Vascular Biology |
From the Ludwig Institute for Cancer Research (O.L., K.P., A.Ö., C.-H.H.), and the Department of Surgery (O.L., M.-A.C., D.B.), Uppsala, Sweden; Nexstar Pharmaceuticals Inc (N.J., C.V., L.S.G.), Boulder, Colo; and the Wallenberg Laboratory (M.L.), Sahlgrenska University Hospital, Göteborg University,, Göteborg, Sweden.
Correspondence to Carl-Henrik Heldin, Ludwig Institute for Cancer Research, Box 595, S-75124, Uppsala, Sweden. E-mail C-H.Heldin{at}licr.uu.se
| Abstract |
|---|
|
|
|---|
Key Words: balloon dilatation intimal hyperplasia PDGF antagonist
| Introduction |
|---|
|
|
|---|
PDGF, a growth factor and chemoattractant for cells of mesenchymal
origin (reviewed in Reference 11 ), occurs as 3 dimeric
isoforms, PDGF-AA, -AB, and -BB, composed of disulfide-linked A-, B-,
and C-chains. PDGF signals are mediated by structurally related
-
and ß-receptors. In the vascular smooth muscle cell proliferation and
migration that occur in intimal hyperplasia, most PDGF effects are
mediated by the PDGF-ß receptor.12 13 15 16 The
ß-receptor is activated by PDGF-BB and, when coexpressed with
the PDGF-
receptor, also by PDGF-AB.
The development of specific PDGF antagonists has allowed studies on the effects of such compounds in different animal models of intimal hyperplasia formation. Promising results have been obtained in rat, porcine, and nonhuman primate models of arterial injury through the use of, for example, antibodies against PDGF and PDGF receptors, receptor antisense oligonucleotides, low-molecular-weight PDGF receptor tyrosine kinase inhibitors, and local overexpression of PDGF-ß receptor extracellular domains.12 14 15 16 17 18 19 20 21 22 23 24 However, one remaining question to be answered is the durability of the observed beneficial effect.
To investigate the in vivo potency of PDGF aptamers in the rat carotid injury model and to investigate the durability of the therapeutic benefit of PDGF antagonists on lesion formation, we characterized the short- and intermediate-term effects on lesion formation of PDGF SELEX aptamers. This is a novel type of PDGF antagonist, ie, aptamers identified by the SELEX (systematic evolution of ligands by exponential enrichment) process (reviewed in Reference 25 ), that bind to the receptor-binding epitopes of the PDGF B-chain and thereby neutralize the biological effects of PDGF-AB and -BB.26
| Methods |
|---|
|
|
|---|
The left common carotid artery was injured with a 2F embolectomy catheter (Baxter Healthcare) basically as described by Clowes et al.27 Animals were anesthetized through intraperitoneal administration of 0.33 mL/100 g body weight of 1 part fentanyl/fluanisone (fluanisone 10 mg/mL, fentanyl 0.2 mg/mL; Hypnorm Vet, Janssen Pharmaceutica), 1 part midazolam (5 mg/mL Dormicum; Roche), and 2 parts sterile water.
Aptamers (NX1975, PDGF B-chainspecific aptamers conjugated to 40-kDa
polyethylene glycol, and NX2210, the inverted-sequence analogue used as
a control) were dissolved in sterile PBS shortly before
intraperitoneal administration. A chemically
modified form of the originally described aptamer was used, in which
the modifications aimed at increasing nuclease resistance and plasma
residence time; the modifications include conjugation to polyethylene
glycol, substitution of some of the nucleotides with
spacers, and exchange of some deoxyribonucleotides by the
corresponding 2'-O-methyl or 2'-fluoro
derivatives.28 The first 2 doses were given 14 hours
and 2 hours before injury, and the subsequent doses
every 12 hours
thereafter.
Eight, 14, or 56 days after injury, the animals were anesthetized as above. Twenty minutes before surgery, the animals received an intravenous injection of 1.0 mL of 0.5% Evans blue dye (Sigma) in PBS to allow identification of the deendothelialized vessel segment. The animals were euthanatized with an intravenous overdose of the anesthetic agent and the vasculature was cleared of blood by perfusion with ice-chilled PBS at 100 mm Hg. The distal halves of the right and left common carotid arteries were snap-frozen and stored at 85°C until further processing. Immediately thereafter, the remaining proximal segments were perfused with 2.5% glutaraldehyde in phosphate buffer, pH 7.3, and were then prepared for paraffin-embedding.
Histopathological and Immunohistological Analyses
From each fixed vessel beginning from the distal cut end, 5
sections approximately 1 mm apart were stained with Verhoeffs
elastin stain and hematoxylin-eosin for measurements of vessel wall
areas; with Harries hematoxylin for determination of the number of
nuclei in each vessel compartment; and with Picro-Sirius red collagen
stain to visualize the extracellular matrix component. All sections
from the left carotid artery were confined to the Evans bluestained
region corresponding the deendothelialized
vessel segment. Areas for the lumen, tunica intima, and tunica media
and the number of nuclei in these 2 vessel wall layers were measured
(KS 400 image acquisition and analysis system, Carl Zeiss), and
intima-to-media and area-per-nuclei ratios were calculated. The mean
value of all 5 sections was used for statistical analysis.
Cryostat sections from a single segment corresponding to the proximal cut end of the frozen specimen were fixed in 4% formaldehyde at 4°C, rinsed in 0.1 mol/L PBS, and stained as follows. Sections of the vessels where proliferating cells were to be quantified were incubated with the primary antibody against nuclear antigen Ki-67 (Ki-67 antigen MIB-5, Immunotech) at a dilution of 1:50 in PBS containing 1% bovine serum albumin and 5% horse serum. Omission of primary antibody was used as a negative control. A rat-absorbed biotinylated anti-mouse IgG antibody (1:200, Vector Laboratories) served as a secondary antibody. For detection of the secondary antibodies, an avidin-biotin complex (Vectastain Elite ABC kit, Vector Laboratories) was applied. The peroxidase reaction was carried out with 3,3'-diaminobenzidine tetrahydrochloride as the substrate. Counterstaining was performed with Mayers hematoxylin. The positive nuclei were counted in the tunica intima and tunica media (Q500IW image acquisition and analysis system, Leica), and the fraction of the positively stained nuclei was calculated.
After brief formaldehyde fixation, cryosections from animals where apoptotic cells were to be quantified were incubated in 3% H2O2 in PBS at 20°C. The terminal deoxynucleotidyl transferasemediated dUTP nick end-labeling (TUNEL) reaction was performed for 1 hour at 37°C in a humidified chamber with 6 U (0.3 U/µL) of terminal transferase (Boehringer Mannheim) per slide in reaction buffer containing 2.5 mmol/L CoCl2 and digoxigenin-UTP. After quenching of the reaction, sections were blocked in 2% bovine serum albumin a in PBS. After incubation with anti-digoxigenin antibodies conjugated to horseradish peroxidase (Boehringer Mannheim) and washing of the slides, the color reaction was performed with diaminobenzidine- H2O2 (Sigma). Counterstaining was performed with Mayers hematoxylin. The fraction of positively stained nuclei was calculated as above.
All instrumentation and measurements were performed by one individual (O.L.) blinded to treatment allocation. Two observers (M.L., M.-A.C.) controlled the measurements from randomly selected multiple samples without knowledge of the origin of the sections.
Statistical Analysis
All data are presented as mean±SD. Statistical
comparisons were determined by unpaired Students t test or
1-way ANOVA with the Bonferroni-Dunn test. Statistical significance was
accepted at the 95% confidence level.
| Results |
|---|
|
|
|---|
50% (P<0.001). This is similar to the 40% to 50%
reduction that has been observed with neutralizing antibodies against
PDGF or PDGF receptors.15 16 17 18 22 We conclude that the
NX1975 acts as a PDGF-AB/-BB antagonist in vivo after
intraperitoneal administration.
|
|
The same protocol was subsequently followed with NX1975 at lower
concentrations, ranging between 2 and 12 mg
kg-1 d-1. As shown in
Figure 1
A, reduction of the dose to 2 mg
kg-1 d-1 did not
diminish the effect of NX1975 obtained with the dose of 20 mg
kg-1 d-1. The
inverted-sequence control aptamer (NX2210) at the dose of 12 mg
kg-1 d-1 had no effect
(P=NS; Figure 1
A). There was no statistical
difference in body weight within identical observation times, as
measured at the time of injury, at harvest, or as reported as a change
of body weight, between the B-chain aptamer, control aptamer, or
PBS-treated animals. Data for animals treated with 12 mg
kg-1 d-1 NX1975 are
shown in Table 1
.
|
To investigate the mechanism whereby treatment with NX1975 reduced the
intima-media ratio, sections from lesions of treated and untreated
animals were analyzed with regard to numbers of cells in the
intima and media, cellular density (reported as area per cell), and
percentage of proliferating cells (Table 1
). In the intimal
layer, the number of cells was reduced to 63% in NX1975-treated (12 mg
kg-1 d-1) animals
compared with control animals (P=0.016). Neither cellular
density nor the fraction of proliferating smooth muscle cells in the
lesions, as monitored by Ki-67 staining after 2-week treatment,
differed significantly between treated and control animals (Table 1
).
Furthermore, we investigated whether apoptotic cell death was
involved in the inhibition of lesion formation by PDGF B-chain
aptamers. Ten animals were randomized into 2 groups, subjected to
embolectomy catheter injury, and treated with PBS or NX1975 (5 mg
kg-1 d-1) for 8 days (a
time point showing a high level of TUNEL-positive cells in this animal
model),29A and the vessels were analyzed
immediately thereafter. A trend toward a reduction in the intima-media
ratio and the number of cells in the intima was seen, but these
observations were not statistically significant (data not shown).
However, the apoptotic index of smooth muscle cells in the
intima as well as the media was significantly higher after NX1975
treatment compared with controls, as visualized by TUNEL staining
(Table 2
). Together, these findings
suggest that apoptotic cell death of smooth muscle cells is an
important mechanism in NX1975-mediated reduction of lesion formation in
this animal model.
|
Shortening of Treatment to 1 Week Reduces the
Inhibitory Effect of NX1975 on Intimal Hyperplasia
Formation
To investigate whether treatment with NX1975 during the first week
after injury was sufficient to obtain significant reduction in lesion
formation, a study was performed wherein animals received NX1975 (12 mg
kg-1 d-1) or carrier
for the first week and carrier for the second week after balloon
injury. The vessels were analyzed 2 weeks after the injury was
performed (Figure 1
B). Vessel wall areas, numbers of cells,
cellular density, or percentage of proliferating cells did not
significantly differ from animals treated with carrier only (Table 1
).
Two-Week Treatment With PDGF SELEX Aptamer Fails to Give Sustained
Effects on the Development of Intimal Hyperplasia
A critical issue that will determine the therapeutic efficacy of
PDGF antagonists for lesion formation is whether treatment
leads to a permanent reduction of intimal hyperplasia or rather simply
delays the process. Previous studies with PDGF antagonists
in the rat carotid artery model have used end points that coincided
with the end of treatment17 20 21 22 23 or when the
antagonist still exerted its beneficial
effect12 14 16 18 19 24 and have therefore not studied
this aspect of the effects of PDGF antagonists. To address
this issue, a study was performed in which NX1975 (12 mg
kg-1 d-1) was
administered during the first 2 weeks after injury, and the effects of
lesion formation were analyzed 6 weeks later.
As shown in Figures 1
C, 2D, and 2E, only a slight
reduction in intima-media ratio was observed at 8 weeks after injury
compared with control animals at the same time point, and the
difference was not statistically significant. Furthermore, in animals
that had received the PDGF antagonist for 2 weeks, the
intima-media ratio was 2- to 3-fold higher 8 weeks after injury
(P<0.001) compared with 2 weeks after injury (Figure 1
C). This indicates that 2-week treatment with PDGF B-chain
antagonists in this animal model fails to permanently block
the development of intimal hyperplasia.
As in the other experiments, lesions were further analyzed to
investigate the cellular processes underlying lesion formation (Table 1
). The increase in intima area (P=0.017) and
intima-media ratio (P=0.006) in control animals at 8 weeks
after injury, compared with that at 2 weeks after injury, was caused
predominantly by a statistically significant increase in area-cell
ratio (P=0.009) rather than an increase in cell number
(P=NS; Table 1
). This is presumably a consequence of
extracellular matrix deposition, also indicated by a more intense
collagen staining of the lesions (Figures 2
B and 2D). Also,
the statistically significant difference in cell number observed 2
weeks after injury between the PDGF antagonist and control
groups (P=0.016) did not remain after 8 weeks
(P=NS; Table 1
). Finally, a significant increase in
the intimal area-cell ratio (P=0.001) rather than cell
number (P=NS) was found when the 2-week and 8-week end
points in NX1975-treated animals were compared, indicating continued
extracellular matrix deposition by the cells in treated lesions after
removal of the inhibitor (Table 1
). Together these
observations suggest that the effect of the PDGF antagonist
on intima area, intima-media ratio, and intimal cell number at 2 weeks
is predominantly achieved through a transient delay of the lesion
formation process, and that this process recurs after removal of the
inhibitor.
| Discussion |
|---|
|
|
|---|
50% when the
compound is given throughout the study period of 2 weeks, a common end
point in this animal model. Aptamers typically interact with their
protein targets with nanomolar and picomolar affinities, and in this
regard, they are comparable to antibodies. Compared with antibodies,
however, the aptamers have a major advantage of not evoking any
detectable immune responses (D. Drolet et al, unpublished data,
1999). In the present study, full effect was observed with
as little as 2 mg kg-1d-1. Aptamers are thus attractive PDGF
antagonists with potential clinical interest.
Parenterally administered SELEX aptamers against
vascular endothelial growth factor and topoisomerase I
are currently being studied in clinical trials (data from Reference
29B ). By using a 10-times-higher dose in the first set of experiments,
we were unable to reach the effect reported in the study by Sirois et
al12 , in which perivascular delivery of 2 different PDGF
ß-receptor antisense oligonucleotides resulted in a
respective 60% and 80% inhibition of lesion formation in the same
animal model and time point. However, the level of reduction in our
results is consistent with several other studies in which PDGF
antibodies,17 22 PDGF ß-receptor
antibodies,15 16 18 and PDGF receptor kinase
inhibitors14 19 20 21 23 have been shown to
inhibit intimal hyperplasia by 35% to 50% in various models in rats,
pigs, and nonhuman primates. Our study shows that the PDGF B-chain
aptamer NX1975 can be used as an antagonist in vivo,
consistent with recent results by Floege et al,28
who demonstrated that an identical aptamer inhibited glomerulonephritis
in a rat model. At the time points of analysis, we found no significant effect of the PDGF antagonist on smooth muscle cell proliferation or cellular density. In contrast, a significant increase in apoptosis was seen in antagonist-treated animals. Our data thus support the notion that an important effect of PDGF in intimal hyperplasia formation is to prevent apoptosis of smooth muscle cells. Given the fact that PDGF-AB and -BB have a powerful chemotactic effect on smooth muscle cells and other cells in vitro30 31 and PDGF-BB an in vivo effect after its intravascular infusion,32 it is possible that NX1975 treatment also inhibits migration of smooth muscle cells from the media to the intima layer of the vessel wall. Interestingly, our results with 1-week treatment and a 2-week end point with a PDGF antagonist were similar to the data shown by Bendeck et al,33 who blocked migration by using a matrix metalloproteinase inhibitor. In their study, administration of the drug during the first 7 days after injury also failed to inhibit intimal thickening at 2 weeks and was associated with a trend toward increased smooth muscle cell proliferation.
An important finding in the present study was that the beneficial effect of the 2-week treatment was lost when administration was discontinued and the effect measured 6 weeks later. The extent of intimal hyperplasia in both untreated and treated animals was significantly greater at 8 weeks compared with 2 weeks after injury, largely as a consequence of continued matrix deposition. Moreover, treatment with aptamers for the first week after balloon dilatation is too short a duration to achieve a significant reduction of intimal thickening in this animal model at 2 weeks, consistent with reports showing an increased level of PDGF ß-receptor activation during the second week after injury.13 34 The recurrence of intimal hyperplasia after removal of PDGF inhibition, as shown in this communication in rats, may also explain results from a nonhuman primate study in which a short course of therapy with PDGF ß-receptor antibodies had only a limited effect on intimal hyperplasia at 30 days after injury.35 Our findings indicate that the durability of therapeutic benefit of various compounds against PDGF, an important molecular target in vascular pathology, needs to be taken into consideration in designing future studies in higher species, with models that better resemble human disease.
It is relevant to note in this context that Floege et al28 have recently observed better therapeutic effect with an identical aptamer in a rat mesangioproliferative glomerulonephritis model. Together with our observations, these results indicate a more central role for PDGF B-chain in the pathophysiology of kidney glomeruli compared with that of large blood vessels. Even though PDGF may have different functions during embryogenesis than in disease in adult animals,36 37 this notion is in agreement with knockout studies; in PDGF B-chain and ß-receptor knockout mice, mesangial cells (and the mesangium) are absent38 39 ; in contrast, although blood vessels are dilated in PDGF B-chain knockout mice, smooth muscle cell hypoplasia is not observed, and major veins and arteries appear normal in PDGF ß-receptor knockout animals.
Together, these results suggest that efficient inhibition of intimal hyperplasia formation by PDGF antagonists may require either prolonged treatment or even combination with other treatments. Interestingly, Rutherford et al22 found that the combination of PDGF antibodies and antibodies against bFGF caused dramatically decreased intimal hyperplasia at day 8 after injury, suggesting that both PDGF and bFGF have important roles in the early stages of this process. Whether intimal hyperplasia recurs after this combined intervention is not known. Furthermore, continued matrix deposition in the treated lesions at 8 weeks after injury suggests that inhibition of the agents that induce the synthesis of matrix components (eg, transforming growth factor-ß) may be useful adjunctive therapy to PDGF antagonists.40 It is also possible that combining PDGF antagonists with factors that promote endothelial cell coverage of the injury site41 will lead to permanent inhibition of intimal hyperplasia.
| Acknowledgments |
|---|
Received February 14, 2000; accepted August 3, 2000.
| References |
|---|
|
|
|---|
and -ß blockade on flow-induced
neointimal formation in endothelialized
baboon vascular grafts. Circ Res. 2000;86:77986.This article has been cited by other articles:
![]() |
K. Kappert, J. Sparwel, A. Sandin, A. Seiler, U. Siebolts, O. Leppanen, S. Rosenkranz, and A. Ostman Antioxidants Relieve Phosphatase Inhibition and Reduce PDGF Signaling in Cultured VSMCs and in Restenosis Arterioscler. Thromb. Vasc. Biol., December 1, 2006; 26(12): 2644 - 2651. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Uutela, J. Lauren, E. Bergsten, X. Li, N. Horelli-Kuitunen, U. Eriksson, and K. Alitalo Chromosomal Location, Exon Structure, and Vascular Expression Patterns of the Human PDGFC and PDGFD Genes Circulation, May 8, 2001; 103(18): 2242 - 2247. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2000 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |