Inflammation 2010: New Adventures of an Old Flame

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Leading Edge Overview Essay

Ruslan Medzhitov1,* 1Department of Immunobiology and Howard Hughes Medical Institute, Yale University School of Medicine, New Haven, CT 06520, USA *Correspondence: [email protected] DOI 10.1016/j.cell.2010.03.006

Inflammation is an essential immune response that enables survival during infection or injury and maintains tissue homeostasis under a variety of noxious conditions. Inflammation comes at the cost of a transient decline in tissue function, which can in turn contribute to the pathogenesis of diseases of altered homeostasis.

Inflammation has been known to human- cesses. Virchow’s main contribution to mechanisms of action. We now know kind for at least a few thousand years, in inflammation research was to establish that inflammation comes in many dif- part because it accompanied two major the cellular basis of pathology, a dra- ferent forms and modalities, which are scourges of the past, wounds and infec- matic departure from the traditional view governed by different mechanisms of tions, and in part because it is rather con- of disease as an imbalance of the four induction, regulation, and resolution. In spicuous. Although references to inflam- humors, which had dominated medicine the past few decades, the spectrum of mation can be found in ancient medical since the time of Hippocrates. prevailing inflammatory conditions has texts, apparently the first to define its Another major milestone was the dis- shifted from acute inflammatory reac- clinical symptoms was the Roman doc- covery of phagocytosis by Elie Metch- tions in response to wounds and infec- tor Cornelius Celsus in the 1st century nikoff and his theory of cellular immu- tions to chronic inflammatory states AD. These symptoms came to be known nity developed in 1892. Metchnikoff that accompany, for example, type 2 as the four cardinal signs of inflamma- emphasized the beneficial aspects of diabetes, atherosclerosis, asthma, neu- tion: rubor et tumor cum calore et dolore inflammation and pointed out the key rodegenerative diseases, and cancer. It (redness and swelling with heat and role of macrophages and microphages is as if surviving wounds and infections pain). Celsus mentions these signs in his (neutrophils) both in host defense and in early in life almost guarantees encoun- treatise De medicina, while describing the maintenance of tissue homeostasis tering chronic inflammatory diseases at procedures for treating chest pain, and (Tauber, 2003). We now appreciate the an advanced age. It is therefore hard to in so doing became an oft-quoted medi- importance of these concepts, as we overestimate the role of inflammatory cal celebrity (Majno, 1975). learn more about the biological func- processes in human health and disease. The physiological basis of the four tions of inflammation and the roles of And as we cannot completely escape cardinal signs of inflammation was macrophages in diseases of impaired from inflammation, we should at least try revealed much later by Augustus Waller homeostasis, such as obesity and ath- to understand it well enough to be able (1846) and Julius Cohnheim (1867), who erosclerosis. Meanwhile, Paul Ehrlich to avoid its more unpleasant aspects. discovered leukocyte emigration from was busy developing the humoral theory the blood vessels and other vascu- of immunity following the discovery of A Spectrum of Inflammatory lar changes characteristic of an acute serum therapy against diphtheria and Responses inflammatory response. Analyzing living tetanus toxins by Emil von Behring and A typical inflammatory response con- tissues under the microscope, Cohn- Shibasaburo Kitasato in 1890. The role sists of four components: inflammatory heim observed vasodilation, leakage of of serum components in immunity was inducers, the sensors that detect them, plasma, and migration of leukocytes out further supported by the discovery of the inflammatory mediators induced by of blood vessels and into the surround- complement by Jules Bordet in 1896. the sensors, and the target tissues that ing tissue (Majno and Joris, 2004). Finally, the establishment of the germ are affected by the inflammatory media- The fifth cardinal sign, functio laesa theory of disease in the late 19th century tors (Figure 1). Each component comes (disturbance of function), was added by Robert Koch and Louis Pasteur was in multiple forms and their combina- by Rudolph Virchow in 1858 in his book crucial for appreciating microbial agents tions function in distinct inflammatory Cellularpathologie (Majno, 1975). Nota- as major inducers of the acute inflamma- pathways. The type of pathway induced bly, although the four cardinal signs of tory response. under given conditions depends on Celsus only apply to acute inflammation Subsequent advances included the the nature of the inflammatory trigger. accompanying wounds and infections, identification of different classes of Thus, bacterial pathogens are detected functio laesa is the only universal sign inflammatory mediators, the pathways by receptors of the innate immune sys- that accompanies all inflammatory pro- that control their production, and their tem, such as Toll-like receptors (TLRs),

Cell 140, March 19, 2010 ©2010 Elsevier Inc. 771 Figure 1. Inflammatory Pathway Components The inflammatory pathway consists of inducers, sensors, mediators, and target tissues. Inducers initiate the inflammatory response and are detected by sensors. Sensors, such as Toll-like receptors (TLRs), are expressed on specialized sentinel cells, such as tissue-resident macrophages, dendritic cells, and mast cells. They induce the production of mediators, including cytokines, chemokines, bioactive amines, eicosanoids, and products of proteolytic cascades, such as bradykinin. These inflammatory mediators act on various target tissues to elicit changes in their functional states that optimize adaptation to the noxious condition (e.g., infection or tissue injury) associated with the particular inducers that elicited the inflammatory response. The specific components shown represent only a small sample of the myriad different sensors, mediators, and target tissues involved in the inflammatory response. which are expressed on tissue-resident I interferons (IFN-α, IFN-β) by infected 2009). Inflammation and nociception macrophages and induce the produc- cells and the activation of cytotoxic lym- are functionally linked at multiple levels: tion of inflammatory cytokines (e.g., TNF, phocytes, whereas infections with para- exudate formation, tissue swelling, and IL-1, IL-6) and chemokines (e.g., CCL2 sitic worms lead to the production of his- inflammatory mediators are responsible and CXCL8), as well as prostaglandins. tamine, IL-4, IL-5, and IL-13 by mast cells for “inflammatory pain,” and nocicep- These inflammatory mediators then and basophils. The latter response can tion complements inflammatory sen- act on target tissues, including local also be triggered by allergens, resulting sors in monitoring tissue homeostasis. blood vessels, to induce vasodilation, in allergic inflammation that affects pri- In addition, prostaglandins can lower the extravasation of neutrophils, and leak- marily the mucosal epithelium, smooth threshold of pain sensation by increasing age of plasma into the infected tissue. muscles, and vasculature. the sensitivity of nociceptors. Notably, Neutrophils recruited from the circula- In the case of sterile tissue injury in the sensing of the inflammatory milieu by the tion, tissue-resident macrophages, and absence of infection, acute inflammation vagus nerve triggers an “inflammatory mast cells seek and destroy invading promotes tissue repair and helps to pre- reflex,” which is involved in the negative pathogens. This process is aided by vent colonization of the damaged tissues control of inflammation (Tracey, 2002). plasma components, including antibod- by opportunistic pathogens. The molec- The acute inflammatory response is nor- ies and complement. In addition, IL-1, ular identities of the triggers and sensors mally terminated once the triggering insult TNF, and IL-6 can have systemic effects involved in the inflammatory response is eliminated, the infection is cleared, and when secreted in sufficient amounts. to tissue injury are incompletely under- damaged tissue is repaired. Termination They induce liver cells (hepatocytes) to stood, although molecules released of the inflammatory response and transi- produce acute phase proteins such as from dying cells, breakdown products tion to the homeostatic state is an active C-reactive protein and coagulation fac- of the extracellular matrix, and products and highly regulated process known as tors, and they activate brain endothelium of the proteolytic cascades activated the resolution of inflammation. Several key to produce prostaglandins, including the by vascular damage are thought to be regulatory mechanisms of resolution have major proinflammatory prostaglandin, involved. Tissue damage is detected been identified including the switch from PGE2. Locally produced PGE2, in turn, by both tissue-resident macrophages, proinflammatory prostaglandins to anti- induces specific populations of neurons which induce inflammatory and repara- inflammatory, resolution-inducing lipoxins. in the central nervous system to pro- tive responses, and by pain receptors This switch, in turn, orchestrates a transi- mote so-called sickness behavior: fever, (nociceptors) that enable pain sensation from neutrophil to monocyte recruit- anorexia, fatigue, sleepiness, and social tion in the affected area. Interestingly, ment that results in clearance of the dead withdrawal (Pecchi et al., 2009). both types of tissue damage sensors cells and other debris and initiation of tis- Depending on the type of infection can be activated by some of the same sue repair at the affected site (Serhan and (bacterial, viral, or parasitic), the sensors, signals that are produced upon injury, Savill, 2005). If the inflammatory trigger is mediators, and target tissues vary such for example, extracellular ATP released not eliminated by the acute inflammatory that the appropriate type of inflammatory from dying cells and bradykinin gener- response or persists for any other reason, response is induced. For example, viral ated by a proteolytic cascade induced the resolution phase may not be appro- infections induce the production of type- by vascular damage (Basbaum et al., priately induced and a chronic inflamma-

772 Cell 140, March 19, 2010 ©2010 Elsevier Inc. tory state may ensue. This state can be conditions. Both beneficial and detri- and, if persistent, may have detrimental caused by chronic infections, unrepaired mental aspects of inflammation can be consequences. At the level of the organ- tissue damage, persistent allergens, undi- explained from this perspective. To better ism, unfavorable environmental condi- gestable foreign particles, or endogenous understand inflammation and inflamma- tions, such as cold temperature, nutri- crystals, such as monosodium urate tory diseases, we need to consider how ent deprivation (caloric restriction), or (Kumar et al., 2003; Majno and Joris, 2004). inflammation is related to other adaptive dehydration, affect the switch between The chronic inflammatory response in processes that operate at different levels antagonistic physiological processes these cases is typically localized to the site in metazoan organisms. that promote either reproductive fitness where the inflammatory inducer is present The notion of homeostasis was first or somatic maintenance. This switch is and often results in different types of local proposed in 1865 by Claude Bernard, controlled by the IGF-1-FOXO pathway, tissue remodeling. For example, persis- who described the constancy of the with activation of the transcription factor tent infection can lead to the formation of internal environment as a “condition of FOXO leading to increased stress resis- granulomas and the generation of tertiary free and independent life.” The idea of tance. In some animals, this transition is lymphoid organs at the site of infection. homeostasis was further developed by taken to an extreme by triggering entry Similarly, persistent airway inflammation Metchnikoff. He viewed this process into a state of suspended animation induced by allergens can lead to respira- as being controlled and maintained by such as the dauer state in nematodes tory epithelial tissue remodeling resulting phagocytes (Metchnikoff, 1892) and or hibernation in certain mammals. in asthma. proposed the concept of “physiologi- However, this transition occurs at the In addition, a growing number of cal inflammation” to describe the role expense of normal functions, principally chronic inflammatory conditions have of phagocytes in the active mainte- reproduction. Similarly, the inflammatory been described where the initiating nance of “harmony” (i.e., homeostasis). response (excessive or not) invariably trigger is not well defined but does not These concepts are important because occurs at a temporary cost to normal tis- seem to involve infection or tissue dam- inflammation is essentially an adaptive sue function and therefore is universally age. These inflammatory conditions response that aims to restore homeosta- accompanied by functio laesa. are of particular interest because they sis. But if this is so, then why is inflam- The second reason for the detrimen- accompany many diseases of industri- mation associated with, and sometimes tal potential of inflammation concerns alized countries, including obesity and rightfully blamed for, so many diseases, a particular mode of adaptation to cer- type 2 diabetes, atherosclerosis, neu- particularly the very diseases that are tain types of persistent or extreme con- rodegenerative diseases, and cancer. caused by a loss of homeostasis? The ditions. Specifically, one can envision Interestingly, in these cases of chronic usual answer to this question is that two types of adaptation strategies: One inflammation there appear to be vicious inflammation is beneficial in appropriate promotes the restoration of homeosta- cycles connecting inflammation and the amounts but can easily become detri- sis by returning the regulated variables pathological process it accompanies. mental when excessive because of its to homeostatic set points. The second Thus, obesity can lead to inflammation, tissue-damaging potential. This is cer- strategy is to switch the homeostatic set whereas chronic inflammation can pro- tainly true, but at least two other reasons points to different values that are better mote obesity-associated diabetes in part may account for the pathological poten- suited to deal with the extreme or per- by inducing insulin resistance (Hotamis- tial of inflammatory processes. sistent abnormal conditions. The inflam- ligil, 2006). Similar positive feedback The first reason is the consequence matory response can engage in both loops are present in atherosclerosis, of a general feature of any adaptive modes of adaptation, although the sec- cancer, and other chronic inflammatory response to noxious conditions: it ond mode has been underappreciated. diseases. Indeed, this type of recipro- always occurs at the expense of normal Nevertheless, changes in homeostatic cal relationship may be responsible, at function. For example, cells deal with set points may be particularly important least in part, for the chronic nature of various types of noxious conditions by in chronic inflammatory diseases, such these inflammatory conditions and dis- inducing appropriate stress responses as obesity and type 2 diabetes. Indeed, tinguishes them from the first type of that ensure adaptation and survival in inflammation can induce a switch in chronic inflammation, which is caused the face of an abnormal cellular environ- metabolic homeostatic set points, for by persistence of the inflammatory ment. A number of dedicated sensors example, through its effect on insulin inducer. To understand the origin of have evolved to detect different stres- sensitivity, which may be intended for these inflammatory processes we need sors and to induce appropriate adaptive the reallocation of nutrients under condi- to take a broader view of inflammation responses. Thus, heat shock, hypoxia, tions of stress or infection. This intended and its relationship with other systems of high levels of reactive oxygen species, and beneficial mechanism can, however, homeostatic control. and glucose and amino acid deprivation become detrimental not because inflam- are sensed by HSF-1, HIF-1α, NRF-2, mation is excessive, but because the Inflammation: An Adaptive AMPK, and ATF4, respectively, result- changes in homeostatic set points can Response to Noxious Conditions ing in alterations in cellular physiology become maladaptive in, and perpetu- Despite the diversity of inflammatory that allow for adaptation to abnormal ated by, the “unnatural” environment of phenomena, fundamentally, inflamma- conditions. These adaptations occur at industrialized countries. In other words, tion is an adaptive response to noxious the expense of normal cellular functions the intended, beneficial role of this mode

Cell 140, March 19, 2010 ©2010 Elsevier Inc. 773 of adaptation can become have many trophic functions detrimental because of the that are essential for tissue mismatch between the cur- maintenance and homeostasis rent environment and the (Pollard, 2009), and presum- evolutionary pressures of the ably for stress adaptation as past (Gluckman and Hanson, well. Depending on the nature 2004). and extent of the problem, the tissue-resident macrophages Inflammation: A Tissue may recruit additional cells Stress Response (e.g., different types of mono- The analogy between inflam- cytes) to the affected tissue mation and stress responses even when the problem is is illustrative for an additional not associated with infection reason: local inflammation or injury. Thus, macrophages can be thought of as a tis- infiltrate the adipose tissue sue stress response, whereas of obese animals, a phenom- systemic inflammation is enon with well-appreciated clearly a specialized type of pathological consequences stress response that occurs at (insulin resistance) but an the level of the whole organ- unknown physiological pur- ism. In fact, in some cases it pose (Schenk et al., 2008). is not obvious where to draw The tissue stress response the line between the classical orchestrated by macrophages cellular stress response and and mast cells is intermediate inflammation. The main dis- between the cell-autonomous tinction may be that cellular stress response and the bona stress responses are largely fide inflammatory response cell-autonomous adaptations, and can be referred to as whereas inflammation typi- “parainflammation” (Medzhi- cally operates at the tissue or tov, 2008) (Figure 2). organismal level. However, the origin of the inflammatory Regulation of the response may become clearer Inflammatory Response if we consider that even The inflammatory response classical stress responses can be controlled at multiple have non-cell-autonomous levels (Nathan, 2002), but the components. For example, Figure 2. Inflammation and the Stress Response regulatory principles are still hypoxia is sensed by HIF-1α Inflammation is an adaptive response to noxious conditions. incompletely understood, in and induces cellular adapta- (A) Under normal conditions, tissue-resident macrophages maintain tissue part due to the complexity of homeostasis by removing dead cells and other debris and by producing tion to a shortage of oxygen, growth factors. the inflammatory response but it also results in produc- (B) Under noxious conditions, the cellular stress response is activated and re- and the multitude of compo- tion of vascular endothelial sults in a cell-autonomous adaptation. But it may also involve communication nents involved. One way to between stressed cells and other cells in the tissue environment, including growth factor, which induces resident macrophages. deconvolute this complex- tissue adaptation through (C) If the nature or the extent of stress is such that it affects not only individual ity is to distinguish among angiogenesis. Viral infections cells but the entire tissue (e.g., hypoxia or hyperthermia), then a tissue-level different checkpoints in the stress response, or parainflammation, is elicited by the resident macrophages induce cell-autonomous anti- (and in some tissues by mast cells). Depending on the degree of the nox- inflammatory response and viral responses but also lead ious condition, this response can involve the recruitment of different types to consider the different to production of IFN-β, which of inflammatory monocytes from the circulation. Parainflammation may also modes of action of regula- involve low-level release of plasma components into the affected tissue. induces tissue level adapta- (D) Finally, if the condition is severe enough (e.g., infection or injury), an acute tory signals. tion to the virus by inducing an inflammatory response ensues. This is characterized by the recruitment of In principle, the inflam- antiviral state in the surround- neutrophils and specialized subsets of monocytes from the circulation that matory response can be ing cells. Indeed most, if not help to protect the host from infection and promote tissue repair and resto- controlled at four levels, cor- ration of homeostasis. (Only a few examples of inflammatory mediators are all, cellular stress responses, shown.) responding to the four com- in addition to cell-autonomous ponents of the inflammatory adaptive changes, produce secreted fac- cialized sentinel cells that sense not only pathway: inducers, sensors, mediators that affect other cells in the tissue, injury and infection but also other types of tors, and target tissues. One key con- including resident macrophages. In fact, noxious conditions such as hypoxia and trol point, which is regulated by major tissue-resident macrophages may be spe - metabolic stress. Indeed, macrophages anti-inflammatory signals (e.g., IL-10,

774 Cell 140, March 19, 2010 ©2010 Elsevier Inc. TGF-β, glucocorticoids), is production microbial, tissue repair, and metabolic responsible for many chronic inflam- of inflammatory mediators. However, genes. These components may need matory states are not clearly defined, anti-inflammatory signals could also to be regulated independently of each although recent advances have impli- act on the target tissue itself. First, the other, which presumably explains why cated certain stress response compo- responsiveness of the target tissue to major anti-inflammatory signals act in a nents (Hotamisligil, 2006). Connections inflammatory mediators can be regu- component-specific manner. Mechanis- between cellular stress and inflamma- lated at the level of receptors and signal- tically, this type of control is commonly tion warrant further study. In particu- ing pathways activated by the mediators. performed at the level of gene tran- lar, little is known about the non-cell- This mode of regulation affords an extra scription. Elucidating the physiological autonomous aspects of cellular stress level of specificity by controlling which rationale underpinning this specificity of responses. Analyses of cell commu- tissues respond, as well as the extent regulation is an important challenge for nication during stress responses are and duration of the response, to a given future studies. likely to reveal new principles of tissue mediator. Notably, many inflammatory Another important aspect of the con- homeostasis and adaptation to noxious mediators, for example PGE2, can signal trol of the inflammatory response is the conditions. Although many molecules through multiple receptor subtypes that way in which anti-inflammatory signals have been implicated as inflammatory may have opposite effects on target tis- are produced. For example, IL-10 is inducers during tissue injury, their rela- sues. In other cases, the action of inflam- produced as a result of an inflamma- tive contributions are poorly defined and matory mediators can be dampened by tory response and is a component of an which of them, if any, are essential for the decoy receptors, again providing spe- activity-dependent negative-feedback induction of the inflammatory response cific control of target tissues. Switch- loop. IL-10 typically inhibits the response is unclear. ing the expression of different types of that initiated its own production and The relationship between inflamma- receptors may contribute to this mode acts on cells that produce inflammatory tory and adaptive immune responses is of regulation. Second, anti-inflammatory mediators, such as macrophages or T also incompletely understood. Although signals can act by reversing the effects cells. Although glucocorticoids can be inflammation is likely to be required of the mediators on the target tissue. For induced by inflammation (Besedovsky for the induction of adaptive immune example, noradrenaline can reverse the and del Rey, 2000), they also can be pro- responses, it is clearly not sufficient. It effects of histamine and bradykinin on duced in response to signals unrelated is important to note that inflammatory bronchiole smooth muscle by inducing to the inflammatory response (psycho- responses to infection and to sterile tis- bronchodilation. Vasoconstriction and logical stress, the circadian clock) and sue injury have different purposes: the vasodilation can also be controlled by can act on almost any cell, including former aims to protect the host from pro- and anti-inflammatory signals with- cells that are targets of the inflammatory infection and can be coupled with the out affecting the production of inflam- response. Thus, IL-10 and glucocorti- induction of adaptive immunity, whereas matory mediators. The advantage of this coids (and several nuclear receptors) act the latter primarily serves to promote type of control mechanism is that it can in a manner that is intrinsic or extrinsic to tissue repair. A detailed analysis of the selectively regulate responsiveness of the inflammatory pathway, respectively. inflammatory mediators induced under different target tissues without affect- Not surprisingly, IL-10 and glucocorti- the two conditions should help to eluci- ing the overall strength or duration of the coids have nonredundant roles in the date the distinct features of inflammation inflammatory response. control of inflammation, and it would be induced by infection and injury. Regulatory mechanisms are often useful to compare their distinct regula- An excessive inflammatory response specific for a particular component of tory functions under different inflamma- is detrimental due to its negative effect the inflammatory response. Thus, rather tory conditions. on tissue function and, when extreme, than shutting down the response com- results in overt tissue damage. Apart pletely, many anti-inflammatory signals Challenges and Perspectives from the acute phase of the inflamma- selectively inhibit certain aspects of the Inflammation comprises a diverse range tory response, however, other stages response. Examples include the effects of processes that affect every aspect of of the inflammatory process can also of IL-10 and glucocorticoids on TLR- normal physiology and pathology. The become dysregulated. A dysregulated induced inflammatory responses in field has become too large and amor- tissue repair response with accompa- macrophages. These anti-inflammatory phous to fit into a single discipline, let nying tissue remodeling, fibrosis, and signals inhibit expression of only 10% alone to be covered in a brief overview. persistent tissue metaplasia can lead to to 15% of TLR-inducible target genes. Although individual branches of inflam- the decline or complete loss of normal The same appears to be true for many mation research have proceeded at dif- tissue function as happens, for example, other anti-inflammatory signals. The ferent paces and in ever more divergent in asthma. One can imagine that dys- reason for this selectivity of regulation directions, there are still some common regulation of the resolution of inflamma- is that the response induced by inflam- principles that remain to be elucidated. tion may similarly contribute to tissue matory sensors, such as TLRs, typically For example, there is no satisfactory the- pathology. It would also be interesting has multiple functional components. ory that explains the chronic inflamma- to investigate whether pathogens that Thus, in addition to inflammatory medi- tory states associated with diseases of establish chronic infections, such as the ators, TLRs induce expression of anti- homeostasis. The inflammatory inducers bacterium Mycobacterium tuberculosis,

Cell 140, March 19, 2010 ©2010 Elsevier Inc. 775 may trigger a dysregulated resolution their coordinate functions in the context Majno, G. (1975). The Healing Hand - Man and response as part of their immune eva- of intact tissues and their multiple modes wound in the ancient world (Cambridge, MA: Har- vard University Press). sion strategy. of regulation are poorly defined. Inflammatory tissue damage and Inflammation research has come a Majno, G., and Joris, I. (2004). Cells, Tissues, and immunopathology incur high fitness long way from the first description of its Disease, Second Edition (Oxford: Oxford University Press). costs, and the fact that they have been cardinal signs by Celsus almost two mil- maintained during evolution may reflect lennia ago. Although much time has been Medzhitov, R. (2008). Nature 454, 428–435. a trade-off with the beneficial, often spent studying inflammation in patho- Metchnikoff, E. (1892). Бopьбa зa cyщecтвoвaниe life-saving, functions of the inflamma- logical contexts, increasingly, inflamma- чacтeй живoтнoгo opгaнизмa. In Пoмoщь tory response. The question remains, tion occupies a central position in many гoлoдaющим (Mocквa: Hayчнo-литepaтypный cбopник), pp. 321–326. however, of whether specialized mecha- branches of biology. This is reflected by nisms exist that have evolved to minimize the broad range of subjects discussed in Nathan, C. (2002). Nature 420, 846–852. inflammatory tissue damage. If so, these the Reviews and Essays in this Cell spe- Pecchi, E., Dallaporta, M., Jean, A., Thirion, S., and mechanisms may operate without affect- cial issue on inflammation. Troadec, J.D. (2009). Physiol. Behav. 97, 279–292. ing the protective aspects of inflamma- Pollard, J.W. (2009). Nat. Rev. Immunol. 9, 259–270. tion, for example, by controlling target References tissue responsiveness to inflammatory Schenk, S., Saberi, M., and Olefsky, J.M. (2008). J. mediators, or by increasing tissue resis- Basbaum, A.I., Bautista, D.M., Scherrer, G., and Ju- Clin. Invest. 118, 2992–3002. lius, D. (2009). Cell 139, 267–284. tance to inflammatory damage through Seixas, E., Gozzelino, R., Chora, A., Ferreira, A., induction of cytoprotective mechanisms Besedovsky, H.O., and del Rey, A. (2000). Z. Rheu- Silva, G., Larsen, R., Rebelo, S., Penido, C., Smith, matol. 59 (Suppl. 2), II/26–30. R.N., Coutinho, A., et al. (2009). Proc. Natl. Acad. (Seixas et al., 2009). Sci. USA 106, 15837–15842. One of the most daunting aspects of Gluckman, P.D., and Hanson, M.A. (2004). Science studying inflammation is the diversity and 305, 1733–1736. Serhan, C.N., and Savill, J. (2005). Nat. Immunol. 6, 1191–1197. complexity of the inflammatory media- Hotamisligil, G.S. (2006). Nature 444, 860–867. tors and their effects on target tissues. Tauber, A.I. (2003). Nat. Rev. Mol. Cell Biol. 4, Kumar, V., Cotran, R.S., and Robbins, S.L. (2003). 897–901. Although mechanistic aspects of their Robbins Basic Pathology, Seventh Edition (New effects are relatively well understood, York: Saunders). Tracey, K.J. (2002). Nature 420, 853–859.