Gene Expression Profiling in Chronic Myeloid LeukemiaPatients Treated with Hydroxyurea

HANA BRUCHOVA, TEREZA BOROVANOVA, HANA KLAMOVA and RADIM BRDICKA*

Institute of Hematology and Blood Transfusion, Department of Molecular Genetics, U nemocnice 1, 128 20 Prague 2, Czech Republic

(Received 25 November 2001)

Using array technology that allows the simultaneous detection of gene expression of hundreds of genes,four patients with chronic myeloid leukemia (CML) were investigated at diagnosis and after startingadministration of hydroxyurea. To detect the gene expression of peripheral blood mononuclears andgranulocytes Human Cancer cDNA Array (CLONTECH) with 588 gene probes was used. Geneexpression mononuclear and granulocyte profiles of patients at diagnosis were compared with thecontrol profiles. The significant expression changes observed in most patients seemed to be important.Increased expression of c-jun N-terminal kinase 2 (JNK2), integrin alpha E, MMP-8, MMP-9 wasdetected in both fractions of most patients. In some samples PCNA, HDGF, MAPK p38, CD59increased expressions were found. Significant down-regulation of expression in patients was detected ingenes CDK4 inhibitor A, PURA, notch1 in mononuclears; STAT2, STAT5, RAR-alpha, MCL-1, junB,caspase 4 in granulocytes; CDK6, GADD153, ERBB-3, cadherin 5 in both fractions.

Expression profiles detected in patients at diagnosis did not differ markedly from those after one-week treatment with hydroxyurea. Only in a few genes were significant changes after hydroxyureaadministration observed and inter-individual expression differences were rather common.

Keywords: Gene expression; Array technology; CML; Hydroxyurea treatment

INTRODUCTION

Although many genes have been assigned to functional

classes, the roles which they play in various biological

processes have yet to be elucidated. An important step

toward understanding these roles is to define gene

expression profiles, and to compare patterns of expression

in different tissues and development stages, in both

healthy and diseased organisms etc. A promising approach

for expression analysis is the hybridisation of entire cDNA

populations to nucleic acid array. Array technology has a

wide range of applications, including the investigation of

normal biological and disease processes. Chronic myeloid

leukemia (CML) is a myeloproliferative disorder charac-

terized by a biphasic or triphasic clinical course in which a

relatively benign chronic phase is followed by transform-

ation into an accelerated and blastic phase. At the

cytogenetic and molecular levels, most patients with CML

demonstrate bcr/abl fusion gene in hematopoietic

progenitor cells, which results from a reciprocal

translocation between chromosomes 9 and 22. This

translocation often leads to a shortened chromosome 22,

called the Philadelphia chromosome (Ph). Translation of

the fusion gene mRNA yields chimeric proteins of

variable sizes with increased tyrosine kinase activity.

Tyrosine kinase activates many proteins that are

involved in several signal pathways in which oncogenes

like ras, myc, jun, and bcl-2 participate. The activation of

these signal pathways causes a decreased dependence of

CML cells on growth factors, adhesion changes [1,2], a

higher proliferation ability, and longer lifespan because of

their increased resistance to the induction of apoptosis [3–

5]. It is quite clear that during CML development other

genes must play key roles, without which the transform-

ation mediated by bcr/abl fusion gene would not be

sufficient.

Hydroxyurea chemotherapy is widely used to control

the chronic phase of myeloid leukemia by stabilizing the

blood cell count and splenomegaly, which may be

achieved for up to 80% of those patients treated at

diagnosis [6,7]. Hydroxyurea, a structurally simple

antimetabolite that interferes with DNA synthesis by

ISSN 1042-8194 print/ISSN 1029-2403 online q 2002 Taylor & Francis Ltd

DOI: 10.1080/10428190290026358

*Corresponding author. Tel.: þ420-2-21977219. Fax: þ420-2-21977371. E-mail: molgen@uhkt.cz

Leukemia and Lymphoma, 2002 Vol. 43 (6), pp. 1289–1295

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

McM

aste

r U

nive

rsity

on

05/0

3/13

For

pers

onal

use

onl

y.

inhibiting ribonucleotide reductase, has become most

popular in recent years [8]; it is considered the best

alternative for conventional CML therapy as it lacks toxic

effects on pulmonary and other organs and induces rapid

disease control [9]. On the other hand, hydroxyurea is not

able to eliminate Phþ stem cells from the bone marrow [7]

and its effect reflects individuality of the disease.

Hydroxyurea is one of the few drugs selectively inhibiting

the cell cycle in the S-phase; this inhibition is usually fast

and in normal cell populations treated with therapeutic

doses is reversible. The remission acquired by hydro-

xyurea treatment in CML patients is of course time

limited, and hydroxyurea is thus employed in most cases

for an introduction into remission, at which point other

drugs such as interferon alpha are applied.

The aim of this study was to follow gene expression in

CML patients and to identify the genes contributing to

CML development by comparing patient expression

profiles with control ones. Gene expression changes

induced by hydroxyurea treatment were also investigated,

and attempts made to ascertain which genes are influenced

by hydroxyurea, and may play important roles in the

therapeutic effect.

MATERIALS AND METHODS

Blood samples from four patients with CML were used in

this study. Total peripheral blood leukocytes were

separated out into mononuclears and granulocytes by

Ficoll–Paque (Amerscham Pharmacia Biotech) centrifu-

gation [11]. At the time of diagnosis all patients were

bcr/abl fusion gene positive (Table I), and cytogenetic

analysis confirmed the presence of the Ph chromosome in

all of the examined mitosis. The second blood samples

were taken from patients one week after hydroxyurea

treatment initiation. Hydroxyurea was administered orally

to patients in doses of 30–50 mg/kg/day with regard to the

leukocyte count. Two mixtures of the same amount of

RNA, isolated from granulocytes and mononuclears,

obtained from six healthy individuals of both sexes and

three age categories (three males aged 68, 33, 25 and three

females aged 63, 36, 25) served as control (standard)

samples.

For the detection of gene expression profiles, Atlas

Human Cancer cDNA Expression Arrays (Clontech

7742-1, Palo Alto, USA) with 588 genes involved in

various biological processes were used (Fig. 1). Total

RNA was isolated from peripheral blood leukocytes by the

method of Chomczynsky and Sacchi [10], and 1mg of

total RNA was reverse-transcribed into cDNA (Clontech

kit 7742-1, MMLV RT), which procedure included

labelling with 32P (Amersham, Buckinghamshire, Eng-

land, PB10204- [a32P] dATP,10Ci/l:3000 Ci/mmol).

CHROMA SPIN-200 DEPC–H2O columns were used to

purify 32P-labelled cDNA from unincorporated32P-labelled nucleotides and small cDNA fragments.

Radioactively labelled cDNAs were hybridised to Atlas

Human Cancer Array overnight, and autoradiographed.

Gene activity was evaluated with AtlasImage 1.5 software

(Clontech), used in accordance with the manufacturer’s

instructions.

The utility of Atlas Arrays for accurately assessing gene

expression in a reproducible manner is well established

(CLONTECHniques 1997). In addition, a control

experiment to confirm the reproducibility of the Atlas

FIGURE 1 Atlas human cancer cDNA expression array—Gene expression profiles of CML patient 1 at diagnosis: (a) mononuclears and (b)granulocytes.

TABLE I Patient characteristics

Leukocyte count (109/l)Patient number Sex Age (y) bcr/abl Fusion type Diagnosis Treatment

1 M 56 Positive b3 a2 23.4 7.22 M 23 Positive b3 a2 57.2 58.73 M 39 Positive b2 a2 137.0 209.24 M 57 Positive b2 a2 443.2 308.0

H. BRUCHOVA et al.1290

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

McM

aste

r U

nive

rsity

on

05/0

3/13

For

pers

onal

use

onl

y.

technology was conducted: the same (Control) sample was

hybridised to two identical arrays, and the obtained

expression patterns were identical.

RESULTS

Expression profiles of patients at diagnosis and after one

week hydroxyurea treatment were compared with the

standard profiles. Results of the evaluation of gene

expression changes have been summarized in Tables II–

IV. Two-fold and higher increase or decrease of gene

expression was considered to be significant (according to

recommendation of software producer). The protein/gene

names were retained in the tables according to the

CLONTECH protein/gene database.

Gene Expression At Diagnosis

Gene expression profiles obtained from patients at the

time of diagnosis were compared with the control profile.

In terms of gene expression changes, the most important

seem to be those observed in majority of the patients, e.g.

c-jun N-terminal kinase 2 (JNK2), MMP-9, MMP-8,

integrin a E, PCNA (increased expression) and CDK6,

p16 INK4A, ERBB3, cadherin 5 etc. (decreased

expression).

JNK2 belongs among MAP-kinases involved in signal

transduction from cell surface to the nucleus. JNK2

activates the transcription factor c-jun that regulates the

expression of specific genes such as IL-2 [12]. JNK

signal cascade may be one of the pathways that are

responsible for transformation mediated by BCR/ABL

[13], and Raitano et al. [14] have shown that the JNK

pathway can also be activated by the bcr/abl oncogene.

We detected JNK2 gene activity only in the patients (in

all patients’ granulocytes and in two mononuclear

samples). MMP8 and MMP-9 are metalloproteinases

degrading components of the extracellular matrix; their

increased expression has been detected in many advanced-

stage cancers (mainly in solid tumours), and they

generally contribute to tumour progression [15–17]. The

patients in this study demonstrated significantly increased

MMP-8 and MMP-9 expressions too, moreover, MMP-8

was not expressed in the control samples at all. In both

patients’ cell fractions (except patient 4 granulocytes)

integrin a E was expressed, while it was not detected in

the control samples. Integrins are adhesion molecules

mediating interactions of cells with extracellular matrix

molecules as well as with endothelial and epithelial cells.

Significantly higher expression of integrin alpha E was

TABLE II Gene expression changes in CML patients. The gene expression changes in mononuclears and granulocytes of CML patients obtained bycomparison with the standard profile are represented by a colour, according to the colour scale (Fig. 2). Expressed genes are clustered into two groupsaccording to their function

GENE EXPRESSION IN CML 1291

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

McM

aste

r U

nive

rsity

on

05/0

3/13

For

pers

onal

use

onl

y.

observed on peripheral blood leukocytes (especially on

T-cells) of systemic lupus erythematosus patients [18]

and it is expressed also on hairy leukemic cells [19].

Other gene showing increased expression in the patients

was PCNA, that was strongly expressed in all patients’

mononuclears and in granulocytes of patient 1. In

control samples, its expression was under detectable

level. PCNA is upregulated in actively proliferating

cells. It is a useful immunochemical proliferation

marker in acute leukemia [20] and it has been already

used for detection of blast cells in CML crisis and

other cells associated with blast crisis [21]. Also in

chronic phase of CML, the high PCNA expression is

probably related with proliferation activity of cancer

cells. Elevated expression of MAPK p38 in mono-

nuclears can also contribute to proliferation because

MAPK p38 pathway can be activated in BCR/ABL

transformed cells [22].

TABLE III Gene expression changes in CML patients. The gene expression changes in mononuclears and granulocytes of CML patients obtained bycomparison with the standard profile are represented by a colour, according to the colour scale (Fig. 2). Expressed genes are clustered into two groupsaccording to their function

TABLE IV Gene expression changes in CML patients. The gene expression changes in mononuclears and granulocytes of CML patients obtained bycomparison with the standard profile are represented by a colour, according to the colour scale (Fig. 2). Expressed genes are clustered into two groupsaccording to their function

H. BRUCHOVA et al.1292

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

McM

aste

r U

nive

rsity

on

05/0

3/13

For

pers

onal

use

onl

y.

In some samples hepatoma-derived growth factor

(HDGF), CD59 (membrane attack complex inhibition

factor), basigin precursor (BSG), CDC42 GTPase-

activating gene and akt 1 increased expressions were

found (Tables II–IV). CD59 regulates complement

cascade at the final step, inhibiting formation of

membrane attack complex. In malignant tissues, it may

protect cancer cells from complement-mediated lysis [23].

We observed its higher expression only in mononuclears.

The second gene group includes those genes in which

the expression in patients was lower or was not detected at

all in comparison with the controls (e.g. CDK6, p16

INK4A, PDCD2, cadherin 5, ERBB3, interleukin-1 beta

precursor etc.). Nearly in all patients’ cell fractions, any

CDK6 (positive cell G1/S transition regulator) expression

was not detected. The significant decrease of p16 INK4A

expression in all mononuclear fractions was also observed.

This gene is cyclin-dependent kinase inhibitor and p16

mutations, deletions and inactivation by methylation are

frequently found in acute leukemias, but also in CML

altered p16 gene was detected [24,25]. Transcription

activity reduction of STAT2, STAT5, GADD153 in all

examined granulocytes were found. Although STAT5

activation appears to contribute to malignant transform-

ation in CML [26,27], in our study, we did not found its

overexpression. GADD153 is growth arrest and DNA-

inducible gene which encodes the nuclear protein CHOP

10 inhibiting cell cycle [28]. In our patients, it was

expressed at low level that correlated with CML

progression. ERBB3 belongs to the type 1 family of

growth factor receptors with intrinsic tyrosine kinase

activity. This receptor, which has been implicated in the

development of variety of normal and malignant tissues, is

activated through ligand-mediated homo- and heterodi-

merization [29]. While overexpression of ErbB-3 has been

found commonly in solid human tumours [30], in all the

cell fractions there was no expression observed. Cadherin

5, that mediates endothelial cell–cell contact in the

vascular endothelium and may regulate vascular per-

meability [31], was strongly expressed in control samples

and only patient 1 showed the same expression level.

Expression of interleukin-1 beta was detected in two

mononuclear samples. Interleukin-1 beta (IL-1), primarily

monocyte-derived cytokines, is a proinflammatory cyto-

kine with related and overlapping spectra of activities. It

was recently demonstrated that interleukin (IL)-1 beta

protein levels are elevated in chronic myelogenous

leukemia (CML) and that IL-1 inhibitors can suppress

CML clonogenic growth. Moreover, IL-1 beta levels in

leukocytes were significantly higher in patients in

accelerated/blastic crisis phases of the disease compared

with patients in chronic phase, and the high IL-1 beta

levels correlated with increased blasts. But interleukin-1

receptor antagonist (IL-1RA) levels did not differ between

chronic-phase CML patients and healthy volunteers [32].

Although in our study we did not observed elevated IL-1

beta expression, the expression of IL-1RA in patients’

granulocytes showed the same level as the standard

sample too. IL-1 beta expression correlated with

expression of interleukin-1 beta convertase (IL-1BC)

that processes maturation of IL-1 precursor [33].

Gene Expression After Hydroxyurea Treatment

Gene expression was studied in patients also after one

week of treatment with hydroxyurea. After this treatment,

the number of leukocytes had fallen only in patients 1 and

4, while in patient 2 the leukocyte count changed very

little and in patient 3, it actually increased (Table I). The

gene expression profiles obtained after treatment were

compared with standard ones. Only a small number of

genes changed their activity after hydroxyurea adminis-

tration, majority of them followed behaviour of those at

diagnosis.

Generally, patient 1 showed different expression profile

already at diagnosis and the most changes after treatment

were observed in this patient (e.g. CSPCP, MMP-9,

CDC42 GTPase, IL-1, STAT5, cdc25B, caspase-4, akt 1 in

granulocytes; MMP-9 in mononuclears). It could be

associated with the low leukocyte count that decreased

after treatment at normal level.

Increased expression after treatment in patients was

revealed in interferon-gamma receptor beta, chromatin

assembly factor 1, GADD45, MAPK p38 gluthatione-S-

transferase in mononuclears and CDC42 GTPase activat-

ing protein, RARA in granulocytes. CDC42 GTPase

activating protein (GAP) enhances the intrinsic GTP

hydrolysis of the GTP-binding proteins, thereby ensuring

signal termination. It appears to be an essential component

in the transduction of intracellular signals that induce

actin–cytoskeleton reorganization and gene activation

[34]. At diagnosis, we detected low expression level of

RARA in granulocytes but after treatment, it reached the

level of control sample. RARA is retinoic acid receptor

FIGURE 2

GENE EXPRESSION IN CML 1293

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

McM

aste

r U

nive

rsity

on

05/0

3/13

For

pers

onal

use

onl

y.

and serves as ligand-activated transcription factor.

Ligand-induced activation of RAR alpha is sufficient to

induce differentiation in myeloid cells [35].

Interferon-gamma receptor beta showed elevated gene

activity in mononuclears, it has potent antiproliferative

and apoptotic effects in T-cells. GADD45, increased in

mononuclears after treatment, is stress-inducible gene

associated with cell growth inhibition [36], its expression

could be induced by hydroxyurea. Chromatin assembly

factor 1 p48 is one of the major proteins that binds to a

putative functional domain at the carboxy terminus of the

Rb protein and may mediate suppression of cell growth

[37].

Significant down-regulation after treatment was

observed in VEGFR-1 in mononuclears that plays

important roles in the angiogenesis and in monocyte/

macrophage migration [38]. Most of the genes showed

expression decrease in only some samples (e.g. MMP-9,

integrin alpha E, integrin beta8, c-raf protooncogene).

Reduction of integrin alpha E gene activity was detected in

mononuclears of only patients 1 and 4 with decreased

leukocyte count. Distinct expression behaviour of genes in

patients indicates that the patients respond differently to

the initial period of treatment, that is evident also in their

leukocyte counts.

DISCUSSION

This study of four CML patients investigated at the time of

diagnosis and one week after the initiation of anticancer

treatment revealed similarity and heterogeneity, both of

which were clinically well-known and to be expected. The

efficacy of hydroxyurea treatment also follows an

individual-dependent pattern. Thus, it is not only the

differences in gene activity characteristic for the

individual cell types analysed, but also polymorphisms

in metabolic and signalling pathways have to be taken into

account. The situation is rather complex, and laying bare

more or less important gene relations is not simple. A

certain advantage has been brought by “biochip”

technology, which may be able to contribute to the

elucidation of the complete genetic background of

leukemias, and this would point to other questions that

would need to be solved (the 588 genes on the Clontech

Cancer Array represent only about 0.5% of all human

genes). Some approaches have already been suggested by

the results of the separation of leukocytes into granulocyte

and mononuclear fractions. On the basis of the results of

this study, it should be possible to concentrate attention on

those genes in which significant gene expression changes

were observed in most patients—such as JNK2,

metalloproteinases, integrins, PCNA, CDK6, p16, cad-

herin 5 etc.— which may contribute to CML development.

In each case, it would be necessary to investigate a larger

number of patients with various types of gene rearrange-

ment (major, minor and micro), and to analyse a larger set

of genes. The observations in this study underline the

individuality of the disease process. For a better under-

standing of the effect of hydroxyurea, it is a requisite to

test other sets of genes that could, via their products, affect

the metabolism and function of ribonucleotide reductase

(e.g. superoxide dismutase, catalase and others). It would

be also interesting to follow gene expression in patients

during treatment (in longer intervals) when it is achieved

haematological and molecular remission.

Acknowledgements

The authors would like to thank J. Moravcova, J. Polak

and K. Michalova for molecular and cytogenetic data.

This project was supported by grant no. NM5901-3 of the

Internal Grant Agency of the Ministry of Health of the

Czech Republic.

References

[1] Bazzoni, G., Carleso, N., Griffin, J.D. and Hemler, M.E. (1996)“Bcr/Abl expression stimulates integrins function in hematopoieticcell lines”, J. Clin. Investig. 98, 521–528.

[2] Zhao, R.C., Tarone, G. and Verfaillie, C.M. (1997) “Presence of theadhesion inhibitory b1B integrin isoform on CML but not normalprogenitors is at least in part responsible for decreased CMLprogenitor adhesion”, Blood 90(1), 393a.

[3] Bedi, A., Zehnbauer, B.A., Barber, J.P., Sharkis, S.J. and Jones, R.J.(1994) “Inhibition of apoptosis by BCR/ABL in chronic myeloidleukemia”, Blood 83, 2034–2044.

[4] Cortez, D., Stoica, G., Pierce, J.H. and Pendergast, A.M. (1996)“The BCR/ABL tyrosine kinase inhibits apoptosis by activation aRas-dependent signaling pathway”, Oncogene 13, 2589–2594.

[5] Cambier, N., Chopra, R., Strasser, A., Metcalf, D. and Elefanty,A.G. (1998) “BCR/ABL pathways mediating cytokine indepen-dence and protection against apoptosis in murine hematopoieticcells in dose-dependent manner”, Oncogene 16, 335–348.

[6] Giralt, S.A., Kantarjian, H.M. and Talpaz, M. (1995) “Treatment ofchronic myelogenous leukemia”, Semin. Oncol. 22, 396–404.

[7] Hehlmann, R. and Heimpel, H. (1996) “Current aspects of drugtherapy in Philadelphia—positive CML: correlation of tumorburden with survival”, Leuk. Lymphoma 22, 161–167.

[8] Hehlmann, R., Heimpel, H. and Hasford, J. (1994) “Randomizedcomparison of interferon alpha with bushulfan and hydroxyurea inchronic myelogenous leukemia”, Blood 84, 4064–4077.

[9] Thiele, J., Kvasnicka, H.M., Schmitt-Graeff, A., Bundschuh, S.,Biermann, T., Roessler, G., Wasmus, M., Diehl, V., Zankovich, R.and Shaefer, H.E. (2000) “Effects of chemotherapy (hydroxyurea–busulfan) and interferon alpha on bone marrow morphologicfeatures in chronic myelogenous leukemia”, Am. J. Clin. Pathol.114, 57–65.

[10] Chomczynsky, P. and Sacchi, N. (1987) “Single step method ofRNA isolation by acid quanidin–isothiocyanate–phenol–chloro-form extraction”, Anal. Biochem. 162, 156–159.

[11] Boyum, A. (1968) “Isolation of mononuclear cells and granulocytesfrom human blood”, Scand. J. Clin. Lab. Investig. 21(Suppl. 97),77–89.

[12] Hardy, K. and Chaudhri, G. (1997) “Activation and signaltransduction via mitogen-activated protein (MAP) kinases in Tlymphocytes”, Immunol. Cell. Biol. 75, 528–545.

[13] Chopra, R., Pu, Q.Q. and Elefanty, A.G. (1999) “Biology of BCR–ABL”, Blood Rev. 13, 211–229.

[14] Raitano, A.B., Halpern, J.R., Hambuch, T.M. and Sawyers, C.L.(1995) “The Bcr–Abl leukaemia oncogene activates Jun kinase andrequires Jun for transformation”, Proc. Natl. Acad. Sci. USA 92,11746–11750.

[15] Rooprai, H.K. and McCormick, D. (1997) “Proteases and theirinhibitors in human brain tumours”, Anticancer Res. 17,4151–4162.

[16] O-charoenrat, P., Rhys-Evans, P., Court, W.J., Box, G.M. andEccles, S.A. (1999) “Differential modulation of proliferation,

H. BRUCHOVA et al.1294

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

McM

aste

r U

nive

rsity

on

05/0

3/13

For

pers

onal

use

onl

y.

matrix metalloproteinase expression and invasion of human headand neck squamous carcinoma cells by c-erbB ligands”, Clin. Exp.Metastasis 17(7), 631–639.

[17] Kossakowska, A.E., Urbanski, S.J., Watson, A., Hayden, L.J. andEdwards, D.R. (1993) “Patterns of expression of metalloproteinasesand their inhibitors in human malignant lymphomas”, Oncol. Res. 5,19–28.

[18] Pang, M., Abe, T., Fujihara, T., Mori, S., Tsuzaka, K., Amano, K.,Koide, J. and Takeuchi, T. (1998) “Up-regulation of alphaEbeta7, anovel integrin adhesion molecule, on T cells from systemic lupuserythematosus patients with specific epithelial involvement”,Arthritis Rheum. 41(8), 1456–1463.

[19] Troussard, X., Maloisel, F. and Flandrin, G. (1998) “Hairy cellleukemia. What is new forty years after the first description?”,Hematol. Cell. Ther. 40(4), 139–148.

[20] Ito, M., Tsurusawa, M., Zha, Z., Kawai, S., Takasaki, Y. andFujimoto, T. (1992) “Cell proliferation in childhood acute leukemia.Comparison of Ki-67 and proliferating cell nuclear antigenimmunocytochemical and DNA flow cytometric analysis”, Cancer69(3), 655–661.

[21] Takasaki, Y., Robinson, W.A. and Tan, E.M. (1984) “Proliferatingcell nuclear antigen in blast crisis cells of patients with chronicmyeloid leukemia”, J. Natl. Cancer Inst. 73(3), 655–661.

[22] Deininger, M.W.N., Goldman, J.M. and Melo, J.V. (2000) “Themolecular biology of chronic myeloid leukemia”, Blood 96(10),3343–3356.

[23] Maio, M., Brasoveanu, L.I., Coral, S., Sigalotti, L., Lamaj, E.,Gasparollo, A., Visintin, A., Altomonte, M. and Fonsati, E. (1998)“Structure, distribution, and functional role of protectin (CD59) incomplement-susceptibility and immunotherapy of human malig-nancies”, Int. J. Oncol. 13(2), 305–318.

[24] Chen, W., Zhu, J., Liu, J. and Tan, S. (1998) “Methylation of p16gene in hematological malignancies”, Chin. Med. J. 111(11),1028–1030.

[25] Guran, S., Bahce, M., Beyan, C., Korkmaz, K. and Yalcin, A. (1998)“P53, p15INK4B, p16INK4A and p57KIP2 mutations during theprogression of chronic myeloid leukemia”, Haematologica 29(3),181–193.

[26] de Groot, R.P., Raaijmakers, J.A., Lammers, J.W., Jove, R. andKoenderman, L. (1999) “STAT5 activation by BCR–ABLcontributes to transformation of K562 leukemia cells”, Blood 94,1108–1112.

[27] Shuai, K., Halpern, J., ten Hoeve, J., Rao, X. and Sawyers, C.L.(1996) “Constitutive activation of STAT5 by the BCR–ABL

oncogene in chronic myelogenous leukemia”, Oncogene 13(2),247–254.

[28] Eymin, B., Dubrez, L., Allouche, M. and Solary, E. (1997)“Increased gadd153 messenger level is associated with apoptosis inhuman leukemic cells treated with etoposide”, Cancer Res. 57(4),686–695.

[29] Earp, H.S., Dawson, T.L., Li, X. and Yu, H. (1995) “Hetero-dimerization and functional interaction between EGF receptorfamily members: a new signaling paradigm with implications forbreast cancer research”, Breast Cancer Res. Treat. 35(1), 115–132.

[30] Gullick, W.J. (1996) “The c-erbB3/HER3 receptor in humancancer”, Cancer Surv. 27, 339–349.

[31] Groten, T., Kreienberg, R., Fialka, I., Huber, L. and Wedlich, D.(2000) “Altered subcellular distribution of cadherin-5 in endothelialcells caused by the serum of pre-eclamptic patients”, Mol. Hum.Reprod. 6(11), 1027–1032.

[32] Wetzler, M., Kurzrock, R., Estrov, Z., Kantarjian, H., Gisslinger, H.,Underbrink, M.P. and Talpaz, M. (1994) “Altered levels ofinterleukin-1 beta and interleukin-1 receptor antagonist in chronicmyelogenous leukemia: clinical and prognostic correlates”, Blood84(9), 3142–3147.

[33] Howard, A.D., Chartrain, N., Ding, C.F., Kostura, M.J., Limjuco,G., Schmidt, J.A. and Tocci, M.J. (1991) “Probing the role ofinterleukin-1 beta convertase in interleukin-1 beta secretion”,Agents Actions Suppl. 35, 77–83.

[34] Symons, M. (1996) “Rho family GTPases: the cytoskeleton andbeyond”, Trends Biochem. Sci. 21(5), 178–181.

[35] Scott, L.M., Mueller, L. and Collins, S.J. (1996) “E3, ahematopoietic-specific transcript directly regulated by the retinoicacid receptor alpha”, Blood 88(7), 2517–2530.

[36] Santucci, M.A., Ripalti, A., di Paola, M.C., Mianulli, A.M., Iacurti,E., Campanini, F., Gamberi, B. and Tura, S. (1999) “Procedure forthe quantitation of Gadd45 expression levels in clonal hematopoie-tic progenitor cells by competitive polymerase chain reaction”,Clin. Biochem. 32(1), 1–8.

[37] Qian, Y.W., Wang, Y.C., Hollingsworth, Jr, R.E., Jones, D., Ling, N.and Lee, E.Y. (1993) “A retinoblastoma-binding protein related to anegative regulator of Ras in yeast”, Nature 346(6438), 648–652.

[38] Gerber, H.P., Condorelli, F., Park, J. and Ferrara, N. (1997)“Differential transcriptional regulation of the two vascularendothelial growth factor receptor genes. Flt-1, but not Flk-1/KDR, is up-regulated by hypoxia”, J. Biol. Chem. 272(38),23659–23667.

GENE EXPRESSION IN CML 1295

Leu

k L

ymph

oma

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

McM

aste

r U

nive

rsity

on

05/0

3/13

For

pers

onal

use

onl

y.