DOI QR코드

DOI QR Code

Alteration of the Metabolome Profile in Endothelial Cells by Overexpression of miR-143/145

  • Wang, Wenshuo (Department of Cardiac Surgery, Zhongshan Hospital of Fudan University and Shanghai Institute of Cardiovascular Diseases) ;
  • Yang, Ye (Department of Cardiac Surgery, Zhongshan Hospital of Fudan University and Shanghai Institute of Cardiovascular Diseases) ;
  • Wang, Yiqing (Department of Cardiothoracic Surgery, Huashan Hospital of Fudan University) ;
  • Pang, Liewen (Department of Cardiothoracic Surgery, Huashan Hospital of Fudan University) ;
  • Huang, Jiechun (Department of Cardiothoracic Surgery, Huashan Hospital of Fudan University) ;
  • Tao, Hongyue (Department of Radiology, Huashan Hospital of Fudan University) ;
  • Sun, Xiaotian (Department of Cardiothoracic Surgery, Huashan Hospital of Fudan University) ;
  • Liu, Chen (Department of Cardiac Surgery, Zhongshan Hospital of Fudan University and Shanghai Institute of Cardiovascular Diseases)
  • Received : 2015.07.31
  • Accepted : 2015.11.16
  • Published : 2016.03.28

Abstract

Communication between endothelial cells (ECs) and smooth muscle cells (SMCs) via miR-143/145 clusters is vital to vascular stability. Previous research demonstrates that miR-143/145 released from ECs can regulate SMC proliferation and migration. In addition, a recent study has found that SMCs also have the capability of manipulating EC function via miR-143/145. In the present study, we artificially increased the expression of miR-143/145 in ECs, to mimic a similar change caused by miR-143/145 released by SMCs, and applied untargeted metabolomics analysis, aimed at investigating the consequential effect of miR-143/145 overexpression. Our results showed that miR-143/145 overexpression alters the levels of metabolites involved in energy production, DNA methylation, and oxidative stress. These changed metabolites indicate that metabolic pathways, such as the SAM cycle and TCA cycle, exhibit significant differences from the norm with miR-143/145 overexpression.

Keywords

Introduction

The endothelial cell (EC) is the major cell type in the vascular intima, playing an important role in maintaining vessel biology and hemostasis [6]. Its dysfunction can lead to the development of vascular diseases, such as atherosclerosis [19]. Endothelial behavior is closely related to angiogenesis, such as proliferation, migration, and basement membrane degeneration [26]. Blood vessels generated from the differentiation of EC progenitors are associated with a process called vasculogenesis [17]. Once vessels form, ECs experience tissue-specific modifications, making vessels functionally different from each other [4].

The maturation process of the endothelium is affected by signals from other neighboring cell types. In fact, the formation of a vascular structure involves an interaction between ECs and surrounding cells, including smooth muscle cells (SMCs) and pericytes. Under physiological conditions, the association between these cells results in vascular maturation and stabilization [1,16]. The interaction between cells depends on various growth factors that show different profiles in physiological and pathological conditions.

MicroRNAs (miRNAs, miRs) are 22~26-nucleotide-long noncoding RNAs, derived from a relatively long precursor through Dicer catalysis, which guide inhibitory complexes to specific mRNAs [3]. MicroRNAs have been proven to participate in various physiological and pathological processes, such as those of vessels [8,12]. Previous studies have provided solid evidence that miR-143/145 can be transmitted from ECs to SMCs and regulate the phenotype and function of the SMCs [14]. For example, miR-143/145 were found to have collaboratively targeted a network of transcription factors, including Elk-1 (ELK1, a member of the ETS oncogene family), myocardin, and Kruppel-like factor 4, to promote the stabilization of SMCs [9]. Interestingly, a recent study demonstrates that miR-143/145 released by SMCs can affect the proliferation as well as differentiation of ECs, and then modulate vessel stabilization [7]. However, the mechanism behind the effect of miR-143/145 on ECs remains unclear.

Metabolism is involved in nearly all aspects of cellular functioning, directly or indirectly. Metabolome profiling, a method to screen small molecule production of biochemical processes, has been used to acquire information about various metabolic pathways [23]. Here, we investigated the metabolome of ECs via the overexpression of miR-143/145, thereby demonstrating the metabolic changes that occur in cellular communication and attempting to shed light on the hitherto unclear link between microRNAs and metabolic pathways.

 

Materials and Methods

Materials

All standards for identification of metabolites and normalization were procured from Sigma-Aldrich (St. MO, USA). Acetonitrile, ultrapure water, formic acid, ammonium fluoride and EDTA were also acquired from Sigma-Aldrich. Dulbecco modified Eagle’s medium (DMEM), dialyzed fetal bovine serum (FBS), and penicillin-streptomycin (PS) were sourced from Gibco (Life Technologies, NY, USA).

Lentiviral Transfection of Endothelial Cells

Endothelial cells were plated and cultured in a 6-well plate with DMEM containing 10% FBS and 1% PS until 30-40% confluence. After that, the cells were infected overnight with 2 ml/well lentiviral supernatant containing the EGFP gene and miR-143/145 precursor or negative control in the presence of 8 μg/ml polybrene. The next day, the viral supernatant was replaced with fresh growth medium and incubated at 37℃ and 5% CO2. At 72 h after incubation, the cells were observed at an excitation wavelength of 488 nm and the percentage of successfully infected cells was estimated.

PCR Assay

To evaluate the relative concentration of miR-143/45, Trizol reagent (Invitrogen) was used to extract the total RNA. To quantitate miR-143 and miR-145 concentrations, an NCode miRNA First Strand cDNA synthesis kit (Invitrogen) was applied to polyadenylate and the total RNA was reversely transcribed. Quantitative real-time PCR (qPCR) was performed with SYBR Green PCR master mix (Applied Biosystems) on an ABI 7300 system. The primers were CCCTCTAACACCCCTTCTCC (miR-143 forward), TCTCAGACTCCCAACTGACCA (miR-143 reverse), CCAGAGGGTTTCCGGTACTT (miR-145 forward), and CGGATG TGGCTTATTGCTCT (miR-145 reverse). Every sample was normalized according to the internal control, U6 snRNA. Fold changes were calculated using the method of relative quantification (2-ΔΔCt).

Metabolite Extraction

On the day of harvest, the medium was carefully aspirated from the wells with transfected ECs (miR-143/145 mimic or blank plasmid) and the cells were washed. Then 1 ml of precooled (4℃) ultrapure water with 1 mM HEPES and 1 mM EDTA was added into the wells. After that, the cells were scratched from the plates, sonicated for 1 min, and treated with three freeze-thaw cycles using liquid nitrogen. Next, the samples were cooled for 1 h at −20℃, followed by 20 min centrifugation at 13,000 rpm (4℃). HPLC vials were utilized to collect the supernatant.

LC-MS/ MS Analysis

Analysis was mainly performed by the method of Benton et al. [5]. In brief, an Agilent 1290 liquid chromatography system (Agilent Technologies, Santa Clara, CA, USA) and an AB 5600 TripleTOF were used to quantity and identify the metabolites. A 2.1 mm × 50 mm Agilent Eclipse Plus-C18 1.8 μm particle column with a 2.1 mm × 30 mm Agilent Zorbax SB-C18 3.5 μm particle guard column was equipped to separate the different metabolites. The separation was achieved under a column temperature of 40℃ using a controlled gradient of mobile phase A, which consisted of 0.1% (v/v) formic acid in water, and mobile phase B, composed of 0.1% formic acid in acetonitrile, at a flow rate of 0.4 ml/min. The gradient flow was first set at 5% (v/v) B for 2 min, linearly increased to 95% B over 11 min, and maintained at this composition for an additional 2 min. The flow rate was set at 50 μl/min and the injection volume of the sample was 5 μl.

ESI source configurations for TripleTOF were set as follows: ion source gas 1 (GS1) as 35, ion source gas 2 (GS2) as 35, curtain gas (CUR) as 30, source temperature as 550℃, and ionspray voltage floating as −4,500 V in negative mode. The instrument was set to acquire over the m/z range 50-1,000 Da for TOF MS scans and the m/z range 25-1,000 Da for the production of ion scans in auto MS/MS acquisition. The accumulation time for TOF MS scans and the production ion scans were set at 0.25 sec/spectra and 0.05 sec/spectra, respectively, and the cycle time was 1 sec. The production ion scan was based on information-dependent acquisition, and was triggered when the full-scan experiment detected small molecules with m/z values between 50 and 1,000 Da. The collision energy of the production ion scan was set at −30 V with ±15 V spread, and the declustering potential was set at −100 V.

Data Processing and Statistical Analysis

For PCR data, the data were shown as the mean values ± SEM and a t-test was used to evaluate the statistical significance using SPSS 22.0 software (SPSS, Inc., Chicago, IL, USA) and p < 0.05 was considered significantly different. On the part of the metabolomics profile, the raw data were converted to mzXML files and uploaded to XCMS online for analysis. The optimized model for AB 5600 TripleTOF was applied to all parameters needed during the process of analysis. All samples were normalized based on the total area. The data, having undergone quantification and normalization, were imported into SIMCA-p 12.0 software (Umetrics AB, Umea, Sweden) for multivariate analysis. Principal component analysis (PCA) and orthogonal partial least-squares discriminant analysis (OPLS-DA) were applied to discover the global differences between the two cell groups. The variable importance in the projection (VIP) value for each identified metabolite was calculated by the validated OPLS-DA model. Meanwhile, the t-test for each metabolite was performed via SPSS 22.0 software to acquire the p-values. The screen criteria of VIP > 1 and p < 0.05 were adopted for the final selection of a panel of metabolites. The exact mass (m/z), retention time, and MS/MS data of significantly altered metabolites were then compared with both M E TLIN (http://metlin.scripps.edu/) and the Human Metabolome Project (http://redpoll.pharmacy.ualberta.ca/hmdb/HMDB/) for identification.

 

Results

Metabolome Profile of Endothelial Cells With or Without Overexpression of miR-143/145

Lentiviral transfection successfully introduced target genes into the endothelial cells (Fig. 1A) and significantly increased the expression of miR-143/145 (Fig. 1B). A total of 738 metabolites were detected in both the overexpression and control groups. The metabolome signature of cells with miR-143/145 overexpression was different from that of the control. A total of 25 metabolites differed between the overexpression group and control group, accounting for 3.52% of the total metabolites detected (Figs. 1C and 1D). Many members belonging to biochemical families showed the same trend of change with 12 nucleic acid catabolites, 5 citric acid cycle and glycolysis components, 4 amino acid catabolites, and 3 free fatty acids having significantly changed in the overexpression group. These data demonstrate that a change in miR-143/145 expression can alter the endothelial metabolome profile.

Fig. 1.miR-143/145 change the metabolome profile of ECs. (A) GFP expression in ECs. The blue fluorescence represents the nuclei of endothelial cells, and the green fluorescence was emitted from the GFP. (B) The alteration of miR-143/145 expression after transfection. (C) The heat map of metabolite features in two groups. The data are shown in log 2 scale and clustered by Spearman distance. (D) The percentage of metabolite feature differences caused by miR-143/145 overexpression. Error bars represent the Standard Error of Mean (SEM). * p < 0.05.

Metabolome Differences Caused by Overexpression of miR-143/145 Reveal Significant Pathways

Aiming to explore further the metabolites that had changed with miR-143/145 overexpression, we monitored the metabolome of EC with miR-143/145 overexpression. Each data group of feature intensities was displayed by a PCA plot (Fig. 2A), so that the relative positions between each data point indicated the total similarities between the two cell types. The PCA plot revealed two separate clusters of points, indicating that the metabolome profiles of these two cell populations were strikingly different.

Fig. 2.Metabolic differences between the overexpression and control groups indicate the new pathways vital for cellular communication. (A) Similar cell populations were closer than dissimilar cell populations in PCA score. Each group contained six biological replicates. Dmodx (distance to the model of X-space) is a method that measures the fitness of observations by the PCA model. (B) Metabolites mainly related to oxidative stress decreased in the overexpression group as identified via accurate mass and tandem MS data. Fold values that compare the median full peak intensities are listed. (C) Metabolites mainly associated with the SAM cycle were increased in the overexpression group compared with the control group.

After finding differences in the metabolome profiles, we identified metabolites that were significantly different between the two cell groups. We identified several metabolites that had a greater than 2-fold difference plus a p-value <0.01 between endothelial cells transfected with miR-143/145 precursor and negative control, respectively. In these metabolites, we discovered that the relative concentrations of unsaturated fatty acid metabolites, such as linoleic acid, adrenic acid, and arachidonic acid, had decreased, when compared with the control, by 80.04%, 88.87%, and 84.80%, respectively. This comparatively lower level resulting from the overexpression of miR-143/145 suggested that changes in the metabolites had occurred after manipulating the oxidative pathways (Fig. 2B). An increased level of linoleic acid has been proven to augment oxidative stress and then lead to endothelial damage [24]. Adrenic acid, an elongated product of arachidonic acid, performs as a surrogate for oxidative damage to myelin membranes [22]. Oxidative stress is one of the most important causes of inflammation and even apoptosis. In addition, our data also indicated changes in metabolites associated with the S-adenosyl methionine (SAM) cycle, such as S-adenosylmethionine and 5’-methylthioadenosine. 5’-Methylthioadenosine is a derivative of SAM, and its concentration is associated with an alteration of the SAM cycle [10]. Specifically, the levels of these metabolites rose as the expression of miR-143/145 increased (Fig. 2C). Considering that transmethylation reactions like DNA methylation require the methyl group of SAM as the most common substrate, we speculate that miR-143/145 play a role in manipulating the epigenetic changes in endothelial cells.

Fig. 3.Quantification of metabolites related to the metabolism of nucleic acids and the TCA cycle. Data points and bars show the integrated peak intensity for the respective samples and the median intensity value for each group. Fold values indicate the difference in median values for the chosen metabolites in the overexpression and control groups. The identification of metabolites was performed based on accurate mass (m/z), retention time, and MS/MS data. Each group was analyzed with six biological and experimental replicates.

Variation in Metabolites Associated with Cellular Respiration Represents the Effect of miR-143/145 Overexpression

Apart from the above eight metabolites, we identified 17 more metabolites that led ECs with miR-143/145 overexpression to be different from that without overexpression. Metabolites with a significant difference were structurally identified on the basis of accurate mass, retention time, and MS/MS data. These metabolites mainly belong to the nucleoside phosphate family and the tricarboxylic acid family. The pattern of change in nucleoside phosphate was indicative of restricted RNA synthesis. We speculate that this change is associated with the increased level of DNA methylation and that it explains the low level of EC proliferation and migration under conditions of miR-143/ 145 overexpression. The metabolites involved in the tricarboxylic acid cycle, such as citric acid and malic acid, exhibited the same trends as nucleoside phosphate; however, the fold values were relative low when compared with nucleoside phosphate. Considering the TCA cycle as the major source of energy, it is reasonable to hypothesize that the miR-143/145 altered the conditions for energy production. However, it is not clear whether miR-143/145 had a direct effect on the proteins participating in the TCA cycle, or if miR-143/145 reduced the cellular demand for energy, resulting in a corresponding decrease in energy production. Since ECs rely on glycolysis to obtain energy support, we subsequently presume that an alteration in glycolytic metabolism might occur under conditions of miR-143/145 overexpression.

 

Discussion

To our best knowledge, this research is the first to have explored the metabolic profile related to the overexpression of miR-143/145. We applied an untargeted metabolomics method with liquid chromatography plus TripleTOF MS to study the relative abundance of metabolites in ECs with different expression of miR-143/145. The altered pattern of metabolism influenced by the overexpression of miR-143/ 145 was characterized by the variation in metabolites involved in oxidative stress, cellular respiration, transmethylation, and energy production. These changed metabolites indicate that metabolic pathways, such as the SAM and TCA cycles, exhibit significant differences. This study provides new insight into how an EC reacts with high levels of miR-143/ 145 caused by cellular communication.

The alteration of cell metabolism is vital for ECs with elevated miR-143/145 expression, as suggested by the results that metabolites related to transmethylation, cellular respiration, and energy production are all found to influence the subsequent reactions after cellular communication. In a recent study, miR-143/145 were found to suppress the EC’s capacity to proliferate and form vessel-like structures [7]. Our findings bridge the gap between genes and cellular behavior, as the changes in metabolites tell us the association between the bioenergetic state and endothelial function. We speculate the existence of a mechanism whereby miR-143/145 regulate the amount of energy that is essential for ECs to divide and differentiate, and then modulate the intima thickness, which mainly depends on the number and morphology of ECs.

Notably, a similar mechanism exists in tumor biology. Many studies have demonstrated that miR-143/145 are abnormally downregulated in cancer development [2,15], coupled with the resultant malformation of vessels, characterized by immature ECs with a high proliferation index and lost SMCs, frequently observed in pathological examinations [13]. The metabolites we identified as being significant in the consequence of miR-143/145 overexpression have also been proven to play a key role in tumor angiogenesis [17]. For example, metabolites involved in glycolysis have become a current focus in anticancer research [18,25]. All these studies agree with our hypothesis that metabolic alteration is the bridge between miR-143/ 145 and endothelial function.

Moreover, epigenetic changes in EC modulate angiogenesis [21]. Our results showed that the level of SAM changed owing to miR-143/145 overexpression. It has been proven that SAM can suppress angiogenesis through participation in the maintenance of genome stability, especially the modification of the status of angiogenesis-related genes [20]. Meanwhile, epigenetic inhibitors can decrease the blood supply to tumor tissue by reversing abnormal demethylation of VEGF and other genes that accelerate EC proliferation [11]. Thus, we surmise that miR-143/145 regulate gene modification via changes in SAM levels, thereby inhibiting EC proliferation.

Our work explored the metabolic mechanism for miR-143/145’s modulation of endothelial function, shedding light on the interaction between ECs and their surrounding cells, such as SMC. These findings may have relevant implications for cardiovascular disease and cancer.

References

  1. Abramsson A, Lindblom P, Betsholtz C. 2003. Endothelial and nonendothelial sources of PDGF-B regulate pericyte recruitment and influence vascular pattern formation in tumors. J. Clin. Invest. 112: 1142-1151. https://doi.org/10.1172/JCI200318549
  2. Akao Y, Nakagawa Y, Kitade Y, Kinoshita T, Naoe T. 2007. Downregulation of microRNAs-143 and -145 in B-cell malignancies. Cancer Sci. 98: 1914-1920. https://doi.org/10.1111/j.1349-7006.2007.00618.x
  3. Bartel DP. 2004. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116: 281-297. https://doi.org/10.1016/S0092-8674(04)00045-5
  4. Benjamin LE, Hemo I, Keshet E. 1998. A plasticity window for blood vessel remodelling is defined by pericyte coverage of the preformed endothelial network and is regulated by PDGF-B and VEGF. Development 125: 1591-1598.
  5. Benton HP, Ivanisevic J, Mahieu NG, Kurczy ME, Johnson CH, Franco L, et al. 2014. Autonomous metabolomics for rapid metabolite identification in global profiling. Anal. Chem. 87: 884-891. https://doi.org/10.1021/ac5025649
  6. Bonetti PO, Lerman LO, Lerman A. 2003. Endothelial dysfunction: a marker of atherosclerotic risk. Arterioscler. Thromb. Vasc. Biol. 23: 168-175. https://doi.org/10.1161/01.ATV.0000051384.43104.FC
  7. Climent M, Quintavalle M, Miragoli M, Chen J, Condorelli G, Elia L. 2015. TGFβ triggers miR-143/145 transfer from smooth muscle cells to endothelial cells, thereby modulating vessel stabilization. Circ. Res. 116: 1753-1764. https://doi.org/10.1161/CIRCRESAHA.116.305178
  8. Condorelli G, Latronico MV, Cavarretta E. 2014. MicroRNAs in cardiovascular diseases: current knowledge and the road ahead. J. Am. Coll. Cardiol. 63: 2177-2187. https://doi.org/10.1016/j.jacc.2014.01.050
  9. Cordes KR, Sheehy NT, White MP, Berry EC, Morton SU, Muth AN, et al. 2009. miR-145 and miR-143 regulate smooth muscle cell fate and plasticity. Nature 460: 705-710.
  10. Du Z, Zhou Y, Lu X, Li L, Lu C, Li L, et al. 2016. Octreotide prevents liver failure through upregulating 5’-methylthioadenosine in extended hepatectomized rats. Liver Int. 36: 212-222. https://doi.org/10.1111/liv.12863
  11. Ebos JM, Lee CR, Cruz-Munoz W, Bjarnason GA, Christensen JG, Kerbel RS. 2009. Accelerated metastasis after short-term treatment with a potent inhibitor of tumor angiogenesis. Cancer Cell 15: 232-239. https://doi.org/10.1016/j.ccr.2009.01.021
  12. Elia L, Quintavalle M, Zhang J, Contu R, Cossu L, Latronico MV, et al. 2009. The knockout of miR-143 and -145 alters smooth muscle cell maintenance and vascular homeostasis in mice: correlates with human disease. Cell Death Differ. 16: 1590-1598. https://doi.org/10.1038/cdd.2009.153
  13. Goel S, Duda DG, Xu L, Munn LL, Boucher Y, Fukumura D, Jain RK. 2011. Normalization of the vasculature for treatment of cancer and other diseases. Physiol. Rev. 91: 1071-1121. https://doi.org/10.1152/physrev.00038.2010
  14. Hergenreider E, Heydt S, Tréguer K, Boettger T, Horrevoets AJ, Zeiher AM, et al. 2002. Atheroprotective communication between endothelial cells and smooth muscle cells through miRNAs. Nat. Cell Biol. 14: 249-256. https://doi.org/10.1038/ncb2441
  15. Iorio MV, Ferracin M, Liu CG. 2005. MicroRNA gene expression de-regulation in human breast cancer. Cancer Res. 65: 7065-7070. https://doi.org/10.1158/0008-5472.CAN-05-1783
  16. Jain RK. 2003. Molecular regulation of vessel maturation. Nat. Med. 9: 685-693. https://doi.org/10.1038/nm0603-685
  17. Monjazeb AM, High KP, Connoy A. 2006. Arachidonic acid-induced gene expression in colon cancer cells. Carcinogenesis 27: 1950-1960. https://doi.org/10.1093/carcin/bgl023
  18. Pelicano H, Martin DS, Xu RH. 2006. Glycolysis inhibition for anticancer treatment. Oncogene 25: 4633-4646. https://doi.org/10.1038/sj.onc.1209597
  19. Ross R. 1999. Atherosclerosis — an inflammatory disease. N. Engl. J. Med. 340: 115-126. https://doi.org/10.1056/NEJM199901143400207
  20. Sahin M, Sahin E, Gumuslu S, Erdogan A, Gultekin M. 2010. DNA methylation or histone modification status in metastasis and angiogenesis-related genes: a new hypothesis on usage of DNMT inhibitors and S-adenosylmethionine for genome stability. Cancer Metastasis Rev. 29: 655-676. https://doi.org/10.1007/s10555-010-9253-0
  21. Shiva Shankar TV, Willems L. 2014. Epigenetic modulators mitigate angiogenesis through a complex transcriptomic network. Vascul. Pharmacol. 60: 57-66. https://doi.org/10.1016/j.vph.2014.01.003
  22. Sonnen JA, Gray S, Larson EB, Montine TJ. 2008. Oxidative damage to human cortex in aging an antioxidant use. FASEB J. 22:167-162.
  23. Sreekumar A, Poisson LM, Rajendiran TM. 2009. Metabolome profiles delineate potential role for sarcosine in prostate cancer progression. Nature 457: 910-914. https://doi.org/10.1038/nature07762
  24. Turpeinen A, Basu S, Mutanen M. 1998. A high linoleic acid diet increases oxidative stress in vivo and affects nitric oxide metabolism in humans. Prostaglandins Leukot. Essent. Fatty Acids 59: 229-233. https://doi.org/10.1016/S0952-3278(98)90067-9
  25. Vander Heiden MG, Cantley LC, Thompson CB. 2009. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324: 1029-1033. https://doi.org/10.1126/science.1160809
  26. Velazquez OC. 2007. Angiogenesis and vasculogenesis: inducing the growth of new blood vessels and wound healing by stimulation of bone marrow-derived progenitor cell mobilization and homing. J. Vasc. Surg. 45: 39-47. https://doi.org/10.1016/j.jvs.2007.02.068