The Center for Immunology and the Departments of Veterinary PathoBiology and Laboratory Medicine and Pathology, University of Minnesota Academic Health Center, University of Minnesota, St. Paul
Correspondence: Dr. Bruce Walcheck, University of Minnesota, 295j, AS/VM Bldg., 1988 Fitch Ave., St. Paul, MN 55108. E-mail: walch003{at}umn.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: inflammation • L-selectin • margination
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Early work revealed that leukocyte–leukocyte interactions accounted for up to 70% of leukocyte accumulation on E- and P-selectin substrates and activated endothelium [4 , 7 ]. These in vitro modeling studies were performed using isolated leukocytes. Studies in vitro and in vivo indicate that red blood cells assist in leukocyte accumulation, in part by increasing leukocyte collisions with the endothelium through forced margination [13 14 15 16 17 18 ]. However, in a recent study, Mitchell et al. [19 ] showed that in the presence of red blood cells, the involvement of indirect capture was negligible during leukocyte accumulation on E-selectin.
In vivo, L-selectin has been extensively shown to account for the majority of the leukocyte rolling flux (60–80%) along small venules (<45 µm diameter) in various inflammatory settings [20 21 22 23 24 25 26 27 28 29 ]. In consideration of this, we sought to examine further the mechanisms of leukocyte capture in whole blood. Using a shear flow assay to model circulatory hydrodynamics with P-selectin as a substrate, our data demonstrate that under various stringencies (substrate concentrations and shear stresses), indirect capture considerably augmented leukocyte accumulation.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Whole blood collection and cell isolation
Peripheral blood was collected from normal, healthy donors in sodium heparin. These procedures were performed in accordance with a protocol approved by the Institutional Review Board: Human Subjects Committee at the University of Minnesota (St. Paul). Total leukocytes and red blood cells were isolated by dextran sedimentation, and neutrophils were isolated by an additional ficoll-hypaque centrifugation, as described previously [4
, 31
, 35
]. Isolated leukocytes and red blood cells were suspended in Hanks’ balanced saline solution (HBSS) and were used immediately.
Substrate preparation
Membrane P-selectin was isolated from human platelets as described previously [4
]. P-selectin in 50 mM n-octyl-ß-glucopyranoside (Sigma Chemical Co., St. Louis, MO) was diluted below the detergent’s critical micelle concentration and adsorbed to 35 mm petri dishes. The plates were then blocked with 2.5% bovine serum albumin. Neutrophils uniformly express PSGL-1, and in our hands, 
90% will attach to P-selectin upon induction of a shear stress of 2 dynes/cm2. A dose curve was generated to determine the minimum concentration of P-selectin that resulted in a functional-site density that supported the attachment of 90% of neutrophils under a shear stress of 2 dynes/cm2. A functional-site density assesses the availability of P-selectin molecules in the correct functional orientation for leukocyte binding, whereas conventional-site density determination does not. The minimum concentration of P-selectin was determined to be 0.11 µg/mL. Soluble recombinant E-selectin (human) was purchased from R&D Systems (Minneapolis, MN) and adsorbed to 35 mm petri dishes at a concentration of 3 µg/mL. A higher concentration of E-selectin compared with P-selectin was likely required since membrane-isolated selectins adsorb at a higher density than selectins in a soluble form [32
, 36
].
Hydrodynamic shear flow assays
Leukocyte adhesion mechanisms were examined under hydrodynamic shear stress using a parallel plate flow chamber obtained from Glycotech (Rockville, MD), as described previously [4
, 31
]. Whole blood (undiluted or diluted) or isolated leukocytes in combination with isolated red blood cells were perfused into the flow chamber. Diluted blood was used to visualize leukocyte capture events on the substrate. With undiluted blood, the accumulated leukocytes could be enumerated following the perfusion of HBSS, as described by others [36
]. The shear stresses used in our assays ranged from 2 to 5 dynes/cm2, which has been described to be within the physiological shear-stress range that occurs in post-capillary venules [37
]. Shear stress was calculated per the manufacturer’s instructions (Glycotech), and a blood viscosity of 0.025 Poise was used [38
]. For HBSS, a viscosity of 0.01 Poise was used [37
]. To calculate the shear stress of perfused, diluted blood, the viscosity of whole blood and HBSS was weighed proportionately. For instance, blood diluted one-fifth was determined to have an average viscosity of 0.013 Poise (one part at 0.025 Poise+four parts at 0.01 Poise=0.065/five Poise or 0.013 Poise). When isolated leukocytes were mixed with a physiological ratio of isolated red blood cells, a viscosity of 0.025 Poise was used.
Fluorochrome cell labeling
Isolated leukocytes were labeled with 5(6)-carboxyfluorescein diacetate, succinimidyl ester (CFSE) dye (Molecular Probes, Eugene, OR). CFSE (1 µL; 5 mM in dimethyl sulfoxide) was added to 1 mL isolated leukocytes (4x106 cells) for 10 min at 37°C in the dark with occasional agitation. Labeled leukocytes were washed twice at room temperature with HBSS. Immediately before their use, the labeled leukocytes were mixed with red blood cells, and the HBSS was adjusted to 2 mM CaCl2.
Antibody labeling and flow cytometry
These procedures were performed as described previously [35
, 39
]. Briefly, whole blood was aliquoted at 50 µl, and the red blood cells were removed with a lysing solution containing fixative (Becton Dickinson, San Jose, CA). Fc receptor and nonspecific antibody-binding sites were blocked by an initial incubation of the cells with wash buffer (phosphate-buffered saline containing 1% goat serum and 5 mM NaN3). One fluorescent-parameter flow cytometry was performed in which the cells were stained with the biotinylated mAb CHO-131, LM2, or 215. Primary antibody staining was revealed by the incubation of cells with PE-conjugated streptavidin (Jackson Immunoresearch). In addition, cells were stained directly with PE-conjugated LAM1-116. Cells were washed with wash buffer after each staining step. Isotype-matched, negative control mAb were used to evaluate levels of background staining. For each sample, 10,000 antibody-labeled cells were analyzed by flow cytometry on a FACSCalibur instrument (Becton Dickinson).
Statistics
The data are represented as mean ± SD. The Student’s t-test was used to compare results between groups.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The studies below directly examined the role of L-selectin in leukocyte accumulation on P-selectin. L-selectin and P-selectin have been demonstrated in numerous studies not to interact [4 , 25 , 29 , 40 41 42 ], and thus, an L-selectin mAb can be used to disrupt leukocyte–leukocyte interactions but not direct capture on P-selectin [4 , 32 ]. In Figure 1A 1B 1C 1D , whole blood diluted one-fifth was used, which maintains a physiological ratio of red blood cells to white blood cells, and was the minimum dilution that allowed for the visualization of leukocyte capture over time. By this approach, we observed that 267 leukocytes from the perfused blood (shear stress=5 dynes/cm2) accumulated on P-selectin (0.11 µg/mL) at 3 min. However, in the presence of an L-selectin mAb (DREG-200), 115 leukocytes accumulated on P-selectin at 3 min (Fig. 1A) . A similar effect was observed when diluted blood containing an isotype-matched, negative control mAb or DREG-200 was compared (Fig. 1B) , indicating that the effects of DREG-200 were a result of blocking L-selectin function and not a result of the exogenous whole antibody causing leukocyte clearing. This is consistent with the specificity of DREG-200 in other studies on L-selectin function [4 , 30 31 32 ]. The presence of a P-selectin mAb in this assay resulted in essentially no leukocyte accumulation on P-selectin (data not shown, and ref. [4 ]). When the diluted blood was perfused at a lower shear stress (3 dynes/cm2), more leukocytes accumulated on P-selectin, and again, leukocyte accumulation was considerably reduced in the presence of an L-selectin mAb (Fig. 1C) . Using a twofold lower concentration of P-selectin (0.06 µg/ml) and a shear stress of 3 dynes/cm2, the effect by an L-selectin mAb was consistent (Fig. 1D) . These findings demonstrate that L-selectin function accounted for 55–66% of the leukocytes that accumulated on P-selectin under various stringencies and assay conditions (Table 1 ). These findings were not merely a result of the use of diluted blood, as we observed a similar level of reduction (66±20% SD) in leukocyte accumulation on P-selectin after 3 min by an L-selectin mAb when undiluted blood was used as well (Fig. 1E) .
 View larger version (24K): [in a new window] |
Figure 1. The accumulation of leukocytes in whole blood on P-selectin is reduced by an L-selectin mAb. Whole blood diluted one-fifth in HBSS in the absence or presence of the indicated mAb (10 µg/mL) was perfused at a specified shear stress into the flow chamber (A–D), as described in Materials and Methods. (A and B) P-selectin adsorbed at 0.11 µg/mL; shear stress = 5 dynes/cm2. (C) P-selectin adsorbed at 0.11 µg/mL; shear stress = 3 dynes/cm2. (D) P-selectin adsorbed at 0.06 µg/mL; shear stress = 3 dynes/cm2. Leukocyte accumulation was allowed to equilibrate for 1 min, and the attached leukocytes were then enumerated over time for a total of 3 min at a 200X original magnification. Representative data from multiple repetitions are shown. (E) Undiluted whole blood in the absence or presence of DREG-200 was perfused into the flow chamber (3 dynes/cm2) containing P-selectin (0.11 µg/mL) for 3 min, as described in Materials and Methods. Leukocytes were enumerated at a 200X original magnification, and data are expressed as a percent of control. Values represent the mean ± SD of three independent experiments.
|
|
View this table: [in a new window] |
Table 1. Average Percent Decrease in the Accumulation of Leukocytes from Whole Blood Diluted One-Fifth in the Presence of an L-Selectin mAb after 3 Mina
|
 View larger version (23K): [in a new window] |
Figure 2. The accumulation of leukocytes in whole blood on E-selectin is reduced by an L-selectin mAb. Undiluted whole blood in the presence of DREG-200 or an isotype-matched, negative control mAb was perfused into a flow chamber (3 dynes/cm2) containing E-selectin (3 µg/mL) for 3 min, as described in Materials and Methods. Leukocytes were enumerated at a 200X original magnification, and data are expressed as a percent of control. Values represent the mean ± SD of three independent experiments.
|
 View larger version (21K): [in a new window] |
Figure 3. The presence of DREG-200 in whole blood does not affect leukocyte adhesion molecule expression levels. DREG-200 (bold line) or an isotype-matched, negative control mAb (dotted line) was added to undiluted whole blood at 10 µg/mL, as performed in the shear flow assays. The red blood cells were then lysed, and the leukocytes were stained for Mac-1 (LM2), PSGL-1 (215), L-selectin (LAM1-116), and P-selectin glycan ligand (CHO-131), as indicated. Nonspecific antibody labeling was determined using the appropriate isotype-matched, negative control mAb (thin line). PE-conjugated F(ab)'2 goat anti-mouse IgG was used to indicate the binding of DREG-200 to the leukocytes. Cell-staining levels were examined by flow cytometry, and 10,000 cells were examined per sample. Representative data from multiple repetitions are shown.
|
 View larger version (21K): [in a new window] |
Figure 4. Occurrences of indirect and direct leukocyte-capture events on P-selectin. Whole blood was diluted one-fifth in HBSS containing 2 mM CaCl2 in the absence or presence of the anti-L-selectin mAb DREG-200 (10 µg/mL) as indicated and perfused at a shear stress of 3 dynes/cm2 into a flow chamber containing P-selectin (0.11 µg/mL). Leukocytes that accumulated on P-selectin within a field at 400X original magnification were examined by image analysis to determine the number that attached by direct or indirect capture (fields consisted of 30–50 cells). Direct capture denotes free-flowing leukocytes that directly bound to the adhesive substrata, whereas indirect capture denotes free-flowing leukocytes that initially contacted an attached leukocyte, rolled over it, and deposited on the substrata no more than two cell lengths downstream. Values are expressed as a percentage of total capture events (indirect+direct) and represent the mean ± SD of five independent experiments. *, P < 0.005 when compared with the percentage of indirect capture events of untreated leukocytes.
|
 View larger version (79K): [in a new window] |
Figure 5. Leukocyte rolling on a monolayer of leukocytes occurs in the presence of red blood cells. Neutrophils isolated from peripheral blood were allowed to form a monolayer on P-selectin (0.11 µg/mL; left panel). Leukocytes isolated from peripheral blood were labeled with CFSE, mixed with red blood cells at a physiological ratio of one leukocyte to 500 red blood cells, and then perfused into the flow chamber (2 dynes/cm2). The frames from a 15-s video segment (300X original magnification) were superimposed to reveal tracks of labeled leukocytes rolling over the monolayer (right panel, solid arrow). For comparison, labeled leukocytes that attached to the P-selectin substrate rolled only a very short distance in this time frame (right panel, open arrow). Data are representative of three experiments.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In contrast to our findings, Mitchell et al. [19 ] have reported that in the presence of red blood cells, the contribution of indirect capture during leukocyte accumulation was negligible. For instance, leukocyte–leukocyte interactions were prevented in human blood diluted up to one-tenth or by the addition of red blood cells to isolated human leukocytes at a ratio as low as 10:1. In their study, however, soluble recombinant E-selectin was used as a substrate, and thus, a direct comparison to our study is problematic. For instance, membrane-isolated adhesion molecules adsorb at a higher density than soluble adhesion molecules [32 , 36 ], and unlike P-selectin, E-selectin and L-selectin have been demonstrated to directly interact [43 44 45 46 47 48 ]. Using conditions described by Mitchell et al. [19 ], including soluble recombinant E-selectin from the same source and concentration (5 µg/mL), the accumulated leukocytes demonstrated considerable firm adhesion and spreading (data not shown). At 3 µg/mL, the proportion of accumulated leukocytes that rolled was greatly increased, and leukocyte accumulation was significantly diminished in the presence of an L-selectin mAb (Fig. 2) . It is possible, however, that a comparison of identical forms of P-selectin and E-selectin (i.e., soluble or membrane-isolated) may reveal that indirect capture does not occur as efficiently on E-selectin. For instance, accumulation on E-selectin has been shown to stimulate leukocytes [53 ], which would likely result in L-selectin shedding—a process that occurs very rapidly upon leukocyte activation [54 ].
Our results are highly consistent with a 60–80% reduction in the leukocyte rolling flux that occurs in small venules (<45 µm diameter) upon blocking L-selectin in diverse inflammatory settings (e.g., type 1 hypersensitivity, cytokine-induced, lipopolysaccharide-induced, trauma, and reperfusion injury) [20
21
22
23
24
25
26
27
28
29
]. In addition to leukocyte accumulation along the vascular wall, their recruitment into the inflamed peritoneal cavity was also reduced by a similar level in L-selectin-deficient mice [55
]. L-selectin ligands, however, can be induced on activated endothelial cells [23
, 24
, 56
], and therefore, the effects of inhibiting L-selectin function in vivo could be the result of blocking indirect and/or direct leukocyte capture. Mice genetically deficient in E- and P-selectin expression should be capable of expressing ligands for L-selectin on the vascular endothelium that would mediate direct capture. This is taking into account that E-selectin and L-selectin in the mouse have been shown not to interact [48
]. Jung and Ley [28
] reported that E-/P-selectin-deficient mice reconstituted with wild-type leukocytes demonstrated an 90% reduction in the leukocyte rolling flux along venules in 6-h tumor necrosis factor 
(TNF-)-stimulated tissue, presumably leaving 10% of the rolling flux to be accounted for by L-selectin ligands on the endothelium and other adhesion mechanisms. However, in the same study, L-selectin-deficient mice or wild-type mice reconstituted with L-selectin-deficient leukocytes demonstrated an 80% reduction in the leukocyte rolling flux along venules in 6-h TNF--stimulated tissue [28
]. It can be inferred from these findings that L-selectin is participating primarily in indirect leukocyte capture.
The involvement of indirect capture during leukocyte accumulation will likely vary depending on the inflammatory setting and the types of blood vessels examined. Indeed, Eriksson et al. [11 ] have demonstrated that leukocyte–leukocyte interactions are prominent in large blood vessels, such as arteries in which indirect capture contributed up to 50% of total capture. In contrast, leukocyte–leukocyte interactions were reported to be infrequent in small venules [11 , 57 ]. It is important to note that analyses of leukocyte accumulation in small venules revealed that the capture of free-flowing leukocytes rarely occurred [11 , 57 ]. For instance, Kunkel et al. [57 ] reported that of 476 leukocytes attached along the walls of small venules, only seven of these cells were recruited by capture (of which six were recruited by indirect capture), and the rest accumulated at some point upstream. Accumulation has been proposed to occur as leukocytes initially enter the venule from the capillaries [11 , 57 , 58 ]. Red blood cells likely facilitate this process by forcing leukocytes toward the endothelium [13 14 15 ]. The involvement of indirect capture during leukocyte accumulation in immediate post-capillary venules at sites of inflammation has not been directly examined. It is tempting to speculate that leukocyte–leukocyte interactions may be a significant mechanism of leukocyte capture at this location, thus explaining the dramatic reduction in the leukocyte rolling flux occurring downstream in the small venules upon blocking L-selectin [20 , 21 , 24 25 26 27 28 29 ].
High endothelial cells lining post-capillary venules in secondary lymphoid organs bulge into the luminal space and likely increase the probability of collisions with leukocytes. The endothelial lining of blood vessels at other anatomical locations is more flattened. It is possible that at sites of inflammation, leukocytes attached to the endothelium provide temporary extensions into the blood vessel lumen to increase collisions with free-flowing leukocytes, which may be augmented during margination. Leukocytes individually and as a monolayer facilitate indirect capture, and this appears to accelerate and sustain leukocyte accumulation, respectively. In addition, leukocyte accumulation on activated endothelium at sites of inflammation shows spatial variability [57 ]. Indirect capture may then be a mechanism for increasing the density of leukocytes within these discrete sites of activated endothelium, as illustrated in Figure 6 .
 View larger version (15K): [in a new window] |
Figure 6. Indirect capture may be a mechanism for increasing the density of accumulated leukocytes. Direct binding to P-selectin occurs randomly (open circles). Next, leukocyte–leukocyte interactions mediate two means of indirect capture. Individual leukocytes attached to P-selectin capture free-flowing leukocytes (gray circles) during initial leukocyte accumulation. In addition, a monolayer of attached leukocytes will capture free-flowing leukocytes, which roll across the leukocyte cluster (black circles containing arrows) and then deposit within gaps or at its leading edge (solid circles).
|
Received October 15, 2002; revised January 7, 2003; accepted January 13, 2003.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Y. Wang, A. H. Herrera, Y. Li, K. K. Belani, and B. Walcheck Regulation of Mature ADAM17 by Redox Agents for L-Selectin Shedding J. Immunol., February 15, 2009; 182(4): 2449 - 2457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Ni, J. J. Campbell, G. Niehans, and B. Walcheck The Monoclonal Antibody CHO-131 Identifies a Subset of Cutaneous Lymphocyte-Associated Antigen T Cells Enriched in P-Selectin-Binding Cells J. Immunol., October 1, 2006; 177(7): 4742 - 4748. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Li, J. Brazzell, A. Herrera, and B. Walcheck ADAM17 deficiency by mature neutrophils has differential effects on L-selectin shedding Blood, October 1, 2006; 108(7): 2275 - 2279. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. E. Mattila, C. E. Green, U. Schaff, S. I. Simon, and B. Walcheck Cytoskeletal interactions regulate inducible L-selectin clustering Am J Physiol Cell Physiol, August 1, 2005; 289(2): C323 - C332. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Burdick and K. Konstantopoulos Platelet-induced enhancement of LS174T colon carcinoma and THP-1 monocytoid cell adhesion to vascular endothelium under flow Am J Physiol Cell Physiol, August 1, 2004; 287(2): C539 - C547. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Walcheck, S. R. Alexander, C. A. St. Hill, and E. Matala ADAM-17-independent shedding of L-selectin J. Leukoc. Biol., September 1, 2003; 74(3): 389 - 394. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
