Therapeutic interventions with metformin and Tideglusib activate AMPK and promote resolution of lung injury after P. aeruginosa-induced lung injury.
To investigate the process of resolution after bacteria-induced lung injury, mice received a significant, but a non-lethal dose of wild-type strain K P. aeruginosa (PAK; 2.5 × 107/50 µl PBS/mouse; orthopharyngeal administration). Lung sections and bronchoalveolar (BAL) fluids were processed from control and groups of mice subjected to PAK for 24, 48, 72 and 96 hours. Acute lung injury (ALI) is evidenced by infiltration of inflammatory cells, in particular neutrophil influx and accumulation of cellular debris within alveolar spaces (Fig. 1a). This is associated with a thickened septum in mice subjected to PAK, as evidenced 24 hours post infection. Notably, lung injury persists for 48 and 72 hours and substantially decrease 96 hours post-infection (Fig. 1b). Next, we tested the therapeutic potential of AMPK activators metformin and Tideglusib. Metformin is a commonly used drug for T2DM; however, it indirectly activates AMPK, while Tideglusib is a potent, selective and irreversible non-ATP-competitive suppressor of the AMPK inhibitor GSK-3β, with previously established efficacy in animal models and in recent clinical trials for AD [37]. Mice received metformin (65 mg/kg; i.p.) or Tideglusib (50 mg/kg; i.p.) during an established pneumonia-induced ALI, i.e., 24 hours after exposure to PAK. Both, metformin and Tideglusib effectively accelerated the recovery from lung injury, as evidenced by improved lung architecture, reduced amounts of neutrophil and proteinaceous material in alveolar space at the 48, 72 hour, as compared to PAK alone (Fig. 1e,f and Fig. 2a). This recovery is associated with a significant decrease in barrier permeability, indicated by reduced protein content in BAL fluids in either metformin or Tideglusib-treated mice, in particular 72 hours post PAK injection (Fig. 2b). Notably, the enhanced permeability and accumulation of inflammatory cells mostly subsided in all groups of mice, including mice exposed to PAK alone for 96 hours (Fig. 2b).
Metformin and Tideglusib-dependent recovery of AMPK activity during pneumonia- established lung injury.
We have previously demonstrated that metformin and GSK-3β inhibitors are potent activators of AMPK, i.e., they increased phosphorylation of Thr172-AMPK [17, 21, 38]. However, it is not known if these compounds can therapeutically effective for recovery of AMPK activity after bacterial pneumonia-induced lung injury. This is an important issue, given that AMPK phosphorylation is significantly reduced in mice treated with PAK for 24, 48, and 72 hours, as compared to control (untreated) mice (Fig. 3a-c). Metformin (Fig. 3b,c) and Tideglusib (Fig. b,d) effectively increased AMPK phosphorylation at the all-time points post-PAK treatment. Notably, metformin was more effective as compared to Tideglusib, likely due to the stimulatory effects of metformin versus the ability of Tideglusib to prevent AMPK de-phosphorylation.
Metformin and Tideglusib promote recovery from ER stress in alveolar epithelial cells after PAK-induce lung injury.
Exposure to PAK resulted in significant and persistent (i.e., 24–96 hour time period) damage of AECs, as indicated by the accumulation of alveolar epithelial type I cell surface membrane marker T-1α in the BAL fluids (Fig. 4a,b). This result demonstrate a prolonged injury of alveolar epithelium after PAK instillation. Notably, such a prolonged injury persisted despite decrease in initial inflammatory responses (Supplementary Fig. 1). In regards to AECs injury, metformin or Tideglusib effectively reduced and further nearly completely normalized the levels of T-1α in the BAL fluids, as early as administration of AMPK activators for 24 hours, compared to groups of mice subjected to PAK alone (Fig. 4c,d).
ER stress is typically associated with the accumulation of CCAAT/enhancer-binding protein-homologous protein (CHOP). In particular, the most pronounced accumulation of CHOP was evidenced 72 hours after PAK-induced lung injury (Fig. 5). Metformin and Tideglusib reduced a number of cells positive for CHOP, as depicted by co-immunofluorescence staining CHOP and alveolar epithelial cells (type I) in lung sections from mice treated with PAK for 72 hours (Fig. 6a,b). To further investigate the effects of AMPK activators on ER-stress, CHOP is measured in whole lung homogenates. This assay is not exclusive for AECs, although important given nearly a 70% of AECs type are present in lungs. We found that metformin and Tideglusib have a significant impact on ER stress (Fig. 7a), in particular CHOP is diminished after exposure to AMPK activators for 24, 48, 72 and 96 hours post PAK (Fig. 7b,c).
AMPK activation reduces AECs apoptosis and promotes clearance of apoptotic cells in lung of mice after P. aeruginosa-induced lung injury.
Given that lung injury is associated with ER stress-related cell death, we investigated if AMPK activators affect AECs apoptosis. At the time of established ER-stress, i.e., 72 hours post exposure to PAK, there is a significant accumulation of apoptotic alveolar cells and other cell populations, as indicated by fluorescence images of (TUNEL staining) from lung sections (Fig. 8a). Importantly, metformin or Tideglusib effectively diminished the abundance of apoptotic cells (Fig. 8a,b). It is important to note that the accumulation of apoptotic cells is not only dependent on the rate of dying cells, but also affected by the ability of lung macrophages to neutralize apoptotic cells, also known as efferocytosis [39, 40]. Given that efferocytosis plays a crucial role in resolution from inflammatory conditions and wound healing, clearance of apoptotic cells was measured upon exposure of peritoneal mature macrophages to metformin (0 or 0.5 mM) and Tideglusib (0 or 30 µM) for 18 hours. The efferocytosis assay was conducted by inclusion of the first dose of apoptotic alveolar epithelial cells (primary murine AECs Type I) labeled with PKH-26 fluorescent dye for 2 hours. Next, the second dose of apoptotic AECs-PKH-67 were included for an additional 45 minutes followed by flow cytometry of macrophage with lineage marker T-1α. Metformin and Tideglusib preserved a continuous clearance of apoptotic AECs (Fig. 8c,d). These results indicate that along with a reduced ER-stress in AECs, the number of apoptotic cells can also be affected by AMPK-dependent enhancement of efferocytosis.