| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Biomedical Engineering and 2Department of Pharmacology and Physiology, University of Rochester, Rochester, New York 14642
Submitted 5 May 2004 ; accepted in final form 18 August 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
leukocyte recruitment; leukocyte adhesion cascade; adhesion molecule expression in vivo; venular hemodynamics
Selectin-mediated leukocyte rolling is characterized by transient adhesion and detachment that has been described as a series of ratchet-like steps (27). Reported values of leukocyte rolling velocity cover a wide range, even in well-defined in vitro systems (8); P-selectin-mediated rolling is typically described as highly variable. The erratic nature of leukocyte rolling is even more prominent in vivo, where the velocity and concentration of rolling leukocytes in postcapillary venules can vary with position and time (6, 16). This raises the important question of what determines this variability, and its impact on expected leukocyte-endothelial interactions. Factors that are likely to be primarily responsible for this variability in leukocyte rolling behavior are variations in shear-related binding kinetics due to local differences in shear forces in vivo and localized variations in adhesion molecule expression levels. Understanding the contributions of each of these to leukocyte rolling behavior in vivo is an essential step in understanding the manifestation of the inflammatory response in the native blood-perfused venular environment.
Shear dependence of leukocyte rolling has been reported both in isolated cell systems (8, 18) and in vivo (4, 13). The shear dependence of selectin bond kinetics (2, 3) suggests that leukocyte rolling velocity should be predictable from local shear forces. However, wide ranges of translational velocities can be observed for a single rolling leukocyte in a constant shear environment (6), suggesting the involvement of additional influences on leukocyte rolling behavior. Importantly, the effects of local shear forces on leukocyte rolling behavior in vivo are somewhat controversial in that some studies report a strong flow dependence of rolling velocity (4, 13), whereas elsewhere it has been reported that while mean rolling velocity varies approximately linearly over a wide range of shear rates, there is in fact substantial variability in velocity at any given shear rate and relative insensitivity to flow at lower shear stresses (6).
In addition to the possible contributions of local variations in shear-dependent behavior (24), leukocyte rolling interactions with the venular wall will necessarily be influenced by the presence of the appropriate adhesion molecule on the endothelial surface. Observations of clustering of rolling leukocytes in localized vessel regions and local similarities in rolling velocity in different venules have led to the suggestion that selectin expression might differ in different venular regions (6, 16). Variability in adhesion molecule expression has been reported between vessels of different organ systems (7) and in whole mount preparations of mouse cremaster microvasculature (10). These studies suggest that regions of variable adhesion molecule (selectin) expression are inherent to the intact microvasculature and that such differences in selectin density could locally alter the behavior of interacting leukocytes. To date, this hypothesis has not been tested directly.
Thus the goals of this study were twofold. The first was to determine how the erratic nature of rolling behavior is related to the fluid shear forces experienced in the native venular environment, and the second was to test whether the spatial distribution of P-selectin in these vessels varied in a way that could directly contribute to leukocyte rolling events in vivo.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Intravital microscopy. Images of venules were acquired at 30 frames/s using an Olympus BX50WI microscope through an Olympus PlanFl immersion objective (x40, 0.80 numerical aperture). Images were recorded to 3/4-in. videotape (SONY VO9600) for off-line analysis. Bright-field images used to track leukocyte interactions with the venular wall were acquired with a charge-coupled device (CCD) video camera (Dage MTI, CCD72S). Fluorescence images were acquired through a Nipkow disk scanning confocal head (Yokogawa) connected to an intensified CCD camera (XR Mega10, Solamere Technology Group) using a 20-mW argon laser (National Laser) to provide excitation at 488 nm; laser power and camera gain settings were held constant for all image acquistion. In protocols where images of leukocyte interactions were matched to fluorescence images in the same venule, bright-field images were collected through the confocal light path using the XR Mega 10 intensified CCD camera.
Immunofluorescence labeling. P-selectin, expressed on the endothelial surface of intact blood-perfused microvessels, was labeled by localized perfusion using micropipette cannulation. Cannulating micropipettes were pulled from borosilicate capillary tubes (WPI) and beveled on an extra fine grinding wheel (Sutter) to create a needle-like tip with an 8–12-µm-diameter opening. Pipettes were filled with the appropriate solutions and connected to a pressure reservoir. A hydraulic micromanipulator (Narishige) was used to position the sharpened pipette and cannulate an arteriole feeding the microcirculatory network of the cremaster muscle, thus allowing the complete perfusion of cremaster microvessels with the pipette contents. When necessary, a blunted glass microoccluding rod was placed using a second micromanipulator and was used to reduce blood flow into the cannulated arteriole to eliminate the presence of blood from the perfused solution. A primary monoclonal antibody against P-selectin (50 µg/ml, RB40.34, BD Biosciences) was perfused for up to 20 min. This was followed by a second micropipette cannulation that was used to perfuse an Alexa 488-conjugated polyclonal secondary antibody for 20 min (50 µg/ml, Alexa 488 anti-rat, Molecular Probes). Normal blood flow was then reestablished, and images were acquired using confocal microscopy. In separate control experiments, the matched isotype control (50 µg/ml, IgG2a, BD Biosciences) was used instead of the primary antibody in the dual-antibody perfusion sequence: these controls showed no detectable fluorescent labeling above background (mean intensity of the sampled control tissue was 25.4 ± 2.8 vs. 24.8 ± 2.9 grayscale units sampled on a glass surface without the tissue). To further define the overall variability inherent in confocal imaging of the immunofluorescent labeled P-selectin, we also imaged P-selectin that had been uniformly coated onto an artificial surface and then labeled with the primary and secondary antibodies as described above. The standard deviation of P-selectin intensity imaged as described above was 11.6% of the mean pixel intensity.
Data acquisition and analyses. Videotaped sequences were digitized to 8-bit tiff images using a CG-7 frame grabber (Scion) on an Apple G4 computer and analyzed with NIH Image (version 1.61). The hydrodynamic critical velocity as defined by Ley and Gaehtgens (20) was used to identify rolling leukocytes. A rolling leukocyte was defined as one that was seen to be translating along the venular wall in contact with the endothelium and was moving more slowly than the critical hydrodynamic velocity. Translational velocity for both leukocytes and fluorescent flow tracers (0.5 µm diameter, Fluorescebrite, Polysciences, injected through a jugular catheter in tracer quantity) was calculated from displacements over defined time intervals. Newtonian wall shear rate was calculated as 8Vavg/D from the measured bead velocities (Vavg) and vessel diameter (D) (12, 13, 20). Bead velocities were also used to calculate the critical velocity as above.
P-selectin expression levels were analyzed (using NIH Image) from confocal images of axial cross sections of each labeled vessel. Fluorescence intensities were sampled for 2 x 5-µm regions of interest drawn around a region of vessel wall that was in clear focus. These were expressed relative to a fluorescence standard solution (0.1 mg/ml FITC-albumin, Sigma-Aldrich) that was continuously perfused into the microvasculature by micropipette cannulation after capture of all P-selectin-labeled images.The FITC-albumin standard solution completely filled each sampled microvessel; the images were used to normalize for vessel specific attenuation of the fluorescence signal due to localized variability in the optical properties of the tissue. Intensity measurements of the fluorescence standard were acquired for each sampled vessel in a region of interest located in the center of the vessel at least 5 µm away from the labeled vessel wall.
Statistics. Statistical tests were performed using Graphpad Prism (version 3.0) to undertake t-tests, ANOVA, linear regression, or correlation analyses as appropriate. Differences were considered to be significant when P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Variability in leukocyte velocity during rolling. The measured translational velocity for leukocytes undergoing rolling behavior can vary widely as a function of time and position in any observed vessel. To define the breadth of variability in this translational velocity, measurements were made from displacements sampled at 100-ms intervals for the population of rolling leukocytes observed during a 2-min interval. The frequency distributions for the sampled minimum and maximum velocities across the observed population show that translational velocities ranged from 0 to >200 µm/s for this representative rolling population (Fig. 1A). This variability was also demonstrated by the velocity trajectory of a representative cell (Fig. 1B), which illustrates the transient high velocity translations and periodic pauses that typify P-selectin-mediated leukocyte rolling. Importantly, we found that the variability in velocity can be related to identifiable local regions of the venular endothelium, as demonstrated in Fig. 1C, which shows a compilation of the average velocity for all leukocytes at each axial location as a function of position along the sampled length of vessel wall. Along the 200-µm sampled length of a typical vessel shown in Fig. 1C, regions that supported slow velocity rolling were found to be interspersed with regions supporting high velocity translation or no visible rolling leukocytes at all. This spatial discontinuity in leukocyte rolling behavior across a straight length of venule that is free of flow disturbances strongly suggests that leukocyte rolling behavior is directly related to the localized characteristics of the endothelial surface and could, for example, be an outcome of locally variable levels of P-selectin.
|
100–400 s–1) that we typically observed in these venules. As illustrated in Fig. 1, leukocyte rolling actually consists of a variety of behaviors ranging from transient pauses to intermittent periods of high velocity translations. We hypothesized that the local fluid shear environment could affect the overall average velocity during rolling either by altering the rolling velocity during periods of consistent contact between the leukocyte and the endothelial surface or, alternatively, by increasing the occurrence of periods of high velocity translations during which the leukocyte transiently loses contact with the endothelium and translates rapidly along the wall. Displacements of 256 rolling leukocytes were tracked at 33-ms intervals over a 1-s period in 13 postcapillary venules from 7 mice. Newtonian wall shear rates sampled for each vessel ranged from 112 to 381 s–1 for native flows in these venules. In addition to measuring the overall average translational velocity (from the total displacement during a 1-s sampling interval), we also measured a "continuous contact" translational velocity from the displacements that occurred while the leukocyte was observed to be directly interacting (i.e., in contact) with the endothelial surface. This continuous contact velocity excluded leukocyte displacements that were associated with transient high velocity periods during which the cell exceeded twice the average velocity for all sampled cells. During this latter interval, leukocytes are translating along the vessel wall in the flowing hydrodynamic streamline and are therefore assumed not to be undergoing the P-selectin-mediated interactions that occur during the continuous contact displacements. There was a statistically significant increase in the overall average velocity with increasing wall shear rate (r2 = 0.56, P < 0.005), whereas, in contrast, the relationship between continuous contact translational velocity and wall shear rate was not statistically significant (Fig. 2A). Comparisons between the two pools of velocity values showed that the continuous contact translational velocity was significantly lower than the overall average velocity (27 ± 2 vs. 39 ± 4 µm/s, means ± SE, paired t-test, P < 0.001) with the only difference between the two measures being the inclusion of the transient high velocity displacements that occur while the leukocyte is not in continuous contact with the endothelium. Thus our data show that the direct effect of an increase in shear forces appears to be specifically related to the high velocity translations during transient detachment rather than being a reflection of variability in the leukocyte behavior during P-selectin-mediated rolling.
|
100 s–1) but that the detachment fraction increased dramatically as a function of wall shear rate, particularly in the range between 200 and 250 s–1, to a point where nearly all rolling leukocytes exhibited transient detachments at the highest observed wall shear rates (>300 s–1). These data clearly show that elevations in local wall shear rate are related to an increase in the occurrence of transient detachments. In turn, it is these transient detachments that account for the erratic nature of leukocyte rolling; the velocity of leukocytes during P-selectin-mediated interaction is relatively invariant, as illustrated in Fig. 2B. From these observations, we hypothesized that variable leukocyte rolling behavior was likely to be related to differences in P-selectin expression level or its variable spatial distribution in different local regions of the venule. P-selectin expression in postcapillary venules. To address whether there is a direct relationship between P-selectin-mediated leukocyte rolling and the localized expression patterns of P-selectin on vascular endothelium, we first asked whether P-selectin expression varied significantly in different regions of the microcirculation. Using immunofluorescent labeling and confocal microscopy, we undertook detailed characterizations of the spatial distribution of P-selectin in the intact blood-perfused microvasculature. Immunofluorescence labeling of P-selectin on the luminal surface of the endothelial membrane revealed a punctate pattern (Fig. 3) that is consistent with that observed in cultured endothelial cells (5, 9, 26). P-selectin expression was detectable only in venules and not in true capillaries or arterioles, consistent with observations on perfusion-fixed microvascular networks (10).
|
Even though generalized differences in P-selectin expression were not observed across vessels experiencing different wall shear rates, the high spatial detail available from the in vivo labeling procedure combined with confocal observations showed that there was considerable local variability in P-selectin expression within an individual venule (Fig. 4). Regions of high P-selectin expression in venules displayed fluorescence intensities more than twofold greater than regions of low expression (Fig. 4, bottom). Typically, the regions of high versus low fluorescence intensity in situ were on the order of 50 µm in length or less. Furthermore, the axial length dimension of the high expression regions that we observed is qualitatively consistent with the regions of slow leukocyte rolling identified earlier using the high temporal resolution analyses of P-selectin-mediated rolling (Fig. 1C). By extension, we infer that the regions of slow leukocyte rolling might be directly attributable to the observed regions of high P-selectin expression.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Leukocyte rolling velocity.
P-selectin-mediated leukocyte rolling and tethering are critical events for the propagation of the inflammatory response. Many studies have detailed biophysical aspects of bond formation between P-selectin and its primary counterligand, P-selectin glycoprotein ligand 1, including kinetic rate constants (3) and the effects of ligand valence (23). This raises questions about how P-selectin directly affects leukocyte interactions with the endothelial surface in a physiologically relevant environment, specifically postcapillary venules. Leukocyte rolling as usually measured in vivo describes a compilation of erratic behaviors that characterize the translation of transiently adhesive leukocytes in the sometimes variable environment of blood flow-induced shear forces. Thus a better understanding of the involvement of shear forces in the regulation of the rolling step of the leukocyte adhesion cascade clearly requires a more detailed analysis of rolling behavior. The mean rolling velocity that we report here is similar to mean velocities reported elsewhere in both venules and isolated cell systems (4, 6, 8, 18, 20). However, we show here that the average leukocyte rolling velocity can be further divided into qualitatively distinct components. In particular, separation of those interactions where the leukocyte is in continuous contact with the endothelial surface from those where the leukocyte appears to "skip" along the surface identifies the influences of shear forces versus localized adhesion molecule distributions that are hidden in the commonly reported values of average rolling velocity. Our data reveal two categories of translational behavior for rolling leukocytes: one consisting of a slow, steady translation where contact between the leukocyte and endothelium is maintained via P-selectin-mediated interactions, and the other consisting of a high velocity translation where direct contact with the endothelium is not apparent. We note that if different hemodynamic criteria were used to identify the rolling leukocyte population, that population would be slightly different from the one we studied; importantly, the division into two categories of translational behavior is independent of this consideration. Our study thus helps to clarify how the prevailing shear forces affect the rolling behavior of leukocytes under flow. Characterizations of shear-dependent selectin binding kinetics (1, 2) suggest that rolling velocity should vary with shear forces. This idea has been tentatively supported by in vivo observations of average velocity (Refs. 4, 6, 13, and 21 and the present study). However, we found that the mechanism underlying the apparent increase in average velocity was more subtle than first anticipated, because the fluid shear forces affected P-selectin-mediated rolling behavior primarily by increasing the occurrence of transient detachments rather than by affecting the translational velocity during consistent (P-selectin mediated) contact with the wall. These findings suggest that elevations in average rolling velocity with increasing shear forces are derived from two causal factors: the increase in the occurrence of transient detachments and the local flow velocity during the detachment, both leading to increased leukocyte translational velocity compared with periods where consistent endothelial contact is maintained. Notably, the occurrence of transient detachments increases rapidly at wall shear rates near 250 s–1, which suggests that there is a threshold level of force where P-selectin bonds are not sustained by the site densities native to the vasculature. Our study shows that, overall, the time that an individual leukocyte will spend interacting with the venular wall is indeed a function of the shear environment in the sense that this will determine both the frequency and characteristic velocity of the skips that occur between the periods of P-selectin-mediated rolling interaction with the wall.
Spatial localization of interactions.
Our velocity analyses showed that there were localized regions of consistent slow rolling and localized regions where rolling leukocytes were absent. As discussed, possible causes for these observations could be localized differences in shear forces and/or endothelial cell adhesion molecule levels; interestingly, Damiano and colleagues (6) reported that leukocyte adhesion energy varies locally in venules and hypothesized that this might reflect heterogeneity in P-selectin expression on the venular wall. Our data show that, indeed, differences in P-selectin expression can account directly for spatial differences in leukocyte rolling behavior.
Interestingly, the length dimension of the regions of slow rolling velocity and, accordingly, high P-selectin expression generally match the characteristic axial length of a venular endothelial cell (50 µm, J. C. Wojciechowski and I. H. Sarelius, unpublished observations), suggesting that an individual endothelial cell can exhibit dramatically different P-selectin expression compared with adjacent cells. We have also seen this localized difference in P-selectin expression among endothelial cells in histamine-stimulated human umbilical vein endothelial cells (11). The underlying cause of these localized differences in P-selectin levels is not obvious as cells in such close proximity are likely to experience similar levels of shear forces and inflammatory stimuli.
What might be the physiological consequences of these local variations in P-selectin expression? We note that on an even smaller length scale than one endothelial cell, P-selectin expression on the surface of individual cells is heterogeneous in that it appears punctate (Fig. 3), an observation reported earlier by others (9, 26). In preliminary modeling studies (14), we found that this punctate distribution supports slower leukocyte rolling than would occur if the same level of P-selectin was distributed uniformly on the endothelial cell surface, perhaps by optimizing local adhesion molecule surface density and hence bond formation and kinetics. We speculate that the P-selectin-rich venular regions that we describe in the present study may also support more efficient leukocyte capture in the venule or optimize later events (adhesion, transmigration) of the inflammatory cascade, for example, by altering adhesion molecule expression, junctional associations, or cytoskeleton arrangements in neighboring endothelial cells.
In summary, the behavior of rolling leukocytes that is mediated by P-selectin is directly related to the heterogeneous spatial expression pattern of P-selectin in intact postcapillary venules. Shear forces in this physiological environment affect the overall translational velocity of rolling leukocytes by increasing the occurrence of transient detachments from the endothelium at higher shear forces, thereby leading to brief periods of high velocity translation near the wall in the free fluid stream. These findings indicate that spatially localized elevations of adhesion molecule expression can regulate patterns of leukocyte rolling, a critical aspect of leukocyte recruitment in inflammation. This finding identifies an important mechanism by which leukocyte recruitment to the venular wall can be controlled in the in vivo environment.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Sumagin, E. Lomakina, and I. H. Sarelius Leukocyte-endothelial cell interactions are linked to vascular permeability via ICAM-1-mediated signaling Am J Physiol Heart Circ Physiol, September 1, 2008; 295(3): H969 - H977. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Miles, E. Liaskou, B. Eksteen, P. F. Lalor, and D. H. Adams CCL25 and CCL28 promote {alpha}4{beta}7-integrin-dependent adhesion of lymphocytes to MAdCAM-1 under shear flow Am J Physiol Gastrointest Liver Physiol, May 1, 2008; 294(5): G1257 - G1267. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Sumagin and I. H. Sarelius A role for ICAM-1 in maintenance of leukocyte-endothelial cell rolling interactions in inflamed arterioles Am J Physiol Heart Circ Physiol, November 1, 2007; 293(5): H2786 - H2798. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Sumagin and I. H. Sarelius TNF-{alpha} activation of arterioles and venules alters distribution and levels of ICAM-1 and affects leukocyte-endothelial cell interactions Am J Physiol Heart Circ Physiol, November 1, 2006; 291(5): H2116 - H2125. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.J. Woollard, D. Kling, S. Kulkarni, A.M. Dart, S. Jackson, and J. Chin-Dusting Raised Plasma Soluble P-Selectin in Peripheral Arterial Occlusive Disease Enhances Leukocyte Adhesion Circ. Res., January 6, 2006; 98(1): 149 - 156. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. H. Sarelius, J. M. Kuebel, J. Wang, and V. H. Huxley Macromolecule permeability of in situ and excised rodent skeletal muscle arterioles and venules Am J Physiol Heart Circ Physiol, January 1, 2006; 290(1): H474 - H480. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
). C: fraction of leukocytes undergoing detachments from the endothelium and translation in the free flow stream increases with wall shear rate. The spline fit line for leukocyte detachment rate sampled in venular regions with different characteristic shear rates shows that the rapid increase in the rate of detachment occurs when the shear rate is in the region of 250 s–1.
