DOI: 10.1161/CIRCULATIONAHA.114.012285
Upregulation of MG53 Induces Diabetic Cardiomyopathy via Transcriptional Activation of PPAR-Į
Running title: Liu et al.; MG53 Induces Diabetic Cardiomyopathy via PPAR-Į Fenghua Liu, PhD1,2*; Ruisheng Song, PhD1,2*#; Yuanqing Feng, BS1,2*; Jiaojiao Guo, BS1,2; Yanmin Chen, BS2,4; Yong Zhang, BS1,2; Tao Chen, BS3; Yanru Wang, BS1,2; Yanyi Huang, PhD3; Chuan-Yun Li, PhD1,4; Chunmei Cao, MD, PhD1,2; 1,2,4,5 ,2,4, ,4, 4,5 5 Yan Zhang, MD, PhD1,2; Xinli Hu, PhD1,2; Rui-ping Xiao, MD, PhD1,2
1
Institute of Molecular Medicine; 2State Key Laboratory of Biomembrane and Membrane
Biotechnology; Biot Bi otec ot echn ec hnol hn olog ol o y;; 3B Biodynamic iodynamic Optical Imaging C Center; enter; 4Center for L en Life iffe Sc S Sciences; iences; 5Beijing City Key Ke Laboratory Laboorato to ory of of Cardiometabolic Card Ca rdio rd i me io meta tabo ta b li bo licc Molecular Mole Mo lecu le c la cu l rM Medicine, edi diici cine ne,, Pe ne Peking eking ng U University, nive ni v rs rsit ity, it y, B Beijing, e ji ei jing ng, Ch ng Chin China inaa in *contributted d eq qually y; #C Curre rent ntt aaddress: ddresss:: De ept off Bi Biol olog ol oggicaal C hem mistrry aand nd Mo Mole lecu cu ular *contributed equally; #Current Dept Biological Chemistry Molecular Ph har armaaco c lo logy gy,, Ha gy Harv rvar rv ardd Me ar Medi ica c l Sc Scho hool ho ol, Bost ol B ost ston on,, MA on Pharmacology, Harvard Medical School, Boston,
Add Ad dress ffor or C orrespondence: d Address Correspondence: Rui-Ping Xiao, MD, PhD or Xinli Hu, PhD Institute of Molecular Medicine, Peking University 5 Yiheyuan Road Beijing 100871 China Tel: 011-86-10-6275-7243 Fax: 011-86-10-6276-7143 E-mail
[email protected] or
[email protected] Journal Subject Codes: Hypertension:[16] Myocardial cardiomyopathy disease, Basic science research:[130] Animal models of human disease, Atherosclerosis:[140] Energy metabolism, Atherosclerosis:[142] Gene expression 1 Downloaded from http://circ.ahajournals.org/ at CONS CALIFORNIA DIG LIB on February 1, 2015
DOI: 10.1161/CIRCULATIONAHA.114.012285
Abstract Background—Diabetic cardiomyopathy, which contributes to more than 50% diabetic death, is featured by myocardial lipid accumulation, hypertrophy, fibrosis, and cardiac dysfunction. The mechanism underlying diabetic cardiomyopathy is poorly understood. Recent studies have shown that a striated muscle-specific E3 ligase Mitsugumin 53 (MG53, or TRIM72) constitutes a primary causal factor of systemic insulin resistance and metabolic disorders. Although it is most abundantly expressed in myocardium, the biological and pathological roles of MG53 in triggering cardiac metabolic disorders remain elusive. Methods and Results—Here we show that cardiac-specific transgenic expression of MG53 induces diabetic cardiomyopathy in mice. Specifically, MG53 transgenic mouse develops severe y p y at 20 weeks of age, g , as manifested by y insulin resistance,, compromised p diabetic cardiomyopathy glucose uptake, increased lipid accumulation, myocardial hypertrophy, fibrosis, aand nd car ardi ar diac di ac cardiac dysfunction. Overexpression of MG53 leads to insulin resistant via destabilizing insulin receptor and insulin receptor recept ptor substrate 1. More importantly, importantly y, we identified a novel role of MG53 in ran ansscri scriptio iona nal up na upre regu gula latiion off pe pero oxiso s me m prol o if ifeera erationn ac acti t vateed rec cep epto tor al lph phaa (P ((PPAR-Į) PA ARR Į) and transcriptional upregulation peroxisome proliferation-activated receptor alpha ts target ta genes es, resulting resu sult ltin ingg in lipid lip ipid id accumulation acc ccuumulatio i n and io annd nd lipid lip pid d toxicity, tox xicit ityy, thereby the here reeby y ccontributing onttrib on tribut u in ut ingg to o ddiabetic iaabeetic iabe its genes, ca ard rdio i my io m oppat athy hy.. cardiomyopathy. Conc nclu lusion ions— —Ou Ourr re result ltss su ugg ggesst that ooverexpression vere ve rexp xpre ressio on off m yocard yo rdia iall MG MG53 53 iiss su ssufficient fficcie ientt tto o Conclusions—Our results suggest myocardial nduce diabe etiic ca card rdio rd iomy io m oppat my athy hy vvia iaa ddual uall me ua mech chan ch anis an isms is ms iinvolving nvvollvi v ng n upr preg pr egul eg ulat ul atio at i n of P io PAR PA R-Į and induce diabetic cardiomyopathy mechanisms upregulation PPAR-Į impairment of insulin signaling. These findings not only reveal a novel function of MG53 in regulating cardiac PPAR-Į gene expression and lipid metabolism, but also underscore MG53 as an important therapeutic target for diabetes and associated cardiomyopathy. Key words: MG53, MG53, peroxisome proliferator-activated receptor alpha, peroxisome proliferator-activated receptor alpha, diabetic cardiomyopathy, insulin resistance
2 Downloaded from http://circ.ahajournals.org/ at CONS CALIFORNIA DIG LIB on February 1, 2015
DOI: 10.1161/CIRCULATIONAHA.114.012285
Introduction Diabetes is an emerging global threat to human health. It is estimated that the total number of people suffering from diabetes is reaching 366 million in 20301. While hypertension and coronary narrowing are major pathogenic factors of acquired cardiomyopathy, diabetes has been established as an independent risk factor since 19722. Diabetic cardiomyopathy, as a major complication, is the leading cause of morbidity and mortality for diabetic patients. Epidemiological studies have demonstrated that diabetic people have a 2-5 folds increase of risk in developing heart failure compared with age-matched healthy subjects after adjusting for age, blood pressure, weight, cholesterol level and coronary artery disease3-7. Rodent models are of particular value in study of diabetic cardiomyopathy hy y because bec ecau ause au se there there her iss minimal confounding involvement of coronary atherosclerosis. It has been demonstrated that diabetic di iab bet etic ic mice mic icee and an nd rats ra fed with high fat diet (HFD (HFD) FD) display cardiac FD cardiaac remodeling remo moddeling and dysfunction8mo 11 1
. Recently Re w wee and and others ottheerss have hav avee sh shown how ownn that that mice mice ov overex overexpressing expr ex preesssing singg Mitsugumin Mit itsu suugu umiin 53 ((MG53) MG53 MG 5 ) 53
develop dev de velo velo lopp obesity, obbes esit ity, y,, systemic syyste ystemi miic insulin inssuli in suli linn resistance, resi re sist sttan nce ce,, ddyslipidemia, y liipiideemi ys mia, a, aand nd hhyperglycemia, yper yp ergl er glyccem gl mia ia,, wh while hil ile MG MG53 53 deficiency prevents preeve vent ntss HFD-induced nt HFDHF D in indu duce du ceed metabolic m taabo me boli licc disorders, li d so di sord rdder ers, s including inc nclu ludi lu diing obesity, obe besi s ty si ty,, dyslipidemia dysl dy sllip ipid idem id emia em i and hyperglycemia12, 13. MG53 was originally shown to play an important role in membrane repair14, 15
. We have shown that MG53 acts as an E3 ligase targeting insulin receptor (IR) and insulin
receptor substrate1 (IRS1) for ubiquitin-dependent degradation, resulting in insulin resistance and metabolic disorders12. However, it is possible that MG53 regulates not only insulindependent glucose metabolism but also lipid metabolism. In addition, we and others have reported that MG53 is abundantly expressed in the heart as well as skeletal muscle and involved in ischemia-mediated cardiac preconditioning and postconditioning16, 17. But it is unclear whether MG53 plays a role in the regulation of myocardial glucose and lipid metabolism. In particular,
3 Downloaded from http://circ.ahajournals.org/ at CONS CALIFORNIA DIG LIB on February 1, 2015
DOI: 10.1161/CIRCULATIONAHA.114.012285
we seek to determine the potential function of MG53 in regulating myocardial lipid metabolism and its malfunction in the pathogenesis of diabetic cardiomyopathy. To understand the exact role of MG53 in the myocardial metabolism, in this study we have generated transgenic mice with cardiac-specific overexpression of MG53 (MG53 h-TG) via alpha myosin heavy chain (Į-MHC) promoter. The MG53 h-TG mice developed typical diabetic cardiomyopathy symptoms characterized by myocardial insulin resistance, defective substrate utilization, cardiac fibrosis, ventricular hypertrophy, and cardiac dysfunction. Mechanistically, in addition to the compromised insulin signaling, the expression levels of PPAR-Į and its downstream target genes were markedly increased in the MG53 h-TG hearts, while downregulated contributes egulated in the MG53-deficient hearts. These findings suggest that MG53 contr rib ibut utess tto ut o th thee pathogenesis of diabetic cardiomyopathy via, at least in part, activation of PPAR-Į signaling pathway. pa ath hwa way. y.
Material Mate Mate teriial and nd Methods Mettho hod ds ds Animals Anim An imal im alss al All animal procedures were carried out in compliance with the protocols approved by the Institute of Animal Care and Use Committee of Peking University, and in accordance with the Guide for the Care and Use of Laboratory Animals (NIH publication No. 86-23, revised 1985). Generation of the MG53 h-TG mice The MG53 transgenic construct was generated by cloning the MG53cDNA into an expression vector containing the Į-MHC promoter (a gift from Dr. Jeff Robbins). The MG53 h-TG construct was injected into one-cell embryos of C57BL/6 mice. The transgenic founders were further crossed with C57BL/6. Age matched male mice of MG53 h-TG and their wild type littermates (WT, C57BL/6) were used in this study. 4 Downloaded from http://circ.ahajournals.org/ at CONS CALIFORNIA DIG LIB on February 1, 2015
DOI: 10.1161/CIRCULATIONAHA.114.012285
Echocardiography Mice were anesthetized with 4% chloral hydrate (1ml/100g body weight) and echocardiography assessment was performed using a VEVO-2100 machine (Visual Sonics) with an M-mode Doppler. In vitro palmitate uptake assay Neonatal rat ventricular myocytes (NRVMs) were infected with Ad-MG53-GFP or Ad-MG53Myc virus 24h prior to palmitate stimulation. Cells in the control groups were infected with AdGFP or Ad-ȕ-gal, respectively. Before images were obtained, NRVMs were incubated in DMEM with 2% BSA with or without 100ȝM palmitate for 6h. Then, lipid uptake was determined using a stimulated Raman scattering (SRS) system18, 19. In vitro glucose uptake assay Male Ma le adult adu dult lt rat rat a ventricular ven enttricular tr myocytes (ARVMs) infected infec nf cted with either eitheer Ad Adv-GFP dvv-G GFP or Adv-MG53-GFP w were erre re incubated ed with with Kreb’s-Ringer K eb Kr eb’s ’s-R ’s -Riinge -R ingeer phosphate phhossphaatee buff buffer ffeer ((128mM 12 28mM NaCl, NaCl, l 1.4mM 1.44mM M CaCl CaC Cl2, 1. 11.4mM .4m 4m mM Mg MgSO gSO4, 5.2mM 5.2m 5. 2mM 2m M KC K KCl, Cl, l, aand nd 10m nd 10mM 0mM 0m M Na2HP HPO O4, pH7.4) pH H7. 7.4) 4) co containing onta onta tain inin in ng gluc gglucose luc ucos o e (3 os (3mM), 3mM mM), ), aand nd the then heen tr trea treated eaateed with 100nM iinsulin nsul ns ullin ffor o 330 or 0 mi min at a 337°C, 7 C, aand 7° nd eexposed xpos xp osed os ed tto o 222-[1,2[1,2 [1 ,2--3H] ,2 H]-deoxy-D-glucose H -deo -d eoxy eo xy-D xy -D D-g glu luco cose co se ((1μCi/ml, 1μCi/ml, 0.02μmol/L) for the final 5 min. Then, the cells were washed three times with ice-cold PBS and solubilized with 0.5M NaOH. The cell associated radioactivity was determined by scintillation counting. Non-specific counts, determined in the presence of 20μM cytochalasin B, were subtracted from each value as previously described20. Real-time PCR Total RNA was isolated with Trizol reagent (Invitrogen) and 2ȝg was reverse-transcribed into cDNA using RT-MLV reverse transcriptase (Promega). The RT reaction mixture was used as template to perform real-time PCR (Stepone Plus Real-Time PCR System, Applied Biosystems).
5 Downloaded from http://circ.ahajournals.org/ at CONS CALIFORNIA DIG LIB on February 1, 2015
DOI: 10.1161/CIRCULATIONAHA.114.012285
The relative mRNA levels were determined by normalizing to the 18S rRNA level. The primers used are listed in Supplementary Table 1. Small-interfering RNA NRVMs were transfected with small-interfering RNA (siRNA) (RiboBio Co. Ltd.) specifically targeting PPAR-Į using Lipofectamine RNAi MAX (Invitrogen) according to the manufacturer’s instructions. The sequences of siRNAs against PPAR-Į are listed in Supplementary Table 2. RNA-Seq Total RNA was extracted from MG53 h-TG or WT mice. RNA-Seq libraries were prepared from four hearts from each group. Next-generation sequencing was performed on an Illumina RNA-Seq HiSeq2000 sequencing system according to the manufacturer's instructions. RNA NA A-S Seq q rreads eads ea ds were mapped to the mouse genome (version mm9) by TopHat (version v2.0.8), according to 2 23 22, computational co omp mput utat atio tio iona nal pipelines na piipe peli l nes as reported previously211. Tools Tools such as DAVID D VI DA VID D22 and Qiagen’s
Ingenuity ngeenu n ity® Pathway Pat athw hwaay hw ay Analysis Anaalysi lysi siss ((IPA ((IP ((IP PA®, QI QIAG QIAGEN GEN Redwood Redwoo wood od City, Cit i y, www.qiagen.com/ingenuity) www ww..qiageen. n.co com m/in m/in inge geenuuity) ity) y were weere r used use sedd for for functional func fu ncti nc tion on nall enrichment enr nric ichm ic hmen hm entt analyses en an nal alys ysees and ys and pathway pat a hw hway ay network net etwo work rk analyses, analy lyses ysess, re rrespectively. spec sp ectivvelly. ec ly ChIP-PCR ChIP assay was conducted using a ChIP kit (Millipore). All the procedure was performed following the protocol provided by the manufacturer. Primers used for ChIP-PCR were listed in Supplementary Table 3. Statistical Analysis Data are presented as mean±SEM. Statistical analysis was performed with prism5.0 or SPSS 17.0. Data sets were tested for normality of distribution with Kolmogorov-Smirnov tests. Data groups with normal distribution were compared using unpaired Student t test. The MannWhitney U test was used for nonparametric data. Comparisons between multiple groups were
6 Downloaded from http://circ.ahajournals.org/ at CONS CALIFORNIA DIG LIB on February 1, 2015
DOI: 10.1161/CIRCULATIONAHA.114.012285
assessed by 1-way ANOVA with Bonferroni post hoc analysis. For the oxygen consumption experiment, repeated measures ANOVA was used to compare the oxygen consumption rate between the wild type and MG53 h-TG cardiac myocytes over the time course. A p0.5 and P
