Adhesion of Sickle Red Blood Cells and Damage to Interleukin-lp ...

Adhesion of Sickle Red Blood Cells and Damage to Interleukin-lp Stimulated Endothelial Cells Under Flow In Vitro

By Maya Natarajan, Mark M. Udden, and Larry V. Mclntire

Two factors that are hypothesized to contribute to vasoocclusive crises in sickle cell anemia are increased sickle red blood cell-endothelial cell interactions and damage to endothelium. Despite considerable study, the mechanisms by which erythrocyte-endothelial interactions occur and the role of endothelial damage have not yet been fully eluci- dated. In this report, we demonstrate that adhesion and damage may be related in a model of vasoocclusion in sickle cell anemia. Phase contrast microscopy coupled to digital image processing was used to determine the adhesion of sickle red blood cells to 1-, 4-, and 24-hour interleukin-lp (IL-1s) stimulated endothelial cells in a parallel plate flow chamber. Morphological alterations to activated endothelial cells after the perfusion of sickle erythrocytes were also identified. Pretreatment of monolayers with 50 pg/mL of

ICICLE CELL DISEASE is marked by periodic vasooc- S clusive crises that are accompanied by pain and end organ damage. Enhanced sickle red blood cell-endothelial cell (SSRBC-EC) interactions may be a contributing factor in the initiation and progression of these vasoocclusive crises. Increased adhesion of sickle erythrocytes to endothelium has been demonstrated in both static and dynamic in vitro experimental and shown to be mediated by several mechanisms.

Acute pain crises in sickle cell anemia are usually pre- ceded by or associated with infection (both viral and bacte- rial) and/or inflammation.' During inflammation, agonists such as histamine and thrombin and cytokines like interleukin-Ip (IL-1p) and tumor necrosis factor-a (TNF-a), are released. These activate endothelial cells leading to a time dependent upregulation of receptors of the selectin and im- munoglobulin Peak expression of E-selectin occurs approximately 3 to 4 hours after cytokine stimulation of endothelial cells and returns to basal levels in 24 hours. Intercellular adhesion molecule- 1 (ICAM- 1) and vascular cell adhesion molecule-1 (VCAM-1) are upregulated maxi- mally after 9 hours of cytokine stimulation of endothelial cells and stay expressed up to 24 hours and beyond. E- selectin has been shown to mediate leukocyte rolling on endothelial cells under flow conditions,' whereas ICAM-1 and VCAM-1 support firm adhesion and transmigration of leukocytes to stimulated The counter receptors for ICAM- 1 and VCAM- 1, which are found on circulating blood cells, include members of the integrin family. Integrins are heterodimeric proteins that mediate cell-cell and cell-extracellular matrix interaction^."-'^ Extracellular matrix proteins often interact with integrins via Arginine- glycine-aspartic (RGD) peptides that are located on the basal side of endothelial cells. Recent reports show that integrin expression can also be demonstrated on the apical side of endothelial cells in ~u l tu re . ' ~ . ' ~ Although a number of investi- gations have shown that these receptor families support leukocyte adhesion to endothelial cell^,^^'^^'^-'^ relatively few studies document their role in sickle erythrocyte-endothelial interactions. A role for stimulation of endothelial cells by cytokines in sickle cell disease is suggested by the work of

IL- lp for 1, 4, and 24 hours caused approximately 16-fold increases in adhesion of sickle cells to activated endothelium at all time points. Results with an Arginine-glycine aspartic acid (RGD) peptide and monoclonal antibodies indicated a role for three different endothelial cell receptors: a& after 1 hour of IL-1p stimulation; E-selectin after 4 hours of IL- lp stimulation; and vascular cell adhesion molecule-1 after prolonged exposure t o cytokines. Perfusion of sickle, but not normal, erythrocytes resulted in alteration of endothelial morphology. Approximately 6% t o 8% damage was observed on 4- and 24-hour IL-lp stimulated endothelial cells after the perfusion of sickle cells. Damage to 24-hour activated endothelial cells showed a positive correlation ( r = .899) with the number of adherent sickle erythrocytes. 0 1996 by The American Society of Hematology.

Swerlick et al." In that study, it was demonstrated that 9- hour TNF-a stimulation of endothelial cells resulted in increased adhesion that was blocked by antibodies to a4pl on sickle erythrocytes and VCAM-1 on endothelial cells. Evidence for increased adhesion of sickle cells to cytokine stimulated endothelial cells after short-term periods of stimulation is lacking. In this report, we demonstrate that short- term stimulation of endothelial cells with IL-1p also increases sickle erythrocyte-endothelial interactions. In addition, we characterize the contributions of specific adhesion receptors on endothelial cells using monoclonal antibody (MoAb) blocking studies and flow cytometry at three different time points- 1,4, and 24 hours post-IL-lp stimulation.

Damage to the endothelium may also result from adhesion of sickle cells. Morphologic changes consistent with endothelial injury have been described in the circulation of the spleen and brain of sickle patients.2'.22 Increased circulating endothelial cells have also been detected in the blood of sickle cell patients during painful crisis.'' The exposure of human umbilical vein endothelial cells (HUVEC) to sickle erythrocytes in vitro results in increased prostacyclin produc- tion by the monolayers" and inhibition of DNA synthesis by these cells.*' Sickle erythrocytes, therefore, appear to directly affect endothelial cell metabolism and function. In this study,

~ ~~~

From the Cox Laboratory for Biomedical Engineering, Rice Uni- versity, Houston; and the Department of Medicine, Baylor College of Medicine, Houston, 7X.

Submitted March 31s 1995; accepted January 24, 1996. Supported by the National Institutes of Health, Grant No. ROI-

HL 50601, Texas Advanced Technology Program Grant No. 003604, and a grant from the Butcher Fund.

Address reprint requests to Larry V. Mclntire, PhD, Chair, Insti- tute of Biosciences and Bioengineering, G.R. Brown Hall, Rice Uni- versity, Houston, 7X 77251 - 1892.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advedsement" in accordance with I8 U.S.C. section 1734 solely to indicate this fact. 0 1996 by The American Society of Hematology. 0006-497I/96/87I I-OO15$3.OO/O

Blood, Vol 87, No 11 (June I ) , 1996: pp 4845-4852 4845

For personal use only.on March 26, 2018. by guest www.bloodjournal.orgFrom

http://www.bloodjournal.org/

http://www.bloodjournal.org/site/subscriptions/ToS.xhtml

4846 NATARAJAN, UDDEN, AND MclNTlRE

endothelial cell morphology after contact with sickle erythrocytes under flow conditions was examined to assess alterations associated with perfusion and adhesion of sickle cells.

MATERIALS AND METHODS

Peptides and MoAbs. GRGDSNPKY, a peptide sequence found in fibronectin, was kindly donated by Dr Zhong Yun (University of Texas, M.D. Anderson Cancer Center, Houston, TX).

The MoAbs used in this study were all preparations of IgGs and included: anti-CD51, recognizes the a, chain of the vitronectin receptor (VnR), Immunotech, Larrabee, ME; LM609, interferes with the RGD binding site of the a,@, dimer, kindly donated as ascites by Dr David Cheresh (Scripps Research Institute, La Jolla, CA); OKM5, binds GPIV, kindly donated by Ortho Diagnostics (Raritan, NJ); 6D1 and Ibl, anti-GPIb antibodies, kindly donated by Dr Barry Coller (Mt. Sinai School of Medicine, New York, NY) and Dr Zaverio Ruggeri (Scripps Research Institute), respectively; 7E3 and 10E5, anti GPIIbnIIa antibodies, kindly donated Dr Barry Coller; anti-CDw49e antibody, recognizes the as chain of the fibronectin receptor (FnR), Immunotech; CL2/6, binds E-selectin and blocks receptor-binding function, and R15.7, an antLCD18 antibody, kindly donated by Dr Wayne Smith (Baylor College of Medicine, Houston, TX); HP2/1, binds a, chain of VLA,, Immunotech; and E1/6, anti- VCAM-1 antibody, Becton Dickinson (San Jose, CA).

The antibodies that were used for immunocytometry of endothelial cells included LM609, anti-CDw49e antibody, CL2/6, R15.7, and fluorescein isothiocyanate (F1TC)-conjugated goat antimouse IgG, a fluorescent marker, Sigma (St Louis, MO).

RBC were separated from heparin anticoagulated venous blood of healthy adult and sickle cell anemia donors by centrifugation and were resuspended in serum free media (SFM; pH 7.2, containing 5.0 pg/mL bovine insulin, 5.0 pglmL human transferrin, 0.2% wt/vol human albumin, 0. I O pg/mL penicillin and streptomycin; and 0.292 mg/mL glutamine in M199). Blood that was drawn was centrifuged at lOOg for 10 minutes and plasma and buffy coat removed. Packed RBC were then centrifuged in a 30” angle-head rotor at 30,OOOg for 1 hour at 32°C. The height of the RBC column was then divided into three equal fractions and the top fraction was carefully removed and resuspended at a I % hematocrit in SFM.4.2h Previous work demonstrates that most of the adhesive cells are present in this top fraction4

RBC were diluted to 1 X IO6 cellslmL and incubated with the dye thiazole orange (TO) at room temperature for 1 hour. Reticulocyte percentage was determined using a FACS- Scan flow cytometer (Becton Dickinson, San Jose CA).

HUVECs were harvested as previously described.27 Primary venular endothelial cells were pooled from 5 to I O cords and suspended in complete M199 [cM199; medium M199 (GIBCO-BRL, Grand Island, NY) supplemented with 10% heat-inactivated fetal calf serum (Hyclone, Logan, UT), 0.1 mglml penicillin-streptomycin antibiotic (GIBCO-BRL, Grand Island, NY), and 0.292 mglml glutamine (GIBCO-BRL), 1 pg/mL low molecular weight heparin (Sigma) and 1 pg/mL hydrocortisone (Sigma)]. Cells were plated onto 35 mm tissue culture dishes and maintained at 37T , 5% COz humidified atmosphere for 3 to 5 days. Visually confluent monolayers were used for flow studies and flow cytometry experiments. In most cases, HUVECs were treated with 50 pg/mL L I P (R&D Systems, Minneapolis, MN) for I , 4, and 24 hours before the perfusion of RBC.

The HUVEC monolayer was placed in a parallel plate flow chamber, rinsed with SFM for 5 minutes to precondition the endothelial cells, perfused with a 1% hematocrit RBC suspension for 10 minutes, and rinsed again with SFM for 15 minutes to remove any nonadherent RBC. The SFM

Isolation of red blood cells (RBC).

Reticulocyte counts.

Endothelial cell culture.

Adhesion assay under fiow conditions.

Table 1. Reticulocyte Counts in Blood Samples

Whole RBC Top Fraction

Normal - 0.856 i- 0.23 Sickle 11.90 i- 2.91 16.76 2 3.26

Reticulocyte counts were obtained using thiazole orange staining. The mean percentage of reticulocytes was determined using flow cytometry as described in Materials and Methods. The number of experiments in each column was five.

and RBC suspensions were passed through the flow chamber at a wall shear stress of 1 dynelcm’, a typical postcapillary venular shear stress. All flow studies were performed at 37°C. Interactions between RBC and endothelium were observed using phase-contrast video microscopy (Diaphot-TMD microscope; Nikon, Inc, Garden City, NY; Panasonic camera WV-BL200, Yokohama, Japan) and analyzed using a digital image processing system (Sun Microsystems, Moun- tain View, CA; Inovision, Inc, Durham, NC). A computer controlled stage coupled to the digital image processing system was used to acquire images before and after perfusion of RBC over monolayers. RBC that adhered to the endothelium appeared as phase-bright ob- jects and were counted using digital imaging processing software. The number of RBC adherent at the end of a run was used as the adhesion value and converted to RBC per square millimeter. Inhibi- tion studies were performed with MoAbs incubated with sickle reticulocytes and/or enaothelial cells for 30 minutes at 37°C before perfusion, unless otherwise stated.

Endothelial cells were removed from the tissue culture dishes using 10 m m o K EDTA solu- tion and centrifuged at 200g for I O minutes. The effluent was dis- carded and all subsequent steps on the remaining pellet were performed at 4°C. The pellet was washed with Dulbecco’s phos- phate-buffered saline (D-PBS) (Ca+z/Mg+z free; Sigma) twice and incubated with LM609 (anti-avp3 antibody) or with anti-CDw49e antibody ( a n t i d antibody) or with CL2/6 (anti-E-selectin antibody) or R15.7 (anti-CD18 MoAb) for 30 minutes. E-selectin served as the positive control for endothelial cell activation, whereas CDIS served as the negative control. Background binding was detected using only the secondary antibody, goat antimouse Ig-FITC conju- gate. All antibodies were used at a final 10 pg/mL concentration. After the 30-minute incubation period, the pellet was once again washed with D-PBS and FITC-conjugated goat antimouse IgG (Sigma) was used as the second antibody for fluorescent labeling and detection. The cells were then fixed using 1% paraformaldehyde and were analyzed using a FACS-Scan flow cytometry system (Bec- ton Dickinson).

Quunt$cution of morphological ulterations to endothelial mono- luyers using digital image processing. Morphological alterations on stimulated endothelial monolayers were quantified using digital image processing. The loss of refractility and the absence of typical “cobblestone” morphology of endothelial cells were used as markers of damage. Percentage damage was calculated using AJA, X

100 where Ad is the damaged area and A, is the total area occupied by the endothelium. Ad was determined by digitally drawing a contour line around the damaged region and recording the number of pixels within the contour. A, corresponded to the total area occupied by the frame, ie, 512 X 484 pixels.

Results are presented as mean 5 SEM, where h is the number of separate experiments. Statistical assessments were made using a two-tailed Student’s t-test or Kolmogorov-Smimov statistics as appropriate. P values less than .05 ( P < .05) were considered as significant in all data shown.

Flow cytometry of endothelial cells.

Data presentation.




SICKLE RBC ADHESION AND DAMAGE TO ACTIVATED EC 4847

Normal 0 Sickle 140 I n=6*

n=13* I T 1' n=6**

24 0 1 4 Time of Stimulation of Endothelial Cells (hn)

Fig 1. Adhesion of RBCs t o unstimulated and stimulated endothelial cells under conditions of flow. Red blood adhesion lcellslsquare mm 2 SEMI was determined from images of stimulated endothelium acquired before and after perfusion of RBCs and analyzed using digital image processing. n = number of experiments. P values were calculated for differences between adhesion to stimulated and unstimulated endothelial cells: +P < .01 and **P .02.

RESULTS

Sickle erythrocyte adhesion to IL- 10 stimulated endothelial monolayers. Flow cytometric determinations of reticulocyte counts done on whole and top fraction sickle blood showed a modest enrichment of younger RBC in the top fraction (Table I) . The adhesion of top fraction normal and sickle RBC was studied on both unstimulated and stimulated endothelial monolayers (Fig I ). Sickle erythrocyte adhesion to monolayers was dramatically increased by stimulation of endothelial cells with 50 pg/mL IL-IO. Although the number of adherent cells varied considerably for the different sickle donors, approximately 16-fold increases in adhesion were observed. The level of adhesion did not appear to depend on the time of cytokine exposure. Perfusion of top density fraction normal RBC over unstimulated and IL-10 stimulated endothelial monolayers resulted in only modest increases in adhesion. These increases in adhesion, though small, were statistically significant on the 4- and 24-hour IL- I0 activated endothelium.

Effect of a monoclonal anti-a, antibody on sickle erythrocyte adhesion. Top fraction sickle erythrocytes were incubated with 50 pg/mL HP2/1 (which binds the U, chain of a4PI) for 30 minutes at 37°C before perfusion over I - , 4-, 9-, and 24-hour IL-I0 stimulated endothelial cells (Fig 2). No significant inhibition of adhesion was observed with the anti-U, antibody when sickle erythrocytes were perfused over 1- and 4-hour IL-10 stimulated endothelial cells. However, there was significant inhibition of SSRBC binding to 9- and 24-hour IL- 10 stimulated endothelial cells. Prolonged activation (9 hours) of endothelial cells with TNF-U also resulted in increased adhesion that was inhibited by incubation of sickle RBC with anti-a4 antibody (data not shown). Treat- ment of sickle erythrocytes with an irrelevant, but isotype matched antibody, had no effect on I-, 4-, 9-, and 24-hour stimulated adhesion in each of two experiments.

100

80

c .- 2 60 .e - - E

40 B

20

0

Fig 2.

1 4 9 24 Time of Stimulation of Endothelial Cells (hrs)

Inhibition of adhesion using HP2/1 antibody after IL-1p stimulation of endothelial cells. Sickle reticulocytes were incubated with HP2/1 MoAb (binds the a4 chain of a4pl) at a final concentration of 50 pglmL for 30 minutes at 37°C and then perfused over 1, 4, 9, and 24 hour IL- lp stimulated endothelial cells. P values were calculated with respect to percent inhibition of control: ' P < .01, * *P < .05, and SP > .05. Number of separate determinations shown as n.

Effect qf peptide inhibition cJf integrins on sickle erythrocyte adhesion. Addition of a peptide GRGDSNPKY to sickle RBC for 30 minutes at 37°C before perfusion over 1 -hour IL- I0 stimulated endothelial cells partially blocked enhanced SSlEC adhesion. Experiments performed with 300 pmol/L, 500 pmol/L, and 1,300 pmol/L of the peptide resulted in similar inhibition of adhesion of sickle RBCs to 1- hour IL-ID activated endothelial cells (37.6 5 14.7, 52.3 2 27.5,40.9 2 12.3%, respectively); whereas experiments with 150 pmol/L of the peptide showed no inhibition of adhesion. Addition of 50 pmol/L GRGDSNPKY peptide to endothelial

Table 2. Receptor Expression on IL-lp Stimulated Endothelial Cells

Time of Stimulation (hours)

0 1 4

LM609' 37.89 2 4.82 45.35 t 3.975 40.07 2 4.21 CDw49et 51.17 2 5.77 47.73 2 7.15 54.23 2 3.89 CL2/6S 16.73 t 4.76 25.52 -c 4.585 147.8 2 29.38§

Indirect immunofluorescence flow cytometry of untreated and 1- and 4-hour IL-18 pretreated endothelial cells. Confluent monolayers of HUVECs were pretreated with IL-18 for 1 and 4 hours, and cell surface expression of a& dimer (LM 609), m5 (CDw49e). and E-selectin (CL2/6) was measured by flow cytometric assays, as described in the Materials and Methods. Mean fluorescence as a function of time of stimulation of endothelial cells are shown above. Values represent mean 2 SD. Background levels of fluorescence (in the absence of primary antibody but in the presence of a FITC-labeled irrelevant secondary antibody) were -12. Pvalues were calculated with respect to expression on unstimulated cells.

* n = 6 . t n = 5. S n = 7 . 5 P < .05.




4848 NATARAJAN. UDDEN, AND MclNTlRE

monolayers, on the other hand, decreased adhesion by 45.4% ? 7.2%.

Because vitronectin and fibronectin receptors are the major RGD- dependent integrins present on endothelial cells, flow cytometry experiments were performed to determine if there was upregulation of expression of these molecules after stimulation with IL-1p (Table 2). The expression of a, showed a biphasic trend. One-hour IL- lp treated endothelial cells showed a small, but significant, increase in the level of expression of av as compared with unstimulated monolayers when incubated with LM 609. This expression returned to basal levels within 4 hours. The expression of a5 was also determined using an anti-CDw49e antibody. Both 1- and 4- hour IL-1p stimulated endothelial cells showed no significant increases in the level of expression of a5 as compared with unstimulated monolayers. The fluorescence of the positive control, E-selectin, increased by 1.5- and %fold after 1 and 4 hours of stimulation respectively. The fluorescence of the negative control, CD18, showed no changes at the 1- and 4-hour time points of stimulation of endothelial cells compared with unstimulated ECs. Comparison of individual experiments showed variability in the expression of a, and a5 from different cord pools. For example, changes in expression of a, were observed only in four of six samples. Variable patterns of expression were also observed with a5. In contrast, endothelial cell activation with IL-lp for 1 and 4 hours always resulted in a significant increase in expression of E-selectin.

Effect of a monoclonal anti-a,, antibody on sickle erythrocyte adhesion to I - and 4-hour IL-lp stimulated endothelium. Perfusion of sickle RBC over 1 hour IL-lp activated endothelium that had been incubated with 10 pg/mL anti- CD51 (a,) antibody or 10 pg/mL LM609 resulted in significant inhibition of SSRBC binding (Fig 3) . However, there were a number of flow experiments where no inhibition was observed in the presence of these two antibodies (anti-CDS 1 MoAb; n = 4 and LM 609, n = 2). Flow and flow cytometry experiments performed with endothelial cells from the same cord pools showed a strong positive correlation between increased surface expression of a, and increased inhibition of SSFU3C adhesion to endothelial cells after treatment with LM 609. In monolayers that did not show upregulation of a”, there was no effect of the same antibody on adhesion (Fig 4). Perfusion of sickle reticulocytes over 4- (n = 3 ) and 24 (n = 2)-hour IL-lp stimulated endothelial cells that had been treated with 10 p g / d of LM 609 did not show significant inhibition of adhesion (data not shown). Treatment of stimulated endothelial monolayers with an irrelevant, but isotyped matched antibody, had no effect on sickle cell adhesion in each of two experiments.

Effect of a monoclonal anti-E-selectin antibody on sickle cell adhesion to IL-1 p stimulated endothelium. Maximum upregulation of E-selectin using immunocytometry was observed after 4 hours of IL-10 stimulation. Treatment of endothelial cells with 50 pg/mL. anti E-selectin MoAb, CL2/6, for 30 minutes at 37°C before perfusion of sickle erythrocytes inhibited SSRBC binding by 27.5 & 6.56% (Fig 3). The effect of CL216 inhibition was statistically significant relative

Flow cytometric analysis of endothelial cells.

100

80

E

60 B 5 8 4 0 e k

c w

20

0

1 hour

n=3**

i

anti CD 51 ab LM 609

4 hours 1 24 hours

CL216 E116

MAb Treatment

Fig 3. Inhibition of adhesion of sickle erythrocytes to IL-lfl stimulated endothelid db. Endothelial cells that had been stimulated with IL-1fl for 1 hour were incubated with 10 pglmL anti-CD51 (anti- a.)MoAb or 10 pglmL LM 809 (interferes with ROD binding site of the nvfl3 dimer) and perfused with sickle erythrocytm for 10 minutes. In other experiments, 4- and &hour IL-lp activated endothelial cells were incubated with 50 pg/mL CU/6 (anti-E-sekctin antibody) or E116 (anti-VCAM-I antibody), respectively and then perfused with sickle erythrocytes. Pvalues were calculated with respect to percent inhibition of control: *P< .01, **P c .05. n = number of experiments.

to control. Perfusion of sickle erythrocytes over 1- and 24- hour IL-lP stimulated endothelial cells treated with 10 pgl mL CL2/6 showed no significant inhibition of adhesion. Treatment of 4-hour IL-lp stimulated endothelial cells with an irrelevant, but isotyped matched antibody, also had no effect on SSRBCEC adhesion in each of two experiments.

Effect of an anti-VCAM-1 antibody on 24-hour IL-lp stimulated endothelial cells. Peak expression of VCAM-1 on endothelial cells is induced after 9 to 24 hours of IL-lp stimulation. Perfusion of sickle erythrocytes over 24-hour E - l p activatdendothelial cells treated with 10 pg/mL El / 6 (anti-VCAM-1 antibody) resulted in 51.8% 2 12.2% inhibition (n = 2) (Fig 3). No significant inhibition of adhesion was observed after perfusion of sickle erythrocytes over 1- and 4-hour IL- lp stimulated endothelial cells treated with 10 pg/mL EU6. Treatment of 24-hour IL-lp stimulated endothelial cells with an irrelevant, but isotype matched antibody, also had no effect on SSRBCEC adhesion in each of two experiments.

Morphological changes to surfaces of endothelial cells due to the pelfusion of RBC. RBCs were perfused over I-, 4-, and 24-hour JLlp stimulated endothelial cells for 10 minutes followed by digital image processing to determine morphological changes. The loss of refractility and the absence of typical “cobblestone” morphology of endothelial cells were used as markers of damage. As can be seen in Fig 5, the perfusion of sickle cells over activated endothelial monolayers resulted in easily detectable damage to the endothelium. Approximately 6% to 8% damage was observed on the 4- and 24-hour IL-lp stimulated endothelial cells: whereas little damage (<1%) was observed on the 1-hour stimulated (data not shown) and unstimulated endothelial cells. The perfusion of normal RBCs over both unstimulated





50 50 1 I C

Fig 4. Immunocytometry and flow experiments of 1- and Chour 11-18 stimulated endothelial cells. Endothelial cells from the same cord pools were stimulated with IL-18 for 1 and 4 hours. Data from one cord pool are shown in (A) and IC), while data from a second cord pool are shown in (E) and (D). Cell surface expression of a& was measured by flow cytometry (A and 6). The thin line indicates expression of a. on unstimulated endothelial cells; the bold line indicates expression of a. on 1-hour IL-18 stimulated endothelial cells and the dotted line indicates expression of a, on 4-hour 11-1 stimulated endothelial cells. Kolmogorov-Smirnov statistics were used to determine if detected changes were statistically significant. In (A), the expression of a. significantly (P < .001) increased after 1 hour of IL-1 stimulation and returned t o basal levels at the Qhour time point. In (E), no significant changes ( P z ,051 in the expression of a" were observed. Flow experiments were performed t o determine percent inhibition of adhesion of sickle cells t o stimulated endothelial cells (C and D). Sickle erythrocytes were perfused over 1 hour IL-1p stimulated endothelial cells (control) and over 1 hour IL-lp stimulated endothelial cells incubated with 10 pglmL LM609. In (C), inhibition of SSRBC binding was observed paralleling the increase in a, expression in (A). In ID), no inhibition of erythrocyte-endothelial interactions was observed consistent with the observations in (E).

and stimulated monolayers did not cause significant morphological changes to the surfaces of endothelial cells at any of the time points (Fig 6). The damage value from each individual experiment was plotted with its corresponding adhesion value to determine any correlations between the two parame- ters (Fig 7). A strong correlation between adhesion and damage was observed on the 24-hour IL-I stimulated (n = 0.899).

DISCUSSION

The abnormal adhesion of sickle cells to endothelium is hypothesized to play an important initiating role in the sequence of events leading to vasoocclusion. The molecular mechanisms that actually mediate this binding are not yet well established because adhesion is the end result of a complex process that seems to be dependent on intrinsic charac- teristics of sickle RBCS:'.~' endothelial cell determinants, and the presence of ligand proteins such as thrombospondin (TSP)29.3" and von Willebrand factor (vWF).".~'

Stimulation of the endothelium with cytokines such as IL- Io or TNF-a leads to a time-dependent differential expression of adhesion receptors. Increased adhesion of sickle cells to cytokine stimulated endothelial cells under both static32 and flow conditions2' has been shown previously. All data in these earlier studies were taken after prolonged stimulation of endothelium (>4 hours). The results presented here

provide evidence that short-term cytokine stimulation also promotes SSRBC binding to endothelial cells, and that this binding is mediated via different receptors at the early time points.

RBC expression of adhesion receptors has been shown to be the greatest on the most immature reticulocytes and decreases as they mature into erythrocytes. The receptors that erythrocyte precursors express include the integrins a4Pl, asp I , and a,-&. As these precursors mature and enter the peripheral circulation, most of these integrins disappear with the exception of a4PI .20.2'.33 Recent studies also demonstrate augmented expression of the nonintegrin GPIV (CD36) on reticulocytes." Identification of these two receptors, a4PI and GPIV, on young RBCs has important implications regarding the potential mechanisms of sickle cell adhesion to the endothelium.

The experiments in this study were designed to elucidate the role of the a4Pl integrin in mediating SSRBC binding to IL- ID stimulated endothelial cells. Results indicate that there was little contribution by the a4PI integrin at the early time points, but after 9 and 24 hours of IL-I@ stimulation of endothelial cells, the levels of SSRBC adhesion were similar as were the levels of inhibition by the anti-a4Pl antibody. The incubation of an anti-VCAM-1 antibody with 24-hour IL- Io stimulated endothelial monolayers also resulted in significant inhibition of SSRBC binding. Previous





Fig 5. Photomicrographs of morphological changes to IL-1p stimulated endothelial cells due to the perfusion of erythrocytes. Normal and sickle RBCs were perfused over 4- and 24-hour IL-1p stimulated monolayers for 10 minutes and resulting damage was determined using digital image processing techniques. The monolayer shown above was stimulated with IL-1p for 4 hours. (Top left1 Preperfusion of normal reticulocytes; (bottom left) preperfusion of sickle reticulocytes; (top right) postperfusion of normal reticulocytes, and (bottom right) postperfusion of sickle reticulocytes. Significant damage is only visable in (bottom right) after the perfusion of sickle erythrocytes over stimulated monolayers.

work by Swerlick et alZo had demonstrated that SSRBC adhesion to 9-hour TNF-CY stimulated endothelial cells was a4PI I VCAM-1-mediated and our results with IL-1P are consistent with these observations?'

14

12

4

2

0

Normal

Sickle

0 4 24 Time of Stimulation (hrs)

Fig 6. Quantification of morphological changes on stimulated endothelium. Damage was quantified on unstimulated and 4- and 24- hour IL-1p stimulated endothelium after perfusion of RBCs for 10 minutes. Results are presented as mean & SEM. Pvalues were calculated with respect to damage due to sickle cells on unstimulated monolayers, * P < .01, **P < .05. n = number of experiments.

- y = 6.6404 + 0.019433~ R= 0.15803 20

A a a

5 1 0 0 50 100 150 200 250

Adhesion (celldsquare mm)

- y = 1.8182 + 0.043011~ R= 0.89918 10

a 2

0 0 20 40 60 80 100 120 140 160

Adhesion (cellskquare mm)

Fig 7. Correlation between damage and adhesion. Adhesion values were plotted with corresponding damage values on (A) &hour and (B) 24-hour 11-18 stimulated endothelium.

RGD sequences on extracellular matrix proteins are recognized specifically by several integrins. In this study, the addition of GRGDSNPKY peptide to sickle reticulocyte popula- tions resulted in approximately 50% inhibition of SSRBC binding to endothelium. Given that neither the integrin a4PL nor the nonintegrin GPIV (only known sickle cell receptors) recognizes specific RGD sequences, these findings suggested an interaction between GRGDSNPKY and an integrin on the endothelial cell surface resulting in blocking of adhesion. This conclusion is also consistent with the differential con- centrations of peptide required to inhibit adhesion when incubation was performed with SSRBC versus the endothelial cell monolayer.

Endothelial cells express multiple integrin receptors that promote their attachment to the subendothelial matrix proteins. Among these receptors, the a& seems to be the most abundantly expressed and can bind to many different matrix and plasma proteins. Conforti et alls have shown using surface iodination and immunoprecipitation techniques that this receptor is not exclusively located on the basal side of endo-





thelia1 cell membranes, but may also be exposed on the apical (blood) side. In this report, increased expression of a,P3 was observed after 1 hour of IL-10 stimulation of endothelium and was shown to be responsible for increased SSRBC adhesion. The expression of avP3 returned to near basal levels after 4 hours of IL-lP stimulation and corresponded in time with the absence of inhibition by blocking MoAbs to a,P3 in flow experiments.

At the 4-hour time point, E-selectin expressed on endothelial monolayers contributes to sickle cell binding under conditions of flow. Peak expression of E-selectin occurs 3 to 4 hours postcytokine activation of endothelial cells, and our data agree with this time course of expression. Anti-E-selectin pretreatment of 4-hour IL- 1P activated endothelial monolayers caused a moderate, but significant, reduction in adhesion. The perfusion of neutrophils over 4-hour cytokine stimulated HUVECs treated with the same antibody also showed similar levels of inhibition of adhesion. Anti-E- selectin MoAbs diminished neutrophil adhesion to IL- 1P stimulated endothelial cells by only -30%.9 In the leukocyte-endothelial cell model, selectins are responsible for the initial rolling of white cells along the vascular Recent reviews show that some tumor cells also experience rolling before adhering to vascular endothelium under physiologic flow condition^.^^ In contrast, in the current RBC model, little or no rolling of sickle erythrocytes was observed, although SSRBC-endothelial adhesion was partially mediated by E-selectin.

The counter-receptors on sickle RBCs mediating adhesion to 1- and 4-hour IL-lP treated endothelial cells could not be discerned from these experiments. The a, receptor is promiscuous and can interact with a variety of extracellular matrix proteins such as vWF and TSP that are recognized by the GPIb and GPIIbDIIa and the CD36 receptors, respectively. Based on additional flow experiments, it was con- cluded that the GPIV/CD36 (recognized by the antibody OKM 5), GPIb (recognized by 6D1 and lbl), and GPIIb/ IIIa (recognized by 7E3 and 10E5) were not involved under the conditions tested (data not shown). In a study by Oku- mura et al,” burst forming units (BFU-E) were found to react with MoAbs against the a& complex in flow cytometry experiments. However, Joneckis et alZK did not detect any aIIp3 on reticulocytes using indirect immunofluorescence methods, and this result was supported by experiments that did not show the aIIP3 mRNA using polymerase chain reac- tion analysis with a primer specific for the integrin subunit a11b.38 On the other hand, in two different systems, in vitro using a parallel plate flow chamber4 and ex vivo using rat mesocaec~m,~~ ultra large forms of vWF were potent media- tors of adhesion of sickle cells to endothelial monolayers. This adhesion seemed to involve GPIb and GPIIbLIIa-like receptors on sickle reticulocytes and GPIb and a,P3 on endothelial cells.40 However, the absence of consistent expression of GPIb, GPIIbIIIa, a&, or avP528 receptors for VWF on sickle reticulocytes indicates that other receptors/factors may be involved in this adhesion. Possible candidate receptors that have been suggested include sulfatides and sulfated glycolipids since, in isolated form, they bind vWF and TSp.28.41.42

Endothelial integrity is required for normal functioning of the vascular wall. Alteration of the endothelium has been described as a determinant of vasoocclusive morbidity in sickle cell patients.21.22 Morphological changes to the cytokine activated endothelium that we have reported here, due to perfusion at postcapillary venular shear stress levels of sickle, but not normal RBCs, may be a potentially important finding in the description of vasooclusion in sickle cell disease. Correlation between adhesion and damage, especially after prolonged stimulation of endothelial cells, suggests that damage may be secondary to erythrocyte-endothelial interaction. At this point, it is unknown if these changes are due to the release of chemical substances (for example, free he- moglobin S or oxidant radicals) from sickle RBCs that damage the endothelium, or because of mechanical forces exerted by the weakly adherent sickle RBCs during detachment. De- tachment of adherent sickle cells, however, was not routinely seen from either 4- or 24-hour IL-lP activated endothelium. These observations are potentially significant because morphologic changes consistent with endothelial injury have been described in the arterial and/or arteriolar circulation of spleen and brain of patients with sickle cell anemia. In addition, significant numbers of increased circulating endothelial cells have been detected in blood of patients during pain crises.21 22.43 Our results were obtained after 10-minute perfusion of sickle cells over IL-lP stimulated endothelium. Un- der physiologic conditions, sickle erythrocytes are in repeti- tive contact with endothelium for much longer periods of time. Such long-term erythrocyte-endothelial interactions could lead to much higher levels of damage to endothelium. Consistent with endothelial cell activation, there could also be the release of growth factors from damaged endothelium that may be responsible for smooth muscle cell hyperplasia often seen in sickle patients’ vasculature at autopsy.

In summary, adhesion of sickle cells to I-, 4-, and 24- hour IL-1P activated monolayers is mediated by at least three different endothelial cell receptors at the three different time points. In addition, after 4 or more hours of IL-10 stimulation, contact between sickle erythrocytes and activated monolayers results in significant morphological alteration of the endothelium. These results suggest a model of vasoocclusion that relates adhesion to damage in sickle cell anemia. Damaged endothelial cells may release cytokines and growth factors that not only augment sickle erythrocyte- endothelial interactions, but also lead to further alteration of endothelial surfaces. Repeated damage-adhesion-damage events could conceivably cause vasoocclusion and smooth muscle cell hyperplasia seen in patients with sickle cell anemia.

ACKNOWLEDGMENT

The authors thank Dr David Sears and Dr Joel Moake for their insightful suggestions during the course of this project. The expert technical assistance of Cynthia Patton, Marcella Estrella, and Nancy Turner is gratefully acknowledged.

REFERENCES 1. Barabino CA, McIntire LV, Eskin SG, Sears DA, Udden M:

Endothelial cell interactions with sickle cell, sickle trait, mechani-





cally injured, and normal erythrocytes under controlled flow. Blood 70:152, 1987

2. Grabowski EF: Sickle erythrocytes adhere to endothelial cell monolayers (ECMs) exposed to flowing blood, in Nagel RL (eds): Pathophysiological Aspects of Sickle Cell Vaso-Occlusion. New York, NY, L i s , 1987, p 167

3. Hebbel RP, Yamada 0, Moldow CF, Jacob HS, White JG, Eaton JW: Abnormal adherence of sickle erythrocytes to cultured vascular endothelium. J Clin Invest 65:154, 1980

4. Wick T, Moake JL, Udden MM, Eskin SG, Sears DA, McIntire LV: Unusually large von Willebrand factor multimers increase adhesion of sickle erythrocytes to human endothelial cells under controlled flow. J Clin Invest 80:905, 1987

5. Lipowsky HH, Chien S: Role of leukocyte-endothelium adhesion in affecting recovery from ischemic episodes. Ann NY Acad Sci 565:308, 1989

6. Dustin M1, Rothlein R, Bhan AK, Dinarello CA, Springer TA: Induction by IL-1 and IFN-y: Tissue distribution, biochemistry, and function of a natural adherence molecule (ICAM-I). J Immunol 137:245, 1986

7. McEver RP: Selectins: Novel receptors that mediate leukocyte adhesion during inflammation. Thromb Haemost 65:223, 1991

8. Osbom L, Hession C, Tizard R, Vassallo C, Luhowskyj S, Chi-Rosso G, Lobb R: Direct expression cloning of VCAM- 1 , a cytokine-induced endothelial protein that binds to lymphocytes. Cell 59:1203, 1989

9. Abbassi 0, Kishimoto TK, McIntire LV, Anderson DC, Smith CW: E-selectin supports neutrophil rolling in vitro under conditions of flow. J Clin Invest 92:2719, 1993

10. Lawrence MB, Smith CW, Eskin SG, McIntire LV: Effect of venous shear stress on CDI 8-mediated neurophil adhesion to cultured endothelium. Blood 75:227, 1990

1 I . Albelda SM, Buck CA: Integrins and other cell adhesion molecules. FASEB J 4:2868, 1990

12. Buck CA, Honvitz AF: Integrin, a transmembrane glycoprotein complex mediating cell-substratum adhesion. J Cell Sci Suppl 8:231, 1987

13. Hynes RO: Integrins: Versatility, modulation, and signaling in cell adhesion. Cell 69:l I , 1992

14. Smyth SS, Joneckis CC, Parise LV: Regulation of vascular integrins. Blood 81:2827, 1993

15. Conforti G, Dominguez-Jimenez C, Zanetti A, Gimbrone MA, Cremona 0, Marchisio PC, Dejana E: Human endothelial cells express integrin receptors on the luminal aspect of their membrane. Blood 80:437, 1992

16. Favaloro EJ, Moraitis N, Bradstock K, Koutts J: Co-expression of haematopoietic antigens on vascular endothelial cells: A detailed phenotypic analysis. Br J Haematol 74:385, 1990

17. Lawrence MB, McIntire LV, Eskin SG: Effect of flow on polymorphonuclear leukocyte/endothelial cell adhesion. Blood 70:1284, 1987

18. Lawrence MB, Springer TA: Leukocytes roll on a selectin at physiologic flow rates: Distinction from and prerequisite for adhesion through integrins. Cell 65:859, 1991

19. Beekhuizen H, van Furth R: Monocyte adherence to human vascular endothelium. J Leukoc Biol 54:363, 1993

20. Swerlick RA, Ekman JR, Kumar A, Jeitler M, Wick TM: a4P 1 -1ntegrin expression on sickle reticulocytes: Vascular cell adhesion molecule- 1 dependent binding to endothelium. Blood 82: 189 1, 1993

21. Bridgers WH: Cerebral vascular disease accompanying sickle cell anemia. Am J Pathol 15:353, 1939

22. Merkel KHH, Ginsberg PL, Parker JC, Post MJD: Cerebro- vascular disease in sickle cell anemia: A clinical, pathological and radiological correlation. Stroke 9:45, 1978

23. Klug PP, Kaye N, Jensen WN: Endothelial cell and vascular damage in the sickle disorders. Blood Cells 8:175, 1982

24. Wautier JL, Pintigny D, Maclouf J, Wautier MP. Corvazier E, Caen IP: Release of prostacyclin after erythrocyte adhesion to cultured vascular endothelium. J Lab Clin Med 12: 127, I987

25. Weinstein R, Zhou MA, Bartlett PA, Wenc K: Sickle erythrocytes inhibit human endothelial cell DNA synthesis. Blood 762146, 1990

26. Murphy JR: Influence of temperature and method of centrifugation on the separation of erythrocytes. J Lab Clin Med 82334, 1973

27. Warren JB: Large vessel endothelial isolation, in Warren JB (ed): The Endothelium: An Introduction to Current Research. New York, NY, Wiley, 1990, p 263

28. Joneckis CC, Ackley RL, Orringer EP, Wayner EA, Parise LV: Integrin alpha 4 beta 1 and glycoprotein IV (CD36) are expressed on circulating reticulocytes in sickle cell anemia. Blood 82:3548, 1993

29. Brittain HA, Eckman JR, Swerlick RA, Howard RJ, Wick TM: Thrombospondin from activated platelets promotes sickle erythrocyte adherence to human microvascular endothelium under physiologic flow: A potential role for platelet activation in sickle cell vasoocclusion. Blood 81:2137, 1993

30. Sugihara K, Sugihara T, Mohandas N, Hebbel R P Thrombos- pondin mediates adherence of CD36+ sickle reticulocytes to endothelial cells. Blood 80:2634, 1992

3 1. Kaul DK, Nagel RL, Chen D, Tsai HM: Sickle erythrocyte- endothelial interactions in microcirculation: The role of von Wille- brand factor and implications for vasoocclusion. Blood 81 :2429, 1993

32. Sowemimo-Coker S, Haywood LJ, Meiselman HJ. Francis RB: Interleukin I increases adherence of normal and sickle erythrocytes to endothelial cells. Blood 74:313, 1989 (abstr)

33. Papayannopoulou T, Brice M: Integrin expression profiles during erythroid differentiation. Blood 79: 1686, 1992

34. Lay K, Gaehtgens P, Fennie C, Singer MS, Lasky LA, Rosen SD: Lectin-like cell adhesion molecule 1 mediates leukocyte rolling in mesenteric venules in vivo. Blood 77:2553, 1991

35. Smith CW, Marlin SD, Rothlein R, Toman C, Anderson DC: Cooperative interactions of LFA-1 and Mac-I with ICAM-I in facili- tating adherence and transendothelial migration of human neutrophils in vitro. J Clin Invest 83:2008, 1989

36. Ciavazzi R, Foppolo M, Dossi R, Remuzzi A: Rolling and adhesion of human tumor cells on vascular endothelium under physi- ological flow conditions. J Clin Invest 92:3038, 1993

37. Okumura N, Tsuji K, Nakahata T: Changes in cell surface antigen expressions during proliferation and differentiation of human erythroid progenitors. Blood 80642, 1992

38. Bray PF, Leung CSL, Shuman MA: Human platelets and megakaryocytes contain alternately spliced glycoprotein 1Ib mRNAs. J Biol Chem 265:9587, 1990

39. Kaul DK, Nagel RL, Chen D, Tsai HM: Sickle erythrocyte- endothelial interactions in microcirculation: The role of von Wille- brand factor and implications for vasoocclusion. Blood 81:2429, 1993

40. Wick T, Moake JL, Udden MM, McIntire LV: Unusually large von Willebrand factor multimers preferentially promote young sickle and nonsickle erythrocyte adhesion to endothelial cells. Am J Hematol 42:284, 1993

41. Ginsburg V, Robert? DD: Glycoconjugates and cell adhesion: The adhesive proteins laminin, thrombospondin and von Willebrand factor bind specifically to sulfated glycolipids. Biochemie 70:1651,1988

42. Roberts DD, Rao CN, Liotta LA, Gralnick HR, Ginsburg V: Comparison of the specificities of laminin, thrombospondin, and von Willebrand Factor for binding to sulfated glycolipids. J Biol Chem 261:6872, 1986

43. Sowemimo-Coker S, Meiselman HJ, Francis RB Jr: Increased circulating endothelial cells in sickle cell crisis. Am J Hematol 31:263, 1989




1996 87: 4845-4852

M Natarajan, MM Udden and LV McIntire stimulated endothelial cells under flow in vitroAdhesion of sickle red blood cells and damage to interleukin-1 beta

http://www.bloodjournal.org/content/87/11/4845.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

Copyright 2011 by The American Society of Hematology; all rights reserved.Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American


http://www.bloodjournal.org/content/87/11/4845.full.html

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests

http://www.bloodjournal.org/site/misc/rights.xhtml#reprints

http://www.bloodjournal.org/site/subscriptions/index.xhtml





Adhesion of Sickle Red Blood Cells and Damage to Interleukin-lp ...

Documents