http://www.elsevier.com/locate/bba

Biochimica et Biophysica A

The distinct erythropoietin functions that promote cell survival and

proliferation are affected by aluminum exposure through mechanisms

involving erythropoietin receptor

Daniela Vittori*, Nicolas Pregi, Gladys Perez, Graciela Garbossa, Alcira Nesse

Laboratorio de Analisis Biologicos, Departamento de Quımica Biologica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires,

Pabellon II, Piso 4, Ciudad Universitaria, Ciudad de Buenos Aires (C1428EHA), Argentina

Received 5 March 2004; received in revised form 23 July 2004; accepted 6 August 2004

Available online 21 August 2004

Abstract

Erythropoietin (Epo) promotes the development of erythroid progenitors by triggering intracellular signals through the binding to its

specific receptor (EpoR). Previous results related to the action of aluminum (Al) on erythropoiesis let us suggest that the metal affects Epo

interaction with its target cells. In order to investigate this effect on cell activation by the Epo–EpoR complex, two human cell lines with

different dependence on Epo were subjected to Al exposure. In the Epo-independent K562 cells, Al inhibited Epo antiapoptotic action and

triggered a simultaneous decrease in protein and mRNA EpoR levels. On the other hand, proliferation of the strongly Epo-dependent UT-7

cells was enhanced by long-term Al treatment, in agreement with the upregulation of EpoR expression during Epo starvation. Results provide

some clues to the way by which Epo supports cell survival and growth, and demonstrate that not all the intracellular factors needed to

guarantee the different signaling pathways of Epo-cell activation are available or activated in cells expressing EpoR. This study then suggests

that at least one of the mechanisms by which Al interfere with erythropoiesis might involve EpoR modulation.

D 2004 Elsevier B.V. All rights reserved.

Keywords: Erythropoietin; Aluminum; Erythropoietin receptor; K562 cell; UT-7 cell line; Apoptosis

1. Introduction

Erythropoietin (Epo) is the main factor that promotes

viability, proliferation, and differentiation of mammalian

erythroid progenitor cells, functions that are transduced by

the specific cell surface Epo receptor (EpoR) [1–3].

In the course of our investigation focused on the

mechanisms by which aluminum (Al) exposure could

impair erythropoiesis, we found some clues to the possible

interference of the metal with Epo activity. The main

physiological target cells for Epo, late erythroid progeni-

tors CFU-E (colony-forming units-erythroid), showed a

remarkable inhibition of in vitro response to the growth

factor, after being exposed to Al [4,5]. The steep reduction

0167-4889/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.bbamcr.2004.08.004

* Corresponding author. Tel./fax: +54 011 4576 3342.

E-mail address: dvittori@qb.fcen.uba.ar (D. Vittori).

of CFU-E development due to this treatment only occurred

under Epo stimulation, and Al affected cellular mecha-

nisms of progenitor cells in such a way that the action of

the metal could not be overcome by increasing Epo doses

[4]. Furthermore, a similar negative effect upon bone

marrow cells was reproduced in ex vivo assays after

experimental Al exposure [6–9].

Since EpoR was detected in many different cells and

tissues, evidence has been accumulated to show that Epo

activity is not restricted to the erythroid compartment [10–

13]. In addition, there is increasing knowledge supporting

the idea that Epo signals that promote proliferation could

be separated from the signals that promote protection

against apoptosis [14]. Nevertheless, despite the available

data on EpoR-activated signal transduction molecules,

little is known about the specific signals regulating this

process.

cta 1743 (2005) 29–36

D. Vittori et al. / Biochimica et Biophysica Acta 1743 (2005) 29–3630

To further explore the common and/or unique signals

triggered by Epo in human cells, we employed a model

involving two Al-overloaded cell lines, K562 and UT-7,

both expressing EpoR but showing different dependence on

the hormone. Erythroleukemic K562 cells express low

EpoR amounts [15] and are non-Epo-dependent. On the

other hand, UT-7 cells, which express a large number of Epo

binding sites, grow in response to Epo [16]. As it was stated

before, Al is likely to affect erythroid cell activation by Epo

[4,5,8]. Therefore, since K562 and UT-7 cells show different

Epo dependence, we assumed that the effect of the metal on

the response of these cell lines to the growth factor would be

a potentially useful model to investigate whether the

multiple functions attributed to Epo are mediated by distinct

intracellular signals. Moreover, this study may contribute to

further understand the mechanisms by which Al affects

erythroid cell response to Epo.

2. Materials and methods

2.1. Materials

All chemicals used were of analytical grade. RPMI-

1640 medium, bovine serum albumin (BSA), 2,7-diamino-

fluorene (DAF), Hoechst 33258 dye, sodium o-vanadate,

phenylmethylsulfonyl fluoride (PMSF), aprotinin, leupep-

tin, and pepstatin A were obtained from Sigma-Aldrich;

Iscove’s Modified Dulbecco’s Medium (IMDM) and

specific primers for EpoR and glyceraldehyde-3-phosphate

dehydrogenase (GAPDH) from Invitrogen Life Technolo-

gies; polyclonal anti-EpoR antibody (SC-697) from Santa

Cruz Biotechnology; monoclonal anti-phosphotyrosine

(anti-PY) antibody and Protein A-agarose from BD

Transduction Laboratories; Trizol Reagent from Gibco

BRL; nitrocellulose membranes (Hybond), chemiluminis-

cent system kit (ECL) and Ready To Go T-Primed First-

Strand Kit from Amersham Biosciences; agarose from

Promega; ethidium bromide from Mallinckrodt; sodium

dodecylsulfate (SDS), acrylamide, bis-acrylamide, alumi-

num chloride, Triton X-100, and Tween 20 from Merck;

fetal bovine serum (FBS) (Bioser) and penicillin–strepto-

mycin (PAA Laboratories) from GENSA and recombinant

human erythropoietin (Epo, Hemax) from Biosidus

(Argentina).

2.2. Cell lines and cultures

Human erythroleukemic K562 cells, purchased from

American Type Culture Collection (Manassas, VA), were

grown in HEPES-buffered RPMI-1640 medium (pH

7.0F0.3), supplemented with 10% heat-inactivated FBS

and 100 U/ml penicillin–100 Ag/ml streptomycin [17].

UT-7 cell line, initially established from bone marrow

cells obtained from a patient with acute megakarioblastic

leukemia [16], was kindly provided by Dr. Patrick Mayeux

(Cochin Hospital, Paris, France). Stock cultures were

maintained in IMDM supplemented with 10% FBS and 1

U/ml Epo.

Cell cultures were developed at 37 8C in an atmosphere

containing 5% CO2 and 100% humidity, and one half of

the medium was replaced every 3–4 days. Proliferation and

cell viability were evaluated by the Trypan blue exclusion

test, and cell differentiation was estimated by counting

hemoglobin-positive cells after DAF–hydrogen peroxide

reaction [18].

2.3. Aluminum cell loading

Al citrate was freshly prepared in 0.1 M Tris–HCl (pH

7.3) by mixing Al chloride and sodium citrate solutions

(1:1.5 molar ratio), and added to culture media at 100 AMfinal concentration. This Al amount proved to produce in

vitro toxic effects upon CFU-E cells similar to those

attributed to in vivo Al overload [5,8].

2.4. Fluorescent nuclear stain of apoptotic cells

Cell cultured on slide covers were stained as follows:

(a) addition of five drops of Carnoy solution (methanol/

acetic acid, 3:1) for 2 min and complete removal of

medium; (b) fixation with 1–2 ml Carnoy solution for 5

min (step repeated twice); (c) drying at 20 8C, after

solution withdrawal; (d) addition of 1 ml of 1 Ag/ml

Hoechst 33258 dye prepared in Mc Ilvaine buffer (0.04 M

citric acid, 0.12 M disodium phosphate, pH 5.5); (e)

washing thrice with distilled water; (f) mounting by using

Mc Ilvaine buffer. Fluorescent nucleus with apoptotic

characteristics were detected by microscopy under UV

light at 365 nm (Axioplan Fluorescent Microscope, Zeiss).

Differential cell counting was performed by analyzing at

least 400 cells.

2.5. Cell lysis

Cells were washed with ice-cold phosphate-buffered

saline (PBS) solution containing 1 mM sodium o-vanadate,

and lysed with hypotonic buffer (50 mM Tris, pH 8.0, 150

mM NaCl, 1% Triton X-100) containing protease inhib-

itors (1 mM PMSF, 4 AM leupeptin, 2 AM pepstatin A, 1

Ag/ml aprotinin) and 1 mM sodium o-vanadate, in a ratio

of 200 Al/107 cells. After 30 min of incubation on ice,

insoluble material was removed by centrifugation at

15,000�g for 15 min.

2.6. Immunoprecipitation

Cell extracts were incubated with 3 Ag/ml polyclonal

anti-EpoR antibody during 1 h at 4 8C. Then, Protein A-

agarose was added and, after overnight incubation at 4 8C in

a rotating shaker, immunoprecipitates were collected by

centrifugation at 15,000�g during 15 min and washed twice

Fig. 1. Proliferative response of K562 and UT-7 cells to Epo. Cells were

initially plated at 2�105 cells/ml in the presence of increasing Epo

concentrations. The viable cell number was determined after 3 days by the

Trypan blue exclusion test. Results are expressed as meanFS.E.M. of five

independent experiments. Pearson coefficient between UT-7 viable cell

number and Epo concentration in the range between 0 and 1.0 U/ml was

r = 0.83 (P b 0.001).

D. Vittori et al. / Biochimica et Biophysica Acta 1743 (2005) 29–36 31

with the lysis buffer, containing the mentioned protease

inhibitors and o-vanadate.

2.7. Western blotting

Immunoprecipitates were boiled for 3 min in the

Laemmli buffer [19], resolved by SDS–polyacrylamide

gel electrophoresis (T =8%) and then, electroblotted onto a

Hybond nitrocellulose membrane during 1.5 h (transfer

buffer: 25 mM Tris, 195 mM glycine, 0.05% SDS, pH 8.3,

and 20% (v/v) methanol). Residual binding sites on the

membrane were blocked with 5% ECL membrane blocking

agent in Tris-buffered saline (25 mM Tris, 137 mM NaCl,

3 mM KCl, pH 7.4) containing 0.1% Tween 20 (TBS–

Tween) for 1 h at room temperature. The blots were then

incubated with the appropriate concentration of mono-

clonal anti-PY antibody during 1 h at 4 8C, washed three

times for 10 min each with TBS–Tween, and probed with

a 1:1000 dilution of anti-mouse horseradish peroxidase-

conjugated antibody for 1 h at 20 8C. After washing, theblots were incubated with the enhanced chemiluminiscence

substrate (ECL kit) and the bands detected by using a

Fujifilm Intelligent Dark Box II (Fuji) equipment coupled

to a LAS-1000 digital camera. To visualize the bands, the

Image Reader LAS-1000 and LProcess V1.Z2 programs

were employed. The blots were then stripped with 62.5

mM Tris–HCl (pH 6.8), 2% SDS, 100 mM 2-mercaptoe-

thanol at 50 8C for 30 min, washed, blocked, and reprobed

with anti-EpoR antibody.

2.8. Reverse transcriptase–polymerase chain reaction

(RT–PCR)

Total RNA was isolated by means of Trizol Reagent and

its concentration estimated by measuring the optical density

at 260 nm [20]. cDNA was synthesized by reverse

transcription using Ready To Go T-Primed First-Strand

Kit, starting from a sample of total RNA (2.5 Ag). An

aliquot of cDNA was amplified by 25 PCR amplification

cycles for UT-7 cells and 30 cycles for K562 cells (94 8C for

20 s, primer annealing at 64 8C for 30 s, extension at 72 8Cfor 40 s) and a final incubation at 60 8C for 7 min. Specific

primers were employed for EpoR [21] and for the internal

standard GAPDH [22]. The PCR products were examined

by electrophoresis on 1.5% agarose gel containing ethidium

bromide. Gels were photographed and analyzed through the

ArrayGauge and ImageGauge software.

2.9. Statistics

Results are expressed as meanFS.E.M. When corre-

sponding, the non-parametric Mann–Whitney U-Test or the

Kruskal–Wallis One-Way Analysis of Variance Test was

employed. At least differences with P b0.05 were considered

the criterion of statistical significance. Correlation between

variables was described by the Pearson r coefficient.

3. Results

3.1. Response of cell lines to Epo

To study the degree of Epo dependence of K562 and UT-

7 cell lines, erythroid differentiation, cell growth, and

survival were analyzed.

Three-day cultures were used to determine the response

to Epo regarding cell viability and proliferation in dose–

effect assays, in the range between 0.1 and 10.0 U Epo/ml

(Fig. 1). UT-7 cultures showed a close relationship between

Epo dose (between 0 and 1.0 U Epo/ml) and viable cell

number. The Epo dependence of these cells was clearly

described by the Pearson coefficient (r =0.83, Pb0.001). No

further linear increase of cell growth was obtained with Epo

concentration of 10.0 U Epo/ml.

On the contrary, K562 cells grew independently of the

Epo concentration present in the culture medium (Fig. 1).

The effect of Epo on cell differentiation was measured by

the development of cells containing hemoglobin after three

days of treatment. In doses up to 10 U/ml, the hormone did

not play any role in erythroid differentiation of both cell

lines, since percentages of hemoglobinized cells showed no

significant differences with respect to spontaneous differ-

entiation (without Epo). Results of cell differentiation in

cultures stimulated with 10 U/ml Epo were: 10.3F1.8% vs.

8.5F0.8% (n=9) for K562 cells, and 4.3F1.7% vs.

5.5F1.2% (n=5) for UT-7 cells.

3.2. Effect of aluminum on Epo antiapoptotic activity

From the experimental evidence described above, it is

clear that K562 cells are Epo-independent to grow and

differentiate. Thus, we assumed that EpoR expression in

these cells would be related to the prevention of

programmed cell death. To analyze this hypothesis, we

D. Vittori et al. / Biochimica et Biophysica Acta 1743 (2005) 29–3632

examined the ability of Epo to inhibit hemin-induced

apoptosis. Apoptotic cells were detected by fluorescence

microscopy after Hoechst staining in cultures developed in

the presence of hemin and Epo (Fig. 2). Whereas the mean

value of spontaneous apoptosis was low (2.4F0.3%), it

increased almost 10 times under hemin induction

(24.6F3.6%), and the latter effect decreased 45% due to

Epo influence, being the percentage of apoptotic cells

13.9F2.4%.

As we already mentioned, one of the purposes of this

work was to analyze whether the model of cells exposed to

Al might be useful to study different mechanisms of Epo

activation. Therefore, we evaluated Al ability to modulate

Epo antiapoptotic action. When K562 cells induced to

apoptosis by hemin were cultured with the simultaneous

presence of a high Epo dose and Al, the protective effect of

the hormone was counteracted, being the amount of

Fig. 2. Effect of aluminum on the Epo antiapoptotic activity. Apoptosis was

measured in K562 cells cultured during 5 days in the presence of 50 AMhemin; 50 AM hemin +10 U/ml Epo or 50 AM hemin +10 U/ml Epo +100

AM Al citrate, as well as in control cells incubated without any treatment or

in the presence of 100 AM Al citrate (upper panel). In UT-7 cell cultures,

apoptosis was achieved by a 3-day Epo deprivation while the hormone

antiapoptotic effect was demonstrated in 1 U/ml Epo stimulated cultures,

regardless of whether 100 AM Al citrate was present or absent (lower

panel). Apoptotic cells were detected by fluorescence microscopy after

Hoechst staining. Each bar represents percentage value (meanFS.E.M.) of

apoptotic cells with respect to total cell number. *Significant differences

with respect to both (Hemin) and (Hemin + Epo + Al) (P b 0.05, n = 5).

**Significant increase with respect to cultures performed in the presence of

Epo (P b 0.01, n = 3).

apoptotic cells 25.5F3.3% (Fig. 2). It is worth mentioning

that Al per se did not affect cell death (3.0F1.0%).

Since UT-7 cells are Epo-dependent to survive, apoptosis

can be achieved by Epo deprivation. As expected, cells

deprived from the hormone for three days suffered a high

degree of apoptosis, which was almost prevented by 1 U/ml

Epo (76.8F3.8% vs. 12.2F1.6%, P b 0.01) (Fig. 2). On the

other hand, no Al effect was observed upon the Epo

protective action. The apoptosis originated by Epo starva-

tion was prevented by the hormone in such a way that

minimal changes introduced by Al could be concealed.

3.3. Aluminum effect on cell proliferation induced by Epo

Since Al seemed to alter Epo action and UT-7 cells

proved to be Epo-dependent (Fig. 1), this cell line was used

to investigate whether the metal might alter cell prolifer-

ation. Figure 3 shows the results of cell growth and viability

determined in three-day cultures carried out under different

conditions: (a) cells incubated without Epo (�Epo), and (b)

cells incubated for 3 days with 1 U/ml Epo, with or without

the addition of Al (Epo 1U+Al 3d and Epo 1U). Since this

acute Al exposure proved not to affect cell proliferation, this

3-day assay was repeated with cells previously treated with

Al for 30 days (Epo 1U+Al 30d). The interest to investigate

cell behavior after such long-term exposure was based on

the slow and silent effects on different tissues reported for

the non-essential metal. Unlike the short-term Al treatment,

chronic exposure to the metal induced a significant increase

in the proliferative UT-7 cell response to Epo (Fig. 3: Epo

1U+Al 30d vs. Epo 1U, P b0.001). This assay demonstrated

that in cells chronically exposed to Al the mean period

necessary to duplicate cell population was 26F0.6 h,

whereas in cells cultured without the metal it was 37F0.9

h, thus showing statistically significant differences between

them (P b 0.005). In order to determine whether this

unexpected response could be attributed to the behavior of

Al as a proliferative factor, an assay without Epo was

developed in the presence of the metal (Fig. 3: Al 3d).

Results showed that Al did not act as a proliferative inducer

(Al 3d vs. �Epo, NS) in UT-7 cells.

To further investigate possible changes in sensitivity to

Epo of cells after long-term Al treatment, cultures were

stimulated with lower Epo amounts (0.1 U/ml). Figure 3

shows that chronically Al-overloaded cells cultured in the

presence of 0.1 U/ml Epo reached the proliferation rate of

Al-untreated cells stimulated with 1 U/ml Epo (Epo

0.1U+Al 30d vs. Epo 1U, NS). In contrast, the lack of

effect due to a 3-day Al exposure (Epo 0.1U+Al 3d)

emphasized the chronic effect of Al upon these UT-7 cells.

3.4. Aluminum effect on EpoR expression

In order to determine whether the inhibition of the Epo

antiapoptotic effect upon K562 cells was related to the

interference of Al with EpoR activation signals, we

Fig. 3. UT-7 cell proliferation under the effect of Epo and aluminum. UT-7 cell cultures were performed under different conditions. Cells without any previous

treatment were incubated for 3 days without Epo (�Epo) and with 0.1 or 1 U/ml Epo (Epo 0.1U and Epo 1U). Similar assays were also made in the presence of

100 AM Al citrate (Al 3d, Epo 0.1U +Al 3d and Epo 1U +Al 3d). Cells previously exposed to 100 AM Al citrate during 30 days were stimulated with 0.1 U/ml

(Epo 0.1 U +Al 30d) or 1 U/ml Epo (Epo 1 U +Al 30d) for 3 days. *Significant differences with respect to Epo 1U ( P b 0.001, n=5). **Significant differences

with respect to Epo 0.1U ( P b 0.005, n = 5).

D. Vittori et al. / Biochimica et Biophysica Acta 1743 (2005) 29–36 33

examined receptor expression and phosphorylation. Cells

were cultured for 5 days in the presence of Al. Then,

treated and untreated cells were activated adding 10 U/ml

Epo for 10 min. Cell lysates were immunoprecipitated

with anti-EpoR antibody and Western blots were revealed

using anti-PY antibody to detect tyrosine-phosphorylated

proteins associated to the activated EpoR. This association

was further confirmed through immune reaction with anti-

EpoR antibody in the stripped blots. As can be seen in

Figure 4, several proteins are recognized by the anti-EpoR

antibody. They have been reported as multimeric com-

plexes of EpoR with different polipeptides as well as

Fig. 4. Effect of aluminum on EpoR expression. K562 cells (left) were cultured in

then stimulated by a 10-min pulse of 10 U/ml Epo. UT-7 cells (right) exposed for

were overnight deprived of Epo and, then, stimulated with 10 U/ml Epo for 10

antibody, and immunoprecipitates analyzed by Western blotting (WB) with anti-P

receptor, reprobing of stripped blots with anti-EpoR antibody (lower panel) was ass

patterns displayed are representative of five assays with K562 cells and seven as

derived from EpoR by glycosylation and phosphorylation

[23–25]. However, the 66 kDa EpoR has been considered

the most certain product immunoprecipitated by specific

antisera to the cloned EpoR [25]. Figure 4 (left),

representative of five different assays, shows lower

amounts of EpoR and weaker EpoR phosphorylation in

K562 cells previously treated with Al. On the other hand,

to further explore whether variations in UT-7 cell response

to Epo under conditions of chronic Al exposure were

related to EpoR modulation, the following experiments

were carried out. UT-7 cells exposed to Al for long

periods (30 days) and untreated cells were deprived of

the presence (Al+) or absence (Al�) of 100 AM Al citrate for 5 days, and

long periods (30 days) to 100 AM Al citrate (Al+) and untreated cells (Al�)

min at 37 8C. Cell lysates were immunoprecipitated (IP) with anti-EpoR

Y antibody (upper panel). To characterize phosphorylation associated to the

essed. The protein band corresponding to the 66 kDa EpoR is indicated. The

says with UT-7 cells.

Fig. 5. Effect of aluminum on EpoR mRNA levels. K562 cells were cultured in the presence (Al+) or absence (Al�) of 100 AMAl citrate for 5 days (left). UT-7

cells that had been chronically (30 days) exposed to 100 AM Al citrate and untreated cells were studied at t=0 after growth factor starvation for 18 h (lanes 2

and 5), and at t=24 after being starved for 18 h and reexposed to Epo for additional 24 h (lanes 3 and 6). Control cells were maintained in 1 U/ml Epo (lanes 1

and 4). Total RNA was extracted and EpoR mRNA amplified by RT–PCR. Signals were analysed in comparison with those of the GAPDH internal standard

and expressed as arbitrary units of band density (lower panels). Error bars indicate the standard error of three determinations.

D. Vittori et al. / Biochimica et Biophysica Acta 1743 (2005) 29–3634

Epo overnight and, then, stimulated with 10 U/ml Epo for

10 min. Cell lysates were immunoprecipitated with anti-

EpoR antibody, and EpoR-associated tyrosine-phosphory-

lated proteins were detected through immune reaction.

Figure 4 (right) shows a Western blotting pattern

representative of seven different assays. EpoR expression

and associated tyrosine phosphorylation were higher in

UT-7 cells chronically exposed to Al in response to Epo

after overnight growth factor starvation.

To further understand the mechanism that led to the

regulation of EpoR expression in Al exposed cells, we

examined EpoR mRNA levels by RT–PCR (Fig. 5). In

agreement with the reduced protein expression, a fainter

band corresponding to EpoR mRNA was observed in

K562 cells grown in Al-rich medium compared to that of

untreated cells (Fig. 5, left). On the other hand, the

response of UT-7 cells chronically exposed to Al (Al 30d)

was compared to that of Al-unexposed cells (Fig. 5, right).

Cells were deprived of Epo overnight, and then, sub-

sequently incubated for additional 24 h with 1 U/ml Epo.

Standard cultures with 1 U/ml Epo stimulation were used

as controls (Fig. 5, lanes 1 and 4). A transient increase of

Epo mRNA level, which represents the normal response of

UT-7 cells to growth factor starvation, can be observed

when comparing the first three lines and the corresponding

bars. A similar pattern was detected after incubation in the

presence of Al for three days (data not shown). However,

EpoR mRNA was overexpressed in Epo starved cells

chronically exposed to Al (Fig. 5, lane 5), whereas a

subsequent addition of Epo made EpoR mRNA return to

basal levels (Fig. 5, lane 6).

4. Discussion

It is well known that the interaction of Epo with its

targets through the binding to its specific receptor EpoR

leads to cell survival, proliferation and/or differentiation.

However, the control of the receptor expression, as well as

the pathways by which the Epo–EpoR complex mediates

the multiple effects triggered by the hormone, is not

completely understood.

In order to investigate possible Al interference with Epo

activity through mechanisms related to EpoR activation,

two human cell lines expressing EpoR — K562 and UT-

7 — were employed. The hypothesis to consider was that

different cell dependence on Epo might be related to

different signaling pathways.

Results of Epo-induced cell growth and differentiation

showed that not all the cells expressing EpoR are influenced

in the same way by the hormone. Epo did not act as a

differentiation stimulus to K562 or UT-7 cells, even though

the former are able to undergo erythroid differentiation in

the presence of other inducers [17,26]. Epo, however, might

have other effects upon these cell lines. Indeed, cell growth

in UT-7 cultures showed to be strongly dependent on Epo

(Fig. 1), whereas hemin-induced apoptosis of K562 cells

was prevented by the hormone (Fig. 2).

D. Vittori et al. / Biochimica et Biophysica Acta 1743 (2005) 29–36 35

Al exposure affected the response to Epo in both cell

lines. The presence of Al counteracted Epo’s ability to

protect K562 cells from apoptosis (Fig. 2), thus supporting

the hypothesis that the hormone would not be assigned to

fulfill erythroid lineage. Rather, it would work mainly as

an antiapoptotic factor maintaining cells alive to let them

follow their maturation program [3]. Al per se did not

affect K562 cell survival (Fig. 2). Therefore, its interfer-

ence in the antiapoptotic process would be associated to

Epo-mediated mechanisms of cell activation. In fact, the

negative action of Al on the Epo-protective effect in K562

cells was related to a parallel reduction in EpoR

expression and the concomitant EpoR phosphorylation

(Fig. 4). These effects agreed with the lower EpoR mRNA

level observed (Fig. 5), suggesting that the activation

pathway mediated by the Epo–EpoR complex was

depressed by Al exposure. Since Bcl-XL, an antiapoptotic

protein of the Bcl-2 family, is highly expressed in K562

cells [27], the apoptosis associated to Al-induced down-

modulation of EpoR expression described herein might be

explained by mechanisms involving the reduced expres-

sion of protective proteins.

The difference in behavior exhibited by both cell lines

due to Al exposure supports the hypothesis on the existence

of different signaling pathways involved in Epo functions. It

seemed that long Al treatment of UT-7 cells, unlike short Al

exposure, exerted a synergistic effect on Epo-induced

proliferation. This was attributed to an increased sensitivity

to the growth factor as shown by the response of Al-

pretreated cells to low Epo levels, as well as by the inability

of Al to behave as a proliferative factor per se (Fig. 3).

These data coincided with EpoR positive modulation under

similar cell conditions of Al exposure (Fig. 4 and 5). The

analysis of EpoR regulation at the mRNA level agreed with

previous reports [28,29] in which a transient increase of this

level as a consequence of Epo starvation was followed by

the return to basal levels after Epo addition (Fig. 5, lanes 1

to 3). However, the threefold transient increase in EpoR

mRNA level after growth factor starvation of chronic Al

exposed UT-7 cells (Fig. 5, lane 5) suggested higher Epo

dependence at these cell conditions. In fact, cells sensitive

enough to respond to low Epo levels will be immediately

aware of this growth factor deprivation.

The fact that Al did not prevent UT-7 cell activation by

Epo (Fig. 3) let us assume that the metal does not affect the

binding of the hormone to its receptor. However, cell

stimulation by Epo appeared to be necessary to make

evident the action of Al, since the metal did not act as a

proliferative factor in the absence of Epo (Fig. 3).

Considering that EpoR was related to Al effect, and that it

seemed unlikely that the metal interfere in Epo–EpoR

binding, its action might be exerted within the cell. It has

been reported that Epo induces upregulation of transferrin

receptors [30,31], the main transport pathway through

which Al enters the cell [26]. Therefore, it can be assumed

that Epo stimulation would improve cell Al uptake, thus

favoring its interference with Epo activity at signal trans-

duction and/or transcriptional levels.

It is well known that many important phosphate

containing biomolecules easily bind to Al ions [32] through

their phosphate groups, and so the interaction with the metal

changes the susceptibility of those groups to enzymes

specifically acting on them [33]. Since Epo–EpoR signaling

is controlled by both phosphorylation and dephosphoryla-

tion of many transduction molecules, alteration of these

processes may occur due to cell Al overload.

Alternatively, other mechanisms might account for Al-

induced EpoR modulation. It has been reported that the

regulation of EpoR mRNA levels in UT-7 cells undergoing

proliferation was associated with serial events controlled by

the cell cycle or related to it [29]. In this regard, Al was able

to influence osteoblast replication inducing the transition

from G0 to S phase in the cell cycle [34], while it triggered

astrocyte apoptosis which was closely associated with an

increase in the G2/M phase [35]. The apparent discrepancy

resulting from different requirements of the cell types

assayed may account for the differences found in our study.

In summary, the experimental models of Al-exposed

human cells herein described proved to be suitable to study

some of the multiple events developed in Epo-activated

cells, since the effects of the hormone observed in both cell

lines were quite distinct. To a great extent, Al blocked the

antiapoptotic effect exerted by Epo upon K562 cells

associated to the downregulation of EpoR expression. On

the other hand, chronic Al exposure of UT-7 cells induced a

sharp increase of EpoR mRNA levels during growth factor

starvation without introducing significant changes during

Epo stimulation periods. It seems that mechanisms involved

in Epo activity — rather than those related to Al action —

are of different nature in both cell types. Since Epo is an

essential factor in cell growth but not in differentiation of

UT-7 cells, its effects on EpoR expression might be

considered representative of Epo-induced regulatory pro-

cesses of cell proliferation. Results suggest that in K562

cells Epo activates pathways mainly related to the preven-

tion of programmed cell death, whereas in UT-7 cells, the

hormone affects pathways involved in the stimulation of cell

proliferation, thus explaining the different response to Epo

shown by the two cell lines.

In conclusion, this work demonstrates that Al affects cell

response to Epo by mechanisms involving EpoR modu-

lation. This would be of interest in studies of clinical

resistance to recombinant human erythropoietin therapy.

Furthermore, the intracellular alteration caused by Al

contributes to support the existence of different signal

transduction mechanisms by which Epo promotes either cell

proliferation or apoptosis prevention. Moreover, this study

provides some evidence that would explain how Epo

supports cell survival and growth, and demonstrates that

not all the intracellular factors needed to guarantee the

different signaling pathways of cell activation by Epo are

available in cells expressing EpoR.

D. Vittori et al. / Biochimica et Biophysica Acta 1743 (2005) 29–3636

Acknowledgements

The authors are grateful to Dr. Patrick Mayeux (Cochin

Hospital, Paris, France) for his generous gift of the UT-7

cell line, to Biosidus (Argentina) for supplying human

recombinant erythropoietin (Hemax), to Santa Cruz Bio-

technology for providing anti-erythropoietin receptor anti-

body, and to Laura Gutierrez for her advice on the English

translation. This research was supported by grants from the

University of Buenos Aires and the Consejo Nacional de

Investigaciones Cientıficas y Tecnicas (CONICET, Argen-

tina). Results included in this work were presented at the

Fifth Keele Meeting on Aluminium, Stoke-on-Trent, United

Kingdom, 2003.

References

[1] F. Verdier, P. Walrafen, N. Hubert, S. Chretien, S. Gisselbrecht, C.

Lacombe, P. Mayeux, Proteasomes regulate the duration of eryth-

ropoietin receptor activation by controlling down-regulation of cell

surface receptors, J. Biol. Chem. 275 (2000) 18375–18381.

[2] L. Mulcahy, The erythropoietin receptor, Semin. Oncol. 28 (2001)

19–23.

[3] M.J. Koury, S.T. Sawyer, S.J. Brandt, New insights into erythropoi-

esis, Curr. Opin. Hematol. 9 (2002) 93–100.

[4] G. Garbossa, A. Gutnisky, A. Nesse, The inhibitory action of

aluminum on mouse bone marrow cell growth: evidence for an

erythropoietin- and transferrin-mediated mechanism, Miner. Electro-

lyte Metab. 20 (1994) 141–146.

[5] D. Vittori, G. Garbossa, C. Lafourcade, G. Perez, A. Nesse, Human

erythroid cells are affected by aluminium. Alteration of membrane

band 3 protein, Biochim. Biophys. Acta 1558 (2002) 142–150.

[6] G. Garbossa, A. Gutnisky, A. Nesse, Depressed erythroid progenitor

cell activity in aluminum-overloaded mice, Miner. Electrolyte Metab.

22 (1996) 214–218.

[7] G. Garbossa, G. Galvez, M.E. Castro, A. Nesse, Oral aluminum

administration to rats with normal renal function. 1. Impairment of

erythropoiesis, Hum. Exp. Toxicol. 17 (1998) 312–317.

[8] D. Vittori, A. Nesse, G. Perez, G. Garbossa, Morphologic and

functional alterations of erythroid cells induced by long term ingestion

of aluminium, J. Inorg. Biochem. 76 (1999) 113–120.

[9] A. Nesse, G. Garbossa, Aluminium toxicity on erythropoiesis.

Mechanisms related to cellular dysfunction in Alzheimer’s Disease,

in: C. Exley (Ed.), Aluminium and Alzheimer’s Disease. The

Science that Describes the Link, Elsevier Science B.V., Amsterdam,

2001, pp. 261–277.

[10] S.E. Juul, D.K. Anderson, R.D. Christensen, Erythropoietin and

erythropoietin receptor in the developing human central nervous

system, Pediatr. Res. 43 (1998) 40–49.

[11] S. Sela, R. Shurtz-Swirski, R. Sharon, J. Manaster, J. Chezar, G.

Shkolnik, G. Shapiro, S.M. Shasha, S. Merchav, B. Kristal, The

polymorphonuclear leukocyte: a new target for erythropoietin,

Nephron 88 (2001) 205–210.

[12] A. Nagai, E. Nakagawa, H.B. Choi, K. Hatori, S. Kobayashi, S.U.

Kim, Erythropoietin and erythropoietin receptors in human CNS

neurons, astrocytes, microglia and oligodendrocytes grown in culture,

J. Neuropathol. Exp. Neurol. 60 (2001) 386–392.

[13] T.R. Lappin, A.P. Maxwell, P. Johnston, EPO’s alter ego: erythro-

poietin has multiple actions, Stem Cells 20 (2002) 485–492.

[14] A.E. Lawson, H. Bao, A. Wickrema, S.M. Jacobs-Helber, S.T.

Sawyer, Phosphatase inhibition promotes antiapoptotic but not

proliferative signaling pathways in erythropoietin-dependent HCD57

cells, Blood 96 (2000) 2084–2092.

[15] J.K. Fraser, F.-K. Lin, M.W. Berridge, Expression and modulation of

specific, high affinity binding sites for erythropoietin on the human

erythroleukemic cell line K562, Blood 71 (1988) 104–109.

[16] N. Komatsu, H. Nakauchi, A. Miwa, T. Ishihara, M. Eguchi, M.

Moroi, M. Okada, Y. Sato, H. Wada, Y. Yawata, T. Suda, Y. Miura,

Establishment and characterization of a human leukemic cell line with

megakaryocytic features: dependency on granulocyte–macrophage

colony-stimulating factor, interleukin-3, or erythropoietin for growth

and survival, Cancer Res. 51 (1991) 341–348.

[17] G. Perez, G. Garbossa, B. Sassetti, C. Di Risio, A. Nesse, Interference

of aluminium on iron metabolism in erythroleukaemia K562 cells, J.

Inorg. Biochem. 76 (1999) 105–112.

[18] R.E. Worthington, J. Bossie-Condreanu, G. Van Zant, Quantification

of erythroid differentiation in vitro using a sensitive colorimetric assay

for hemoglobin, Exp. Hematol. 15 (1987) 85–92.

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

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

[20] J. Sambrook, D. Russell, Molecular Cloning. A Laboratory Manual,

Cold Spring Harbor Laboratory Press, New York, 2001.

[21] R. Shimizu, N. Komatsu, Y. Miura, Dominant negative effect of a

truncated erythropoietin receptor (EPOR-T) on erythropoietin-induced

erythroid differentiation: possible involvement of EPOR-T in ineffec-

tive erythropoiesis of myelodysplastic syndrome, Exp. Hematol. 27

(1999) 229–233.

[22] M. McKinney, M. Robbins, Chronic atropine administration up-

regulates rat cortical muscarinic m1 receptor mRNA molecules:

assessment with the RT/PCR, Brain Res. Mol. Brain Res. 12 (1992)

39–45.

[23] H. Youssoufian, G. Longmore, D. Neumann, A. Yoshimura, H.

Lodish, Structure, function, and activation of the erythropoietin

receptor, Blood 81 (1993) 2223–2236.

[24] P. Mayeux, S. Pallu, S. Gobert, C. Lacombe, S. Gisselbrecht, Structure

of the erythropoietin receptor, Proc. Soc. Exp. Biol. Med. 206 (1994)

200–204.

[25] J.W. Fisher, Erythropoietin: physiology and pharmacology update,

Exp. Biol. Med. 228 (2003) 1–14.

[26] G. Perez, G. Garbossa, C. Di Risio, D. Vittori, A. Nesse, Disturbance

of cellular iron uptake and utilisation by aluminium, J. Inorg.

Biochem. 87 (2001) 21–27.

[27] A. Benito, M. Silva, D. Grillot, G. Nunez, J.L. Fernandez-Luna,

Apoptosis induced by erythroid differentiation of human leukemia cell

lines is inhibited by Bcl-xL, Blood 87 (1996) 3837–3843.

[28] R. Migliaccio, Y. Jiang, G. Migliaccio, S. Nicolis, S. Crotta, A.

Ronchi, S. Ottolenghi, J.W. Adamson, Transcriptional and posttran-

scriptional regulation of the expression of the erythropoietin receptor

gene in human erythropoietin-responsive cell lines, Blood 82 (1993)

3760–3769.

[29] N. Komatsu, K. Kirito, Y. Kashii, Y. Furukawa, J. Kikuchi, N.

Suwabe, M. Amamoto, Y. Miura, Cell-cycle-dependent regulation of

erythropoietin receptor gene, Blood 89 (1997) 1182–1188.

[30] S.B. Krantz, Erythropoietin, Blood 77 (1991) 1661–1665.

[31] G. Weiss, T. Houston, S. Kastner, K. Jfhrer, K. Grqnewald, J.H.Brock, Regulation of cellular iron metabolism by erythropoietin:

activation of iron-regulatory protein and upregulation of transferrin

receptor expression in erythroid cells, Blood 89 (1997) 680–687.

[32] R.B. Martin, The chemistry of aluminum as related to Biology and

Medicine, Clin. Chem. 32 (1986) 1797–1806.

[33] P. Nayak, A.K. Chatterjee, Differential response of certain brain

phosphoesterases to aluminium in dietary protein adequacy and

inadequacy, Food Chem. Toxicol. 39 (2001) 587–592.

[34] L.D. Quarles, R.J. Wenstrup, S.A. Castillo, M.K. Drezner,

Aluminum-induced mitogenesis in MC3T3-E1 osteoblasts: potenial

mechanism underlying neoosteogenesis, Endocrinology 128 (1991)

3144–3151.

[35] G.W. Guo, Y.X. Liang, Aluminum-induced apoptosis in cultured

astrocytes and its effect on calcium homeostasis, Brain Res. 888

(2001) 221–226.