All experimental protocols were reviewed and approved by the Committee for Animal Experimentation of the Medical University of Vienna and the Austrian Ministry of Education, Science and Research (“Role of the renin-angiotensin system in long-term ventilator-induced lung injury: An experimental study in mice,” permit number 2020-0.009.485 on 21 January 2020). All experiments were performed in the research laboratories at the Center for Biomedical Research of the Medical University of Vienna, Vienna, Austria, in compliance with National Institutes of Health guidelines.

Healthy C57BL/6 mice of both sexes, 8–12 weeks of age, with a body weight ranging from 20 to 27 g were randomly subdivided into 7 groups (n = 14), denoted as sham group (unventilated controls), low tidal volume (LVT, 6 mL/kg) and high tidal volume group (HVT, 30 mL/kg), as well as HVT + lisinopril groups ± ALT-00 at dose levels of 2.7, 10 or 100 μg/kg/min. The animals were housed at a room temperature of 22°C ± 2°C, an air humidity of 55% ± 10%, a 12:12 h dark:light regime in Makrolon^® cages on softwood litter. Animals were anesthetized with 3% isoflurane for about 2–3 min, injected intraperitoneally (i.p.) with ketamine (80 mg/kg) and xylazine (6 mg/kg), and placed on a heating pad. A tracheostomy was performed, and the mice were connected to a rodent ventilator (VentElite, Harvard Apparatus, Boston, MA) in volume control mode with the following initial settings: In the LVT group with a VT of 6 mL/kg, positive end-expiratory pressure (PEEP) of 1 cm H₂O and respiratory rate of 135 breaths/min, and in the HVT group with a VT of 30 mL/kg, PEEP of 0 cm H₂O and respiratory rate of 40 breaths/min; the fraction of inspired oxygen FiO₂ was 0.5 and I:E = 1:2 in both groups. A bolus dose of rocuronium 1 mg/kg was given i.p., to prevent dyssynchrony with the ventilator.

A polyethylene catheter was inserted into the right carotid artery for continuous MAP monitoring as well as for blood gas analysis and blood sampling at the end of the experiment. Another polyethylene catheter was inserted into the right jugular vein for intravenous fluid and drug administration. Solution 1 (S1) for fluid therapy contained 80% Ringer’s lactate and 20% glucose 5%, solution 2 (S2) for maintenance of anesthesia contained 0.9% saline with ketamine 5 mg/mL, xylazine 0.3 mg/mL, and rocuronium 0.05 mg/mL, and solution 3 (S3) for experimental therapy contained lisinopril 0.15 μg/kg/min (a dose eliciting significant shifts in systemic RAS in pharmacodynamic pilot experiments) ± ALT-00 at one of three different dose levels (2.7, 10, and 100 μg/kg/min). All of these solutions were infused via a syringe pump (Kent Scientific, Torrington, CT). The S1 infusion rate was set at 30 mL/kg/h for 45 min to permit hemodynamic stabilization and was subsequently reduced to 15 mL/kg/h to maintain hemodynamic stability during ventilation. The S2 infusion rate was set at 10 mL/kg/h throughout the 4 h period in which the mice were exposed to ventilation. The S3 infusion rate was set for lisinopril at 0.15 μg/kg/min, and for ALT-00 at 2.7, 10 or 100 μg/kg/min, respectively.

The tidal volumes and duration of the experiment were chosen based on results from pilot experiments to induce VILI in wild type mice with significant alteration in lung mechanics. During HVT ventilation, with PEEP = 0 and peak inspiratory pressure = 30 cm H₂O, carbon dioxide (CO₂) was added to the inspiratory gas to avoid hypocapnia. The added CO₂ concentration was regulated to achieve an end-tidal CO₂ of 30–40 mmHg (4.0–5.3 kPa) that was monitored with a capnograph type 340 (Hugo-Sachs Elektronik - Harvard Apparatus, March, Germany). This helped to avoid respiratory acidosis or alkalosis, which may otherwise have influenced inflammatory mediators or could have contributed to the pathophysiology of acute lung injury (Laffey and Kavanagh, 2002; Takeshita et al., 2003; Doerr et al., 2005; Mao et al., 2022). During mechanical ventilation, each group was also subjected to additional recruitment maneuvers (35 cm H₂O, 5 s) every 60 min throughout the experiment.

All mice were connected to a Powerlab System to monitor the electrocardiogram, heart rate (HR), MAP, and body temperature throughout the experiment that lasted 5 h in total: One hour for animal preparation (anesthesia, tracheostomy, and insertion of catheters) and 4 h of ventilation.

Baseline hemodynamic data was recorded in all ventilated groups after surgical preparation and stabilization. ALT-00 (synthesized at the Technical University Graz, Institute of Organic Chemistry) and lisinopril (Sigma-Aldrich, St. Louis, MO) were solved in saline. This was followed by administering a continuous intravenous infusion of lisinopril at 0.15 μg/kg/min, alone or combined with ALT-00 at 2.7, 10, or 100 μg/kg/min for 4 h in the respective treatment groups.

The hemodynamic parameters were recorded at 4 h.

Each experimental group (n = 14) was divided into two alternating sampling subgroups of n = 7 mice. Gender of the mice was alternated within sampling subgroups to achieve an equal distribution of males and females within each experimental group and sampling subgroup. In one sampling subgroup, the right lung was removed and used to measure the wet-to-dry weight ratio, while bronchoalveolar lavage fluid (BALF) was obtained from the left lungs and analyzed by ELISA. In the other subgroup of mice the left lungs were instilled with 4% formalin at a hydrostatic pressure of 20 cmH₂O for 15 min, then removed and stored in 4% formalin for lung histology. Paraffin-embedded lung tissue was sectioned at 3.5 µm and stained with hematoxylin-eosin. No animals or data points were excluded from analysis. All analyses of the samples were performed in an observer blinded manner.

At the end of the experiment, 70 μL of blood was collected from the carotid artery catheter for blood gas analysis, and heparin plasma from another 0.6 mL of blood was collected after centrifugation at 3,000 rpm (4°C) for 10 min and stored at −80°C for measuring the RAS parameters.

The equilibrium levels of angiotensin metabolites (Ang I, Ang II, Ang 1-7, Ang 1-5) in murine heparinized plasma samples were quantified with liquid chromatography-tandem mass spectrometry (LC-MS/MS) performed at Attoquant Diagnostics (Vienna, Austria) using previously validated and described methods (Kovarik et al., 2015; Basu and Yogasundaram, 2017). The lower limits of quantification (LLOQ) for Ang I, Ang II, Ang III, Ang IV, Ang 1-7 and Ang 1-5 were 4, 2, 3, 2, 3 and 4 pmol/L, respectively. The plasma was analyzed following equilibration at 37°C for 30 min. Samples where stabilized by chaotropic inactivation and spiked with stable isotope-labeled internal standards for individual angiotensin metabolites. Subsequently these samples underwent C-18-based solid-phase-extraction and were analyzed with LC-MS/MS, using a reversed-phase analytical column compatible with a Xevo TQ-S triple quadruple mass spectrometer (Waters, Milford, MA). Internal standards were used to correct peptide concentrations for recovery efficiency of the sample preparation for each analyte in every sample. Analyte concentrations were calculated from in-run calibration curves and reported in pmol/L.

We calculated the Ang concentration-based markers PRA-S (Ang I + Ang II concentrations) (Burrello et al., 2020; Guo et al., 2020), ACE-S (ratio of Ang II/Ang I concentration) (Burrello et al., 2020) and ALT-S [(Ang 1-7+Ang1-5 concentrations)/(Ang I+Ang II+Ang 1-7+Ang 1-5 concentrations)] to evaluate renin activity, ACE activity and activation of the alternative RAS axis, respectively (Mao et al., 2022).

Plasma renin activity (PRA) was measured with an LC-MS/MS-based Ang I generation assay (Burrello et al., 2020). Heparin plasma was supplemented with an Ang I-stabilizing inhibitor mix (Attoquant Diagnostics) in phosphate-buffered saline (Dulbecco’s PBS, pH 7.4, Sigma-Aldrich). Of these mixtures, one aliquot was put on ice, while the other was incubated at 37°C for 60 min. Next, samples were stabilized and LC-MS/MS-based quantification of Ang I was performed as described above. The PRA was calculated by subtracting Ang I levels measured in samples on ice from Ang I levels determined in incubated samples as an intrinsic Ang I formation rate given in (ng Ang I/mL/h). The LLOQ of this assay was 0.02 ng Ang I/mL/h.

In one sampling group of animals (n = 7), after 4 h of ventilation, 3 aliquots of phosphate buffered saline, 30 mL/kg each, were instilled into the left lungs and withdrawn slowly 3 times with a BALF recovery of > 90%. The recovered fluid was centrifuged at 3,000 rpm (4°C) for 5 min. The supernatant from the first sample was stored at −80°C for the determination of inflammatory cytokines, including interleukin (IL)-1β, IL-6, keratinocyte-derived cytokine (KC/CXCL1), and macrophage inflammatory protein-2 (MIP2/CXCL2) using Duoset ELISA kits (R&D Systems; Biotechne, MN) according to the manufacturer’s instructions.

Hematoxylin/eosin-stained lung tissue sections were graded in an observer-blinded fashion with the experimental acute lung injury score (Matute-Bello et al., 2011; Kulkarni et al., 2022). The score involved numbers of neutrophils, presence of hyaline membranes and proteinaceous debris as well as alveolar wall thickening, and was calculated as described in the workshop report of the American Thoracic Society published by Matute-Bello et al., in 2011.

The sample size was based on our previous experience with this design (Mao et al., 2022) and on results of pilot studies. Data were analyzed by one-way analysis of variance with Tukey posttest, using GraphPad Prism 8.0 (GraphPad software, San Diego, CA). A p-value < 0.05 was considered significant. For angiotensin peptide levels below the LLOQ, the respective LLOQ divided by the square root of two was used for calculations (Croghan and Egeghy, 2003; Antlanger et al., 2017).

Hemodynamic parameters (MAP and HR) were monitored in all HVT groups for up to 4 h of ventilation. We calculated the ΔMAP = (MAP at 4 h)—(MAP at baseline) shown in Figure 1A, and the ΔHR = (HR at 4 h)—(HR at baseline) shown in Figure 1B. The ΔMAP was greatly decreased by treatment with lisinopril versus untreated LVT and HVT mice (p < 0.001), while additional treatment with ALT-00 at 10 and 100 μg/kg/min reduced the drop in MAP occurring in the group with HVT + lisinopril alone (p = 0.006 and p = 0.046, respectively). This protective effect of additional treatment with ALT-00 at 10 and 100 μg/kg/min on the drop in MAP induced by lisinopril was also observed in animals ventilated at LVT (Supplementary Figure S1). The ΔHR showed no differences between the groups.

The partial pressure of arterial carbon dioxide (PaCO₂) and pH were well maintained after 4 h without significant differences between the groups (Figure 2).

To assess the amount of pulmonary edema, the protein content in BALF and the lung wet-to-dry (W/D) weight ratios were measured. The BALF protein content (Figure 3A) in the HVT group was increased compared to the sham (p = 0.0006) and LVT (p = 0.002) groups. Treatment with lisinopril and ALT-00 decreased the BALF protein content compared to the untreated HVT group (all p < 0.01). The lung W/D weight ratio of the HVT group was increased versus the sham and LVT group (p = 0.044 and p = 0.045, respectively), and slightly reduced by treatment with lisinopril and ALT-00, as compared to HVT controls (Figure 3B).

The evaluation of intra-alveolar inflammation with BALF concentrations of IL-6, KC, MIP2 and IL-1β showed similar patterns. Significant increases of all cytokines only occurred in the HVT group versus sham (p < 0.01), while treatment with lisinopril and ALT-00 at 2.7 and 10 μg/kg/min, but never at 100 μg/kg/min, reduced the BALF concentrations of individual inflammatory cytokines, as compared to the untreated HVT group (p < 0.05, Figures 4A–D). Specifically, lisinopril alone decreased MIP2 and IL-1β levels versus the untreated HVT group. With additional ALT-00 at 2.7 μg/kg/min IL-6, KC, MIP2 and IL-1β levels were decreased, and with ALT-00 at 10 μg/kg/min MIP2 levels were decreased, as compared to the untreated HVT group.

Figure 5 shows a summary of RAS Fingerprints from equilibrium analysis of plasma samples. These RAS Fingerprints include median equilibrium plasma levels of six angiotensin metabolites [Ang I, Ang II, Ang III (= Ang 2-8) and Ang IV (= Ang 3-8), as well as the alternative RAS metabolites Ang 1-7 and Ang 1-5]. The Ang I, Ang II, Ang 1-7 and Ang 1-5 plasma levels in the HVT group were increased compared to the sham group (p < 0.01, Figures 6A–D). Also LVT ventilation somewhat increased these angiotensin metabolite concentrations, but no significant differences compared to the sham group were found. Treatment with lisinopril ± ALT-00 decreased Ang II and Ang 1-5, and increased Ang I and Ang 1-7 concentrations in the respective HVT treatment groups versus the untreated HVT group (Figures 6A–D). The comparison of the HVT group receiving lisinopril with the groups receiving additional ALT-00 showed that Ang II concentrations were higher than in the group treated with lisinopril alone if ALT-00 was administered at 10 μg/kg/min (p = 0.008). Ang II concentrations in the group treated with lisinopril + ALT-00 at 10 μg/kg/min were also higher than in the group treated with lisinopril + ALT-00 at 2.7 μg/kg/min (p = 0.014). Ang II levels in the group treated with additional ALT-00 at 100 μg/kg/min were increased versus lisinopril alone, but without statistical significance. The angiotensin metabolite levels of mice ventilated at LVT under treatment with lisinopril and lisinopril + ALT-00 at 2.7 μg/kg/min are shown in Supplementary Figure S2.

PRA-S as marker of plasma renin activity was increased following treatment with lisinopril + ALT-00 at 10 and 100 μg/kg/min versus the untreated HVT group, lisinopril treatment alone, and lisinopril + ALT-00 at the lowest dose level (p < 0.05 for ALT-00 at 10 μg/kg/min, p < 0.0001 for ALT-00 at 100 μg/kg/min; Figure 7A). A separate measurement of PRA in an Ang I generation assay showed that PRA was highest in the HVT group treated with lisinopril + ALT-0 at 100 μg/kg/min (p < 0.05 versus all other HVT groups), while there was no difference between the untreated HVT group and the other HVT groups receiving lisinopril and ALT-00 (data not shown). ACE activity in plasma (Ang II/Ang I ratio, ACE-S) was downregulated by lisinopril ± all dose levels of ALT-00 (Figure 7B), as compared to the untreated HVT group (all p < 0.0001). The alternative RAS activation marker ALT-S was increased by treatment with lisinopril combined with ALT-00 at 100 μg/kg/min (Figure 7C) versus the sham, LVT and HVT groups (p = 0.001, p = 0.030 and p = 0.042, respectively).

The histological lung injury score was highest in the HVT group (p < 0.0001 versus sham and LVT group, Figure 8A), and was reduced by treatment with lisinopril ± ALT-00 (p < 0.01). Exemplary images of lung histology from each experimental group are shown in Figure 8B. The sham group showed normal lung tissue structures and morphology (Figure 8B/A). In the LVT group, the histological appearance resembled that of the sham group with a few hyaline membranes and neutrophils (Figure 8B/B). The HVT group showed neutrophil infiltration in the alveolar and interstitial space, alveolar septal thickening, and extensive diffuse congestion with proteinaceous debris in the airspaces (Figure 8B/C). In the HVT group receiving lisinopril a few patches of alveolar septal thickening and proteinaceous debris in the airspaces were observed (Figure 8B/D). Combined treatment with lisinopril and ALT-00 at all three dose levels led to a similar histological pattern as with lisinopril alone (Figure 8B/E–G).