyrate) than in the present study [41].
Cells grown on glucose, acetic acid and butyric acid were observed under a fluorescence microscope. The cells cultivated on butyric acid had a significantly larger size (41.23 ± 6.75 µm) and more lipid droplets compared to those cultivated on acetic acid, which was of intermediate size (20.09 ± 3.38 µm), and those grown on glucose, which were significantly smaller (11.11 ± 2.31 µm) (Figure 3). The cell number were high in the case of cells grown in glucose than the C2 and C4 cultivated cells while the lipid droplets per cell were high in the case of C4 grown cells. The largest cells were estimated to be almost 40 to 45 µm in diameter and, to the best of our knowledge, this is the largest oleaginous microorganism ever reported. While the exact mechanism for such a dramatic increase in size remains unknown, a possible reason may be the profound effect of butyric acid on metabolic flux and fatty acid synthesis [42]. sludge whose total chemical oxygen demand was 3840 mg/L, including acetic acid (1.2 g/L), propionic acid (0.45 g/L), butyric acid (0.23 g/L), isobutyric acid (0.24 g/L), valeric acid (0.36 g/L) and isovaleric acid (0.14 g/L) [40]. Waste activated sludge yielded a cell dry weight of 0.34 g/L; which was only slightly lower than that obtained on 15 g/L of pure glucose (0.63 g/L) or waste activated sludge enriched with selenite (0.50 g/L) [40]. Another freshwater microalga, Chlorella sorokiniana, was also heterotrophically cultivated on acetate, butyrate and a mixture of butyrate and acetate, but the concentration of these substrates was considerably lower (500-mM acetate and butyrate) than in the present study [41]. Cells grown on glucose, acetic acid and butyric acid were observed under a fluorescence microscope. The cells cultivated on butyric acid had a significantly larger size (41.23 ± 6.75 µm) and more lipid droplets compared to those cultivated on acetic acid, which was of intermediate size (20.09 ± 3.38 µm), and those grown on glucose, which were significantly smaller (11.11 ± 2.31 µm) ( Figure  3). The cell number were high in the case of cells grown in glucose than the C2 and C4 cultivated cells while the lipid droplets per cell were high in the case of C4 grown cells. The largest cells were estimated to be almost 40 to 45 µm in diameter and, to the best of our knowledge, this is the largest oleaginous microorganism ever reported. While the exact mechanism for such a dramatic increase in size remains unknown, a possible reason may be the profound effect of butyric acid on metabolic flux and fatty acid synthesis [42].

Effect of Various Volatile Fatty Acids on the Fatty Acid Profile of Aurantiochytrium sp. T66
According to a literature survey by Morabito et al. (2019), more than 65% w/w of total lipids accumulated by thraustochytrids is made of C16:0 (palmitic acid) and C22:6 (DHA), which corresponds to one of the highest DHA contents among PUFA-producing microorganisms [43]. The total amount of lipids and their profile varies among thraustochytrid species, with the genera Aurantiochytrium, Thraustochytrium and Schizochytrium considered the most productive for DHA [44].

Effect of Various Volatile Fatty Acids on the Fatty Acid Profile of Aurantiochytrium sp. T66
According to a literature survey by Morabito et al. (2019), more than 65% w/w of total lipids accumulated by thraustochytrids is made of C16:0 (palmitic acid) and C22:6 (DHA), which corresponds to one of the highest DHA contents among PUFA-producing microorganisms [43]. The total amount of lipids and their profile varies among thraustochytrid species, with the genera Aurantiochytrium, Thraustochytrium and Schizochytrium considered the most productive for DHA [44].
When cultivation was shifted from 2 to 10 g/L of VFAs, the main fatty acids were C14:0, C15:0, C16:0, C17:0, C18:0, C18:1, as well as long-chain fatty acids such as C20:4n6, DPA and DHA ( Table 2). The DHA content (w/w total lipids ) was 9.57%, 28.75%, 7.28%, 18.64%, 7.88%, 6.01% and 4.14% when cultivation was carried out in C1, C2, C3, C4, C5, C6 and glucose, respectively ( Table 2). The highest DHA content was observed with C2 as substrate followed by C4. A possible reason is the direct production of acetyl-CoA, a key intermediate in fatty acid synthesis, from VFA via a non-conventional route [54]. When oleaginous microorganisms are grown on glucose as sole carbon source, the cells start assimilating glucose for glycolysis and form pyruvate, which is transported to mitochondria and fed into the tricarboxylic acid cycle. This yields citrate, which is transported back to the cytosol, where it is converted into acetyl-CoA by ATP:citrate lyase [55]. In contrast, when oleaginous cells assimilate acetic acid, this is converted directly into acetyl-CoA in the cytosol by acetyl-CoA synthetase in a single step [47,54]. Table 1. Fatty acid composition (wt% of total fatty acids) of Aurantiochytrium sp. T66 cultivated on 2 g/L of various VFAs (C1:0, formic acid; C2:0, acetic acid; C3:0, propionic acid; C4:0, butyric acid; C5:0, valeric acid and C6:0, caproic acid). Glucose (GC) was used as control.  Table 2. Fatty acid composition (wt% of total fatty acids) of Aurantiochytrium sp. T66 cultivated on 10 g/L of various VFAs (C1:0, formic acid; C2:0, acetic acid; C3:0, propionic acid; C4:0, butyric acid; C5:0, valeric acid and C6:0, caproic acid). Glucose (GC) was used as control. Finally, Aurantiochytrium sp. T66 was shifted to an even higher concentration of the substrate to study the effect on DHA production of 20 g/L and 40 g/L of C2 or C4. At 20 g/L of C2, C4 or glucose, DHA content increased to 30.40%, 36.25% and 29.58%, respectively (Table 3); the other major fatty acids being produced were C15:0, C16:0, C17:0, C18:0, C18:3n3 and DPA. When cells are cultivated on C2 compounds such as acetic acid, the tricarboxylic acid cycle cannot provide sufficient carbon skeleton building blocks, ATP and reducing power (NADH). This shortage prompts the induction of another metabolic pathway, the glyoxylate cycle, and the activation of enzymes, such as isocitrate lyase and malate synthase [56,57]. It has been suggested that the oleaginous microalgae Chlorella vulgaris can express the glyoxylate cycle enzymes even when cultivated on glucose, however they are not active unless the microorganism is grown on acetate [58]. The activity and concentration of isocitrate lyase was found to be increased in another microalga, Scenedesmus obliquus, after being cultivated on acetate in the dark; moreover, the concentration of this enzyme increased with increasing concentration of acetate [59]. In the present study, when cells were grown in 40 g/L C2 and C4, DHA content increased to 36.93% and 42.63%, respectively, while on glucose the DHA concentration was 38.72% ( Table 4). The highest DHA concentration (2.81 g/L) and DHA yield (0.23 g/g CDW ) were observed when cells were cultivated in 40 g/L of C4, followed by 40 g/L of glucose (2.07 g/L and 0.17 g/g CDW , respectively) ( Figure 4). Addition of 1% v/v n-dodecane (C12) to the medium of the marine microalga C. cohnii was shown to enhance cell dry weight and lipid accumulation, including DHA content (51% w/w of total lipids) [60]. The authors suggested that the hydrocarbon in the medium acted as an oxygen vector and could improve oxygen availability to the microorganisms, thus enhancing DHA productivity. At the same time, the total content of unsaturated fatty acids (16:1n7, 18:1n9 and 22:5n3) remained unaffected, suggesting that the increase in DHA was independent of the oxygen-dependent desaturase [60]. Table 3. Fatty acid composition (wt% of total fatty acids) of Aurantiochytrium sp. T66 cultivated on 20 g/L of acetic acid (C2) or butyric acid (C4). Glucose (GC) was used as control.

Conclusions
The present study indicates that VFAs can be utilized efficiently as a sole carbon source for the cultivation of oleaginous marine thraustochytrids. Cell dry weight and lipid production were particularly high when Aurantiochytrium sp. T66 was cultivated on butyric acid as sole carbon source. The fatty acid profile of this strain revealed up to 42.63% of DHA in total lipids after cultivation on VFAs. The maximum DHA concentration (2.81 g/L) and DHA yield (0.23 g/gCDW) were attained when cells were cultivated in 40 g/L butyric acid, and were even higher than those observed in 40 g/L glucose (2.07 g/L and 0.17 g/gCDW, respectively). The important finding of this study was the dramatic increment in cell size after cultivation on 40 g/L of butyric acid that provides a foundation for the further exploration of the assimilation pathway and metabolic flux in thraustochytrids. Moreover, replacing costly refined sugars as carbon source with VFAs derived from abundant waste streams will make microbial PUFA synthesis more economical at an industrial scale.
Funding: This work was part of the project 'Tuned volatile fatty acids production from organic waste for biorefinery platforms (VFA biorefinery)' funded by the Swedish Research Council (FORMAS) with reference number 2018-00818.

Conflicts of Interest:
The authors declare no conflict of interest.

Conclusions
The present study indicates that VFAs can be utilized efficiently as a sole carbon source for the cultivation of oleaginous marine thraustochytrids. Cell dry weight and lipid production were particularly high when Aurantiochytrium sp. T66 was cultivated on butyric acid as sole carbon source. The fatty acid profile of this strain revealed up to 42.63% of DHA in total lipids after cultivation on VFAs. The maximum DHA concentration (2.81 g/L) and DHA yield (0.23 g/g CDW ) were attained when cells were cultivated in 40 g/L butyric acid, and were even higher than those observed in 40 g/L glucose (2.07 g/L and 0.17 g/g CDW , respectively). The important finding of this study was the dramatic increment in cell size after cultivation on 40 g/L of butyric acid that provides a foundation for the further exploration of the assimilation pathway and metabolic flux in thraustochytrids. Moreover, replacing costly refined sugars as carbon source with VFAs derived from abundant waste streams will make microbial PUFA synthesis more economical at an industrial scale.