Sphingosine-1-phosphate phosphohydrolase regulates endoplasmic reticulum-to-golgi trafficking of ceramide

Abstract

Previous studies demonstrated that sphingosine-1-phosphate (S1P) phosphohydrolase 1 (SPP-1), which is located mainly in the endoplasmic reticulum (ER), regulates sphingolipid metabolism and apoptosis (H. Le Stunff et al., J. Cell Biol. 158:1039-1049, 2002). We show here that the treatment of SPP-1-overexpressing cells with S1P, but not with dihydro-S1P, increased all ceramide species, particularly the long-chain ceramides. This was not due to inhibition of ceramide metabolism to sphingomyelin or monohexosylceramides but rather to the inhibition of ER-to-Golgi trafficking, determined with the fluorescent ceramide analog N-(4,4-difluoro-5,7-dimethyl-4-bora-3a,4a-diaza-s-indacene-3-pentanoyl)-d-erythro-sphingosine (DMB-Cer). Fumonisin B1, an inhibitor of ceramide synthase, prevented S1P-induced elevation of all ceramide species and corrected the defect in ER transport of DMB-Cer, readily allowing its detection in the Golgi. In contrast, ceramide accumulation had no effect on either the trafficking or the metabolism of 6-([N-(7-nitrobenzo-2-oxa-1,3-diazol-4-yl)amino]hexanoyl)-sphingosine, which rapidly labels the Golgi even at 4 degrees C. Protein trafficking from the ER to the Golgi, determined with vesicular stomatitis virus ts045 G protein fused to green fluorescent protein, was also inhibited in SPP-1-overexpressing cells in the presence of S1P but not in the presence of dihydro-S1P. Our results suggest that SPP-1 regulates ceramide levels in the ER and thus influences the anterograde membrane transport of both ceramide and proteins from the ER to the Golgi apparatus.

FIG. 1.

ESI-MS/MS measurements of uptake and cellular levels of S1P and dihydro-S1P. HEK 293 cells expressing vector or SPP-1 were incubated in serum-free media with vehicle (□), 5 μM S1P (▪), or 5 μM dihydro-S1P () for 24 h as described in Materials and Methods. Lipids were extracted and phosphorylated (A and B) and free sphingoid bases (C and D) were analyzed by ESI-MS/MS. The data are averages of triplicate determinations and are expressed as pmol lipid per mg of protein. , P < 0.05 compared to vector.

FIG. 2.

Uptake and metabolism of dihydro-S1P and S1P. (A) HEK 293 cells overexpressing SPP-1 were incubated in serum-free media with 5 μM [³²P]S1P (□) or [³²P]dihydro-S1P (▪) for the indicated times. Lipids were extracted and phosphorylated sphingoid bases separated by TLC as described in Materials and Methods. The data are expressed as the percentage of ³²P-labeled sphingoid base remaining in the medium. (B) HEK 293 cells expressing vector (solid lines) or SPP-1 (dashed lines) were incubated in serum-free media with [³²P]S1P (triangles) or [³²P]dihydro-S1P (squares) for the indicated times. Lipids were extracted from cells and analyzed by TLC. The data are expressed as pmol of phosphorylated sphingoid base per mg of protein and are means ± the SD of triplicate determinations. , P < 0.05 compared to vector. (Inset) Cell-associated phospholipids from duplicate cultures after 24 h of treatment with [³²P]S1P or [³²P]dihydro-S1P (DH-S1P) were extracted and analyzed by TLC. Bands are labeled based on comigration with authentic unlabeled standards. LcbP, sphingoid base phosphates. A representative autoradiograph of a TLC plate is shown.

FIG. 3.

S1P, but not dihydro-S1P, markedly increases specific ceramide species. HEK 293 cells expressing vector (□) or SPP-1 (▪) were incubated in serum-free media with vehicle, S1P, or dihydro-S1P for 24 h. Lipids were extracted, and ceramide species were determined by ESI-MS/MS. The data are averages of triplicate determinations and are expressed as pmol of lipid per mg of protein. , P < 0.05 compared to vector.

FIG.4.

S1P, but not dihydro-S1P, alters ER-to-Golgi trafficking of DMB-Cer in SPP-1-expressing cells. (A) HEK 293 cells overexpressing SPP-1 were transfected with DsRed-Monomer Golgi and incubated with DMB-Cer (5 μM) for 30 min at 4°C. After being washed, the cells were incubated for 30 min at 37°C in DMEM containing 0.34 mg of BSA/ml and fixed, and localization of fluorescence was then determined by confocal microscopy. The right panels show the superimposed merged pictures, with yellow indicating colocalization. The Pearson correlation coefficients for colocalization of DMB fluorescence with that of DsRed without and with S1P treatment were 0.91 ± 0.04 and 0.26 ± 0.1, indicating significant overlap and low overlap, respectively. (B) Vector- and SPP-1-expressing HEK 293 cells were preincubated in DMEM containing vehicle, S1P (1.5 μM), or dihydro-S1P (1.5 μM) for 24 h at 37°C and then incubated with DMB-Cer (5 μM) for 30 min at 4°C. After being washed, the cells were incubated for 30 min at 37°C in DMEM containing 0.34 mg of BSA/ml, fixed, and then examined by confocal fluorescence microscopy. (C) Duplicate experiments were quantified by scoring the number of cells that showed typical Golgi staining after pulse-labeling with DMB-Cer. A total of 90% ± 6% (n = 75) of vector- and SPP-1-transfected cells showed DMB-Cer targeting to the Golgi, even after treatment with dihydro-S1P, compared to less than 29% ± 5% (n = 75) of cells transfected with SPP-1 and treated with S1P for 30 min. (D) SPP-1-expressing cells grown on gridded coverslips were treated for 24 h without (control) or with 1.5 μM S1P, labeled with DMB-Cer as described in panel B, washed, and then warmed to 37°C for 30 min. The cells were then fixed, and DMB fluorescence images (green) were collected from known regions of the grid by confocal microscopy. The cells were then permeabilized, stained with antibodies to the Golgi protein GM130 (red), and reimaged. The right panels show the superimposed merged pictures, with yellow indicating colocalization.

FIG.4.

S1P, but not dihydro-S1P, alters ER-to-Golgi trafficking of DMB-Cer in SPP-1-expressing cells. (A) HEK 293 cells overexpressing SPP-1 were transfected with DsRed-Monomer Golgi and incubated with DMB-Cer (5 μM) for 30 min at 4°C. After being washed, the cells were incubated for 30 min at 37°C in DMEM containing 0.34 mg of BSA/ml and fixed, and localization of fluorescence was then determined by confocal microscopy. The right panels show the superimposed merged pictures, with yellow indicating colocalization. The Pearson correlation coefficients for colocalization of DMB fluorescence with that of DsRed without and with S1P treatment were 0.91 ± 0.04 and 0.26 ± 0.1, indicating significant overlap and low overlap, respectively. (B) Vector- and SPP-1-expressing HEK 293 cells were preincubated in DMEM containing vehicle, S1P (1.5 μM), or dihydro-S1P (1.5 μM) for 24 h at 37°C and then incubated with DMB-Cer (5 μM) for 30 min at 4°C. After being washed, the cells were incubated for 30 min at 37°C in DMEM containing 0.34 mg of BSA/ml, fixed, and then examined by confocal fluorescence microscopy. (C) Duplicate experiments were quantified by scoring the number of cells that showed typical Golgi staining after pulse-labeling with DMB-Cer. A total of 90% ± 6% (n = 75) of vector- and SPP-1-transfected cells showed DMB-Cer targeting to the Golgi, even after treatment with dihydro-S1P, compared to less than 29% ± 5% (n = 75) of cells transfected with SPP-1 and treated with S1P for 30 min. (D) SPP-1-expressing cells grown on gridded coverslips were treated for 24 h without (control) or with 1.5 μM S1P, labeled with DMB-Cer as described in panel B, washed, and then warmed to 37°C for 30 min. The cells were then fixed, and DMB fluorescence images (green) were collected from known regions of the grid by confocal microscopy. The cells were then permeabilized, stained with antibodies to the Golgi protein GM130 (red), and reimaged. The right panels show the superimposed merged pictures, with yellow indicating colocalization.

FIG. 5.

Metabolism of DMB-Cer in SPP-1-expressing cells. (A) Lipids were extracted from Vector- and SPP-1-expressing HEK 293 cells treated as described in Fig. 5B and separated by TLC. DMB-SM was quantified as described in Materials and Methods. The data are means ± the SD. (B) Vector- and SPP-1-expressing cells were preincubated without or with 1 μg of BFA/ml for 20 min at 37°C before being labeled with 5 μM DMB-Cer for 30 min at 4°C. After being washed, the cells were further incubated in medium containing 0.34 mg of BSA/ml without or with 1 μg of BFA/ml for the indicated times at 37°C. Cellular lipids were extracted and DMB-SM quantified. Symbols: ○, vector cells treated with S1P; •, SPP-1 cells treated with vehicle; ▵, SPP-1 cells treated with S1P; ▴, SPP-1 cells treated with BFA plus S1P.

FIG. 6.

Lack of effect of S1P and dihydro-S1P on intracellular trafficking and metabolism of NBD-Cer. (A) Vector- and SPP-1-expressing HEK 293 cells were preincubated in DMEM containing vehicle, S1P (1.5 μM), or dihydro-S1P (1.5 μM) for 24 h at 37°C and then incubated with NBD-Cer (5 μM) for 30 min at 4°C. After being washed, the cells were incubated for 30 min at 37°C in DMEM containing 0.34 mg of BSA/ml and then examined by confocal fluorescence microscopy. (B) Lipids were extracted from duplicate cultures and separated by TLC.

FIG.7.

Effects of FB1 on intracellular ceramide levels and ER-to-Golgi trafficking of DMB-Cer. (A) Vector- and SPP-1-transfected HEK 293 cells were incubated for 24 h without or with 1.5 μM S1P in the absence or presence of 25 μM FB1. Ceramide content was determined by mass spectrometry as described in Materials and Methods. The data are expressed as pmol per 10⁶ cells and are means ± the SD of triplicate determinations. (B and C) Vector- and SPP-1-transfected HEK 293 cells were incubated for 24 h without or with 1.5 μM S1P in the absence or presence of 25 μM FB1. Cells were then incubated with DMB-Cer (5 μM) for 30 min at 4°C. After being washed, the cells were incubated for 30 min (B) or for the indicated times (C) at 37°C in DMEM containing 0.34 mg of BSA/ml and then examined by confocal fluorescence microscopy. For panel C, duplicate experiments were quantified by scoring the number of cells that showed typical Golgi staining after pulse-labeling with DMB-Cer. (D) Vector (circles)- and SPP-1 (triangles)-transfected cells were preincubated with 1.5 μM S1P in the absence (open symbols) or presence (filled symbols) of 25 μM FB1 for 24 h at 37°C and then labeled with 5 μM DMB-Cer for 30 min at 4°C. After being washed, cells were further incubated in medium containing 0.34 mg of BSA/ml for the indicated times at 37°C. Cellular lipids were extracted, separated by TLC and DMB-SM quantified. Symbols: ○, vector cells treated with S1P; ▵, SPP-1 cells treated with S1P; ▴, vector cells treated with S1P plus FB1; ▴, SPP-1 cells treated with S1P plus FB1.

FIG.7.

Effects of FB1 on intracellular ceramide levels and ER-to-Golgi trafficking of DMB-Cer. (A) Vector- and SPP-1-transfected HEK 293 cells were incubated for 24 h without or with 1.5 μM S1P in the absence or presence of 25 μM FB1. Ceramide content was determined by mass spectrometry as described in Materials and Methods. The data are expressed as pmol per 10⁶ cells and are means ± the SD of triplicate determinations. (B and C) Vector- and SPP-1-transfected HEK 293 cells were incubated for 24 h without or with 1.5 μM S1P in the absence or presence of 25 μM FB1. Cells were then incubated with DMB-Cer (5 μM) for 30 min at 4°C. After being washed, the cells were incubated for 30 min (B) or for the indicated times (C) at 37°C in DMEM containing 0.34 mg of BSA/ml and then examined by confocal fluorescence microscopy. For panel C, duplicate experiments were quantified by scoring the number of cells that showed typical Golgi staining after pulse-labeling with DMB-Cer. (D) Vector (circles)- and SPP-1 (triangles)-transfected cells were preincubated with 1.5 μM S1P in the absence (open symbols) or presence (filled symbols) of 25 μM FB1 for 24 h at 37°C and then labeled with 5 μM DMB-Cer for 30 min at 4°C. After being washed, cells were further incubated in medium containing 0.34 mg of BSA/ml for the indicated times at 37°C. Cellular lipids were extracted, separated by TLC and DMB-SM quantified. Symbols: ○, vector cells treated with S1P; ▵, SPP-1 cells treated with S1P; ▴, vector cells treated with S1P plus FB1; ▴, SPP-1 cells treated with S1P plus FB1.

FIG.8.

S1P, but not dihydroS1P, alters the intracellular trafficking of VSVG-ts045 protein in SPP-1-overexpressing cells. HEK 293 cells stably expressing vector (A) or SPP-1 (B) were transfected with VSVG-ts045 and incubated in DMEM in the presence of vehicle, 1.5 μM S1P, or 1.5 μM dihydro-S1P for 20 h at 40°C, shifted to 32°C for the indicated times, and examined by confocal microscopy. (C) The percentage of cells with distinct Golgi staining was determined. The data are means ± the SD.

FIG.8.

S1P, but not dihydroS1P, alters the intracellular trafficking of VSVG-ts045 protein in SPP-1-overexpressing cells. HEK 293 cells stably expressing vector (A) or SPP-1 (B) were transfected with VSVG-ts045 and incubated in DMEM in the presence of vehicle, 1.5 μM S1P, or 1.5 μM dihydro-S1P for 20 h at 40°C, shifted to 32°C for the indicated times, and examined by confocal microscopy. (C) The percentage of cells with distinct Golgi staining was determined. The data are means ± the SD.