α/β hydrolase domain-containing protein 1 acts as a lysolipid lipase and is involved in lipid droplet formation

Abstract

Lipid droplets (LDs) are the major sites of lipid and energy homeostasis. However, few LD biogenesis proteins have been identified. Using model microalga Chlamydomonas, we show that ABHD1, an α/β-hydrolase domain-containing protein, is localized to the LD surface and stimulates LD formation through two actions: one enzymatic and one structural. The knockout mutants contained similar amounts of triacylglycerols (TAG) but their LDs showed a higher content of lyso-derivatives of betaine lipid diacylglyceryl-N,N,N-trimethylhomoserine (DGTS). Over-expression of ABHD1 increased LD abundance and boosted TAG content. Purified recombinant ABHD1 hydrolyzed lyso-DGTS, producing a free fatty acid and a glyceryltrimethylhomoserine. In vitro droplet-embedded vesicles showed that ABHD1 promoted LD emergence. Taken together, these results identify ABHD1 as a new player in LD formation by its lipase activity on lyso-DGTS and by its distinct biophysical property. This study further suggests that lipases targeted to LDs and able to act on their polar lipid coat may be interesting tools to promote LD assembly in eukaryotic cells.

Keywords: algae; betaine lipid; lipid droplet; lysolipid; α/β hydrolase.

Figure 1.

ABHD1 protein is located on the surface of LDs and its quantity increases with prolonged N starvation. (A) ABHD1 protein level as detected by immunoblot analysis using anti-ABHD1 antibodies. Total protein extracts (from 0.4 million cells) were loaded onto each lane. Ponceau's stained proteins transferred onto nitrocellulose membrane were shown as the control for protein load (image height of whole lane was compressed). Abbreviations: α, antibody; d, day; N, nitrogen; TAP, tris-acetate-phosphate media. (B) Confocal imaging. Cells were cultured in TAP-N (1 d) and stained with BODIPY to observe LDs. Pseudo-colors were used: BODIPY-stained LDs in yellow, chlorophyll autofluorescence in blue, mCherry and mCherry-tagged ABHD1 in red. DIC: differential interference contrast. Bar = 10 µm.

Figure 2.

ABHD1 over-expression results in LD formation and a higher TAG content. (A) LD imaging in cells over-expressing ABHD1 in nitrogen-replete mixotrophic growth, i.e. TAP medium (three independent lines are shown, bar = 5 µm). (B) TAG amount in WT versus ABHD1 over-expressing lines during nitrogen replete mixotrophic growth. Four independent experiments, for six independent lines (three technical replicates for each line), are shown. Error bars show mean and standard deviation, Mann-Whitney test: ***P = 0.0002. (C) The distribution of LD numbers per cell during nitrogen replete mixotrophic growth (>200 cells examined per group). Histograms represent mean and standard deviation of three biological replicates for WT and five independent lines for OE (ABHD1 overexpressor). (D) TAG quantification in the H1246 yeast strain expressing the Chlamydomonas ABHD1 gene. Yeast transformants expressing the respective gene were harvested and analyzed for total TAG amount using gas chromatography–mass spectrometry (GC-MS). Data are means of three biological replicates with standard deviation bars shown. A significance of q = 0.13 for empty vector (EV) vs. +CrABHD1, q = 0.02 for EV vs. +ScDGA1 and q = 0.13 for +ScDGA1 vs. +CrABHD1 was determined using the Kruskal–Wallis test followed by a two-stage step-up Benjamini, Krieger and Yekutieli false discovery rate procedure for multiple comparison correction.

Figure 3.

Isolation and characterization of lipids for the two abhd1-1 and abhd1-2 mutants. (A) The site of the AphVIII cassette insertion in the ABHD1 gene for abhd1-1 and abhd1-2 mutants. (B) RT-PCR analysis of ABHD1 transcript. (C) Immunoblot analysis of ABHD1 protein on isolated LDs from the three strains. LDs were extracted from cells N-starved for 2 d. Proteins were extracted then loaded into each lane, based on a fixed number of fatty acid methyl esters (FAMEs) (i.e. ca. 55 μg FAME equivalent of LDs loaded). Silver-stained SDS-PAGE of LD-isolated proteins acts as a loading control. (D) Membrane coat composition of isolated LDs from nitrogen starved cultures. Levels of lyso-DGTS species detected in the polar lipid fraction of isolated LDs. Signal from ultra-high performance liquid chromatography with quadrupole time-of-flight mass spectrometry (LC-MS) was normalized by FAME quantities in the polar lipid fraction. (E) Changes in LD lyso-DGTS species in the mutants in comparison to WT during nitrogen starvation. For both (D) and (E): data are means of five biological replicates from three independent LD isolations. Bars indicate standard deviation. P values were determined, after passing Bartlett's test for homoscedasticity and Shapiro-Wilk's test for normality, by a two-way ANOVA with Bonferroni correction: *P < 0.05; **P < 0.01. Abbreviations: CBLP: Chlamydomonas beta subunit-like polypeptide; FAME, fatty acid methyl ester; MGDG, monogalactosyldiacylglycerol; DGDG, digalactosyldiacylglycerol; PG, phosphatidylglycerol; SQDG, sulfoquinovosyldiacylglycerol; DGTS, diacylglyceryl-N,N,N-trimethylhomoserine; PE, phosphatidylethanolamine; PI, phosphatidylinositol.

Figure 4.

Enzyme activity analysis of the refolded rABHD1 protein. (A) SDS-PAGE of rABHD1 purified from E. coli inclusions bodies and refolded. (B) Activity test of rABHD1 on abhd1-1 total lipid extracts identified lyso-DGTS as potential ABHD1 substrate. LC-MS detected over 10 000 m/z peaks for both the reactions with and without rABHD1. The full spectrum from LC-MS is provided in Fig. S10. (C) The lipase activity of rABHD1 toward purified lyso-DGTS. (D) The lipase activity of rABHD1 toward commercial lyso-PC. (E) Activity assay on lyso-PE, lyso-PG, MAG, PC and DGTS. But, no activity could be detected on PC or on DGTS for rABHD1. Data are the mean of three independent experiments and error bars refer to standard deviation. Student's t-test is applied after passing Bartlett's test for homoscedasticity and Shapiro-Wilk's test for normality: **P < 0.01, ***P < 0.001. Free fatty acid (FFA) formation was quantified using 19:0 fatty acid as an internal standard. Reaction conditions for all activity tests: Teorell Stenhagen universal buffer pH 7.5, with 100 mM NaCl and Triton-X100 at 1 CMC. During the enzymatic reaction (incubation of 2 h), during the preparation of lysolipid substrate or during lipid extraction, there is always some autolysis of the substrate, which is responsible for the presence of a certain amount of free fatty acids in the negative control.

Figure 5.

Addition of the rABHD1 protein to the GUV boosts curvature formation. (A) Schematic view of the experimental protocol. GUVs are put in contact with a thin emulsion of NBD-labeled triacylglycerol (TAG) that forms droplets at the lipid bilayer. When placed on a glass coverslip, membrane rupture results in a 2D lipid bilayer at the surface with droplets of TAG incorporated in. Right: typical image of a DEV after rupture is shown with Rho-PE in red and TAG NBD in green. Scale bar is 10 µm. (B) The droplet initially forms a spherical cap and is budding from the membrane 45 min after rABHD1 addition in the observation buffer (100 µg mL⁻¹). (C) Time lapse of a TAG droplet at the membrane plane (NBD TAG signal in upper panel and bright field in lower panel), showing the decrease of the surface after addition of rABHD1. Scale bars are 2 µm. (D) Control time lapse of a TAG droplet at the membrane plane without addition of protein. Scale bars are 2 µm. (E, F) Measurement of the droplet surface at the membrane plane according to time in two different lipid compositions: (E) DOPC/Lyso-PC/RhoPE and (F) DOPC/RhoPE. The measured area of each droplet is normalized by the initial surface occupied at t = −5 min. Dashed lines represent the time traces of individual droplets and thick lines are the average trend. Light yellow lines correspond to the control in the absence of ABHD1 [(E) N = 3 droplets analyzed, (F) N = 4], and dark red lines to the addition of rABHD1 at 100 µg mL⁻¹ [(E) N = 16, (F) N = 13].

Figure 6.

A model explaining the multifaceted actions of lyso-DGTS and ABHD1 protein in promoting LD budding, growth and TAG synthesis. ABHD1 catalyzes the hydrolysis of lyso-DGTS to produce a free fatty acid and a GTS moiety. ABHD1 is present at basal levels during optimal growth, and it increases in transcription and protein amount as N starvation is initiated, paralleling the increases in both LD number and size [67]. During LD budding, the formation of lyso-DGTS and the synthesis of ABHD1 protein both promote positive membrane curvature and therefore LD budding. Later on, ABHD1 activity may facilitate LD growth as less curvature is needed by degrading lyso-DGTS. The resulting fatty acid (FA) can be activated to acyl-CoA by LCS2 and contribute to TAG synthesis. Additionally, by consuming lyso-DGTS, ABHD1 is pulling the flow of acyls from either a putative PDAT or a PLA2, whose activity drives either TAG or FA production, respectively. Indeed, in two knockout mutants, LCS2 protein is present in a reduced amount, likely because the flow of free FA is reduced without ABHD1. Taken together, the major role of ABHD1 lies in LD coat remodeling and maintaining LD stability. Abbreviations: BTA1, betaine lipid synthase 1; DAG, diacylglycerol; DGAT, diacylglycerol acyltransferase; DGTS, diacylglyceryl-3-O-4'-(N,N,N-trimethyl)-homoserine; FAS, fatty acid synthase; G3P, glycerol-3-phosphate; GPAT, glycerol-3-phosphate acyltransferase; MLDP, major lipid droplet protein; LCS2, long chain acyl-CoA synthetase; Lyso-PA, lysophosphatidic acid; LPLAT, lysophospholipid acyltransferase; PA, phosphatidic acid; PDAT, phospholipid: diacylglycerol acyltransferase; PLA2, phospholipase A2; TAG, triacylglycerol.