AJP - Heart AJP: Lung Cellular and Molecular Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 286: H359-H367, 2004. First published September 11, 2003; doi:10.1152/ajpheart.00491.2003
0363-6135/04 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
286/1/H359    most recent
00491.2003v1
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 (32)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, X.
Right arrow Articles by Goligorsky, M. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhang, X.
Right arrow Articles by Goligorsky, M. S.

Atomic force microscopy measurement of leukocyte-endothelial interaction

Xiaohui Zhang,1 Aileen Chen,1 Dina De Leon,2 Hong Li,4 Eisei Noiri,2 Vincent T. Moy,1 and Michael S. Goligorsky4

1Department of Physiology and Biophysics, University of Miami School of Medicine, Miami, Florida 33101; 2Department of Medicine, The University of Tokyo, Tokyo 113-8655, Japan; 3Department of Pediatrics, State University of New York, Stony Brook 11794; and 4Departments of Medicine and Pharmacology, New York Medical College, Valhalla, New York 10595

Submitted 28 May 2003 ; accepted in final form 10 September 2003


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Leukocyte adhesion to vascular endothelium is a key initiating step in the pathogenesis of many inflammatory diseases. In this study, we present real-time force measurements of the interaction between monocytic human promyelocytic leukemia cells (HL-60) cells and a monolayer of human umbilical vein endothelial cells (HUVECs) by using atomic force microscopy (AFM). The detachment of HL-60-HUVEC conjugates involved a series of rupture events with force transitions of 40–100 pN. The integrated force of these rupture events provided a quantitative measure of the adhesion strength on a whole cell level. The AFM measurements revealed that HL-60 adhesion is heightened in the borders formed by adjacent HUVECs. The average force and mechanical work required to detach a single HL-60 from the borders of a tumor necrosis factor-{alpha}-activated HUVEC layer were twice as high as those of the HUVEC bodies. HL-60 adhesion to the monolayer was significantly reduced by a monoclonal antibody against {beta}1-integrins and partially inhibited by antibodies against selectins ICAM-1 and VCAM-1 but was not affected by anti-{alpha}V{beta}3. Interestingly, adhesion was also inhibited in a dose-dependent manner (IC50 {approx} 100 nM) by a cyclic arginine-glycine-aspartic acid (cRGD) peptide. This effect was mediated via interfering with the VLA-4-VCAM-1 binding. In parallel measurements, transmigration of HL-60 cells across a confluent HUVEC monolayer was inhibited by the cRGD peptide and by both anti-{beta}1 and anti-{alpha}V{beta}3 antibodies. In conclusion, these data demonstrate the role played by {beta}1-integrins in leukocyte-endothelial adhesion and transmigration and the role played by {alpha}V{beta}3 in transmigration, thus underscoring the high efficacy of cRGD peptide in blocking both the adhesion and transmigration of monocytes.

integrins; cell-cell adhesion; arginine-glycine-aspartic acid peptide

INTERACTIONS OF LEUKOCYTES with the vascular endothelium culminating in their diapedesis represent not only a crucial event in immune surveillance and defense but are also critically involved in the pathogenesis of many inflammatory diseases including atherosclerosis. The mechanics of leukocyte-endothelial cell adhesion has been extensively investigated during the past two decades. The process of leukocyte egress, as observed using intravital microscopy, has been tentatively subdivided into stages of rolling, cell activation, firm adhesion, and transendothelial migration with each stage engaging a different set of adhesion molecules and their activation (6, 15, 23). The identities of these adhesion molecules present on the surface of the interacting cells have been elucidated: they include L-selectin; carbohydrate ligands of selectins, {beta}1-, {beta}2-, {beta}7-integrins; PECAM-1 on leukocytes and P- and E-selectins; ICAM-1 and ICAM-2; VCAM-1, MadCAM-1, and PECAM-1; and {beta}1-integrins on endothelial cells (6). Therapeutic strategies based on inhibiting some of these interactions with neutralizing antibodies have been successfully applied to the animal models of ischemia-reperfusion and acute inflammatory processes (16, 29).

To study the intimate molecular mechanisms of leukocyte-endothelial interaction, a quantitative evaluation of cell adhesion is necessary. The commonly used methods for the evaluation of leukocyte adhesion are various forms of cell adhesion assays, which quantify adhesion by counting the number of bound cells. Although these assays provide a convenient means to estimate adhesion, they do not reveal the underlying biophysical mechanisms of cell-cell interaction. The main reason is due to the fact that in these methods, the bound cell numbers are counted a posteriori. Therefore, the detailed information on the dynamics of cell-cell interaction is missing.

In this work, we have employed the atomic force microscopy (AFM) (4) to quantify whole cell adhesion. A single live leukocyte, attached to end of the AFM cantilever, served as a functionalized probe that monitored leukocyte-endothelial interaction in real time. Taking advantage of the high temporal, spatial, and force resolutions of this technique, we have measured the adhesive interaction of the leukocyte over different locations in the endothelial monolayer and evaluated the relative contribution of different adhesion molecules in supporting cell adhesion. The inhibitory potency of different function-blocking antibodies and a cyclic arginine-glycine-aspartic acid (cRGD) peptide was also further investigated.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cells and cell culture. The human promyelocytic leukemia cells (HL-60) were maintained in RPMI 1640 medium supplemented with 10% FBS, 1% glutamine, 50 U/ml penicillin, and 50 µg/ml streptomycin in 5% CO2 at 37°C. HL-60 cells were treated with PMA (10 nM) for 3 to 4 days to induce monocytic phenotype. HUVECs were purchased from Cambrex (Walkersville, MD) and cultured in endothelial cell basal medium (EBM-2) supplemented with 2% FBS, hydrocortisone, human fibroblast growth factor, VEGF, IGF-1, ascorbic acid, human epidermal growth factor, gentamicin-amphotericin, and heparin. HUVECs were plated on 35-mm tissue culture dishes (Becton-Dickinson; Franklin Lakes, NJ) coated with 25 µg/ml human fibronectin (overnight at 25°C). Before the AFM measurements, confluent HUVECs were stimulated with tumor necrosis factor-{alpha} (TNF-{alpha}, 100 ng/ml) for 6 h at 37°C. HUVECs were used between passages 5 and 7.

Proteins and other reagents. Human TNF-{alpha}, human recombinant P-selectin-Fc, and VCAM-1-Fc fusion proteins were purchased from R&D Systems (Minneapolis, MN). cRGD and cyclic Arg-Ala-Asp-D-Phe-Val (cRAD) were purchased from Bachem (King of Prussia, PA), and CellTracker CM-Dil (C-7000) was purchased from Molecular Probes (Eugene, OR). Human function-blocking antibodies BBIG-E4 (anti-E-selectin), BBIG-I1 (anti-ICAM-1), and AF809 (anti-VCAM-1) were purchased from R&D Systems. LM-609 (anti-{alpha}V{beta}3) and JB1A (anti-{beta}1) were from Chemicon International (Temecula, CA). WAPS-12.2 (anti-P-selectin) was purified from culture supernatant of WAPS-12.2 hybridomas (ATCC; Manassas, VA) by protein A column. All other reagents were from Sigma (St. Louis, MO).

Protein immobilization. VCAM-1/Fc (30 µl) and P-selectin/Fc (25 µg/ml) in 0.1 M NaHCO3 (pH 8.6) were adsorbed overnight at 4°C on the center of a 35-mm tissue culture dish (Falcon, Becton-Dickinson). Unbound proteins were removed, and bovine serum albumin (BSA, Sigma) at 100 µg/ml in PBS was used to block the exposed surface of the dish.

Measurement of leukocyte adhesion by atomic force microscopy. The AFM instrument used was as described earlier (7). Briefly, cells were localized using an inverted optical system. The position of the AFM tip was set by a piezotranslator with a strain gauge position sensor (Physik Instrumente; Waldbronn, Germany). The interaction between the AFM tip and the substrate was determined from deflection of the AFM cantilever. Cantilevers were calibrated by thermal fluctuation analysis according to the method of Hutter and Bechhoefer (12) and had a spring constant of ~0.010 Nm–1.

Individual HL-60 cells were attached to the tip of a cantilever through Concanavalin A (Con A)-mediated linkages (see Fig. 1) (2, 30). The cantilevers were first incubated in 0.5 mg/ml biotinamidocaproyl-labeled BSA overnight at 37°C and then rinsed with PBS and incubated in 0.5 mg/ml streptavidin for 10 min at room temperature. Finally, the cantilevers were incubated in 0.5 mg/ml biotinylated Con A. All AFM studies were conducted at 25°C in phenol red-free RPMI 1640 medium containing 10 mM HEPES. For cell attachment, the tip of the Con A-functionalized cantilever was positioned above the center of a cell and gently lowered onto the cell for ~1 s. The HL-60 firmly adhered to the Con A on the AFM tip and was then used to probe adhesion. To measure the unbinding of the HL-60 from endothelial cells, the strength of cantilever-leukocyte linkage and the endothelial cell-substrate linkage must be much larger than the strength of the leukocyte-endothelial cell interaction. To obtain an estimate of the strength of the cell-cantilever linkage, we allowed the attached HL-60 cell to interact for 1 min with a tissue culture dish coated with Con A. On retraction of the cantilever, separation always (N > 30) occurred between the cell and the Con A-coated surface. This is likely due to the fact that the contact area formed between the cantilever and the cell is larger than that between the cell and the substrate. The average force needed to induce separation was >2.3 nN, which was at least two times larger than the detachment forces recorded in this study (<700 pN). As for the other linkage in the experimental system, i.e., the linkage between HUVEC and the petri dish, Sagvolden et al. (22) have shown that the deadhesion force between an adherent cell and the dish is at least 20 nN. In fact, we never observed the detachment of HUVEC from the petri dish during our force measurements. In the experiments using function-blocking antibodies or peptides, immobilized proteins or HUVECs were incubated with different antibodies (50 µg/ml each) or different concentrations of peptides for 30 min before the force measurements.



View larger version (72K):
[in this window]
[in a new window]
  		Fig. 1. A: micrograph of a human promyelocytic leukemia (HL-60) cell attached to the tip of an atomic force microscopy (AFM) cantilever. B: micrograph of a monolayer of human umbilical vein endothelial cells (HUVECs). Typical cell-cell contact zones for AFM force measurements are indicated by circles (solid line, HUVEC cell body; dashed line, HUVEC cell junction). C: micrograph of a HL-60 cell attached to the tip of an AFM cantilever over a HUVEC monolayer. Bars are 20 µm.

 

Leukocyte adhesion assay. Falcon culture plates (24-well) were coated with 0.2% gelatin overnight, sterilized under UV light for 20 min, and plated with HUVEC. Confluent HUVEC were then stimulated with TNF-{alpha} (100 ng/ml) for 6 h at 37°C. HL-60 cells were loaded with CM-Dil cell tracker, washed with PBS twice, and resuspended in RPMI 1640 medium. After washing was completed, 20,000 cells in 0.7 ml medium (RPMI/Hanks) were added to each well with a confluent monolayer of TNF-{alpha}-stimulated HUVEC. Different concentrations of cRGD (10 nM-10 µM) or cRAD (1 µM) were tested by adding them to each well (one well was kept as control with no additions). In a separate series of experiments, monoclonal blocking antibodies (specified in RESULTS) were added to stimulated HUVEC monolayers coincubated with cell tracker-labeled monocytes. Monocyte-endothelial cell interaction was quantified after 30 min. Nonadherent monocytes were removed by gentle washing with PBS.

Leukocyte transmigration assay. Transmigration assays were performed using Boyden chemotaxis chamber (Neuro Probe; Gaithersburg, MD). RPMI (30 µl) with N-formyl-Met-Leu-Phe (10 nM) were added to each well of the lower chamber. Polycarbonate membrane with 5-µm pores was placed gently on top of the lower chamber. Membrane in each well was coated with gelatin (10 mg/ml) and allowed to dry for 2 h. HUVEC were stimulated with TNF-{alpha} (100 ng/ml) for 6 h, trypsinized, and resuspended in a nonphenolated RPMI 1640 medium, and 40 µl of the suspension were added to each well of the upper chamber for 2-h incubation at 37°C and 5% CO2. Differentiated HL-60 cells (1 x 105 CM-Dil labeled) were added to each well for 4-h transmigration. To assess the integrity of HUVEC monolayers, cells on polycarbonate filters were stained with Wright-Giemsa at the completion of the transmigration assays. Transmigrated monocytes were counted under low-power magnification in 5 fields/well. Each experiment was performed in triplicate, and experiments were repeated at least three times.

Statistical analysis was performed by two-tailed t-test or ANOVA, with P < 0.05 considered statistically significant. The data were presented as means ± SE.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
HL-60 cells bind specifically to substrates coated with adhesion molecules. To validate the use of AFM in the studies of HL-60-HUVEC interactions, we first examined HL-60 adhesion to individual adhesive partners. PMA-differentiated HL-60 cell line was employed as a model system for monocyte adhesion. Direct force measurements were carried out between a HL-60 cell attached to the end of an AFM cantilever and purified adhesion molecules immobilized on a tissue culture dish (Fig. 1A). Important endothelial-derived adhesion molecules, including P, E-selectins, ICAM-1, and VCAM-1 were tested in these experiments. As illustrated in Fig. 2A, a complete cycle of the AFM force scan begins with the downward translation of the cantilever to a substrate. After a predetermined duration of cell-substrate contact, the cycle is completed following the upward displacement of the cantilever to its initial level. A typical force-displacement record of the interaction between a HL-60 cell and immobilized VCAM-1 is presented in Fig. 2B, top trace. At the beginning of the force measurement, before the cell made contact with the substrate, there was no force on the cantilever. Expansion of the piezoelectric translator lowered the cell onto the substrate. After cell-substrate contact, a compression force was gradually built on the cell and caused an upward deflection of the cantilever. The maximal compression force was set at 200 pN and was held constant for 10 s. On retraction of the cantilever, if there were adhesive interactions established between the cell and the substrate, the cantilever would be pulled downward. In the example illustrated in Fig. 2B, top trace, the deadhesion process involved a series of six rupture events. Each of these rupture events resulted in a rapid jump in force (indicated by arrows) and may represent the unbinding of one or more adhesive ligand-receptor bonds. The magnitudes of these ruptures were between 40 and 100 pN, consistent with the reported unbinding force of single integrin-ligand bonds (17, 30). After the addition of a function-blocking antibody against VCAM-1, the number of the rupture events was largely reduced (Fig. 2B, middle trace). The AFM measurements thus revealed that cell deadhesion mediated by adhesion molecules is a complex process involving multiple rupture events.



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 2. AFM measurements of HL-60-substrate adhesion. A: schematics of the principal events during an AFM force vs. distance measurement: 1) approach of the HL-60 cell to a surface coated with adhesion molecules, 2) contact between cell and substrate, 3) retraction of the HL-60 cell, and 4) separation of the cell from the substrate. Arrows indicate the direction of cantilever movement. B: typical force spectrum traces for HL-60 cells bound to different substrates. Strong adhesion occurred between HL-60 and immobilized VCAM-1 (top trace). Adhesion between HL-60 and immobilized VCAM-1 was markedly reduced by a function-blocking antibody against VCAM-1 (middle trace). HL-60 adhered weakly to immobilized BSA (bottom trace). Measurements were acquired with a compression force of 200 pN, 10 s contact, and a cantilever retraction speed of 3 µm/s. Shaded area in the top trace is the "work of deadhesion." Arrows in the top trace point to rupture events; breakage of adhesive bond(s). Dashed lines indicate zero forces.

 

To obtain a quantitative measure of the adhesion strength on a whole cell level, we introduced two adhesive indexes: one being the detachment force and the other the work of deadhesion (Fig. 2B). Detachment force is the maximal adhesive force detected by the cantilever during cell-substrate separation. The work of deadhesion, as indicated by the shaded area in Fig. 2B, top trace, is the work done by the cantilever to separate the cell and was derived from integrating the adhesive force over the distance traveled by the cantilever up to the point of the last bond rupture. It included the work done to break the molecular linkages and the work done to stretch the cells. For the experimental conditions shown in Fig. 2B, our data revealed an average detachment force of 320 ± 30 pN and the work of deadhesion of 7.8 ± 1.3 x 10–16 J. In the presence of a function-blocking antibody against VCAM-1, both the detachment force and work of deadhesion were substantially reduced to 80 ± 8 pN (75% decrease) and 3 ± 0.8 x 10–17 J (96% decrease), respectively. Hence, both indexes seem to be specific in reflecting the adhesion between HL-60 and the substrate. More significant difference was found if adhesion is measured by the work of deadhesion. This is probably due to the fact that the work of deadhesion reflects both the difference in force and the difference in the stretching distance during cell-substrate separation. Similar results were seen on HL-60 cell adhesion to other adhesion molecules, including P-selectin, E-selectin, and ICAM-1 (data not shown).

We next tested whether AFM method is suitable for a surface presenting different types of adhesion molecules, which closely resemble activated endothelium. Dishes coated with both VCAM-1 and P-selectin were studied (Fig. 3). In these experiments, anti-VCAM-1 or anti-P-selectin alone only partially inhibited adhesion. The adhesion was completely blocked in the presence of both antibodies but was not affected by nonspecific mouse Ig (P > 0.05) (Fig. 3). Hence, these experiments indicated that the AFM method could specifically reveal the adhesion strength of the HL-60 cell interacting with different endothelial-derived adhesion molecules and should be suitable for studying leukocyte adhesion to vascular endothelial cells. In addition, these experiments proved the potency of each of function-blocking antibodies against specific adhesion molecules and validated their usage for blocking adhesion on endothelial cells.



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 3. A: detachment forces. B: work of deadhesion of HL-60 cells bound to immobilized BSA or to immobilized VCAM-1 and P-selectin. Inhibitory antibodies used were AF809 (anti-VCAM-1) and WAPS (anti-P-selectin). Experimental conditions were the same as in Fig. 2. Error bar is the standard error with N > 15 in each case. *P < 0.05 compared with the control (P-selectin and VCAM-1). #P < 0.01 compared with anti-VCAM-1 alone or with anti-P-selectin alone; ns, Not significant (P > 0.05).

 

AFM measurement of leukocyte-HUVEC interaction. Experiments were carried out between a HL-60 cell attached to the end of an AFM cantilever and a HUVEC monolayer plated on a tissue culture dish (Fig. 1C). The HUVEC monolayer was activated by TNF-{alpha}, a well-defined in vitro model of the activated endothelium. A complete cycle of the AFM force scan is illustrated in Fig. 4A and is similar to that of Fig. 2A. Figure 4B, top trace, represents a typical force curve of the interaction between a HL-60 cell and the cell body of an activated HUVEC. The deadhesion process involved multiple rupture events before finial separation. Several well-defined rupture events are highlighted in Fig. 4B. The magnitude of these rupture events was between 40 and 100 pN, which is consistent with the reported unbinding force of selectin or integrin bonds (9, 17, 24, 30). Figure 4B, middle trace, presents a record of the interaction between HL-60 and unstimulated HUVEC, which showed significantly fewer bond-rupturing events and a lower work of deadhesion. The enhanced adhesion of HUVEC following TNF-{alpha} activation has been attributed to an upregulation of adhesion molecules such as E-selectin, VCAM-1, and ICAM-1 (3, 20). A significant decrease in adhesion between HL-60 and TNF-{alpha}-activated HUVEC was observed in the presence of a cocktail of function-blocking antibodies against P, E-selectins, VCAM-1, and ICAM-1. (Fig. 4B, bottom trace).



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 4. AFM measurements of HL-60-HUVEC adhesion. A: principal events of AFM measurements: 1) approach of the HL-60 to the HUVEC, 2) contact occurring between HL-60 and HUVEC, 3) retraction of the HL-60, and 4) its separation from HUVEC. Arrows indicate the direction of cantilever movement. B: typical force spectrum traces for HL-60 cells bound to the cell body of HUVECs. Strong adhesion was found between HL-60 and TNF-{alpha} activated HUVECs (top trace). HL-60 adhered weakly to the resting HUVECs (middle trace). HL-60 adhesion to activated HUVEC was largely reduced by of a cocktail of monoclonal antibodies against P-, E-selectins, ICAM-1, and VCAM-1 (bottom trace). Arrows in top trace point to some of the well-defined ruptures of adhesive bond(s). Experimental conditions were the same as in Fig. 2. Dashed lines indicate zero forces. Shaded area in top trace is the "work of deadhesion."

 

Past single molecule studies demonstrated that the unbinding force of individual ligand-receptor bonds varies more or less with the loading rates applied to the bonds. Experimentally, the loading rate is adjusted by changing the separation speeds. To find out whether our experimental result is altered by separation speeds (loading rates), we tested three different separation speeds: 0.5, 3, and 6 µm/s (Fig. 5). As shown, varying the speed 10-fold only induced a subtle difference in adhesion (P > 0.05). This result demonstrated that in the loading regime being tested in this study, the loading rate is not a key factor that significantly affects the experimental results. In an in vivo study using intravital microscopy, Kunkel et al. (14) showed that the speed of leukocyte rolling on a TNF-{alpha}-activated endothelium was between 0.3 and 9 µm/s, with an average speed of 3.9 µm/s. Hence, the separation speed used in this study, i.e., 3 µm/s, is close to the physiological separation speed of leukocyte-endothelial interaction. The loading regime being tested in Fig. 5 is physiologically relevant.



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 5. Effect of separation speed on detachment force (A) and work of deadhesion (B). HL-60 cell adhesion was tested on the cell bodies of activated HUVECs (squares) and immobilized VCAM-1 and P-selectin (circles). Error bar is the standard error and N > 10 in each case.

 

Another factor that may affect the loading rate is the mechanics of both the HL-60s and the HUVECs. The mechanical properties of these two cell types were measured by AFM indentation experiments. The measurements were made with indentations of <1 µm on HL-60s and <300 nm on HUVECs. Estimates of Young's modulus were based on the Hertz model (27). The average value for Young's modulus (N > 100; 10–15 cells) of the HL-60s was 0.4 kPa, and the Young's modulus of the resting and activated HUVECs ranged from 5 to 10 kPa. Because the HL-60 cells are at least one order of magnitude softer than the HUVECs, the loading rate is mainly determined by the elasticity of the HL-60s during their separation from HUVEC.

Monocytes adhered more tightly to the cell border of a HUVEC monolayer. Past observations by Gopalan et al. (10) demonstrated that neutrophils preferentially adhered to the cell-cell junctions of a HUVEC monolayer. Hence, we decided to characterize the adhesive properties of PMA-differentiated HL-60 to the cell body of individual HUVEC and to the cell borders of the HUVEC monolayer by direct force measurements. Before the measurements, the HL-60 cells were carefully positioned above the cell bodies of individual HUVECs or above the cell-cell junctions formed by two adjacent HUVECs (Fig. 1B). During each force measurements, the contact zones between HL-60s and HUVECs were visualized using a light microscope equipped with a x40 objective to ensure that the cell-cell contacts occurred within HUVEC cell bodies or at cell-cell junctions.

Figure 6 plots the detachment force and work of deadhesion of HL-60 bound to either the cell body or cell border of a HUVEC monolayer. The AFM force measurement revealed that the adhesion depended on the site where the HL-60 makes contact with the endothelial cell monolayer. When HUVEC was activated by TNF-{alpha}, the work of deadhesion was 105% higher at HUVEC cell borders compared with that of the cell bodies. However, in the resting HUVEC, the adhesion in cell bodies and borders only differed by 59%. Thus our measurements revealed that after activation, the enhancement of adhesion was more significant at the HUVEC cell borders.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 6. Adhesion strength between HL-60s and HUVEC monolayers measured by detachment force (A) and work of deadhesion (B). Open and closed bars represent HUVEC borders and bodies, respectively. Inhibitory antibodies used were WAPS (anti-P-selectin), BBIG-E4 (anti-E-selectin), BBIG-I1 (anti-ICAM-1), and AF809 (anti-VCAM-1). Data are presented as means ± SE of 5–12 experiments. *P < 0.05 compared with the respective control [tumor necrosis factor (TNF)-{alpha}-stimulated HUVEC]. #P < 0.05 compared with anti-selectin alone, anti-ICAM-1 alone, or with anti-VCAM-1 alone.

 

Monocyte-HUVEC adhesion is partially inhibited by antibodies against P, E-selectins, ICAM-1, and VCAM-1. It has been previously reported that TNF-{alpha} increases the expression level of E-selectin, ICAM-1, and VCAM-1 on vascular endothelial cells (3, 20). To more precisely define the relative contribution of these three molecules to leukocyte-HUVEC adhesion, the effect of function-blocking antibodies against these molecules was tested separately. Addition of antibody against P, E-selectins, ICAM-1, or VCAM-1 yielded a partial inhibition of the adhesion strength between HL-60 and TNF-{alpha} activated HUVEC (Fig. 6). Regarding the work of deadhesion, anti-selectins and anti-ICAM-1 led to a significant inhibition of work of deadhesion in HUVEC cell boders but did not show significant blockade in the HUVEC cell bodies (Fig. 6B). In contrast, anti-VCAM-1 could inhibit HL-60 adhesion to both HUVEC cell bodies and cell borders significantly. Anti-VCAM-1 has the most potent inhibitory effect among these function-blocking antibodies, reducing 47% and 30% of adhesion strength in HUVEC cell borders and cell bodies, respectively. Surprisingly, a cocktail of all the four antibodies only resulted in a 65% and 55% blockade of adhesion strength in HUVEC cell borders and cell bodies, respectively, suggesting that ~40% of adhesion strength could be contributed by molecules other than P, E-selectins, ICAM-1, and VCAM-1. A similar result was also found when adhesion strength is measured by detachment force (Fig. 6A). The adhesion between HL-60 and activated HUVECs was not affected by mouse Ig used as a negative control.

Monocyte adhesion is inhibited by a cRGD peptide. We have previously observed that a cRGD peptide exerted an inhibitory effect on leukocyte-endothelial cell adhesion (21). Together with the above demonstration of an only partial adhesion blockade achieved with all four neutralizing antibodies, we were prompted to investigate in depth the role of RGD peptides in leukocyte-endothelial interactions.

Addition of the cRGD peptide resulted in a dose-dependent decrease in work of deadhesion with an IC50 {approx} 100 nM (Fig. 7A). The effects of cRGD peptide were independent of the site of adhesion to HUVEC. This effect was not observed with the cRAD peptide. The maximum blockade, performed by 10 µM of cRGD, led to a 60% reduction of adhesion strength in HUVEC borders and a 53% reduction in HUVEC bodies. This inhibitory potency is almost identical to that of the sum of function-blocking antibodies against P-, E-selectins, ICAM-1, and VCAM-1.



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 7. A: inhibitory effect of different concentrations of cRGD peptide on the interaction of HL-60 and TNF-{alpha}-stimulated HUVEC (open symbols). Squares and circles are measurements obtained from the cell body and cell borders of HUVEC, respectively. Solid symbols represent measurements obtained with 10 µM cRGD. B: similar effect of cRGD (10 µM) and anti-{beta}1 on HL-60-HUVEC adhesion. Adhesion is not affected by anti-{alpha}V{beta}3 (P > 0.05). Open and shaded bars represent HUVEC borders and bodies, respectively. Data are presented as means ± SE of 5–12 experiments. C: dose-dependent inhibition of cell adhesion by cRGD detected using a standard cell adhesion assay. *P < 0.05 compared with the respective control (TNF-{alpha}-stimulated HUVEC).

 

The known RGD recognizing integrins involved in leukocyte-endothelial interactions are the {beta}1-integrins and {alpha}V{beta}3 (28). In an attempt to identify the integrins involved in cRGD blockade, we tested two function-blocking monoclonal antibodies, i.e., JB1A (anti-{beta}1) and LM609 (anti-{alpha}V{beta}3) for their ability to inhibit adhesion. Both JB1A and LM609 are well-characterized reagents and have been shown to block cell adhesion mediated by {beta}1-integrins and {alpha}V{beta}3, respectively (5, 11, 13, 18). Neutralizing antibody against the {beta}1-integrin significantly suppressed HL-60-HUVEC interactions. However, neutralizing antibodies against the {alpha}V{beta}3-integrin did not change HL-60 adhesion to HUVEC monolayers (Fig. 7B). The combined effect of anti-{beta}1 and cRGD was not significantly different from the inhibitory action of each individual compound (P < 0.05) (Fig. 7B), implying that the cRGD exerts this effect predominantly via the {beta}1-integrins. It should be noted that inhibition of {beta}1-integrins reduced adhesion by 60%, better than any other specific function-blocking antibodies tested. Hence, our data suggest that {beta}1-integrins play an essential role in HL-60-HUVEC adhesion.

Because the cRGD seems to block the interaction of {beta}1-integrins, we decided to test whether it inhibits VLA-4-VCAM-1 interaction. VLA-4 is one of the {beta}1-integrins expressed in HL-60 cells (confirmed by FACS, data not shown). In the presence of 10 µM cRGD and the function-blocking monoclonal antibody to VCAM-1, we observed a 65% inhibition in work of deadhesion at HUVEC borders and a 52% inhibition at HUVEC bodies. This effect is very similar to that of cRGD alone (P > 0.05) (Fig. 7B), suggesting that cRGD did interfere with VLA-4-VCAM-1 interaction. However, the effect of cRGD is better than that of the VCAM-1 antibody alone, indicating that cRGD also has other targets. In another experiment, we added 10 µM cRGD and monoconal antibodies against P-, E-selectins, ICAM-1, and VCAM-1 (Fig 7B). In this experiment, work of deadhesion was further reduced to 75% and 65% in cell borders and bodies, respectively. Taken together, these results demonstrate that the action of cRGD is mediated through a blockade of VLA-4-VCAM-1 interaction and of some other {beta}1-integrin interactions.

Using a standard adhesion assay, we were able to confirm the above findings. Adhesion of PMA-differentiated HL-60 cells to endothelial monolayers pretreated with TNF-{alpha} was three to four times higher than to inactivated endothelial cells. Addition of cRGD peptide resulted in a dose-dependent inhibition of leukocyte adhesion with IC50 {approx} 10 nM. In contrast, the cRAD peptide had no effect on HL-60-HUVEC adhesion (Fig. 7C).

Monocyte transmigration is inhibited by cRGD, anti-{beta}1, and anti-{alpha}V{beta}3. Transmigration of HL-60 cells was studied using a modified Boyden chamber. In the presence of cRGD, HL-60 transmigration was reduced by 60%. Addition of neutralizing anti-{alpha}V{beta}3 or anti-{beta}1 antibodies or their combination resulted in a partial but significant inhibition of transmigration, which was not significantly different from the inhibitory effect of cRGD peptide (P > 0.05) (Fig. 8). The peptide inhibited transmigration with the IC50 {approx} 100 nM. cRAD peptide did not affect the transmigration of PMA-differentiated HL-60 (P > 0.05).



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 8. A: dose-dependent inhibition of transmigration by cRGD. B: inhibition of HL-60 transmigration by antibodies against {beta}1-integrins and {alpha}V{beta}3. *P < 0.05 compared with the control (TNF-{alpha}-activated HUVEC). ns, Not significant (P > 0.05).

 


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In this work, we have modified the AFM to quantitatively measure the molecular forces involved in the course of adhesive interactions between monocytes and endothelial cells in culture. The data demonstrated that cell adhesion occurs predominantly at the borders of endothelial cells, and TNF-{alpha} significantly enhances the adhesive strength at these sites, in accord with previous reports (10, 20). A cRGD peptide attenuates adhesion by interfering with VLA4-VCAM-1 interaction.

An important difference between the AFM and other cell adhesion assays is that the AFM measures in real time the change of force and energy instead of counting the bound cell numbers afterward. Because the bound cell numbers do not reflect any biophysical properties, they can only be considered as a rough estimate of cell adhesion. The AFM method has overcome the previous technical difficulties in cell adhesion assays that prohibited examining the detailed dynamics of leukocyte-endothelial interaction. The spatial resolution of the AFM also allowed us to measure adhesion of a single leukocyte to different regions in an endothelial monolayer. Using the AFM approach, we have provided the first direct evidence of higher adhesive force developed in the HUVEC borders compared with the cell bodies. The mechanical work for separating a HL-60 cell from the cell borders of an activated HUVEC monolayer has been found to be twice as much as that of the HUVEC bodies. It has been shown that a vast majority of leukocyte transmigration occurred at the endothelial cell-cell junctions (28). Higher adhesive force at the endothelial cell borders may facilitate the accumulation of leukocytes into the proximity of endothelial cell junctions, thus favoring transendothelial migration. Certain factors could potentially account for this preferential distribution of adhesive forces: different expression levels of adhesion molecules, affinity or avidity activation of some adhesion molecules, and the geometry of the HUVEC monolayer. Past studies (1) using the AFM-imaging technique have revealed the surface contour of a bovine aortic endothelial monolayer. The boundaries between cells were visualized, and the average differential from the highest point of the cells (i.e., the nuclear region) to the lowest point at cell-cell junctions was averaging several microns. Hence, when a HL-60 cell was brought to interact with the junction formed by two HUVECc, it is conceivable that the round HL-60 cell (10–15 µm in diameter) could fit into the groove formed by adjacent HUVECs nicely. This would lead to an increase of the contact area between HL-60 and HUVECs and could potentially account for the enhanced adhesion in HUVEC borders. However, this hypothesis remains to be experimentally verified.

When cell adhesion on a surface presenting different adhesion molecules is measured, the AFM also provides an estimate of the relative contribution of each molecule in supporting whole cell adhesion. In the case illustrated in Fig. 3B, the adhesive system includes two adhesion molecules: VCAM-1 and P-selectin. Because the addition of function-blocking antibodies can completely eliminate adhesion, it can be estimated from the work of deadhesion that P-selectin and VCAM-1 contribute to 40% and 60% of total adhesion, respectively. In the case of HL-60-HUVEC interaction (Fig. 6), although the total adhesion was not completely blocked by a cocktail of all function-blocking antibodies, we could still roughly estimate the ratio of adhesion contributed by selectins, ICAM-1, and VCAM-1 as 1:2:2.5 in HUVEC borders and 1:4:6 in HUVEC bodies. Our data revealed that VCAM-1 supports monocyte adhesion to HUVEC better than ICAM-1 and selectins. This finding is consistent with a recent report suggesting a major role for VCAM-1 in the early stages of atherosclerosis (8).

Another major finding reported herein is an unexpected inhibitory effect of nanomolar concentrations of cRGD peptide on the adhesion of monocytic HL-60 to the activated endothelial cells and their transmigration across endothelial monolayer. The current paradigm on adhesive interactions engaged in these processes does not predict such an action; nonetheless, the observed effects of cRGD peptide (but not cRAD peptide) were so profound and confirmed using direct force measurements that they could not be ignored or ascribed to an artifact. The finding that effects of cRGD peptide could not be mimicked by the neutralizing anti-{alpha}V{beta}3 antibody would indicate that this receptor is not directly involved in the initial leukocyte tethering and adhesion. In contrast, anti-{beta}1-integrin neutralizing antibody mimicked the effect of cRGD and their combined effect was nonadditive, suggesting that the cRGD peptide mediates its effect on monocyte-endothelial adhesion via {beta}1-integrins. Consistently, the data revealed that cRGD peptide interfered with the VLA-4-VCAM-1 interaction. However, most likely this is not the only target that the cRGD inhibits. It is conceivable that the cRGD peptide exerts its action via a receptor(s) previously not known to recognize the RGD motif. Indeed, the recent finding that disintegrins affect leukocyte-endothelial interactions (25) provides additional support to the above observation.

Our data also show that HL-60 transmigration is inhibited by both anti-{beta}1 and anti-{alpha}V{beta}3 antibodies. Therefore, it is not surprising that cRGD blocked HL-60 transmigration. Interestingly, no additional inhibition was found when {alpha}4{beta}1 and {alpha}v{beta}3 antibodies were added together. This may suggest that {alpha}4{beta}1 and {alpha}v{beta}3 are engaged sequentially during monocyte transmigration. It is also conceivable that the ligands of {alpha}v{beta}3, such as L1 or PECAM, are localized within the endothelial cell-cell junctions (19, 26) and are not accessible to {alpha}v{beta}3 during the initial adhesive contact. Our data suggest that {alpha}v{beta}3 plays a significant role at a step after leukocyte firm adhesion to vascular endothelium.

In conclusion, we have developed a highly sensitive method to study the adhesive force in leukocyte-endothelial interaction. This technique is not limited to the chosen system and can potentially be used to study a broad range of cell-cell or cell-substrate interactions. The data demonstrate that the interaction of {beta}1-integrins and their ligands play an important role in both leukocyte-endothelial adhesion and leukocyte transmigration. Both processes were substantially suppressed by a synthetic cRGD peptide. Our findings extend the previous usage of RGD peptides in blocking cell-to-matrix adhesion and raise the possibility of their therapeutic applicability in suppressing monocytic infiltration of the vascular wall, the necessary precursor in the formation of "fatty streaks."


   ACKNOWLEDGMENTS
 
GRANTS

These studies were supported in part by ACS and National Institutes of Health Grants GM-55611 (to V. T. Moy), DK-45462, DK-54602, and DK-45695 (to M. S. Goligorsky). X. Zhang was supported by a fellowship (0215139B) from the American Heart Association. H. Li was supported by the training grant from the National Institutes of Health T32DK-07521-14.


   FOOTNOTES
 

Address for reprint requests and other correspondence: V. T. Moy, Dept. of Physiology and Biophysics, Univ. of Miami School of Medicine, Miami, FL 33101-6430 (E-mail: vmoy{at}miami.edu) or M. S. Goligorsky, Dept. of Medicine, New York Medical College, Valhalla, NY 10595 (E-mail: Michael_Goligorsky{at}nymc.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
 REFERENCES
 

  1. Barbee KA, Mundel T, Lal R, and Davies PF. Subcellular distribution of shear stress at the surface of flow-aligned and nonaligned endothelial monolayers. Am J Physiol Heart Circ Physiol 268: H1765–H1772, 1995.[Abstract/Free Full Text]
  2. Benoit M. Cell adhesion measured by force spectroscopy on living cells. Methods Cell Biol 68: 91–114, 2002.[Web of Science][Medline]
  3. Bevilacqua MP. Endothelial-leukocyte adhesion molecules. Annu Rev Immunol 11: 767–804, 1993.[CrossRef][Web of Science][Medline]
  4. Binnig G, Quate CF, and Gerber C. Atomic force microscope. Physiol Rev Lett 56: 930–933, 1986.
  5. Brooks PC, Clark RA, and Cheresh DA. Requirement of vascular integrin alpha v beta 3 for angiogenesis. Science 264: 569–571, 1994.[Abstract/Free Full Text]
  6. Carlos TM and Harlan JM. Leukocyte-endothelial adhesion molecules. Blood 84: 2068–2101, 1994.[Abstract/Free Full Text]
  7. Chen A and Moy VT. Cross-linking of cell surface receptors enhances cooperativity of molecular adhesion. Biophys J 78: 2814–2820, 2000.[Web of Science][Medline]
  8. Cybulsky MI, Iiyama K, Li H, Zhu S, Chen M, Iiyama M, Davis V, Gutierrez-Ramos JC, Connelly PW, and Milstone DS. A major role for VCAM-1, but not ICAM-1, in early atherosclerosis. J Clin Invest 107: 1255–1262, 2001.[Web of Science][Medline]
  9. Evans E, Leung A, Hammer D, and Simon S. Chemically distinct transition states govern rapid dissociation of single L-selectin bonds under force. Proc Natl Acad Sci USA 98: 3784–3789, 2001.[Abstract/Free Full Text]
  10. Gopalan PK, Burns AR, Simon SI, Sparks S, McIntire LV, and Smith CW. Preferential sites for stationary adhesion of neutrophils to cytokine-stimulated HUVEC under flow conditions. J Leukoc Biol 68: 47–57, 2000.[Abstract/Free Full Text]
  11. Grzeszkiewicz TM, Lindner V, Chen N, Lam SC, and Lau LF. The angiogenic factor cysteine-rich 61 (CYR61, CCN1) supports vascular smooth muscle cell adhesion and stimulates chemotaxis through integrin alpha(6)beta(1) and cell surface heparan sulfate proteoglycans. Endocrinology 143: 1441–1450, 2002.[Abstract/Free Full Text]
  12. Hutter JL and Bechhoefer J. Calibration of atomic-force microscope tips. Rev Sci Instrum 64: 1868–1873, 1993.[CrossRef]
  13. Kaul DK, Tsai HM, Liu XD, Nakada MT, Nagel RL, and Coller BS. Monoclonal antibodies to alphaVbeta3 (7E3 and LM609) inhibit sickle red blood cell-endothelium interactions induced by platelet-activating factor. Blood 95: 368–374, 2000.[Abstract/Free Full Text]
  14. Kunkel EJ, Dunne JL, and Ley K. Leukocyte arrest during cytokine-dependent inflammation in vivo. J Immunol 164: 3301–3308, 2000.[Abstract/Free Full Text]
  15. Lasky LA. Selectins: interpreters of cell-specific carbohydrate information during inflammation. Science 258: 964–969, 1992.[Abstract/Free Full Text]
  16. Lefer DJ. Pharmacology of selectin inhibitors in ischemia/reperfusion states. Annu Rev Pharmacol Toxicol 40: 283–294, 2000.[CrossRef][Web of Science][Medline]
  17. Li F, Redick SD, Erickson HP, and Moy VT. Force Measurements of the alpha(5)beta(1) Integrin-Fibronectin Interaction. Biophys J 84: 1252–1262, 2003.[Web of Science][Medline]
  18. Ni H and Wilkins JA. Localisation of a novel adhesion blocking epitope on the human beta 1 integrin chain. Cell Adhes Commun 5: 257–271, 1998.[Web of Science][Medline]
  19. Piali L, Hammel P, Uherek C, Bachmann F, Gisler RH, Dunon D, and Imhof BA. CD31/PECAM-1 is a ligand for alpha v beta 3 integrin involved in adhesion of leukocytes to endothelium. J Cell Biol 130: 451–460, 1995.[Abstract/Free Full Text]
  20. Risau W. Differentiation of endothelium. FASEB J 9: 926–933, 1995.[Abstract]
  21. Romanov V, Noiri E, Czerwinski G, Finsinger D, Kessler H, and Goligorsky MS. Two novel probes reveal tubular and vascular Arg-Gly-Asp (RGD) binding sites in the ischemic rat kidney. Kidney Int 52: 93–102, 1997.[Web of Science][Medline]
  22. Sagvolden G, Giaever I, Pettersen EO, and Feder J. Cell adhesion force microscopy. Proc Natl Acad Sci USA 96: 471–476, 1999.[Abstract/Free Full Text]
  23. Springer TA. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76: 301–314, 1994.[CrossRef][Web of Science][Medline]
  24. Tees DF, Waugh RE, and Hammer DA. A microcantilever device to assess the effect of force on the lifetime of selectin-carbohydrate bonds. Biophys J 80: 668–682, 2001.[Web of Science][Medline]
  25. Vassilev TL, Kazatchkine MD, Van Huyen JP, Mekrache M, Bonnin E, Mani JC, Lecroubier C, Korinth D, Baruch D, Schriever F, and Kaveri SV. Inhibition of cell adhesion by antibodies to Arg-Gly-Asp (RGD) in normal immunoglobulin for therapeutic use (intravenous immunoglobulin, IVIg). Blood 93: 3624–3631, 1999.[Abstract/Free Full Text]
  26. Voura EB, Ramjeesingh RA, Montgomery AM, and Siu CH. Involvement of integrin alpha(v)beta(3) and cell adhesion molecule L1 in transendothelial migration of melanoma cells. Mol Biol Cell 12: 2699–2710, 2001.[Abstract/Free Full Text]
  27. Wojcikiewicz EP, Zhang X, Chen A, and Moy VT. Contributions of molecular binding events and cellular compliance to the modulation of leukocyte adhesion. J Cell Sci 116: 2531–2539, 2003.[Abstract/Free Full Text]
  28. Worthylake RA and Burridge K. Leukocyte transendothelial migration: orchestrating the underlying molecular machinery. Curr Opin Cell Biol 13: 569–577, 2001.[CrossRef][Web of Science][Medline]
  29. Yusuf-Makagiansar H, Anderson ME, Yakovleva TV, Murray JS, and Siahaan TJ. Inhibition of LFA-1/ICAM-1 and VLA-4/VCAM-1 as a therapeutic approach to inflammation and autoimmune diseases. Med Res Rev 22: 146–167, 2002.[CrossRef][Web of Science][Medline]
  30. Zhang X, Wojcikiewicz E, and Moy VT. Force spectroscopy of the leukocyte function-associated antigen-1/intercellular adhesion molecule-1 interaction. Biophys J 83: 2270–2279, 2002.[Web of Science][Medline]



This article has been cited by other articles:


Home page
 FASEB J.Home page
A. P. Gunning, S. Chambers, C. Pin, A. L. Man, V. J. Morris, and C. Nicoletti
Mapping specific adhesive interactions on living human intestinal epithelial cells with atomic force microscopy
FASEB J, July 1, 2008; 22(7): 2331 - 2339.
[Abstract] [Full Text] [PDF]

Home page
 JNCI J Natl Cancer InstHome page
J. H. Rhim and G. Tosato
Targeting the Tumor Vasculature to Improve the Efficacy of Oncolytic Virus Therapy
J Natl Cancer Inst, December 5, 2007; 99(23): 1739 - 1741.
[Full Text] [PDF]

Home page
 BloodHome page
S. Subramanian, R. Parthasarathy, S. Sen, E. T. Boder, and D. E. Discher
Species- and cell type-specific interactions between CD47 and human SIRP{alpha}
Blood, March 15, 2006; 107(6): 2548 - 2556.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Renal Physiol.Home page
S. Elitok, S. V. Brodsky, D. Patschan, T. Orlova, K. M. Lerea, P. Chander, and M. S. Goligorsky
Cyclic arginine-glycine-aspartic acid peptide inhibits macrophage infiltration of the kidney and carotid artery lesions in apo-E-deficient mice
Am J Physiol Renal Physiol, January 1, 2006; 290(1): F159 - F166.
[Abstract] [Full Text] [PDF]

Home page
 Nephrol Dial TransplantHome page
M. S. Goligorsky
Whispers and shouts in the pathogenesis of acute renal ischaemia
Nephrol. Dial. Transplant., February 1, 2005; 20(2): 261 - 266.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
286/1/H359    most recent
00491.2003v1
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 (32)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, X.
Right arrow Articles by Goligorsky, M. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhang, X.
Right arrow Articles by Goligorsky, M. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2004 by the American Physiological Society.