Angiotensin-(1–7) treatment blocks lipopolysaccharide-induced organ damage, platelet dysfunction, and IL-6 and nitric oxide production in rats

Sepsis can lead to shock, multiple organ failure, and even death. Platelets play an active role in the pathogenesis of sepsis-induced multiple organ failure. Angiotensin (Ang)-(1–7), a biologically active peptide, counteracts various effects of Ang II and attenuates inflammatory responses, reactive oxygen species production, and apoptosis. We evaluated the effects of Ang-(1–7) on organ injury and platelet dysfunction in rats with endotoxaemia. We treated male Wistar rats with saline or lipopolysaccharide (LPS, 10 mg, intravenously) then Ang-(1–7) (1 mg/kg, intravenous infusion for 3 h beginning 30 min after LPS administration). We analysed several haemodynamic, biochemical, and inflammatory parameters, as well as platelet counts and aggregation. Ang-(1–7) improved hypotension and organ dysfunction, and attenuated plasma interleukin-6, chemokines and nitric oxide production in rats after LPS administration. The LPS-induced reduction in platelet aggregation, but not the decreased platelet count, was restored after Ang-(1–7) treatment. The protein expression of iNOS and IκB, but not phosphorylated ERK1/2 and p38, was diminished in Ang-(1–7)-treated LPS rats. The histological changes in liver and lung were significantly attenuated in Ang-(1–7)-treated LPS rats. Our results suggest that Ang-(1–7) ameliorates endotoxaemic-induced organ injury and platelet dysfunction, likely through the inhibition of the inflammatory response and nitric oxide production.


Results
We did not observe significant differences in the haemodynamic or biochemical parameters, platelet counts, or inflammatory markers among the groups at baseline, or between the control and Ang-(1-7) groups during the experimental period (Figs. 1, 2, 3, 4).

Ang-(1-7) attenuated endotoxin-induced delayed hypotension in rats.
Systemic LPS administration resulted in a biphasic fall in mean arterial pressure. The initial fall in arterial pressure appeared at 1 h, followed by a partial recovery, and then a further progressive fall from 4 to 6 h after LPS administration (Fig. 1a). Meanwhile, endotoxin caused a significant increase in heart rate from 2 to 6 h after administration (p < 0.001 vs. control group; Fig. 1b). Treatment with Ang-(1-7) significantly improved LPS-induced late hypotension (p = 0.009 at 4 h and p = 0.004 at 6 h vs. LPS group; Fig. 1a), but not tachycardia.
After LPS administration, the levels of blood glucose initially increased at 1 h (p < 0.001) and subsequently declined below the baseline values (at 6 h: p < 0.001 vs. control group; Fig. 2e). Ang-(1-7) treatment attenuated LPS-induced late hypoglycaemia at 4 and 6 h after administration (at 4 h: p = 0.001 and at 6 h: p < 0.001 vs. LPS group; Fig. 2e), but not the initial hyperglycaemia.

Ang-(1-7) attenuated LPS-induced increases in plasma IL-6 and NO. LPS significantly raised
plasma IL-6 levels at 6 h post-administration (p < 0.001 vs. control group; Fig. 4a). This change was attenuated by Ang-(1-7) treatment (p = 0.001 vs. LPS group). In addition, LPS increased plasma NO levels with time (p < 0.001 vs. control group). The plasma NO increases at 4 and 6 h after LPS administration were also attenuated by Ang-(1-7) (at 4 h: p = 0.003 and at 6 h: p = 0.001 vs. LPS group; Fig. 4b). In lung and liver tissues, LPS also caused elevated NO levels in endotoxaemic rats (lung: p = 0.016 and liver: p = 0.016 vs. control group; Fig. 5a,b). Treatment with Ang-(1-7) attenuated the NO levels in tissues, and there was no significant difference of NO level between LPS + Ang-(1-7) and control group (Fig. 5a,b). As biomarkers of oxidative stress, the 4-hydroxynonenal (4-HNE) and protein carbonyl levels in lung and liver tissues were no significant differences among all groups ( Fig. 5c-f).

Histological examination/immunohistochemistry analysis. In histologic examination, LPS
induced extensive alveolar interstitial fibrosis in lungs ( Fig. 8a) and portal fibrosis with inflammation in liver tissues (Fig. 8b). However, treatment with Ang-(1-7) attenuated these histopathological changes. Immunohistochemistry (IHC) analysis was carried out to detect phosphorylated NF-κB expression in lung and liver tissue. We found that phosphorylated NF-κB was not expressed in lung ( Fig. 9a) and liver ( Fig. 9b) tissues in the control group. Moreover, the areas of phosphorylated NF-κB expression were increased in lung and liver tissues after LPS, and treatment with Ang-(1-7) in LPS rats decreased these areas ( Fig. 9).

Discussion
We found that Ang-(1-7) administration inhibited endotoxin-induced hypotension, hypoglycaemia, thrombocytopathy, and multiple organ injury in rats. Along with the findings that Ang-(1-7) altered inflammatory and signalling molecule expression, these results suggest that Ang-(1-7) ameliorates LPS-induced inflammation and platelet dysfunction, thereby protecting against organ injury. This hypothesis and the results of the study are summarised in Fig. 10. The underlying mechanism through which Ang-(1-7) mediates its protective effects against sepsis remains unresolved. In the classical pathway, NF-κB proteins are bound and inhibited by IκB proteins. Upon stimulation by LPS, IκB is phosphorylated and degraded, and thus allowing NF-kB activation. Active NF-κB complexes are further activated by phosphorylation, and translocate to the nucleus, where it regulates the target gene expression. The ACE2/Ang-(1-7)/Mas axis protects pancreatic acinar cells from inflammatory damage by inhibiting the p38 MAPK/NF-κB pathway 28 , but previous studies showed that Ang-(1-7) counteracts LPS-induced skeletal muscle atrophy 29 and hepatocyte injury 30 partially by signalling through the p38 MAPK pathway. Furthermore, Ang-(1-7) inhibits ERK1/2 and NF-κB activation to protect against LPS-induced acute lung injury in rats 31 . However, we found that Ang-(1-7) increased IκB protein and suppressed phosphorylated NF-κB expression, but not activation of the ERK1/2 and p38 pathway in livers and lungs of endotoxaemic rats. Our findings suggest that the anti-inflammatory mechanism of Ang-(1-7) may differ by cell type and in physiological versus pathological conditions 32 . www.nature.com/scientificreports/ Among three isoforms of NOS, nNOS and eNOS produce a nanomolar amount of NO for seconds to minutes, while iNOS generate large amounts of NO (micromolar range) by stimulating with proinflammatory cytokines and LPS, sometimes for hours [33][34][35] . Indeed, we found LPS administration increased the plasma NO levels and significantly increased iNOS expression in liver tissues, which were attenuated by Ang-(1-7). NF-κB signalling modulates the transcription of inflammatory genes, including pro-inflammatory cytokines, chemokines and iNOS 36,37 . Thus, Ang-(1-7) may act through the IκB/NF-κB pathway to attenuate IL-6, chemokines and iNOS production in the setting of sepsis.
Sepsis induces complex regulation of platelet functions, such as alterations in aggregation 38 . Incubation of human platelets with LPS for 1 h inhibited aggregation by suppressing Ca 2+ mobilisation and protein kinase C activation 39 , whereas 6-h incubation of rat platelets with LPS did not induce activation and aggregation 40 . We found that platelet aggregation induced by collagen, but not by ADP, was suppressed in endotoxaemic rats at 6 h  www.nature.com/scientificreports/ post-administration, which is comparable to the findings of human sepsis studies 38,41 . Two mechanisms may account for this finding: (1) only hyporesponsive platelets remained at the end stage of endotoxaemia in rats, or (2) platelets were desensitised to various agonists during the course of sepsis 42,43 . Further studies are necessary to clarify the possible mechanisms that modulate platelet responses to various agonists in sepsis. Activated platelets release thrombin, leading to platelet aggregation and three-dimensional thrombus formation, which can be inhibited by ACE inhibitors and angiotensin type I receptor blockers through Ang-(1-7) production 44 . Furthermore, Ang-(1-7) exerts antithrombotic activity by signalling through the Mas receptor to stimulate NO release from platelets and endothelial cells 22 . Interestingly, platelets do not have the nuclear transcriptional regulatory mechanisms of the iNOS gene, however, iNOS knock-out and iNOS inhibitors modulate collagen-induced platelet aggregation in vitro 45 . In addition, platelets are endowed with the  www.nature.com/scientificreports/ NF-κB/I-κBα complex. NF-κB regulates platelet phospholipase C-β2 46 , and inhibitors of NF-κB modulate platelet activation 47 . Ang-(1-7) administration ameliorated the endotoxin-induced inhibition of platelet aggregation in our study, while dampening iNOS expression and consequent NO overproduction. These findings indicate that Ang-(1-7) affects platelet aggregation in endotoxaemic rats through its global effects rather than through its antithrombotic activity. Although Ang-(1-7) administration improved LPS-induced suppression of collagen-activated platelet aggregation, it did not attenuate LPS-induced thrombocytopaenia. Altered platelet production and scavenging in the circulation during sepsis and increased platelet consumption by DIC have all been attributed to thrombocytopaenia 40 . These complicated mechanisms of thrombocytopaenia action may explain why Ang-(1-7) did not improve endotoxin-induced thrombocytopaenia.
In a preliminary study, we found that 2 mg/kg of Ang-(1-7) resulted in haemodynamic instability and even early death in endotoxaemic rats (data not shown). Thus, we used 1 mg/kg of Ang-(1-7) infused over 3 h, a dosage comparable to our previous study in a caecal ligation/perforation (CLP) model 14 . This dosage did not exert marked effects on haemodynamic or biochemical variables in control rats during the experimental period.
CLP-induced polymicrobial sepsis and endotoxaemia promote microvascular thrombosis through different mechanisms 26 . Kinasewitz et al. reported that endotoxin rather than CLP leads to thrombocytopenia and coagulation derangement, which were comparable to the haemostasis characteristics of human sepsis 25 . Thus, we chose the animal model of endotoxaemia, which induces prolonged thrombocytopaenia, platelet sequestration, and pro-inflammatory cytokine responses, to assess the effects of Ang-(1-7) on platelet function, inflammatory responses, and the related mechanism in sepsis 41 . In addition, we used only male rats in this study to avoid the confounding effects of estrous cycle and sex hormones in female rats. Further studies using female animals are warranted to investigate the role of Ang-(1-7) in sepsis.
In conclusion, Ang-(1-7) treatment inhibited multiple organ injury in endotoxaemic rats by attenuating inflammation, diminishing NO production, and restoring LPS-induced platelet dysfunction. The IκB/NF-κB signalling pathway, rather than the ERK1/2 and p38 pathway, appears to mediate the protective effects of Ang-(1-7) in this model. Our observations suggest that the Ang-(1-7)/Mas axis should be considered a putative target for the development of a new class of drugs to treat sepsis. Additional studies are needed to explore the molecular mechanism of the Ang-(1-7)/Mas pathway in platelets during sepsis, such as studies using Mas −/− platelets in Mas +/+ mice.

Materials and methods
Animals and experimental design. Thirty-six adult male Wistar rats (260 to 280 g; BioLASCO Taiwan, Taipei, Taiwan) were used in this study. The rats were housed in 12-h light/dark conditions with free access to food and tap water. The animal studies complied with the National Institutes of Health guidelines for ethical animal treatment, and were approved by the Committee on the Ethics of Animal Experiments of Taipei Veterans General Hospital (IACUC-2015-164).
We anesthetised the rats by intraperitoneal injection of sodium pentobarbital (50 mg/kg). We placed a polyethylene catheter in the left carotid artery, for arterial pressure monitoring, and in the right jugular vein, for drug administration. The catheters were cannulated, exteriorised, and fixed to the back of the neck. The cannulated animals were allowed to recover to their normal condition overnight with ad libitum access to standardised pellet food and tap water.
After we recorded baseline haemodynamics and collected blood, we randomly divided the rats into 4 groups. The control group received intravenous infusion of normal saline (1 ml/kg) for 15 min from 0 h, the Ang-(1-7) group received intravenous infusion of Ang-(1-7) (1 mg/kg for 3 h) starting 30 min after saline infusion, the LPS group received intravenous infusion of Escherichia coli-derived LPS (10 mg/kg for 15 min) starting from 0 h, and the LPS + Ang-(1-7) group received infusion of Ang-(1-7) (1 mg/kg for 3 h) starting 30 min after the same LPS regimen. The dosage of Ang-(1-7) was based on our previous study 14 .
Bacterial LPS (E. coli serotype 0127:B8; L3127) was purchased from Sigma Chemical (St Louis, MO, USA) and Ang-(1-7) was obtained from Tocris Bioscience (Bristol, UK). The experiments were performed on pairs of www.nature.com/scientificreports/ rats, and the haemodynamic parameters were monitored for 6 h. At the 0, 1, 2, 4 and 6 h-time points, we obtained 0.8 ml of blood from experimental rats for biochemistry, cytokines, and platelet counts studies. Each volume of blood withdrawn was replaced with an equal volume of saline. We assessed signs of distress in the animals hourly during the experimental period. At 6 h after LPS treatment or upon signs of imminent death (i.e. failure to respond to external stimuli, inability to maintain upright position/tremors, prolonged/deep hypothermia, and/ or agonal breathing), rats were sacrificed by overdose of pentobarbital (100 mg/kg, intravenous). We excised the lungs and livers to analyse the levels of iNOS, IκB, and ERK1/2 protein expression.
Measurement of haemodynamic parameters. We monitored haemodynamics in all rats by connecting the arterial catheter to a pressure transducer (P23ID, Statham, Oxnard, CA, USA). At 0, 0.5, 1, 2, 4, and 6 h after LPS treatment, we recorded phasic blood pressure and heart rate on a multichannel recorder (XctionView Data Acquisition System, SINGA Technology Corporation, Taipei, Taiwan).

Quantification of organ function and injury.
We analysed the blood glucose levels in the samples (10 μl) using the OneTouch II blood glucose monitoring system (Lifescan, Milpitas, CA, USA). After immediate centrifugation at 16,000 g for 3 min, some samples were analysed for plasma levels of LDH, ALT, BUN, and creatinine using a Fuji DRICHEM 3030 (Fuji Photo Film, Tokyo, Japan). Each volume of blood withdrawn was replaced with an equal volume of saline. Washed platelet preparation. It was essential to examine platelet aggregation using washed platelets to isolate the cells from the effects of the plasma environment 48 . We collected and centrifuged (100 g for 15 min) whole blood samples to obtain platelet-rich plasma (PRP) at the end of the experiment (6 h post-LPS administration). Platelets were pelleted by further centrifugation of the PRP in the presence of apyrase (0.02 U/ml) and prostaglandin I 2 (1 μg/ml) at 700 g for 10 min. The resulting pellets were resuspended twice in modified Tyrode's buffer (134 mM NaCl, 2.9 mM KCl, 12 mM NaHCO 3 , 0.34 mM NaH 2 PO 4 , 1 mM MgCl 2 , 20 mM HEPES, 0.1% glucose, and 0.35% bovine serum albumin). The washed platelets were adjusted to a concentration of 3 × 10 5 cells/μL.

Platelet aggregometry.
Platelet aggregation was monitored in 96-well plates using a modified light transmission method 49 . Briefly, we placed washed platelets (100 μl aliquots) into individual wells of 96-well plates. We added various concentrations of ADP (0.3-30 μM) and collagen (0.1-10 μg/ml) to the wells, which we shook on a microplate shaker (BioShake IQ, Quantifoil Instruments, Großlöbichau, Germany) at 1200 rpm for 5 min at 37 °C to activate the platelets. We measured the absorbance at 595 nm with a 96-well plate reader (Tecan Sunrise; Tecan, Weymouth, UK). We calculated the percentage of platelet aggregation using the absorbance of modified Tyrode's buffer as a blank.
Measurement of plasma IL-6 concentrations. Rat plasma samples obtained at 0 and 6 h of treatment were used to measure IL-6 levels with an enzyme-linked immunosorbent assay kit (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's instructions.
Tissue homogenization and protein quantification. The lungs and livers obtained after euthanasia were homogenized in lysis buffer and centrifuged for 25 min at 18,000 × g (4 °C). Protein in the supernatant portion was quantitated by BCA Protein Assay Kit (Thermo scientific, Rockford, IL, USA). We used aliquots of tissue homogenates for tissue chemokine levels, NO assay, oxidative stress and Western blot analysis.

Measurement of chemokine levels in tissue.
The CXCL1 and MIP-3α levels in lung and liver tissues were determined using specific ELISA kits from R&D Systems (Abingdon, UK) according to manufacturer's instructions.
Measurement of NO concentrations. We measured total nitrite and nitrate concentrations in plasma and tissue homogenates to determine NO concentration. We incubated plasma (30 μl) and tissue homogenates with 95% ethanol for 30 min, and then centrifuged the samples at 16,000 g for 6 min. The nitrite/nitrate concentrations in all samples were analysed by chemiluminescence as described in previous studies 50,51 . Briefly, to measure the amounts of nitrate, the supernatant portion was added to a reducing agent (0.8% VCL 3 in 1 N HCl) in the purge vessel to convert nitrate to NO·. NO· was stripped from the samples using helium purge gas, and was then drawn into an NO analyser (Sievers NOA-280i; Sievers, Boulder, CO, USA). Nitrate concentrations in plasma and tissues were calculated after comparison with standard solutions of sodium nitrate (Sigma Chemical, St Louis, MO, USA).
Oxidative stress analysis. The 4-HNE adducts and protein carbonyl levels in lung and liver tissues were measured using ELISA kits from Mybiosource (San Diego, CA, USA) according to manufacturer's instructions. www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.