High Levels of Constitutive WAF1/Cipi Protein Are ... · Overexpression of WAF1/Cipi has been shown to inhibit growth of colon cancer cells, leukemia cells, and brain tumor cells

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

http://clincancerres.aacrjournals.org/

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).



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



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




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-



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




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.



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

[email protected] at

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



http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

High Levels of Constitutive WAF1/Cipi Protein Are ... · Overexpression of WAF1/Cipi has been shown to inhibit growth of colon cancer cells, leukemia cells, and brain tumor cells