Previous studies have demonstrated a protective effect of the Ang-(1–7)/Mas receptor axis on pathological cardiac hypertrophy. Also, the involvement of Mas receptor in exercise-induced cardiac hypertrophy has been suggested. However, the role of the Ang-(1–7)/Mas receptor on pregnancy-induced cardiac remodelling remains unknown. The objective of the present study was to evaluate the participation of the Mas receptor in the development of the cardiac hypertrophy and fibrosis induced by gestation. Female Wistar rats were divided in three groups: control, pregnant and pregnant treated with Mas receptor antagonist A-779. Wild-type (WT) and Mas-knockout (KO) mice were distributed in non-pregnant and pregnant groups. Systolic blood pressure (SBP) was measured by tail-cuff plethysmography. The medial part of the left ventricle (LV) was collected for histological analysis. Echocardiographic analysis was used to evaluate cardiac function. SBP was not changed by pregnancy or A-779 treatment in the Wistar rats. Pharmacological blockade or genetic deletion of Mas receptor attenuates the pregnancy-induced myocyte hypertrophy. The treatment with A-779 or genetic deletion of the Mas receptor increased the collagen III deposition in LV from pregnant animals without changing fibroblast proliferation. KO mice presented a lower ejection fraction (EF), fractional shortening (FS) and stroke volume (SV) and higher end systolic volume (ESV) compared with WT. Interestingly, pregnancy restored these parameters. In conclusion, these data show that although Mas receptor blockade or deletion decreases physiological hypertrophy of pregnancy, it is associated with more extracellular matrix deposition. These alterations are associated with improvement of cardiac function through a Mas-independent mechanism.
In the present study, we have evaluated the participation of the Mas receptor activation in the development of cardiac remodeling induced by gestation. We observed that pharmacological blockade or genetic deletion of Mas receptor attenuated the pregnancy-induced myocyte hypertrophy and increased the collagen III deposition in LV from pregnant animals.
Interestingly, the pregnancy restored systolic dysfunction observed in KO mice. Many studies have suggested the involvement of the ACE2/Ang-(1–7)/Mas receptor axis in the development of the physiological cardiac hypertrophy induced by exercise. In the present study, we have demonstrated that Ang-(1–7)/Mas receptor participates in the development of pregnant-induced cardiac remodeling.
These data can be useful to improve understanding of the mechanism involved in the physiological hypertrophy development. Indeed, alterations in the ACE2/Ang-(1–7)/Mas receptor axis can be related to cardiac complications observed during pregnancy.
Cardiac hypertrophy can be defined as a quantitative increase in the myocyte size  in response to pressure or volume overload, leading to an increase in contractility and normalization of cardiac wall stress [2–5].
Left ventricular hypertrophy, based on the type and duration of stimuli , results in two classes of functional cardiac phenotypes, pathological or physiological . In prolonged pathological diseases, such as hypertension, aortic valve stenosis , myocardial infarction and genetic mutations , the hypertrophied heart displays functional impairments, mainly due to an increase in fibrosis and stiffness . On the other hand, physiological hypertrophy is present during postnatal growth, chronic physical exercise and in the maternal heart during pregnancy .
Throughout pregnancy, the heart develops eccentric ventricular hypertrophy in response to volume overload caused by increased blood volume. This hypertrophy is characterized by reversible growth with normal or enhanced cardiac function . The changes in circulating sexual hormones levels in pregnancy can be related to ventricular hypertrophy development . One example is the increase in protein synthesis in the cardiac muscle in response to progesterone [13,14].
The heart is one of the most important targets for the actions of the ACE2/Ang-(1–7)/Mas axis and a growing body of research has reported an important relationship between cardiac hypertrophy development and the renin–angiotensin system (RAS). Angiotensin (Ang) II and Ang-(1–7) are the main effectors of the RAS and have counter-regulatory actions binding to AT1 and Mas receptors, respectively. AT1 receptor activation increases blood pressure and causes cardiac hypertrophy and fibrosis [15,16]. Mas receptor mRNA or protein is expressed in aorta , heart [18,19] and other tissues, such as ovary , brain, testis, kidney , liver  and lung . The activation of the Mas receptor by Ang-(1–7) or AVE, a non-peptide Mas receptor agonist, was able to prevent isoprenaline (isoproterenol)-, Ang II- and hypertension-induced cardiac remodelling [18,24,25]. Ang-(1–7) was able to prevent Ang-II-induced cardiomyocyte hypertrophy by modulating the calcineurin/NFAT signalling cascade mediated by the NO/cGMP pathway . In addition, Mercure et al.  reported that overexpression of Ang-(1–7) in rat hearts decreases the Ang-II-induced phosphorylation of c-Src and p38 kinase, whereas the increase in ERK1/2 phosphorylation was unaffected.
Recently, it was demonstrated that exercise training increases Ang-(1–7) levels and up-regulates Mas receptor in hypertrophied rat hearts , suggesting that ACE2/Ang-(1–7)/Mas receptor axis plays a role in the development of physiological cardiac hypertrophy.
Interestingly, a cross-link between RAS and ovarian hormones has been observed during pregnancy. In the gestational period, oestrogens cause an increase in the RAS activity, which augments the levels of angiotensinogen and rennin . In addition, increased Ang II levels seem to contribute to the pathophysiology of preeclampsia  and the blockade of AT1 receptor prevents cardiac remodelling associated hypertension in pregnancy .
Nevertheless, whether or not Ang-(1–7)/Mas receptor axis is involved in the genesis and development of gestational cardiac hypertrophy remains to be unravelled. In this regard, the present study aimed to evaluate the role of Mas receptor activation in the development of cardiac remodelling induced by gestation.
MATERIALS AND METHODS
Male (280–350 g) and female (180–210 g) Wistar rats were provided by the animal facilities of the Federal University of Goiás, Brazil. Male and female C57Bl/6 Wild-type (WT) and Mas-knockout (KO) mice weighing 18–20 g were obtained from the transgenic animal facilities of the Laboratory of Hypertension, Federal University of Minas Gerais, Brazil. All animals were kept in temperature-controlled rooms with 12 h light/dark cycles and the animals had free access to water and food. All animal procedures were performed in accordance with institutional guidelines approved by local authorities (ethical committee number 039/14).
Female Wistar rats were randomly assigned to three groups: control (W-NP), pregnant (W-P) and pregnant treated with Mas receptor antagonist A-779 (W-P + A-779). Several studies have shown that A-779 is a potent and selective antagonist for Ang-(1–7). A-779 was able to inhibit the antidiuretic effect of Ang-(1–7) in water-loaded rats and changes in blood pressure produced by Ang-(1–7) . In addition, it was demonstrated that specific 125I-Ang-(1–7) binding to Mas-transfected CHO cells was displaced with high affinity both by A-779 [31,32]. In agreement with these studies, A-779 blocked the cardiac effect of Ang-(1–7) or AVE 0991 [33–35].
Two days before mating (diestrus II), the rats were anesthetized with tribromoethanol (0.1 ml/10 g body weight of 2.5% solution) and received a subcuatneous implant of osmotic minipump (Alzet Osmotic Pumps, Cupertino CA) containing Mas receptor antagonist A-779 (120 μg/day/Kg). The infusion was kept over the gestational period. WT and KO mice were divided into non-pregnant (WT and KO) and pregnant (WT-P and KO-P) groups. The presence of a vaginal plug and sperm in the vagina confirmed successful mating and indicated the first day of gestation. On day 20 of gestation the animals were weighed and euthanatized by decapitation and plasma and tissues were collected.
General parameters of pregnancy
In order to assess possible changes in the fetuses and fertility we performed a midline laparotomy. The ovaries were removed and the number of corpora lutea, implantation sites and reabsorptions were determined. An implantation site with a shrunken placenta and a dissolved or discoloured brown embryo was defined as reabsorption. Pre-embryonic loss was assessed by calculating the difference between the number of corpora lutea (representing the number of eggs ovulated) and implantations sites for each female; pre-embryonic loss was set as the lack of fertilization or death of the embryo prior to implantation sites, which is defined as embryotoxic effect. The post-embryonic loss, an index that demonstrates embryotoxicity, was calculated by the following formula: (number of reabsortion /number of implantations) × 100. At the end of pregnancy, fetuses were sexed, weighed and decapitated.
Systolic blood pressure
Systolic blood pressure (SBP) was measured in the rats by noninvasive tail-cuff plethysmography (PowerLab/400 ADInstruments) at the 20th day of pregnancy.
Ang-(1–7) and Ang II measurements
Blood samples were collected in tubes containing 1 mmol/l p-hydroxymercuribenzoate, 30 mmol/l 1,10-phenanthroline, 1 mmol/l PMSF, 1 mmol/l pepstatin A and 7.5% EDTA (50 μl/ml of blood). After centrifugation, plasma samples were frozen in dry ice and stored at minus 80°C. The plasma peptide measurements were carried out using proprietary methods of the LabFar company. Briefly, the plasma peptides were extracted on to a Bond-Elut phenyl/silica cartridge (Varian) as previously described . Ang-(1–7) and Ang II levels were measured by mass spectrometry.
Morphometrical and stereological analyses
The hearts were collected and submerged in potassium chloride solution (4 mmol/l). The hearts were dissected and the medial part of the left ventricle (LV) was separated for histological analysis. LV samples were fixed in methacarn, dehydrated in ethanol, clarified in xylene and embedded in paraffin. The tissue fragments were sectioned with thicknesses of 4 μm. Haematoxylin and eosin (HE) staining was used to evaluate the cross-sectional area of myocytes (MCSA) and general morphological studies. The measures of MCSA were performed in the region corresponding to the nucleus and with visible cell limit. Picrosirius red (direct red 80) staining was used for histological analysis of total collagen (Col) deposition. The images were obtained in a system coupled with a polarizing filter, facilitating the identification of collagen fibres. Gomori's reticulin cytochemistry was also performed to specify type I and III collagens fibres.
Stereological analyses were performed on a multipoint test system with 130 points  to quantify the relative frequency of total Col and type I and III Col. All histological images were obtained under a microscope Zeiss Axio Scope A1, coupled with Zen Lite software 2012. The morphometrical and stereological analyses were performed using the software Image Pro-Plus v6.1 for Windows (Media Cybernetics).
Left ventricular sections were subjected to immunohistochemistry in order to detect the proliferating cell nuclear antigen (PCNA). The sections were deparaffinized, rehydrated through alcohol of decreasing series, and antigen retrieval was performed in a citrate buffer (pH 6.0). For detection of primary antibodies, the Leica BIOSYSTEMS NovoLink Polymer Detection System (RE7150-K) kit was employed. Endogenous peroxidase activity was neutralized with Novocastra Peroxidase Block, followed by Novocastra Protein Block to reduce the nonspecific binding of the primary antibody and polymer. Primary antibodies reactive to PCNA (mouse monoclonal IgG2a, SC 56, Santa Cruz Biotechnology) were used at a dilution of 1:100, overnight at 4°C. The Novocastra Post Primary and NovoLink Polymer were used as secondary antibodies. The sections were stained with Novocastra DAB Chromogen, and finally counterstained with haematoxylin. The histological sections were analysed using an Olympus BX43 light microscope (Olympus). For PCNA quantification, 30 microscopy fields were employed for each experimental group. The number of PCNA-positive cells of the connective tissue was obtained as a relative frequency in relation to the total number of cells counted in this compartment (PCNA-positive and PCNA-negative cells). These PCNA-positive cells were taken as being fibroblasts, considering they are the most common cells of the normal connective tissue in myocardium. Moreover, cells with nuclei showing morphological aspects different from those observed for most of the fibroblasts (flattened and elongated nuclei) were not counted.
Cardiac morphology and function in adult mice (10–12 weeks old) were assessed noninvasively using a high-frequency, high-resolution echocardiographic system consisting of a VEVO 2100 ultrasound machine equipped with a 30–40 MHz bifrequencial transducer (Visual Sonics). The mice were anaesthetized with 5% isoflurane during 1 min for induction. Anaesthesia was maintained via a nose cone with 1.0 to 1.25% isoflurane. The anterior chest was shaved and the mice were placed in supine position on an imaging stage equipped with built-in electrocardiographic electrodes for continuous heart rate monitoring and a heater to maintain the body temperature at 37°C. Images of short and long axis views of the heart were acquired using bidimensional and M-modes. LV chamber dimensions and wall thicknesses were measured. The parameters of cardiac function were: ejection fraction (EF), fractional shortening (FS), stroke volume (SV), end systolic volume (ESV), cardiac output (CO) and isovolumetric relaxation time (IVRT). All measurements and calculations were made according to the guidelines of the American Society of Echocardiography.
The results are presented as the mean ± S.E.M. General parameters of pregnancy were compared by Student's t test. One-way analysis of variance followed by the Newman–Keuls post-test was used to analyse blood pressure, morphometrical and stereological analysis and echocardiographic data, and PCNA quantification. All statistical analyses were considered significant at P<0.05.
Effects of the Mas receptor blockade on pregnancy-induced cardiac changes
Pregnancy did not alter SBP in the rats at the late phase of the gestation period (110±7.2 compared with 105±6.6 mmHg in pregnant rats). Indeed, treatment with A-779 did not change SBP in pregnant rats (105±6.6 compared with 105 ± 5.9 mmHg in W-P + A-779 animals).
As expected, pregnancy induced an increase in MCSA from Wistar rats. However, the Mas receptor blockade attenuates the myocyte hypertrophy (Figures 1A and 1B). Thereafter, Col deposition in the LV was evaluated. Pregnancy did not change the total Col deposition in Wistar rats. However, A-779 treatment induced an increase in the Col deposition in the LV from Wistar rats (Figures 2A and 2B). To evaluate whether the changes in Col deposition were related to fibroblast proliferation, staining of PCNA was performed. The number of PCNA-stained positive cells was not modified by pregnancy or by A-779 treatment (Supplementary Figures S1A and S1B).
Mas receptor blockade attenuates the pregnancy-induced myocyte hypertrophy in Wistar hearts
Mas receptor blockade increase the Collagen deposition in Wistar rats
Plasma levels of Ang-(1–7) and Ang II
Plasma levels of Ang-(1–7) and Ang II
Effects of the genetic deletion of the Mas receptor on pregnancy-induced cardiac changes
In order to confirm the changes observed by pharmacological blockade of the Mas receptor, we investigated the pregnancy-induced cardiac changes in Mas-KO mice. The cardiomyocyte hypertrophy promoted by pregnancy was also observed in WT mice hearts. Similarly to pharmacological blockade, the deletion of the Mas receptor attenuated the hypertrophy induced by pregnancy (Figures 4A and 4B).
Deletion of the Mas receptor attenuates the hypertrophy induced by pregnancy
Previous studies have demonstrated that male KO mice present a fibrotic profile in the heart . Here, no difference was found in the total Col deposition in non-pregnant female KO mice hearts. However, pregnancy increased the total Col deposition in KO-P, but not in WT mice hearts (Figures 5A and 5B). The deposition of Col I (Figure 6A) and Col III (Figure 6B) did not differ between hearts from non-pregnant KO and WT mice. However, we observed an increased Col III deposition in KO-P mice hearts. The expression of Col I was not changed by pregnancy (Figures 6A–6C). Pregnancy did not alter the number of PCNA-stained positive cells in WT or KO mice (Supplementary Figures S2A and S2B).
Pregnancy increase the total Collagen deposition in KO mice hearts
Pregnancy increase the Col III deposition in KO mice hearts
The deletion of Mas receptor induced a significant reduction in systolic cardiac function in female mice. As observed in Figure 6, KO mice presented a lower EF (Figure 7A), FS (Figure 7B), SV (Figure 7C) and higher ESV (Figure 7D) compared with WT. Interestingly, pregnancy restored these parameters. CO was significantly higher in KO-P animals compared with KO (Figure 7E). Diastolic function was not different among the groups as observed by IVRT (Figure 7F).
Echocardiographic analysis in WT and Mas-knockout mice
General parameters of pregnancy
Treatment with Mas receptor antagonist or genetic deletion of Mas receptor did not affect the fertility of the animals (Supplementary Tables S1 and S2) or gestational body weight gain (Supplementary Figures S3A and S3B). Mas receptor blockade decreased placental weight without changing fetal weight, which resulted in increase of the fetal to placenta ratio (Supplementary Figures S4A, S4C and S4E). In contrast, placental weight was not changed in KO mice, but fetal weight was lower in these animals, thus resulting in a decreased fetal to placenta ratio (Supplementary Figures S4B, S4D and S4F).
The major findings of the present paper were: (i) blockade of Mas receptor in Wistar rats or its deletion in mice attenuated the myocyte hypertrophy development in pregnant animals; (ii) hearts from pregnant KO mice displayed an exacerbated Col deposition in LV; (iii) female KO mice presented systolic dysfunction (lower EF, FS, SV and higher ESV compared with WT), but these parameters were restored by pregnancy in KO mice.
Ang-(1–7)/Mas receptor axis is involved in the development of cardiac hypertrophy in physiological and pathological conditions. Increase in circulating Ang-(1–7) levels in transgenic rats reduced the cardiac hypertrophy induced by either isoprenaline  or Ang II . Similarly, Ang-(1–7) treatment prevented Ang II- and hypertension-induced cardiac hypertrophy [34,39]. Furthermore, the involvement of the Ang-(1–7) and Mas receptor in the development of the exercise-induced cardiac hypertrophy has been suggested . Swimming training increases Mas receptor expression , ACE2 activity and protein expression and Ang-(1–7) levels in the heart . Pregnancy is another physiological stimulus that results in cardiac hypertrophy . In our study, we observed that myocyte hypertrophy was attenuated in both KO mice and A-779 treated Wistar rats. We could hypothesize that these results are due to changes in the plasma concentration of Ang-(1–7) or Ang II, since it has been observed that the Ang-(1–7) and Ang II increase during gestation in women [42,43]. However, in our study, we did not observe any difference in the plasma concentration of these peptides in the rats groups. Accordingly, previous studies have also demonstrated that, in contrast with the human pregnant subjects, there are no significant changes in plasma concentration of Ang II or Ang-(1–7) at the end of pregnancy in rats  or mice . Thus, we can assert that the phenotypic changes in the heart induced by A-779 were not due to changes of the Ang II or Ang-(1–7) levels.
Previous studies have described that Ang-(1–7) prevents cardiac fibrosis induced by chronic infusion of Ang II , isoprenaline  or in DOCA-salt model of hypertension . Also, genetic deletion of Mas receptor leads to higher levels of extracellular matrix proteins in both right and left ventricles from male mice . In our study, we did not observe difference in total Col deposition between non-pregnant KO and WT mice hearts; however, pregnancy induced Col deposition only in KO hearts. This suggests that the cardiac fibrosis may be a gender driven feature, probably governed by ovarian hormones. In this regard, female mice submitted to transverse aortic constriction presented lower fibrosis than male mice . Interestingly, pregnancy increased Col III deposition in KO mice hearts, but not in WT hearts. Despite the mechanisms underlying Col deposition in heart during pregnancy remaining unknown, the increase in Col III expression observed in KO hearts might be related to the matrix metalloproteinases activity, since it was previously demonstrated that the active form of the matrix metalloproteinases-2, which cleaves Col III, is down-regulated in hearts of neonatal and adult KO mice . Indeed, the increase in Col deposition observed in hearts from pregnant KO mice and pregnant Wistar rats treated with A-779 could be related to fibroblast proliferation. However, the number of PCNA-stained positive fibroblast was not modified in any groups. This result suggests that the increase in collagen deposition was probably due to higher fibroblast activity in the collagen synthesis.
Previous studies have demonstrated the involvement of the ovarian hormones in regulating heart remodelling. Progesterone increased neonatal rat ventricular myocytes cell size  and cardiac mass of non-pregnant female mice . Furthermore, progesterone and 17-β-oestradiol inhibited female cardiac fibroblast proliferation  and collagen synthesis [47,48]. Also, Costa et al.  showed that the mean levels of Ang-(1–7) in proestrus and estrus are significantly higher than those in metestrus and diestrus and Ang-(1–7) induced an increase in oestradiol and progesterone production in the ovary of immature rats. Thus, it is feasible to hypothesize that the deletion or blockade of the Mas receptor attenuates the increase of the ovarian hormone in pregnant animals thereby resulting in reduced myocyte hypertrophy and higher collagen deposition.
In order to evaluate whether the changes in morphology, specifically Col deposition, would influence the cardiac function, the echocardiography analyses were performed in pregnant and non-pregnant KO mice. We observed a significant impairment in systolic function (EF, FS, SV and ESV) in non-pregnant KO mice. Interestingly, these parameters were restored in the late phase of the gestational period. Previous studies have described that genetic deletion of Mas, in male mice, leads to a marked impairment of cardiac function [50,51]. These alterations appear to be caused, at least in part, by severe alterations in Col protein expression in LVs . Here, we did not observe differences in Col deposition between WT and KO female mice hearts, suggesting a dysfunction in the contractile machinery of the cardiomyocytes from KO hearts. Even so, pregnant KO mice presented an improvement in the cardiac function. This data can be, at least in part due to the higher steroid sex hormones levels in the gestational period, since previous studies have indicated that oestradiol can improve cardiac muscle contractility . In addition, the increase in Col III can also participate in the improvement of the cardiac function, given that this Col forms a reticular network with compliant profile. So, further studies are necessary to better clarify the mechanism involved in the improvement of the cardiac function observed in pregnant KO mice.
In addition to the alterations in the KO mice hearts, we also observed that fetal weight was significantly lower in KO mice as compared with WT mice. Similar results were previously observed in two different backgrounds of Mas-knockout mice . This finding can be related to impairment in umbilical blood flow, which can affect the nutrients transfer to the fetus. Accordingly, it was demonstrated that Mas deficient mice present systemic endothelium dysfunction . Also, pregnant ACE 2 knockout mice, which have reduced plasma concentration of Ang-(1–7), developed uterine artery dysfunction associated with placental hypoxia and reduced umbilical blood flow velocity .
Taken together, these data show that although Mas receptor blockade or deletion decrease physiological hypertrophy of pregnancy, it is associated with more extracellular matrix deposition. These alterations are associated with improvement of the cardiac function through Mas-independent mechanism.
Cintia do Carmo e Silva performed the treatment with A-779 and histological analyses in wistar rats. Jônathas Almeida conducted the experiments in WT and KO mice. Larissa Macedo and Gustavo Pedrino performed the plethysmography. Diego Colugnati, Fernanda Alcantara dos Santos and Manoel Biancardi developed and interpreted the histochemical analyses. Marcos Melo performed and interpreted the echocardiography. Robson Souza dos Santos, Renata Mazaro-Costa and Adryano de Carvalho helped with the pregnancy model, designed and revised the paper. Elizabeth Mendes designed the experiments, interpreted the data and revised the paper. Carlos de Castro designed the study, interpreted the data and wrote the paper.
This work was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (Edital Universal n° 14/2012); and the Fundação de Amparo à Pesquisa do Estado de Goiás (Edital Universal n° 05/2012). RMC is recipient of PET SESu MEC fellowship.
end systolic volume
isovolumetric relaxation time
myocytes cross-sectional area
proliferating cell nuclear antigen
systolic blood pressure