Metabolic and Structural Consequences of GM3 Synthase Deficiency: Insights from an HEK293-T Knockout Model

Abstract

Background: GM3 Synthase Deficiency (GM3SD) is a rare autosomal recessive neurodevelopmental disease characterized by recurrent seizures and neurological deficits. The disorder stems from mutations in the ST3GAL5 gene, encoding GM3 synthase (GM3S), a key enzyme in ganglioside biosynthesis. While enzyme deficiencies affecting ganglioside catabolism are well-documented, the consequences of impaired ganglioside biosynthesis remain less explored. Methods: To investigate GM3SD, we used a Human Embryonic Kidney 293-T (HEK293-T) knockout (KO) cell model generated via CRISPR/Cas9 technology. Lipid composition was assessed via high-performance thin-layer chromatography (HPTLC); glycohydrolase activity in lysosomal and plasma membrane (PM) fractions was enzymatically analyzed. Lysosomal homeostasis was evaluated through protein content analysis and immunofluorescence, and cellular bioenergetics was measured using a luminescence-based assay. Results: Lipidome profiling revealed a significant accumulation of lactosylceramide (LacCer), the substrate of GM3S, along with increased levels of monosialyl-globoside Gb5 (MSGb5), indicating a metabolic shift in glycosphingolipid biosynthesis. Lipid raft analysis revealed elevated cholesterol levels, which may impair microdomain fluidity and signal transduction. Furthermore, altered activity of lysosomal and plasma membrane (PM)-associated glycohydrolases suggests secondary deregulation of glycosphingolipid metabolism, potentially contributing to abnormal lipid patterns. In addition, we observed increased lysosomal mass, indicating potential lysosomal homeostasis dysregulation. Finally, decreased adenosine triphosphate (ATP) levels point to impaired cellular bioenergetics, emphasizing the metabolic consequences of GM3SD. Conclusions: Together, these findings provide novel insights into the molecular alterations associated with GM3SD and establish the HEK293-T KO model as a promising platform for evaluating potential therapeutic strategies.

Keywords: GM3 synthase deficiency; gangliosides; lysosomes; neurodegeneration; plasma membrane.

Figure 7

Effect of GM3SD on sphingoid-base lipid metabolism in HEK293-T cells. GM3S metabolizes LacCer into GM3, the precursor for subsequent a- and b-series gangliosides. Alternatively, LacCer can be used as a substrate for the synthesis of 0-series gangliosides. Lipid species overexpressed in ST3GAL5 KO cells and catabolic hyperactivated enzymes are shown in red, while the downregulated ones are shown in blue. Enzymes that were not tested in this study are indicated in grey. The light-blue box highlights the a-series gangliosides that are not expressed in ST3GAL5 KO cells; the grey box indicates b-series and 0-series gangliosides that are not expressed in WT or ST3GAL5 KO cells. Oligosaccharide sugar symbols were selected according to Varki et al. [43]. ST3GAL2: ST3 beta-galactoside alpha-2,3-sialyltransferase 2; ST3GAL3: ST3 beta-galactoside alpha-2,3-sialyltransferase 3; ST3GAL5: ST3 beta-galactoside alpha-2,3-sialyltransferase 5; Neu: neuraminidase; B3GALT4: beta-1,3-galactosyltransferase 4; B4GALNT1: be-ta-1,4-N-acetyl-galactosaminyltransferase 1; ST8SIA1: ST8 alpha-N-acetyl-neuraminide alpha-2,8-sialyltransferase 1; ST6GALNAC5: ST6 N-acetylgalactosaminide alpha-2,6-sialyltransferase 5; ST6GALNAC6: ST6 N-acetylgalactosaminide alpha-2,6-sialyltransferase 6; B3GALT5: beta-1,3-galactosyltransferase 5; B3GALNT1: be-ta-1,3-N-acetylgalactosaminyltransferase 1; A4GALT: alpha 1,4-galactosyltransferase; GCS: glucosylceramide synthase; CDase: ceramidase; CerS: ceramide synthase; SMS: sphingomyelin synthase; Cer: ceramide; SM: sphingomyelin.

Figure 1

Lipid composition in WT and ST3GAL5 KO cells. An endogenous lipid pattern was obtained by extraction with chloroform/methanol/water, 20:10:1 (v/v/v), and a two-phase partitioning of total cell lysates from WT and ST3GAL5 KO cells. (a) The levels of sialylated glycosphingolipids were inferred by treating samples with Vibrio cholerae sialidase and visualized by Ehrlich reagent (for details, see Supplementary Figure S1), (b) neutral sphingolipids and ceramide were visualized by anisaldehyde reagent, (c) phospholipids and sphingomyelin were visualized by phosphorus reagent, and (d) cholesterol was visualized by anisaldehyde reagent. Lipids were identified by co-migration with specific lipid standards. The HPTLC images shown are representative of five independent experiments (n = 5) and have been equally adjusted for brightness and contrast. Data are displayed as fold changes with respect to WT cells (set to 1.0) and are presented as the mean ± SEM (* p < 0.05, ** p < 0.01 by Mann–Whitney test, WT vs. KO). SM: sphingomyelin; Cer: ceramide; Cho: cholesterol.

Figure 2

Purification of DRM fractions. Cell lysates obtained from WT and ST3GAL5 KO HEK293-T cells fed with [1-³H] sphingosine to the steady state were subjected to ultracentrifugation in a sucrose gradient to isolate plasma membrane microdomains. Twelve fractions were collected, starting from the top of the tube, with fractions 5–6 corresponding to the DRM fractions and fractions 10–12 to the HD fractions. (a) The radioactivity distribution in the different fractions was evaluated by liquid scintillation counting by a beta-counter and expressed as disintegrations per minute (d.p.m.). (b) Levels of specific protein markers in the WT (top) and ST3GAL5 KO (bottom) cells in the fractions were analyzed by WB to ensure proper DRM purification. Calnexin was used as an HD marker, while flotillin, Fyn proto-oncogene Src Family Tyrosine Kinase (Fyn), and prion protein (Prp) were used as DRM markers. Data are representative of five independent experiments (n = 5).

Figure 3

DRM lipid composition in WT and ST3GAL5 KO cells. Cell sphingolipids were metabolically labelled by [1-³H]sphingosine. Following cell lysis with 1% TX-100 at 4 °C, the HD and lipid rafts (DRMs) were isolated by ultracentrifugation on a discontinuous sucrose gradient. Extracted lipids, obtained from the organic phase and aqueous phase, corresponding to equal radioactivity amounts, were separated using HPTLC, visualized using digital autoradiography, and measured by M3 software (1.0.8.1187; Biospace Lab Inc; https://biospacelab.com/, accessed on 1 July 2019). (a) LacCer and (b) ganglioside (GM3 and GM2) and globoside (MSGb5 and DSGb5) levels were obtained by quantifying the specific lipid bands and normalizing them over the total radioactivity of the lipid species in the lane. (c) Cholesterol levels were visualized by means of anisaldehyde reagent and expressed as percentages of total cholesterol (HD + DRM). The HPTLC images shown are representative of five independent experiments (n = 5) and have been equally adjusted for brightness and contrast. Values are expressed as means ± SEMs (*** p < 0.001, ** p < 0.01, * p < 0.05 by Mann–Whitney test, WT vs. KO). Cho: cholesterol.

Figure 4

Activity of sphingolipid hydrolases in WT and ST3GAL5 KO cells. Enzymatic activity of (a) lysosomal and (b) PM-associated hydrolases. Data are displayed as fold changes with respect to WT cells (set to 1.0) and are presented as the means ± SEMs of six independent experiments (**** p < 0.0001, ** p < 0.01, * p < 0.05 by Student’s t-test, WT vs. KO, n = 6).

Figure 5

Consequences of GM3SD for lysosomal homeostasis. (a) Enzymatic activity of lysosomal and PM-associated β-Man hydrolase in WT and ST3GAL5 KO cells. (b) Lamp-1 protein levels were evaluated by Western blot (WB) analysis (left) and immunofluorescence (right, 100× magnification) in WT and ST3GAL5 KO cells. WB analysis of (c) LC3 and (d) p62 proteins in cells subjected to amino acid starvation in WT and ST3GAL5 KO cells. WB data were obtained by normalizing the signal intensity of LC3-II and p62 to that of GAPDH, used as a loading control; immunofluorescence results are expressed as the integrated density (IntDen, mean fluorescence x area measured of each cell; corrected for the background). Images shown are representative of five independent experiments and have been equally adjusted for brightness and contrast. Histograms depict the means ± SEMs of values obtained from five independent experiments (n = 5) and expressed as percentages with respect to WT controls (set to 1.0; A.U.: arbitrary units). Data were compared by Mann–Whitney tests ((a,b) **** p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.05, WT vs. KO) and one-way ANOVA followed by Tukey’s post hoc test ((c,d) ** p < 0.01, * p < 0.05 vs. WT before starvation, 0 h).

Figure 6

Effect of GM3SD on mitochondrial proteins and energy production. (a) The levels of mitochondrial proteins in WT and ST3GAL5 KO cells evaluated by WB. The WB images shown are representative of six independent experiments. Histograms display the levels of Oxphos complexes and TOM20. WB data were obtained by dividing the signal intensity of the band associated with specific proteins by that of GAPDH, used as the loading control. (b) Total ATP levels of WT and ST3GAL5 KO cells. Data are expressed as the mean ± SEM of the percentage of WT controls (set to 1.0) and were obtained from six independent experiments (* p < 0.05 by Mann–Whitney test, WT vs. KO, n = 6).