Redirection of sphingolipid metabolism drives cytoskeletal defects in SPLIS and reveals ROCK inhibition as therapy
Adam Majcher, Ranjha Khan, Kathrin Buder, Florence Bourquin, Julie D. Saba, Thorsten Hornemann
Adam Majcher, Ranjha Khan, Kathrin Buder, Florence Bourquin, Julie D. Saba, Thorsten Hornemann
View: Text | PDF
Research Article Genetics Metabolism Nephrology

Redirection of sphingolipid metabolism drives cytoskeletal defects in SPLIS and reveals ROCK inhibition as therapy

Abstract

Sphingosine-1-phosphate lyase (SPL) insufficiency syndrome (SPLIS), also known as nephrotic syndrome type 14, is an autosomal recessive multisystem disorder caused by loss-of-function mutations in SGPL1, encoding the enzyme responsible for the terminal degradation of sphingosine-1-phosphate (S1P). We investigated a patient carrying a previously undescribed c.1084T>A (p.Ser362Thr) SGPL1 variant and analyzed the metabolic and cellular consequences of SPL deficiency, using patient fibroblasts, SGPL1-KO HEK293T cells, and Sgpl1–/– and Sgpl1rosa+fl/fl mice. Metabolic stable isotope labeling revealed that SPL deficiency does not invariably result in S1P accumulation. Instead, SPL-deficient cells maintain near-normal S1P levels through (a) feedback regulation of de novo sphingolipid synthesis via the ORMDL–ceramide axis and (b) increased diversion of excess ceramides into glycosphingolipids. However, perturbation of sphingolipid homeostasis, either by exogenous sphingolipid load or disruption of compensatory regulation, induces pathological intracellular S1P accumulation. In vivo, Sgpl1–/– mice had pronounced urinary S1P excretion and renal S1P enrichment, accompanied by cytoskeletal disorganization and impaired epithelial morphogenesis. Mechanistically, we identify aberrant Rho/ROCK signaling as a key mediator of S1P-driven cytoskeletal dysregulation. Pharmacological ROCK inhibition with fasudil mitigated renal cytoskeletal defects in Sgpl1–/– and Sgpl1rosa+fl/fl mice and partially restored epithelial architecture. These findings redefine the metabolic consequences of SPL deficiency and identify S1P-driven Rho/ROCK hyperactivation as a tractable therapeutic target in SPLIS.

Authors

Adam Majcher, Ranjha Khan, Kathrin Buder, Florence Bourquin, Julie D. Saba, Thorsten Hornemann

×

Figure 2

SPT regulation via the ORMDL/Cer axis prevents toxic S1P accumulation in SGPL1-deficient cells.

Options: View larger image (or click on image) Download as PowerPoint
SPT regulation via the ORMDL/Cer axis prevents toxic S1P accumulation in...
(A) Schematic overview of de novo SL synthesis and its regulation by SPT. (B) Reduced SPT activity in SPLIS primary skin fibroblasts lacking SGPL1 (SGPL1–) compared with control fibroblasts (SGPL1+). Cells were supplemented with exogenous SLs (+) or vehicle (–) to assess homeostatic regulation. (C) SPT activity in HEK293T SGPL1 KO cells compared with WT cells, with or without d7-So supplementation. (D) Total SL levels after ORMDL1–3 knockdown (KD) in HEK293T SGPL1 KO and WT cells. Cells were transfected with scrambled (ORMDL+) or ORMDL-targeting siRNAs (ORMDL–) using Lipofectamine 3000 for 72 hours before analysis. (E) Total Sa and sphinganine-1-phosphate (Sa1P) levels in HEK293T SGPL1 KO and WT cells treated with the Cer synthase inhibitor FB1. (F) De novo formation of Sa1P in SGPL1 KO cells treated with vehicle, FB1, FB1 plus cell-permeable C6Cer, or FB1 plus the SPT inhibitor myriocin. (G) FB1 toxicity assay in HEK293T SGPL1 KO and WT cells. Cells were treated with FB1 for 72 hours and ATP levels were quantified using the CellTiter-Glo assay. (H) Total SL levels in SPLIS fibroblasts (SGPL1–) and control fibroblasts (SGPL1+) treated with vehicle (MeOH), FB1, or FB1+myriocin. SPT activity was determined by measuring incorporation of d3-15N-serine after 24 hours. Plots represent mean ± SD (n = 3). HexCers represent the sum of glucosylceramide and galactosylceramide. SL levels were quantified by LC-MS/MS after lipid extraction. Statistical significance was assessed using Student’s t test with multiple-testing correction using the 2-stage step-up method of Benjamini, Krieger, and Yekutieli. **P < 0.01; ***P < 0.001. Toxicity assay data were normalized to the mean of vehicle-treated cells (n = 4). Created in BioRender.