|Home | About | Journals | Submit | Contact Us | Français|
There is presently no proven pharmacological therapy for the acute respiratory distress syndrome. Recently, we and others discovered that the heptapeptide angiotensin‐(1–7) [Ang‐(1–7)] shows significant beneficial effects in preclinical models of acute lung injury (ALI). Here, we aimed to identify the best time window for Ang‐(1–7) administration to protect rats from oleic acid (OA) induced ALI.
The effects of i.v. infused Ang‐(1–7) were examined over four different time windows before or after induction of ALI in male Sprague–Dawley rats. Haemodynamic effects were continuously monitored, and loss of barrier function, inflammation and lung peptidase activities were measured as experimental endpoints.
Ang‐(1–7) infusion provided the best protection against experimental ALI when administered by continuous infusion starting immediately after 30 min OA infusion till the end of the experiment (30–240 min). Both pretreatment (−60 to 0 min before OA) and short‐term therapy (30–90 min) also had beneficial effects although less pronounced than the effects achieved with the optimal therapy window. Starting infusion of Ang‐(1–7) 60 min after the end of OA treatment (90–240 min) did not protect barrier function or haemodynamics but still reduced myeloperoxidase activity and increased ACE2/ACE activity ratio respectively.
Our findings indicate that early initiation of therapy after ALI and continuous drug delivery are most beneficial for optimal therapeutic efficiency of Ang‐(1–7) treatment in experimental ALI and, presumably accordingly, in clinical acute respiratory distress syndrome.
|Enzymes a||GPCRs b|
|ACE, angiotensin converting enzyme||AT1 receptor|
|ACE2, angiotensin converting enzyme 2||Mas receptor|
These Tables list key protein targets and ligands in this article which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Pawson et al., 2014) and are permanently archived in the Concise Guide to PHARMACOLOGY 2015/16 (a,bAlexander et al., 2015a, 2015b).
The acute respiratory distress syndrome (ARDS) is a critical illness that is characterized by acute hypoxic respiratory failure, bilateral pulmonary infiltrates on chest radiography, and pulmonary oedema that is not due to a cardiac cause (Ranieri et al., 2012). With an age‐adjusted incidence of 86.2 per 100 000 person‐years, approximately 75 000 persons die of ARDS in the USA each year, a figure that is comparable with the number of adult deaths attributed to breast cancer (Rubenfeld et al., 2015). Despite intense research efforts and a multitude of clinical trials over the past decade, the mortality for ARDS remains unacceptably high at approximately 30–40% (Johnson and Matthay, 2010; Rubenfeld et al., 2015), and no pharmacological therapy has so far proved effective to improve patient outcome (Boyle et al., 2013; McAuley et al., 2014). Moreover, ARDS survivors experience long‐term sequelae, which in turn reduce quality of life and increase societal costs (Herridge et al., 2011).
The crucial feature of ARDS is disruption of both the epithelial and endothelial barriers in the lung, leading to the formation of a protein‐rich alveolar oedema fluid, as well as a massive immune reaction characterized by pulmonary cytokine secretion, and the infiltration of large numbers of leukocytes (Matthay et al., 2012). ARDS can arise as a result of a host of insults, including those originating in the lung (direct ARDS, e.g. as a result of pneumonia, aspirated gastric contents or inhaled toxic gases), or the systemic inflammation (indirect or extrapulmonary ARDS, as caused, e.g. by sepsis or severe trauma). Acute respiratory failure typically necessitates mechanical ventilation as a life‐saving supportive strategy to ensure sufficient oxygen delivery; however, mechanical ventilation characteristically bears the risk of adding ventilator‐induced lung injury (VILI) as an addition to the pre‐existing disease (Slutsky and Ranieri, 2013). Because the term ARDS is exclusively used to refer to the condition in humans, the pathophysiological correlate in experimental models and animal studies is typically referred to as acute lung injury (ALI).
The renin–angiotensin system (RAS) is a humoral system that is essentially involved in the regulation of blood pressure, electrolyte homeostasis, and water and sodium intake. Angiotensin II (AngII), the main biologically active peptide of the RAS, has been proposed to contribute to the pathogenesis of ARDS/ALI through stimulation of its AT1 receptor subtype. This attribution is based on the observation that experimental ALI in mice can be attenuated by treatment with a recombinant form of the human angiotensin converting enzyme 2 (ACE2), which converts AngII into its heptapeptide metabolite angiotensin‐(1–7) [Ang‐(1–7)], thereby presumably reducing the activation of AT1 receptors by AngII (Imai et al., 2005). Importantly, however, the AngII metabolite Ang‐(1–7) itself possesses potent biological activity and has been shown to counteract many of the detrimental actions of AngII (Santos, 2014).
We previously identified Ang‐(1–7) as an important signalling molecule that binds to the GPCR known as Mas (Santos et al., 2003). Furthermore, we recently reported proof of principle for a highly effective therapeutic effect of Ang‐(1–7) in different preclinical models of ALI/ARDS including the model of oleic acid (OA)‐induced ALI in rats, and models of overventilation‐induced and acid instillation‐induced lung injury in mice (Klein et al., 2013). Meanwhile, these beneficial effects have been successfully reproduced by others, demonstrating the validity and reproducibility of this therapeutic effect in different models and laboratories respectively (Li et al., 2015; Zambelli et al., 2015). As the heptapeptide Ang‐(1–7) has already been approved by the Food and Drug Administration for use in a series of clinical trials (Balingit et al., 2012; Pham et al., 2013), testing of this peptide as a therapeutic strategy for the treatment of ARDS is becoming a realistic and promising scenario. However, prior to the planning and initiation of a clinical trial, important details of the therapy including the optimal time window of drug delivery, the best dosing, the optimal formulation and delivery route and the risk of potential adverse effects specific to ARDS patients need to be addressed in appropriate preclinical disease models.
In the present study, we have addressed the first of these crucial questions, namely, the optimal time window for the treatment with Ang‐(1–7). Time windows for the treatment with Ang‐(1–7) were selected based on our previous finding that continuous infusion of Ang‐(1–7) starting at the time of lung injury until the end of the experiment conferred significant protection. Specifically, we aimed to answer the following three questions: (i) Can pretreatment with the heptapeptide initiate similar protective effects? (ii) Does the therapeutic benefit require treatment until the end of the experiment or is Ang‐(1–7) equally beneficial when given only for a short duration during the initial development of ALI? and (iii) Can Ang‐(1–7) still positively affect pathological parameters of ALI when given only after substantial injury has already developed?
All animal care and experimental procedures complied with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85‐23, revised 1996) and were approved by the local government authorities. Studies involving animals are reported in accordance with the ARRIVE guidelines for reporting experiments involving animals (Kilkenny et al., 2010; McGrath and Lilley, 2015).
Male Sprague–Dawley rats (strain code 400) with an age ranging between 10 and 12 weeks (330–360 g ) were purchased from Charles River (Sulzfeld, Germany, or St. Constant, QC, Canada, respectively). All animals were housed in enriched standard cages and were maintained at 21°C under 12 h light–dark cycles. The animals were fed standard diet and water ad libitum. A minimum time of 48 h was permitted for acclimatization in the vivarium before animals entered the experiment. A total of 98 animals were used in the experiments.
Experiments were performed as a prospective, randomized and controlled study. Randomization was performed by lottery, on the morning of each experiment, and two experiments from the same group were never performed consecutively. Blinding of the researcher conducting the actual animal experiment was not possible, because different TWs for the drug application were used, necessitating individual start points for Ang‐(1–7) administration. Post hoc analyses, however, including biochemical assays and histological assessment were performed in a blinded fashion. For this study, rats have been selected because the rat model of OA‐induced ALI is well established. The study has no implication for replacement, refinement or reduction.
Rats were randomly (see above) assigned to the following experimental groups (Figure 1): the first control group (C) received no pharmacological interventions, while the second control group received a continuous infusion of Ang‐(1–7) (50 pmol·kg−1·min−1; Bachem GmbH, Weil am Rhein, Germany) over 210 min [C + Ang‐(1–7)]. In all other groups, ALI was induced by i.v. injection of OA [0.2 mg·kg−1 body weight (bw); Sigma, Munich, Germany] over 30 min. One ALI group received no treatment, while four ALI groups received infusion of Ang‐(1–7) over different therapeutic time windows (TW1–TW4). All TWs refer to time point 0 min as the start of ALI induction and time point 30 min as the end of the OA infusion. In treatment group TW1, infusion of Ang‐(1–7) was initiated 60 min before ALI induction and stopped at time point 0, when ALI induction was initiated (TW1: −60 to 0 min). In treatment group TW2, Ang‐(1–7) administration started immediately after ALI induction and continued till the end of the experiment (TW2: 30–240 min). The Ang‐(1–7) administration in the TW3 group started also immediately after ALI induction, but Ang‐(1–7) administration was stopped after 60 min (TW3: 30–90 min). In TW4, Ang‐(1–7) administration started 60 min after ALI induction and continued until the end of the experiment (TW4: 90–240 min).
Group size was initially projected for n = 11 in each experimental group, but several animals had to be excluded retrospectively from the analyses due to a compromised haemodynamic status at baseline, defined as a mean arterial pressure of <60 mmHg prior to infusion of OA or Ang‐(1–7) respectively. Accordingly, animal numbers are n = 9 for the first control group (C), n = 8 for the second control group receiving only Ang‐(1–7) [C + Ang‐(1–7)], n = 10 for the OA damage group without treatment, n = 9 for the TW1 group, n = 11 for TW2 and TW3 and n = 10 for TW4. For histological assessment, a separate set of experiments was performed with n = 3 animals per group.
For anesthesia, a combination of fentanyl (0.05 mg·kg−1 bw; Fentanyl®‐Janssen 0.05 mg·mL−1, Neuss, Germany), medetomidine (0.5 mg·kg−1 bw, Dormitor® 1 mg·mL−1; Graeub AG, Basel, Switzerland) and midazolam (5 mg·kg−1 bw; Dormicum® 5 mg·mL−1, Roche, Basel, Switzerland) was applied. For i.v. infusion, Ang‐(1–7) was dissolved in 0.9% NaCl. Animals in the untreated OA group received an equivalent volume of physiological saline. For determination of ACE activity in lung tissue homogenates, 5.7 mmol·L−1 Hip‐His‐Leu as substrate (Sigma‐Aldrich, Steinheim, Germany) was dissolved and stored in aliquots at −20°C until use; o‐phthalaldehyde (SERVA Electrophoresis GmbH, Heidelberg, Germany) was freshly dissolved in methanol (20 mg·mL−1). For ACE2 activity measurements, Mca‐APK (Dnp) as fluorogenic substrate (Biosynthan GmbH, Berlin, Germany) was dissolved in DMSO (50 μmol·L−1, final concentration).
Rats were anesthetized by an initial intramuscular injection of fentanyl, medetomidine and midazolam as described previously (Bueltmann et al., 2009), and anesthesia was maintained by continuous i.v. infusion of 50% of the initial induction dose per hour. Rats were placed in supine position and maintained at 38 ± 0.3°C body temperature by a heating lamp coupled to a rectal thermo probe. Following tracheotomy and tracheal cannulation, rats were ventilated with room air in pressure‐controlled mode (Advanced Animal Respirator; TSE Systems GmbH, Bad Homburg, Germany) with peak inspiratory and positive end‐expiratory pressures of 14 and 3 cmH2O, respectively, at 65 breaths per minute. Saline‐filled polyethylene catheters (0.58 mm inner diameter; Smiths Medical, Dublin, OH, USA) were placed in the left carotid artery and right internal jugular vein for continuous monitoring of arterial blood pressure (AP) and for fluid replacement and i.v. Ang‐(1–7) delivery respectively. For measurement of pulmonary arterial and left atrial pressure, a median thoracotomy was performed, and catheters were surgically introduced into the left atrium via the left auricle and into the pulmonary artery via the right ventricle respectively. A 2.5 mm ultrasonic flow probe (Transonic; Transonic Systems, Ithaca, NY, USA) was placed around the ascending aorta distal to the branching of the coronary arteries for continuous recording of aortic flow and subsequent calculation of pulmonary vascular resistance (PVR) as arteriovenous pressure difference over flow. Following surgical preparation and stabilization, haemodynamic data were recorded (DasyLab® 32, DasyLab, Moenchengladbach, Germany) at 30, 60, 120, 180 and 240 min after ALI induction. Animals in the control groups [C and C + Ang‐(1–7)] underwent the same surgical instrumentation and were infused in the same manner as the injury and time window groups respectively. After 4 h, rats were killed by exsanguination, the right main bronchus was ligated and the right lung was excised and processed for determination of wet‐to‐dry lung weight ratio and myeloperoxidase (MPO) activity. For determination of the wet‐to‐dry lung weight ratio, the wet lung was weighed and then dried in an oven at 70°C overnight until its weight had stabilized (dry weight), and the ratio of wet‐over‐dry lung weight was calculated as an index of lung oedema formation. The left lung was lavaged four times with 2.5 mL saline of which >90% was recovered as BALF. For histological analysis, lungs from a separate set of experiments were fixed in 4% paraformaldehyde and paraffin embedded, 5 μm sections were cut and stained with haematoxylin and eosin.
Lung oedema and protein extravasation were quantified as wet‐to‐dry lung weight ratio and total protein content in BALF respectively (Bueltmann et al., 2009). MPO activity in lung homogenates as a measure of neutrophil accumulation was determined by a photometric assay and expressed as units per gram lung tissue (U·g−1) (Kuebler et al., 1996). ACE (Faber et al., 2006) and ACE2 (Gembardt et al., 2005) activities in lung tissue were fluorimetrically measured. For this assay, the lung tissue homogenates were prepared at 4°C in 50 mmol·L−1 Tris buffer, pH 7.4, filtered through nylon gauze and stored at −80°C. Protein concentration was determined using the BCA protein assay (Thermo Fisher Scientific, Bonn, Germany). ACE activity was fluorimetrically measured using Hip‐His‐Leu as substrate and His‐Leu as standard reagent. Fluorescence arising from His‐Leu after reaction with o‐phthalaldehyde was measured at 365 (excitation) and 500 nm (emission) and ACE activity expressed as nmol His‐Leu·min−1·mg·protein−1. Fluorimetric measurement of ACE2 activity was performed with Mca‐APK (Dnp). ACE2 activity is expressed as nmol Mca‐AP·min−1·mg·protein−1. Concentration of TNF‐α in BALF was quantified by a commercially available elisa kit (Hölzel Diagnostika GmbH, Cologne, Germany).
These studies comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2015). All data are presented as mean ± SEM. Two‐way repeated measures (RM) ANOVA was used to test for differences between arterial pressure in different groups over the entire experimental protocol. All other data were tested for differences between groups by one‐way ANOVA test with Dunnett's post hoc testing (SigmaStat 3.10; Systat Software, San Jose, CA, USA). Data from assays of TNF‐α in bronchoalveolar lavage fluid (BALF) were log10 transformed due to heteroscedasticity. Statistical significance was assumed at P < 0.05.
Pulmonary and haemodynamic effects of Ang‐(1–7) were investigated in a rat model of OA‐induced ALI. Four hours after the start of OA infusion, rats had developed considerable lung oedema as indicated by a marked increase in the wet‐to‐dry weight ratio in OA as compared with control rats. However, due to a failed equality of variance test, no post hoc tests were run on this data set. Graphic depiction of the data (Figure 2A) indicates that lung oedema was slightly diminished by Ang‐(1–7) pretreatment (TW1) but was fully reversed when treatment with Ang‐(1–7) was initiated immediately after 30 min ALI induction and persisted over the course of the 4 h experiment (TW2; 30–240 min) and partly reversed when treatment was stopped after 60 min (TW3; 30–90 min). A later start of the treatment (60 min after the end of 30‐min OA infusion), although running until the end of the experiment (TW4; 90–240 min), had no beneficial effect on the wet‐to‐dry weight ratio. Consistent with the wet‐to‐dry lung weight ratio measurements, the ALI‐induced increase in BALF protein content could be fully reversed when treatment with Ang‐(1–7) was initiated immediately after ALI induction and run until the end of the experiment, and partly reversed when Ang‐(1–7) treatment was stopped after 60 min (Figure 2B), while treatment starting 60 min after ALI induction had no detectable effect. In contrast to lung oedema, the OA‐induced increase in BALF protein was, however, blunted by pretreatment with Ang‐(1–7). Infusion of Ang‐(1–7) alone in the absence of OA [control + Ang‐(1–7)] had no detectable effect on lung vascular barrier function.
As a measure of the characteristic inflammatory response in ALI, we determined the MPO activity as a surrogate marker of neutrophil invasion into the lung. OA‐induced ALI caused a significant increase in lung MPO activity, indicative of a marked infiltration of neutrophils (Figure 3A). Administration of Ang‐(1–7) over any of the four tested TWs reversed this effect, as the OA‐induced increase in MPO activity was completely blocked. In line with the early inflammatory phase of ALI, the concentration of the pro‐inflammatory cytokine TNF‐α increased markedly, by approximately 20‐fold, in the BALF in response to OA. There was a clear trend towards a reduction in BALF TNF‐α by any of the tested protocols for Ang‐(1–7) treatment, but only the early initiated and prolonged treatment regimen (TW2) resulted in a significant reduction of TNF‐α in BALF (Figure 3B). In uninjured control rats, infusion of Ang‐(1–7) had no detectable effect on MPO activity or TNF‐α levels. Lung injury was also evident in representative haematoxylin and eosin‐stained histological sections of OA‐treated rats, demonstrating increased alveolar septal thickening and cellular infiltration relative to control lungs (Figure 4). Infusion of Ang‐(1–7) attenuated these pathological findings in all studied time windows.
A summary of haemodynamic parameters for the different experimental groups as assessed at the end of the 4 h experimental protocol is provided in Table 1. OA infusion caused an increase in PVR that was prevented either by pretreatment with Ang‐(1–7) or by Ang‐(1–7) infusion starting immediately after ALI induction until the end of the experiment (Figure 5A). Short‐term or later infusion of Ang‐(1–7) (TW3 and TW4) also showed a trend towards a reduction in PVR, but without reaching significance, compared with the OA group.
In the control group, AP was maintained at approximately 90 mmHg over the time course of the experiment. Infusion of Ang‐(1–7) into control rats caused a slight decrease in AP by approximately 10 mmHg, which, however, was not significant, relative to the untreated control group. In contrast, AP was 20–30% lower, compared with control in all lung injury groups immediately after the end of the OA infusion. In addition, AP changed over time as a function of the different Ang‐(1–7) treatment windows (Figure 5B). In the OA group, AP showed a total decrease of 40% over the 3.5 h interval following OA infusion. This decrease was primarily attributable to a reduced cardiac output as indicated by a significantly lower aortic flow in the OA as compared with the control group, while systemic vascular resistance remained largely unchanged (Table 1). Pretreatment with Ang‐(1–7) showed no significant alleviation of this decline in AP with a total decrease of 27%. Ang‐(1–7) administration immediately after OA infusion increased the AP in the TW2 and TW3 groups at 90 min. While stopping the Ang‐(1–7) infusion after 60 min in TW3 caused AP to decline over the next hour to a level similar to the pretreatment group, the continuous infusion of Ang‐(1–7) in TW2 stabilized AP over the full experimental period with a total decrease in AP of only 2%. Later start of Ang‐(1–7) infusion, as in TW4, did not significantly alleviate the decline in AP and resulted overall in an AP decrease by 25%.
Finally, we investigated the effects of OA‐induced ALI and treatment with Ang‐(1–7) over different time windows on the activity of the two RAS‐related peptidases, ACE and ACE2. Neither induction of ALI nor infusion of Ang‐(1–7) in healthy control rats [control + Ang‐(1–7)] altered the activities of ACE (Figure 6A) or of ACE2 (Figure 6B) in lung tissue compared with control rats and consequently did not lead to changes in the ratio between ACE2 and ACE activities (Figure 6C). All four tested TWs showed a trend towards a reduction in ACE activity, which, however, only reached significance for TW3 and TW4, relative to control or OA respectively. In contrast, ACE2 activity increased in response to Ang‐(1–7), relative to either the control or the ALI group, but only in groups TW1 and TW2, while short‐term treatment (TW3) and later treatment (TW4) after OA had no effect on ACE2 activity. Accordingly, the greatest increase in the ACE2/ACE ratio was evident in the TW2 group, that is, when Ang‐(1–7) was administered from the time of OA infusion over the entire experimental time. Nevertheless, all other treatment groups also showed a significant increase in the ACE2/ACE ratio.
In this study, we aimed to identify the optimal TW for Ang‐(1–7) treatment in experimental ALI. The unchanged mortality of ARDS over the past decades (Phua et al., 2009) highlights the need for novel therapeutic interventions and their optimal timing in the disease process. Here, we tested four different TWs including pre‐injury and post‐injury treatments with different times of onset after injury and duration in an established model of OA‐induced lung injury and compared outcome parameters to untreated injured animals and uninjured control rats. In this model of ALI, all tested TWs for Ang‐(1–7) showed some form of therapeutic benefit, in that they alleviated OA‐induced increases in permeability, inflammatory markers or haemodynamic alterations. This finding is in line with prior work in which we and others have recently demonstrated the therapeutic potential of Ang‐(1–7) for the treatment of experimental ALI (Klein et al., 2013; Li et al., 2015; Zambelli et al., 2015). Of the four different time windows tested, the best protection in terms of lung vascular barrier function, inflammatory and haemodynamic parameters was provided by a continuous infusion of Ang‐(1–7) starting at the time of injury and continuing until the end of the experimental period (TW2). Pretreatment and short‐term treatment with Ang‐(1–7) also provided partial benefits, which were however largely absent in the later treatment group (TW4).
Our experimental model of ARDS shows the characteristic features of experimental ALI as outlined in the recent consensus statement from the American Thoracic Society (Matute‐Bello et al., 2011), including impaired alveolo‐capillary barrier function (evident as increase in wet‐to‐dry lung weight ratio and BALF protein), a strong inflammatory response (as indicated by an elevated MPO activity in lung tissue and a marked increase in TNF‐α in the BALF) and altered haemodynamics. In a series of experiments, we and others have identified a critical role of the RAS in the pathogenesis of ALI/ARDS. A first indication for a key role of Ang‐(1–7) in ALI was provided by Imai et al. (2005) who demonstrated the protective effects of recombinant ACE2 in murine models of acid‐induced and sepsis‐induced ALI, an effect which the authors linked to reduced AngII concentrations and, consequently, reduced stimulation of AT1. However, metabolism of AngII by ACE2 concomitantly leads to the generation of Ang‐(1–7), which counteracts many of the effects of the AngII‐AT1 receptor axis by signalling through its own receptor, Mas. In previous work, we and others have provided experimental proof‐of‐principle demonstrating not only that exogenous Ang‐(1–7) exerts similar or even more pronounced protective effects as recombinant ACE2 or pharmacological blockade of AT1 receptors respectively (Klein et al., 2013; Li et al., 2015; Zambelli et al., 2015), but also that the protective effect of AT1 receptor blockade is largely attributable to an enhanced signalling via the Ang‐(1–7)/Mas axis (Klein et al., 2013). The cellular mechanisms underlying the protective effects of Ang‐(1–7) in ALI are still incompletely understood but are likely to involve endothelial barrier‐stabilizing effects via cGMP‐dependent and cAMP‐dependent pathways, given that inhibition of NO synthase by L‐NAME inhibited the barrier‐protective effect of Ang‐(1–7) in lung microvascular endothelial cells challenged by thrombin (Klein et al., 2013) and that Ang‐(1–7) increased intracellular cAMP levels (Liu et al., 2012), which can attenuate endothelial leak (Moore et al., 1998) and reduce inflammatory cell infiltration (Derian et al., 1995). The emerging role for a therapeutic potential of Ang‐(1–7) in the treatment of lung disease is further substantiated by similar protective effects in experimental models of lung fibrosis or pulmonary hypertension (Shenoy et al., 2010; Chen et al., 2011; Wösten‐van Asperen et al., 2011). Ang‐(1–7) treatment presents not only a promising but also a realistic strategy given that the drug has been approved by the Food and Drug Administration for clinical trials. Importantly, with respect to its prospective clinical use, Ang‐(1–7) infusion in uninjured control rats did not cause any detectable adverse effects with respect to lung vascular barrier function or inflammation. However, crucial factors such as optimal dosage, administration route and time window all need to be analysed in preclinical studies prior to the initiation of a first‐in‐man trial. In the present study, we sought the optimal time window for Ang‐(1–7) therapy in ARDS by testing four different treatment strategies, namely, a pretreatment (TW1), a continuous treatment (TW2), a short‐term strategy (TW3) and a late‐term (TW4) strategy.
Prophylactic treatment (TW1) showed some protective effects with respect to protein and TNF‐α concentration in BALF, as well as in MPO, ACE and ACE2 activities. However, these protective effects were less pronounced than those seen with a continuous drug infusion (TW2). A prophylactic approach is also of limited clinical relevance, given the poor predictability of ARDS, which – in conjunction with its moderate effectiveness – makes this approach less promising. An immediate but short‐term infusion of Ang‐(1–7) (TW3) showed a similar trend towards protection in terms of permeability, inflammation and haemodynamic stabilization as seen with the continuous infusion in group TW2, but to an overall lesser degree. Hence, continuation of Ang‐(1–7) infusion throughout the acute inflammatory phase of ARDS seems critical for an optimal treatment effect, a finding that is probably of little surprise given the short half‐life of Ang‐(1–7) of only 20–30 min in plasma (Mordwinkin et al., 2012). Of interest, there was a noticeable increase in arterial pressure after the start of Ang‐(1–7) infusion in both the TW2 and TW3 groups. This finding may initially seem surprising as Ang‐(1–7) is known to exert direct vasodilatory effects (Chappel et al., 1998; Peiró et al., 2007; Peiró et al., 2013), which was also evident as a slight decrease in arterial pressure in the control group receiving Ang‐(1–7) infusion. The Ang‐(1–7)‐induced increase in arterial pressure in ALI rats is thus likely to reflect an overall stabilization of cardiovascular haemodynamics due to the anti‐edematous and anti‐inflammatory effects of the heptapeptide as well as additional positive inotropic effects of Ang‐(1–7) (Zhou et al., 2015). However, after stopping Ang‐(1–7) infusion in the TW3 group, AP decreased rapidly again to levels that were not statistically different from the untreated OA group. This finding confirms the notion that cardiovascular stabilization was a direct result of Ang‐(1–7) infusion, and emphasizes the need for continuous drug infusion during the acute inflammatory phase of ALI/ARDS. Finally, later administration of Ang‐(1–7) had no significant positive effects on systemic haemodynamics and could not reverse all signs of ALI at this advanced stage. However, the later administration still prevented the OA‐induced increase in MPO activity and the concomitant decrease in ACE activity. The former finding is in line with the relatively long time, at least 2 h, that PMN take to emigrate from the vasculature into the alveolar space (Reutershan et al., 2005), which leaves a longer time window for Ang‐(1–7) therapy to reduce neutrophil infiltration. In comparison, endothelial injury and barrier failure occur within minutes after OA infusion (Schuster, 1994), which explains why the later Ang‐(1–7) treatment had no detectable beneficial effects on parameters of lung barrier function such as BALF protein concentration or lung wet‐to‐dry weight ratio.
Interestingly, the experimental groups with the best protective effects of Ang‐(1–7) showed also the highest ACE2/ACE activity ratio, which may therefore represent a potential biomarker for monitoring the therapeutic efficiency of ALI/ARDS treatment by Ang‐(1–7). In subsequent studies, it will be of interest to monitor in parallel the ACE2/ACE activity ratio in circulating blood cells and to correlate it with therapeutic effectiveness, as measurements that can be made with whole blood would markedly increase their utility as potential biomarkers. Functionally, this relation indicates that exogenous Ang‐(1–7) shifts the balance from the detrimental ACE/AngII/AT1 receptor pathway towards the protective endogenous ACE2/Ang‐(1–7)/Mas axis in ALI, thereby establishing a positive therapeutic feedback loop.
A potential limitation of our study lies in the fact that we only tested a single concentration of Ang‐(1–7) (50 pmol·kg−1·min−1) in one experimental model of ALI. In previous work, however, we had demonstrated the excellent efficacy of the applied concentration of Ang‐(1–7) in three different experimental models of ALI, namely, acid aspiration, OA and VILI (Klein et al., 2013), and therefore focused in the present study exclusively on the optimal time window for Ang‐(1–7) therapy.
In conclusion, although there is no proven pharmacological therapy for the treatment of ALI/ARDS, at present, we and others have already identified i.v. infusion of Ang‐(1–7) as a promising and highly effective therapeutic strategy in a variety of preclinical models of ALI. Here, we have demonstrated that the best protection is achieved when Ang‐(1–7) infusion is initiated immediately after induction of lung injury and continued throughout the inflammatory/exudative phase of the disease. While immediate administration may not be achievable in a clinical setting, the present findings stress the need for Ang‐(1–7) infusion to be started as soon as possible after ARDS diagnosis, in any future clinical trials. The extent to which Ang‐(1–7) infusion should be continued beyond the initial inflammatory/exudative phase into the fibroproliferative phase of ARDS also deserves further exploration, particularly in view of preclinical data that also attribute therapeutic effects to Ang‐(1–7) in the context of lung fibrosis (Shenoy et al., 2010; Meng et al., 2014; Meng et al., 2015).
All authors performed the research; T.W. and W.M.K. designed the research; all authors analysed the data; F.K., T.W. and W.M.K. wrote the paper.
T.W. and W.M.K. are the inventors of the patent application ‘Use of an Ang‐(1–7) receptor agonist in acute lung injury’, which has been submitted to the European Patent Office and to national patent offices in the USA, Canada, Brazil, Japan, China and Korea.
This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research recommended by funding agencies, publishers and other organizations engaged with supporting research.
We thank Esther‐Pia Jansen for excellent technical assistance and Nadine Klein for training and technical support. The present work was supported by grants of the Deutsche Forschungsgemeinschaft (KU 1218/7 and WA 1441/22‐1&2).
Supé S., Kohse F., Gembardt F., Kuebler W. M., and Walther T. (2016) Therapeutic time window for angiotensin‐(1–7) in acute lung injury. British Journal of Pharmacology, 173: 1618–1628. doi: 10.1111/bph.13462.