Journal: Nature Communications

Article Title: SPNS2 exports sphingosine-1-phosphate and imports glucose

doi: 10.1038/s41467-026-71659-7

Figure Lengend Snippet: a ConSurf analysis of hSPNS2 protein. b Pairwise Alignment Scores for hSPNS2 protein and DNA sequences with putative homologs arranged according to the degree of sequence identity. c S1P levels in blood (n = 13, 15), lymph fluid (n = 10, 12), perfused liver, lung, kidney, heart (n = 3, 3). d Glucose levels in blood (n = 10,14), lymph fluid (n = 7, 13), perfused liver (n = 7, 5), lung (n = 7, 10), kidney (n = 5, 5), heart (n = 5, 5). e Glucose levels in urine (n = 7, 6), feces (n = 10, 8), tibialis anterior muscle (n = 5, 5), and gonadal adipose tissue (n = 4, 5). f Body weights (n = 20, 20). Body composition expressed as percentage of fat mass (n = 12, 12). g Food consumption, energy balance, energy expenditure, and locomotor activity (Beam breaks/h) determined with the PhenoMaster (n = 9,10; N = 3). h Blood hemoglobin concentration (n = 17, 9) and percentage of glycosylated hemoglobin, HbA1c (n = 7, 7). i Oral glucose tolerance test (GTT) and area under the curve (AUC) (n = 5, 5; N = 2). j Plasma levels of fasting insulin (n = 14, 14), glucagon (n = 8, 8) and thyroid hormone triiodothyronine (T3). Data are means ± s.e.m. k – n PET-CT imaging analysis of 2-FDG in Spns2 –/– mice. PET tracer 2-FDG (270–550 µCi) was gavaged prior to anesthetization of WT or Spns2 –/– mice for PET-CT scans. (n = 3, 3, N = 2). k , l At 60 min, tracer activity of target organs was quantified in volumes of interest (VOI). Data are percentage of whole-body activity for the right kidney (RK), heart, and bladder (Bl). l Coronal sections (0.4 mm thick) of PET images are presented according to a spectral scale for tracer activity, from red (highest), to green (intermediate), to blue (lowest). m The distribution of 2-FDG in representative WT at the specified times following gavage is shown to demonstrate the assessment of gastric emptying and intestinal absorption. St, stomach; In, intestines. n Representative images of the VOI in bladder to determine % of urinary excretion. o % of urinary excretion, gastric emptying and intestinal absorption. Data are means ± s.d. Two-tailed unpaired t-test. Source data are available for this figure in the Source Data file.

Article Snippet: Cells were seeded (200,000 cells/6-well) and 24 h later, the cell culture medium was replaced with glucose-free DMEM (Gibco, # A14430-01, phenol-red and glucose-free) with 10% FBS, 1 mM sodium pyruvate, and 1 × GlutaMax for 4 h. The medium was then replaced by DMEM without or with 1 g/L or 4.5 g/L glucose (#G8644, Sigma) for 24 h. In other experiments, cells were treated with 200 μM of the S1P lyase inhibitor A6770 (#29972, Cayman Chemical Company), 2 μM of the SPNS2 inhibitor SLF1081851 (#HY-149004, MedChemExpress), or the corresponding vehicles in medium with 4.5 g/L glucose for 24 h. Where indicated, cells were treated with 500 nM or 10 μM S1P (#860492, Avanti Polar Lipids) for 30 min, 100 nM insulin (#C-52310, PromoCell) for 4 h, 10 nM BAY-876 (#HY-100017, MedChemExpress) for 4 h, or with 1 μM LB-100 (#HY-18597, MedChemExpress), 10 μM DT-061 (#HY-112929, MedChemExpress), 2 nM Calyculin A (#HY-18983, MedChemExpress), or 100 nM Endothall (#HY-113976A, MedChemExpress) for 24 h in medium with 4.5 g/L glucose.

Techniques: Sequencing, Activity Assay, Concentration Assay, Clinical Proteomics, Positron Emission Tomography-Computed Tomography, Imaging, Two Tailed Test

Journal: Nature Communications

Article Title: SPNS2 exports sphingosine-1-phosphate and imports glucose

doi: 10.1038/s41467-026-71659-7

Figure Lengend Snippet: a Immunofluorescence localization of SPNS2 (red) on plasma membrane of SVEC4-10. b Glucose increases S1P secretion by SVEC4-10 cells (n = 3, N = 3). c SVEC4-10 cells were treated with S1P lyase inhibitor A6770 (200 µM) or with SPNS2 inhibitor SLF1081851 (2 µM) and levels of S1P in cells (n = 8–10, N = 5) and medium (n = 9, N = 3) as well as glucose uptake were determined (n = 7, N = 3). d SPNS2 expression in two SPNS2 stably overexpressing SVEC4-10 cell lines generated by CRISPR activation plasmids (CTL1, SPNS2-OE1) or lentiviral activation particles (CTL2, SPNS2-OE2) and in two SPNS2 deleted SVEC4-10 cell lines (SPNS2-KO1 generated with double nickase plasmids, and SPNS2-KO2 via CRISPR/Cas9 knock-out and homology-directed repair plasmids) compared to their controls. e – h S1P levels in cells and medium, and glucose uptake were measured in SPNS2-OE1 ( e ), SPNS2-OE2 ( f ), SPNS2-KO1 ( g ), SPNS2-KO2 ( h ). (n = 6, N = 3). i , j SPNS2-OE1, SPNS2-KO1 cells and their controls were treated with S1P (500 nM), and phosphorylation of p42/44 ( i ) and glucose uptake ( j ) were determined (n = 6, N = 3). k , l Glucose uptake in SPNS2-OE1, SPNS2-KO1 and their control cells treated with insulin (100 nM) or GLUT1 inhibitor BAY-876 (10 nM) (n = 3, N = 3). Data are means ± s.d. b One-way analysis of variance test followed by Šídák’s multiple comparisons test. c – l two-tailed unpaired t-test. Source data are available for this figure in the Source Data file.

Article Snippet: Cells were seeded (200,000 cells/6-well) and 24 h later, the cell culture medium was replaced with glucose-free DMEM (Gibco, # A14430-01, phenol-red and glucose-free) with 10% FBS, 1 mM sodium pyruvate, and 1 × GlutaMax for 4 h. The medium was then replaced by DMEM without or with 1 g/L or 4.5 g/L glucose (#G8644, Sigma) for 24 h. In other experiments, cells were treated with 200 μM of the S1P lyase inhibitor A6770 (#29972, Cayman Chemical Company), 2 μM of the SPNS2 inhibitor SLF1081851 (#HY-149004, MedChemExpress), or the corresponding vehicles in medium with 4.5 g/L glucose for 24 h. Where indicated, cells were treated with 500 nM or 10 μM S1P (#860492, Avanti Polar Lipids) for 30 min, 100 nM insulin (#C-52310, PromoCell) for 4 h, 10 nM BAY-876 (#HY-100017, MedChemExpress) for 4 h, or with 1 μM LB-100 (#HY-18597, MedChemExpress), 10 μM DT-061 (#HY-112929, MedChemExpress), 2 nM Calyculin A (#HY-18983, MedChemExpress), or 100 nM Endothall (#HY-113976A, MedChemExpress) for 24 h in medium with 4.5 g/L glucose.

Techniques: Immunofluorescence, Clinical Proteomics, Membrane, Expressing, Stable Transfection, Generated, CRISPR, Activation Assay, Knock-Out, Phospho-proteomics, Control, Two Tailed Test

Journal: Nature Communications

Article Title: SPNS2 exports sphingosine-1-phosphate and imports glucose

doi: 10.1038/s41467-026-71659-7

Figure Lengend Snippet: a Proliferation of Spns2 overexpressing (SPNS2-OE1) or Spns2 deleted cells (SPNS2-KO1) and controls cells. 10 5 cells were cultured in medium containing 4.5 g/L glucose without or with 500 nM S1P as indicated and cell numbers measured after 72 h (n = 3). b – f Cells were cultured in medium without or with 4.5 g/L glucose and/or 500 nM S1P as indicated. b , c Migration of cells in wound healing assays 24 h after creating a gap in a confluent monolayer and change to media containing glucose or S1P as indicated. b Representative images at 0 or 24 h after initiation of migration and ( c ) percentages of wound closures determined in cell migration assays in the presence of aphidicolin (n = 4). d – f ECIS measurements of the resistance of the indicated cells. d Representative continuous resistance measurements. e normalized endpoint resistance of the indicated cells cultured for 118 h. Resistance was normalized to the value measured at 12 h. f Normalized endpoint resistance of the indicated cells cultured for 118 h in the absence or presence of glucose and/or S1P (n = 3). g FITC-dextran leakage from the indicated cell monolayers cultured in the present of glucose (n = 3, 6). h FITC-dextran leakage from the indicated cell monolayers in the absence or presence of glucose and/or S1P (n = 4). a , c , e – h Data are means ± s.d. ns, not significant; One-way analysis of variance test followed by Sidak's multiple comparisons test or ( h ) Welch’s ANOVA multiple comparisons test. Source data are available for this figure in the Source Data file.

Article Snippet: Cells were seeded (200,000 cells/6-well) and 24 h later, the cell culture medium was replaced with glucose-free DMEM (Gibco, # A14430-01, phenol-red and glucose-free) with 10% FBS, 1 mM sodium pyruvate, and 1 × GlutaMax for 4 h. The medium was then replaced by DMEM without or with 1 g/L or 4.5 g/L glucose (#G8644, Sigma) for 24 h. In other experiments, cells were treated with 200 μM of the S1P lyase inhibitor A6770 (#29972, Cayman Chemical Company), 2 μM of the SPNS2 inhibitor SLF1081851 (#HY-149004, MedChemExpress), or the corresponding vehicles in medium with 4.5 g/L glucose for 24 h. Where indicated, cells were treated with 500 nM or 10 μM S1P (#860492, Avanti Polar Lipids) for 30 min, 100 nM insulin (#C-52310, PromoCell) for 4 h, 10 nM BAY-876 (#HY-100017, MedChemExpress) for 4 h, or with 1 μM LB-100 (#HY-18597, MedChemExpress), 10 μM DT-061 (#HY-112929, MedChemExpress), 2 nM Calyculin A (#HY-18983, MedChemExpress), or 100 nM Endothall (#HY-113976A, MedChemExpress) for 24 h in medium with 4.5 g/L glucose.

Techniques: Cell Culture, Migration

Journal: Nature Communications

Article Title: SPNS2 exports sphingosine-1-phosphate and imports glucose

doi: 10.1038/s41467-026-71659-7

Figure Lengend Snippet: a Molecular dynamics simulations of the inward-facing open conformation of hSPNS2 (PDB ID 8EX4) show that S1P can slither up as its hydrocarbon tail becomes vertically aligned from its kinked structure. b Changes in the vertical distance of the S1P phosphate headgroup from its original location near S232 of SPNS2. c The upward translocation of S1P causes noticeable structural changes at the extracellular vestibule of SPNS2. d The root-mean-squared deviation (RMSD) of the SPNS2 structure as S1P slithers upward during the unbiased simulation. The set of hydrophobic residues shown in Fig. 4a pose a barrier for the upward movement illustrated by the free-energy plot shown in Supplementary Fig. . Once the phosphate group makes it past these residues, further upward movement is relatively easier, and the hydrophobic residues close in to obstruct its downward movement. e – g Snapshots of the representative, inward-facing open structures of SPNS2-S1P complexes at various times during the MD simulation in which S1P moves upward. Transient glucose-binding pockets were obtained by molecular docking of glucose into 2000 frames extracted from the 1000 ns simulation. h – j Close-up views of the glucose-binding pockets are shown in ( e – g ). Residues near the most stable glucose-binding pockets are indicated. Lower panels highlight residues within 3.5 Å of glucose in the top (site 1), central (sites 2 and 3), and bottom (site 4) glucose-binding pockets and H-bonds are indicated with dotted lines. k Glucose molecule hydrogen bonds with S1P and nearby protein residues before transitioning of glucose to the intracellular side (during 150–300 ns in the distance plot shown in m ). l Trace the glucose pathway as it enters the intracellular side from the extracellular side. m Changes in the vertical distance of glucose from its initial position at the central site 3 binding pocket as it traverses through the channel and passes into the intracellular side. Source data are available for this figure in the Source Data file.

Article Snippet: Cells were seeded (200,000 cells/6-well) and 24 h later, the cell culture medium was replaced with glucose-free DMEM (Gibco, # A14430-01, phenol-red and glucose-free) with 10% FBS, 1 mM sodium pyruvate, and 1 × GlutaMax for 4 h. The medium was then replaced by DMEM without or with 1 g/L or 4.5 g/L glucose (#G8644, Sigma) for 24 h. In other experiments, cells were treated with 200 μM of the S1P lyase inhibitor A6770 (#29972, Cayman Chemical Company), 2 μM of the SPNS2 inhibitor SLF1081851 (#HY-149004, MedChemExpress), or the corresponding vehicles in medium with 4.5 g/L glucose for 24 h. Where indicated, cells were treated with 500 nM or 10 μM S1P (#860492, Avanti Polar Lipids) for 30 min, 100 nM insulin (#C-52310, PromoCell) for 4 h, 10 nM BAY-876 (#HY-100017, MedChemExpress) for 4 h, or with 1 μM LB-100 (#HY-18597, MedChemExpress), 10 μM DT-061 (#HY-112929, MedChemExpress), 2 nM Calyculin A (#HY-18983, MedChemExpress), or 100 nM Endothall (#HY-113976A, MedChemExpress) for 24 h in medium with 4.5 g/L glucose.

Techniques: Translocation Assay, Binding Assay

Journal: Nature Communications

Article Title: SPNS2 exports sphingosine-1-phosphate and imports glucose

doi: 10.1038/s41467-026-71659-7

Figure Lengend Snippet: a , b Identification of key SPNS2 residues involved in glucose transport. a Western blots and representative images of localization of hSPNS2-TurboGFP and its mutants. b Glucose uptake activities of SPNS2 variants with mutations in potential key residues involved in glucose or S1P engagement. hSPN S2 , GLUT1 vector, or the indicated mutants were overexpressed in SPNS2-KO1 cells lacking endogenous SPNS2. Glucose uptake was normalized to SPNS2 expression determined by GFP fluorescence ( n = 5, N = 3). Data are means ± s.e.m. One-way analysis of variance test followed by Dunnett’s multiple comparison test. c – f Direct glucose and S1P transport by SPNS2 proteoliposomes. c Illustration of cell-free preparation of SPNS2 proteoliposomes for functional transport analysis. d Comparable levels of hSPNS2 and its variants by western blots. e , f Proteoliposomes of WT hSPNS2 and variants were loaded without or with glucose ( e ) or S1P ( f ) as indicated and uptake of 1 μM NBD-S1P ( e ) or 1 µM NBD-glucose ( f ) determined. Arbitrary units (a.u.) (n = 3, N = 3). Nonspecific uptake was measured using protein-free liposomes (empty), vector containing proteoliposomes (vector), and T1R1 containing proteoliposomes (control). (n = 3, N = 3). One-way analysis of variance test followed by Dunnett’s multiple comparisons test. g Illustration of SPNS2-mediated export of S1P out of cells while transporting glucose inward. Illustrations in panels c and g created by Luciana Giono. h – j SPNS2-mediated D-[3H]glucose uptake. h Time-dependent specific uptake of D-[ 3 H]glucose into hSPNS2-containing proteoliposomes that were loaded without or with S1P (n = 3–5, N = 3). Nonspecific uptake measured using protein-free liposomes (empty) was subtracted from the specific uptake. i Uptake of [ 3 H]glucose by hSPNS2 or empty liposomes at 40 sec (n = 4, N = 3). j Kinetics of D-glucose uptake by hSPNS2. Specific uptake measured at 40 sec was calculated by subtraction of nonspecific [ 3 H]glucose uptake by empty liposomes and fitted to a non-linear regression analysis using Michaelis–Menten enzyme kinetics plot with K M , V max , and k cat values calculated (n = 4, N = 4). Data are means ± s.e.m. of independent experiments. Source data are available for this figure in the Source Data file.

Article Snippet: Cells were seeded (200,000 cells/6-well) and 24 h later, the cell culture medium was replaced with glucose-free DMEM (Gibco, # A14430-01, phenol-red and glucose-free) with 10% FBS, 1 mM sodium pyruvate, and 1 × GlutaMax for 4 h. The medium was then replaced by DMEM without or with 1 g/L or 4.5 g/L glucose (#G8644, Sigma) for 24 h. In other experiments, cells were treated with 200 μM of the S1P lyase inhibitor A6770 (#29972, Cayman Chemical Company), 2 μM of the SPNS2 inhibitor SLF1081851 (#HY-149004, MedChemExpress), or the corresponding vehicles in medium with 4.5 g/L glucose for 24 h. Where indicated, cells were treated with 500 nM or 10 μM S1P (#860492, Avanti Polar Lipids) for 30 min, 100 nM insulin (#C-52310, PromoCell) for 4 h, 10 nM BAY-876 (#HY-100017, MedChemExpress) for 4 h, or with 1 μM LB-100 (#HY-18597, MedChemExpress), 10 μM DT-061 (#HY-112929, MedChemExpress), 2 nM Calyculin A (#HY-18983, MedChemExpress), or 100 nM Endothall (#HY-113976A, MedChemExpress) for 24 h in medium with 4.5 g/L glucose.

Techniques: Western Blot, Plasmid Preparation, Expressing, Fluorescence, Comparison, Functional Assay, Liposomes, Control

Journal: The FASEB Journal

Article Title: ORMDL Proteins Turnover via Proteasome and Autophagy Is Cell‐Type Dependent and Tied to Ceramide Homeostasis

doi: 10.1096/fj.202502924RR

Figure Lengend Snippet: ORMDLs reduction does not correlate with phosphorylated sphingolipid levels and is unaffected by direct dhS1P or S1P treatment. (A) Quantification of phosphorylated sphingolipids by LC‐ESI‐MS/MS in HEK293 cells treated with low or high concentrations of myriocin, FB 1 , or their combination, (B) together with relative quantification of ORMDLs, SPTLC1, SPTLC2, and phosphorylated AKT (S473) protein levels. (C) Analysis of ORMDLs, SPTLC1, SPTLC2, and the pAKT S473 /AKT ratio in HEK293 cells following a 6‐h desensitization period and subsequent 30‐ or 60‐min stimulation with dhS1P or S1P. In both (A) and (B), protein levels were normalized to total protein loading based on Ponceau S staining. (D) Representative time‐lapse images of HEK293 cells showing morphological responses to dhS1P or S1P stimulation. Scale bar: 100 μm. Data in (A‐C) are presented as geometric means ± GSEM ( N = 3). Statistical significance was assessed by one‐way ANOVA followed by Tukey's post hoc test. Significance levels: p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), p < 0.0001 (****); n.s., not significant.

Article Snippet: Briefly, 50 μL of homogenate was transferred to a 1.5 mL microcentrifuge tube (Cat# 780420, Brand GmbH, Wertheim, Germany), mixed with 25 μL of MilliQ water, and loaded with internal standards: 50 ng dhS1P d17:0 (Cat# 860655), 50 ng S1P d17:1 (Cat# 860641), and 25 ng C1P d18:1/C12:0 (Cat# 860531), each from Avanti Polar Lipids, dissolved in 5 μL of chloroform/methanol solution (2:1, v/v).

Techniques: Tandem Mass Spectroscopy, Quantitative Proteomics, Staining

Journal: The FASEB Journal

Article Title: ORMDL Proteins Turnover via Proteasome and Autophagy Is Cell‐Type Dependent and Tied to Ceramide Homeostasis

doi: 10.1096/fj.202502924RR

Figure Lengend Snippet: ORMDLs reduction does not correlate with phosphorylated sphingolipid levels and is unaffected by direct dhS1P or S1P treatment. (A) Quantification of phosphorylated sphingolipids by LC‐ESI‐MS/MS in HEK293 cells treated with low or high concentrations of myriocin, FB 1 , or their combination, (B) together with relative quantification of ORMDLs, SPTLC1, SPTLC2, and phosphorylated AKT (S473) protein levels. (C) Analysis of ORMDLs, SPTLC1, SPTLC2, and the pAKT S473 /AKT ratio in HEK293 cells following a 6‐h desensitization period and subsequent 30‐ or 60‐min stimulation with dhS1P or S1P. In both (A) and (B), protein levels were normalized to total protein loading based on Ponceau S staining. (D) Representative time‐lapse images of HEK293 cells showing morphological responses to dhS1P or S1P stimulation. Scale bar: 100 μm. Data in (A‐C) are presented as geometric means ± GSEM ( N = 3). Statistical significance was assessed by one‐way ANOVA followed by Tukey's post hoc test. Significance levels: p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), p < 0.0001 (****); n.s., not significant.

Article Snippet: Sphingolipid concentrations of S1P, dhC1P, and C1P were determined by single‐point calibration using external standards [ ]: 50 ng S1P d18:1 (Cat# 860492), 50 ng dhC1P d18:0/C16:0 (Cat# 860522), and 50 ng C1P d18:1/C16:0 (Cat# 860533), all from Avanti Polar Lipids.

Techniques: Tandem Mass Spectroscopy, Quantitative Proteomics, Staining

Journal: The FASEB Journal

Article Title: ORMDL Proteins Turnover via Proteasome and Autophagy Is Cell‐Type Dependent and Tied to Ceramide Homeostasis

doi: 10.1096/fj.202502924RR

Figure Lengend Snippet: Regulation of ORMDLs stability in human RPE‐1 cells and mouse BMMCs. (A) Levels of ORMDLs, SPTLC1, SPTLC2, and total d18:1 ceramides (measured by LC‐ESI‐MS/MS) were assessed in human RPE‐1 cells and mouse BMMCs after 24‐h treatment with 10 μM myriocin or 10 μM FB 1 , and compared with untreated controls. (B) Quantification of ORMDLs, SPTLC1, SPTLC2, and the LC3‐II/LC3‐I ratio was performed in untreated RPE‐1 cells and BMMCs, and compared with cells treated for 24 h with 50 μM CQ or 10 μM MG132, as determined by immunoblotting. (C) Effects of p97/VCP inhibition by CB‐5083 on ORMDLs, SPTLC1, SPTLC2, the LC3‐II/LC3‐I ratio, ATF4, and p62 levels were assessed in RPE‐1 cells and BMMCs. Protein levels in (A‐C) were normalized to total protein loading using Ponceau S staining. Data are presented as geometric mean ± GSEM ( N = 4). Statistical analysis was performed using one‐way ANOVA with Tukey's post hoc test. Significance levels: p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), p < 0.0001 (****); n.s., not significant.

Article Snippet: Sphingolipid concentrations of S1P, dhC1P, and C1P were determined by single‐point calibration using external standards [ ]: 50 ng S1P d18:1 (Cat# 860492), 50 ng dhC1P d18:0/C16:0 (Cat# 860522), and 50 ng C1P d18:1/C16:0 (Cat# 860533), all from Avanti Polar Lipids.

Techniques: Tandem Mass Spectroscopy, Western Blot, Inhibition, Staining