INVITED EDITORIAL
Invited Editorial on "Tumor necrosis factor- in ischemia and reperfusion injury in rat lungs"

Institute for Environmental Medicine, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104-6068

	ARTICLE

Top Article References

THE ISCHEMIA-REPERFUSION (I/R) phenomenon is a paradoxical increase in tissue injury during the reperfusion period in an organ that had sustained relatively minor damage during a period of ischemia (10). Considerable evidence has accumulated to indicate that generation of reactive oxygen species (ROS) is a major factor in initiation of tissue injury with I/R (6). Although the precise mechanisms are still under investigation, the currently accepted paradigm is that ROS generation occurs with reoxygenation during reperfusion, as a consequence of priming coincident with ischemia-induced tissue anoxia. The biochemical changes responsible for priming may vary with different organs, and possibilities include xanthine dehydrogenase conversion to oxidase, liberation of free Fe³⁺ and its reduction to Fe²⁺, activation of membrane-bound NADPH oxidase, and reduction of mitochondrial and microsomal components.

The changes that occur during I/R and form the basis for reperfusion injury apply to organs with systemic circulation where I/R is accompanied by anoxia-reoxygenation. On the other hand, lung ischemia does not lead to lung tissue anoxia, and reperfusion does not mean reoxygenation if ventilation is maintained throughout the I/R period. Pulmonary embolism is a clinical example of "ventilated" ischemia. The maintenance of tissue oxygenation with animal models of ventilated ischemia has been confirmed by measurements of lung tissue ATP content that is essentially unchanged from control values (7). Despite the absence of a cycle of anoxia-reoxygenation with I/R in the ventilated lung, published studies have indicated the occurrence of oxidative I/R injury and of ROS generation that is initiated during the ischemic period (2, 7). Our recent studies (3, 16) indicate that activation of endothelial NADPH oxidase may be the primary mechanism for ROS generation during oxygenated lung ischemia. Conversely, anoxia-reoxygenation can be imposed experimentally on lungs by alternate N₂/O₂ ventilation and results in ROS generation and tissue oxidation during reoxygenation in the absence of ischemia; ROS generation in this model is not via NADPH oxidase (16). NADPH oxidase activation with ischemia may be a physiological response related to cellular signaling, although pathological effects such as lipid peroxidation and protein oxidation can result (4, 7).

Although ROS generation through the NADPH oxidase pathway in the oxygenated lung appears to cease with reperfusion, manifestations of tissue injury may progress (5). These and other studies suggest a role for secondary inflammation in I/R injury subsequent to the initiating events. This inflammatory phase can result in significant amplification of cellular damage. Inflammation-mediated injury is also associated with generation of ROS so that its biochemical footprints may be difficult to distinguish from those of the initiating oxidative insult. The mechanisms for initiation of the inflammatory phase during reperfusion are still under investigation, and many cellular and humoral factors, including platelets, cytokines, the complement system, cell-adhesion molecules, and polymorphonuclear neutrophils (PMNs), may all combine. For example, I/R induces translocation of P-selectin, a cell-adhesion molecule, to the endothelial cell membrane that may result in rolling and adherence of circulating PMNs (15). Endothelial cell-derived interleukin-8, endothelin, leukotriene B₄, tumor necrosis factor- (TNF-), and activated components of the complement cascade can all serve as other potential mediators of the inflammatory phase resulting in the tissue recruitment of PMNs. Marginated and migrated PMNs can release oxidants (e.g., O₂· in the respiratory burst via NADPH oxidase and hypochlorous acid via myeloperoxidase), proteolytic enzymes (e.g., elastase, heparanase), and vasoconstricting agents (e.g., leukotriene B₄). Endothelial cells can also release platelet-activating factor with ischemia or hypoxia that may lead to release of a potent vasoconstrictor thromboxane A₂ (11). Vasoconstriction and adherence of aggregated PMNs and platelets to the endothelium may result in microvascular blockade to cause the "no-reflow" phenomenon during reperfusion and further aggravate the injury. It is certainly not surprising that modification of the cascade with amelioration of reperfusion injury can be demonstrated with a broad variety of potential therapeutic agents.

There has not as yet been a definitive study of the mediators of the reperfusion component of injury in a ventilated lung model. In this regard, the report by Khimenko et al. (8) represents a valuable addition to the literature. The study evaluated the effects of nonventilated (inflated) and air-ventilated ischemia for 45 min, followed by 90 min of reperfusion with air ventilation, on vascular permeability [measured by filtration coefficient (K_fc)] and on perfusate TNF- concentration in isolated rat lungs. With nonventilated I/R, the authors found that K_fc increased fivefold and TNF- increased 1.8-fold within the first 30 min of reperfusion. With air-ventilated I/R, the increases in K_fc and TNF- were significantly greater, indicating that ventilation during ischemia exacerbates the injury in an oxygenated lung. Administration of recombinant TNF- at 50,000 U (a dose twice the maximum perfusate TNF- release with ventilated I/R) potentiated the permeability changes in ischemic but not in control lungs. Antibody to TNF- administered intraperitoneally before lung isolation prevented K_fc increase with both models of injury. Thus these results indicate that ROS generated with ischemia can damage the lung and lead to TNF- release, which can potentiate the lung injury. Increased injury with the ventilated lung model may have been secondary to increased TNF- release, although cause and effect are not yet clear. Potentiation of injury by TNF- in the isolated lung was not through PMN recruitment, suggesting a direct effect of the cytokine. The mechanism for this effect remains to be determined.

The role of proinflammatory cytokines in reperfusion-induced lung injury is getting increasing attention. Recently, it has been shown that reperfusion after prolonged pulmonary ischemia in the isolated lung results in a significant elevation of local tissue levels of TNF- (1). TNF- has been also implicated in mediation of secondary lung injury associated with ischemia in other organs. Serum TNF- increased rapidly during lower extremity ischemia and caused increased production of nitric oxide (NO) from rat lungs by upregulating inducible NO synthase (13). Increased levels of TNF- were related to lung damage with intestinal I/R in rats (12) and were associated with higher lung myeloperoxidase levels after 30 min of supraceliac aortic occlusion followed by 2 h of reperfusion in mice (14). Pretreatment with anti-TNF- antibody or TNF--binding protein was protective in those experiments. In a recent study (9), it was found that plasma levels of TNF- and other proinflammatory cytokines increased significantly during the early postoperative period in lung-transplantation patients with early hemodynamic failure. Clearly, lung transplantation also represents a paradigm for I/R. Based on the report by Khimenko et al. (8), it will be important, in order to understand mechanisms in these models, to separate the direct effects of TNF- from those mediated through secondary inflammation associated with PMN recruitment.

In summary, lung ischemia initiates a complex cascade characterized by generation of ROS, followed on reperfusion by an inflammatory phase that amplifies tissue injury. The amplification phase is the result of complex interaction among cellular (PMNs, endothelial cells, platelets, and others) and humoral factors. The study by Khimenko et al. (8) provides important evidence that TNF- is not only a mediator in the initiation of inflammation but also may have direct effects on tissue damage by ROS. Further study of mechanisms of ROS generation in lung I/R and mechanisms for amplification holds the promise for development of specific interventions to prevent the manifestation of acute lung injury.

	REFERENCES

Top Article References

1.   Abolhoda, A., A. Brooks, M. Choudhry, Y. Kaneda, D. Liu, H. Cheng, and M. Burt. Characterization of local inflammatory response in an isolated lung perfusion model. Ann. Surg. Oncol. 5: 87-92, 1998[Abstract].

2.   Al-Mehdi, A. B., H. Shuman, and A. B. Fisher. Intracellular generation of reactive oxygen species during nonhypoxic lung ischemia. Am. J. Physiol. 272 (Lung Cell. Mol. Physiol. 16): L294-L300, 1997[Abstract/Free Full Text].

3.   Al-Mehdi, A. B., G. Zhao, C. Dodia, K. Tozawa, K. Costa, V. Muzykantov, C. Ross, F. Blecha, M. Dinauer, and A. B. Fisher. Endothelial NADPH oxidase as the source of oxidants with lung ischemia or high K⁺. Circ. Res. 83: 730-737, 1998[Abstract/Free Full Text].

4.   Ayene, S. I., A. B. Al-Mehdi, and A. B. Fisher. Inhibition of lung tissue oxidation during ischemia/reperfusion by 2-mercaptopropionylglycine. Arch. Biochem. Biophys. 303: 307-312, 1993[Medline].

5.   Bishop, M. J., E. Y. Chi, M. Su, and F. W. Cheney. Dimethylthiourea does not ameliorate reperfusion lung injury in dogs or rabbits. J. Appl. Physiol. 65: 2051-2056, 1988[Abstract/Free Full Text].

6.   Cross, C. E., B. Halliwell, E. T. Borish, W. A. Pryor, B. N. Ames, R. L. Saul, J. M. McCord, and D. Harman. Oxygen radicals and human disease. Ann. Intern. Med. 107: 526-545, 1987.

7.   Fisher, A. B., C. Dodia, Z. Tan, I. Ayene, and R. G. Eckenhoff. Oxygen-dependent lipid peroxidation during lung ischemia J. Clin. Investig. 88: 674-679, 1991.

8.   Khimenko, P. L., G. J. Bagby, J. Fuseler, and A. E. Taylor. Tumor necrosis factor- in ischemia and reperfusion injury in rat lungs. J. Appl. Physiol. 85: 2005-2011, 1998[Abstract/Free Full Text].

9.   Mal, H., M. Dehoux, C. Sleiman, J. Boczkowski, G. Leseche, R. Pariente, and M. Fournier. Early release of proinflammatory cytokines after lung transplantation. Chest 113: 645-651, 1998[Abstract/Free Full Text].

10.   McCord, J. M. Oxygen-derived free radicals in postischemic tissue injury. N. Engl. J. Med. 312: 159-163, 1985[Abstract].

11.   Milhoan, K. A., T. A. Lane, and C. M. Bloor. Hypoxia induces endothelial cells to increase their adherence for neutrophils: role of PAF. Am. J. Physiol. 263 (Heart Circ. Physiol. 32): H956-H962, 1992[Abstract/Free Full Text].

12.   Sorkine, P., O. Szold, P. Halpern, M. Gutman, M. Greemland, V. Rudick, and G. Goldman. Gut decontamination reduces bowel ischemia-induced lung injury in rats. Chest 112: 491-495, 1997[Abstract/Free Full Text].

13.   Tassiopoulos, A. K., R. E. Carlin, Y. Gao, A. Pedoto, C. M. Finck, S. K. Landas, D. G. Tice, W. Marx, T. S. Hakim, and D. J. McGraw. Role of nitric oxide and tumor necrosis factor on lung injury caused by ischemia/reperfusion of the lower extremities. J. Vasc. Surg. 26: 647-656, 1997[Medline].

14.   Welborn, M. B., W. G. Douglas, Z. Abouhamze, T. Auffenburg, A. S. Abouhamze, J. Baumhofer, J. M. Seeger, J. H. Pruitt, P. D. Edwards, R. Chizzonite, D. Martin, L. L. Moldawer, and T. R. Harward. Visceral ischemia-reperfusion injury promotes tumor necrosis factor (TNF) and interleukin-1 (IL-1) dependent organ injury in the mouse. Shock 6: 171-176, 1996[Medline].

15.   Weyrich, A. S., X. Y. Ma, D. J. Lefer, K. H. Albertine, and A. M. Lefer. In vivo neutralization of P-selectin protects feline heart and endothelium in myocardial ischemia and reperfusion injury. J. Clin. Invest. 91: 2620-2629, 1993.

16.   Zhao, G., A. B. Al-Mehdi, and A. B. Fisher. Anoxia/reoxygenation vs. ischemia in isolated rat lungs. Am. J. Physiol. 273 (Lung Cell. Mol. Physiol. 17): L1112-L1117, 1997[Abstract/Free Full Text].