www.elsevier.com/locate/ybcmd

Blood Cells, Molecules, and Diseases 32 (2004) 262–269

Interferon-alpha suppresses proliferation of chronic myelogenous

leukemia cells K562 by extending cell cycle S-phase without

inducing apoptosis

Dana Grebenova,a Katerina Kuzelova,a Ota Fuchs,a Petr Halada,b Vladimır Havlıcek,b

Iuri Marinov,a and Zbynek Hrkala,*

aDepartment of Cellular Biochemistry, Institute of Hematology and Blood Transfusion, 128 20, Prague-2, Czech Republicb Institute of Microbiology AS CR, Prague-2, Czech Republic

Submitted 17 September 2003; revised 23 October 2003

(Communicated by M. Lichtman, M.D., 30 October 2003)

Abstract

We examined the effects of interferon-alpha (IFN-a) treatment on the growth, cell cycle, proliferation, and apoptotic parameters as well as

adhesive properties and proteome of chronic myelogenous leukemia (CML)-derived K562 cells. IFN-a treatment (200 to 600 U/ml, 24 to

72 h) suppressed growth and caused accumulation of K562 cells in the S-phase of cell cycle (increase in S-phase cells by up to 52% in

comparison with the untreated controls) at the expenses of cells in G1-phase. No transition of cells to G0-phase occurred as followed from

Ki-67 protein determination. Although the level of chimeric gene product, BCR-ABL mRNA coding for BCR-ABL protein with anti-

apoptotic properties, decreased by 30%, apoptosis was not triggered as judged from Annexin-V, APO2.7, and TUNEL assays. Adhesion of

K562 cells to fibronectin-coated surfaces increased by up to 52% as determined by calcein assay. The proteomic analysis (2-D electrophoresis

in combination with mass spectrometry, MALDI-MS) revealed a single protein, ubiquitine cross-reactive protein (UBCR), whose level

markedly increased due to IFN-a treatment. The ubiquitination-like directed degradation processes may thus play a role in the mechanism of

IFN-a antiproliferative effects.

D 2003 Elsevier Inc. All rights reserved.

Keywords: K562; IFN-a; Cell cycle; Apoptosis; Adhesion; Proteomic analysis

Introduction with IFN-a therapy are its toxicity at the optimal therapeutic

IFN-a belongs to a family of proteins secreted by

eukaryotic cells following challenge by viruses and other

infectious agents [1]. In addition to the antiviral effects [2],

interferon-alpha (IFN-a) exerts pleiotropic cellular effects

including inhibition of cell proliferation [3], induction of

apoptosis [4,5], and modulation of the immune system [6].

IFN-a is able to induce a state of tumor dormancy and to

control the chronic phase of CML. IFN-a causes complete

genetic remission in 10% to 20% patients with chronic

phase CML [7], in some patients (approximately 7%) even

without further therapy [8]. The main problems encountered

1079-9796/$ - see front matter D 2003 Elsevier Inc. All rights reserved.

doi:10.1016/j.bcmd.2003.10.008

* Corresponding author. Department of Cellular Biochemistry, Institute

of Hematology and Blood Transfusion, U Nemocnice 1, 128 20 Prague-2,

Czech Republic. Fax: +420-224918390.

E-mail address: hrkal@uhkt.cz (Z. Hrkal).

doses and low efficiency in blast phase of CML. Although

the therapeutic efficacy of IFN-a for the treatment of

chronic phase CML is well established, the mechanism by

which IFN-a exerts the direct antitumor action on CML

cells is an open question. It has been reported that IFN-a

binding to type I IFN receptor activates JAK/STAT pathway

leading to the formation of interferon-stimulated gene factor

3 (ISGF3) complex, which activates the interferon-stimulat-

ed responsive element (ISRE) sequences on the related

genes [9]. Induction of cell cycle inhibitors is the commonly

accepted mechanism of IFN-a antiproliferative effects [10].

Frequently IFN-a treatment leads to G1 or G0 arrest in IFN-

a responsive cells or induces apoptosis [5,10].

The chronic myelogenous leukemia (CML)-derived

K562 cells [11] bear an abnormal Ph chromosome charac-

terized by the abnormal fusion gene BCR-ABL of the b3a2

type [12]. This gene expresses mRNA, which is translated to

D. Grebenova et al. / Blood Cells, Molecul

the chimeric 210-kDa fusion protein BCR-ABL whose

constitutively elevated Abl kinase activity is responsible

for the known resistance of K562 cells to anticancer drug-

induced apoptosis [13]. In addition to abnormal gene BCR-

ABL, K562 cells also have mutated gene for the p53 protein

[14] whose proapoptotic or cell cycle regulatory function is

thus annulled. These combined mutations make of K562

cells a suitable model for the ‘in vitro’ studies of CML blast

phase treatment. In our recent paper, we deal with the

mechanism of cytotoxic effects imposed on K562 cells by

the photodynamic treatment showing that an early apoptotic

process, which is triggered by this procedure, does not

proceed to the terminal phase and is followed by the cell

necrosis [15]. In this communication, we report on the

growth inhibitory effect imposed on wild-type K562 cells

by IFN-a treatment and its mechanism.

Materials and methods

Chemicals

Human recombinant interferon-aA/D, propidium iodide,

RPMI-1640, mouse monoclonal antibodies to cyclin A, h-actin, and p21WAF1/Cip1, rabbit antibodies to cyclins D1 and

E, were purchased from Sigma (Prague, Czech Republic)

and rabbit antibody to retinoblastoma protein (pT821) was

purchased from BioSource Europe, S.A., Belgium. The

APO 2.7 kit was purchased from Coulter/Immunotech

A.S., Prague, Czech Republic; FITC-conjugated mouse

anti-human Ki-67 antibody set was purchased from BD

Biosciences; Annexin-V-FLUOS staining kit and In situ cell

death detection kit, fluorescein, were purchased from Roche

Applied Science (Mannheim, Germany); Vybrant (TM) cell

adhesion kit was purchased from Molecular Probes Europe,

BV (Leiden, The Netherlands); and FIX and PERM cell

permeabilization kit was purchased from An Der Grub,

Kaumberg, Austria.

Cell culture

K562 cells were purchased from the European Collection

of Animal Cell Cultures (Salisbury, UK) and cultured in

RPMI 1640 medium supplemented with 10% fetal calf

serum (FCS), 100 units/ml penicillin, 50 Ag/ml streptomy-

cin, and 2 mM L-glutamine at 37jC in 5% CO2 humidified

atmosphere. Cells were diluted to a density of 2 � 105 cells

per ml three times a week.

Cell viability

The effect of IFN-a treatment on the cell viability was

assessed by flow cytometry of propidium iodide (PI)-stained

cells using Coulter Epics XL flow cytometer as previously

described for photodynamic therapy experiments [16] and

by visual cell counting (trypan blue exclusion test).

3H-Thymidine proliferation assay

Four sample replicates of IFN-a-treated cells as well as of

the appropriate controls (without IFN-a) were pulsed with

24 kBq (6-3H)-thymidine (Institute for Research, Develop-

ment and Application of Radioisotopes, Prague, Czech

Republic, specific activity 980 Gbq/mmol) for 4 h and the

cells were collected using Scatron cell harvester. The incor-

porated radioactivity into the newly synthesized DNA was

measured with the beta scintillation counter. The mean

values of the quadruplicates expressed in counts per minute

were calculated, and the suppression of DNA synthesis

caused by IFN-a was given as the percentage of control

values [17].

Cell cycle analysis

The cells (1 � 106) were collected by centrifugation,

suspended in 70% ethanol, and incubated for 30 min at room

temperature in 1 ml of the modified Vindelovs propidium

iodide buffer (10 mM Tris, pH 8, 1 mM NaCl, 0.1% Triton

X-100, 20 Ag/ml PI, and 10 Kunitz units of ribonuclease A).

The red fluorescence excited at 488 nm was then measured

using Coulter Epics XL flow cytometer. The histograms

were analyzed using the G1/G2M Only Fit method.

Detection of cell cycle antigen Ki-67

Ki-67 protein expression was analyzed using the direct

immunofluorescence method. After the treatment with IFN-

a for defined periods (0 to 72 h), 1 � 106 cells were

harvested and stained following the protocol for Intracellu-

lar Staining With FIX and PERM (BIOZOL Diagnostica

Vertrieb GmbH, Eching, Germany) using FITC-conjugated

anti-Ki-67 antibody. The fraction of Ki-67 antigen express-

ing cells was determined by flow cytometry.

Exposition of mitochondrial membrane antigen 7A6

Expression of mitochondrial membrane neoantigen 7A6

appearing in apoptotic cells was measured in IFN-a-treated

cells as well in appropriate controls by flow cytometry

employing antibody APO 2.7 [18].

Assessment of apoptosis by TUNEL assay

The TUNEL assay was performed employing the In situ

cell death detection kit, fluorescein (Roche Applied Sci-

ence), following the standard manufacturer’s protocol. The

extent of DNA labeling with fluorescein-dUTP was deter-

mined by flow cytometry.

Annexin-V apoptosis assay

The analysis of phosphatidylserine exposition on the

outer side of plasma membrane of IFN-a-treated cells as

es, and Diseases 32 (2004) 262–269 263

D. Grebenova et al. / Blood Cells, Molecules, and Diseases 32 (2004) 262–269264

well as of controls was performed using Annexin-V-FLUOS

staining kit (Roche Applied Science), employing flow

cytometry. The kit consists of Annexin-V-fluorescein and

propidium iodide for the differentiation of apoptotic and

necrotic cells. The recommended manufacturer’s protocol

was obeyed.

Electrophoresis and Western blotting

The cells (5 � 106) were suspended in a lysis buffer

containing 0.15 M NaCl, 1 mM PMSF (phenylmethanesul-

fonyl fluoride), 1 Al/100 Al of phosphatase inhibitor cocktail2, and 0.5% Triton X-100 (Sigma) and kept for 20 min on

ice. Protein samples were heated to 100jC for 4 min in the

presence of 5% 2-mercaptoethanol, chilled, and subjected to

one-dimensional electrophoresis (PAGE) in 12% or 15% gel

with SDS according to Laemmli [19]. As a rule, 10 Ag total

protein was applied to each well. Proteins were transferred

to Hybond-ECL membrane at 10 V for 30 min. Nitrocellu-

lose membranes were blocked with 5% nonfat milk in TBS-

T (Tris-buffered saline, 0.1% Tween 20) and incubated for

1 h with the appropriate antibody in TBS-T. Immunoblots

were reacted in parallel with anti-h-actin as the IFN-a

unaffected controls. After washes with TBS-T, the mem-

branes were incubated with horseradish peroxidase-conju-

gated anti-mouse or anti-rabbit secondary antibodies. The

antigens were detected using the enhanced chemilumines-

cence Western blotting detection system ECL + PLUS

(Amersham Pharmacia Biotech, UK) according to the man-

ufacturer’s instructions and visualized by autoradiography

on X-ray film. Intensity of protein bands was compared by

densitometry evaluation using AIDAVersion 2.1, 1D-Eval-

uation system (Raytest GmbH, Germany).

Cell adhesion assay

Adhesion of IFN-a-treated K562 as well as control cells

to fibronectin surface was determined using Vybrantk cell

adhesion kit employing the manufacturer’s protocol. The

microtitration plate wells (Microfluor 2 Flat Bottom Micro-

titer Plates, Dynex Technologies, Chantilly, VA, USA) were

coated with fibronectin (100 Al fibronectin, 25 Ag/ml in

PBS) at 4jC overnight. The cells (5 � 106/ml) were labeled

with calcein-AM (5 AM, 37jC, 30 min). The cell density

was adjusted to 1 � 106 cells/ml and 100 Al aliquots of cellsuspension were transferred to the wells and incubated for

2 h at 37jC. Finally, fluorescence of adhering cells was

measured (excitation 485 nm, emission 520 nm) using

FluoStar Galaxy fluorescence microplate reader (BMG

Labtechnologies, GmbH, Offenburg, Germany).

Analysis of BCR-ABL gene transcript

Total RNAwas isolated from K562 cells using RNazol B

(Tel-Test Inc, Friendswood, TX, USA) according to manu-

facturer’s recommendations. Semiquantitative RT-PCR was

performed on cDNA templates generated by reverse tran-

scription (Superscript II, Gibco, Life Technologies, Rock-

ville, MD, USA) from 2 Ag of DNase I treated total RNA in

a 20-Al reaction volume. The BCR-ABL fusion transcript

was amplified using bcr-2-ex-sense primer 5V-TTCA-GAAGCTTCTCCCTG-3V and abl-2-ex-antisense primer

5V-CTCCACTGGCCACA-AAT-3V. For PCR amplifications,

the PTC-200 Peltier Thermal Cycler (MJ Research Inc.,

Waltham, Massachusetts) was employed. PCR products

were examined after electrophoresis in ethidium bromide

stained 1.5% agarose gel, and fluorescence signals were

evaluated with Fuji FLA-2000 phosphoimager (Raytest

GmbH) and quantitated by AIDA version 2.1 software

(Raytest GmbH). The results are the average of three

independent experiments.

Two-dimensional electrophoresis (2-DE)

The linear immobilized pH gradient gels (180� 3.3� 0.5

mm), ReadyStripk IPG strips pH 3 to 10 (BioRad Hercules,

CA, USA), were placed in the Immobiline DryStrip Reswel-

ling Tray (Amersham Biosciences, Vienna, Austria) and

rehydrated overnight in 315 Al of the cell lysate containing

150 Ag proteins, 8 M urea, 2% CHAPS, 65 mM DTT, 0.2%

Bio-Lyte 3-10, and 0.01% bromphenol blue. Isoelectric

focusing (IEF) was performed using Multiphor II system

(Amersham Biosciences) at 6000 V for a total 70 kV/h at

20jC. After the first dimension run, the strips were equili-

brated with a solution containing Tris–HCl pH 8.8 (0.375

mM), urea (6 M), glycerol (20% v/v), DTT (2% w/v), SDS

(2% w/v), and 0.01% bromphenol blue. Resulting free SH

groups were blocked in the second equilibration step in

which DTT was replaced with iodoacetamide (2.5% w/v).

The second dimension run was performed using a

Protean II xi vertical electrophoresis system (BioRad Her-

cules), gel size 200 � 200 � 1.5 mm, in a Laemmli-SDS-

discontinuous buffer system [19] and polyacrylamide gel

gradient (9% to 16% T, 2.6% C). The equilibrated first

dimension gel was placed directly onto the second dimen-

sion gel and overlaid with a solution containing 0.4%

agarose in Tris–glycine–SDS pH 8.3 buffer (25 mM Tris,

192 mM glycine, 0.1% w/v SDS) heated to 70jC. Electro-phoresis was performed at a constant current 35 mA/gel for

5 h at 10jC. The gels were fixed in a solution containing

ethanol (40% v/v) and acetic acid (10% v/v) before staining

with silver nitrate. The analytical gels (protein load 150 Ag)were silver stained as described by Bjellqvist et al. [20] and

the preparative gels (protein load 1500 Ag) according to

Schevchenko et al. [21].

2-D PAGE gel imaging

Silver-stained gels were scanned using a scanner UMAX

PowerLook III (1200 � 2400 dpi) (Umax Systems GmbH,

Germany). The 2-DE image computer analysis was carried

out employing AIDA Proteomix 2D Protein Gel Matching

Fig. 1. Effect of IFN-a on the growth of K562 cells. The cells were cultured

in the absence and presence of 200 to 800 U/ml IFN-a for 72 h and

regularly counted.

D. Grebenova et al. / Blood Cells, Molecules, and Diseases 32 (2004) 262–269 265

Version 6.00 software (Raytest GmbH). Individual gels of

IFN-a-treated cells were compared with the gels of untreat-

ed cells and vice versa. The differences in spot volumes

greater than 50% (integrated optical density over spot area)

were considered significant. The findings are based on the

evaluation of at least ten 2-D PAGE experiments. The

isoelectric points and molecular weights of individual pro-

teins were approximated using polypeptide 2-D SDS PAGE

Standards (BioRad, Richmond, USA).

MALDI mass spectrometry and protein identification

Silver-stained protein spots were excised from the gel

and the gel pieces reconstituted in a cleavage buffer con-

taining 2-sulfanylethan-1-ol, 50 mM 4-ethylmorpholine

Fig. 2. Effect of IFN-a on the proliferation of K562 cells assayed by 3H-thymidine

to 72 h, and the cell aliquots in quadruplicates were incubated with 3H-thymidine

measured in counts/min. The data are normalized to the untreated controls (100%

acetate, 1 mM CaCl2, 10% acetonitrile (MeCN), and se-

quencing grade trypsin (50 ng/Al; Promega, Madison, WI,

USA). The digestion was carried out overnight and resulting

peptides were extracted with 30% MeCN/0.5% trifluoro-

acetic acid [22]. One microliter aliquot of each peptide

mixture was deposited on the target, and 1 Al of a matrix

solution (saturated a-cyano-4-hydroxycinnamic acid in 30%

MeCN/0.2% trifluoroacetic acid) was added. Positive ion

mass spectra were measured on an MALDI reflectron time-

of-flight mass spectrometer BIFLEX (Bruker-Franzen, Bre-

men, Germany) equipped with a nitrogen laser (337 nm) and

gridless delayed extraction ion source. Protein spots were

identified by searches of peptide mass maps in a nonredun-

dant database NCBI using the search program ProFound

(http://prowl.rockefeller.edu/cgi-bin/ProFound) [22].

Results

We examined the effects of IFN-a treatment on the

growth of wild-type K562 cells. In Fig. 1, the time depen-

dence of cell density is shown for several IFN-a doses. It

follows that doses of 200 U/ml and higher suppressed

progressively the cell growth. The decrease in cell growth

was not due to the IFN-a cytotoxicity since fractions of

propidium iodide-positive cells (around 5%) as well as

number of Trypan blue-positive cells (3% to 4%) found in

untreated controls did not increase during 48 h incubation

with IFN-a at concentrations 200 to 800 U/ml.

To examine the effect of IFN-a on the proliferation of

K562 cells, we used 3H-thymidine incorporation into the

cell DNA. The extent of 3H-thymidine incorporation was

reduced by 40% to 50% following 48 h and by approxi-

incorporation. The cells were incubated with 200 to 800 U/ml IFN-a for up

for 4 h. The cells were harvested and the 3H-thymidine radioactivity was

).

D. Grebenova et al. / Blood Cells, Molecules, and Diseases 32 (2004) 262–269266

mately 75% following 72 h incubation with 200 to 800 U/ml

IFN-a in comparison with the untreated controls (Fig. 2).

As apoptosis is considered a possible mechanism by

which IFN-a executes its anticancer effects, we determined

several apoptotic parameters of K562 cells during their

treatment with IFN-a. The amount of TUNEL-positive cells

Fig. 3. DNA content analysis of the controls (panels 1 to 4) and IFN-a-treated K5

up to 72 h, and in time intervals the flow cytometry histograms were obtained

RNAse). The control cells were treated equally with the exception of IFN-a add

(see enclosures).

was approximately 1% in the control and their number did

not increase during 72 h cultivation in the presence of 200

and 400 U/ml IFN-a. Similarly, the subG1 peak was not

observed on cell cycle histograms (see Fig. 3). Further,

almost no AnnexinV-positive cells have been detected by

flow cytometry (fractions of AnnexinV-positive cells were

62 cells (panels 5 to 7). The cells were incubated with 200 U/ml IFN-a for

of permeabilized cells stained with propidium iodide solution (containing

ition. The cell cycle analysis was performed by G1/G2M Only Fit method

Fig. 5. Effect of IFN-a on the adhesive properties of K562 cells. The cells

were treated with IFN-a (200 and 400 U/ml) for 72 h. In time intervals, the

relative numbers of cells adhering to microtitration plate wells previously

coated with fibronectin were determined by fluorescence measurement of

calcein-AM-stained cells (Vybrant cell adhesion assay). The data were

normalized to the adherent fractions of untreated K562 cells (100%).

D. Grebenova et al. / Blood Cells, Molecules, and Diseases 32 (2004) 262–269 267

about 1%), and the fraction of APO2.7-positive cells (cells

exhibiting the mitochondrial apoptosis-related antigen 7A6)

was only 3% to 5% in both cases after 48 h incubation with

200 to 800 U/ml IFN-a. We conclude that apoptosis is not a

mechanism responsible for the suppressed growth of wild-

type K562 cells caused by IFN-a treatment.

Another possible mechanism by which cell growth can

be limited is the transition of cells from G1 to G0 phase. We

therefore determined the fraction of cells in the overall

active phases of cell cycle (G1, S, G2, and M) by flow

cytometry using the antibody against the proliferation-relat-

ed protein Ki-67 (MKI-67 protein). It followed that the

active phases of cell cycle were not affected by IFN-a

treatment as the fraction of Ki-67-positive cells represented

91% and 96% both in the control and IFN-a-treated cells

(200 U/ml, 48 and 72 h). Thus, no transition of cells from

G1 to G0 phase occurred. Subsequently, we examined the

effect of IFN-a treatment on the cell cycle distribution of

K562 cells. In Fig. 3, a series of flow cytometry histograms

are presented showing changes in DNA distribution during

K562 treatment with 200 U/ml IFN-a. Analysis of DNA

distribution by G1/G0-G2/M Only Fit method showed

pronounced changes that occurred during IFN-a treatment.

The fractions of cells in G1 as well as G2/M phase

decreased while the fraction in S phase increased (see

enclosures). Since we assumed that suppressed proliferation

of K562 cells may be due to the perturbation of certain cell

cycle regulatory proteins, we determined by Western blot-

ting the effects of IFN-a treatment on the expression of

cyclins D1, E, and A, the cell cycle inhibitor, protein p21,

and on the phosphorylation state of Rb protein (Fig. 4). We

found no increased expression of cyclins D1 and E and the

cell cycle inhibitor p21. On the other hand, an increase by

30% in the level of cyclin A as well as in the phosphory-

lation state of Rb protein has been observed.

Fig. 4. Effect of IFN-a on the expression of cell cycle proteins and inhibitors dete

72 h, they were lysed, subjected to SDS electrophoresis, and blotted to nitrocellu

developed with peroxidase-conjugated secondary antibody, and detected by ECL

The clinically relevant question, whether IFN-a can

affect expression of BCR-ABL gene in blast phase CML

cells, has been addressed by determining BCR-ABL mRNA

level following bcr-abl gene amplification by PCR. The

amounts of BCR-ABL mRNA decreased by 30.4% F 4.4%

following K562 cells IFN-a treatment (200 U/ml, 48 h). We

further investigated the effect of IFN-a treatment on the

adhesive properties of K562 cells. The fraction of cells

adhering to fibronectin surface increased by up to 53% due

to IFN-a treatment in dependence of IFN-a concentration

and incubation time (Fig. 5). Finally we attempted by 2-D

electrophoresis to uncover proteins that were differentially

expressed in K562 cells due to the IFN-a treatment. Fig. 6

shows the section of 2-D electrophoretic polypeptide map of

rmined by Western blotting. The cells were incubated with IFN-a for up to

lose membrane. The blots were incubated with specific primary antibodies,

.

Fig. 6. Proteomic analysis of IFN-a-treated K562 cells. Two-dimensional

electrophoresis (a section) of the lysate of K562 cells untreated (panel A)

and treated with IFN-a (panel B; 200 U/ml, 48 h) stained with silver. The

protein spot newly appearing on the electrophoreogram of IFN-a-treated

cells is depicted by an arrow.

D. Grebenova et al. / Blood Cells, Molecules, and Diseases 32 (2004) 262–269268

K562 cells that reveals a newly appearing protein spot after

treatment with 200 U/ml IFN-a for 48 h. This protein was

identified by mass spectrometry (MALDI-MS) as the ubiq-

uitine cross-reactive protein (UCRP).

Discussion

We employed the K562 cell line as the model of CML

blast phase, as this cell line not only bears the Philadelphia

chromosome and expresses BCR-ABL gene, but it also

possesses an additional mutation in gene p53 resulting in

the expression of nonfunctional protein p53. This protein

normally functions as the anti-oncogenic compound induc-

ing G1/S cell cycle arrest following DNA damage. IFN-a

treatment suppressed growth of K562 cells; however, nei-

ther apoptosis was triggered nor necrosis occurred. In the

recent paper dealing with IFN-a pro-apoptotic effects on

K562 cells [23], the authors have shown pronounced

apoptosis in these cells due to IFN-a treatment and even

in the untreated controls. In contrast to these results, we did

not find any increase of apoptosis above the (low) level

found in untreated controls. The explanation we offer is that

cell lines behavior is much dependent on the particular

strand employed; our cells obtained from European Collec-

tion of Animal Cell Cultures, UK, are probably closer to the

original type established by Lozzio and Lozzio [11].

Possible transition of cells from G1 to G0 phase has been

excluded by the measurements of proliferation-related pro-

tein Ki-67, which is present in all active phases of the cell

cycle. The fraction of cells expressing this protein

approached completion and was not affected by IFN-a

treatment, suggesting that no transition of cells to G0 phase

occurred. The suppressive effect must therefore rely in a

perturbation of the cell cycle. The decrease in the G1 phase

as well as G2/M phase cells due to IFN-a treatment was

compensated by an increase of S-phase cells. Since the cell

density as well as overall DNA replication rate decreased,

the possible explanation is that the cells were either arrested

in S-phase or the DNA replication was slowed down. We

tried to uncover the molecular mechanism, which is behind

the antiproliferative effects of IFN-a. However, neither

cyclins D1 and E, which drive cells through G1/S phase

restriction point, nor the cell cycle inhibitor protein p21

were up-regulated. Thus, the IFN-a antiproliferative mech-

anism does not involve the G1 phase of the cell cycle. To the

contrary, an increase of S-phase cells supports the idea on

slowing down the DNA replication leading to accumulation

of cells in S-phase. In accordance with this theory, we

observed an increased level of cyclin A and increased

phosphorylation of retinoblastoma protein due to IFN-a

treatment. The decrease of G2/M phase cells (Fig. 3)

resulting from the delay of cells in S phase leads to the

lower number of replicating cells and therefore to the

suppressed proliferation.

Cell adhesion to extracellular matrix is mediated by the

transmembrane proteins, the integrins. Following interaction

with extracellular ligands h1-integrin associates with severalcytoskeletal proteins including a-actinin, talin, vinculin, and

tensin to form multimolecular ‘focal adhesions’ complex.

Abnormal circulation and unregulated proliferation of CML

progenitors is related, at least in part, to BCR-ABL protein-

induced abnormalities in h1 integrin-mediated adhesion and

signaling [24]. CML progenitors were shown to adhere

significantly less to a4h1 and a5h1 binding regions of

fibronectin, indicating that h1 integrin receptor function is

abnormal in CML. The BCR-ABL protein exhibits en-

hanced binding to F-actin, which may alter the integrin–

cytoskeletal interactions [25]. The decreased level of BCR-

ABL kinase in K562 cells due to IFN-a treatment leads

presumably to partial restoration of h1 integrin–cytoskeletalinteractions, which contribute to their increased adhesion to

fibronectin surfaces.

Employing two-dimensional electrophoresis, we revealed

the single distinct protein spot in K562 cell lysate, which

was markedly enhanced on IFN-a treatment. Using mass

spectrometry (MALDI-MS), this protein was identified as

the ubiquitine cross-reacting protein (UCRP/ISG15). Induc-

D. Grebenova et al. / Blood Cells, Molecules, and Diseases 32 (2004) 262–269 269

tion of this 17 kDa ubiquitine-like protein and its subsequent

conjugation to the cellular targets is the earliest response of

cells to the type I interferons [26]. Proteins conjugated to

UCRP are either modulated or targeted for processing

through the proteasome [27]. It was reported that conjugates

of UCRP distribute in a cytoskeletal pattern in both unin-

duced and interferon-treated A549 lung carcinoma cells,

while significant increase in the sequestration of UCRP

conjugates on intermediate filaments accompanied interfer-

on induction [28]. It can be speculated that increased

expression of UCRP in K562 cells following IFN-a treat-

ment may be related to IFN-a antiproliferative activity

mediated by a specific degradation of intermediate filaments

protein compounds.

Acknowledgments

The authors wish to thank Mrs. J. Sedlmaierova for the

expert technical assistance. The work was supported by

grants of the Grant Agency of the Czech Republic No 303/

01/1445 and Internal Grant Agency of the Ministry of

Health, Czech Republic No NL-7681-3.

References

[1] D. Grander, O. Sangfelt, S. Erickson, How does interferon exert its

cell growth inhibitory effect? Eur. J. Haematol. 59 (1997) 129–135.

[2] C.E. Samuel, Antiviral actions of interferons, Clin. Microbiol. Rev. 14

(2001) 778–809.

[3] O. Sangfelt, S. Erickson, D. Grander, Mechanisms of interferon-in-

duced cell cycle arrest, Front. Biosci. 5 (2000) 479–487.

[4] L. Thyrell, S. Erickson, B. Zhivotovsky, K. Pokrovskaja, O. Sangfelt,

J. Castro, S. Einhorn, D. Grander, Mechanisms of interferon-alpha

induced apoptosis inmalignant cells, Oncogene 21 (2002) 1251–1262.

[5] O. Sangfelt, H. Strander, Apoptosis and cell growth inhibition as anti-

tumor effector functions of interferons, Med. Oncol. 18 (2001) 3–14.

[6] S. Erickson, O. Sangfelt, J. Castro, M. Heyman, S. Einhorn, D.

Grander, Interferon-a inhibits proliferation of human T lymphocytes

by abrogation of interleukin 2-induced changes in cell cycle-regula-

tory proteins, Cell Growth Differ. 10 (1999) 575–582.

[7] B.N. Jahagirdar, J.S. Miller, A. Shet, C.M. Vertaillie, Novel thera-

pies for chronic myelogenous leukemia, Exp. Hematol. 29 (2001)

543–556.

[8] M. Talpaz, Interferon-alpha-based treatment of chronic myeloid leu-

kemia and implications of signal transduction inhibition, Semin.

Hematol. 38 (Suppl. 18) (2001) 22–27.

[9] L.C. Platanias, E.N. Fish, Signaling pathways activated by interfer-

ons, Exp. Hematol. 27 (1999) 1583–1592.

[10] O. Sangfelt, S. Erickson, J. Castro, T. Heiden, A. Gustafsson, S.

Einhorn, D. Grander, Molecular mechanisms underlying interferon-

alpha-induced G0/G1 arrest: CKI-mediated regulation of G1 Cdk-

complexes and activation of pocket proteins, Oncogene 18 (1999)

2798–2810.

[11] C.B. Lozzio, B.B. Lozzio, Human chronic myelogenous leukemia

cell line with positive Philadelphia chromosome, Blood 45 (1975)

321–334.

[12] H.G. Drexler, The leukemia-lymphoma cell line, FactsBook, Aca-

demic Press, London, 2001, pp. 632–633.

[13] A. McGahon, R. Bissonnette, M. Schmitt, K.M. Cotter, D.R. Green,

T.G. Cotter, BCR-ABL maintains resistance of chronic myelogenous

leukemia cells to apoptotic cell death, Blood 83 (1994) 1179–1187.

[14] J.C. Law, M.K. Ritke, J.C. Yalowich, G.H. Leder, R.E. Ferrell, Muta-

tional inactivation of the p53 gene in the human erythroid leukemia

K562 cell line, Leuk. Res. 17 (1993) 1045–1050.

[15] K. Kuzelova, D. Grebenova, M. Pluskalova, I. Marinov, Z. Hrkal,

Early apoptotic features of K562 cell death induced by 5-aminolae-

vulinic acid-based photodynamic therapy, J. Photochem. Photobiol.,

B Biol. (in press).

[16] D. Grebenova, H. Cajthamlova, J. Bartosova, J. Marinov, H. Klamova,

O. Fuchs, Z. Hrkal, Selective destruction of leukaemic cells by

photo-activation of 5-aminolaevulinic acid-induced protoporphyrin

IX, J. Photochem. Photobiol., B Biol. 47 (1998) 74–81.

[17] J. Soucek, P. Pouckova, J. Matousek, P. Stockbauer, J. Dostal, M.

Zadinova, Antitumor action of bovine seminal ribonuclease, Neoplas-

ma 43 (1996) 335–340.

[18] Z. Zhang, Z. Ao, A. Seth, S.F. Schlossman, A mitochondrial mem-

brane protein defined by a novel monoclonal antibody is preferen-

tially detected in apoptotic cells, J. Immunol. 157 (1996) 3980–3987.

[19] U.K. Laemmli, Cleavage of structural proteins during the assembly of

the head of bacteriophage T4, Nature 227 (1970) 680–685.

[20] B. Bjellqvist, Ch. Pasquali, F. Ravier, J.Ch. Sanchez, D.F. Hoch-

strasser, A nonlinear wide-range immobilized pH gradient for two-

dimensional electrophoresis and its definition in a relevant pH scale,

Electrophoresis 14 (1993) 1357–1365.

[21] A. Schevchenko, M. Wilm, O. Vorm, M. Mann, Mass spectrometric

sequencing of proteins from silver-stained polyacrylamide gels, Anal.

Chem. 68 (1996) 850–858.

[22] P. Halada, P. Man, D. Grebenova, Z. Hrkal, V. Havlıcek, Identification

of HL60 proteins affected by 5-aminolevulinic acid-based photody-

namic therapy using mass spectrometric approach, Collect. Czecho-

slov. Chem. Commun. 66 (2001) 1720–1728.

[23] G. Saydam, H.H. Aydin, F. Sahin, N. Selvi, G. Oktem, E. Terzioglu,

F. Buyukkececi, S.B. Omay, Involvement of protein phosphatase 2A

in interferon-a-2b-induced apoptosis in K562 human chronic myelog-

enous leukaemia cells, Leuk. Res. 27 (2003) 709–717.

[24] R. Bhatia, C.M. Verfaillie, Inhibition of BCR-ABL expression with

antisense oligonucleotides restores h1 integrin-mediated adhesion and

proliferation inhibition in chronic myelogenous leukemia hemato-

poietic progenitors, Blood 91 (1998) 3414–3422.

[25] R. Bhatia, H.A. Munthe, C.M. Verfaillie, Role of abnormal integrin-

cytoskeletal interactions in impaired h1 integrin function in chronic

myelogenous leukemia hematopoietic progenitors, Exp. Hematol. 27

(1999) 1384–1396.

[26] N. Reich, B. Evans, D. Levy, D. Fahey, R. Knight Jr., J.E. Darnell

Jr., Interferon-induced transcription of a gene encoding a 15 kDa

protein depends on an upstream enhancer element, Proc. Natl. Acad.

Sci. U. S. A. 84 (1987) 6394–6398.

[27] J. Narasimhan, J.L. Potter, A.L. Haas, Conjugation of the 15-kDa

interferon-induced ubiquitin homolog is distinct from that of ubiqui-

tin, J. Biol. Chem. 271 (1996) 324–330.

[28] K.R. Loeb, A.L. Haas, Conjugates of ubiquitin cross-reactive pro-

tein distribute in a cytoskeletal pattern, Mol. Cell. Biol. 14 (1994)

8408–8419.