Transcriptome analysis-identified long noncoding RNA CRNDE in maintaining endothelial cell proliferation, migration, and tube formation

Abstract

Obesity is a leading risk factor for type-2 diabetes. Diabetes often leads to the dysregulation of angiogenesis, although the mechanism is not fully understood. Previously, long noncoding RNAs (lncRNAs) have been found to modulate angiogenesis. In this study, we asked how the expression levels of lncRNAs change in endothelial cells in response to excessive palmitic acid treatment, an obesity-like condition. Bioinformatics analysis revealed that 305 protein-coding transcripts were upregulated and 70 were downregulated, while 64 lncRNAs were upregulated and 46 were downregulated. Gene ontology and pathway analysis identified endoplasmic reticulum stress, HIF-1 signaling, and Toll-like receptor signaling as enriched after palmitic acid treatment. Moreover, we newly report enrichment of AGE-RAGE signaling pathway in diabetic complications, IL-17 signaling, and cysteine and methionine metabolism by palmitic acid. One lncRNA, Colorectal Neoplasia Differentially Expressed (CRNDE), was selected for further investigation. Palmitic acid induces CRNDE expression by 1.9-fold. We observed that CRNDE knockdown decreases endothelial cell proliferation, migration, and capillary tube formation. These decreases are synergistic under palmitic acid stress. These data demonstrated that lncRNA CRNDE is a regulator of endothelial cell proliferation, migration, and tube formation in response to palmitic acid, and a potential target for therapies treating the complications of obesity-induced diabetes.

Figure 1

Distribution of RNA-seq data. (A) Analysis workflow for RNA-seq data. (B) Percentage of the different types of RNA identified by RNA-seq. (C) Percentages of the different types of lncRNA identified in (B).

Figure 2

Differentially expressed mRNAs. (A) Expression plot of identified mRNAs. Expression is determined by average FPKM. Blue are those mRNAs significantly differentially expressed. (B) Heatmap of the top 20 upregulated and top 20 downregulated mRNAs. Data was plotted as z-score of the log2 fold change. (C) Highly enriched KEGG pathways from differentially expressed genes. Protein coding genes are identified by gene symbols.

Figure 3

Differentially expressed lncRNAs. (A) Expression plot of identified lncRNAs. Expression is determined by average FPKM. Blue and red are those lncRNAs significantly differentially expressed; red is CRNDE. (B) Heatmap of the top 20 lncRNAs by expression that are differentially expressed. Data was plotted and ranked as z-score of the log2 fold change. (C) Effect of palmitic acid on gene expression. HUVECs were treated with palmitic acid or vehicle control for 12 hours. The expression of CRNDE, HMOX1, and VEGF-A was analyzed by RT-qPCR, *p < 0.05, **p < 0.001.

Figure 4

CRNDE knockdown inhibits endothelial cell proliferation and tube formation. (A) CRNDE knockdown reduces wound closure in HUVECs. CRNDE was treated with CRNDE shRNA (shCRNDE) or non-targeting control shRNA (shNT) for 24 h before wounding. Images were taken from fixed cells 12 h after wounding. Representative images of 3 independent experiments were shown. Scale bars are 500 µm. (B) Quantification of wound closure, *p = 0.027. (C) CRNDE knockdown reduces tube formation. HUVECs were treated as in a before being seeded into Matrigel-coated wells. Images were taken from fixed wells after 6 hours. Representative images of 3 independent experiments were shown. (D) Quantification of tube length in millimeters *p = 3.2 * 10⁻⁶. (E) Quantification of number of segments (*p = 9.9 * 10⁻⁴) and number of nodes (*p = 6.5 * 10⁻⁴).

Figure 5

CRNDE knockdown inhibits the progression of cell cycle. (A) Histogram of DNA content in HUVECs after CRNDE knockdown. HUVECs were treated with shNT or shCRNDE and then stained with propidium iodide. DNA content was measured by flow cytometry at 488 nm. Representative histograms of three independent experiments were shown. (B) Quantification of cell cycle phase in three independent experiments, *p < 0.05. (C) Western blot analysis of p21 expression and quantification of three independent experiments (*p = 1.4 * 10⁻³).

Figure 6

CRNDE knockdown potentiates the inhibitory effects of palmitic acid on endothelial cell migration and tube formation. (A) CRNDE knockdown potentiates the inhibitory effects of palmitic acid on wound closure. HUVECs were treated with shNT or shCRNDE for 24 h and then palmitic acid or vehicle control for 12 h before wounding. Images were taken from fixed cells 12 h after wounding. Representative images of 3 independent experiments were shown. Scale bars are 500 µm. (B) Quantification of (A) (*p < 0.001). (C) CRNDE knockdown potentiates the inhibitory effects of palmitic acid on tube formation. HUVECs were treated as for wound healing before being seeded into Matrigel-coated wells. Images were taken from fixed wells after 6 h. Representative images of 3 independent experiments were shown. (D) Quantification of tube formation in (C). Tube length in millimeters, the number of segments, and the number of nodes (*p < 0.05, **p < 0.01, ***p < 1 * 10⁻⁴). (E) Quantification of tube formation in (C). Number of branches (*p < 0.05, **p < 0.001, ***p < 1 * 10⁻⁴).