Chemokines: Sirens of Neutrophil Recruitment—but Is It Just One Song?Braedon McDonald1 and Paul Kubes1, ,
Available online 26 August 2010.
|Refers to:||Lipid-Cytokine-Chemokine Cascade Drives Neutrophil Recruitment in a Murine Model of Inflammatory Arthritis|
Immunity, Volume 33, Issue 2, 27 August 2010, Pages 266-278,
Richard C. Chou, Nancy D. Kim, Christian D. Sadik, Edward Seung, Yinan Lan, Michael H. Byrne, Bodduluri Haribabu, Yoichiro Iwakura, Andrew D. Lu
Neutrophil trafficking to inflamed tissues requires the integration of multiple chemoattractant guidance signals. In this issue of Immunity, Chou et al. (2010) demonstrate that collaborative “cascades” of chemoattractant mediators control neutrophil recruitment to arthritic joints in mice.
Main TextChemoattractants represent an essential group of molecular guidance signals that choreograph the transit of leukocytes out of the mainstream of blood and into tissues at sites of inflammation. Neutrophils are uniquely sensitive to a vast array of chemoattractants including complement fragments (C5a, C3a), lipid mediators (LTB4, PAF), a multitude of chemokines (in mice, MIP-1α, MIP-1β, MIP-2, KC, and others), as well as exogenous mediators produced by pathogens (N-formylated peptides). The ability to respond to multiple chemoattractants with seemingly overlapping functions may represent a mechanism to ensure timely neutrophil recruitment during host defense. Alternatively, the apparent redundancy among chemoattractant signals may simply represent a default hypothesis resulting from a lack of evidence for nonredundant functions.A multitude of chemokines can direct neutrophil recruitment, including (in mice) ligands of the two main chemokine receptors expressed on neutrophils, CCR1 and CXCR2. Studies using deletion or blocking strategies to investigate the roles of CCR1 or CXCR2 in vivo have historically resulted in only partial inhibition of neutrophil recruitment, leading many to presume that these receptors (and their chemokine ligands) possess overlapping functions (Jacobs et al., 2010). However, such observations may also be explained by the alternate hypothesis that precise temporal and/or spatial control of chemoattractant production may yield unique roles for different chemokines and their receptors. Furthermore, signals from other mediators such as lipids and complement fragments must be integrated with chemokines to control neutrophil trafficking.
Chou et al. (2010) test the hypothesis that neutrophil recruitment into inflamed joints is controlled by temporally distinct cascades of chemoattractants, in which the response to one chemoattractant initiates pathways that lead to the expression of additional mediators, yielding nonredundant functions for each guidance signal (Chou et al., 2010). Previous investigations by the same group revealed that the lipid chemoattractant LTB4 and its receptor BLT1 are critical for neutrophil infiltration and disease pathology during autoantibody-induced arthritis (Kim et al., 2006). Interestingly, the LTB4-BLT1 axis was required to attract a first wave of neutrophils, but once these initial responders had infiltrated into the joint, further neutrophil recruitment proceeded independently of BLT1. In this issue, Chou et al. (2010) report that neutrophils recruited via BLT1 signaling early in arthritogenesis produce cytokines including IL-1β, which act on resident cells within the joint to induce the expression of chemokine ligands of CCR1 (including MIP-1α, MIP-1β, and RANTES), followed soon after by CXCR2-ligands (including KC, MIP-2, and LIX) (Figure 1).
Figure 1. Lipid-Cytokine-Chemokine Cascade Choreographs Neutrophil Recruitment into the Inflamed JointAfter transfer of autoantibody-containing serum from K/BxN transgenic mice into recipients, immune complexes deposit within joints and initiate an inflammatory response. Initially, production of the lipid chemoattractant LTB4 by synovial leukocytes (mast cells, macrophages, neutrophils) attracts neutrophils into the joint via signaling through BLT1 receptors. Infiltrating neutrophils release IL-1β, which stimulates resident cells (including endothelium, macrophages, and synoviocytes) to express chemokines that amplify and sustain neutrophil recruitment. Expression of CC-chemokines (including MIP-1α, MIP-1β, and RANTES) recruits neutrophils via CCR1 signaling during the initiation phase of arthritis. CXC-chemokines (including KC, MIP-2, and LIX) sustain neutrophil recruitment during the maintenance phase of disease via signaling though CXCR2.
Together with the temporally defined pattern of chemokine expression, the authors also found that CCR1 and CXCR2 mediated temporally distinct, nonredundant roles during arthritis development. By using CCR1-, CXCR2-, as well as doubly-deficient mice, it was found that CCR1 provided the guidance signals for early neutrophil recruitment, whereas CXCR2 promoted later and sustained infiltration. Previous studies have demonstrated that cytokine-mediated control of chemokine and receptor expression also serves to diversify guidance signals over time. During chronic inflammation, neutrophils begin to express new receptors and respond to chemokines normally thought to recruit monocytes (such as MCP-1) in response to cytokines such as IFN-γ ([Bonecchi et al., 1999] and [Johnston et al., 1999]). Findings such as these raise the interesting question of whether all neutrophils possess the ability to adapt to changing chemoattractant environments, or whether continued infiltration represents the progressive recruitment of phenotypically and functionally unique subsets of neutrophils. For example, BLT1-expressing neutrophils may be particularly adept at IL-1 production, positioning them as a unique sentinel population that stimulates the recruitment of further neutrophil subsets. Indeed, mice have been reported to possess multiple neutrophil subsets with unique surface receptor phenotypes and effector functions (Tsuda et al., 2004). Furthermore, the existence of a putative population of tissue-resident or perhaps continuously recirculating “pioneer” neutrophils (for which there is currently no direct evidence) may explain why many studies including this have struggled to understand how the very first neutrophils arrive at sites of inflammation.
The study by Chou et al. (2010) highlights the underappreciated complexity of chemical guidance signals that direct neutrophil trafficking and challenges our concepts of biological redundancy among inflammatory mediators. In addition to the temporally unique roles for BLT1, CCR1, CXCR2, and their ligands identified in this study, further layers of complexity exist among chemoattractants that are spatially separated in vivo. Leukocytes must traffic through multiple microenvironments en route to sites of inflammation, each presenting distinct gradients of chemoattractants expressed by different cell types along the way. When faced with opposing gradients of different chemoattractants, neutrophils preferentially migrate toward end-target chemoattractants (C5a, N-fomylated peptides) and away from intermediary chemoattractants (chemokines, LTB4) (Foxman et al., 1997). The ability to perform such cellular decision-making involves intracellular signal hierarchies that allow end-target chemoattractant signaling to dominate over other directional cues (Heit et al., 2008). Indeed, during autoantibody-induced arthritis, disruption of the neutrophils' ability to preferentially migrate toward end-target chemoattractants by deletion of PTEN within neutrophils attenuated recruitment into joints as well as clinical disease (Heit et al., 2008).
The existence of a positive-feedback loop involving lipid-cytokine-chemokine cascades that drive neutrophil recruitment and development of arthritis raises a serious problem for the host: how can this self-perpetuating process be turned off? A key avenue for future research will involve discovering how these mediator cascades are modulated and/or downregulated to initiate resolution of inflammation, or alternatively, how dysregulation of these cascades results in sustained inflammation and disease. Events such as lipid mediator “class switching” (describing the gradual reduction of proinflammatory prostaglandins and leukotrienes in favor of proresolution mediators such as resolvins and protectins) may stimulate downstream “cascades” of cytokines and chemokines that actually reduce neutrophil infiltration and contribute to resolution of inflammation (Serhan et al., 2008). Furthermore, understanding how inappropriate or overexuberant chemoattractant cascades contribute to disease pathogenesis may reveal targets for the development of anti-inflammatory therapies. The findings by Chou et al. (2010) suggest that therapies aimed at chemokines or their receptors present the challenge of trying to hit a moving target, whereas blockade of factors upstream in the lipid-cytokine-chemokine cascade may be of greater benefit. This evidence may explain, in part, the effectiveness of anticytokine therapies in human rheumatoid arthritis.
ReferencesBonecchi et al., 1999 R. Bonecchi, N. Polentarutti, W. Luini, A. Borsatti, S. Bernasconi, M. Locati, C. Power, A. Proudfoot, T.N. Wells and C. Mackay et al., J. Immunol. 162 (1999), pp. 474–479. View Record in Scopus | Cited By in Scopus (124)
Chou et al., 2010 R.C. Chou, N.D. Kim, C.D. Sadik, E. Seung, Y. Lan, M.H. Byrne, B. Haribabu, Y. Iwakura and A.D. Luster, Immunity 33 (2010), pp. 266–278 this issue. Article | PDF (1766 K)
Foxman et al., 1997 E.F. Foxman, J.J. Campbell and E.C. Butcher, J. Cell Biol. 139 (1997), pp. 1349–1360. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (260)
Heit et al., 2008 B. Heit, S.M. Robbins, C.M. Downey, Z. Guan, P. Colarusso, B.J. Miller, F.R. Jirik and P. Kubes, Nat. Immunol. 9 (2008), pp. 743–752. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (24)
Jacobs et al., 2010 J.P. Jacobs, A. Ortiz-Lopez, J.J. Campbell, C.J. Gerard, D. Mathis and C. Benoist, Arthritis Rheum. 62 (2010), pp. 1921–1932. View Record in Scopus | Cited By in Scopus (0)
Johnston et al., 1999 B. Johnston, A.R. Burns, M. Suematsu, T.B. Issekutz, R.C. Woodman and P. Kubes, J. Clin. Invest. 103 (1999), pp. 1269–1276. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (96)
Kim et al., 2006 N.D. Kim, R.C. Chou, E. Seung, A.M. Tager and A.D. Luster, J. Exp. Med. 203 (2006), pp. 829–835. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (52)
Serhan et al., 2008 C.N. Serhan, N. Chiang and T.E. Van Dyke, Nat. Rev. Immunol. 8 (2008), pp. 349–361. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (158)
Tsuda et al., 2004 Y. Tsuda, H. Takahashi, M. Kobayashi, T. Hanafusa, D.N. Herndon and F. Suzuki, Immunity 21 (2004), pp. 215–226. Article | PDF (427 K) | View Record in Scopus | Cited By in Scopus (76)