AJP - Heart AJP: Advances in Physiology Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 293: H2786-H2798, 2007. First published August 17, 2007; doi:10.1152/ajpheart.00720.2007
0363-6135/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
293/5/H2786    most recent
00720.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sumagin, R.
Right arrow Articles by Sarelius, I. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sumagin, R.
Right arrow Articles by Sarelius, I. H.

A role for ICAM-1 in maintenance of leukocyte-endothelial cell rolling interactions in inflamed arterioles

Ronen Sumagin1 and Ingrid H. Sarelius2

1Department of Biomedical Engineering and 2Department of Pharmacology and Physiology, University of Rochester, Rochester, New York

Submitted 20 June 2007 ; accepted in final form 13 August 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
A key endothelial receptor in leukocyte-endothelial cell (EC) interactions is ICAM-1. ICAM-1 is constitutively expressed at low levels on vascular ECs, and its levels significantly increase following stimulation with many proinflammatory agents. This study provides evidence that in inflamed arterioles of anesthetized mice (65 mg/kg ip Nembutal), ICAM-1 mediates leukocyte rolling, in contrast to its expected role of mediating firm adhesion in venules. The number of leukocytes rolling on arteriolar ECs is decreased in ICAM-1 knockout (KO) compared with wild-type (WT) mice (KO, 6.0 ± 0.9; WT, 12.0 ± 1.0 leukocytes/40 s; P < 0.05), whereas the leukocyte-rolling number in venules remains unaffected (KO, 5.6 ± 0.9; WT, 7.0 ± 0.7 leukocytes/40 s; n = 13–15 sites). We also show that the fraction of leukocytes that is rolling on arteriolar ECs does so with a higher characteristic velocity (>70 µm/s), and, furthermore, that the distance over which rolling contacts with the arteriolar wall are maintained is ICAM-1 dependent. In ICAM-1 KO animals or in WT mice in the presence of ICAM-1-blocking antibody, leukocytes rolled significantly shorter distances over the sampled 200-µm vessel length compared with WT (68 ± 6.7 and 55 ± 9.4 vs. 85 ± 12.9% total, respectively, n = 4 sites, P < 0.05). We also found evidence that in ICAM-1 KO mice, a significant fraction of leukocyte rolling and adhesive interactions with arteriolar ECs could be accounted for by upregulation of another adhesion molecule, VCAM-1, providing an important illustration of how expression of related proteins can be altered following genetic ablatement of a target molecule (in this case ICAM-1).

intracellular adhesion molecule-1 knockout mice; intracellular adhesion molecule-1; vascular cell adhesion molecule-1; leukocyte adhesion; leukocyte rolling; tumor necrosis factor-{alpha}; in vivo; intravital confocal microscopy

RECRUITMENT OF CIRCULATING leukocytes from the bloodstream to specific areas of damaged tissue is a complex, multistep process orchestrated by a number of specific interactions between adhesion molecules expressed by endothelial cells (EC) and their counterligands on the leukocytes. A key endothelial receptor in leukocyte-EC interactions is ICAM-1. ICAM-1 is a transmembrane glycoprotein with five extracellular IgG-like domains and a short cytoplasmic tail that associates with cytoskeletal proteins (29). It binds to the family of beta2-integrins, leukocyte function antigen-1 (LFA-1, CD11a/CD18), and Mac-1 (CD11b/CD18), thereby mediating cell-cell adhesive interactions and mediating signal-transduction pathways in both leukocytes and ECs. In addition to the endothelium, ICAM-1 is expressed on antigen-presenting cells (1, 17, 47) and on lung epithelial cells (23). It plays a critical role not only in immune and inflammatory responses but also in embryonic and nervous-system development (14).

ICAM-1 is constitutively expressed at low levels on vascular ECs, and its levels significantly increase following stimulation with many proinflammatory agents, such as IL-1 and -6 (8, 48), TNF-{alpha} (42, 48), IFN-{gamma} (35), and LPS (4).

The role of ICAM-1 in mediating leukocyte interactions with venular ECs has been studied extensively in both isolated cells and in vivo models. A great deal is known about the role of ICAM-1 in mediation of leukocyte adhesion (25, 29, 49) and transmigration (16, 49). There is also evidence that it can contribute to selectin-mediated leukocyte rolling (11, 19, 38).

Arterioles on the other hand have been somewhat neglected, and, as a consequence, not a lot is known about the way leukocytes interact with ECs in arterioles and the regulatory mechanisms involved. Previously, we (42) have shown that during inflammation, the expression of adhesion molecules such as ICAM-1 is highly upregulated not just in venules but also in arterioles. Likewise, we (42) and others (26, 43) have shown that leukocyte rolling in arterioles is induced on proinflammatory stimulation, suggesting a role for arterioles in the inflammatory cascade. Importantly, we found that the expression levels and distribution patterns of ICAM-1 in TNF-{alpha}-activated arterioles were different from those in venules; we also observed differences in leukocyte behavior, in that, unlike in venules, leukocyte rolling in arterioles did not result in firm adhesion and consequent transmigration, despite the fact that the expression of ICAM-1 was dramatically increased in the presence of TNF-{alpha} (42).

These observations suggest that in arterioles, both leukocyte rolling and increased levels of ICAM-1, rather than leading to adhesion and transmigration, might have a different functional outcome. A number of previously published studies (14, 33, 45) have shown that, in in vitro settings, ICAM-1 ligation with monoclonal antibodies induces activation of a variety of signaling molecules such as the MAP kinases Raf1, p38, and ERK1/2; caveolin; Src; and PKC; these in turn evoke multiple cellular responses, including for example rearrangements of junctional and cytoskeletal proteins. Thus it is reasonable to speculate that interaction of leukocytes with ICAM-1 expressed on the surface of arteriolar ECs might stimulate one or more of these signaling molecules and thus facilitate participation of arterioles in some aspect of the inflammatory response. That arterioles possess the capacity to participate in inflammatory responses is indicated, for example, by the observation that they increase their permeability in response to appropriate inflammatory stimuli (15, 39). To address the possibility of such a role for leukocyte interactions with arteriolar ECs, the first essential step is to establish whether in arterioles there is indeed a component of the observed leukocyte rolling that is dependent on ICAM-1. Previously, P-selectin has been identified as a key adhesion molecule mediating leukocyte rolling in arterioles (26), but recently we reported that in in situ arterioles, regions of immunofluorescently labeled ICAM-1 could be correlated not only with the number of adhered leukocytes in venules but also, importantly, with the number of rolling leukocytes in arterioles (42), implying that ICAM-1 may indeed mediate leukocyte rolling in arterioles. In support of the idea that ICAM-1 may be mediating arteriolar leukocyte rolling, earlier work has shown that CD18/ICAM-1 interactions are capable of sustaining a fraction of leukocyte-rolling interactions on venular endothelium (11, 38).

Hence, the goal of the present study was to establish that ICAM-1 expression levels in arterioles directly affect leukocyte behavior, specifically by mediating a significant fraction of leukocyte rolling. To achieve this goal, we have examined leukocyte-rolling interactions in arterioles of wild-type (WT) mice under control conditions and after stimulation by the proinflammatory cytokine TNF-{alpha}, as well as in arterioles of mice lacking EC surface expression of ICAM-1 or where the ability to ligate ICAM-1 on the EC surface was modified.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Animal Preparation

All procedures were approved by the Institutional Review Board of the University of Rochester.

Male WT (C57BL/6J; Jackson Laboratories) or ICAM-1 knockout (KO) mice (B6.129S4-Icam1tm1Jcgr/J; Jackson Laboratories) between 12 and 15 wk old were initially anesthetized with sodium pentobarbital (65 mg/kg ip) and maintained on supplemental anesthetic as needed throughout the experiment via a jugular catheter. An endotracheal tube was inserted to insure a patent airway during the experiment, and body temperature was maintained by placing the animal on a warmer. The cremaster muscle was prepared for intravital microscopy as previously described (20, 27). Briefly, the right cremaster muscle was exteriorized and gently pinned over a quartz pedestal for visualization by microscopy. During preparation and observation, the tissue was continuously superfused with warmed physiological solution with the following composition (in mM): 131.9 NaCl, 4.7 KCl, 2.0 CaCl, 1.2 MgSO4, 18 NaHCO3, pH 7.4 at 36°C, and equilibrated with gas containing 0% O2, 5% CO2, and 95% N2 to maintain tissue PO2 < 15 Torr. On completion of the protocols, each animal was euthanized by anesthetic overdose.

To induce inflammation, selected animals were locally treated by intrascrotal injection of mouse recombinant TNF-{alpha} (0.5 µg TNF-{alpha} in 0.25 ml saline; Sigma-Aldrich) 3 h before the start of the surgical preparation (21, 25). Microcirculatory observations were made 4–5 h after the TNF-{alpha} injection.

Intravital Microscopy

An Olympus BX61WI microscope with an Olympus PlanF1 immersion objective (x20, 0.65 numerical aperture) was used to acquire images. Bright-field images used to track leukocyte interactions with the vessel wall were acquired by using a charge-coupled device (CCD) camera (Dage-MTI CD72). Fluorescence images were acquired by illuminating the tissue with a 20-mW argon laser and imaging with a Nipkow disk confocal head (CSU 10; Yokogawa) and intensified CCD camera (XR Mega 10; Stanford Photonics). Laser power and camera gain settings were unchanged throughout all the experiments. Images were either recorded to a DVD recorder (Sony DVO100MD) or to a VCR (Sony VO9500) for offline analysis. The spatial resolution in this imaging system is 0.5 µm.

In Situ Immunofluorescence Labeling of Adhesion Molecules

Adhesion molecules expressed on the surface of the endothelium were labeled by localized perfusion by using micropipette cannulation. This approach to in vivo labeling and imaging of blood-perfused microvessels has been described elsewhere (22, 42, 46). Briefly, after the surgery was completed, the preparation was placed under the microscope and was allowed to stabilize for at least 15 min. Sharpened micropipettes (42) were used to cannulate a main arteriole upstream of the microvascular region targeted for observation; this avoided damage in the region to be observed. A separate glass occluding rod (modified micropipette) was used to occlude the main inflow arteriole to the tissue upstream of the cannulation site at the time of the antibody loading to facilitate complete perfusion of the downstream microvasculature with the antibody solution. At the completion of all antibody loading (see below), the cannulating and occluding micropipettes were removed and blood flow was reestablished in the targeted microvascular region. The tone and diameter of cannulated arterioles before and following cannulation were closely monitored; by 3 min after cannulation, the diameter of the targeted microvessel was not different from that before cannulation, indicating that the vessel had regained its initial tone and suggesting that its functional capabilities were unchanged.

In all cases, the vessel was initially perfused with a primary rat anti-mouse monoclonal antibody in saline. Antibodies used were ICAM-1 (YN/1.7.4; eBioscience), P-selectin (RB40.34; Pharmingen), E-selectin (10E9.6; Pharmingen), and VCAM-1 (429; Pharmingen). The selected antibody was perfused at a concentration of 50 µg/ml for 15 min; the perfusion pipette was then withdrawn, and blood perfusion was briefly restored before a second cannulation with a pipette containing goat anti-rat secondary fluorescent polyclonal antibody in saline (Alexa 488 anti-rat, 50 µg/ml; Molecular Probes), which was perfused for another 15 min. The second micropipette was then withdrawn, blood flow was reestablished in the target microvascular region, and the experimental protocol was undertaken. Finally, after completion of all imaging protocols, a third cannulation was used to perfuse the target vessels with fluorescent standard solution (0.05 mg/ml FITC-dextran in saline, 150 kDa mol wt; Sigma-Aldrich) to normalize for differences in fluorescence produced by localized variability in the optical properties of the tissue. We used the same technical approach to evaluate nonspecific binding of the primary or secondary antibody: in separate control protocols, the cremaster muscle vasculature was perfused with 1) the relevant IgG-binding antibody followed by the secondary or 2) the secondary antibody alone. In each case, mean fluorescence intensity was 24 ± 0.9 and 26 ± 1.8 intensity units, respectively, which was not different from tissue background (28 ± 1.2 intensity units). In addition, we confirmed the lack of ICAM-1 expression in ICAM-1 KO mice by performing the same procedure in these mice using the anti ICAM-1 antibody; mean fluorescence intensity in microvessels in these animals was 27 ± 0.8 intensity units, which is not different from tissue background.

Analyses

Leukocyte-EC interactions. All images of intact blood-perfused microvessels and leukocyte-EC interactions were analyzed by using either ImageJ or NIH Image software. Analog images acquired to VCR were first digitalized to 8-bit TIFF images by using a CG-7 frame grabber (Scion); images acquired to DVD were analyzed directly.

Leukocyte-EC interactions were sampled in arterioles and venules ranging from 30 to 60 µm unless stated differently. Rolling leukocytes were defined as any leukocytes observed translating along the vessel wall in continuous contact with the endothelium that were moving slower than the hydrodynamic critical velocity as defined by Ley and Gaehtgens (31). The number of rolling leukocytes on the vessel wall (the rolling flux) was calculated by counting leukocytes rolling past a line perpendicular to the vessel axis per 40-s time interval. The fraction of rolling leukocytes (%total) was calculated by normalizing the number of rolling leukocytes to the leukocyte delivery, which was the total number of leukocytes observed in the vicinity of the vessel wall, including both firmly adhered leukocytes and those that were not in direct contact with the wall but were translating in the free stream close to the wall. Leukocyte-velocity distribution profiles were obtained by tracking a total of a 100 rolling leukocytes for a period of time >0.15 s (at least 10 frames) within a 50-µm length defined for each observed vessel (n = 6 vessels).

To asses the role of ICAM-1 in the ability of leukocytes to interact with the arteriolar endothelium, rolling leukocytes were tracked over 200-µm segments of arteriolar wall. The distance rolled and the location where leukocytes detached and disappeared from the field of view were measured, and the data are presented as percent of the observed 200-µm length that supports rolling. Additionally, representative leukocyte-rolling profiles were obtained in arterioles by tracking the displacement of leukocytes over 200 µm and determining the average translational velocity as a function of axial position (at successive 5-µm intervals).

To explore the contributions of both hemodynamics and adhesion-molecule expression to the ability of leukocytes to firmly adhere in arterioles, the shear rate was locally modulated (reduced) in selected arterioles by placing an occluding rod upstream from the site of data collection (at least 1 mm away) and lowering it in four consecutive steps. For each step, shear rate and leukocyte interactions were quantified. Newtonian shear rates were calculated as described elsewhere (21) by using 0.5-µm-diameter fluorescent beads as tracers. Shear rate was calculated as 8Vb/D from the measured mean bead velocities (Vb) and vessel diameter (D). Firmly adhered leukocytes were defined as cells that remained stationary for ≥30 s.

Adhesion-molecule intensity. Fluorescence-intensity levels, as an index of expression levels of P- and E-selectin and VCAM-1, were analyzed by using NIH Image software as previously described (22, 42). Briefly, fluorescence was measured as grayscale units, where 0 = black and 255 = white, and was normalized and expressed as intensity relative to the intensity of the infused FITC-dextran solution sampled as described earlier. For all intensity measurements, the laser power and camera gain settings were held constant, and the camera response was verified to be linear over the range used for these acquisitions. We also verified that the relationship between measured intensity and concentration of fluorochrome was linear when using this acquisition system (r2 was between 0.996 and 0.999 for the range of gain settings used). Fluorescence intensity was measured on the vessel wall at the central plane (focal plane positioned in the middle of the vessel, demonstrated in Refs. 22 and 42) of the vessel cross-section by projecting a line 5 pixels wide along the wall and obtaining the intensity profile along that line. We verified that the sampled vessel region was contained within the same confocal plane by moving the sampling plane vertically and observing that the entire selected region equally went out of focus.

Statistics. Statistical tests were performed by using GraphPad Prism (v. 4.0) to undertake t-tests, ANOVA, linear regression, or correlation analyses as appropriate. Statistical significance was set at P < 0.05.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
To understand the role of ICAM-1 in leukocyte rolling in the arterioles and venules, we characterized leukocyte-rolling behavior under selected conditions in which ICAM-1 function was altered by using ICAM-1 KO mice and an ICAM-1 function-blocking monoclonal antibody. Observations were made in both arterioles and venules (diameter range, 30–60 µm) in control animals and in tissues after activation with TNF-{alpha} as described in MATERIALS AND METHODS.

In Arterioles but Not Venules, the Number of Rolling Leukocytes Is Dependent on ICAM-1

Initial leukocyte-EC rolling interactions in venules are a key step in leukocyte recruitment (22, 24, 29–31, 34). In our study, leukocyte-EC rolling interactions were observed as expected in control and TNF-{alpha}-activated venules in WT mice. In control animals, a small number of leukocytes were found to be firmly adhered, but as expected, the majority of leukocytes interacting with the venular endothelium were rolling (15 ± 1.0 leukocytes/40 s, n = 15 venular sites). Under inflammatory conditions (TNF-{alpha} treatment), the balance between leukocyte rolling and adhesion interactions was, again as expected, dramatically altered. Under these conditions, the majority of leukocytes were firmly adhered to the venular wall, thus reducing the number of delivered leukocytes, occupying substantial wall area, and in consequence contributing to the significant decrease in the number of rolling leukocytes (7 ± 0.7 leukocytes/40 s, n = 15 sites, P < 0.001, Fig. 1A).


Figure 1
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 1. In arterioles (art) but not venules (ven), the number of rolling leukocytes is affected by ICAM-1. A: number of rolling leukocytes was calculated by counting leukocytes rolling past a line perpendicular to vessel axis during a 40-s time period. [Note that in arterioles, leukocyte-endothelial cell (EC) interactions did not occur under control conditions and were not quantified.] B: fraction of rolling leukocytes was calculated by normalizing number of rolling leukocytes to leukocyte delivery as defined in MATERIALS AND METHODS. Fractions of rolling leukocyte were calculated to account for elevated number of circulating leukocytes in ICAM-1 knockout (KO) mice. In arterioles of ICAM-1 KO animals, number and fraction of rolling leukocytes in TNF-{alpha}-activated arterioles was significantly lower (n = 13–15, P < 0.05) compared with wild-type (WT) animals. In venules of ICAM-1 KO mice, number of rolling leukocytes under control conditions was significantly (P < 0.001) increased due to a greater number of circulating leukocytes, but fraction of leukocyte rolling was not different from that in WT mice in both control and TNF-{alpha} venules, suggesting that ICAM-1 plays an essential role in arterioles but not in venules. *Significantly different from each other (P < 0.05); #significantly different from each other (P < 0.001); +significantly different from control group but not from each other (P < 0.001).

 
To determine whether there was a role for ICAM-1 in these rolling interactions in venules, we quantified leukocyte-rolling interactions in ICAM-1 KO mice. Despite the fact that in control venules of ICAM-1 KO animals the number of rolling leukocytes was about twofold higher than in WT mice [32 ± 1.9 (n = 13 sites) vs. 15 ± 1.0 (n = 15 sites) leukocytes/40 s, P < 0.001], the leukocyte-rolling fraction was not different from that of WT mice (Fig. 1B), suggesting that lack of ICAM-1 did not affect the leukocyte population that was rolling under these conditions. On the basis of previous work showing that all subsets of circulating leukocytes are increased two- to threefold over WT in these KO mice (6, 40), we inferred that the increase in rolling leukocytes that we observed was also made up of proportionate increases in all subpopulations. The fraction of rolling leukocytes was calculated by normalizing the number of rolling leukocytes to the total number of leukocytes observed in the proximity of the vessel wall (the leukocyte delivery as defined in MATERIALS AND METHODS) to account for the elevated number of circulating leukocytes in ICAM-1 KO animals. The difference in the number of rolling leukocytes between KO and WT mice could therefore be due to the elevated number of circulating leukocytes (about twofold, measured by hemocytometer from blood sampled via toe clip; data not shown). However, despite the much higher number of rolling leukocytes in control venules in the KO mice, TNF-{alpha} treatment produced a dramatic decrease, bringing the number of rolling leukocytes in venules to levels similar to those observed in WT animals [5.6 ± 0.9 (n = 13 sites) vs. 7 ± 0.7 (n = 15 sites) leukocytes/40 s, respectively, Fig. 1A]. Moreover, similar to control conditions, the fraction of rolling leukocytes in TNF-{alpha}-activated venules was not significantly different between ICAM-1 KO and WT mice (Fig. 1B). As expected, in TNF-{alpha}-activated venules in both KO and WT mice the fraction of rolling leukocytes was significantly reduced [73.9 ± 3.3 to 43.7 ± 3.0% (n = 15, WT) and 80.0 ± 1.3 to 38.5 ± 3.4% (n = 14, KO)] compared with control venules. These findings directly support previous conclusions (25, 28, 40) that in venules, ICAM-1 is not vital for sustaining leukocyte-rolling interactions.

In contrast to our findings in venules, our data indicate that ICAM-1 plays a significant role in leukocyte interactions with the arteriolar wall. In control conditions for both WT and ICAM-1 KO animals, leukocyte interactions (both rolling and adhesion) did not occur and were not quantified. However, activation with TNF-{alpha} was associated with the occurrence of leukocyte-rolling events in WT arterioles (12.0 ± 1.0 leukocytes/40 s, n = 15 sites). We established, using both intravenously administered fucoidin (25 mg/kg) or a combination of P-and L-selectin-blocking antibodies (50 µg/ml each), that 34% of arteriolar rolling was not due to selectin-mediated interactions (Fig. 2), directly supporting the hypothesis that ICAM-1 has an important role in leukocyte rolling in arterioles. Importantly, in venules, only 5% of leukocytes remained rolling following selectin block (Fig. 2), confirming the efficacy of the selectin block and indicating that other adhesion molecules only play a minimal role in sustaining leukocyte-rolling interactions in venules. In contrast to our observation in activated venules, in arterioles of KO animals, the number of rolling leukocytes in TNF-{alpha}-activated vessels was significantly lower (6.0 ± 0.9 leukocytes/40 s, n = 13 sites, P < 0.05, Fig. 1A) compared with WT animals. Similarly, the fraction of rolling leukocytes in TNF-{alpha}-activated arterioles was significantly lower in ICAM-1 KO mice compared with WT (37.8 ± 2.9 vs. 83.2 ± 2.2%, P < 0.001, Fig. 1B), again indicating a role for ICAM-1 in mediating leukocyte rolling. We conclude from these data that arterioles and venules regulate leukocyte-EC interactions differently, with ICAM-1 playing an essential role in arterioles and an effectively nonessential role in venules.


Figure 2
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 2. Arterioles exhibit selectin-independent leukocyte rolling. Mice were pretreated with TNF-{alpha} 4 h before data collection to induce leukocyte rolling in arterioles. Number of rolling leukocytes was quantified in selected arterioles before and following administration of either fucoidin or a combination of P- and L-selectin-blocking antibodies. Data were collected 5 min after intravenous drug administration. There was no significant difference between treatment groups; hence, pooled data from the fucoidin and selectin-blocking experiments are presented relative to the number of rolling leukocytes before treatment (n = 6 for each group). In arterioles, 34% of leukocyte rolling was independent of selectins, whereas in venules almost no rolling was observed following selectin block.

 
Average Leukocyte-Rolling Velocity and the Velocity-Distribution Profiles Differ in Arterioles and Venules of WT Mice Compared With Vessels That do not Express ICAM-1 on the EC Surface

The results summarized above, together with previously published data from our laboratory (42), clearly suggest that there is a significant role for ICAM-1 in mediating leukocyte rolling in arterioles. To explore this further, we quantified leukocyte-rolling velocities in TNF-{alpha}-activated arterioles in the presence or absence of ICAM-1 as an initial test of the hypothesis that the distribution of rolling velocities on the endothelial surface would reflect the distribution of ICAM-1 and that, in the absence of ICAM-1-mediated rolling, velocities of rolling leukocytes would typically be greater (i.e., ICAM-1 mediates a slow rolling component).

To test this concept, we quantified leukocyte-velocity distributions in WT and ICAM-1 KO mice. In each animal model, the velocities of 100 rolling leukocytes were measured in arterioles ranging from 30 to 60 µm in diameter, and velocity histograms were constructed. In support of the main thrust of our hypothesis, we found that both mean (or median) leukocyte-rolling velocities and velocity-distribution profiles were dramatically different in ICAM-1 KO mice compared with WT (Fig. 3, A and B). However, the change in mean velocity was not in the direction we expected. As illustrated in Fig. 3, we found that average leukocyte-rolling velocity in ICAM-1 KO arterioles was significantly decreased compared with WT (32.2 ± 1.7 vs. 53.5 ± 2.7 µm/s, P < 0.001, Fig. 3, A and B). Inspection of the histograms in Fig. 3 shows that in ICAM-1 KO animals, the rolling population has lost the fraction of leukocytes that roll with velocities >70 µm/s; in addition, the velocity-distribution profile in the KO mice has changed to bimodal, indicating that there are two populations of rolling leukocytes in the arterioles of these animals. We conclude from these data that ICAM-1 mediates rolling with a velocity that is characteristically high relative to the total population.


Figure 3
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 3. Leukocyte-rolling velocity distributions in arterioles from WT (A), ICAM-1 KO (B), and WT + ICAM-1 function-blocking antibody (WT + Ab; C) mice. Mice were pretreated with TNF-{alpha} 4 h before data collection. Velocity histograms were constructed by sampling velocities of 100 rolling leukocytes in arterioles ranging from 30 to 60 µm in diameter. Average leukocyte-rolling velocity in ICAM-1 KO arterioles was significantly decreased compared with WT (P < 0.001), and velocity-distribution profile in KO mice became bimodal, indicating that there are 2 populations of rolling leukocytes. In presence of an ICAM-1-blocking antibody, the velocity profile in WT mice was similar to the skewed shape of distribution from unblocked WT arterioles, with consistently lower average rolling velocity (n = 10 sites, P < 0.05) compared with unblocked WT animals.

 
Previously published data using similar experimental procedures (19, 40, 41) determined that ICAM-1 can act synergistically with selectins to mediate optimal leukocyte rolling in venules. In agreement with these studies, we found that average leukocyte-rolling velocity measured in venules of ICAM-1 KO mice under control conditions was significantly higher than that of WT mice (45.7 ± 1.9 vs. 39.8 ± 1.3 µm, 13 leukocytes/site, n = 10 sites, P < 0.001), although the distributions were not different (Fig. 4, A and C). As expected, after TNF-{alpha} stimulation, both WT and KO mice had significantly reduced mean leukocyte-rolling velocities compared with unactivated control venules (Fig. 4). Importantly, in ICAM-1 KO mice these rolling velocities were significantly higher than in venules from WT mice (19.4 ± 1.0 vs. 11.6 ± 0.5 µm, 13 leukocytes/site, n = 10 sites, P < 0.001). These data provide strong support for the concept that in venules, interactions between beta2-integrins and ICAM-1 contribute to the slower leukocyte rolling (Fig. 4, B and D) that eventually results in firm arrest, and given our finding that in arterioles, ICAM-1-mediated rolling has a higher characteristic velocity, the data again emphasize that the regulation of leukocyte rolling is fundamentally different in arterioles and venules. Our finding that the rolling population in venules of KO mice has a high-velocity component that is absent in their WT counterparts is consistent with the hypothesis that other adhesion molecules (selectins) on the venular surface have been engaged and contribute to the observed rolling interactions.


Figure 4
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 4. Leukocyte-rolling velocity distributions in venules from WT and ICAM-1 KO mice in control conditions (A and C) and with TNF-{alpha} (B and D). Mice were pretreated with TNF-{alpha} 4 h before data collection where indicated. Velocity histograms were constructed by sampling velocities of 100 rolling leukocytes in venules ranging from 20 to 60 µm in diameter. Average leukocyte-rolling velocity measured in KO mice under control and TNF-{alpha}-stimulation conditions was significantly higher than that of WT mice (n = 10 sites, P < 0.05 and P < 0.001, respectively), although distributions were not different.

 
Because the characteristics of rolling in arterioles contrasted with our hypothesis and, furthermore, were opposite to those that we observed in venules, we wondered whether the fact that the sampled arteriolar population was limited with respect to size (30- to 60-µm diameter) had influenced our results (for example, due to possible differences in adhesion-molecule expression or hemodynamic conditions in arterioles from different branch orders). Thus we evaluated whether the overall decrease in rolling velocity that we observed in arterioles of KO mice compared with WT was true for a wider range of vessel sizes. We found that leukocyte-rolling velocity was consistently lower throughout the entire range of sampled arterioles (25–85 µm) and was not different in arterioles of different sizes (data not shown). This reinforces the suggestion that in the KO mice, lower average rolling velocities and the bimodal velocity profile, are likely due to a compensatory shift in the expression of other adhesion molecules following genetic intervention (knocking out the ICAM-1 gene). We tested this idea further by quantifying rolling velocities in TNF-{alpha}-activated arterioles of WT mice in the presence of an ICAM-1-blocking antibody. Under this condition, the velocity profile lost the high-velocity component but retained the skewed shape of the velocity distribution, similarly to that observed in WT arterioles (Fig. 3C). Likewise, the average rolling velocity in the ICAM-1-blocked WT arterioles remained consistently lower than that in untreated WT animals (38.4 ± 1.9 µm/s, 10 leukocytes/site, n = 10 sites, P < 0.001), giving further support to our earlier conclusion that ICAM-1 supports leukocyte rolling at a high velocity relative to the total rolling population.

Leukocyte Adhesion in ICAM-1 KO Arterioles and Venules

The observation that there was effectively no leukocyte adhesion in TNF-{alpha}-activated arterioles of WT mice underlay our hypothesis that ICAM-1 plays an important role in leukocyte rolling in arterioles. In the previous section, we demonstrated the importance of ICAM-1 in leukocyte-rolling behavior by using both the ICAM-1 KO model and a function-blocking ICAM-1 antibody. Despite the significantly lower fraction of interacting leukocytes in TNF-{alpha}-activated arterioles, a closer look revealed additional phenomena. Whereas numbers of leukocytes adhering to the arteriolar wall of both KO and WT mice were extremely low, the number of adhered leukocytes in the arterioles of KO mice was nevertheless 3.2-fold higher than in WT mice (2.1 ± 0.3 vs. 0.64 ± 0.2 leukocyte/100 µm, n = 12 vessel sites, P < 0.05, Fig. 5). Clearly, these adhesion events in the ICAM-1 KO animals are not ICAM-1 mediated. The observation illustrates how the maintenance of balance between adhesion molecules expressed on the endothelium under different conditions determines leukocyte-EC interactions. Moreover, it further supports the concept that in the KO animals, other adhesion molecules are upregulated, and, from the observed increase in numbers of adhesion events over WT, we infer that upregulated molecules may be even more potent than ICAM-1 in mediation of adhesion in arterioles. Published findings on the role of VCAM-1 in mediation of adhesion (32, 36) suggest that this molecule is a likely candidate for mediation of adhesion in TNF-{alpha}-activated arterioles of the KO mice. This will be addressed in a later section.


Figure 5
View larger version (9K):
[in this window]
[in a new window]

  		Fig. 5. Leukocyte adhesion in ICAM-1 KO arterioles and venules. All mice were pretreated with TNF-{alpha} 4 h before data collection. Number of adhered leukocytes in WT arterioles treated with ICAM-1-blocking antibody (wt +Ab) was not different from untreated WT arterioles, indicating that in KO mice increase in adhesion is due to upregulation of another adhesion molecule. In venules, leukocyte adhesion was significantly decreased in both KO and WT + Ab mice. *Different from WT and each other, P < 0.05; #different from WT, P < 0.01.

 
To further explore the possibility that the genetic modification resulted in upregulation and involvement of other adhesion molecules in leukocyte-EC interactions in ICAM-1 KO mice, we extended previously published work from our laboratory (42), in which we showed that leukocyte adhesion to the venular wall was significantly reduced but not abolished in ICAM-1 KO mice. In Fig. 5 we show, in confirmation of our earlier finding and the work of others (5, 6, 10), that in ICAM-1 KO mice adhesion to the venular wall was significantly reduced compared with WT mice, although not abolished (3.8 ± 0.2 vs. 5.2 ± 0.4 leukocytes/100 µm, respectively, n = 15 sites, P < 0.01). Using ICAM-1-blocking antibody in WT mice, hence preserving the balance of normal adhesion-molecule expression, we found that in TNF-{alpha}-activated venules of WT mice, leukocyte adhesion was even further reduced relative to controls (2.5 ± 0.3 leukocytes/100 µm, n = 15 sites, P < 0.01, Fig. 5) but was still not abolished. This new evidence clearly shows that although ICAM-1 indeed has an important role in leukocyte adhesion in venules, it is clear that other adhesion molecules contribute to these interactions.

Overall, data from the ICAM-1 KO animals, together with the use of a function-blocking antibody in WT animals, has provided insight into the contribution of ICAM-1 to leukocyte-endothelial interactions in both arterioles and venules. These findings provide clear evidence that ICAM-1 is involved in mediation of leukocyte rolling in arterioles. Our next goal was to define more closely the role of ICAM-1 in determining the extent of leukocyte-rolling interactions with the arteriolar wall.

Heterogeneity of Leukocyte Rolling on Arteriolar ECs is ICAM-1 Dependent

As we showed in Fig. 1, in arterioles but not in venules, the fraction of rolling leukocytes was decreased in the absence of ICAM-1, implying that ICAM-1 contributes to these interactions. We hypothesized that this dependence will be a function of a leukocyte's ability to interact with the endothelium, as well as the duration of these interactions. To asses this, we first measured the distance leukocytes actually remained in rolling contact with the arteriolar wall in WT mice, ICAM-1 KO mice, and WT mice treated with ICAM-1-blocking antibody. Figure 6 shows that in ICAM-1 KO animals leukocytes rolled significantly shorter distances over the sampled 200-µm vessel length compared with WT (68 ± 6.7 vs. 85 ± 12.9% total, respectively, n = 4 sites, P < 0.05). Moreover, in WT vessels in the presence of a monoclonal ICAM-1-blocking antibody, the distances rolled by leukocytes became even shorter (55 ± 9.4%, n = 4 sites, P < 0.05, Fig. 6), implying that the presence of ICAM-1 on the surface of the ECs is crucial to sustain a significant fraction of the observed leukocyte rolling in arterioles. We note that the higher mean rolling distance in ICAM-1 KO mice compared with WT mice in which ICAM-1 was blocked by using an antibody, again supports the conclusion that other adhesion molecules capable of mediating leukocyte-EC rolling interactions have been upregulated in ICAM-1 KO mice, as we suggested in the previous section.


Figure 6
View larger version (8K):
[in this window]
[in a new window]

  		Fig. 6. Ability of leukocytes to roll on arteriolar ECs is ICAM-1 dependent. Leukocytes were tracked over 200-µm segments of arteriolar wall, and the distance that leukocytes remained in continuous rolling contact with endothelium was measured in WT mice, ICAM-1 KO mice, and WT mice treated with an ICAM-1-blocking antibody. In ICAM-1 KO animals and in WT mice in presence of blocking antibody, leukocytes rolled significantly shorter distances over the sampled 200-µm vessel length compared with WT (15 leukocytes/site, n = 4 sites, P < 0.05), implying that presence of ICAM-1 on surface of ECs contributes significantly to observed leukocyte rolling. *Different from WT and each other, P < 0.05.

 
In a second protocol, we tracked the velocity of all rolling leukocytes in consecutive 5-µm sections of sampled 200-µm lengths of vessel wall in arterioles from WT, ICAM-1 KO, and WT + ICAM-1-blocking antibody animals (Fig. 7). In WT arterioles, the majority of rolling leukocytes maintained continuous rolling over the entire sampled 200-µm distance, and a variety of rolling behavior such as slow rolling, fast rolling, and high-velocity translation were observed (Fig. 7A), consistent with the idea that different regions of the wall have different characteristic adhesion-molecule properties that determine rolling kinetics.


Figure 7
View larger version (9K):
[in this window]
[in a new window]

  		Fig. 7. Leukocyte-rolling behavior is affected by absence of ICAM-1. Velocity of all rolling leukocytes was tracked in consecutive 5-µm lengths of sampled 200-µm lengths of vessel wall in arterioles from WT (A), ICAM-1 KO (B), and WT + ICAM-1-blocking antibody animals (C). In WT arterioles, the majority of rolling leukocytes maintained continuous rolling over entire sampled 200 µm. In ICAM-1 KO mice and when ICAM-1 was blocked in WT mice, leukocyte rolling was interrupted by frequent detachments from endothelium, where leukocytes disappeared from view and reemerged further along the vessel to continue rolling.

 
In ICAM-1 KO mice, both slow and fast rolling velocities were observed, but the continuity of rolling as well as the highest velocities were lost. Under these conditions, leukocyte rolling was interrupted by frequent detachments from the endothelium, where leukocytes disappeared from view and emerged back after a period of time to continue rolling (Fig. 7B). We conclude that the absence of ICAM-1 from regions of the wall is responsible for the lack of leukocyte-rolling interactions in these regions.

Furthermore, when ICAM-1 was blocked in WT mice, the gaps where leukocytes disappeared from the view became longer (Fig. 7C). In other words, in arterioles from both ICAM-1 KO and WT + ICAM-1-blocking antibody-treated animals, the high velocities were lost and the leukocyte's ability to roll at any location on the arteriolar ECs was impaired. These findings strongly support the idea that leukocytes roll on ICAM-1-expressing regions of the arteriolar wall at a characteristic, relatively high velocity.

Expression of E-Selectin, P-Selectin, and VCAM-1 in ICAM-1 KO Compared With WT Mice

To establish the relative expression levels of the major adhesion molecules other than ICAM-1 that might mediate rolling (E- and P-selectin) and adhesion (VCAM-1), we used a previously established immunofluorescent-labeling technique as earlier described (22, 42). Interestingly, and confirming our hypothesis that surface expression of other molecules capable of mediating leukocyte-endothelial interactions was upregulated in the ICAM-1 KO mice, we found that VCAM-1 fluorescence intensity was much higher in both arterioles and venules of ICAM-1 KO mice (Fig. 8A) relative to vessels of WT mice. In agreement with previously published data (32), we found that basal VCAM-1 expression in control WT arterioles was very low [measured fluorescence intensity was not significantly different from background fluorescence (25.6 ± 1.0 vs. 29.2 ± 1.4, respectively, P < 0.05)], but the intensity significantly increased (1.7-fold) following TNF-{alpha} activation. Arterioles in ICAM-1 KO animals, on the other hand, displayed VCAM-1 intensity that was 2.7-fold higher than background (P < 0.05) under control conditions. Following TNF-{alpha} activation, VCAM-1 fluorescence intensity increased an additional 2.1-fold, which was significantly higher than in control arterioles of KO mice (P < 0.05) and significantly higher than the expression in activated arterioles of WT mice (P < 0.05). Similarly, in venules of control KO animals, the relative expression of VCAM-1 was 2.0-fold higher than that measured in control WT venules, and an additional 2.1-fold increase followed TNF-{alpha} activation (Fig. 8A). The relative VCAM-1 expression level in TNF-{alpha}-activated venules was significantly higher than in control venules of KO mice (P < 0.05) and was significantly higher than that in activated venules of WT mice (P < 0.05).


Figure 8
View larger version (43K):
[in this window]
[in a new window]

  		Fig. 8. Expression of VCAM-1 but not E- or P-selectin was upregulated in ICAM-1 KO mice. Vessels were perfused with appropriate primary and secondary fluorescent polyclonal antibody in sequence for 15 min each (see MATERIALS AND METHODS). VCAM-1 (A), P-selectin (B), and E-selectin (C) relative expression was quantified as a function of fluorescence intensity. Relative expression is presented as intensity units (grayscale) normalized to intensity of a known concentration of FITC-dextran (see MATERIALS AND METHODS). Expression of VCAM-1 in both arterioles and venules was highly upregulated under control and TNF-{alpha} conditions in ICAM-1 KO mice compared with WT mice. *Different from WT, P < 0.05; +different from control group, P < 0.05. Images below each graph are corresponding representative fluorescently labeled venules (top) and arterioles (bottom) taken from ICAM KO mice following TNF-{alpha} stimulation. Bar on left-most venule image is 10 µm and is the same for all images. Note difference in labeling patterns among adhesion molecules.

 
These results provide strong support for our hypothesis that in the absence of functional ICAM-1, upregulation of other adhesion molecules enables maintenance of proper leukocyte trafficking. The findings reported in this and earlier sections indicate that VCAM-1 is a likely candidate molecule for maintenance of increased leukocyte adhesion in arterioles of ICAM-1 KO mice.

Since leukocyte rolling was a major feature in TNF-{alpha}-activated arterioles (Fig. 1), we asked whether this could be a result of different expression level of molecules known to be prominent mediators of leukocyte rolling, e.g., E- and P-selectin.

We found that for both E-selectin and P-selectin, fluorescence intensity was almost not detectable in both WT and KO arterioles in control conditions (0.28 ± 0.02 and 0.27 ± 0.02 intensity units, respectively, for P-selectin; 0.26 ± 0.01 and 0.32 ± 0.01 intensity units, respectively, for E-selectin) but their expression significantly increased approximately twofold in both WT and KO mice [to 0.51 ± 0.03 and 0.52 ± 0.04, P < 0.05 (WT and KO), respectively, for P-selectin and to 0.51 ± 0.02 and 0.50 ± 0.02, P < 0.05 (WT and KO), respectively, for E-selectin] following TNF-{alpha} treatment (Fig. 8, B and C). We also found that the expression levels of P-selectin, but not E-selectin, in control venules (0.47 ± 0.02 and 0.50 ± 0.04 intensity units, WT and KO, respectively, for P-selectin and 0.34 ± 0.02 and 0.29 ± 0.01 intensity units, WT and KO, respectively, for E-selectin) of both WT and ICAM-1 KO mice were significantly higher (P < 0.05) compared with control arterioles of WT and ICAM-1 KO mice, respectively. Following TNF-{alpha} treatment, venules (as expected) showed a 1.4-fold significant (P < 0.05) increase in the relative expression of both molecules (0.76 ± 0.06 and 0.70 ± 0.03 intensity units, WT and KO, respectively, for P-selectin and 0.51 ± 0.03 and 0.57 ± 0.03 intensity units, WT and KO, respectively, for E-selectin).

Importantly, the expression of both selectin molecules in control and TNF-{alpha}-activated arterioles and venules was not significantly different in ICAM-1 KO compared with WT mice. Thus these data suggest that the differences in rolling behavior in ICAM-1 KO compared with WT mice cannot be due to differences in the expression of either of these two selectins. Our findings suggested that in addition to the differences in leukocyte adhesion, the differences in leukocyte-rolling behavior between WT and KO animals may also be attributed to changes in the expression of VCAM-1.

VCAM-1 May Be Responsible for Increased Leukocyte Adhesion and the Differences in Leukocyte-Rolling Velocities

To address this, we conducted experiments to quantify leukocyte-endothelial interactions in which the protocols were similar to those described earlier but using a VCAM-1 function-blocking antibody in place of the anti-ICAM-1 antibody used earlier.

In confirmation of our hypothesis, by blocking VCAM-1 in the arterioles of ICAM-1 KO animals, we were able to significantly reduce the number of adhered leukocytes to a level similar to that observed in arterioles of WT mice (0.69 ± 0.2 vs. 0.64 ± 0.2 leukocytes/100 µm, Fig. 9A). Furthermore, the leukocyte-velocity distribution profile in TNF-{alpha}-activated arterioles of ICAM-1 KO mice following VCAM-1 block was restored to a skewed shape, similar to that observed in WT arterioles in the presence of ICAM-1-blocking antibody (Figs. 9B and 3C vs. Fig. 3B). The mean leukocyte velocities in both cases were similar to each other but were significantly different from measured velocities in untreated WT mice [37 ± 1.6 (WT + VCAM-1 block) and 36.2 ± 1.5 (WT + ICAM-1 block) vs. 53.5 ± 2.8 (WT) µm/s, P < 0.001], again supporting our hypothesis that VCAM-1 plays a role both processes.


Figure 9
View larger version (9K):
[in this window]
[in a new window]

  		Fig. 9. VCAM-1 affects leukocyte-rolling behavior and adhesion in ICAM-1 KO mice. All mice were pretreated with TNF-{alpha} 4 h before data collection. Arterioles were treated with VCAM-1 function-blocking antibody (100 µg in PBS) as described in MATERIALS AND METHODS. A: number of adhered leukocytes per 100-µm vessel length (n = 9 sites) in arterioles of WT mice, ICAM-1 KO mice, and ICAM-1 KO mice treated with function-blocking VCAM-1 antibody (WT + Ab). B: velocity-distribution histogram constructed by sampling velocities of 100 rolling leukocytes in arterioles (n = 10 sites) of ICAM-1 KO mice following treatment with VCAM-1-blocking antibody. In presence of VCAM-1-blocking antibody, number of adhered leukocytes in KO mice was significantly decreased to levels similar to those observed in untreated WT mice. Mean leukocyte-rolling velocity in KO + VCAM-1 Ab mice was not different from WT mice treated with ICAM-1-blocking antibody. Velocity-distribution profile in VCAM-1-blocked ICAM-1 KO mice shifted from bimodal (Fig. 3B) to a skewed shape that was not different from WT or ICAM-1-blocked WT mice (Fig. 3, A and C). *Different from WT, P < 0.05.

 
To address the speculation from our previous publication (42), where we suggested that the absence of leukocyte adhesion in TNF-{alpha}-activated arterioles might be due to insufficient expression of ICAM-1 in these vessels, and to further explore VCAM-1-mediated compensatory leukocyte adhesion in ICAM-1 KO mice, we looked at the leukocyte adhesion in the same regions of selected arterioles under different shear conditions. We found that reducing the shear rate resulted in increased leukocyte adhesion in control arterioles in WT but not ICAM-1 KO mice [Fig. 10A; note the absence of leukocyte adhesion under initial (basal) shear conditions as discussed earlier]. This finding clearly indicates that ICAM-1 has the capacity to mediate leukocyte adhesion in control (unstimulated) arterioles under appropriate shear conditions. Clearly, both adhesion molecules and the shear environment together play a role in leukocyte adhesion; in the absence of sufficient levels of adhesion molecules, the prevailing force that determines a leukocyte's ability to adhere is the hemodynamic environment.


Figure 10
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 10. Local shear rate influences leukocyte adhesion in arterioles. In arterioles of WT and ICAM-1 KO mice in either control or TNF-{alpha}-activated conditions (A and B) and in ICAM-1 KO mice with TNF-{alpha} ± VCAM-1-blocking antibody (C), shear rate was reduced in 4 consecutive steps by placing an occluding rod upstream from site of data collection. Shear rate and leukocyte interactions were measured at each step. Reducing the shear rate resulted in increased leukocyte adhesion in control arterioles in WT but not ICAM-1 KO mice. Following TNF-{alpha} treatment, reduction in local shear rate increased the number of adhered leukocytes in WT vessels, and, importantly, a significant number of adhered leukocytes were found in KO mice. This increase was blocked by VCAM-1-blocking antibody.

 
When the same experiment was performed in TNF-{alpha}-treated mice, we observed contrary results (Fig. 10B). Reducing the shear rate increased the number of adhered leukocytes in WT as expected, but more importantly, a significant number of adhered leukocytes were found in KO mice, again implying that there was VCAM-1-mediated leukocyte adhesion. To confirm that the increase in leukocyte adhesion in ICAM-1 KO mice was indeed VCAM-1 mediated, we explored the consequences of blocking VCAM-1 function (using a blocking antibody) in the ICAM-1 KO mice. VCAM-1 blockade resulted in significant reduction in the number of adhered leukocytes (Fig. 10C), confirming our hypothesis.


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
This study provides evidence that in inflamed arterioles, ICAM-1 mediates a significant amount (one third) of leukocyte rolling, unlike in venules, where almost all leukocyte rolling is mediated by the selectin family of adhesion molecules. We show that the fraction of leukocytes rolling on arteriolar ECs is decreased in ICAM-1 KO mice, whereas the leukocyte-rolling fraction in venules remains unaffected. We also show that the fraction of leukocytes that is rolling on arteriolar ECs does so with a higher characteristic velocity, and, furthermore, that the distance over which rolling contacts with the arteriolar wall are maintained is ICAM-1 dependent. We found evidence that in ICAM-1 KO mice, a significant fraction of leukocyte rolling and adhesive interactions with arteriolar ECs could be accounted for by upregulation of another adhesion molecule, VCAM-1, providing an important illustration of how expression of related proteins can be altered following genetic ablatement of a target molecule (in this case, ICAM-1). This finding confirms and extends previously published work, which showed that chronic deficiency of one adhesion molecule can alter the expression of other related molecules on ECs (7).

The role of ICAM-1 has been widely explored in venular microcirculation. ICAM-1 has been identified as an important molecule in mediation of leukocyte adhesion (29) and has been shown to contribute to stable leukocyte rolling on selectin molecules (19). In this study, we confirmed its importance in leukocyte adhesion in venules following proinflammatory stimuli as well as its possible contribution to selectin mediating rolling. Importantly, in this study we explore the differences between arterioles and venules in the regulation of leukocyte-EC interactions via adhesion molecules and show that ICAM-1 is important in arterioles to sustain leukocyte rolling.

In previous work (42, 46), we showed that although in venules leukocytes roll under control conditions and undergo firm adhesion followed by transmigration during inflammation, in arterioles leukocyte rolling can be observed following TNF-{alpha} stimulation but no adhesion or transmigration occurs. Leukocyte rolling in arterioles is not followed by significant adhesion or transmigration despite a dramatic increase in the expression of ICAM-1 in TNF-{alpha}-activated arterioles as detected by immunofluorescent labeling. This suggests to us that ICAM-1-dependent leukocyte interactions with the arteriolar wall may have outcomes other than immediate egress of leukocytes from these microvessels into the tissue. For example, as was mentioned earlier, arterioles have the capacity to participate in inflammatory responses by increasing their solute permeability in response to appropriate inflammatory stimuli (15, 39), and hence it is reasonable to speculate that leukocyte-EC interactions in arterioles might subserve these or other related responses. Although a direct test of this idea is beyond the scope of the present study, it clearly warrants exploration in future work. We also showed that another major difference between arterioles and venules is the expression levels and patterns of adhesion molecules, such as ICAM-1, which also contribute directly to the observed differences in leukocyte behavior (42).

We found that the fraction of rolling leukocytes in TNF-{alpha}-activated arterioles was significantly reduced in ICAM-1 KO mice, supporting our hypothesis that ICAM-1 has an explicit role in leukocyte rolling in arterioles. Unexpectedly, we discovered that leukocyte-rolling velocities were significantly reduced in ICAM-1 KO mice over a wide size range (25- to 85-µm diameter) of sampled vessels. This observation contradicts the accepted model generated in venules, where ICAM-1 is thought to contribute to a selectin-mediated decrease in leukocyte-rolling velocity to allow firm adhesion (40). The fact that our venular data agree with that model supports our earlier speculation that different mechanisms underlie leukocyte-EC interactions in arterioles vs. venules, allowing different roles for the same molecule (ICAM-1) in these two vessel types. It is presumed that in arterioles and venules the ICAM-1 molecules are structurally identical, but differences in biological effects could arise from several possible sources. We previously reported (42) that the surface density of ICAM-1 differs between arterioles and venules, which we speculate could in turn be associated with activation of different signaling cascades. Recently, it was reported that different conformations of beta2-integrins on leukocytes have differing affinities for ICAM-1 and thus mediate rolling vs. adhesive interactions (38). Thus if the conformational state of beta2-integrins is altered by local hemodynamic conditions, which are characteristically different in arterioles and venules, this could lead to rolling in one environment and adhesion in the other. Finally, it is reasonable to assume that, given that there are established differences in phenotype between arteriolar and venular ECs (12), ICAM-1 ligation may in each case lead to different downstream signals and, ultimately, different responses.

We also found that the rolling-velocity distribution profile in ICAM-1 KO mice changed to bimodal, suggesting that in these mice leukocytes divided into populations of slow-rolling (19.3 ± 3.5 µm/s) and fast-rolling (49.4 ± 7.0 µm/s) cells. Based on previously published literature, which assigns characteristically slow rolling to E-selectin (18, 37) and faster rolling to P-selectin (2, 18), we hypothesized that in the absence of ICAM-1, the slow and faster populations observed in ICAM-1 KO mice would be due to upregulated expression of E- and P-selectin, respectively. Contrary to our expectation, immunostaining of these molecules revealed that the relative expression of both molecules was not significantly different in ICAM-1 KO compared with WT mice. Although it is possible that in the absence of ICAM-1 both E- and P-selectin molecules exercise greater influence on leukocyte rolling, thus changing the velocity-distribution profiles compared with WT vessels, we suggest that it is more likely that the differences in leukocyte-rolling behavior between WT and KO animals may be attributable to observed changes in the relative expression of VCAM-1. We conclude this from results presented in Fig. 8B, which shows that following VCAM-1 block the bimodal velocity distribution was smoothed and resembled that in WT in the presence of ICAM-1-blocking antibody. In further support of this conclusion, others have shown that VCAM-1 is capable of mediating leukocyte rolling in conduit arteries (3, 32) and EC monolayers (9) and furthermore can affect leukocyte-rolling velocity (36).

Interestingly, the observed expression patterns and localization of these three adhesion molecules are very different (Fig. 8). These differences are likely contributing significantly to the differences in their role in mediating leukocyte rolling, as mentioned above. VCAM-1, similarly to ICAM-1 (42), exhibits heterogeneity in its expression on the EC surface and appears to be excluded from cell borders. P-selectin uniquely exhibits a punctuate distribution as documented previously (13, 22), and, finally, E-selectin is present on the entire EC surface, including the junctional regions.

We observed a significant increase in leukocyte adhesion in TNF-{alpha}-activated arterioles of ICAM-1 KO mice (Fig. 9A). It is possible that KO mice are more susceptible to inflammation (TNF-{alpha} treatment) or, alternatively, that leukocytes in KO mice are constantly in a more activated state, but these alternatives appear unlikely because under similar conditions we saw a decrease in leukocyte adhesion in venules. We suggest that the increase in adhesion in arterioles was mediated by VCAM-1, because the increase could be blocked with blocking VCAM-1 antibody in the ICAM-1 KO animals (Fig. 9A). The finding that increased VCAM-1 expression levels could support leukocyte adhesion in arterioles extends one of our conclusions from previously published data (42), where we suggested that one of the reasons for the lack of leukocyte adhesion in arterioles was that there was an insufficient expression level of ICAM-1.

On average, rolling leukocyte velocity is lower in ICAM-1 KO mice. Even if we discard velocities >70 µm/s from the WT population to match the velocity range seen in KO mice (Fig. 3B), the mean velocity in WT arterioles remains significantly higher than in KO vessels (39.5 ± 1.9 vs. 32.2 ± 1.7 µm/s, P < 0.001). Although the absolute numbers are small, we found on average that there were two adhered leukocytes per 100-µm vessel wall in ICAM-1 KO mice, which was significantly higher than in WT arterioles. We asked whether this density of adhered leukocytes could potentially disturb the blood flow (shear rate) and/or serve as a physical barrier that rolling leukocytes have to overcome. Moreover, potential leukocyte-leukocyte interactions (via L-selectin) could play a role in slowing down passing leukocytes. Preliminary observations (data not shown) suggested that the velocities of rolling leukocytes following an encounter with a leukocyte adhered to the surface of the vessel wall were indeed significantly lower than the velocities before the encounter, supporting the idea that leukocyte-leukocyte interactions may have contributed to the reduction in leukocyte velocity in ICAM-1 KO mice.

Our study has identified a role for ICAM-1 in leukocyte-EC interactions in inflamed arterioles and confirms that this appears to be separate from its contribution to adhesion and subsequent transmigration of leukocytes in the venular microcirculation. Thus an important question that remains unanswered is what is the physiological consequence(s) of ICAM-1-mediated leukocyte rolling and upregulation of ICAM-1 expression in arterioles.

ICAM-1 has been identified as a signaling molecule, capable of mediating changes in cellular functions via outside-inside signaling (14) that may contribute to other aspects of the inflammatory response. For example, ICAM-1-dependent changes in intracellular free Ca2+, myosin contractility, and induction of tyrosine phosphorylation of cytoskeletal proteins that are involved in rearrangement of the interendothelial junctions (33) have been identified, thus raising the possibility that ICAM-1 may contribute to changes in arteriolar barrier function, perhaps as part of an inflammatory response in atherosclerotic regions. It is also known that ICAM-1 ligation increases EC stiffness, causing increased production of reactive oxygen species (44), again implying that it has some role in an inflammatory cascade. We speculate that such a signaling contribution to inflammatory responses is indeed the main role of ICAM-1 in arterioles; clearly, further studies will be required to address these ideas.

In summary, using a KO mouse model together with function-blocking antibodies has allowed us to better understand the role of ICAM-1 in arterioles. The differences in leukocyte behavior observed in ICAM-1 KO mice and in WT mice treated with ICAM-1-blocking antibody emphasize how maintenance of the balance between expression levels of different adhesion molecules on the EC surface is a key process in regulation of leukocyte-EC interactions. We conclude that first, ICAM-1 mediates leukocyte rolling in arterioles at characteristically high velocities and is important in determining the duration of these interactions; second, changes in adhesion density and rolling velocity distributions in ICAM-1 KO arterioles are due to a compensatory shift in the expression of VCAM-1; and third, although in venules the major role of ICAM-1 is its direct contribution to leukocyte recruitment to the tissue during inflammation, in arterioles, ICAM-1 has a different role that appears most likely to be related to other inflammatory signaling mechanisms that are a function of ICAM-1 ligation by rolling leukocytes.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by National Heart, Lung, and Blood Institute Grants HL-18208 and HL-75186 and American Heart Association Grant 0615677T.


   ACKNOWLEDGMENTS
 
We thank J. M. Kuebel and P. A. Titus for their experience and contributions.


   FOOTNOTES
 

Address for reprint requests and other correspondence: I. H. Sarelius, Dept. of Pharmacology and Physiology, Univ. of Rochester, Medical Center, Box 711, Rochester, NY 14642 (e-mail: ingrid_sarelius{at}urmc.rochester.edu)

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
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Anderson ME, Siahaan TJ. Targeting ICAM-1/LFA-1 interaction for controlling autoimmune diseases: designing peptide and small molecule inhibitors. Peptides 24: 487–501, 2003.[CrossRef][Web of Science][Medline]
  2. Borgstrom P, Hughes GK, Hansell P, Wolitsky BA, Sriramarao P. Leukocyte adhesion in angiogenic blood vessels. Role of E-selectin, P-selectin, and beta2 integrin in lymphotoxin-mediated leukocyte recruitment in tumor microvessels. J Clin Invest 99: 2246–2253, 1997.[Web of Science][Medline]
  3. Chen C, Mobley JL, Dwir O, Shimron F, Grabovsky V, Lobb RR, Shimizu Y, Alon R. High affinity very late antigen-4 subsets expressed on T cells are mandatory for spontaneous adhesion strengthening but not for rolling on VCAM-1 in shear flow. J Immunol 162: 1084–1095, 1999.[Abstract/Free Full Text]
  4. Dal Secco D, Moreira AP, Freitas A, Silva JS, Rossi MA, Ferreira SH, Cunha FQ. Nitric oxide inhibits neutrophil migration by a mechanism dependent on ICAM-1: role of soluble guanylate cyclase. Nitric Oxide 15: 77–86, 2006.[CrossRef][Web of Science][Medline]
  5. Dunne JL, Ballantyne CM, Beaudet AL, Ley K. Control of leukocyte rolling velocity in TNF-alpha-induced inflammation by LFA-1 and Mac-1. Blood 99: 336–341, 2002.[Abstract/Free Full Text]
  6. Dunne JL, Collins RG, Beaudet AL, Ballantyne CM, Ley K. Mac-1, but not LFA-1, uses intercellular adhesion molecule-1 to mediate slow leukocyte rolling in TNF-alpha-induced inflammation. J Immunol 171: 6105–6111, 2003.[Abstract/Free Full Text]
  7. Eppihimer MJ, Russell J, Anderson DC, Wolitzky BA, Granger DN. Endothelial cell adhesion molecule expression in gene-targeted mice. Am J Physiol Heart Circ Physiol 273: H1903–H1908, 1997.[Abstract/Free Full Text]
  8. Etienne-Manneville S, Manneville JB, Adamson P, Wilbourn B, Greenwood J, Couraud PO. ICAM-1-coupled cytoskeletal rearrangements and transendothelial lymphocyte migration involve intracellular calcium signaling in brain endothelial cell lines. J Immunol 165: 3375–3383, 2000.[Abstract/Free Full Text]
  9. Feigelson SW, Grabovsky V, Shamri R, Levy S, Alon R. The CD81 tetraspanin facilitates instantaneous leukocyte VLA-4 adhesion strengthening to vascular cell adhesion molecule 1 (VCAM-1) under shear flow. J Biol Chem 278: 51203–51212, 2003.[Abstract/Free Full Text]
  10. Foy DS, Ley K. Intercellular adhesion molecule-1 is required for chemoattractant-induced leukocyte adhesion in resting, but not inflamed, venules in vivo. Microvasc Res 60: 249–260, 2000.[CrossRef][Web of Science][Medline]
  11. Gaboury JP, Kubes P. Reductions in physiologic shear rates lead to CD11/CD18-dependent, selectin-independent leukocyte rolling in vivo. Blood 83: 345–350, 1994.[Abstract/Free Full Text]
  12. Garcia-Cardena G, Comander J, Anderson KR, Blackman BR, Gimbrone MA Jr. Biomechanical activation of vascular endothelium as a determinant of its functional phenotype. Proc Natl Acad Sci USA 98: 4478–4485, 2001.[Abstract/Free Full Text]
  13. Hattori R, Hamilton KK, Fugate RD, McEver RP, Sims PJ. Stimulated secretion of endothelial von Willebrand factor is accompanied by rapid redistribution to the cell surface of the intracellular granule membrane protein GMP-140. J Biol Chem 264: 7768–7771, 1989.[Abstract/Free Full Text]
  14. Hubbard AK, Rothlein R. Intercellular adhesion molecule-1 (ICAM-1) expression and cell signaling cascades. Free Radic Biol Med 28: 1379–1386, 2000.[CrossRef][Web of Science][Medline]
  15. Huxley VH, Williams DA. Role of a glycocalyx on coronary arteriole permeability to proteins: evidence from enzyme treatments. Am J Physiol Heart Circ Physiol 278: H1177–H1185, 2000.[Abstract/Free Full Text]
  16. Issekutz AC, Rowter D, Springer TA. Role of ICAM-1 and ICAM-2 and alternate CD11/CD18 ligands in neutrophil transendothelial migration. J Leukoc Biol 65: 117–126, 1999.[Abstract]
  17. Jenks SA, Eisfelder BJ, Miller J. LFA-1 co-stimulation inhibits T(h)2 differentiation by down-modulating IL-4 responsiveness. Int Immunol 17: 315–323, 2005.[Abstract/Free Full Text]
  18. Jung U, Ley K. Mice lacking two or all three selectins demonstrate overlapping and distinct functions for each selectin. J Immunol 162: 6755–6762, 1999.[Abstract/Free Full Text]
  19. Kadono T, Venturi GM, Steeber DA, Tedder TF. Leukocyte rolling velocities and migration are optimized by cooperative L-selectin and intercellular adhesion molecule-1 functions. J Immunol 169: 4542–4550, 2002.[Abstract/Free Full Text]
  20. Kim MB, Sarelius IH. Distributions of wall shear stress in venular convergences of mouse cremaster muscle. Microcirculation 10: 167–178, 2003.[CrossRef][Web of Science][Medline]
  21. Kim MB, Sarelius IH. Regulation of leukocyte recruitment by local wall shear rate and leukocyte delivery. Microcirculation 11: 55–67, 2004.[CrossRef][Web of Science][Medline]
  22. Kim MB, Sarelius IH. Role of shear forces and adhesion molecule distribution on P-selectin-mediated leukocyte rolling in postcapillary venules. Am J Physiol Heart Circ Physiol 287: H2705–H2711, 2004.[Abstract/Free Full Text]
  23. Krunkosky TM, Jarrett CL. Selective regulation of MAP kinases and chemokine expression after ligation of ICAM-1 on human airway epithelial cells. Respir Res 7: 12, 2006.[CrossRef][Medline]
  24. Kunkel EJ, Chomas JE, Ley K. Role of primary and secondary capture for leukocyte accumulation in vivo. Circ Res 82: 30–38, 1998.[Abstract/Free Full Text]
  25. Kunkel EJ, Jung U, Bullard DC, Norman KE, Wolitzky BA, Vestweber D, Beaudet AL, Ley K. Absence of trauma-induced leukocyte rolling in mice deficient in both P-selectin and intercellular adhesion molecule 1. J Exp Med 183: 57–65, 1996.[Abstract/Free Full Text]
  26. Kunkel EJ, Jung U, Ley K. TNF-alpha induces selectin-mediated leukocyte rolling in mouse cremaster muscle arterioles. Am J Physiol Heart Circ Physiol 272: H1391–H1400, 1997.[Abstract/Free Full Text]
  27. Lau KS, Grange RW, Isotani E, Sarelius IH, Kamm KE, Huang PL, Stull JT. nNOS and eNOS modulate cGMP formation and vascular response in contracting fast-twitch skeletal muscle. Physiol Genomics 2: 21–27, 2000.[Abstract/Free Full Text]
  28. Lawrence MB, Berg EL, Butcher EC, Springer TA. Rolling of lymphocytes and neutrophils on peripheral node addressin and subsequent arrest on ICAM-1 in shear flow. Eur J Immunol 25: 1025–1031, 1995.[Web of Science][Medline]
  29. Lawrence MB, Springer TA. Leukocytes roll on a selectin at physiologic flow rates: distinction from and prerequisite for adhesion through integrins. Cell 65: 859–873, 1991.[CrossRef][Web of Science][Medline]
  30. Ley K, Allietta M, Bullard DC, Morgan S. Importance of E-selectin for firm leukocyte adhesion in vivo. Circ Res 83: 287–294, 1998.[Abstract/Free Full Text]
  31. Ley K, Gaehtgens P. Endothelial, not hemodynamic, differences are responsible for preferential leukocyte rolling in rat mesenteric venules. Circ Res 69: 1034–1041, 1991.[Abstract/Free Full Text]
  32. Ley K, Huo Y. VCAM-1 is critical in atherosclerosis. J Clin Invest 107: 1209–1210, 2001.[Web of Science][Medline]
  33. Lorenzon P, Vecile E, Nardon E, Ferrero E, Harlan JM, Tedesco F, Dobrina A. Endothelial cell E- and P-selectin and vascular cell adhesion molecule-1 function as signaling receptors. J Cell Biol 142: 1381–1391, 1998.[Abstract/Free Full Text]
  34. Muller WA. Migration of leukocytes across endothelial junctions: some concepts and controversies. Microcirculation 8: 181–193, 2001.[CrossRef][Web of Science][Medline]
  35. Naik SM, Shibagaki N, Li LJ, Quinlan KL, Paxton LL, Caughman SW. Interferon gamma-dependent induction of human intercellular adhesion molecule-1 gene expression involves activation of a distinct STAT protein complex. J Biol Chem 272: 1283–1290, 1997.[Abstract/Free Full Text]
  36. Norman MU, Van De Velde NC, Timoshanko JR, Issekutz A, Hickey MJ. Overlapping roles of endothelial selectins and vascular cell adhesion molecule-1 in immune complex-induced leukocyte recruitment in the cremasteric microvasculature. Am J Pathol 163: 1491–1503, 2003.[Abstract/Free Full Text]
  37. Renard M, Heutte F, Boutherin-Falson O, Finet M, Boisseau MR. Induced changes of leukocyte slow rolling in an in flow pharmacological model of adhesion to endothelial cells. Biorheology 40: 173–178, 2003.[Web of Science][Medline]
  38. Salas A, Shimaoka M, Phan U, Kim M, Springer TA. Transition from rolling to firm adhesion can be mimicked by extension of integrin alphaLbeta2 in an intermediate affinity state. J Biol Chem 281: 10876–10882, 2006.[Abstract/Free Full Text]
  39. Sarelius IH, Kuebel JM, Wang J, Huxley VH. Macromolecule permeability of in situ and excised rodent skeletal muscle arterioles and venules. Am J Physiol Heart Circ Physiol 290: H474–H480, 2006.[Abstract/Free Full Text]
  40. Steeber DA, Campbell MA, Basit A, Ley K, Tedder TF. Optimal selectin-mediated rolling of leukocytes during inflammation in vivo requires intercellular adhesion molecule-1 expression. Proc Natl Acad Sci USA 95: 7562–7567, 1998.[Abstract/Free Full Text]
  41. Steeber DA, Tang ML, Green NE, Zhang XQ, Sloane JE, Tedder TF. Leukocyte entry into sites of inflammation requires overlapping interactions between the L-selectin and ICAM-1 pathways. J Immunol 163: 2176–2186, 1999.[Abstract/Free Full Text]
  42. Sumagin R, Sarelius IH. 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 291: H2116–H2125, 2006.[Abstract/Free Full Text]
  43. Thorlacius H, Lindbom L, Raud J. Cytokine-induced leukocyte rolling in mouse cremaster muscle arterioles in P-selectin dependent. Am J Physiol Heart Circ Physiol 272: H1725–H1729, 1997.[Abstract/Free Full Text]
  44. Wang Q, Doerschuk CM. Neutrophil-induced changes in the biomechanical properties of endothelial cells: roles of ICAM-1 and reactive oxygen species. J Immunol 164: 6487–6494, 2000.[Abstract/Free Full Text]
  45. Wang Q, Pfeiffer GR 2nd, Stevens T, Doerschuk CM. Lung microvascular and arterial endothelial cells differ in their responses to intercellular adhesion molecule-1 ligation. Am J Respir Crit Care Med 166: 872–877, 2002.[Abstract/Free Full Text]
  46. Wojciechowski JC, Sarelius IH. Preferential binding of leukocytes to the endothelial junction region in venules in situ. Microcirculation 12: 349–359, 2005.[Web of Science][Medline]
  47. Wojcikiewicz EP, Abdulreda MH, Zhang X, Moy VT. Force spectroscopy of LFA-1 and its ligands, ICAM-1 and ICAM-2. Biomacromolecules 7: 3188–3195, 2006.[CrossRef][Web of Science][Medline]
  48. Wung BS, Ni CW, Wang DL. ICAM-1 induction by TNFalpha and IL-6 is mediated by distinct pathways via Rac in endothelial cells. J Biomed Sci 12: 91–101, 2005.[CrossRef][Web of Science][Medline]
  49. Yang L, Froio RM, Sciuto TE, Dvorak AM, Alon R, Luscinskas FW. ICAM-1 regulates neutrophil adhesion and transcellular migration of TNF-alpha-activated vascular endothelium under flow. Blood 106: 584–592, 2005.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 BloodHome page
N. A. Watkins, A. Gusnanto, B. de Bono, S. De, D. Miranda-Saavedra, D. L. Hardie, W. G. J. Angenent, A. P. Attwood, P. D. Ellis, W. Erber, et al.
A HaemAtlas: characterizing gene expression in differentiated human blood cells
Blood, May 7, 2009; 113(19): e1 - e9.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
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]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
293/5/H2786    most recent
00720.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sumagin, R.
Right arrow Articles by Sarelius, I. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sumagin, R.
Right arrow Articles by Sarelius, I. H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2007 by the American Physiological Society.