Resolving the problem of persistence in the switch from acute to chronic inflammation
- *Pulmonary and Critical Care Medicine, Brigham and Women's Hospital and Harvard Medical School, 75 Francis Street, Boston, MA 02115; and
- †Rheumatology Research Group, Medical Research Council Centre for Immune Regulation, Institute for Biomedical Research, University of Birmingham, Vincent Drive, Edgbaston, Birmingham B15 2TT, United Kingdom
The rapid induction of an inflammatory response is crucial for effective host defense. A wide range of proinflammatory agents ensure that this response occurs. Prominent among these are members of the prostaglandin and leukotriene family that are enzymatically derived from arachidonic acid (AA, C20:4) (1). At the onset of inflammation, AA is metabolized via either cyclooxygenases (COX) to produce prostaglandins (PGs) and thromboxanes or lipoxygenases to generate leukotrienes (LTs) (1). These ecicosanoids induce and help establish the inflammatory infiltrate. However, at some point during the inflammatory cascade, inhibition and ultimately resolution of the inflammatory response must occur; otherwise, tissue homeostasis cannot occur and inflammation persists.
Until relatively recently, the resolution of inflammation was thought to occur mainly as a result of the passive withdrawal of proinflammatory mediators. It is now clear that the inflammatory program contains key checkpoints as well as temporal and spatial switches that regulate its onset, maintenance, and resolution (2). A key finding in recent years is that the biosynthetic pathways required for inflammation to resolve often require the prior production of lipid mediators produced during the induction phase of inflammation. In particular, PGE2 and PGD2, which are responsible for inflammation induction, subsequently stimulate antiinflammatory circuits by inducing the lipoxin-producing enzyme 15-lipoxygenase in neutrophils. This class switch in eicosanoid production, from LTs and PGs to lipoxins (3), provides temporal and spatial checkpoints during the inflammatory cascade (see Fig. 1). In this issue of PNAS, Rajakariar et al. (4) provide yet another twist in the resolution story with the demonstration that class switching may not always be required for resolution to be initiated. They provide compelling evidence that, within a single eicosanoid family pathway initiated by COX-2 and subsequently regulated by the enzyme hematopoietic PGD2 synthase (hPGD2S), proresolving mediators PGD2 and 15-deoxyΔ12–14 PGJ2 (15d-PGJ2) are produced at sufficient levels in vivo to drive the resolution of acute inflammation.
Temporal and spatial regulation of resolution. (A) Current paradigm: Breaches in host defense such as pathogen invasion, host trauma, or loss of barrier function initiate the inflammatory response with AA rapidly converted to PGs and LTs (e.g., LTB4) via COX enzymes. LTB4 is a potent chemoattractant of neutrophils to the site of inflammation (red arrow). As the inflammatory response progresses, a temporal switch from initiation to resolution occurs with the transcellular generation of lipoxins (LXA4) that inhibit the further recruitment of polymorphonuclear neutrophils (PMNs) (blue line). As the inflammatory response continues, monocytes are recruited via the release of the monocyte chemotactic protein-1. Macrophages release proinflammatory cytokines IL-6 and TNFα, but engulfment of apoptotic PMNs causes macrophages to change to a nonphlogistic phenotype releasing TGF-β, IL-10, and lipoxin. At the end of an acute inflammatory response, macrophages and lymphocytes leave the site of inflammation via the draining lymphatics, stimulated by lipoxins (blue arrow). (B) New addition to the paradigm: At the onset of the inflammatory response, PGD2 is generated from AA via COX to PGH2 and then converted to PGD2 by hPGD2S. PGD2 is metabolized nonenzymatically to 15-dPGJ2. PGD2 acts on the DP1 receptor expressed on macrophages and lymphocytes, whereas 15-dPGJ2 is likely to interact with peroxisome proliferator-activated receptor-γ. Working in tandem, they decrease proinflammatory cytokines such as TNFα and increase antiinflammatory cytokines such as IL-10, but like the lipoxin family, they also promote egress of macrophages and lymphocytes to the draining lymphatics (blue arrows).
The concept that antagonists that limit the extent of a biological cascade are generated as the cascade develops is familiar in other self-limiting pathways such the complement and coagulation cascades. However, homeostasis takes an unexpected twist in inflammation. Events at the onset of acute inflammation establish biosynthetic circuits for a series of chemical mediators that later serve not only as antagonists but also as agonists for an opposing program: resolution. These proresolution agonists don't just inhibit the cascade, they actively dismantle it. Therefore, antiinflammation and proresolution are not equivalent. The agonists that actively promote resolution (an emerging family of proresolving lipid mediators, including lipoxins, resolvins, and protectins) are inherently different from the antagonists that limit the extent of the inflammatory response, at both the molecular and the cellular level (2).
Ever since the discovery that aspirin exerts many of its antiinflammatory actions through inhibition of the COX enzymes (5), there has been keen interest in developing specific inhibitors of these enzymes. However, the logic for such an approach did not account for the fact that the inducible COX-2 enzyme is also responsible for the production of antiinflammatory eicosanoids, some of which are important in maintaining vascular homeostasis (6). In a provocative paper published in 1999, Gilroy et al. (7) suggested that COX-2 may have antiinflammatory properties. Subsequent findings by Fukunaga et al. (8) and Schwab et al. (9) have since demonstrated that blocking prostaglandin biosynthesis with COX-2 inhibitors, although effective at the very early stages of the inflammatory cascade, delays the onset of resolution (8) and/or impairs phagocyte removal of dead cells (9). In effect, these potent COX-2 selective agents lead to friendly fire by being “resolution toxic.”
In a series of seminal papers, Gilroy and colleagues have built on their original observations to present a convincing case that some of the downstream products from the COX-2-derived PGD2 family, and in particular 15d-PGJ2, are important in the resolution of acute inflammatory responses (9, 10). Critics of these studies have pointed out that a molecular mechanism that detailed how PGD2 leads to resolution was missing. More damning were the comments that the antibody-based methods used to measure these metabolites, and the negligible levels of 15d-PGJ2 found in various biological systems, made the biological importance of their original observations questionable (11). Using both genetic and biochemical studies, Rajakariar et al. (4) now provide compelling evidence, complete with a molecular mechanism that should silence the critics.
hPGD2S metabolizes COX-derived PGH2 to PGD2. However, the role of hPGD2S in host defense remains unclear because it and its metabolites appear to display both pro- and antiinflammatory properties, depending on the disease model and the way in which they are administered (12). This ambiguous role for hPGD2S is even more problematic because PGD2 is further metabolized nonenzymatically to produce PGs of the J series, including PGJ2 and 15d-PGJ2, whose relevance to pathophysiology has remained highly controversial. In an attempt to clarify the role of hPGD2S in inflammation, Rajakariar et al. (4) used mice deficient in the enzyme to unequivocally demonstrate that hPGD2S is responsible for the production of all PGD2 and 15d-PGJ2 during zymosan-induced resolving peritonitis. Using liquid chromatography–tandem MS (LC-MS/MS), they provide physical evidence for in vivo formation of 15d-PGJ2 in resolving exudates in a peritonitis model of inflammation at levels up to 5 ng/ml (≈15 nM). They noticed that macrophage emigration from the inflamed tissues to draining lymph nodes was delayed in hPGD2S knockout mice. This delay was rescued by addition of DP1-specific agonists, emphasizing the important role of DP1 in controlling both the onset of acute inflammation and its resolution. Surprisingly, lymphocyte numbers increased in the hPGD2S knockout mice at a time when resolution was at its peak in wild-type mice. Although the authors were unable to pinpoint why this increase occurred, they suggest that it was not due to changes in proliferation or cell death. Clearly these issues will need to be addressed in the future, particularly as pharmacological intervention with exogenous 15d-PGJ2 leads to accelerated leukocyte apoptosis, but at doses that are substantially higher than those formed naturally in the peritoneal exudates (13).
Although these results will excite those who work in the field of inflammation biology, and in particular its pathobiology and molecular pharmacology, they are of broad interest to investigators in many other fields, including host defense and cancer. For example, cyclopentone isoprostanes derived from PGs of the A and D series have been shown to possess antiviral and anticancer properties working through the activation of nuclear-membrane-bound peroxisome proliferator-activated receptors (12). However, the most exiting implication of these new findings is that they will encourage a shift in focus from the development of even more antiinflammatory, and potentially “resolution-toxic” drugs, toward “resolution-friendly” replacement therapies (14). Perhaps only then will the next generation of aspirin- and steroid-like compounds with the ability to both inhibit the induction of inflammation and jump-start its resolution emerge. It would appear that in the field of inflammation resolution, Cinderella has finally arrived at the ball.
Footnotes
- ‡To whom correspondence should be addressed. E-mail: c.d.buckley{at}bham.ac.uk
-
Author contributions: O.H. and C.D.B. wrote the paper.
-
The authors declare no conflict of interest.
-
See companion article on page 20979.
- © 2007 by The National Academy of Sciences of the USA
