Thrombopoietin improved ventricular function and regulated remodeling genes in a rat model of myocardial infarction

International Journal of Cardiology 167 (2013) 2546–2554

Contents lists available at ScienceDirect

International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard

Thrombopoietin improved ventricular function and regulated remodeling genes in a rat model of myocardial infarction☆ Kathy Yuen Yee Chan b, 1, Ligang Zhou c, 1, Ping Xiang d, 1, Karen Li a, b, 1, Pak Cheung Ng b, Chi Chiu Wang a, e, Ming Li f, Nga Hin Pong b, Liu Tu g, Haiyan Deng h, Carrie Ka Lai Kong b, Rita Yn Tz Sung a, b,⁎ a

Li Ka Shing Institute of Health Sciences, The Chinese University of Hong Kong, Hong Kong Department of Paediatrics, The Chinese University of Hong Kong, Hong Kong Neonatal Intensive Care Unit, Chongqing Health Center for Women and Children, Chongqing, China d Department of Cardiology, Children's Hospital of Chongqing Medical University, Chongqing, China e Department of Obstetrics & Gynaecology, The Chinese University of Hong Kong, Hong Kong f Department of Physiology, The Chinese University of Hong Kong, Hong Kong g Department of Physiology, Chongqing Medical University, Chongqing, China h Department of Cardiology, Children's Hospital of Fudan University, Shanghai, China b c

a r t i c l e

i n f o

Article history: Received 8 November 2011 Received in revised form 3 May 2012 Accepted 9 June 2012 Available online 6 July 2012 Keywords: Thrombopoietin Myocardial infarction Microarray profiling Cardiac remodeling Endothelial progenitor cells

a b s t r a c t Background: Thrombopoietin (TPO) protects against heart damages by doxorubicin-induced cardiomyopathy in animal models. We aimed to investigate the therapeutic efficacy of TPO for treatment of myocardial infarction (MI) in a rat model and explored the mechanisms in terms of the genome-wide transcriptional profile, TPO downstream protein signals, and bone marrow endothelial progenitor cells (EPCs). Methods: Sprague–Dawley rats were divided into 3 groups: Sham-operated, MI (permanent ligation of the left coronary artery) and MI + TPO. Three doses of TPO were administered weekly for 2 weeks, and outcomes were assessed at 4 or 8 weeks post-injury. Results and conclusions: TPO treatment significantly improved left ventricular function, hemodynamic parameters, myocardium morphology, neovascularization and infarct size. MI damage upregulated a large cohort of gene expressions in the infarct border zone, including those functioned in cytoskeleton organization, vascular and matrix remodeling, muscle development, cell cycling and ion transport. TPO treatment significantly reversed these modulations. While phosphorylation of janus kinase 2 (JAK2), signal transducer and activator of transcription 3 (STAT3) and protein kinase B (AKT) was modified in MI animals, TPO treatment regulated phosphorylation of STAT3 and extracellular signal-regulated kinases (ERK), and bone morphogenetic protein 1 (BMP1) protein level. TPO also increased EPC colonies in the bone marrow of MI animals. Our data showed that TPO alleviated damages of heart tissues from MI insults, possibly mediated by multi-factorial mechanisms including suppression of over-reacted ventricular remodeling, regulation of TPO downstream signals and mobilization of endothelial progenitor cells. TPO could be developed for treatment of cardiac damages. © 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Thrombopoietin (TPO) is an endogenous cytokine known to promote hematopoietic progenitor cells and the megakaryocytic/platelet lineage. TPO mimetics have recently been approved for treatment of thrombocytopenia [1–3]. In previous studies, we provided evidence ☆ Grant support: This project was supported by the Li Ka Shing Institute of Health Sciences Grant, The Chinese University of Hong Kong, and Earmarked Grant CUHK4521/05 M and Direct Grant 2008.1.070 from the Research Grant Council of Hong Kong. ⁎ Corresponding author at: Department of Paediatrics, The Chinese University of Hong Kong, 6th Fl, Clinical Sciences Block, The Prince of Wales Hospital, Shatin, NT, Hong Kong. Tel.: + 852 2632 7823; fax: + 852 2636 0020. E-mail address: [email protected] (R.Y.T. Sung). 1 These authors contributed equally to this manuscript. 0167-5273/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2012.06.038

that TPO reduced cardiac damage caused by acute or chronic treatments of doxorubicin in animal models [4,5]. In these animals, TPO significantly improved ventricular performance and ameliorated severe changes in gene expressions relevant for maintenance of heart functions, including apoptosis, blood vessel and matrix remodeling, cell division, ion channels and contractile proteins [4]. The protective mechanism might also involve phosphatidylinositol 3-kinase (PI3K)/ protein kinase B (AKT) and p42/p44 extracellular signal-regulated kinases (ERK 1/2) activation and bone marrow endothelial progenitor cell (EPC) induction [4]. Since these mechanisms are strongly implicated in heart damages caused by myocardial infarction (MI) in human and animal models [6–9], we pursued to investigate the efficacy of TPO treatment in a model of ischemic cardiomyopathy and the associated expression profile using a genome-wide microarray assay. Our results provided the first evidence that TPO improved

K.Y.Y. Chan et al. / International Journal of Cardiology 167 (2013) 2546–2554

ventricular morphology and functions post-MI, possibly mediated by down-regulation of excessive remodeling responses. 2. Materials and methods 2.1. Rat model of MI The study was approved by the Animal Research Ethics Committee, The Chinese University of Hong Kong, Hong Kong and in compliance with the ethical policy of US National Institutes of Health (NIH A5613-01). Male Sprague–Dawley rats weighing 260–300 g (Laboratory Animal Services Centre, The Chinese University of Hong Kong) were allocated into 3 treatment groups: Sham-operated, MI and MI + TPO. Animals were subjected to left coronary artery ligation or sham surgery under anesthesia with single injection of i.p. ketamine (Alfasan, Woerden, Holland) and xylazine (Alfasan) at 75 mg/kg and 10 mg/kg, respectively. This dosage of anesthesia was adequate to prevent the animals from conscious activity and response to external stimuli. To MI + TPO animals, TPO (PeproTech, Rocky Hill, NJ) was administered at 10 μg/kg i.p. immediately after induction of MI, and 3 doses weekly for 2 weeks. The Sham-operated and MI animals were given saline under the same schedule. At week 4 or week 8, twodimensional echocardiography and hemodynamic parameters [5,10] were measured on animals under anesthesia with i.p. ketamine (75 mg/kg) and xylazine (10 mg/kg). Animals were sacrificed after the echocardiography or hemodynamic procedure by cervical dislocation and under anesthesia. Heart tissues at the MI border zone were then harvested for morphology, capillary density, molecular and protein analyses. Endothelial progenitor cells in the bone marrow of the animals were also determined.

2.2. Heart function assessment by echocardiography The left ventricular (LV) function of animals was evaluated at baseline (1 day before any treatment) and before sacrifice at week 4 or week 8 [4,5]. Transthoracic echocardiography was performed on animals under anesthesia, using Sonos 7500 (Philips Ultrasound, MA) with a 7.5–12 MHz probe. M-mode echocardiography of the LV was conducted at the papillary muscle level, as guided by two-dimensional short-axis images. Left ventricular end-diastolic dimensions (LVEDD) and left ventricular endsystolic dimensions (LVESD) were measured on M-mode tracings. The LV fractional shortening (% FS) was calculated as [(LVEDD − LVESD) / LVEDD] × 100. Cardiac output (CO) was calculated by the M-mode Teichholz formula.

2.3. Hemodynamic measurements Animals were anesthetized with ketamine (75 mg/kg) and xylazine (10 mg/kg), and a microtip pressure transducer catheter (Millar Instruments, Houston, TX) was inserted into the left ventricle. Various hemodynamic parameters, including indices of contractility and relaxation, the maximal rates of increase and decrease in the left ventricular pressure (dp/dt + and dp/dt −) were recorded using the MacLab instrument (AD Instruments, NSW, Australia) [10].

2.4. Infarct size Heart tissues were fixed in 4% formaldehyde, and 4 μm-thick paraffin sections were stained with Masson reagent (accustain trichrome) (Sigma) [11]. The infarct size was measured by the SPOT Advanced Software (Diagnostic Instruments, Sterling Heights, MI, USA) and presented as the average percentage of infarct epicardial length/epicardial circumference. Results are expressed as the average of 6 sections per heart.

2.5. Capillary density Heart tissues were fixed in 4% paraformaldehyde. Five micrometer sections were stained with α‐smooth muscle actin antibody (SMA, Chemicon, Millipore, Billerica, MA, USA) at 1:500 dilution overnight at 4 °C in a humidified chamber. The slides were then incubated with a secondary antibody-linked HRP polymer (Zymed, San Francisco, CA, USA) for 45 min. Signals were developed in DAB solution (Zymed, San Francisco, CA, USA). The sections were counterstained with Mayer hematoxylin. The control slide was processed with all reagents except the primary antibody. Arteriolar length density was calculated from 3 randomly selected fields at the MI border zone and septum, in a blinded manner [12].

2.6. Electronic microscopy At week 4 post-MI, cardiac tissues were immersion-fixed overnight in 2.5% phosphate-buffered glutaraldehyde (pH 7.4), postfixed for 1 h with 1% osmium tetroxide, dehydrated through a graded ethanol series, and embedded in Epon medium. Ultrathin sections were stained with uranyl acetate and lead citrate. Ultrastructural changes of cardiomyocyte, myofibril arrangement and mitochondria were observed by transmission electron microscopy (CM120, Philips, The Netherlands).

2547

2.7. Bone marrow EPC Bone marrow mononucleated cells were cultured on fibronectin (Invitrogen, Carlsbad, CA) at 5 × 106 cells per mL IMDM, supplemented with murine growth factors (PeproTech): vascular endothelial growth factor, 20 ng/mL and basic-fibroblast growth factor, 5 ng/mL. After 48 h, non-adherent cells were transferred to fresh medium. Adherent (endothelial cell colony-forming unit, CFU-EC) and non-adherent cells (endothelial colony forming cells, ECFC) were further cultured for 7 days and colonies were counted under light microscopy as previously described [4].

2.8. RNA extraction, microarray and pathway analysis Heart tissues collected at week 4 were homogenized in TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA) and purified using an RNeasy Mini kit (Qiagen, Valencia, CA) as previously described [4]. RNA samples were labeled using the Agilent Low Input Linear Amplification Labeling Kit (Agilent Technologies Inc., Palo Alto, CA). The Cy3-labeled cRNA (1.65 μg) was hybridized onto rat whole genome Agilent microarrays (4 × 44 K format, Agilent) for 17 h at 65 °C. The slides were scanned on an Agilent DNA Microarray Scanner (Agilent). The extracted feature intensities from each array were processed by the GeneSpring software version 10 (Silicon Genetics, San Carlos, CA) for normalization and log2 transformation. Per chip and per gene normalizations were done according to Agilent's recommendation using a 75th percentile shift and baseline transformation at the median of each array. Differentially expressed gene lists were generated with a two-fold cutoff between comparisons of SHAM versus MI and MI versus MI + TPO groups. Pathway analysis of differentially expressed genes including Agilent probe ID and log fold change values was performed using the MetaCore software (GeneGo, St. Joseph, MI). The gene ontology annotations of the differentially expressed genes were assigned using the web-based tool version 6 (http://david.abcc.ncifcrf.gov/) [13]. A link has been created to allow review of the microarray data GSE22489: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? token=lxothmwckakksni&acc=GSE22489.

2.9. Quantitative RT-PCR The identified targets were validated in the same set of RNA samples and an independent cohort of samples was validated by quantitative RT-PCR (qPCR) using predesigned TaqMan assays and PDH (pyruvate dehydrogenase) as the endogenous control [4,14]. Amplification was performed for 40 cycles with denaturation at 95 °C for 15 s, annealing at 60 °C for 1 min. The emission intensity was detected by the ABI realtime system 7300 (Applied Biosystems). Relative quantification values expressed as threshold cycle (Ct) were averaged and subsequently used to determine the relative expression ratios between cases.

2.10. Western-blot analysis Frozen heart tissues harvested at 4 weeks post-MI were powdered in liquid nitrogen and suspended in radioimmunoprecipitation assay lysis buffer (Sigma Chemical Co, St. Louis, MO) containing protease and phosphatase inhibitors (Roche, Basel, Switzerland). The samples were homogenized and the lysates were centrifuged at 14,000 rpm for 15 min, and the supernatant was stored at − 80 °C. The protein concentration was measured by the Bio-Rad DC protein assay (Bio-Rad Laboratories, Hercules, CA), using bovine serum albumin as a standard. Protein samples were subjected to SDS-PAGE and transferred to polyvinylidene difluoride membrane (Amersham International, Buckinghamshire, England) for antibody binding. Specific antibodies against phospho-ERK, total ERK, phospho-AKT, total AKT, phospho-JAK2, total JAK2, phospho-STAT3, total STAT3 were products of Cell Signaling Technology Inc. (Boston, MA), whereas those against BMP-1 and biglycan were from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). The blots were visualized by means of chemiluminescence (ECL, Amersham), and the signals were quantified by densitometry. The membranes were reprobed with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) which served as the loading control as described previously [15]. Expression levels are presented as the ratio of phospho-specific kinases to total kinases, or relative to those of GAPDH expression.

2.11. Statistical analysis Survival rates of animals were compared by the log rank test. Heart and endothelial progenitor cell parameters were compared by the Mann–Whitney test. Statistical analysis of the microarray data were performed by GeneSpring software. One-way Analysis of Variance (ANOVA) test with a P value b 0.01 was considered significant and the Post-Hoc analysis was performed using Student Newman–Keuls (SNK) test to reveal the difference between groups (P b 0.05). Data with P values of b 0.01 or expressions with a fold change of greater or less than 2 were further analyzed. mRNA and protein expression data were analyzed by the Mann–Whitney U test using SPSS version 17.0 (SPSS Inc, Chicago, IL). Data were presented as mean ± standard deviation (SD). The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology.

2548

K.Y.Y. Chan et al. / International Journal of Cardiology 167 (2013) 2546–2554

3. Results

Sham

10000

MI

3.1. Animal model and echocardiography There was no difference in the mortality rates of Sham-operated, MI and MI + TPO animals, which were 4.16%, 7.41% and 8.33% respectively. These animals had similar body weight throughout the study period. Echocardiographic parameters demonstrated that cardiac functions were significantly compromised in MI animals compared with Sham-operated animals at week 4 and week 8 post-surgery, as indicated by values of LVEDD, LVESD and fractional shortening (FS) (all **P values b 0.001; Table 1). The treatment with TPO for two weeks significantly reduced LEVDD (both #P b 0.05) and LVESD (both ##P = 0.001) and increased FS (both ##P b 0.001) measured at 4 and 8 weeks, compared with MI animals without TPO treatment (Table 1). Cardiac parameters at baseline were similar in the 3 groups except slightly but significantly higher LVESD and lower FS in the MI + TPO animals compared with the MI animals (#P b 0.05). This could have happened by chance during animal randomization and would not influence the interpretation of results on the positive effects of TPO on cardiac functions.

dp/dt Max (mm Hg/s)

MI+TPO 8000

##

**

6000

##

* 4000

2000

0 dp/dt(+)

dp/dt(-)

Fig. 1. Hemodynamic analysis of maximum pressure ascending and descending rates in the MI model. At week 8 post-MI, dp/dt(+) (**P = 0.006) and dp/dt(−) (*P = 0.032) were significantly lower in MI (n = 9) animals, compared with Sham-operated animals. TPO treatment (n = 5) increased these parameters (##P b 0.001).

3.5. Neovascularization 3.2. Hemodynamic parameters At week 8 post-treatments, there were no differences in systolic blood pressure, diastolic blood pressure and heart rate (HR) between the 3 groups of animals. The maximum pressure ascending [dp/ dt(+)] and descending [dp/dt(−)] rates were significantly compromised in MI animals (n = 9) (**P = 0.006 and *P = 0.032 respectively, Fig. 1). MI + TPO-treated animals (n = 12) had significantly higher dp/dt(+) and dp/dt(−) values compared with the MI animals (both ##P b 0.001, Fig. 1).

The arteriolar vessel length density as measured by staining of αsmooth muscle actin was increased in tissues from the left ventricular MI border zone (##P = 0.005) in MI + TPO animals (n = 5) at week 4, compared with that of the MI group (n = 4) (Fig. 4A). The arteriolar density in septum tissues was not altered by MI or TPO treatment. 3.6. Bone marrow EPC

MI + TPO rats (n = 17) had significantly reduced relative infarct size compared with that in MI animals (n = 12) at week 8 post-MI by over 30% (##P = 0.002, Fig. 2).

At week 4, bone marrow EPC as represented by CFU-adherent, CFU-non-adherent and total CFU were similar in the Sham-operated (n = 7) and MI (n = 9) animals. MI + TPO animals (n = 9) had significantly increased CFU-adherent (##P = 0.004; Fig. 4B), CFU-nonadherent (##P = 0.01) and total CFU (##P = 0.001), compared with their MI counterparts. A pilot study showed that TPO administration in Sham-operated animals (n = 9) did not affect their EPC levels (data not shown).

3.4. Electron microscopy

3.7. Expression array, qPCR validation and network proposal

Severe morphological damages were observed in the MI border tissues in MI animals at week 4 as indicated by degenerative cardiomyocytes with lysis of myofibrils (Fig. 3B). MI + TPO animals had milder damages, showing relatively normal cardiomyocytes with slight mitochondrial vacuolization (Fig. 3C) compared with normal morphology of Sham-operated animals (Fig. 3A).

Compared with the Sham-operated control, 5181 transcripts were up- or down-regulated over 2-fold after MI treatment, whereas 5543 transcripts were differentially expressed between MI and MI + TPO groups. Of the two comparisons, 2231 transcripts were overlapping, of which 96.5% were regulated at an inverse manner, reflecting possible cardiac protective mechanisms of TPO against MI-induced damage.

3.3. Infarct size

Table 1 Echocardiography in the rat model of MI. Time point

Group

n

LVEDD (mm)

LVESD (mm)

HR (beats/min)

FS (%)

CO (mL/min)

Baseline

Sham MI MI + TPO Sham MI MI + TPO Sham MI MI + TPO

23 26 33 24 27 39 9 14 21

6.62 ± 0.35 6.68 ± 0.31 6.72 ± 0.24 7.24 ± 0.57 9.50 ± 0.70⁎⁎ 9.06 ± 0.92# 7.34 ± 0.34 9.69 ± 0.59⁎⁎ 9.25 ± 0.65#

3.50 ± 0.42 3.48 ± 0.47 3.73 ± 0.24# 4.11 ± 0.51 7.83 ± 0.78⁎⁎

299 ± 21.6 305 ± 23.2 302 ± 17.6 282 ± 25.4 278 ± 19.3 283 ± 22.0 287 ± 25.0 272 ± 16.8 279 ± 16.8

47.4 ± 4.47 48.0 ± 5.64 44.8 ± 2.63# 43.3 ± 4.80 17.6 ± 4.10⁎⁎

167 ± 26.2 174 ± 19.8 169 ± 16.7 191 ± 38.0 200 ± 43.0 218 ± 47.2 196 ± 32.9 213 ± 36.6 232 ± 40.7

Week 4

Week 8

MI vs. Sham: ⁎P b 0.05, ⁎⁎P b 0.001. MI vs. MI + TPO: #P b 0.05, ##P b 0.001.

7.08 ± 1.03## 4.44 ± 0.41 7.98 ± 0.64⁎⁎ 7.19 ± 0.68##

22.1 ± 4.20## 40.6 ± 3.35 17.8 ± 3.18⁎⁎ 22.4 ± 3.53##

K.Y.Y. Chan et al. / International Journal of Cardiology 167 (2013) 2546–2554

2549

MI

MI MI+TPO

0.5

##

Relative Infarct Size

0.4

0.3

0.2

0.1

0 MI

MI+TPO

Treatment Fig. 2. Infarct size of MI rats. At week 8 post-MI, heart sections were stained with Masson reagent and the infarct size was determined as the average percentage of epicardial and endocardial circumference occupied by the infarct. Results are expressed as the average of 6 sections per heart. The panels represent serial transverse sections from apex to base through the hearts of MI and MI + TPO animals. The infarct size of MI + TPO animals (n = 17) was significantly smaller than that of MI animals (n = 12) (##P = 0.002).

Of thirty functionally annotated genes selected for confirmation by qPCR analysis, 26 genes showed a similar direction and magnitude of regulation as the microarray analysis (Table 2). These genes fell into the functional categories of cytoskeleton organization, blood vessel and matrix remodeling, muscle development, cell cycle and differentiation, inflammation, ion transport and metabolic process. They were all up-regulated in the MI group (mean 25.5 ± 22.8 fold), compared with the Sham-operated group. Treatment with TPO down-regulated their expressions to levels that were similar with or slightly higher than those of the Sham-operated animals (mean 3.63 ± 3.07 fold). The microarray results of the 26 targets were further analyzed for construction of a network of possible links in their biologic functions. The resulting “super-network” was shown to comprise of more than two thousand objects. A sub-network highlighting the possible role of TPO as a key regulator of the JAK2, STAT3, ERK and AKT pathways is shown in Fig. 5. 3.8. Protein expression of selected target genes In heart tissues harvested at week 4 post-MI injury, our data showed that ratios of pJAK2/total JAK2, pSTAT3/total STAT3, pAKT/ total AKT were significantly regulated by MI compared with those in Sham-operated animals (*P b 0.05; Fig. 6). Treatment with TPO significantly increased the pERK/total ERK ratio (#P = 0.043), and decreased the pSTAT3/total STAT3 ratio and BMP1 expression (#P = 0.029) (n = 3–4). There was a trend of lower expression of biglycan in MI animals compared with the Sham-operated counterparts (P = 0.083).

Fig. 3. Representative ultra-structures of infarct border zone. Electron microscopy showed noticeable differences in the tissue morphology of the 3 groups of animals at 4 weeks post-MI. A: Normal myocardium in Sham-operated rats. B: MI animals had severe morphological damage, including degenerative cardiomyocytes with lysis of myofibrils. C: Regular myofibril arrangement with mild mitochondrial vacuolization in MI + TPO animals. The scale bar represents 1 micrometer.

4. Discussion Our results demonstrated that TPO treatment alleviated cardiac damages on an in vivo model of MI, as evidenced by improvements in ventricular function and morphology. The genome-wide analysis revealed extensive gene activation, especially those involved in vascular and matrix remodeling, ion transportation and muscle development,

2550

K.Y.Y. Chan et al. / International Journal of Cardiology 167 (2013) 2546–2554

(A) Arteriolar Density (Counts/mm2)

25 ##

LV W all 20

Se ptum

15

10

5

0 Sham

MI

MI+TPO

Treatment

MI

Sham

MI +TPO

(B) 60

MI MI+TPO

CFU/107 MNC

50 40

##

Sham

## ##

30 20 10 0 CFU-ad

CFU-non-ad

Total

Bone Marrow EPC Fig. 4. Arteriolar density and bone marrow EPC of the MI model. At week 4, the arteriolar density and bone marrow EPC of the MI model were analyzed. A: The arteriolar density at the MI border and septum of the animals were examined by immunohistologic staining with α-smooth muscle actin antibody. TPO treatment significantly increased the arteriolar density at the left ventricular border zone (##P = 0.005), but not at the septum (magnification 200×). Sham (n = 2), MI (n = 4), MI + TPO (n = 5). B: EPC colonies in bone marrow MNC were similar in Sham-operated (n = 7) and MI (n = 9) animals. MI + TPO animals (n = 9) had significantly increased CFU-ad (## P = 0.004), CFU-non-ad (##P = 0.01) and total CFU (##P = 0.001) compared with those of MI animals.

as a result of ischemia caused by permanent coronary artery ligation. TPO exerted reversal effects on these gene expressions and thus might have eased the over-reacted process of ventricular remodeling. The protective mechanism of TPO might be mediated by regulation of downstream signals STAT3, ERK and BMP1, and may also involve bone marrow EPC induction. Our results showed that LVEDD, LVESD and FS values were severely compromised in animals at week 4 post-MI compared with their Sham-operated litter mates (Table 1). Results of ventricular functions and other parameters including hemodynamic data, infarct size and ultrastructure of peri-infarct tissues indicated that cardiac damages were persistent at weeks 4 and 8. Treatment with TPO for 2 weeks in MI animals significantly reduced cardiac damages, although the improvement observed at week 4 and week 8 did not recover these

parameters to levels of Sham-operated animals. It would be of interest in future studies to investigate the efficacy of a prolonged TPO treatment on increasing beneficial cardiac outcomes. The method of Yoder et al. [16] was used for quantifying CFU-EC and ECFC in the bone marrow of the animals. In accordance with the reported data, the number of EPC in the bone marrow was not significantly affected by a MI insult [17]. However, EPCs were significantly increased by TPO treatment (Fig. 4B). The data are in line with our previous study showing that TPO increased bone marrow EPC in a rat model of chronic cardiomyopathy induced by doxorubicin [4]. It is significant that we also observed increased neovascularization in the MI border zone of the animal group treated with TPO (Fig. 4A). Circulating and bone marrow EPCs have been implicated in the outcome of heart diseases. EPC levels in patients with coronary artery disease

K.Y.Y. Chan et al. / International Journal of Cardiology 167 (2013) 2546–2554

2551

Table 2 Validation of target gene expressions in the MI model. Symbol

Actin cytoskeleton organization Cnn1 Ctss Cald1 Blood vessel remodeling and matrix Bgn Col1a2 Col3a1 Col18a1 Hapln1 Lum Nid1 Cartilage and muscle development Bmp1 Clec3a Mmp13 Nrg1 Cell cycle and differentiation Cdkn1C Dmbt1 Id2 Inflammation Reg3b Ion transport and binding Cacnb3 Prkci Lrp1 Ltbp3 Gda Myo5a Metabolic process Pip4k2a Acsl4

Accession

Gene name

Relative fold change Microarray (n = 2)

qPCR (n = 3–4)

MI vs Sham

MI + TPO vs MI

MI vs Sham

MI + TPO vs MI

NM_031747 NM_017320 NM_013146

Calponin 1, basic, smooth muscle Cathepsin S Caldesmon 1

7.38 2.71 13.03

0.22 0.32 0.22

53.95⁎ 14.18⁎ 8.55⁎

0.05⁎ 0.21⁎ 0.25⁎

NM_017087 NM_053356 NM_032085 NM_053489 NM_019189 NM_031050 XM_213954

Biglycan Collagen, type I, alpha 2 Collagen, type III, alpha 1 Collagen, type XVIII, alpha 1 Hyaluronan and proteoglycan link protein 1 Lumican Nidogen-1

3.58 3.91 3.8 4.81 21.09 2.23 2.19

0.26 0.2 0.26 0.09 0.05 0.36 0.41

41.86⁎ 25.29⁎ 43.11⁎ 16.48⁎ 25.14⁎ 10.63⁎ 6.3⁎

0.14⁎ 0.17⁎ 0.13⁎ 0.13⁎ 0.07 0.29 0.56

NM_031323 NM_001108899 NM_133530 NM_031588

Bone morphogenetic protein1 C-type lectin domain family 3, member a Matrix metallopeptidase 13 Neuregulin 1

2.91 25.64 9.28 5.43

0.42 0.03 0.04 0.16

8.55⁎ 83.65 36.21⁎ 82.24⁎

0.25⁎ 0.01⁎ 0.44 0.07

NM_182735 NM_022849 NM_013060

Cyclin-dependent kinase inhibitor1C Deleted in malignant brain tumors1 Inhibitor of DNA binding 2

3.03 11.66 2.88

0.3 0.19 0.41

13.29⁎ 41.32⁎ 6.71⁎

0.24⁎ 0.17⁎ 0.30⁎

NM_053289

Regenerating islet-derived 3 beta

2.44

0.05

37.35⁎

0.11

NM_012828 NM_032059 NM_001130490

2 3.17 4.03

0.48 0.42 0.33

7.75⁎ 2.74⁎ 17.46⁎

0.24 0.41⁎ 0.21⁎

EV770136 NM_031776 NM_022178

Calcium channel, voltage-dependent, beta 3 subunit Protein kinase C, iota Low density lipoprotein-related protein 1 (alpha-2-macroglobulin receptor) Latent transforming growth factor beta binding protein 3 Guanine deaminase Myosin Va

4.62 2.77 2.24

0.3 0.38 0.44

11.95⁎ 14.73⁎ 4.29⁎

0.22⁎ 0.22⁎ 0.42⁎

NM_053926 NM_053623

Phosphatidylinositol-5-phosphate 4-kinase, type II, alpha Acyl-CoA synthetase long-chain family member 4

2.22 2.06

0.38⁎⁎ 0.35⁎⁎⁎

11.39⁎ 7.58⁎

0.22⁎ 0.24⁎

⁎

P b 0.05. P b 0.01. ⁎⁎⁎ P b 0.001. ⁎⁎

could predict the occurrence of cardiovascular events and death [8]. Treatment with human EPC increased neovascularization of the infarct border zone, reduced adverse LV remodeling and improved ventricular function of rodent models [7,18]. It would be of importance to further explore the contribution of TPO-induced bone marrow EPC to neoangiogenesis and subsequent heart repair process in our animal models. From microarray analysis, we demonstrated that a MI insult led to significant upregulation of gene expressions relevant to heart functions. These included gene expressions known to be involved in compensatory matrix and basement membrane remodeling in human and animal models, such as specific collagens [19–21], Mmp13, the small leucine-rich proteoglycan family members Bgn [20] and Lum [22–25], Nid1 [22,26], Ctss [23] and Bmp1 [27]. Col18a1 has been indicated in the regulation of EPC [28,29]. In MI tissues, we also observed over-expression of vascular remodeling genes such as Cnn1 [30–32] and Cald1 [33–35]. Lrp1 has been shown to regulate vascular smooth muscle cell proliferation and protect against atherosclerosis [36]. Yet it was also implicated in the progression of atherosclerotic lesion [37]. In MI + TPO animals, all these gene expressions were reduced when compared with the MI group, yet their levels remained higher (1.76–15.9 fold) than those of Sham-operated animals. The Nrg1 gene was strongly upregulated in MI animals, indicating that cellular repair was switched on. Treatment with TPO reduced Nrg1 expression

to 5.76 fold of that in the Sham-operated counterparts. The Nrg1/ErbB family has been strongly implicated in cardiomyocyte proliferation and repair [38,39] and has been applied in clinical trials for treating patients with chronic heart failure [40]. We also observed upregulation of novel matrix gene expressions such as Hapln1 and Clec3a, which have not been described in ischemia or other heart conditions. We observed significant increases of cell cycle inhibitor Cdkn1C in MI tissues, which might be involved in stress modulating functions [41]. We also demonstrated changes in expressions of electrical remodeling genes in MI tissues such as increased Cacnb3 [42], Id2 [43], Prkci [44] and Myo5a [45] which are important players of the cardiac conduction system. Other expressions such as Dmbt1, Reg3b, Pip4k21, Ascl4 and Ltbp3 are less known to be involved in the normal or ischemic heart. Invariably, our data suggest that in TPO-treated animals, a milder compensational repairing process was in order, compared with the drastic pathologic remodeling activities in response to an infarct. Of significance is that expressions of these genes appeared to be responsive to MI or TPO in a concerted manner and exhibited regulatory links downstream of the TPO-mediated JAK2, STAT3, ERK and AKT phosphorylation pathways (Fig. 5) [46–48]. Protein levels of 6 targeted signals have been confirmed by Western blot analysis (Fig. 6). The ERK1/2 and AKT pathways play important roles in heart development and diseases [6,9,49]. The phosphorylation cascades of ERK and AKT have been suggested as the Reperfusion Injury

2552

K.Y.Y. Chan et al. / International Journal of Cardiology 167 (2013) 2546–2554

Fig. 5. A proposed biological network of MI- and TPO-regulated genes. A biological network of shortest paths between the 26 target genes differentially regulated by MI and TPO was analyzed using the network-building algorithm of the MetaCore software. The gene symbol or name of each target gene was shown next to the object. Green, red and gray arrows between target genes represented positive, negative and unspecified interactions, respectively. Target genes were colored according to their proposed functional groups.

Salvage Kinase (RISK) pathways which mediate cell survival during ischemia heart reperfusion injury [6]. Recent studies have suggested a potential role of TPO as a physiological regulator of coronary flow, mediated by P13K/Akt-dependent eNOS phosphorylation [50]. Our hypothesis is that the MI insult activated STAT3 and AKT phosphorylation, leading to downstream recruitment of a battery of remodeling gene activities. In TPO treated animals, ERK phosphorylation was increased, resulted in dampening down of the over-reacted compensational activities through reduced STAT3 phosphorylation, as well as down-regulation of BMP-1 and other functional genes. Although TPO treatment had clearly ameliorated the severity of changes in many signals, there existed complexity in the temporal dynamics, positive and negative feedback activities, post-transcriptional controls, interactions with the microenvironment and external stimuli (e.g. circulating EPC) which had created a very complicated network of regulation. For example, we did not observe a significant effect of TPO on the protein level of biglycan despite over 40 fold increase in mRNA level, possibly due to other gene interactions, such as cleavage by MMP-13 [51] and competitive inhibition by Col18a1 [52]. A limitation of our model is that it only represents morphologic and functional outcomes at 4 and 8 weeks post-MI. The immediate protective mechanism of TPO remains to be investigated at earlier time points post-injury. Nevertheless, our study is in line with a recent report [53] on rat models that demonstrated the direct protective effect of TPO on isolated hearts or anesthetized animals subjected to 25–30 min ischemia and followed up for 3 h post-reperfusion. Their

results showed that TPO treatment improved cardiac function and reduced infarct size possibly mediated by JAK-2, p42/44MAK and KATP channels. As most human patients suffer from chronic heart failure, it would be of interest and importance to further investigate the efficacy of TPO treatment in animal models of chronic heart failure. In summary, we presented the first evidence on the long term protective effect of TPO on a rat model of ischemic injury and proposed a regulated remodeling network involving JAK2, STAT3, ERK phosphorylation and BMP1. To date, heart failure post-MI remains a global and significant cause of morbidity and mortality, despite treatment options such as cardiac assist device and surgical intervention. Our results support further the development of TPO and related compounds as cardiac protective agents, to be used in adjunct to established treatment protocols. References [1] Kantarjian HM, Giles FJ, Greenberg PL, et al. Phase 2 study of romiplostim in patients with low- or intermediate-risk myelodysplastic syndrome receiving azacitidine therapy. Blood 2010;116:3163–70. [2] Kuter DJ. Thrombopoietin and thrombopoietin mimetics in the treatment of thrombocytopenia. Annu Rev Med 2009;60:193–206. [3] Perugini M, Varelias A, Sadlon T, D'Andrea RJ. Hematopoietic growth factor mimetics: from concept to clinic. Cytokine Growth Factor Rev 2009;20:87–94. [4] Chan KY, Xiang P, Zhou L, et al. Thrombopoietin protects against doxorubicin-induced cardiomyopathy, improves cardiac function, and reversely alters specific signalling networks. Eur J Heart Fail 2011;13:366–76. [5] Li K, Sung RY, Huang WZ, et al. Thrombopoietin protects against in vitro and in vivo cardiotoxicity induced by doxorubicin. Circulation 2006;113:2211–20.

K.Y.Y. Chan et al. / International Journal of Cardiology 167 (2013) 2546–2554

1.0

Total STAT3 pERK Total ERK pAKT(Thr)

0.05 0.00

Sham

* 1.0

#

0.5

0.0

MI

BMP1/GAPDH

# 0.8 0.6 0.4 0.2 0.0

Sham

MI+TPO

MI+TPO

1.5

Total AKT

Biglycan

MI

1.5

Sham

BMP1

pERK/total ERK/total ERK ratio

pSTAT3(Tyr)

* 0.10

pAKT(Thr)/total AKT rati ratio

Total JAK2

0.15

MI

MI+TPO

0.20

*

0.15 0.10 0.05 0.00

Sham

MI

MI+TPO

Sham

MI

MI+TPO

0.5

1.0

#

0.5

GAPDH 0.0

Biglycan/GAPDH

pJAK2(Tyr)

pSTAT3/total STAT3 ratio pJAK2/total JAK2 ratio

0.20

Sham MI MI+TPO

2553

0.4 0.3 0.2 0.1 0.0

Sham

MI

MI+TPO

Fig. 6. Western blot analysis of selected targeted genes in heart tissues. At week 4 post-MI injury, heart tissues were harvested for Western blot analysis, measuring protein levels of total and phosphorylated JAK2, STAT3, ERK, AKT as well as BMP1 and biglycan. Values are presented as the ratio of phospho-specific kinases to total kinases, or relative to GAPDH expression. Data show that ratios of pJAK2/total JAK2, pSTAT3/total STAT3, pAKT/total AKT were significantly regulated by MI compared with those in Sham-operated animals (*P b 0.05). Treatment with TPO significantly increased pERK/total ERK ratio (#P = 0.043), and decreased pSTAT3/total STAT3 ratio and BMP1 expression (#P = 0.029) (n = 3–4). [6] Hausenloy DJ, Yellon DM. New directions for protecting the heart against ischaemia– reperfusion injury: targeting the Reperfusion Injury Salvage Kinase (RISK)-pathway. Cardiovasc Res 2004;61:448–60. [7] Kocher AA, Schuster MD, Szabolcs MJ, et al. Neovascularization of ischemic myocardium by human bone-marrow-derived angioblasts prevents cardiomyocyte apoptosis, reduces remodeling and improves cardiac function. Nat Med 2001;7: 430–6. [8] Werner N, Kosiol S, Schiegl T, et al. Circulating endothelial progenitor cells and cardiovascular outcomes. N Engl J Med 2005;353:999–1007. [9] Yin H, Zhang J, Lin H, et al. p38 mitogen-activated protein kinase inhibition decreases TNFalpha secretion and protects against left ventricular remodeling in rats with myocardial ischemia. Inflammation 2008;31:65–73. [10] Pacher P, Nagayama T, Mukhopadhyay P, Batkai S, Kass DA. Measurement of cardiac function using pressure–volume conductance catheter technique in mice and rats. Nat Protoc 2008;3:1422–34. [11] Zhang J, Ding L, Zhao Y, et al. Collagen-targeting vascular endothelial growth factor improves cardiac performance after myocardial infarction. Circulation 2009;119: 1776–84. [12] Christman KL, Vardanian AJ, Fang Q, Sievers RE, Fok HH, Lee RJ. Injectable fibrin scaffold improves cell transplant survival, reduces infarct expansion, and induces neovasculature formation in ischemic myocardium. J Am Coll Cardiol 2004;44:654–60. [13] Dennis Jr G, Sherman BT, Hosack DA, et al. DAVID: database for annotation, visualization, and integrated discovery. Genome Biol 2003;4:3. [14] Andersen CL, Jensen JL, Orntoft TF. Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Res 2004;64:5245–50. [15] Xiang P, Deng HY, Li K, et al. Dexrazoxane protects against doxorubicin-induced cardiomyopathy: upregulation of Akt and Erk phosphorylation in a rat model. Cancer Chemother Pharmacol 2009;63:343–9. [16] Yoder MC, Mead LE, Prater D, et al. Redefining endothelial progenitor cells via clonal analysis and hematopoietic stem/progenitor cell principals. Blood 2007;109:1801–9. [17] Delgaudine M, Gothot A, Beguin Y. Spontaneous and granulocyte colony-stimulating factor-enhanced marrow response and progenitor cell mobilization in mice after myocardial infarction. Cytotherapy 2010;12:909–18. [18] Schuh A, Liehn EA, Sasse A, et al. Transplantation of endothelial progenitor cells improves neovascularization and left ventricular function after myocardial infarction in a rat model. Basic Res Cardiol 2008;103:69–77. [19] Stanton LW, Garrard LJ, Damm D, et al. Altered patterns of gene expression in response to myocardial infarction. Circ Res 2000;86:939–45.

[20] Simkhovich BZ, Kloner RA, Poizat C, Marjoram P, Kedes LH. Gene expression profiling — a new approach in the study of myocardial ischemia. Cardiovasc Pathol 2003;12:180–5. [21] Barrans JD, Allen PD, Stamatiou D, Dzau VJ, Liew CC. Global gene expression profiling of end-stage dilated cardiomyopathy using a human cardiovascular-based cDNA microarray. Am J Pathol 2002;160:2035–43. [22] Roy S, Khanna S, Kuhn DE, et al. Transcriptome analysis of the ischemia-reperfused remodeling myocardium: temporal changes in inflammation and extracellular matrix. Physiol Genomics 2006;25:364–74. [23] Strom CC, Kruhoffer M, Knudsen S, et al. Identification of a core set of genes that signifies pathways underlying cardiac hypertrophy. Comp Funct Genomics 2004;5:459–70. [24] Tyagi SC. Extracellular matrix dynamics in heart failure: a prospect for gene therapy. J Cell Biochem 1998;68:403–10. [25] Naito Z. Role of the small leucine-rich proteoglycan (SLRP) family in pathological lesions and cancer cell growth. J Nippon Med Sch 2005;72:137–45. [26] Angel PM, Nusinow D, Brown CB, et al. Networked-based characterization of extracellular matrix proteins from adult mouse pulmonary and aortic valves. J Proteome Res 2011;10:812–23. [27] He W, Zhang L, Ni A, et al. Exogenously administered secreted frizzled related protein 2 (Sfrp2) reduces fibrosis and improves cardiac function in a rat model of myocardial infarction. Proc Natl Acad Sci USA 2010;107:21110–5. [28] Capillo M, Mancuso P, Gobbi A, et al. Continuous infusion of endostatin inhibits differentiation, mobilization, and clonogenic potential of endothelial cell progenitors. Clin Cancer Res 2003;9:377–82. [29] Khan ZA, Melero-Martin JM, Wu X, et al. Endothelial progenitor cells from infantile hemangioma and umbilical cord blood display unique cellular responses to endostatin. Blood 2006;108:915–21. [30] Kubo M, Umemoto S, Fujii K, et al. Effects of angiotensin II type 1 receptor antagonist on smooth muscle cell phenotype in intramyocardial arteries from spontaneously hypertensive rats. Hypertens Res 2004;27:685–93. [31] el-Mezgueldi M. Calponin. Int J Biochem Cell Biol 1996;28:1185–9. [32] Cai WJ, Kocsis E, Wu X, et al. Remodeling of the vascular tunica media is essential for development of collateral vessels in the canine heart. Mol Cell Biochem 2004;264:201–10. [33] Lin JJ, Li Y, Eppinga RD, Wang Q, Jin JP. Chapter 1: roles of caldesmon in cell motility and actin cytoskeleton remodeling. Int Rev Cell Mol Biol 2009;274: 1–68. [34] Mayanagi T, Sobue K. Diversification of caldesmon-linked actin cytoskeleton in cell motility. Cell Adh Migr 2011;5:150–9.

2554

K.Y.Y. Chan et al. / International Journal of Cardiology 167 (2013) 2546–2554

[35] Slomp J, Gittenberger-de Groot AC, Glukhova MA, et al. Differentiation, dedifferentiation, and apoptosis of smooth muscle cells during the development of the human ductus arteriosus. Arterioscler Thromb Vasc Biol 1997;17:1003–9. [36] Boucher P, Herz J. Signaling through LRP1: protection from atherosclerosis and beyond. Biochem Pharmacol 2011;81:1–5. [37] Llorente-Cortes V, Badimon L. LDL receptor-related protein and the vascular wall: implications for atherothrombosis. Arterioscler Thromb Vasc Biol 2005;25: 497–504. [38] Gu X, Liu X, Xu D, et al. Cardiac functional improvement in rats with myocardial infarction by up-regulating cardiac myosin light chain kinase with neuregulin. Cardiovasc Res 2010;88:334–43. [39] Pentassuglia L, Sawyer DB. The role of neuregulin-1beta/ErbB signaling in the heart. Exp Cell Res 2009;315:627–37. [40] Gao R, Zhang J, Cheng L, et al. A phase II, randomized, double-blind, multicenter, based on standard therapy, placebo-controlled study of the efficacy and safety of recombinant human neuregulin-1 in patients with chronic heart failure. J Am Coll Cardiol 2010;55:1907–14. [41] Haley SA, Zhao T, Zou L, Klysik JE, Padbury JF, Kochilas LK. Forced expression of the cell cycle inhibitor p57Kip2 in cardiomyocytes attenuates ischemia–reperfusion injury in the mouse heart. BMC Physiol 2008;8:4. [42] Murakami M, Ohba T, Xu F, et al. Modified sympathetic nerve system activity with overexpression of the voltage-dependent calcium channel beta3 subunit. J Biol Chem 2008;283:24554–60. [43] Moskowitz IP, Kim JB, Moore ML, et al. A molecular pathway including Id2, Tbx5, and Nkx2-5 required for cardiac conduction system development. Cell 2007;129: 1365–76. [44] Rohr S, Bit-Avragim N, Bdelilah-Seyfried S. Heart and soul/PRKCi and nagie oko/Mpp 5 regulate myocardial coherence and remodeling during cardiac morphogenesis. Development 2006;133:107–15.

[45] Rao VS, La Bonte LR, Xu Y, Yang Z, French BA, Guilford WH. Alterations to myofibrillar protein function in nonischemic regions of the heart early after myocardial infarction. Am J Physiol Heart Circ Physiol 2007;293:H654–9. [46] Engelbrecht AM, Ellis B. Apoptosis is mediated by cytosolic phospholipase A2 during simulated ischaemia/reperfusion-induced injury in neonatal cardiac myocytes. Prostaglandins Leukot Essent Fatty Acids 2007;77:37–43. [47] Faucher FA, Gannier FE, Lignon JM, Cosnay P, Malecot CO. Roles of PKA, PI3K, and cPLA2 in the NO-mediated negative inotropic effect of beta2-adrenoceptor agonists in guinea pig right papillary muscles. Am J Physiol Cell Physiol 2008;294: C106–17. [48] Pavoine C, Defer N. The cardiac beta2-adrenergic signalling a new role for the cPLA2. Cell Signal 2005;17:141–52. [49] Wang Y. Mitogen-activated protein kinases in heart development and diseases. Circulation 2007;116:1413–23. [50] Ramella R, Gallo MP, Spatola T, Lupia E, Alloatti G. A novel role of thrombopoietin as a physiological modulator of coronary flow. Regul Pept 2011;167:5–8. [51] Monfort J, Tardif G, Reboul P, et al. Degradation of small leucine-rich repeat proteoglycans by matrix metalloprotease-13: identification of a new biglycan cleavage site. Arthritis Res Ther 2006;8:R26. [52] Zeng X, Chen J, Miller YI, Javaherian K, Moulton KS. Endostatin binds biglycan and LDL and interferes with LDL retention to the subendothelial matrix during atherosclerosis. J Lipid Res 2005;46:1849–59. [53] Baker JE, Su J, Hsu A, et al. Human thrombopoietin reduces myocardial infarct size, apoptosis, and stunning following ischaemia/reperfusion in rats. Cardiovasc Res 2008;77:44–53.