Disrupted Lipid Metabolism Aggravates Ischemic Brain Injury: Targeting FDFT1 for Stroke Therapy

Abstract

Lipid metabolism disorder has been established as a contributing factor to the exacerbation of ischemic stroke (IS) damage. Conditions such as metabolic dysfunction-associated steatotic liver disease (MASLD) and atherosclerosis are known to elevate the risk of IS. Therefore, elucidating the association between potential risk factors of IS and the pathogenesis of IS from the perspective of lipid regulation may provide new insight for the prevention and treatment. In our study, we obtained Gene Expression Omnibus Series (GSE) from NCBI (National Center for Biotechnology Information) GEO (Gene Expression Omnibus). Through the analysis of the datasets in MASLD and IS patients, we found that abnormal lipid metabolism is a potential pathway for stroke induced by MASLD as a risk factor. Furthermore, we established a middle cerebral artery occlusion-reperfusion (MCAO/R) model in mice, measured atherosclerotic lesions in ApoE-deficient mice, and performed RNA-seq analysis to identify differentially expressed genes (DEGs) following IS. Our findings indicate that the DEGs are associated with lipid metabolism signaling pathways and inflammatory response pathways. The ApoE mice exhibited more severe IS injury. edaravone, a free radical scavenger clinically used for acute ischemic stroke treatment, was employed here to investigate whether its neuroprotective pathways intersect with lipid metabolism regulation. We found that treatment with edaravone rectified metabolic disorders and mitigated IS damage. Furthermore, we observed that the expression of the hub gene Fdft1 was upregulated in both the brain and liver post-IS injury and significantly reduced following edaravone treatment. These findings suggest that lipid regulation is a promising avenue for IS therapy, and Fdft1 may emerge as a critical target for modulating lipid metabolism in the aftermath of IS.

Keywords: Atherosclerosis; Disrupted lipid metabolism; Fdft1; Ischemic stroke; MASLD.

Fig. 1

Flowchart of the study

Fig. 2

Patients with IS and MASLD both exhibit significant abnormalities in lipid metabolism A Volcano plot representation of differential gene expression of MASLD in GSE80632. B Volcano plot representation of differential gene expression in integrated IS dataset. C KEGG enrichment analysis of differential genes in GSE80632. D KEGG enrichment analysis of differential genes in integrated IS dataset. E Differential activity of lipid metabolism pathways in MASLD vs. healthy controls. F Differential activity of lipid metabolism pathways in IS vs. healthy controls. Heatmap displaying GSVA enrichment scores (row-normalized) for individual lipid metabolism pathways across each sample. Stick plot displaying differential pathway activity analysis. Statistical comparison of GSVA scores between groups was performed for each lipid metabolism pathway using the limma package

Fig. 3

Screening and functional enrichment of shared lipid-related DEGs. A Venn diagram showing overlap of lipid-related genes and DEGs from both disease datasets. B Heatmap representation of log2(Foldchange) of shared DEGs in the two disease datasets. C GO enrichment (BP, CC, MF) for shared upregulated DEGs. D GO enrichment (BP, CC, MF) for shared downregulated DEGs. E Top KEGG pathways enriched in shared upregulated DEGs. F Top KEGG pathways enriched in shared downregulated DEGs

Fig. 4

Identify hub genes based on machine learning algorithms in the MASLD and IS datasets. A The heatmap illustrates the optimal tuning parameter lambda (lambda_lse and lambda_min) corresponding to different alpha values in the MASLD dataset. B Enet cross-validation to select optimal tuning parameter (λ) in the MASLD dataset. C Lasso cross-validation to select optimal tuning parameter (λ) in the MASLD dataset. D The intersection genes of Enet and Lasso results in the MASLD dataset. E ROC curves of intersection genes in the MASLD dataset. F The heatmap illustrates the optimal tuning parameter lambda (lambda_lse and lambda_min) corresponding to different alpha values in the integrated IS dataset. G Enet cross-validation to select optimal tuning parameter (λ) in the integrated IS dataset. H Lasso cross-validation to select optimal tuning parameter (λ) in the integrated IS dataset. I The intersection genes of Enet and Lasso result in the integrated IS dataset. J ROC curves of intersection genes in the integrated IS dataset

Fig. 5

Validation of shared hub genes in external datasets. A Venn diagram identifying two shared hub genes. B–I Differential expression of shared hub genes (Fdft1 and Sqle) across datasets: B GSE164760, C GSE130970, D Fdft1 in GSE126848, E Sqle in GSE126848, F GSE162694, G GSE213177, H GSE30655, and I GSE114652. J, K Cell-type-specific expression in GSE163752: J Fdft1 and K Sqle

Fig. 6

Observation of ischemic injury in metabolic disorder mice. A The levels of TG in WT and ApoE mice (n = 6). *P-value < 0.05. B The levels of CHO in WT and ApoE mice (n = 6). C The levels of LDL-C in WT and ApoE mice (n = 6). D The levels of HDL-C in WT and ApoE mice (n = 6). E The plaque deposition of mice in each group was detected by oil O staining (n = 4). F The inflammatory infiltration of the brain in each group was detected by HE staining (n = 4). G Neuronal damage in each group was detected by Nissl staining (n = 4)

Fig. 7

DEG analysis by RNA-seq in WT and ApoE mice. A Volcano plot for DEGs in the WT + I/R group compared to the WT group (n = 4). B KEGG pathway enrichment of DEGs in the WT + I/R group (n = 4). C Volcano plot for DEGs in the ApoE + I/R group compared to the ApoE group (n = 4). D KEGG pathway enrichment of DEGs in the ApoE + I/R group (n = 4)

Fig. 8

Observation of ischemic injury after edaravone intervention. A The levels of TG in each group (n = 6). B The levels of CHO in each group (n = 6). C The levels of LDL-C in each group (n = 6). D The levels of HDL-C in each group (n = 6). E The inflammatory infiltration of brain in each group was detected by HE staining (n = 4). F Neuronal damage in each group was detected by Nissl staining (n = 4)

Fig. 9

DEG analysis by RNA-seq treatment by edaravone. A Volcano plot for DEGs in the EDA + I/R group compared to the EDA group (n = 4). B KEGG pathway enrichment of DEGs in the EDA + I/R group (n = 4). C Volcano plot for DEGs in the EDA + I/R group compared to the WT + I/R group (n = 4). D KEGG pathway enrichment of DEGs in the EDA + I/R group compared to the WT + I/R group (n = 4)

Fig. 10

The expression of FDFT1 in each group of mice. A The protein expression of FDFT1 in the brain in each group (n = 4). B The protein expression of FDFT1 in the liver in each group (n = 4). C The protein expression of FDFT1 in the brain after treatment by edaravone (n = 4). D The protein expression of FDFT1 in the liver after treatment by edaravone (n = 4)