Metabolomics analysis reveals 6-benzylaminopurine as a stimulator for improving lipid and DHA accumulation of Aurantiochytriumsp.

BACKGROUND: Phytohormones are chemical messengers that have a positive effect on biodiesel production of microalgae at low concentrations. However, the effect of phytohormone 6-benzylaminopurine on lipid and docosahexaenoic acid (DHA) production in marine DHA-producer Aurantiochytrium has never been reported. In this study, a GC-MS-based metabolomics method combined with a multivariate analysis is applied to reveal the metabolic mechanism of 6-benzylaminopurine enhancing production of lipid and DHA in Aurantiochytrium sp.YLH70.
RESULTS: In total, 71 metabolites were identified by GC-MS. The PCA model revealed that 76.9% of metabolite variation was related to 6-benzylaminopurine treatment, and overall metabolomics profiles between the 6-benzylaminopurine and control groups were clearly discriminated. Forty-six metabolites identified by the PLS-DA model were responsible for responding to 6-benzylaminopurine. Metabolic analysis showed that 6-benzylaminopurine could accelerate the rate of utilization of glucose in Aurantiochytrium sp. YLH70, and the metabolic flux from glycolysis, TCA cycle and mevalonate pathway to fatty acids biosynthesis was promoted. Moreover, the anti-stress mechanism in Aurantiochytrium sp.YLH70 might be induced by 6-benzylaminopurine.
CONCLUSION: Metabolomics is a suitable tool to discover the metabolic mechanism for improving lipid and DHA accumulation in a microorganism. 6-benzylaminopurine has the potential to stimulate lipid and DHA production of Aurantiochytrium sp.YLH70 for industrial purposes.

INTRODUCTION

Docosahexaenoic acid (DHA), a type of ω‐3 polyunsaturated fatty acid (PUFA), has attracted considerable attention due to its structural and physiological functions.1 DHA is an essential compound in the cellular membrane of nervous tissues, such as brain and retina, and can improve development of these tissues. Moreover, DHA has beneficial effects in preventing cardiovascular diseases, cancer, schizophrenia, and Alzheimer's disease.2 Among many microorganisms, Aurantiochytrium sp., which is a marine fungus‐like heterotrophic Labyrinthulomycetes, is an effective fermentative microorganism for DHA production because it grows fast and accumulates high levels of DHA as a main fatty acid contained in microbial lipids.3 The microbial oil from Aurantiochytrium sp., containing high levels of DHA, was considered one of the most feasible DHA sources and could alleviate the problems caused by the traditional fish oil source.3 Cultivation of Aurantiochytrium sp. for DHA production was the most fundamental step because it affects cost and quality of DHA. Apart from essential macronutrients (carbon, nitrogen sources, and sea salt), some micronutrients, such as vitamins and trace elements, are required for growth of Aurantiochytrium sp. and are added to the medium to improve the biomass and DHA yield of Aurantiochytrium sp..4, 5 These micronutrients could reduce cost and increase quality of DHA significantly at low concentrations. However, the effects of phytohormone on biomass and DHA yields in Aurantiochytrium sp. and its mechanisms have never been reported.

Plant hormones (phytohormones) are natural or synthetic chemical messengers, which regulate growth and development of plants and are effective at very low concentrations. Plant hormones have been divided into many groups and they stimulate a variety of biological processes, such as cell growth and metabolism of fatty acids and lipids.6, 7 Apart from plants, some non‐plant microorganisms, including fungi and microalgae, have also been reported to be affected by plant hormones. Chatterjee et al. reported that 0.1 mg L−1 gibberellic acid could enhance mycelial growth and chitosan yield in Rhizopus oryzae.8 In plant–fungi symbiosis systems, some plant hormones, as signal molecules, may affect the symbiosis positively or adversely.9 Microalgae are primitive eukaryotic plant cells and have a close evolutional relationship with plants. Thus, there are various reports on effects of different plant hormones on algal growth and production of metabolites. Uses of plant substrates, such as auxins, cytokinins, abscisic acid (ABA), polyamines, brassinosteroids, jasmonides, and salicylic acid, were able to stimulate growth of Chlorella sp., whether for the production of lipids for biofuels, the production of bioproducts, the treatment of wastewater, or a variety of other reasons.10 Salama et al. found that phytohormones, such as indole‐3‐acetic acid (IAA) and diethyl aminoethyl hexanoate (DAH), could accelerate growth of Scenedesmus obliquus and induce the quality and quantity of fatty acid content for biodiesel production.11

As a plant hormone, 6‐Benzylaminopurine (6‐BAP) is a cytokinin that stimulates plant growth and development, setting blossoms and stimulating fruit maturity by accelerating cell division.12 It is also an inhibitor of respiratory kinase in plants, and increases post‐harvest life of green vegetables.13, 14 However, few studies on the effects of 6‐BAP on the growth of microorganisms have been reported. Thus, 6‐BAP should be considered a growth stimulator in fermentation media for improvements of biomass and DHA yield in Aurantiochytrium sp. Moreover, Labyrinthulomycetes are often prevalent in marine ecology and inhabit marine plants as commensals or mutualists.15 The effect of plant hormone on Aurantiochytrium sp. and its mechanism can deepen our understanding of the relationship between plants and Aurantiochytrium sp., and have considerable ecological significance.

Metabolomics has provided new insights into understanding intracellular metabolites and discovering potential biomarkers.16 This technology has been applied in microalgae to improve lipid productivity.17 Moreover, metabolomics was also used to reveal the global response of biochemical network reactions to environmental, genetic, or developmental signals.18 Gas chromatography–mass spectrometry (GC‐MS) was applied to investigate the metabolic profile of Aurantiochytrium sp. under different oxygen conditions.18 This metabolomics method was also used to analyze the metabolomics profile of the final fermentative cells of Schizochytrium sp.19

In this study, effects of 6‐BAP on growth, lipid and DHA yield in Aurantiochytrium sp. were investigated. Comparative metabolomics analysis of Aurantiochytrium sp. under 6‐BAP‐treated and control groups (without 6‐BAP) at different time points was performed through a GC‐MS‐based metabolomics method and multivariate analysis. This study could provide a better understanding of the mechanism underlying 6‐BAP enhancement of lipid and DHA biosynthesis in Aurantiochytrium sp.

MATERIALS AND METHODS

Chemicals

The nonadecanoic acid, docosahexaenoic acid methyl esters, 6‐benzylaminopurine, adonitol, methoxyamine hydrochloride, N‐methyl‐N‐trimethylsilyl‐trifluoroacetamide and trimethychlorosilane were purchased from Sigma‐Aldrich (St. Louis, USA). BF3‐methanol was purchased from ANPEL Laboratory Technologies Inc. (Shanghai, China). All other chemicals were obtained from Aladdin Industrial Inc. (Shanghai, China).

Strains, media and culture conditions

Aurantiochytrium sp. YLH70 (CCTCC No: M2014215), which was isolated from the mangrove ecosystem in Yueqing Bay and was collected in the China Centre for Type Culture Collection (CCTCC), was used in the present study. The strain was preserved in 20% glycerol at −80 °C and was maintained in GPY medium, containing 2% glucose, 1% polypeptone, 0.5% yeast extract and 20% sea salt. The strain was inoculated into a 500 mL flask containing 300 mL of GPY medium and cultured at 28 °C for 2 days as seed culture. The seed culture was then inoculated into 5 L fermentation reactors containing 3 L of fermentation medium, composed of 6% glucose, 1% yeast extract, 0.5% peptone and 20 g L−1 sea salt, and cultured at 28 °C for 5 days. For 6‐BAP‐treated cells, different concentrations of 6‐BAP (0, 1, 3, 5, 7, and 9 mg mL−1) were added to the fermentation medium before cultivation. Samples of the control and 6‐BAP‐treated groups were collected every 24 h until the total fermentation time reached 120 h, and the biomass, lipid content, fatty acid profile and metabolic profile were analyzed as described below.

Biomass determination

The biomass was measured in terms of the dried cell weight (DCW). An aliquot of 50 mL of cell suspension was centrifuged at 4 °C and 7000 g for 5 min and then washed twice with distilled water. The cell pellets were lyophilized to a constant weight at −50 °C for approximately 48 h.

Lipid extraction and fatty acids composition analysis

The lipid extraction and fatty acid analysis were performed according to the method reported by Gao et al. with moderate modifications.20 The lyophilized cell was ground by mortar and pestle into a fine powder under liquid nitrogen. The cell powder was extracted into 100 mL of chloroform/methanol (2:1, v/v) at room temperature. The lipid extract was dried by evaporation and then weighed. For fatty acid analysis, 0.1 g cell powder was mixed with 5 mL of 0.4 mol L−1 methanolic KOH and incubated at 60 °C for 1 h, followed by an esterification procedure in 5 mL of BF3‐methanol (14%, w/w) reagent at 60 °C for another 1 h. The fatty acid methyl esters (FAMEs) were extracted with 5 mL hexane and analyzed by Agilent 6890 N GC using an HP‐INNOWAX column (30 m × 0.25 mm, 0.25 µm film thickness, Agilent Technologies) with He as carrier gas. The temperature‐control procedure was as follows: from 100 to 240 °C at 15 °C per min and maintained at 240 °C for another 10 min; the temperature of the injection port and flame ionization port was 250 °C. Nonadecanoic acid methyl ester was used as an internal standard.

Sampling, quenching and extraction of intracellular metabolites

Cells were sampled every 24 h until 120 h during fermentation. The collected cell was quenched and extracted according to the methods reported by Li et al. and Hu et al. with moderate modifications.18, 21 The sample was quickly mixed with cold 60% methanol (−40 °C, v/v) to capture the metabolism of cells instantaneously and then centrifuged at 5000 g and −4 °C for 3 min. The cell pellet was washed twice with cold physiological saline (0.9% sodium chloride solution) and ground by mortar and pestle into a fine powder under liquid nitrogen. 1 mL of pre‐chill methanol (−40 °C) was added to the cell powder (100 mg) and mixed thoroughly for 30 s. The mixture was centrifuged at 5000 g and −4 °C for 5 min and the supernatant collected. An aliquot of 0.75 mL of pre‐chill methanol (−40 °C) was added to the cell pellet and shaken violently for 30 s followed by centrifugation at 8000 g and −4 °C for 5 min. Both supernatants were pooled together and preserved at −80 °C for further analysis. Six independent repeats were prepared for the control and 6‐BAP‐treated groups at each time point.

Derivatization and GC‐MS performance

For GC‐MS analysis, derivatization of samples was performed according to the method of Hu et al. with moderate modifications.21 The 0.5 mL sample obtained above and 50 μL internal standard (adonitol in water, 1 mg mL−1) were dried in a vacuum centrifuge dryer. 200 μL of methoxyamine hydrochloride in pyridine (20 mg mL−1) was added to the dried sample and incubated at 30 °C for 2 h. Subsequently, 200 μL of MSTFA (N‐methyl‐N‐trimethylsilyl‐trifluoroacetamide) containing trimethychlorosilane (1%, v/v) was added to the sample, and incubated at 37 °C for another 1 h. The resulting samples were preserved at −40 °C and adjusted to room temperature before injection.

The GC‐MS analysis was performed on an Aglilent 7890‐5975C GC‐MS solution system (Agilent, USA) equipped with a DB‐5 capillary column (30 m × 0.25 mm, 0.25 µm film thickness; Agilent J&W Scientific, Folsom, CA). Under a constant flow rate of 1 mL min−1, helium was used as the carrier gas. 1 µL of samples were injected into the DB‐5 capillary column at a split radio of 1:3. The GC oven temperature was kept at 70 °C for 2 min, raised to 290 °C at a rate of 8 °C min−1, and then kept for 3 min. The temperatures of injection, transfer line and ion source were 300 °C, 280 °C and 250 °C, respectively. A mass scan range (50–600 m/z) at a rate of 20 scans s−1 was used. The real time analysis software in the GC‐MS system (Agilent, USA) was used to obtain mass spectrometric data.

Data analysis

All detected metabolites were identified by alignments of mass spectra in NIST 05 (National Institute of Standards and Technology, Gaithersburg, MD) installed in the GC‐MS system. The data set was normalized using division of the intensity of the peaks in each sample by the intensity of the internal standard peak, and then by the quality of each sample on the same chromatograph. The generated normalized metabolite contents (variables) were imported into the SIMCA package (Version 10.0, Umetrics, Umea, Sweden) for multivariate analysis. Principal component analysis (PCA), a nonsupervised method was initially performed to obtain an overview of the GC‐MS data from the control groups from the 6‐BAP‐treated groups. This model can also discern the control groups from the 6‐BAP‐treated groups. Differences between the samples were detected in the PCA score plots, in which each point represents a linear combination of all the metabolites from every sample. Distance between the groups of samples shows a measure of the overall difference among the metabolic profiles under different conditions. A supervised PLS‐DA was subsequently performed to identify the metabolites contributing to the differences between the control and 6‐BAP‐treated groups. A coefficient loading plot of the PLS‐DA models was used to determine the differential metabolites that could separate the sample groups. Based on the loading plots and variable importance in the projection (VIP) value threshold (VIP > 1), metabolites that were contributing to differing the control and the 6‐BAP‐treated groups were identified through the PLS‐DA model.22 The altered metabolites, which were elicited by 6‐BAP and selected from the PLS‐DA model, were further subjected to hierarchical cluster analysis (HCA). HCA was performed by Euclidian distance and complete linkage grouping with the help of Gene Cluster software, and visualized through TreeView software. The raw data set was filtered using the default set with all metabolites passed. The filtered data was adjusted one cycle of log transform. Then, two cycles of median center genes and arrays were performed, which repeated five times. Finally, the data sets were applied to three cycles of normalize genes and arrays, which repeated five times. One‐way analysis of variance (ANOVA) t tests were performed using SPSS 12.0 software to analyze the distribution of the concentrations and logarithmically transformed concentrations for metabolites in this study.

RESULTS

Effects of 6‐BAP treatment on the fermentative parameters in Aurantiochytrium sp. YLH70

To investigate the effects of 6‐BAP on biomass, lipid and DHA yields for Aurantiochytrium sp. YLH70, different concentrations of 6‐BAP (0, 1, 3, 5, 7 and 9 mg mL−1) were added to the medium before fermentation and the 6‐BAP‐free group (0 mg mL−1) was used as the control group. As shown in Fig. 1, the optimal biomass, lipid and DHA production (20 g L−1, 52.8% biomass and 4.6 g L−1)1 were obtained when 3 mg mL−1 6‐BAP was used. There was almost no difference in biomass for Aurantiochytrium sp. YLH70 with addition of 6‐BAP, however, compared with the control group, addition of 6‐BAP caused up to 25.9% and 52.8% increases in lipid and DHA production in Aurantiochytrium sp. YLH70. Combining the results from the biomass, lipid and DHA productions, we concluded that 3 mg mL−1 6‐BAP added to medium was optimal. For this dose, we examined the time courses of lipid and DHA yields by Aurantiochytrium sp. YLH70 (Fig. 2). During 120 h fermentation, the lipid and DHA yields were significantly enhanced by 6‐BAP treatment.

Figure 1

Effects of different concentrations of 6‐benzylaminopurine on biomass, lipid content and DHA yield in Aurantiochytrium sp. YLH70. All data are means of three replicates; vertical bars represent error bars with the value equal to the standard error of the mean.

Figure 2

Time course of lipid content (A) and DHA yield (B) in Aurantiochytrium sp. YLH70 with and without 6‐benzylaminopurine addition. All data are means of three replicates; vertical bars represent error bars with the value equal to the standard error of the mean.

Comparative metabolite profile analysis of fermentation process with and without 6‐BAP treatment

The metabolic profiles of Aurantiochytrium sp. YLH70 from all samples were analyzed by a PCA scores plot (Fig. 3). After GC–MS analysis, mass spectral deconvolution and removal of the internal standard and pseudo‐positive peaks, such as those caused by noise, the number of peaks were 104 and 156 in the control and 6‐BAP‐treatment groups, respectively. From these peaks, a total of 71 intracellular metabolites were deduced. The PCA analysis showed that all samples from the control and 6‐BAP treatment groups were well separated, suggesting different metabolic profiles among these samples. In all cases, 76.9% of the total variance in data was described as R = 0.769 by the 1–3 principle components. This indicates that 76.9% of variation in the data was related to the 6‐BAP treatment and 6‐BAP was likely to be responsible for the metabolic variation.

Figure 3

PCA‐derived scores plot for all groups with and without 6‐benzylaminopurine addition. (a) Control group at 24 h; (b) 6‐BAP treatment group at 24 h; (c) control group at 48 h; (d) 6‐BAP treatment group at 48 h; (e) control group at 72 h; (f) 6‐BAP treatment group at 72 h; (g) control group at 96 h; (h) 6‐BAP treatment group at 96 h; (i) control group at 120 h; (j) 6‐BAP treatment group at 120 h.

Through PLS‐DA pairwise comparison, a significant separation of metabolic profiles between the control and 6‐BAP treatment groups at each time point was observed (Fig. 4(A), (B), (C), (D) and (F)). As shown in Fig. S1 (Supporting material), VIP plots showed that some of the identified metabolites were responsible for the group separation. The metabolites with a VIP value more than 1 were considered the most effective metabolites which could explain the difference.22 Thus, based on the VIP value (>1) and P‐value (<0.05), a total of 46 intracellular metabolites were selected and 38 of them were identified with the aid of the NIST 05 database. These metabolites were mainly composed of fatty acids, amino acids, organic acids, carbohydrates, alcohols, squalene, cholesterol, and so on, most of which were related with the fatty acids metabolism and growth in Aurantiochytrium sp. YLH70.

Figure 4

PLS‐DA derived plots for the control and 6‐BAP treatment groups at each time point. (A) Control versus 6‐BAP‐treated group at 24 h; (B) control versus 6‐BAP‐treated group at 48 h; (C) control versus 6‐BAP‐treated group at 72 h; (D) control versus 6‐BAP‐treated group at 96 h; (E) control versus 6‐BAP‐treated group at 120 h.

HCA was performed to confirm the results from the PCA model and PLS‐DA model analyses, and to reflect clustering patterns among metabolites. As shown in Fig. 5, a clear separation between the first two‐days and the last three‐day groups was observed. Furthermore, the control and 6‐BAP‐treated groups were well subdivided. These results verified the reliability of the PCA and PLS‐DA model and indicated that the effect of 6‐BAP on Aurantiochytrium sp.YLH70 was also related to time.

Figure 5

Hierarchical cluster analysis (HCA) for the 46 identified metabolites. (a) Control group at 24 h; (b) 6‐BAP treatment group at 24 h; (c) control group at 48 h; (d) 6‐BAP treatment group at 48 h; (e) control group at 72 h; (f) 6‐BAP treatment group at 72 h; (g) control group at 96 h; (h) 6‐BAP treatment group at 96 h; (i) control group at 120 h; (j) 6‐BAP treatment group at 120 h.

DISCUSSION

In this study, addition of phytohormone 6‐BAP had almost no effect on the growth of Aurantiochytrium sp. YLH70, however, it caused a significant increase in lipid and DHA accumulation. Since the lipid and DHA productivity was the most important characteristic for Aurantiochytrium sp. in the fermentation process, 6‐BAP could be a regulator which enhanced fermentation performance. However, the metabolic regulation mechanism was still unclear. A GC‐MS method combined with multivariate analyses was used to investigate changes of metabolites under exposure to 6‐BAP at the optimal concentration (3 mg mL−1). These metabolites responsible for responding to 6‐BAP belonged to different categories of components, including fatty acids, amino acids, organic acids, carbohydrates, alcohols, squalene, cholesterol, and so on. These metabolites could explain changes of metabolic fluxes following 6‐BAP treatment.

Addition of 6‐BAP increased yields of lipid and DHA (Figs 1 and 2). Meanwhile, other fatty acids, such as tetradecanoic acid (C14:0), heptadecanoic acid (C17:0), octadecanoic acid (C18:0), oleic Acid (C18:1) and arachidonic acid (C20:4), were also induced by 6‐BAP treatment. These results suggested that the metabolic flux to fatty acids was promoted by 6‐BAP treatment for lipid accumulation (Fig. 6). Research in Jatropha curcas showed that 6‐BAP could modify the fatty acid profile and induce linoleic and linolenic acids.22 Moreover, other phytohormones, such as indole‐3‐acetic acid, gibberellic acid, kinetin, 1‐triacontanol, and abscisic acid were often used as regulator for improvement of biodiesel production by microalgae.10, 11, 23 Thus, this study is the first report that 6‐BAP promotes the fatty acids and lipid accumulation in microorganism. Fatty acids were the main components of the cell membrane and affect the fluidity and permeability of cell membrane.24 Some fatty acids, especially unsaturated fatty acids, were induced by phytohormones to increase the fluidity and permeability of cell membrane as a response to phytohormone treatment of the microorganism. As a result, uptake of the substrate in medium into the cell might be accelerated and metabolism might be promoted. Zhu et al. found that odd‐numbered fatty acids C15:0 and C17:0 were increased for responses to environmental factors, such as temperature and salinity in Schizochytrium limacinum OUC88.25 These results indicated that changes in fatty acid composition induced by 6‐BAP might be correlated with the anti‐stress mechanism in Aurantiochytrium sp. YLH70.

Figure 6

Metabolic changes induced by 6‐benzylaminopurine in Aurantiochytrium sp. YLH70.

At each time point, the content of glucose was decreased by 6‐BAP, indicating that the glycolysis was induced by 6‐BAP and the rate of utilization of glucose was accelerated for production of other metabolites (Table 1 and Fig. 6). As shown in Fig. 6, the glycerol pathway was a metabolic bypass in glyceraldehyde‐3‐phosphate. The decrease of the content of glycerol under 6‐BAP treatment indicated that the glycerol pathway was suppressed and more carbon flux was directed to pyruvate. Pyruvate was the precursor for many metabolites, such as valine and ethanol, which were decreased in response to 6‐BAP, and more metabolic flux was directed to acetyl‐CoA (Table 1 and Fig. 6).

Table 1

The metabolites responsible for response to 6‐benzylaminopurine in Aurantiochytrium sp. YLH70 at five different time points

Metabolites	24 h		48 h		72 h		96 h		120 h
	Control	6‐BAP	Control	6‐BAP	Control	6‐BAP	Control	6‐BAP	Control	6‐BAP
Unknown	0.66 ± 0.02	0.71 ± 0.16	0.65 ± 0.03	1.30 ± 0.22**	1.38 ± 0.08	2.42 ± 0.13**	2.10 ± 0.11	3.21 ± 0.35**	3.54 ± 0.85	5.11 ± 0.70
Tetradecanoic acid	0.45 ± 0.06	0.63 ± 0.24	1.86 ± 0.06	1.33 ± 0.40**	2.60 ± 0.35	2.97 ± 0.38	3.42 ± 0.22	3.94 ± 0.71	4.03 ± 0.08	5.40 ± 0.43**
Acetic acid	2.53 ± 0.50	2.86 ± 0.31	3.55 ± 0.31	4.51 ± 0.46**	5.11 ± 0.15	7.45 ± 0.41**	5.85 ± 0.12	5.94 ± 0.31	6.60 ± 0.20	4.82 ± 0.60**
Heptadecanoic acid	0.14 ± 0.02	0.22 ± 0.07*	0.26 ± 0.03	0.37 ± 0.07**	0.34 ± 0.04	0.45 ± 0.04**	0.35 ± 0.02	0.60 ± 0.06**	0.35 ± 0.03	0.49 ± 0.06**
Myo‐Inositol	3.40 ± 0.31	5.11 ± 0.79**	4.85 ± 0.52	6.22 ± 0.78**	7.02 ± 0.65	8.51 ± 0.79**	7.80 ± 1.03	12.54 ± 0.98**	11.04 ± 1.08	16.23 ± 0.92**
Ethanol	6.10 ± 0.43	5.60 ± 0.51	6.09 ± 0.47	7.36 ± 0.35**	5.74 ± 0.52	5.76 ± 0.61	6.43 ± 0.70	7.24 ± 0.66	6.11 ± 0.39	4.93 ± 0.55**
Octadecanoic acid	2.60 ± 0.26	2.61 ± 0.46	2.98 ± 0.73	4.26 ± 0.31**	7.44 ± 0.84	9.28 ± 0.90**	9.87 ± 1.56	11.52 ± 0.84*	11.30 ± 1.84	9.78 ± 0.58
Unknown	1.72 ± 0.17	1.54 ± 0.29	1.68 ± 0.10	2.24 ± 0.32**	1.77 ± 0.22	3.01 ± 0.38**	1.69 ± 0.17	2.65 ± 0.21**	1.76 ± 0.11	2.30 ± 0.38**
Squalene	3.04 ± 0.44	0.00 ± 0.00**	6.52 ± 0.77	1.42 ± 0.43**	7.68 ± 2.22	4.26 ± 0.58**	8.31 ± 2.60	4.44 ± 0.14**	9.96 ± 1.91	2.98 ± 0.54**
Doconexent acid	8.73 ± 0.60	9.99 ± 0.81*	11.51 ± 1.56	12.58 ± 0.83	14.41 ± 1.31	17.05 ± 1.20**	17.23 ± 0.99	21.73 ± 2.06**	20.37 ± 1.30	35.36 ± 2.32**
Cholesterol	0.92 ± 0.28	0.00 ± 0.00**	2.08 ± 0.40	0.29 ± 0.14**	3.16 ± 0.37	0.75 ± 0.17**	4.71 ± 0.46	1.25 ± 0.24**	5.38 ± 0.25	1.49 ± 0.25**
Unknown	0.44 ± 0.14	0.60 ± 0.23	0.40 ± 0.13	0.47 ± 0.19	0.32 ± 0.12	0.98 ± 0.16**	0.30 ± 0.11	0.85 ± 0.12**	0.36 ± 0.11	0.33 ± 0.07
Pyruvamide	1.68 ± 0.20	1.74 ± 0.30	2.31 ± 0.17	2.61 ± 0.51	2.84 ± 0.18	3.65 ± 0.34**	3.54 ± 0.25	4.23 ± 0.49**	3.35 ± 0.11	4.55 ± 0.26*
Unknown	0.39 ± 0.11	1.12 ± 0.62*	0.36 ± 0.09	1.25 ± 0.57**	0.32 ± 0.06	1.11 ± 0.50**	0.36 ± 0.12	1.01 ± 0.48*	0.25 ± 0.04	1.38 ± 0.29**
Unknown	0.22 ± 0.05	0.28 ± 0.06	0.40 ± 0.06	0.74 ± 0.14**	0.48 ± 0.05	1.15 ± 0.24**	0.64 ± 0.03	0.87 ± 0.32**	0.74 ± 0.03	0.68 ± 0.15**
Unkown	0.04 ± 0.03	0.00 ± 0.00**	0.17 ± 0.04	0.00 ± 0.00**	0.26 ± 0.03	0.38 ± 0.18	0.36 ± 0.03	0.65 ± 0.30*	0.34 ± 0.04	0.84 ± 0.11
Oleic Acid	0.72 ± 0.20	1.50 ± 0.23**	1.08 ± 0.20	2.02 ± 0.30**	1.82 ± 0.13	2.66 ± 0.27**	1.97 ± 0.12	4.15 ± 0.66**	2.43 ± 0.11	5.85 ± 0.83**
Unknown	1.29 ± 0.12	2.08 ± 0.29**	1.27 ± 0.18	1.52 ± 0.29	0.80 ± 0.06	0.54 ± 0.27*	1.20 ± 0.21	0.00 ± 0.00**	0.00 ± 0.00	0.00 ± 0.00**
Unknown	0.26 ± 0.11	0.21 ± 0.11	0.42 ± 0.06	0.62 ± 0.10**	0.61 ± 0.07	0.86 ± 0.12**	0.69 ± 0.06	1.17 ± 0.22**	0.81 ± 0.08	0.87 ± 0.11**
Arachidonic acid	0.09 ± 0.01	0.16 ± 0.05**	0.12 ± 0.02	0.31 ± 0.07**	0.17 ± 0.01	0.26 ± 0.04**	0.11 ± 0.01	0.05 ± 0.02	0.08 ± 0.01	0.00 ± 0.00
Pentanoic acid	1.13 ± 0.05	0.13 ± 0.05**	1.21 ± 0.02	0.00 ± 0.00**	0.32 ± 0.03	0.00 ± 0.00**	0.39 ± 0.04	0.00 ± 0.00**	0.44 ± 0.03	0.00 ± 0.00**
Butanoic acid	0.07 ± 0.01	0.00 ± 0.00**	0.11 ± 0.01	0.00 ± 0.00**	0.17 ± 0.03	0.00 ± 0.00**	0.09 ± 0.01	0.00 ± 0.00**	0.05 ± 0.01	0.00 ± 0.00**
Carbonochloridic acid	2.23 ± 0.19	2.63 ± 0.21**	1.80 ± 0.11	1.68 ± 0.33	1.56 ± 0.04	0.53 ± 0.37**	1.09 ± 0.06	0.63 ± 0.31**	0.74 ± 0.11	0.00 ± 0.00**
Butane	1.25 ± 0.14	3.00 ± 0.73**	1.69 ± 0.20	1.91 ± 0.64	2.48 ± 0.20	1.24 ± 0.50**	2.91 ± 0.26	1.21 ± 0.52**	1.80 ± 0.16	0.19 ± 0.07**
Pyridinecarboxylic acid	0.03 ± 0.00	0.54 ± 0.20**	0.01 ± 0.01	0.07 ± 0.03**	0.00 ± 0.00	0.04 ± 0.02**	0.00 ± 0.00	0.00 ± 0.00**	0.00 ± 0.00	0.00 ± 0.00*
Propanoic acid	1.10 ± 0.07	0.00 ± 0.00**	1.37 ± 0.12	0.00 ± 0.00**	0.88 ± 0.06	0.00 ± 0.00**	0.66 ± 0.04	0.00 ± 0.00**	0.34 ± 0.10	0.00 ± 0.00*
Valine	0.89 ± 0.06	0.00 ± 0.00**	1.33 ± 0.04	0.00 ± 0.00**	1.71 ± 0.07	0.00 ± 0.00**	1.90 ± 0.05	0.00 ± 0.00**	1.60 ± 0.06	0.00 ± 0.00
Acrylanilide	0.09 ± 0.01	0.11 ± 0.03	0.11 ± 0.01	0.16 ± 0.03**	0.13 ± 0.01	0.26 ± 0.03**	0.17 ± 0.01	0.25 ± 0.03**	0.13 ± 0.02	0.04 ± 0.03**
Phosphoric acid	2.11 ± 0.08	1.06 ± 0.51**	2.41 ± 0.10	0.92 ± 0.32**	1.91 ± 0.05	0.33 ± 0.13**	1.63 ± 0.02	0.05 ± 0.04**	1.06 ± 0.04	0.00 ± 0.00**
Proline	4.10 ± 0.05	3.26 ± 0.49**	4.36 ± 0.06	1.72 ± 0.35**	4.62 ± 0.06	1.19 ± 0.27**	4.91 ± 0.07	0.64 ± 0.26**	5.30 ± 0.19	0.10 ± 0.06**
Arabitol	2.12 ± 0.06	2.49 ± 0.31*	3.09 ± 0.12	3.43 ± 0.52	3.78 ± 0.13	3.56 ± 0.37	2.14 ± 0.21	2.79 ± 0.11	1.33 ± 0.15	1.08 ± 0.39
Malic acid	0.56 ± 0.12	0.86 ± 0.09**	0.20 ± 0.17	0.53 ± 0.16**	0.00 ± 0.00	0.05 ± 0.02**	0.00 ± 0.00	0.00 ± 0.00**	0.00 ± 0.00	0.00 ± 0.00*
Fumaric acid	4.19 ± 0.14	4.16 ± 0.22	5.37 ± 0.22	2.52 ± 0.29**	6.22 ± 0.08	1.84 ± 0.45**	4.14 ± 0.05	1.10 ± 0.46**	2.36 ± 0.16	0.48 ± 0.29**
Lysine	2.18 ± 0.13	2.20 ± 0.25	3.66 ± 0.21	1.60 ± 0.50**	5.41 ± 0.23	0.99 ± 0.27**	6.25 ± 0.18	0.58 ± 0.32**	2.34 ± 0.23	0.02 ± 0.01**
Alanine	1.47 ± 0.24	1.24 ± 0.15	2.70 ± 0.14	0.70 ± 0.31**	4.29 ± 0.20	0.05 ± 0.02**	4.90 ± 0.05	0.00 ± 0.00**	5.56 ± 0.20	0.00 ± 0.00**
Glutamine	0.73 ± 0.23	0.78 ± 0.29	1.32 ± 0.06	1.81 ± 0.34**	1.90 ± 0.07	2.87 ± 0.32**	2.33 ± 0.12	3.48 ± 0.37**	2.76 ± 0.15	4.44 ± 0.30**
Fructose	0.71 ± 0.26	0.46 ± 0.27	1.77 ± 0.14	1.35 ± 0.25**	2.12 ± 0.05	2.36 ± 0.36	0.28 ± 0.15	1.42 ± 0.27	0.00 ± 0.00	0.35 ± 0.15
Xylose	0.50 ± 0.17	0.47 ± 0.21	0.14 ± 0.06	1.69 ± 0.59**	0.00 ± 0.00	0.66 ± 0.30**	0.00 ± 0.00	0.13 ± 0.05**	0.00 ± 0.00	0.00 ± 0.00*
Hexanoic acid	2.28 ± 0.19	2.45 ± 0.42	1.52 ± 0.05	1.73 ± 0.16*	1.19 ± 0.15	0.67 ± 0.20**	0.46 ± 0.06	1.62 ± 0.27	0.11 ± 0.07	2.57 ± 0.45
Mannonic acid	0.35 ± 0.09	0.54 ± 0.25	0.85 ± 0.05	1.44 ± 0.25**	1.32 ± 0.05	2.51 ± 0.53**	1.70 ± 0.09	1.29 ± 0.08**	2.12 ± 0.06	1.51 ± 0.52
Mannose	0.42 ± 0.12	1.12 ± 0.22**	0.86 ± 0.06	1.57 ± 0.27**	1.10 ± 0.09	1.97 ± 0.30**	1.50 ± 0.09	2.79 ± 0.42**	1.73 ± 0.13	2.42 ± 0.26**
Galactose	0.80 ± 0.24	0.65 ± 0.19	1.39 ± 0.09	1.45 ± 0.37	1.80 ± 0.11	2.67 ± 0.15**	2.14 ± 0.04	1.86 ± 0.57**	2.27 ± 0.04	1.54 ± 0.27*
Glucopyranose	3.22 ± 0.17	3.18 ± 0.33	2.84 ± 0.14	1.94 ± 0.52**	2.25 ± 0.21	1.64 ± 0.29**	1.50 ± 0.08	1.07 ± 0.16*	0.92 ± 0.08	0.36 ± 0.14*
Mannitol	0.49 ± 0.05	0.90 ± 0.18**	0.68 ± 0.07	1.57 ± 0.21**	0.91 ± 0.06	1.93 ± 0.33**	1.39 ± 0.20	1.06 ± 0.50**	1.84 ± 0.07	2.69 ± 0.30*
Glycerol	2.25 ± 0.07	2.02 ± 0.34	2.96 ± 0.03	1.57 ± 0.30**	3.33 ± 0.09	0.72 ± 0.25**	2.12 ± 0.03	0.33 ± 0.14**	1.35 ± 0.20	0.00 ± 0.00**
Glucose	3.72 ± 0.16	3.43 ± 0.21*	3.16 ± 0.04	2.11 ± 0.17**	2.20 ± 0.05	0.96 ± 0.32**	1.78 ± 0.17	0.05 ± 0.02**	0.66 ± 0.13	0.00 ± 0.00**

The VIP scores of all metabolites are higher than 1. The data represents the contents of metabolites and are showed as the mean ± SD. All data are means of six replicates.

P < 0.05 compared with the control.

P < 0.01 compared with the control.

Acetyl‐CoA is a very important precursor for many metabolic pathways, including mevalonate biosynthesis, tricarboxylic acid cycle (TCA) and fatty acids synthesis (Fig. 6). Squalene and cholesterol, which are the active components in the mevalonate biosynthesis pathway of Aurantiochytrium sp. YLH70, were both decreased by 6‐BAP treatment. Squalene is a triterpene hydrocarbon and an important ingredient used in cosmetics, pharmaceutical and medical industries. Aurantiochytrium sp. has been used for production of squalene.26 A gene encoding squalene synthase, which condensed two molecules of farnesyl pyrophosphate into a squalene, was cloned in Schizochytrium limacinum, and several cis‐acting elements in the promoter were also predicted, suggesting that synthesis of squalene was regulated by some factors at transcriptional level.27 Thus, it will be necessary to study the regulation mechanism of 6‐BAP on squalene synthesis in future research. Cholesterol is one of the most important sterols in Aurantiochytrium sp. and was synthesized by further catalysis of squalene.28 However, we found that unsaponifiable matter, such as squalene and cholesterol, competed acetyl‐CoA with fatty acids synthesis pathway. Thus, the mevalonate pathway in Aurantiochytrium sp. could be inhibited to improve fatty acids synthesis by metabolic engineering. Besides, a portion of acetyl‐CoA flow to the tricarboxylic acid cycle (TCA) in Aurantiochytrium sp., producing metabolites associated with this pathway. A 6‐BAP treatment caused decreases in fumaric acid, which was an intermediate of the TCA cycle (Fig. 6). Moreover, some amino acids, such as proline, alanine and lysine, which could be converted from α‐ ketoglutaric acid, were all reduced by 6‐BAP treatment. These results indicated that the metabolic flux from glycolysis, TCA cycle and mevalonate pathway to the fatty acids biosynthesis was promoted by 6‐BAP addition.

As shown in Table 1, myo‐inositol was increased by 6‐BAP addition. Some polylols, such as myo‐inositol and quercitol, were also found to make a great contribution to responses to environmental factors, such as oxygen and osmotic pressure in Schizochytrium sp. strains.18, 29 Further investigation will be necessary to determine the relationship between 6‐BAP addition and myo‐inositol metabolism. Addition of 6‐BAP could cause a significant decrease in phosphoric acid (Table 1). Ren et al. found that a phosphate‐limitation condition could maintain higher activities of malic enzyme and glucose‐6‐phosphate dehydrogenase, which favored accumulations of lipid and DHA.30 In the present study, increases in lipid and DHA could be explained by a decrease in phosphate induced by 6‐BAP treatment in Aurantiochytrium sp., which resulted in changes in enzyme activities.

CONCLUSIONS

In the current research, a GC‐MS‐based metabolomics method was applied to investigate the metabolic response to 6‐BAP of Aurantiochytrium sp. For metabolites profile, group classification and pairwise discrimination between the control and the 6‐BAP groups were obviously observed. Some metabolite, such as fatty acids, amino acids, carbohydrates and polyols, were responsible for the response to 6‐BAP in Aurantiochytrium sp.YLH70. Our findings indicated that with addition of 6‐BAP, fatty acids biosynthesis in Aurantiochytrium sp.YLH70 was promoted and the metabolic flux to TCA, mevalonate pathway and glycerol metabolism were diminished. Moreover, the anti‐stress mechanism in Aurantiochytrium sp.YLH70 might be induced by 6‐BAP. This study showed that the metabolomics method based on GC‐MS analysis was an appropriate tool to investigate the response to environmental or fermentative regulators in the molecular mechanism of microorganism, and 6‐BAP had potential as a stimulator for improving lipid and DHA production in Aurantiochytrium sp.YLH70.

Fig. S1 Variable importance of the projection plots for the intracellular metabolites along component. (A) Control versus 6‐BAP‐treated group at 24 h using PLS‐DA model, (B) control versus 6‐BAP‐treated group at 48 h using PLS‐DA model, (C) control versus 6‐BAP‐treated group at 72 h using PLS‐DA model, (D) control versus 6‐BAP‐treated group at 96 h using PLS‐DA model, (E) control versus 6‐BAP‐treated group at 120 h using PLS‐DA model.

Click here for additional data file.