| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
2
* Pulmonary, Allergy and Critical Care Division, Department of Medicine,
Institute of Environmental Medicine and
Department of Pharmacology, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104-6068, USA
2Correspondence: Institute of Environmental Medicine, University of Pennsylvania Medical Center, 1 John Morgan Building, 36th Street and Hamilton Walk, Philadelphia, PA 19104-6068, USA. E-mail: muzykant{at}mail.med.upenn.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: drug targeting • lung • PECAM-1 • ß-galactosidase • streptavidin
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Some progress has been made toward achieving effective routes for
cellular entry of proteins (protein transduction). One approach uses
genetically engineered fusion proteins using membrane-permeating
peptides coupled to foreign enzymes (1
2
3)
. As an example,
Schwarze et al. recently demonstrated the ubiquitous expression of an
active enzyme in vivo after intraperitoneal injection of a
denatured fusion protein composed of the protein transduction domain of
the human immunodeficiency virus TAT protein fused to ß-galactosidase
(4)
. Although useful, this strategy currently lacks
specificity, that is, protein expression cannot be targeted to a
desired tissue. A potentially even more useful application of protein
transduction would combine effective intracellular entry with
cell-selective targeting of the desired protein in vivo.
Vascular immunotargeting theoretically offers such an approach
(5
, 6)
. The endothelium represents a large, accessible,
and important therapeutic target for the treatment of a spectrum of
cardiovascular and pulmonary diseases. However, a successful vascular
immunotargeting strategy will depend on two processes. First,
conjugates of effector compounds (e.g., enzymes) with antibodies
against specific endothelial surface antigens must bind selectively to
endothelium. Second, these conjugates must then become internalized and
routed to an intracellular compartment that allows the active
conjugated drug to act. Successful targeting of the conjugates to
pulmonary endothelium has been demonstrated previously. Because of the
extremely large surface area and effective circulation, carrier
antibodies accumulate in the lungs and have been shown to bring about
preferential pulmonary uptake of radiolabeled enzymes after intravenous
(i.v.) administration in intact animals (6
, 7
, 8
, 9)
.
However, the second part of the process, intracellular delivery of an
active enzyme, has not been demonstrated in intact animals.
Our recent studies show that antibodies to platelet endothelial cell
adhesion molecule-1 (anti-PECAM) conjugated with streptavidin (SA, a
biotin binding protein that cross-links biotinylated enzymes and
antibodies) can serve as a carrier for intracellular delivery of
conjugated foreign cargo compounds (such as enzymes and genes) to
endothelium. We have found that streptavidin markedly facilitates
intracellular uptake of anti-PECAM in endothelial cell culture in
vitro and its pulmonary targeting in vivo in intact
animals (9)
, and that glucose oxidase conjugated with
anti-PECAM/SA carrier enters endothelial cells and generates
H2O2 intracellularly in
cell cultures (10)
. We have also observed that
radiolabeled glucose oxidase conjugated with an anti-PECAM/SA carrier
accumulate in the lungs after intravascular injection in intact animals
(9
, 11)
.
Based on these results, we postulated that anti-PECAM/SA could deliver active enzymes inside endothelial cells in vivo. In the present study we tested this hypothesis directly by characterizing the functional activity, along with the tissue and cellular localization using confocal and electron microscopy, of a reporter enzyme (ß-galactosidase, ß-Gal) conjugated to an anti-PECAM carrier. The results indicate that anti-PECAM permits cell-selective, intracellular delivery of an active foreign enzyme, ß-Gal, to endothelium in cell culture and intact animals. The kinetics of this process were also defined. Our results validate the feasibility of vascular immunotargeting of enzymes and imply that this strategy may be useful for the treatment of acute pulmonary or vascular diseases.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; and
2) nonspecific rat IgG (Calbiochem, San Diego, Calif.). The
preparation of ß-galactosidase covalently coupled to streptavidin
(SA-ß-Gal, with specific ß-Gal activity 400 units/mg) was purchased
from Sigma Chemicals (St. Louis, Mo.).
Biotinylation, radiolabeling, and conjugation of proteins
Proteins were biotinylated with 6-biotinylaminohexanoic acid
N-hydroxysuccinimide ester (NHS-LC-Biotin, Pierce, Rockford,
Ill.) without detectable reduction of their functional activity as
described previously (6)
and designated as b-IgG or
b-anti-PECAM. Prior to conjugation with biotinylated proteins,
SA-ß-Gal was dissolved at 0.5 mg/ml in phosphate-buffered saline
(PBS), pH 7.3 and labeled with 125Iodine (NEN
Dupont) using Iodogen (Pierce), according to a standard manufacturer’s
instructions. Either b-IgG or b-anti-PECAM was added dropwise to
[125I]SA-ß-Gal in the vortex at room
temperature (RT). The molar ratio
[125I]SA-ß-Gal to biotinylated proteins
varied from 0.25 to 2 in order to obtain nonaggregated, stable
conjugates. Size of the resulting conjugates was determined using a
Dynamic Light Scattering machine (Brookhaven Instruments, N.Y.), with
60°, 90°, and 120° angles. The radiolabeled conjugates are
designated in the text as
anti-PECAM/[125I]SA-ß-Gal or
IgG/[125I]SA-ß-Gal, whereas nonlabeled
counterparts are designated as anti-PECAM/SA-ß-Gal and
IgG/SA-ß-Gal.
The measurement of ß-Gal activity in vitro and
in vivo
ß-Galactosidase activity in the conjugate preparations, cell
lysates, or organ homogenates was determined using a ß-Gal enzyme
assay kit from Promega (Madison, Wis.). Organs were homogenized in 3 ml
of the reporter lysis buffer (RLB) 1x from the kit containing protease
inhibitor mixture (10 µl/ml, Sigma) and centrifuged at 4°C, 4000
RPM (3000 g) for 45 min. Enzymatic ß-Gal activity was
determined in the supernatants at various dilutions. The BCA Protein
Assay Kit (Pierce) was used to measure the protein concentration in the
samples.
Binding and uptake of the conjugates in cell culture
Human mesothelioma REN cells transfected with cDNA encoding
murine PECAM (REN/PECAM cells, obtained as described (13)
,
were used in the study. Nontransfected REN cells served as a cell
type-matched control. The conjugates were incubated with the cells in
RPMI medium (GibcoBRL, Grand Island, N.Y.) with 10% fetal calf serum
for 1 h at 37°C. Afterward, in some wells, according to the
ß-Gal enzyme assay protocol, cells were washed with PBS
(Ca2+ and Mg2+ free) and
incubated with RLB for 15 min at RT. The cell lysates were collected,
centrifuged, and ß-Gal activity was determined in the supernatants.
In the parallel wells, cells were washed with RPMI and fixed with 2%
paraformaldehyde for 10 min at RT. In some wells, cells were incubated
for 10 min with 0.1% Triton X, in order to attain complete
permeabilization of the plasma membrane. Permeabilized and
nonpermeabilized cells were further incubated with rabbit antibody
against ß-Gal (1:250 dilution, 5Prime
3Prime, Inc., Boulder,
Colo.), washed, and incubated with Texas red-labeled goat anti-rabbit
antibody (1:250 dilution, Jackson ImmunoResearch Lab., Inc, Bar Harbor,
Maine) for 30 min at RT, then washed with Dulbecco’s PBS. Images were
taken using a fluorescent microscope (Olympus, Tokyo, Japan) or using
an inverted epifluorescence Nikon TE300 microscope with 60x
oil-immersion objective, appropriate filter sets, and a confocal
attachment (Radiance 2000, Bio-Rad, Hercules, Calif.).
Biodistribution and tissue localization of the conjugates in
intact animals
The biodistribution of radiolabeled conjugates in animals was
studied as described previously (6)
. One hour after the
tail vein injection of
anti-PECAM/[125I]SA-ß-Gal or
IgG/[125I]SA-ß-Gal conjugates in anesthetized
BALB/c mice (8 µg of
[125I]SA-ß-Gal/animal), animals were killed,
internal organs were harvested, rinsed with saline, and the
125I in tissues was determined in a gamma counter
(Wallac-LKB). The lung is a very vascularized organ and as much as
10–15% of the wet lung weight may belong to a residual blood. Thus,
theoretically, the value of lung-associated conjugate may be
contaminated by the blood pool. However, in previous studies we
observed that flushing of pulmonary vasculature with saline did not
change significantly values of the specific pulmonary uptake of
immunoconjugates in animals in vivo (6
, 7)
.
Thus, in the present study we did not flush lungs with saline. However,
to account for potential contribution of the residual blood in every
organ, we determined the blood level of 125I in
each experiment, calculated tissue-to-blood ratio (localization ratio),
and compared this parameter for
anti-PECAM/[125I]SA-ß-Gal or
IgG/[125I]SA-ß-Gal conjugates.
The tissue localization of enzymatically active ß-Gal in mice was
studied by the ß-Gal activity assay in the tissue homogenates and by
histological analysis of X-Gal staining. After injection of 30–50 µg
of the conjugates into the tail vein of BALB/c mice, animals were
killed at different time intervals. The internal organs were cleared of
blood by intra-cardiac injection of 20 ml of PBS via the right
ventricle, harvested, and rinsed externally with PBS. Organs were
frozen either in RLB for the ß-Gal activity assay performed as
described above or in OCT (VWR Scientific Products, Irving, Tex.) for
the X-Gal staining. Frozen sections of tissues were rinsed in PBS,
fixed in 0.5% glutaraldehyde, and incubated in X-Gal solution for
4 h at RT (14)
.
Electron microscopy of the endothelial uptake of
anti-PECAM/SA-ß-Gal conjugate
No data available.
Routine embedding for immunoelectron microscopy
Lungs were fixed in ice-cold fixative (2%
para-formaldehyde/0.1% glutaraldehyde) in 0.1 M cacodylate buffer pH
7.4 for 30 min, cut into 1 mm3 pieces, and
allowed to fix further under vacuum for several hours. After three
washings in cacodylate buffer, lungs were incubated in 0.1 M glycine
for 1 h at RT. Lungs were then washed three times with cacodylate
buffer, followed by dehydration in a series of graded alcohols [15%,
35%, 70%, 95%, 100% (3x)] for 10 min in each step. Sections were
infiltrated with LR White resin (EM Sciences, Malvern, Pa.) and allowed
to polymerize at 55°C for 3 days.
Immunoelectron microscopy
Ultra-thin sections (100 nm) were cut with a diamond knife
(Diatome, Switzerland) using a Leica ultra-microtome (Chicago, Ill.).
LR White sections were collected on 200 mesh nickel grids (Electron
Microscopy Sciences) and first rinsed on drops of water, followed by
etching in 10% NaIO4 and 0.1 M HCl, each for 15 min, and then by
blocking in 5% serum each for 15 min. Grids were floated on drops of
PBS solution containing the primary antibody (rabbit polyclonal
anti-ß-Gal, 1:100 dilution) overnight at 4°C. Nonbound anti-ß-Gal
was washed with PBS containing 0.5% bovine serum albumin, 0.1% fish
gelatin, and 0.05% Tween 20. Grids were floated for 30 min in the
drops of PBS solution containing the secondary antibody (gold
conjugated, goat-anti-rabbit IgG-Au10, Amersham,
1:10 dilution). Sections were rinsed as described previously and
postfixed in 2% glutaraldehyde/PBS for 2 min, followed by washing with
water, and postcontrasted with 20% uranyl acetate/0.5% lead citrate.
The sections were inspected in a Hitachi H-600 transmission electron
microscope (Nissey Sangyo, Gaithersburg, Md.) at 75 KV.
Statistics
Analysis of statistically significant differences
(P<0.05) between groups was performed using a t
test or a one-way analysis of variance (SigmaStat 2.0.). Post
hoc testing was performed using the Fischer least square
difference test (7)
. If not otherwise indicated, all data
are expressed as mean ± SE.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Binding of the conjugates to the target cells in
vitro
To characterize the behavior of anti-PECAM/SA-ß-Gal conjugates
in an in vitro model of endothelium, we studied an
endothelium-like cell line (human mesothelioma REN cells) transfected
with murine PECAM. This model has advantages of stability,
reproducibility, and allows the use of nontransfected REN cells as a
cell type-matched negative control.
Figure 1
shows fluorescent microscope images of the conjugates binding to the
cells revealed in the permeabilized cells after 1 h incubation by
a polyclonal antibody directed against ß-Gal (anti-ß-Gal). The
control conjugate, IgG/SA-ß-Gal, showed no significant binding to
either PECAM-negative or PECAM-positive cells (Fig. 1A
, B
).
In contrast, intense cellular fluorescence was observed after binding
of anti-PECAM/SA-ß-Gal to REN/PECAM, but not control REN cells
(compare panels C and D, Fig. 1
). These data show
the specificity of anti-PECAM/SA-ß-Gal recognition of the
antigen-positive cells.
|
To more precisely characterize the cellular localization of
anti-PECAM/SA-ß-Gal in REN/PECAM cells, we compared fluorescence
images obtained with anti-ß-Gal in nonpermeabilized vs. permeabilized
cells (Fig. 2
). Confocal microscope sections of the nonpermeabilized cells revealed
bright fluorescence on the luminal surface (Fig. 2A
) and in
the plasma membrane (Fig. 2B
), but not in the intracellular
compartment. In contrast, permeabilized cells displayed an intense
intracellular anti-ß-Gal fluorescence that increased with the depth
of the cell sections (Fig. 2C
, D
). Therefore, REN/PECAM
cells internalize anti-PECAM/SA-ß-Gal.
|
Anti-PECAM/SA-ß-Gal associated with REN/PECAM cells was
enzymatically active (Fig. 3
). After 1 h incubation of anti-PECAM/SA-ß-Gal with REN/PECAM
cells at 37°C, ß-Gal activity in the wells attained 80 mU/mg,
whereas it was an order of magnitude lower in the control wells (e.g.,
8.6±0.2 mU/mg in REN cells). Likely due to the high sensitivity of the
enzymatic assay, we detected a significant level of ß-Gal activity in
the REN/PECAM cells incubated with IgG/SA-ß-Gal (
20 mU/mg).
Nevertheless, it was fourfold lower than that in the REN/PECAM cells
incubated with anti-PECAM/SA-ß-Gal conjugate. Incubation of REN/PECAM
cells (preincubated with anti-PECAM/SA-ß-Gal for 1 h at 37°C)
with a glycine buffer, pH 2.5 (an elution procedure that eliminates the
surface-bound conjugate) led to only a 30% reduction of the
cell-associated ß-Gal activity (not shown). Together with the
confocal microscopy pattern of anti-ß-Gal staining shown on Fig. 2
,
these data indicate that 1) REN/PECAM cells internalize the
major proportion (60–70%) of the cell-associated
anti-PECAM/SA-ß-Gal and 2) intracellular ß-Gal is
active. These findings corroborate with our previous in
vitro observations that endothelial and REN/PECAM cells
internalize an active glucose oxidase conjugated with anti-PECAM/SA
carrier (9
, 10)
.
|
Immunotargeting of anti-PECAM/[125I]SA-ß-Gal to
the vascular endothelium in intact animals
In the next series of experiments, we measured the uptake of
[125I]SA-ß-Gal conjugated with either
anti-PECAM or IgG in the internal organs of mice 1 h after i.v.
administration (Table 1
). The conjugates initially showed accumulation of
125I in the liver [22–26% injected dose/g of
tissue (% injected dose/g)] and spleen (13–15% injected dose/g).
The hepatic and splenic immunospecificity index (ISI, ratio between the
tissue uptake of immune and nonimmune counterparts) was close to 1.
This result indicates equal uptake of the immune and nonimmune
conjugates and likely reflects PECAM-independent, Fc receptor-mediated
uptake by macrophages in these organs. Even though the blood level of
the immune conjugate was similar to and did not exceed that of
nonimmune counterpart, uptake of
anti-PECAM/[125I]SA-ß-Gal doubled that of
IgG/[125I]SA-ß-Gal in heart and kidney
(P<0.01). Thus, renal and cardiac ISI attained a value of
2.3–2.5. This result likely reflects binding of
anti-PECAM/[125I]SA-ß-Gal to renal and
cardiac endothelium. The most striking difference between the levels of
tissue uptake of immune and nonimmune conjugates was observed in the
lungs. The lung is the most vascularized organ in the body and
endothelium-specific antibodies accumulate in the pulmonary vasculature
(5
6
7
8
9)
. Level of 125I in murine
lungs was an order of magnitude higher after injection of
anti-PECAM/[125I]SA-ß-Gal than
IgG/[125I]SA-ß-Gal (Fig. 4
). High pulmonary uptake of
anti-PECAM/[125I]SA-ß-Gal cannot be explained
by lung contamination with a residual blood, since blood levels of the
anti-PECAM conjugate did not exceed that of the control IgG conjugate.
To account for a residual blood, we calculated lung-to-blood ratio
(localization ratio) for immune and nonimmune conjugates. The ratio of
these two parameters gives the ISI (normalized to blood) that exceeds
10 (see Table 1
). In fact, pulmonary uptake of the immune conjugate
markedly exceeded that of any other organ, likely reflecting binding of
the conjugate to the pulmonary endothelium. Thus, anti-PECAM/SA-ß-Gal
injected i.v. accumulates preferentially in the lungs.
|
|
Anti-PECAM/SA-ß-Gal causes organ-selective elevation ß-Gal
activity in vivo
To evaluate whether anti-PECAM/SA carrier delivers enzyme to
endothelium in an active form, ß-Gal activity was measured in the
internal organs after i.v. injection of the conjugates. Table 2
shows the profile of ß-Gal activity in various organs 1 and 4 h
after conjugate injection. Similar to the 125I
distribution data, 1 h after injection of either the immune or
nonimmune conjugate, significant ß-Gal activity was detected in the
liver and spleen. The renal ß-Gal activity after
anti-PECAM/SA-ß-Gal injection doubled that seen after IgG/SA-ß-Gal
injection. Again, the most striking difference between the immune and
nonimmune conjugates was observed in the lung. After
anti-PECAM/SA-ß-Gal injection, pulmonary ß-Gal activity attained
the highest level of any organ and exceeded that seen after
IgG/SA-ß-Gal injection by an order of magnitude. No significant
ß-Gal activity was observed in the lungs after injection of
nonconjugated SA-ß-Gal (data not shown).
|
ß-Gal activity in the liver and spleen was reduced to a background
level during the time interval from 1 to 4 h postinjection. One
explanation for this rapid decline of activity is that the conjugates
are degraded by the resident reticuloendothelial system cells. In
contrast, ß-Gal activity in the lung tissue was reduced by only 30%
during this time interval and still was an order of magnitude higher
than the control level (Table 2)
. The specific (i.e., provided by
anti-PECAM/SA-ß-Gal, but not IgG/SA-ß-Gal) and prolonged character
of ß-Gal activity in the lung, together with its localization in the
tissue (see below), implies that anti-PECAM delivers active ß-Gal to
the pulmonary vascular endothelium.
Figure 5
compares the kinetics of ß-Gal activity in the lung and spleen.
Anti-PECAM/SA-ß-Gal i.v. injection caused very rapid elevation of
ß-Gal activity in both organs, reaching a peak level 15–20 min
postinjection. However, in contrast to a rapid decline of ß-Gal
activity in spleen, pulmonary ß-Gal activity displayed a two-phase
decline: a rapid reduction to 50% of the peak level within first hour,
followed by a slower reduction (to 30% of the peak level 4 h
postinjection). As long as 8 h postinjection, ß-Gal activity in
the lung tissue remained at an elevated level (almost threefold higher
than the background level, P<0.01). One day after a single
bolus injection of anti-PECAM/SA-ß-Gal conjugate, ß-Gal activity in
the lung tissue returned to a background level.
|
These data indicate that i.v. administration of anti-PECAM/ß-Gal causes a more prominent and more persistent elevation of ß-Gal activity in the lungs than in other organs. Enzymatic activity in the lung was detectable for at least 8 h after injection.
Endothelial localization of ß-Gal activity in
vivo
To define the tissue localization of active ß-Gal in the lung
and other organs, frozen sections of the organs were analyzed after
standard X-gal staining. Figure 6
shows representative results of this analysis in organs harvested
4 h after conjugate injection. Consistent with the quantitative
data shown in Table 2
, minimal ß-Gal activity was detected in the
sections of the liver tissue 4 h postinjection of either
anti-PECAM/SA-ß-Gal or IgG/SA-ß-Gal. Residual ß-Gal activity was
detected at the periphery of the follicles in specific areas of the
spleen. This activity was presumably due to nonspecific uptake, since
the pattern of staining was similar for anti-PECAM/SA-ß-Gal and
IgG/SA-ß-Gal. In the kidney tissue sections, specific ß-Gal
activity was detected in the renal glomeruli and in the lumen of blood
vessels after injection of anti-PECAM/SA-ß-Gal, but not
IgG/SA-ß-Gal. Intense ß-Gal activity was detected in the lung
tissue after injection of anti-PECAM/SA-ß-Gal, but not
IgG/SA-ß-Gal.
|
The photomicrographs shown in Fig. 7
illustrate points relating to the localization of ß-Gal activity in
the lungs after injection of anti-PECAM/SA-ß-Gal. First, ß-Gal
activity was homogeneously disseminated throughout the highly
vascularized alveolar compartment containing the major fraction of the
pulmonary capillaries (Fig. 7A
). Second, ß-Gal activity
was also detected in the lumen, but not in the tunica media and
adventitia of large blood vessels (Fig. 7A
, C
). Third, no
ß-Gal activity was detected in the airways and interstitial
compartments (Fig. 7B
). Therefore, these data indicate that
anti-PECAM carrier delivers an active ß-Gal selectively to
endothelium in intact animals. Endothelial cells in pulmonary
vasculature and the glomerular capillaries seem to be preferential
targets for the PECAM-directed delivery after i.v. administration.
|
Intracellular uptake of anti-PECAM/SA-ß-Gal in pulmonary
endothelium in vivo
To determine the subcellular localization of ß-Gal in the lungs,
we used immuno-gold electron microscopy with a ß-Gal antibody. Rare
cases of binding of gold particles to white blood cells in the
pulmonary vessels were observed after injection of the conjugates,
likely reflecting Fc receptor-mediated uptake by resident leukocytes
(see as an example in Fig. 8A
). No specific binding of gold particles in the lung tissue
could be detected after injection of saline or IgG/SA-ß-Gal (data not
shown). Rare gold particles found in the lung tissue (no more than one
or two particles per field) could not be attributed to any specific
structure in the lung tissue. This result corroborates with the data
documenting low levels of the pulmonary uptake of IgG/SA-ß-Gal (see
Tables 1
and 2
) and confirms the specificity of secondary anti-ß-Gal
and immuno-gold conjugate used in the study.
|
Figure 8
shows the typical patterns of the immuno-gold particles
binding to sections of murine lungs obtained after anti-PECAM/ß-Gal
injection. In this type of experiment (as opposed to direct staining of
tissue sections with primary and secondary antibodies), only a portion
of antigen molecules can accommodate the conjugate and gold particles.
Therefore, one could expect a relatively modest labeling of the tissue.
However, as many as 20–30 gold particles could be detected per
capillary vessel in most of the fields, revealing anti-PECAM/ß-Gal in
the pulmonary vasculature. Only rare gold particles were found in
association with epithelial cells, red blood cells or components of the
interstitial compartment. This result confirms that anti-PECAM carrier
delivers ß-Gal selectively to the endothelial cells. In contrast,
numerous gold particles could be found in association with endothelial
cells. We were able to visualize three patterns of immuno-gold labeling
of endothelial cells. First, some gold particles were associated with
endothelial plasma membrane (labeled as small arrowheads in Fig. 8A
, B
, C
, D
). Even in relatively clear areas (see Fig. 8D
), the resolution of our method did not allow us to
attribute these single particles to either intracellular or
extracellular compartment. Second, small clusters of gold particles
were localized in the endothelial junctions (indicated as large
arrows), the site of preferential PECAM expression in the endothelial
cells and likely site of the cellular entry for the
anti-PECAM/SA-ß-Gal conjugate (see Fig. 8B
). Finally, the
most common pattern was the presence of numerous gold particles and
their clusters localized in the intracellular endothelial vacuoles,
indicated by large arrowheads (Fig. 8A
, B
, C
). For example,
Fig. 8A
outlines a cross section of a pulmonary capillary
vessel. The insert clearly shows a large vesicular structure in the
endothelial cell filled with gold particles. In fact, 38 of a total of
43 gold particles associated with this cell localize in this vesicle.
This type of vesicular structure filled with conjugate was a common
finding in the lung sections. Particles associated with the plasma
membrane invaginations and vacuoles in the cytoplasm imply that the
endothelial cells are actively internalizing the conjugate. In general,
electron microscopy findings in the lung tissue after injection of
anti-PECAM/SA-ß-Gal were similar to those in the endothelial cells
incubated with anti-PECAM/SA/ferritin in cell culture (9)
;
both revealed internalization of antiPECAM conjugates by
endothelial cells. Therefore, the results of electron microscopy
indicate that 1) anti-PECAM delivers ß-Gal selectively to
the endothelial cells and 2) a significant proportion of the
delivered enzyme undergoes intracellular uptake.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 7
, 8
, 9
, 15
, 16)
. We and others have shown that selective pulmonary delivery
of radiolabeled conjugates is possible (6
7
8
9)
. Very
recently, we have also generated data in mice using anti-PECAM/SA
conjugates of glucose oxidase (GOX, a
H2O2-producing enzyme) to
show that functional enzymes can be delivered to the lung using this
approach. We found that anti-PECAM/SA/GOX accumulated in the lungs
after i.v. injection and caused severe oxidative injury in the
pulmonary vasculature (11)
. However, rapid and extensive
injury caused by H2O2
generated in the lung made the visualization of the enzyme and kinetic
studies of activity of the targeted enzyme impossible. Thus, to date, a
number of questions that are key to the ultimate utility of vascular
immunotargeting remain unanswered. These include 1) the
precise tissue localization of the delivered conjugates, 2)
the ability of the conjugates to move intracellularly in
vivo, and 3) the time course and kinetics of enzymatic
activity after conjugate delivery. To answer these questions, we studied the delivery of a well-characterized reporter enzyme, ß-galactosidase, which allows measurement of its activity and localization in tissues. Using this approach, we could show that 1) anti-PECAM/SA-ß-Gal recognizes the vascular endothelium and accumulates preferentially in the lungs after i.v. administration; 2) endothelium-associated ß-Gal displays high enzymatic activity that is detectable within several minutes after a single bolus injection and lasts for at least 8 h, and 3) a proportion of the injected ß-Gal conjugates appear to accumulate inside the endothelial cells in cell culture and likely in vivo. This study, therefore, demonstrates for the first time that vascular immunotargeting can permit cell-selective, intracellular delivery of an active foreign enzyme to endothelium in intact animals.
The ability to actually visualize and track the localization of active
enzyme provides some important insights. Clearly, the endothelium of
the lung represents a major target for the conjugates. The distribution
of enzyme was homogeneous: both large and small vessel endothelium
showed activity. The other area to show specific uptake was the
glomerular region of the kidney. A rapid, nonspecific uptake of both
anti-PECAM and control IgG conjugates was seen in liver and spleen.
This result is not surprising, given the major role of these organs in
the clearance of immune complexes. In fact, all radiolabeled antibody-
or IgG-based conjugates tested in our and other laboratories have
displayed high hepatic and splenic uptake (6
7
8
9)
. However,
ß-Gal enzymatic activity in these organs was very short-lived
compared to the lung. We postulate that this is due to rapid Fc
receptor-mediated uptake and degradation of the conjugates by the
‘professional’ phagocytes residing in the liver and spleen.
Additional evidence to indicate that the fate of the internalized
conjugates is different in liver and spleen vs. lung and kidney is
provided by our recent study using anti-PECAM/GOX
conjugates (11)
. In this study, marked hepatic and splenic
uptake of both anti-PECAM/[125I]GOX and
IgG/[125I]GOX was noted 1 h postinjection,
yet these organs remained uninjured. The injury was induced only by
antiPECAM/GOX and was localized in the lung (the organ where
anti-PECAM/[125I]GOX, but not
IgG/[125I]GOX, displayed a preferential uptake)
and, to a lesser extent, in the renal glomeruli (11)
.
A prolonged, two-phase decrease of the conjugate activity in the lung
may be explained in several ways. It could perhaps reflect a
‘reutilization’ or reuptake of the conjugate released by liver and
spleen. We do not favor this scenario because the conjugate would
likely loose its antigen binding and enzymatic activity after passage
through macrophages in these organs. Another possibility is that an
initial rapid phase of pulmonary degradation is due to uptake and
degradation of the conjugate by resident mononuclear cells or
macrophages in the lungs (see, for example, Fig. 8A
).
However, the very homogeneous distribution of the conjugate throughout
the pulmonary vasculature (Fig. 7A
) indirectly argues
against this scenario. Another explanation could be a two-component
mechanism of the conjugate degradation in the pulmonary endothelium
(e.g., one rapid, such as the lysosomal pathway, and another that is
slower). The mechanism(s) of the conjugate degradation in the lung is
not clear at the present time, but it may be different for different
types of conjugated enzymes. The topic represents a subject for more
in-depth investigation. Nevertheless, our results clearly indicate that
the anti-PECAM-conjugated reporter enzyme displays significant activity
in the pulmonary vasculature for at least many hours after i.v.
administration. This time interval may be sufficient to complement a
gene delivery system.
The generalizability of these observations is an area of active
investigation. In the present study, we used antibodies directed
against PECAM-1 to provide immunotargeting to lung endothelial cells.
PECAM-1 was chosen because 1) it is highly expressed on the
surface of endothelium; 2) it mediates transmigration of
leukocytes from blood to the tissues; and 3) inflammatory
mediators do not suppress its expression by endothelium
(17)
. Blocking of PECAM by antibodies or conjugates may
inhibit leukocyte infiltration and provide a secondary therapeutic
benefit in inflammatory settings (18)
. Preliminary studies
also indicate that, at least in part, pulmonary deposition of
anti-PECAM conjugate is a ‘first-pass’ phenomenon and that selective
catheterization may facilitate vascular delivery to specific organs
such as the heart. However, other targeting antibodies have also been
evaluated and appear to function as effective carriers
(5)
. For example, we and others have successfully used
anti-ACE and anti-thrombomodulin antibodies to deliver radiolabeled
enzymes to pulmonary endothelium in rat and mouse models, although the
tissue localization and functional activity of the delivered enzymes
remain to be characterized (5
, 7
, 8)
.
The choice of targeted enzymes and the physical structure of the
conjugates also require further study. Our preliminary data from
in vitro studies suggest that the majority of conjugated
enzyme is initially trafficked to large vacuoles that eventually fuse
with lysosomes. Enzymes that generate or detoxify highly diffusible
oxidants (such as hydrogen peroxide) may be well suited for this type
of delivery, as precise cytoplasmic localization may not be necessary.
There are, however, data that conjugates can escape from lysosomes and
move to other cellular compartments, such as the nucleus. We have
generated anti-PECAM/polylysine/green fluorescent protein DNA
conjugates that selectively transfect PECAM-1-expressing cells
(19)
. Effective delivery of conjugates may also depend on
the size and aggregation status of the conjugates. Analysis of
conjugate size using dynamic light scattering has indicated that
smaller particles (100–200 nm in diameter) are more efficient in
intracellular delivery of enzymes in vitro. Studies to
correlate size and delivery destination/efficiency in vivo
are under way.
The ability to transiently deliver active enzymes at high levels to the vasculature has a number of potentially important therapeutic implications. For example, conjugates of antioxidant enzymes such as catalase or superoxide dismutase could be perfused prophylactically through donor organs to modulate ischemia/reperfusion injury associated with transplantation. Endothelial oxidant injury is also thought to play a major role in a number of other pulmonary diseases such as acute respiratory distress syndrome, radiation lung injury and oxygen toxicity and could thus potentially be approached using this technology.
In conclusion, our data indicate that anti-PECAM carrier permits cell-selective, intracellular delivery of an active foreign enzyme to endothelium in intact animals. These results validate the feasibility of vascular immunotargeting of enzymes and suggests that this strategy may be useful for the treatment of acute pulmonary or vascular diseases.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication January 21, 2000.
Revision received August 2, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  V. V. Shuvaev, S. Tliba, M. Nakada, S. M. Albelda, and V. R. Muzykantov Platelet-Endothelial Cell Adhesion Molecule-1-Directed Endothelial Targeting of Superoxide Dismutase Alleviates Oxidative Stress Caused by Either Extracellular or Intracellular Superoxide J. Pharmacol. Exp. Ther., November 1, 2007; 323(2): 450 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
|
|
Y. Zhang and W. M. Pardridge Delivery of {beta}-Galactosidase to Mouse Brain via the Blood-Brain Barrier Transferrin Receptor J. Pharmacol. Exp. Ther., June 1, 2005; 313(3): 1075 - 1081. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Moghimi, A. C. Hunter, and J. C. Murray Nanomedicine: current status and future prospects FASEB J, March 1, 2005; 19(3): 311 - 330. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Muro, X. Cui, C. Gajewski, J.-C. Murciano, V. R. Muzykantov, and M. Koval Slow intracellular trafficking of catalase nanoparticles targeted to ICAM-1 protects endothelial cells from oxidative stress Am J Physiol Cell Physiol, November 1, 2003; 285(5): C1339 - C1347. [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]
|
|
|
|
|
|
J.-C. Murciano, S. Muro, L. Koniaris, M. Christofidou-Solomidou, D. W. Harshaw, S. M. Albelda, D. N. Granger, D. B. Cines, and V. R. Muzykantov ICAM-directed vascular immunotargeting of antithrombotic agents to the endothelial luminal surface Blood, May 15, 2003; 101(10): 3977 - 3984. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Muro, R. Wiewrodt, A. Thomas, L. Koniaris, S. M. Albelda, V. R. Muzykantov, and M. Koval A novel endocytic pathway induced by clustering endothelial ICAM-1 or PECAM-1 J. Cell Sci., April 15, 2003; 116(8): 1599 - 1609. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Scherpereel, J. J. Rome, R. Wiewrodt, S. C. Watkins, D. W. Harshaw, S. Alder, M. Christofidou-Solomidou, E. Haut, J.-C. Murciano, M. Nakada, et al. Platelet-Endothelial Cell Adhesion Molecule-1-Directed Immunotargeting to Cardiopulmonary Vasculature J. Pharmacol. Exp. Ther., March 1, 2002; 300(3): 777 - 786. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Wiewrodt, A. P. Thomas, L. Cipelletti, M. Christofidou-Solomidou, D. A. Weitz, S. I. Feinstein, D. Schaffer, S. M. Albelda, M. Koval, and V. R. Muzykantov Size-dependent intracellular immunotargeting of therapeutic cargoes into endothelial cells Blood, February 1, 2002; 99(3): 912 - 922. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Danilov, V. D. Gavrilyuk, F. E. Franke, K. Pauls, D. W. Harshaw, T. D. McDonald, D. J. Miletich |
