| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Cutaneous Biology Research Center, Department of
Dermatology,*
Massachusetts General Hospital and Harvard
Medical School, Charlestown; and the Department of
Pathology,
Beth Israel Deaconess Medical
Center and Harvard Medical School, Boston, Massachusetts
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In contrast, much less is known about the expression and biological role of endogenous inhibitors of angiogenesis during carcinogenesis. Several naturally occurring angiogenesis inhibitors have been identified, including thrombospondin-1 (TSP-1),6 TSP-2,7 angiostatin,8 and endostatin.9 TSP-1 is a 420-kd homotrimeric matricellular glycoprotein that regulates attachment, proliferation, migration, and differentiation of various cell types (for review, see Ref. 10 ). TSP-1 inhibits proliferation and migration of vascular endothelial cells in vitro and inhibits neovascularization in vivo, contributing to the normal quiescence of the vasculature.11 However, controversial data have been reported regarding the role of TSP-1 in epithelial tumor growth and metastasis. TSP-1 protein expression was shown to be inversely correlated to cellular differentiation in several squamous cell carcinoma (SCC) cell lines,12 and was shown to induce SCC proliferation, adhesion, migration, and invasion of cells in vitro.13-15 Enhancement of in vitro tumor cell invasion by TSP-1 has also been reported for breast, lung, and pancreatic carcinoma cell lines.16-19 Based on the observation that antisense inhibition of TSP-1 in SCC resulted in suppression of tumor growth in vivo,20 it was suggested that TSP-1 may promote tumor growth.21 In contrast, other studies reported that TSP-1 expression was inversely correlated with malignant progression in human lung, breast, and bladder carcinoma cell lines.22,23
In human skin, TSP-1 is deposited in the basement membrane,24 contributing to the antiangiogenic barrier that separates the avascular epidermis from the vascularized dermis. Recently, we found that TSP-1 expression was down-regulated in squamous cell carcinomas (SCC) of the skin, richly vascularized malignant tumors derived from epidermal keratinocytes (Detmar M, Velasco P, Tognazzi K, Brown LF, unpublished data). Based on these findings, we studied the biological role of TSP-1 for cutaneous carcinoma growth, using stable tumor cell transfectants in an intradermal xenograft model. Here, we report that TSP-1 overexpression reduced intradermal tumor growth of A431 epidermoid carcinoma cells and completely inhibited intradermal tumor formation of SCC-13 squamous cell carcinomas, although no direct effect on tumor cell proliferation was detected in vitro and in vivo. TSP-1 overexpressing A431 tumors were characterized by extensive areas of necrosis and by decreased tumor vessel numbers and sizes. TSP-1 overexpression also up-regulated the TSP-1 receptor CD36, leading to enhanced A431 cell adhesion to TSP-1. These findings establish TSP-1 as a potent inhibitor of angiogenesis and growth of cutaneous squamous cell carcinomas.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human epidermoid carcinoma cell line A431 was obtained from the American Type Culture Collection (Rockville, MD) and was maintained in Dulbecco's minimal essential medium (DMEM; Gibco BRL, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS) and 1% L-glutamine (all purchased from Gibco BRL).
The human cutaneous squamous carcinoma cell line SCC-13,25 kindly provided by Dr. James Rheinwald (Harvard Medical School, Boston), was maintained in DMEM/Ham's F-12 medium (3:1) supplemented with 5% FBS, 1% L-glutamine, 50 µg/ml insulin, 0.1 U/ml epidermal growth factor, 18.2 mg/ml adenine, and 0.4 µg/ml hydrocortisone (all purchased from Gibco BRL). Human dermal microvascular endothelial cells (HDMEC) were isolated from neonatal foreskins and cultivated as recently described.26
Cell Transfection and Selection
A 3.643-kb human TSP-1 cDNA sequence, comprising the full TSP-1 coding sequence (nucleotides 67–3689 of the human TSP-1 sequence; GenBank accession code X04665), was kindly provided by Dr. Luisa Iruela-Arispe, University of California, Los Angeles. The TSP-1 cDNA was cloned into a pGEM7Z vector (Promega, Madison, WI). The sequence was verified by restriction mapping and by direct sequencing using the Sanger dideoxy method. After restriction digestion with SacI, SphI, and PvuI (Gibco BRL), a 3.6-kb fragment was gel purified, blunted, and ligated into a pcDNA3.1Zeo(-) expression vector (Invitrogen, San Diego, CA) which contains a CMV-enhancer-promoter and a Zeocin selection cassette. Subconfluent cultures were stably transfected either with pcDNA3.1Zeo(-) vector containing the full-length human TSP-1 cDNA or with pcDNA3.1Zeo(-) vector alone using the SuperFect transfection reagent (Qiagen, Chatsworth, CA) according to the manufacturer's protocol. Forty-eight hours after transfection, cells were split 1:3 into their full growth medium containing 250 µg/ml Zeocin (Invitrogen) to select transfectants. Stably transfected clones were expanded, and 10 clones were characterized for TSP-1 mRNA and protein expression.
RNA Isolation and Northern Blot Analysis
Total cellular RNA was isolated from stable transfectants and from intradermal tumors using the RNeasy kit (Qiagen), according to the manufacturer's instructions. The isolated RNA was subjected to electrophoresis and transferred to Biotrans nylon supported membranes (ICN Pharmaceuticals, Costa Mesa, CA). 32P-radiolabeled cDNA probes were prepared with a random primed synthesis kit (Multiprime; Amersham, Arlington Heights, IL). We used a 4.1-kb TSP-1 cDNA probe and a 300-bp human VEGF cDNA probe which recognizes all known VEGF variants. A 2.0-kb human ß-actin cDNA probe purchased from Clontech (Palo Alto, CA) was used as a control for equal RNA loading. Blots were washed at high stringency as described27 and exposed to X-OMAT MR film (Kodak, Rochester, NY) or a PhosphorImager screen (Molecular Dynamics, Sunnyvale, CA). mRNA expression was quantitated with a Molecular Dynamics scanning densitometer using the ImageQuant software.
Western Blot Analysis
Western Blot analyses were performed on cell lysates and conditioned media from stably transfected A431 cells, SCC-13, and HDMEC. Cells were grown to confluence in 100-mm dishes, washed with phosphate buffered saline (PBS), and lysed as described.28 Cell lysates were homogenized using a cell shredder (Qiagen), and protein concentrations were determined using the Bio-Rad protein assay (Bio-Rad, Hercules, CA). Conditioned media were obtained from confluent cells grown for 48 hours in serum-free culture medium. TSP-1 was concentrated using heparin beads (Sigma, St. Louis, MO). All samples were boiled in denaturating sample buffer, and equal amounts according to the protein assay were electrophoresed on polyacrylamide gels under reducing conditions.29 Proteins were blotted onto polyvinylidene difluoride membranes (Bio-Rad). To verify equal protein loading membranes were stained with 0.1% Ponceau red (Sigma) diluted in 5% acetic acid. Membranes were incubated overnight in PBS containing 0.1% Tween-20 and 3% bovine serum albumin to block nonspecific binding. Membranes were then incubated with primary antibodies directed against human TSP-1 (clone 133; Genzyme, Cambridge, MA), human CD36 (clone 1A7; NeoMarkers, Fremont, CA), or human VEGF (clone 2352, kindly provided by Dr. Don Senger, Beth Israel Deaconess Medical Center, Boston, MA), washed in PBS/Tween, incubated with horseradish peroxidase-conjugated anti-mouse IgG (Amersham), and analyzed by the enhanced-chemiluminescence system (Amersham). Protein expression was quantitated with a Molecular Dynamics scanning densitometer using the ImageQuant software.
Cell Growth and Apoptosis Assays
To determine whether TSP-1 overexpression influences tumor cell proliferation, we measured anchorage-dependent and -independent cell growth rates. 1 x 105 A431 cells were plated in duplicate into 100-mm culture dishes and total cell numbers per dish were determined after 2, 4, and 6 days using a hemocytometer. Anchorage-independent growth rates were determined by using a soft-agar assay as described.30 Ten thousand control transfected or TSP-1 transfected A431 cells were transferred into six 30-mm cell culture dishes with a 2-mm grid (Nunc, Naperville, IL). The dishes were incubated at 37°C and 5% CO2, and colonies were counted after 8 days. The effects of conditioned media from TSP-1 transfectants or from vector transfected controls on endothelial cell proliferation were determined using the BrdU labeling and detection kit (Boehringer, Mannheim, Germany). Human dermal microvascular endothelial cells were grown in 24-well plates in the presence of cell culture medium or conditioned media, supplemented with 10% FBS for 24 hours. HDMEC were also co-incubated with either TSP-1 (25 µg/ml) and/or a TSP-1 neutralizing antibody (50 µg/ml, clone AB-1, Neomarkers) or an isotype-specific control antibody (clone MOPC 21, Sigma). Cell proliferation was assayed according to the manufacturer's instructions. The absorbance was determined at 405 nm using a microtiter plate reader (Titertek, Huntsville, AL). The results represent the mean values ± SD of four dishes per group. Apoptosis induced by serum withdrawal was studied in subconfluent A431 cell clones after 6 days in serum-free medium. The percentage of apoptotic cells was determined as described,31 using the Fluorescein-FragEL DNA fragmentation kit (Oncogene, Cambridge, MA) according to the manufacturer's instructions, and a Beckton-Dickinson FACS-Scan (Franklin Lanes, NJ).
Cell Adhesion Assay
Twenty-four-well plates were coated with 20 µg/ml human collagen type I (Sigma) or with 20 µg/ml platelet-derived human TSP-1 (Sigma) for 1 hour at 4°C, followed by 100 mg/ml bovine serum albumin (Sigma) to block remaining protein binding sites. A431 cells (1 x 106 cells/ml) in serum-free DMEM medium (Gibco BRL) were mixed with a blocking mouse monoclonal antibody to human CD36 (Neomarkers) or with mouse control IgG (Sigma) at 5 µg/ml. One hundred µl of the cell suspensions were added to each well and were incubated at 37°C for 60 minutes. Unattached cells were removed by three gentle washes with PBS. Attached cells were fixed with 0.1% glutaraldehyde (Sigma), stained with a 0.1% crystal violet solution (Sigma), and washed three times with PBS. Crystal violet adsorbed onto cells was solubilized with 0.2% Triton X-100 (Sigma), and absorbance was measured at 590 nm in a Titertek microtiter plate reader. The absorbance of 1 x 105 fixed cells served as control (100% value).32 The unpaired Student's t-test was used for statistical analysis of the results.
Growth of Xenografts in Nude Mice
Confluent A431 or SCC-13 cells, stably transfected with a human
TSP-1 expression vector or with the expression vector alone, were
trypsinized and resuspended in serum-free DMEM medium (Gibco BRL) at a
density of 2 x 107
cells/ml. Two million tumor cells
of each type were injected intradermally into both flanks of five
8-week-old female BALB/c (nu/nu) mice. Two control clones and 3 TSP-1
overexpressing clones were investigated. The smallest and largest tumor
diameter were measured weekly, using a digital caliper, and tumor
volumes were calculated using the following formula:
 |
In Situ Hybridization and Immunohistochemistry
In situ hybridization was performed on 6-µm paraffin sections of tumor xenografts as described.27 The sense and antisense single-stranded RNA probes for human VEGF were transcribed from a pGEM-3Zf(+) vector containing a 204-bp polymerase chain reaction fragment common to all known VEGF splicing variants. A RNA probe to human TSP-1 was transcribed from a pBluescript II KS+ vector containing a 240-bp polymerase chain reaction fragment of the coding region of human TSP-1. Immunohistochemical staining was performed on 6-µm frozen or paraffin sections of tumor xenotransplants as previously described,33 using monoclonal antibodies against mouse CD31 (Pharmingen, San Diego, CA), human TSP-1 (Genzyme), human CD36 (Neomarkers), and human PCNA antigen (Zymed Laboratories, San Francisco, CA).
Computer-Assisted Morphometric Analysis of Tumor Vessels
To determine the degree of tumor-induced angiogenesis, cryostat sections of tumor xenografts were stained with an anti-mouse CD31 monoclonal antibody. Representative sections obtained from five tumors from each cell clone were analyzed, using a Nikon E-600 microscope (Nikon, Melville, NY). Images were captured with a Spot digital camera (Diagnostic Instruments, Sterling Heights, MI), and morphometric analyses were performed using the IP LAB software program (Scanalytics Inc.; Fairfax, VA). Three different fields at 60x magnification were examined on each section, and the number of vessels per mm2, the size distribution, the average, and the total area of tumor blood vessels were determined. The unpaired t-test was used to analyze differences in the vessel areas.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human squamous cell carcinoma cell lines A431 and SCC-13 were
chosen with respect to their endogenous TSP-1 and VEGF secretion and
their in vivo growth characteristics. A431 cells are
characterized by strong secretion of VEGF but little or no TSP-1
secretion (Figure 1)
, and form fast
growing and highly vascularized tumors in
vivo.34
SCC-13 cells express low levels of VEGF but
synthesize substantial amounts of TSP-1 (Figure 1)
, and grow slowly as
well differentiated, rounded tumors. The TSP-1 expression levels of
multiple TSP-1 and control transfected clones were determined by
Northern Blot analyses (Figure 1A)
. The highest levels of TSP-1 mRNA
were detected in A431 clones 10, 12, and 19 and SCC-13 clones 1, 2, and
3. Western blot analyses confirmed that increased TSP-1 mRNA levels
correlated with increased amounts of TSP-1 protein. In TSP-1
transfected A431 cell clones, strong expression of the 180 kd TSP-1
protein was found in cell lysates and in culture supernatants,
confirming efficient secretion of TSP-1 (Figure 1B)
. In contrast,
little or no TSP-1 protein was detected in A431 cells transfected with
vector only. In SCC-13 cell clones, TSP-1 was already expressed by
control transfectants (Figure 1B)
; three TSP-1 transfected clones
showed markedly increased TSP-1 secretion although the amount of
cell-associated TSP-1 was not significantly increased. TSP-1 expression
was also detected in cell lysates but not in conditioned medium
harvested from human dermal microvascular endothelial cells (Figure 1B)
. The biological activity of transfected human TSP-1 was confirmed
in HDMEC proliferation assays. HDMEC proliferation was significantly
inhibited after a 24 hour incubation with conditioned media harvested
from TSP-1 overexpressing A431 and SCC-13 cell clones, as compared to
HDMEC incubated with unconditioned media (Figure 1, C and E)
.
Supernatants obtained from control transfected A431 clones stimulated
HDMEC proliferation. Incubation with a TSP-1 neutralizing antibody but
not with an isotype-control antibody abolished the inhibition of HDMEC
proliferation by culture medium supplemented with TSP-1 and by
conditioned medium obtained from TSP-1 overexpressing cell clones (1D).
|
Our results showed no differences in anchorage-dependent cell
proliferation between TSP-1 overexpressing A431 clones and control
clones growing in plastic culture dishes (Figure 2A)
. Anchorage-independent cell growth
was studied by determination of colony numbers in a soft agar assay. No
significant differences in the number of colonies were observed between
the different cell clones (Figure 2B)
. To compare the susceptibility to
induction of apoptosis by serum withdrawal, A431 clones were cultured
under serum-free conditions for 6 days. No significant differences in
the percentage of apoptotic cells were found between TSP-1
overexpressing A431 clones (13.67% ± 4.04%) and control A431 clones
(12.33% ± 1.45%). To determine the biological effects of TSP-1
overexpression on the orthotopic tumor growth of A431 and SCC-13 cells
in vivo, tumor cells were injected intradermally into the
flanks of immunodeficient nude mice. Control transfected A431 cell
clones formed rapidly growing tumors (Figure 2E)
, reaching a volume of
2000–3000 mm3
after 4 weeks. Stable overexpression of
TSP-1 resulted in a significant inhibition of tumor growth by 50 to
75% after 4 weeks, as compared to control tumors (Figure 2E)
. Northern
Blot analysis of RNA extracted from two representative tumors of each
A431 cell clone confirmed that TSP-1 transfected tumor cell clones
maintained TSP-1 mRNA expression in vivo (data not shown),
with the highest TSP-1 mRNA expression and the slowest tumor growth in
A431 clone T10. Northern analysis of the same RNA samples demonstrated
equal levels of VEGF mRNA expression in TSP-1 overexpressing and
control tumors. The effects of transfected TSP-1 on in vivo
tumor growth were confirmed in transfected SCC-13 cell clones. After 2
weeks, control transfected SCC-13 clones formed slowly growing
intradermal tumors (Figure 2F)
. In contrast, TSP-1 overexpression in
SCC-13 cells led to a complete inhibition of in vivo tumor
growth in all clones tested (clones T2 and T3) over an observation
period of up to 12 weeks (Figure 2F)
.
|
Extensive areas of necrosis were detected in TSP-1 overexpressing
tumors (see Figure 4B
), whereas only occasional small necrotic foci
were found in control tumors (see Figure 4A
). However, the fraction of
proliferating cells within the viable tumor areas, as determined by
staining for the PCNA antigen, was unchanged (data not shown). The
expression and distribution of TSP-1 within A431 tumors was assessed by
in situ hybridization and by immunohistochemistry. Only weak
TSP-1 mRNA expression was detected in control tumor cells (Figure 3, A and B)
, and TSP-1 protein expression
was predominantly found in the dermal-epidermal basement membrane zone
of adjacent normal skin and in blood vessels, but not in tumor cells
(see Figure 5A
). In contrast, strong TSP-1 mRNA expression was detected
in TSP-1 overexpressing tumor cells (Figure 3, C and D)
, and
immunohistochemistry demonstrated massive TSP-1 deposits in the tumor
stroma (see Figure 5B
). In accordance with the Northern Blot results,
no major differences of VEGF mRNA expression were found between TSP-1
overexpressing and control tumors by in situ hybridization
(Figure 3, E–H)
. Up-regulation of VEGF mRNA expression was observed
adjacent to areas of necrosis in tissue samples derived from TSP-1
overexpressing tumors and control tumors.
|
|
|
The morphology of tumor supplying blood vessels was studied on the
surface of tumors of identical size derived from TSP-1 transfected A431
clones after 4 weeks and control transfectants after 3 weeks (Figure 2, C and D)
. A significant rarefaction of large blood vessels was seen
selectively on the surface of TSP-1 overexpressing tumor xenografts. To
determine the microvascular density within TSP-1 transfected and
control A431 tumors, frozen sections of 5 different tumors derived from
the same A431 clone were stained with an antibody against mouse CD31
(Figure 4, C and D)
. Morphometric
analysis revealed decreased microvessel densities (Figure 4E)
in
3-week-old tumors derived from TSP-1 overexpressing clones T10 (28
± 10 vessels/mm2), T12 (46 ± 12
vessels/mm2), and T19 (32 ± 7
vessels/mm2), as compared to control clone C2 (60 ±
19 vessels/mm2) and C3 (52 ± 20
vessels/mm2). Moreover, the average area per vessel was
significantly (P < 0.001) smaller in TSP-1
overexpressing tumors (Figure 4G)
derived from clone T10 (729 ±
213 µm2), T12 (551 ± 157 µm2), and
T19 (755 ± 189 µm2), as compared to control tumors
derived from clone C2 (1057 ± 307 µm2) and C3
(1298 ± 504 µm2). The fraction of large blood
vessels with an area of more than 2000 µm2
was decreased
to fewer than 1% of all vessels in TSP-1 overexpressing tumors, as
compared to more than 10% in control tumors (Figure 4H)
. In accordance
with these data, the total area covered by blood vessels (Figure 4F)
was significantly (P < 0.001) smaller in tumors
derived from TSP-1 transfected clones T10 (1.7 ± 0.6%), T12
(2.6 ± 1.2%) and T19 (2.2 ± 0.5%), as compared to control
transfected clones C2 (5.8 ± 2%) and C3 (5.4 ± 1.7%).
Up-Regulation of the TSP-1 Receptor CD36 in TSP-1 Transfected Tumor Cells and Xenografts
It has been suggested that the antiangiogenic effect of TSP-1 is
mediated through the CD36 receptor on endothelial cells.35
We studied the expression of CD36 in TSP-1 overexpressing A431 tumors
and controls by immunohistological staining. While CD36 expression was
nearly absent in control transfected tumors (Figure 5C)
, strong CD36 expression was found in
a substantial number of TSP-1 overexpressing A431 tumor cells (Figure 5D)
. These findings were confirmed by Western blot analysis,
demonstrating strong up-regulation of CD36 in cell lysates obtained
from TSP-1 overexpressing A431 (Figure 5E)
and SCC-13 (Figure 5F)
clones as compared to control clones. When subconfluent A431 control
cells (clones 1, 2, and 3) were incubated for 24 hours with conditioned
media taken from TSP-1 overexpressing A431 clone T19, a 1.5 to ninefold
induction of CD36 expression was detected by Western blot, as compared
to incubation with conditioned media from control clones (Figure 5G)
.
In contrast, CD36 was already strongly expressed in untreated HDMEC,
and incubation with TSP-1 containing conditioned media did not further
increase CD36 protein expression levels (data not shown). To study the
biological significance of increased CD36 expression on TSP-1
transfected A431 clones, we studied cell adhesion to immobilized TSP-1
and collagen type I. TSP-1 overexpressing A431 cells showed
significantly enhanced cell adhesion onto a TSP-1 matrix
(P < 0.001) but not onto a collagen type I
matrix (Figure 5, H and I)
. Incubation with a blocking antibody against
the CD36 receptor significantly reduced adhesion of TSP-1
overexpressing A431 cells to TSP-1 (P < 0.001),
but not to collagen type I (Figure 5, H and I)
. In contrast, the
adhesion of A431 control cells to TSP-1 or collagen type I was not
affected. These results suggest that overexpression of TSP-1 in A431
and SCC-13 tumor cells leads to enhanced expression of its receptor
CD36 with potential consequences for tumor cell adhesion.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Overexpression of TSP-1 in A431 cell xenotransplants potently decreased tumor growth, as compared to control tumors transfected with vector only. Increased TSP-1 secretion by stable transfectants was confirmed by Western blot analyses of conditioned media, and the bioactivity of transfected TSP-1 was confirmed in a proliferation assay using human dermal microvascular endothelial cells which by themselves did not secrete substantial amounts of TSP-1. Addition of conditioned media from TSP-1 transfected A431 clones significantly inhibited HDMEC proliferation, whereas conditioned media from control A431 clones stimulated HDMEC growth. The potent inhibitory effect of conditioned media from TSP-1 transfected cell clones on HDMEC proliferation was TSP-1 specific since it was abolished by co-incubation with a TSP-1-neutralizing antibody. In situ hybridizations of tumor xenotransplants demonstrated that TSP-1 mRNA expression was maintained at high levels in TSP-1 transfected tumor cell clones. Together, these data provide evidence for a potent inhibitory effect of TSP-1 on skin cancer growth. In accordance with our results, TSP-1 has been reported to inhibit the in vivo growth of MDA-MB-435 breast carcinoma cells36 and of v-src transformed NIH 3T3 fibroblasts.37 Furthermore, TSP-1 negatively affected the tumor forming ability of transformed mouse endothelial cells,38 of human glioblastoma cells,39 and of human B16/F10 melanoma cells.40 However, a previous investigation using the human squamous cell carcinoma cell line 11B reported that reduction of TSP-1 secretion by antisense transfection inhibited tumor growth in vivo, and the authors suggested that TSP-1 overexpression in squamous cell carcinomas might lead to enhanced tumor growth.20 Our data provide strong evidence against this hypothesis and demonstrate, furthermore, that TSP-1 may even completely prevent the formation of cutaneous squamous cell carcinomas. We could demonstrate that stable TSP-1 transfection of human SCC-13 cells, a slowly growing human squamous cell carcinoma line with formation of highly differentiated tumors in vivo, induced a complete inhibition of any detectable tumor growth over an observation period of 12 weeks after xenotransplantation.
The tumor growth inhibition induced by TSP-1 in cutaneous squamous cell carcinomas was not due to direct TSP-1-mediated inhibition of tumor cell growth. We used three independent in vitro assays to investigate TSP-1 effects on A431 cells: (1) Anchorage-dependent cell growth in monolayer culture; (2) anchorage-independent cell growth, as determined by the ability to form colonies in soft agar; and (3) susceptibility to induction of apoptosis by serum withdrawal. No significant differences between TSP-1 transfected A431 clones and control transfected A431 clones were detected in any of these studies. These findings are in accordance with previous findings, showing that TSP-1 transfection did not influence the in vitro growth of human MDA-MB-453 breast cancer cells36 and B16/F10 melanoma cells.40 However, they are in contrast to a previous investigation, reporting a direct correlation between the levels of TSP-1 secretion by several squamous cell carcinoma cell lines and enhanced in vitro cell growth and invasion.41 Moreover, another study reported that antisense-mediated down-regulation of TSP-1 expression in a squamous cell carcinoma cell line led to decreased in vitro cell proliferation, suggesting TSP-1 as a promoter of tumor growth.20 In 3T3 fibroblasts, antibodies to TSP-1 were reported to inhibit cell proliferation, and TSP-1 overexpression induced serum- and anchorage-independent growth.42 While there is no obvious explanation for these discrepancies, varying susceptibilities of different cell types (eg, mesenchymal versus epithelial) or of distinct cell lines to TSP-1 may play a role. It has been reported that TSP-1 inhibited proliferation of A2058 malignant melanoma cells in vitro,43 and certain TSP-1 effects seem to be specific for distinct cell lines. Our results, obtained by using three independent in vitro growth assays, strongly indicate that A431 cell growth is not influenced by TSP-1. This is further supported by the findings that cell proliferation rates of TSP-1 overexpressing A431 xenotransplants were indistinguishable from control transfected A431 xenotransplants, as measured by expression of the proliferation-related antigen PCNA. Similarly, tumor treatment with the angiogenesis inhibitor angiostatin also led to reduced tumor size without changing tumor cell proliferation rates.44
The CD36 receptor has been reported to be the receptor on vascular endothelial cells that mediates the anti-angiogenic effects of TSP-1.35 We could not detect differences in endothelial cell CD36 expression between control tumors and TSP-1 overexpressing tumors by immunostaining. However, TSP-1 overexpressing A431 tumors showed increased CD36 expression in tumor cells, as compared to no or little CD36 expression in control tumors. TSP-1 transfected A431 and SCC-13 cell clones also showed highly increased CD36 expression in vitro. Moreover, incubation of control A431 clones with conditioned media obtained from TSP-1 overexpressing clones significantly induced CD36 expression. These results suggest that soluble TSP-1 induces the CD36 receptor on tumor cells. They are in accordance with the recent findings that adhesion to TSP-1 induced CD36 in embryonic fibroblasts.45 Our results further demonstrate that up-regulation of CD36 expression in TSP-1 transfected A431 cells moderately enhanced cell adhesion to immobilized TSP-1 in vitro, in accordance with a previous report demonstrating that HT-1080 fibrosarcoma cells adhered to TSP-1 via their CD36 receptor. The functional consequences of enhanced tumor cell expression of CD36 in vivo remain unknown since we did not detect any differences in tumor cell proliferation between control and TSP-1/CD36 overexpressing A431 xenotransplants. However, it is conceivable that the CD36/TSP-1 system might play a role for the invasive or metastatic properties of malignant tumor cells in vivo.
Overexpression of TSP-1 in A431 xenotransplants resulted in extensive areas of tumor cell necrosis, possibly due to anti-angiogenic effects of TSP-1. To define the effects of TSP-1 on tumor vasculature, we performed morphometric analyses on CD31 stained tumor sections by using a semiautomated image analysis program. In TSP-1 overexpressing tumors, a moderate reduction of the microvascular density was observed, as measured by the average number of CD31-positive tumor vessels per mm2. Similar results have been reported in human breast cancer xenotransplants.36 However, when we compared vessel sizes, we found that the average vessel area was significantly diminished in TSP-1 overexpressing tumors with absence of larger vessels with surface areas of greater that 2,000 µm2. These findings demonstrate that TSP-1 reduced the characteristically increased size of tumor vessels and suggest that measuring vascular densities alone may not be sufficient to detect changes in tumor vasculature. They also suggest that determining the total vascular area per unit area may serve as a more sensitive parameter to measure tumor angiogenesis.
It is of interest that the first vascular changes observed during
treatment of experimental tumors with an antibody to the angiogenesis
factor VEGF also consisted of a dramatic reduction of blood vessel
diameters and tortuosity.46
Moreover, overexpression of
VEGF in the skin of transgenic mice33
or in MEL-57
melanoma xenotransplants4
led to the development of
tortuous and dilated blood vessels, and inhibition of the
VEGF-inducible 
1- and 2-integrins significantly inhibited
VEGF-driven tumor angiogenesis in vivo, most prominently
through reduction of average blood vessel diameters.47
To
exclude that the reduction in vessel sizes observed in TSP-1
overexpressing A431 xenotransplant tumors was due to down-regulation of
VEGF expression, we performed in situ hybridizations of
tumor xenotransplants. These studies demonstrated unchanged levels of
VEGF mRNA expression in TSP-1 overexpressing tumors versus
controls. Therefore, the reduction in vessel sizes reflects an
important role of TSP-1 on the formation of tumor vasculature and
demonstrates that similar vascular effects can be obtained by
overexpression of TSP-1 or by inhibition of VEGF, suggesting antipodal
roles of the two molecules in tumor angiogenesis. In summary, TSP-1
induced a potent growth inhibition of malignant epithelial skin cancer
with complete inhibition of tumor development of highly differentiated
SCC-13 tumors. The anti-tumoral effect of TSP-1 was not due to direct
inhibition of tumor cell proliferation, but was associated with
significant inhibition of tumor angiogenesis. Our results suggest that
quantitation of both microvascular densities and of total vascular
areas provides a much more sensitive parameter for tumor angiogenesis
than determination of vessel densities alone.
|
Footnotes |
|---|
Supported by NIH/NCI grant CA69184 (to M.D.), by Deutsche Forschungsgemeinschaft grant De483/3–2 (to M.D.), by Deutscher Akademischer Austauschdienst (to M.St.), by the Human Frontier Science Program (M.Sk.), by NIH/HLBI grant HL28749 (to J.L.), and by the Cutaneous Biology Research Center through the Massachusetts General Hospital/Shiseido Co. Ltd. Agreement (M.D.).
Accepted for publication April 11, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5ß1-fibronectin receptor in HT29 colon-cancer cells reduces activity of C-SRC: increase of C-SRC activity by attachment on fibronectin. Int J Cancer 1998, 76:91-98[Medline]
5ß1: a potent inhibitor of experimental lung metastasis. Clin Exp Metastasis 1998, 16:427-435[Medline]
1ß1, and 2ß1integrins. Proc Natl Acad Sci USA 1997, 94:13612-13617This article has been cited by other articles:
 |
|
 |
|
  M.-P. Wu, M.-J. Young, C.-C. Tzeng, C.-R. Tzeng, K.-F. Huang, L.-W. Wu, and C.-Y. Chou A novel role of thrombospondin-1 in cervical carcinogenesis: inhibit stroma reaction by inhibiting activated fibroblasts from invading cancer Carcinogenesis, June 1, 2008; 29(6): 1115 - 1123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. Sreekumar, J. Zhou, J. Sohn, C. Spee, S. J. Ryan, B. J. Maurer, R. Kannan, and D. R. Hinton N-(4-hydroxyphenyl) Retinamide Augments Laser-Induced Choroidal Neovascularization in Mice Invest. Ophthalmol. Vis. Sci., March 1, 2008; 49(3): 1210 - 1220. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Hashimoto, M. Parsons, and J. C. Adams Dual Actin-bundling and Protein Kinase C-binding Activities of Fascin Regulate Carcinoma Cell Migration Downstream of Rac and Contribute to Metastasis Mol. Biol. Cell, November 1, 2007; 18(11): 4591 - 4602. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Li, C. Wang, X. Jiao, Y. Lu, M. Fu, A. A. Quong, C. Dye, J. Yang, M. Dai, X. Ju, et al. Cyclin D1 Regulates Cellular Migration through the Inhibition of Thrombospondin 1 and ROCK Signaling. Mol. Cell. Biol., June 1, 2006; 26(11): 4240 - 4256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Masli, B. Turpie, and J W. Streilein Thrombospondin orchestrates the tolerance-promoting properties of TGF{beta}-treated antigen-presenting cells Int. Immunol., May 1, 2006; 18(5): 689 - 699. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Zhang, E. Galardi, M. Duquette, J. Lawler, and S. Parangi Antiangiogenic Treatment with Three Thrombospondin-1 Type 1 Repeats versus Gemcitabine in an Orthotopic Human Pancreatic Cancer Model Clin. Cancer Res., August 1, 2005; 11(15): 5622 - 5630. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. G. Frangogiannis, G. Ren, O. Dewald, P. Zymek, S. Haudek, A. Koerting, K. Winkelmann, L. H. Michael, J. Lawler, and M. L. Entman Critical Role of Endogenous Thrombospondin-1 in Preventing Expansion of Healing Myocardial Infarcts Circulation, June 7, 2005; 111(22): 2935 - 2942. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Nyberg, L. Xie, and R. Kalluri Endogenous Inhibitors of Angiogenesis Cancer Res., May 15, 2005; 65(10): 3967 - 3979. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Zhang, E. Galardi, M. Duquette, M. Delic, J. Lawler, and S. Parangi Antiangiogenic Treatment with the Three Thrombospondin-1 Type 1 Repeats Recombinant Protein in an Orthotopic Human Pancreatic Cancer Model Clin. Cancer Res., March 15, 2005; 11(6): 2337 - 2344. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Pohl, C.-L. Ho, R. J. Kurman, R. Bristow, T.-L. Wang, and I.-M. Shih Inactivation of the Mitogen-Activated Protein Kinase Pathway as a Potential Target-Based Therapy in Ovarian Serous Tumors with KRAS or BRAF Mutations Cancer Res., March 1, 2005; 65(5): 1994 - 2000. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sund, Y. Hamano, H. Sugimoto, A. Sudhakar, M. Soubasakos, U. Yerramalla, L. E. Benjamin, J. Lawler, M. Kieran, A. Shah, et al. Function of endogenous inhibitors of angiogenesis as endothelium-specific tumor suppressors PNAS, February 22, 2005; 102(8): 2934 - 2939. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. G. Spinale Cell-Matrix Signaling and Thrombospondin: Another Link to Myocardial Matrix Remodeling Circ. Res., September 3, 2004; 95(5): 446 - 448. [Full Text] [PDF]
|
|
|
|
|
|
S. Adams, G. T. Miller, M. I. Jesson, T. Watanabe, B. Jones, and B. P. Wallner PT-100, a Small Molecule Dipeptidyl Peptidase Inhibitor, Has Potent Antitumor Effects and Augments Antibody-Mediated Cytotoxicity via a Novel Immune Mechanism Cancer Res., August 1, 2004; 64(15): 5471 - 5480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. O. Yee, M. Streit, T. Hawighorst, M. Detmar, and J. Lawler Expression of the Type-1 Repeats of Thrombospondin-1 Inhibits Tumor Growth Through Activation of Transforming Growth Factor-{beta} Am. J. Pathol., August 1, 2004; 165(2): 541 - 552. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Segrelles, S. Ruiz, M. Santos, J. Martinez-Palacio, M. F. Lara, and J. M. Paramio Akt mediates an angiogenic switch in transformed keratinocytes Carcinogenesis, July 1, 2004; 25(7): 1137 - 1147. [Abstract] [Full Text] [PDF]
|
|
