Multi-Omic analyses characterize the ceramide/sphingomyelin pathway as a therapeutic target in Alzheimer's disease

Abstract

Dysregulation of sphingomyelin and ceramide metabolism have been implicated in Alzheimer's disease. Genome-wide and transcriptome-wide association studies have identified various genes and genetic variants in lipid metabolism that are associated with Alzheimer's disease. However, the molecular mechanisms of sphingomyelin and ceramide disruption remain to be determined. We focus on the sphingolipid pathway and carry out multi-omics analyses to identify central and peripheral metabolic changes in Alzheimer's patients, correlating them to imaging features. Our multi-omics approach is based on (a) 2114 human post-mortem brain transcriptomics to identify differentially expressed genes; (b) in silico metabolic flux analysis on context-specific metabolic networks identified differential reaction fluxes; (c) multimodal neuroimaging analysis on 1576 participants to associate genetic variants in sphingomyelin pathway with Alzheimer's disease pathogenesis; (d) plasma metabolomic and lipidomic analysis to identify associations of lipid species with dysregulation in Alzheimer's; and (e) metabolite genome-wide association studies to define receptors within the pathway as a potential drug target. We validate our hypothesis in amyloidogenic APP/PS1 mice and show prolonged exposure to fingolimod alleviated synaptic plasticity and cognitive impairment in mice. Our integrative multi-omics approach identifies potential targets in the sphingomyelin pathway and suggests modulators of S1P metabolism as possible candidates for Alzheimer's disease treatment.

Fig. 1. Overview of sphingolipid pathway manually curated from the Recon3D model.

The metabolites participating in reactions are represented in boxes. The arrows for reactions A–K are colored based on the direction in the pathway. Some reactions are not reversible (single arrows). The table on the right lists the catalyzing enzymes in the sphingolipid pathway in humans and is denoted with the same color code as the reaction arrow.

Fig. 2. In silico flux analysis for metabolic reactions in the sphingolipid pathway.

Box plot of normalized reaction fluxes for a serine palmitoyl transferase (SPT), b sphingomyelin synthase (SMS), and c ceramide kinase (CERK) reactions. The orange, mustard yellow, and blue bars correspond to Alzheimer’s Disease (AD), mild cognitive impairment (MCI), and no cognitive impairment (NCI). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ns is non-significant.

Fig. 3. Association of genetic variants in SPTLC3 and SGMS1 with structural (MRI) and molecular (FDG-PET) neuroimaging phenotypes.

a Gene-based association analysis of SPTLC3 with cognitive performance (Rey auditory verbal learning test total score). d Gene-based association analysis of SGMS1 with global brain glucose metabolism. b and e Surface-based whole brain analysis of cortical thickness (brain atrophy measured from MRI scans) for SPTLC3 and SGMS1. c and f Voxel-based whole brain analysis of brain glucose metabolism measured from FDG-PET scans for SPTLC3 and SGMS1. Red color suggests a decrease in glucose metabolism. chr chromosome, FDG fluorodeoxyglucose, MRI magnetic resonance imaging, PET positron emission tomography, SNP single nucleotide polymorphism.

Fig. 4. Hybrid network of genetic associations revealed by gene-based association studies and significant partial correlations of detected sphingomyelins^, .

The six identified genes can be grouped into two categories: global sphingomyelin synthesis and synthesis and degradation of sphingosine-1-phosphate. The selected SM ratio is colored orange, other SM species are in green (light green: non-targeted metabolomics in Shin et al.; dark green: targeted metabolomics in ADNI and Draisma et al.), and genes are in dark yellow. S1P Sphingosine-1-phosphate, SM sphingomyelin species.

Fig. 5. Fingolimod (FTY720) ameliorates memory and synaptic impairment in APP/PS1 mice.

a Exploration time spent on the novel object in the NOR test session. Data are expressed as a discrimination index ± SEM. FTY720 treatment significantly enhances the discrimination index of the APP/PS1 mice at 9 mo. b Barnes maze performance during training days. Acquisition learning trials were performed, and the time it took to locate and enter the escape box is reported in seconds. The average performance of four trials per day is expressed as mean ± SEM. A shorter latency indicates faster spatial learning. No significant difference across trials between APP/PS1 treated and untreated was found. c Probe trial was performed on day 5 of the Barnes Maze protocol, during which the escape box was removed. The time spent in the target/escape hole is plotted ±SEM. A larger percentage of time indicates better spatial memory. FTY720 mitigated the spatial learning deficits of the APP/PS1 at 9 mo. d CA3 to CA1 synapse LTP. The four small line graphs are representative analog traces of evoked EPSPs before (light colors) and after (dark colors) high-frequency stimulation (HFS). The large plot graph is an LTP timeline. Plotted are normalized evoked excitatory postsynaptic potentials (EPSPs) slopes (Y) vs. recording time (X). The first 15 min of evoked responses were normalized and used as the baseline responses of LTP. e The magnitude of LTP was determined according to the responses between 60 and 75 min after the HFS. Data represent mean fEPSP Slope ± SEM. A rescue of LTP at the CA3-CA1 synapse in APP/PS1 mice at 9 mo is observed after chronic FTY720 treatment. f LECII to LECII synapse LTP. The four small line graphs are representative analog traces of evoked EPSPs before (light color) and after (dark color) HFS. The large plot graph is an LTP timeline. g LTP magnitude between 60 and 75 min after the HFS. Data represent mean fEPSP Slope ±SEM. FTY720 treatment rescues LTP at the LECII–LECII synapse in APP/PS1 mice at 9 mo. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. fEPSP Field excitatory post-synaptic potentials, WT Wild type.