Downregulation of NAGLU in VEC Increases Abnormal Accumulation of Lysosomes and Represents a Predictive Biomarker in Early Atherosclerosis

Abstract

Cardiovascular diseases (CVDs), predominantly caused by atherosclerosis (AS), are the leading cause of mortality worldwide. Although a great number of previous studies have attempted to reveal the molecular mechanism of AS, the underlying mechanism has not been fully elucidated. The aberrant expression profiling of vascular endothelial cells (VECs) gene in early atherosclerosis (EAS) was analyzed according to the dataset (GSE132651) downloaded from the Gene Expression Omnibus (GEO) database. We primarily performed functional annotation analysis on the downregulated genes (DRGs). We further identified that α-N-acetylglucosaminidase (NAGLU), one of the DRGs, played a critical role in the progression of EAS. NAGLU is a key enzyme for the degradation of heparan sulfate (HS), and its deficiency could cause lysosomal accumulation and lead to dysfunctions of VECs. We found that siRNA knockdown of NAGLU in human umbilical vein endothelial cell (HUVEC) aggravated the abnormal accumulation of lysosomes and HS. In addition, the expression of NAGLU was reduced in the EAS model constructed by ApoE ^-/- mice. Furthermore, we also showed that heparin-binding EGF-like growth factor (HB-EGF) protein was upregulated while NAGLU knockdown in HUVEC could specifically bind to vascular endothelial growth factor receptor 2 (VEGFR2) and promote its phosphorylation, ultimately activating the phosphorylation levels of extracellular signal-regulated kinases (ERKs). However, the application of selective VEGFR2 and ERKs inhibitors, SU5614 and PD98059, respectively, could reverse the abnormal lysosomal storage caused by NAGLU knockdown. These results indicated that downregulation of NAGLU in HUVEC increases the abnormal accumulation of lysosomes and may be a potential biomarker for the diagnosis of EAS.

Keywords: ERK; NAGLU; VEGFR2; bioinformatics analysis; early atherosclerosis; lysosome; vascular endothelial cell.

FIGURE 1

Normalization and repeatability analysis of the dataset. (A) RMA algorithm normalization of samples from the GSE132651 dataset. Left: data before RMA; right: data after RMA. (B) Pearson’s correlation analysis of NG and ANG samples from the GSE132651 dataset. The color reflects the intensity of the correlation, where the darker the color, the stronger the correlation intensity. (C) PCA of samples from the GSE132651 dataset. Each point represents a sample in the figure, and the distance between the two samples represents the difference in gene expression patterns of the two samples. (D) The volcano plot illustrates the differences between NG and ANG samples after analysis of the GSE132651 dataset using R.

FIGURE 2

Construction of co-expression modules by WGCNA. (A) The relatively balanced scale independence and mean connectivity in Scale-free topology analysis. The power value was 6, which was the lowest power for the scale-free topology fit index of 0.9. (B) Construction of co-expression modules by the WGCNA package through R. Genes are represented by branches, and co-expression modules are shown in different colors below in figure. (C) Network heatmap of the interaction relationship of different modules. (D) Eigengene adjacency heatmap plot of different modules. (E) Heatmap of the correlation between module eigengenes and the disease status of EAS. The red module was the most negatively correlated with status. NG: normal endothelial function groups; ANG: abnormal endothelial function groups.

FIGURE 3

Aberrant gene expression profiling and functional annotation enrichment analyses. (A) Heatmap of aberrant gene expression profiling. (B) GO (including BP, CC, and MF) and KEGG analyses of aberrant gene expression profiling. (C) GO BP enrichment of DRGs using R. (D) KEGG pathways enrichment of DRGs using R. (E) Construction of the PPI network of DRGs via Cytoscape software. (F) Multiple GSEA of varied expression of NAGLU using GO BP analysis (c5). (G) Multiple GSEA of varied expression of NAGLU using KEGG analysis (c2).

FIGURE 4

NAGLU reduction in the EAS animal model and NAGLU knockdown in HUVEC result in defective lysosomal storage. (A) The mRNA expression levels of NAGLU in HUVEC after transfection by qRT-PCR. si-NAGLU-8 was used for subsequent research. (B,C) The protein expression levels of NAGLU in HUVEC after transfection with si-NAGLU and si-NC as measured by Western blotting; TUBB was used as a control. (D) Representative images of lysosomes labeled with the LysoTracker probe in HUVEC si-NAGLU and HUVEC si-NC. Scale bars: 10 μm. (E) Representative images of HS labeled with lectin probe in HUVEC si-NAGLU and HUVEC si-NC. Scale bar: 7.5 μm. (F) Representative images of ApoE ^−/− mice aortic arches stained with DAPI, NAGLU, and CD31. NCD: normal chow diet; HFD: high-fat diet. *p < 0.05, **p < 0.01, Student’s t-test.

FIGURE 5

NAGLU knockdown causes aberrantly lysosomal accumulation in HUVEC targeting VEGFR2/ERKs and promotes EAS. (A) The protein expression levels of HB-EGF in HUVEC after transfection with si-NAGLU and si-NC as measured by Western blotting. TUBB was used as a control. (B) VEGFR2 and ERKs phosphorylation levels in HUVEC si-NAGLU and HUVEC si-NC were measured by Western blot. To monitor the equal loading of protein in the gel lanes, the upper blot was stripped and tested using anti-VEGFR2 and anti-ERKs antibodies, respectively. (C) VEGFR2 inhibition reduces ERK1/2 phosphorylation levels in HUVEC si-NAGLU. HUVEC si-NAGLU and HUVEC si-NC were both treated or untreated with 30 μM VEGFR2 inhibitor SU5614 for 12 h and then measured by Western blot. (D) VEGFR2 inhibition reduces aberrantly lysosomal accumulation in HUVEC si-NAGLU. Representative images of lysosomes labeled by the LysoTracker probe in HUVEC si-NAGLU and HUVEC si-NC, both treated with 30 μM SU5614 for 12 h. Scale bars: 10 μm. (E) MAPK/ERKs inhibition reduces aberrant lysosomal accumulation in HUVEC si-NAGLU. Representative images of the lysosomes labeled by the LysoTracker probe in HUVEC si-NAGLU and HUVEC si-NC, both treated or untreated with 50 μM PD98059 for 24 h. Scale bars: 10 μm.