Local coordination verses systemic disregulation: complexities in leukocyte recruitment revealed by local and systemic activation of TLR4 in vivo Steven M. Kerfoot and Paul Kubes1 Immunology Research Group, Department of Physiology and Biophysics, University of Calgary, Alberta, Canada
Abstract: The recruitment of leukocytes to a tissue is a critical step in the inflammatory response. Toll-like receptor 4 (TLR4) is an important receptor involved in the initiation of inflammatory responses. Administration of the ligand for TLR4, lipopolysaccharide, is often used to model inflammation—local responses to stimuli within a specific tissue and systemic responses such as those observed during endotoxic or septic shock. Here, we review work, which demonstrates that in response to local activation of TLR4, highly coordinated and multistep processes are initiated, ultimately resulting in the leukocyte’s arrival at the inflamed tissue. In contrast, systemic activation of TLR4 results in nonspecific accumulation of leukocytes within the lung capillaries and liver sinusoids through mechanisms profoundly different than those involved in local tissue recruitment. Contrary to current dogma, leukocyte accumulation in the lung is dependent on endothelial rather than leukocyte activation. Finally, we discuss recent evidence suggesting that activation of leukocytes through TLR4, although still in the circulation, effectively paralyzes inflammatory cells, rendering them incapable of appropriate trafficking to inflamed tissues. J. Leukoc. Biol. 77: 862– 867; 2005. Key Words: sepsis 䡠 LPS
INTRODUCTION The series of coordinated events collectively termed “inflammation” is the body’s way of dealing with a broad list of traumatic events including infection by various pathogens. Inflammation is essential to control these infections, contributing directly to clearance of pathogen during early innate responses to the infection and later, as an effector arm of the adaptive response. Despite this, inflammation has gained a rather poor reputation as a result of its pathological involvement in a variety of disorders, including stroke or other ischemic episodes, and autoimmune diseases, such as multiple sclerosis or rheumatoid arthritis. Relatively speaking, however, these are rare occurrences. Rather, the primary role of inflammation is in clearing bacteria that have breached skin or mucosal barriers. In fact, a well-controlled and coordinated 862
Journal of Leukocyte Biology Volume 77, June 2005
inflammatory response is responsible for limiting the proliferation and spread of infection, thereby preventing these relatively minor situations from escalating to catastrophic and potentially fatal events. The flip-side is that if infection is not contained but becomes systemic, the resulting uncontrolled, global activation of inflammatory processes produces a syndrome known as “septic shock” [1, 2], which can itself be lethal over and above the danger posed by the original infection. For a number of years, our laboratory has been interested in the role of Toll-like receptor 4 (TLR4) in the initiation of inflammatory responses, in particular, the differential consequences of local versus systemic activation. Here, we will survey our findings and those by others, which demonstrate collectively that although localized activation of TLR4 produces a highly controlled and coordinated response, systemic activation literally “paralyzes” inflammatory cells, effectively rendering them incapable of responding to pathogens.
MODELING TLR4-INDUCED INFLAMMATION WITH LIPOPOLYSACCHARIDE (LPS) TLR4 is a transmembrane protein expressed by numerous immune and nonimmune cells [3, 4]. Activation of TLR4 produces powerful inflammatory responses by inducing the production of inflammatory mediators including cytokines, chemokines, and others [3, 5, 6]. Although a number of activators of TLR4 have been identified, its primary ligand is LPS, which is a major component of gram-negative bacterial cell walls and as such, is recognized by the immune system (via TLR4) as an indicator of infection [3, 4]. The dependence on TLR4 to mediate inflammatory responses to LPS in vivo has been demonstrated clearly in three different TLR4-deficient strains of mice. Two of these strains, C3H/HeJ and C57BL/ 10ScCR, contain a naturally occurring point mutation (C3H/ HeJ) or deletion (C57BL/10ScCR) of the TLR4 gene and have been used extensively in studies of LPS-induced inflammation and sepsis. Both are unresponsive to LPS administration in vivo [7, 8]. A third strain, in which the TLR4 gene has been
1
Correspondence: Immunology Research Group, Department of Physiology and Biophysics, University of Calgary, 3330 Hospital Drive N.W., Calgary, Alberta, Canada T2N 4N1. E-mail:
[email protected] Received October 22, 2004; revised February 7, 2005; accepted February 13, 2005; doi: 10.1189/jlb.1004607.
0741-5400/05/0077-862 © Society for Leukocyte Biology
deleted by homologous recombination, has an identical phenotype [9]. For the above reasons, LPS administration alone is commonly used to model basic inflammatory mechanisms. Indeed, activation of TLR4 through administration of LPS to a particular tissue produces a typical localized inflammation within that tissue [10, 11]. In contrast, systemic activation of TLR4 through intraperitoneal or intravenous (i.p. or i.v., respectively) administration of LPS produces a shock syndrome (endotoxic shock) [12] similar to that observed during systemic bacterial infection (septic shock) [1, 2]. Indeed, endotoxin is often found in the blood of septic patients [13, 14]. However, LPS administration, locally or systemically, should not be mistaken for sepsis or septic shock resulting from bacterial infection. Bacteria-induced shock involves the activation of multiple inflammatory pathways in addition to TLR4. In fact, use of whole bacteria makes it impossible to make conclusions about the role of TLR4 in vivo. Nevertheless, local and systemic LPS have been used to mimic aspects of sepsis, including activation of TLR4, increased leukocyte/endothelial interactions locally, and neutrophil trapping in lungs and liver when LPS is given systemically (see below). It is important to state that this review is primarily focused on the role TLR4 plays in vivo. Therefore, LPS is used as a selective activating agent of the TLR4 pathway. Central to any inflammatory response is the recruitment of leukocytes from the circulation (see refs. [15–18] for in-depth reviews of the mechanisms of leukocyte recruitment). This process is ultimately dependent on the activation of the endothelial cells, which line blood vessels, to express the adhesion molecules and other mediators required to facilitate interactions with blood-borne leukocytes. Circulating leukocytes can then tether to and roll along the vascular wall. Both of these events are predominantly mediated by selectins (P- and Eselectin in nonlymphoid tissues), although ␣4-integrins may also mediate these interactions in some circumstances. If rolling leukocytes encounter appropriate stimulatory molecules— chemokines, for example—adhesion is rapidly induced through the activation of ␣4- or CD18-integrins. Once adherent, leukocytes can become motile and actively transmigrate across the endothelial monolayer and enter the tissue. Transmigration and the subsequent directional migration of cells to the site of infection occur in response to various chemotactic agents recognized by leukocytes [18]. This well-organized cascade of events was worked out and described initially in isolated in vitro systems mimicking vasculature conditions [19 –21]. Each step has been considered to be a prerequisite for the next—the implication being that interference with any one step should prevent a leukocyte from reaching its target. As we will outline here, subsequent investigations in vivo have challenged this somewhat over-simplified and linear view. It is ironic that the move to in vivo assays of leukocyte recruitment required a step back in time. In the mid-1800s, Cohnheim [22] made the original observations of trafficking leukocytes within blood vessels of a frog’s tongue that had been stretched out thinly to allow for transillumination of the tissue. In more recent times, we and others have adapted this same assay to observe leukocyte trafficking in other animal models, primarily mice. Like the frog tongue, thin tissues such
as the cremaster muscle and mesentery were initially used in intravital microscopy. Early experiments largely verified the in vitro model of a linear, multi-step cascade dependent on selectins and integrins. It was not until techniques were developed permitting the visualization of thicker tissues that complexities of tissue-specific and stimulus-specific inflammatory mechanisms were revealed. Over the course of the last decade, our laboratory has applied these technologies to understand inflammatory mechanisms induced by local or systemic activation of TLR4.
LEUKOCYTE RECRUITMENT IN RESPONSE TO TLR4 LOCALIZED ACTIVATION Initially, we used intravital microscopy of the cremaster microcirculation to study localized inflammation initiated through TLR4. Under baseline conditions, there was some leukocyte rolling but no adhesion or emigration into the tissue. Upon application of LPS to the tissue, rapid changes in the interactions between leukocytes and the endothelium were observed [23, 24]. Within 15 min, the number of rolling cells increased, and the velocity at which they rolled decreased—a sign of activation. Leukocyte adhesion was also induced, and subsequently, leukocytes could be observed within the tissue. Activation of these processes did not occur in TLR4-deficient mice, confirming its central role in inducing inflammation in this model [10]. No systemic hemodynamic or humoral changes were observed [10, 11]. The rapid nature of the response to local stimulation of TLR4 suggested the involvement of P-selectin. In most tissues, P-selectin is constitutively synthesized by the endothelium and stored intracellularly in specialized Weibel-Palade bodies [16]. It can be mobilized to the cell surface within minutes in response to inflammatory stimuli such as histamine, oxidants, and LPS [16]. Indeed, administration of a P-selectin-blocking antibody completely prevented LPS-induced leukocyte-endothelial interactions for the first 90 min [23]. This is consistent with the linear, multi-step model of leukocyte recruitment described above; the blockade of upstream rolling prevented the recruitment of leukocytes to the inflamed tissue. However, after 90 min, a second wave of recruitment that could not be blocked by anti-P-selectin was initiated, suggesting the involvement of additional adhesion molecules [23]. Indeed, observations made 4 h following LPS administration in P-selectin-deficient mice revealed that numbers of adherent cells had been fully restored, despite the reduction in rolling compared with wild-type mice (our unpublished observations). In fact, adhesion was only prevented in mice deficient in both endothelial selectins (E-selectin and P-selectin) in which rolling was almost completely eliminated [25]. Although on the surface, these in vivo observations are largely consistent with the accepted model outlined above, they also reveal the first complexity from the simplified cascade. The relationship between the number of rolling cells and the number of adherent cells is not linear. Indeed, rolling must be blocked by at least 90% to have an impact on adhesion [25, 26]. Similar to the sequential cascade of events mediating leukocyte-endothelial interactions described above, recent studies Kerfoot and Kubes TLR4-induced leukocyte recruitment
863
suggest that leukocyte transmigration and migration through the tissue during inflammation are also a multi-step processes. Once adherent, leukocytes can become motile and then follow chemotactic stimuli out of the vessel lumen into the tissue. Activation through TLR4 induces the local production of chemotactic agents, including chemokines and others (plateletactivating factor, leukotriene B4) by tissue cells and inflammatory cells within the tissue. The result is the accumulation of large numbers of inflammatory cells within a tissue following administration of LPS [11, 23]. However, over the course of this process, migrating leukocytes are exposed to a series of different chemotactic agents generated from a number of sources. If bacteria were present in the inflamed tissue, chemotactic bacterial products [formyl-Met-Leu-Phe (fMLP), for example] or other agents such as C5a (generated through the classical or alternative pathways in response to bacterial recognition) would also be present in the tissue. In the face of numerous, often conflicting chemotactic gradients, the ultimate goal of an infiltrating leukocyte would be to target the infectious agent. To reach the final target, leukocytes must somehow differentiate these different agents and prioritize their movements accordingly. A hierarchy for chemotactic agents was described originally by Campbell et al. [27], who demonstrated that migration to a chemokine (interleukin-8) was inhibited by the bacterial product fMLP but that the reverse did not occur. Using an in vitro assay of neutrophil migration, we recently characterized this hierarchy further and showed that given a choice and irrespective of the concentration of either agent, migrating neutrophils will always move toward an agent generated at the source of the infection (for example, fMLP or C5a, termed an “end-target chemoattractant”) in preference to other agents (for example, a chemokine or leukotriene, termed an “intermediary chemoattractant”) [28]. This preference was explained by the intracellular signaling pathways responsible for migration to the different classes of chemotactic agent. Migration toward “intermediary” agents, such as chemokines or leukotrienes, was dependent on signaling through the phosphatidylinositol-3 kinase (PI-3K) pathway. In contrast, migration to end-target agents, such as fMLP or C5a, was dependent on signaling through the p38 mitogen-activated protein kinase (MAPK) pathway, which inhibited signaling through the PI-3K pathway, resulting in preferential migration of neutrophils toward the end-target chemotactic agent [28]. We hypothesize that this would allow migrating neutrophils to leave an area with a high concentration of intermediary chemoattractants (adjacent to the endothelium, for example) and follow a weaker gradient of end-target agents to the site of infection. By combining these studies, a basic model of localized inflammation induced through TLR4 can be outlined (Fig. 1). Stimulation of TLR4 on endothelium induces rapid adhesion molecule expression. This results in increased leukocyte rolling and adhesion. In addition, other TLR4-expressing cells, such as tissue resident macrophages, are activated to produce additional inflammatory agents such as cytokines and chemokines, which also act on the endothelium and migrating cells. Leukocyte adhesion and initial transmigration are driven by intermediary chemotactic agents produced by the endothelium and tissue-resident cells. In a more complex system in which 864
Journal of Leukocyte Biology Volume 77, June 2005
Fig. 1. Leukocyte recruitment to a localized site of inflammation occurs through highly organized and coordinated mechanisms. In this model, inflammation is initiated when endothelial cells or tissue resident macrophages (M) are activated through TLR4 by local LPS administration. Endothelial cells are activated to express adhesion molecules resulting in the capture of circulating leukocytes from the blood flow. These leukocytes roll along the vascular wall until they encounter an appropriate chemotactic stimulus, often a chemokine, produced by the endothelium or tissue-resident macrophage. The leukocyte then transmigrates across the endothelial monolayer and into the surrounding tissue in response to the chemotactic agent. If bacteria are present in the inflamed site (for example, gram-negative bacteria expressing LPS), end-target chemotactic agents will be generated. Once the migrating leukocyte encounters an end-target chemotactic agent, migration toward the original, intermediary agent is inhibited, and the leukocyte migrates to the source of the bacterial infection.
bacteria are also present in the tissue (gram-negative bacteria, for example), an infiltrating leukocyte would then preferentially migrate toward end-target chemotactic agents generated at the site of infection and ignore the intermediary agents. Through these highly organized processes, circulating leukocytes are able to find their way to a site of localized inflammation and ultimately, to the site of a localized infection so that the pathogen can be cleared effectively.
THE EFFECTS OF SYSTEMIC TLR4 ACTIVATION ON LEUKOCYTE RECRUITMENT An often-fatal endotoxic shock syndrome results from the systemic activation of TLR4 through systemic administration of LPS. The central role of TLR4 in endotoxic shock is clearly demonstrated by the strains of deficient mice, which are protected from shock-related symptoms and death [3, 7–9]. During endotoxic shock, massive numbers of neutrophils and other leukocytes accumulate in the lung and the liver [2, 29 –32], a process entirely dependent on TLR4 [10, 33]. Similar TLR4mediated mechanisms would also be induced during septic shock induced by gram-negative bacteria. Indeed, leukocyte accumulation in the lung and liver is also observed in human septic patients. Damage to the liver and especially the lung caused by these cells is a major cause of pathology and death in sepsis [2, 34, 35]. For this reason, it is important to understand TLR4-mediated mechanisms responsible for leukocyte http://www.jleukbio.org
accumulation in these tissues, with the hope that accumulation and subsequent injury could be prevented. The liver vasculature has a very different architecture than that of peripheral tissues. Almost the entire tissue is devoted to a highly organized and densely packed sinusoidal system. These are small, capillary-like vessels, and leukocytes must traverse though these tributaries in single file. The sinusoids are fed by the portal vein and empty into collecting venules for return to the vena cava. Intravital microscopy was used to investigate the mechanisms of leukocyte accumulation in the liver following systemic activation of TLR4 via i.p. administration of LPS. Under baseline conditions, leukocytes moved smoothly through the sinusoidal network. Little leukocyte rolling was observed in the collecting venules. Reminiscent of the recruitment observed in the cremaster microvasculature following localized TLR4 activation, systemic activations of TLR4 resulted in significant adhesion in the collecting venules of the liver [25]. Indeed, as in the cremaster muscle, adhesion in the collecting venules was prevented in E/P-selectin, double-deficient mice. These findings are entirely consistent with the standard leukocyte recruitment cascade described above. However, in addition to the adhesion observed in the venules, a tremendous amount of leukocyte adhesion also occurred within the sinusoids [25]. Cells did not roll but immediately adhered. This adhesion was not prevented in selectin-deficient mice, and the deletion or blockade of any other standard adhesion molecule could not reduce the accumulation of leukocytes within the sinusoids (ref. [25] and our unpublished observations). This revealed a second complexity in leukocyte recruitment: The mechanisms and molecular requirements for recruitment can differ dramatically depending on the specific tissue and vascular architecture (reviewed in ref. [36]). Similar to the liver, the lung is highly vascular and contains a dense capillary network surrounding the alveoli. Again, systemic activation of TLR4 results in tremendous trapping of leukocytes within lung capillaries [2, 33, 34]. Because of movement during respiration, it is difficult of perform intravital microscopy studies on the lung. Our preliminary studies of this tissue revealed that under baseline conditions, some leukocytes do adhere within the lung. However, in the inflamed lung, tremendous leukocyte accumulation can be observed within capillaries and venules (our unpublished observations; Fig. 2). Because of the technical difficulties associated with performing intravital microscopy of the lung, most in vivo studies of leukocyte recruitment to the tissue have relied on assays for the neutrophil-specific enzyme myeloperoxidase (MPO). In response to systemic LPS activation of TLR4, MPO levels within the lung increase two- to threefold [10, 11, 33]. As in the liver, this increase is completely independent of selectins or other standard adhesion molecules [33, 37, 38]. A leading hypothesis to explain the accumulation of leukocytes in the lung and liver following systemic administration of LPS is that the membrane of a circulating leukocyte stiffens when it is activated through TLR4 [32]. As mentioned above, lung capillaries and liver sinusoids have narrow diameters, and even under normal conditions, leukocytes have to squeeze through one at a time. It is thought that by virtue of their less-pliable membranes, activated cells are not able to squeeze
Fig. 2. Intravital microscopy of the lung showing leukocyte accumulate following systemic activation of TLR4 through administration of LPS. (A) Light microscopy of the lung microvasculature. The contour of a post-capillary venule is outlined. The asterisk marks an alveolus. Circulating leukocytes were labeled fluorescently by i.v. administration of rhodamine 6G. Fluorescence microscopy was then used to observe labeled leukocytes in the lung microvasculature. In untreated, control animals (B), some leukocyte rolling (arrow) but little adhesion was observed. Following i.p. administration of LPS, large numbers of leukocytes rolled and adhered in postcapillary venules (C; open arrows). In addition, many leukocytes were trapped in capillaries (closed arrows).
through the narrow vessels and become physically trapped. This process would completely bypass the typical recruitment cascade and the requirement for specialized, adhesion molecule-dependent interactions. We made use of TLR4-deficient mice to begin to explore the “trapping” phenomenon more closely. Our recent studies have challenged this widely held view. LPS-induced neutrophil trapping in the lung is completely dependent on TLR4 in that no increase in MPO is observed in TLR4-deficient mice [10, 33]. We generated bone-marrow chimeric mice in which all circulating (and tissue resident) leukocytes were deficient in TLR4, and tissue cells such as the endothelium expressed TLR4 or the reverse, in which leukocytes were wild-type, and endothelium was TLR4-deficient [33]. If leukocyte trapping in the lung is the result of nonpliable, activated leukocytes in the circulation, one would expect that TLR4-expressing leukocytes would accumulate in the lung even in the absence of endothelial TLR4. In fact, we observed precisely the opposite; leukocyte trapping in the lungs was observed only if the endothelium expressed TLR4, even if leukocytes were TLR4-deficient [33]. This suggests an active, endothelium-driven mechanism of leukocyte trapping in the lung rather than a passive, physical trapping event. The process is clearly different from the mechanisms normally associated with leukocyte recruitment, as Kerfoot and Kubes TLR4-induced leukocyte recruitment
865
rolling does not occur nor are the standard adhesion molecules required. We are currently investigating these mechanisms and have recently identified a novel adhesion molecule, vascular adhesion protein-1 (VAP-1) [39], which mediates lymphocyte adhesion within liver sinusoids (Claudine S. Bonder, Mark G. Swain, Lori D. Zbytnuik, Sirpa Jalkaner, and Paul Kubes, submitted). The mechanism by which neutrophils are recruited into the sinusoids is not VAP-1-dependent and remains unknown. It has yet to be determined if the lung uses a similar VAP-1 mechanism. Systemic TLR4 activation with LPS also has profound consequences to the recruitment of leukocytes to peripheral tissues. In contrast to the lung and liver, almost no leukocyte recruitment occurs in other tissues following systemic application of LPS. For example, i.p. administration of LPS rapidly results in its dissemination throughout the body via the circulation. When we assessed neutrophil accumulation to the peritoneal cavity (modeling recruitment to the original source of the inflammatory stimulus), no accumulation was observed [11]. To understand why this was the case, we returned to the cremaster muscle and observed that following systemic administration of LPS, leukocyte rolling was drastically reduced even when compared with baseline levels of recruitment [11, 33]. Although leukocyte adhesion was significantly greater than baseline levels, it was not nearly as great as observed in response to local activation [11]. Similar effects of systemic TLR4 activation on recruitment to the skin were also observed [10]. The paucity in leukocyte/endothelial interactions was not a result of a lack of endothelial activation within the tissue, as increased adhesion molecule expression was detected in virtually all tissues including the skin and cremaster muscle [10, 11]. Rather, it appears to be largely a result of the fact that nearly all leukocytes are trapped in the lung and liver; effectively removing them from the circulation [10, 11, 33]. The implication is that although inflammatory cells are stuck in the lung and mediating self-destruction of the tissue, they are no longer able to traffic to peripheral tissues. The same process of leukocyte trapping in the lung occurs during bacterial-induced septic shock [2, 34, 35], although it is not yet known if this is similarly dependent on endothelial activation. However, if the mechanism could be identified and specifically blocked, it might be possible to not only prevent lung injury but also to redirect inflammation back to the periphery to facilitate bacterial clearance. Our recent studies suggest that the defect in leukocyte recruitment observed following systemic activation of TLR4 goes even further than a simple depletion of circulating leukocytes. In bone marrow chimeric mice in which leukocytes, but not endothelial cells, expressed TLR4, no significant drop in the number of circulating leukocytes was observed in response to systemic activation (i.e., they were not trapped in the lung). Despite this, leukocyte rolling in the cremaster muscle remained impaired [33], suggesting that activation of circulating leukocytes through TLR4 results in down-regulation of rolling mechanisms. In addition, despite the presence of at least some leukocyte adhesion, virtually no transmigration into peripheral tissues was observed following systemic TLR4 activation [11], even if a chemotactic stimulus was added locally in an attempt to draw leukocytes out of the vasculature (our 866
Journal of Leukocyte Biology Volume 77, June 2005
Fig. 3. Systemic activation of TLR4 through LPS administration results in the systemic activation of inflammatory processes. Two separate mechanisms contribute to the resulting disregulation of leukocyte recruitment. (A) Leukocyte accumulation in the narrow vessels of the lung and liver is a prominent, pathological feature of endotoxic shock. This not only results in damage to the local tissue but effectively removes the majority of inflammatory cells from the circulation so that they are unavailable to be recruited to peripheral tissues. The mechanism of accumulation is dependent on endothelial but not leukocyte activation through TLR4 and may be mediated by nonclassical adhesion molecules. (B) Activation of circulating leukocytes through TLR4 results in a down-regulation of the mechanisms mediating leukocyte/endothelial interactions and those mediating migration toward chemotactic stimuli. Therefore, leukocytes exposed to LPS while still in the circulation may be unable to enter tissues effectively to clear an infection.
unpublished observations). Indeed, in our in vitro assay of neutrophil migration to chemotactic stimuli, prior exposure to LPS to activate TLR4 initiates a p38-dependent inhibition of migration to intermediary [28] or end-target agents (our unpublished observations). Together, these studies demonstrate that if a leukocyte is activated through exposure to LPS while still in the circulation, recruitment mechanisms are shut down prematurely. Under normal circumstances, leukocytes would not encounter LPS until they had entered a tissue infected with a gram-negative bacteria. We are currently pursuing the in vivo mechanisms and implications of this process. Our working hypothesis is that p38 MAPK is activated in neutrophils from endotoxemic mice, and these neutrophils no longer respond to PI-3K or p38 MAPK-activating chemotactic stimuli. In summary, we propose a model of disregulated leukocyte recruitment resulting from systemic activation of TLR4 by LPS (Fig. 3). Activation of endothelial TLR4 results in the activation of the endothelium throughout the body to express adhesion molecules. This results, for reasons we are only beginning to understand, in leukocyte accumulation specifically within the narrow vessels of the liver and especially the lung, causing damage to these tissues (Fig. 3A). Circulating leukocytes are also activated through TLR4, but this is not responsible for leukocyte trapping in the lung as previously postulated. Instead, the premature activation of leukocytes in the circulation shuts down their migratory mechanisms, preventing them from effectively gaining access to an inflamed tissue in the periphery (Fig. 3). Therefore, systemic activation of TLR4 (both on endothelial cells and circulating leukocyte) effectively paralyzes the inflammatory response, potentially limiting the appropriate clearance of a bacterial infection in peripheral tissues. http://www.jleukbio.org
REFERENCES 1. Bone, R. C. (1991) Sepsis, the sepsis syndrome, multi-organ failure: a plea for comparable definitions. Ann. Intern. Med. 114, 332–333. 2. Welbourn, C. R., Young, Y. (1992) Endotoxin, septic shock and acute lung injury: neutrophils, macrophages and inflammatory mediators. Br. J. Surg. 79, 998 –1003. 3. Medzhitov, R., Janeway, C. J. (2000) Innate immune recognition: mechanisms and pathways. Immunol. Rev. 173, 89 –97. 4. Takeda, K., Akira, S. (2003) Toll receptors and pathogen resistance. Cell. Microbiol. 5, 143–153. 5. Janeway Jr., C. A., Medzhitov, R. (2002) Innate immune recognition. Annu. Rev. Immunol. 20, 197–216. 6. Ulevitch, R. J., Tobias, P. S. (1995) Receptor-dependent mechanisms of cell stimulation by bacterial endotoxin. Annu. Rev. Immunol. 13, 437– 457. 7. Poltorak, A., He, X., Smirnova, I., Liu, M. Y., Van Huffel, C., Du, X., Birdwell, D., Alejos, E., Silva, M., Galanos, C., Freudenberg, M., Ricciardi-Castagnoli, P., Layton, B., Beutler, B. (1998) Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282, 2085–2088. 8. Takeuchi, O., Hoshino, K., Kawai, T., Sanjo, H., Takada, H., Ogawa, T., Takeda, K., Akira, S. (1999) Differential roles of TLR2 and TLR4 in recognition of gram-negative and gram-positive bacterial cell wall components. Immunity 11, 443– 451. 9. Hoshino, K., Takeuchi, O., Kawai, T., Sanjo, H., Ogawa, T., Takeda, Y., Takeda, K., Akira, S. (1999) Cutting edge: Toll-like receptor 4 (TLR4)deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the LPS gene product. J. Immunol. 162, 3749 –3752. 10. Andonegui, G., Goyert, S. M., Kubes, P. (2002) Lipopolysaccharideinduced leukocyte-endothelial cell interactions: a role for CD14 versus Toll-like receptor 4 within microvessels. J. Immunol. 169, 2111–2119. 11. Yipp, B. G., Andonegui, G., Howlett, C. J., Robbins, S. M., Hartung, T., Ho, M., Kubes, P. (2002) Profound differences in leukocyte-endothelial cell responses to lipopolysaccharide versus lipoteichoic acid. J. Immunol. 168, 4650 – 4658. 12. Wagner, J. G., Roth, R. A. (1999) Neutrophil migration during endotoxemia. J. Leukoc. Biol. 66, 10 –24. 13. Danner, R. L., Elin, R. J., Hosseini, J. M., Wesley, R. A., Reilly, J. M., Parillo, J. E. (1991) Endotoxemia in human septic shock. Chest 99, 169 –175. 14. van Deventer, S. J., Buller, H. R., ten Cate, J. W., Sturk, A., Pauw, W. (1988) Endotoxaemia: an early predictor of septicaemia in febrile patients. Lancet 1, 605– 609. 15. Kubes, P. (2002) The complexities of leukocyte recruitment. Semin. Immunol. 14, 65–72. 16. Patel, K. D., Cuvelier, S. L., Wiehler, S. (2002) Selectins: critical mediators of leukocyte recruitment. Semin. Immunol. 14, 73– 81. 17. Alon, R., Feigelson, S. (2002) From rolling to arrest on blood vessels: leukocyte tap dancing on endothelial integrin ligands and chemokines at sub-second contacts. Semin. Immunol. 14, 93–104. 18. Johnston, B., Butcher, E. C. (2002) Chemokines in rapid leukocyte adhesion triggering and migration. Semin. Immunol. 14, 83–92. 19. Springer, T. A. (1995) Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Annu. Rev. Physiol. 57, 827– 872. 20. Springer, T. A. (1994) Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76, 301–314. 21. Lawrence, M. B., Springer, T. A. (1991) Leukocytes roll on a selectin at physiologic flow rates: distinction from and prerequisite for adhesion through integrins. Cell 65, 859 – 873.
22. Cohnheim J. (1877) Vorlesungen u¨ber Allgemeine Pathologie, Berlin, August Hirschwald Verlag. 23. Johnston, B., Walter, U. M., Issekutz, A. C., Issekutz, T. B., Anderson, D. C., Kubes, P. (1997) Differential roles of selectins and the ␣4-integrin in acute, subacute, and chronic leukocyte recruitment in vivo. J. Immunol. 159, 4514 – 4523. 24. Hickey, M. J., Sharkey, K. A., Sihota, E. G., Reinhardt, P. H., Macmicking, J. D., Nathan, C., Kubes, P. (1997) Inducible nitric oxide synthasedeficient mice have enhanced leukocyte-endothelium interactions in endotoxemia. FASEB J. 11, 955–964. 25. Wong, J., Johnston, B., Lee, S. S., Bullard, D. C., Smith, C. W., Beaudet, A. L., Kubes, P. (1997) A minimal role for selectins in the recruitment of leukocytes into the inflamed liver microvasculature. J. Clin. Invest. 99, 2782–2790. 26. Kubes, P., Jutila, M., Payne, D. (1995) Therapeutic potential of inhibiting leukocyte rolling in ischemia/reperfusion. J. Clin. Invest. 95, 2510 –2519. 27. Campbell, J. J., Foxman, E. F., Butcher, E. C. (1997) Chemoattractant receptor cross talk as a regulatory mechanism in leukocyte adhesion and migration. Eur. J. Immunol. 27, 2571–2578. 28. Heit, B., Tavener, S., Raharjo, E., Kubes, P. (2002) An intracellular signaling hierarchy determines direction of migration in opposing chemotactic gradients. J. Cell Biol. 159, 91–102. 29. Meyrick, B., Brigham, K. L. (1983) Acute effects of Escherichia coli endotoxin on the pulmonary microcirculation of anesthetized sheep structure: function relationships. Lab. Invest. 48, 458 – 470. 30. Williams Jr., J. H., Patel, S. K., Hatakeyama, D., Arian, R., Guo, K., Hickey, T. J., Liao, S. Y., Ulich, T. R. (1993) Activated pulmonary vascular neutrophils as early mediators of endotoxin-induced lung inflammation. Am. J. Respir. Cell Mol. Biol. 8, 134 –144. 31. Worthen, G. S., Schwab III, B., Elson, E. L., Downey, G. P. (1989) Mechanics of stimulated neutrophils: cell stiffening induces retention in capillaries. Science 245, 183–186. 32. Erzurum, S. C., Downey, G. P., Doherty, D. E., Schwab III, B., Elson, E. L., Worthen, G. S. (1992) Mechanisms of lipopolysaccharide-induced neutrophil retention. Relative contributions of adhesive and cellular mechanical properties. J. Immunol. 149, 154 –162. 33. Andonegui, G., Bonder, C. S., Green, F., Mullaly, S. C., Zbytnuik, L., Raharjo, E., Kubes, P. (2003) Endothelium-derived Toll-like receptor-4 is the key molecule in LPS-induced neutrophil sequestration into lungs. J. Clin. Invest. 111, 1011–1020. 34. Sheridan, B. C., McIntyre Jr., R. C., Moore, E. E., Meldrum, D. R., Agrafojo, J., Fullerton, D. A. (1997) Neutrophils mediate pulmonary vasomotor dysfunction in endotoxin-induced acute lung injury. J. Trauma 42, 391–396. 35. Holman Jr., J. M., Saba, T. M. (1988) Hepatocyte injury during postoperative sepsis: activated neutrophils as potential mediators. J. Leukoc. Biol. 43, 193–203. 36. Liu, L., Kubes, P. (2003) Molecular mechanisms of leukocyte recruitment: organ-specific mechanisms of action. Thromb. Haemost. 89, 213–220. 37. Tasaki, O., Goodwin, C., Mozingo, D. W., Cioffi Jr., W. G., Ishihara, S., Brinkley, W. W., Dubick, M. A., Smith, R. H., Srivastava, O., Pruitt Jr., B. A. (1999) Selectin blockade worsened lipopolysaccharide-induced lung injury in a swine model. J. Trauma 46, 1089 –1095. 38. Burns, J. A., Issekutz, T. B., Yagita, H., Issekutz, A. C. (2001) The ␣ 4  1 (very late antigen (VLA)-4, CD49d/CD29) and ␣ 5  1 (VLA-5, CD49e/CD29) integrins mediate  2 (CD11/CD18) integrin-independent neutrophil recruitment to endotoxin-induced lung inflammation. J. Immunol. 166, 4644 – 4649. 39. Salmi, M., Jalkanen, S. (1992) A 90-kilodalton endothelial cell molecule mediating lymphocyte binding in humans. Science 257, 1407–1409.
Kerfoot and Kubes TLR4-induced leukocyte recruitment
867
