High-resolution mapping reveals topologically distinct cellular pools of phosphatidylserine

Abstract

Phosphatidylserine (PS) plays a central role in cell signaling and in the biosynthesis of other lipids. To date, however, the subcellular distribution and transmembrane topology of this crucial phospholipid remain ill-defined. We transfected cells with a GFP-tagged C2 domain of lactadherin to detect by light and electron microscopy PS exposed on the cytosolic leaflet of the plasmalemma and organellar membranes. Cytoplasmically exposed PS was found to be clustered on the plasma membrane, and to be associated with caveolae, the trans-Golgi network, and endocytic organelles including intraluminal vesicles of multivesicular endosomes. This labeling pattern was compared with the total cellular distribution of PS as visualized using a novel on-section technique. These complementary methods revealed PS in the interior of the ER, Golgi complex, and mitochondria. These results indicate that PS in the lumenal monolayer of the ER and Golgi complex becomes exposed cytosolically at the trans-Golgi network. Transmembrane flipping of PS may contribute to the exit of cargo from the Golgi complex.

Figure 1.

Plasma membrane localization of expressed GFP-Lact-C2. (a, top) A431 cells cotransfected with GFP-Lact-C2 (left) and RFP-PH-PLCδ (middle), as well as a merged image (right). (a, bottom) A431 cells transfected with mRFP-Lact-C2 and stained with Alexa Fluor 488–labeled cholera Toxin B. (b) Representative frozen section of cells expressing GFP-Lact-C2 and immunogold labeled for the GFP tag. Note labeling over the PM of the expressing cell, but not the neighboring cell, and the lack of cytosolic labeling. The nucleus (N) and nuclear envelope are unlabeled but multivesicular endosomes (E) show significant labeling. (c–e) Cells transfected with GFP-Lact-C2 double labeled for PTRF/cavin-1 (small gold) and GFP (large gold). Note the significant labeling for GFP in the regions enriched in PTRF/cavin-1–labeled caveolae (arrowheads). (f and g) PM lawns were prepared from GFP-Lact-C2–transfected cells and labeled with anti-GFP antibody gold conjugates. GFP-Lact-C2 is clustered in undifferentiated regions of the PM (f) and also labels vesicular profiles (g). Arrows indicate membrane domains consistent with caveolae. Cluster analysis suggests peak clustering at ∼22 nm (inset). Bars: (b–g) 200 nm.

Figure 2.

Comparison of expressed GFP-LactC2 with postprocessing immunolabeling. (a, top) A431 cells cotransfected with mRFP-Lact-C2 (left) and sec61α-GFP (middle). The images are overlaid on the right. (a, bottom) A431 cells cotransfected with mRFP-Lact-C2 and GFP-KDEL. (b) Binding specificity of GST-Lact-C2: overlay assays were performed using lipid strips from Echelon. PA, phosphatidic acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphatidylinositol; SM, sphingomyelin; TAG, triacylglycerol. (c) BHK cells transiently transfected with GFP-Lact-C2, frozen sectioned, and stained using anti-GFP antibody followed by incubation with gold-labeled protein A. Note the labeling of the PM and endosomes (E), but negligible labeling of ER and mitochondria (M). (d and e) Mouse liver samples or BHK cells (inset) were high pressure-frozen, freeze-substituted, and embedded in UV-polymerized Lowicryl at low temperature. Sections were incubated with purified GST-Lact-C2 and labeled as described in Materials and methods. Images were pseudocolored to highlight gold particles, which were color-coded to indicate assignment to specific compartments: red, mitochondria; yellow, RER; black, unassigned. Note the labeling of mitochondria, including cristae (inset), and the RER, particularly in close apposition to the ER. Unmodified images of d and e are provided in Fig. S5. Bars: (a) 4 µm; (c–e) 200 nm.

Figure 3.

On-section ultrastructural localization of PS. (a) A431 cells transiently cotransfected with GFP-Lact-C2 (left) and mito-RFP (middle). The images are overlaid on the right. (b–e) Mouse liver (b and c) and A431 samples (d and e) were processed using HPF/FS/LTE. Sections were incubated with purified GST-Lact-C2 (b, d, and e) or with GST (c and inset) and labeled as described in Materials and methods. Images were color-coded to indicate specific compartments: green, PM; red, mitochondria; yellow, RER; black, unassigned. Note the specific labeling of mitochondria, the RER, and the PM with GST-Lact-C2, compared with GST alone. Unmodified images of b–e are provided in Fig. S5. G, Golgi complex; LD, lipid droplet; S, sinusoidal membrane; BC, bile canaliculus. Bars: (a) 4 µm; (b–e) 200 nm.

Figure 4.

Localization of PS to the Golgi complex and endosomes. (a, top) HeLa cells transiently cotransfected with mRFP-Lact-C2 (left) and GalT-GFP (middle). (a, middle) HeLa cells transfected with mRFP-Lact-C2 were fixed, permeabilized, and immunostained for TGN46. (a, bottom) HeLa cells transiently transfected with GFP-Lact-C2 and incubated with rhodamine-transferrin for 30 min. (b–d) Cultured cells were processed using HPF/FS/LTE. Sections were incubated with purified GST-Lact-C2 (b and c) or GST (d) and labeled as described in Materials and methods. c′ shows the image in c pseudocolored to highlight gold particles, color-coded to indicate specific compartments: orange, Golgi complex; blue, endosomes (End). (e and f) BHK cells transiently transfected with GFP-Lact-C2 and a TGN marker, GCC88-myc, fixed, frozen-sectioned, and double-labeled for TGN (arrows, large gold, 10 nm) and GFP (arrowheads, small gold, 5 nm). The expressed probe associates with membranes close to the Golgi but does not label the entire Golgi stack (e and f). In contrast, labeling with the overlaid probe is throughout the Golgi complex, including cisternae (b–c’). Bars: (a) 4 µm; (b–f) 200 nm.

Figure 5.

Distribution of PS in the endocytic pathway. A431 cells doubly transfected with GFP-Lact-C2 and RFP-Rab5 (a) or GFP-Lact-C2 and Lamp1-RFP (b), or transfected with GFP-Lact-C2, then incubated in the presence of tetramethylrhodamine (TMR) dextran for 16 h and chased for 1 h before visualization (c), as indicated. In a and b, arrows identify representative endosomal structures labeled by both markers. Inset shows enlarged views of the merged image. Bars, 5 µm.

Figure 6.

Ultrastructural localization of PS in the endocytic pathway. Cells transfected with GFP-Lact-C2 and GFP detected on frozen sections (a–c), or processed using HPF/FS/LTE and overlaid with purified GST-Lact-C2 (d–f). (a–c) The cytoplasmically expressed probe labels the PM but not the ER. Multivesicular endosomes (E) show strong labeling, particularly on the cytoplasmic aspect (e.g., arrowheads in a). Strong labeling is also evident in large vesicles close to the PM (arrows in B). (d–f) On-section labeling reveals labeling of internal membranes of MVB including putative late endosomes/lysosomes (L) and early endosomes (EE). Arrowheads indicate ILVs. Bars, 200 nm.

Figure 7.

Ultrastructural localization of the cytoplasmically expressed GFP-C2 probe; HPF-FS method. BHK cells expressing GFP-Lact-C2. Cells were high-pressure frozen, freeze-substituted, and embedded in UV-polymerized Lowicryl at low temperature. Sections were immunogold-labeled for GFP. (a) Low-magnification overview showing specificity of labeling (top cell transfected, bottom cell untransfected). Note the high labeling on the PM and on multivesicular endosomes (E) and the absence of labeling of the nuclear envelope. (b–f) Gallery of representative images showing labeling of endosomes but low labeling of mitochondria (M) and Golgi cisternae (G). (b) A putative early endosome (EE), recognizable by the bilayered clathrin patch (arrow), shows significant labeling. (c–e) Individual multivesicular endosomes show distinct labeling patterns including cytoplasmic staining (c and d, arrows), predominantly internal labeling (f), a mixture of the two, or unlabeled (c, asterisk). Note the striking labeling of a specific subpopulation of internal membranes in c, d, and the inset (arrowheads). (e and f) Golgi cisternae show negligible labeling, but vesicles and tubules close to the Golgi, putative TGN elements, including clathrin-coated buds (arrows), show significant labeling. Bars, 200 nm.

Figure 8.

PS becomes cytosolically exposed during progression through the secretory pathway. COS7 were cells double-transfected with VSVG-GFP and mRFP-Lact-C2 (a–c), then incubated as indicated to label the ER (40°C overnight), the Golgi complex (20°C for 1 h after 40°C overnight), and post-Golgi compartments (37°C for 40 min, after 40°C overnight followed by 1 h at 20°C). Insets show enlargement of indicated merged image. Cells expressing VSVG-GFP (d) were grown overnight at 40°C, then shifted to 20°C for 1 h, fixed, and immunostained for TGN46. Insets show enlarged views of the boxed region in the merged image. Bars, 5 µm.

Figure 9.

Distribution of PS following elevation of cytosolic Ca²⁺ and induction of apoptosis. (a and b) HeLa cells transfected with GFP-Lact-C2 (a) were incubated with phosphatidylethanolamine-conjugated annexin V, or GFP-Lact-C2 and PM-RFP (b), and imaged before (top) or 10 min after treatment with 10 µM ionomycin in the presence of 2.5 mM Ca²⁺ (bottom). (c and d) Cells were transfected with GFP-Lact-C2 (c) or cotransfected with RFP-Lact-C2 and GFP-KDEL (d). Cells were exposed to anti-FAS IgM (1.5 µg/ml) and images were acquired after 4 h. Apoptotic cells were identified using Alexa Fluor 647–conjugated annexin V. Cells that had undergone rounding are deemed to be in a more advanced state of apoptosis. Images in a and d are representative of three experiments of each type. Bars, 10 µm. (e) Pearson’s correlation coefficients calculated to assess the colocalization of RFP-Lact-C2 and GFP-KDEL. Values obtained separately for 12 untreated (*) and 12 apoptotic cells (•) are plotted; the mean for each dataset is indicated by the horizontal line.