Inflammation ( # 2014) DOI: 10.1007/s10753-014-9834-0
Rosiglitazone Dampens Pulmonary Inflammation in a Porcine Model of Acute Lung Injury Valbona Mirakaj,1 Christian Mutz,2 Dierk Vagts,2 Janek Henes,1 Helene A. Haeberle,1 Susanne Husung,2 Tony König,2 Gabriele Nöldge-Schomburg,2 and Peter Rosenberger1,3
Abstract—The hallmarks of acute lung injury (ALI) are the compromised alveolar-capillary barrier and the extravasation of leukocytes into the alveolar space. Given the fact that the peroxisome proliferatoractivated receptor-γ agonist rosiglitazone holds significant anti-inflammatory properties, we aimed to evaluate whether rosiglitazone could dampen these hallmarks of local pulmonary inflammation in a porcine model of lung injury. For this purpose, we used a model of lipopolysaccharide (LPS, 50 μg/kg)induced ALI. One hundred twenty minutes following the infusion of LPS, we started the exposure to rosiglitazone through inhalation or infusion. We found that intravenous rosiglitazone significantly controlled local pulmonary inflammation as determined through the expression of cytokines within the alveolar compartment. Furthermore, we found a significant reduction of the protein concentration and neutrophil activity within the alveolar space. In summary, we therefore conclude that the treatment with rosiglitazone might dampen local pulmonary inflammation during the initial stages of ALI. KEY WORDS: acute lung injury; inflammation; rosiglitazone; cytokine.
INTRODUCTION Acute lung injury (ALI) and the acute respiratory distress syndrome (ARDS) remain devastating disorders of the lung that are associated with a high mortality [1]. A hallmark of ALI is the disruption of the alveolar-capillary barrier, the extravasation of fluid from the vascular space causing pulmonary edema, and the trafficking of leukocytes into the alveolar space [2, 3]. These inflammatory changes are crucial for the development of ALI, a fact supported by the finding that experimental lung injury can be substantially reduced through a depletion of neutrophils [4]. The chemokines that are released during this
1
Department of Anesthesiology and Intensive Care Medicine, Tübingen University Hospital, Eberhard-Karls University, Tübingen, Germany 2 Department of Anesthesiology and Intensive Care Medicine, Rostock University Hospital, Rostock, Germany 3 To whom correspondence should be addressed at Department of Anesthesiology and Intensive Care Medicine, Tübingen University Hospital, Eberhard-Karls University, Tübingen, Germany. E-mail: [email protected]
process activate alveolar macrophages, which then further aggravate the intra-alveolar inflammatory changes [4, 5]. This process results in a self-propagating inflammation within the alveolar space that determines the severity of pulmonary organ injury and compromises pulmonary oxygen exchange [6–8]. Strategies for the treatment of ALI are aimed at the reduction of pulmonary inflammation which alleviates these symptoms and improves overall pulmonary function [9, 10]. Anti-inflammatory effects were described for the peroxisome proliferator-activated receptor (PPAR)-γ. PPARγ is a member of the nuclear hormone receptor superfamily and rosiglitazone is its specific agonist [11]. Several studies have demonstrated the anti-inflammatory potential of rosiglitazone in vitro [12, 13]. Liu et al. demonstrated an anti-inflammatory role for rosiglitazone during lung injury when animals were pretreated with rosiglitazone prior to the induction of lung injury [14, 15]. This anti-inflammatory and protective role for rosiglitazone was also shown in a model of pancreatitis-associated lung injury [16]. However, to date, no study has tested the effects of rosiglitazone as a potential treatment intervention following the onset of pulmonary inflammation and lung injury. To fill this gap of knowledge, we conducted this study in a large animal
0360-3997/14/0000-0001/0 # 2014 Springer Science+Business Media New York
Mirakaj, Mutz, Vagts, Henes, Haeberle, Husung, König, Nöldge-Schomburg, and Rosenberger model of lung injury and used rosiglitazone intravenously or through inhalation to test its effect on the degree of pulmonary and systemic inflammation during lung injury.
METHODS Anesthesia After approval of the study by the local Ethics Committee Rostock on Animal Research and in concordance with the Helsinki Convention for the Care and Use in Animals, domestic pigs were premedicated i.m. with 8 mg kg −1 azaperon (Janssen-Cilag), 60 mg kg −1 ketamine (Bela-Pharm), 0.25 mg kg −1 midazolam (Merckle), and 16 μg kg−1 atropin (B. Braun), after overnight fasting and receiving water ad libitum. After placing an intravenous line, anesthesia was induced i.v. with 3 μg kg−1 fentanyl (Janssen), 16 μg kg−1 ketamine (DeltaSelect), 0.3 μg kg−1 flunitrazepam (Roche Pharma), and 0.3 μg kg−1 pancuronium (Hikma Farmaceutica). The trachea was intubated and anesthesia maintained by continuous i.v. infusion of 85 μg kg−1 h−1 flunitrazepam, 0.2 mg kg−1 h−1 pancuronium, and 7 mg kg−1 h−1 ketamine. Mechanical ventilation was provided by pressurecontrolled ventilation with a Cato ventilator (Dräger). Instrumentation After placement of a three-lumen central venous catheter (Certofix) and an 8.5-F introducer (Arrow) into the right internal jugular vein, a Swan-Ganz thermodilution catheter (Edwards Laboratories) was introduced into the pulmonary artery. Next, the right femoral artery was cannulated with a 4.5-F introducer set (Arrow) and a COLD® catheter (Pulsion Medical Systems) was advanced 30 cm to reach the distal aorta. A suprapubic cystofix catheter (Cystofix® CH 10, B. Braun Melsungen AG, Melsungen, Germany) was inserted for measurement of diuresis. To provide adequate fluid replacement, animals received fullelectrolyte solution 10 ml kg−1 h−1 i.v. (Jonosteril®, Fresenius-Klinik, Bad Homburg v.d.H., Germany) during the study period.
Cardiac output was determined by the thermodilution technique (Baxter CO-computer). The mean value of three injections of 10 ml ice-cooled saline was considered to estimate actual cardiac output if the measurements were within a range of ±10 % of the calculated mean. Intrathoracic blood volume (ITBV), global end-diastolic volume (GEDV), extravascular lung water (EVWL), and indocyanine green (ICG) clearance were determined with the double indicator dilution technique, using the COLD® catheter. Blood gas values were measured using an ABL3 autoanalyzer (Radiometer Copenhagen, Denmark). Bronchoscopy At the beginning of the experimental procedure and at 6 h following the LPS, infusion bronchoscopy was performed by identifying the left lobe, and lavage with 10 ml saline solution was performed. The solution was then centrifuged and the supernatant obtained for further analysis. Endotoxin Infusion and Experimental Protocol Escherichia coli 055:B5 lipopolysaccharide (Sigma Chemicals) was infused at a concentration of 50 μg/kg for 2 h. Animals were divided into four groups as follows: (1) no LPS infusion, (2) LPS infusion + vehicle treatment, (3) LPS infusion + rosiglitazone i.v. (4 mg over 2 h), and (4) LPS infusion + rosiglitazone inhalation (4 mg over 2 h). For details about the experimental protocol, please see Fig. 1. BAL Myeloperoxidase Activity and Protein Content BAL was performed as described above. Cell and protein measurements were performed according to standard methods. Pulmonary myeloperoxidase activity was quantified by enzymatic assay for the azurophilic neutrophil granule protein myeloperoxidase (MPO) as described previously [17]. Measurement of BAL Cytokine Concentration Cytokine concentration was measured in supernatant of BAL and in serum of animals with standard ELISA kits (R&D Systems).
Measurement and Calculations
Statistical Analysis
All intravascular catheters were connected to pressure transducers and signals recorded online (PO-NE-MAH®, Digital Acquisition Analysis and Archive Systems). Measured variables included heart rate (HR), MAP, CVP, PAP, and pulmonary capillary wedge pressure (PCWP).
Statistical analysis was carried out using JMP® software package (SAS Inc. Cary, NC, USA). We performed statistical analysis using Student’s t test (two sided, α
Rosiglitazone Dampens Pulmonary Inflammation in a Porcine Model of Acute Lung Injury Valbona Mirakaj,1 Christian Mutz,2 Dierk Vagts,2 Janek Henes,1 Helene A. Haeberle,1 Susanne Husung,2 Tony König,2 Gabriele Nöldge-Schomburg,2 and Peter Rosenberger1,3
Abstract—The hallmarks of acute lung injury (ALI) are the compromised alveolar-capillary barrier and the extravasation of leukocytes into the alveolar space. Given the fact that the peroxisome proliferatoractivated receptor-γ agonist rosiglitazone holds significant anti-inflammatory properties, we aimed to evaluate whether rosiglitazone could dampen these hallmarks of local pulmonary inflammation in a porcine model of lung injury. For this purpose, we used a model of lipopolysaccharide (LPS, 50 μg/kg)induced ALI. One hundred twenty minutes following the infusion of LPS, we started the exposure to rosiglitazone through inhalation or infusion. We found that intravenous rosiglitazone significantly controlled local pulmonary inflammation as determined through the expression of cytokines within the alveolar compartment. Furthermore, we found a significant reduction of the protein concentration and neutrophil activity within the alveolar space. In summary, we therefore conclude that the treatment with rosiglitazone might dampen local pulmonary inflammation during the initial stages of ALI. KEY WORDS: acute lung injury; inflammation; rosiglitazone; cytokine.
INTRODUCTION Acute lung injury (ALI) and the acute respiratory distress syndrome (ARDS) remain devastating disorders of the lung that are associated with a high mortality [1]. A hallmark of ALI is the disruption of the alveolar-capillary barrier, the extravasation of fluid from the vascular space causing pulmonary edema, and the trafficking of leukocytes into the alveolar space [2, 3]. These inflammatory changes are crucial for the development of ALI, a fact supported by the finding that experimental lung injury can be substantially reduced through a depletion of neutrophils [4]. The chemokines that are released during this
1
Department of Anesthesiology and Intensive Care Medicine, Tübingen University Hospital, Eberhard-Karls University, Tübingen, Germany 2 Department of Anesthesiology and Intensive Care Medicine, Rostock University Hospital, Rostock, Germany 3 To whom correspondence should be addressed at Department of Anesthesiology and Intensive Care Medicine, Tübingen University Hospital, Eberhard-Karls University, Tübingen, Germany. E-mail: [email protected]
process activate alveolar macrophages, which then further aggravate the intra-alveolar inflammatory changes [4, 5]. This process results in a self-propagating inflammation within the alveolar space that determines the severity of pulmonary organ injury and compromises pulmonary oxygen exchange [6–8]. Strategies for the treatment of ALI are aimed at the reduction of pulmonary inflammation which alleviates these symptoms and improves overall pulmonary function [9, 10]. Anti-inflammatory effects were described for the peroxisome proliferator-activated receptor (PPAR)-γ. PPARγ is a member of the nuclear hormone receptor superfamily and rosiglitazone is its specific agonist [11]. Several studies have demonstrated the anti-inflammatory potential of rosiglitazone in vitro [12, 13]. Liu et al. demonstrated an anti-inflammatory role for rosiglitazone during lung injury when animals were pretreated with rosiglitazone prior to the induction of lung injury [14, 15]. This anti-inflammatory and protective role for rosiglitazone was also shown in a model of pancreatitis-associated lung injury [16]. However, to date, no study has tested the effects of rosiglitazone as a potential treatment intervention following the onset of pulmonary inflammation and lung injury. To fill this gap of knowledge, we conducted this study in a large animal
0360-3997/14/0000-0001/0 # 2014 Springer Science+Business Media New York
Mirakaj, Mutz, Vagts, Henes, Haeberle, Husung, König, Nöldge-Schomburg, and Rosenberger model of lung injury and used rosiglitazone intravenously or through inhalation to test its effect on the degree of pulmonary and systemic inflammation during lung injury.
METHODS Anesthesia After approval of the study by the local Ethics Committee Rostock on Animal Research and in concordance with the Helsinki Convention for the Care and Use in Animals, domestic pigs were premedicated i.m. with 8 mg kg −1 azaperon (Janssen-Cilag), 60 mg kg −1 ketamine (Bela-Pharm), 0.25 mg kg −1 midazolam (Merckle), and 16 μg kg−1 atropin (B. Braun), after overnight fasting and receiving water ad libitum. After placing an intravenous line, anesthesia was induced i.v. with 3 μg kg−1 fentanyl (Janssen), 16 μg kg−1 ketamine (DeltaSelect), 0.3 μg kg−1 flunitrazepam (Roche Pharma), and 0.3 μg kg−1 pancuronium (Hikma Farmaceutica). The trachea was intubated and anesthesia maintained by continuous i.v. infusion of 85 μg kg−1 h−1 flunitrazepam, 0.2 mg kg−1 h−1 pancuronium, and 7 mg kg−1 h−1 ketamine. Mechanical ventilation was provided by pressurecontrolled ventilation with a Cato ventilator (Dräger). Instrumentation After placement of a three-lumen central venous catheter (Certofix) and an 8.5-F introducer (Arrow) into the right internal jugular vein, a Swan-Ganz thermodilution catheter (Edwards Laboratories) was introduced into the pulmonary artery. Next, the right femoral artery was cannulated with a 4.5-F introducer set (Arrow) and a COLD® catheter (Pulsion Medical Systems) was advanced 30 cm to reach the distal aorta. A suprapubic cystofix catheter (Cystofix® CH 10, B. Braun Melsungen AG, Melsungen, Germany) was inserted for measurement of diuresis. To provide adequate fluid replacement, animals received fullelectrolyte solution 10 ml kg−1 h−1 i.v. (Jonosteril®, Fresenius-Klinik, Bad Homburg v.d.H., Germany) during the study period.
Cardiac output was determined by the thermodilution technique (Baxter CO-computer). The mean value of three injections of 10 ml ice-cooled saline was considered to estimate actual cardiac output if the measurements were within a range of ±10 % of the calculated mean. Intrathoracic blood volume (ITBV), global end-diastolic volume (GEDV), extravascular lung water (EVWL), and indocyanine green (ICG) clearance were determined with the double indicator dilution technique, using the COLD® catheter. Blood gas values were measured using an ABL3 autoanalyzer (Radiometer Copenhagen, Denmark). Bronchoscopy At the beginning of the experimental procedure and at 6 h following the LPS, infusion bronchoscopy was performed by identifying the left lobe, and lavage with 10 ml saline solution was performed. The solution was then centrifuged and the supernatant obtained for further analysis. Endotoxin Infusion and Experimental Protocol Escherichia coli 055:B5 lipopolysaccharide (Sigma Chemicals) was infused at a concentration of 50 μg/kg for 2 h. Animals were divided into four groups as follows: (1) no LPS infusion, (2) LPS infusion + vehicle treatment, (3) LPS infusion + rosiglitazone i.v. (4 mg over 2 h), and (4) LPS infusion + rosiglitazone inhalation (4 mg over 2 h). For details about the experimental protocol, please see Fig. 1. BAL Myeloperoxidase Activity and Protein Content BAL was performed as described above. Cell and protein measurements were performed according to standard methods. Pulmonary myeloperoxidase activity was quantified by enzymatic assay for the azurophilic neutrophil granule protein myeloperoxidase (MPO) as described previously [17]. Measurement of BAL Cytokine Concentration Cytokine concentration was measured in supernatant of BAL and in serum of animals with standard ELISA kits (R&D Systems).
Measurement and Calculations
Statistical Analysis
All intravascular catheters were connected to pressure transducers and signals recorded online (PO-NE-MAH®, Digital Acquisition Analysis and Archive Systems). Measured variables included heart rate (HR), MAP, CVP, PAP, and pulmonary capillary wedge pressure (PCWP).
Statistical analysis was carried out using JMP® software package (SAS Inc. Cary, NC, USA). We performed statistical analysis using Student’s t test (two sided, α
