Alteration of E2F2 Expression in Governing Endothelial Cell Senescence

Endothelial cell senescence has a vital implication for vascular dysfunction, leading to age-related cardiovascular disease, especially hypertension and atherosclerosis. E2F transcription factor 2 (E2F2) plays a critical role in cell proliferation, differentiation, and DNA damage response. Up to date, no study has ever connected E2F2 to vascular endothelial cell senescence. Here, we demonstrate that E2F2 is involved in endothelial cellular senescence. We found that E2F2 expression is decreased during the replicative senescence of human umbilical vein endothelial cells (HUVECs) and the aortas of aged mice. The knockdown of E2F2 in young HUVECs induces premature senescence characterized by an increase in senescence-associated β-galactosidase (SA-β-gal) activity, a reduction in phosphorylated endothelial nitric oxide synthase (p-eNOS) and sirtuin 1 (SIRT1), and the upregulation of senescence-associated secretory phenotype (SASP) IL-6 and IL-8. The lack of E2F2 promoted cell cycle arrest, DNA damage, and cell proliferation inhibition. Conversely, E2F2 overexpression reversed the senescence phenotype and enhanced the cellular function in the senescent cells. Furthermore, E2F2 deficiency downregulated downstream target genes including CNNA2, CDK1, and FOXM1, and overexpression restored the expression of these genes. Our findings demonstrate that E2F2 plays an indispensable role in endothelial cell senescence.


Introduction
Vascular endothelial cells compose the inner layer of blood vessels, which is fundamental for homeostasis and is extremely vulnerable to hemodynamic stress [1,2]. During the aging process, endothelial cells undergo senescence, decreased proliferation, and a loss of vascular repair capability, thereby leading to a decline in endothelial-dependent vasomotor function [3]. The impairment of aging vessels is attributed to DNA damage, telomere attrition, and cellular senescence [4][5][6]. The phenotype of vascular aging is illustrated by enlarged vascular lumen, a thickened intima-media layer, decreased small vessel density, increased collagen, and decreased elastin deposition [7]. It is known that vascular aging is associated with an increased incidence of cardiovascular disease, especially hypertension and atherosclerosis, and an advanced age accelerates vascular endothelial dysfunction [8][9][10][11]. Endothelial senescence is considered as a significant mediator for vascular aging and age-related disease. The senescent cells exhibit common features including elevated SA-β-galactosidase (SA-β-gal) activity, the increased expression of

Cell Culture
Human umbilical vein endothelial cells (HUVECs) were extracted from an umbilical cord and cultured in Endothelial Cell Medium (ScienCell, San Diego, CA, USA). All cells were maintained at an incubator with 5% CO 2 at 37 • C.

Mice
Male C57BL/6 mice were acquired from the Model Animal Research Center of Nanchang University and kept on 12 h light/dark cycles with free access to food and water in ventilated cages. All experimental procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals from the Human Aging Research Institute of Nanchang University.
The Ct values were achieved based on the amplification curve. The relative gene expression was estimated by the 2 −∆∆Ct method.

SA-β-gal Staining
The detection of SA-β-gal activity was carried out by the Senescence β-Galactosidase Staining Kit (Beyotime, Shanghai, China). The HUVECs were fixed by a fixative solution at room temperature for 10 min. Then, the cells were incubated with an X-gal staining solution at 37 • C for 16 h. The percentage of blue-stained cells was calculated under a microscope.

Cell Cycle Analysis
Cell cycle analysis was performed with the Cell Cycle and Apoptosis Analysis kit (Beyotime, Shanghai, China). The HUVECs were harvested and washed once with PBS. Then, the cells were resuspended with 70% ethanol at 4 • C overnight. After the ethanol was removed, the cells were washed twice with PBS. The cells were incubated with a propidium iodide (PI) solution containing RNase for 30 min. The percentage of cells in each phase of the cell cycle was detected by FACSverse (BD Biosciences, San Jose, CA, USA).

EdU Cell Proliferation Assay
The proliferation of HUVECs was detected by the EdU-488 Cell Proliferation Detection Kit (Beyotime, Shanghai, China). The HUVECs were incubated with the EdU solution for 2 h in an incubator with 5% CO 2 at 37 • C. The cells were fixed and stained with Azide-fluor 488. The nuclear counterstaining was carried out by Hoechst33342. Photos were taken with ImageXpress Confocal HT.ai (Molecular Devices, San Jose, CA, USA).

ELISA
The concentration of interleukin 6 (IL6) and interleukin 8 (IL8) from the conditioned medium was measured by the homo IL6 and IL8 ELISA kit (Proteintech, Wuhan, Hubei, China). The medium was collected and centrifuged at 1000× g for 5 min to remove cell debris. After adding the stop solution, the optical density values were determined using SpectraMax i3x (Molecular Devices, San Jose, CA, USA) at 450 nm.

Statistical Analysis
The statistical analysis was executed using GraphPad PRISM version 8.4.2 (La Jolla, CA, USA). All data are displayed as the mean ± SD. One-way ANOVA with Dunnett's multiple comparison was applied for comparing three or more groups, and a two-tailed Student's t-test was applied for comparing two groups. Two-way ANOVA was used for multiple comparisons with two variables. A p value < 0.05 is statistically significant.

E2F2 Expression Is Decreased in the Senescent Endothelial Cells and the Aorta of Aged Mice
To investigate whether E2F2 expression changes with aging, we carried out the experiments using a replicative senescence model at different population doubling levels (PDL), as well as with aged mice. PDL12 was defined as young cells, and PDL20, 28, and 36 were defined as senescent cells [26]. We found that the E2F2 protein expression level was lower in the HUVECs at PDL20-36 than that of PDL12 ( Figure 1A). A decrease in the E2F2 level was also observed in the aorta of aged mice (15 months) compared to that of young mice (3 months) ( Figure 1B). We verified the HUVECs with CD31, which is a specific marker for endothelial cells [27]. Furthermore, the senescence marker P21 was increased in PDL20-36 cells in comparison with that in PDL12 cells ( Figure 1C). SA-β-gal activity, a biomarker for the identification of senescent cells [28], was elevated and characterized by more blue-stained cells at PDL20-36 ( Figure 1D). The senescence-related genes p-eNOS and SIRT1 showed lower levels in PDL20-36 HUVECs than in PDL12 cells ( Figure 1E,F). The cell cycle progression was suppressed at the G0/G1 phase in the senescent cells ( Figure 1G). All the data indicate that E2F2 expression is decreased during cellular senescence and aging.
E2F2 level was also observed in the aorta of aged mice (15 months) compared to that of young mice (3 months) ( Figure 1B). We verified the HUVECs with CD31, which is a specific marker for endothelial cells [27]. Furthermore, the senescence marker P21 was increased in PDL20-36 cells in comparison with that in PDL12 cells ( Figure 1C). SA-β-gal activity, a biomarker for the identification of senescent cells [28], was elevated and characterized by more blue-stained cells at PDL20-36 ( Figure 1D). The senescence-related genes p-eNOS and SIRT1 showed lower levels in PDL20-36 HUVECs than in PDL12 cells ( Figure 1E, F). The cell cycle progression was suppressed at the G0/G1 phase in the senescent cells ( Figure 1G). All the data indicate that E2F2 expression is decreased during cellular senescence and aging.

E2F2-Knockdown Induces Endothelial Cell Senescence
In light of these findings, we inspected whether a deficiency of E2F2 results in endothelial cell senescence. We knocked down the E2F2 gene with two sets of siRNA (siE2F2-1 and siE2F2-2) in young HUVECs at PDL12. SA-β-gal activity and SASP IL-6 and IL-8 from a conditioned medium were found to be significantly increased in E2F2-knockdown cells compared to those of the scramble (Figure 2A-C). The protein levels of p-eNOS and SIRT1 were decreased when knocking down E2F2 ( Figure 2D,E). However, there was no difference observed in P21 expression between cells with or without the treatment of siRNA ( Figure 2F). Western analysis verified a higher efficiency of E2F2 knockdown in HUVECs ( Figure 2G). These results suggest that a lack of E2F2 leads to endothelial cell senescence. thelial cell senescence. We knocked down the E2F2 gene with two sets of siRNA (siE2F2-1 and siE2F2-2) in young HUVECs at PDL12. SA-β-gal activity and SASP IL-6 and IL-8 from a conditioned medium were found to be significantly increased in E2F2-knockdown cells compared to those of the scramble (Figure 2A-C). The protein levels of p-eNOS and SIRT1 were decreased when knocking down E2F2 ( Figure 2D,E). However, there was no difference observed in P21 expression between cells with or without the treatment of siRNA ( Figure 2F). Western analysis verified a higher efficiency of E2F2 knockdown in HUVECs ( Figure 2G). These results suggest that a lack of E2F2 leads to endothelial cell senescence.

E2F2-Knockdown Impairs Endothelial Cellular Function
We next examined the function of HUVECs at PDL12 upon E2F2 deficiency. Flow cytometry analysis showed that the cell cycle was arrested at the G0/G1 phase in the cells transfected with siE2F2s as compared to the scramble ( Figure 3A). We further evaluated the DNA damage level by detecting the intensity of γH2AX, a DNA damage marker [29], in E2F2-knockdown HUVECs and found an increased γH2AX level of cells ( Figure 3B). In addition, the EdU incorporation assay showed that cell proliferation was inhibited by the knockdown of E2F2 ( Figure 3C). These experiments demonstrate that a lack of E2F2 promotes endothelial cellular dysfunction. transfected with siE2F2s as compared to the scramble ( Figure 3A). We further evaluated the DNA damage level by detecting the intensity of γH2AX, a DNA damage marker [29], in E2F2-knockdown HUVECs and found an increased γH2AX level of cells ( Figure 3B). In addition, the EdU incorporation assay showed that cell proliferation was inhibited by the knockdown of E2F2 ( Figure 3C). These experiments demonstrate that a lack of E2F2 promotes endothelial cellular dysfunction.

Overexpression of E2F2 Attenuates the Senescence Phenotype of Endothelial Cells
To verify the role of E2F2 in endothelial cell senescence, we overexpressed E2F2 by ADV-E2F2 in the senescent HUVECs (PDL36). The efficiency of overexpression was evaluated by Western analysis ( Figure 4A). The percentage of SA-β-gal positive cells was remarkably reduced in PDL36 cells transfected with ADV-E2F2 compared to ADV-GFP control cells ( Figure 4B). Likewise, lower expression levels of p-eNOS and SIRT1 were recovered in the senescent cells treated with ADV-E2F2 ( Figure 4C,D). Moreover, the IL6 level was decreased in the conditioned medium of senescent HUVECs with E2F2 overexpression ( Figure 4E). The results indicate that the overexpression of E2F2 reverses the senescence phenotype in endothelial cells. uated by Western analysis ( Figure 4A). The percentage of SA-β-gal positive cells was remarkably reduced in PDL36 cells transfected with ADV-E2F2 compared to ADV-GFP control cells ( Figure 4B). Likewise, lower expression levels of p-eNOS and SIRT1 were recovered in the senescent cells treated with ADV-E2F2 ( Figure 4C,D). Moreover, the IL6 level was decreased in the conditioned medium of senescent HUVECs with E2F2 overexpression ( Figure 4E). The results indicate that the overexpression of E2F2 reverses the senescence phenotype in endothelial cells.

E2F2 Overexpression Reverses Endothelial Dysfunction in Senescent Endothelial Cells
We further tested whether E2F2 overexpression improves cellular dysfunction in senescent HUVECs. By flow cytometry, we found that G0/G1 phase accumulation in the senescent cells was attenuated after overexpressing E2F2 ( Figure 5A). DNA damage detected by γH2AX immunocytochemistry was lessened in the senescent cells transfected with ADV-E2F2 compared to that with ADV-GFP ( Figure 5B). Similarly, the cell proliferation was restored in senescent HUVECs with the transfection of ADV-E2F2 ( Figure 5C). These findings suggest that the overexpression of E2F2 reinstates cellular function in senescent endothelial cells.

E2F2 Overexpression Reverses Endothelial Dysfunction in Senescent Endothelial Cells
We further tested whether E2F2 overexpression improves cellular dysfunction in senescent HUVECs. By flow cytometry, we found that G0/G1 phase accumulation in the senescent cells was attenuated after overexpressing E2F2 ( Figure 5A). DNA damage detected by γH2AX immunocytochemistry was lessened in the senescent cells transfected with ADV-E2F2 compared to that with ADV-GFP ( Figure 5B). Similarly, the cell proliferation was restored in senescent HUVECs with the transfection of ADV-E2F2 ( Figure 5C). These findings suggest that the overexpression of E2F2 reinstates cellular function in senescent endothelial cells.

Differentially Expressed Genes (DEGs) Are Identified in E2F2-knockdown and Senescent Endothelial Cells
Our data illustrated that E2F2 plays a part in endothelial senescence. Furthermore, we attempted to identify the molecular pathways downstream from E2F2. We analyzed two datasets retrieved from the publicly available GEO. One is GSE172094 (siRNAmediated E2F2 silencing in growth-stimulated LNCaP prostate cancer cells), and the other is GSE155680 (young and senescent HUVECs). Overall, 339 DEGs were identified as overlapping between E2F2-silencing LNCaP prostate cancer cells and the senescent HU-VECs ( Figure 6A). KEGG and GO enrichment analysis showed that these DEGs were clustered in various pathways including cellular senescence, cell cycle, and DNA repair ( Figure 6B), in which a total of 13 genes were screened out in the pathway of cellular senescence ( Figure 6C). Among them, 10 genes were found to interact with each other by network analysis ( Figure 6D). qPCR validated that the mRNA levels of these 10 genes were decreased in the replicative senescence model ( Figure 6E). After knocking down E2F2, CCNA2 (cyclin A), CDK1 (cyclin dependent kinase 1), and FOXM1 (forkhead box M1) were significantly downregulated in the PDL12 HUVECs ( Figure 6F). Notably, in the senescent cells, only these three genes repressed by E2F2 silencing were rescued by the overexpression of E2F2 ( Figure 6G). All the results confirmed that E2F2 modulates the expression of specific downstream genes related to cellular senescence and the cell cycle, suggesting the possible signaling of E2F2 in senescence.

Differentially Expressed Genes (DEGs) Are Identified in E2F2-knockdown and Senescent Endothelial Cells
Our data illustrated that E2F2 plays a part in endothelial senescence. Furthermore, we attempted to identify the molecular pathways downstream from E2F2. We analyzed two datasets retrieved from the publicly available GEO. One is GSE172094 (siRNA-mediated E2F2 silencing in growth-stimulated LNCaP prostate cancer cells), and the other is GSE155680 (young and senescent HUVECs). Overall, 339 DEGs were identified as overlapping between E2F2-silencing LNCaP prostate cancer cells and the senescent HUVECs ( Figure 6A). KEGG and GO enrichment analysis showed that these DEGs were clustered in various pathways including cellular senescence, cell cycle, and DNA repair ( Figure 6B), in which a total of 13 genes were screened out in the pathway of cellular senescence CCNA2 (cyclin A), CDK1 (cyclin dependent kinase 1), and FOXM1 (forkhead box M1) were significantly downregulated in the PDL12 HUVECs ( Figure 6F). Notably, in the senescent cells, only these three genes repressed by E2F2 silencing were rescued by the overexpression of E2F2 ( Figure 6G). All the results confirmed that E2F2 modulates the expression of specific downstream genes related to cellular senescence and the cell cycle, suggesting the possible signaling of E2F2 in senescence.

Discussion
In the present study, we found that E2F2 expression is decreased during the replicative senescence of endothelial cells as well as the natural aging of aortas. The knockdown of E2F2 in endothelial cells induces premature senescence and cellular dysfunction in young HUVECs, while E2F2 overexpression rescues this phenotype and enhances the cell function in the senescent cells. Additionally, E2F2 deficiency downregulated target genes, and overexpression promotes the expression of these genes. Our findings demonstrate that E2F2 plays an indispensable role in endothelial cell senescence.
The development of cellular senescence is crucial for endothelial dysfunction related to vascular aging. Endothelial cells experience replicative senescence, which is characterized by a declined proliferation capacity that can be assessed by PDL [7]. We introduced an endothelial replicative senescence model as an in vitro setting into our study, and this model is verified by the classic senescence indicators such as SA-β-gal, P21, cell cycle progression, and the senescence signatures eNOS and SIRT1 [12,30,31] The fundamental feature of senescent cells is the persistent arrest of the cell cycle, leading to diminished proliferation [32,33]. Upon knocking down E2F2 in young HUVECs, we observed similar phenotypes to the replicative senescence, including elevated SA-β-gal activity, increased IL6 and IL8 levels, deceased p-eNOS and SIRT1 expression, and cell cycle arrest. Interestingly, the senescence marker P21 was not affected in E2F2-deficient cells. The available evidence shows that E2F2 is controlled by cyclin-dependent kinases and their inhibitors. However, E2F2 induces the expression of cell cycle protein such as CCNA2 but does not increase the P21 level [34,35]. The previous report indicates that P21 expression may not be necessary for human fibroblasts senescence [36]. Further study reveals that the senescence-associated phenotype and cell cycle arrest are not definitely tethered in senescent P21-deficeint fibroblasts; however, a low level of P21 is sufficient to facilitate the senescent cell cycle arrest [37]. It is found that neural stem cells treated with irradiation exhibit the senescence phenotype without the significant upregulation of P21 [38]. The senescence induced by interleukin-32θ shows a decreased level of E2F2 and unaltered P21 expression [39]. Thus, endothelial cell senescence induced by E2F2 deficiency may be irrespective of P21 expression. This implies a novel mechanism of E2F2 in modulating senescence-possibly P21, as upstream signaling impacting E2F2 [40]. Moreover, our study shows that the knockdown of E2F2 suppresses the cell cycle at the G0/G1 phase in young HUVECs, which is in agreement with a G1 arrest in senescent endothelial cells, thereby stopping the entrance into the S phase [28,35]. The overexpression of E2F2 in the senescent cells recaptures the function.
The other hallmark of cellular senescence is excessive SASP secretion. Upon senescent cells buildup, they release plenty of SASP proteins to induce neighboring normal cell senescence and to promote cellular function deterioration, which exacerbates tissue aging. On the other hand, SASP can eliminate the senescent cells through the immune system and repair damaged tissues [41,42]. As a result, the SASP proinflammatory phenotype is presented in arteries with an advanced age [43]. The most consistently age-related biomarkers are IL6 and IL-8 [10,44]. Our data showed that IL6 and IL8 levels were increased in E2F2-deficient young cells and reversed in E2F2-overexpressed senescent cells. SASP production is also stimulated by a persistent DNA damage response [45]. Accumulative unrepaired DNA damage contributes to vascular aging [46]. DNA damage response and genomic instability are recognized as characteristics of cellular senescence and aging [4,47]. E2F2 has been reported to prevent DNA damage and maintain genomic stability in a number of cell types, including neuronal cells, embryonic fibroblasts, and T lymphocytes [48,49]. We detected that a lack of E2F2 resulted in more DNA damage in young HUVECs; in contrast, the overexpression of E2F2 counteracts increased DNA damage in the senescent cells. Such evidence highlights a role of E2F2 in endothelial cell senescence.
It is well-documented that senescence changes gene expression in vascular endothelial cells. In this study, bioinformatics analysis identified three downstream target genes of E2F2 including CNNA2, CDK1, and FOXM1. We found that CNNA2, a member of the cyclin family, is expressed in young HUVECs and decreases with an increasing PDL, which is consistent with previous reports [35,50]. As in the case of replicative senescence, CNNA2 mRNA expression is inhibited in E2F2-knockdown cells. Conversely, the overexpression of E2F2 enhances the CNNA2 level in the senescent cells. CDK1 belongs to the protein kinase superfamily, which presents a low expression and activity in the T lymphocytes of aged subjects [51]. There is evidence showing that CDK1 deficiency promotes cellular senescence [52]. The study also indicates that the knockdown of CCNA2 impacts CDK1 [52]. FOXM1 is a transcription activator which is involved in controlling cellular aging and improving the lifespan [53]. Hence, it is possible that E2F2 may drive vascular endothelial cell senescence from interacting with these senescence-associated target genes.
This study has some limitations. The cellular senescence induced by E2F2 seems to not be interrelated with alterations in the expression of p21. More studies are needed to address the underlying mechanism and define how E2F2 mediates endothelial senescence. The role of E2F2 in senescence at the cellular level requires validation in aged animal models and in aged blood vessels from humans. Informed Consent Statement: This study was approved by the Ethics Committee of Nanchang University (Protocol No. HARI-SC-0061, approved on 18 September 2016), and informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The publicly available RNAseq datasets analyzed in this study are available at NCBI's Gene Expression Omnibus (GSE172094 and GSE155680).

Conflicts of Interest:
The authors have no conflicts of interest to declare.