Nature Biotechnology
19, 29 - 34 (2001)
doi:10.1038/83471
Local endostatin treatment of gliomas administered by microencapsulated
producer cellsTracy-Ann Read1, Dag R. Sorensen2, Rupavathana Mahesparan1, Per Ø. Enger1, Rupert Timpl3, Bjørn R. Olsen4, Mari H.B Hjelstuen5, Olav Haraldseth6
& Rolf Bjerkvig11 Department of Anatomy and Cell Biology, University
of Bergen, Norway. 2 Department of Comparative Medicine, The National Hospital,
University of Oslo, Norway. 3 Max-Planck-Institut für Biochemie,
Martinsried, Germany. 4 Harvard Medical School, Boston,
MA, USA. 5 SINTEF Unimed, MR Center, 7565
Trondheim, Norway. 6 Department of Anaesthesia and Medical Imaging, The
Norwegian University of Science and Technology, Trondheim,
Norway.
Correspondence should be addressed to Tracy-Ann Read Tracy-Ann.Read@PKI.UIB.NOWe describe a technique for the treatment of malignant brain tumors based
on local delivery of the anti-angiogenic protein endostatin from genetically
engineered cells encapsulated in ultrapure sodium alginate. Alginate consists
of L-guluronic and D-mannuronic acid, which in the presence of divalent cations
forms an extended gel network, in which cells reside and remain immunoisolated,
when implanted into the rat brain. Here, we show that endostatin-transfected
cells encapsulated in alginate maintain endostatin secretion for at least
four months after intracerebral implantation in rats. During the implantation
period 70% of the encapsulated cells remained viable, as opposed to 85% in
in vitro-cultured capsules. Rats that received transplants of BT4C glioma
cells, together with endostatin-producing capsules (0.2  g/ml per capsule),
survived 84% longer than the controls. The endostatin released from the capsules
led to an induction of apoptosis, hypoxia, and large necrotic avascular areas
within 77% of the treated tumors, whereas all the controls were negative.
The encapsulation technique may be used for many different cell lines engineered
to potentially interfere with the complex microenvironment in which tumor
and normal cells reside. The present work may thus provide the basis for new
therapeutic approaches toward brain tumors.
alginateencapsulationproducer cellsbrain tumorsAngiogenesis and blood−brain barrier dysfunction are characteristic
features of malignant brain tumors1, making them strong candidates
for anti-angiogenic therapy2. Several endogenous proteins have
been reported to have anti-angiogenic effects3,
4. One such
protein is endostatin, a 20 kDa proteolytic fragment of collagen XVIII that
has been shown to inhibit the growth of various ectopic experimental tumors
in mice5,
6,
7,
8. The results of such studies and most other
anti- angiogenic therapy studies are based on systemic delivery of the active
compound. However, because gliomas tend to recur at the resection site9, locally applied treatment may well yield better results than treatment
delivered systemically10,
11.
One way of administering tumor-controlling/suppressing agents locally is
via alginate bioreactors12. Such bioreactors are composed of
genetically engineered producer cells encapsulated in an immunoisolating substance
such as sodium alginate13,
14,
15. The cells may be engineered
to produce and secrete recombinant proteins that can interfere with several
of the biological pathways that are involved in glioma progression, including
angiogenesis, hence offering a delivery vehicle for specific and multifocal
therapy.
We have previously demonstrated that encapsulated producer cells survive
and maintain their transgene expression for more than four months within the
brain, inducing a minimum immune reaction toward the bioreactors (beads)16. Furthermore, it has been shown that recombinant proteins can successfully
be delivered to the brain by polymers, in which the spatial protein distribution
is expected to be between 1 and 2 mm from the implantation site in addition
to distant distribution through the perivascular and subarachnoid space17,
18,
19.
The following work aimed at studying the effects of endostatin-producing
bioreactors on intracerebral rat BT4C glioma growth. For this purpose human
fetal kidney 293 cells, transfected with a pCEP-Pu episomal expression vector
containing human endostatin (293-endo), were encapsulated in sodium alginate.
Western blotting and radioimmunoassays were performed to assess the release
of endostatin from the encapsulated cells. We injected BT4C glioma cells directly
into the right cerebral hemisphere of rats, together with the endostatin-producing
bioreactors. Treatment efficacy was assessed by determining the surviving
fraction and by visualizing tumor growth by high-resolution magnetic resonance
imaging (MRI) techniques.
The spatial distribution of free endostatin and its anti- angiogenic effects
were further assessed by immunohistochemistry, in which we sought to determine
blood vessel density, apoptosis, hypoxia, and vascular endothelial growth
factor (VEGF) expression, using both light and scanning confocal microscopy.
Results
293-endostatin producer cells encapsulated in alginate.
293-endo and 293-EBNA (Epstein−Barr virus nuclear antigen) cells
were evenly distributed within the beads on the day of encapsulation. During
two to three weeks in culture, spheroids developed gradually within the beads
(Fig. 1). Western blots of conditioned medium from encapsulated
293-endo cells showed a strong single band of  20 kDa (
Fig. 2A). In the cellular fraction, two weaker bands were seen at 20
and 23 kDa, indicating that most of the protein was released from the cells.
Radioimmunoassay of growth medium conditioned for 48 h by 10 alginate beads
containing 293-endo cells showed an increase in production from the beads
during the first three weeks (10 beads release 2  g/ml endostatin
during a 48 h period; Fig. 2B). During the period of
treatment, no antibodies against human endostatin were detected in serum samples
collected from the rats (data not shown).
Encapsulated 293-endo cells remain viable within the brain for prolonged
periods and reduce intracerebral BT4C tumor growth.
MRI scans
and macroscopic evaluation (10 treated and 10 control animals per experiment,
duplicate experiments) showed a reduced tumor volume in treated rats compared
to the controls (Fig. 3A−D). Furthermore, they
indicate a disturbance in vascular permeability, as seen by the presence of
contrast enhancement fluid beyond the tumor margin (Fig. 3D
). There was a significant difference in the survival times between
animals that received alginate beads containing 293-endo cells as compared
to those receiving encapsulated 293 mock-transfected cells (293-EBNA). The
rats in the treated group survived significantly longer (84%,
P < 0.01) than the controls, and some of the rats became long-term
survivors (Fig. 3E). Verification of endostatin biodistribution
was performed by western blot analysis of cerebrospinal fluid (CSF) samples
taken from endostatin-treated animals, which showed a single band of 20 kDa,
when incubated with anti-human endostatin antibody (Fig. 3F
). Immune response against human endostatin was not detected in the
immunoblots of rat serum samples (data not shown).
 | | |
A fluorescence viability assay revealed that an average of 70% ( 12.5
s.d.) of the encapsulated cells (beads ranging from 300 to 500 m in diameter)
remained viable after four months in vivo, indicated by green fluorescence
emitted from the intracellular esterase-converted calcein (
Fig. 4A, B). Encapsulated cells maintained in culture for four months
showed 85.7% (10.5 s.d.) viability.
Light microscopy of hematoxylin and eosin-stained cryosections showed that
77% of the tumors treated with encapsulated 293-endo cells contained large
central necrotic areas located in some animals at a distance to the implanted
alginate beads (Fig. 4C−E). The large central
necrosis constituted between 20 and 46% of the total tumor volume and was
collectively absent in the controls.
Encapsulated 293-endo cells reduce blood vessel formation and induce
apoptosis and hypoxia within the tumors.
Immunostaining with
anti-von Willebrand factor antibodies showed large avascular areas within
the tumors treated with encapsulated 293-endo cells, corresponding to the
necrotic areas seen on the histological sections (Fig. 5A).
Parallel sections stained for CD 31 (P-CAM) gave similar results (data not
shown). The vessels surrounding the necrotic areas appeared scarcer and generally
had smaller diameters than those seen in the control tumors. In the transition
area between necrotic and normal tumor tissue, a considerable number of apoptotic
cells were identified using terminal deoxynucleotidyltransferase (TdT)-mediated
dUTP nick end-labeling (TUNEL assay) (Fig. 5B). Flow-cytometric
DNA measurements confirmed an induction of apoptosis in the treated tumors
(Fig. 6B). Immunostaining for the hypoxic marker pimonidazole
revealed a strong induction of hypoxia within the apoptotic areas (
Fig. 5C). Fluorescence intensity plots showed that the pimonidazole
staining intensity was at least twice the background level. Tumors treated
with cell-free alginate beads or beads containing 293-EBNA cells showed minimal
apoptosis (Fig. 6A) and hypoxia. Immunostaining showed
VEGF to be mainly expressed in the tumor periphery, which is common for these
rapid-growing, low necrotic tumors. However, in the endostatin-treated tumors,
individual tumor cells expressing VEGF were also identified adjacent to the
necrotic/hypoxic areas within the tumor mass (Fig. 5D).
Immunostaining for endostatin showed local deposits of free endoststatin within
the necrotic/apoptotic areas as well as around tumor blood vessels (Fig. 5E). To determine that the free endostatin actually was
released from the beads and not as a part of collagen XVIII turnover, we also
utilized a polyclonal antibody recognizing human endostatin only.
Figure 5F shows an endostatin-producing alginate bead releasing human
endostatin into the tumor tissue. Endostatin could be seen at a distance of
1−1.7 mm around the bioreactors, in addition to distant deposits located
within the necrotic/apoptotic areas. Endostatin-positive cells were also observed
within the beads. Figure 5G shows a control tumor stained
with the human endostatin antibody, indicating absence of human endostatin.
| | Figure 5. Biological effects of local endostatin treatment on intracranial tumors.
| | |  | (A) Composite image of several confocal images obtained from an endostatin-treated
tumor (T) cryosectioned and stained with anti-von Willebrand factor (FITC-labeled
in green) and propidium iodide nuclear staining (in red). The section reveals
a large nonvascular area (N) corresponding to the necrotic area seen in the
histological sections (Fig. 4D). (B) Section in parallel
to (A), stained for apoptosis using the TUNEL assay, which identifies DNA
fragmentation in the transition zone between vascular and avascular tissue.
(C, D) Areas of (C) hypoxia (anti-pimonidazole staining) and (D) VEGF expression
(FITC-labeled in green), in central areas of the tumor, observed exclusively
in endostatin-treated tumors. A fluorescence intensity plot is provided over
the hypoxic area, represented by a blue line. (E) Parallel section to (A)
showing a three-dimensional reconstruction of 32 confocal sections stained
for endostatin using rabbit antiserum against human endostatin (FITC-labeled)
and propidium iodide nuclear staining. Endostatin is seen to be associated
with blood vessels (collagen XVIII) and appears as free protein in the tumor
tissue. (F) An alginate bioreactor releasing endostatin. The sections were
stained with anti-human endostatin antibody, revealing free human endostatin
in the tumor tissue and endostatin-positive 293 cells within the beads. (G)
A corresponding control tumor (treated with 293-EBNA encapsulated cells) showing
no human endostatin. Magnification (A) 120 , (B) 200, (C) 125,
(D) 480, (E−G) 180.
 Full Figure and legend (192K) |
| |
| | |
Encapsulated 293-endo cells show no effect on BT4C tumor cell growth
in vitro.
The BT4C monolayers that were exposed to the alginate
bioreactors containing 293-endo cells or 293-EBNA cells showed the same growth
as the controls, indicating no direct toxic effects of the 293-endo bioreactors
on the tumor cells (data not shown).
Discussion
At present, malignant brain tumors have a poor prognosis and are commonly
treated by surgery before radiotherapy and/or chemotherapy9.
Following treatment these tumors tend to recur at their primary site and show
severe phenotypic heterogeneity. Therefore, local multifocal targeting of
various tumor cell types and their numerous molecular pathways should be considered
as a possible therapeutic strategy.
In the present work, we have demonstrated that recombinant proteins with
anti-angiogenic potential can be continuously delivered to intracerebral tumors
by cells encapsulated in alginate. By using a two-color fluorescence viability
assay and immunohistochemical analyses, we show that alginate-encapsulated
endostatin-producing cells stay viable for at least four months following
intracerebral implantation and that they continue to produce endostatin after
encapsulation and implantation.
The data presented herein and data that we have published16,
19
show that an equilibrium between cell proliferation and death within the beads
is reached at roughly three weeks post encapsulation. At this point multicellular
spheroids can be observed accompanied by a gradual increase in protein production16,
19 (Fig. 2B). The stabilization of protein
production at three weeks post encapsulation probably reflects (as reported
in ref. 19) that the cells have successfully
adapted to the enforced environment, hence establishing normalization of transgene
expression and protein synthesis.
The viability of encapsulated cells (70% after four months in vivo)
is higher than that reported in other alginate encapsulation systems, which
also show reduced transgene expression over time20. Variations
in the type of alginate used, encapsulation procedures, cell lines, and stability
of the transgene expression may in part explain this discrepancy. In previous
studies the investigators used potassium alginate and sodium poly-L-lysine
alginate for encapsulation, whereas we used ultrapure sodium alginate, which
has an endotoxin level below 4 IU. Such ultrapure alginate gels have not previously
been used for encapsulation. Histological and immunohistochemical analyses
of the tumors showed that endostatin is indeed released from the alginate
bioreactors in vivo and asserts its biological effects in the animals,
at a distance to the cell implants as seen by large nonvascular necrotic areas
within the treated tumors (Fig. 4D, E). The spatial
distribution of endostatin in the brain/tumor tissue was identified at 1−1.7
mm from the bioreactors, with some distant deposits (4−6 mm), in addition
to being present in the CSF (Fig. 3F), supporting previous
studies17,
18,
19. The presence of endostatin in the CSF indicates
a possible combination of local as well as systemic administration of the
protein, since CSF is found in the ventricular systems and notably in the
perivascular space around major blood vessels. Indeed, CSF is present around
all arterioles and venules within the brain parenchyma. Therefore, if the
concentration gradient is right within the CSF, the active molecule may also
exert its action at some distance from the implantation site.
It is well known that interstitial pressure and concentration gradients
within the host tissue often are affected by an increasing tumor mass and
hence may affect convection and distribution of molecules therein21,
22.
Furthermore, local delivery of proteins can lead to concentration gradients
that may affect protein activity in the brain17,
18. It is therefore
possible that concentration gradients and the physiological imbalance caused
by intracerebral tumor growth may account for the distant deposits and effects
of endostatin as observed in the treated tumors. The apparent biological effects
of endostatin were seen whether tumor inoculation occurred simultaneously
with (as presented herein) or one week before bioreactor implantation (data
not shown). The exact mechanisms controlling the anti- angiogenic activity
of endostatin have not yet been established; however, several reports of its
inhibitory effects on the growth of various experimental tumors have been
published. It has for example been shown that endostatin may induce apoptosis
in endothelial cells and may not compete for fibroblast growth factor-2 or
VEGF binding in vitro23,
24,
25. In this study we observed
large central areas of necrosis, apoptosis, and hypoxia exclusively in endostatin-treated
tumors, indicating that this is a specific anti-tumor effect of endostatin
also in vivo. Furthermore, the in vitro data show that endostatin
has no direct effect on tumor cells, suggesting that the necrosis/apoptosis
of tumor cells is rather a secondary effect resulting from a reduced vascularization
of the tumor.
Recently, it has been shown that hypoxia, due for example to impaired tumor
microcirculation, may induce apoptosis in tumor cells26 and
is often associated with ischemic necrosis in glioblastomas27.
Indeed, anti-pimonidazole staining of the treated tumors revealed numerous
hypoxic areas close to the necrotic regions. We therefore hypothesize that
endostatin through its anti-angiogenic activity induces hypoxic conditions
in the tumor, which ultimately leads to apoptosis and tumor cell death (Figs 5C and 6B). This hypothesis is
further supported by the presence of individual VEGF-expressing cells (Fig. 5D) within the endostatin-induced hypoxic areas28.
Working from the notion that the highly invasive characteristic of brain
tumors predicts the failure of locally applied drugs29, anti-angiogenic
therapy of experimental brain tumors has mainly been applied systemically.
Recently, a brief communication by Lichtenbeld et al. described the
advantage of local anti-VEGF treatment of adenocarcinomas, grown in dorsal
skin-fold chambers in SCID mice30. The authors show that vessel
responses to locally applied low-dose anti-VEGF were as rapid as systemically
administrated high-dose (20-fold increase) responses. Bearing in mind that
many anti-angiogenic compounds exert their vascular effects on the abluminal
side of the endothelium, it makes sense to apply such compounds at an extra-
rather than intravascular site. The survival data and MRI scans presented
here suggest that this indeed is the case, because the animals treated with
endostatin-producing alginate beads lived significantly longer (84%) and had
generally smaller tumors than the controls (Fig. 3A−D
). In these experiments treatment and tumor inoculation occurred simultaneously,
thereby imitating a single cell invasion model, which would indeed be the
case if this treatment was applied as intended (following surgical debulking
of the tumor) in humans. However, the survival of the animals was not significantly
influenced when animals received delayed treatment (that is, one week after
tumor inoculation; data not shown). This may in part be explained by the rate
at which this tumor grows; untreated animals die after only a few weeks. The
rapid growth of the tumor may also explain the lack of antibody production
against human endostatin in the treated animals. The MRI scans also indicate
a disturbance in vascular permeability, because the contrast enhancement fluid
was seen beyond the tumor margin, which was verified by histological examination
of the scanned tumors (data not included). Current research is aimed at encapsulating
different cell lines that produce recombinant proteins that may interfere
with tumor growth and progression. In this way a "library" of
encapsulated cells can be made with the aim of tailoring local therapy to
specific biological parameters that regulate tumor growth. Encapsulation therapy
may have advantages over systemic injections in that it renders sustained,
local, multifocal therapy/control of the tumor.
Experimental protocol
Encapsulation of 293-endostatin producer cells in alginate.
Human fetal kidney 293 cells (293-EBNA) expressing the Epstein−Barr
virus nuclear antigen (EBNA)−1 were obtained from Invitrogen (Carlsbad,
CA). The cells were liposome transfected with the episomal expression vector
pCEP-Pu containing the gene encoding human endostatin25.
The transfected cells (293-endo) were grown to confluency in 175 cm
2 culture flasks (Nunc, Roskilde, Denmark) containing Dulbecco's modified
Eagle medium (DMEM; Bio Whittaker, Walkersville, MD) supplemented as described25. The mock-transfectants were generated by transfecting 293 cells
with the pCEP-Pu vector alone. The BT4C tumor cell line is an ethylnitrosourea-induced
rat glioma, established in syngeneic in BD-IX rats31. The cells
(passage 26) were grown in 80 cm2 culture flasks with complete
growth medium consisting of DMEM supplemented as previously recommended31.
The method of cell encapsulation has been described in detail elsewhere15,
16. Briefly, droplets of cells dispersed in 1.5 % sodium alginate
(2 107 cells/ml alginate) were released into a 0.1
M CaCl2 solution prepared from 0.13 M NaCl stock solution initiating
gel formation by cross-linking.
After gel formation, the alginate beads (300−500 m in diameter)
were washed three times in Dulbecco's PBS (DPBS, Sigma, St Louis, MO), and
once in the previously described growth medium. The encapsulated cells were
cultured in 175 cm2 culture flasks containing 50 ml growth
medium and kept in a standard tissue culture incubator at 37°C, 100% humidity,
95% air, and 5% CO2.
Quantification of endostatin release from encapsulated cells.
To determine whether endostatin was released from the beads, conditioned
medium from encapsulated endo-293 and 293-EBNA cells was collected and used
for standard SDS−PAGE western blotting32 (anti-endostatin
concentration 1:100).
Cell fractions were prepared according to standard procedures32.
Briefly, the alginate beads were dissolved in 8 ml sodium citrate solution
prepared from 4.41 g trisodium citrate dihydrate in 150 ml DMEM, and promptly
centrifuged for 4 min (1500 r.p.m.). The proteins were extracted from the
cell pellet by adding extraction buffer32.
The amount of endostatin released from 10 alginate beads over a 48 h incubation
period was determined by radioimmunoassay of 1 ml conditioned medium following
standard protocols33.
SDS−PAGE western blotting was also used to identify endostatin in
the CSF samples and for detection of human endostatin antibodies in serum
from treated and untreated animals following the procedures described by Takeoka
et al.34 and Visted et al.35. The primary
antibody (anti-endostatin concentration 1:50) was detected using the chemiluminescent
horseradish peroxidase (Pierce, Rockford, IL) technique36. The
horseradish peroxidase (DAKO, Glostrup, Denmark) was diluted 1:5,000 in blocking
buffer32.
Intracerebral implantation of encapsulated 293-endo cells.
Adult inbred BD-IX rats37 (200−250 g) of both sexes
were anesthetized by intraperitoneal injections of pentobarbital at a dosage
of 0.4 ml/100 g body weight. The rats were immobilized in a Kopf small animal
stereotaxic frame (David Kopf Instruments, Tujunga, CA), and the beads were
implanted as described16. Subsequently, 8 10
3 BT4C glioma cells were injected 1 mm lateral to the alginate beads
at a depth of 2 mm. The alginate beads contained endostatin-producing 293
cells, 293-mock transfected cells (293-EBNA), or no cells (clear) (13 recipient
animals for each type of bead). In addition, eight animals were injected with
BT4C cells alone and six animals received alginate beads containing 293-endo
cells alone and were used for evaluation of the in vivo viability of
the cells within the beads. These experiments were performed in duplicate,
six months apart. A separate group of animals consisting of three treated
animals and three controls (i.e., 293-EBNA cells in alginate) were used for
serum sampling in order to determine the presence of rat antibodies against
human endostatin.
During the experimental period the animals were housed in pairs at constant
temperature and humidity, fed a standard pellet diet, and provided tap water
at liberty.
The six animals that received alginate beads containing 293-endo cells
alone were allowed to live four months before they were killed by CO
2 inhalation.
The remaining animals were killed upon signs of pathology, at which time
six animals (three treated, three mock transfectant controls) received intraperitoneal
injections of 60 mg pimonidazole (Natural Pharmasia Inc., Baltimore, MD) for
detection of hypoxic tissue38, 1 h before euthanasia. These
animals were excluded from the survival study.
Samples of CSF were taken from the remaining animals by puncture of the
ventricle in the hemisphere contralateral to the implantations, using a self-designed
syringe.
Magnetic resonance imaging.
The size and localization
of the tumors at 17 and 28 days post surgery were detected in rats using contrast-enhanced
MRI. The MRI was performed with a 2.35 T Bruker Biospec (Bruker Medizinteknik,
Ettlingen, Germany) using a specially designed rat head coil, with an inner
diameter of 4.5 cm. The rats were anesthetized with 1−2% isoflurane
in 70%/30% N2/O2, delivered through a mask throughout
the entire MRI examination, and placed supine on a board with heated circulating
fluorocarbons to maintain the body temperature at 37°C. The MRI started 2
min after an intravenous bolus injection of 0.5 mmol/kg body weight of the
contrast agent gadodiamide (OMNISCAN, Nycomed Amersham, Little Chalfont, UK).
We acquired 10 successive transaxial slices through the rat head, covering
the whole brain between the olfactory bulb and cerebellum. The parameters
for the T1-weighted spin-echo pulse sequence were TE/TR 13 ms/400 ms, acquisition
matrix 256 192, field of view 35 mm, slice thickness 1 mm, slice separation
0.2 mm, and number of excitations 4.
In vivo viability of alginate-encapsulated cells.
The viability of the cells within the alginate beads was investigated using
a two-color fluorescence viability assay (Live/Dead TM Viability/Cytotoxity
Assay, Molecular Probes, Eugene, OR) according to the protocol provided by
the manufacturer. Fluorescence was measured in optical sections through the
alginate using a confocal scanning laser microscope with a krypton-argon laser
(Leica TCS-NT, Heidelberg, Germany), using tetramethylrhodamine (TRITC) and
fluorescein isothiocyanate (FITC) filter optics. Fluorescence was recorded
in a plane between 70 and 120 m inside the beads. In each experiment six
beads were retrieved from the animals (n = 6) and
compared with six beads that had been in culture for four months. The experiments
were performed in duplicate.
Histology.
After dissection, the brains were either
mounted on stubs, embedded in Tissue-Tek (Miles Laboratories Inc., Naperville,
IL), and frozen in 2-methylbutane (E. Merck, Darmstadt, Germany) cooled with
liquid N2, or fixed in 4% formaldehyde for paraffin embedding.
Serial axial sections (10 m and 60 m) were cut and mounted on slides
precoated with poly-L-lysine (Sigma). The sections were stained with Harris
hematoxylin and eosin G (H&E; Merck) according to standard procedures,
mounted in Entellan (Merck), and examined with a Nikon Diaphot light microscope.
The necrosis seen in the treated tumors was estimated as percentage of total
tumor volume by measuring the two largest diameters of the tumor and central
necrosis seen in the serial histological sections.
Assessment of blood vessel formation, apoptosis, and hypoxia within
the tumors.
The sections were analyzed with regards to VEGF expression,
endostatin distribution, apoptosis (TUNEL assay), hypoxia, and blood vessel
density (von Willebrand factor and CD31). The immunostaining was performed
according to standard procedures39. The following antibodies
were used (1:100 dilution in DPBS): anti-VEGF (Santa Cruz Biotechnology, Santa
Cruz, CA), anti-von Willebrand factor (DAKO), anti-CD31 (Becton Dickinson,
Heidelberg, Germany), anti-pimonidazole (hybridoma extract containing mouse
mAb (clone 4.3.11.3) against pimonidazole adducts in hypoxic tissue (Natural
Pharmisia, Inc.), Rabbit antiserum against human endostatin and anti-human
endostatin, polyclonal (Chemicon International, Temecula, CA). Corresponding
FITC-conjugated secondary antibodies (DAKO) were used for primary antibody
detection (1:30 dilution in DPBS). For nuclear staining the sections were
treated with ribonuclease (Sigma; 1 mg/ml in 10% DPBS) followed by a short
exposure to propidium iodide (Sigma; 50 g/ml in DPBS). Finally the sections
were washed in DPBS and mounted with Vectashield (Vector Laboratories Inc,
Burlingame, CA).
Staining for apoptosis was performed using the TUNEL assay kit and following
the protocol supplied by the manufacturer (Boehringer Mannheim, Mannheim,
Germany). All sections were viewed and evaluated using the confocal laser
scanning microscope (Leica) applying TRITC and FITC filter optics.
Flow cytometry.
Tumors (three treated and three controls)
were dissected out of the rat brains, cut into fine pieces, and washed in
PBS. Thereafter they were fixed in 100% ice-cold ethanol and incubated in
0.5% (wt/vol) pepsin/HCl (wt/vol) for 15 min at 37°C. The isolated nuclei
were washed in 0.9% (wt/vol) NaCl, filtered and incubated with ribonuclease
(1 mg/ml in 0.9% NaCl) for 5 min. DNA staining was obtained by propidium iodide
staining (50 g/ml). DNA histograms distinguishing ploidy were established
by running the samples on a FACSORT flow cytometer (Becton Dickinson, Palo
Alto, CA).
Effect of endostatin on BT4C tumor cells in vitro.
In order to evaluate whether endostatin had any cytotoxic effect on the
BT4C tumor cells, monolayers were exposed to endostatin released from alginate
beads. BT4C cells were seeded in to 24-well culture plates (8,000 cells/well,
five parallels at each time point with five time points) and were allowed
to adhere before being exposed to either clear alginate beads or beads containing
endo-293 or 293-EBNA cells. The tumor cells were counted on day 2, 4, 6, 8,
and 10 using a Coulter counter (Coulter Electronics Ltd, Herpenden Herts,
England), and growth curves were established (data not shown).
Received 6 June 2000; Accepted 18 September 2000
REFERENCES
- Scherer, H.J. Structural development in gliomas. Am. J. Cancer 34, 333–351 (1938).
- Plate, K.H. & Risau, W. Angiogenesis in malignant gliomas. Glia 15, 339–347 (1995). | PubMed | ISI | ChemPort |
- Cao, R. et al. Suppression of angiogenesis and tumor growth by the inhibitor K1-5 generated by plasmin-mediated proteolysis. Proc. Natl. Acad. Sci. USA 96, 5728–5733 (1999). | Article | PubMed | ChemPort |
- Chung, J., Gao, A.G. & Frazier, W.A. Thrombospondin acts via integrin-associated protein to activate the platelet integrin alphaIIbbeta3. J. Biol. Chem. 272, 14740–14746 (1997). | Article | PubMed | ISI | ChemPort |
- O'Reilly, M.S. et al. Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 88, 277–285 (1997). | Article | PubMed | ChemPort |
- Chen. Q.R., Kumar, D., Stass, S.A. & Mixson, A.J. Liposomes complexed to plasmids encoding angiostatin and endostatin inhibit breast cancer in nude mice. Cancer Res. 59, 3308–3312 (1999). | PubMed | ISI | ChemPort |
- Dhanabal, M. et al. Endostatin: yeast production, mutants, and antitumor effect in renal cell carcinoma. Cancer Res. 59, 189–197 (1999). | PubMed | ISI | ChemPort |
- Blezinger, P. et al. Systemic inhibition of tumor growth and tumor metastases by intramuscular administration of the endostatin gene. Nat. Biotechnol. 17, 343–348 (1999). | Article | PubMed | ISI | ChemPort |
- Kleihues, P., Burger, P.C., Plate, K.H., Ohgaki, H. & Cavanee, W.K. Glioblastoma. In Pathology and genetics of tumours of the central nervous system. (eds Kleihues, P. & Cavanee, W.K.) 16–24 (The International Agency for Research on Cancer, Lyons, France; 1997).
- Olivi, A., DiMeco, F., Bohan, E. & Brem H. Developing new methods for the treatment of malignant brain tumours: local delivery of anti-neoplastic agents using biodegradable polymers. Forum (Genova) 10, 152–165 (2000). | PubMed | ChemPort |
- Hottinger, A.F. & Aebischer. P. Treatment of diseases of the central nervous system using encapsulated cells. Adv. Tech. Stand. Neurosurg. 25, 3–20 (1999). | PubMed | ChemPort |
- Lang, M.S., Hovenkamp, E., Savelkoul, H.F.J., Knegt, P. & van Ewijk, W. Immunotherapy with monoclonal antibodies directed against the immunosuppressive domain of p15E inhibits tumour growth. Clin. Exp. Immunol. 102, 468–475 (1995). | PubMed | ISI | ChemPort |
- Lim, F. & Sun, A.M. Microencapsulated islets as bioartificial endocrine pancreas. Science 210, 908–910 (1980). | PubMed | ISI | ChemPort |
- Martinsen, A., Skjåk-Bræk, G. & Smidsrød, O. Alginate as immobilization material: I. Correlation between chemical and physical properties of alginate gel beads. Biotechnol. Bioeng. 33, 79–89 (1989). | ISI | ChemPort |
- Martinsen, A., Storrø, I. & Skjåk-Bræk, G. Alginate as immobilization material: III. Diffusional properties. Biotechnol. Bioeng. 39, 186–194 (1992). | ISI | ChemPort |
- Read, T-A. et al. Cells encapsulated in alginate: a potential system for delivery of recombinant proteins to malignant brain tumours. Int. J. Dev. Neurosci. 17, 653–663 (1999). | Article | PubMed | ISI | ChemPort |
- Krewson, C.E. & Saltzman, W.M. Transport and elimination of recombinant human NGF during long-term delivery to the brain. Brain Res. 15, 169–181 (1996). | Article |
- Mahoney, M.J. & Saltzman, W.M. Millimeter-scale positioning of a nerve-growth-factor source and biological activity in the brain. Proc. Natl. Acad. Sci. USA 13, 4536–4539 (1999). | Article |
- Thorsen, F., Read, T-A., Lund-Johansen, M, Tyssnes, B.B. & Bjerkvig, R. Alginate encapsulated producer cells: a potential new approach to the treatment of malignant brain tumours. Cell Transplant. (December 2000). | PubMed |
- Ross, C.J., Ralph, M. & Chang, P.L. Delivery of recombinant gene products to the central nervous system with nonautologous cells in alginate microcapsules. Hum. Gene Ther. 10, 49–59 (1999). | Article | PubMed | ISI | ChemPort |
- Boucher, Y., Salehi, H., Witwer, B., Harsh, G.R & Jain, R.K. Interstitial fluid pressure in intracranial tumours in patients and in rodents. Br. J. Cancer 75, 829–836 (1997). | PubMed | ISI | ChemPort |
- Steen, R.G., Kromhout-Schiro, S. & Graham, M.M. Relationship of perfusion to edema in the 9L glioma. J. Neurooncol. 16, 81–87 (1993). | PubMed | ChemPort |
- Dhanabal, M. et al. Endostatin induces endothelial cell apoptosis. J. Biol. Chem. 274, 11721–11726 (1999). | Article | PubMed | ISI | ChemPort |
- Chang, Z., Choon, A. & Friedl, A. Endostatin binds to blood vessels in situ independent of heparan sulfate and does not compete for fibroblast growth factor-2 binding. Am. J. Pathol. 155, 71–76 (1999). | PubMed | ISI | ChemPort |
- Yamaguchi, N. et al. Endostatin inhibits VEGF-induced endothelial cell migration and tumour growth independently of zinc binding. EMBO J. 18, 4414–4423 (1999). | Article | PubMed | ISI | ChemPort |
- Yao, K.S., Clayton, M. & O'Dwyer, P. Apoptosis in human adenocarcinoma HT29 cells induced by exposure to hypoxia. J. Natl. Cancer Inst. 87, 117–122 (1995). | PubMed | ChemPort |
- Tohma, Y. et al. Necrogenesis and Fas/APO-1 (CD95) expression in primary (de novo) and secondary glioblastomas. Neuropathol. Exp. Neurol. 57, 239–245 (1998). | ChemPort |
- Plate, K.H., Breier, G., Weich, H.A. & Risau, W. Vascular endothelial growth factor is a potential tumor angiogenesis factor in human gliomas in vivo. Nature 359, 845–848 (1992). | Article | PubMed | ISI | ChemPort |
- Goldbrunner, R.H., Bernstein, J.J. & Tonn, J.C. Cell-extracellular matrix interaction in glioma invasion. Acta Neurochir. (Wien) 141, 295–305 (1999). | Article | PubMed | ChemPort |
- Lichtenbeld, H.C., Ferarra, N., Jain, R.K. & Munn, L.L. Effect of local anti-VEGF antibody treatment on tumor microvessel permeability. Microvasc. Res. 57, 357–362 (1999). | Article | PubMed | ISI | ChemPort |
- Laerum, O.D. & Rajewsky, M.F. Neoplastic transformation of fetal rat brain cells in culture after exposure to ethylnitrosourea in vivo. J. Natl. Cancer Inst. 55, 1177–1187 (1975). | PubMed | ChemPort |
- Blake, M.S., Johnston, K.H., Russell-Jones, G.J.& Gotschlich, E.C. A rapid, sensitive method for detection of alkaline phosphatase-conjugated anti-antibody on Western blots. Anal. Biochem. 136, 175–179 (1984). | PubMed | ISI | ChemPort |
- Sasaki, T. et al. Structure, function and tissue forms of the C-terminal globular domain of collagen XVIII containing the angiogenesis inhibitor endostatin. EMBO J. 17, 4249–4256 (1998). | Article | PubMed | ISI | ChemPort |
- Takeoka, T., Shinohara, Y., Furumi, K. & Mori, K. Characteristic protein fractions of cerebrospinal fluid disc electrophoretic analysis. Brain Res. 29, 147–150 (1980). | Article |
- Visted, T. et al. acZ-neoR transfected glioma cells in syngeneic rats: growth pattern and characterization of the host immune response against cells transplanted inside and outside the CNS. Int. J. Cancer 15, 228–235 (2000). | Article |
- Walker, G.R., Feather, K.D., Davis, P.D. & Hines, K.K. SuperSignalMT CL-HPR: a new enhanced chemiluminescent substrate for the development of horseradish peroxidase label in western blotting applications. J. NIH Res. 7, 76 (1995).
- Druckrey, H. Genotypes and phenotypes of ten inbred strains of BD-rats. Arzneimittelforschung 21, 1274–1278 (1971). | PubMed | ChemPort |
- Varia, M.A. et al. Pimonidazole: a novel hypoxia marker for complementary study of tumour hypoxia and cell proliferation in cervical carcinoma. Gynecol. Oncol. 71, 270–277 (1998). | Article | PubMed | ISI | ChemPort |
- Mahesparan, R. et al. Extracellular matrix-induced cell migration from glioblastoma specimens in vitro. Acta Neuropathol. 97, 231–239 (1999). | Article | PubMed | ISI | ChemPort |
Acknowledgments
We thank the Norwegian Cancer Society, the National Gene Therapy Program,
the Norwegian Research Counsel and Innovest, and the University of Bergen
for financial support toward this study. Furthermore, we thank Bodil Hansen,
Tove Drange Johannsen, and Tore Jacob Raa for excellent technical assistance.
Finally we thank Dr. Rupert Timpl at the Max-Planck institute, Martinsried,
Germany for supplying the endostatin antiserum.
|
