close

Вход

Забыли?

вход по аккаунту

?

931

код для вставкиСкачать
2230
Expression of Transforming Growth Factor-b (TGF-b)
Isoforms in Osteosarcomas
TGF-b3 Is Related to Disease Progression
Peter Kloen, M.D.1
Mark C. Gebhardt, M.D.1
Antonio Perez-Atayde, M.D.2
Andrew E. Rosenberg, M.D.3
Dempsey S. Springfield, M.D.1
Leslie I. Gold, Ph.D.4
Henry J. Mankin, M.D.1
BACKGROUND. Transforming growth factor-b (TGF-b) is a multipotent growth factor affecting development, homeostasis, and tissue repair. In addition, increased
expression of TGF-b has been reported in different malignancies, suggesting a role
1
Department of Orthopaedic Surgery, Massachusetts General Hospital, Harvard Medical
School, Boston, Massachusetts.
2
Department of Pathology, Children’s Hospital,
Harvard Medical School, Boston, Massachusetts.
3
Department of Pathology, Massachusetts General Hospital, Harvard Medical School, Boston,
Massachusetts.
4
Department of Pathology, New York University
Medical Center, New York, New York.
Presented at the 42nd Annual Meeting of the
Orthopaedic Research Society, Atlanta, Georgia,
February 18-22 1996.
for this growth factor in tumorigenesis.
METHODS. Using immunohistochemistry, the expression, prevalence, and distribution of TGF-b isoforms were evaluated in 25 high grade human osteosarcomas.
The Cox proportional hazards models and Kaplan-Meier curves were calculated
correlating disease free survival with TGF-b expression.
RESULTS. Expression of one or more TGF-b isoforms was found in all the osteosarcomas. Immunoreactivity for TGF-b1 and TGF-b3 generally was stronger than for
TGF-b2. The cytoplasm of the tumor cells showed stronger staining than their
surrounding extracellular stroma. Most notably, osteoclasts showed strong to intense staining for all three isoforms. In 11 of 25 specimens angiogenic activity was
noted with staining of multiple small vessels in the tumor stroma. Expression of
TGF-b3, but not of TGF-b2 or TGF-b1, related to disease progression, such that
there was a statistically significant decrease in the disease free interval as the
immunoreactivity for TGF-b3 increased.
CONCLUSIONS. All osteosarcomas expressed TGF-b in the cytoplasm of the tumor
cells as well as in their extracellular stroma. The presence of TGF-b in the endothelial and perivascular layers of small vessels in the tumor stroma suggests angiogenic
activity of this growth factor. The expression of TGF-b3 was correlated strongly
with disease progression (P Å 0.027). These data suggest that increased expression
of TGF-b isoforms, especially TGF-b3, may play a role in osteosarcoma progression.
Cancer 1997;80:2230–9. q 1997 American Cancer Society.
Supported by an institutional grant of the Orthopaedic Research and Education Foundation.
KEYWORDS: osteosarcoma, transforming growth factor-b, immunohistochemistry,
disease progression.
The authors thank Ilsoon Yang, Ph.D., Department of Biostatistics, Harvard School of Public
Health, for her help with the statistical evaluation of the data.
T
Dr. Springfield’s current address: Department
of Orthopaedic Surgery, Mount Sinai Medical
Center, New York, New York.
Address for reprints: Peter Kloen, M.D., Department of Orthopaedic Surgery, Massachusetts
General Hospital, Gray 6, 55 Fruit Street, Boston, MA 02114.
Received March 6, 1997; revision received June
13, 1997; accepted June 13, 1997.
ransforming growth factor-b(TGF-b), originally described for its
ability to transform the phenotype of fibroblasts, is a member of
the large TGF-b superfamily including TGF-b1-5, bone morphogenetic proteins 2-8 (BMP 2-8), inhibins, activins, and müllerian inhibiting
substance.1 Of the 5 different TGF-b isoforms, only TGF-b1, -b2, and b3 are found in mammalian tissue. Although the highest concentration of TGF-b is found in blood platelets, on a tissue mass basis, bone
is the most abundant source of these polypeptides.2 The different
TGF-b isoforms share many of the biologic activities involved in normal processes such as morphogenesis and differentiation in embryogenesis, tissue repair, and wound healing.1,3,4 In vitro and in vivo
studies have shown that TGF-b1, -b2, and -b3 can regulate bone
formation by inducing osteoblast and osteoclast proliferation, extra-
q 1997 American Cancer Society
/ 7b89$$1515
11-25-97 12:52:41
cana
W: Cancer
TGF-b Expression in Osteosarcomas/Kloen et al.
cellular matrix formation, and bone-related protein
production.5,6 The direct application of TGF-b to bone
induced intramembranous bone repair and repair of
calvarial defects.7 – 9 The involvement of TGF-b in both
osteogenic and chondrogenic bone formation occurs
through its ability to recruit osteoblasts and osteoclasts to the injured areas and through its ability to
stimulate the growth and production of extracellular
matrix protein by these cells.5,9 With its substantial
production and presence in bone and its known role
in skeletal development and remodeling affecting both
absorption (osteoclasts) and production (osteoblasts)
of bone (at picomolar concentrations), a tight control
mechanism of TGF-b expression is needed.10
In addition to its ubiquitous presence in normal
tissues, increased expression of TGF-b has been reported in many different tumors.11 – 26 Furthermore, increased expression of any one of the isoforms has been
associated with disease progression and decreased
survival.13 Because of the stimulatory effect on the
constituents of stromal tissue, angiogenesis,27,28 chemotaxis, and enhancement of cell migration and adhesion, TGF-b can provide tumor cells with a growth
advantage. TGF-b also is an immunosuppressant29
and its inhibiting effect on natural killer cells and lymphokine-activated cytotoxic T cells would result in a
decrease in immunosurveillance and immune cellmediated tumor destruction.
No information currently is available on the prevalence and distribution of TGF-b isoforms in osteosarcoma, a malignant bone tumor characterized by osteoid production of tumor cells. It is the most common
primary tumor of bone, affecting people in their second decade, males more often than females. The etiology of osteosarcoma still is unknown, although many
suggestions have been made, including mutations in
suppressor-genes (p53 and retinoblastoma (Rb)),30 – 32
and increased expression of growth factors including
insulin-like growth factor (IGF ),33 platelet-derived
growth factor,34 TGF-b,35 and BMPs.36 – 38 Although significant improvements have been made in the survival
rates of osteosarcoma patients using combinations of
adjuvant chemotherapy, surgery, and radiation therapy, it still is important to find markers identifying
patients with a poor prognosis so that therapy can be
adjusted. The current study was undertaken to document the prevalence and distribution of the TGF-b1, b2, and -b3 in high grade osteosarcoma, seeking a
possible correlation with disease progression.
MATERIALS AND METHODS
Tissue Specimens
High grade, Stage 2B39 formalin fixed, paraffin embedded appendicular skeleton primary classic osteosarco-
/ 7b89$$1515
11-25-97 12:52:41
cana
2231
mas were selected (n Å 25) from the Departments of
Pathology at the Massachusetts General Hospital, Boston, and Children’s Hospital, Boston. Rather than
studying tumors of different grade and/or stage, this
study focused on high grade Stage 2B lesions because
this subgroup would benefit from a potential additional prognostic marker such as TGF-b that would
identify those tumors with a worse outcome. Specimens were decalcified prior to paraffin embedding.
Informed consent was obtained regarding the use of
tissue for research purposes. The male:female ratio
was 16:9, with patient ages ranging from 9 – 68 years.
The mean standard deviation for age was 33 { 22 years
for females and 25 { 16 years for males. Based on
availability of the paraffin blocks, specimens were
taken from biopsies (n Å 21) or resections prior to
chemotherapy (n Å 4). Primary tumors were located
at the following sites: proximal humerus (n Å 2), distal
humerus (n Å 1), proximal femur (n Å 2), distal femur
(n Å 17), proximal tibia (n Å 1), distal tibia (n Å 1),
and distal fibula (n Å 1). The specimens were classified
based on their histology as osteoblastic, chondroblastic, fibroblastic, teleangiectatic, giant cell, or mixed osteosarcoma (see Table 1 for patient data). All patients
were treated with adjuvant chemotherapy. For patients age õ 30 years this was comprised of a Pediatric
Oncology Group protocol of doxorubicin, cisplatin,
bleomycin, cyclophosphamide, actinomycin D, and
methotrexate. Patients age ú 30 years received an individualized protocol using similar drugs at a lower concentration.
Immunohistochemistry
Sections of 4 mm were cut and transferred to silanized
(3-aminopropyltriethoxysilane; Sigma, St. Louis, MO)
Fisher-Plus slides (Fisher Scientific, Fair Lawn, NJ).
Immunostaining was performed as described previously.4 Briefly, the slides were baked overnight at
58 7C and the tissue was deparaffinized in xylene, hydrated in sequential gradients of ethanol, and washed
with 0.3% (volume/volume [v/v]) Triton-X-100/trisbuffered saline (TBS) for 15 minutes at room temperature (RT). Endogenous peroxidase was quenched with
0.6% hydrogen peroxide in methanol (v/v) for 45 minutes, and the slides subsequently were treated with
hyaluronidase (1 mg/ml in 100 mM sodium acetate0.85% NaCl [pH 5.5]; Sigma) at 37 7C. The slides then
were washed with TBS-0.1% bovine serum albumin
(BSA) and blocked with goat serum immunoglobulin
(Ig) G at 1 mg/ml in TBS-1.0% BSA for 20 minutes at
RT on a rocker platform. The tissue sections then were
incubated overnight with antibodies to TGF-b1, -b2,
and -b3 at a concentration of 1.0 mg/ml, at 4 7C, in
humido. The antibodies were raised to synthetic pep-
W: Cancer
2232
CANCER December 15, 1997 / Volume 80 / Number 12
TABLE 1
Patient Data
Patient
no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Gender
Age
(yrs)
Site
Histology
Surgery
F
M
M
M
M
F
F
M
M
F
M
F
F
M
M
F
M
M
F
M
M
F
M
M
M
31
21
16
17
48
61
9
21
10
56
14
12
20
12
30
25
10
37
19
14
68
65
31
31
21
PH
PFe
DH
DFe
DFe
DFe
DFe
DFe
PH
DFe
DFe
DFe
DFe
DT
DFi
DFe
DFe
DFe
DFe
PT
DFe
DFe
DFe
DFe
PFe
Ost
Ost
Ost
Ost
Ost
Ost
Ost
Ost
Ost
Ost
Ost
Ost
Ost
Ost
M
M
Chon
Fibr
Fibr
Fibr
Fibr
Fibr
GC
Tel
Tel
LS
LS
LS
LS
LS
LS
LS
LS
LS
Amp
LS
Amp
LS
Amp
LS
LS
LS
LS
LS
LS
LS
LS
LS
LS
LS
Metastases (at
recurrence)
L
L
L/B
L
L
L
L
L/B
L/A
L
L
L
L
L
L/B
TGF-b1
staining
TGF-b2
staining
TGF-b3
staining
Disease free
interval
(mos)
Status
3/
1/
2/
2/
2/
2/
3/
3/
3/
1/
3/
4/
3/
2/
2/
1/
4/
2/
2/
3/
2/
2/
4/
1/
3/
3/
2/
2/
1/
NA
1/
2/
3/
2/
2/
2/
4/
2/
NA
2/
2/
3/
4/
1/
2/
2/
2/
NA
1/
2/
2/
2/
1/
3/
3/
2/
3/
2/
3/
1/
3/
3/
3/
3/
1/
3/
3/
3/
2/
3/
3/
2/
4/
2/
2/
20
12
22
21
4
11
7
48
8
45
23
9
13
37
66
3
49
32
17
32
60
59
14
87
33
Alive
Alive
DOD
DOD
DOD
DOD
Alive
Alive
DOD
Alive
Alive
DOD
DOD
Alive
Alive
DOD
Alive
DOD
Alive
Alive
Alive
Alive
Alive
Alive
Alive
TGF: transforming growth factor; F: female; M: male; P: proximal; D: distal; H: humerus; Fe: femur; T: tibia; Fi: fibula; Ost: osteoblastic; M: mixed; Chon: chondroblastic; Fibr: fibroblastic; GC: giant cell; Tel:
teleangiectatic; LS: limb salvage; Amp: amputation; L: lung; B: bone; A: abdomen; NA: not available; DOD: dead of disease.
tides of each TGF-b isoform and the IgG was peptide
affinity purified. The preparation, characterization,
and isoform specificity of the antibodies, by Western
blot analyses, were all described elsewhere.4 These antibodies have been shown to be specific for each isoform of TGF-b and are non cross-reacting. The antibody concentration used previously was optimized using dilutions ranging from 0 – 5 mg/ml. Nonimmune
rabbit IgG (Immunon, Pittsburgh, PA) at the same concentration as the TGF-b antibodies was used as a con-
trol to show antibody specificity. After incubation with
the antibodies, the slides were washed extensively with
TBS-0.l% BSA, and biotinylated goat antirabbit secondary antibody (Vectastain ELITE ABC Kit; Vector
Laboratories, Inc., Burlingame, CA) was applied to
slides for 1 hour at RT, followed by washing with TBS0.1% BSA, and then incubation with avidin-biotin
complex reagent for 1 hour at RT. To develop the color
reaction, 0.05% 3,3’-diaminobenzidine (DAB) (Sigma)
in 0.01% hydrogen peroxide and 0.05 M Tris was added
c
FIGURE 1.
Immunohistochemical localization of transforming growth factor-b (TGF-b) in osteosarcoma. Immunoreactivity is represented by a brown
color developed by avidin-biotin complex and 3,3’diaminobenzidine (Vectastain ELITE ABC Kit, Vector Laboratories, Burlingame, CA). A) TGF-b3 immunoreactivity in an osteoblastic osteosarcoma (Patient 9). There is strong staining (3/) in tumor cells, with light (1/) to moderate staining (2/) in tumor
stroma and an absence of staining in mineralized tumor matrix (1 100). B) Anti-TGF-b3 immunoreactivity in a fibroblastic osteosarcoma (Patient 18).
There is strong staining (3/) in tumor cells and moderate staining (2/) in tumor stroma (1 100). C) Immunohistochemical staining for TGF-b2 in
endothelial and perivascular smooth musculature of multiple small vessels in tumor stroma (Patient 4) (1 80). D) Control specimen, nonimmune rabbit
immunoglobulin G. Osteoblastic osteosarcoma (Patient 3). There is an absence of staining indicating that the results are specific (1 100). The individual
figures may not always correctly represent the mean immunoreactivity intensities represented in Table 1.
/ 7b89$$1515
11-25-97 12:52:41
cana
W: Cancer
TGF-b Expression in Osteosarcomas/Kloen et al.
/ 7b89$$1515
11-25-97 12:52:41
cana
W: Cancer
2233
2234
CANCER December 15, 1997 / Volume 80 / Number 12
to the slides for 3 minutes at RT. Finally, the slides were
counterstained for 5 minutes in Gill’s #2 hematoxylin
(Fisher Scientific), and 10 seconds in 0.6% lithium carbonate (Fisher Scientific); the tissue was dehydrated
in graded ethanol and the slides were mounted using
Permount (Fisher Scientific). Staining was evaluated
by three investigators that were blinded to the disease
progression. Staining was graded on a scale of 0 – 4/,
with 0 representing no staining, and 4/ representing
intense staining. Increased staining at the edges of the
tissue was considered artefact and disregarded during
evaluation.
Statistical Evaluation
Follow-up on the patients was obtained through their
office charts. Data in the statistical evaluation included
the disease free interval, which represented time from
first surgery to first evidence of metastasis or recurrence if observed; otherwise it represented the time
from the first surgery to the end of the study. Overall
survival was calculated as the time from the first surgery to the time of death. For statistical analysis immunoreactivity was divided in two groups: low immunoreactivity (1//2/) and high immunoreactivity (3//
4/) for each TGF-b isoform. The data were analyzed
using Cox proportional hazards models and KaplanMeier curves40 to determine whether immunoreactivity for TGF-b isoforms was related significantly to disease progression.
RESULTS
Immunoreactivity for TGF-b Isoforms in Osteosarcomas
Expression of one or more TGF-b isoforms was found
in all the Stage 2B classic osteosarcomas studied (Table 1) (Fig. 1A and 1B). (One small cell osteosarcoma
specimen did not immunostain for any TGF-b isoform; this tumor was not included in our analysis).
Absence of immunostaining of the classic osteosarcomas using nonimmune rabbit IgG indicated that the
results obtained were specific (Fig. 1D). Immunoreactivity for TGF-b1 and -b3 generally was stronger than
for TGF-b2. Minimal to moderate staining for TGF-b1
was observed in 14 of 25 tumors (56%), for TGF-b2 in
17 of 22 tumors (77%), and for TGF-b3 in 11 of 25
tumors (44%). Strong to intense immunoreactivity was
observed in 11 of 25 tumors (44%), 5 of 22 tumors
(23%), and 14 of 25 tumors (56%) for TGF-b1, -b2,
and -b3, respectively.
TGF-b isoforms were not present in mineralized
bone of osteosarcomas. Uninvolved bone in the resected specimens also had absence of staining of mineralized matrix. In general the osteocytes embedded
in the inner areas of the trabeculae in osteosarcoma
bone were negative for TGF-b. Normal unmineralized
/ 7b89$$1515
11-25-97 12:52:41
cana
bone matrix showed little or no staining, whereas
newly formed unmineralized pathologic bone showed
light to moderate staining. The cytoplasm of the osteosarcoma cells generally showed stronger staining for
TGF-b compared with the surrounding extracellular
stroma. Most notably, the tumor osteocytes embedded
in woven tumor bone were more strongly immunoreactive for all TGF-b isoforms compared with the osteocytes embedded in normal lamellar bone. The cytoplasmic expression of TGF-b in occasional chondrocytes observed in either pathologic fracture callus or in
growth plate was inconsistent, with immunostaining
ranging from minimal to moderate. Actively resorbing
osteoclasts, which were observed mainly at the outer
edges of pathologic bone and in lesser quantity in normal mineralized bone, showed strong to intense staining for all three isoforms. In nearly 50% of the tumors
(11 of 25) angiogenic activity was noted by the presence of multiple small vessels in and around tumor
stroma. The endothelial cells and perivascular muscle
cells of these vessels stained for all TGF-b isoforms,
although immunoreactivity was more intense for TGFb1 and TGF-b3 than for TGF-b2 (Fig. 1C). Although
vessels in normal stroma showed a similar staining
pattern, there were significantly fewer vessels present
in the normal tissue compared with tumor stroma.
Grading of all specimens was found to be consistent
and reproducible between observers and between observations.
Intense Immunoreactivity for TGF-b3 Is Related to
Disease Progression
For statistical calculation, the specimens were divided
into two groups: those with low (1/ to 2/) and those
with high (3/ to 4/) immunoreactivity. We then analyzed the relation of TGF-b immunostaining to disease
progression separately for each isoform. First we included the different variables given each TGF-b isoform. We found that age (P Å 0.35), gender (P Å 0.20),
tumor site (P ú 0.2), histology (P ú 0.05), and surgery
(limb salvage vs. amputation) (P ú 0.1) were not statistically significantly related to disease progression. Interaction effects were considered, but found not to be
significant.
Next, Cox proportional hazards models were fitted
with each TGF-b isoform, adjusting for age and gender
differences between the groups with low and high
TGF-b immunoreactivity. Table 2 shows the Cox proportional hazards models that best explain the relationships between TGF-b isoforms and disease progression. The levels of TGF-b1 and -b2 immunostaining were not related significantly to disease
progression (P Å 0.75 and P Å 0.15, respectively). In
contrast, immunoreactivity for TGF-b3 was related to
W: Cancer
TGF-b Expression in Osteosarcomas/Kloen et al.
DISCUSSION
TABLE 2
Multivariate Cox Proportional Hazards Model on Disease Free Timea
Variable
Hazard
ratio
P value
95% CI
Male
Age
High TGF-b1
Male
Age
High TGF-b2
Male
Age
High TGF-b3
0.27
0.97
1.23
0.11
0.93
0.3
0.19
0.99
5.07
0.052
0.189
0.745
0.008
0.006
0.145
0.013
0.411
0.027
(0.07, 1.01)
(0.93, 1.01)
(0.36, 4.19)
(0.02, 0.56)
(0.89, 0.98)
(0.06, 1.51)
(0.05, 0.70)
(0.96, 1.02)
(1.20, 21.4)
95% CI: 95% confidence interval; TGF: transforming growth factor.
a
Cox proportional hazards model that best explains the relationships for each transforming growth
factor-b isoform and disease-free time.
disease progression, given that there was a statistically
significant decrease in the disease free period as staining increased from low to high (P Å 0.027) (Table 2).
The hazard ratio calculated for patients in this study
with high levels of TGF-b3 showed that they were 5.07
times more likely to have disease progression than
those with low levels of TGF-b3, and that this risk for
females was 5.26 times that for males (Table 2).
Kaplan-Meier disease free survival curves plotted separately for males and females indicate a relation between TGF-b3 and gender (data not shown), but this
interaction was not significant (P Å 0.182). KaplanMeier disease free survival curves also were plotted
versus the level of each TGF-b isoform in the tumor.
High levels of TGF-b3 correlated with disease progression (Fig. 2C) (P Å 0.027). Although a similar trend was
observed for TGF-b1 (Fig. 2A), this was not statistically
significant (P Å 0.75) (Table 2). No correlation was
observed between the expression of TGF-b2 and disease progression (P Å 0.15) (Fig. 2B) (Table 2).
When we analyzed the effects of TGF-b isoforms
on survival (time from first surgery to death), using
Cox proportional hazards models, neither TGF-b1 (P
Å 0.43) nor TGF-b2 (P Å 0.86) were significantly related to survival. For TGF-b3, we only considered patients with recurrent or metastatic disease when evaluating survival, given the relationship between this isoform and disease progression (Table 2) (Fig. 2C).
According to the analysis, the level of TGF-b3 staining
was not significantly related to patient survival (P Å
0.39), although it is noted that these results may be
due to the small sample size (n Å 15). The findings
imply that low immunoreactivity for TGF-b3 indicates
longer disease free survival, but does not necessarily
correlate with survival time.
/ 7b89$$1515
11-25-97 12:52:41
2235
cana
TGF-b has multiple roles in normal processes such as
development, growth, differentiation, and tissue repair
with complex effects on almost all cell types.1 Effects
of TGF-b on cellular proliferation characteristically are
inhibitory for epithelium-derived cells, and stimulatory for mesenchyme-derived cells. Although no significant qualitative or quantitative differences have
been identified between the three TGF-b isoforms regarding their effects on the skeletal system, their differential expression during mouse embryogenesis and
development suggest that they each have specific
roles.3,4 Previous immunohistochemical studies have
shown that TGF-b2 and TGF-b3 generally were expressed in transitional chondrocytes, whereas TGF-b1
was expressed more highly in osteogenesis areas.3,4,41
In vivo studies showed that injection of TGF-b1 in
the subperiosteal region of rat femurs resulted in the
stimulation of new bone formation resembling embryologic bone formation and fracture repair.9
The biologic activities of TGF-b that enable normal growth control, repair, and homeostasis, such as
both stimulatory and inhibitory effects on cell proliferation, angiogenesis, cell migration, chemotaxis, and
cell adhesion, may become dysregulated, leading to
tumorigenesis.1,27,42 This could be the result of changes
in the expression of one or more of its isoforms and/
or its receptors or postreceptor pathways. These observations, in addition to the immunosuppressive activity29 and the auto-inductive43 and cross-regulatory
capacity of TGF-b isoforms, have led to the evaluation
and subsequent implication of TGF-b in the development and progression of different tumors.11 – 26 For example, using transfection studies resulting in increased expression of active TGF-b1 in fibrosarcoma
cells in vivo, it was shown that increased TGF-b1 synthesis provides a significant advantage for tumor formation.44 Furthermore, the different TGF-b isoforms
have been correlated with disease progression and decreased patient survival in various neoplasms (Table
3). The seemingly paradoxic increased expression of
TGF-b in epithelium-derived tumors (because TGF-b
inhibits epithelial cell proliferation) can be explained
by the loss of inhibitory control in addition to paracrine effects on stroma production, angiogenesis, and
immunosuppression. Less is known regarding autocrine loops involving TGF-b in mesenchyme-derived
tumors, in which the increased TGF-b theoretically
could stimulate cell proliferation. An immunohistochemical study by McCune et al.19 on the expression of
TGF-b in small round cell tumors affecting the skeletal
system in children reported that in both Ewing’s sarcoma and the (mesenchymally derived) rhabdomyosarcomas there was increased immunostaining for
W: Cancer
2236
CANCER December 15, 1997 / Volume 80 / Number 12
FIGURE 2.
Kaplan-Meier plots of the disease free time in patients with high grade osteosarcoma that had low versus high
expression of A) transforming growth factor (TGF)-b1 (n Å 25); B) TGF-b2 (n Å 23); or C) TGF-b3 (n Å 25).
TGF-b1 and TGF-b3, but no staining for TGF-b2. In
contrast to these findings, we found weak to moderate
staining for TGF-b1 in 9 of 9, light to moderate staining
for TGF-b2 in 6 of 9, and moderate to strong immunoreactivity for TGF-b3 in 9 of 9 Ewing’s sarcomas (unpublished data). Because TGF-b2 is most highly associated with bone and has a higher specific activity for
bone induction,5 one would expect to find TGF-b2 in
bone. However, the different antibodies used in these
studies could be related to the different results. Recently, two studies reported on the immunohistochemical localization of TGF-b in mesenchymally derived Kaposi sarcoma.11,26
We previously showed that all prerequisites for an
autocrine loop involving TGF-b (presence of TGF-b
receptors, production of active and latent TGF-b, effect on DNA synthesis by adding exogenous TGF-b,
and obviating effect of TGF-b by antibodies to TGFb) are present in vitro in human osteosarcoma cell
lines.35 These findings are supported by the current
study, in which it was shown that human osteosarcoma tissue expresses TGF-b. In general there was
higher expression of TGF-b1 and TGF-b3 than TGF-
/ 7b89$$1515
11-25-97 12:52:41
cana
b2 in the osteosarcoma cells and their surrounding
stroma. The absence of TGF-b in mineralized bone
also has been noted in another immunohistochemical
study.41 It is possible that once TGF-b becomes embedded in the calcified matrix the antibody is unable
to bind to the protein. The fact that bone is an abundant source of TGF-b supports this. Alternatively, TGFb may not be present in ossified bone. It is interesting
to note that although TGF-b has been shown to stimulate extracellular matrix production, it inhibits matrix
mineralization for reasons that currently are unclear.10
Areas of active remodeling and/or tumor invasion
showed high expression of TGF-b in the local osteoclasts. The presence of TGF-b in the osteoclasts may
be due to autocrine production by the osteoclasts, or
to the binding of TGF-b stored in the matrix. Because
TGF-b is secreted as a latent molecule, the regulation
of TGF-b activity is dependent on whether it is locally
activated. Although the antibody used in these studies
cannot differentiate active from latent TGF-b, it is
likely that at least part of the TGF-b released by solubilization of the matrix becomes activated by the acidic
pH and lysosome-like microenvironment below the
W: Cancer
TGF-b Expression in Osteosarcomas/Kloen et al.
TABLE 3
Expression of TGF-b Isoforms in Cancer Related to Disease
Progressiona
Pancreatic carcinoma12
Predominantly autocrine mode of action
TGF-b2 associated with advanced tumor stage
Lack of expression of all TGF-b isoforms correlates with increased
survival
Colon carcinoma21,25
Intense immunoreactivity for TGF-b1 correlates with progression
Endometrial carcinoma14
Switch from paracrine/autocrine to predominantly paracrine
mechanism of action
Glandular cells lost ability to respond to stromal cells
TGF-b1 and TGF-b2 expression is associated with malignancy
Breast carcinoma15
Intense immunoreactivity for TGF-b1 correlates with progression
Gliomas (unpublished data)
Predominantly autocrine mode of action
TGF-b1 as a marker for carcinogenesis
High expression correlates with short term survival in glioblastoma
Osteosarcomasb
Intense immunoreactivity for TGF-b3 correlates with progression
TGF-b: transforming growth factor-b.
a
Modified from Gold LI, Korc M. Expression of transforming growth factor-b1, 2 and 3 mRNA and
protein in human cancers. Dig Surgery 1994;11:150–6.
b
Current study.
ruffled border of the osteoclast.45 A number of mechanisms can be responsible for the increase of TGF-b in
the tumorous bone and osteosarcoma cells, including
increased production by the osteosarcoma cells induced by an unknown factor, release of bound TGFb by solubilization of the mineralized bone and tumor
matrix by osteoclasts and tumor cells, and/or uptake
from the bloodstream. The last explanation is unlikely,
because of the three isoforms only TGF-b1 is detectable at physiologically significant levels in human
plasma.46 It is known that osteoblast and osteosarcoma cells have high affinity receptors for TGF-b,6 and
that they proliferate and produce extracellular matrix
proteins in response to TGF-b at levels as low as 0.1
ng/ml.5,9 Thus, whether the increased presence of
TGF-b results from increased autocrine production of
TGF-b by the osteosarcoma cells or from release of
stored TGF-b from the matrix, a significant effect on
the osteosarcoma cells and the normal osteoblasts can
be expected. Moreover, TGF-b isoforms are auto-inductive and cross-inductive,43 adding an endogenous
amplification loop. The angiogenic activity of TGF-b
has been well documented.27,28 The presence of multiple small vessels that were immunoreactive for the
different TGF-b isoforms in both the endothelial and
perivascular layers suggest that this growth factor also
/ 7b89$$1515
11-25-97 12:52:41
cana
2237
may have angiogenic activity in osteosarcomas, further promoting the growth of these tumors.
Currently, the expression and distribution of
TGF-b in human osteosarcoma tissue has not been
reported. Two groups have examined expression of
certain growth factors in human osteosarcomas;
BMP-2 and BMP-4, which are members of the TGFb superfamily, were found to be expressed in a subset of human osteosarcomas, with correlation to
histologic subtype and poor clinical outcome and
response to chemotherapy.36 – 38 More recently, increased expression of IGF-1, IGF-2, and IGF-1 receptors (IGF-1 and IGF-2) was found in human osteosarcomas, but not in a nonosseous normal cell
line. 33 The most important prognostic variable in
high grade nonmetastatic conventional osteosarcoma was reported to be tumor necrosis after neoadjuvant chemotherapy.47 Whereas mutations in
tumor suppressor genes such as p53 and Rb have
been shown to be involved in carcinogenesis, p53
mutations have not yet been proven to be of prognostic value in osteosarcoma patients.31 However,
loss of heterozygosity of the Rb gene appeared to
be an early predictive feature of osteosarcoma30 although no information regarding the grade of the
tumor was given. Of recent significance, elevated
expression of P-glycoprotein, a membrane-bound
channel important for drug binding and transport
out of the cell, has been shown to be related to
disease progression.48
This study presented data on the expression,
prevalence, and distribution of TGF-b isoforms in
human osteosarcomas. Significant levels of TGF-b
isoforms are present in this neoplasm, suggesting
an autocrine/paracrine role for this protein in malignancy. Expression of TGF-b 3 was shown to be
related to disease progression, and therefore could
be used as a possible (prognostic outcome) marker
to subclassify high grade osteosarcomas. Whether
the increased expression of this multipotential
growth factor is the cause or merely the result of
increased cellular proliferation is unclear from this
study. Most likely it is part of a complex multistep
process involving alterations in the response to
TGF-b at different levels of tumorigenesis and progression.
REFERENCES
1.
2.
Roberts AB, Sporn MB. Peptide growth factors and their
receptors. In: Roberts AB, Sporn MB, editors. Handbook of
experimental pharmacology. Heidelberg: Springer-Verlag,
1990:419–72.
Assioan RK, Komoriya A, Meyers CA, Miller DM, Sporn MB.
Transforming growth factor-beta in human platelets. J Cell
Biochem 1983;258:7155–66.
W: Cancer
2238
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
CANCER December 15, 1997 / Volume 80 / Number 12
Heine UI, Munoz EF, Flanders KC, Ellingsworth LR, Lam HYP, Thompson NL, et al. Role of transforming growth factorß in the development of the mouse embryo. J Cell Biol 1987;
105:2861–76.
Pelton RW, Saxena B, Jones M, Moses HL, Gold LI. Immunohistochemical localization of TGFß1, TGFß2, and TGFß3 in
the mouse embryo: expression patterns suggest multiple
roles during embryonic development. J Cell Biol 1991;115:
1091–105.
Centrella M, Horowitz MC, Wozney JM, McCarthy TL. Transforming growth factor-ß gene family members and bone.
Endocr Rev 1994;15:27–39.
Robey PG, Young MF, Flanders KC, Roche NS, Kondaiah
P, Reddi HA, et al. Osteoblasts synthesize and respond to
transforming growth factor-type ß (TGF-ß) in vitro. J Cell
Biol 1987;105:457–63.
Beck LS, Amento EP, Deguzman L, Lee WP, Ngyen T, Gillett
NA. TGF-ß1 induces bone closure of skull defects: temporal
dynamics of bone formation in defects exposed to rh-TGFß1. J Bone Miner Res 1993;8:753–61.
Beck LS, Deguzman L, Lee WP, Xu Y, McFatridge LA, Gillett
NA, et al. TGF-ß1 induces bone closure of skull defects. J
Bone Miner Res 1990;6:1257–65.
Joyce ME, Roberts AB, Sporn MB, Bolander ME. Transforming growth factor-ß and the initiation of chondrogenesis and osteogenesis in the rat femur. J Cell Biol 1990;110:
2195–207.
Erlebacher A, Derynck R. Increased expression of TGF-ß2 in
osteoblasts results in an osteoporosis-like phenotype. J Cell
Biol 1996;132:195–210.
Ciernik FI, Krayenbuhl-Ciernik BH, Cockerell CJ, Minna JD,
Gazdar AF, Carbone DP. Expression of transforming growth
factor ß receptors on AIDS-associated Kaposi’s sarcoma.
Clin Cancer Res 1995;1:1119–24.
Friess H, Yamanaka Y, Büchler M, Ebert M, Beger HG, Gold
LI, et al. Enhanced expression of transforming growth factor
ß isoforms in pancreatic cancer correlates with decreased
survival. Gastroenterology 1993;105:1846–56.
Gold LI, Korc M. Expression of transforming growth factorß1,2 and 3 mRNA and protein in human cancers. Dig Surgery
1994;11:150–6.
Gold LI, Saxena B, Mittal KR, Marmor M, Goswami S, Nactigal L, et al. Increased expression of transforming growth
factor ß isoforms and basic fibroblast growth factor in complex hyperplasia and adenocarcinoma of the endometrium:
evidence for paracrine and autocrine action. Cancer Res
1994;54:2347–58.
Gorsch SM, Memoli VA, Stukel TA, Gold LI, Arrick BA. Immunohistochemical staining for transforming growth factor ß1
associates with disease progression in human breast cancer.
Cancer Res 1992;52:6949–52.
Jennings MT, Kaariainen IT, Gold L, Maciunas RJ, Commers
PA. TGF-ß1 and TGF-ß2 are potential growth regulators for
medulloblastomas, primitive neuroectodermal tumors, and
ependymomas: evidence in support of an autocrine hypothesis. Hum Pathol 1994;25:464–75.
Jennings MT, Maciunas RJ, Carver R, Bascom CC, Juneau P,
Misulis K, et al. TGF-ß1 and TGF-ß2 are potential growth
regulators for low-grade and malignant gliomas in vitro: evidence in support of an autocrine hypothesis. Int J Cancer
1991;49:129–39.
Johnson MD, Federspiel CF, Gold LI, Moses HL. Transforming growth factor-ß and Transforming growth factor-ß-
/ 7b89$$1515
11-25-97 12:52:41
cana
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
receptor expression in human meningioma cells. Am J Pathol 1992;141:633–42.
McCune BK, Patterson K, Chandra RS, Kapur S, Sporn MB,
Tsokos M. Expression of transforming growth factor-ß isoforms in small round cell tumors of childhood. Am J Pathol
1993;142:49–59.
Nakatsukasa H, Evarts RP, Hsia C, Marsden E, Thorgeirsson
SS. Expression of transforming growth factor ß1 during
chemical hepatocarcinogenesis in the rat. Lab Invest 1991;
65:511–7.
Robson H, Anderson E, James RD, Schofield PF. Transforming growth factor ß1 expression in human colorectal
tumours: an independent prognostic marker in a subgroup
of poor prognosis patients. Br J Cancer 1996;74:753–8.
Schwarz LC, Wright JA, Gingras M, Kondaiah P, Danielpour
D, Pimentel M, et al. Aberrant TGF-ß production and regulation in metastatic malignancy. Growth Factors 1990;3:115–
27.
Steiner MS, Zhou Z, Tonb DC, Barrack ER. Expression of
transforming growth factor-ß1 in prostate cancer. Endocrinology 1994;135:2240–7.
Takanami I, Imamura T, Hashizume T, Kikuchi K, Yamamoto Y, Kodaira S. Transforming growth factor ß1 as a prognostic factor in pulmonary adenocarcinoma. J Clin Pathol
1994;47:1098–100.
Tsushima H, Kawata S, Tamura S, Ito N, Shirai Y, Kiso S, et
al. High levels of transforming growth factor ß1 in patients
with colorectal cancer: association with disease progression.
Gastroenterology 1996;110:375–82.
Williams AO, Ward JM, Li JF, Jackson MA, Flanders KC. Immunohistochemical localization of transforming growth factor-ß1 in Kaposi’s sarcoma. Hum Pathol 1995;26:469–73.
Folkman J. Clinical applications of research on angiogenesis.
N Engl J Med 1995;333:1757–63.
Roberts AB, Sporn MB, Assoian RK. Transforming growth
factor type ß: rapid induction of fibrosis and angiogenesis
in vivo and stimulation of collagen formation in vitro. Proc
Natl Acad Sci USA 1986;83:4167–71.
Torre-Amione G, Beauchamp RD, Koeppen H, Park BH,
Schreiber H, Moses HL. A highly immunogenic tumor transfected with a murine transforming growth factor type b1
cDNA escapes immune surveillance. Proc Natl Acad Sci USA
1990;87:1486–90.
Fuegeas O, Guriec N, Babin-Boilletot A, Marcellin L, Simon
P, Thyss A. Loss of heterozygosity of the RB gene is a poor
prognostic factor in patients with osteosarcoma. J Clin Oncol
1996;14:467–72.
Grundmann E, Rossner A, Ueda Y, Schneider-Stock R, Radig
K. Current aspects of the pathology of osteosarcoma. Anticancer Res 1995;15:1023–33.
Miller CM, Aslo A, Tsay C, Slamon D, Ishizaki K, Toguchida
J, et al. Frequency and structure of P53 rearrangements in
human osteosarcoma. Cancer Res 1990;50:7950–4.
Sekyi-Otu A, Bell RS, Ohashi C, Pollak M, Andrulis IL. Insulin-like growth factor 1 (IGF-1), and IGF-2 are expressed in
primary human sarcomas. Cancer Res 1995;55:129–34.
Graves DT, Owen AJ, Barth RK, Tempst P, Winoto A, Fors
L. Detection of c-cis transcripts and synthesis of PDGF-like
proteins by human osteosarcoma cells. Science 1984;226:
972–4.
Kloen P, Jennings CL, Gebhardt MC, Springfield DS, Mankin
HJ. Expression of transforming growth factor-beta (TGF-ß)
receptors, TGF-ß1 and TGF-ß2 production and autocrine
growth control in osteosarcoma cells. Int J Cancer 1994;58:
440–5.
W: Cancer
TGF-b Expression in Osteosarcomas/Kloen et al.
36. Bauer FCH, Urist MR. Human osteosarcoma-derived soluble
bone morphogenetic protein. Clin Orthop 1980;154:291–5.
37. Yoshikawa H, Rettig WJ, Lane JM, Takaoka K, Alderman E,
Rup B, et al. Immunohistochemical detection of bone morphogenetic proteins in bone and soft-tissue sarcomas. Cancer 1994;74:842–7.
38. Yoshikawa H, Rettig WJ, Takaoka K, Alderman E, Rup B,
Rosen V, et al. Expression of bone morphogenetic proteins
in human osteosarcoma. Cancer 1994;73:85–91.
39. Enneking WF, Spanier SS, Goodman MA. A system for the
surgical staging of musculoskeletal sarcoma. Clin Orthop
1980;153:44–6.
40. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–81.
41. Thorp BH, Anderson I, Jakowlew SB. Transforming growth
factor-ß1, -ß2, -ß3 in cartilage and bone cells during endochondral ossification in the chick. Development 1992;114:
907–11.
42. Roberts AB, Sporn MB. TGF-ß: possible roles in carcinogenesis. Br J Cancer 1988;57:594–600.
43. VanObberghen-Schilling E, Roche NS, Flanders KC, Sporn
MB, Roberts AB. Transforming growth factor-ß1 positively
/ 7b89$$1515
11-25-97 12:52:41
cana
44.
45.
46.
47.
48.
2239
regulates its own expression in normal and transformed
cells. J Biol Chem 1988;263:7741–6.
Chang H, Gillett N, Figari I, Lopez AR, Palladino MA, Derynck R. Increased transforming growth factor ß expression
inhibits cell proliferations in vitro, yet increases tumorigenicity and tumor growth of Meth A sarcoma cells. Cancer
Res 1993;53:4391–8.
Barton R, Neff L, Louvard CA, Courtoy PJ. Cell-mediated
extracellular acidification and bone resorption: evidence for
a low pH in resorbing lacunae and localization of a 100
kD lysosomal membrane protein at the osteoclast ruffled
border. J Cell Biol 1985;101:2210–22.
Wakefield LM, Letterio JJ, Chen T, Danielpour D, Allison
RSH, Pai LH, et al. Transforming growth factor-ß1 circulates
in normal human plasma and is unchanged in advanced
metastatic breast cancer. Clin Cancer Res 1995;1:129–36.
Davis AM, Bell RS, Goodwin RJ. Prognostic factors in osteosarcoma: a critical review. J Clin Oncol 1994;12:423–31.
Baldini N, Scotlandi D, Barbanti-Brodani G, Manara MC,
Aurici D, Bacci G, et al. Expression of P-glycoprotein in highgrade osteosarcoma in relation to clinical outcome. N Engl
J Med 1995;23:1380–5.
W: Cancer
Документ
Категория
Без категории
Просмотров
3
Размер файла
744 Кб
Теги
931
1/--страниц
Пожаловаться на содержимое документа