|Home | About | Journals | Submit | Contact Us | Français|
We identified a novel muscle-restricted putative coiled-coil protein, MURC, which is evolutionarily conserved from frog to human. MURC was localized to the cytoplasm with accumulation in the Z-line of the sarcomere in the murine adult heart. MURC mRNA expression in the heart increased during the developmental process from the embryonic stage to adulthood. In response to pressure overload, MURC mRNA expression increased in the hypertrophied heart. Using the yeast two-hybrid system, we identified the serum deprivation response (SDPR) protein, a phosphatidylserine-binding protein, as a MURC-binding protein. MURC induced activation of the RhoA/ROCK pathway, which modulated serum response factor-mediated atrial natriuretic peptide (ANP) expression and myofibrillar organization. SDPR augmented MURC-induced transactivation of the ANP promoter in cardiomyocytes, and RNA interference of SDPR attenuated the action of MURC on the ANP promoter. Transgenic mice expressing cardiac-specific MURC (Tg-MURC) exhibited cardiac contractile dysfunction and atrioventricular (AV) conduction disturbances with atrial chamber enlargement, reduced thickness of the ventricular wall, and interstitial fibrosis. Spontaneous episodes of atrial fibrillation and AV block were observed in Tg-MURC mice. These findings indicate that MURC modulates RhoA signaling and that MURC plays an important role in the development of cardiac dysfunction and conduction disturbance with increased vulnerability to atrial arrhythmias.
The heart is constantly exposed to biomechanical and neurohumoral stress, even under physiological conditions, and increased stress on the heart leads to hypertrophy and apoptosis of cardiomyocytes and ultimately heart failure (HF). Atrial fibrillation (AF) is one of the most common arrhythmias that produces substantially excess cardiovascular morbidity and mortality (13). Clinically, increased vulnerability to AF is associated with underlying heart disease, such as valvular heart disease, HF, coronary artery disease, and hypertension, particularly when left ventricular hypertrophy is present (13). The Z-disc in striated muscle constitutes an anchoring site for actin, titin, and nebulin filaments and plays a critical role in muscle structure and function, including sarcomeric assembly and organization, sarcolemmal membrane integrity, and muscle force generation and transmission (8). The Z-disc connects to the costamere, the basement membrane at periodic membrane-associated plaques, which serves to transmit force from the Z-disc to the sarcolemma and extracellular matrix (9). In addition to its role in muscle contraction, the Z-disc works as a biomechanical sensor that can respond to changes in tension in the sarcolemma. Various signaling molecules have been identified as components of the Z-disc, and a large number of the Z-disc-associated proteins have a dynamic distribution in muscle cells and shuttle between the Z-disc and other subcellular locations to transmit signals (8, 22, 35). Thus, the Z-disc is not only simply the structural border of the sarcomere but also functions in sensing and transmitting external and internal signals. Since the Z-disc is a multiprotein complex, the identification of the precise molecular mechanisms of the Z-disc and its role in signaling has become critical for understanding the regulation of cardiac function and the design of therapeutic strategies to prevent the progression to HF.
Rho GTPases are molecular switches that control a variety of signaling pathways in eukaryotic cells (10). The Rho family GTPase RhoA controls the formation of actin structures, and the RhoA-actin signaling pathway regulates serum response factor (SRF) transcriptional activity. SRF regulates serum-inducible and muscle-specific gene expression by binding to the serum response element (SRE) (also referred to as the CArG box). ROCK (Rho-kinase) is one of the downstream effectors of Rho GTPases (2, 32). In the heart, overexpression of RhoA resulted in sinus and atrioventricular (AV) nodal dysfunction, AF, and ventricular contractile failure with chamber enlargement and interstitial fibrosis (37). On the other hand, inhibition of Rho family protein activities by overexpression of Rho GDP dissociation inhibitor α resulted in an AV block with atrial enlargement and ventricular hypertrophy (47). These results suggest that fine-tuning of Rho GTPase signaling is required for maintaining cardiac rhythm, conduction, and structure.
In this study, we sought to identify novel cardiac-restricted molecules that participate in cardiac homeostasis and pathogenesis. We identified MURC (muscle-restricted coiled-coil protein) and characterized its function in cardiomyocytes and hearts.
Total RNA was extracted from cells or tissues using an RNeasy Mini kit (Qiagen) or Trizol reagent (Invitrogen). Total RNA of human heart was purchased from Ambion. cDNA synthesis and kinetic real-time PCR were performed as described previously (33). Primers used were as follows: mouse MURC (mMURC) forward primer (5′-ACAGTCACACAGCAATACGGGCTA-3′) and mMURC reverse primer (5′-TTCTCGGGCAGGCTTCTGTCTTTA-3′); mouse atrial natriuretic peptide (ANP) forward primer (5′-AACCTGCTAGACCACCTGGA-3′) and mouse ANP reverse primer (5′-TGCTTTTCAAGAGGGCAGAT-3′); mouse brain natriuretic peptide (BNP) forward primer (5′-CTGAAGGTGCTGTCCCAGAT-3′) and mouse BNP reverse primer (5′-CCTTGGTCCTTCAAGAGCTG-3′); mouse serum deprivation response (mSDPR) forward primer (5′-ATGAGGAAGCCCTGGAAGAT-3′) and mSDPR reverse primer (5′-CCCAGATGATGCTTTCTGGT-3′); mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH) forward primer (5′-TTGTGATGGGTGTGAACCACGAGA-3′) and mouse GAPDH reverse primer (5′-CATGAGCCCTTCCACAATGCCAAA-3′); mouse α-myosin heavy chain (αMHC) forward primer (5′-GAGGACCAGGCCAATGAGTA-3′) and mouse αMHC reverse primer (5′-GCTGGGTGTAGGAGAGCTTG-3′); mouse β-myosin heavy chain (βMHC) forward primer (5′-TCGATTTGGGAAATTCATCC-3′) and mouse βMHC reverse primer (5′-CGCATAATCGTAGGGGTTGT-3′); mouse sarcoplasmic reticulum Ca2+ ATPase 2 (SERCA2) forward primer (5′-CTGTGGAGACCCTTGGTTGT-3′) and mouse SERCA2 reverse primer (5′-CAGAGCACAGATGGTGGCTA-3′); mouse transforming growth factor β1 (TGF-β1) forward primer (5′-TTGCTTCAGCTCCACAGAGA-3′) and mouse TGF-β1 reverse primer (5′-TGGTTGTAGAGGGCAAGGAC-3′); mouse TGF-β2 forward primer (5′-CAGCGCTACATCGATAGCAA-3′) and mouse TGF-β2 reverse primer (5′-CCTCGAGCTCTTCGCTTTTA-3′); mouse TGF-β3 forward primer (5′-GATGAGCACATAGCCAAGCA-3′) and mouse TGF-β3 reverse primer (5′-ATTGGGCTGAAAGGTGTGAC-3′); mouse procollagen type I α1 (Col1a1) forward primer (5′-GAGCGGAGAGTACTGGATCG-3′) and mouse Col1a1 reverse primer (5′-GCTTCTTTTCCTTGGGGTTC-3′); mouse Col1a2 forward primer (5′-CCGTGCTTCTCAGAACATCA-3′) and mouse Col1a2 reverse primer (5′-GAGCAGCCATCGACTAGGAC-3′); mouse Col3a1 forward primer (5′-GTCCACGAGGTGACAAAGGT-3′) and mouse Col3a1 reverse primer (5′-GATGCCCACTTGTTCCATCT-3′); rat MURC (rMURC) forward primer (5′-ACTGAAGATGAAGACCAGGACGCA-3′) and rMURC reverse primer (5′-TGTTAACAACGTAGCCCGTGTTGC-3′); rat ANP forward primer (5′-ATACAGTGCGGTGTCCAACA-3′) and rat ANP reverse primer (5′-CGAGAGCACCTCCATCTCTC-3′); rat BNP forward primer (5′-GGAAATGGCTCAGAGACAGC-3′) and rat BNP reverse primer (5′-CGATCCGGTCTATCTTCTGC-3′); and rat GAPDH forward primer (5′-ATGGGAAGCTGGTCATCAAC-3′) and rat GAPDH reverse primer (5′-GTGGTTCACACCCATCACAA-3′).
Serial analysis of gene expression (SAGE) was performed as described previously (33). The SAGE libraries were constructed essentially following the I-SAGE long kit protocol (Invitrogen) using total RNA extracted from adult mouse hearts. Double-stranded cDNAs were digested with NlaIII, and the restriction enzyme was replaced by MmeI after linker ligation. Ditags produced from 400 PCRs were isolated, cleaved with NlaIII, and cloned into pZErO. All sequence files were processed using SAGE2000 version 4.5 software. The extracted tags were further processed to determine the identity of associated genes through several stringent filters using the CGAP website (http://cgap.nci.nih.gov/SAGE).
The corresponding cDNA fragments for human MURC (hMURC), mMURC, and rMURC were cloned by PCR from human heart cDNA, mouse heart cDNA, and rat cardiomyocyte cDNA templates, respectively. PCR was performed using the following primers: hMURC forward primer (5′-ATGGAACATAATGGGTCTGC-3′) and hMURC reverse primer (5′-TTACGATGAGTGCTTTAAATCTAAC-3′); mMURC forward primer (5′-ATGGAACACAACGGATCAGCT-3′) and mMURC reverse primer (5′-CTATTTGTAGTCTGAGGACTGCTTTAGCTCCA-3′); rMURC forward primer (5′-ATGGAACACAATGGATCTGC-3′) and rMURC reverse primer (5′-CTATGAGGACTGCTTTAATTCCAAC-3′). The cDNAs encoding mMURC and hMURC with a C-terminal Flag epitope were cloned into pcDNA3 (Invitrogen) to generate pcDNA3-mMURC and pcDNA3-hMURC, respectively. The corresponding cDNA fragment for human SDPR (hSDPR) was cloned by PCR from human heart cDNA template. PCR was performed using hSDPR forward primer (5′-ATGGGAGAGGACGCTGACAGGC-3′) and hSDPR reverse primer (5′-TCACGGCAGTCTGATCCACAT-3′). The cDNA encoding hSDPR with a C-terminal hemagglutinin (HA) epitope was cloned into pcDNA3 to generate pcDNA3-hSDPR. The cDNA encoding hMURC was cloned into pGBKT7 (Clontech) to generate a bait vector, pGBKT7-hMURC. The RNA interference (RNAi) target sequences for rMURC (5′-TTCGAGTAACCAAAGTCGAAA-3′), rat SDPR (rSDPR, 5′-GAAGCAGTGTGTACAGGTGAA-3′), and green fluorescent protein (GFP; 5′-CGTAAACGCCCACAAGTTC-3′) were cloned into the BamHI-EcoRI sites of the RNAi-Ready-pSIREN-RetroQ vector (Clontech) as an inverted repeat with a hairpin loop spacer to generate RNAi-Ready-pSIREN-RetroQ-rMURC, RNAi-Ready-pSIREN-RetroQ-rSDPR, and RNAi-Ready-pSIREN-RetroQ-GFP (used as a control), respectively.
Total RNA was isolated from tissues with Trizol reagent (Invitrogen). Total RNA was size fractionated by electrophoresis in a 1.3% agarose gel containing 2.2 M formaldehyde and transferred to nylon membranes. A HindIII fragment of pcDNA3-mMURC (nucleotides 1 to 797 of the mMURC open reading frame) was used as a probe.
Rabbit immunization was conducted by Medical & Biological Laboratories Co., Ltd. (Nagoya, Japan) using synthetic peptides spanning fragments of mMURC with N-terminal acetylation (GERLRQSGERFKKSISC). For immunostaining and Western blot analysis, immunoglobulin G (IgG) was purified from antisera with protein A-Sepharose beads.
Specimens were fixed in 4% paraformaldehyde and stained with rabbit polyclonal anti-MURC antibody, mouse monoclonal anti-α-actinin antibody (Sigma), mouse monoclonal anti-α-smooth muscle actin antibody (Sigma), rat monoclonal anti-mouse CD31 antibody (BD Biosciences), rat monoclonal anti-HA antibody (Roche), mouse monoclonal anti-Flag antibody (Sigma), or rabbit polyclonal anti-Flag antibody (Sigma). Secondary antibodies were conjugated with Alexa Fluor 488, 555, or 594 (Invitrogen), and nuclei were visualized using 4′,6-diamino-2-phenylindole (DAPI; Invitrogen).
Rat neonatal cardiomyocytes, cultured from 1-day-old Sprague-Dawley rats, were prepared as described previously with slight modifications (42, 44). Briefly, ventricles were digested enzymatically, and cardiomyocytes were purified over a Percoll gradient. The culture medium was changed to serum-free medium after 24 h. Neonatal cardiomyocytes were cultured under serum-free conditions for 24 h before experiments. Adult cardiomyocytes were isolated from the hearts of male C57BL6 mice at 8 weeks of age. Ventricles were minced roughly and subsequently placed into 0.05% trypsin-EDTA (Invitrogen) at 4°C. After overnight incubation, ventricular tissue segments were put into Hanks' balanced salt solution buffer (Invitrogen) containing 0.1% (wt/vol) type 2 collagenase (Worthington Biochemical Corporation) at 37°C for 2 h. Dissociated adult cardiomyocytes were plated on 0.1% gelatin-coated slides and cultured in Dulbecco's modified Eagle's medium-F-12 medium with 5% bovine serum.
Male mice were anesthetized with 2,2,2-tribromoethanol (0.25 mg/g of body weight; Aldrich). A midline abdominal incision was used to expose the suprarenal abdominal aorta. The aorta was tied with a 6-0 silk suture against a blunt needle (26 gauge). The needle was immediately removed, leaving the aortic lumen constricted to the diameter of the needle. Sham-operated mice were subjected to the same procedure without the aortic banding. Seven days after surgery, mice were sacrificed and total RNA was extracted from the heart.
A Saccharomyces cerevisiae two-hybrid screen was performed using a Matchmaker Gal 4 two-hybrid system 3 (Clontech) and a Matchmaker pretransformed human heart cDNA library (Clontech) according to the manufacturer's instructions.
The cDNAs encoding mMURC with a C-terminal Flag epitope and hSDPR with a C-terminal HA epitope were inserted into a pAxCAwtit cosmid vector in an adenovirus expression vector kit (Dual Version; Takara Bio Inc., Otsu, Japan). RNAi-Ready-pSIREN-RetroQ-rMURC and RNAi-Ready-pSIREN-RetroQ-luciferase (Clontech), which was used as a control, were digested with BglII and EcoRI to obtain the U6 promoter with each of the target sequences, and blunted fragments were inserted into a promoterless pAxcwit cosmid vector with an adenovirus expression vector kit (Dual Version). Recombinant adenoviruses expressing Flag-tagged mMURC (Ad-MURC), HA-tagged hSDPR (Ad-SDPR), LacZ (Ad-LacZ), MURC shRNA (Ad-rMURC shRNA), and Luc shRNA (Ad-Luc shRNA) were generated as described previously (43). Twenty-four hours after seeding, cardiomyocytes were infected with Ad-MURC, Ad-SDPR, Ad-LacZ, Ad-rMURC shRNA, or Ad-Luc shRNA diluted in the culture medium at a multiplicity of infection (MOI) of 10 or 20 and incubated at 37°C for 1 h. The viral suspension was removed, and cardiomyocytes were cultured with serum-depleted culture medium. Phenylephrine (PE) or Y-27632 was added after infection.
COS cells were plated in 60-mm dishes. The following day, the cells were transfected with 1 μg of pcDNA3-hMURC and/or pcDNA3-hSDPR. The total plasmid amount was adjusted to 2.0 μg with an empty vector plasmid. Cardiomyocytes were plated in 60-mm dishes. The following day, the cells were infected with Ad-MURC and/or Ad-SDPR. The total MOI was adjusted to 20 with Ad-LacZ. Cells were cultured for another 48 h and lysed with a lysis buffer (20 mM HEPES, pH 7.7, 100 mM NaCl, 5 mM MgCl2, 1% Nonidet P-40, 0.1 mM Na3VO4, 2 μg/ml aprotinin, 0.7 μg/ml pepstatin A, 0.1 mM phenylmethylsulfonyl fluoride, and 1 mM dithiothreitol). Cell lysates were incubated with anti-Flag M2 affinity gel (Sigma) or an anti-HA antibody and protein A-Sepharose beads (GE Healthcare) at 4°C. After the beads were extensively washed with the lysis buffer, the bound proteins were eluted by boiling the beads in sodium dodecyl sulfate (SDS) sample buffer and subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE), followed by Western blot analysis.
Cell lysates were extracted with a lysis buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 1× protease inhibitor cocktail (Pierce), 1 mM Na3VO4, and 1 mM NaF. Cell lysates were electrophoresed in 10% SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes (Millipore). Membranes were subsequently incubated with primary antibodies against Flag (Sigma), HA, and MURC. Horseradish peroxidase-conjugated anti-rabbit IgG, anti-rat IgG, and anti-mouse IgG (GE Healthcare) were used as secondary antibodies.
RhoA activity was determined from protein isolated from adenovirus-infected neonatal cardiomyocytes or heart tissues of transgenic mice using an absorbance-based G-LISA RhoA activation assay biochemistry kit (Cytoskeleton) according to the manufacturer's instructions. Cell protein was isolated on day 2 postinfection using the provided cell lysis buffer. Heart tissue protein of mice at 6 weeks of age was also isolated by homogenizing in cell lysis buffer. Cell or tissue lysates were clarified by centrifugation at 10,000 rpm at 4°C for 2 min, and these extracts were processed rapidly on ice and snap-frozen until the time of assay. Protein concentration was determined according to the manufacturer's protocol, and extracts were equalized to contain total protein concentrations of 1 mg/ml for the assay. Signals were measured at an absorbance of 490 nm using a microplate spectrometer as suggested by the manufacturer.
COS cells were plated in six-well plates. The following day, the cells were transfected with 200 ng of ANP luciferase reporter construct (−638 ANP Luc; kindly provided by Kenneth R. Chien, Massachusetts General Hospital, Boston, MA) or no-SRE1/SRE2 mutant in the ANP luciferase reporter construct (kindly provided by Andrew Thorburn, Wake Forest University School of Medicine, NC). The cells were also transfected with 1.0 μg of expression vectors containing hMURC, mMURC, RhoA Val14, hSDPR, GFP shRNA, or rSDPR shRNA, 500 ng of a C3 expression vector, and 200 ng of pTKβ-Gal using FuGene6 reagent (Roche). The total plasmid amount was adjusted to 1.9 μg with an empty vector plasmid. An expression vector containing RhoA V14 (the mutant of Gly to Val at codon 14), which is a point-mutated active form of RhoA, was kindly provided by Yoshimi Takai (Osaka University, Suita, Japan). A C3 expression vector was kindly provided by Seigo Izumo (Novartis Institutes for Biomedical Research, Cambridge, MA). Cells were cultured for another 48 h, lysed with 200 μl of reporter lysis buffer (Promega), and assayed for luciferase activity (by using a Promega assay) and β-galactosidase activity. Luciferase activity was normalized against β-galactosidase activity.
Ad-LacZ- or Ad-MURC-infected cardiomyocytes were incubated in serum-free medium. Y-27632 was added after infection. After 96 h of 10 μM Y-27632 treatment, cells were fixed in 4% paraformaldehyde and stained with fluorescein isothiocyanate-conjugated phalloidin (Sigma) for the detection of actin filaments. Ad-Luc shRNA- or Ad-rMURC shRNA-infected cardiomyocytes were incubated in serum-free medium at 37°C for 48 h prior to stimulation with 100 μM PE. After 48 h of PE stimulation, cells were fixed and stained with fluorescein isothiocyanate-conjugated phalloidin.
The cDNA encoding Flag-tagged mMURC was cloned into the third 5′-untranslated exon of αMHC promoter plasmid clone 26 (a generous gift from Jeffrey Robbins, Cincinnati Children's Hospital Medical Center, Cincinnati, OH) (16), and transgenic mice were generated as described previously (25). All of the aspects of animal care and experimentation performed in this study were approved by the Institutional Animal Care and Use Committee of Kyoto University.
Echocardiographic examination of mice was performed as described previously (25, 26, 39). Briefly, mice were anesthetized with 2,2,2-tribromoethanol (0.20 mg/g), and M-mode recordings of the left ventricle (LV) were obtained at the level of the papillary muscles from a parasternal window using a Hewlett-Packard (Andover) Sonos 5500 equipped with a 12-MHz probe. Mice anesthetized with 2,2,2-tribromoethanol (0.20 mg/g) by intraperitoneal injection were analyzed by multilead-surface electrocardiogram (ECG). ECG recordings were performed using an ECG-9902 (Nihon Kohden, Tokyo, Japan).
For left ventricular catheterization with a 1.0-F high-fidelity micromanometer-tipped catheter (model SPR-1000; Millar Instruments), mice were anesthetized with 2,2,2-tribromoethanol (0.25 mg/g), subsequently anesthetized with isoflurane (2.5% [vol/vol] in O2), and placed on a 36.5° table. A microtip catheter was inserted into the LV via the right carotid artery. To determine cardiac contractile and diastolic function, signals for LV pressure and the maximal rates of systolic pressure increase (maximum dP/dt) and isovolumetric relaxation (minimum dP/dt) were recorded at 1,000 Hz and analyzed by using a PowerLab system (AD Instruments).
Total RNA isolated from LV tissue was extracted using an RNeasy Mini kit and treated with the DNase I (Qiagen) according to the manufacturer's instructions. DNA microarrays used in these experiments were produced using the Mouse Genome Oligo 4.0 set of 70-mer oligonucleotides (Operon). Hybridization, processing, and the scanning process were done by Filgen Incorporated (Nagoya, Japan). Scan data images were analyzed using Microarray Data Analysis Tool version 2.0 software (Filgen).
All experiments were performed at least three times. Data are expressed as means ± standard errors and were analyzed by one-way analysis of variance with post hoc analysis. A P value of <0.05 was considered significant.
To identify unknown cardiac-specific genes, we performed SAGE (45) using RNA isolated from the hearts of adult mice and searched expressed sequence tag databases for novel sequences found only in cardiac cDNA libraries. Sequences that did not correspond to known genes were then used as probes for Northern blot analysis of adult mouse tissues, in order to confirm their cardiac-specific expression. One of the cDNAs identified in this screen appeared to be a muscle-restricted gene, and the extended cDNA sequence encoded a novel protein of 362 amino acids (Fig. (Fig.1).1). We then isolated human and rat orthologues of this gene by using human heart and rat cardiomyocyte cDNAs. Northern blot analysis of RNA from adult mouse tissues revealed abundant expression of this transcript in the heart and skeletal muscle (Fig. (Fig.2A).2A). Weaker signals were found in the aorta and lung, suggesting expression of this transcript in vascular smooth muscle cells. We refer to this gene as MURC (muscle-restricted coiled-coil protein), because the predicted open reading frame of MURC contains a putative coiled-coil motif. Database searches revealed that hMURC is encoded by a gene located on human chromosome 9q31. A cDNA encoding a MURC-like protein was also identified in the Xenopus laevis expressed sequence tag database (accession number NM_001097894). These findings suggest that MURC is an evolutionarily conserved protein in vertebrates.
To identify a cell source expressing MURC in the heart, we separated cardiomyocytes from noncardiomyocytes using a rat primary culture and performed real-time RT-PCR. As shown in Fig. Fig.2B,2B, the MURC transcript was found to be much more abundant in cardiomyocytes than in noncardiomyocytes, indicating that the major source of MURC in the heart is cardiomyocytes. We then performed immunostaining of adult mouse heart sections with an anti-MURC antibody to investigate the subcellular localization of MURC in cardiomyocytes. MURC staining was detected in the cytoplasm with a striated and periodic staining pattern and was partly colocalized with α-actinin in the Z-line (Fig. (Fig.3A),3A), suggesting that MURC interacts with the Z-disc. Besides cardiomyocytes, MURC staining was also detected in vascular smooth muscle cells with a lower intensity than in cardiomyocytes, but not in other surrounding cells, including vascular endothelial cells (Fig. 3B and C), confirming the muscle-specific expression. In vascular smooth muscle cells, MURC staining was detected diffusely in the cytoplasm.
To evaluate whether MURC expression is regulated during heart development, we analyzed mRNA expression in hearts of embryo, neonate, and adult mice. MURC mRNA expression in the heart was detected at embryonic day 10.5 (E10.5) (Fig. (Fig.4A,4A, left). Real-time RT-PCR with RNA isolated from embryo (E10.5, E13. 5, E15.5, and E17.5), neonate, and adult hearts revealed that MURC mRNA expression increased during the developmental process from the embryonic stage to adulthood (Fig. (Fig.4A,4A, right). We then investigated whether MURC expression might be regulated under pathological conditions in the postnatal heart. We performed abdominal aortic banding to make hypertrophied hearts in mice. Seven days after abdominal aortic constriction, the heart weight normalized to body weight ratio (HW/BW ratio) (band, 4.47 ± 0.03; sham, 4.09 ± 0.04; P < 0.01) and the heart weight normalized to tibial length ratio (HW/TL ratio) (band, 5.90 ± 0.06; sham, 5.21 ± 0.07; P < 0.01) were increased in banded mice compared with sham-operated mice, indicating that cardiac hypertrophy had indeed developed in the banded mice. As shown in Fig. Fig.4B,4B, MURC mRNA expression significantly increased in hypertrophied hearts (2.27-fold ± 0.12-fold; P < 0.01 compared with sham-operated mice).
To elucidate the molecular mechanism involving MURC, we screened over 5 × 107 clones of a human heart cDNA library with hMURC as the “bait” using the yeast two-hybrid system. In this screen, we obtained 13 independent clones. Sequence analysis showed that five isolated cDNAs encoded copper metabolism domain containing 1 (COMMD1) and two isolated cDNAs encoded leucine zipper protein 1 (LUZP1). COMMD1 and LUZP1 were excluded during verification of the protein interaction. The eight other isolated cDNAs encoded SDPR, which is a phosphatidylserine (PS)-binding protein (5, 17, 18, 30). An amino acid homology search revealed that hSDPR was 31% identical to hMURC. A database search for conserved domains revealed that hSDPR also has a predicted coiled-coil domain. To examine the association between MURC and SDPR in mammalian cells, we transfected COS cells with constructs encoding Flag-tagged hMURC and HA-tagged hSDPR. Western blot analysis showed that hMURC was coimmunoprecipitated with hSDPR (Fig. (Fig.5A).5A). Furthermore, we performed immunoprecipitation using cardiomyocytes infected with Ad-MURC and/or Ad-SDPR. As shown in Fig. Fig.5B,5B, Flag-tagged MURC was coimmunoprecipitated with HA-tagged SDPR in cardiomyocytes, confirming the interaction of MURC with SDPR in cardiomyocytes. In human and mouse tissues, SDPR mRNA has been shown to be highly expressed in heart and lung (17, 18). To assess the subcellular localization of SDPR, isolated adult cardiomyocytes were infected with Ad-SDPR. As shown in Fig. Fig.5C,5C, exogenous HA-tagged SDPR was localized to the cytoplasm, and MURC was localized to the cytoplasm and the Z-line. We then examined SDPR mRNA expression during heart development. As shown in Fig. Fig.5D,5D, SDPR mRNA expression in hearts gradually increased during embryonic stages and reached a maximum in neonates. Furthermore, SDPR mRNA expression increased significantly in hypertrophied hearts (Fig. (Fig.5E)5E) (1.78-fold ± 0.14-fold; P < 0.01 compared with sham-operated mice).
SDPR is a PS-binding protein (5, 17, 18, 30). PS is implicated in signaling pathways, and PS-binding proteins contribute to a wide range of cellular process (40). Among PS-binding proteins, conventional protein kinase C (PKC) and Raf-1 have been reported to be implicated in the development of cardiac hypertrophy and heart failure (7, 31, 40). The observation of the interaction of MURC with SDPR prompted us to investigate the effect of MURC on signaling pathways in cardiomyocytes. The ANP gene is well-suited for this purpose, as it has been used as a molecular marker to explore signaling pathways that regulate cardiac gene expression during development and disease, and reactivation of ANP expression is part of a conserved adaptive change in molecular phenotypes in response to heart failure (24). Overexpression of MURC in cardiomyocytes using Ad-MURC induced ANP mRNA expression (Fig. (Fig.6A).6A). ANP mRNA expression in cardiomyocytes is known to be regulated by the Rho/ROCK pathway (23). Furthermore, the cardiac phenotype of transgenic mice expressing Flag-tagged MURC (Tg-MURC), as described below, was reminiscent of that of transgenic mice expressing RhoA (37). Therefore, we examined whether MURC might alter RhoA activity in cardiomyocytes. RhoA was significantly activated by MURC in cardiomyocytes (Fig. (Fig.6B)6B) (1.41-fold ± 0.07-fold; P < 0.05 compared with Ad-LacZ-infected cardiomyocytes). MURC-induced ANP mRNA expression was attenuated by a ROCK inhibitor, Y-27632, in a dose-dependent manner (Fig. (Fig.6C6C).
To further examine the effect of MURC signaling on the transcription of ANP, we performed a luciferase reporter assay using the ANP promoter (−638 ANP Luc), which has two SRF-binding sites (21). hMURC transactivated −638 ANP Luc, and the extent of activation by mMURC was similar to that caused by hMURC (Fig. (Fig.6D).6D). As shown in Fig. Fig.6E,6E, an activated mutant of RhoA, RhoA V14, transactivated −638 ANP Luc. When Rho signaling was inhibited by C3, which specifically ADP-ribosylates Rho and inhibits its function (3), or Y-27632, MURC-induced transactivation of −638 ANP Luc was attenuated. We examined the requirement for the SRE sites in the ANP promoter for responsiveness to MURC signaling using no-SRE1/SRE2 Luc, which has mutations in both SREs in −638 ANP Luc (21, 41). As RhoA signaling has been reported to regulate SRF transcriptional activity (6, 14, 20, 44), the action of RhoA V14 was abolished in no-SRE1/SRE2 Luc, and the action of MURC was also abolished in no-SRE1/SRE2 Luc (Fig. (Fig.6F6F).
We then examined the role of SDPR in the MURC action in cardiomyocytes. When SDPR was transfected with MURC, SDPR augmented MURC-induced transactivation of −638 ANP Luc (Fig. (Fig.6G).6G). RNAi using rSDPR shRNA was performed in neonatal rat cardiomyocytes to determine whether SDPR was required for the action of MURC. To evaluate the effect of rSDPR RNAi, we transfected COS cells with plasmids encoding Flag-tagged rSDPR and rSDPR shRNA. The protein expression of Flag-tagged rSDPR was inhibited by rSDPR RNAi (data not shown). rSDPR RNAi attenuated MURC-induced transactivation of −638 ANP Luc in cardiomyocytes (Fig. (Fig.6H).6H). To exclude off-target effects of the rSDPR shRNA used, we used another shRNA sequence targeting SDPR and obtained similar effects on −638 ANP Luc (data not shown).
PE, an α1-adrenergic receptor agonist, induced MURC mRNA expression in cardiomyocytes (Fig. (Fig.6I).6I). To directly examine the biological effects of MURC, we made a recombinant adenovirus expressing rMURC shRNA (Ad-rMURC shRNA) and examined the effect of MURC on PE-induced ANP mRNA expression. MURC protein expression was repressed in a time-dependent manner in cardiomyocytes infected with Ad-rMURC shRNA (Fig. (Fig.6J).6J). When PE-induced upregulation of MURC protein was inhibited by Ad-MURC shRNA (Fig. (Fig.6K),6K), PE-induced ANP mRNA expression was attenuated (Fig. (Fig.6L).6L). We used another shRNA sequence targeting MURC and obtained a similar effect on −638 ANP Luc (data not shown).
The Rho/ROCK pathway has been also shown to mediate myofibrillar organization in cardiomyocytes (23). Therefore, we examined whether MURC could affect myofibrillar organization. Overexpression of MURC in cardiomyocytes using Ad-MURC induced myofibrillar organization, and MURC-induced myofibrillar organization was inhibited by Y-27632 (Fig. (Fig.7A).7A). Furthermore, PE-induced myofibrillar organization was attenuated in cardiomyocytes infected with Ad-rMURC shRNA (Fig. (Fig.7B).7B). Thus, MURC also affected myofibrillar organization in cardiomyocytes.
To assess the functional effect of MURC in postnatal hearts, cardiac-specific transgenic mice expressing Flag-tagged MURC, Tg-MURC mice, were generated under the control of the αMHC promoter (16). Western blot analysis using an anti-Flag antibody and an anti-MURC antibody revealed that four independent lines of transgenic mice expressed comparable levels of transgenic protein (Fig. (Fig.8A).8A). Immunostaining showed that Flag-tagged MURC was localized to the cytoplasm and the Z-line in cardiomyocytes of Tg-MURC mice (Fig. (Fig.8B).8B). This finding supports the subcellular localization of MURC in cardiomyocytes, which was shown in Fig. Fig.3A.3A. We examined RhoA activity in the hearts of Tg-MURC mice. Consistent with the in vitro data, RhoA activity increased significantly in the hearts of Tg-MURC mice at 13 weeks of age (Fig. (Fig.8C)8C) (2.11-fold ± 0.12-fold; P < 0.01 compared with nontransgenic [NTg] mice).
We measured heart weight at 13 weeks of age in Tg-MURC mice (Table (Table1).1). Although the HW/BW ratio in Tg-MURC mice was significantly lower than that in NTg mice, no significant difference in the HW/TL ratio was observed between Tg-MURC and NTg mice. The hearts of Tg-MURC mice at 13 weeks of age showed atrial chamber enlargement with a reduction in ventricular wall thickness (Fig. 9A and B). Histological assessment of cardiac pathology demonstrated a large organized thrombus in the left atrium (Fig. (Fig.9B).9B). Diffuse interstitial fibrosis was detected in both atria and ventricles of Tg-MURC mice (Fig. 9C and D).
To address the functional consequences of MURC expression in vivo, echocardiography was performed on 12-week-old Tg-MURC and NTg mice, which confirmed reduced posterior wall thickness and depressed left ventricular systolic function in Tg-MURC mice compared with NTg mice (Table (Table2).2). Furthermore, we assessed systolic and diastolic LV pressures, contractility, and relaxation in Tg-MURC mice at 13 weeks of age. In Tg-MURC mice, the maximum LV dP/dt was significantly lower compared with that in NTg mice (Table (Table3).3). Thus, Tg-MURC mice showed cardiac contractile dysfunction associated with cardiac fibrosis.
We then examined the heart sections of Tg-MURC mice at 5 weeks of age to determine the earlier responses of cardiomyocytes to the MURC action in vivo. As shown in Fig. Fig.9E,9E, the cross-sectional area of cardiomyocytes in Tg-MURC mice was bigger than that in NTg mice (1.46-fold ± 0.02-fold; P < 0.01 compared with NTg mice), and various sizes of cardiomyocytes were observed in Tg-MURC mice. These findings suggest that the cardiac dysfunction in 12-week-old Tg-MURC mice is likely to be the end point of a pathological hypertrophy.
ECGs were obtained from NTg and Tg-MURC mice at 9 and 12 weeks of age. Among 12 Tg-MURC mice analyzed by ECG at 9 weeks of age, 2 Tg-MURC mice (numbers 436 and 443) showed AF, 1 Tg-MURC mouse (429) showed complete AV block, and the other Tg-MURC mice showed AV conduction defects evidenced by prolongation of the PR interval (Fig. 10A). At 12 weeks of age, in Tg-MURC mouse 436 AF was sustained, whereas in Tg-MURC mouse 429 it showed a shift from complete AV block to AF, and Tg-MURC mouse 443 showed a shift from AF to complete AV block. The ECGs of other Tg-MURC mice with sinus rhythm (e.g., number 400) at 12 weeks of age demonstrated slow heart rate and prolongation of the PR interval (Fig. 10B). Furthermore, 16 Tg-MURC mice showed cardiac arrhythmias, including AF and AV block, among 38 Tg-MURC mice up to 20 weeks of age, whereas among 51 NTg mice, no mice showed cardiac arrhythmia.
To identify a molecular signature for the susceptibility to heart failure and atrial arrhythmias conferred by MURC, we compared microarray expression profiles of the ventricular myocardium from Tg-MURC and NTg mice and verified the findings by real-time PCR. ANP, BNP, βMHC, and fibrosis-related genes, such as TGF-β1, TGF-β2, TGF-β3, and procollagens, were upregulated in Tg-MURC mice compared with NTg mice, whereas SERCA2 was downregulated in Tg-MURC mice compared with NTg mice (Table (Table44).
To examine the role of MURC in ANP expression in vivo, we compared ANP mRNA expression in embryonic (E10.5), neonatal, and adult hearts of NTg and Tg-MURC mice by real-time RT-PCR. As shown in Fig. Fig.11,11, ANP mRNA expression in embryonic and neonatal hearts of Tg-MURC mice was not altered compared with that in embryonic and neonatal hearts of NTg mice, respectively. However, ANP mRNA expression increased in adult hearts of Tg-MURC mice compared with NTg mice. Furthermore, although in NTg mice ANP mRNA expression decreased in adult hearts compared with neonatal hearts, which is consistent with a previous report (24), in Tg-MURC mice ANP mRNA expression increased in adult hearts compared with neonatal hearts. Since we used the αMHC promoter to generate Tg-MURC mice, most of the exogenous MURC gene driven by the αMHC promoter in the hearts of Tg-MURC mice would be expressed after birth. Taken together, our findings suggest that sustained overexpression of MURC in the postnatal heart regulates ANP expression in vivo.
We demonstrated MURC as a novel molecule localized to the cytoplasm and partly in the Z-line of the sarcomere in the heart. MURC mRNA was expressed in heart, skeletal muscle, and vascular smooth muscle. In vascular smooth muscle cells, which have no Z-disc, MURC protein was diffusely localized to the cytoplasm, suggesting that MURC is normally localized to the cytoplasm in addition to the Z-disc. We also identified SDPR as a MURC-binding protein. Exogenous SDPR was observed in the cytoplasm of cardiomyocytes. SDPR has been shown to be phosphorylated by PKC and to be localized to the cytoplasm (18). SDPR has been identified also as a PKCα-binding protein involved in the targeting of PKCα to caveolae (30). Caveolae have been suggested to be implicated in many cellular processes, including transcytosis of macromolecules, cholesterol transport, and signal transduction (15). Signaling through the caveolae is considered to be involved in the pathogenesis of the cardiovascular system (15). In cardiomyocytes, initiation and transduction of stretch-induced RhoA activation through caveloae have been reported (27). We demonstrated in the present study that MURC induced RhoA activation in vitro and in vivo. MURC also induced ANP mRNA expression and myofibrillar organization in cardiomyocytes, both of which were mediated by Rho/ROCK pathways. Furthermore, overexpression of SDPR in cardiomyocytes augmented MURC-induced transactivation of the ANP promoter, while RNAi-mediated knockdown of SDPR attenuated it. These findings suggest that in cardiomyocytes MURC functions as a molecule that shuttles between the Z-disc and other subcellular locations to transmit signals and that MURC requires SDPR to function efficiently in cardiomyocytes.
In addition to the activation of RhoA induced by MURC in vitro and in vivo, the cardiac phenotype of Tg-MURC mice was similar to that of transgenic mice expressing RhoA (37). These findings suggest that RhoA acts as one of the important downstream effectors of MURC in the heart. The Rho GTPase cycles between GDP-bound inactive and GTP-bound active forms (12). The nucleotide state of Rho GTPases is regulated by three kinds of regulators: guanine nucleotide exchange factors, GTPase-activating proteins, and Rho GDP dissociation inhibitors. Since MURC contains no recognizable protein motifs except for the coiled-coil motif, MURC probably interacts with molecules that directly or indirectly regulate the activity of upstream regulators for RhoA. The RhoA signaling pathway is activated by G protein-coupled receptor agonists angiotensin II and PE in cardiomyocytes (3, 23). The results shown here that MURC mediates PE-induced ANP mRNA expression and myofibrillar organization suggest the involvement of a MURC-regulated Rho/ROCK pathway. Taken together, our findings indicate that MURC per se causes signaling events to induce gene expression and myofibrillar organization, and MURC also serves to transmit signals from cell surface receptors.
Responsiveness to MURC signaling in the ANP promoter required the SRE sites, suggesting that MURC signaling regulates SRF transcriptional activity. Cardiac-specific transgenic mice expressing SRF develop severe heart failure accompanied by activation of the fetal gene program, cardiac hypertrophy, chamber dilatation, and interstitial fibrosis (49). Cardiac-specific overexpression of a mutant form of SRF that reduces the binding activity to SRE sites causes dilated cardiomyopathy (50), and deletion of SRF from the adult heart also shows dilated cardiomyopathy (34). These studies indicate that the function of SRF is critical for maintaining cardiac function in the postnatal heart. Since Tg-MURC mice also exhibited a partly similar cardiac phenotype to transgenic mice expressing SRF, SRF transcriptional activation by MURC could contribute to the cardiac phenotype observed in Tg-MURC mice.
The striated muscle activator of Rho signaling (STARS) protein has been identified as a muscle-specific actin-binding protein localized to the I-band and the M-line of the sarcomere (4). STARS stimulates the transcriptional activity of SRF. STARS expression in the heart has been shown to be upregulated in mouse models of cardiac hypertrophy and in human cardiomyopathy, and cardiac overexpression of STARS sensitizes the heart to biomechanical stress induced by transverse aortic banding and calcineurin signaling (28). Since the localization of STARS abuts the Z-line on both sides (4), STARS appears not to be colocalized with MURC in cardiomyocytes. However, with regard to the Rho/ROCK pathway, MURC signaling overlaps with STARS signaling. Transgenic mice expressing STARS without biomechanical stress have been reported to show no cardiac dysfunction and no histological abnormality (28). Tg-MURC mice spontaneously showed cardiac dysfunction and histological abnormality. The phenotypic difference between these mice may be caused by the action on the Rho/ROCK pathway and/or the functions that are dependent on their localization.
MURC expression in the heart increased during the developmental process from the embryonic stage to adulthood. Furthermore, in the postnatal heart MURC expression was upregulated in the hypertrophied heart. These results indicate that MURC expression is regulated under both physiological and pathological conditions. Tg-MURC mice showed cardiac contractile dysfunction and conduction disturbances with increased vulnerability to atrial arrhythmias, which was associated with an altered cardiac gene expression profile resembling that in pressure-overloaded mice but not in exercised mice (29). In Tg-MURC mice at 5 weeks of age, various sizes of cardiomyocytes were observed in Tg-MURC mice. These findings suggest that sustained overexpression of MURC in the postnatal heart causes pathological gene expression and that the expression level of MURC in the postnatal heart is critical for the maintenance of cardiac function and arrhythmogenesis. Increased inducibility of AF has been demonstrated in animal models of aging, congestive HF (CHF), atrial fibrosis, cardiac-specific overexpression of Rho A and constitutively active Rac1, and connexin 40 deficiency (1, 11, 19, 37). Atrial fibrosis increases with age in humans and is observed in patients with AF and in animal models of aging, CHF, and cardiac-specific overexpression of TGF-β1, tumor necrosis factor alpha, and angiotensin-converting enzyme (11, 36, 38, 46, 48). Genes involved in fibrosis, such as TGF-β and procollagens, were upregulated in the hearts of Tg-MURC mice, which also exhibited diffuse atrial fibrosis and spontaneous episodes of AF. Therefore, susceptibility to fibrosis is likely attributed to an increased vulnerability to AF in MURC-Tg mice. Our findings also suggest the critical role of MURC in the development of a vulnerable substrate for AF, especially in the setting of CHF.
In conclusion, the present study demonstrates that MURC is a Z-line-localizing molecule and it modulates the Rho/ROCK signaling pathway. Sustained overexpression of MURC facilitates functional deterioration, including cardiac function and conduction disturbances. Further investigation of the role of MURC will provide insights into the molecular mechanisms of the Z-disc that regulate cardiac homeostasis and pathogenesis.
We thank M. Kuramoto, M. Nishikawa, A. Kosugi, and A. Yasui for their technical assistance. We also thank the following investigators for their kind gifts of plasmids: Jeffrey Robbins, Kenneth R. Chien, Andrew Thorburn, Yoshimi Takai, and Seigo Izumo.
This work was supported by grants-in-aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, Japan Association for the Advancement of Medical Equipment, Takeda Science Foundation, and Mitsubishi Pharma Research Foundation.
Published ahead of print on 10 March 2008.