Seipin and Nem1 establish discrete ER subdomains to initiate yeast lipid droplet biogenesis

Abstract

Lipid droplets (LDs) are fat storage organelles that originate from the endoplasmic reticulum (ER). Relatively little is known about how sites of LD formation are selected and which proteins/lipids are necessary for the process. Here, we show that LDs induced by the yeast triacylglycerol (TAG)-synthases Lro1 and Dga1 are formed at discrete ER subdomains defined by seipin (Fld1), and a regulator of diacylglycerol (DAG) production, Nem1. Fld1 and Nem1 colocalize to ER-LD contact sites. We find that Fld1 and Nem1 localize to ER subdomains independently of each other and of LDs, but both are required for the subdomains to recruit the TAG-synthases and additional LD biogenesis factors: Yft2, Pex30, Pet10, and Erg6. These subdomains become enriched in DAG. We conclude that Fld1 and Nem1 are both necessary to recruit proteins to ER subdomains where LD biogenesis occurs.

Graphical abstract

Figure 1.

The TAG-synthase Lro1 reveals sites of nascent LD biogenesis in the ER. (A) Topology of TAG-synthase Lro1. N, nucleus. (B) Design of the MBP-GFP-FFAT-sLro1 chimera, which binds the ER-resident Scs2 through a FFAT motif and has its catalytic domain exposed to the cytoplasm. (C and D) GFP-FFAT-sLro1 is enzymatically active. Cells of indicated genotypes were grown in SC raffinose media, diluted in SC galactose containing [³H]palmitic acid, and grown for the indicated period of time. Lipids were extracted and separated by TLC. TAG-to-phospholipids ratio is plotted in C, and TLC image of 21-h time point is shown in D. Data represent mean ± SD of three independent experiments. (E and F) GFP-FFAT-sLro1 localizes to discrete ER subdomains. 4ΔKO cells expressing genomic Erg6-mCherry and coexpressing GFP-FFAT-sLro1 (E) or GFP-FFAT (F) from a galactose-inducible promoter. Arrowheads indicate colocalization between GFP-FFAT-sLro1 and Erg6-mCherry. Scale bar, 5 µm. (G) Western blot showing expression of GFP-FFAT-sLro1 or Kar2 after galactose induction for the indicated period of time. (H) Quantification of colocalization between GFP-FFAT-sLro1 and Erg6-mCherry. Percentage of GFP-FFAT-sLro1 puncta that colocalize with Erg6-mCherry after galactose induction. Data represent mean ± SD, n > 50 cells; ***, P < 0.0005. (I) GFP-FFAT-sLro1 colocalizes with LDs. CLEM imaging of 4ΔKO cells expressing GFP-FFAT-sLro1. Cells were induced in galactose for 1 h. GFP-FFAT-sLro1 is enriched at a discrete site in the ER by fluorescence imaging and colocalizes with an electron-translucent LD by EM. Yellow arrow demarcates ER, white arrow denotes GFP-FFAT-sLro1 puncta, and black arrow denotes an LD. N, nucleus.

Figure S1.

Data associated with Fig. 1. TAG-synthase sLro1 needs to be ER associated for its activity. (A and B) sLro1-containing TMDs from Sec63 localize to ER subdomains. Design of the TAG-synthase chimera containing TMDs 1–3 from Sec63 (aa 1-250), facing the active site to the cytoplasm [Sec63(1–250)-GFP-sLro1] (A). 4ΔKO cells expressing Erg6-mCherry and coexpressing Sec63(1–250)-GFP-sLro1 from a galactose-inducible promoter were grown in raffinose media and switched to galactose media for the indicated period of time (B). White arrowheads indicate colocalization of Sec63(1–250)-GFP-sLro1 punctae with Erg6-mCherry. Scale bar, 5 µm. (C) GFP-FFAT-sLro1 localizes to LDs in WT cells. WT cells expressing Erg6-mCherry and coexpressing GFP-FFAT-sLro1 from a galactose-inducible promoter were grown as described in B. White arrowheads indicate colocalization of GFP-FFAT-sLro1 punctae with Erg6-mCherry. The pink arrowhead denotes a GFP-FFAT-sLro1 puncta not associated with Erg6-mCherry. Scale bars, 5 µm. (D–F) The TAG-synthase sLro1 needs to be anchored to the ER membrane for its activity. Design of the soluble fusion protein GFP-sLro1 localized in the cytosol (D), and ssKar2-GFP-sLro1-HDEL localized in the ER lumen (E). 4ΔKO cells expressing Erg6-mCherry and coexpressing either GFP-sLro1 or ssKar2-GFP-sLro1-HDEL from a galactose-inducible promoter. Cells were grown in raffinose media and switched to galactose-containing media for 3 h (F). Scale bars, 5 µm. (G) Nonmembrane-associated sLro1 fails to catalyze TAG synthesis. 4ΔKO cells expressing genomic Erg6-mCherry and coexpressing either native Lro1, GFP-FFAT-sLro1, GFP-sLro1, or ssKar2-GFP-sLro1-HDEL from a galactose-inducible promoter were grown in galactose media containing [³H]palmitic acid for 21 h. Lipids were extracted and separated by TLC. FFA, free fatty acid.

Figure S2.

Data associated with Fig. 2. Cells lacking Nem1, Spo7, Pah1, or Fld1 accumulate neutral lipids in the ER and have aberrant droplets. (A) Diagram depicting the regulation of triglyceride synthesis. The Nem1-Spo7 complex dephosphorylates and thereby activates the phosphatidate phosphatase Pah1, resulting in formation of DAG from phosphatidic acid (PA). DAG then serves as a substrate for the TAG-synthesizing enzymes Dga1 and Lro1. (B) BODIPY accumulates in the ER of nem1Δ, spo7Δ, and pah1Δ cells. Yeast cells of the indicated genotype expressing genomic Erg6-mCherry were stained with BODIPY. White arrowheads denote BODIPY accumulation in the ER membrane. Scale bar, 5 µm. (C) Fld1-deleted cells accumulate numerous small and a few supersized LDs. fld1Δ cells expressing genomic Erg6-mCherry were stained with BODIPY. White arrows denote colocalization between BODIPY and Erg6-mCherry. Yellow arrows denote BODIPY-stained supersized puntae that do not colocalize with Erg6-mCherry. Two examples are shown (I, II). Scale bar, 5 µm. (D) WT cells do not show BODIPY accumulation in the ER. BODIPY staining of WT cells expressing genomic Erg6-mCherry. White arrows denote colocalization between BODIPY and Erg6-mCherry. Scale bar, 5 µm. (E) Pah1-7P rescues the LD biogenesis defect of nem1Δ, spo7Δ, and pah1Δ mutant cells. BODIPY staining of nem1Δ, spo7Δ, and pah1Δ mutants expressing Erg6-mCherry and Pah1-7P. White arrows denote colocalization between BODIPY and Erg6-mCherry. Scale bar, 5 µm.

Figure S3.

Data associated with Figs. 2 and 3. Fld1 and Nem1 localize to punctae in cells devoid of neutral lipids or LDs. (A) Localization of Fld1 or Nem1 to ER foci is independent of neutral lipid synthesis or the presence of LDs. BODIPY staining of 4ΔKO cells expressing genomic Fld1-mCherry or Nem1-mCherry. Scale bar, 5 µm. (B and C) Time-lapse images of Fld1-GFP and Nem1-GFP. 4ΔKO cells expressing genomic Fld1-GFP (B) or Nem1-GFP (C) were grown to early stationary phase, and time-lapse images were recorded. The boxed region in B and C is shown in higher magnification. Snapshots of images at the indicated time points are shown. White arrowheads indicate Fld1 or Nem1 punctae that stay immobile in the ER membrane. Yellow arrowheads denote rapidly moving Fld1 punctae. Scale bars, 5 µm. (D) Fld1 colocalizes with Nem1. WT cells expressing Nem1-GFP and coexpressing Fld1-mCherry were imaged. White arrowheads denote colocalization between Fld1 and Nem1. Scale bar, 5 µm. (E) Quantification of colocalization between Nem1-GFP and Fld1-mCherry. Data are means ± SD, n > 50 cells.

Figure 2.

Fld1 and Nem1 define ER subdomains for the recruitment of the TAG-synthase. (A and B) 4ΔKO cells expressing Fld1-mCherry (A) or Nem1-mCherry (B) together with GFP-FFAT-sLro1 from a galactose-inducible promoter. Fld1-mCherry (A) and Nem1-mCherry (B) show punctate localization in the absence of LDs (0 h). Upon 1 h of induction, GFP-FFAT-sLro1 colocalizes with Fld1-mCherry (A) or Nem1-mCherry (B). White arrowheads indicate colocalization of sLro1 with Fld1 (A) and Nem1 (B), respectively. Two examples of the 1 h time points are shown (I, II) for A and B. Scale bars, 5 µm. (C and E) Line scan of signal intensity along the white line shown in A (I) and B (II), respectively. (D and F) Quantification of colocalization between Fld1 (D) and Nem1 (F) punctae with sLro1. Data are means ± SD, n > 50 cells; ***, P < 0.0005. (G) Fld1, Nem1, and Pah1 are required to recruit the TAG-synthase into discrete ER foci. Cells lacking Fld1, Nem1, or Pah1 expressing GFP-FFAT-sLro1 and coexpressing mCherry-HDEL to mark the ER. Scale bar, 5 µm. (H) Overexpression of Fld1 in fld1Δ cells creates ER sites that recruit GFP-FFAT-sLro1. Fluorescence microscopy of fld1Δ cells harboring Fld1-mCherry and coexpressing GFP-FFAT-sLro1. White arrowheads denote colocalization between Fld1 and sLro1. Scale bar, 5 µm. (I) Cartoon showing the recruitment of sLro1 at Fld1 and Nem1 sites.

Figure 3.

ER sites marked by Fld1 and Nem1 recruit native Lro1. (A and B) ER subdomains marked by Fld1 and Nem1 recruit native Lro1. 4ΔKO cells expressing Fld1-mCherry (A) or Nem1-mCherry (B) and coexpressing native GFP-Lro1 from a galactose-inducible promoter. Cells show punctate distribution of Fld1-mCherry and Nem1-mCherry (0 h). Upon induction, GFP-Lro1 colocalizes with Fld1-mCherry and Nem1-mCherry (1 h). White arrowheads indicate colocalization of Lro1 with Fld1 (A) and Nem1 (B), respectively. Two examples of the 1 h time points are shown (I, II) for A and B. Scale bars, 5 µm. (C and E) Line trace of signal intensity along the white line shown in A (I) and B (I), respectively. (D) Quantification of colocalization between Fld1 or Nem1 foci with Lro1. Data are means ± SD, n > 50 cells; **, P < 0.005. (F) Cells missing Fld1, Nem1, or Pah1 fail to recruit GFP-Lro1 into discrete foci. Cells expressing native GFP-Lro1 and coexpressing mCherry-HDEL to mark the ER were cultured in galactose-containing media for 1 h. Scale bars, 5 µm. (G and H) Colocalization between Fld1 and Nem1 during LD biogenesis. 4ΔKO cells expressing genomic Nem1-GFP and coexpressing Fld1-mCherry under a constitutive ADH1 promoter, together with Lro1. Colocalization between Nem1 and Fld1 denoted by white arrowheads (G) was quantified after galactose induction of Lro1 for 2 h (H). Data are means ± SD, n > 50 cells; ***, P < 0.0005. Scale bars, 5 µm. (I) Cartoon illustrating recruitment of native Lro1 at Fld1 and Nem1 sites.

Figure S4.

Data associated with Figs. 4, 5, 6, 7, and 8. Enrichment of DAG at Fld1 and Nem1 ER subdomains. (A) ER subdomains marked by Fld1 and Nem1 show enrichment of DAG. WT cells expressing Fld1-mCherry or Nem1-mCherry and coexpressing the GFP-tagged ER-DAG sensor were grown in SC media to early stationary phase and imaged live. White arrowheads indicate colocalization of the ER-DAG sensor with ER punctae marked with Fld1 or Nem1. Scale bars, 5 µm. (B) Fld1- and Nem1-marked ER subdomains show enrichment of DAG in LD-deficient cells. 4ΔKO cells expressing Fld1-mCherry or Nem1-mCherry and coexpressing the ER-DAG sensor were grown as described in A. White arrowheads indicate colocalization of the ER-DAG sensor with punctae marked by Fld1 or Nem1. Scale bar, 5 µm. (C) Lack of Nem1, Spo7, or Pah1 results in uniform distribution of the ER-DAG sensor. nem1Δ, spo7Δ, or pah1Δ mutant cells expressing the ER-DAG sensor were grown as described in A. ER was visualized by mCherry-HDEL. Scale bar, 5 µm. (D) BODIPY staining of 4ΔKO cells expressing MBP-Scarlet-FFAT-sLro1. Cells were grown in glucose media to mid-log phase. Scale bar, 5 µm. (E) Coomassie blue staining of an SDS-PAGE gel showing purified WT and mutant sLro1. (F) Rate of TAG synthesis. Time-dependent incorporation of [³H]palmitic acid into TAG in the indicated yeast mutant cells. Cells were labeled and collected at 10-, 30-, 60-, 90-, and 120-min time points. Data are means ± SD of three independent experiments. (G) An ER membrane protein that is not involved in LD biogenesis does not show enrichment at Fld1-marked ER subdomains. The ER protein Sec63 does not become enriched at Fld1 sites during induction of LD biogenesis. GAL-LRO1 3ΔKO cells expressing Fld1-mCherry and coexpressing Sec63-GFP were grown in raffinose media and switched to galactose-containing media for 1.5 h. Scale bar, 5 µm. (H) Lack of Pet10 does not affect Nem1 localization and recruitment of TAG-synthase. The recruitment of GFP-FFAT-sLro1 to Nem1 sites is not affected in pet10Δ mutant cells. pet10Δ mutant cells expressing Nem1-mCherry and coexpressing GFP-FFAT-sLro1 from a galactose-inducible promoter were shifted to galactose media for the indicated period of time. Arrowheads denote colocalization of sLro1 with Nem1 foci. Scale bar, 5 µm. (I) Neutral lipids accumulate in the ER membrane in seipin mutant cells. fld1Δ mutant cells (fld1Δ GAL-LRO1 3ΔKO) expressing Nem1-mCherry were grown in raffinose media and transferred to galactose-containing media for the indicated period of time, stained with BODIPY, and imaged. White arrowheads indicate BODIPY accumulation in the ER. Yellow arrows indicate BODIPY punctae that do not colocalize with Nem1 foci. Scale bar, 5 µm.

Figure 4.

Enzymatic activity of TAG-synthase sLro1 is required for its recruitment to DAG-enriched ER subdomains. (A) Induction of LD formation by OA results in elevated DAG levels at Fld1 and Nem1 sites. WT cells expressing Fld1-mCherry or Nem1-mCherry and coexpressing the GFP-tagged ER-DAG sensor were diluted and incubated in SC media containing 0.1% OA for 1 h. White arrowheads indicate colocalization of the ER-DAG sensor with Fld1 or Nem1 punctae. Scale bars, 5 µm. (B) Quantification of colocalization between ER-DAG sensor punctae with Fld1 and Nem1 1 h after OA addition. Data are means ± SD, n > 50 cells; ***, P < 0.0005. (C) Cytosolic sLro1 and native Lro1 colocalize with the DAG sensor. 4ΔKO cells expressing mCherry-tagged ER-DAG sensor and coexpressing GFP-FFAT-sLro1/GFP-Lro1 under galactose promoter. White arrowheads denote colocalization between the ER-DAG sensor and either sLro1 or Lro1. Scale bars, 5 µm. (D) Mutant sLro1 fails to produce TAG. 4ΔKO cells expressing WT sLro1 (Scarlet-FFAT-sLro1) or mutant sLro1 (Scarlet-FFAT-sLro1^H618A) were grown in galactose media containing [³H]palmitic acid for 21 h. Lipids were extracted and separated by TLC. FFA, free fatty acid. (E) A catalytically inactive mutant of sLro1 fails to produce LDs. 4ΔKO cells expressing WT or mutant version of sLro1 from a galactose-inducible promoter were stained with BODIPY. White arrowheads denote colocalization between BODIPY-marked LDs and WT sLro1 punctae. Scale bars, 5 µm. (F and G) WT and mutant version of sLro1 bind DAG. In vitro binding of the short chain DAG (C8:0) by purified WT and mutant sLro1 was measured by MST. conc, concentration; Fnorm, normalized fluorescence. (H) Enzymatic activity of sLro1 is required for its colocalization with Fld1. 4ΔKO cells expressing WT or mutant version of sLro1 together with Fld1-GFP were switched to galactose media for 2 h. White arrowheads denote colocalization between WT sLro1 and Fld1. Scale bars, 5 µm.

Figure 5.

Fld1 and Nem1 are both required for recruitment of the TAG-synthase. (A and B) Fld1 is required to recruit both native and cytosolic Lro1 to Nem1 sites. fld1Δ cells expressing Nem1-mCherry and coexpressing native GFP-Lro1 (A) or cytosolic GFP-FFAT-sLro1 (B) from a galactose-inducible promoter. Scale bars, 5 µm. (C and D) Nem1 is required to recruit Lro1 to Fld1 sites. nem1Δ cells expressing Fld1-mCherry and coexpressing native GFP-Lro1 (C) or cytosolic GFP-FFAT-sLro1 (D) from a galactose-inducible promoter. Scale bars, 5 µm. (E and F) Fld1 and Nem1 together constitute functional ER sites for LD biogenesis. Cartoon depicting ER subdomains containing both Fld1 and Nem1 as a prerequisite for the recruitment of the native GFP-Lro1 or the cytosolic GFP-FFAT-sLro1. (G) Expression of Pah1-7P in nem1Δ cells rescues the defect in recruitment of the TAG-synthase. nem1Δ cells expressing Fld1-mCherry, and coexpressing both Pah1-7P, and GFP-FFAT-sLro1. White arrowheads denote colocalization between GFP-FFAT-sLro1 and Fld1-mCherry. Scale bar, 5 µm.

Figure 6.

ER subdomains marked by Fld1 and Nem1 recruit Yft2. (A and B) Yft2 concentrates at Fld1- and Nem1-containing ER sites. 3ΔKO cells expressing Fld1-mCherry (A) or Nem1-mCherry (B) and coexpressing Yft2-sf-GFP were transferred to galactose media. White arrowheads indicate colocalization of Yft2 with Fld1 (A) or Nem1 (B). (C) Quantification of colocalization of Yft2 with Fld1 and Nem1. Data are means ± SD, n > 50 cells. (D) Cartoon depicting ER subdomain containing Fld1, Nem1, and Yft2. (E) Localization of Yft2 at Nem1 sites is independent of Fld1. fld1Δ 3ΔKO cells expressing Nem1-mCherry and coexpressing Yft2-sf-GFP were grown as described in A. White arrowheads indicate colocalization of Yft2 with Nem1. (F) Cartoon showing that the colocalization of Yft2 with Nem1 is independent of Fld1. (G and H) Yft2 requires Nem1 at sites of LD biogenesis. NEM1 (G) and nem1Δ (H) cells expressing Fld1-mCherry and coexpressing Yft2-sf-GFP were grown in SC media, resuspended into fresh media containing 0.1% OA, and visualized after 1.5 h. White arrowheads denote colocalization of Yft2 with Fld1 foci in G. Scale bar, 5 µm. (I) Cartoon depicting colocalization between Yft2 and Fld1 is dependent on Nem1. (J) DAG levels affect the localization of Yft2 and Fld1. nem1Δ cells expressing Fld1-mCherry, Yft2-sf-GFP, and coexpressing Pah1-7P were grown to early stationary phase in SC media. White arrowheads denote colocalization of Yft2 with Fld1 foci. Scale bar, 5 µm.

Figure 7.

Epistatic relation of proteins at LD biogenesis sites. (A) Lack of FIT proteins does not abrogate recruitment of GFP-FFAT-sLro1 at Fld1 sites. yft2Δ scs3Δ double mutant cells expressing Fld1-mCherry and coexpressing GFP-FFAT-sLro1 were switched to galactose media for 1 h. White arrowheads indicate colocalization of Fld1 with sLro1. (B) Cells lacking FIT proteins have elevated DAG levels at Fld1 sites. WT and yft2Δ scs3Δ double mutant cells expressing Fld1-mCherry and coexpressing the GFP-tagged ER-DAG sensor were grown to early stationary phase. Arrowheads indicate colocalization of Fld1 foci with the DAG-sensor punctae. (C) Cartoon showing accumulation of DAG at Fld1- and Nem1-marked ER subdomains when these sites are missing Yft2. (D and E) Pex30 acts downstream of Fld1 and Nem1 in recruiting GFP-FFAT-sLro1 to sites of LD biogenesis. pex30Δ mutant cells expressing Fld1-mCherry (D) or Nem1-mCherry (E) and coexpressing GFP-FFAT-sLro1 were switched to galactose-containing media for the indicated time. (F) Yft2 functions independently of Pex30 at sites of LD biogenesis. pex30Δ mutant cells expressing Fld1-mCherry or Nem1-mCherry and coexpressing Yft2-sf-GFP were diluted into fresh SC media containing 0.1% OA and imaged after 1 h. White arrowheads indicate colocalization of Yft2 with Fld1 or Nem1 punctae. (G and H) Cartoon illustrating the localization of Fld1, Nem1, Yft2, and Pex30 at ER subdomains to constitute functional LD biogenesis sites (G). Absence of Pex30 from these sites does not affect the localization of Fld1, Nem1, and Yft2, but these sites fail to recruit the sLro1 (H). (I) Lack of Pet10 does not impair the recruitment of GFP-FFAT-sLro1 at Fld1 sites. pet10Δ mutant cells expressing Fld1-mCherry and coexpressing GFP-FFAT-sLro1 were grown as described in D. Arrowheads denote colocalization of sLro1 with Fld1 foci. Scale bar, 5 µm. (J) Cartoon illustrating localization of sLro1 at Fld1 sites in the absence of Pet10.

Figure 8.

Absence of Fld1 from LD biogenesis sites results in ectopic droplet formation. (A and B) LDs form at Fld1- and Nem1-marked ER sites. 3ΔKO expressing Fld1-mCherry (A) or Nem1-mCherry (B) and coexpressing Erg6-GFP were transferred to galactose-containing media for the indicated time. White arrowheads denote colocalization between Erg6 with Fld1 (A) or Nem1 (B). (C) Quantification of colocalization of Erg6 with Fld1 and Nem1. Data are means ± SD, n > 50 cells. (D) LDs fail to form at Nem1 sites in cells lacking Fld1. fld1Δ mutant cells (fld1Δ 3ΔKO) expressing Nem1-mCherry and coexpressing Erg6-GFP were grown as described in A. White arrowheads indicate Erg6 punctae that do not colocalize with Nem1. Scale bar, 5 µm. (E and F) Quantification of cells with at least 1 LD (E), or number of LDs per cell (F) for images shown in D. Data are means ± SD, n > 50 cells; ***, P < 0.0005. (G) Cartoon depicting lack of Fld1 from LD biogenesis sites does not affect the localization of Nem1 and Yft2 but results in mislocalization of Pex30 and failure to recruit the TAG-synthase. (H) Cartoon showing that ER subdomains containing Fld1, Nem1, Yft2, and Pex30 define sites for the localization of the TAG-synthase and droplet formation. Erg6 and Pet10 bind to newly formed LDs from the cytosolic side.

Figure S5.

Induction of TAG-synthase Dga1 results in LD formation at Fld1 and Nem1 sites. (A and B) Fld1 and Nem1 sites are stained with BODIPY upon TAG synthesis. 4ΔKO cells expressing Fld1-mCherry (A) or Nem1-mCherry (B) and coexpressing Dga1 from a galactose-inducible promoter were switched to galactose media for 1 h and stained with BODIPY. White arrowheads indicate colocalization between BODIPY and Fld1/Nem1 punctae. Scale bars, 5 µm. (C and D) ER subdomains marked by Fld1 and Nem1 become enriched with LD marker protein. 4ΔKO cells expressing Fld1-mCherry (C) or Nem1-mCherry (D), Erg6-GFP, and coexpressing Dga1 from a galactose-inducible promoter were grown as described in A. White arrowheads indicate colocalization between Erg6 and Fld1/Nem1 punctae. Scale bars, 5 µm. (E and F) Dga1 colocalizes with Fld1/Nem1 sites. 4ΔKO cells expressing Fld1-GFP (E) or Nem1-GFP (F) and coexpressing Dga1-mCherry from a galactose-inducible promoter were switched to galactose for the indicated time. White arrowheads denote colocalization between Dga1-mCherry and Fld1-GFP/Nem1-GFP punctae. Scale bars, 5 µm. (G) Cartoon illustrating the recruitment of Dga1-mCherry to Fld1 and Nem1 ER sites, where droplets begin to form.

Figure 9.

Model of LD formation at discrete ER subdomains defined by Fld1 and Nem1. Cartoon showing the order of events (steps a to h) occurring at predefined ER subdomains for efficient LD formation. Fld1- and Nem1-marked ER subdomains undergo sequential enrichment with LD biogenesis factors, thereby promoting localized TAG production and droplet assembly.