Reduction in skeletal muscle fibrosis of spontaneously hypertensive rats after laceration by microRNA targeting angiotensin II receptor

Regeneration of injured skeletal muscles is affected by fibrosis, which can be improved by the administration of angiotensin II (AngII) receptor (ATR) blockers in normotensive animals. However, the role of ATR in skeletal muscle fibrosis in hypertensive organisms has not been investigated yet. The tibialis anterior (TA) muscle of spontaneously hypertensive (SHR) and Wistar rats (WR) were lacerated and a lentivector encoding a microRNA targeting AngII receptor type 1 (At1) (Lv-mirAT1a) or control (Lv-mirCTL) was injected. The TA muscles were collected after 30 days to evaluate fibrosis by histology and gene expression by real-time quantitative PCR (RT-qPCR) and Western blot. SHR’s myoblasts were analyzed by RT-qPCR, 48 h after transduction. In the SHR’s TA, AT1 protein expression was 23.5-fold higher than in WR without injury, but no difference was observed in the angiotensin II receptor type 2 (AT2) protein expression. TA laceration followed by suture (LS) produced fibrosis in the SHR (23.3±8.5%) and WR (7.9±1.5%). Lv-mirAT1 treatment decreased At1 gene expression in 50% and reduced fibrosis to 7% 30 days after. RT-qPCR showed that reduction in At1 expression is due to downregulation of the At1a but not of the At1b. RT-qPCR of myoblasts from SHR transduced with Lv-mirAT1a showed downregulation of the Tgf-b1, Tgf-b2, Smad3, Col1a1, and Col3a1 genes by mirAT1a. In vivo and in vitro studies indicate that hypertension overproduces skeletal muscle fibrosis, and AngII-AT1a signaling is the main pathway of fibrosis in SHR. Moreover, muscle fibrosis can be treated specifically by in loco injection of Lv-mirAT1a without affecting other organs.


Introduction
Skeletal muscles weight about 33-40% of total body weight of non-obese men and women, being responsible for posture and voluntary movements [1]. Muscle injury can cause muscle contracture, atrophy, and fibrotic scar tissue, which can affect physical movements [2,3]. PLOS  Complete muscle recovery after a serious injury is usually affected by the establishment of a fibrotic scar tissue at the site of injury. Sportsmen who demand highly efficient muscle functioning can suffer with even small fibrosis, and recurrent injuries can occur at the same site during physical exercises [4,5]. Fibrosis is part of the final step of muscle recovery and is characterized by intense synthesis of extracellular matrix (ECM) proteins by myofibroblasts, which results in the formation of connective tissue scar [6]. In cases of severe muscle injury, fibrogenesis is exacerbated and results in the formation of a dense scar, besides irregular and malfunctioning tissue [3,7]. Therapy based on anti-inflammatory drugs, ice, and rest have limited efficacy in preventing or treating muscle fibrosis after injury [8,9].
Several studies have shown that hypertension promotes fibrosis, and angiotensin II (AngII) receptor type 1 (AT1) is the key factor in hypertension and fibrosis [10,11]. Reduction of fibrosis by AT1 blockers is an important evidence of the fibrotic activity of AngII-AT1 [12][13][14]. Even in the chronic muscular genetic diseases, such as Marfan syndrome and Duchenne muscular dystrophy, the AT1 antagonist losartan was able to reduce fibrosis, improving skeletal muscle regeneration [13].
TGF-β1 is a pleiotropic factor that orchestrate various cell activities. It promotes the negative proliferation of myeloid, epithelial, and lymphoid tumor cells [15,16], but it also stimulates fibroblast proliferation [17] and ECM synthesis in myofibroblasts [18]. Due to these last activities, TGF-β1 is considered the most important fibrotic factor [19,20]. It is known that AngII-AT1 and TGF-β1 signaling cross-talk activates Smad2/3 and upregulates the fibrogenic factor connective tissue growth factor (CTGF) and collagen I expression [21]. However, the pro-fibrotic AngII-AT1 activity can also occur via TGF-β1 in an independent manner [22,23].
According to the World Health Organization (http://www.who.int), there are about 1 billion hypertensive people in the world. In addition, hypertension is one of the most important risk factors for cardiovascular diseases, which affect heart, brain, peripheral limbs, and other organs [24,25]. Although there are several significant numbers of studies showing the effect of AngII-AT1 signaling in skeletal muscle fibrosis, the knowledge about this signaling in skeletal muscles of hypertensive organisms is not clear [14,[26][27][28].
The spontaneously hypertensive rat (SHR) has been largely used to study physiology and pathophysiology related to human hypertension (over 13,000 Medline references were found using both "SHR" and "hypertension" words) due to the genetic predisposition to hypertension and age dependence to manifest hypertension, as occurs in humans [29]. The spontaneous cardiac fibrosis is a major complication of hypertension in SHR, and the development of fibrous tissue is related to overexpression of TGF-β1, which is stimulated by AngII [30][31][32]. The AT1 blockers or angiotensin converting enzyme (ACE) inhibitors have been successfully used to reduce hypertrophy and cardiac fibrosis in this model [33][34][35]. Although the AT1 blocker losartan showed significant benefits in the treatment of muscle fibrosis, this medication has a systemic action and can cause several adverse effects [36][37][38]. Therefore, the search for a localized therapy for patients with muscle injury and fibrosis is necessary.
Here, we report a study on the relationship between fibrogenesis and AT1 in the skeletal muscle of SHR after a severe injury, using the normotensive Wistar rat (WR) as a control. To establish a deep injury in the skeletal muscle, we chose laceration followed by suture because it is a common practice among orthopedists to treat deep muscle injury, which can occur among sportsmen and patients who suffered of vehicle accidents. In addition, as it was commented by Bedair et al [14], laceration is a highly reproducible muscle injury model, its histopathology is very similar to that observed in strain injury patterns, and any treatment benefits seen in this model would likely bring equal or greater benefit in less severe strain injuries. We hypothesized that the skeletal muscle of SHR should express more ATR and, should thus produce more fibrotic tissue after injury. Down-regulation of the AT1 receptor in the SHR muscle by miRNA should reduce fibrosis and promote better muscle regeneration. This strategy may provide a useful in loco treatment of patients with deep muscle injuries.

Animals and muscle laceration
The SHR were obtained from the animal facility of the Institute of Biomedical Sciences, University of São Paulo (USP), and WR were obtained from the National Institute of Pharmacology, Federal University of São Paulo (UNIFESP). All rats (35 SHR and 15 WR) were 16-weekold males (weight: 300-350 g). All procedures involving animals were performed only after the project protocol received approval by the Research Ethics Committee of UNIFESP (Approval number: 0347/12). The completed ARRIVE (Animal Research: Reporting In Vivo Experiments) Checklist is in S1 File.
For muscle laceration, SHR and WR were initially anesthetized with intraperitoneal (ip) injection of ketamine (40 mg/kg) and xylazine (20 mg/kg). Muscles were surgically lacerated as previously described [39], with some modifications. Briefly, the tibialis anterior (TA) muscle from both legs were exposed and transversely cut with a scissor in the middle of the muscle (Fig 1A). The muscles were sutured (laceration-suture group, LS) or not (laceration group, L) after the injury. In the LS group, a PDSII 5-0 suture (Ethicon; NJ, USA) was placed at the medial edge of the lacerated site. After laceration, the skin was closed with black silk 4-0 suture (Ethicon). Four weeks after laceration, all animals were euthanized to evaluate muscle healing and fibrosis. Five muscles from each group were evaluated by histology and other five muscles were evaluated by real-time quantitative PCR (RT-qPCR) and Western blot.

Primary muscle cell culture from SHR and WR
Primary muscle cells were obtained by enzymatic disaggregation of leg muscle from 1-day-old SHR (CEDEME animal facility, UNIFESP) or WR (INFAR animal facility, UNIFESP) and cultured as previously described previously [40][41][42]. Eight rats were used in this study. Isolated cells were then suspended in Dulbecco's modified Eagle's medium (DMEM; Life Technologies; NY, USA) with 10% fetal bovine serum (FBS; Life Technologies), 10% horse serum (Sigma Chemical Co.; MO, USA), 0.5% chick embryo extract (United States Biological; MA, USA), penicillin/streptomycin (100 U/mL; Life technologies) and 2 mM L-glutamine (Life technologies), and maintained at 37˚C with 5% CO 2 . Cells were then plated in collagen-coated flasks. Two hours after plating, the supernatant was removed from the flask and replated in a fresh collagen-coated flask. Within 2-h incubation, mostly fibroblasts adhered to the collagen coat. The serial replating of non-adherent cells was repeated daily for 4 days to establish a primary myoblast cell culture [40,41]. The rat skeletal muscle cell line L6 (ATCC; VA, USA) was used as a control.
AngII effect in primary muscle cells SHR and WR primary muscle cells were seeded on a 24-well plate (3x10 4 cells/well) and 24 hours later AngII (kindly donated by Dr. Julie Bossuyt, University of California, Davis) was added to a final concentration of 100 nM/well [43,44]. After 24 hours, the total RNA was extracted and genes were quantified following methods described in the Gene expression analysis by Real-time quantitative PCR (RT-qPCR) section.

Lentivector construction, production, and transduction
The mirAT1a (micro RNA to AT1a receptor) and mirCTL (micro RNA control) sequences are the same as those from Fan et al [45]. They were obtained from GenScript (NJ, USA) and were cloned into the pUC57 vector. The mirAT1a and mirCTL sequences were removed from the pUC57 vector using MluI/ClaI and cloned into the pLVTHM vector (Addgene; MA, USA), which was previously treated with the same enzymes. These vectors were named as pLv-mir-AT1a and pLv-mirCTL. All plasmids were amplified using the Plasmid Midi Kit 1 (Qiagen; Hilden, Germany) and sequenced in the 3100 Avant Genetic Analyzer (Life Technologies).
To determine the viral titer, which was expressed as transducing units (TU) per mL, HEK-293T cells were transduced with different concentrations of lentiviral vectors in the presence of Polybrene (8 μg/mL; Sigma). After 3 days, the HEK-293T cells expressing GFP were counted (>10,000 events analyzed) using a FACScan (Becton Dickinson, BD; CA, USA) cytometer and the data was analyzed using the FlowJo (TreeStar Inc.; OR, USA) software. The titer is given by GFP positive cells (%) X total number of cells on the day of transduction / volume of virus.
For transduction of L6 and rat primary muscle cells from SHR (passage 4), these cells were seeded on a 6-well plate at 70% confluence, with DMEM supplemented with 10% FBS. After 24 h, the medium was replaced with a fresh one containing Polybrene (8 μg/mL; Sigma), and Lv-mirAT1a or Lv-mirCTL vectors were added to the media with multiplicity of infection (MOI) to transduce L6 (MOI = 10) or rat primary muscle cells (MOI = 50). After 24 h, all media were replaced with fresh ones.

Gene therapy
The SHR and WR were anesthetized with ip injection of ketamine (40 mg/kg) and xylazine (20 mg/kg), and muscles were lacerated as described above. Soon after laceration, 1.2 x 10 7 TU of Lv-mirAT1a or Lv-mirCTL in DMEM (100 μL) were injected in each TA muscle. Four weeks after the treatment, the muscles were removed for histology, gene, and protein expression analysis.

Histological analysis
After the TA muscles were excised, they were weighted and fixed in formaldehyde (10%; 24 h). Next, the muscles were cut (Fig 1A), dehydrated, and embedded in paraffin. Four-micrometer sections were obtained and stained with hematoxylin-eosin (HE) or Picrosirius Red (PR).
Digital images were acquired using a microscope, and 20 fields per slide were analyzed using the Image Pro Plus (Media Cybernetics, Inc.; MD, USA) software. Fibrotic areas were expressed in percentage of PR stained area per total injured area, constituted by the middle portion of TA where the injury was established.

Gene expression analysis by real-time quantitative PCR (RT-qPCR)
Total RNA from TA muscles, heart, primary culture of muscle cells, or L6 cells were extracted with the Trizol reagent (Life Technologies) and treated with DNAse I (Invitrogen).

Statistical analysis
All statistical analyses were performed using Student t-tests (two-tail comparisons) or one-way analysis of variance (ANOVA) with post hoc Bonferroni's test using Prism 6 software (Graphpad; CA, USA). Differences between conditions were considered significant if P < 0.05.

Quantitative histological analysis of muscle fibrosis after laceration
To study the relationship between hypertension and fibrosis after muscle injury, two muscle injury methods were first compared: Laceration (L) and Laceration followed by Suture (LS) (Fig 1A). Four weeks after the L and LS procedures were performed in SHR and WR, the muscles were isolated for analyses. Macroscopic analysis showed that animals in the L group exhibited a bigger injured area as compared to those in the LS group (Fig 1B). The TA muscles from WR were 18.2% heavier than those from SHR ( Fig 1C). Laceration caused a significant (15%) decrease in the TA muscle weight in the SHR/L group as compared to that in the SHR/CTL group, however, in the WR/L group the TA muscle weight was not significantly different in comparison to the non-lacerated group WR/CTL (Fig 1C). Tissue fibrosis, as assessed by PR staining, showed that fibrosis was clearly established after four weeks in both injury models. The laceration area was characterized by an intense reddish staining (Fig 1D). The cells in the scar were more disorganized in animals of the L group as compared to those in animals of the LS group. The animals in the SHR/LS group showed a significant increase in the fibrotic area as compared to those in the WR/LS group (Fig 1E). However, no difference was observed in the fibrotic area between the LS and L groups, within the WR group.

Expression of AngII receptors in the skeletal and cardiac muscles of hypertensive and normotensive rats
The expression of AT1 and AT2 receptor proteins in the TA and heart muscles from animals of the of the SHR and WR groups without injury was quantified by Western blot analysis (Fig  2A) to correlate increase in fibrosis in animals of the SHR group with expression of AT1 receptors. As compared to animals of the WR group, those in the SHR group showed a much higher AT1 expression in the TA muscle (25-fold) and heart (2.4-fold). Regarding the AT2 receptor, no gene or protein expression was detected in the SHR and WR hearts. In the TA muscle, the AT2 receptor gene expression in the WR group was about 2-fold higher than SHR, however no difference was observed between SHR and WR groups in terms of protein.
We also compared gene expression of AngII receptors in the TA muscle and heart of animals of the SHR and WR groups by RT-qPCR (Fig 2B). In the TA muscle, animals of the WR group showed a significantly higher expression of Agtr1a (1.4-fold) in comparison to those of the SHR group. However, expression of Agtr1b in animals of the SHR group expressed 2.4-fold higher than in those of the WR group. Therefore, the higher expression of AT1 receptor protein by animals of the SHR group should be of the Agtr1b gene expression. On the other side, Agtr2 gene expression in animals of the WR group was higher (5.6-fold).

Fig 2. Expression of AngII receptors in the skeletal and cardiac muscles. (A)
The AT1 and AT2 receptors in the TA and heart muscles from animals of the SHR and WR groups were quantified by Western Blot analysis. Band intensities were quantified by the Image J program and the ATR/β-actin ratios were plotted (n = 3 rats per group). In the heart, AT2 receptor expression was not detected in both rat strains. (B) Expression of the Agtr1a, Agtr1b, and Agtr2 genes in the TA and heart muscles from animals of the SHR and WR groups was quantified by RT-qPCR (n = 5 rats per group). Expression of the Agtr1b and Agtr2 genes was not detected in the heart from both rat strains. * p<0.05. TA: tibialis anterior. SHR: spontaneously hypertensive rat. WR: Wistar rat.
In the heart, no difference was observed in the Agtr1a gene expression between animals of the SHR and WR groups (Fig 2B). The Agtr1b gene was expressed only in animals of the WR group, and the Agtr2 gene was not expressed in any of the rats.

Fibrogenic gene expression profile in SHR and WR myoblasts after treatment with AngII
We investigated the fibrotic effect of AngII in SHR and WR myoblasts by RT-qPCR. AngII upregulated the fibrogenic genes Col1a1, Col1a3, TGFbr1 and Smad3 on SHR cells (Fig 3A), whereas only Tgfbr1 and Smad3 were upregulated on WR cells (Fig 3B). Smad2 was downregulated on both cells in the presence of AngII (Fig 3A and 3B).

Down-regulation of the AT1a receptor reduces fibrosis in the SHR skeletal muscle
Initially, the efficiency of lentivectors to down-regulate the Agtr1a gene was assessed in rat myoblast L6 cells, which express the Agtr1a gene normally [47]. Lv-mirAT1a transduced L6 cells had a reduction in Agtr1a mRNA expression in comparison with non-transduced (91.0%) or transduced (88.5%) Lv-mirCTL (Fig 4A), showing high knockdown efficacy. No statistical difference between non-transduced and transduced with Lv-mirCTL groups was observed.
To assess the effect of Lv-mirAT1a in fibrosis, this lentivector was injected just after laceration and the TA muscles were analyzed four weeks after the treatment. No significant differences in muscle weight were observed among groups (Fig 4B). In histology, the injured rats or the treated with Lv-mirCTL showed an intense PR staining in a large area, whereas the group treated with Lv-mirAT1a showed a less stained area and a pattern similar to that of noninjured muscle (Fig 4C and 4D). Fibrosis was reduced after treatment with Lv-miAT1a, in comparison with the groups injured (16%) or treated (18%) with Lv-mirCTL (Fig 4D). In addition, muscle architecture of Lv-mirAT1a rats was more similar to that of non-injured rats, which was very different from the irregular muscle architecture as observed in rats injured or treated with Lv-mirCTL (Fig 4C).
No difference was observed among groups in AT1 receptor protein expression (Fig 4E). However, a significant reduction (53%) was observed in Agtr1a mRNA expression in the group treated with Lv-mirAT1a in comparison with that one of the non-injured group ( Fig  4F). No difference was observed in the Agtr1b and Agtr2 mRNA expressions between groups (Fig 4G and 4H).

Gene expression analysis of myoblast primary culture after treatment with Lv-mirAT1a
Finally, to investigate the genes regulated by the AT1a receptor, we established a primary culture of myoblasts from SHR neonates and the expression of myogenic markers (Myod, Des, and Myf5) was evaluated by RT-qPCR in different passages, using rat myoblast cell line L6 as a positive control. These cells were positive for the myogenic genes until passage 6 ( Fig 5A).
These cells were then transduced with Lv-mirAT1a, and the gene expression profile was analyzed 48 h later (Fig 5B). The Agtr1a expression was reduced (50%) after transduction with Lv-mirAT1a as compared to the animals of the non-transduced group. In contrast, Agtr1b expression was increased after transduction with Lv-mirAT1a. Regarding Agtr2, no significant difference was observed between groups.
The Tgfb1, Tgfbr2, Col1a1, Col3a1, and Smad3 genes related to fibrosis were down-regulated after transduction with Lv-mirAT1a (Fig 5B). The Tgfbr1 and Smad2 genes were also down-regulated, but the variations were not statistically significant. The Ctgf gene was the only one up-regulated after transduction with Lv-mirAT1a.

Discussion
Angiotensin II plays a central role in blood pressure control via AT1, also binding to AT2 with similar affinity. However, several studies suggest that AT2 can antagonize AT1-mediated activities, in addition to having activities not related to AT1 signaling [48]. In rats, there are two AT1, AT1a, and AT1b subtypes, which have similar affinity for AngII. However, studies with AT1a and AT1b KO mice show that the block of AT1a gene affects much more blood pressure than of AT1b, showing the importance of AT1a in the blood control than AT1b [49,50]. These genes are expressed differently in tissues [51]. Since the ATR distribution in rat tissues is variable, we first investigated the expression pattern of angiotensin receptors in the skeletal muscle, using the heart as an experimental control due to the availability of data on angiotensin receptor gene expression in this tissue [52][53][54].
Our gene expression data show higher AT1 expression in the heart and the skeletal muscle of SHR as compared to WR (Fig 2) and absence of AT2 expression in the heart of both adult rats, which explains in parts the spontaneous hypertension in SHR and AT1 specificity in the heart. The AT2 receptor is highly expressed during embryogenesis, but such expression drops down after birth, and its expression is maintained after birth only in some tissues [55]. In addition, its expression can vary with age and pathophysiological conditions, thus leaving the literature data apparently inconsistent. For example: AT2 expression was seen in the heart of rat   fetus by immunohistology, and its expression still could be detected 4 weeks after birth [56]; in addition, very low level of AT2 expression was detected by RT-PCR in the aorta of SHR, which increased 2-fold after long-term treatment with losartan [57]. Therefore, the absence of AT2 expression in the heart of 16-week old SHR and WR, as seen by two very sensitive methods (RT-qPCR and western blot), is not contradictory to other data but reinforce the idea that AT2 expression decreases with age in the heart of healthy rats.
Another significant difference between the heart and TA muscles of SHR is that AT1a is the unique form found in the heart, whereas presence of AT1b is about 3-fold higher than AT1a in the TA. As was commented above, AT1a is the main regulator of arterial pressure and AT1b contributes minimally [49,50], which is why the application of a miRNA specific to AT1a significantly decreased fibrosis, even though the expression of AT1b is much higher than AT1a in the TA muscle. Fibrosis reduction after AT1a knock-down is evident (Fig 4D and 4F), and the total amount of AT1 protein found 30 days after treatment with Lv-mirAT1a had the lowest level in comparison to the control groups (Fig 4E), although it was not significant statistically. As AT1 antibody does not distinguish between two AT1 subtypes, the AT1 band shown in the Fig 4E is the sum of AT1a and AT1b protein concentration. This fact should be the cause of a less impressive decrease of AT1 protein expression than its gene expression.
The larger extension of fibrosis as observed in SHR after laceration in comparison with WR (Fig 1D and 1E) and fibrosis reduction after treatment with Lv-mirAT1a (Fig 4D and 4F) were the first evidence of correlation between fibrosis and hypertension in the skeletal muscle, indicating that AT1-mediated hypertension should be the main cause of fibrosis in the skeletal muscle. Furthermore, the more expressive upregulation of fibrogenic genes in SHR myoblast than WR myoblast after treatment with AngII (Fig 3A and 3B) corroborates this assumption.
There are several evidences showing that interfering TGFβR with signaling by AT1 signaling is one of the mechanisms that organisms control fibrosis [12][13][14]. In order to verify the influence of AT1 on the expression of fibrotic genes, we used myoblasts isolated from SHR, once these cells are the main source to skeletal muscle regeneration after injury [42]. Our in vitro study show a clear influence of AT1 on the expression of Tgfb1, Tgfbr2, and Smad3, and fibrotic genes controlled by Col1a1 and Col3a1 genes (Fig 5), except for the profibrotic Ctgf, which was upregulated after mirAT1a treatment. CTGF is a pleiotropic factor that under normal circumstances is involved in angiogenesis and cellular differentiation, but in pathological conditions CTGF modulates, among others, wound healing and fibrosis [58][59][60]. Although CTGF regulates such large biological activities, its gene expression is modulated mainly by TGFß1 [61][62][63] but AngII can also induce TGFß1 expression independently [64]. Thus, although AT1 signaling and TGF-ß1 receptor signaling have cross-talk, the knockdown of AT1 with mirAT1a does not necessarily interrupt TGF-ß1 signaling and Ctgf expression. Therefore, it seems that skeletal muscle fibrosis in hypertensive organisms follows the same mechanism ones of the normotensive organisms, but more markedly.
Losartan, an angiotensin receptor antagonist, has been shown to be a good inhibitor of skeletal muscle fibrosis [14,65] and cardiac muscle fibrosis after injury [12,34]. Even losartan is a FDA-approved drug and is widely used to treat hypertension, its oral use to treat skeletal muscle fibrosis after injury can bring unnecessary undesired side effects, such as upper respiratory infections, dizziness, hyperkalemia, and other symptoms [36][37][38]. Knockdown of AT1 with Lv-miRNA or a similar vector system can avoid these side-effects since the vector is delivered locally. To maintain the injected vectors in the muscle, avoiding vector leaking and spreading, these lentivectors can be prepared with alginate hydrogel and injected directly into the muscle. Just one injection can last more than two months of transgene expression in the muscle [66], a sufficient time to heal the injured muscles, without targeting other tissues [39].
Collectively, our in vivo and in vitro studies indicate that hypertension promotes more skeletal muscle fibrosis, and AngII-AT1a signaling is the main pathway of fibrosis in SHR, and muscle fibrosis can be specifically treated with in loco injection of Lv-mirAT1a without affecting other organs. Our studies also indicate that the skeletal muscle injury in hypertensive people is prone to have more muscle fibrosis than in normotensive ones, thus contributing to the development of muscle specific gene therapy vectors to improve muscle healing after injury with less fibrosis.
Supporting information S1