| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Pulmonary, Allergy, and Critical Care Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
Correspondence and requests for reprints should be addressed to Augustine M. K. Choi, M.D., Division of Pulmonary, Allergy, and Critical Care Medicine, University of Pittsburgh School of Medicine, 3459 Fifth Avenue, MUH 628, Pittsburgh, PA 15213. E-mail: choiam{at}upmc.edu
ABSTRACT
Acute respiratory distress syndrome (ARDS) and acute lung injury (ALI) have been the major cause of morbidity and mortality in intensive care units (ICU) over the past decades despite advances in therapeutic modalities. This syndrome is characterized by noncardiogenic pulmonary edema, and pulmonary and systemic inflammation resulting in respiratory failure (1, 2). Both exudative and proliferative organizing phases of ARDS/ALI have been described pathologically (3, 4). The exudative phase is often called diffuse alveolar damage (DAD) characterized by inflammation and hyaline membrane composed of fibrin and cellular debris (3–5). Pulmonary alveolar cell death is the major pathologic change during the exudative phase in DAD. Repair and remodeling of injured lung cells occur during the proliferative phase, characterized by hyperplasia of alveolar type II cells and fibroblast proliferation (3–5). A variety of cellular insults can cause ALI/ARDS including but not limited to sepsis, trauma, drugs, high concentration of oxygen therapy, and mechanical ventilation (1, 2, 5). The broad spectrum of insults that potentially cause ARDS highlights the complexity of pathogenesis of this syndrome.
Key Words: carbon monoxide • cytoprotection • heme oxygenase • lung injury
Accumulating data indicate that antioxidant enzymes and/or stress response genes play important functional roles in conferring cytoprotection in acute lung injury (6, 7). One of these stress inducible genes is heme-oxygenase 1 (HO-1) (6). HO-1, first identified in 1968 (8), is the inducible isoform of heme oxygenases (HO). HO-2 is the constitutive isoform; HO-3 is a recently identified constitutive isoform that has not been well characterized to date. HO catalyzes heme degradation to generate carbon monoxide (CO), biliverdin, and free iron (8, 9). In the early years after its discovery, much of the investigations of HO-1 were focused on the regulation of heme degradation, since HO represented the rate-limiting enzyme in the breakdown of heme moieties. However, in the late 1980s up to the present, reports were published demonstrating that HO-1 can be induced by multiple stimuli and cellular stresses, and expressed ubiquitously in various cell types and tissues (6, 10). These observations that HO-1 can be induced by cellular stresses opened a new line of investigations that HO-1 indeed could function as a stress response molecule in addition to its major function of degrading heme. Oxidative stress and endotoxin, both potent mediators in ALI, markedly increase HO-1 expression in various pulmonary cell types including epithelial, endothelial, fibroblasts and macrophages (11). Accumulating data demonstrate that HO-1 and its product, carbon monoxide (CO), may play protective roles in ARDS/ALI. This review will focus on the cytoprotection by HO-1 and CO in preclinical in vivo models and cell culture in vitro models of ALI.
EXPRESSION OF HO-1
ARDS is a complex syndrome involving many factors such as inflammatory cytokines and reactive oxygen species. Many investigators have focused on the regulation and function of stress-inducible gene products in the pathogenesis of ARDS/ALI. HO-1, as a stress-inducible gene product, is one such stress response gene that has been extensively investigated in ALI/ARDS. Camhi and coworkers showed that HO-1 expression and activity are induced by LPS (11) in macrophages and also demonstrated that LPS-induced HO-1 expression is transcriptionally regulated via activation of AP-1 DNA binding transcription (11). The upregulation of HO-1 expression was observed in macrophages and bronchoalveolar epithelial cells, suggesting the potential roles of HO-1 in ALI induced by sepsis (11). Hyperoxia-induced lung injury is a well-established model for ALI/ARDS (12) and induces high levels of HO-1 expression in vitro and in vivo (13). Keyse and Tyrrell also showed in 1989 that HO-1 is highly induced by oxidative stress (14). Recently, Liu and colleagues demonstrated upregulation of HO-1 in a rat model of ARDS induced by oleic acid (OA) (15). Interestingly, CO, which represents a major by-product of HO-1 catalysis of heme, was highly detectible in pulmonary arterial and bronchoalveolar lavage fluid (BALF) in this study. The concentration of CO in pulmonary artery was markedly higher at 2 h and maintained at high levels thereafter. Concentration of CO in BALF increased after 0.5 to 24 h and subsided by 72 h. HO-1 is also expressed in pulmonary blood vessel walls, bronchial epithelial cells, alveolar epithelial cells, and inflammatory cells in OA-induced lung injury in rats (15). Zegdi and coworkers also showed increased pulmonary HO activity and HO-1 expression in a rat model of post-extracorporeal circulation–induced ALI (16). The expression of inducible heme oxygenase is mainly localized in inflammatory cells. In this study, exhaled CO concentration increased after extracorporeal circulation from 0.16 ± 0.05 ppm at baseline to 0.7 ± 0.2 ppm at the end (p = 0.0001) (16). Furthermore, in patients with ARDS, Mumby and colleagues studied the patients in intensive care units meeting the diagnostic criteria of ARDS. HO-1 protein levels were elevated in the lungs of patients with ARDS in both BALF and/or lung tissue (from patients with lung resection) compared with those of control patients (17). These observations further confirm the induction of HO-1 expression in ARDS and highlight the potential functional role of HO-1 in ARDS.
FUNCTIONAL ROLE OF HO-1/CO IN VITRO
Accumulating reports have demonstrated the protective roles of HO-1/CO in various models of lung injury. Lee and coworkers showed that HO-1 is cytoprotective against hyperoxic stress (18). In their study, A549 lung epithelial cells overexpressing HO-1 were generated genetically to better understand the function of HO-1 in response to hyperoxia. They observed that cells overexpressing HO-1 exhibited increased resistance to hyperoxia-induced cell death (18). Suttner and colleagues also showed that HO-1 overexpression in rat fetal lung cells (RFL-6) transiently transfected with a full-length rat HO-1 cDNA construct conferred decreased cell death, protein oxidation, and lipid peroxidation, suggesting resistance to oxidative injury (19). Petrache and coworkers successfully showed that overexpression of HO-1 prevents tumor necrosis factor (TNF)-
–induced apoptosis (20). Interestingly, in the same study, Petrache and collegues demonstrated that CO provides similar protection against TNF-–induced apoptosis in fibroblasts (20). Additional reports further suggest the potent antiapoptotic effects of CO (21). Brouard and colleagues demonstrated that CO prevents apoptosis in endothelial cells (EC), independent of HO-1 expression. Exposure of EC to CO prevents apoptosis similarly to the effect of HO-1 (21) even in the presence of tin protoporphyrin, a potent chemical inhibitor of HO activity. The precise mechanism or signaling pathway by which CO prevents apoptosis is not clear, but activation of the p38 mitogen-activated protein kinase (MAPK) may be involved in CO-induced antiapoptotic effects (22). Zhang and coworkers also showed that the p38 MAPK is involved in the signal transduction pathway of CO-induced cytoprotection in an ischemia–reoxygenation lung injury model (22).
CO, besides preventing apoptosis, is cytoprotective against oxidative stress, as shown in the studies by Otterbein and colleagues (23). CO confers potent antiinflammatory effects, downregulating proinflammatory cytokines such as TNF-, interleukin (IL)-6, and IL-1ß and at the same time increasing antiinflammatory cytokines such as IL-10 (24, 25). This antiinflammatory effect was also observed similarly in vivo (24, 25). CO also imparts potent antiproliferative effects in various cell types, including lymphocytes, fibroblasts, and smooth muscle cells (26–28).
FUNCTIONAL ROLE OF HO-1/CO IN VIVO
Recent reports have demonstrated the cytoprotective role of HO-1 and/or CO in various in vivo models of acute lung injury. Otterbein and coworkers first demonstrated that hemoglobin-induced HO-1 confers cytoprotection in a rat model of sepsis (29). The same group, in their subsequent studies, further showed that exogenous administration of HO-1 by adenovirus–HO-1 gene transfer confers protection against hyperoxia-induced lung injury. Exogenous HO-1 was delivered and expressed by Ad5–HO-1, a recombinant adenovirus expressing HO-1. In this study, lung injury was assessed by BAL cell count, pleural effusion, and histologic analyses (30). Otterbein and colleagues further showed that inhaled CO at low concentration (50–500 ppm) protected against hyperoxia-induced lung injury in rats (31). Hashiba and coworkers also demonstrated that overexpression of HO-1 in a murine model of ALI caused by the type A influenza virus significantly decreased the inflammatory response in the injured lung and reduced the level of DNA damage in lung epithelial cells via a caspase-8 mediated pathway (32). Sarady and coworkers illustrated the potential mechanism of the protective effects of CO in acute lung injury induced by sepsis/LPS (33). In this study, HO-1 or low concentration of CO, inhibited LPS-induced GM-CSF production in macrophages. Furthermore, Sarady and colleagues also demonstrated that inhaled CO could increase the survival of rats exposed to lethal endotoxemia, and interestingly showed that CO can differentially affect inducible nitric oxide synthase (iNOS) expression in a tissue-dependent manner. These authors showed that CO can inhibit iNOS expression in lung and alveolar macrophages but at the same time increase hepatic iNOS expression in vitro and in vivo (34). This tissue specific differential effect of CO may account for the cytoprotection against endotoxin. Moreover, the same group also observed that CO-induced cytoprotection in TNF-–induced hepatitis and liver failure may be mediated by CO-induced, nuclear factor-
B–dependent expression of iNOS and HO-1 (35).
Another independent group showed that in a rat model of hyperoxia-induced ALI, intratracheal administration of hemogloblin significantly induces HO-1 and subsequently protects rats from hyperoxia-induced lung injury. They also concluded that these protective effects are not mediated via increased HO-1 activity (36), but rather that ferritin, a major by-product of HO catalysis of heme, may mediate HO-1 cytoprotection. Yang and colleagues also showed in their study, using replication-defective retroviral vector, that overexpression of HO-1 in lung microvessel endothelium confers significant resistance to oxidative stress (37). In a model of ozone-induced lung injury, Li and coworkers found that HO-1 is induced by ozone and may play the protective role in this model (38). In their study, tin protoporphyrin IX dichloride (SnPP), a potent chemical HO inhibitor, reverses the protective effects of HO-1 in ozone-induced injury (39). In a model of ischemia-induced lung injury, Fujita and colleagues demonstrated that HO-1–deficient (Hmox1–/–) mice exhibit lethal ischemic lung injury. Furthermore, inhaled CO can rescue these mice from lethal lung injury. These authors suggest that CO may be mediating these cytoprotective effects by depressing the fibrinolytic pathway (40, 41). Zhou and coworkers also showed that heme-induced HO-1 plays a protective role in a rat model of limb ischemia–induced lung injury (39). Inoue and colleagues illustrated a potential mechanism by which HO-1 may provide cytoprotection (40). They first developed a model, using direct, intratracheal inoculation of an adenovirus encoding heme oxygenase 1 (Ad.HO-1) to increase exogenous HO-1. Further studies demonstrated that overexpression of HO-1 in macrophages resulted in high levels of IL-10 production. As the increase in IL-10 in the macrophages is known to be critical for the resolution of neutrophilic migration in the lung when exposed to LPS (40), overexpression of exogenous HO-1 is thought to provide its cytoprotection at least via the mediation of IL-10 production (40).
Minamino and coworkers studied the pulmonary vasculature in hypoxia-induced lung injury (42). HO-1 plays crucial cytoprotective roles in preventing both pulmonary inflammation, pulmonary hypertension, and vessel wall hypertrophy induced by hypoxia. Using transgenic mice that constitutively overexpress HO-1 in the lung, proinflammatory cytokines and chemokines induced by hypoxia are significantly attenuated by overexpression of HO-1, suggesting the protection by HO-1 in pulmonary vascular constriction and inflammation (42).
Most reported studies point to the protective roles of HO-1 in ALI/ARDS, and it is generally believed that HO-1 functions through its major downstream product, CO. However, in vivo studies on the protective effects of CO remain controversial. Dennery and colleagues showed in their study that HO-1 null mice are less susceptible to hyperoxia-induced lung injury by decreasing lung reactive iron and iron-associated proteins, NADPH cytochrome cp450 reductase activity, and lung peroxidase activity (43). Surprisingly, administration of tin protoporphyrin, an HO inhibitor, in the wild-type mice decreased ALI caused by hyperoxia. HO-1 null mice survive chronic hyperoxia as well as wild-type controls. Transduction of human HO-1 in the HO-1 null mice decreases the survival under hyperoxia (43). Clayton and coworkers studied the role of CO in protecting ALI in vivo using a rat model of ALI caused by hyperoxia. Lung wet-to-dry weight ratio (W/D), pleural fluid volume, myeloperoxidase (MPO) activity, histology, expression of HO-1, and manganese superoxide dismutase (Mn SOD) proteins were compared in their study. Low concentration of CO decreased W/D significantly. Surprisingly, pleural fluid accumulation was not affected by CO treatment. CO treatment has negligible effect on lung MPO and expression of oxidant responsive proteins Mn SOD and HO-1. Furthermore, rats treated with low concentration of inhaled CO showed significant apoptosis in the cortex and hippocampus (44). Most recently, Ghosh and colleagues investigated the effects of inhaled CO in three different in vivo mouse models of ALI: hyperoxia-, LPS-, and OA-induced ALI. They found that inhaled CO did not yield significant changes in the measured physiologic or immunologic parameters such as lavage fluid neutrophils, total protein, and cytokines (45). The differential effects of inhaled CO in certain models of tissue injury are not clear at this time. The mechanism(s) by which HO-1 and/or CO biochemistry and biology regulates stress responses in tissue and cells to maintain homestasis is obviously complex. The complexity of the mechanism of HO-1 or CO's role in lung injury is further highlighted by recent report that CO provides potent cytoprotection in a model of ventilator-associated lung injury. Dolinay and coworkers showed a protective role of inhaled CO in preventing inflammation caused by ventilator injury, and the protective effect of inhaled CO is via the p38 mitogen-activated protein kinase pathway but independent of activator protein-1 and nuclear factor-B pathways (46).
In summary, since the discovery and characterization of HO-1 in 1968, the study of HO-1 has been transformed significantly. The HO-1 field has moved from being a "mere" enzyme that degraded heme to its by-products bilirubin and ferritin to a molecule that is highly induced by cellular stresses and provides potent cytoptection against cellular and tissue injury. The breakthroughs of nitric oxide in biology and application in human diseases, and the fact that another endogenously derived gaseous molecule, CO, a major by-product of HO-1, possesses intriguing cellular functions, have stimulated much interest in HO-1 and/or CO biology in human diseases. Challenges lie ahead in our attempts to better define the mechanism(s) by which HO-1 and CO regulate stress responses to help maintain cellular and tissue homeostasis. With the known toxicity of CO at high concentration in inducing tissue hypoxia by the avid binding of CO to hemoglobin and CO binding to heme moieties especially in the mitochondria, formidable challenges lies ahead in any attempts to use inhaled CO in human diseases. Rigorous pharmacokinetics and dose–response studies are needed in humans to even entertain the idea of using inhaled CO akin to the current use of inhaled NO in a variety of human diseases. The recent report by Motterlini and colleagues demonstrating that CO releasing compounds can confer similar physiologic effects as gaseous CO in vascular systems provides intriguing possibilities for future studies (47).
There is much work to be done!
FOOTNOTES
Conflict of Interest Statement: Neither of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
(Received in original form March 24, 2005; accepted in final form May 9, 2005)
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  R. A. Oeckler and R. D. Hubmayr Ventilator-associated lung injury: a search for better therapeutic targets Eur. Respir. J., December 1, 2007; 30(6): 1216 - 1226. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
