Impaired AMPK Activity Drives Age-Associated Acute Lung Injury after Hemorrhage

Trauma is a significant cause of mortality worldwide. Hemorrhage is responsible for 30–40% of trauma mortality, and hemorrhagic shock is the second leading cause of death among the injured (1). Acute lung injury after hemorrhage is associated with an overwhelming systemic inflammatory response characterized by elevated systemic cytokine levels and activation of the innate immune system (2). In hemorrhagic shock, advanced age has been associated with more severe organ failure and an increased number of days spent on the ventilator and in the intensive care unit (3). It is also well known that older individuals (>65 yr old) have greater morbidity and mortality with all etiologies of acute lung injury (4). However, the mechanisms linking age to worse outcomes in acute lung injury are poorly understood.

In this issue of the Journal, Klingbeil and colleagues (pp. 585–596) (5) explore the role of the 5′ adenosine monophosphate–activated protein kinase (AMPK) in age-related acute lung injury after hemorrhagic shock. AMPK is a sensor of cellular energy status that responds to decreases in energy (increased AMP/ATP or ADP/ATP ratios) by stimulating mitochondrial catabolic activity. Aging has been shown to decrease AMPK activity and signaling, which results in reduced mitochondrial biogenesis and function (6). These investigators conducted an in vivo/in vitro analysis of whether the predominant catalytic AMPK subunit in the lung (AMPKα1) was associated with age-related impaired mitochondrial biogenesis and autophagy in response to hemorrhage-induced lung injury.

Arterial hemorrhage to a fixed blood pressure (from 60 to 30 mm Hg) was performed equally in young (3–5 mo; approximately 10–17-yr-old human equivalent) and older (9–12 mo; 30–40-yr-old approximate human equivalent) wild-type mice, followed by fluid/blood resuscitation, then treatment ± an AMPK activator. Hemorrhage induced convincing acute lung injury, as assessed by increases in bronchoalveolar lavage (BAL) total protein, BAL inflammatory cells, wet/dry lung weights, histopathologic lung injury scores, and myeloperoxidase activity in lung homogenates, along with increases in systemic cytokine levels. Overall, there was a more severe lung injury response in the older mice compared with that in younger mice. Furthermore, there was a greater percentage of structurally altered mitochondria in older mice.

Treatment with a pharmacologic AMPK activator (AICAR, an analog of AMP) after resuscitation had a restorative normotensive effect selectively in young mice. Whereas the specific AICAR treatment effects on hemorrhage-induced lung injury and cytokines were similarly beneficial in younger and older mice, the resultant injury phenotype remained more deranged in older mice. Concordant with the beneficial effects of AMPK activation, mice lacking AMPK 1α (AMPK KO) had a greater inflammatory response to hemorrhage (myeloperoxidase, lung pathologic changes, and plasma IL-1β and TNF-α levels) compared with wild-type mice, and AMPK KO mice had no response to AICAR. Taken together, these data demonstrate a beneficial effect of AMPK activation in younger mice, greater than that in older mice, and suggest that the AICAR’s beneficial effects were a result of AMPK activation, rather than off-target effects. This work extends the similar prior findings of the role of AMPK in LPS-induced lung injury (7, 8) to hemorrhage-induced lung injury.

To begin to identify the mechanism of AMPK’s age-dependent effect on lung injury, the putative role of AMPK in both mitochondrial biogenesis and autophagy were measured. Prior work had demonstrated that energy depletion-induced activation of AMPK (by phosphorylation) is impaired with aging (6). This study extends this impaired response to alveolar epithelial cells in mice. AMPK activation is linked to mitochondrial biogenesis through its ability to activate peroxisome proliferator-activated receptor (PPAR)-γ coactivator 1-alpha (PGC-1α) and the deactylase sirtuin-1 (SIRT1) (9). PGC-1α is deacetylated by SIRT1, thereby increasing its activity in the nucleus and inducing a mitochondrial biogenetic gene expression program. These investigators demonstrate that AMPK phosphorylation was impaired in primary alveolar epithelial cells of older mice. Further, PGC-1α and SIRT1 nuclear localization was decreased after hemorrhage and restored with AICAR, but to a lesser extent in older mice. This suggests that the observed impairment of AMPK phosphorylation drives the increase in structurally altered mitochondria in aged mice upon hemorrhage-induced lung injury. AICAR had a modest effect in restoring mitochondrial structure in both groups, with a greater effect in younger mice, suggesting the AMPK-initiated pathway is refractory to activation in older mice.

Although lung injury increased the autophagy marker protein LC3-B in both young and older mice, lung injury, age, and AICAR treatment did not have an effect on the number of autophagic vesicles. These apparently contradictory results have yet to be resolved. Similarly, some of the beneficial effects of AICAR on lung injury could be a result of increased blood flow selectively in young mice on AICAR treatment. AMPK has pleotropic effects in multiple cell types that play a role in lung injury, so future studies could be directed toward identifying the specific cause, cell type, and pathway by which AMPK activation mediates lung injury protection.

Decreased AMPK activation capacity with aging has been attributed to reduced AMPK phosphorylation, but the exact mechanisms of the reduced phosphorylation have not been elucidated (6). Concordant with the current study, the downstream effects of impaired AMPK activation include impaired longevity/autophagy/mitochondrial biogenesis pathways (mTOR, FoxO, PCG-1α, SIRT1), impaired stress resistance pathways (SIRT1, HIF1), and increased inflammatory/immune responses (NF-κB) (10). The reduction of SIRT1 activity and elevation of NF-κB in older mice in this study extends these mechanistic pathways to lung injury after hemorrhage.

As with most key regulatory proteins, there are consequences of both impaired and overexuberant AMPK activation. For example. AMPK is dysregulated in several cancers and can promote cancer cell survival under conditions of metabolic stress, such as hypoxia and glucose deprivation (11–13). Furthermore, AMPK is involved in a diverse panoply of metabolic and physiologic processes such as lipid and glucose metabolism, protein synthesis, and redox regulation. For example, it has been shown that hypercapnia plays a role in the AMPK-dependent atrophy of skeletal muscle in patients with COPD (14). As AMPK activation is cell type process- and context-dependent, one should be cautious in considering AMPK manipulations for therapeutic intent before thorough study.

In addition to acute lung injury, AMPK has been shown to play diverse roles in other inflammatory lung diseases. Metformin has been shown to suppress eosinophilic airway inflammation and reduce peribronchial fibrosis via AMPK activation in asthmatic mice (15). AMPK-induced MMP8 secretion in neutrophils has been implicated in causing lung destruction in human tuberculosis (16). AMPKα1 KO mice exposed to cigarette smoke have increased production of IL-8, leading to increased lung inflammation and emphysema (17). Finally, in experimental pulmonary hypertension, AMPK activation has been shown to be both beneficial (18) and deleterious (19). The role of aging on the AMPK-mediated pathways in the aforementioned lung diseases has yet to be evaluated, and is an exciting area for future research.

The results of this study agree with numerous other studies that demonstrate that AMPK basal activity or phosphorylation is impaired with aging in diverse tissues, including myocardium, skeletal muscle, and endothelium. For example, the capacity for AMPK activation in skeletal muscle in response to exercise was decreased in older rats. As in this study, AICAR administration only increased the activity of AMPK in younger rats (20). In contrast to studies involving LPS-induced lung injury (21), AICAR administration in this study did not affect NF-κB or TNF-α levels after hemorrhage-induced lung injury.

In summary, for the first time, the authors show that hemorrhage-induced lung injury is more severe in older mice than younger mice. Further, the level of AMPK activation in response to AICAR and its beneficial effects are impaired in older mice. The age-dependent effect of AMPK activation on selected measures of lung injury suggests that AMPK is an important determinant of the age-dependent effect of hemorrhage-induced lung injury (Figure 1). It remains to be seen whether AMPK activation is regulatable in older humans. Studies directed into the mechanism by which AMPK activation is impaired in aging would be potentially therapeutically transforming. This work adds to the literature showing that AICAR treatment improves LPS-induced lung injury, and cardiovascular and gastrointestinal injury and survival after hemorrhagic shock in young rats/pigs (22–24). As aspirin, metformin, and statins have all been shown to activate AMPK, the potential for effective therapeutic interventions in hemorrhage-induced lung injury may be on the horizon.

Figure 1.

Working model for the role of 5′ adenosine monophosphate–activated protein kinase (AMPK) in aging and hemorrhage-induced lung injury. Hemorrhage and aging conspire to reduce AMPK phosphorylation, leading to the increased inflammation and impaired mitochondrial regulation that drives lung injury. NF-κB, nuclear factor-κB; PCG-1α, peroxisome proliferator-activated receptor-γ coactivator 1-α; SIRT1, sirtuin 1.