C1q/TNF-Related Protein-9 Attenuates Palmitic Acid-Induced Endothelial Cell Senescence via Increasing Autophagy

Background: Autophagy is an important process in the pathogenesis of atherosclerosis. C1q/tumor necrosis factor-related protein 9 (CTRP9) is the closest adiponectin paralog. CTRP9 has purported anti-aging and anti-atherogenic effects, but its roles in autophagy and endothelial senescence are unknown. The aim of this study was to evaluate whether CTRP9 prevents palmitic acid (PA)-induced endothelial senescence by promoting autophagy. Methods: After no treatment or pre-treatment of human umbilical vein endothelial cells with CTRP9 prior to PA treatment, the level of senescence was measured by senescence associated acidic β-galactosidase staining and the level of hyperphosphorylated pRB protein. Autophagy was evaluated by LC3 conversion and the level of p62/SQSTM1, a protein degraded during autophagy. Autophagosome–lysosome fusion was detected by uorescence microscopy. Results: Pre-treatment with CTRP9 attenuated PA-induced endothelial senescence. CTRP9 increased the conversion of LC3-I to LC3-II, and decreased the level of p62 in time- and dose-dependent manners. Although both CTRP9 and PA treatment increased the LC3 conversion, treatment of PA increased p62 and decreased the fusion of autophagosomes and lysosomes, which represented decreased autophagic ux. However, pre-treatment with CTRP9 recovered the autophagic ux inhibited by PA. AMP-activated kinase (AMPK) activation was involved in LC3 conversion and decreased p62 induced by CTRP9. Conclusion: CTRP9 inhibits PA-induced endothelial senescence by recovering autophagy and autophagic ux through AMPK activation.

anti-atherogenic effects [10,11]. In mouse studies, CTRP9 levels were associated with coronary atherosclerosis disease (CAD) [12], and treatment of CTRP9 inhibited neointimal formation, which is one of the processes that transform plaques into a vulnerable form that is prone to rupture [9]. In other studies, CTRP9 treatment decreased vascular in ammation [11] and improved endothelial dysfunction [13]. However, the possible effect of CTRP9 on cellular senescence has not been fully investigated.
Autophagy is a regenerative survival process that is involved in clearing long-lived or injured proteins and organelles. Autophagy also provides cells with alternate nutrients produced by the recycling of cellular proteins [14]. Decreased autophagic activity with aging and accumulation of intracellular damage has been reported to determine the level of cellular senescence [14,15]. In this context, declining autophagic function causes the senescence of vascular endothelial cells, which has been strongly associated with the pathogenesis of cardiovascular disease (CVD) [16].
Presently, we examined whether CTRP9 inhibited endothelial senescence induced by lipotoxicity, such as PA. Considering that autophagy is an important preventive process of cellular senescence, we hypothesized that autophagy and autophagic ux mediate this inhibitory effect of CTRP9 on endothelial senescence. In addition, because CTRP9 has been reported to attenuate endothelial in ammation mediated AMP-activated kinase (AMPK) signaling [11], we examined AMPK activation might mediate the effect of CTRP9 on autophagy induced inhibitory effects of cellular senescence.

Cell culture and treatment
Human umbilical vein endothelial cells (HUVECs) obtained from Lonza Inc. (C2517A, Walkersville, MD) were cultured in endothelial basal medium (EBM-2, CC-3162, Lonza) supplemented with 2% fetal bovine serum (FBS) and various growth factors required for the growth of endothelial cells at 37℃ in a humidi ed incubator in an atmosphere of 5% CO 2 . Cells were used at six or less passages in all experiments. The cells were transferred to a medium containing 2% FBS and incubated for 24 hours in a medium containing 500 µM PA (P0500, Sigma-Aldrich, St. Louis, MO). In all experiments, 2% bovine serum albumin was used as control.
CTRP9 (00081-04-100, Aviscera Bioscience Inc., Santa Clara, CA) and phosphate buffered saline was used as the vehicle. Cells were transformed to a medium containing 2% FBS and incubated in a medium containing various concentrations of CTRP9 for the indicated times before treatment with PA. To assess autophagic ux, HUVECs were treated with 10 nmol/L of ba lomycin A1, an inhibitor of vacuolar H + adenosine triphosphatase (B1793, Sigma-Aldrich) Compound C (AMP-activated protein kinase inhibitor, P5499, Sigma-Aldrich) was used as pre-treatment (10 µmol/L) to evaluate the involvement of AMPK signaling pathway in autophagic induction by CTRP9.
Real-time PCR analysis cDNAs were synthesized using the ReverTra Ace qPCR RT kit (FSQ-101, Toyobo, Osaka, Japan). Real-time analysis was conducted on an ABI 7500 Fast RT-PCR system with the Fast SYBR® Green Master Mix (4385612, Applied Biosystems, Foster City, CA). Each sample was assayed in duplicate in a 20 µL reaction volume containing 1 µL cDNA (corresponding to 100 ng of total RNA input), 10 µL of 2 × SYBR Green Master Mix (4309155, Applied Biosystems), and 1 µL of forward and reverse primers (10 pmol/µL for each). Negative controls (no template or RNA) were included to ensure the absence of contamination. Ampli cation of 18S rRNA was used as the internal control. The ratio between the expression levels of the target gene and 18S rRNA was calculated using a relative quanti cation method (ΔΔ cycle value (Ct) method) as described previously (User Bulletin No. 2, Applied Biosystems). In brief, the ampli cation plot is a plot of uorescence versus PCR cycle number. The threshold Ct is the fractional PCR cycle number at which the uorescent signal reaches the detection threshold. Therefore, the input cDNA copy number and Ct are inversely related. Data were analyzed using the Sequence Detector System software version 2.

Immuno uorescence
HUVECs were transfected with a green uorescent protein fused to the N-terminus of LC3 (GFP-LC3) reporter construct using Lipofectamine 2000. After transfection for 24 hours, cells were incubated with 100 µmol/L LysoTracker Red (L-7528, Invitrogen) for one hour. Coverslips were adhered to glass slides with uorescence Mounting Medium (S3023, Dako, Carpinteria, CA). Colocalization of autophagy marker GFP-LC3 and LysoTracker was observed by laser scanning confocal microscopy using an LSM 780 microscope (Carl Zeiss, Oberkochen, Germany). Images were analyzed using Zen software (v2012, Carl Zeiss). Pearson's coe cient was used to measure the correlation between the signals from the two different marker proteins. Mander's coe cient was used to calculate the coe cient of overlap as previously described [18].

Statistical analyses
All data are presented as the mean ± SEM (standard error of the mean). Student's t-test was used to compare two groups. One-way analysis of variance (ANOVA) followed by a post-hoc analysis using the Tukey's multiple comparison test were used to compare multiple groups. P-values < 0.05 were regarded as statistically signi cant. All experiments were conducted at least ve times. All statistical analyses were performed using SPSS 21.0 for Windows (IBM SPSS, Chicago, IL).

CTRP9 attenuates PA-induced endothelial cell senescence
HUVECs were treated with 500 µM of PA for 24 hours to examine the effects of PA on endothelial senescence. The number of SA-β-Gal-positive cells was increased by treatment with PA (Fig. 1A, right Page 6/26 upper). Treatment only with CTRP9 did not change the frequency of SA-β-Gal-positive cells compared to controls (Fig. 1A, left lower). However, when HUVECs were pre-treated with 3 µg/mL of CTRP9 for 1 hour before PA treatment, the number of SA-β-Gal-positive cells was signi cantly decreased compared to the cells treated with PA alone (Fig. 1A, right lower). These results indicated that CTRP9 treatment can decrease PA-induced endothelial senescence in HUVECs. In addition, as senescence arrest in vascular endothelial cells are related to reduced levels of the hyperphosphorylated form of the retinoblastoma gene product (pRB) [18], the hyperphosphorylated pRB (ppRB) protein level was measured as an indicator of senescence. After PA treatment, the expression of ppRB decreased. However, pre-treatment with CTRP9 considerably increased the level of ppRB (Fig. 1B).
Inhibition of autophagy alleviates preventive effect of CTRP9 on PA-induced endothelial senescence Next, to assess whether autophagy was involved in the inhibition of PA-induced endothelial senescence by CTRP9, HUVECs were transfected with LC3 siRNA or control siRNA. The decreased expression SA-β-Gal positive cells, demonstrating the preventive effect of CTRP9 against PA-induced endothelial senescence, was attenuated in LC3 siRNA-transfected cells ( Fig. 2A). Similarly, the increased expression of ppRb by CTRP9 was attenuated in LC3 siRNA-transfected cells (Fig. 2B). Collectively, these results suggested that autophagy might be involved in the CTRP9's anti-senescent effect in vascular endothelial cells.

CTRP9 promotes autophagy and autophagic ux in HUVECs
To con rm the effects of CTRP9 on autophagy and autophagic ux, we measured protein expression of LC3-I/ LC3-II and p62/SQSTM1. The conversion of LC3-I to LC3-II, re ecting autophagosome formation [19], was estimated to assess whether CTRP9 increased the level of autophagy in time-and dosedependent manners. The time-and dose-dependent protein level of p62, an indicator of autophagic ux [20], was also estimated to determine the effect of CTRP9 on the level of autophagic ux. When HUVECs were treated with 3 µg/mL of CTRP9 for 4-24 hours, conversion of LC3-I to LC3-II increased and p62 expressions decreased over time (Fig. 3A). Likewise, CTRP9 increased LC3 conversion and decreased p62 expression in a dose-dependent manner (Fig. 3B). These ndings indicated that CTRP9 increased autophagy as well as autophagic re ux.
Additionally, we measured the mRNA and protein levels of ATGs involved in autophagosome formation. The mRNA expressions and protein levels of ATG levels (ATG7 and ATG12) were measured by real-time PCR analysis and western blot analysis, respectively. Both mRNA and protein levels of ATG7 and ATG12 were increased by CTRP9 treatment in time- (Fig. 4A) and dose-dependent manners (Fig. 4B).
CTRP9 recovers autophagy and autophagic ux inhibited by PA Based on the ndings that CTRP9 promoted autophagy and its ux in vascular endothelial cells, we next evaluated whether CTRP9 affected PA-induced changes in autophagic process in HUVECs. Both CTRP9 and PA treatment increased the LC3 conversion, but PA treatment increased p62 level, indicating the reduction of autophagic ux (Fig. 5A). Distinct from PA treatment, CTRP9 prior to PA treatment decreased p62 level, suggesting that CTRP9 increased autophagy and also restored autophagic ux. Furthermore, the autophagic ux was examined by the treatment with ba lomycin A1, a pharmacological inhibitor of the fusion between autophagosomes and lysosomes [21]. After treatment with ba lomycin A1, CTRP9 treatment further increased LC3 conversion compared with cells treated with PA alone (Fig. 5B).
To verify the fusion of the autophagosome and the lysosome, HUVECs were transfected with GFP-LC3 and LysoTracker Red and visualized by uorescence microscopy. The fusion between autophagosomes and lysosomes was shown by the numbers of colocalized yellow foci, which were signi cantly increased in the CTRP9 treatment group compared with the PA group (Fig. 6). The colocalization coe cients also demonstrated an increasing trend in the CTRP9 treatment group in the same manner (Fig. 6).

CTRP9 promotes autophagy and autophagic ux via AMPK pathways
The mechanisms of the autophagic process have been intensively studied. Several protein kinases are considered to be involved in this process [17]. Among these, the activation of AMPK has been reported to mediate the initiation of the autophagy process [22]. Furthermore, considering the important role of AMPK activation by CTRP9 in its cellular effect [11,[23][24][25], we treated HUVECs with compound C, a speci c AMPK inhibitor.
After inhibiting AMPK activity with a compound C, the CTRP9-dependent increase in LC3 conversion and decrease in p62 expression were attenuated (Fig. 7A). In addition, HUVECs were pre-treated with the AMPK inhibitor and colocalized with GFP-LC3 and LysoTracker Red to demonstrate the role of AMPK activation on the fusion of the autophagosome and the lysosome. The inhibition of AMPK decreased the formation of autophagolysosomes induced by CTRP9 (Fig. 7B). Increased colocalization coe cients induced by CTRP9 treatment were also abolished after the treatment of AMPK inhibitor (Fig. 7B).
Inhibition of AMPK attenuates preventive effects of CTRP9 on PA-induced endothelial cell senescence Finally, we tested whether the inhibitory effect of CTRP9 on PA-induced endothelial senescence required the activation of AMPK. After HUVECs were pre-treated with the AMPK inhibitor (compound C), the number of SA-β-Gal positive cells was decreased. The results showed that the preventive effect of CTRP9 disappeared (Fig. 8A). Similarly, the increased expression of ppRB by CTRP9 treatment was also attenuated after AMPK inhibition (Fig. 8B).

Discussion
We rst demonstrated that CTRP9 inhibited endothelial senescence induced by the PA FFA by promoting autophagy and autophagic ux. Activation of AMPK pathway was responsible for anti-senescent effect of CTRP9 due to the stimulation of autophagy and autophagic ux. CTRP9 has been suggested to act as regulator of vascular function, with bene cial effects on atherosclerosis and CVD [10,26]. In several previous animal studies, CTRP9 contributed to an advantageous vascular effect, especially concerning atherosclerosis [27][28][29][30]. CTRP9 was demonstrated to mediate endothelium-dependent vasorelaxation [27] and suppress vascular smooth muscle cell proliferation and neointimal formation [28]. Furthermore, elevated levels of CTRP9 improved the vascular plaque stability and alleviated the development of atherosclerosis in ApoE knockout mice by decreasing pro-in ammatory cytokines in macrophages [29,30]. The current ndings are consistent with previous results [27][28][29][30] in terms of the protective effects of CTRP9 in the pathogenesis of atherosclerosis.
Adipokines are bioactive substances secreted by adipose tissue. These adipokines include adiponectin, leptin, omentin, and CTRPs [10]. CTRP9 has attracted a lot of attention since it was discovered in 2009 [31]. Accumulating evidence has indicated the protective effects of CTRP9 against atherosclerosis through multiple mechanisms including inhibition of the in ammatory response [11,29,30,32,33] and amelioration of endothelial dysfunction [10,32,34]. The ndings implicate the CTRP9 adipokine as a promising target to prevent and treat atherosclerosis-associated disease. CTRP9 has also been reported to have bene cial effects on glucose metabolism, cellular survival, oxidative stress regulation, and inhibition of endoplasmic reticulum stress, all of which are associated with the aging mechanism of endothelial cells [35]. A previous study on the effect of CTRP9 on cellular senescence demonstrated that CTRP9 treatment prevented the oxidation reaction associated with aging in mesenchymal stem cells [23]. Although the anti-aging role of CTRP9 has been of interest, the protective role of CTRP9 in FFA-induced endothelial senescence had not been fully investigated. We presently con rmed the preventive effect of CTRP9 on this process by checking the level of the senescence mediator ppRB (Fig. 1B), as well as by SAβ-GAL staining (Fig. 1A). Endothelial senescence is believed to be associated with atherosclerosis [1]. Thus, our current ndings also support the anti-atherogenic effects of CTRP9.
Autophagy is an intracellular self-digestion mechanism that plays an important role in the degradation of cellular debris and exhausted organelles [14]. The process of autophagy is believed to mediate anti-aging effects [36]. Recent evidence has shown that the autophagic process is crucial in the pathogenesis of atherosclerosis [37]. However, whether autophagy reduces or promotes atherosclerosis is unclear. Autophagy maintains atherosclerotic plaques against oxidative stress by degrading damaged components and contributing to cellular retrieval [38]. Thereby, favorable autophagy can prohibit the apoptosis of macrophages and stabilize the atherosclerotic lesion [38]. In addition, the important role of autophagy in the control of apolipoprotein B (apoB) secretion from the liver was recently described. Accumulation of apoB in vessel walls is the inevitable initiating event in the progression of atherosclerosis [39]. Moreover, by diminishing intracellular damage, autophagy has been recognized as having an anti-aging function and has been shown to have protective effect in vascular endothelial senescence [40,41]. However, excess activated autophagy or impaired autophagy may lead apoptotic death of vascular smooth muscle cells, causing plaque instability [42].
Based on the previous results on the association of both CTRP9 and autophagy with cellular senescence and atherosclerosis [24,30], we assessed whether autophagy allowed CTRP9 to prevent endothelial senescence. We demonstrated that CTRP9 reversed the suppression of autophagy that had resulted from PA treatment and recovered autophagic ux (Fig. 5, 6). The inhibition of autophagy by transfection of LC3 siRNA supported an effect of autophagy on the anti-aging role of CTRP9 in vascular endothelial cells (Fig. 2). Although the effect of autophagy in atherosclerosis was not fully investigated, our results suggest another crucial anti-atherogenic mechanism of CTRP9 through its anti-aging functions via autophagy in vascular endothelial cells.
Endothelial senescence attenuates the abilities of vascular repair, regeneration, and nitric oxide (NO) production [1]. In endothelial cells, senescence increases oxidative stress and expression of proin ammatory cytokines, mainly tumor necrosis factor-alpha [18] and nuclear factor-kappa B [43]. In addition, endothelial aging is induced by various factors including impaired mitochondrial function [44], elevated serum glucose [4], genetic factors [5], and lipotoxicity [6]. Among these factors that induce endothelial senescence, PA can cause lipotoxicity when present in excess in non-fat cells [18]. Elevated concentrations of FFA impair endothelial function through the increase in IKKβ activity, which is the main signaling pathway responsible for cellular senescence, and are critical role in impairment of NO production in endothelial cells [7,8]. We also observed that PA treatment affected cellular senescence (Fig. 1).
Presently, exposure to PA decreased the autophagic ux (Fig. 5A, lower) and inhibited the fusion of autophagosomes and lysosomes (Fig. 6). However, con icting ndings have been reported regarding the function of FFAs in autophagy induction. In previous studies using pancreatic β-cells [45] and mouse embryonic broblasts [46], PA induced autophagy to protect cells from lipotoxicity. However, other reports including mouse broblasts [47] and human aortic endothelial cells [41] showed consistent ndings with our results. PA exposure inhibited autophagosome-lysosome fusion, and thus alleviated autophagy [41,47]. These con icting results might be due to multiple factors such as different cell types, different degree of lipotoxicity, and study design [46]. Further studies are required to explain the heterogeneity in the lipotoxicity-induced autophagic induction in various conditions.
A number of signaling pathways are involved in regulating the initiation and maturation of autophagy [48]. One pathway includes mammalian target of rapamycin (mTOR), which acts as a gate-keeper in the inhibition of autophagy [48]. Inhibition of the mTOR pathway is related to anti-atherogenic effects and atherosclerotic plaque stabilization [49]. Besides mTOR inhibitors, mitogen-activated protein kinase (MAPK) pathways including c-Jun N-terminal kinase (JNK), p38, and extracellular signal-regulated kinase (ERK) are also play crucial roles in anti-atherogenic effects in autophagy [49]. Although we did not investigate mechanisms other than the AMPK pathway, previous studies have shown that AMPK activation is a positive regulator of autophagy by attenuating its downstream target of mTOR [50].
AMPK is an energy sensor that has various anti-atherogenic effects and is bene cial to endothelial function [51]. CTRP9 exerts its anti-atherogenic effects via the AMPK pathway, which include an antiin ammatory effect in vascular endothelial cells [11,37]. Based on the previous relationship between CTRP9 and the AMPK pathway, we evaluated whether the AMPK pathway was involved in autophagic induction by CTRP9. We observed that the inhibition of AMPK signi cantly reduced LC3 conversion induced by CTRP9 treatment (Fig. 7A). In accordance with previous studies, the activation of AMPK has been demonstrated to be crucial in autophagic induction by CTRP9 [14,24,37]. When vascular endothelial cells were pre-treated with the AMPK inhibitor, signi cant decreases of the CTRP9-induced autophagy and autophagic ux were observed (Fig. 7A, 7B). In addition, pre-treatment with the AMPK inhibitor also attenuated the protective role of CTRP9 on PA-induced endothelial cell senescence (Fig. 8A,  8B). These ndings suggest a mechanism in which AMPK activation mediates the effect of CTRP9 in preventing endothelial senescence via the induction of autophagy and autophagic ux. Further investigations will be needed to determine whether other mechanisms might be involved in the autophagic induction by CTRP9.

Conclusion
CTRP9 can attenuate PA-induced endothelial senescence by recovering autophagy and autophagic ux. AMPK activation is involved in this protective effect of CTRP9 on cellular senescence. Further research using in vivo models will be needed to expand these ndings and to identify CTRP9-based therapies for atherosclerosis.
Abbreviations ANOVA, one-way analysis of variance; AMPK, AMP-activated protein kinase; ATG, autophagy related genes; CTRP9, C1q/TNF-related protein-9; CVD, cardiovascular disease; DMSO, dimethyl sulfoxide; eNOS, endothelial nitric oxide synthase; ERK, extracellular signal-regulated kinase; FBS, fetal bovine serum; FFA, free fatty acid; GFP-LC3, green uorescent protein fused to the N-terminus of LC3; HUVEC, human umbilical vein endothelial cells; JNK, c-Jun N-terminal kinase; MAPK, mitogen-activated protein kinase; mTORC1, mammalian target of rapamycin complex 1; NO, nitric oxide; PA, palmitic acid; pRB, retinoblastoma gene product; ROS, reactive oxygen species; SA-β-Gal, senescence-associated acidic βgalactosidase; siRNA, small interfering RNA; SQSTM1, sequestosome 1 Declarations Ethics approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and materials The analyzed datasets generated during the study are available from the corresponding author on reasonable request.       The AMPK signaling pathway is involved in CTRP9-induced autophagy in HUVECs. (A) The protein levels of LC3-II/LC3-I and p62 after pre-treatment with AMPK inhibitor (Compound C) followed by treatment with