Chemokines and leukocyte traffic
Federica Sallusto1 & Marco Baggiolini2
Institute for Research in Biomedicine, CH-6500 Bellinzona, Switzerland.
Swiss National Supercomputing Center, CH-6928 Manno, Switzerland.
Correspondence to: Federica Sallusto1 e-mail: sallusto@irb.unisi.ch
Abstract
Twenty years after the discovery of chemokines is an appropriate time to review leukocyte traffic and to assess the knowledge and opportunities that have arisen from countless studies of the large and tight-knit family of chemotactic proteins.
Introduction
"Chemokines and leukocyte traffic" was the title of a Nature progress paper in 1998 that reviewed the highlights of the first decade of research in this field1. In introducing the focus presented here, we would like to provide a reminder of the excitement about the discovery of chemotactic proteins that were originally thought to be selective for neutrophils, monocytes, eosinophils or basophils. The first years seemed like a 'gold rush', with several laboratories competing in a race for the discovery of new chemokines and new chemokine receptors. While the 'miners' were mostly busy with granulocytes, two fundamental observations were made about T lymphocytes. The first finding was that some chemokines inhibit the infection of CD4+ T cells by human immunodeficiency virus 1 (ref. 2); the second, more far-reaching finding was that T lymphocytes can express multiple chemokine receptors and that receptor expression and responsiveness to chemokines is regulated after activation or differentiation3, 4, 5. Chemokines suddenly became attractive to virologists and immunologists. The field expanded considerably, and fundamental questions about the trafficking of white cells that ensures immunity and can cause inflammation were answered in very rapid succession.
Chemokines provided a relevant complement to knowledge about the function of adhesion proteins in leukocyte extravasation and the seminal multistep paradigm based on the sequential use of selectins, chemokine receptors and integrins to induce tethering and rolling, firm adhesion and diapedesis6. The selective migration of naive and memory T lymphocytes was explained by the combined expression of selectins, integrins and chemokine receptors7. L-selectin and the chemokine receptor CCR7 were found to be essential for the migration of naive T cells to peripheral lymph nodes, whereas expression of the adhesion molecule CLA with CCR4 and of the integrin 47 with CCR9 confers on T cells the ability to migrate to skin and gut.
Mark Sokoloff
Artist's rendition of a chemokine receptor's seven-transmembrane structure.
figure
In the past decade, impressive progress has been made. The idea that different chemoattractant receptors are engaged consecutively to accomplish distinct migratory steps in tissues received a new 'twist' with the discovery that regulated expression of S1P1, the receptor for the lipid chemoattractant sphinogosine 1-phosphate present in serum and lymph, controls the exit of cells from thymus and lymph nodes8. That and other studies further broadened the spectrum of chemoattractants to include lipid mediators and even antibacterial peptides9. Studies of T lymphocytes and B lymphocytes showed that chemokines not only are messengers of innate immunity but also have 'housekeeping' functions: they are expressed constitutively at specific sites and in specialized compartments and regulate the homing, localization and survival of thymocytes and mature T lymphocytes, B lymphocytes and plasma cells and contribute to the generation and continuous renewal of lymphoid tissues.
Classical genetic approaches, such as knockout and transgenic experiments, have been instrumental in studies of the function of chemokines and their receptors at the level of the organism. Although in some cases such studies have identified considerable redundancy of the system, in others they have indicated unique functions for chemokine receptors in the immune system. For example, mice lacking CCR7 or its ligand CCL21 and mice lacking CXCR5 have defects in the entry and positioning of T lymphocytes and B lymphocytes in secondary lymphoid organs10, 11, 12. CCR7-deficient mice also have impaired migration of dendritic cells from tissues to lymph nodes. Genetic studies have also shown that the chemokine system is linked to a wider range of biological processes than first envisaged. The most notable examples were provided by the findings that the evolutionarily conserved chemokine receptors CXCR4 and CXCR7 and their ligand SDF-1 (now called CXCL12) have essential functions not only in hematopoiesis but also in organogenesis, vascularization and embryogenesis13, 14, 15.
Many animal studies have confirmed the involvement of various chemokine receptors or ligands in inflammatory disease models, from asthma to rheumatoid arthritis, sepsis and transplantation, which makes chemoattractant receptors promising targets of anti-inflammatory drugs16. In addition, drugs that target other classes of molecules involved in cell migration, such as integrins, have been successfully advancing through the clinic, the most notable example being the use of antibody to 4 integrin to treat multiple sclerosis17. The five state-of-the-art reviews in this focus provide a comprehensive update on progress in the field. Some of the topics are introduced below.
Seeing is believing
The development of intravital microscopy to visualize cell motility in a live organism has provided a new dimension in leukocyte trafficking18. The body of a mouse is becoming a 'Wunderkammer' in which the dynamic interactions of T cells with B cells and dendritic cells and the development of an immune response can be monitored in real time19. The new techniques of visualization not only have confirmed textbook ideas but also, more importantly, have shown unanticipated aspects of leukocyte activity.
Friedl and Weigelin provide a comprehensive overview of the use of imaging techniques to study leukocyte motility in tissues and the dynamics of cell-cell interation20. In intact lymph nodes, T cells and B cells migrate along the network of fibroblastic reticular cells using a fast amoeboid movement and establish many brief contacts with dendritic cells and follicular stromal cells. Antigen signaling prompts T cells and B cells to stop migrating and to establish strong adhesive cell contacts that may last from minutes to hours21. The duration of the immunological synapse between a T cell and an antigen-presenting dendritic cell is dependent on sustained intracellular signaling and affects the subsequent fate of the interacting cells. T cell priming results from stable contacts between T cells and mature immunostimulatory dendritic cells, whereas T cell deletion results from brief contacts of T cells with tolerogenic immature dendritic cells22, 23. Those in vivo observations support the signal-strength and serial-engagement models of T cell activation that were initially based on in vitro experiments24, 25. All cellular interactions after those events, such as those between T helper cells and cytotoxic T cells or between follicular T helper cells and B cells, involve chemokines and chemokine receptors26, 27. Such experiments led to the suggestion that activation-dependent changes in chemokine sensitivity ensure efficient cell cooperation.
Intravital microscopy may eventually help to elucidate the migration and retention of leukocytes in inflamed tissues after exposure to a variety of chemokines that can act as agonists, partial agonists or antagonists for receptors that are continually modulated. Furthermore, association with glycosaminoglycans28, trimming by proteases29 and possibly the formation of homodimers and heterodimers30 are all processes that may substantially change the potency, specificity and spectrum of activity of the chemokines.
Specialization at its best
T lymphocytes have the daunting task of patrolling almost every tissue of the body. They scan dendritic cells in lymphoid organs in search of specific antigens and, after being primed, they migrate to follicles to help B cells produce antibodies and to sites of antigen exposure to deliver the appropriate effector or regulatory function, thus inducing or dampening inflammation. Once the antigen is eliminated, central memory and effector memory T cells patrol lymphoid organs and peripheral nonlymphoid tissues to react promptly if a second attack occurs. Some T cells retain memory not only for the antigen but also for the site where they encountered it. This is true for memory T cells in the skin and the gut. In their reviews, Luster and colleagues31 and Sigmundsdottir and Butcher32 discuss how by orchestrating T cell traffic, chemokine receptors best serve this functional specialization; they also focus on advances in the understanding of how homing programs are 'imprinted' in T cells.
Chemokine receptors have been particularly useful in the identification of distinct subsets of effector and memory T lymphocytes with distinct migratory abilities and effector functions. Such studies, mainly of the human system, serve as the basis for the present distinction of central memory, follicular helper and effector memory T cells and for the ability to use surface markers to distinguish among T helper type 1 cells, T helper type 2 cells and interleukin 17–producing T helper cells33. It is now known that CCR6 can be used to identify the last cells, which represent the new lineage of cells called 'TH-17 cells', in blood and inflamed tissues34.
An emerging idea in the field is that T lymphocyte trafficking is tightly controlled in the steady state but becomes less restricted in conditions of inflammation. For example, the lymph node–homing receptors CD62L and CCR7 become dispensable for the entry of T lymphocytes into inflamed lymph nodes when high endothelial venules are induced to transiently express new types of chemokines and adhesion molecules, such as CXCL9 (ref. 35) and P-selectin (A. Martín-Fontecha and F.S., unpublished data). CCR7-negative effector memory T cells and natural killer cells, which are mostly excluded from lymph nodes in the steady state, migrate into inflamed lymph nodes, where they participate in the ongoing immune response, for example, by producing interferon- or killing antigen-presenting dendritic cells35, 36.
The constitutive migration of antigen-experienced T lymphocytes in the skin and small intestine is controlled by distinct sets of chemokines and adhesion molecules. The homing of T cells to the skin provides an example of how different receptors can be used sequentially in the process of extravasation and tissue positioning. The entry of circulating T cells is mediated by interaction of the adhesion molecules CLA, 41 and LFA-1 with vascular E-selectin, vascular cell adhesion molecule and intercellular adhesion molecule1, respectively. CCR4 (ref. 37) and possibly CCR8 (ref. 38) are also required in the transition from blood to dermis. In contrast, CCR10 is required for the targeting of T cells to the epidermis, where its ligand CCL27 is produced by keratinocytes. Similarly, CCR9 targets T cells to the small intestine at sites of CCL25 production by intestinal epithelial cells. This steady-state migration of lymphocytes serves an immune surveillance function and probably applies to all bodily tissues, including the central nervous system, which is considered an 'immune-privileged' site. However, the homing determinants for many tissues, such as the colon, lungs and brain, remain unknown.
An intriguing issue relates to the molecular mechanisms that regulate in T lymphocytes the constitutive and activation-dependent expression of homing receptors. T-bet, GATA-3 and RORt act as 'master transcription factors' to 'imprint' cytokine expression patterns on T helper type 1 cells, T helper type 2 cells and interleukin 17–producing T helper cells, respectively, and may be involved in regulating the expression of chemokine receptors as well. As discussed by Sigmundsdottir and Butcher32, the evolution of gut and skin has adopted external 'cues' from food (vitamin A) and sunlight (vitamin D3) to 'imprint' the homing of lymphocytes to the small intestine and to the epidermis, respectively. This 'education' is mediated by dendritic cells, which process these vitamins into the active 'cues' (retinoic acid and 1,25-dihydroxycholecalciferol) that act on nuclear hormone receptors for 'imprinting' lymphocyte migratory properties as a function of the environment.
'Insider information'
Chemokines act through rhodopsin-like, class A, seven-transmembrane-domain receptors coupled to heterotrimeric G proteins of the Gi type. Studies of structure-activity relationships have shown that chemokines interact with their receptors at two separate sites. A conformationally rigid domain adjacent to the two first conserved cysteine residues serves as recognition site and ensures the docking of the chemokine onto externally exposed loops of the heptahelical receptor structure. After docking, the conformationally disordered, short amino-terminal sequence preceding the first cysteine residue positions itself in the central binding pocket between the helices and initiates receptor activation and the signal-transduction cascade39. As shown for several CXC and CC chemokines, the amino-terminal sequence is critical; single amino acid deletions or substitutions cause loss of activity or turn agonists into antagonists (analogs that still bind but do not trigger a response)40. Such antagonists can block the responses of isolated leukocytes and inhibit inflammation in animal models41.
G proteins mediate the initial step of chemokine-induced signaling by splitting into the subunits G and G, which have distinct activities. G is particularly important, as it activates two signaling pathways. One of these involves phosphatidylinositol 3-OH kinase and protein kinase B, and the other involves phospholipase C, which generates two messenger molecules, inositol trisphosphate and diacylglycerol. The first mediates a transient increase in cytosolic free calcium, and the second activates protein kinase C. Signal transduction was first studied in human neutrophils; this showed that interleukin 8 (called 'NAF' then) elicits the same pattern of responses as known attractants, the complement fragment C5a and N-formyl-methionyl peptides. The responses to interleukin 8 were inhibited by Bordetella pertussis toxin, which inactivates Gi proteins, 17-hydroxywortmannin (later found to block phosphatidylinositol 3-OH kinase) and the protein kinase C inhibitor staurosporin42. Experiments with those and related reagents eventually showed that all chemokines act on the same class of receptors and initiate a similar signal-transduction cascade.
Thelen and Stein present an impressive overview of findings and suggestions on chemokine signaling43. The new results do not challenge early evidence about the mode of action of chemokines, in particular that chemokines are active as monomers binding to single receptors and that they all share a common mechanism of signal transduction with some selective requirements for the initiation of different functions. Of course, the signaling cascade is probably more complicated than that. The review presents a rich selection of papers reporting the formation of homodimers and heterodimers or oligomers by chemokines or by chemokine receptors, but the function of these phenomena remains speculative, as pointed out by the authors, for different aspects of leukocyte migration. Of particular interest are the discussions of the function of Rho and Ras GTPases, guanine-exchange factors and GTPase activation proteins in the context of cell polarization and migration, as the induction of such responses seems to rely on different mechanisms in lymphocytes and neutrophils.
The challenge
Since the early days of this field, chemokines have been viewed as promising targets for new anti-inflammatory drugs directed against the migration and activation of selected populations of inflammatory cells. The demonstration that chemokines act through G protein–coupled, seven-transmembrane-domain receptors was thus particularly helpful, as many synthetic low-molecular-weight inhibitors of this largest class of receptors have been successfully used in the treatment of a variety of diseases. Charles Mackay reviews new therapeutic strategies for inflammatory conditions, with particular attention paid to the development of low-molecular-weight antagonists for chemokine receptors44. Inflammation is a challenging target. It could be regarded as an area of pathology and medicine in which little progress has been made since its recognition millennia ago. The diagnosis is based on signs defined in ancient times, and therapy still relies mainly on aspirin-like drugs, despite the myriad possible mediators described. In his review, Mackay is optimistic about a turnaround, thanks to new drug candidates to treat inflammation and autoimmunity being developed as inhibitors of leukocyte migration. The clinical use of monoclonal antibodies to adhesion molecules and classical inflammatory cytokines interleukin 1 and tumor necrosis factor may be considered a first 'proof-of-concept'. The emphasis now, however, is on agents that block chemokine receptors, which are expected to be more selective than the monoclonal antibodies mentioned, and on preventing the migration and activation of selected subpopulations of leukocytes.
Studies of structure-activity relationships have provided guidance for the development of chemokine receptor–blocking agents. Such studies indicate that responses may be prevented by interference with the recognition step by which chemokines dock on the receptor or by the blockade of triggering in the central receptor-binding pocket. The docking site may be successfully occupied by monoclonal antibodies, whereas small synthetic compounds are more likely to compete successfully by blocking the triggering site. The effects of the substitution or deletion of single amino-terminal residues indicate that the triggering site of chemokine receptors is highly restricted. Lipophilic, highly diffusible organic compounds may thus be the structures of choice. It is nevertheless notable that 20 years after the recognition of chemokines and chemokine receptors as desirable drug targets, only a few synthetic receptor antagonists are ready for clinical application. This will change as promising chemical leads are identified.
Antibodies have also been considered for the inactivation of chemokines. However, as inflammatory leukocytes usually respond to multiple chemokines sharing the same receptor, such strategies seem less promising that those directed against receptors. Also less likely is interference with signal transduction, as the biochemical signaling process is similar for all chemokines, and analogies are also present in the whole class of rhodopsin-type receptors that, despite the large variety of antagonist structures and differences in receptor-binding mode, seem to share a common molecular activation mechanism45.
The future
Ideas about the function of chemokines in leukocyte traffic that were proposed after discovery of this class of chemotactic proteins have remained mostly valid. The main task for the future will continue to be searching for chemokine receptor antagonists in the hope of developing new therapeutic strategies for chronic inflammatory conditions. Some interesting questions remain unanswered, however. Few chemokines stand out in the large, uniform family of leukocyte attractants, and the study of their functions may provide new perspectives about the role of cell migration in biology.
The function of CXCL12 and its selective receptor CXCR4 is probably the 'hottest' issue in chemokine biology. This ligand and receptor are constitutively expressed in many conditions and in most tissues, and deletion of the gene encoding the chemokine or its receptor is lethal, which indicates they have vital developmental functions. CXCL1, the first chemokine to be identified (and called 'interleukin 8'), may also be considered atypical. It acts like a classical inflammatory chemokine, as it is induced in most tissues in pathological conditions, but it is released in an apparently constitutive way by exocrine glands. The interferon--regulated chemokines CXCL9, CXCL10 and CXCL11 are increasingly recognized as key mediators of inflammation, but they may also be involved in angiogenesis, tissue repair and new tissue generation46.
As highlighted in the reviews of this focus, there is still much to be learned about chemokines and leukocyte traffic. For example, the finding that CCR5 deficiency in humans increases the risk of symptomatic West Nile virus infection47 suggests that this system, rather than producing redundancy, may instead rely on a very large number of highly specialized chemokine-receptor pairs. Many of the present paradigms will probably be replaced as insights into the complexity of the system are gained.
Nenhum comentário:
Postar um comentário