Vol.

176,

April

No.

BIOCHEMICAL

1, 1991

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

15, 1991

Pages

473-478

A 41 kDa TRANSFERRIN RELATED MOLECULE ACTS AS AN AUTOCRINE GROWTH FACTOR FOR HL-60 CELLS1 Klaus Molecular nich

Oncology

Medical

H. Dittmann

Laboratory,

School

Received

January

Department

Grophadern, Munich

29,

and Petro

and

70,

E. Petrides

III,

of Medicine

Institute

for

Clinical

Marchioninistr.l5/25,

University

of Mu-

Hematology,

GSF, 8

Germany

1991

HL-60 cells produce an autostimulatory growth factor. Since the stimulatory effect of HL-60 conditioned medium is only observed in the absence of exogenous transferrin we have assayed HL-60 cells for the production of transferrin and found that they produce polypeptides which react with trans35S-methionine labelling, immunoprecipitation and subseferrin antibodies. quent separation by SDS-gel electrophoresis reveals the presence of a major transferrin related 41 f 2 kDa species released by HL-60 cells. Physiological levels of iron salts completely abolish the requirement of exogenous transferrin which indicates that the endogenous transferrin related polypeptides in the presence of exogenous inorganic iron salts are sufficient for the proinsulin or related growth factors are liferation of HL-60 cells provided present. The addition of transferrin receptor antibodies inhibits the stimulatory action of the endogenous transferrin related activity. Q 1991 Academic Press, 1°C.

For

survival

supplemented

with

line,

serum

can

these

cells

transferrin growth

are

and growth serum.

most

be replaced grown

eukaryotic

In HL-60 with

in suspension

and supraphysiological to occur

'This work meinschaft

[1,2].

was supported (Pe 258-10/l)

2To whom correspondence

Through

cells

in culture

require

a human promyelocytic

chemically

defined

medium

culture.

In such

a medium

concentrations the

in part and from should

cells,

action

of

by grants from GSF (Fe 71967).

be addressed

insulin

of transferrin

cell

(e.g.IMDM) the

when

presence

is mandatory

the

the

a medium

leukemia

Deutsche

cells

are

of for

provi-

Forschungsge-

49.89.70958137).

(Fax:

Abbreviations: IGF, insulin like growth factor; MTT, 3-(4,5-dimethylthiazolATCC, American Type Culture Collec2-yl) 2,5-diphenyl-tetrazoliumbromide; IMDM, tion; PBS, Phosphate buffered Saline; PPO, 2,5-diphenyloxazole; ELISA, Enzyme Linked Immunoadsorbent AsIscove’s modified Dulbecco Medium; EGF, epidermal growth factor. say; $1.50

0006-291X/91 Copyright

473

All

right5

oj

0

1991

reprodrtcrim

by Academic in arzy form

Press,

Inc.

reserved.

Vol.

176,

No.

ded with cleotide

iron

which

reductase

prerequisite

the

AND

important

action

is

the

physiological

extrinsic

are

leukemic

completely

cells

stimulation

which

possess

stimulatory

antibodies

mediated

BIOPHYSICAL

the

of the

growth

of

of required,

IGF-I

receptor

upon

extrinsic

own specific

as ribonuwhich

transferrin

are

their

such

receptors

growth

[5].

COMMUNICATIONS

enzymes

insulin

stimulus

dependent produce

of

transferrin

effect

inhibit

through

RESEARCH

constituent

cells

concentrations

its

crine

an

HL-60 growth

receptor

supraphysiological

populations

is [3].

for

transferrin

also

BIOCHEMICAL

1, 1991

are

because

HL-60

cells it

anti-

[4].

Since

was argued

and

proven

Although

most

that

that

IGF-I

leukemic

growth

factors

there

growth

factors

for

a

is

cells are auto-

[6]. MATERIALS

AN0 METHODS

HL-60 cell culture. The HL-60 cell line was obtained from the ATCC (Rockville, Vg.). The cells were cultivated in IMDM supplemented with 5% fetal calf serum (both from Gibco) or under serum free conditions [1,7] in IMDM supplemented with 0.5 pg/ml bovine insulin (Sigma) and 5 p,g/ml human transferrin (Sigma). The iron content of this non-iron supplemented medium is 10.14 kg/L (as determined by atomic absorption spectroscopy). The cell line was incubated at 37°C and 6% CO, and split twice a week. Every three months the cultures were evaluated for mycoplasma contamination. Gel chromatography of conditioned medium. Medium of HL-60 cells conditioned in the absence of exogenous transferrin and insulin was concentrated 50x by speedvac evaporation and dialyzed against PBS containing calcium and magnesium ions (0.1 g/l) for 8 h and then centrifuged at 6500 x g. 2 ml of the concentrate were loaded on a Sephadex 675 column (Pharmacia). The column was then eluted with PBS at a flow rate of 14 ml/h. 3 ml fractions were collected, filter sterilized and 100 JL~ aliquots assayed for biological activity and immunoreactivity. Quantitative growth assay. 0.125 x 10' cells/100 Pl were incubated in IMDM in the presence of insulin (0.5 pg/ml) and 100 pl aliquots of the 675 column fractions eluate in 0.32 cm* wells. After 5 days the cell number was quantified by the MTT-assay [8]. In other experiments where iron salts [Fe(NO ) ] interfered with this assay cells were counted in a hemocytometer under "tie microscope. Solid phase was obtained

assay for transferrin. Rabbit polyclonal transferrin from Sigma. The ELISA was performed as described in

antiserum [9].

Labelling of HL-60 polypeptides with 35S-methionine and imrrnoprecipitation with a polyclonal transferrin antiserum. 1 x lo6 cells per 1 ml batch were incubated with 50 p,Ci %S-methionine (Amersham, specific activity 1410 Curies/mmol) for 3 days. The conditioned medium was freed of cells through centrifugation at 600 x g and transferred into a Centricon (Amersham) microconcentrator (cutoff 10 kDa). The immune precipitations were exactly carried out as described [lo]. For comparative purposes human transferrin (Sigma) was electrophoresed on a parallel lane. After electrophoresis proteins were fixed in ethanol/acetate, the gels were soaked in a PPO-solution (Sigma) and dried. Autoradiography was carried out for 7 days. Incubation of HL-60 cells in the presence of transferrinreceptor-blocking antibody. Partially purified transferrin related polypeptides (675 fractions 10 and 11) were incubated with 1~10~ HL-60 cells in 100 l.~l IMDM containing insulin in the presence of 50 Pl of the monoclonal transferrin receptor anti474

Vol.

176,

No.

BIOCHEMICAL

1, 1991

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

body 42/6 (100 pg/ml)[ll] commercially available from Oianova (Hambur ). As a control a monoclonal human EGF antibody produced in our laboratory 9 121 was used. After 3 days growth stimulation was quantified by the MTT-assay [8]. RESULTS When medium transferrin

was tested

observed

when

(data

not

of growth

the

fractions an activity

indeed

produced of of

35S-methionine

medium

whereas

cell

lecular

species

ciated the

with

the

medium.

implicate rin.

However,

in

concentrations

ding

provided growth

of the in

(Fig.4).

its

the blocks

growth Monoclonal

caused

bands

in

only

the

when

higher

of iron

is

was present

in the

(Fractions of

the

proliferative

the

stimulating

by transferrin

related

effect was

however, activity. 475

was present insulin

major

was

role

of

to

the

added

concentrations

of

exogenous to bind

medium

exceetransferrin

and uptake

(Fig.3).

11 of

receptor of

transferrin

reduced

to

had no effect

the

the

10 pg/ml

transferrin

by day 5. When partially 10 and

would

transfer-

containing

iron

into

molecule

to be the

culture

asso-

one released

50 r.Lg/ml exogenous

transferrin

effect

EGF antibody,

supposed

sufficient

as determined

presence

IMDM

requirement

as 5 or

on SDS41 kDa mo-

transferrin

Fe(NO,),

were

pre-

transferrin

to this

related

in

the for

the

of exogenous

additional the

in

separated

the

absence

culture.

in

radioactivity to

is

41 + 2 kDa species

in addition

exogenous

and

same effect

activity

then

pre-

assayed

released

of a transferrin

sources

cells

incubation

as compared

abolished

HL-60

after

was 21:l

suspension

iron

incubated

one major

of

medium

were

transferrin-bound

grow

the

still

and

corre-

To ascertain

conditioned

as cell

When the

Assay

cells

of

pg/ml

endogenous

had the

specifically cells

the

related

incubated

in

completely enough

nitrate

several

and uptake

system

II

shown).

HL-60

revealed

as iron-transferrin)

cells

that

transferrin

lation

binding

1 pg/ml

iron

0.05

10 and not

the

antiserum

ratio

was observed of

HL-60

indicating

the

growth

was only

revealed

up while

as well

cells

should

and

assay

fractions.

in

taken

production

cells

(usually of

ferric

cells

HL-60

Since

transferrin medium

leukemic

The endogenous that

(Fig.2).

contained The

promoting

passages)

transferrin

(Fig.1).

insulin

effect

in fractions

polypeptide:

by HL-60

extracts

not

associated

with

conditioned

of

indicator

immunoreactivity

growth

this

cell

the

of 50 kDa (data

identified

and

of

immunoprecipitated

gels:

these

earlier

absence

stimulatory

from

was observed

(in

production

a growth

weight

cells

the

had to be present in the medium. and separated by gel chromatography

transferrin

by the

in

omitted

transferrin

in

transferrin

endogenous

sence

for

immunoreactive

presence

was

activity molecular

of the

cells

was concentrated

individual

that

HL-60 however,

promoting

such

cells

HL-60

was also

to an apparent

sence

the

upon

Insulin,

medium

sponding

by

transferrin

shown).

conditioned peak

conditioned

675

antibody [ll])

on

purified column) 42/6 with

12% over

the

on the

growth

was (which HL-60

control stimu-

Vol.

176,

No.

BIOCHEMICAL

1, 1991

CM

Cells

AND

BIOPHYSICALRESEARCH

COMMUNICATIONS

HT kDa

02

01

04 0

Fig.l. proteins (cells).

SDS-gel electrophoresis and secreted by HL-60 cells 5 pg 80 kDa human transferrin

Fiq.2. trations.

Growth

of

HL-60

cells

in

1

2

3daya

4

autoradiography of 35S-methionine (CM) and associated with (HT) as a control.

the

presence

of

various

/ 6

5

HL-60

transferrin

labelled cells

concen-

DISCUSSIDN HL-60 iron

cells

grow

can be taken

thesized

and

in

the

up by the

secreted

absence cells

by the

of exogenous via

transferrin

endogenous

leukemic

cells.

iron We have

provided

binding

enough

molecules

identified

syn-

a major

41

0.1 2.5-

2.0~6 U-F z

03

. 0

O.ug/ml O.Olgg/ml

. . l

O.lyg/ml lug/ml ioJJg/ml

1.5..

O’i: 0

1

Fiq.3. ter the

2

Growth addition

0.08 +

3 days

of of

4

5

i”‘“iMu!I1

6

0 4

HL-60 cells in the absence of various amounts of Fe(NO,),.

control

EGFMOAB

exogenous

TRMOAB

transferrin

FlO/li

FlO/H TRMOAB

and

af-

Fig.4. Growth of HL-60 cells in the presence of monoclonal antibodies directed against EGF (EGF-MOAB) or transferrin receptor (TR-MOAB), in the presence of fractions 10 and 11 of the 675 column chromatography and EGF-MOAB or TR-MOAB (n= 3 f. standard deviations).

476

Vol. 176,

No.

1, 1991

BIOCHEMICAL

kDa transferrin

related

also

with

associated

molecular hour

weight

halves thereby

gaining

derived

from

such

ciated

incubation

cells

may

cells:

such

molecule

since

fragment

[13].

could is

because

account

its

which

tumor

antibodies

for

by [4])

are,

human plasma

iron

levels

that

the

growth

take

into

the

ve action

of

contrast

and

transferrin

(which

is the

iron

This chelators

to

the is

pounds

is the

removed

action

of exogenous

from

endogenous

indicates

that

cells

[16]. [2]

as proliferation

the

serum

41

kDa actions

may be interconnected

free

culture

477

action

than of the

to

the

if

only in other

the

either Under

the

indicates

systems

iron

up-

are

in par-

for

insulin

of

these

appropriate can

presence

related

one

antiproliferati-

molecule in

41

physiological

requirement

medium.

transferrin

as seen

since

concentration the

Our observations

related but

produced

an increased

showing that

cells

transfer-

80 kDa transferrin through

ceases

transferrin of

for

by studies

here

HL-60

proliferation

that

Collins

therefore

polypeptide

in a higher

kDa and

reported

The observation

mediated

a 40

and

similar

exogenous

41 kD

antibodies.

concentrations

is

80 kDa transferrin the

related

and

proteins

react

of cellular

(used

the

produces

with

of

absolute

to

that

yet

polypeptides

presence

of

a molecule of fetal

in

iron

in HL-60

asso-

expressed

2 pg/ml).

statement

(data

are

transferrin

Prerequisite

supported

case

polypeptide

about

is

the

not

promotor

effect

80

related

effect.

stimulating

exogenous

antisera

is

we

species

binding

with

conin

produce

papain iron

be

moreover,

by transferrin it

inorganic

not

in membranes

or of

similarity

antibody

mitogenic

cell.

trypsin

produce

is

as as a precursor

family

could However,

from

cells

occurs

The 41 kDa transferrin

however,

supply

molecules

lines

leukemic

transferrin

blocking

abrogate

tial

the

as an autocrine

its

kDa molecule

with

to

however,

receptor

inhibits

p97

the

which

the

duplication

cells;

transferrin

serve

recognized

151.

acts

could

structural

is

cell

[e.g.

cells

a transferrin

This

of

as origin

Several

by HL-60

in

of

belongs

80 kDa lactotransferrin;

[14]. rin

p97

that

carboxy-

subsequently

be produced

cells

80 kDa

proposed

proteases. HL-60

a 72

the and

been

and

low

after of

intragenic

cells

also

p97 which

a molecule

has by

is

and

The 41 kDa species

molecular

suggest

it

with

HL-60

high

related

digestion

molecule

with that

amino-

the

extracellular

should

41 kDa species

Since

by HL-60 or

the

melanoma

Another

of

high

be recognized half

site.

associated

transferrin

crossreacts

binding

molecule

minor

41 kDa is

ancestor

any 80 kDa species

to the

can

of

COMMUNICATIONS

This

several

homology

produced

the

addition

internal iron

The observation

with

similar

agent.

a smaller

intra-

upon

shown).

an

a 80 kDa species

kDa transferrin

responsible

mass

the

detect

RESEARCH

substances

from

influence

BIOPHYSICAL

by hepatocytes.

show evolved

a situation

not

in

molecular

an additional

under not

but

produced

transferrin

propably

as the

cells

The

normally of

could

the

period.

protein

verted

molecule

immunoprecipitable

labelling

transferrin

AND

and

iron

fulfil of

insulin

where

comthe

insulin. related

the

cycling

Vol.

of

176,

No.

transferrin

is regulated

1, 1991

receptors by growth

BIOCHEMICAL

between factors

AND

the such

BIOPHYSICAL

plasma

RESEARCH

membrane

COMMUNICATIONS

and endosomal

membranes

as IGF I [17].

ACKNOWLEDGMENTS We thank Ursula Wallner and Daniele Knaus-Dittmann stance and Andreas Brachmann for the preparation

for expert technical of the figures.

assi-

REFERENCES 1.

::

4. 5. 6. 7. 8. 9. 10. 11. 12. 13. ::: :;:

Breitman, T.R., Collins, S.J. and Keene, B.R. (1980) Exp. cell. Res. 126, 494 - 497 Collins, S.J. (1987) Blood, 70 1233 - 1244 Cory, JiG. (1988) Adv.in Enzyme Regulation 27, 437 - 455 Taetle, R., Rhyner, K., Castagnola, J., To D. and Mendelsohn, J. (1988) J. Clin. Invest. 75, 1061 - 1067 Kellerer, M., Obermaier-Kusser, B., Ermel, B Wallner, U., Haring, H.U., and Petrides, P.E. (1990) J.Biol.Chem. 265 ii40 - 9345 Young, D.C., Wagner, K. and Griffin, J.D. i 1987) J.Clin.Invest. 79, 100 106 Taketasu, F., Kubota, K., Kajigaya, S., Sh ionoya, S., Motoyoshi, K., Sai to, M., Takaku, F. and Miura, Y. (1984) Cancer Res. 44, 531 - 535 Mossmann, T. (1983) J.immunol.Methods 65, 55 - 63 Harlow, E. and Lane, D. (1988) Immunoassay. In: E. Harlow and D. Lane (eds.), Antibodies. A lab manual, pp. 563 - 612. Cold Spring Harb. Lab, NY Firestone, G.L. and Winguth, S.D. (1990) In: S.P. Colowick and N.O. Ka plan (eds.), Methods in Enzymology, Immunoprecipitation of proteins, Vol. 182 pp. 688 - 700. Orlando FL: Academic Press, Inc. Trowbridge, I.S. (1988) In: H. Waldmann (ed.), Monoclonal antibody thera py, Vol. 45 pp. 121 - 146 Malordy, B., George-Nasciemento, C. and Petrides, P.E. (1987) J.Cancer Res. Clin.Oncol. 113, 6 Brown, J.P., Nishiyama, K., Hellstrom, I. and Hellstrom, K.E. (1981) J.Immun. 127, 539 - 544 Rado, T.A., Wei, X. and Benz, E.J. (1987) Blood 70, 989 - 993 Nakanishi, Y., Cuttitta, F., Kasprzyk, P.G., Avis, I., Steinberg, S.M., Gazdar, A.F. and Mulshine, J.L. (1988) Exp. Cell.Biol. 56, 74 - 85 Becton, D.L.and Roberts, B. (1989) Cancer Res. 49, 4809 - 4812 Davis, R.J., Faucher, M., Racaniello, L.K., Carruthers, A. and Czech, M.P. (1987) J.Biol.Chem. 262. 13126 - 13134

478