| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Bioengineering, The Whitaker Institute for Biomedical Engineering, University of California San Diego, La Jolla
Correspondence: Dr. Geert W. Schmid-Schönbein, Department of Bioengineering, The Whitaker Institute for Biomedical Engineering, University of California San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0412. E-mail: gwss{at}bioeng.ucsd.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: shear stress • microcirculation • pseudopod
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Centrifugation, which is almost indispensable for isolation of blood subtypes [2 3 4 ], may cause damage to leukocytes [5 6 7 ]. This may not be surprising, because the acceleration to which leukocytes are exposed during even modest centrifugation is large compared with any acceleration in a normal circulation. The question then is what could be the effect of such relatively large accelerations on the fluid shear response, a mechanism that may depend on delicate membrane transport mechanisms. Thus, we examined the fluid shear stress response in the form of pseudopod formation on individual leukocytes after centrifugation.
Furthermore, to explore the question of whether the behavior of leukocytes in the microcirculation can remain normal after their shear response has been compromised, we also designed a study to investigate the role of the shear stress response of leukocytes in the microcirculation in vivo. Using a combined set of in vitro experiments on human and rat leukocytes and in vivo observations in the rat mesentery, we show here that centrifugation attenuates the fluid shear response of leukocytes. Reintroduction of centrifuged leukocytes without normal fluid shear response into the circulation leads to microcirculatory entrapment, suggesting that the shear stress response is a requirement for normal passage of leukocytes through the capillary network.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Leukocyte centrifugation
Fresh leukocytes were collected from nonsymptomatic volunteers with ammonium heparin (30 units per 1 ml blood). After red cell sedimentation at room temperature (1 G for 30 min), the native plasma, including the supernatant mixture of platelets and leukocytes as well as sporadic erythrocytes, was diluted by Plasma-Lyte (Baxter Health Care, Mundelein, IL; 1:20 dilution with 2.5 mmol/l Ca2+) and simultaneously spilt into two centrifuge tubes (1 ml in each tube). The cell suspension in the first tube was not centrifuged and served as controls. The cell suspension in the second tube was centrifuged at 100, 600, or 900 G for 5, 10, 15, or 20 min and was then resuspended in the same suspension in the same tube by gentle mixing. In each group, centrifugation was set to be completed 40 min after the blood collection. The leukocyte shear response was examined at several time intervals after centrifugation. Each cell suspension was kept in each centrifuge tube just before it was deposited into a small chamber with a transparent glass bottom.
In vitro test of fluid shear response
To examine the possible effect of plasma on the shear stress response of normal, noncentrifuged leukocytes, the plasma without leukocytes was collected from the blood sample after centrifugation at 600 G for 20 min. The supernatant mixture of platelets and leukocytes after red cell sedimentation at 1 G was resuspended in the centrifuged plasma (1:20 dilution) as the Noncentrifuged Leukocyte with Centrifuged Plasma group. Noncentrifuged cells with noncentrifuged plasma were used as controls.
Fluid shear response in a single leukocyte
One hour after the blood collection, 100 µl of each cell suspension was deposited into a small plastic chamber with a transparent glass bottom on an inverted microscope (Leitz Diavert, Welzler, Germany) with 50x objective (numerical aperture=1.4, oil immersion; Leitz Diavert). In selected groups of cells (noncentrifuged cells and centrifuged cells at 900 G for 20 min), the Ca2+ channel blocker, diltiazem (Sigma Chemical Co., St. Louis, MO; 300 µmol/l), was applied to the cell suspension.
The microscope light source had a heat filter, and all experiments were carried out at room temperature. The microscope eyepiece was connected to a closed circuit television system that included a black-and-white-coupled charge device camera (Model JE 2362, Javelin, Tokyo, Japan). Micropipettes with 4–6 µm internal tip diameter were fabricated using a micropipette puller (Sutter Instruments, Novato, CA) filled with Plasma-Lyte with 2.5 mmol/l Ca2+ and connected to a reservoir with hydrodynamic pressure adjustment. The micropipettes were positioned above individual leukocytes (neutrophils, monocytes, and lymphocytes) spread on the glass surface so that a fluid flow could be applied over the cell surface. The micropipette orientation was kept at an angle of 30° to the glass slide, and the horizontal distance between cell surface and pipette tip was kept at about 8 µm. The magnitude of the fluid surface stress on the leukocytes was computed numerically by solution of Stokes approximation of the equation of motion for plasma as a Newtonian viscous incompressible fluid with a coefficient of viscosity of 1.2 centipoise. The boundary conditions were selected based on the actual cell and pipette size, its angle, the distance between pipette tip and cell surface, and velocity of fluid out of the pipette [1 ]. All leukocytes were exposed to similar shear stresses on the order of 1.5 dyn/cm2 for 3 min. The observations were completed 4 h after the blood collection to avoid further aging of the cells in vitro.
In vitro experiments in a cone-and-plate shear field
As observations on single cells depend on their adhesion to a substrate, we also used a technique that exposes leukocytes in free suspension to well-controlled shear stresses in a cone-and-plate device [8
]. The femoral veins and arteries of 12 mature male Wistar rats (280–330 g) were cannulated after general anesthesia (sodium pentobarbital, 50 mg/kg; Abbot Laboratories, Chicago, IL). The animals were maintained at 37°C. All protocols were reviewed and approved by the University of California-San Diego Animal Subject Committee.
Arterial blood samples (0.6 ml) with ammonium heparin (30 units per 1 ml blood) were collected from each animal. In the centrifugation groups, the blood was spun at 600 G for 15 min and immediately remixed. The blood was then divided into two groups: blood samples with and without shear stress application. Diltiazem (75 µmol/l) was applied to selected blood groups 20 min after the blood collection. The whole blood (0.3 ml) was sheared in a cone-and-plate device 50 min after collection [9 ] at a level of 5.0 dyn/cm2 for 10 min. The blood was fixed with 2% glutaraldehyde for determination of pseudopod projection or with 0.4% paraformaldehyde for CD18 or CD29 labeling immediately after shear stress application (60 min after the blood collection). Pilot studies had shown that pseudopods were fixed rapidly by this procedure (within less than 10 s) without sufficient time for retraction. Unsheared control samples (0.3 ml) were also fixed at the same time in the same way.
For morphological identification, the cells (including neutrophils, lymphocytes, and monocytes) were stained with 0.02% crystal violet. The number of leukocytes with pseudopods was counted by light microscopy. To determine CD18 and CD29 membrane expression, fluorescein isothiocyanate-labeled monoclonal antibody against rat CD18 (5.43 µl/ml; PharMingen, San Diego, CA), CD29 (8.57 µl/ml; PharMingen), or rat immunoglobulin G isotype (5.43 µl/ml; PharMingen) was applied to each blood group for 30 min. Erythrocytes were removed by fluorescein-activated cell sorter (FACS) lysing solution (Becton Dickinson, San Jose, CA). CD18 and CD29 expression was measured with a flow cytometer (Becton Dickinson FACS analyzer).
In vivo experiments
The femoral arteries and veins of 24 mature male Wistar rats (260–330 g) were cannulated under sodium-pentobarbital anesthesia (Abbot; 50 mg/kg, intraperitoneally). Booster dosages (5 mg/kg, intravenously) were administered as required after testing reflexes. The animals were placed on a heating pad, covered with a blanket, and maintained at 37°C. A blood volume equal to 0.3% of their body weight was taken and centrifuged for 15 min at 600 G.
The following experimental groups were formed:
Blood Centrifugation group
Whole blood (0.3% vol per body weight) was taken from the animals, centrifuged, and then remixed and labeled with 14.56 nmol/l carboxyfluorescein succinimidyl ester (CFSE; Molecular Probes, Junction City, OR), a stable actin-bound label for circulating and migrating leukocytes for 20 min ex vivo [10
].
Diltiazem Treatment group
Diltiazem (74 µmol/l; 1.0 mg per kg of weights of the animals) was applied to the centrifuged blood and labeled by CFSE in the same way as in the Blood Centrifugation group.
Leukocyte Centrifugation group
Only centrifuged leukocytes from the buffy coat were resuspended in physiological saline and labeled with 14.56 nmol/l CFSE.
Erythrocyte Centrifugation group and Plasma Centrifugation group
Another 0.3% vol whole blood (per body weight) was taken from the animals and labeled with CFSE without centrifugation. Erythrocytes and plasma were collected from 0.3% vol whole blood (per body weight) after centrifugation and mixed with the CFSE-labeled, noncentrifuged whole blood.
Control group
Whole blood (0.3% vol per body weight) was taken from the animals and labeled with CFSE without centrifugation.
The ileocecal portion of the mesentery was exteriorized, draped over a pedestal, and superfused with a Krebs-Henseleit bicarbonate-buffered solution saturated with a 95% N2 and 5% CO2 mixture (37°C, pH 7.4). Forty-five minutes after the blood was collected, the blood in each group was reinjected into the same animal via the femoral vein, and the mesenteric microcirculation was visualized through an intravital fluorescence microscope (25x objective) with a SIT camera (Model 66, Dage, Michigan City, IN) and a 200-watt mercury lamp.
To elicit fluorescent images, the light was passed through a quartz collection heat filter, a 480 nm wavelength excitation filter (Leitz Diavert). The fluorescence images of the microvessels were digitized [10 ]. All fluorographs were recorded with a fixed gain control setting on the SIT camera controller.
The exposure period to the fluorescent light excitation for a single measurement was limited to 30 s to minimize quenching. The number of migrating leukocytes (including adherent cells on the endothelium for 30 s) was measured in 20 observation fields (400 µm by 300 µm) every 15 min for 60 min in each animal.
Data analysis
The measurements are summarized in the form of mean ± SD. Differences among groups were analyzed by analysis of variance and Fischer’s protected least significant difference test. A value of P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
The ß2 integrin (CD18) on neutrophils was down-regulated in response to shear stress at 5.0 dyn/cm2 [8 ]. In the presence of fluid shear, centrifugation had no significant effect on the expression of CD18 on leukocytes (Fig. 6B) , although in the absence of fluid shear stress, centrifugation had a significant tendency to increase CD18 expression (Fig. 6B) .
Contrary to the CD18 expression, fluid shear stress at 5 dyn/cm2 caused up-regulation of ß1 integrin (CD29) on neutrophils (Fig. 6C) and lymphocytes (Fig. 6D) . Lymphocytes had two peaks in the FACS analyzer histograms whereas neutrophils had one. Centrifugation led to suppression of CD29 up-regulation by shear application on neutrophils, but not on lymphocytes, and in the absence of shear stress, there was no significant difference in CD29 expression between the Control group and the Centrifuged group (Fig. 6C and 6D) .
The behavior of leukocytes after centrifugation in the microcirculation
After centrifugation and reintroduction into the circulation, the abnormal shear stress response in the leukocytes compromised their ability to circulate. The number of labeled leukocytes migrating into the tissue from mesenteric microvessels under normal blood flow in the Control group did not increase even 1 h after cell reinjection (Figs. 7
and 8
). These uncentrifuged cells continued to circulate and exhibited low tendency to adhere to the endothelium and migrate into the tissue. In contrast, in the Blood Centrifugation group, the number of migrating cells increased significantly 60 min after return of the cells into the circulation (Figs. 7
and 8)
. Among the Erythrocyte Centrifugation group, the Plasma Centrifugation group, and the Leukocyte Centrifugation group, a significant increase in the number of migrating cells could only be seen in the Leukocyte Centrifugation group as compared with the Control group (Fig. 8)
. The number of the cells migrating into the tissue in the presence of diltiazem was significantly smaller than that in the Blood Centrifugation group (Fig. 8)
.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In contrast to normal, noncentrifuged leukocytes, the shear stress response of leukocytes after centrifugation was significantly attenuated and, in part, completely under these experimental conditions. The annihilation of the shear response depended on the duration and the magnitude of the acceleration applied during centrifugation (Figs. 2 and 3) . The effect is visible at relatively modest accelerations, used in many practical protocols for isolation of leukocytes, down to values as low as 100 G for 5 min (Fig. 2) . Moreover, centrifugation at relatively high accelerations or for a longer duration often not only reduced the ability of leukocytes to retract pseudopods, but also induced a projection of pseudopods in response to shear stress (Figs. 1 2 3 and 6) .
Neutrophils exposed to shear stress in vitro at a level between 75 and 150 dyn/cm2 for 10 min are subject to morphological changes, and shear stress at 600 dyn/cm2, which is more than 10 times higher than that in vivo, leads to marked cell destruction [5 ]. Higher mechanical stresses other than those in a physiological state may cause damage to leukocytes. Therefore, centrifugation may directly cause mechanical damage to the shear stress response elements of leukocytes. During centrifugation, the leukocytes in a blood sample are forced to sediment to the bottom of their container, and thereafter are lifted by the heavier red cells into the buffy coat [11 ]. An order of magnitude of the fluid stresses on the membrane can be estimated as follows. If a leukocyte (10 µm in diameter; density, 0.08 gm/cm3 higher than that of buffer) sediments at 600 G, its sedimentation velocity is 0.26 cm/s, according to Stokes formula for motion of a sphere in a uniform velocity field. The shear stress on the surface membrane of the cell during this sedimentation is nonuniform. Its peak value at this sedimentation velocity according to Stokes solution of a flow over a sphere moving in a medium with viscosity of 0.01 poise is 7.8 dyn/cm2 (3/2xviscosityxsedimentation velocity/radius of the cell). Thus, the peak shear stress during 600 G centrifugation is of the same order of magnitude as the fluid shear stress we applied during the experiments with the pipette or the cone-and-plate shear device. This analysis suggests that during centrifugation, no excessive shear stress may be responsible for the loss of the shear response. Instead, there is also the stress in the interior structures of the cell, as different molecular components have different densities. This stress is suddenly magnified by 600-fold during centrifugation and is potentially the main reason for the loss of the shear stress response. Heavier elements in the cell cytoplasm and membrane (e.g., proteins with heavy metal) may try to separate from the lighter elements (e.g., lipids). In the process, internal membranes and the plasma membrane may be subjected to restructuring, which may compromise the fluid shear response.
There could also be other possibilities by which centrifugation may attenuate the shear stress response. A stimulator or toxic mediator, such as proteases, may be released by centrifugation, or centrifugation may lead to up-regulation of some adhesion molecules on leukocytes, such as ß1 and ß2 integrins, which in turn may influence the shear response. To shed more light on this issue, we first examined the shear stress response of normal leukocytes by resuspension in centrifuged plasma in vitro (Fig. 4) . Noncentrifuged leukocytes in the presence of serum isolated from centrifuged blood could still respond to shear stress, suggesting that centrifuged serum is less likely to be involved in the prevention of the normal shear stress response by centrifugation. In addition, in the rat mesenteric microcirculation, an enhanced number of leukocytes migrating into the tissue were observed in the Leukocyte Centrifugation group, and this was not observed in the Erythrocyte or Plasma Centrifugation group (Fig. 8) . This evidence suggests that suppression of the fluid shear response by centrifugation may not be a result of release of any cell stimulator, but it appears to be an effect on the leukocytes themselves.
The expression of integrins on leukocytes, particularly ß1 and ß2 integrins, is closely associated with the regulation of pseudopod formation and cell spreading [12 , 13 ]. We have shown that inflammatory stimulators, such as platelet-activating factor and formyl-Met-Leu-Phe, which induce expression of ß1 and/or ß2 integrins on leukocytes, suppress the shear stress response of leukocytes in a dose-dependent way [8 ]. The expression of ß1 integrin leads to suppression of the shear stress response [14 ]. Moreover, it is known that expression of several adhesion molecules on leukocytes can be altered by centrifugation [15 ]. Thus, there is the possibility that the integrin(s) may be up-regulated on leukocytes after centrifugation, which may lead to inhibition of the shear stress response. To clarify this hypothesis, we examined the expression of ß1 and ß2 integrins with and without centrifugation in the absence and in the presence of shear stress.
It is interesting that ß1 integrin is up-regulated in response to fluid shear stress (Fig. 6) in contrast with down-regulation of ß2 integrin by fluid shear [8 ] (Fig. 6B 6C 6D) . Centrifugation does not cause up-regulation of ß1 integrin in the presence and in the absence of shear stress (Fig. 6C and 6D) . Centrifugation also had no significant effect on ß2 integrin expression in the presence of shear stress, although ß2 integrin was up-regulated by centrifugation compared with the Control group in the absence of shear stress. It is therefore less likely that ß1 or ß2 integrin is associated with the prevention of shear stress response by centrifugation.
It remains to be determined what kind of shear stress response elements in leukocytes, if any, is damaged by centrifugation. We found that the pseudopod projection as an abnormal response to shear stress could be inhibited by diltiazem, a voltage-dependent, Ca2+ channel blocker (Fig. 5) . Therefore, this reversed shear stress response of leukocytes after centrifugation may require Ca2+ influx via voltage-dependent Ca2+ channels. The normal shear stress response of noncentrifuged leukocytes is suppressed by K+ channel blockers, such as glybenclamide and tetraethyl ammonium, and is completely inhibited by calcium chelation by ethylenediaminetetraacetic acid [1 ]. Because closure of K+ channels leads to opening voltage-dependent Ca2+ channels [16 ], this group of Ca2+ channels may be involved in the shear stress response of leukocytes. We also have obtained evidence that the intracellular-free Ca2+ level in leukocytes, in which the shear response is reversed, is increased during pseudopod projection by fluid shear whereas the free Ca2+ level in normal leukocytes is decreased slightly during pseudopod retraction in response to shear stress (unpublished data).
In addition to adhesion molecules, pseudopod projection is essential for migration and phagocytosis because leukocytes in a spherical state without pseudopods cannot spread or tightly adhere on the endothelium or migrate into the tissue [17 ]. Recently, we have demonstrated that the fluid shear response of leukocyte plays an inhibitory role in inflammation [8 ]. The fluid shear response serves to prevent pseudopod projection of leukocytes adhered to postcapillary venules during normal blood flow (8), suggesting that the abnormal fluid shear response of leukocytes can enhance the inflammatory reaction in the microcirculation. To examine the fluid shear response of leukocytes in vivo, we examined the behavior of CFSE-labeled leukocytes after centrifugation in the rat mesenteric microcirculation. The centrifuged leukocytes migrated in large numbers into the tissues compared with normal, noncentrifuged cells (Figs. 7 and 8) . Moreover, diltiazem had the ability to reduce the number of centrifuged leukocytes migrating into the tissue (Fig. 8) . The Ca2+ channel blocker may improve the circulation of centrifuged leukocytes due to inhibition of pseudopod projection in response to shear stress. Pseudopod projection instead of pseudopod retraction in response to fluid shear and blood flow does lead to an accelerated migration and an increase in the membrane contact area between leukocytes and endothelium. This in turn can contribute to the migration of the cells across the vessel wall and to the development of inflammation. The evidence supports the hypothesis that the shear stress response of leukocytes may play an anti-inflammatory role and may be necessary for the regulation of inflammation.
Pseudopod retraction in response to shear stress can be a key requirement for normal passage of leukocytes through the microcirculation. These in vitro and in vivo data suggest that loss of a normal fluid shear stress response of leukocytes can lead to a shift in microcirculatory insufficiency. Cell function in vivo, blood kinetics, and in vivo chemotaxis of transfused neutrophils collected by filtration leukophoresis (including centrifugation at 400 G for 15 min) are severely impaired compared with cells collected by phlebotomy without the centrifugation procedure [6 ]. Although the leukopheresis-induced alterations are potentially due to activating processes other than centrifugation in this case, impairment of the cells may also be partly due to centrifugation. Even in the absence of inflammatory stimulators, centrifuged leukocytes can adhere to the endothelium of postcapillary venules and migrate into the tissues (Figs. 7 and 8) . Thus, leukocytes without normal fluid shear response may spend less time circulating in vivo, a process that may lead to suppression of a normal immune response because of a lack of circulating leukocytes. New approaches may need to be developed to preserve the shear stress response in biological, physiological, and hemorheological studies. In turn, observing the behavior of centrifuged leukocytes under fluid shear can be useful to clarify the role of the shear stress response of leukocytes in the circulation. The observations may also be relevant with respect to the in vivo behavior of leukocytes after isolation for the purpose of transplantation.
|
ACKNOWLEDGEMENTS |
|---|
|
FOOTNOTES |
|---|
Received November 26, 2001; revised February 11, 2002; accepted February 15, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. J. Powner, R. M. Payne, T. R. Pettitt, M. L. Giudici, R. F. Irvine, and M. J. O. Wakelam Phospholipase D2 stimulates integrin-mediated adhesion via phosphatidylinositol 4-phosphate 5-kinase I{gamma}b J. Cell Sci., July 1, 2005; 118(13): 2975 - 2986. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. Coughlin and G. W. Schmid-Schonbein Pseudopod Projection and Cell Spreading of Passive Leukocytes in Response to Fluid Shear Stress Biophys. J., September 1, 2004; 87(3): 2035 - 2042. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Fukuda, T. Yasu, N. Kobayashi, N. Ikeda, and G. W. Schmid-Schonbein Contribution of Fluid Shear Response in Leukocytes to Hemodynamic Resistance in the Spontaneously Hypertensive Rat Circ. Res., July 9, 2004; 95(1): 100 - 108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Fukuda, H. Mitsuoka, and G. W. Schmid-Schonbein Leukocyte fluid shear response in the presence of glucocorticoid J. Leukoc. Biol., April 1, 2004; 75(4): 664 - 670. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Fukuda and G. W. Schmid-Schonbein Regulation of CD18 expression on neutrophils in response to fluid shear stress PNAS, November 11, 2003; 100(23): 13152 - 13157. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), 100 (
), 600 (x), and 900 G (•) gravitational force on the fluid shear stress response of leukocytes measured in the time interval between (A) 0 and 1 h, (B) 1 and 2 h, and (C) 2 and 3 h after centrifugation. *, vs. Control (P<0.05).
