© 1995 American Heart Association, Inc.
Articles |
Immunoselective Targeting of an Anti-Thrombin Agent to the Surface of Cytokine-Activated Vascular Endothelial Cells
From the Vascular Research Division, Departments of Pathology, Brigham and Women's Hospital and Harvard Medical School, Boston, Mass.
Correspondence to Michael A. Gimbrone, Jr, MD, Vascular Research Division, Department of Pathology, Brigham and Women's Hospital, 221 Longwood Ave, LMRC-401, Boston, MA 02115-5817.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract An immunoconjugate was designed to target hirudin, a potent and specific inhibitor of thrombin, to the surface of activated endothelial cells. Hirudin was covalently cross-linked to the monoclonal antibody H18/7 that recognizes the extracellular domain of E-selectin (CD62E), an endothelium-leukocyte adhesion molecule that is expressed only on cytokine-activated endothelium. The hirudin-H18/7 immunoconjugate selectively bound to interleukin-1–activated but not to unactivated cultured human umbilical vein endothelial cells with a temporal profile similar to that of inducible cell-surface procoagulant activity. When bound to activated endothelial cells, the hirudin-H18/7 immunoconjugate significantly inhibited endogenous thrombin activity generated from coincubated human plasma and fibrin clot formation on the monolayer surface. Cellular responses that are mediated via the thrombin receptor, such as increases in cytoskeletal F-actin content, also were significantly downregulated, and monolayers were protected from thrombin-induced disruption by this treatment. The ability to selectively antagonize thrombin-dependent processes at the endothelium-blood interface may provide new insights into complex pathophysiological processes, such as thrombosis, inflammation, and atherogenesis. These studies also demonstrate the general feasibility of selective targeting of therapeutic agents to endothelial cells based on recognition of an activation-dependent surface phenotype.
Key Words: immunotargeting • vascular endothelium • hirudin • E-selectin (CD62E) • thrombin
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Activation of cultured human ECs by inflammatory cytokines, such as IL-1 and tumor necrosis factor, or Gram-negative bacterial endotoxin (lipopolysaccharide) leads to a transient procoagulant state that is partially due to the surface expression of tissue factor.1 2 This procoagulant activity can interact with plasma components to form a prothrombinase complex that cleaves prothrombin to generate thrombin.3 4 Thrombin, in addition to its ability to enzymatically convert fibrinogen to fibrin during clot formation, is itself a pluripotent effector molecule for ECs, acting via a high-affinity membrane receptor.5 Known effects of thrombin on cultured ECs include the modulation of expression of various growth factors, as well as the generation of lipid mediators, such as platelet-activating factor and prostacyclin.6 7 8 Thrombin also induces degranulation of Weibel-Palade bodies, leading to secretion of the platelet-adhesive protein von Willebrand factor and transient surface expression of P-selectin, an EC-leukocyte adhesion molecule.9 10 In an in vitro monolayer model, thrombin induces EC shape change and enhances macromolecular permeability in association with alterations in the EC actin cytoskeleton.11 12 13 In addition to these actions directed at the endothelium, thrombin, acting as a secretogogue and chemoattractant,14 15 also stimulates platelets and leukocytes. In view of these various actions, thrombin generation at the vessel wall–blood interface in vivo would be predicted to be an important component of the pathophysiology of "response-to-injury" processes, such as acute and chronic inflammation, wound healing, arterial intimal hyperplasia, and atherogenesis.
To further investigate the role of thrombin in vascular responses to injury, we have attempted to modulate the actions of thrombin at the blood–vessel wall interface by selectively targeting hirudin, the natural thrombin inhibitor, to the surface of activated ECs. For this purpose, we constructed an immunoconjugate composed of hirudin covalently cross-linked to the E-selectin–specific monoclonal antibody H18/7.16 E-selectin (CD62E) was chosen as the targeted activation antigen because it is typically not expressed in unactivated ECs; however, on activation by cytokines and bacterial endotoxin, E-selectin is abundantly expressed on the cell surface, with a temporal pattern of induction similar to that of tissue factor.17 18 Both E-selectin and tissue factor are products of activation-dependent genes, with peak cell-surface expression in cultured human ECs typically occurring at 4 to 6 hours, followed by a rapid return to near-background levels by 24 hours. Hirudin, a potent, specific, and direct-acting inhibitor of thrombin, is an approximately 10 000-molecular-weight protein produced in the salivary gland of the leech Hirudo medicinalis. Hirudin forms a high-affinity complex with thrombin by binding to its anion-binding exosite, thereby blocking the ability of thrombin to cleave fibrinogen and activate its receptor.19 20 We reasoned that selective targeting of this inhibitor should allow us to evaluate the contribution of thrombin to the pathophysiological processes that occur in the vicinity of the activated EC surface.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Cells and Reagents
Human ECs were isolated from segments of pooled normal-term umbilical cord veins and cultured in Medium 199 (BioWhittaker) supplemented with 20% fetal bovine serum (Hyclone), endothelial mitogen (50 µg/mL; Biomedical Technologies, Inc), penicillin (100 U/mL), and streptomycin (100 µg/mL, BioWhittaker), as described previously.21 For experimental use, HUVECs were plated at passage 2 in culture wells coated with 0.1% gelatin (DIFCO). The murine monoclonal antibody H18/7 (IgG2a) recognizes a functional epitope on human E-selectin,16 22 whereas Rb1/9 (IgG1) recognizes rabbit vascular adhesion cell molecule–1 but not its human homologue.23 Recombinant desulfatohirudin was obtained from CIBA-GEIGY. The cross-linker SPDP was purchased from Pierce, and the chromogenic substrate Spectrozyme TH was purchased from American Diagnostica Inc. Purified human thrombin was purchased from Haematologic Technologies, Inc. Purified recombinant human interleukin-1ß (rhIL-1ß) was a gift from Biogen. Rhodamine/phalloidin was purchased from Molecular Probes.
Whole blood was drawn from healthy human volunteers into CCD solution (1:9, vol/vol, of 100 mmol/L sodium citrate and 136 mmol/L dextrose, pH 6.5). Plasma was isolated after centrifugation (700g for 5 minutes) and then depleted of platelets by a 2-minute microcentrifuge spin (16 000g). Dulbecco's phosphate-buffered saline (DPBS), which contains 1.1 mmol/L Ca2+ and 0.5 mmol/L Mg2+, was purchased from BioWhittaker.
Cross-linking of Monoclonal Antibody H18/7 to Hirudin
Recombinant hirudin was cross-linked to a purified
immunoglobulin of the monoclonal anti–E-selectin antibody H18/7
(IgG2a) by using the heterobifunctional cross-linker SPDP
[3-(2-pyridyldithio)propionate N-hydroxysuccinimide
ester]. Adapting the manufacturer's suggested protocol, we added
2-pyridyl disulfide groups to the hirudin by incubating 10 mg hirudin
dissolved in 1.0 mL borate-buffered saline (pH 9.0) with 20 µL SPDP
(20 mmol/L in DMSO) and to the purified immunoglobulin by incubating 25
mg IgG in 1.5 mL of borate buffer with 200 µL of 20 mmol/L SPDP. The
hirudin-bound PDP disulfide bonds were specifically reduced with 12 mg
DTT to expose the thiol groups. The thiolated hirudin was incubated
overnight with the PDP-derivatized IgG in DPBS to allow cross-linking
to occur. Free hirudin was then removed by gel filtration in DPBS on a
G-75 Sephadex sizing column, and the immunoconjugate was concentrated
by centrifugation in a Centricon-30 microconcentrator
(Amicon). Immunoreactivity of the cross-linked preparation was
evaluated by measuring the binding to unactivated or
IL-1–activated HUVEC monolayers by standard techniques in a
quantitative fluorescence immunobinding assay and compared with
uncoupled H18/7 monoclonal antibody. The ability of the hirudin
immunoconjugate to inhibit the activity of a known concentration of
thrombin was also measured in a chromogenic assay and
compared with a standard inhibition curve for free hirudin (see below).
In preliminary experiments, the anti-thrombin activity of hirudin,
before and after SPDP addition and DTT treatment, was compared;
although there was a slight shift in the inhibition curve, treated
hirudin retained at least 80% of its original activity (data not
shown).
Assay for Thrombin Activity
Thrombin activity was quantitated in a microtiter plate with an
amidolytic assay and a thrombin-specific chromogenic
substrate Spectrozyme TH. Cleavage of Spectrozyme TH releases
p-nitroaniline, which can be quantitated by measuring an
increase in absorbance at 405 nm. A source of thrombin, either diluted
fresh human plasma incubated with activated ECs
(endogenous source) or purified human thrombin (exogenous
source) in DPBS, was incubated with Spectrozyme TH substrate for 15 to
30 minutes at 37°C in a water bath, and the assay was stopped by
addition of glacial acetic acid. Thrombin enzymatic activity, which was
linear over this time range, was determined by spectrophotometrically
measuring the amount of cleaved substrate on a microplate reader
(Biorad model 3550). In certain assays, various concentrations of
recombinant hirudin were added to the incubation mixture, and the
resulting inhibition of thrombin activity was measured.
Generation of Thrombin by Activated EC
Cultures
HUVEC monolayers cultured in microtiter wells were
activated by incubation for 4 hours with rhIL-1ß (10 U/mL).
Monolayers were washed once with DPBS and then incubated at 37°C in
100 µL DPBS with a 2x final concentration of human platelet-poor
plasma and 100 µL of the chromogenic substrate. The
amount of cleaved substrate was determined by measuring the
OD405 in a microplate reader. Thrombin activity could be
detected within 15 minutes and was linear over 30 minutes. At least 1%
plasma was required to generate a detectable level of thrombin
activity. At 10% plasma, thrombin activity was also generated at
detectable levels with unactivated HUVEC monolayers. Therefore,
1% plasma was routinely used for this assay. In assays that tested the
effect of the immunoconjugate, hirudin-H18/7 (100 µg/mL in RPMI 1640
with 1% fetal bovine serum) was incubated with an activated EC
monolayer on ice for 30 minutes, and the unbound hirudin
immunoconjugate was removed by washing prior to the addition of plasma
and the chromogenic substrate.
Plasma Clotting Assay on Activated EC
Cultures
Clot formation on the surface of intact, IL-1–activated
HUVECs was measured in a plasma recalcification assay that uses a
microplate reader to detect increases in opacity over time as plasma
fibrinogen is converted to fibrin.24 Confluent HUVEC
monolayers in microtiter wells were activated with rhIL-1ß
for 4 hours at 37°C, incubated with either the control antibody or
the hirudin-H18/7 conjugate (1 mg/mL) for 30 minutes on ice, and washed
three times with Tris buffer (0.02 mol/L Tris in 0.15 mol/L NaCl, pH
7.2) prior to assay. Citrated platelet-poor plasma was added in
combination with Tris buffer to yield final assay concentrations
ranging from 1% to 10% whole plasma. In some cases, various
concentrations of hirudin were also added. Clotting was initiated by
recalcification with CaCl2 (final concentration, 12.5
mmol/L), the plates were incubated at room temperature, and at selected
times the OD405 was measured.
Quantitation of F-actin in EC Monolayers
The F-actin content of HUVECs was quantitated essentially as
described previously.13 In brief, after activation with
rhIL-1ß (10 U/mL for 4 hours), replicate confluent monolayers in
microtiter wells were washed, incubated on ice for 30 minutes with the
hirudin immunoconjugate, washed four times with DPBS containing 0.5%
(wt/vol) BSA, and then incubated for 15 minutes at 37°C with either
diluted human plasma (as an endogenous thrombin source) or
purified human thrombin in DPBS-BSA buffer. The monolayers were then
fixed with 3.7% (vol/vol) formaldehyde in DPBS for 5 minutes at room
temperature, washed three times, permeabilized with a
buffer containing 1.4% (vol/vol) formaldehyde in DPBS and 0.1%
(vol/vol) NP-40 for 90 seconds at room temperature, and stained with
rhodamine/phalloidin (1:10, vol/vol, in DPBS) for 40 minutes at
room temperature. After three additional washes, 100 µL methanol was
added to each well, and F-actin content was measured with a microplate
fluorescence reader (Pandex) with an excitation wavelength of
545 nm and an emission wavelength of 575 nm.
Microscopic Evaluation of Monolayer Integrity
In certain experiments, monolayer integrity was evaluated
morphologically by phase-contrast and
immunofluorescence microscopy. For this purpose,
after plasma coincubation or thrombin treatment with and without
hirudin-immunoconjugate pretreatment, HUVECs were grown to confluence
on gelatin-coated (0.1%, wt/vol) Permanox Lab-Tek chamber slides
(Nunc) and fixed with 3.7% (vol/vol) formaldehyde in DPBS for 40
minutes at room temperature. To visualize the organization of the
cytoskeleton, staining of permeabilized monolayers with
rhodamine/phalloidin was carried out as described above.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Hirudin Immunoconjugate Exhibits Anti-Thrombin Activity
After the cross-linking procedure was performed, free hirudin was separated from the cross-linked compound (and any unreacted immunoglobulin) by gel filtration on a G-75 Sephadex sizing column (1.5x40 cm, Fig 1A
). Selected fractions were tested for
hirudin (anti-thrombin) activity by using the Spectrozyme
chromogenic assay. The hirudin activity of test samples was
quantified by comparison with known amounts of uncoupled hirudin that
are required to neutralize a human thrombin standard. The bulk of
hirudin activity was found in the low-molecular-weight fractions,
corresponding to free hirudin. Significant hirudin activity, however,
was also recovered in the IgG-containing void volume. When this peak
was rechromatographed, hirudin activity again remained in
the void volume. No activity was detected in fraction 35 or 40, where
dimers or trimers of hirudin would be expected to elute. To determine
whether hirudin oligomers could be produced during the cross-linking
procedure, DTT-treated, SPDP-derivatized hirudin was incubated
overnight without H18/7 to allow formation of hirudin complexes. When
this material was chromatographed, all of the anti-thrombin
activity was recovered in the 10 000 included peak, whereas the void
volume, where the immunoconjugates typically eluted, contained no
detectable anti-thrombin activity. Taken together, these data indicate
that after monoclonal antibody cross-linking, high-molecular-weight
hirudin did not contain oligomerized, unconjugated hirudin.
|
)
with the Spectrozyme chromogenic assay. Column
precalibration showed elution of unconjugated IgG in fractions 28-32
(void volume) and free hirudin (standard) in a broad peak spanning
fractions 45-62. Data represent mean values from 1 of 8 similar
preparations. B, Electrophoresis of void-volume activity peak. Pooled
fractions 28-32 were analyzed by SDS-PAGE (4-15%) under
reducing conditions. Lane 1, unconjugated H18/7 IgG; lane 2,
immunoconjugate. Note presence of released hirudin (arrow) in reduced
immunoconjugate sample.
When the immunoconjugate-containing peak was evaluated by 4% to 15%
SDS-PAGE under nonreducing conditions, only one band was seen at
approximately 150 kD. Under reducing conditions, three bands appeared,
one at 
50 kD (IgG heavy chain), the second at 25 kD (IgG light
chain), and the third at 10 kD (free hirudin, Fig 1B
). Eight
separate H18/7-hirudin conjugates were prepared, the anti-thrombin
activities of which ranged from 27 to 115 µg per milligram protein,
corresponding to an average specific activity of roughly 1 molecule of
hirudin per 1 molecule of IgG. At least two different preparations of
immunoconjugate were utilized in each experimental study described
below.
Hirudin Immunoconjugate Retains Selective
Immunoreactivity
The cross-linking procedure did not affect the ability of the
E-selectin antibody to interact with the surface of
cytokine-activated HUVECs, as demonstrated in a
quantitative immunobinding assay (Fig 2A). When
preparations of H18/7-hirudin immunoconjugate were compared with
uncoupled H18/7 IgG, equivalent dose-dependent, saturable binding to
IL-1–activated HUVEC monolayers was observed, with maximum
binding achieved at added antibody levels above 1.0 µg/mL. Neither
the immunoconjugate nor uncoupled IgG bound detectably to
unactivated HUVECs. Binding of both compounds was detected
within 2 hours of IL-1 activation, with maximum levels seen by 4 hours
(Fig 2B), consistent with the previously reported temporal
pattern of inducible E-selectin surface expression in this cell
system.16
|
) or hirudin-H18/7 immunoconjugate (
)
were incubated with activated (10 U/mL recombinant human [rh]
IL-1ß for 4 hours) HUVEC monolayers for 30 minutes on ice; unbound
antibody was then washed out and FITC-conjugated goat anti-mouse IgG
was added. After incubation on ice for 30 minutes, wells were washed
and fluorescence quantitated in a fluorescence
microplate reader. Immunobinding data are presented as mean±SD
percent maximum fluorescence from 5 separate experiments with
triplicate determinations in each. Triangles (open, H18/7; closed,
hirudin-H18/7) illustrate lack of binding to unactivated HUVECs
at the highest concentration of reagent tested (100 µg/mL). B,
Temporal pattern of E-selectin expression on activated HUVECs
is similar to that of inducible thrombin-generating activity. HUVECs
activated with rhIL-1ß (10 U/mL) for different times were
tested for binding of H18/7 IgG (
Hirudin Immunoconjugate Inhibits Thrombin Generation From
Plasma
Cytokine and/or endotoxin activation of HUVECs results in
the induction and surface expression of tissue factor activity and
renders these cultured EC monolayers prothrombotic.1 2 18
When HUVEC monolayers were treated with rhIL-1ß, the capability to
generate thrombin activity from added human plasma developed in a
temporal pattern similar to that of inducible H18/7 immunobinding
(E-selectin cell-surface expression, Fig 2B). By comparison of
dose-response curves for free hirudin inhibition of an exogenous
thrombin standard, the amount of endogenous thrombin
activity generated from 1% human platelet-poor plasma in this cell
system was on the order of 1 U/mL (Fig 3A). In three
separate experiments with two different preparations of hirudin
immunoconjugate, significant (79±22%, P<.05) inhibition
of plasma-generated thrombin activity was observed (Fig 3B). In
contrast, treatment with uncoupled H18/7 IgG did not have any
inhibitory effect. When free hirudin was preincubated in
parallel with activated HUVEC monolayers and then washed out
before the addition of plasma, no residual inhibition of
endogenously generated thrombin activity was seen (data not
shown).
|
Hirudin Immunoconjugate Delays Plasma Clotting Time
Clot formation was detected by a plasma recalcification assay that
was performed on HUVEC monolayers in microtiter wells. In this in vitro
model system, IL-1–activated HUVEC monolayers serve as a
source of cell-surface–available tissue factor activity, which
initiates the coagulation cascade in the coincubated plasma, thus
generating the thrombin that then cleaves fibrinogen to form fibrin. In
preliminary experiments, when 10% human plasma was added to the
activated EC monolayers and recalcified, a change in optical
density due to fibrin generation was detected at 10 minutes and was
maximal at 30 minutes. Within this time interval, free (uncoupled)
hirudin dose-dependently inhibited plasma clot formation over a broad
concentration range, with virtually complete inhibition observed at 1.0
µg/mL added hirudin. In three separate experiments, pretreatment with
the hirudin-H18/7 immunoconjugate resulted in significant
(P<.005) inhibition of clot formation (ranging from 22% to
49%) at all time points examined from 5 to 30 minutes (Fig 4). The level of inhibition observed with the
H18/7-hirudin immunoconjugate (120 mg hirudin activity per milligram)
was comparable to that obtained in parallel assays with 0.1 µg/mL
free hirudin. When preparations of hirudin cross-linked to the
nonbinding murine monoclonal antibody Rb1/9 (anti-rabbit vascular cell
adhesion molecule–1, 225 µg hirudin activity per milligram) were
substituted for the H18/7 immunoconjugate, no decrease in clot
formation was observed (data not shown). Over time, inhibition of clot
formation by either free hirudin or the cell-surface–bound hirudin
immunoconjugate was overcome, presumably because of the redundant and
cascading nature of the coagulation process. At later time points, clot
formation was also detectable on unactivated EC monolayers (Fig 4).
|
), or no antibody
(
) and the time course of
clot formation on unactivated endothelium
(
Hirudin Immunoconjugate Inhibits Thrombin-Induced Increases in
F-actin Content and Promotes Monolayer Integrity
In in vitro model systems,11 12 thrombin has been
reported to induce changes in EC shape and alterations in EC monolayer
permeability to macromolecules, presumably via activation of specific
thrombin receptors. Both of these effects are accompanied by dramatic
changes in the organization of the cytoskeleton, in particular, actin
microfilaments.13 In preliminary experiments, the F-actin
content of both unactivated and IL-1–activated HUVEC
monolayers, as measured by rhodamine/phalloidin binding, was detectably
increased by the addition of as little as 0.03 U/mL purified human
thrombin. The addition of human plasma (2%, recalcified) to HUVEC
monolayers also caused an increase in F-actin content (comparable to
the maximal stimulation obtained with 0.1 U/mL exogenous thrombin; Fig 5), but only if the monolayers were cytokine
activated (10 U/mL IL-1 for 4 hours) to allow
endogenous thrombin generation to occur (data not shown).
In three separate experiments, thrombin-induced increases in HUVEC
F-actin content were significantly antagonized by the addition of
uncoupled hirudin (0.1 µg/mL, 90±9% inhibition, exogenous thrombin;
74±4% inhibition, plasma-generated thrombin activity) or the
specifically bound hirudin-H18/7 immunoconjugate (60±6% and 68±3%
inhibition, respectively) but not the nonbinding (control)
hirudin-Rb1/9 immunoconjugate (Fig 5).
|
Microscopic examination of confluent IL-1–activated HUVEC
monolayers, after a 15-minute incubation with 2% recalcified human
plasma, revealed cell retraction and disruption of monolayer integrity
(Fig 6A and 6B). This response was inhibited by the
addition of free hirudin (data not shown), suggesting that this
resulted from the generation of endogenous thrombin in the
vicinity of the procoagulant-activated EC surface. Pretreatment
with hirudin immunoconjugate after IL-1 activation but prior to
incubation with plasma essentially ablated this thrombin-induced effect
(Fig 6C).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The vascular endothelial lining is a biologically active interface at which multiple effector molecules involved in inflammation, thrombosis, and atherogenesis interact as components of various pathophysiological balances.25 26 Activation of the endothelium by cytokines and bacterial products, such as Gram-negative endotoxins, typically results in a shift of the set point of these balances to a proinflammatory, prothrombotic state.26 Thrombin, via its actions as both a humoral protease and cellular agonist, has emerged as a key mediator in many of these activation-dependent processes.5 27 Therefore, factors that determine the effective concentration of thrombin in the vicinity of the luminal surface of the endothelium may have important pathophysiological relevance. Such factors would include thrombin generation by the membrane-associated prothrombinases of activated ECs, leukocytes, and platelets; hemodynamic conditions that influence boundary-layer diffusion and convection; and neutralizing mechanisms, such as thrombomodulin and heparin-like glycosaminoglycans, that act to reduce thrombin activity at the endothelium-blood interface. Given the localized nature of these activation-dependent processes and their potential for sequestration in so-called "black holes"28 at points of cell-cell or cell-matrix contact, it would be desirable to design an anti-thrombin agent that could be selectively targeted to such sites. In addition, because multiple cell types in circulating blood (eg, leukocytes and platelets) and within the vessel wall (eg, endothelium and smooth muscle) express receptors for thrombin,5 29 cell-selective targeting could also provide a therapeutic selectivity not currently attainable.
E-selectin, an endothelium-specific, cytokine-inducible leukocyte adhesion molecule, is rapidly upregulated on the surface of cultured human vascular ECs in a temporal profile that coincides with the expression of tissue factor procoagulant activity.1 16 18 Coincubation of activated EC monolayers with human plasma thus would be predicted to result in the generation of thrombin activity at the EC surface in close proximity to the extracellular domains of E-selectin molecules. In our current studies, we have attempted to exploit these properties of the activated endothelial phenotype in the design of an immunoselectively targeted agent to antagonize thrombin-dependent processes in a cell-specific and context-defined fashion.
Murine monoclonal antibody H18/7 is an adhesion-blocking,
E-selectin–specific reagent that does not interact with
unactivated human ECs. Hirudin, a potent, naturally occurring
anticoagulant, forms a tight, highly stable, noncovalent complex with
thrombin that blocks both its proteolytic cleavage of fibrinogen (and
other substrates) and its ability to activate cellular
receptors. Covalent cross-linking of these two components, which can be
achieved by using well-established methodologies,30
yielded an immunoconjugate (Fig 1A and 1B) that retained appropriate
immunoselectivity and anti-thrombin activity. This immunoconjugate
preparation, like the unconjugated monoclonal antibody, exhibited
dose-dependent, saturable binding to activated HUVEC monolayers
but failed to bind to unactivated HUVEC monolayers, thus
satisfying the requirement of activation-dependent
endothelial targeting (Fig 2A and 2B).
Although individual hirudin-H18/7 immunoconjugate preparations were not
purified to homogeneity by the removal of unconjugated immunoglobulin
(which may account partially for variations in their anti-thrombin
activity), all were effective in neutralizing both exogenously added
thrombin and endogenously generated thrombin in the
proximity of the activated EC surface (Fig 3B). On the basis of
an average coupling efficiency of 1 hirudin molecule per IgG molecule
(activity measurements) and an estimated density of
1x106 E-selectin molecules per
cytokine-activated HUVEC (M.I. Cybulsky, unpublished
data, 1995), the calculated density of targeted hirudin molecules on
the surface of a confluent monolayer (1 to 2x104 cells per
microtiter well) would be on the order of 10 to 20x10-15
mol per test well. Based on further comparisons with free hirudin and
exogenous thrombin standard, the estimated concentration of thrombin
generated by incubation of the activated
endothelial monolayer with 1% human platelet-poor
plasma is on the order of 1 U/mL (Fig 3A). Thus, in multiple
experiments with different hirudin immunoconjugate preparations, these
relatively small amounts of surface-targeted hirudin exhibited
anti-thrombin activities comparable to those of fluid-phase
concentrations of free hirudin in the range of 10 to 30 nmol/L (Fig 3A and 3B). The lack of effect of unconjugated H18/7 immunoglobulin or
residual free hirudin (after sham addition and wash-out) confirmed that
selective targeting of the anti-thrombin immunoconjugate to the
endothelial surface was in fact responsible for this
observed inhibition (Fig 3B).
Studies of the kinetics of clotting in recalcified human plasma on
activated EC surfaces provided further evidence of the ability
of immunotargeted hirudin to antagonize thrombin-dependent interface
reactions. Again, femtomole amounts of surface-targeted hirudin
immunoconjugate were equivalent to 10 nmol/L fluid-phase concentrations
of free hirudin in inhibiting the generation of fibrin in this in vitro
model system (Fig 4). This result is all the more striking, given the
enzymatic amplification inherent in the coagulation cascade from the
level of tissue factor activation to the end point of fibrin generation
and the absence of any fluid movement that might enhance
thrombin-inhibitor complex formation at the monolayer
surface. The general strategy of immunotargeting the
activators of the fibrinolytic system to the fibrin and
platelet components of a developing thrombus to limit its clinical
impact has been extensively explored in various experimental systems in
vitro and in vivo.30 31 Hirudin immunoconjugates that bind
selectively to fibrin after fibrinopeptide cleavage
have been shown to significantly reduce fibrin deposition on the
surface of experimental clots.32 A preliminary report of
viral vector-mediated transduction of cultured bovine aortic ECs with a
hybrid gene that encodes a secretable form of hirudin has also recently
appeared.33 To our knowledge, however, our data provide
the first example of activation-dependent immunotargeting of an
anti-thrombotic agent to the vascular endothelium.
Immunotargeting of hirudin was also effective in antagonizing certain
intracellular effects of thrombin in cultured EC monolayers (Figs 5 and 6), presumably mediated via cell-surface receptors.5 The
increases in F-actin content that accompany thrombin-induced
reorganization of the endothelial
cytoskeleton,13 mediated by either exogenous thrombin or
an endogenous (plasma) thrombin-generating source (Fig 5),
were significantly inhibited by hirudin-H18/7 in an immunospecific
fashion and to a degree comparable to that of free hirudin.
Furthermore, the striking cell retraction and monolayer disruption
induced by plasma clotting and thrombin generation on the
activated endothelial monolayer were also
effectively antagonized by the hirudin-H18/7 immunoconjugate (Fig 6).
Activation of thrombin receptors in vascular ECs typically results in a
transient rise in intracellular Ca2+
levels11 ; in preliminary experiments, hirudin-H18/7
immunoconjugate also significantly inhibited this thrombin-induced
response, as measured in fura 2–loaded HUVECs (F.W. Luscinskas, PhD,
and J.-M. Kiely, MS, unpublished observations, 1995). The ability to
selectively inhibit these thrombin-mediated cellular events may provide
pathogenetic insights into the endothelial dysfunction
that is associated with acute inflammatory and thrombotic states in
which thrombin generation at the vessel wall–blood interface
occurs.2 6 7 8 9 26 29
In conclusion, we have demonstrated that the immunoselective targeting of the antithrombin agent hirudin to the surface of human ECs in an activation-dependent fashion can significantly reduce the thrombogenicity of this interface, as well as modify thrombin-mediated cellular events. This experimental approach, if applied in appropriate in vitro and in vivo models, may provide a useful tool for probing the multiple roles of thrombin as a pleiotropic mediator in vascular pathophysiology. In addition, the principle of activation-dependent immunotargeting to the vascular endothelium in vivo conceivably could be extended to other potential therapeutic interventions (eg, anti-oxidant, anti-adhesive, anti-complement, and immunosuppressive agents) in various disease settings (eg, acute and chronic inflammation, vasculitis, sepsis, transplant rejection, and atherosclerosis) in which endothelial activation plays a role.23 26 34 35 36
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This research was supported by grants from the National Institutes of Health (P01-HL36028 and P01-HL48743 to Drs Gimbrone and Cybulsky). M.A. Gimbrone, Jr, is a recipient of an unrestricted grant for cardiovascular research from the Bristol-Myers Squibb Research Institute. The authors wish to thank Dr Joseph Loscalzo for helpful discussions and for providing recombinant hirudin and to acknowledge the expert assistance of William Atkinson and Kay Case in cell culturing and the advice of Dr Masayuki Yoshida for the F-actin staining. Monoclonal antibody H18/7 has been deposited with the American Type Culture Collection (ATCC HB11684).
Received November 30, 1994; accepted May 26, 1995.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Bevilacqua M, Pober J, Majeau G, Cotran R, Gimbrone MA Jr. Interleukin-1 (IL-1) induces biosynthesis and cell surface expression of procoagulant activity in human vascular endothelial cells. J Exp Med. 1984;160:618-623.
2.
Parry GCN, Mackman N. Transcriptional
regulation of tissue factor expression in human
endothelial cells. Arterioscler Thromb
Vasc Biol. 1995;15:612-621.
3.
Nemerson Y. Tissue factor and
hemostasis. Blood. 1985;71:1-8.
4. Morrissey JH, Fakhrai H, Edgington TS. Molecular cloning of the cDNA for tissue factor, the cellular receptor for the initiation of the coagulation protease cascade. Cell. 1987;50:129-135. [Medline] [Order article via Infotrieve]
5. Coughlin S, Vu T-KH, Hung D, Wheaton V. Characterization of a functional thrombin receptor. J Clin Invest. 1992;78:587-591.
6.
Daniel TO, Gibbs VC, Milfay DF, Garavoy MR, Williams
LT. Thrombin stimulates c-sis gene expression in microvascular
endothelial cells. J Biol
Chem. 1986;261:9579-9582.
7. Zimmerman GA, McIntyre TM, Prescott SM. Thrombin stimulates neutrophil adherence by an endothelial-dependent mechanism. Ann N Y Acad Sci. 1986;485:349-368. [Medline] [Order article via Infotrieve]
8.
Jaffe E, Grukich J, Weksler BB, Hampel G, Wantanabe K.
Correlation between thrombin-induced prostacyclin
production and inositol triphosphate and cytosolic free calcium
levels in cultured human endothelial cells.
J Biol Chem. 1987;262:8557-8565.
9. Sporn L, Marder V, Wagner D. Inducible secretion of large, biological potent von Willebrand factor multimers. Cell. 1986;46:185-190. [Medline] [Order article via Infotrieve]
10. Collins P, Macey M, Cahill M, Newland A. Von Willebrand factor release and P-selectin expression is stimulated by thrombin and trypsin but not IL-1 in cultured human endothelial cells. Thromb Haemost. 1993;70:346-350. [Medline] [Order article via Infotrieve]
11. Sago H, Iinuma K. Cell shape change and cytosolic Ca2+ in human umbilical-vein endothelial cells stimulated with thrombin. Thromb Haemost. 1992;67:331-334. [Medline] [Order article via Infotrieve]
12. Garcia J, Siflinger-Birnboirn A, Bizios R, Del Vecchio P, Fenton J, Malik A. Thrombin induced increase in albumin permeability across the endothelium. J Cell Physiol. 1986;128:96-104. [Medline] [Order article via Infotrieve]
13. Thurston G, Turner D. Thrombin induced increase of F-actin in human umbilical vein endothelial cells. Microvasc Res. 1994;47:1-20. [Medline] [Order article via Infotrieve]
14. Davey MG, Luscher EF. Actions of thrombin and other coagulant and proteolytic enzymes on blood platelets. Nature (Lond). 1967;216:857-858. [Medline] [Order article via Infotrieve]
15.
Bar-Shavit R, Kahn A, Wilner GD, Fenton JW III.
Monocyte chemotaxis: stimulation by specific exosite region in
thrombin. Science. 1983;220:728-731.
16.
Bevilacqua MP, Pober JS, Mendrick DL, Cotran RS,
Gimbrone MA Jr. Identification of an inducible
endothelial-leukocyte adhesion molecule.
Proc Natl Acad Sci U S A. 1987;84:9238-9242.
17. Pober JS, Bevilacqua MP, Mendrick DL, Lapierre LA, Fiers W, Gimbrone MA. Two distinct monokines, interleukin 1 and tumor necrosis factor, each independently induce biosynthesis and transient expression of the same antigen on the surface of cultured human vascular endothelial cells. J Immunol. 1986;136:1680-1687. [Abstract]
18. Bevilacqua MP, Gimbrone MA Jr. Inducible endothelial functions in inflammation and coagulation. Semin Thromb Hemost. 1987;13:425-433. [Medline] [Order article via Infotrieve]
19. Stone SR, Maraganore JM. Hirudin interactions with thrombin. In: Berliner L, ed. Thrombin: Structure/Function. New York, NY: Plenum Press; 1992:92-104.
20.
Rydel TJ, Ravichandran KG, Tulinsky A, Bode R, Huber C,
Roitsch C, Fenton JW III. The structure of a complex of
recombinant hirudin and human alpha-thrombin.
Science. 1990;249:277-280.
21. Luscinskas FW, Cybulsky M, Kiely JM, Peckins CS, Davis VM, Gimbrone MA Jr. Cytokine activated human endothelial monolayers support enhanced neutrophil transmigration via a mechanism involving both endothelial-leukocyte adhesion molecule-1 and intercellular adhesion molecule-1. J Immunol. 1991;146:1617-1625. [Abstract]
22.
Bevilacqua MP, Stengelin S, Gimbrone MA Jr, Seed B.
Endothelial-leukocyte adhesion molecule 1: an
inducible receptor for neutrophils related to complement regulatory
proteins and lectins. Science. 1989;243:1160-1165.
23.
Cybulsky MI, Gimbrone MA Jr.
Endothelial expression of a mononuclear
leukocyte adhesion molecule during atherogenesis.
Science. 1991;251:788-791.
24. Chang P, Aronson D. A microtiter plate reader assay for factor VIII. Thromb Res. 1992;66:599-602. [Medline] [Order article via Infotrieve]
25. Gimbrone MA Jr. Vascular endothelium: nature's blood-compatible container. Ann N Y Acad Sci. 1987;516:5-11. [Medline] [Order article via Infotrieve]
26. Gimbrone MA Jr. Vascular endothelium in health and disease. In: Haber E, ed. Molecular Cardiovascular Medicine. New York, NY: Scientific American Medicine; 1995. In press.
27. Harker LA. Strategies for inhibiting the effects of thrombin. Blood Coagulation Fibrinol. 1994;5:S47-S58.
28. Huang AJ, Silverstein SC. Mechanisms of neutrophil migration across endothelium. In: Simionescu N, Simionescu M, eds. Endothelial Cell Dysfunctions. New York, NY: Plenum Press; 1992:201-231.
29. Nelken NA, Soifer SJ, O'Keefe J, Vu T-KH, Charo IF, Coughlin SR. Thrombin receptor expression in normal and atherosclerotic human arteries. J Clin Invest. 1992;90:1614-1621.
30. Haber E, Quertermous T, Matseuda GR, Runge MS. Innovative approaches to plasminogen activator therapy. Science. 1989;243:52-54.
31.
Dewerchin M, Lijnen HR, Stassen JM, DeCock F,
Quertermous T, Ginsberg MH, Plow EF, Collen D. Effect of
chemical conjugation of recombinant single-chain urokinase-type
plasminogen activator with monoclonal
antiplatelet antibodies on platelet aggregation and on plasma
clot lysis in vitro and in vivo. Blood. 1991;78:1005-1018.
32.
Bode C, Hudelmayer M, Mehwald P, Bauer S, Freitag M,
von Hodenburg E, Newell JB, Kübler W, Haber E, Runge M.
Fibrin-targeted recombinant hirudin inhibits fibrin deposition
on experimental clots more efficiently than recombinant
hirudin. Circulation. 1994;90:1956-1963.
33. Rade JJ, Lee SW, Dichek DA. Viral vector-mediated expression of biologically active hirudin in cultured endothelial cells. Circulation. 1993;88(pt 2):I-418. Abstract.
34. Drake TA, Cheng J, Chang A, Taylor FB Jr. Expression of tissue factor, thrombomodulin, and E-selectin in baboons with lethal E. coli sepsis. Am J Pathol. 1993;142:1458-1470. [Abstract]
35.
Pober JS, Cotran RS. Cytokines and
endothelial cell biology. Physiol
Rev. 1990;70:427-451.
36. Briscoe DM, Schoen FJ, Rice GE, Bevilacqua MP, Ganz P, Pober JS. Induced expression of endothelial-leukocyte adhesion molecules in human cardiac allografts. Transplantation. 1991;51:537-547.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  B.-S. Ding, T. Dziubla, V. V. Shuvaev, S. Muro, and V. R. Muzykantov Advanced drug delivery systems that target the vascular endothelium. Mol. Interv., April 1, 2006; 6(2): 98 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B.-S. Ding, C. Gottstein, A. Grunow, A. Kuo, K. Ganguly, S. M. Albelda, D. B. Cines, and V. R. Muzykantov Endothelial targeting of a recombinant construct fusing a PECAM-1 single-chain variable antibody fragment (scFv) with prourokinase facilitates prophylactic thrombolysis in the pulmonary vasculature Blood, December 15, 2005; 106(13): 4191 - 4198. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Chen, K. Giannopoulos, P. G. Shiels, Z. Webster, J. H. McVey, G. Kemball-Cook, E. Tuddenham, M. Moore, R. Lechler, and A. Dorling Inhibition of intravascular thrombosis in murine endotoxemia by targeted expression of hirudin and tissue factor pathway inhibitor analogs to activated endothelium Blood, September 1, 2004; 104(5): 1344 - 1349. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Christofidou-Solomidou, A. Scherpereel, R. Wiewrodt, K. Ng, T. Sweitzer, E. Arguiri, V. Shuvaev, C. C. Solomides, S. M. Albelda, and V. R. Muzykantov PECAM-directed delivery of catalase to endothelium protects against pulmonary vascular oxidative stress Am J Physiol Lung Cell Mol Physiol, August 1, 2003; 285(2): L283 - L292. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Yoshida, T. Sawada, H. Ishii, R. E. Gerszten, A. Rosenzweig, M. A. Gimbrone Jr, Y. Yasukochi, and F. Numano HMG-CoA Reductase Inhibitor Modulates Monocyte-Endothelial Cell Interaction Under Physiological Flow Conditions In Vitro : Involvement of Rho GTPase-Dependent Mechanism Arterioscler Thromb Vasc Biol, July 1, 2001; 21(7): 1165 - 1171. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Riesbeck, D. Chen, G. Kemball-Cook, J. H. McVey, A. J. T. George, E. G. D. Tuddenham, A. Dorling, and R. I. Lechler Expression of Hirudin Fusion Proteins in Mammalian Cells : A Strategy for Prevention of Intravascular Thrombosis Circulation, December 15, 1998; 98(24): 2744 - 2752. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. D. Spragg, D. R. Alford, R. Greferath, C. E. Larsen, K.-D. Lee, G. C. Gurtner, M. I. Cybulsky, P. F. Tosi, C. Nicolau, and M. A. Gimbrone Jr. Immunotargeting of liposomes to activated vascular endothelial cells: A strategy for site-selective delivery in the cardiovascular system PNAS, August 5, 1997; 94(16): 8795 - 8800. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||