Reduced ACTC1 Expression Might Play a Role in the Onset Of

Circulation Journal ORIGINAL ARTICLE Official Journal of the Japanese Circulation Society http://www.j-circ.or.jp Pediatric Cardiology and Adult Congenital Heart Disease Reduced ACTC1 Expression Might Play a Role in the Onset of Congenital Heart Disease by Inducing Cardiomyocyte Apoptosis Hong-Kun Jiang, MD; Guang-Rong Qiu, MD, PhD; Jesse Li-Ling, MD, PhD; Na Xin, MSc; Kai-Lai Sun, MD, PhD

Background: The Cardiac α actin 1 gene (ACTC1) has been related to familial atrial septal defects. This study was set to explore a potential role of this gene in the formation of sporadic congenital heart disease (CHD).

Methods and Results: Assessment of cardiac tissue samples from 33 patients with sporadic CHD (gestational age (GA) 18 weeks – 49 months) with real-time RT-PCR, Western blotting and immunohistochemistry has revealed a markedly decreased ACTC1 expression in the majority of samples (78.8%) compared with autopsied normal heart tissue from aged-matched subjects (GA 17 weeks – 36 months). Also, as shown by terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) assay, the proportion of apoptotic cardiomyocytes in samples featuring down-regulated ACTC1 expression (Group 1) was significantly greater than those with normal expression (Group 2) and the controls (P<0.01). The proportion of apoptotic cells strongly correlated with the expression of ACTC1 (r=–0.918, P<0.01). A study of 2 essential genes involved in apoptosis, Caspase-3 and Bcl-2, confirmed that the former has significantly increased expression, whilst the latter has decreased expression in Group 1 than in the other groups (P<0.01). Transfection of a small interfering RNA targeting, Actc1 (Actc1- siRNA), to a cardiomyocyte cell line, H9C2, also detected more apoptotic cells.

Conclusions: Reduced ACTC1 expression might play a role in the onset of CHD through induction of cardiomyocyte apoptosis. (Circ J 2010; 74: 2410 – 2418)

Key Words: Apoptosis; Bcl-2; Cardiac α actin 1; Caspase-3; Congenital heart disease

eart development is a complex multi-stage process as α-myosin heavy chain (MYH6) have also been identified consisting of a series of precisely regulated molec- in familial atrial septal defect (ASD).6 H ular and morphogenetic events, which include cell specification, looping, chamber segmentation and growth, and Editorial p 2297 septum formation.1 As the most common type of birth defects, congenital heart disease (CHD) occurs in the forms of septal Recently, a M123V mutation of Cardiac α actin 1 gene defects, patent ductus arteriosus (PDA), pulmonary stenosis, (ACTC1) has been found to be responsible for an autosomal tetralogy of Fallot (TOF), etc. While most CHD occur spo- dominant type of secundum ASD.7 As a cytoskeletal pro- radically, some might appear as part of a syndrome.1–4 Recent tein, actin is ubiquitously expressed in eukaryotic cells and studies have begun to uncover the genetic bases for particu- is involved in an extraordinary array of cellular functions. lar forms of CHD, and have provided new insights into how Besides its classic role in muscle contraction, actin also plays dysregulation of heart development might lead to CHD. A important roles in cellular processes such as gene transcrip- large number of genes in connection with familial and spo- tion, chromosome morphology, cell cycle control, modulation radic forms of CHD have been identified. Notably, most of of a variety of membrane responses, translation of several such genes, for example, TBX5, NKX2.5 and GATA4, encode mRNA species, modulation of enzyme activity and intra-cel- transcription factors that regulate specific events in the heart lular localization, and cellular apoptosis.8,9 development.5 In addition, mutations of structural genes such Actin filaments involved in the above processes display

Received March 12, 2010; revised manuscript received June 21, 2010; accepted July 1, 2010; released online October 15, 2010 Time for primary review: 24 days Department of Pediatrics, The First Affiliated Hospital of China Medical University, Shenyang (H.-K.J.); Department of Medical Genetics, China Medical University, Shenyang (H.-K.J., G.-R.Q., J.L.-L., N.X., K.-L.S.), China Mailing address: Hong-Kun Jiang, MD, Department of Pediatrics, The First Affiliated Hospital of China Medical University, 155 North Nanjing Street, Shenyang 110001, China. E-mail: [email protected] or Kai-Lai Sun, MD, PhD, Department of Medical Genetics, China Medical University, 92 Bei’er Rd, Shenyang 110001, China. E-mail: [email protected] ISSN-1346-9843 doi: 10.1253/circj.CJ-10-0234 All rights are reserved to the Japanese Circulation Society. For permissions, please e-mail: [email protected]

Circulation Journal Vol.74, November 2010 Reduced ACTC1 Expression in CHD 2411

variances in organization, intracellular location, motor pro- Table 1. Clinical Findings of the Patients and Controls tein interactions and dynamics.10 Six primary actin isoforms have been identified in higher vertebrates. As the main iso- Case Age Sex Diagnosis form in the adult heart, ACTC1 has shown to be the pre- Controls dominant form in early muscle development in most cultured 1 GA 17w F Neurofibromatosis cell lines as well as in the late stages of fetal development.8,11 2 GA 18w M Neurofibromatosis Matsson et al reported that Morpholino knockdown of 3 GA 23w+1d M Spontaneous abortion ACTC1 in chick embryos can result in delayed looping and 4 GA 26w+2d M Congenital microtia reduced atrial septa, and concluded that reduced ACTC1 5 GA 28w+4d M Spontaneous abortion 7 expression might lead to isolated ASD. However, the level 6 GA 32w F Dysplasia of right kidney of ACTC1 expression in the sporadic-type of human CHD 7 8 h F Hydrocephalus still remains unknown. 8* 19 h F Hydrocephalus In this study, we have investigated the expression of ACTC1 9* 17 days F Intracranial hemorrhage in the heart tissues from patients with sporadic types of CHD, and correlated it with cardiomyocyte apoptosis, which might 10* 2 months F Intracranial hemorrhage underlie the formation of CHD.12 We also studied the expres- 11* 12 months M Surgical trauma sion of 2 essential genes, Caspase-3 and Bcl-2, in the cardiac 12* 36 months M Surgical trauma tissues from such patients. To clarify whether reduced ACTC1 Patients with CHD expression might induce apoptosis of cardiomyocytes, a small 1 GA 18w M TOF interfering RNA targeting, Actc1 (Actc1-siRNA), was also 2 GA 22w+5d F VSD, single atrium, transfected into a rat embryonic cardiomyocytes cell line, persistent truncus H9C2. Thereafter, apoptotic cells were detected. arteriosus 3 GA 24w+1d F VSD 4 GA 26w+3d M VSD Methods 5 GA 26w+5d M VSD Surgical Pathology/Autopsied Specimens 6 GA 27w+1d M AVSD Cardiac tissues were obtained as iced waste and paraffin-em- 7 GA 28w+2d F VSD bedded surgical pathology/autopsy materials from 33 patients 8 GA 29w M VSD with CHD (Patient group, gestational age (GA) 18 weeks – 9 GA 29w+3d F AVSD 49 months) and 12 age-matched autopsies where no structural 10 GA 30w+3d F AVSD or hemodynamic abnormalities of the heart were detected (Control group, GA 17 weeks – 36 months) were obtained. 11 GA 30w+4d F TOF Clinical data were retrieved from patient records. Tissue 12 GA 31w+1d M VSD specimens were obtained from the free wall of the left ven- 13 GA 32w M VSD, double outlet tricle or atrial appendage. Eleven samples were obtained 15– right ventricle 40 min after the patients’ death, whilst the remaining samples 14 GA 32w+2d F AVSD were obtained during surgical operation or within 15 min 15 GA 33w M VSD after the patient’s death (see Table 1 for details). Patients with 16 GA 33w+5d F TOF syndromic CHD and other developmental anomalies were 17 GA 38w F TOF excluded. All samples were fixed with 10% buffered for- 18* 11 months F VSD malin (pH =7.4), and then embedded in paraffin for light 19* 13 months F ASD microscopy. Four-μm-thick paraffin sections were stained 20* 16 months F ASD with hematoxylin-eosin and Masson’s trichrome for histo- 21* 19 months F ASD pathological examination. Specimens with histopathological 22* 20 months M ASD findings of autolysis were also excluded. The remaining 23** 21 months M ASD tissue samples were stored at –80°C. Informed consent was obtained from the guardian of each patient, and the study 24* 22 months F ASD protocol has conformed to the ethical guidelines of the 1975 25 23 months M ASD Declaration of Helsinki, as reflected in a priori approval by 26** 25 months F ASD the institution’s human research committee. 27** 28 months F ASD 28** 33 months M ASD RNA Isolation, Reverse Transcription and Real-Time PCR 29** 35 months F ASD RNA was extracted from the heart tissues or H9C2 cells using 30 35 months F ASD the TRIzol Reagent (Invitrogen, Shanghai, China) according 31** 36 months M ASD to the manufacturer’s protocol. cDNA was synthesized from 32 38 months M ASD 3 g of RNA using a RNA PCR kit (TaKaRa, Dalian, China). μ 33** 49 months M ASD With 8 pairs of primers (Table 2), real-time RT-PCR was performed on an ABI 7500 system (Applied Biosystems, Foster GA, gestational age; w, week; F, female; M, male; d, days; TOF, tetralogy of Fallot; VSD, ventricular septal defect; AVSD, atrio- City, CA, USA) in a 25μ l reaction mixture containing 12.5μ l ventricular septal defect; ASD, atrial septal defect. of SYBR Green PCR Master mix (Applied Biosystems) and *Samples were obtained approximately 15–40 min after the pa- 1μ l cDNA. Relative cDNA concentrations were calculated tients had deceased. The remainder were obtained within 15 min from a standard curve using sequential dilutions of correspond- after ex vivo during surgical operation or dying. **Samples obtained from the surgically removal atrial append- ing PCR fragments. The amplification program involved age. The remainder were from autopsy specimens obtained from initial denaturation at 95°C for 10 min, 40 cycles of dena- the free wall of the left ventricle. turation at 95°C for 10 s, and annealing and extension at 60°C

Circulation Journal Vol.74, November 2010 2412 JIANG HK et al.

Table 2. PCR Primers for the Target Genes Amplicon size Target/control gene Primer sequences (base pair, bp) ACTC1 (human) F: 5’-GCCCTGGATTTTGAGAATGA-3’ 162 R: 5’-ATGCCAGCAGATTCCATACC-3’ Actc1 (rat) F: 5’-AGAGCACGGCATTATCAC-3’ 146 R: 5’-TCATCTTCTCACGGTTGG-3’ Caspase-3 (human) F: 5’-AGAACTGGACTGTGGCATTGAG-3’ 191 R: 5’-GCTTGTCGGCATACTGTTTCAG-3’ Caspase-3 (rat) F: 5’-CAAGTCGATGGACTCTGGAA-3’ 129 R: 5’-GTACCATTGCGAGCTGACAT-3’ Bcl-2 (human) F: 5’-CGAGTGTCTCAAGCGCATC-3’ 171 R: 5’-GCAAAGTAGAAAAGGGCGACAAC-3’ Bcl-2 (rat) F: 5’-CCGGGAGAACAGGGTATGATAA-3’ 81 R: 5’-CCCACTCGTAGCCCCTCTG-3’ β-Actin (human) F: 5’-CCGTCTTCCCCTCCATCG-3’ 155 R: 5’-GTCCCAGTTGGTGACGATGC-3’ β-Actin (rat) F: 5’-CCCATCTATGAGGGTTACGC-3’ 150 R: 5’-TTTAATGTCACGCACGATTTC-3’ for 1 min. Fluorescence was measured at the end of each images was analyzed to assess the percentage of isoactin- extension step. The threshold cycle (Ct) was subsequently positive cardiomyocytes using KS400 software (Kontron determined. Relative concentrations of mRNA were calcu- System; Zeiss Vision, Oberkochen, Germany). lated based on the Ct values and normalized to the level of β -actin of each sample. All reactions were performed with Detection of Apoptotic Cardiomyocytes in Tissue Sections appropriate negative controls (template-free). Melting curve A terminal deoxynucleotidyl transferase-mediated dUTP nick analyses were conducted after the completion of cycling end-labeling (TUNEL) assay was used to identify double- with the aid of a temperature ramp (from 45°C to 95°C at stranded DNA fragmentation; characteristics of DNA degra- 0.5°C/2 s) and continuous fluorescence monitoring to deter- dation due to apoptosis. Briefly, tissue slides were deparaf- mine the specificity of PCR, which was confirmed with con- finized and treated with proteinase K (20μ g/ml) for 20 min ventional gel electrophoresis. All reactions were repeated at room temperature. The slides were then quenched in 2.0% twice to ensure the reproducibility of the results. hydrogen peroxide. After rinsing in phosphate-buffered saline (PBS, pH 7.4), specimens were incubated in 1× equilibration Western Blotting buffer for approximately 10–15 s. The slides were then incu- Frozen heart tissues from patients and controls and cultured bated with terminal deoxynucleotidyl transferase (rTdT) for H9C2 cells were lysed in buffer. The protein concentration 1 h at 37°C, blocked with a stop/washing buffer, and incu- of each lysate was determined with a bicinchoninic acid bated with peroxidase-conjugated anti-digoxigenin antibody (BCA) kit (Keygen Biotech Co Ltd, Nanjing, China) accord- for 30 min at room temperature. Finally, the slides were ing to the manufacturer’s instruction. Total protein (30μ g) stained with diaminobenzidine (DAB; Promega). Nuclear was applied to a 12% SDS-polyacrylamide gel. After elec- staining by hematoxylin was performed as the counter-stain. trophoresis, polyvinylidene fluoride (PVDF) membranes A commercially available human lymph node section was were washed in Tris-buffered saline containing 0.1% Tween- used as the positive control. A negative control of the TUNEL 20, and then incubated with a primary antibody (monoclonal assay was set by staining human heart tissue in the same antibody to ACTC1; Santa Cruz, CA, USA, 1:200 diluted). manner but without rTdT enzyme incubation. Subsequently, The membrane was incubated with a secondary antibody cardiac-specific immunostaining was performed using a (1:4,000 diluted) and immuno-stained bands were detected primary anti-ACTC1 antibody (Santa Cruz Biotechnology with a ProtoBlot II AP System with a stabilized substrate Inc) and a secondary reaction using the streptavidin-alkaline (Promega, Madison, WI, USA). Beta-actin was used as the phosphatase system to clarify the type of positive cells. For internal control. each specimen, the proportion of TUNEL-positive cardiomyocytes was expressed as the ratio of cardiomyocytes Immunohistochemistry expressing TUNEL over the total number of cardiomyocytes Immunohistochemistry was performed in order to clarify the per field. For each case, 100 random fields were counted. expression of ACTC1 protein in the disease tissue. For the Approximately 3–5 sections obtained at a distance of 100μ m start, heart tissues from patients were fixed with 10% buffered for each tissue block were used. To avoid potential bias due formalin (pH =7.4) and embedded in paraffin. Subsequently, to selection of fields with different cardiomyocytes density, 4-μm sections were prepared by sequential incubation with an approximately 1,500 cardiomyocyte nuclei were investigated anti-ACTC1 monoclonal antibody (Santa Cruz Biotechnology, for each optical field (objective lens,× 20). The total numbers Inc, Santa Cruz, CA, USA) at a 1:100 dilution. Normal mouse of cardiomyocytes counted ranged from 450×103 to 750×103. serum (Santa Cruz, CA, USA) at the same dilution was used Only actin-positive cells with well-organized sarcomeric stria- as the negative control. After the staining, images were tion were identified as cardiomyocytes, whereas actin-nega- acquired with an Axiophot microscope equipped with a high tive cells without striation (such as vascular smooth-muscle sensibility color camera (Axiocam, Carl Zeiss). A set of cells of an intramyocardial arteriole and rare interstitial cells)

Circulation Journal Vol.74, November 2010 Reduced ACTC1 Expression in CHD 2413

Figure 1. Expression of the Cardiac α actin 1 gene (ACTC1) as determined by the abundance of mRNA and protein in the cardiac tissues from patients with congenital heart disease (CHD). (A) As shown, the mRNA of ACTC1 was significantly reduced in the disease tissues (lanes 2, 4, 6) compared with the controls (lanes 1, 3, 5). The specificity of real-time RT-PCR was confirmed with conventional gel electrophoresis. (B) ACTC1 proteins, as detected by Western blotting, was reduced accordingly (lanes 4–6) compared with the controls (lanes 1–3). **P<0.01 vs control. (C) The patient group was divided into 2 sub-groups based on the expression of ACTC1. Samples with reduced ACTC1 expression were designated as Group 1, and the remainder were designated as Group 2. (D) As confirmed by Western blotting, the abundance of ACTC1 protein was consistent with that of ACTC1 mRNA from the same patients. **P<0.01 vs control and Group 2. No significant change was detected between Group 2 and the control. (E) Expression of the ACTC1 gene in the cardiac tissues determined by immunohistochemistry. Staining was mainly restricted in the cytoplasm, which was intense in cardiac tissues of controls and Group 2 patients (b and d) but decreased in Group 1 (c) patients. Pictures were amplified to ×400. Scale=30μ m. Negative control was also shown (a). ○, control; ▲, CHD group; ▲, Group 1; △, Group 2.

Circulation Journal Vol.74, November 2010 2414 JIANG HK et al.

Figure 2. Apoptosis detection in cardiac tissues. (A–C) TUNEL and immunohistochemical staining of ACTC1 (×1,000). Detec- tion of TUNEL-positive cardiomyocytes (shown by arrows) in a control sample (A), patients with down-regulated ACTC1 (B) or normal ACTC1 expression (C). Relative proportions of TUNEL-positive cardiomyocytes are also shown in (D). ACTC1 proteins co-localized with markers of apoptosis. Double positive staining (nuclear staining for DNA fragmentation at TUNEL, and cyto- plasmic staining for ACTC1) within the same cardiomyocytes is shown. The correlation between TUNEL-positive cardiomyocytes and ACTC1 expression in patients with CHD is shown in (E) (▲, Group 1; ▲, Group 2; r=–0.918, P<0.01). Expression of Caspase-3 and Bcl-2 genes, as determined by real-time RT-PCR, in the diseased tissues with down-regulated ACTC1 (Group 1) or normal expression (Group 2) and controls is shown in (F) and (G). **P<0.01 vs control and Group 2. ACTC1, Cardiac α actin 1 gene; ns, not significant; TUNEL, transferase-mediated dUTP nick end-labeling.

Circulation Journal Vol.74, November 2010 Reduced ACTC1 Expression in CHD 2415 were excluded. The nuclei were counted by 2 observers who were blinded to the experimental conditions.

Cell Culture and Actc1-siRNA Transfection H9C2 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal bovine serum. After the cells had grown to confluence, they were placed in a quies- cent medium (0.5% fetal bovine serum) for 16 h. To silence Actc1 gene expression, we performed siRNA transfection using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). A short interfering RNA targeting rat Actc1 (Actc1-siRNA) was designed and synthesized by GenePharma, Shanghai, China. The sequences of Actc1-siRNA were as follows: 5’-GAAACUACUUAUAACAGCATT-3’ and 5’-UGCUG- UUAUAAGUAGUUUCGT-3’. siRNA transfection was performed according to the manufacturer’s instruction (Targeting Systems, San Diego, CA, USA). Actc1-siRNA or scramble siRNA (negative control) was transfected into H9C2 cells at the final concentration of 100 nmol/L. The efficacy of Actc1 knockdown was assessed by real-time RT-PCR (48 h after transfection) and Western blotting (72 h after transfection).

Detection of Apoptotic H9C2 Cells Apoptosis in H9C2 cells 72 h after transfection with Actc1- siRNA was assayed with an Annexin V-FITC Apoptosis Detection Kit (Miltenyi, Germany). At the end of incubation, the cells were gently washed once with PBS and incubated for 15 min at room temperature with 120μ l of annexin- binding buffer containing 10μ l of fluorescein isothiocyanate (FITC)-conjugated annexin V and 1μ l of 100μ g/ml PI. The staining mixture was then removed and replaced with 120μ l of annexin-binding buffer. The cells were viewed immedi- ately using an Olympus IX-70 inverted fluorescent microscope equipped with filters appropriate for fluorescein and rhoda- mine. Thereafter, samples were analyzed by flow cytometry (FACScaliber, Becton Dickinson, Franklin Lakes, NJ, USA) for viable (annexin V-negative and PI-negative), early apoptotic (annexin V-positive, PI-negative), and late apoptotic/ secondary necrotic cells (annexin V-positive and PI-positive). The extent of apoptosis was quantified as the percentage of annexin V-positive cells. The findings in Actc1-silenced Figure 3. H9C2 cells transiently transfected with Actc1-spe- H9C2 cells were confirmed by TUNEL assay using a com- cific small interfering RNA (Actc1-siRNA) or a scramble siRNA (negative control). Actc1 mRNA (A) was determined by real- mercial kit (Promega). Briefly, after removal of the floating time PCR, and Actc1 protein was detected by Western blot- cells, adherent cells in 2-well chamber slides were fixed and ting (B). Beta-actin was used as controls. **P<0.01 vs nega- stained according to the manufacturer’s instructions. Apop- tive control. ACTC1, Cardiac α actin 1 gene. tosis was defined as TUNEL-positive nuclei in cells with morphological features of cell death (cell shrinkage, aggre- gation of chromatin into dense masses, and cell fragmentation). The mean number of apoptotic cells (TUNEL positive) was significantly down-regulated accordingly. We have fur- from 3 random fields ×( 40) in each well was counted. ther divided our patients into 2 sub-groups based on the mRNA levels of ACTC1 normalized to the mean value of the Statistical Analysis control group. Those with reduced ACTC1 expression were All values were recorded as mean ± SD. Statistical significance designated as Group 1 (n=26, tetralogy of Fallot (TOF) 4, between the groups was determined by a 1-way ANOVA. ASD 11, ventricular septal defect (VSD) 7, atrioventricular A P<0.05 was considered to be statistically significant. All septal defect (AVSD) 4), and the remainder were designated statistical analyses were performed with by using SPSS soft- as Group 2 (n=7, ASD 4, VSD 3). As confirmed by Western ware (version 13.0; SPSS Inc, USA). blotting, the level of ACTC1 protein was consistent with that of ACTC1 mRNA in samples from the same patients. How- ever, no significant changes were found between Group 2 Results and the controls (Figures 1C,D). Reduced ACTC1 in the Heart Tissues From CHD Patients As shown in Figure 1, 26 out of the 33 samples (78.8%) Elevated TUNEL-Positive Cardiomyocytes in CHD Patients from patients with CHD showed reduced ACTC1 mRNA TUNEL-positive cardiomyocytes were clearly detected in compared with controls. Western blotting and an immuno- both controls (Figure 2A) and patients featured down-reg- histochemistry assay also confirmed that the ACTC1 protein ulated ACTC1 expression (Group 1, n=26, Figure 2B) or

Circulation Journal Vol.74, November 2010 2416 JIANG HK et al.

Figure 4. Apoptosis induced by Actc1-siRNA. (A) Nuclei of apoptotic cells were stained dark-brown (shown by arrows). The proportion of apoptotic cells is shown as a bar graph (B). H9C2 cells were stained with Annexin V-FITC and PI. The percentage of the gated area is shown as a bar graph (C,D). Expression of Caspase-3 and Bcl-2 genes, as determined by real-time RT-PCR, after the transfection is shown. **P<0.01 vs negative control. Scale bar=10μ m. ACTC1, Cardiac α actin 1 gene.

normal expression (Group 2, n=7, Figure 2C). To clarify 1.25±0.81%, P<0.01, respectively; Figure 2D). Again, no dif- whether apoptotic death of cardiomyocytes has increased in ference was found between Group 2 and the control group in CHD patients, we also determined the proportion of TUNEL- regard to the proportion of TUNEL-positive cardiomyocytes positive cardiomyocytes in each group. As found, the ratio (1.45±0.04% vs 1.25±0.08%, P>0.05; Figure 2D). of the latter to the total of cardiomyocytes was significantly Next, we attempted to correlate the proportion of TUNEL- higher in patients with down-regulated ACTC1 compared positive apoptotic cardiomyocytes with the expression of with Group 2 and the controls (5.77±2.23% vs 1.45±0.44%, ACTC1 in diseased samples. As shown in Figure 2E, a high

Circulation Journal Vol.74, November 2010 Reduced ACTC1 Expression in CHD 2417 proportion of TUNEL-positive apoptotic cardiomyocytes was as a key intermediate between altered outflow tract morpho- strongly correlated with the mRNA level of ACTC1 (r= genesis and signaling events during remodeling of branchial –0.918, P<0.01). arch arteries.22 Based on our findings, we propose that the reduced expression of ACTC1 might result in decreased car- Expression of Caspase-3 and Bcl-2 Genes in Cardiac Tissues diomyocyte proliferation, which might lead to the disturbance As shown in Figures 2F,G, the expression of Caspase-3 was of hemodynamics during embryonic development, resulting higher in diseased tissues featuring decreased ACTC1 expres- in abnormal morphogenesis of the heart. sion than in controls and in Group 2 (P<0.01). Meanwhile, the Besides cell proliferation, apoptosis has also been proposed expression of Bcl-2 in patients with down-regulated ACTC1 as a key process during organogenesis.23–25 Transgenic and was lower than that in Group 2 and in the controls (P<0.01). knockout studies both provided strong evidence for aberrant patterns of apoptosis resulting in CHD, including septation Reduced Actc1 Expression Might Induce H9C2 Apoptosis anomalies.25 Studies have revealed that actin plays a key role Reduced Actc1 mRNA was detected at 48 h by RT-PCR and in apoptosis regulation.26 As discovered recently, the death at 72 h by Western blotting after the transfection (Figures 3A, of adult myocytes as terminal cells is not genetically pro- B). As shown in Figure 4A, more TUNEL-positive cardio- grammed as with apoptosis.27–29 In this study, cardiomyocyte myocytes were detected in Actc1-silenced H9C2 cells than apoptosis was detected in embryonic or infantile tissues. The in control cells (58.79±0.32 vs 1.89±0.13, P<0.01). Apop- proportion of apoptotic cardiomyocytes appeared to be higher totic cells were also evaluated by flow cytometry. Annexin in patients with decreased ACTC1, and is strongly correlated V-positive cells had increased remarkably in cells transfected with the latter. Furthermore, more TUNEL-positive/Annexin with Actc1-siRNA compared with the negative control (52.92± V-positive cardiomyocytes were detected in Actc1-silenced 0.29 vs 2.16±0.09, P<0.01; Figure 4B). Furthermore, as deter- H9C2 cells. Together with an altered expression of Caspase-3 mined by real-time RT-PCR, the expression of the Caspase-3 and Bcl-2, these have clearly indicated that reduced Actc1 gene had increased after the transfection, whilst that of the expression might induce H9C2 apoptosis in vitro. Bcl-2 gene had decreased compared with the controls (P< As reported, homozygous deletion of Actc1 in mice can 0.01; Figures 4C,D). lead to functional and structural disturbances of the heart, which might be due to disorganized development of actomyosin filaments in the affected cardiomyocytes because of Discussion the induced apoptosis in the defective cardiomyocytes and As essentially the only actin expressed in embryonic heart disrupted differentiation.12 Disruption of the actin-filament muscle,8,13 ACTC1 has a remarkably conserved amino acid network in the hypodynamic heart has been associated with sequence, probably more so than any other proteins. This apoptosis.12,30 Notably, reduced expression of the ACTC1 alone seems to suggest that genetic lesions in ACTC1, though gene might also evoke a signal for apoptosis and result in the only transiently expressed in cardiomyocytes, might cause observed defects.12,30,31 The addition of a F-actin-stabilizing cardiovascular phenotypes. A rich variety of ACTC1 muta- drug, JASP (Jasplakinolide), to Jurkat T-cells can rapidly tions has been identified in dilated cardiomyopathy and hyper- induce apoptosis, which is accompanied by an increase in trophic cardiomyopathy.14 Among such patients, septal defects caspase-3 activation.32 This seems in keeping with our find- have been identified as additional features.15–17 Patients with ings of the over-expression of Caspase-3 in CHD patients contiguous gene syndromes and 15q deletions spanning the with decreased ACTC1 expression (in vivo) and in Actc1- ACTC1 gene have also been described.18 The presented silenced H9C2 cells (in vitro). Notably, down-regulated phenotypes, including ASD, VSD and TOF, have suggested Bcl-2 expression was also detected in cardiac tissues from the importance of ACTC1 for the formation of cardiac struc- such patients or Actc1-silenced H9C2 cells. Inhibition of ture. In 2 families featuring isolated ASD, Matsson et al have actin depolymerization-stimulated apoptosis and Bcl-2 might identified dominant-type ACTC1 mutations. The screening rescue the cell death induced by direct disruption of the actin of 408 sporadic CHD cases had identified 1 case with ASD cytoskeleton.33,34 A plausible explanation to this might be and a 17-bp deletion in the ACTC1 gene, predicting a non- that the reduced ACTC1 expression might promote apoptosis functional protein. Therefore, mutations or deletions of the by stimulating expression of Caspase-3, the essential execu- ACTC1 gene might affect its interactions with actomyosin tor for apoptosis, whilst inhibiting the expression of Bcl-2, and binding with regulatory proteins.7 In this study, the major- an anti-apoptotic protein.35 The increased apoptosis of car- ity of patients with sporadic CHD (78.8%) showed a reduced diomyocytes and impaired contractility caused by actin loss ACTC1 expression, suggesting the latter might contribute to during cardiogenesis might lead to hemodynamic distur- the onset of CHD. bance, which is responsible for abnormal morphogenesis of As shown by others, Morpholino knockdown of ACTC1 the heart. In return, exposure to the pressure overload caused in chick embryos can lead to delayed bulboventricular loop- by mechanical stress from blood accumulation owing to the ing and reduced atrial septum due to the affected actin cardiac deformities might increase apoptosis. This might help polymerization.7 Down-regulation of ACTC1, due to Smad4 to explain the universally observed increased cardiomyocyte deletion, can lead to a dramatic decrease in cardiomyocyte apoptosis in our patients. proliferation and severe morphological defects including thin compact layer, disorganized trabeculae and VSD.19 Dur- ing the gestation, proliferation of cardiomyocytes is necessary Conclusion for coping with the increasing hemodynamic load.5 As the Based on the above results, we propose that reduced ACTC1 heart has to function through its own development, hemo- expression at a crucial stage during heart development might dynamic forces might also participate in its morphogenesis. contribute to the onset of CHD owing to increased apoptosis In zebrafish, altering hemodynamics mechanically or geneti- of cardiomyocytes. cally can both have a profound impact on heart morphogenesis.20,21 In mice, 1 study had pinpointed altered hemodynamics

Circulation Journal Vol.74, November 2010 2418 JIANG HK et al.

Acknowledgment compaction. Circulation 2008; 117: 2893 – 2901. 18. Erdogan F, Ullmann R, Chen W, Schubert M, Adolph S, Hultschig This work has been jointly supported by grants from the Natural Science C, et al. Characterization of a 5.3 Mb deletion in 15q14 by com- Foundation of China (No. 30400485) and Liaoning Higher Education parative genomic hybridization using a whole genome “tiling path” Project (No. 20060951, 2008754). BAC array in a girl with heart defect, cleft palate, and developmental delay. Am J Med Genet A 2007; 143: 172 – 178. 19. Qi X, Yang G, Yang L, Lan Y, Weng T, Wang J, et al. Essential References role of Smad4 in maintaining cardiomyocyte proliferation during 1. Basson CT, Bachinsky DR, Lin RC, Levi T, Elkins JA, Soults J, et murine embryonic heart development. Dev Biol 2007; 311: 136 – al. Mutations in human TBX5 [corrected] cause limb and cardiac 146. malformation in Holt-Oram syndrome. Nat Genet 1997; 15: 30 – 20. Hove JR, Koster RW, Forouhar AS, Acevedo-Bolton G, Fraser SE, 35. Gharib M. Intracardiac fluid forces are an essential epigenetic 2. Gelb BD, Tartaglia M. Noonan syndrome and related disorders: factor for embryonic cardiogenesis. Nature 2003; 421: 172 – 177. Dysregulated RAS-mitogen activated protein kinase signal trans- 21. Auman HJ, Coleman H, Riley HE, Olale F, Tsai HJ, Yelon D. duction. Hum Mol Genet 2006; 15: R220 – R226. Functional modulation of cardiac form through regionally confined 3. Wiseman FK, Alford KA, Tybulewicz VL, Fisher EM. Down syn- cell shape changes. PLoS Biol 2007; 5: e53. drome--recent progress and future prospects. Hum Mol Genet 2009; 22. Yashiro K, Shiratori H, Hamada H. Haemodynamics determined 18: R75 – R83. by a genetic programme govern asymmetric development of the 4. Wat MJ, Shchelochkov OA, Holder AM, Breman AM, Dagli A, aortic arch. Nature 2007; 450: 285 – 288. Bacino C, et al. Chromosome 8p23.1 deletions as a cause of com- 23. van den Hoff MJ, van den Eijnde SM, Viragh S, Moorman AF. plex congenital heart defects and diaphragmatic hernia. Am J Med Programmed cell death in the developing heart. Cardiovasc Res Genet A 2009; 149A: 1661 – 1677. 2000; 45: 603 – 620. 5. Bruneau BG. The developmental genetics of congenital heart dis- 24. Abdelwahid E, Pelliniemi LJ, Jokinen E. Cell death and differentia- ease. Nature 2008; 451: 943 – 948. tion in the development of the endocardial cushion of the embry- 6. Ching YH, Ghosh TK, Cross SJ, Packham EA, Honeyman L, onic heart. Microsc Res Tech 2002; 58: 395 – 403. Loughna S, et al. Mutation in myosin heavy chain 6 causes atrial 25. Poelmann RE, Gittenberger-de Groot AC. Apoptosis as an instru- septal defect. Nat Genet 2005; 37: 423 – 428. ment in cardiovascular development. Birth Defects Res C Embryo 7. Matsson H, Eason J, Bookwalter CS, Klar J, Gustavsson P, Today 2005; 75: 305 – 313. Sunnegardh J, et al. Alpha-cardiac actin mutations produce atrial 26. Franklin-Tong VE, Gourlay CW. A role for actin in regulating septal defects. Hum Mol Genet 2008; 17: 256 – 265. apoptosis/programmed cell death: Evidence spanning yeast, plants 8. Gunning P, O’Neill G, Hardeman E. Tropomyosin-based regula- and animals. Biochem J 2008; 413: 389 – 404. tion of the actin cytoskeleton in time and space. Physiol Rev 2008; 27. Ferrari R, Ceconi C, Campo G, Cangiano E, Cavazza C, Secchiero 88: 1 – 35. P, et al. Mechanisms of remodelling: A question of life (stem cell 9. Fedorova E, Zink D. Nuclear architecture and gene regulation. production) and death (myocyte apoptosis). Circ J 2009; 73: 1973 – Biochim Biophys Acta 2008; 1783: 2174 – 2184. 1982. 10. Kueh HY, Mitchison TJ. Structural plasticity in actin and tubulin 28. Hosoda T, Kajstura J, Leri A, Anversa P. Mechanisms of myocar- polymer dynamics. Science 2009; 325: 960 – 963. dial regeneration. Circ J 2010; 74: 13 – 17. 11. McNally E, Dellefave L. Sarcomere mutations in cardiogenesis 29. Miura T, Miki T. GSK-3β, a therapeutic target for cardiomyocyte and ventricular noncompaction. Trends Cardiovasc Med 2009; 19: protection. Circ J 2009; 73: 1184 – 1192. 17 – 21. 30. Yamazaki Y, Tsuruga M, Zhou D, Fujita Y, Shang X, Dang Y, et al. 12. Abdelwahid E, Pelliniemi LJ, Szucsik JC, Lessard JL, Jokinen E. Cytoskeletal disruption accelerates caspase-3 activation and alters Cellular disorganization and extensive apoptosis in the developing the intracellular membrane reorganization in DNA damage-induced heart of mice that lack cardiac muscle alpha-actin: Apparent cause apoptosis. Exp Cell Res 2000; 259: 64 – 78. of perinatal death. Pediatr Res 2004; 55: 197 – 204. 31. White SR, Williams P, Wojcik KR, Sun S, Hiemstra PS, Rabe KF, 13. Suurmeijer AJ, Clement S, Francesconi A, Bocchi L, Angelini A, et al. Initiation of apoptosis by actin cytoskeletal derangement in Van Veldhuisen DJ, et al. Alpha-actin isoform distribution in human airway epithelial cells. Am J Respir Cell Mol Biol 2001; 24: normal and failing human heart: A morphological, morphometric, 282 – 294. and biochemical study. J Pathol 2003; 199: 387 – 397. 32. Odaka C, Sanders ML, Crews P. Jasplakinolide induces apoptosis in 14. Laing NG. Congenital myopathies. Curr Opin Neurol 2007; 20: various transformed cell lines by a caspase-3-like protease-depen- 583 – 589. dent pathway. Clin Diagn Lab Immunol 2000; 7: 947 – 952. 15. Capek P, Brdicka R. Hypertrophic cardiomyopathy. Cas Lek Cesk 33. Martin SS, Leder P. Human MCF10A mammary epithelial cells 2006; 145: 93 – 96; Discussion 96 – 97. undergo apoptosis following actin depolymerization that is inde- 16. Monserrat L, Hermida-Prieto M, Fernandez X, Rodriguez I, Dumont pendent of attachment and rescued by Bcl-2. Mol Cell Biol 2001; C, Cazon L, et al. Mutation in the alpha-cardiac actin gene asso- 21: 6529 – 6536. ciated with apical hypertrophic cardiomyopathy, left ventricular 34. Nishida K, Otsu K. Cell death in heart failure. Circ J 2008; 72(Suppl non-compaction, and septal defects. Eur Heart J 2007; 28: 1953 – A): A-17 – A-21. 1961. 35. Sun CK, Yuen CM, Kao YH, Chang LT, Chua S, Sheu JJ, et al. 17. Klaassen S, Probst S, Oechslin E, Gerull B, Krings G, Schuler P, Propylthiouracil attenuates monocrotaline-induced pulmonary arte- et al. Mutations in sarcomere protein genes in left ventricular non- rial hypertension in rats. Circ J 2009; 73: 1722 – 1730.

Circulation Journal Vol.74, November 2010