- Email: [email protected]
Dysregulation of autophagy-related lncRNAs in peripheral blood of coronary artery disease patients
Journal Pre-proof Dysregulation of autophagy-related lncRNAs in peripheral blood of coronary artery disease patients Nader Ebadi, Soudeh Ghafouri-Fard, Mohammad Taheri, Shahram Arsang-Jang, Saeed Alipour Parsa, Mir Davood Omrani PII:
S0014-2999(19)30804-0
DOI:
https://doi.org/10.1016/j.ejphar.2019.172852
Reference:
EJP 172852
To appear in:
European Journal of Pharmacology
Received Date: 12 July 2019 Revised Date:
3 December 2019
Accepted Date: 9 December 2019
Please cite this article as: Ebadi, N., Ghafouri-Fard, S., Taheri, M., Arsang-Jang, S., Parsa, S.A., Omrani, M.D., Dysregulation of autophagy-related lncRNAs in peripheral blood of coronary artery disease patients, European Journal of Pharmacology (2020), doi: https://doi.org/10.1016/ j.ejphar.2019.172852. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier B.V.
Dysregulation of autophagy-related lncRNAs in peripheral blood of coronary artery disease patients Nader Ebadi1, Soudeh Ghafouri-Fard1, Mohammad Taheri2, Shahram Arsang-Jang3, Saeed Alipour Parsa4, Mir Davood Omrani1* 1. Department of Medical Genetics, Shahid Beheshti University of Medical Sciences, Tehran, Iran 2. Urogenital Stem Cell Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran 3. Clinical Research Development Center (CRDU), Qom University of Medical Sciences, Qom, Iran 4. Department of Cardiology, Cardiovascular Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran *Corresponding author: Mir Davood Omrani Tel & Fax: 00982123872572 Email: [email protected]
Abstract Coronary artery disease (CAD) as a major cause of death has been associated with dysregulation of several processes among them is autophagy. In the current study, we assessed expression of autophagy related gene 5 (ATG5) and three ATG5-associated long non-coding RNAs (lncRNAs Chast, HULC and DICER1-AS1) in the peripheral blood of patients with premature CAD and healthy subjects. Expression levels of ATG5, Chast, HULC and DICER1-AS1 were significantly lower in peripheral blood of CAD cases compared with healthy subjects. Receiver Operating Characteristic (ROC) curve analysis showed that HULC and DICER1-AS1 can properly differentiate CAD patients from healthy subjects (area under curve (AUC) values of 0.90 and 0.87, respectively). Expression levels of ATG5 and Chast were inversely correlated with FBS levels (r=0.41, P<0.0001 and r=-0.38, P<0.0001 respectively) but no other biochemical factors. Expression of DICER1-AS1 was inversely correlated with FBS (r=-0.54, P<0.0001), TG (r=-0.29, P<0.0001) and TG/ HDL ratio (r=-0.27, P<0.0001). Expression of HULC was inversely correlated with age (r=-0.24, P<0.0001), FBS (r=-0.62, P<0.0001) and TG (r=-0.31, P<0.0001). There were significant pairwise correlations between expression levels of all genes. The most robust correlations were detected ATG5 and Chast (r=0.81, P<0.0001) and between DICER1-AS1 and HULC (r=0.75, P<0.0001). The current study further verified associations between dysregulation of autophagy and CAD. Moreover, our results indicate appropriateness of two autophagy-related lncRNAs for differentiation of CAD status. Key words: Coronary artery disease, ATG5, lncRNA, Chast, HULC, DICER1-AS1
1. Introduction Cardiovascular Disease (CVD) is the leading cause of mortality across the world. Annually 17.7 million individuals die from CVD. This statistic accounts for one-third of all deaths worldwide. Coronary artery disease (CAD) is the main form of CVD and the leading cause of heart attacks or myocardial infraction (MI) (Organization, 17 May 2017). CAD is an atherosclerotic disease recognized by narrowing or blockage of the coronary arteries. It is manifested by stable angina, unstable angina, MI or sudden cardiac death (Galkina and Ley, 2009). Atherosclerosis as the underlying pathologic event in CAD, is a chronic inflammatory disease characterized by the formation of atherosclerotic plaques in the vessel wall of large- and medium-sized arteries (Bentzon Jacob et al., 2014). During the process of formation of atherosclerotic plaques, circulating monocytes moving into the subendothelium of vessel walls are converted into macrophages. After engulfing oxidized low-density lipoprotein or other modified lipoproteins, these cells are transformed into foam cells which are hallmarks of atherosclerotic lesions. Macrophages secrete multiple inflammatory factors that worsen plaque instability leading to plaque rupture. Consequently, macrophages exert an important role in atherosclerosis (Tabas and Bornfeldt, 2016). Since autophagy promotes cholesterol efflux from cells, evidences suggest that autophagic responses in macrophages play a protective role against atherosclerosis (Shao et al., 2016). Moreover, macrophage autophagy has an important role in maintaining vascular endothelial function by reducing oxidative stress, improving the bioavailability of nitric oxide and decreasing vascular inflammation (Shao et al., 2016). Autophagy is a self-protecting cellular function that relies on lysosomes and promotes cell survival. Dysfunctional organelles, long-lived proteins and wrong-folded proteins are degraded and recycled through the autophagy process (Zhang et al., 2012). Overall inhibition of autophagy can accelerate the pathological
process of atherosclerosis and CAD (Liao et al., 2012; Razani et al., 2012). A competent autophagic process lets cardiomyocytes to survive in ischemic conditions. A previous in vivo investigation demonstrated the effects of an autophagy-enhancer in reducing left ventricular remodeling and cardiac dysfunction after myocardial infarction. The proposed causal mechanism was suppression of p38 MAPK and induction of phosphoAkt signaling (Zhang et al., 2014). Another recent animal study, Sciarretta et al. have shown that a natural non-reducing sugar stimulates autophagy and converses the detrimental structural remodeling of the heart after myocardial infarction and ameliorates cardiac dysfunction. Notably, the beneficial effects of this natural agent were not detected in beclin 1+/– heterozygous mice in which autophagic pathways are deficient (Sciarretta et al., 2018). In the current study, we aimed at identification of expression pattern of autophagy-related long non-coding RNAs (lncRNAs) in the peripheral blood of CAD patients. So, we focused on three lncRNAs which have functional interactions with autophagy related gene 5 (ATG5). ATG5 is the key gene regulating autophagy in macrophages (Ye et al., 2018). Knockout of ATG5 have resulted in aggravation of atherosclerotic lipid plaques. The lncRNA cardiac hypertrophy-associated transcript (Chast) has been shown to inhibit autophagy possibly through suppressing ATG5 expression. This lncRNA inhibits expression of Pleckstrin homology domain-containing protein family M member 1 which is located on reverse strand of Chast, so prevents autophagy and driving hypertrophy in cardiomyocyte (Viereck et al., 2016). The lncRNA highly upregulated in liver cancer (HULC) enhances expression of ubiquitin-specific peptidase (USP22) which increases stability of Sirt1 protein. Sirt1 stimulates protective autophagy through deacetylating ATG5 (Xiong et al., 2017). A functional study in ovarian cancer cells has shown that HULC enhances mitochondria establishment, while its knockdown stimulated construction of autophagosom. As treatment with an autophagy inhibitor
agent reversed the effects of HULC knockdown in prompting apoptosis and hindering cell proliferation, it was suggested that HULC exerts its function through preventing autophagy (Chen et al., 2017). Finally, DICER1 antisense RNA 1 (DICER1-AS1) has been reported to enhance autophagy of osteosarcoma cells through modulation of miR-30b/ATG5 route. DICER1-AS1 silencing has suppressed expression of ATG5, LC3-II and Beclin 1 at protein level. In vivo indicated studies have shown the role of DICER1-AS1 silencing in suppression of the osteosarcoma tumor growth and expression of ATG5 (Gu et al., 2018). In the present project, we investigated expression of ATG5 and three functionally associated lncRNAs (Chast, HULC and DICER1-AS1) in peripheral blood of premature CAD patients and healthy subjects to evaluate autophagy process in these subjects. 2. Material and methods 2.1. Human subjects Totally, 50 patients with premature CAD (32 male and 18 female patients, age (mean±S.D.): 49.46±4.76) and 50 age- and sex-matched healthy controls (31 male and 19 female subjects, age (mean± S.D.): 47.2±4.51) were recruited in the current study. Inclusion criteria for CAD patients were diagnosis before age of 55 years and the presence of at least a single vessel stenosis >50% as diagnosed by angiography. We excluded patients with hypertension or congenital heart disease from study. Subjects recruited for control group aged more than 45 and had no personal or family history of CVD. Exclusion criteria for both cases and controls were personal history of renal disease, malignancy, autoimmune diseases, hematologic disorders or inflammatory disorders. The study protocol was approved by ethical committee of Shahid Beheshti University of
Medical Sciences. Written informed consent forms were signed by all study participants. Table 1 shows demographic and clinical features of study participants.
Table 1. Demographic and clinical features of study participants. Groups
Cases
Controls
Age (mean± S.D.)
49.46±4.76
47.2±4.51
Male/ female ratio
32/ 18
31/ 19
Fasting Blood Sugar (mean± S.D.)
159.42±70.48
84.44±26.84
Total cholesterol (mean± S.D.)
208.7±52.54
192.66±37.13
LDL cholesterol (mean± S.D.)
100.9±28.83
114.8±32.69
HDL cholesterol (mean± S.D.)
51.42±15.01
46.7±15.73
Triglyceride (mean± S.D.)
1971±77.17
137.42±73.18
Parameters
2.2. Expression assays Five ml of peripheral blood was collected from all participants in EDTA-containing tubes. All samples were collected immediately before angiography. Based on the results of angiography, patients were enrolled in case or control groups. Total RNA was isolated from all samples using Hybrid-RTM blood RNA extraction kit (GeneAll, Seoul, South Korea). The quality of RNA was assessed using NanoDrop system (Thermo Fisher Scientific). cDNA was synthetized from RNA samples using the OneStep RT-PCR Series Kit (BioFact™, Seoul, South Korea). Primers and probes used for PCR were designed using the Allele ID 7 for × 64 windows software (Premier Biosoft, Palo Alto, USA). Real time PCR reactions were prepared using RealQ Plus 2x PCR Master Mix Green Without ROX™ PCR Master Mix (Ampliqon, Odense, Denmark). Amplifications were performed in StepOnePlus™ RealTime PCR System (Applied Biosystems, Foster city, CA, USA). To achieve more accurate data for expression analysis, expressions of all genes were normalized by using two housekeeping genes (B2M and ATCB). Table 2 shows the detailed features of primers. Cycle threshold (Ct) values of genes were corrected for efficiency of amplification. For each sample, Ln (Ct Housekeeping gene-Ct target gene) was calculated to describe the relative expression of each gene. Table 2. The detailed features of primers. Genes
HULC
Primer Primer sequence
Primer
Amplicon
type
length
length 75
F
ACGTGAGGATACAGCAAGGC
20
R
AGAGTTCCTGCATGGTCTGG
20
ATG5
F
TTCGAGATGTGTGGTTTGGAC
21
R
CACTTTGTCAGTTACCAACGTCA
23
F
GCAGAGGGTGCCAACTTGTA
20
R
TCTCAGGGAAATCAGATTGCGG
22
DICER1-
F
CCCAGCCTGCTTCCTGTTTTAAC
23
AS1
R
TTCTCTCCCATCTTCACCTTCTCC
24
ATCB
F
CCTGGCACCCAGCACAAT
18
R
GGGCCGGACTCGTCATAC
18
F
AGATGAGTATGCCTGCCGTG
20
R
CGGCATCTTCAAACCTCCA
19
CHAST
B2M
134
109
126
144
104
2.3. Statistical analyses Relative expression of genes was compared between CAD patients and healthy subjects using Bayesian regression model. The effects of independent variables were adjusted in this model. The asymmetric Laplace prior distribution was assumed for parameterization of gene expression ratio with 4000 iteration and 1000 warm-up. P-values for regression model were estimated from Frequentist method. Spearman
correlation was applied to assess correlation between expressions of genes. Analyses were carried out in in R 3.5.2 environment using pROC, qreg, and Stan with loo packages. The diagnostic power of expression levels of genes were determined by depicting Receiver Operating Characteristic (ROC) curves. 3. Results 3.1. Biochemical analyses Levels of fasting blood sugar (FBS), total cholesterol (TC), triglyceride (TG), LDL cholesterol, HDL cholesterol, Non HDL cholesterol and TG/HDL ratio were examined in all study participants. Fig. 1 shows distribution of age and biochemical factors in cases and controls. Cases and controls were significantly different in their FBS levels. 3.2. Expression assays After normalization with either housekeeping genes, expression levels of ATG5, Chast, HULC and DICER1-AS1 were significantly lower in peripheral blood of CAD cases compared with healthy subjects (Fig. 2). The results of normalization of expression of all genes with ATCB and B2M were consistent. The following Figs. and tables are based on the results obtained from normalization with ATCB. Results of Bayesian Regression model showed that after adjustment of the effects of age and gender, expression of all genes were significantly lower in cases compared with controls (Table 3). Table 3.The results of Bayesian Regression model for comparison of expression of genes between case and control groups with adjusting the effects of age and gender (ER: expression ratio, SE: standard error, CrI: Credible Interval, P values are estimated from Frequentist method).
Groups
Genes
ATG5
Variable
Posterior
Chast SE
P-Value
Beta of
95% CrI
Posterior
for ER
Beta of
Group
-1.519
SE
P-
95% CrI for
Posterior
Value
ER
Beta of
ER
ER Total
HULC
0.46
<0.0001
[-2.38, -
-1.038
0.937
0.53
0.649
[-0.12,
0.547
0.5
0.011
0.73
0.677
1.96] Gender
0.021
0.04
.226
[-0.06,
[-2.06, -0.1]
-3.537
0.04
0.25
0.1] Group*
-0.904
0.83
0.718
Gender Male
Group
[-2.47,
[-0.78,
0.124
2.04] -0.022
[-0.11,
-0.029
0.46
0.002
[-2.44, -
0.92
0.304
[-2.27, 1.3]
0.602
0.0004
0.05
0.83
[-0.1,
-1.014
0.49
0.007
-0.028
0.06
0.409
0.08] Female
Group
-2.504
0.78
0.002
[-3.89, -
[-2, -0.09]
-3.39
0.064
0.08
0.409
[-0.11, 0.23]
3.3. ROC curves
[-0.15,
-0.077
0.07] -1.846
0.81
0.036
0.92] Age
Posterior
Value
ER
Beta of
[-3.35, -
0.5
<0.00
4
01
0.6
0.715
[-4.58, -2.49]
[-1.21, 1.4]
-2.629
0.73
P-
95% CrI
Value
for ER
0.4
<0.00
[-3.46, -
5
01
1.71]
0.5
0.975
[-0.29,
0.0
3 0.527
[-0.12, 0.06]
0.032
0.9
0.294
0.478
[-1.28, 2.44]
-0.639
0.7
<0.00
2
01
0.0
0.327
[-4.45, -2.38]
[-0.18, 0.02]
-2.597
0.038
0.871
[-5.39, -2.03]
-3.071
0.4
<0.00
[-3.57, -
8
01
1.65]
0.0
0.984
[-0.07,
0.7
0.13] 0.002
0.32] 0.023
0.09
0.602
[-0.18, 0.19]
[-2.04, 0.9]
5 0.016
[-0.05, 0.11]
5
0.5
0.9
0.0
1.71]
4
5 -3.701
SE
ER
3
0.66] Age
95% CrI for
5
0.71] -1.534
P-
4
0.06] -0.536
SE
ER
0.6] Age
DICER1-AS1
[-4.45, 1.63]
0.068
0.0 8
0.033
[-0.08, 0.22]
0.012
0.0 7
0.954
[-0.13, 0.16]
ROC curve analysis showed the best diagnostic power for HULC and DICER1-AS1 (area under curve (AUC) values of 0.90 and 0.87 respectively). Fig. 3A-D show the ROC curves for ATG5, Chast, DICER1-AS1 and HULC, respectively. 3.4. Correlations between expression levels of genes and biochemical factors Expression levels of ATG5 and Chast were inversely correlated with FBS levels (r=-0.41, P<0.0001 and r=-0.38, P<0.0001 respectively) but no other biochemical factors (Fig. 4A-B). Expression of DICER1-AS1 was inversely correlated with FBS (r=-0.54, P<0.0001), TG (r=-0.29, P<0.0001) and TG/ HDL ratio (r=-0.27, P<0.0001) (Fig. 4C). Expression of HULC was inversely correlated with age (r=-0.24, P<0.0001), FBS (r=-0.62, P<0.0001) and TG (r=-0.31, P<0.0001) (Fig. 4D). There were significant pairwise correlations between some biochemical factors such as between TG and TG/ HDL ratio (r=0.81, P<0.0001) and between TC and non HDL cholesterol (r=0.93, P<0.0001). 3.5. Correlations between expression levels of genes There were significant pairwise correlations between expression levels of all genes. The most robust correlations were detected ATG5 and Chast (r=0.81, P<0.0001) and between DICER1-AS1 and HULC (r=0.75, P<0.0001) (Fig. 5). We detected no difference in gene-gene correlation between CAD populations and healthy controls. 4. Discussion Although the protective role of autophagy against CAD has been acknowledged, most of studies focusing on enhancement of this process have been conducted in the context of cancer. However, increasing evidences show that autophagy is a potentially attractive therapeutic target in cardiopathology as well (Lavandero et al., 2015; Towers and Thorburn, 2016). Autophagy in cardiomyocytes is involved in both main stages of
myocardial injury namely ischemia and reperfusion (Matsui et al., 2007), so therapeutic options for enhancement of autophagy are expected to influence both mentioned phases. An important shortcoming in translating investigational autophagy-targeted modalities into treatment of cardiopathology is the challenge in the evaluation of autophagy in cardiovascular system in the clinical situation (Schiattarella and Hill, 2016). In the current study, we have assessed expression of autophagy-related lncRNAs in the peripheral blood of CAD patients with the supposition that peripheral blood tissue reflects expression of these genes in the coronary arteries as the main pathological location of CAD. Notably, we detected significant down-regulation of ATG5, Chast, HULC and DICER1-AS1 in peripheral blood of CAD patients compared with healthy individuals. ATG5 as a main regulator of autophagy has been shown to be implicated in cardiopathology as well. Impairment of autophagy in Atg5+/- mice has resulted in over-production of Reactive Oxygen Species (ROS), which stimulates nuclear factor κB (NF-κB) pathway in immune cells leading to inflammatory responses in cardiac tissue (Zhao et al., 2014). Meanwhile, ATG5 has a possible role in regulation of glucose tolerance. A previous animal study has shown that transgenic over-expression of Atg5 in mice leads to significantly decreased blood glucose levels, which might be due to higher glucose tolerance or augmented glucose clearance by Atg5. Moreover, Atg5 transgenic mice had higher insulin sensitivity after insulin stimulation (Pyo et al., 2013). Others have shown that dietary glucose has a direct influence on autophagy in mammalian cells. Glucose scarcity stimulates autophagy mostly through AMPK activation and the succeeding suppression of mTORC1 (Moruno et al., 2012). Yet, dysregulated autophagy participates in diabetes and its complications as autophagy controls the typical function of pancreatic β cells and insulintarget tissues (Bhattacharya et al., 2018). Animal studies have shown a crucial role for autophagy in the maintenance of normal structural design of pancreatic β cells (Jung et al., 2008). However, signs of reformed autophagy have been detected in the β cells of the Zucker diabetic fatty rat
and in a β cell line after sustained exposure to high glucose (Kaniuk et al., 2007). Thus, autophagy, high blood glucose and CAD are associated with each other in a complicated way that identification of cause-effect relationship or even the direction of effect is not simple. Only through elaborate functional studies, one can deduce the exact mechanism. In line with these evidences, we detected inverse correlations between FBS levels and expression of autophagy-related lncRNAs including ATG5, Chast, DICER1-AS1 and HULC in the context of CAD. However, as no previous study has addressed the functional role of these lncRNAs in CAD, more data is needed to elaborate the mechanism of the observed associations. LncRNAs usually regulate autophagy through changing transcript levels of ATG genes. Mainly, they act as competing endogenous RNAs to alter expression of microRNAs (miRNAs) involved in autophagy (Yang et al., 2017). However, Chast has been shown to decrease expression of Pleckstrin homology domain-containing protein family M member 1 and inhibit cardiomyocyte autophagy while inducing hypertrophy in these cells (Viereck et al., 2016). Opposite to our expectation, we detected lower levels of Chast in CAD patients compared with healthy subjects. Based on the role of Chast in cardiac remodeling (Viereck et al., 2016), we hypothesize that decreased peripheral levels of this lncRNA is not reflective of its levels in damaged cardiac tissues. Remodeling of the coronary arteries is a process being initiated in the endothelium and gradually spreading towards the atherosclerotic plaque and myocardium (Heusch et al., 2014). HULC promotes autophagy through inhibition of miR-6825-5p, miR-6845-5p and miR-6886-3p and subsequent upregulation of USP22 expression. The final effect of this lncRNA is deacetylation of ATG5 and ATG7 (Xiong et al., 2017). The role of DICER1-AS1 in enhancement of autophagy has been demonstrated in osteosarcoma cells as its silencing has inhibited autophagy in vitro. Such effect has been accompanied by suppression of ATG5, LC3-II and
Beclin 1 protein levels (Gu et al., 2018). Consequently, in the present study, we demonstrated down-regulation of three pro-autophagy genes in peripheral blood of CAD patients. Moreover, we detected down-regulation of an lncRNA whose function is associated with cardiac remodeling. Notably, our results showed appropriateness of HULC and DICER1-AS1 transcript levels for diagnosis of CAD. Such autophagy-related lncRNAs might be putative markers for assessment of autophagy in CAD patients. Future studies are needed to elaborate whether their transcript levels are modulated by autophagy-targeted therapies. We also detected inverse correlations between expression levels of pro-autophagy genes and FBS levels which imply the role of high FBS in suppression of autophagy. Moreover, expressions of DICER1-AS1 and HULC were inversely correlated with TG levels. It is possible that some detrimental effects of chronic high FBS and TG are mediated through suppression of autophagy in cardiac tissues. Animal experiments have shown induction of autophagy in beta cells of pancreas by either a high-fat regimen or a combined high-fat and high-glucose regimen, but not by only high-glucose diet (Sheng et al., 2017). Differences between human and rodents or tissue-specific responses to high glucose and high fat diets should be considered in explanation of these conflicting results. In addition, we detected significant pairwise correlations between expression levels of all genes especially between ATG5 and Chast and between DICER1-AS1 and HULC. The positive correlation between ATG5 and Chast might imply the presence of a feedback loop for regulation of autophagy which decreases the level of an anti-autophagy gene in response to robust suppression of autophagy. Taken together, the current study further verified associations between dysregulation of autophagy and CAD. Moreover, our results indicate appropriateness of two autophagy-related lncRNAs for differentiation of CAD status. Our study had limitations regarding lack of assessment of
expression of genes in atherosclerotic plaques, lack of validation experiments in vitro and in vivo and lack of evidence to prove the direct or sole correlation between the autophagy related lncRNAs and CAD. Authors' contribution NE, MT performed the experiments. NE and MDO conceived and designed the study. SGF and MDO supervised the study and wrote the manuscript. SAJ analyzed the data. SAP assessed the clinical data and supervised patients' enrollment. All authors contributed in the study design and approved the final manuscript. Acknowledgement The current study was supported by a grant from Shahid Beheshti University of Medical Sciences and was conducted as the Ph.D. thesis project of the first author. Conflict of interest None to be declared Availability of data The data that support the findings of this study are available from the corresponding author upon reasonable request. References Bentzon Jacob, F., Otsuka, F., Virmani, R., Falk, E., 2014. Mechanisms of Plaque Formation and Rupture. Circulation Research 114, 1852-1866.
Bhattacharya, D., Mukhopadhyay, M., Bhattacharyya, M., Karmakar, P., 2018. Is autophagy associated with diabetes mellitus and its complications? A review. EXCLI journal 17, 709-720. Chen, S., Wu, D.D., Sang, X.B., Wang, L.L., Zong, Z.H., Sun, K.X., Liu, B.L., Zhao, Y., 2017. The lncRNA HULC functions as an oncogene by targeting ATG7 and ITGB1 in epithelial ovarian carcinoma. Cell Death Dis 8, e3118. Galkina, E., Ley, K., 2009. Immune and inflammatory mechanisms of atherosclerosis (*). Annual review of immunology 27, 165-197. Gu, Z., Hou, Z., Zheng, L., Wang, X., Wu, L., Zhang, C., 2018. LncRNA DICER1-AS1 promotes the proliferation, invasion and autophagy of osteosarcoma cells via miR-30b/ATG5. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie 104, 110-118. Heusch, G., Libby, P., Gersh, B., Yellon, D., Böhm, M., Lopaschuk, G., Opie, L., 2014. Cardiovascular remodelling in coronary artery disease and heart failure. The Lancet 383, 1933-1943. Jung, H.S., Chung, K.W., Won Kim, J., Kim, J., Komatsu, M., Tanaka, K., Nguyen, Y.H., Kang, T.M., Yoon, K.H., Kim, J.W., Jeong, Y.T., Han, M.S., Lee, M.K., Kim, K.W., Shin, J., Lee, M.S., 2008. Loss of autophagy diminishes pancreatic beta cell mass and function with resultant hyperglycemia. Cell Metab 8, 318-324. Kaniuk, N.A., Kiraly, M., Bates, H., Vranic, M., Volchuk, A., Brumell, J.H., 2007. Ubiquitinated-protein aggregates form in pancreatic betacells during diabetes-induced oxidative stress and are regulated by autophagy. Diabetes 56, 930-939. Lavandero, S., Chiong, M., Rothermel, B.A., Hill, J.A., 2015. Autophagy in cardiovascular biology. The Journal of clinical investigation 125, 55-64.
Liao, X., Sluimer, J.C., Wang, Y., Subramanian, M., Brown, K., Pattison, J.S., Robbins, J., Martinez, J., Tabas, I., 2012. Macrophage autophagy plays a protective role in advanced atherosclerosis. Cell metabolism 15, 545-553. Matsui, Y., Takagi, H., Qu, X., Abdellatif, M., Sakoda, H., Asano, T., Levine, B., Sadoshima, J., 2007. Distinct roles of autophagy in the heart during ischemia and reperfusion: roles of AMP-activated protein kinase and Beclin 1 in mediating autophagy. Circ Res 100, 914-922. Moruno, F., Perez-Jimenez, E., Knecht, E., 2012. Regulation of autophagy by glucose in Mammalian cells. Cells 1, 372-395. Organization, W.H., 17 May 2017. Cardiovascular diseases (CVDs). Retrieved from: https://www.who.int/cardiovascular_diseases/en/. Pyo, J.O., Yoo, S.M., Ahn, H.H., Nah, J., Hong, S.H., Kam, T.I., Jung, S., Jung, Y.K., 2013. Overexpression of Atg5 in mice activates autophagy and extends lifespan. Nature communications 4, 2300. Razani, B., Feng, C., Coleman, T., Emanuel, R., Wen, H., Hwang, S., Ting, J.P., Virgin, H.W., Kastan, M.B., Semenkovich, C.F., 2012. Autophagy links inflammasomes to atherosclerotic progression. Cell metabolism 15, 534-544. Schiattarella, G.G., Hill, J.A., 2016. Therapeutic targeting of autophagy in cardiovascular disease. Journal of molecular and cellular cardiology 95, 86-93. Sciarretta, S., Yee, D., Nagarajan, N., Bianchi, F., Saito, T., Valenti, V., Tong, M., Del Re, D.P., Vecchione, C., Schirone, L., 2018. Trehaloseinduced activation of autophagy improves cardiac remodeling after myocardial infarction. Journal of the American College of Cardiology 71, 1999-2010.
Shao, B.Z., Han, B.Z., Zeng, Y.X., Su, D.F., Liu, C., 2016. The roles of macrophage autophagy in atherosclerosis. Acta pharmacologica Sinica 37, 150-156. Sheng, Q., Xiao, X., Prasadan, K., Chen, C., Ming, Y., Fusco, J., Gangopadhyay, N.N., Ricks, D., Gittes, G.K., 2017. Autophagy protects pancreatic beta cell mass and function in the setting of a high-fat and high-glucose diet. Scientific reports 7, 16348. Tabas, I., Bornfeldt, K.E., 2016. Macrophage Phenotype and Function in Different Stages of Atherosclerosis. Circ Res 118, 653-667. Towers, C.G., Thorburn, A., 2016. Therapeutic targeting of autophagy. EBioMedicine 14, 15-23. Viereck, J., Kumarswamy, R., Foinquinos, A., Xiao, K., Avramopoulos, P., Kunz, M., Dittrich, M., Maetzig, T., Zimmer, K., Remke, J., Just, A., Fendrich, J., Scherf, K., Bolesani, E., Schambach, A., Weidemann, F., Zweigerdt, R., de Windt, L.J., Engelhardt, S., Dandekar, T., Batkai, S., Thum, T., 2016. Long noncoding RNA Chast promotes cardiac remodeling. Science translational medicine 8, 326ra322. Xiong, H., Ni, Z., He, J., Jiang, S., Li, X., He, J., Gong, W., Zheng, L., Chen, S., Li, B., Zhang, N., Lyu, X., Huang, G., Chen, B., Zhang, Y., He, F., 2017. LncRNA HULC triggers autophagy via stabilizing Sirt1 and attenuates the chemosensitivity of HCC cells. Oncogene 36, 3528-3540. Yang, L., Wang, H., Shen, Q., Feng, L., Jin, H., 2017. Long non-coding RNAs involved in autophagy regulation. Cell Death Dis 8, e3073. Ye, X., Zhou, X.J., Zhang, H., 2018. Exploring the Role of Autophagy-Related Gene 5 (ATG5) Yields Important Insights Into Autophagy in Autoimmune/Autoinflammatory Diseases. Frontiers in immunology 9, 2334. Zhang, T., Qi, Y., Liao, M., Xu, M., Bower, K.A., Frank, J.A., Shen, H.-M., Luo, J., Shi, X., Chen, G., 2012. Autophagy Is a Cell Self-Protective Mechanism Against Arsenic-Induced Cell Transformation. Toxicological Sciences 130, 298-308.
Zhang, Y.J., Yang, S.H., Li, M.H., Iqbal, J., Bourantas, C.V., Mi, Q.Y., Yu, Y.H., Li, J.J., Zhao, S.L., Tian, N.L., Chen, S.L., 2014. Berberine attenuates adverse left ventricular remodeling and cardiac dysfunction after acute myocardial infarction in rats: role of autophagy. Clinical and experimental pharmacology & physiology 41, 995-1002. Zhao, W., Li, Y., Jia, L., Pan, L., Li, H., Du, J., 2014. Atg5 deficiency-mediated mitophagy aggravates cardiac inflammation and injury in response to angiotensin II. Free radical biology & medicine 69, 108-115. Figure legends Fig. 1. Distribution of age and biochemical factors in cases and controls (P<0.001 for FBS and >0.05 in other comparisons; The central line shows the median value, while the box contains the 25th to 75th percentiles. The black whiskers mark the 5th and 95th percentiles, and values beyond these upper and lower bounds are considered outliers, marked with red + signs). Fig. 2. Relative expressions of ATG5 (P<0.0001), Chast (P=0.011), HULC (P<0.0001) and DICER1-AS1 (P<0.0001) in CAD patients and healthy subjects (The central line shows the median value, while the box contains the 25th to 75th percentiles. The black whiskers mark the 5th and 95th percentiles, and values beyond these upper and lower bounds are considered outliers, marked with red + signs). Fig. 3. ROC curves for assessment of diagnostic power of ATG5 (A), Chast (B), DICER1-AS1 (C) and HULC (D) in differentiation of CAD patients from healthy subjects (True positive rate (Sensitivity) is plotted in function of the false positive rate (100-Specificity) for different cutoff points of expression levels of genes).
Fig. 4. Correlations between expression levels of ATG5 (A), Chast (B), DICER1-AS1 (C) and HULC (D) genes and biochemical factors (The diagonal displays the distribution of certain parameter. The left diagonal has a scatterplot of the two variables overhead it and to its right; the right diagonal designs the correlation coefficient for the same variables, with the font size escalating for larger correlations). Fig. 5. Pairwise correlations between expression levels of autophagy related genes (The diagonal displays the distribution of expression of a gene. The left diagonal has a scatterplot of the two variables overhead it and to its right; the right diagonal designs the correlation coefficient for the same variables, with the font size escalating for larger correlations).
S0014-2999(19)30804-0
DOI:
https://doi.org/10.1016/j.ejphar.2019.172852
Reference:
EJP 172852
To appear in:
European Journal of Pharmacology
Received Date: 12 July 2019 Revised Date:
3 December 2019
Accepted Date: 9 December 2019
Please cite this article as: Ebadi, N., Ghafouri-Fard, S., Taheri, M., Arsang-Jang, S., Parsa, S.A., Omrani, M.D., Dysregulation of autophagy-related lncRNAs in peripheral blood of coronary artery disease patients, European Journal of Pharmacology (2020), doi: https://doi.org/10.1016/ j.ejphar.2019.172852. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier B.V.
Dysregulation of autophagy-related lncRNAs in peripheral blood of coronary artery disease patients Nader Ebadi1, Soudeh Ghafouri-Fard1, Mohammad Taheri2, Shahram Arsang-Jang3, Saeed Alipour Parsa4, Mir Davood Omrani1* 1. Department of Medical Genetics, Shahid Beheshti University of Medical Sciences, Tehran, Iran 2. Urogenital Stem Cell Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran 3. Clinical Research Development Center (CRDU), Qom University of Medical Sciences, Qom, Iran 4. Department of Cardiology, Cardiovascular Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran *Corresponding author: Mir Davood Omrani Tel & Fax: 00982123872572 Email: [email protected]
Abstract Coronary artery disease (CAD) as a major cause of death has been associated with dysregulation of several processes among them is autophagy. In the current study, we assessed expression of autophagy related gene 5 (ATG5) and three ATG5-associated long non-coding RNAs (lncRNAs Chast, HULC and DICER1-AS1) in the peripheral blood of patients with premature CAD and healthy subjects. Expression levels of ATG5, Chast, HULC and DICER1-AS1 were significantly lower in peripheral blood of CAD cases compared with healthy subjects. Receiver Operating Characteristic (ROC) curve analysis showed that HULC and DICER1-AS1 can properly differentiate CAD patients from healthy subjects (area under curve (AUC) values of 0.90 and 0.87, respectively). Expression levels of ATG5 and Chast were inversely correlated with FBS levels (r=0.41, P<0.0001 and r=-0.38, P<0.0001 respectively) but no other biochemical factors. Expression of DICER1-AS1 was inversely correlated with FBS (r=-0.54, P<0.0001), TG (r=-0.29, P<0.0001) and TG/ HDL ratio (r=-0.27, P<0.0001). Expression of HULC was inversely correlated with age (r=-0.24, P<0.0001), FBS (r=-0.62, P<0.0001) and TG (r=-0.31, P<0.0001). There were significant pairwise correlations between expression levels of all genes. The most robust correlations were detected ATG5 and Chast (r=0.81, P<0.0001) and between DICER1-AS1 and HULC (r=0.75, P<0.0001). The current study further verified associations between dysregulation of autophagy and CAD. Moreover, our results indicate appropriateness of two autophagy-related lncRNAs for differentiation of CAD status. Key words: Coronary artery disease, ATG5, lncRNA, Chast, HULC, DICER1-AS1
1. Introduction Cardiovascular Disease (CVD) is the leading cause of mortality across the world. Annually 17.7 million individuals die from CVD. This statistic accounts for one-third of all deaths worldwide. Coronary artery disease (CAD) is the main form of CVD and the leading cause of heart attacks or myocardial infraction (MI) (Organization, 17 May 2017). CAD is an atherosclerotic disease recognized by narrowing or blockage of the coronary arteries. It is manifested by stable angina, unstable angina, MI or sudden cardiac death (Galkina and Ley, 2009). Atherosclerosis as the underlying pathologic event in CAD, is a chronic inflammatory disease characterized by the formation of atherosclerotic plaques in the vessel wall of large- and medium-sized arteries (Bentzon Jacob et al., 2014). During the process of formation of atherosclerotic plaques, circulating monocytes moving into the subendothelium of vessel walls are converted into macrophages. After engulfing oxidized low-density lipoprotein or other modified lipoproteins, these cells are transformed into foam cells which are hallmarks of atherosclerotic lesions. Macrophages secrete multiple inflammatory factors that worsen plaque instability leading to plaque rupture. Consequently, macrophages exert an important role in atherosclerosis (Tabas and Bornfeldt, 2016). Since autophagy promotes cholesterol efflux from cells, evidences suggest that autophagic responses in macrophages play a protective role against atherosclerosis (Shao et al., 2016). Moreover, macrophage autophagy has an important role in maintaining vascular endothelial function by reducing oxidative stress, improving the bioavailability of nitric oxide and decreasing vascular inflammation (Shao et al., 2016). Autophagy is a self-protecting cellular function that relies on lysosomes and promotes cell survival. Dysfunctional organelles, long-lived proteins and wrong-folded proteins are degraded and recycled through the autophagy process (Zhang et al., 2012). Overall inhibition of autophagy can accelerate the pathological
process of atherosclerosis and CAD (Liao et al., 2012; Razani et al., 2012). A competent autophagic process lets cardiomyocytes to survive in ischemic conditions. A previous in vivo investigation demonstrated the effects of an autophagy-enhancer in reducing left ventricular remodeling and cardiac dysfunction after myocardial infarction. The proposed causal mechanism was suppression of p38 MAPK and induction of phosphoAkt signaling (Zhang et al., 2014). Another recent animal study, Sciarretta et al. have shown that a natural non-reducing sugar stimulates autophagy and converses the detrimental structural remodeling of the heart after myocardial infarction and ameliorates cardiac dysfunction. Notably, the beneficial effects of this natural agent were not detected in beclin 1+/– heterozygous mice in which autophagic pathways are deficient (Sciarretta et al., 2018). In the current study, we aimed at identification of expression pattern of autophagy-related long non-coding RNAs (lncRNAs) in the peripheral blood of CAD patients. So, we focused on three lncRNAs which have functional interactions with autophagy related gene 5 (ATG5). ATG5 is the key gene regulating autophagy in macrophages (Ye et al., 2018). Knockout of ATG5 have resulted in aggravation of atherosclerotic lipid plaques. The lncRNA cardiac hypertrophy-associated transcript (Chast) has been shown to inhibit autophagy possibly through suppressing ATG5 expression. This lncRNA inhibits expression of Pleckstrin homology domain-containing protein family M member 1 which is located on reverse strand of Chast, so prevents autophagy and driving hypertrophy in cardiomyocyte (Viereck et al., 2016). The lncRNA highly upregulated in liver cancer (HULC) enhances expression of ubiquitin-specific peptidase (USP22) which increases stability of Sirt1 protein. Sirt1 stimulates protective autophagy through deacetylating ATG5 (Xiong et al., 2017). A functional study in ovarian cancer cells has shown that HULC enhances mitochondria establishment, while its knockdown stimulated construction of autophagosom. As treatment with an autophagy inhibitor
agent reversed the effects of HULC knockdown in prompting apoptosis and hindering cell proliferation, it was suggested that HULC exerts its function through preventing autophagy (Chen et al., 2017). Finally, DICER1 antisense RNA 1 (DICER1-AS1) has been reported to enhance autophagy of osteosarcoma cells through modulation of miR-30b/ATG5 route. DICER1-AS1 silencing has suppressed expression of ATG5, LC3-II and Beclin 1 at protein level. In vivo indicated studies have shown the role of DICER1-AS1 silencing in suppression of the osteosarcoma tumor growth and expression of ATG5 (Gu et al., 2018). In the present project, we investigated expression of ATG5 and three functionally associated lncRNAs (Chast, HULC and DICER1-AS1) in peripheral blood of premature CAD patients and healthy subjects to evaluate autophagy process in these subjects. 2. Material and methods 2.1. Human subjects Totally, 50 patients with premature CAD (32 male and 18 female patients, age (mean±S.D.): 49.46±4.76) and 50 age- and sex-matched healthy controls (31 male and 19 female subjects, age (mean± S.D.): 47.2±4.51) were recruited in the current study. Inclusion criteria for CAD patients were diagnosis before age of 55 years and the presence of at least a single vessel stenosis >50% as diagnosed by angiography. We excluded patients with hypertension or congenital heart disease from study. Subjects recruited for control group aged more than 45 and had no personal or family history of CVD. Exclusion criteria for both cases and controls were personal history of renal disease, malignancy, autoimmune diseases, hematologic disorders or inflammatory disorders. The study protocol was approved by ethical committee of Shahid Beheshti University of
Medical Sciences. Written informed consent forms were signed by all study participants. Table 1 shows demographic and clinical features of study participants.
Table 1. Demographic and clinical features of study participants. Groups
Cases
Controls
Age (mean± S.D.)
49.46±4.76
47.2±4.51
Male/ female ratio
32/ 18
31/ 19
Fasting Blood Sugar (mean± S.D.)
159.42±70.48
84.44±26.84
Total cholesterol (mean± S.D.)
208.7±52.54
192.66±37.13
LDL cholesterol (mean± S.D.)
100.9±28.83
114.8±32.69
HDL cholesterol (mean± S.D.)
51.42±15.01
46.7±15.73
Triglyceride (mean± S.D.)
1971±77.17
137.42±73.18
Parameters
2.2. Expression assays Five ml of peripheral blood was collected from all participants in EDTA-containing tubes. All samples were collected immediately before angiography. Based on the results of angiography, patients were enrolled in case or control groups. Total RNA was isolated from all samples using Hybrid-RTM blood RNA extraction kit (GeneAll, Seoul, South Korea). The quality of RNA was assessed using NanoDrop system (Thermo Fisher Scientific). cDNA was synthetized from RNA samples using the OneStep RT-PCR Series Kit (BioFact™, Seoul, South Korea). Primers and probes used for PCR were designed using the Allele ID 7 for × 64 windows software (Premier Biosoft, Palo Alto, USA). Real time PCR reactions were prepared using RealQ Plus 2x PCR Master Mix Green Without ROX™ PCR Master Mix (Ampliqon, Odense, Denmark). Amplifications were performed in StepOnePlus™ RealTime PCR System (Applied Biosystems, Foster city, CA, USA). To achieve more accurate data for expression analysis, expressions of all genes were normalized by using two housekeeping genes (B2M and ATCB). Table 2 shows the detailed features of primers. Cycle threshold (Ct) values of genes were corrected for efficiency of amplification. For each sample, Ln (Ct Housekeeping gene-Ct target gene) was calculated to describe the relative expression of each gene. Table 2. The detailed features of primers. Genes
HULC
Primer Primer sequence
Primer
Amplicon
type
length
length 75
F
ACGTGAGGATACAGCAAGGC
20
R
AGAGTTCCTGCATGGTCTGG
20
ATG5
F
TTCGAGATGTGTGGTTTGGAC
21
R
CACTTTGTCAGTTACCAACGTCA
23
F
GCAGAGGGTGCCAACTTGTA
20
R
TCTCAGGGAAATCAGATTGCGG
22
DICER1-
F
CCCAGCCTGCTTCCTGTTTTAAC
23
AS1
R
TTCTCTCCCATCTTCACCTTCTCC
24
ATCB
F
CCTGGCACCCAGCACAAT
18
R
GGGCCGGACTCGTCATAC
18
F
AGATGAGTATGCCTGCCGTG
20
R
CGGCATCTTCAAACCTCCA
19
CHAST
B2M
134
109
126
144
104
2.3. Statistical analyses Relative expression of genes was compared between CAD patients and healthy subjects using Bayesian regression model. The effects of independent variables were adjusted in this model. The asymmetric Laplace prior distribution was assumed for parameterization of gene expression ratio with 4000 iteration and 1000 warm-up. P-values for regression model were estimated from Frequentist method. Spearman
correlation was applied to assess correlation between expressions of genes. Analyses were carried out in in R 3.5.2 environment using pROC, qreg, and Stan with loo packages. The diagnostic power of expression levels of genes were determined by depicting Receiver Operating Characteristic (ROC) curves. 3. Results 3.1. Biochemical analyses Levels of fasting blood sugar (FBS), total cholesterol (TC), triglyceride (TG), LDL cholesterol, HDL cholesterol, Non HDL cholesterol and TG/HDL ratio were examined in all study participants. Fig. 1 shows distribution of age and biochemical factors in cases and controls. Cases and controls were significantly different in their FBS levels. 3.2. Expression assays After normalization with either housekeeping genes, expression levels of ATG5, Chast, HULC and DICER1-AS1 were significantly lower in peripheral blood of CAD cases compared with healthy subjects (Fig. 2). The results of normalization of expression of all genes with ATCB and B2M were consistent. The following Figs. and tables are based on the results obtained from normalization with ATCB. Results of Bayesian Regression model showed that after adjustment of the effects of age and gender, expression of all genes were significantly lower in cases compared with controls (Table 3). Table 3.The results of Bayesian Regression model for comparison of expression of genes between case and control groups with adjusting the effects of age and gender (ER: expression ratio, SE: standard error, CrI: Credible Interval, P values are estimated from Frequentist method).
Groups
Genes
ATG5
Variable
Posterior
Chast SE
P-Value
Beta of
95% CrI
Posterior
for ER
Beta of
Group
-1.519
SE
P-
95% CrI for
Posterior
Value
ER
Beta of
ER
ER Total
HULC
0.46
<0.0001
[-2.38, -
-1.038
0.937
0.53
0.649
[-0.12,
0.547
0.5
0.011
0.73
0.677
1.96] Gender
0.021
0.04
.226
[-0.06,
[-2.06, -0.1]
-3.537
0.04
0.25
0.1] Group*
-0.904
0.83
0.718
Gender Male
Group
[-2.47,
[-0.78,
0.124
2.04] -0.022
[-0.11,
-0.029
0.46
0.002
[-2.44, -
0.92
0.304
[-2.27, 1.3]
0.602
0.0004
0.05
0.83
[-0.1,
-1.014
0.49
0.007
-0.028
0.06
0.409
0.08] Female
Group
-2.504
0.78
0.002
[-3.89, -
[-2, -0.09]
-3.39
0.064
0.08
0.409
[-0.11, 0.23]
3.3. ROC curves
[-0.15,
-0.077
0.07] -1.846
0.81
0.036
0.92] Age
Posterior
Value
ER
Beta of
[-3.35, -
0.5
<0.00
4
01
0.6
0.715
[-4.58, -2.49]
[-1.21, 1.4]
-2.629
0.73
P-
95% CrI
Value
for ER
0.4
<0.00
[-3.46, -
5
01
1.71]
0.5
0.975
[-0.29,
0.0
3 0.527
[-0.12, 0.06]
0.032
0.9
0.294
0.478
[-1.28, 2.44]
-0.639
0.7
<0.00
2
01
0.0
0.327
[-4.45, -2.38]
[-0.18, 0.02]
-2.597
0.038
0.871
[-5.39, -2.03]
-3.071
0.4
<0.00
[-3.57, -
8
01
1.65]
0.0
0.984
[-0.07,
0.7
0.13] 0.002
0.32] 0.023
0.09
0.602
[-0.18, 0.19]
[-2.04, 0.9]
5 0.016
[-0.05, 0.11]
5
0.5
0.9
0.0
1.71]
4
5 -3.701
SE
ER
3
0.66] Age
95% CrI for
5
0.71] -1.534
P-
4
0.06] -0.536
SE
ER
0.6] Age
DICER1-AS1
[-4.45, 1.63]
0.068
0.0 8
0.033
[-0.08, 0.22]
0.012
0.0 7
0.954
[-0.13, 0.16]
ROC curve analysis showed the best diagnostic power for HULC and DICER1-AS1 (area under curve (AUC) values of 0.90 and 0.87 respectively). Fig. 3A-D show the ROC curves for ATG5, Chast, DICER1-AS1 and HULC, respectively. 3.4. Correlations between expression levels of genes and biochemical factors Expression levels of ATG5 and Chast were inversely correlated with FBS levels (r=-0.41, P<0.0001 and r=-0.38, P<0.0001 respectively) but no other biochemical factors (Fig. 4A-B). Expression of DICER1-AS1 was inversely correlated with FBS (r=-0.54, P<0.0001), TG (r=-0.29, P<0.0001) and TG/ HDL ratio (r=-0.27, P<0.0001) (Fig. 4C). Expression of HULC was inversely correlated with age (r=-0.24, P<0.0001), FBS (r=-0.62, P<0.0001) and TG (r=-0.31, P<0.0001) (Fig. 4D). There were significant pairwise correlations between some biochemical factors such as between TG and TG/ HDL ratio (r=0.81, P<0.0001) and between TC and non HDL cholesterol (r=0.93, P<0.0001). 3.5. Correlations between expression levels of genes There were significant pairwise correlations between expression levels of all genes. The most robust correlations were detected ATG5 and Chast (r=0.81, P<0.0001) and between DICER1-AS1 and HULC (r=0.75, P<0.0001) (Fig. 5). We detected no difference in gene-gene correlation between CAD populations and healthy controls. 4. Discussion Although the protective role of autophagy against CAD has been acknowledged, most of studies focusing on enhancement of this process have been conducted in the context of cancer. However, increasing evidences show that autophagy is a potentially attractive therapeutic target in cardiopathology as well (Lavandero et al., 2015; Towers and Thorburn, 2016). Autophagy in cardiomyocytes is involved in both main stages of
myocardial injury namely ischemia and reperfusion (Matsui et al., 2007), so therapeutic options for enhancement of autophagy are expected to influence both mentioned phases. An important shortcoming in translating investigational autophagy-targeted modalities into treatment of cardiopathology is the challenge in the evaluation of autophagy in cardiovascular system in the clinical situation (Schiattarella and Hill, 2016). In the current study, we have assessed expression of autophagy-related lncRNAs in the peripheral blood of CAD patients with the supposition that peripheral blood tissue reflects expression of these genes in the coronary arteries as the main pathological location of CAD. Notably, we detected significant down-regulation of ATG5, Chast, HULC and DICER1-AS1 in peripheral blood of CAD patients compared with healthy individuals. ATG5 as a main regulator of autophagy has been shown to be implicated in cardiopathology as well. Impairment of autophagy in Atg5+/- mice has resulted in over-production of Reactive Oxygen Species (ROS), which stimulates nuclear factor κB (NF-κB) pathway in immune cells leading to inflammatory responses in cardiac tissue (Zhao et al., 2014). Meanwhile, ATG5 has a possible role in regulation of glucose tolerance. A previous animal study has shown that transgenic over-expression of Atg5 in mice leads to significantly decreased blood glucose levels, which might be due to higher glucose tolerance or augmented glucose clearance by Atg5. Moreover, Atg5 transgenic mice had higher insulin sensitivity after insulin stimulation (Pyo et al., 2013). Others have shown that dietary glucose has a direct influence on autophagy in mammalian cells. Glucose scarcity stimulates autophagy mostly through AMPK activation and the succeeding suppression of mTORC1 (Moruno et al., 2012). Yet, dysregulated autophagy participates in diabetes and its complications as autophagy controls the typical function of pancreatic β cells and insulintarget tissues (Bhattacharya et al., 2018). Animal studies have shown a crucial role for autophagy in the maintenance of normal structural design of pancreatic β cells (Jung et al., 2008). However, signs of reformed autophagy have been detected in the β cells of the Zucker diabetic fatty rat
and in a β cell line after sustained exposure to high glucose (Kaniuk et al., 2007). Thus, autophagy, high blood glucose and CAD are associated with each other in a complicated way that identification of cause-effect relationship or even the direction of effect is not simple. Only through elaborate functional studies, one can deduce the exact mechanism. In line with these evidences, we detected inverse correlations between FBS levels and expression of autophagy-related lncRNAs including ATG5, Chast, DICER1-AS1 and HULC in the context of CAD. However, as no previous study has addressed the functional role of these lncRNAs in CAD, more data is needed to elaborate the mechanism of the observed associations. LncRNAs usually regulate autophagy through changing transcript levels of ATG genes. Mainly, they act as competing endogenous RNAs to alter expression of microRNAs (miRNAs) involved in autophagy (Yang et al., 2017). However, Chast has been shown to decrease expression of Pleckstrin homology domain-containing protein family M member 1 and inhibit cardiomyocyte autophagy while inducing hypertrophy in these cells (Viereck et al., 2016). Opposite to our expectation, we detected lower levels of Chast in CAD patients compared with healthy subjects. Based on the role of Chast in cardiac remodeling (Viereck et al., 2016), we hypothesize that decreased peripheral levels of this lncRNA is not reflective of its levels in damaged cardiac tissues. Remodeling of the coronary arteries is a process being initiated in the endothelium and gradually spreading towards the atherosclerotic plaque and myocardium (Heusch et al., 2014). HULC promotes autophagy through inhibition of miR-6825-5p, miR-6845-5p and miR-6886-3p and subsequent upregulation of USP22 expression. The final effect of this lncRNA is deacetylation of ATG5 and ATG7 (Xiong et al., 2017). The role of DICER1-AS1 in enhancement of autophagy has been demonstrated in osteosarcoma cells as its silencing has inhibited autophagy in vitro. Such effect has been accompanied by suppression of ATG5, LC3-II and
Beclin 1 protein levels (Gu et al., 2018). Consequently, in the present study, we demonstrated down-regulation of three pro-autophagy genes in peripheral blood of CAD patients. Moreover, we detected down-regulation of an lncRNA whose function is associated with cardiac remodeling. Notably, our results showed appropriateness of HULC and DICER1-AS1 transcript levels for diagnosis of CAD. Such autophagy-related lncRNAs might be putative markers for assessment of autophagy in CAD patients. Future studies are needed to elaborate whether their transcript levels are modulated by autophagy-targeted therapies. We also detected inverse correlations between expression levels of pro-autophagy genes and FBS levels which imply the role of high FBS in suppression of autophagy. Moreover, expressions of DICER1-AS1 and HULC were inversely correlated with TG levels. It is possible that some detrimental effects of chronic high FBS and TG are mediated through suppression of autophagy in cardiac tissues. Animal experiments have shown induction of autophagy in beta cells of pancreas by either a high-fat regimen or a combined high-fat and high-glucose regimen, but not by only high-glucose diet (Sheng et al., 2017). Differences between human and rodents or tissue-specific responses to high glucose and high fat diets should be considered in explanation of these conflicting results. In addition, we detected significant pairwise correlations between expression levels of all genes especially between ATG5 and Chast and between DICER1-AS1 and HULC. The positive correlation between ATG5 and Chast might imply the presence of a feedback loop for regulation of autophagy which decreases the level of an anti-autophagy gene in response to robust suppression of autophagy. Taken together, the current study further verified associations between dysregulation of autophagy and CAD. Moreover, our results indicate appropriateness of two autophagy-related lncRNAs for differentiation of CAD status. Our study had limitations regarding lack of assessment of
expression of genes in atherosclerotic plaques, lack of validation experiments in vitro and in vivo and lack of evidence to prove the direct or sole correlation between the autophagy related lncRNAs and CAD. Authors' contribution NE, MT performed the experiments. NE and MDO conceived and designed the study. SGF and MDO supervised the study and wrote the manuscript. SAJ analyzed the data. SAP assessed the clinical data and supervised patients' enrollment. All authors contributed in the study design and approved the final manuscript. Acknowledgement The current study was supported by a grant from Shahid Beheshti University of Medical Sciences and was conducted as the Ph.D. thesis project of the first author. Conflict of interest None to be declared Availability of data The data that support the findings of this study are available from the corresponding author upon reasonable request. References Bentzon Jacob, F., Otsuka, F., Virmani, R., Falk, E., 2014. Mechanisms of Plaque Formation and Rupture. Circulation Research 114, 1852-1866.
Bhattacharya, D., Mukhopadhyay, M., Bhattacharyya, M., Karmakar, P., 2018. Is autophagy associated with diabetes mellitus and its complications? A review. EXCLI journal 17, 709-720. Chen, S., Wu, D.D., Sang, X.B., Wang, L.L., Zong, Z.H., Sun, K.X., Liu, B.L., Zhao, Y., 2017. The lncRNA HULC functions as an oncogene by targeting ATG7 and ITGB1 in epithelial ovarian carcinoma. Cell Death Dis 8, e3118. Galkina, E., Ley, K., 2009. Immune and inflammatory mechanisms of atherosclerosis (*). Annual review of immunology 27, 165-197. Gu, Z., Hou, Z., Zheng, L., Wang, X., Wu, L., Zhang, C., 2018. LncRNA DICER1-AS1 promotes the proliferation, invasion and autophagy of osteosarcoma cells via miR-30b/ATG5. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie 104, 110-118. Heusch, G., Libby, P., Gersh, B., Yellon, D., Böhm, M., Lopaschuk, G., Opie, L., 2014. Cardiovascular remodelling in coronary artery disease and heart failure. The Lancet 383, 1933-1943. Jung, H.S., Chung, K.W., Won Kim, J., Kim, J., Komatsu, M., Tanaka, K., Nguyen, Y.H., Kang, T.M., Yoon, K.H., Kim, J.W., Jeong, Y.T., Han, M.S., Lee, M.K., Kim, K.W., Shin, J., Lee, M.S., 2008. Loss of autophagy diminishes pancreatic beta cell mass and function with resultant hyperglycemia. Cell Metab 8, 318-324. Kaniuk, N.A., Kiraly, M., Bates, H., Vranic, M., Volchuk, A., Brumell, J.H., 2007. Ubiquitinated-protein aggregates form in pancreatic betacells during diabetes-induced oxidative stress and are regulated by autophagy. Diabetes 56, 930-939. Lavandero, S., Chiong, M., Rothermel, B.A., Hill, J.A., 2015. Autophagy in cardiovascular biology. The Journal of clinical investigation 125, 55-64.
Liao, X., Sluimer, J.C., Wang, Y., Subramanian, M., Brown, K., Pattison, J.S., Robbins, J., Martinez, J., Tabas, I., 2012. Macrophage autophagy plays a protective role in advanced atherosclerosis. Cell metabolism 15, 545-553. Matsui, Y., Takagi, H., Qu, X., Abdellatif, M., Sakoda, H., Asano, T., Levine, B., Sadoshima, J., 2007. Distinct roles of autophagy in the heart during ischemia and reperfusion: roles of AMP-activated protein kinase and Beclin 1 in mediating autophagy. Circ Res 100, 914-922. Moruno, F., Perez-Jimenez, E., Knecht, E., 2012. Regulation of autophagy by glucose in Mammalian cells. Cells 1, 372-395. Organization, W.H., 17 May 2017. Cardiovascular diseases (CVDs). Retrieved from: https://www.who.int/cardiovascular_diseases/en/. Pyo, J.O., Yoo, S.M., Ahn, H.H., Nah, J., Hong, S.H., Kam, T.I., Jung, S., Jung, Y.K., 2013. Overexpression of Atg5 in mice activates autophagy and extends lifespan. Nature communications 4, 2300. Razani, B., Feng, C., Coleman, T., Emanuel, R., Wen, H., Hwang, S., Ting, J.P., Virgin, H.W., Kastan, M.B., Semenkovich, C.F., 2012. Autophagy links inflammasomes to atherosclerotic progression. Cell metabolism 15, 534-544. Schiattarella, G.G., Hill, J.A., 2016. Therapeutic targeting of autophagy in cardiovascular disease. Journal of molecular and cellular cardiology 95, 86-93. Sciarretta, S., Yee, D., Nagarajan, N., Bianchi, F., Saito, T., Valenti, V., Tong, M., Del Re, D.P., Vecchione, C., Schirone, L., 2018. Trehaloseinduced activation of autophagy improves cardiac remodeling after myocardial infarction. Journal of the American College of Cardiology 71, 1999-2010.
Shao, B.Z., Han, B.Z., Zeng, Y.X., Su, D.F., Liu, C., 2016. The roles of macrophage autophagy in atherosclerosis. Acta pharmacologica Sinica 37, 150-156. Sheng, Q., Xiao, X., Prasadan, K., Chen, C., Ming, Y., Fusco, J., Gangopadhyay, N.N., Ricks, D., Gittes, G.K., 2017. Autophagy protects pancreatic beta cell mass and function in the setting of a high-fat and high-glucose diet. Scientific reports 7, 16348. Tabas, I., Bornfeldt, K.E., 2016. Macrophage Phenotype and Function in Different Stages of Atherosclerosis. Circ Res 118, 653-667. Towers, C.G., Thorburn, A., 2016. Therapeutic targeting of autophagy. EBioMedicine 14, 15-23. Viereck, J., Kumarswamy, R., Foinquinos, A., Xiao, K., Avramopoulos, P., Kunz, M., Dittrich, M., Maetzig, T., Zimmer, K., Remke, J., Just, A., Fendrich, J., Scherf, K., Bolesani, E., Schambach, A., Weidemann, F., Zweigerdt, R., de Windt, L.J., Engelhardt, S., Dandekar, T., Batkai, S., Thum, T., 2016. Long noncoding RNA Chast promotes cardiac remodeling. Science translational medicine 8, 326ra322. Xiong, H., Ni, Z., He, J., Jiang, S., Li, X., He, J., Gong, W., Zheng, L., Chen, S., Li, B., Zhang, N., Lyu, X., Huang, G., Chen, B., Zhang, Y., He, F., 2017. LncRNA HULC triggers autophagy via stabilizing Sirt1 and attenuates the chemosensitivity of HCC cells. Oncogene 36, 3528-3540. Yang, L., Wang, H., Shen, Q., Feng, L., Jin, H., 2017. Long non-coding RNAs involved in autophagy regulation. Cell Death Dis 8, e3073. Ye, X., Zhou, X.J., Zhang, H., 2018. Exploring the Role of Autophagy-Related Gene 5 (ATG5) Yields Important Insights Into Autophagy in Autoimmune/Autoinflammatory Diseases. Frontiers in immunology 9, 2334. Zhang, T., Qi, Y., Liao, M., Xu, M., Bower, K.A., Frank, J.A., Shen, H.-M., Luo, J., Shi, X., Chen, G., 2012. Autophagy Is a Cell Self-Protective Mechanism Against Arsenic-Induced Cell Transformation. Toxicological Sciences 130, 298-308.
Zhang, Y.J., Yang, S.H., Li, M.H., Iqbal, J., Bourantas, C.V., Mi, Q.Y., Yu, Y.H., Li, J.J., Zhao, S.L., Tian, N.L., Chen, S.L., 2014. Berberine attenuates adverse left ventricular remodeling and cardiac dysfunction after acute myocardial infarction in rats: role of autophagy. Clinical and experimental pharmacology & physiology 41, 995-1002. Zhao, W., Li, Y., Jia, L., Pan, L., Li, H., Du, J., 2014. Atg5 deficiency-mediated mitophagy aggravates cardiac inflammation and injury in response to angiotensin II. Free radical biology & medicine 69, 108-115. Figure legends Fig. 1. Distribution of age and biochemical factors in cases and controls (P<0.001 for FBS and >0.05 in other comparisons; The central line shows the median value, while the box contains the 25th to 75th percentiles. The black whiskers mark the 5th and 95th percentiles, and values beyond these upper and lower bounds are considered outliers, marked with red + signs). Fig. 2. Relative expressions of ATG5 (P<0.0001), Chast (P=0.011), HULC (P<0.0001) and DICER1-AS1 (P<0.0001) in CAD patients and healthy subjects (The central line shows the median value, while the box contains the 25th to 75th percentiles. The black whiskers mark the 5th and 95th percentiles, and values beyond these upper and lower bounds are considered outliers, marked with red + signs). Fig. 3. ROC curves for assessment of diagnostic power of ATG5 (A), Chast (B), DICER1-AS1 (C) and HULC (D) in differentiation of CAD patients from healthy subjects (True positive rate (Sensitivity) is plotted in function of the false positive rate (100-Specificity) for different cutoff points of expression levels of genes).
Fig. 4. Correlations between expression levels of ATG5 (A), Chast (B), DICER1-AS1 (C) and HULC (D) genes and biochemical factors (The diagonal displays the distribution of certain parameter. The left diagonal has a scatterplot of the two variables overhead it and to its right; the right diagonal designs the correlation coefficient for the same variables, with the font size escalating for larger correlations). Fig. 5. Pairwise correlations between expression levels of autophagy related genes (The diagonal displays the distribution of expression of a gene. The left diagonal has a scatterplot of the two variables overhead it and to its right; the right diagonal designs the correlation coefficient for the same variables, with the font size escalating for larger correlations).