Neuregulin-1 Promotes Myocardial Angiogenesis in Acute Myocardial Infarction Through Regulating PI3K-AKT-eNOS and VEGF/VEGFR2 Signal Pathways

Background: Myocardial angiogenesis is central to the recovery of acute myocardial infarction (AMI). Neuregulin-1 (NRG-1) plays a critical role in cardiac function, although its role in myocardial angiogenesis is still unclear. The aim of this study was to investigate the effects of NRG-1 in myocardial angiogenesis in a rat model of AMI, and elucidate the underlying mechanisms. Methods: AMI was induced by a single ligation of left anterior descending coronary artery, followed by intravenous injection of recombinant human NRG-1 or normal saline for 8 consecutive days. The cardiac function indices were measured using the catheter MPA cardiac function analysis system. Histo-pathological changes were observed by HE. Microvessel density (MVD) was measured by CD31 and α-SMA immunostaining. The expression levels of other proteins were assessed by Western blotting. Results: NRG-1 improved cardiac function and alleviated myocardial damage induced by AMI. Compared to the sham-operated group, the capillary density and arteriole density increased after AMI (P<0.05), and were augmented by NRG-1 which also signicantly increased the left ventricular function (P<0.05). Furthermore, Compared with sham group, PI3K-AKT-eNOS signaling was decreased signicantly (P<0.05) whereas VEGF/ VEGFR2 signaling was signicantly increased (cid:0) P<0.05 (cid:0) in AMI group and both of therm were further upregulated by NRG-1 (P > 0.05). Conclusion: NRG-1 improved cardiac function and promoted myocardial angiogenesis post AMI by up-regulating VEGF and activating the PI3K-Akt-eNOS pathway.


Introduction
Acute myocardial infarction (AMI) is the leading cause of mortality and disability associated with cardiovascular diseases worldwide [1]. The current method for treating AMI is revascularization through percutaneous coronary intervention (PCI) or coronary artery bypass grafting, which however are limited by coronary artery diffuse stenosis [2] and iodinated contrast media hypersensitivity [3], resulting in the high long-term mortality of patients post-AMI [4]. Angiogenesis, or the generation of new blood vessels from pre-existing capillaries, is a vital compensatory mechanism in myocardial ischemia which supplies the ischemic tissue with blood [5]. Therefore, therapeutic angiogenesis is a viable strategy to improve reperfusion and cardiac function after myocardial infarction [6], and angiogenesis induction in AMI has gained considerable attention in recent years [7].
Neuregulin-1 (NRG-1) is a paracrine growth factor synthesized by the endocardium and microvascular endothelium, and is involved in adult heart maintenance [8]. Studies show that NRG-1 increases survival of cardiomyocytes [9], reduces oxidative stress [10] and induces cardiomyocyte proliferation post heart injury [11]. A recent phase II clinical trial revealed that the short-term administration of recombinant human NRG-1 can improve cardiac function after heart failure [12]. In addition, there are reports indicating a pro-angiogenic role of NRG-1 as well. For instance, Ebner et al. found that remote application of recombinant NRG-1b protected the heart against early reperfusion injury without affecting hemodynamics [13]. Similarly, Hedhli et al. reported that NRG-1 released by endothelial cells is necessary for arteriogenesis and angiogenesis, and the administration of exogenous NRG-1 can promote both processes [14]. However, the precise role of NRG-1 in myocardial angiogenesis remains to be elucidated.
In a previous study, we found that NRG-1 promoted myocardial angiogenesis in a rat model of diabetic cardiomyopathy along with other angiogenic factors [15]. Phosphatidylinositol 3 kinase(PI3K) signal pathway is closely related to angiogenesis and vascular endothelial growth factor (VEGF) is the key and strongest angiogenic factor to induce angiogenesis. Law et al. reported a mechanistic link between NRG-1 and the PI3K-AKT signaling pathway in patients with schizophrenia [16]. In addition, Ferrara et al found that NRG-1-induced angiogenesis can be maintained by both VEGF-dependent or independent pathways [17]. We hypothesized that the mechanisms of NRG-1 promoting angiogenesis in AMI may be associated with VEGF or PI3K-AKT-eNOS signaling pathway. In this study, we explored the effects of NRG-1 in myocardial angiogenesis and elucidated the underlying mechanisms in a rat model of AMI. Institute of Health. The animals were randomly divided into the sham-operated, and the untreated and NRG-1-treated AMI groups. The AMI model was established as previously described [18]. Brie y, the rats (n = 24) were anesthetized with an intraperitoneal injection of 4% chloral hydrate solution (1 ml/100 g body weight), and then rapidly intubated and mechanically ventilated (tidal volume, 4 ml/100 g body weight; ventilation rate 80 strokes/min). The chest was then opened gently by left thoracotomy at the 3rd intercostal space, and the left anterior descending coronary artery was ligated 1-2 mm from the tip of the left atrial appendage using a 6 − 0 polyester suture. ST-segment elevation and pathological Q waves were con rmed by electrocardiography. The same procedure was followed for the sham-operated rats (n = 9) except for the ligation step. Recombinant human NRG-1 (rhNRG-1) was administered daily for 8 days at the dose of 10 g/kg via the intravenous route, while the sham-operated and untreated AMI rats were injected with the same volume of normal saline.

Evaluation of cardiac function
The hemodynamic and cardiac functional parameters were evaluated using a catheter MPA cardiac function analysis system as described previously [19]. Eight days after the operation, all rats were weighted and anesthetized with 4% chloralhydrate (1 ml/100 g). After intubation and mechanical ventilation at stable tidal volume (4 ml/100 g), left anterolateral longitudinal thoracotomy was performed to expose the apex, and a catheter was inserted into the left ventricle and connected to the analysis system. Stable conditions were established, and the left ventricular systolic pressure (LVSP), left ventricular end-diastolic pressure (LVEDP) and maximum increase and decrease rates of left ventricular pressure (± dp/dt max ) were detected and recorded.

Cardiac tissue harvest
After evaluating the cardiac function, heart tissues were harvested and weighted. The tissues from the paler infarct regions and peri-infarct regions within 5 mm of the former were collected from the left ventricle, and divided into two parts. One potion of each was snap froze in liquid nitrogen and stored at -80˚C, and the other was xed in 10% neutral buffered formalin, embedded into para n and cut into 5 µm thick section.

Hematoxylin-eosin (HE) staining
Cardiac tissue sections were dewaxed in xylene for 20 min, and dehydrated through an ethanol gradient. The sections were then immersed sequentialy in hematoxylin (5 min), 1% hydrochloric acid (30 s) and 1% eosin -alcohol (5 min), followed by another round of gradient alcohol dehydration. Five random elds from each section were evaluated under 200X magni cation by two investigators who were blinded to the grouping. The area of the diseased myocardium relative to that of the entire myocardium was calculated in terms of in ammation, ischemia and necrosis, and the samples were histo-pathologically scored as follows: 0 -absence of the above signs, 1 -< 25% of the myocardium, 2-25%-50% of the myocardium, 3-50%-75% of the myocardium, and 4 -> 75% of the myocardium showing the above signs.

Immunohistochemistry
Myocardial cross sections were cleared and dehydrated as described, and treated with 3% hydrogen peroxide to quench the endogenous peroxidases. After incubating with rabbit anti-α-SMA (Abcam, 1:200 dilution) and rabbit anti-CD31 (Abcam, 1:100 dilution) antibodies for 2-3 hours, the sections were then treated with real time MaxVision, followed by the secondary antibody (DAKO, 1:14000) for 90 minutes. The sections were developed using DAB solution and counterstained with hematoxylin. Microvessels in the infarct and peri-infarct regions were evaluated in two separate slides from at least 6 random elds under 400X magni cation. Microvessel density (MVD) was calculated by dividing the number of microvessels by the eld area.
The blots were washed thrice and then incubated with horseradish peroxidase-conjugated secondary antibody (DAKO, 1:14000) for 2 hours at 4℃. Enhanced chemiluminescence detection system was used to developed the positive bands, and their densities were analyzed using Image-Pro Plus6.0 software.

Statistical analysis
The continuous variables were expressed as the mean ± SD. All data were analyzed using the SPSS 17.0 statistic software package by two researchers blinded to the study. Two-sample t-test was used to compare two groups, one-way analysis of variance for multiple groups, and least signi cant difference ttest for pair-wise comparison between groups. P values less than 0.05 were considered statistically signi cant.

AMI model was successfully established
The myocardial tissues below the artery ligation line appeared pale during surgery, and 8 days later, the infarcted region was distinctly pale and thin compared to the healthy tissues (Fig. 1A). While all shamoperated animals survived during the experiment, 7 of the 33 rats that underwent arterial ligation were dead within 24 h of surgery. Furthermore, the levels of phosphorylated ErbB receptors increased in the myocardium of healthy animals following administration of rhNRG-1, indicating high e cacy of the latter (Fig. 1B). The indices of cardiac function, including LVSP, +dp/dtmax and -dp/dtmax, were signi cantly reduced while LVEDP was increased in the AMI group compared to the sham-operated group, and restored by rhNRG-1 ( Fig. 1C and Table 1).
NRG-1 alleviated myocardial injury and restored angiogenesis post AMI As shown in Fig. 1D, the myocardial tissues of the sham-operated animals were arranged in an orderly manner, whereas massive in ltration of in ammatory cells was seen in the cardiac tissues following AMI. NRG-1 treatment signi cantly repaired the myocardial injury, as compared to that of the untreated AMI rats .
We next analyzed the in situ expression of CD31 and α-SMA as the respective markers of capillary and arteriole densities [20,21]. Compared to the myocardium of the sham-operated animals, that of the AMI group showed similar capillary density in the infarct region but signi cantly higher density in the periinfarct region. NRG-1 treatment on the other hand signi cantly increased the capillary density in the infarct region as well (Fig. 2). Furthermore, the arteriole density was signi cantly increased in both the infarct and peri-infarct regions of the AMI group, and was augmented further by NRG-1 (Fig. 3). Taken together, NRG-1 treatment not only improved the myocardial architecture but also arteriole density post AMI .
NRG-1 promotes angiogenesis via the VEGF/VEGFR2 and PI3K/AKT/eNOS signaling pathways To elucidate the mechanisms underlying the pro-angiogenic effects of NRG-1, we next examined the expression levels of the VEGF/VEGFR2 and PI3K-AKT-eNOS signaling pathways. Compared to the shamoperated group, VEGF was slightly upregulated in the infarct and peri-infarct regions of the AMI group, whereas VEGFR2 expression levels were slightly increased in the infarct region and unaltered in the periinfarct region. In contrast, both VEGF and VEGFR2 were signi cantly upregulated in the infarct region and only slightly increased in the peri-infarct region of the NRG-1-treated animals (Fig. 4). Furthermore, AMI resulted in a signi cant decrease in the levels of PI3K and p-AKT in the infarct region, and a minor reduction in the peri-infarct region. NRG-1 administration signi cantly upregulated PI3K, p-AKT and p-eNOS in the infarct as well as in the peri-infarct region (Fig. 5). Taken together, the pro-angiogenic effects of NRG-1 are likely mediated via activation of the VEGF/VEGFR2 and PI3K/AKT/eNOS pathways (Fig. 6).

Discussion
We report for the rst time that NRG-1 improves the density of capillaries and arterioles in the different ischemic regions post AMI. In addition, NRG-1 also improved left ventricular function after infarction and inhibited myocardial brosis in a rat model of AMI. Mechanistically, the protective effects of NRG-1 were mediated via activation of the proangiogenic VEGF/VEGFR2 and PI3K/AKT/eNOS pathways in the cardiac tissues.
The rat model of AMI has been established previously [18]. AMI results in myocardial cell apoptosis, cardiac in ammation and myocardial brosis that adversely affect cardiac systolic and diastolic functions [22]. It eventually leads to ventricular wall thinning, left ventricular cavity expansion and cardiac ventricular remodeling, all of which increase the risk of heart failure [23]. We found that NRG-1 signi cantly inhibited myocardial collagen deposition post-AMI and improved cardiac function. The NRG-1/ErbB system is crucial for adapting to cardiac demands, and its disruption reduces tolerance to myocardial ischemia [24]. NRG-1 administration also improves cardiac function in animal models of ischemic heart disease, dilated cardiomyopathy and viral cardiomyopathy [8]. At the cellular and molecular level, it improves cardiomyocyte survival [25], ameliorates myo lament injury [26], prevents mitochondrial dysfunction [27], enhances reparative in ammatory response [28], and increases calcium intake in the sarcoplasmic reticulum [29].
Our study showed a pro-angiogenic effect of NRG-1 in an animal model of AMI, which was the likely basis of improved cardiac function. This is consistent with Hedhli et al. who reported that exogenous injection of NRG-1 improved blood ow in an ischemia model [14], as well as our previous study wherein we found that NRG-1 increased the levels of angiogenic factors in vitro and promoted myocardial angiogenesis in vivo [15]. In the current study, NRG-1 treatment increased capillary density in peri-infarct region and arteriole density in the infarct region. Angiogenesis, or the formation of new blood vessels from pre-existing vascular beds, is activated during ischemia [30]. The ischemic and hypoxic conditions in the core infarct regions urgently require vascular compensation, resulting in increased collateral arterioles following NRG-1 treatment. However, since the injury in the peripheral region is relatively minor, NRG-1 promotes the formation of new capillary cavities that act as reserves of mature blood vessels.
Angiogenesis is a complex process that requires the participation of various angiogenic factors, pathways and cells [31,32]. a rat model of diabetic cardiomyopathy along with other angiogenic factors [15]. PI3K-AKT-eNOS signaling cascade is closely related to angiogenesis and VEGF is the key and strongest angiogenic factor to induce angiogenesis. We found that NRG-1 signi cantly activated this proangiogenic pathway in both the infarct and peri-infarct regions, which is consistent with previous studies showing that NRG-1 induces angiogenic factors in various tissues and cells. For instance, NRG-1β promotes glucose uptake in neonatal rat cardiomyocytes via the PI3K/Akt pathway [33], and induces VEGF secretion by endothelial cells [34]. The PI3K-Akt-eNOS pathway lies downstream of VEGF/VEGFR [35,36], although the relationship is more complex in vivo. Nevertheless, our results clearly show that they are critical to myocardial angiogenesis post AMI.
Our study has several limitations that should be acknowledged. First, although the hemodynamic parameters improved signi cantly following NRG-1 treatment, the actual infarct size was not measured. Second, the use of α-SMA as a marker of angiogenesis is controversial despite recent studies showing that activated broblasts express higher levels of α-SMA following myocardial infarction [5,36]. Third, the effect of NRG-1 on the angiogenic pathways was not validated by genetic or pharmacological inhibition of the said factors. Finally, it remains to be clari ed whether NRG-1 regulates other angiogenic pathways in AMI.
In conclusion, NRG-1 improved cardiac function and promoted myocardial angiogenesis in a rat model of AMI by up-regulating VEGF and activating the PI3K-Akt-eNOS pathway.
Abbreviations AMI acute myocardial infarction;NRG-1:Neuregulin-1;LVEDP:left ventricular end-diastolic pressure; LVSP:left ventricular systolic pressure;±dp/dt max :decrease rates of left ventricular pressure;MVD:Microvessel density ;PI3K:Phosphatidylinositol 3 kinase;VEGF:vascular endothelial growth factor Declarations All data generated or analysed during this study are included in this published article and the datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.      Representative immunoblots showing VEGF/VEGFR2 levels in the three groups 8 days post-MI (n=8). A.