close

Вход

Забыли?

вход по аккаунту

?

Analysis of prognostic factors after the first relapse of Hodgkin's disease in 187 patients

код для вставкиСкачать
1272
Differences between Phosphotyrosine Accumulation
and Neu/ErbB-2 Receptor Expression in Astrocytic
Pro1iferative Processes
Implications for Glial Oncogenesis
Donald A. Kristt, M.D.”’
Yosef Yarden, P ~ . D . ~
‘
Institute of Pathology, Rabin Medical Center,
Beilinson Campus, Petach Tikvah, Israel: Department of Pathology, University of Maryland
School of Medicine, Baltimore, Maryland.
* Department of Chemical Immunology, The
Weizmann Institute, Rohovot, Israel.
Presented in part at the 69th Annual Meeting of
the American Association of Neuropathologists,
Washington, DC, June 3-6, 1993.
Partially supported by the Minna and James
Heineman Foundation, and NIH contract
NICHD28960.
The authors wish to thank Jannitt Simons, Arielle Sontag, and Dr. Mohamed Abdel-Hamid for
technical assistance, and Dr. Ray Koski for the
monoclonal antibody to neu.
Address for reprints: Donald A. Kristt, M.D., Institute of Pathology, Rabin Medical CenterBeilinson Campus, 49100 Petach-Tikvah, Israel.
Received January 15, 1996; revision received
June 7, 1996; accepted June 7, 1996.
0 1996 American Cancer Society
BACKGROUND. Previous work has shown that enhanced growth potential of malignant astrocytomas correlates with increased expression of growth factor receptor
tyrosine kinases. The functional implications of increased receptor expression were
addressed by analyzing possible accumulation of phosphotyrosyl proteins in neoplastic and nonneoplastic astrocytic proliferative processes. The results were compared with the expression of Neu receptor protein (also called ErbB-2 or HER-2).
METHODS. Western immunoblots and immunocytochemistry were utilized to evaluate glioma and carcinoma cell lines, neonatal astrocytic cultures, and human
brain biopsies of graded gliosis and astrocytomas. The effects of three tyrosine
kinase inhibitors on 3H-thymidine uptake and cell proliferation and viability were
examined in cultured glioma cells.
RESULTS. Phosphotyrosine was conspicuously elevated in all three grades of astrocytoma, but remained at low levels in nonneoplastic astrocytic proliferations. Dosedependent decreases in DNA synthesis and proliferation of cultured glioma cells
occurred after inhibition of tyrosine kinase. Neu receptor protein showed increased
expression in malignant astrocytomas (including glioblastomas) and severe reactive gliosis.
CONCLUSIONS. Upregulation of tyrosyl protein phosphorylation enables differentiation of neoplastic from nonneoplastic astrocytic proliferative states. Inhibition of
this phosphorylation impairs growth of glioma cells. Increased Neu receptor protein expression can distinguish malignant from low grade astrocytomas. We speculate that genetic events leading to stably increased phosphotyrosine may be critical
for neoplastic transformation of astrocytes, whereas increased receptor tyrosine
kinase expression could be a factor in the aggressive growth associated with malignancy. Cancer 1996;781272-83. 0 1996 American Cancer Society.
KEYWORDS gliosis, astrocytoma, Neu, ErbB-2, HER-2, tyrosine kinase, phosphotyrosine, astrocyte, tyrosine kinase inhibitor, gliogenesis.
A
strocytes normally are capable of a regulated proliferative response that results in gliosis. Aberrations in the regulatory controls
for astrocytic growth and proliferation give rise to astrocytic neoplasms, astrocytomas. Although the molecular basis for impaired proliferation control in astrocytomas is incompletely understood, considerable previous work focused on growth factor receptors.
Activation of receptor proteins of the cell membrane, in response to ligands with diverse growth factor activities, has pronounced effects on growth and proliferation in many cell types,
including astro~ytes.l-~
One well studied family of receptors with
Phosphotyrosine and Neu in Astrocytoma/Kristt and Yarden
1273
METHODS
intracellular tyrosine kinase activity is collectively
Cell Lines
referred to as receptor tyrosine kinases. The kinase
The U251-MG malignant human glioma cell line was
activity is stimulated when the extracellular domain
kindly made available by Dr. D. D. Bigner of Duke
of the receptor binds to a ligand, which then results
University. A human glioblastoma cell line, A172, and
in phosphorylation of cytoplasmic tyrosyl proteins?-4
the MDA-MB-453 human breast carcinoma cell line
Examples include receptors for epidermal growth
that highly expresses Neu receptor
were obfactor (EGF), Neu differentiation factor/heregulin,
tained from the American Type Culture Collection,
platelet-derived growth factor, and stem cell factor
Rockville, Maryland. U251 cells were grown to conflu(SCF).6-'2
ence in low glucose DMEM (Gibco, Grand Island, NY)
Several lines of evidence suggest that abnormal
with 10% fetal calf serum at 37 "C under 5% carbon
growth regulation in malignant astrocytic neoplasms
dioxide/95% air; A172 cells in normal glucose-DMEM;
is related to increased receptor tyrosine kinase expresand MDA cells in RPMI 1640 (Gibco, Grand Island, NY)
sion. Increased expression of the EGF and other tyroin 20% fetal calf serum. The media contained penicillin
sine kinase receptors occurs preferentially in malignant astrocytomas and g l i o b l a ~ t o m a ~ . ~ AUg~ ~ " ~ - ' ~and streptomycin (Gibco).Cells were lysed and detergent-solubilized as described.2zThe lysis buffer conmented expression is associated with amplification
tained sodium orthovanadate (1mM) to block dephosand rearrangements of the receptor
as well
phorylation. The lysates were rapidly frozen and stored
as oncogenic r n ~ t a t i o n ~ . "Functional
-~~
relevance of
overexpression is supported by receptor inhibition
at -70 "C until used.
studies showing impaired growth of nonglial tumor
cell lines and ~enografts.'~-~'
However, in interpreting
Primary Astrocytk Cultures
these findings, it should be recalled that increased exAstrocytes from the cerebrum of newborn CD/BR rats
pression of receptor tyrosine kinases has been ob(Charles River, Willmington, MA) were cultured for 15
served in purely reactive astrocytic proliferative states
days according to the method of Robbins et
This
in humans and rat^.^!^' It is also likely to occur during
method yields at least 98% glial fibrillary acidic protein
gliogene~is.'~,~~-~~
(GFAP) positive, phenotypic astrocytic cells.
The range of situations associated with increased
receptor expression prompted an investigation of the
Biopsy Samples
functional implications of receptor expression in these
Forty human brain biopsies were obtained intraoperadifferent astrocytic proliferative processes. We hypothtively. This material was fixed, processed for paraffin
esized that it would be possible to distinguish inembedding, cut, stained with hematoxylin and eosin
creased receptor expression in neoplasia from other
(H & E) or immunoreacted. Diagnosis and grade of
astrocytic proliferative lesions, based on the activity
astrocytoma were based on the criteria suggested by
of signal transduction pathways. This study was unthe new World Health Organization International Hisdertaken to initiate an examination of that hypothesis
tological Classification of turn our^.^^ There was no
by comparing expression of the astrocytic tyrosine kihistory of previous treatment in any of these cases.
nase receptor protein, p185""", to phosphotyrosine acThe observations on neoplasms were limited to zones
cumulation. These parameters were evaluated in neoof recognizable tumor, in which the microscopic fields
plastic, reactive, and developmental processes, in vivo
were almost completely occupied by tumor. Two inand in vitro. In complementary experiments, we investensity grades of astrocytosis were operationally detigated the effects of tyrosine kinase inhibitors on the
fined, mild and severe; their clinical settings are noted
growth of malignant glioma cells in vitro. Although
in Table 1.
increased phosphotyrosine has been noted in nonglial
malignant cell
many aspects relevant to glial
Antibodies
oncogenesis remain to be examined.
Three monoclonal antibodies to the Neu gene prodExpression of the Neu/ErbB-2 receptor tyrosine
uct, p185""", were prepared as described previously.28
kinase was analyzed in this study because Neu is stimDetails of the preparation of antisera to ~185"""and
ulated by a family of glial growth factor^.",^^-^^ Mutant
phosphotyrosine (PT4)have been reported.22The antiforms have been characterized that possess constitusera to Neu were made against the carboxy terminal
tive activity with transformation in v i t r ~ . ~Also,
~ - ' ~Neu
of the receptor protein, in which the latter was in the
may exhibit a higher oncogenic potential than the EGF
form of a synthetic peptide." The specificity of the
receptor, because it is a potentiator of the EGF signal4'
antisera was tested by immunoprecipitation and Westand is a more potent mitogenic transducer for some
ern blotting that showed a single band that reflected
cell types.42
their respective specificity. A monoclonal antibody to
1274
CANCER September 15,1996 / Volume 78 / Number 6
TABLE I
Immunoreactivity of Astrocytic Lesions
No
CMd
Age (PI
Sex
Freq
Intens
Freq
Intens
AVM
Seizures
unknown
47
28
39
M
M
F
0
0
0
0
0
0
0
0
0
0
0
0
Metastasis
Absces
Granuloma
Abscess
Trauma"
Recent infarction
Lymphoma
Abscess
Recent infarctiona
Abscess
Recent infarction
Lymphoma
45
29
51
13
6
45
71
58
71
75
47
57
F
M
F
M
M
M
F
M
M
F
0
0
0
lt
0
I+
2t
0
0
0
0
0
0
0
0
10
2t
21
It
3t
31.
2t
3t
2t
lt
2t
2t
4t
3t
It
3t
4t
3t
3131
2t
It
2t
4t
4t
Cerebellar (diffuse, fibrillary)
Frontal
Frontal (gemistocytic)
Parietal
Parietal
5
47
47
13
12
M
F
M
4t
2t
4t
4t
4t
0
F
3t
3t
4t
3t
4t
0
0
It
0
lt
Frontal
Temporal
Frontal
Parietal
Occipital
25
64
55
64
30
M
F
F
M
M
4t
3t
4t
3t
2t
4t
It
2t
4t
It
2t
3t
3t
4t
3t
3t
3t
3t
2t
Occipital
Frontal
Frontal
Temporal
Frontal
Temporal
Thalamus
Temporal
Temporal
Occipital
68
57
45
52
58
55
50
54
48
66
M
M
F
F
F
F
F
M
M
M
4t
3t
2t
3t
4t
4t
2t
4t
4t
4t
3t
2t
2t
4t
4t
4t
2t
4t
4t
4t
2t
4t
4t
3t
It
4t
2t
3t
3t
4t
It
4t
4t
3t
2t
4t
2t
3t
4t
4t
Mild gliosis
1
2
3
Severe gliosis
4
5
6
7
8
9
10
11
12
13
14
15
Astrocpoma
16
17
18
19
20
Anaplasric astrocytoma
21
22
23
24
25
Glioblastoma
26
27
28
29
30
31
32
33
34
35
h.I
IVl
M
lt
It
0
0
0
0
0
0
2t
0
1t
3t
No: specimen number; Freq: relative frequency, Intens: relative intensity of immunostained cells with respect to glid fibrillary acid protein of 4; M male; F female;AW. menovenous malformation.
Miloticallyactive cells present.
AU patients are World Health Organization Grade 2."
phosphotyrosine, 05-321, and a second polyclonal antibody to phosphotyrosine, 06-123,46were obtained
from Upstate Biotechnology Inc. (Lake Placid, NM.
The 06-123 antibody to phosphotyrosine was used at
a 1:5 dilution, which gave results comparable to PT4.
Anti-GFAP was obtained as a mouse antihuman
monoclonal antibody (DAKO, Santa Barbara, CAI.
lmmunoprecipitation and lmmunoblotting
Sepharose-protein A bearing rabbit antimouse immunoglobulin IgG (fc fragment specific) was complexed
to a monospecific antibody, and used for immunoprecipitation. This was followed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis using a 7.5%
gel and subsequent Western immunoblotting with the
above noted antisera. In some experiments, the lysates
were immunoprecipitated using a single monospecific
antibody, but immunoblotted with antisera to different antigens, by separating the lanes after transfer to
nitrocellulose. Isotype control utilized a mouse immunoglobulin, IgG (Sigma, St. Louis, MO). The proteins
were detected on the immunoblot using horseradish
Phosphotyrosine and Neu in AstrocytomalKristt and Yarden
1275
TABLE 2
Statistical Findings (Based on Scores in Table 1): Frequency of Immunoreactive Cells
Phosphotyrosine
SG
A
AA
Neu
A
AA
GBM
A
AA
GBM
P < O.OOla
P < o.oOla
(U = 0)
(U = 0)
P < 0.001“
(U= 2)
-
P > 0.579
P < 0.00P
(U= 1)
P > 0.05
P > 0.05
(U= 23)
P < .illa
P > 0.05
(U= 26)
P < O.Ola
(U= 2)
P > 0.05
-
(U = 15)
-
(U = 16)
P > 0.05
(U = 14)
-
(U= 1)
-
-
(U= 15)
A astrocytoma; AA: anaplastic astrocyioma; GBM: glioblastoma mulliforme; SG: severe gliosis.
a
Statisticallysignificant differences between KIOUPS, Mann-Whitney U test, one-tailed.
cells were allowed to grow for 24 hours, and then incubated for 26 hours with the inhibitor. For genistein
and herbimycin, cells were grown for 34 hours, and
exposed to an inhibitor for the next 16 hours. 3H-thymidine was added at the beginning of the last 8 hours.
Control cultures were grown without exposure to inhibitor; some were exposed to the drugs’ solvent, dimethylsulfoxide, without an inhibitor. All conditions
lmmunocytochemistry
were performed in triplicate. At the termination of the
Phosphotyrosine, the Neu receptor protein ( ~ 1 8 5 ~ ~ ~experiments,
1,
cells were either photographed and
and GFAP were immunocytochemically detected in
counted, assessed for cell viability based on trypan
adjacent sections. Four cases had no GFAP immunoreblue exclusion using a hemocytometer, or measured
activity and were not analyzed further based on the
for radioactivity. Cell counts were made from microassumption of poor antigen preservation. Cell cultures
photographs (x16) of the center of 3 wells/condition
were made into cell blocks, embedded in paraffin, and
in 96 well plates. Results were averaged and expressed
then treated the same as biopsies. The methods have
as percent of nontreated control. Viability estimates
been detailed previou~ly.~
In brief, primary antibody
were based on the average number of cells of three
was detected by a streptavidin-biotin peroxidase
80-cell samples that showed trypan blue exclusion;
method using the Immunon Maxitag kit according
nonviable cells were stained blue. Data are expressed
to the manufacturer’s instructions (Lipshaw, Pittsas percent of sampled cells.
burgh, PA).
Diaminobenzidine hydrochloride was used as the
Thymidine Incorporation Assay
chromogen. Negative controls underwent identical
3H-thymidine uptake was examined after exposure of
processing except for the omission of primary anticells to either tyrphostin or genistein. Noninhibited
body.
cells served as controls. Into each well 0.4 curie (Ci)
of 3H-thymidine (2 Ci/mmol) was added. After incubaTyrosine Kinase Inhibition Studies
tion for an additional 8 hours (as explained above),
Three inhibitors (Gibco Inc., Grand Island, NY) were
the cells were washed with cold phosphate-buffered
used at the doses and duration of exposure suggested
saline and then lightly trypsinized. The radioactivity
by previous studies and our own preliminary work.47 in each well was measured with an LKB betaplate scinCells were exposed to 4 or 5 doses of each inhibitor
tillation counter. Triplicate replications were peras follows: herbimycin A, 1- 15 ,uM4’; tyrphostin, 28formed for each condition.
450 ,uM4’; and genistein, 2-200 ,ug/mL (7.4-740 P M ) . ~ ~
‘IJ251 malignant glioma cells were cultured in 96 well
Microscopic Analysis of Biopsies
plates, at a density of 5 x lo4 cells/well. To study all
Our approach to sampling variability problems in a
treatment conditions after an equal time in culture,
given lesion has been discussed at length previously.’
but different durations of inhibitor exposure, the folIn brief, large sections or multiple biopsies were utilowing schedule was used. For tyrphostin exposure,
lized with good clinical pathologic correlations. In all
peroxidase (HRP)-tagged antibodies or HRP-conjugated to protein A, and visualized using the ECL
chemiluminescence method (Amersham Corp, Arlingion Hts, IL). Total protein content in cell lysates was
determined using a protein assay kit (Pierce Chemical
Co., Rockford, IL) to load approximately equivalent
protein for each sample.
1276
CANCER September 15,1996 / Volume 78 / Number 6
FIGURE 1. Glioblastoma multiforme immunoreacted for Neu receptor
protein. Large numbers of anaplastic cells are stained (~1000).
cases, the immunostained sections exhibited the same
grade of pathology as the H & E diagnostic section.
All the immunostained preparations reported in
Table 1 were evaluated using a semiquantitative scale
of 0 to 4+ (highest) relative to GFAP staining in an
adjacent section. Two parameters were evaluated: relative "frequency" of stained cells and relative "intensity" of reaction product in the majority of stained
cells. In all cases, both the frequency and the intensity
parameters were equal or greater for GFAP than for
phosphotyrosine or ~185""".Absolute counts of frequency of stained cells were also made of representative lesions of each of the four possible positive scores.
The cell counts were made using an ocular micrometer
(5 x 5 grid, 20 ym/box) in a Zeiss standard (Oberkochen, Germany) microscope, at a magnification of
x600. At least 100 cells were counted in 3 or more
randomly chosen fields in regions showing immunoreactivity.
All slides bore only numeric identifiers and were
examined without prior knowledge of the diagnosis.
Many cases were randomly reevaluated by the same
observer with results identical to the original examination. The immunocytochemical reaction for some
cases also was independently evaluated by two observers who reported comparable scores.
Statistical Analysis
The semiquantitative scores of immunocytochemical
staining were compared using the nonparametric
Mann-Whitney U-te~t,~'
as seen in Table 2. Bar graphs
depict the mean and standard deviation.
RESULTS
Three types of experiments were undertaken utilizing
either immunocytochemistry on biopsies and cell
FIGURE 2. Severe astrocytosis, secondary to recent infarction, immunoreacted for Neu receptor protein. Reactive astrocytes are intensely imrnunostained ( ~ 1 0 0 0 ) .
FIGURE 3. Low grade astrocytoma immunoreacted for Neu protein. Rare
astrocyte shows minimal irnrnunoreactivity ( ~ 1 0 0 0 ) .
lines, immunoblotting of proteins in cells lines, or in
vitro inhibition of tyrosyl protein phosphorylation. Because the work on biopsies provides an overview of
the relationship between phosphotyrosine and Neu in
the different astrocytic proliferative states, these results will be described first.
Neu Expression in Biopsies
The antisera to Neu receptor protein, p185""", recognize the carboxy terminal of the receptor protein. Immunocytochemical detection of this protein in brain
biopsies revealed elevated expression in high grade
astrocytomas (Fig. l),and severe reactive gliosis (Fig.
2). It was generally very low or absent in low grade
tumors (Fig. 3 ) . The semiquantitative evaluation is
Phosphotyrosine and Neu in AstrocytomalKristt and Yarden
PHOSPHOTYROSINE I N ASTROCYTIC LESIONS
NEU EXPRESSION IN ASTROCYTIC LESIONS
Frequency of lrnrnunoreactive Cells
Frequency of lrnmunoreactive Cells
4'0
.r
F
1277
-
30
30
n
U
*
VI
>
.
v
W
e
W
8 2 0
g 2 0
Ln
v)
z
z
3
I
10
10
00
MG
SG
A
A1
GBH
DIAGNOSTIC GROUP
FIGURE 4. Mean frequency of cells imrnunoreactive for Neu receptor
protein. The graph is based on the data in Table 1 for each of the diagnostic
groups. The bar for mild gliosis, left-most, is actually at zero. MG: mild
gliosis; SG: severe gliosis; A: astrocytoma (low grade); AA: anaplastic
astrocytoma; GBM: glioblastoma multiforme. The error bars indicate -t
standard deviation.
00
MG
SO
A
AA
GBM
DIAGNOSTIC GROUP
FIGURE 5. Mean frequency of cells immunoreactive for phosphotyrosine. The graph is based on Table 1 for each of the diagnostic groups.
The bar for mild gliosis, left-most, is at zero. MG: mild gliosis; SG: severe
gliosis; A: astrocytoma (low grade); AA: anaplastic astrocytoma; GBM:
glioblastoma rnultiforme.
FIGURE 7. Astrocytoma, low grade, immunoreactive for phosphotyrosine. Almost all astrocytic cells in this tumor show heavy cytoplasmic
immunoreactivity ( ~ 1 5 0 0 ) .
FIGURE 6. Glioblastoma multiforme immunoreacted for phosphotyrosine. There is diffuse, intense irnrnunoreactivity in almost all tumor cells
(x850).
summarized in Table 1and Figure 4. These differences
in expression are statistically significant (Table 2). The
histopathologic findings in relation to the immunoreactivity have been detailed previo~sly.~
Phosphotyrosine in Biopsies
A semiquantitative analysis of the findings is summarized in Table 1 and Figure 5. Both antisera to phos-
photyrosine PT4 and 06-123 provided similar results.
These data show that in both high grade (Fig. 6) and
low grade astrocytomas (Fig. 71, a substantial majority
of histologically neoplastic cells are immunoreactive
for phosphotyrosine, and the intensity of the immunoreactivity per cell is relatively high, Phosphotyrosinerich neoplastic astrocytes are diffusely distributed
across these lesions. No vascular or neuronal structures were stained. Negative controls were similar to
Fig. 8.
Actual cell counts of immunoreactive astrocytes
1278
CANCER September 15,1996 I Volume 78 / Number 6
FIGURE 8. Severe astrocytosis, secondary to an acute abscess, without
immunoreactivity for phosphotyrosine. Despite prominent hypertrophic
astrocytes, no significant cytoplasmic staining is noted. Negative controls
stained similarly ( ~ 1 0 0 0 ) .
indicated a trend comparable to that for the relative
frequency seen in Table 1. A lesion scored as 1+ contained an average of 7% phosphotyrosine-immunoreactive astrocytes (range, 0-11%); 2+: 32% (range, 1245%); 3-t: 51% (range, 33-78%); and 4+: 92% (range,
90- 100%).
In contrast, little or no staining was seen in either
mild or severe gliosis (Fig. 8). Normal astrocytes also
are not immunoreactive. Occasionallyin severe gliosis,
a limited zone within 500 pm of acute hemorrhage,
necrosis, or a nonglial malignancy contained immunoreactive astrocytes. These foci were always smaller
and contained fewer immunoreactive astrocytes than
the region of reactive gliosis defined by Neu receptor
expression or GFAP immunoreactive astrocytes.
Table 2 shows that these differences between reactive and neoplastic lesions are statistically significant.
It should be noted that from a quantitative perspective
alone, a few cases in the severe gliosis group (e.g., Case
10) have a similar irnmunoprofile for phosphotyrosine
and Neu as an occasional anaplastic astrocytomalglioblastoma (e.g., Case 25). In such instances, as noted
above, the qualitative characteristics of the staining
pattern usually affords separating these two types of
processes. Nonetheless, in a given case of a small biopsy (less than 1 mm) taken from a reactive process,
it is possible that immunocytochemical results will not
be diagnostic.
Phosphotyrosine and Neu in Cell Lines
In Western blots of immunoprecipitated phosphotyrosyl proteins, a small number of different sized proteins
FIGURE 9. Western blot comparing U251 cells (Lanes 2, 4, and 5)
with neonatal rat astrocytic cultures (Lanes 1 and 3). All lysates were
immunoprecipitated (IP) with a monoclonal antibody to phosphotyrosine
(PT). After transfer, the nitrocellulose blot was cut into three pieces to
enable individualized immunodetection of proteins. The different antisera
used for irnmunoblotting (16) are indicated beneath the lane numbers.
Approximately equal amounts of protein were loaded in each lane. Molecular weight markers to right are M, ( ~ 1 0 0 0 ) .
were detected from both U251 glioma cells and primary neonatal astrocytic cultures. Higher levels of
these proteins were seen in glioma cells than in neonatal astrocytes for approximately equal amounts of
loaded protein (Fig. 9, Lanes 3 and 4). For instance,
in the 140-145 kilodalton (ma)band, the signal in the
glioma was approximately 10- 15 times greater than in
neonatal astrocytes. The phosphotyrosine-rich protein
band at 185 kDa was examined in a further experiment. Glioma lysates were immunoprecipitated with
antiphosphotyrosine and then immunoblotted with
anti-Neu. It was observed that the p185 band corresponds to Neu (Fig. 9, Lane 5). This is the first clear
indication that Neu is autophosphorylated in malignant glioma cells.
Expression of ~185"""also was examined in the
A172 glioblastoma cell line and MDA-MB-453 breast
carcinoma cell line using Western blots and immunocytochemistry. This carcinoma cell line is known to
. ~ ~
of receptor protein
highly express N ~ u Expression
was seen in both glioma cell lines (U251 and A1721,
but lower than levels in the carcinoma cells (data not
shown).
Effects of Tyrosine Kinase Inhibitors
Three tyrosine kinase inhibitors were used: tyrphostin,
genistein, and herbimycin A. Each has a different
mechanism of inhibition and ~pecificity.~'
The doses
chosen were based on previous studies on nonglial
Phosphotyrosine and Neu in AstrocytomalKristt and Yarden
TABLE 3
Effects of Tyrosine b a s e Inhibitors on Cultured Glioma Cells
~
Inhibitor
None
Tyrphostin
Herbimycin A
~
~~
Dose
Proliferation"
Viabilityb
-
100%
100
71'
53"
45'
94%
96
93
79"
25'
96
92
8gC
27'
28 pM
56
113
450
0.1 jtM
I
Genistein
5
10
15
2 pglmL
I
20
200
89
47=
34c
24'
100
56'
3EC
26'
96
91
86'
37=
'Average of three cell countsicandition of U251-MG cells. Results expressed as percent of nontreated
controls.
Average number of cells that showed trypan blue exclusion, based on 3 cell counts, 80-cell samples;
nonviable cells are stained blue. Data are expressed as percent of sample.
'Statisticaliv significant difference from control, P < 0.001, Mann-Witnev UTest, one-tailed.
neoplastic ce11s26~48~49
and our pilot studies in malignant g l i o m a ~ . ~ ~
In cultured U251 malignant glioma cells, a dosedependent decrease in cell number compared with
control samples was noted with all three drugs, as
shown in Table 3; control samples were grown without
the addition of inhibitor. For instance, cell numbers
were reduced to 53 2 7% of control by 113 pM tyrphostin ( P < 0.001) and 38 2 5% of control by 20 pg/mL
genistein ( P < 0.001) (Fig. 10). Control samples varied
by ? 5%. For the same doses, 3H-thymidine uptake
was reduced to 38% of control levels by tyrphostin ( P
< 0.05) (Fig. 111, and 59%by genistein (P< 0.05) (data
not shown). Viability estimates in these cases, based
on trypan blue exclusion, were 79 2 2% for 113 pM
tyrphostin ( P < 0.001) and 86 2 5% for 20 pglmL
genistein ( P < 0.001) (Table 31. Trypan blue exclusion
assay showed 5-10% nonviable cells in nontreated or
dimethylsulfoxide (vehicle)-treated cells. Herbimycin
produced analogous effects on cell number and viability at a dose of 5 pM ( P < 0.001) (Table 3). For all but
the highest dose, morphology of the cultured cells was
comparable to control cells. At the highest doses, cell
surface was markedly deformed, but apoptosis was not
observed at the single time point evaluated morphologically.
DISCUSSION
In this study, we have found that phosphotyrosine accumulation provided a basis for differentiating
1279
astrocytic neoplasia, of all grades, from gliosis and gliogenesis. The increased phosphotyrosyl protein levels
did not directly correlate with Neu receptor protein
expression throughout these various proliferative processes. The differences between phosphotyrosine accumulation and receptor tyrosine kinase expression
raise a number of issues that have implications for our
understanding of glial oncogenesis.
Distinguishing Astrocytomas
The goal of these studies was to identify possible molecular bases for the impaired growth regulation that
distinguishes astrocytomas from reactive astrocytosis
and developmental gliogenesis. One possible approach to this issue can be based on phosphotyrosine
accumulation. Our evidence suggests that up-regulation of tyrosyl protein phosphorylation occurs to a
greater extent in neoplasia than in nonneoplastic
states. A marked elevation of phosphotyrosine even
discriminates low grade astrocytomas from acute reactive processes. Elevated phosphotyrosine has been reported previously for some malignant cell l i n e ~ .It~ ~ , ~ ~
has not been reported previously for glioma cell lines
or tumors.
The effects of tyrosine kinase inhibitors indicate
some functional consequences of tyrosyl phosphorylation in neoplastic astrocytes. Treated malignant glioma cells in vitro showed a dose-dependent reduction
of DNA synthesis and proliferation. Results were similar for all three inhibitors, each of which functions
with a different mechanism of action." Our findings
are generally comparable to earlier work in nonglial
cell
The data suggest that tyrosyl protein
phosphorylation is probably a critical factor for growth
of astrocytomas. Conversely, the increased levels of
phosphotyrosine in astrocytomas probably do not directly correlate with the rate of proliferation. Observations on mitotic activity and preliminary studies with
cell cycle marker Ki-67 show little cell cycle activity in
both severe gliosis and low grade astrocytomas (unpublished data), despite significant differences in phosphotyrosine accumulation. From these considerations, it is likely that increased tyrosyl protein phosphorylation is an expression of the abnormal control
of proliferation in astrocytomas, but does not directly
affect rate of proliferation.
Three tyrosine kinase inhibitors were used: tyrphostin, genistein, and herbimycin A. Each has a different mechanism of inhibition and specifi~ity.~~
Tyrphostin is a substrate peptide site inhibitor, highly effective against the tyrosine kinase of the EGF
r e ~ e p t o r and
~ ~ , appears
~~
highly specific for protein
tyrosine kinases. Genistein is an adenosine triphosphate site inhibitor with specificityfor protein tyrosine
1280
CANCER September 15,1996 I Volume 78 / Number 6
FIGURE 10. Effects of tyrosine kinase inhibition by genistein on proliferation of malignant glioma cells. At 2 pgimL, average cell counts were
100% of control with 96% of cells excluding trypan blue; at 20 pglmL,
cells counts were 38% of control with 86% excluding trypan blue. Control
cells were grown without inhibitor ( ~ 6 0 0 ) .
kinases (e.g.,EGF receptor, v-fes, v-src) over serine and
threonine k i n a s e ~ . ' ~At. ~the
~ highest dose used here
(200 FglmL), effects on other enzymes have been reported (e.g., topoisomerase and phosphokinase A and
C).26*51
In addition, we found that this dose uniquely
resulted in very high levels of glioma cell death. Herbimycin is known to interact with sulfhydryl groups, and
may lead to alkylation, which inactivates the kinase,
although its exact mode of action is still unresolved. It
has only been reported active against protein tyrosine
kinase~.~*~~'
There also is a clinical implication to these inhibitor studies. Because of the selectively high levels of
phosphotyrosine observed in human gliomas, tyrosine
kinase inhibitors could prove to be an option for therapeutic growth modulation of these tumors. A similar
application of protein kinase C inhibitors already appears to have some clinical ~tility.~'
Our findings regarding neu receptor expression
in malignant gliomas were comparable to previously
reported work on other receptor tyrosine k i n a ~ e s . ~ ~ ~ ~
15.19,20,30
Neu receptor protein was not only increased
in malignant glial tumors, but in reactive processes as
well. A parallel duality of Neu overexpression in both
neoplastic and reactive processes has been reported
for hepatocellular diseases.53The relevant implication
Phosphotyrosine and Neu in Astrocytoma/Kristt and Yarden
ever, €or malignant astrocytomas, increased phosphorylation could be related to increased receptor expression. In the malignant glioma cell lines we examined, the overexpressed Neu receptor contained a
functional kinase. Similar findings have been reported
for the overexpressed SCF receptor in glioblastoma cell
lines,56and EGF receptor in some malignant glioma
cell lines.57
T Y R P H O S T I N INHIBITION OF U251 GLIOMA
EFFECT ON 3H-THYMIDINE UPTAKE
40
lib
OF
0
I
coNc E NTKTI
o N ( p dp0
120
1281
480
FIGURE 11. Effect of tyrphostin as a function of dose on 3H-thymidine
uptake in malignant gliorna cells. Cells were grown in the presence of
inhibitor and then incubated with 3H-thymidine. Control cells were grown
without inhibitor. Each graph point is the average of three replications.
Values for control cells, not exposed to inhibitor, are indicated by the large
triangle; the values for the lowest dose, 28 p M , overlapped the control
values. At 56 p M , 3H-thymidine uptake was decreased to 55% of control,
whereas trypan blue was excluded from 93% of cells. Viability assessments
for other doses are given in Table 3.
here is that receptor expression per se does not provide a basis to distinguish astrocytoma from nonneoplastic proliferative processes. The question of
whether the determinants of augmented receptor expression differ among the various astrocytic proliferative states is still unresolved. The little information
available on this last issue is briefly considered below.
Tumor Grade
Expression of the Neu receptor protein varies as a
function of tumor grade. Low levels are observed in
low grade tumors, whereas distinctly higher levels
characterize malignant astrocytomas and glioblastonias. This finding is common to all receptor tyrosine
kinases that have been evaluated in astrocytic neoplasms.8,9,13,14.15.19.20 Enhanced receptor expression
could represent a m p l i f i ~ a t i o nand/or
~ ~ ’ ~ ~constitutive
oncogenic mutations in the neu protoon~ogene.’~-‘~
Phosphotyrosine accumulation, in contrast, does
not distinguish tumor grade because it is immunocytochemically expressed equally in both low grade and
malignant astrocytomas, including glioblastoma multiforme. We tentatively conclude that low grade astrocytoma is characterized by up-regulation of tyrosyl
protein phosphorylation that is not linked significantly
to the expression of receptor tyrosine kinases. How-
Glial Oncogenesis
Marked elevations in phosphotyrosine distinguished
low grade astrocytic neoplasia from regulated reactive
and developmental astrocytic proliferations. We speculate that genetic events leading to stably increased
phosphotyrosine may be critical for transition to even
low grade astrocytic neoplasia. Indirect support for
this view is in the finding that enhanced dephosphorylation blocks transformation in cell lines bearing a mutant tyrosine kinase gene.‘*
Conversely,the distinction between low grade and
high grade (malignant) astrocytoma can be based on
the dramatic increase in receptor tyrosine kinase expression in malignant t ~ m ~ r ~ .The
~ differen~ ~ ~ ~ ~ tial receptor expression as a function of tumor grade
probably does not result from receptor gene amplifications. The incidence of EGF receptor gene amplifications, for example, is the same in low grade and high
grade astrocyt~mas.’~
Other genetic events presumably underlie the consistently overexpressed receptor
in these malignant tumors, such as constitutive, oncogenic mutations in the neu proto~ncogene.~’-’~
These
latter genotypic alterations may be an important factor
in the increased aggressiveness accompanying the
progression from well differentiated astrocytoma to
malignant status.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
Salomon DS, Brandt R, Ciardiello F, Normanno N. Epidermal growth factor-related peptides and their receptors in
human malignancies. Crit Rev Oncol Hematol 1995;19:183232.
Yarden Y, Ullrich A. Growth factor receptor tyrosine kinases.
Annu Rev Biochem 1988;57:443-78.
Ullrich A, Schlessinger J. Signal transduction by receptors
with tyrosine kinase activity. Cell 1990;61:203-12.
Yarden Y, Kelman Z. Transmembrane signaling receptors
for cytokines and growth factors. Curr Opin Struct Biol
1991;1:582-9.
McCormick F. How receptors turn Ras on. Nature 1993;
363:15-6.
Berdel WE, de Vos S, Maurer J, Oberberg D, von Marschall
Z, Schroeder JK, et al. Recombinant human stem cell factor
stimulates growth of a human glioblastoma cell line expressing c- kit protooncogene. Cancer Res 1992;523498-502.
Bigner SH, Burger PC, Wong AJ, Werner MH, Hamilton SR,
Muhlbaier LH, et al. Gene amplification in malignant human
gliomas and clinical and histopathological aspects. J Neuropath Exp Neurol1988;47:191-205.
1282
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
CANCER September 15, 1996 / Volume 78 / Number 6
Cairncross JG. Astrocytoma. In: Fedoroff S, Vernadakis A,
editors. Astrocytes: cell biology and pathology of astrocytes.
Volume 3. New York Academic Press, 1986:337-56.
Kristt DA, Reedy E, Yarden Y. Receptor tyrosine kinase expression in astrocytic lesions: similar features in gliosis and
glioma. Neurosurgery 1993;33:106- 15.
Peles E, Bacus SS, Koski RA, Lu HS, Wen D, Ogden SG, et
al. Isolation of the NeulHER-2 stimulatory ligand: a 44kd
glycoprotein that induces differentiation of mammary tumor cells. Cell 1992;69:205-16.
Marchionni MA, Goodearl DJ, Chen MS, BerminghamMcDonogh 0, Kirk C, Hendricks M, et al. Glial growth factors
are alternatively spliced erbB2 ligands expressed in the nervous system. Nature 1993;362:312-8.
Richardson WD, Pringle N, Mosley M, Westrnark B, DuboisDalcq M. A role for platelet-derived growth factor in normal
gliogenesis in the central nervous system. Cell 1988;53:30919.
Bruce JN, Duigou GJ, Fisher B. Expression of the epidermal
growth factor receptor in human central nervous system
tumors. In: Furth M, Greaves M, editors. Cancer cells: molecular diagnostics of human cancer. Volume 7. Cold Spring
Harbor: Cold Spring Harbor Press, 1989:363-70.
Jaros E, Perry RH, Adam L, Kelly PJ, Crawford PJ, Kalbag
RM, et al. Prognostic implication of p53 proteins, epidermal
growth factor receptor and Ki-67 labeling in brain tumors.
Br J Cancer 1992;66:373-85.
Orian JM, Vasilopoulos K, Yoshida S, Kaye AH, Chow CW,
Gonazles MF. Overexpression of multiple oncogenes related
to histological grade of astrocytic glioma. Br J Cancer
1992;66:106- 12.
Steck PA, Lee P, Hung M-C, Yung WKA. Expression of an
altered epidermal growth factor receptor by human glioblastoma cells. Cancer Res 1988;48:5433-41.
U HS, Kelley PY, Hatton JD, Shew JY.Proto-oncogene abnormalities and their relationship to tumorigenicity in some
human glioblastomas. J Neurosurg 1989;71:83-90.
Filmus J, Pollak N, Cairncoss JG, Buick RN. Amplified overexpressed and rearranged epidermal growth factor receptor
gene in a human astrocytoma cell line. Biochem Biopkys Res
Commun 1985;131207-15.
Hurt MR, Mossy J, Donovan-Peluso M, Locker J. Amplification of epidermal growth factor receptor gene in gliomas:
histopathology and prognosis. J Neuropatk Exp Neurol
1992;51184-90.
Liebermann TA, Nussbaum HR, Razon N, Kris R, Soreq H,
Whittle N, et al. Amplification, enhanced expression and
possible rearrangement of EGF receptor gene in primary
human brain tumors of glial origin. Nature 1985;313:144-7.
Yamazaki H, Fukui Y, Ueyama Y, Tamaoki N, Kawamoto
T, Tanguchi S, et al. Amplification of the structurally and
functionally altered epidermal growth factor receptor gene
(c-erb B) in human brain tumors. Mol Cell Biol1988;8:181620.
Peles E, Ben-Levy R, Or E, Ullrich A, Yarden Y. Oncogenic
forms of the neulHER2 tyrosine kinase are permanently
coupled to phospholipase Cgamma. EMBO J 1991;10207786.
Kiyokawa N, Yan DH, Brown ME, Hung MC. Cell cycle-dependent regulation of p185neu: a relationship between disruption of this regulation and transformation. Proc Natl
Acad Sci USA 1995;92:1092-6.
Brown-Shimer S, Johnson KA, Hill DE, Bruskin AM. Effect
of protein tyrosine phosphatase 1B expression on transfor-
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
mation by the human neu oncogene. Cancer Res 1992;
52:478-82.
Masui H, Kawamoto T, Sato JD, Wolf B, Sato GH, Mendelsohn J. Growth inhibition of human tumor cells in athymic
mice by anti-epidermal growth factor receptor monoclonal
antibodies. Cancer Res 1984;44:1002-7.
Traganos F, Ardelt B, Halko N, Bruno S, Darzykiewicz Z.
Effects of genistein on the growth and cell cycle progression
of normal human lymphocytes and leukemic MOLT-4 cell
and HL-60 cells. Cancer Res 1992;52:6200-8.
Yayon A, Klagsbrun M. Autocrine transformation by a chimeric signal peptide basic fibroblast growth factor: reversal
by Surarnen. Proc Natl Acad Sci USA 1990;87:5346-50.
Stancovski I, Hurwitz E, Leitner 0, Ullrich A, Yarden Y, Sela
M. Mechanistic aspects of the opposing effects of monoclonal antibodies to the erb-BZ receptor on tumor growth.
Proc Nut1 Acad Sci USA 1991;88:8691-5.
Mendelsohn J. Potential clinical applications of anti-EGF
receptor monoclonal antibodies. In: Furth M, Greaves M,
editors. Cancer cells: molecular diagnostics of human cancer. Volume 7. Cold Spring Harbor: Cold Spring Harbor Laboratory, 1989;359-62.
Nieto-Sampedro M. Astrocyte mitogen inhibitor related to
epidermal growth factor receptor. Science 1988;240:1784-6.
Vartanian R, Corfas G, Li Y, Fischbach GD, Stefansson K. A
role for the acetylcholine receptor-inducing protein ARIA
in oligodendrocyte development. Proc Natl Acad Sci USA
1994;91:11626-30.
Pinkas-Kramarski R, Eilam R, Spiegler 0, Lavi S, Liu N,
Chang D, et al. Brain neurons and glial cells express Neu
differentiation factorlheregulin: a survival factor for astrocytes. Proc Nut1 Acad Sci USA 1994;91:9387-91.
Orr-Urtreger A, Avivi A, Zimrner Y, Givol D, Yarden Y, Lonai
P. Developmental expression of c-kit, a proto-oncogene encoded by the W locus. Development 1990; 109:911-23.
Giordano MF, Di Renzo D, Cirillo L, Naldini L, Chiado P,
Comoglio PM. Proteins phosphorylated on tyrosine as markers of human tumor cell lines. I n t J Cancer 1987;39:482-7.
Ogawa R, Ohtsuka M, Sasadaira H, Hirasa M, Yabe H, Uchida
H, et al. Increase of phosphotyrosine containing proteins in
human carcinomas. Gunn 1985;76:1049-55.
Carroway KL, Soltoff SP, Diamonti AJ, Cantley LC. Heregulin
stimulates mitogenesis and phosphatidylinositol3-kinase in
mouse fibroblasts transfected with erbBZ/neu and erbB3. J
BioZ Ckem 1995;2707111-6.
Sliwkowski W, Schaefer G, Akita RW, Lofgren JA, Fitzpatrick
VD, Nuijens A, et al. Coexpression of erbB2 and erbB3 proteins reconstitutes a high affinity receptor for heregulin. J
BioZ Chem 1994;269:14661-5.
Wen D, Suggs SV, Karunagaran D, Liu N, Cupples RL, Luo
Y, et al. Structural and functional aspects of the multiplicity
of Neu differentiation factors. Mol Cell Biol 1994;14:190919.
Ben-Baruch N, Yarden Y. Neu differentiation factors: a family of alternatively spliced neuronal and mesenchymal factors. Proc SOC Exp Biol Med 1994;206:221-7.
Peles E, Yarden Y. Neu and its ligands: from an oncogene
to neural factors. Bioessays 1993;153315-24.
Graus-Porta D, Beerli RR, Hynes NE. Single-chain antibodymediated intracellular retention of ErbB-2 impairs Neu differentiation factor and epidermal growth factor signaling.
Mol Cell Biol 1995;15:1182-91.
Phosphotyrosine and Neu in Astrocytoma/Kristt and Yarden
42. Di Fiore PP, Helin K, Kraus MH, Pierce JH, Artrip J, Segatto
0, et al. A single amino acid substitution is sufficient to
modify the mitogenic properties of the epidermal growth
factor receptor to resemble that of gp185erbB-2. EMBO I
1992;1113927-33.
43. Bacus SS, Stancovski I, Huberman E, Chin D, Hrwiz E, Mills
GF, et al. Tumor-inhibitory monoclonal antibodies to the
44.
45.
46.
47.
48.
HER2/Neu receptor induce differentiation of human breast
cancer cells. Cancer Res 1992;52:1-10.
Robbins DS, Shirazi Y, Drysdale BE, Lieberman A, Shin HS,
Shin M. Production of cytotoxic factor for oligodendrocytes
by stimulated astrocytes. J Irnrnunol 1987;139:2593-7.
Kleihues P, Burger PC, Scheithauer BW. Histological typing
of tumours of the central nervous system, World Health Organization. International histological classification of tumours. Berlin: Springer-Verlag, 1993.
Pang DT, Sharma BR, Shafer JA. Purification of the catalytically active phosphorylated form of insulin receptor kinase
by affinity chromatography with 0-phosphotyrosyl-binding
antibodies. Arch Biochern Biophys 1985;242:176-81.
Kristt DA, Sontag A, Wolf AL, Fiandaca MS, Yarden Y. Neu,
kit receptor, and phosphotyrosine expression: distinguishing astrocytosis from astrocytoma [abstract] Neurosci Abstracts 1993;19:419.
Fukazawa H, Li P-M, Yamamoto C, Murakami Y, Mizuno S,
Uehara Y. Specific inhibition of cytoplasmic protein tyrosine
kinases by herbimycin A in vitro. Biochern Pharrnacol
1991;42:1661-7.
49. Faaland CA, Mermelstein FH, Hayashi I, Laskin JD. Rapid
uptake of tyrphostin into A431 human epidermoid cells is
1283
followed by delayed inhibition of epidermal growth factor
(EGF)-stimulated EGF receptor tyrosine kinase activity. Mol
Cell Biol 1991;11:2697-703.
50. Siege1 S. Nonparametric statistics for the behavioral sciences. New York McGraw-Hill, 1956.
51. Peles E, Yarden Y. Inhibitors of protein tyrosine kinases. In:
Sandler M, Johnsmith H, editors. Design of enzyme inhibitors as drugs. Volume 2. Oxford Oxford Science Publishers,
1994~127-63.
52. Couldwell WT, Weiss MH, DeGorgio CM, Weiner LP, Hinton
DR, Ehresmann GR, et al. Clinical and radiographic response
in a minority of patients with recurrent malignant gliomas
treated with high-dose tamoxifen. Neurosurgery 1993;
32:485-90.
53. Brunt EM, Swanson PE. lmmunoreactivity for c-erbB2 oncopeptide in benign and malignant diseases of the liver. Am J
Clin Pathol 1992;97:S54-62.
54. Heinz NH, Leslie KO, Rogers LA, Howard PL. Amplification
of the c-erbB2 oncogene and prognosis of breast adenocarcinoma. Arch Pathol Lab Med 1990; 14:160-3.
55. Zhau HE, Zhang X, von Eschenbach AC. Amplification and
expression of the c-erbB-2lneu protooncogene in human
bladder cancer. Mol Carcinog 1990;3:254-7.
56. Yarden Y, Kuang W-J, Yang-Feng T, Coussens L, Munemitsu
S, Dull TJ, et al. Human protooncogene c-kit a new cell
surface receptor tyrosine kinase for an unidentified ligand.
EMBO J1987;6:3341-51.
57. Bigner SH, Humphrey PA, Wong AJ, Vogelstein B, Mark J,
Friedman HS. Characterization of the epidermal growth factor receptor in human glioma cell lines and xenografts. Cancer Res 1990;50:8017-22.
Документ
Категория
Без категории
Просмотров
4
Размер файла
1 268 Кб
Теги
187, prognostic, factors, patients, first, analysis, disease, hodgkin, relapse
1/--страниц
Пожаловаться на содержимое документа