|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Physiology Program, Department of Environmental Health, Harvard School of Public Health, Boston, MA.
Correspondence to: Claire M. Doerschuk, MD, Physiology Program, Harvard School of Public Health, 665 Huntington Ave, Boston, MA 02115-6021; e-mail: cdoersch{at}hsph.harvard.edu
Neutrophils are thought to play an important role in the pathogenesis of ARDS.1 2 3 4 These patients often have a variety of inflammatory stimuli present within their lungs and circulating in their blood. The process through which these inflammatory mediators activate neutrophils and initiate their response occurs through a series of steps. First, the activated neutrophils sequester within the pulmonary microvasculature. They then adhere to the endothelium, a process most likely involving neutrophil-endothelial cell adhesion molecules. Neutrophils then migrate out of the vasculature into the lung parenchyma and airspace, usually in the presence of a lung injury or a chemotactic/haptotaxic gradient. Neutrophil-mediated injury to endothelial cells can occur during adhesion or emigration, although both these processes can occur in the absence of any discernible endothelial cell injury.
There are several differences in the inflammatory process when it occurs in the lungs compared to most other organs. First, the site of neutrophil sequestration and emigration in the systemic circulation is usually the postcapillary venules.5 6 However, in the pulmonary circulation, these processes occur primarily within the pulmonary capillaries.7 8 9 10 11 12 Second, the initial response of neutrophils within the systemic postcapillary venules is their rolling along the endothelial cells, a process that is mediated by members of the selectin family of adhesion molecules.13 14 15 16 17 While rolling occurs within both the arterioles and the venules of the lungs, rolling does not occur within the pulmonary capillary bed.18 Most of the capillary segments are narrower in diameter than spherical neutrophils, and the absence of rolling is most likely due to these spatial restraints.19 20
Neutrophil Traffic Through the Normal Pulmonary Microvasculature
Even in normal lungs, neutrophils are concentrated within the pulmonary capillary bed as compared to the systemic blood.21 22 23 24 25 26 27 28 Neutrophil transit through the pulmonary capillary bed requires considerably longer than the transit of plasma or RBCs. The median pulmonary capillary transit time for neutrophils is 26 s, while the capillary transit times for plasma are 1.4 ± 0.3 s and for RBCs are 1.4 s.21 25 The neutrophils are not merely moving at a slower velocity, but their longer transit time reflects stops made by neutrophils as they pass through the capillaries.21 22 28 About 40 to 50% of neutrophils transit through the lungs without stopping, while about 30 to 40% stop once and 20 to 30% stop twice or more. We and others have suggested that these stops reflect the time neutrophils require to deform and elongate, in order that they may pass through the narrower pulmonary capillary segments.
To begin to test this hypothesis, the shape of neutrophils was measured
as the ratio of the longest axis to the perpendicular axis at the
midpoint.18
19
29
Using several techniques, including
video, confocal, and electron microscopy, most neutrophils within the
arterioles had shape factors within 1.0 to 1.2.18
However,
when neutrophils were in capillaries, their shape measured
1.5 ± 0.3,29
suggesting that they did indeed need to
deform in order to traffic through this capillary bed. In the first
postcapillary venules, many neutrophils were still deformed, and the
shape factors varied from
1.1 to 2.18
Because about 40
to 60% of the capillary segments are smaller than neutrophils, and a
capillary pathway from an arteriole to a venule contains 40 to 100
capillary segments, most neutrophils will encounter a capillary segment
narrower than their spherical diameters in a single transit through the
pulmonary capillary bed.20
30
However, small changes in
shape that reduce the cross-sectional diameter by 15 to 20% require
only a short period of time and are unlikely to require a measurable
stop.20
In contrast, shape changes larger than a 20%
reduction in diameter are likely to result in neutrophil stops and
lengthened transit times. These studies show that the mechanical
properties of neutrophils are important in determining their
trafficking patterns through the pulmonary microvasculature.
Neutrophil Sequestration Within the Pulmonary Capillaries
During inflammation, the number of neutrophils within the pulmonary capillary bed increases dramatically. Many studies investigating the early events in neutrophil sequestration have utilized intravascular infusion of individual inflammatory mediators. For example, the infusion of complement fragments induces a profound neutropenia within 0.5 to 1.0 min.31 32 When the fragments are cleared from the circulation, the neutrophil counts rapidly recover, rebound, and return to baseline levels within about 90 min.32 Initial hypotheses focused on understanding the mechanisms of this sequestration examined the role of adhesion molecules. However, inhibition of either selectins or the CD11/CD18 adhesion complex did not prevent the initial sequestration induced by complement fragments,31 33 34 and the circulating neutrophil counts decreased to similar values within 2 min of infusion of complement fragments (Fig 1) . In contrast, adhesion molecules, particularly L-selectin and CD11/CD18, were required to maintain the sequestered neutrophils within the microvasculature.31 33 34 When the function of either molecule was prevented, the neutropenia lasted for only 4 to 7 min, and the circulating neutrophil counts began to recover despite continued infusion of the complement fragments (Fig 1) . The fourfold to fivefold increase in the number of neutrophils within the pulmonary capillaries that accumulate during a 15-min infusion of complement fragments was completely prevented when the animals were pretreated with fucoidin or anti-CD18 antibodies, but not when pretreated with an anti-P-selectin antibody. Taken together, these data demonstrate that recognized adhesion molecules are not required for the immediate sequestration of neutrophils within the pulmonary capillary bed, but they are required to maintain the sequestered cells within the bed for > 4 to 5 min.
|
Changes in biomechanical properties have been evaluated by many laboratories using a variety of approaches. Early studies showed that the pressure required to pass neutrophils through filters containing 5-µm pores was increased when the neutrophils were stimulated with complement fragments or other mediators35 36 37 38 (Fig 2) . This increase in pressure was not prevented when neutrophils were treated with anti-CD18 antibodies. Studies evaluating the contribution of components of the cytoskeleton showed that when the neutrophils were treated with colchicine to block microtubular reassembly, this complement fragment-induced stiffening was not prevented (Fig 2) . However, pretreatment with cytochalasin B decreased the pressure required to pass quiescent neutrophils through the filter and completely prevented the increase in pressure required to pass stimulated neutrophils through this filter (Fig 2) . These studies and others led to the hypothesis that when stimuli bind to transmembrane receptors on the surface of neutrophils, they induce soluble g-actin to bind to f-actin filaments. This increased cytoskeletal formation occurs at the cell periphery, immediately under the membrane. This change is thought to decrease the deformability of neutrophils and increase their sequestration within the pulmonary capillary, accounting for the rapid onset of neutropenia.29 30 37
|
To determine if complement fragments induced a redistribution of actin within the cells, ultrastructural immunohistochemical studies with colloidal gold labeling were performed using an antibody against actin that recognized both g-actin and f-actin.29 Intravascular complement fragments induced a redistribution of actin from the central, perinuclear regions of intracapillary neutrophils to the peripheral regions under the cell membrane within just 2 min. In addition, intravascular complement fragments also induced the accumulation of actin within the microvillar processes. These processes are present on quiescent neutrophils and flatten to provide the excess membrane necessary to encompass the neutrophils' volume when elongated. This increase in actin might decrease the ability of these microvillar processes to flatten. These studies suggested that perhaps not only the increase in actin beneath the membrane but also the decrease in the ability of microvillar processes to flatten might play a role in the stiffening process.
Neutrophil Adhesion and Migration
Once the neutrophils are sequestered within the capillaries, they adhere to the endothelium and migrate most commonly between two endothelial cells into the thick wall of the capillary loop.11 12 They then migrate through this interstitium into the alveolar space, preferentially between type I and type II alveolar epithelial cells.11 12 This process of adhesion and migration has been termed emigration.
In contrast to neutrophil emigration in the systemic
postcapillary venules where CD11/CD18 is almost always required,
neutrophil emigration can occur within the distal lung parenchyma
through one of two adhesion pathways, one that requires the CD11/CD18
complex and one that does not. Which pathway is selected appears to
depend on the stimulus. Escherichia coli, Pseudomonas aeruginosa,
E coli lipopolysaccharide (LPS), IgG immune complexes,
interleukin-1
, and phorbol myristate acetate elicit neutrophil
emigration through a CD11/CD18-dependent pathway while
Streptococcus pneumoniae, group B streptococcus,
Staphylococcus aureus, hydrochloric acid, hyperoxia, and C5a
induce neutrophil emigration through a CD11/CD18-independent
pathway.39
40
41
42
43
44
45
46
These pathways have been defined using
blocking anti-CD18 monoclonal antibodies. Even in the
CD11/CD18-dependent pathway, there is often a component of
CD11/CD18-independent neutrophil emigration that comprises 15 to 30%
of the emigration. In contrast, neutrophil emigration through systemic
postcapillary venules is virtually always completely blocked by the
same antibodies.
A controversial point has been whether the CD11/CD18-independent pathway and the remaining CD11/CD18-independent component of primarily CD11/CD18-dependent pathways was mediated through an alternative site on the CD11/CD18 complex that was not inhibited by blocking monoclonal antibodies. Initial attempts to evaluate this question utilized CD18 null mice. However, neutrophil emigration in response to E coli, a stimulus eliciting CD18-dependent neutrophil emigration, was actually increased two- to threefold, and neutrophil emigration in response to S pneumoniae, which elicits CD18-independent emigration, was increased fourfold to fivefold in CD18 null mice.47 The circulating neutrophil counts in the CD18 null mice were 15 to 45 times higher than those of wild-type mice, indicating that the number of circulating neutrophils delivered to the inflammatory site was much greater. These high circulating values made predicting the number of neutrophils expected to migrate in the absence of an adhesion defect difficult, if not impossible.
Therefore, an alternative approach was developed to generate chimeric mice with bone marrows producing both CD18 null and wild-type neutrophils.48 C57BL/6 mice were lethally irradiated and given IV injections of homogenates of fetal liver tissue mixed from day 14 CD18 null and wild-type mice. After the bone marrow had reconstituted, the mice were given pneumonias with either LPS or S pneumoniae. After 6 h following E coli LPS instillation, the fraction of CD18 null neutrophils in the BAL fluid was only 30% of that circulating in the blood. In streptococcal pneumonia, there was no defect in the emigration of CD18 null neutrophils compared with wild-type cells. These studies indicate that both the CD11/CD18-independent pathway of neutrophil emigration and the small number of neutrophils that emigrate when primarily CD18-dependent pathways are utilized both occur through adhesion pathways that do not require any region of the CD11/CD18 molecules.48 CD18-independent neutrophil emigration is therefore truly CD18 independent and not mediated by an alternative site on this molecule.
Current work in our laboratory has focused on two questions: what
determines which adhesion system is utilized for emigration and what
adhesion molecules mediate CD18-independent emigration in the lungs. To
begin to identify factors that determine which adhesion system will be
utilized, the messenger RNA (mRNA) and protein expression of
intercellular adhesion molecule-1 (ICAM-1), an important ligand for the
CD11/CD18 adhesion complex, was examined during CD11/CD18-dependent and
-independent pneumonias. These studies showed that ICAM-1 mRNA was
increased by 2 h when neutrophil emigration was induced by E
coli or P aeruginosa, but no upregulation of ICAM-1
mRNA was observed when S pneumoniae was the stimulus.
Ultrastructural immunohistochemical studies showed that ICAM-1
expression was increased on pulmonary capillary endothelial cells
in vivo when E coli LPS but not when S
pneumoniae was the stimulus.49
These data suggested
that induction of ICAM-1 may determine which pathway will be utilized.
Because ICAM-1 is regulated by nuclear factor-
B (NF-
B),
subsequent studies evaluated the translocation of this transcription
factor. In pneumonias induced by E coli but not S
pneumoniae, NF-
B translocation occurred within 2 h.
Because NF-
B translocation both regulates and is regulated by
cytokine production, current studies are focused on understanding the
roles of the proinflammatory cytokines tumor necrosis factor-
,
interleukin-1, and interferon-
in CD11/CD18-dependent and
-independent neutrophil emigration.
The second question, what adhesion molecules mediate CD11/CD18-independent emigration in the lungs, is still an open one. The role of selectins was examined by pretreating either wild-type or E-selectin/P-selectin double deficient mice with fucoidin prior to inducing streptococcal pneumonia. These studies showed that CD11/CD18-independent neutrophil emigration was not prevented by either this simultaneous genetic deficiency or by blockade of selectins by fucoidin.50 These studies suggest that selectins are not mediating CD11/CD18-independent neutrophil emigration. Adhesion molecules that remain to be tested include platelet-endothelial cell adhesion molecule-1 and very late antigen-4.
Release of Neutrophils From the Bone Marrow
The number of neutrophils available to sequester and emigrate into the lungs during ARDS is influenced by the production and release of neutrophils from the bone marrow. As mentioned above, infusion of inflammatory mediators, including complement fragments, induces a peripheral blood neutrophilia that is observed after the complement fragments have been cleared from the bloodstream. Morphometric studies evaluating the number of neutrophils within the pulmonary microvasculature after infusion of complement fragments for 15 min showed that the increase in the total number of intracapillary neutrophils was four to five times greater than the number of neutrophils that were circulating in the blood and immediately available for sequestration in the lungs prior to the infusion.31 51 In addition, studies examining the number of neutrophils in the lungs at the time when the circulating neutrophil counts were elevated following clearance of inflammatory mediators from the blood showed that the reduction in intracapillary neutrophils within the lungs measured about 25%, and this change was unlikely to account for the large increase in circulating cell numbers.32 In fact, there was an increase in neutrophils within the hepatic sinusoids at this time, suggesting that sequestered neutrophils circulate poorly once released.32 Taken together, these data suggest that neutrophils were released rapidly from a pool, most likely within the bone marrow, in the presence of intravascular inflammatory mediators.
Recent studies have demonstrated that the release of neutrophils from the bone marrow occurs much more rapidly than previously appreciated.32 In fact, by 7 to 10 min, the bone marrow is releasing neutrophils into the venous blood which immediately sequester within the pulmonary circulation while the infusion of complement fragments or other mediators is ongoing (Fig 3) . Virtually the entire fourfold to fivefold increase in neutrophils sequestered within the capillaries could be accounted for by the number of neutrophils released from the bone marrow.32 This inflammatory mediator-induced release does not occur for platelets (Fig 3) or lymphocytes, although circulating counts do decrease for all three. Curiously, the circulating platelet counts decreased 1 to 4 min after the neutrophils (Fig 3) , suggesting that platelets might be adhering to sequestered neutrophils. Current studies are focused on understanding how inflammatory mediators induce this release of neutrophils.
|
The mechanisms through which neutrophils respond to inflammatory stimuli within the lungs and the bloodstream are complex. Initial sequestration of neutrophils in response to inflammatory mediators appears to occur through mediator-induced changes in the biomechanical properties of neutrophils, particularly a stiffening of these cells and a reduction in their ability to deform and pass through the pulmonary capillary bed. Subsequent events require neutrophil-endothelial cell adhesion to keep the sequestered neutrophils within the lungs. Adhesion and migration can occur through two pathways, one that requires the CD11/CD18 pathway and one that does not. Which pathway is selected may depend on the balance of cytokines produced and signaling factors that regulate these processes. The presence of acute inflammatory stimuli within the bloodstream also results in the release of neutrophils from the bone marrow, increasing the numbers of neutrophils available to perpetuate lung injury or to aid in the resolution of acute inflammatory stimuli. The use of genetically deficient animals has been helpful in understanding the utilization of and the redundancies between both adhesion pathways and cytokines.
Acknowledgements
The authors wish to thank Arthur L. Beaudet, MD, for generously providing the adhesion molecule-deficient mice, Fumio Takei, PhD, for providing antibodies against murine adhesion molecules, and Bruce Horwitz, MD, PhD, and David Baltimore, PhD, for their expertise in generating bone marrow chimeric mice. We are very grateful to each of these scientists for countless fruitful and exciting discussions.
Footnotes
Supported by PHS HL48160, HL52466, HL33009, a Research Grant from the Cystic Fibrosis Foundation, and a Clinical Scientist Award in Translational Research from the Burroughs Wellcome Fund.
References
This article has been cited by other articles:
![]() |
A. Basit, J. Reutershan, M. A. Morris, M. Solga, C. E. Rose Jr., and K. Ley ICAM-1 and LFA-1 play critical roles in LPS-induced neutrophil recruitment into the alveolar space Am J Physiol Lung Cell Mol Physiol, August 1, 2006; 291(2): L200 - L207. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Matthay and G. A. Zimmerman Acute Lung Injury and the Acute Respiratory Distress Syndrome: Four Decades of Inquiry into Pathogenesis and Rational Management Am. J. Respir. Cell Mol. Biol., October 1, 2005; 33(4): 319 - 327. [Full Text] [PDF] |
||||
![]() |
H. Saito, Y. Minamiya, U. Kalina, S. Saito, and J.-i. Ogawa Effect of antithrombin III on neutrophil deformability J. Leukoc. Biol., September 1, 2005; 78(3): 777 - 784. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Tasaka, H. Koh, W. Yamada, M. Shimizu, Y. Ogawa, N. Hasegawa, K. Yamaguchi, Y. Ishii, S. E. Richer, C. M. Doerschuk, et al. Attenuation of Endotoxin-Induced Acute Lung Injury by the Rho-Associated Kinase Inhibitor, Y-27632 Am. J. Respir. Cell Mol. Biol., June 1, 2005; 32(6): 504 - 510. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. J. Laudes, R.-F. Guo, N. C. Riedemann, C. Speyer, R. Craig, J. V. Sarma, and P. A. Ward Disturbed Homeostasis of Lung Intercellular Adhesion Molecule-1 and Vascular Cell Adhesion Molecule-1 During Sepsis Am. J. Pathol., April 1, 2004; 164(4): 1435 - 1445. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Kiefmann, K. Heckel, M. Dorger, S. Schenkat, M. Stoeckelhuber, J. Wesierska-Gadek, and A. E. Goetz Role of poly(ADP-ribose) synthetase in pulmonary leukocyte recruitment Am J Physiol Lung Cell Mol Physiol, November 1, 2003; 285(5): L996 - L1005. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Thatte, V. Dabak, M. B. Williams, T. J. Braciale, and K. Ley LFA-1 is required for retention of effector CD8 T cells in mouse lungs Blood, June 15, 2003; 101(12): 4916 - 4922. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |