Metabolome Analysis Reveals Betaine Lipids as Major Source for Triglyceride Formation, and the Accumulation of Sedoheptulose during Nitrogen-Starvation of Phaeodactylum tricornutum

Abstract

Oleaginous microalgae are considered as a promising resource for the production of biofuels. Especially diatoms arouse interest as biofuel producers since they are most productive in carbon fixation and very flexible to environmental changes in the nature. Naturally, triacylglycerol (TAG) accumulation in algae only occurs under stress conditions like nitrogen-limitation. We focused on Phaeodactylum strain Pt4 (UTEX 646), because of its ability to grow in medium with low salinity and therefore being suited when saline water is less available or for wastewater cultivation strategies. Our data show an increase in neutral lipids during nitrogen-depletion and predominantly 16:0 and 16:1(n-7) accumulated in the TAG fraction. The molecular species composition of TAG suggests a remodeling primarily from the betaine lipid diacylglyceroltrimethylhomoserine (DGTS), but a contribution of the chloroplast galactolipid monogalactosyldiacylglycerol (MGDG) cannot be excluded. Interestingly, the acyl-CoA pool is rich in 20:5(n-3) and 22:6(n-3) in all analyzed conditions, but these fatty acids are almost excluded from TAG. Other metabolites most obviously depleted under nitrogen-starvation were amino acids, lyso-phospholipids and tricarboxylic acid (TCA) cycle intermediates, whereas sulfur-containing metabolites as dimethylsulfoniopropionate, dimethylsulfoniobutyrate and methylsulfate as well as short acyl chain carnitines, propanoyl-carnitine and butanoyl-carnitine increased upon nitrogen-starvation. Moreover, the Calvin cycle may be de-regulated since sedoheptulose accumulated after nitrogen-depletion. Together the data provide now the basis for new strategies to improve lipid production and storage in Phaeodactylum strain Pt4.

Fig 1. Growth parameters and total fatty acid composition of Phaeodactylum tricornutum control cultures under normal light (blue) or with N-depletion under normal light (red) or high light (green).

In (A) the dry weight content [mg/ml] is depicted and in (B) the chlorophyll a amount in relation to the dry weight [mg/g]. Samples were taken at days 0, 2, 5 and 7 for N-replete growth and at days 0, 2, 3 and 6 for N-depleted growth. These samples were used for determination of growth parameters as well as for lipid and metabolite analysis. The total FA composition in μmol/g is displayed in (C) showing day 0 and the last time point of each condition. Day 0 comprises the mean of all conditions (beige). Day 7 of replete conditions is shown in blue, day 6 of N-deplete with normal light in red and day 6 of N-deplete with high light in green. Data are mean values of 3 biological replicates, for the comprised time point 0d, 4 biological replicates were used. Error bars indicate standard deviation.

Fig 2. DAG and TAG molecular species as well as acyl-CoA pool under replete and N-deplete conditions.

The amounts of the molecular species of diacylglycerol (DAG) und triacylglycerol (TAG) are displayed with a threshold of 0.1 μmol/g in (A). These data were produced by neutral loss scanning which allows the determination of the sum (number of carbon atoms and of double bonds) of two (for DAG) or three FAs (for TAG) but not always the elucidation of the distinct fatty acid composition of the molecular species. Therefore, the possible fatty acid composition of the different molecular species are given in (B). (C): The diagrams show the acyl-CoA pool as comparison of day 0 (beige) and the last day of the growth kinetic of growth under replete conditions (blue), N-depletion with normal light (red) and N-depletion with high light (green). Data are mean values of 3 biological replicates, for the comprised time point 0d, 9 biological replicates were used. Error bars indicate standard deviation.

Fig 3. Main molecular lipid species that may be relevant for TAG synthesis.

The diagrams show the comparison of day 0 (beige) and the last day of the growth kinetic of growth under replete conditions (blue), N-depletion with normal light (red) and N-depletion with high light (green). Data are mean values of 3 biological replicates, for the comprised time point 0d, 9 biological replicates were used. Error bars indicate standard deviation.

Fig 4. Metabolite fingerprinting analysis of Phaeodactylum tricornutum under replete and N-deplete conditions.

Phaeodactylum tricornutum cultures grown under replete condition and normal light (NL), or grown under N-depleted conditions (-N) and normal light (NL) or high light (HL) were harvested at the indicated time points and extracted by two phase partitioning. The fingerprint of metabolites of the polar extraction phase was generated by UPLC-TOF-MS analysis. A subset of 935 high-quality features (FDR < 10⁻⁵) derived from the positive as well as the negative ionization mode were used for clustering and visualization by means of one-dimensional self-organizing map (1D-SOM, http://marvis.gobics.de). Prototype 2 (blue frame) represents features with reduced relative amounts under N-depleted conditions, while features combined in prototype 4–6 (red frame) are enriched under N-depletion. Horizontal and vertical dimensions correspond to prototypes and experimental conditions, respectively. The heat map colors represent average intensity values according to the color map on the right-hand side. The width of each prototype column is proportional to the number of marker candidates assigned to this prototype. Bar plots show mean values with standard deviations of 3 biological replicates for prominent metabolite markers of the selected clusters. The first 4 bar plots (light to dark blue) show the relative amounts of the compounds for replete conditions (0, 3, 5, 7d), while the next bar plots show the data for N-deplete conditions (0, 3, 6d) under normal light (light to dark red) or high light (light to dark green). The identity of the indicated compounds was confirmed by high resolution MS² experiments. Visualization was applied using VANTED 2.1 software [39].

Fig 5. Profiling analysis of the central metabolites of Phaeodactylum tricornutum under replete and N-deplete conditions.

Phaeodactylum tricornutum cultures were analyzed by metabolite profiling. Bar plots show mean values with standard deviations of 3 biological replicates for prominent metabolite markers of the selected clusters. The first 4 bar plots (light to dark blue) show the relative amounts of the compounds for replete conditions (0, 3, 5, 7d), while the next bar plots show the data for N-deplete conditions (0, 3, 6d) under normal light (light to dark red) or high light (light to dark green). The identity of the indicated compounds was confirmed by high resolution MS² experiments. Pathway visualization was applied using VANTED 2.1 software [39].

Fig 6. Summarizing scheme of changes of lipidome and metabolome under N-depletion.

The scheme displays the main observations followed by N-depletion. Regarding lipid species (yellow) the main changes affect MGDG, DGTS and TAG, suggesting a major flux from DGTS towards TAG, but there are also evidence for a participation of MGDG and maybe SQDG. All lysolipid species are decreased. The Chla content is decreasing which suggests together with a decrease of different amino acids towards protein degradation. In addition, the compounds of the TCA cycle are reduced. However, different sugar compounds as well as S-containing metabolites and carnitines are increased.