Phagocyte-derived reactive oxygen species as suppressors of inflammatory disease.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 58, No. 10, October 2008, pp 2931–2935 DOI 10.1002/art.23941 © 2008, American College of Rheumatology EDITORIAL Phagocyte-Derived Reactive Oxygen Species as Suppressors of Inflammatory Disease Lena Björkman, Claes Dahlgren, Anna Karlsson, Kelly L. Brown, and Johan Bylund important not only in terms of the quantity of radical produced, but also in terms of where the production is localized. Neutrophils are phagocytic leukocytes that are central for host defense and for rapid eradication of infecting pathogens. These cells are armed with a variety of potent antimicrobial systems, including the NADPH oxidase that is capable of generating vast amounts of reactive oxygen species (ROS) during the so-called respiratory burst. In addition to playing a vital role in microbial killing, ROS have long been considered important culprits of inflammatory tissue damage. The finding by Ferguson et al, which is reported in this issue of Arthritis & Rheumatism (1), that neutrophils from patients with the autoinflammatory SAPHO syndrome (synovitis, acne, pustulosis, hyperostosis, and osteitis) have a specific defect in intracellular ROS production, is as interesting as it is intriguing; it indicates that decreased ROS production may play a role in the development of this inflammatory disorder. The finding is consistent with several recent reports describing how absent or compromised phagocytic ROS production confers a state of hyperinflammation instead of resulting in a milder inflammatory response, as would have been expected based on existing dogma. We may thus have arrived at a point of reconsideration regarding the role of ROS in inflammatory disease; the previous “bad guys” accused of harming innocent bystanders may in some instances be the “good guys” capable of dampening inflammatory responses and in this way limiting the extent of tissue damage. Ultimately, it may be vitally Phagocyte-derived ROS in inflammation One major function of the respiratory burst became obvious when the defect leading to the syndrome called fatal granulomatous disease of childhood (now known as chronic granulomatous disease [CGD]) was identified in the 1950s (for review, see ref. 2). This rare disease stems from defects in the genes encoding subunits of phagocyte NADPH oxidase (phox) (Figure 1), resulting in phagocytes incapable of ROS production. The disease was originally named to depict the syndrome of recurrent life-threatening infections often combined with the development of sterile granuloma in hollow organs, and only later was this condition demonstrated to be associated with a lack of ROS production during phagocytosis. Since then, there has been a broad, rigid consensus that the ROS formed by phagocytes during the respiratory burst are required to eradicate pathogenic microbes but may also cause “collateral damage” to surrounding cells and tissues. An overproduction of ROS would, according to this hypothesis, contribute to the tissue-destructing activities associated with various inflammatory disease states. This idea has been hard to reconcile with the fact that CGD is often associated with hyperinflammatory syndromes that are remarkably similar to idiopathic inflammatory diseases (2,3). Despite these overt inflammatory phenotypes in CGD, a potential link between radicals and inflammation has been disregarded in favor of the assumption that the inflammatory disorders in CGD result from persistent underlying infections by nonculturable microorganisms rather than from the immunodeficiency itself. The possibility that low-grade infections drive the inflammatory disease becomes less compelling with increasing evidence that ROS have roles in metabolism, cell death, apoptosis, induction of Supported by the Swedish Medical Research Council, the King Gustaf V Memorial Foundation, the Fredrik and Ingrid Thuring Foundation, the Wenner-Gren Foundations, the Swedish Society for Medicine, and the Swedish State under the LUA/ALF agreement. Lena Björkman, MD, Claes Dahlgren, PhD, Anna Karlsson, PhD, Kelly L. Brown, PhD, Johan Bylund, PhD: University of Gothenburg, Gothenburg, Sweden. Address correspondence and reprint requests to Johan Bylund, PhD, Phagocyte Research Laboratory, Department of Rheumatology and Inflammation Research, University of Gothenburg, Guldhedsgatan 10, S-413 46 Gothenburg, Sweden. E-mail: johan. email@example.com. Submitted for publication June 3, 2008; accepted in revised form June 27, 2008. 2931 2932 BJÖRKMAN ET AL Figure 1. Phagocyte NADPH oxidase and its activation. NADPH oxidase consists of 2 membrane-bound subunits (cytochrome b) present in the plasma membrane and in the membranes of the specific granules, as well as at least 3 cytosolic components. Activation of NADPH oxidase occurs by translocation of the cytosolic components to the cytochrome b. An electron from cytoplasmic NADPH is ferried across the membrane, resulting in the formation of superoxide. In response to, for example, stimulation with phorbol myristate acetate, activation occurs both at the plasma membrane, resulting in extracellular release of reactive oxygen species (ROS), and at an intracellular location. The exact nature of this intracellular location is not yet established, but in order to give rise to an intracellular chemiluminescence (CL) reaction, the ROS need access to myeloperoxidase (MPO) that is stored in azurophil granules. Colocalization of ROS and MPO is probably a result of a fusion event involving both specific and azurophil granules. Details on the chemiluminescence techniques used for the detection of ROS generated at different sites are provided in ref. 13. HRP ⫽ horseradish peroxidase; SOD ⫽ superoxide dismutase. host defense genes, oxidative signaling, and the regulation of inflammation (4). Nevertheless, it has been difficult to conclusively determine—and demonstrate— whether ROS-deficient cells drive the development of inflammatory diseases or whether repeated infections that are slow to resolve in CGD patients are the catalyst for the inflammatory disease. The study by Ferguson et al, showing markedly decreased production of ROS specifically in intracellular compartments (discussed below), seems to favor the former possibility (that ROSdeficient cells drive inflammatory disease), since the patients with SAPHO syndrome had no history of recurrent infections (1). Inherent antiinflammatory roles of ROS are consistent with several in vivo studies. For example, rodents with a polymorphism in the Ncf1 gene (encoding the cytosolic oxidase component p47phox) (Figure 1) leading to phagocytes with decreased capacity to produce ROS are more susceptible to the development of autoimmune inflammatory diseases (e.g., arthritis) (for review, see ref. 5). The proposed mechanism for protection of wild-type animals is that ROS released from phagocytes decrease the levels of surface thiol groups on neighboring lymphocytes and thereby inhibit the activity of arthritogenic T cells (5). Another mechanism was recently suggested to explain exaggerated inflammatory responses of CGD (p47phox-knockout) mice, where superoxide was shown to be crucial for normal tryptophan metabolism. A dysfunctional tryptophan conversion in the absence of ROS was shown to result in unrestrained activity of interleukin-17–producing T cells together with decreased activity of regulatory T cells (6). Such complex immunologic mechanisms may also partly explain why CGD (gp91phox-knockout) mice are susceptible to severe, T cell–dependent (collageninduced) arthritis (7) and persistent, severe, sterile hyperinflammation and tissue necrosis (8). We and others have reported that isolated leukocytes from CGD pa- EDITORIAL tients (and mice) produce more proinflammatory mediators (chemokines and cytokines) than ROS-competent cells in vitro (9,10). These cells could be viewed as intrinsically hyperinflammatory, and the effect would thus be independent of the production of extracellular ROS as a means of decreasing T cell activity. Thus, there is in vivo as well as in vitro evidence that ROS can limit inflammatory responses in several ways. Several distinct mechanisms probably account for this, both where the ROS function as signaling molecules between neighboring immune cells and where they act directly on gene expression and/or metabolism of a given cell. It is known, although often overlooked, that the oxidant–antioxidant balance influences immune cell function (for review, see ref. 4). It has been noted that signal transduction pathways and gene transcription are sensitive to cellular redox states. Zinc-finger coordination, for example, is a redox-sensitive cellular event. Since many transcription factors and RNA-(de)stabilizing proteins contain zinc-finger domains, DNA binding, transcription, and messenger RNA stabilization are subject to change with alterations in the cellular redox states. Indeed, several important transcription factors are known to be redox sensitive (11). Redox states can also alter protein oxidation. In particular, this has been shown to alter the activity of redox-sensitive phosphotyrosine kinases. Exactly how ROS (generated in fused granules or elsewhere; see below) would interact with, or even gain access to, the various signaling components in the cytoplasm/nucleus remains to be determined. Hence, redox-sensitive signal transduction is an area that warrants future investigation. Given the scarcity of individuals with a deficiency in ROS production, it is reasonable to use chemical inhibitors of NADPH oxidase to produce ROS-deficient cells. However, it should be noted that these inhibitors often have nonspecific (NADPH oxidase–independent) side effects and are inaccurate models for cells that are naturally deficient in radicals (10). The use of chemical NADPH oxidase inhibitors with nonspecific side effects is likely to explain many of the contradictory results regarding the role(s) of ROS in cellular processes (12). The patients with SAPHO syndrome investigated by Ferguson and coworkers (1) did not completely lack intracellular ROS production, and it is worth noting that small differences in the levels of ROS may have a noticeable impact on cellular function and biologic phenotype. This is illustrated by the fact that inflammatory manifestations of CGD are more commonly seen in the X-linked form (gp91phox deficiency), in which patients’ phagocytes completely lack ROS production, than 2933 in the autosomal-recessive form (most often p47phox deficiency) (3), in which cells are able to produce minute, but detectable, levels of ROS (13). In addition, female carriers of X-linked CGD, whose phagocyte populations represent a mosaic of normal and ROSdeficient cells, have a higher incidence of discoid lupus, an autoimmune disease of the rheumatic family, compared with the general population (3). It should also be mentioned that even in CGD patients with identical genetic defects, the inflammatory disease is clinically heterogeneous, thereby pointing to other factors, perhaps genetic (polymorphisms) or environmental, that may be important in the progression and severity of inflammatory conditions in both CGD and SAPHO syndrome. The link between reduced/absent ROS production and defective microbial killing by neutrophils has been the main focus of research on phagocyte-derived ROS for many years. More recent evidence, including the report by Ferguson et al, challenges conventional thinking on the role(s) of radicals. Today we are in the middle of a paradigm shift, with increasing evidence pointing to the fact that ROS limit the inflammatory response, thus curbing the extent of inflammatory tissue damage and the development of inflammatory disorders. Like many elements of the immune and inflammatory response, ROS appear to be a double-edged sword, contributing directly to tissue damage, yet clearly capable of suppressing the development of inflammatory diseases. Many of the suggested regulatory functions of ROS involve modification of extracellular (or membranebound) proteins/peptides, which may be the easiest way to visualize extracellular ROS released from phagocytes as having an influence on cellular and biologic processes. Interestingly, added scavengers, such as superoxide dismutase (SOD) (which dismutates superoxide to H2O2) or catalase (which removes H2O2), which are known to decrease the severity of inflammatory conditions such as arthritis (for review, see ref. 14), are not cell permeable. Thus, the roles of ROS in inflammation probably differ depending not only on the levels of ROS produced, but also on where the ROS are generated. The report by Ferguson et al (1) correlates the intracellular production of radicals with the incidence of inflammatory disease. Intracellular ROS production has long been synonymous with intraphagosomal production (i.e., ROS being produced inside the phagosomal compartment after uptake of a prey). However, many phagocytes have the ability to generate ROS in the absence of a particulate prey (i.e., in an intracellular compartment distinct from the phagosome) (15). The details are 2934 gradually becoming clearer regarding this activity and the cellular compartment in which it occurs. Production and detection of intracellular ROS Phagocyte NADPH oxidase consists of 2 membranebound components (comprising the cytochrome b) and at least 3 cytosolic components (Figure 1). In the resting neutrophil, a minor fraction of cytochrome b is localized in the plasma membrane, while the major part is found in membranes of intracellular organelles, mainly the specific granules (16). Upon cell activation, the cytosolic proteins translocate to the membrane-bound cytochrome b, and a functional multicomponent electrontransfer system is formed that reduces molecular oxygen to superoxide anion at the expense of cytosolic NADPH (Figure 1). This activation can take place either at the plasma membrane, leading to release of oxygen radicals to the extracellular environment, or at intracellular membranes, leading to ROS formation in intracellular compartments. Neutrophil NADPH oxidase can be activated in response to a number of physiologic and synthetic agonists, and the nature of the cell-activating agent determines where in the cell the NADPH oxidase is activated (i.e., in the plasma membrane or in intracellular membranes). Signaling through chemoattractant receptors, for example, mainly activates the plasma membrane–localized NADPH oxidase that releases ROS extracellularly, while engulfment of a particulate prey primarily generates ROS within the phagosome. It is, however, becoming increasingly clear that a substantial part of the oxidants produced by activated neutrophils are generated at an intracellular site distinct from the phagosome; activation of CR3 on the neutrophil surface is one example of physiologic stimulation that can result in intracellular ROS production in the absence of phagosome formation (15). In cell-free systems, NADPH oxidase assembly and activation can take place on specific granule membranes, but the precise compartment in the intact cell where intracellular (nonphagosomal) activation takes place has yet to be defined. A number of techniques are used to determine cellular ROS production in/from neutrophils (and other cells) (13). The chemiluminescence technique used by Ferguson et al (1) to specifically detect intracellular respiratory burst activity uses a membrane-permeable detector molecule, luminol, which is excited by ROS and emits light when returning to its ground state. To specifically detect intracellular ROS, luminol is combined with SOD and catalase, both of which are unable BJÖRKMAN ET AL to cross biologic membranes and will therefore consume the extracellular radicals (13). The intracellular luminolamplified chemiluminescence reaction in neutrophils is highly dependent on the participation of myeloperoxidase (MPO) that is stored in another granule type, the azurophil granules. Theoretically, ROS generated in specific granules could reach MPO in the azurophil granules by diffusion. However, it is more reasonable to assume that the ROS reach the MPO-containing compartment through fusion between the 2 granule types involved, a process that would bring the 2 entities required for the chemiluminescence reaction (superoxide anion and MPO) into the same compartment (15) (Figure 1). The fusion hypothesis gains support from the facts that a similar process occurs during eosinophil activation (17) and that the intracellular chemiluminescence reaction can be selectively inhibited without alterations in either intracellular NADPH oxidase activity (as measured by an alternative, peroxidase-independent technique) or MPO activity (Karlsson A: unpublished observations). Ferguson et al report normal MPO activity as well as normal amounts and sizes of 4 NADPH components in the phagocytes from their subjects (1), and it is tempting to speculate that the decreased intracellular ROS production could be due to dysfunctional intracellular fusion processes. Ferguson et al launch the hypothesis that the SAPHO syndrome is a genetic disorder resulting in aberrant intracellular ROS production while the extracellular production is intact (i.e., that the 2 pools of NADPH oxidase are affected differently) (1). The intracellular signals that induce NADPH oxidase activation are far from defined. However, the differences between the patterns of extracellular and intracellular production of ROS induced by different stimuli indicate a diversity in the regulating pathways to the 2 pools of NADPH oxidase, and studies of dependencies for intracellular calcium fluxes between the 2 pools support such a scenario (18). Further, the use of signal transduction inhibitors has revealed that activation of the intracellular NADPH oxidase pool by phorbol myristate acetate (a stimulus used by Ferguson et al) is dependent on phosphatidylinositol 3-kinase (PI 3-kinase) activation, while activation of the plasma membrane–localized pool is not, and that different protein kinase C isozymes (upstream of PI 3-kinase) appear to take part in the 2 signal transduction pathways (15). Consequently, a molecular explanation for the SAPHO syndrome may hypothetically be sought among the signal transduction molecules responsible for activation of NADPH oxidase EDITORIAL 2935 in intracellular membranes or, as suggested above, among those leading to intracellular granule fusion. Conclusion The neutrophil respiratory burst has long been considered to be of main importance for the generation of ROS inside the phagosome, where it promotes the killing of microbes, or for the release of extracellular ROS mainly thought to contribute to tissue damage. However, intracellular ROS production often occurs in the absence of phagosome formation through a process that most likely involves fusion events of at least 2 types of distinct granules. In addition, ROS may act as essential signaling molecules with the capacity to affect a variety of cellular processes of importance for immune regulation. The scientific community is gradually opening up to the somewhat counterintuitive idea that ROS may actually curb inflammatory responses. The article by Ferguson et al (1) could be an important contribution to our understanding of the exaggerated inflammatory responses following diminished or absent ROS production. We hope that future research will determine exactly how (and where!) oxygen radicals dampen inflammatory processes, but clearly, the old view of ROS merely as antimicrobials and mediators of tissue damage is about to change. REFERENCES 1. Ferguson PJ, Lokuta MA, El-Shanti HI, Muhle L, Bing X, Huttenlocher A. Neutrophil dysfunction in a family with a SAPHO syndrome–like phenotype. Arthritis Rheum 2008;58:3264–9. 2. Assari T. Chronic granulomatous disease; fundamental stages in our understanding of CGD [review]. Med Immunol 2006;5:4. 3. Winkelstein JA, Marino MC, Johnston RB Jr, Boyle J, Curnutte J, Gallin JI, et al. Chronic granulomatous disease: report on a national registry of 368 patients. Medicine (Baltimore) 2000;79: 155–69. 4. Knight JA. Review: free radicals, antioxidants, and the immune system. Ann Clin Lab Sci 2000;30:145–58. 5. Gelderman KA, Hultqvist M, Olsson LM, Bauer K, Pizzolla A, Olofsson P, et al. Rheumatoid arthritis: the role of reactive oxygen species in disease development and therapeutic strategies. Antioxid Redox Signal 2007;9:1541–67. 6. Romani L, Fallarino F, De Luca A, Montagnoli C, D’Angelo C, Zelante T, et al. Defective tryptophan catabolism underlies inflammation in mouse chronic granulomatous disease. Nature 2008;451:211–5. 7. George-Chandy A, Nordstrom I, Nygren E, Jonsson IM, Postigo J, Collins LV, et al. Th17 development and autoimmune arthritis in the absence of reactive oxygen species. Eur J Immunol 2008;38: 1118–26. 8. Schappi M, Deffert C, Fiette L, Gavazzi G, Herrmann F, Belli D, et al. Branched fungal ␤-glucan causes hyperinflammation and necrosis in phagocyte NADPH oxidase-deficient mice. J Pathol 2008;214:434–44. 9. Lekstrom-Himes JA, Kuhns DB, Alvord WG, Gallin JI. Inhibition of human neutrophil IL-8 production by hydrogen peroxide and dysregulation in chronic granulomatous disease. J Immunol 2005; 174:411–7. 10. Bylund J, Macdonald KL, Brown KL, Mydel P, Collins LV, Hancock RE, et al. Enhanced inflammatory responses of chronic granulomatous disease leukocytes involve ROS-independent activation of NF-B. Eur J Immunol 2007;37:1087–96. 11. Liu H, Colavitti R, Rovira II, Finkel T. Redox-dependent transcriptional regulation. Circ Res 2005;97:967–74. 12. Bowie A, O’Neill LA. Oxidative stress and nuclear factor-B activation: a reassessment of the evidence in the light of recent discoveries. Biochem Pharmacol 2000;59:13–23. 13. Dahlgren C, Karlsson A, Bylund J. Measurement of respiratory burst products generated by professional phagocytes. Methods Mol Biol 2007;412:349–63. 14. Afonso V, Champy R, Mitrovic D, Collin P, Lomri A. Reactive oxygen species and superoxide dismutases: role in joint diseases. Joint Bone Spine 2007;74:324–9. 15. Karlsson A, Dahlgren C. Assembly and activation of the neutrophil NADPH oxidase in granule membranes. Antioxid Redox Signal 2002;4:49–60. 16. Sengelov H. Secretory vesicles of human neutrophils. Eur J Haematol Suppl 1996;58:1–24. 17. Hafez I, Stolpe A, Lindau M. Compound exocytosis and cumulative fusion in eosinophils. J Biol Chem 2003;278:44921–8. 18. Granfeldt D, Samuelsson M, Karlsson A. Capacitative Ca2⫹ influx and activation of the neutrophil respiratory burst: different regulation of plasma membrane- and granule-localized NADPHoxidase. J Leukoc Biol 2002;71:611–7.