Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Vol. 1, 1051-1057, September 1995 Clinical Cancer Research 1051
High Levels of Constitutive WAF1/Cipi Protein Are Associated with
Chemoresistance in Acute Myelogenous Leukemia1
Wei Zhang,2 Steven M. Kornblau,2
Tohru Kobayashi, Anne Gambel, David Claxton,
and Albert B. Deisseroth3
Departments of Neuro-Oncology [W. Z., T. K.], Tumor Biology[W. Z.], and Hematology [W. Z., S. M. K., T. K., A. G., D. C.,A. B. D.], The University of Texas M. D. Anderson Cancer Center,Houston, Texas 77030
ABSTRACT
The WAFJ1C1pJ gene product is an important regulatorat the G1 checkpoint in the cell cycle. WAF1/Cipi expression
can be activated through p53-dependent and p53-indepen-
dent pathways. The WAF1/Cipi protein binds to cychin-
dependent kinase complexes and inhibits the kinase activity
that is required for cell cycle progression. In this prehimi.
nary study, we analyzed with Western blot assays the
steady-state levels of the WAF1/Cipi protein in the leukemia
cells of 100 untreated acute myebogenous leukemia (AML)
patients. Normal bone marrow cells from six donors were
used as a control. The results of these analyses showed that
the levels of the WAF1/Cipi protein were very low in nor-
mal marrow cells and in the leukemia cells of 83 AML
patients. High levels of WAF1/Cipi were detected in 17
patients; these patients with high WAF1/Cipi levels were
significantly hess likely to achieve complete remission (41%
versus 69%, P = 0.03) and were four times as likely to be
resistant to therapy (47% versus 12%, P = 0.003) as patientswith very low levels of WAF1/Cipi. Median survival was 38
weeks for patients having very low expression levels versus
11 weeks for patients having high expression levels (P
0.04). The WAF1/Cipl leveh was an independent predictor
for response but not survival in a stepwise multivariate
regression analysis. Southern blotting analyses did not de-
tect deletion of the WAFJ/Cipi gene in the 12 negativeWAF1/Cipi AML samples tested. Also, the level of WAF1/
Cipi protein expression was not correlated with overexpres-
sion of cyclin Dl, cyclin E, proliferating cell nuclear antigen,
cyclin-dependent kinase 4, or p53 in the heukemia cells.
However, the hevehs of cycin Dl, cydin E, and cyclin-depen-
dent kinase 4 were elevated in most of the AML samples
compared with that in normal marrow. We hypothesize that
Received 4/4/95; revised 4/9/95; accepted 5/10/95.
I This work was supported by the Anderson Chair for Cancer Treatment
and Research, the Bush Leukemia Fund, and the National CancerInstitute AML P01 (CA55164) to A. B. D., and an institutional start-upfund to W. Z.2 Both authors contributed equally to this work.3 To whom requests for reprints should be addressed, at Department ofHematology, Box 24, The University of Texas M. D. Anderson CancerCenter, 1515 Holcombe Boulevard, Houston, TX 77030.
high-level constitutivehy expressed WAF1/Cipi in tumor
cells may result in an indolent state that is refractory to
chemotherapy drugs. We conclude that the WAF1/Cipl ex-
pression level may be an important prognostic factor for
response to therapy and survival in AML patients.
INTRODUCTION
Disruption of the G, checkpoint control in the cell cycle is
one mechanism through which the abnormal proliferation of
tumor cells can evolve. Two families of regulators, positive and
negative, of the G1 checkpoint have been identified by extensive
molecular studies. A functional complex with active kinase
activity is formed from a G1 cychin associated with a member of
the CDK4 family and the PCNA (1-3). Overexpression of
positive regulators, such as cyclin D or CDK, has been shown in
some human tumors (4-9). It is believed that the CDK com-
plexes promote the G1 to S-phase progression by phosphorylat-
ing substrates such as the Rb protein. Phosphorylation of Rb
results in the release of the E2F transcriptional factor, which, in
turn, activates the expression of genes required in DNA synthe-
sis (1-3). Previous research by our group and others has dem-
onstrated that alterations in expression of Rb were associated
with resistance to therapy for AML (10). Such perturbations of
expression of the components of these cell cycle-regulatory
pathways have been shown to be common events with clinical
consequences in human malignancy.
Negative regulators of G1 checkpoint control have been
another focus of extensive investigations. One such regulator,
WAF1/Cipi, was recently identified and cloned (11-15).
WAF1/Cipi protein binds to the CDK complexes and inhibits
their activities (12-17). The WAF1/Cipi protein also directly
interacts with PCNA, blocks the DNA polymerase 8, and inhib-
its DNA replication (18). Overexpression of WAF1/Cipi has
been shown to inhibit growth of colon cancer cells, leukemia
cells, and brain tumor cells (1 1, 19). The WAF1/Cipi gene is
transcriptionally regulated by tumor suppressor p53, and is a
downstream mediator of p53’s antiproliferation function (1 1).
Further studies of WAF1/Cipi regulation demonstrated that
WAFJ/Cipl gene expression is also controlled by p53-indepen-
dent pathways. In primary embryo fibroblasts derived from
homozygous p53 knock-out mice, WAFJ/Cipi expression is
activated by treatment of cells with platelet-derived growth
factor and by fibroblast growth factor (20). In addition, the
expression of WAF1/Cipi protein in p53 null mice showed no
difference from wild-type mice (21). In three leukemia cell lines
devoid of p53 expression, K562, HL6O, and U937, WAF1/Cipi
4 The abbreviations used are: CDK, cychin-dependent kinase; AML,acute myelogenous leukemia; PCNA, proliferating cell nuclear antigen;
Rb, retinoblastoma.
Research. on August 22, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
1052 WAF1/Cipl in AML
expression can be induced by treatment with okadaic acid,
interferons, and 12-O-tetradecanolylphorbol-13-acetate (19).
Several genes functionally related to WAF1/Cipi have also
been identified. p16 forms complexes with cyclin D-CDK4 and
inhibits its kinase activity (22). Initial studies showed that p16 is
frequently deleted in various human tumor cell lines, suggesting
that p16 may be another tumor suppressor gene (23, 24). How-
ever, mutation of p16 in primary tumors may not be as frequent
as in cell lines (25-29). p27, which also inhibits CDK4, mcdi-
ates the G1 arrest induced by transforming growth factor �31 or
cell to cell contact (30-32). p27 shares some structural similar-
ity with WAF1/Cipi (30). Other members of the CDK inhibitor
family include piS, pl8, and the newly reported p57 (33-36).
Because altered expression of other components in this cell
cycle signal transduction pathway has been shown to affect
response to therapy in AML patients (10), we wanted to deter-
mine the expression pattern of WAF1/Cipl and its clinical
importance. We therefore analyzed the steady-state WAF1/Cipi
protein levels in the leukemia cells of 100 untreated AML
patients and evaluated the effects of expression level on re-
sponse and survival. Southern blotting was performed on a
subset of patient samples to look for genomic alterations that
might account for low expression. Additionally, the relationship
between WAF1/Cipi expression and the expression of other
components of this cell cycle-regulatory pathway, including
PS3, PCNA, cychin Dl, cyclin E, and CDK4, was analyzed.
MATERIALS AND METHODS
Cells and Phasmids. K562 cells were obtained from the
American Type Culture Collection (Rockville, MD), and main-
tamed in RPMI 1640 medium supplemented with 10% FCS in a
37#{176}Cincubator containing 5% CO2. K562 is a myelogenous
leukemia cell line that does not express endogenous p53 (37).
Patient Data. WAF1/Cipi expression was determined
by Western blot analyses of whole-cell lysates from 100 newly
diagnosed, untreated, nonconsecutive AML patients. Protein
from these samples was prepared as previously described (10),
from either the mononuclear fraction of peripheral blood or
from peripheral blood pheresis collections. Samples for analysis
were obtained during regularly scheduled diagnostic evaluations
as part of protocols approved by the Institutional Review Board
of The University of Texas M. D. Anderson Cancer Center.
There were 61 men and 39 women in this study. The median age
was 50 (range, 17-83) years. There were patients in all FAB
classification categories: 2 MO, 16 Ml, 23 M2, 4 M3, 33 M4, 12
M5, 1 M6, and 9 unclassifiable patients.
Western Blotting. Protein, extracted from 5 X i0� cells
mixed in boiling hot sample buffer (125 mt�i Tris-HC1 [pH 6.8],
1% SDS, 2% �3-mercaptoethanol, and 0.01% bromophenoh
blue), was loaded onto a 10% SDS-polyacrylamide gel. After
electrophoresis overnight at 45 V, the protein was transferred to
an Immobilon polyvinylidene difluoride membrane (Millipore
Corp., Bedford, MA), blocked with a blocking solution [50 mr�i
Tris-HC1 (pH 7.5), 0.9% NaC1, 3% nonfat dry milk, and 0.05%
Tween 20] for 4 h, and then incubated overnight with anti-
WAF1/Cipi antibody (PharMingen, San Diego, CA), anti-p53
antibody DO-i (Oncogene Science, Inc., Manhasset, NY),
anti-actin antibody (Oncogene Science, Inc.), anti-cyclin E,
anti-cyclin Dl, anti-PCNA, or anti-CDK4 (PharMingen). The
levels of antigens were analyzed using the enhanced chemilu-
minescence system (Amersham Corp., Arlington Heights, IL)
according to the manufacturer’s instructions. The levels of cx-
pression were defined as low for the very low expressors that
can only be seen after an extended period of exposure or high
for clearly visible detection.
Southern Blotting. Ten p.g genomic DNA from cells
were digested with EcoRI overnight. The DNAs were phenol-
chloroform extracted, ethanol precipitated, resuspended, and run
on a 1% agarose gel. The DNAS were transferred onto a nylon
membrane and hybridized with a 32P-labebed WAF1/Cipi cDNA
probe (the 2.1-kb NotI fragment). The labeling procedure was
according to the instructions provided by the random labeling
kit (Boehringer Mannheim). After the final wash at 65#{176}Cwith
1 X SSC, the membrane was exposed to a Kodak X-AR film.
The membrane was stripped and rehybridized with a �3-actin
probe as a control (data not shown).
Statistical Analysis. Distribution of prognostic features
among WAF1/Cipi groups was studied by x2 tests. Survival
distributions were estimated according to the method of Kaplan
and Meier (38), and comparisons were based on the Wilcoxon
and log rank tests (39). Regression methods based on a propor-
tional hazards model were used to provide a test of association
between the WAF1/Cipi bevel and response or survival out-
comes. Characteristics were selected for inclusion in the model
based on their previous recognition as important prognostic
factors and evidence of association with either response or
survival in this population.
RESULTS
Overexpression of WAF1/Cipi Protein in Leukemia
Cells of 17% of AML Patients Studied. To investigate
whether the regulation of WAF1/Cipl is altered in AML, we
analyzed the levels of WAF1/Cipi protein in the leukemia cells
of 100 untreated AML patients. Examples of the results are
shown in Fig. 1, and the results are summarized in Table 1.
WAF1/Cipi protein level was very low (can only be seen after
an extended period of exposure) in normal marrow cells from
eight donors (see Fig. 4, and data not shown). Among the
patients with AML, WAF1/Cipl protein levels were similar to
the normal marrow sampbes, i.e., very low level, in 83 patients.
However, 17 patients had high levels of WAF1/Cipl protein
(Lanes 9 and 19, Fig. 1). Actin protein levels were also probed
as a control for protein loading.
The level of p53 expression was also determined in these
samples, and Rb levels had previously been determined for these
samples (10). Previous studies established that the acquisition of
a p53 mutation is a very infrequent event (5-6%) in AML
patients (40, 41). Overall, there is no correlation between the
basal levels of WAF1/Cipi and the levels of p53 protein in the
cells (r� = 0.238). However, there were several cases with a
missense mutant p53 or no p.53 expression (Lanes 5, 15, 23, and
27, Fig. 1) in which only a trace amount of WAF1/Cipl protein
was detected. The correlation between Rb and WAF1/Cipl
expression was examined and no association was observed (P =
0.75, �2 test).
Research. on August 22, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
WAFI/Cipi -+�
Clinical Cancer Research 1053
5 W. Zhang, unpublished results.
K562
� AML(1-28)� 1 2 3 4 5 6 7 8 9 101112
ac�n=�
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
� � 3 �actin -+
WAFI/Cipi -*
Fig. 1 The levels of WAF1/Cipi and p53 in the leukemia cells of 28
AML patients. Protein extracts from a half-million leukemia cells wereanalyzed by Western blot assays using specific mAbs. The filters werereprobed with anti-actin antibody as a control for protein loading. Fortygig protein extract from K562 cells transfected with wild-type p53 or
CMV.neo vector were also used in Western blot assays as a positive andnegative control for p53 and WAFI/Cipi proteins, respectively.
WAFJ/Cipi Gene Deletion Is Not Detected in AML.
Although the levels of WAF1/Cipi protein which are observed
in normal marrow cells are very low, there may be various
explanations for the lack of WAF/Cipi protein in some AML
samples. One possibility is gene deletion, a permanent genetic
defect. To test whether gene deletion caused the lack of WAF1/
Cipi protein in these AML samples, genomic DNAS were
isolated from the leukemia cells of 12 AML patients with very
low levels of WAF1/Cipl protein. The integrity of the WAFJ/
Cipi gene was analyzed by Southern blotting assays. The rep-
resentative data presented in Fig. 2 show that there is no ho-
mozygous deletion or rearrangement of the WAF1/Cipi gene in
all cases. Another study was performed with genomic DNA
from a different cohort of 1 18 AML samples, and no homozy-
gous deletion or rearrangement of the WAF1/Cipi gene was
detected.5 This is consistent with the cytogenetic data, which
show that deletion of chromosome 6p, in which the WAFJ/Cipi
gene resides, is rare.
Correhation between WAF1/Cipi Protein Levels and
Patient Survival. The distribution of several previously iden-
tified prognostic factors within the very low or high WAF1/Cipl
expression groups is shown in Table 1 . There were no signifi-
cant differences for the following major prognostic features:
cytogenetics, performance status, age, or antecedent hematolog-
ical disorder. Additionally, there were no significant differences
between the two groups concerning the distribution of hemo-
Table I Distribution of clinical and laboratory parameters, res
and outcome of patients by WAF1 expression group
ponse,
WAF1
Very low level High P
No. of patients 83 17
Gender (MiT) 54/29 7/10 0.065FAB type
MO (2) 2.4%Ml (16) 16.8% 11.7%M2 (23) 24.1% 17.6%
M3 (4) 3.6% 5.9%M4 (33) 33.7% 29.4%MS (12) 8.4% 29.4%M6 (1) 1.2%
M unknown (9) 9.6% 5.9%Zubrod performance status
0-1 66% 71%2 29% 18%
3-4 5% 6%Cytogenetics
Favorable 13% 6% 0.39”Intermediate 45% 59% 0.28” 0.49”Unfavorable 42% 35% 0.59”
Prior hematological disorder 24% 24%Median age, yr (range) 50 (17-79) 60.5 (24-83)
ResponseComplete remission 69% 41%Resistant 12% 47% 0.003
Fail 19% 12%
Relapse rate 77% 86% 0.60
Median survival (wk) 38 1 1 O.04Cax2 test for that cytogenetic group vs. the other two combined.
b x2 for all three groups.C Wilcoxon test.
gbobin, platelet count, WBC count, percentage of blasts in either
blood or marrow, serum albumin, serum bilirubin, or fibrinogen.
To investigate whether levels of WAF1/Cipi protein in
leukemia cells of AML patients are predicative of clinical out-
come, we tested for correlations between WAF1/Cipi protein
levels and patient response to therapy or patient survival. Re-
sponse was divided into three categories: complete remissions,
failures (defined as death prior to the time of recovery of counts
or regrowth of leukemia during either cycle 1 or 2 of induction),
and resistance (defined as persistent leukemia after two cycles
of induction therapy). The complete remission rate was signif-
icantly higher for patients with very low levels of WAF1/Cipi
compared to those with high WAF1/Cipl (69% versus 41%,
P = 0.03). Conversely, the rate of primary resistance to two
cycles of chemotherapy was four times greater for patients with
high WAF1/Cipi expression compared to individuals whose
leukemia cells exhibited very low levels of WAF1/Cipi (47%
versus 12%, P 0.003). The rate of death during induction was
similar for both groups. For the group of patients that achieved
remission, the rate of relapse did not appear to be affected by the
level of WAF1/Cipl expression. As would be expected due to
the differences in response, patients with very low levels of
WAF1/Cipl had a significantly longer median survival experi-
ence (38 weeks versus 1 1 weeks, P = 0.04) than patients with
high WAF1/Cipi (Fig. 3).
Our previous studies showed that AML patients with low
Rb had poor prognosis (10). Among the 13 patient samples with
Research. on August 22, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
I 23456789
� �
� � m� E�* � _
Fig. 2 Southern blot analyses of nine AML samples that have very lowlevels of WAF1/Cipi protein. The genomic DNA was digested withEcoRI. The NotI fragment of WAFJ/Cipi cDNA was used as a probe.Both bands are WAF1/Cipi DNA.
�) O�8 �\ � � �- � P =
� 0.7 � A High WAF1.� 0.6
.� 0.5 �
� 0.4 A
n.C 0.3 A-’a. 0.2 A--
0.1 � A------
0 � �-��---�- -�-�----� � �0 26 52 78 104 130 156 182 208 234 260
Survival in Weeks
Fig. 3 The relationship between the levels of WAF1/Cipi and the
survival of AML patients. Kaplan-Meier survival curves for WAF1 very
low level (#{149})and high level (A) expressors. Scale stops at 260 weeks.Tick marks, patients still alive.
1054 WAF1/Cipi in AML
low Rb expression in this study, 2 had high WAF1/Cipl cx-
pression. This sample size was not sufficient to perform a
Kaplan-Meier analysis. There was sufficient sample size in the
high Rb group to evaluate the effect of WAF1/Cipl expression
on survival experience. The WAF1/Cipl expression retained an
independent prognostic impact within the high Rb group, and
the difference in survival was significant by either the Gehan-
Breslow (P = 0.04) or log rank test (P = 0.05). A larger cohort
of patients is currently being examined prospectively for Rb and
WAF1/Cipl to further conform this correlation. This prelimi-
nary result suggests that the prognostic impact of the WAF1/
Cipi expression pattern is independent of the Rb expression
pattern.
A multivariate analysis was conducted to determine factors
independently predicting for response and survival. Factors con-
sidered in the model for response included: therapy, age, per-
formance status, cytogenetics, hemoglobin, platelet count, WBC
Table 2 Multiple regression results for response and
End point Response to therapy
survival
Survival
P P
0.021 0.01
0.050 0.019
Variable
Age (yr)
Performance status
0-2 = 03-4=
WAF1 very low level = 0,High = 1
Cytogenetics
Favorable = +1Intermediate = 0Unfavorable = -
aWAF1/Cipi level was not a significant variable in the final modelfor survival and cytogenetics was not a significant variable in the final
model for response; hence, these cells are blank.
count, albumin and bibirubin bevels, history of antecedent hema-
tobogical disorder, and WAF1/Cipi level. For the survival
model, the analysis was performed with and without response to
therapy as a variable. The final characteristics of each model are
shown in Table 2. The WAF1/Cipl level joined the age and
performance status as factors predicting for response. Patient
age, poor performance status, and cytogenetics were all mdc-
pendent variables in the final model for survival. When response
was included, it replaced performance status in the final model
for survival.
Lack of Correlation between the Levels of WAF1/Cipl
and the Levels of PCNA, Cydlin Dl, Cyclin E, and CDK4 inAML. The above results from clinical correlation studies
posed a paradox. A function of WAF1/Cipl is to inhibit CDK
kinases and cell growth. Therefore, high levels of WAF/Cipl
protein were expected to provide beneficial effects; however,
they do not appear to do so. Several events may explain why this
was not the case among our AML patients. Relative levels of
regulators of the G1 checkpoints may determine whether the cell
divides. In addition, it has been proposed that WAF1/Cipl
inhibits the CDK complex when more than one molecule of
WAF1/Cipl binds to the CDK complex (42, 43). Therefore,
CDKS, cycbins, and PCNA, as positive regulators, may be spe-
cificably overexpressed in leukemia cells with high levels of
WAF1/Cipl protein, and may overcome the otherwise growth
inhibitory effects of WAF1/Cipl. Such a mechanism was pro-
posed in a previous study (9).
To test whether this was occurring in the leukemia cells of
some of our patients, we analyzed levels of PCNA, cyclin Dl,
CDK4, and cyclin E in leukemia cells from patients with very
bow or high WAF1/Cipl. Our results showed that there was no
significant difference between the two WAF1/Cipl groups (Fig.
4). However, when compared to normal marrow cells, cyclin
Dl, cyclin E, and CDK4 are overexpressed in leukemia cells of
most AML patients, whereas PCNA is expressed at similar
levels between normal marrow and AML cells.
DISCUSSION
The characterization of the mechanisms which are respon-
sible for the control of the cell cycle is one of the keys to
deciphering tumorigenesis changes in tumor cells. Increased
1
1
0.004 NS�
0.050
NSa
Research. on August 22, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
Clinical Cancer Research 1055
Normal marrow AML
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
. � -�
- I �
:� P53
.4 WAF1/Cipi
.4 PCNA
I CyclinDi
t... � .. :1 CyclinE
- :1 CDK4
Fig. 4 Expression of p53, WAF1/Cipi, PCNA, cyclin Di, cyclin E, and CDK4 in cells from normal marrow of donors and leukemia cells of AMLpatients.
levels of positive cell cycle regulators and deletion of negative
cell cycle regulators have been found in tumor cells (1). In this
study, we examined one of the negative regulators, the WAFJ/
Cipi gene. This gene may play an important role because it is a
downstream mediator of the tumor suppressor gene p.53, and it
inhibits a broad range of CDKS. Our previous studies demon-
strated that WAF/Cipl expression is very low in several p13-
negative leukemia cell lines, and overexpression of WAF1/Cipl
by transfection of the expression vector into K562 leukemia
cells resulted in the inhibition of colony formation (19). Induc-
tion of WAF1/Cipl by other agents such as interferons and
l2-O-tetradecanoylphorbol-l3-acetate is also correlated with
growth arrest (19). Here, we examined WAF1/Cipl protein
expression in primary leukemia cells from AML patients, and
the results of these studies demonstrated that expression of
WAF1/Cipl is heterogeneous in AML patients. The majority of
AML patients (83%) have very low levels of WAF1/Cipi pro-
tein. WAF1/Cipl protein levels were also very low in normal
marrow cells. Significantly higher levels of the protein were
observed in 17% of the AML patients studied. To test whether
gene deletion was responsible for the low expression levels of
WAF1/Cipl protein in AML cells, we studied the status of the
WAFJ/Cipl genes by Southern blot analyses. Our result did not
show homozygous deletion or rearrangement in more than 100
AML samples; thus, the very low levels of WAF1/Cipl in AML
cells was not due to gene deletion.
The significance of the high levels of WAF1/Cipi protein
in primary AML cells was tested by analyzing the survival and
response to chemotherapy of AML patients with high or low
levels of WAF1/Cipl proteins. High levels of WAF1/Cipl cx-
pression were correlated with primary resistance to AML induc-
tion therapy and consequently were negatively associated with
survival. Resistance was defined by the stringent criteria of
detection of residual leukemia after two courses of induction
therapy. Many patients not surviving two cycles, probably re-
sistant, were counted in the ‘ ‘failure’ ‘ group. The true rate of
resistance is, therefore, likely to be higher in both groups. If all
failures are considered as resistant, this resistance rate is still
significantly higher in the high WAF1/Cipl group. While the
level of WAF1/Cipl protein was an independent prognostic
factor in a multivariate analysis for response to chemotherapy,
cytogenetics data replaced WAF1/Cipl level as a predicting
factor for survival. This is likely due to the fact that a larger
percentage of the population had poor prognosis cytogenetics
than high WAF1/Cipl level.
The association of high expression levels of WAF1/Cipl
with poor prognosis was unexpected initially, since WAF1/Cipl
is an inhibitor of cell proliferation. We considered several pos-
sible explanations. First, the cells in which elevated WAF1/Cipi
protein level are found may also contain elevated levels of
positive cell cycle regulators (e.g., CDKS overexpression),
which may offset the effect of the WAF1/Cipl. He et a!. (9)
suggested that CDK4 gene amplification contributes to transfor-
mation of cell lines. In another study, overexpression of cyclin
Dl reduced growth inhibition of transforming growth factor �3,
which activates p27 expression (44). To test whether this was
occurring in AML cells, we examined the expression of cyclin
Dl, cyclin E, PCNA, and CDK4 in AML samples with very low
or high WAF1/Cipi levels. We did not find any correlation
between the levels of the WAF1/Cipi proteins and the levels of
positive cell cycle regulators we tested. It is important to point
out, however, that the positive cell cycle regulators cyclin Dl,
cyclin E, and CDK4 were overexpressed in the leukemia cells of
the majority of the AML patients that we analyzed.
Other genetic events may also overcome the negative effect
of WAF1/Cipl protein in cancer cells. For example, constitutive
expression of B-myb could bypass the effect of WAF1/Cipi
(45). It is possible that the genetic events may modify WAF1/
Cipl in such a way that it cannot bind to and/or cannot inhibit
the CDK complexes efficiently. Additional studies are needed to
identify such an attenuating event in the patient cells that con-
tam high constitutive levels of WAF1/Cipl protein.
The second possibility is that the WAF1/Cipl protein in the
high expressors is mutant. Future sequencing studies will pro-
Research. on August 22, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
1056 WAF1/Cipi in AML
vide definitive answers to this, although recent studies suggested
that mutation of the WAFJ/Cipl gene is an extremely rare event
in human cancers (46).
The third and, we consider, most likely possibility is that
high basal levels of WAF1/Cipl inhibit the proliferation activity
of the leukemia cells. This indolent state of leukemia cells may
thus reduce the sensitivity of the cells to chemotherapy drugs
that preferentially kill cycling cells. Consistent with this hypoth-
esis, WAF1/CIP1 protein has been found to be expressed at very
low levels in normal brain cells but overexpressed in many
astrocytic tumors that are known to be slow growing and highly
resistant to chemotherapy and radiotherapy.5 Additional studies
are warranted to elucidate the relationship between cell cycle
control and response to chemotherapy and radiotherapy.
ACKNOWLEDGMENTS
We thank Drs. W. Edward Mercer and Paul Chiao for critical
reading and providing comments on this article. We thank Craig
McClain and Shilpen Patel for technical assistance, and Gene Zhang,
Leslie Wildrick, Rosemarie Lauzon, and Joyce Palmer for their editorial
assistance.
REFERENCES
1. Sherr, C. J. Mammalian Gi cyclins. Cell, 73: 1059-1065, 1993.
2. Zhang, H., Xiong, Y., and Beach, D. Proliferating cell nuclearantigen and p21 are components of multiple cell cycle kinase com-
plexes. Mol. Biol. Cell, 4: 897-906, 1993.
3. Kato, J., Matsushime, H., Hiebert, S. W., Ewen, M. E., and Sherr,C. J. Direct binding of cyclin D to the retinoblastoma gene product(pRb) and pRb phosphorylation by the cyclin D-dependent kinaseCDK4. Genes Dcv., 7: 331-342, 1993.
4. Motokura, T., Bloom, T., Kim, H. G., Juppner, H., Ruderman, J. V.,
Kronenberg, H. M., and Arnold, A. A novel cyclin encoded by a
bcl-linked candidate oncogene. Nature (Land.), 350: 512-515, 1991.
S. Lammie, G. A., Fantl, V., Smith, R., Schuuring, E., Brookes, S.,
Michalides, R., Dickson, C., Arnold, A., and Peters, G. D11S128, a
putative oncogene on chromosome 1 1q13 is amplified and expressed in
squamous cell and mammary carcinomas and linked to BCL-1. Onco-
gene, 6: 439-444, 1991.
6. Jiang, W., Kahn, S. M., Tomita, N., Zhang, Y. J., Lu, S. H., andWeinstein, I. B. Amplification and expression of the human cyclin D
gene in esophageal cancer. Cancer Res., 52: 2980-2983, 1992.
7. Buckley, M. F., Sweney, K. J., Hamilton, J. A., Sini, R. L., Manning,D. L., Nicholson, R. I., deFazio A., Watts, C. K., Musgrove, E. A., andSutherland, R. L. Expression and amplification of cyclin genes in human
breast cancer. Oncogene, 8: 2127-2133, 1993.
8. Tsuruta, H., Sakamoto, H., Onda, M., and Terada, M. Amplification
and overexpression of EXP1 and EXP2/cyclin Di gene in human
esophageal carcinomas. Biochem. Biophys. Res. Commun., 196: 1529-
1536, 1993.
9. He, J., Allen, J. R., Collins, V. P., Allalunis-Turner, J., Godbout, R.,Day, R. S., and James, C. D. CDK4 amplification is an alternative
mechanism to pitS gene homozygous deletion in ghioma cell lines.
Cancer Res., 54: 5804-5807, 1994.
10. Kornblau, S. M., Xu, H-J., Zhang, W., Hu, S-X., Beran, M., Smith,
T. L., Hester, J., Estey, E., Benedict, W. F., and Deisseroth, A. B. Levelsof retinoblastoma protein expression in newly diagnosed acute myelog-
enous leukemia. Blood, 84: 256-261, 1994.
1 1. El-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Par-sons, R., Trent, L. M., Lin, D., Mercer, W. E., Kinzler, K. W., and
Vogelstein, B. WAF1, a potential mediator of p53 tumor suppression.
Cell, 75: 817-825, 1993.
12. Harper, J. W., Adami, G. R., Wei, N., Keyomarsi, K., and Elledge,
S. J. The p21 Cdk-interacting protein Cipi is a potent inhibitor of G1
cyclin-dependent kinases. Cell, 75: 805-816, 1993.
13. Xiong, Y., Hannon, G. J., Zhang, H., Casso, D., Kobayashi, R., and
Beach, D. p21 is a universal inhibitor of cyclin kinases. Nature (Lond.),
366: 701-704, 1993.
14. Gu, Y., Turck, C. W., and Morgan, D. 0. Inhibition of CDK2
activity in vivo by an associated 20K regulatory subunit. Nature (Lond.),
366: 707-710, 1993.
is. Noda, A., Ning, Y., Venable, S. F., Pereira-Smith, 0., and Smith,
J. R. Cloning of senescent cell-derived inhibitors of DNA synthesisusing an expression screen. Exp. Cell Res., 211: 90-98, 1994.
16. El-Deiry, W. S., Harper, J. W., O’Connor, P. M., Velculescu, V. E.,Canman, C. E., Jackman, J., Pietenpol, J. A., Burrell, M., Hill, D. E.,
Wang, Y., Wiman, K. G., Mercer, W. E., Kastan, M. B., Kohn, K. W.,Elledge, S. J., Kinzler, K. W., and Vogelstein, B. WAF1/CIP1 is induced
in p53-mediated G1 arrest and apoptosis. Cancer Res., 54: 1169-1174,
1994.
17. Dulic, V., Kaufmann, W. K., Wilson, S. J., Tlsty, T. D., Lees, E.,
Harper, J. W., Elledge, S. J., and Reed, S. I. p53-dependent inhibition of
cyclin-dependent kinase activities in human fibroblasts during radiation-
induced G1 arrest. Cell, 76: 1013-1023, 1994.
18. Waga, S., Hannon, G. J., Beach, D., and Stillman, B. The p21
inhibitor of cyclin-dependent kinases controls DNA replication by in-
teraction with PCNA. Nature (Lond.), 369: 574-578, 1994.
19. Zhang, W., Grasso, L., McClain, C. D., Gambel, A. M., Cha, Y.,Traveli, S., Deisseroth, A. B., and Mercer, W. E. p53-independentinduction of WAF1/Cipi in human leukemia cells is correlated with
growth arrest accompanying monocyte/macrophage differentiation.
Cancer Res., 55: 668-674, 1995.
20. Michiele, P., Chedid, M., Lin, D., Mercer, W. E., and Givol, D.
Induction of WAF1/Cipi by a p53-independent pathway. Cancer Res.,
54: 3391-3395, 1994.
21. Parker, S. B., Eichele, G., Zhang, P., Rawls, A., Sands, A. T.,
Bradley, A., Olson, E. N., Harper, J. W., and Elledge, S. J. p13-independent expression of p21CJP1 in muscle and other terminally dif-
ferentiating cells. Science (Washington DC), 267: 1024-1027, 1995.
22. Serrano, M., Hannon, G. J., and Beach, D. A new regulatory motif
in cell-cycle control causing specific inhibition of cyclin D/CDK4.
Nature (Lond.), 366: 704-707, 1993.
23. Kamb, A., Gruis, N. A., Weaver-Feldhaus, J., Liu, 0., Harshman,K., Tavtigian, S. V., Stockert, E., Day, R. S. I., Johnson, B. E., andSkolnick, M. H. A cell cycle regulator potentially involved in genesis of
many tumor types. Science (Washington DC), 264: 436-440, 1994.
24. Nobori, T., Mirura, K., Wu, D. J., Lois, A., Takabayashi, K., and
Carson, D. A. Deletions of the cyclin-dependent kinase-4 inhibitor gene
in multiple human cancers. Nature (Lond.), 368: 753-756, 1994.
25. Okamoto, A., Demetrick, D. J., Spillare, E. A., Hagiwara, K.,Hussain, S. P., Bennett, W. P., Forrester, K., Gerwin, B., Serrano, M.,Beach, D. H., and Harris, C. C. Mutations and altered expression of
pi6�4 in human cancer. Proc. Natl. Acad. Sci. USA, 91: 11045-
11049, 1994.
26. Cairns, P., Mao, L., Mrlo, A., Lee, D. J., Schwab, D., Eby, Y.,
Tokino, K., van der Reit, P., Blaugrund, J. E., and Sidransky, D. Ratesof p16 (MTS1) mutations in primary tumors with 9p loss. Science(Washington DC), 265: 415-416, 1994.
27. Kamb, A., Shattuck-Eidens, D., Eeles, R., Liu, 0., Gruis, N. A.,Ding, W., Hussey, C., Tran, T., Mild, Y., Weaver-Feldhaus, J., Mc-Clure, M., Aitken, J. F., Anerson, D. E., Bergman, W., Frants, R.,Goldgar, D. E., Green, A., MacLennan, R., Martin, N. G., Meyer, L. J.,
Youl, P., Zone, J. J., Skolnick, M. H., and Cannon-Albright, L. A.Analysis of the pitS gene (CDKN2) as a candidate for the chromosome
9p melanoma susceptibility locus. Nat. Genet., 8: 22-26, 1994.
28. Spruck, C., Gonzalez-Zulueta, M., Shibata, A., Simoneau, A., Lin,M., Gonzales, F., Tsai, Y., and Jones, P. pItS gene in uncultured tumors.
Nature (Lond.), 370: 183-184, 1994.
29. Mori, T., Miura, K., Aoki, T., Nishihira, T., Mori, S., and Naka-mura, Y. Frequent somatic mutation of the MTS1/CDK4 (multiple tumor
Research. on August 22, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
Clinical Cancer Research 1057
suppressor/cychin-dependent kinase 4 inhibitor) gene in esophageal cell
carcinoma. Cancer Res., 54: 3396-3397, 1994.
30. Toyoshima, H., and Hunter, T. pZ7, a novel inhibitor of G1 cyclin-
Cdk protein kinase activity, is related to p21. Cell, 78: 67-74, 1994.
31. Polyak, K., Kato, J. Y., Solomon, M. J., Sherr, C. J., Massague, J.,
Roberts, J. M., and Koff, A. �27K��1, a cyclin-Cdk inhibitor, links
transforming growth factor-beta and contact inhibition to cell cycle
arrest. Genes Dcv., 8: 9-22, 1994.
32. Polyak, K., Lee, M-H., Erdjument-Bromage, H., Koff, A., Roberts,
J. M., Tempst, P., and Massague, J. Cloning of � a cychin-dependent kinase inhibitor and a potential mediator of extracelluhar
antimitogenic signals. Cell, 78: 59-66, 1994.
33. Hannon, G. J., and Beach, D. �15��4B is a potential effector of
TGF-�3 induced cell cycle arrest. Nature (Lond.), 371: 257-261, 1994.
34. Guan, K., Lenkins, C. W., Li, Y., Nichols, M. A., Wu, X., O’Keefe,C. L., Matera, A. G., and Xiong, Y. Growth suppression by p18, a� and � CDK6 inhibitor, correlates with
wild-type pRb function. Genes Dcv., 8: 2939-2952, 1994.
35. Lee, M-H., Reynisdottir, I., and Massague, J. Cloning of p57KIP2, a
cychin-dependent kinase inhibitor with unique domain structure and
tissue distribution. Genes Dcv., 9: 639-649, 1995.
36. Matsuoka, S., Edwards, M. C., Bai, C., Parkers, S., Zhang, P-M.,
Baldini, A., Harper, J. W., and Elledge, S. J. p57K��2, a structurally
distinct member of the p2i’�”� Cdk inhibitor family, is a candidate
tumor suppressor gene. Genes Dcv., 9: 650-662, 1995.
37. Zhang, W., Funk, W. D., Wright, W. E., Shay, J. W., and Deis-
seroth, A. B. Novel DNA binding of p53 mutants and their role in
transcriptional activation. Oncogene, 8: 2555-2559, 1993.
38. Kaplan, E. L., and Meier, P. Nonparametric estimation from incom-
plete observations. J. Am. Stat. Assoc., 53: 457-461, 1987.
39. Mantel, E. Evaluation of survival data and two new rank order
statistics arising in its consideration. Cancer Chemother. Rep., 50:
63-170, 1966.
40. Hu, G., Zhang, W., and Deisseroth, A. B. p53 gene mutations in
acute myelogenous leukemia. Br. J. Haematol., 81: 489-494, 1992.
41. Fenaux, P., Preudhomme, C., Quiquandon, I., Jonveaux, P., Lai,
J. L., Vanrumbeke, M., Loucheux-Lefebvre, M. H., Bauters, F., Berger,R., and Kerchaert, P. Mutations of the p53 gene in acute myeloid
leukemia. Br. J. Haematol., 80: 178-183, 1992.
42. Zhang, H., Hannon, G., and Beach, D. p21-containing cychin ki-
nases exist in both active and inactive states. Genes Dev., 8: 1750-1758,
1994.
43. Harper, J. W., Elledge, S. J., Keyomarsi, K., Dynlacht, B., Bai,
L-H., Zhang, P., Dobrowoski, S., Bai, C., Connell-Growley, L., Swin-dell, E., Fox, M. P., and Wei, N. Inhibition of cyclin-dependent kinases
by p21. Mol. Biol. Cell, 6: 387-400, 1995.
44. Okamoto, A., Jiang, W., Kim, S-J., Spillare, E. A., Stoner, G. D.,
Weinstein, I. B., and Harris, C. C. Overexpression of human cyclin
Dl reduces the transforming growth factor 3 (TGF-�3) type II recep-
tor and growth inhibition by TGF-�31 in an immortalized human
esophageal epithelial cell line. Proc. Natl Acad. Sci. USA, 91:
11576-11580, 1994.
45. Lin, D., Fiscella, M., O’Connor, P. M., Jackman, J., Chen, M., Luo,L. L., Sala, A., Travali, S., Appella, E., and Mercer, W. E. Constitutiveexpression of B-myb can bypass p53-induced Wafi/Cipi-mediated G1
arrest. Proc. Natl. Acad. Sci. USA, 91: 10079, 1994.
46. Li, Y-J., Laurent-Puig, P., Salmon, R. J., Thomas, G., and Hamehin,
R. Polymorphisms and probable lack of mutation in the WAFI-CIPI
gene in colorectal cancer. Oncogene, 10: 599-601, 1995.
Research. on August 22, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from
1995;1:1051-1057. Clin Cancer Res W Zhang, S M Kornblau, T Kobayashi, et al. with chemoresistance in acute myelogenous leukemia.High levels of constitutive WAF1/Cip1 protein are associated
Updated version
http://clincancerres.aacrjournals.org/content/1/9/1051
Access the most recent version of this article at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
.http://clincancerres.aacrjournals.org/content/1/9/1051To request permission to re-use all or part of this article, use this link
Research. on August 22, 2020. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from