Vol. 182, No. 2, 1992 January 31, 1992

BIOCHEMICAL

PLACENTAL HUMAN

Received

Makiya

of Medical

December

RESEARCH COMMUNICATIONS Pages 624-630

ALKALINE PHOSPHATASE IS RELATED TO IgG INTEFWALIZATION IN HEp2 CELLS Bicardo

Department

AND BIOPHYSICAL

10,

and Torgny

Stigbrand

Biochemistry and Biophysics, S-901 87 Ume& Sweden

* University

of Umea,

1991

Summary: The biological function(s) of placental alkaline phosphatase has not yet been unraveled. The low catalytic activity of the enzyme at physiological pH. and the lack of “natural substrates”, bring about the necessity of a more structure-related conception of its role. We have observed an interaction between placental alkaline phosphatase and human IgG. In this report we show that this isozyme is the major membrane protein able to bind IgG in a IgG-internalizing cell line (HEp2). Pretreatment of these cells with Fab fragments of anti placental alkaline phosphatase antibodies blocks the internalization of IgG without perturbing the endocytosis of other ligands. Our results indicate that placental alkaline phosphatase has the ability not only to bind human IgG. but also to promote its internalization in HEpP cells. 0 1992 Academic Press, Inc.

Many

reports

last half century markers

for

incremented 5). The

dealing

(for extensive

different

elucidation Except has been at which

“physiological Some function observation human

reports

for alkaline

during

0006-291X/92

TB$

properties

work

the last years besides

binding

IgG with a dissociation

The abbreviations isothiocyanate: azidosalicilamido) dithiothreitol.

isozymes

enzymes

of these

as

specific

diseases has been

(4, less

to bone mineralization

is still unknown.

at physiological

showed

and malignant

of these

the

of these enzymes

as they

seems to be related

during

The search

proteins.

pH is very

for a

However,

the

low,

the

and

has not yet been identified.

phosphatases

* To whom correspondence

Copyright All rights

enzymes

The usefulness

established,

function(s) which

on the catalytic

these

have been presented

with bone, hepatobiliary

of the remaining

focused

of a distinct

clearly

physiological

role(s)

substrate(s)”

see refs. l-3).

associated

of the

phosphatases

has been

for the bone isoform,

(6). the physiological efficiency

alkaline

reviews

diseases

levels in serum

successful. function

with

between

constant

should

suggested catalysis placental

a structurally (7). Such alkaline

of 3.68 @I (8). Consistent

deduced

a hypothesis phosphatase

alternative led us to the (PLAP) and

with such a constant,

IgG

be addressed.

sed are: PLAP. placental alkaline phosphatase; FITC. fluorescein Tris-buffered saline (pH 7.4): SASD, sulfosuccinimidyl 2-(pethyl- 1,3’-dithiopropionate; TCA, trichloracetic acid: D’IT,

$1.50

0 1992 by Academic Press, of reproduction in any form

Inc. reserved.

624

Vol.

182,

No.

is expected maternal

2,

1992

to interact

BIOCHEMICAL

with

blood surrounding In this

membranes, internalization

report

AND

PLAP in placenta

BIOPHYSICAL

studied

and the importance

COMMUNICATIONS

due to the high IgG concentration

the syncytiotrophoblast

we have

RESEARCH

the IgG interaction

of the PLAP-IgG

in the

(50 to 110 pM). complex

with

PLAP anchored

formation

in relation

to cell to the

of IgG in the HEp2 cell line.

MATERIALS

AND METHODS

Proteins, chemicals and reagents: Placental alkaline phosphatase was purified as earlier described (9). Human IgG was obtained by affinity chromatography on Protein A-Sepharose (Pharmacia, Sweden) of human serum. IgG was iodinated by the method of Greenwood et al. (10) using carrier-free ]1251]Nal (Amersham]. [ 14C]]leucine was from New England Nuclear. Hicin toxin was purchased from Sigma chemicals. Culture media and supplements were purchased from Gibco. Rabbit anti human IgG and goat unspecific IgG were from Dako, Denmark: goat anti PLAP was from Merck and FITC-labeled goat anti rabbit immunoglobulins was from Southern Biotech. IgG-Fab fragments were obtained by papain digestion. Fc fragments were removed by Protein A-Sepharose chromatography or ion exchange in DE52 (Whatman). All other chemicals were of the highest quality available. CeII culture: HEp2 cells (11) were grown at 37 OC. in Dulbecco’s modification of Eagle medium supplemented with 10 % fetal calf serum. The cells were harvested with mild digestion with trypsin (0.02Oh) and EDTA (0.02%) in saline. For microscopic studies, HEp2 cells were directly cultured on glass slides in a petri dish. Internalkation of human IgG in HEp2 cells: HEpP cells were cultured as described in 24-well plastic trays (Costar) for 24 hours. The cells were then cooled to 4oC and cold, serum-free medium, supplemented with 5 mg/ml [ 1251]human IgG (2 x lo6 cpm/mg) were added to each well. The cells were incubated at 37OC for the indicated periods of time. At the end of each incubation, the medium was replaced with 1 ml 10 016TCA. The cell precipitate was further washed twice with 10 % TCA, dissolved in 0.1 M NaOH and transferred to counting vials. Preparation of SASD-coupled IgG: 1OOi.d of SASD solution in phosphate buffer 0.05 M. pH 7.5, containing 0.6 mg crosslinker/ml were iodinated with 20 pCi carrier-free [1251]NaI using a Iodobead (Pierce) prewashed in TBS. The mixture was pipetted, after 30 seconds reaction, into another tube containing 0.3 ml of pure human IgG containing 5.3 mg protein/ml borate buffer 0.1 M, pH 8.4. All the procedures were performed in the dark. After 4 hours reaction the crosslinker-conjugated IgG was ready to be employed in the following experiment. Detection of IgG binding proteins in I-lJ3p2 cell membranes using crosslinker-coupled m: HEpP cells growing to confluency in petri dishes were treated with 0.5 ml of the SASD-IgG preparation in 10 ml serum-free culture medium, at 4oC, in a dark room, for 2 hours. The cells were then exposed to LlV light (350 nm) for 15 minutes, harvested with a rubber policeman. washed 3 times with TBS. reduced with 10 pg/ml D’l”f in TBS. carboxymethylated with 20 pg/ml iodoacetamide in TBS and homogenized in TBS containing 0.5 O/6 NP-40 and 6 M guanidinium-HCl. A sample of this preparation was immunoprecipitated with rabbit anti PLAP antibodies and insolubilized protein A (Calbiochem]. The samples were electrophoresed in a 7.5 % SDS-polyacrylamide gel and autoradiographed. Immunofluorescence studies: Cells growing on glass microscope slides were trated with Fab fractions of goat anti PLAP or goat unspecific antibodies (150) in serum-free medium at 4OC for 1 hour. After washing the antibodies with cold TBS. the cells were incubated with 5 mg/ml protein A-purified human IgG in serum-free medium at 37OC in a humidified incubator for 20 minutes. Immediately after the incubation, the cells were washed with cold TBS and fixed with 3.7 % formaldehyde in TBS for 1 hour. The fixative solution was washed off with TBS containing 0.2 % bovine serum albumin and the cells permeabilized with 0.1 % saponin in TBS for 30 minutes. The slides were covered with rabbit anti human IgG [1:500) in TBS and incubated in a humidified chamber for 2 hours at 37OC in the dark. The fluorescent antibody was removed with 3 washes of TBS. and the preparation mounted under thin coverslips with a drop of 5 O/bN-propyl gallate in 70 % glycerol in TBS. Identical protocols were used when assessing transferrin internalization. Transferrin was previously saturated with ferric ions, incubating 5 mg of the protein in Tris buffer 0.1 M pH 7.4 containing 50 mM ferric chloride and 5 mM sodium bicarbonate. The excess of ferric ions was removed by gel filtration on Sephadex G-50.

625

Vol.

182, No. 2, 1992

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

Internalization of ricin toxin in HJZp2 cells: This experiment was carried out essentially as described by Moya et a1.(12). HEpP cells were cultured on 24-well &sue culture dishes for 24 hours. To each well, 0.5 ml medium containing 40 pg/ml Fab fragments of either rabbit anti PIMP antibodies or rabbit preimmune IgG. After 1 hour incubation the medium was replaced by fresh medium containing the indicated concentrations of ricin toxin. The cells were incubated for 30 minutes before replacing the medium with a fresh one without toxin. After 4 hours incubation, the medium was again replaced with a leucine-free medium supplemented with 0.1 pCi/ml [ 14C]leucine and incubation was performed for 1 hour. The radioactive medium was then removed and 1 ml 10 % TCA was added to each well. The precipitate was washed 3 times with TCA , dissolved in 0.5 ml 0. 1M NaOH and transferred to vials containing 2 ml Aquasol (New England Nuclear) and counted. RESULTS We have interaction,

observed

PLAP

immunoglobulins. Mikulska

behaves

like

Furthermore,

et al. (13.

immunochemically establish

a distinct

the function

binding

between

a Fc receptor,

PLAP

shown

these findings,

placental

studies

human

intracellular reflecting carry

out

basolateral

reached

state

maximum

between

portion

Fc receptor

at the cellular

transcytosis,

because

membranes.

after

of

the

described

by and

level are required

to

surface PLAP (15). HEp2 cells

of absence

and

place

indicating

[Fig. 1). At 37OC, the followed

degradation.

by a plateau The

lysosomal

for IgG in these cells since they are unable of differentiation

At 4OC, IgG characteristically

took

fashion

45 minutes

internalization

seems to be the final destination

no internalization

In this

of PLAP as Fc receptor.

IgG content

compartment

Fc

IgG.

electrophoretically

IgG in a time- and temperature-dependent

a steady

human

the

to be undistinguishable

In this study we have used the HEpP cell line expressing internalize

and

binding

PLAP and the putative

14) were

(8). Besides

AND DISCUSSION

generating

accumulated

the IgG internalization

apical

at the cell surface to be receptor

30? wm

(x103) 25

20

30

40

50

80

70

80

90

time(min)

Figure 1. Internalization and degradation of human [125 1)IgG by HEp2 cells incubation times, at 37 OC. Filled circles. internalized IgG; open circles, degraded

and and

mediated

(not shown).

10

to

at different IgG.

Vol.

182,

No.

2, 1992

In order internalization labeled crosslinker, has

cells,

a reagent one

forms

near neighbour When

intact

at the surface

crosslinks

with bonds

amino

end (= lo-*

grups.

following

capable

ethylThe

ensures

to mediate

of intact

IgG

cells were

molecules.

The

1,3’-dithiopropionate

other

photoactivation. to be splitted

seconds)

COMMUNICATIONS

interacting

salicylamido)

allows the crosslinker

life of the photoactivable with

proteins

specifically

reacting

RESEARCH

potentially

P-[p-azido

covalent

of the molecule,

BIOPHYSICAL

proteins

IgG-binding

which

end

AND

the membrane

sulfosuccinimidyl

radioiodinated, middle

to identify in HEp2

with

(SASD).

BIOCHEMICAL

end,

which

A disulfide

off by reduction.

crosslinking

can

bond

in the

The short half

only between

molecules

relationships. HEp2

reduced,

the radioactive

2). which

was identified

cells were incubated residue

was transferred

with

( 1251)SASD-IgG,

to only one major

as PLAP by hnmunoprecipitation.

photoactivated membrane

Radioactive

protein

116

42

2. Autoradiography

of the IgG binding

proteins

in HEpP cell plasma membranes.

Line A shows the proteins specifically radiolabeled after the interaction with 112511SASDIgG. Proteins immunoprecipitated with anti PW antibodies are presented in line B. The numbers at the right side indicate the migration of molecular mass markers in kDa.

627

and (Fig.

IgG heavy and light

MR (kD)

Figure

be

Vol.

182, No. 2, 1992

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

Vol.

182,

No.

2, 1992

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

% ‘00 SO-

80-

40.

zo-

l-i/ 1

5

50

10

‘00

nM

Figure 4. Detection of the protein synthesis inhibition caused by internalized ricin toxin in HEp2 cells pretreated with anti PLAF’ antibodies (filled circles) or with control unspecific antibodies (open circlesl. The graphic shows % of protein synthesis inhibition. compared to non-treated cells, as a function of the amount of ricin toxin added to the culture medium.

chains,

not completely

responsibility Fab

fragments

the

resembling

lysosomes. with

PL.AP-Fab

These

to the

at the cell surface,

probably

imputes

3C and

antibodies

3D). The

to block

antibodies Alkaline

the

manner,

these

studies

clustered incubation

IgG-PL.AP antibodies

on HEp2

in coarse

different

when with

due to unspecific

cells

granules

the cells were

IgG.

Some

binding,

which

the internalization

because

transfer&r

endocytosis

is a known

(16). and ricin

is clathrin-independent

gave identical

internalization

patterns

of ricin

of transferrin

faint but no

toxin

and ricin

prototype

toxin

of the

seems to undergo

(12). Preincubations in cells internalizing

was followed

with

transferrin

by measuring

of the protein synthesis caused by the toxin intracellularly patterns and the ID56s are undistinguishable in cells treated

membrane

PLAP the

(Fig 3B).

were chosen

type of internalization

inhibition inhibition control

prior

receptor-mediated

anti PLAP or control

appeared was markedly

did not perturb

two ligands

shown

Immunofluorescence

fragments

PL.AP antibodies

clathrin-dependent,

were

of IgG which

IgG could be detected

Anti

(Figs.

This result highly

site (8). In a similar

(Fig. 3A). The pattern

was observed

intracellular

antibodies

IgG internalization.

internalization anti

fluorescence

PLAP

with the binding

to hinder

confirmed

another

of anti

by competing

are expected

toxin.

out. were also observed.

for IgG internalization.

Interaction

treated

washed

the

(12). The protein with anti PLAP or

(Fig. 4). phosphatases

by a glycan

are members

phosphatidyl

of a family

inositol

of proteins

(GPI) anchor.

(17.18).

attached Even

to the plasma secluded

to the

Figure 3. Detection of intracellular IgG by immunofluorescence in HEp2 cells pretreated with unspecific goat antibodies (A) or goat anti PLAP antibodies (B). Detection of intracellular transferrin by the same method in HEpP cells pretreated with unspecific goat antibodies (C) or with goat anti PLAF’ antibodies (D). Magnification=POOX

629

Vol.

BIOCHEMICAL

182, No. 2, 1992

outer

hem&leaflet

of the plasma

from the cytoplasmic internalization

evidences

which

the GPI-tailed

events. They are able to interact

and recycle between

Our results internalization

membrane.

AND BIOPHYSICAL

the plasma

proteins

could

IgG in HEp2 permit

based on a more structural

membrane

cells. The present

an approach

and dynamic

seem not to be isolated

with cytoplasmic

structures,

and endosomes

suggest that PLAP not only has capacity

of human

RESEARCH COMMUNICATIONS

to bind, report

to the disclosure

undergo

(19-23).

but also to promote

provides.

the

for the first time,

of the biological

role of PLAP,

viewpoint.

REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

McComb, R. B., Bowers, G. N. and Posen, S. (1979) In Alkalinephosphatase. New York, Plenum Stigbrand, T. and Fishman, W. H. (1984) In Human alkalinephosphatases. New York, A. R Liss Harris. H. (1989) CIin. Chim. Acta 186, 133-150 Lange, P. H.. MiIIfm, J. L., Stigbrand. T., VesseIIa, R. L., Ruohslahti. E. and Fishman. W. H. (1982) Cancer Res. 42.3244-3247 Jeppsson. A., Wahren. B., Brehmer-Andersson. E., Silfversward. C., Stigbrand. T. and MiIIan. J. L. (1984) Int. J. Cancer 34, 757-761 Fraser, D. (1957) Am. J. Med. XXII, 730-746 Millan, J. L. (1990) Prog. CIin. Biol. Res. 334, 453-475 Makiya, R and Stigbrand. T. (1991) Submitted to Eur. J. Biochem. Hohngren. P. A. and Stigbrand, T. (1976) Biochem. Genet. 14, 777-789 Greenwood, F. C.. Hunter, W. M. and Glover. J. S. (1963) J. Biochem. 89. 114-123 Moore, A E., Sabachewsky. L. and Toolan, H. W. (1955) Cancer Res. 344.453-475 Moya. M., Dautry-Varsat, A , Goud, B., Louvard, D. and Boquet, P. (1985) J. Cell Biol. 101.548-559 Mikulska, J., Boratynski, J.. Niezgodka. M. and Lisowski, J. (1982) Immunol. Letters 5, 137-143 Mikulska, J. and Lisowski, J. (1987) Arch. Immunol. Ther. Exp. 35, 819-829 Riklund, K. E., Makiya, R. A., Sundstrom. B. E.. ThomeII. L-E. and St&brand. T. (1990) Anticancer Res. 10.379-384 Wileman, J. P., Harding, C. and Stahl, P. (1985) Biochem. J. 232, 1-14 Low, M. G. and ZiIversmit, D. B. (1980) Biochemistry 19.3913-3918 Kominami, T., Miki, A. and Ikehara. Y. (1985) Biochem. J. 227. 183- 189 Tausk, F.. Fei. M. and Gigh. I. L. (1989) J. Immunol. 143. 3295-3302 Kammer, G. M.. Waker, E. I. and Medof, M. E. (1988) J. Immunol. 141, 2924-2928 Lemansky, P.. Fatemi, S. H.. Gorican, B., Meyale, S., Rosero, R and Tartakoff, A. M. (1990) J. Ceil Biol. 110, 1525-1531 Schell, D., Evers, R., Preis. D., Ziegelbaver, K.. Kiefer, K., Lottspeich, F., Comehssen. A W. C. A and Overath, P. (1991) EMBO J. 10. 1061-1066 van den Bosch, R A, du Maine, A P. M., Geuze. H. J., van der Ende, A and Strous, G. J. (1988) EMBG J. 7.33453351

630

Placental alkaline phosphatase is related to human IgG internalization in HEp2 cells.

The biological function(s) of placental alkaline phosphatase has not yet been unraveled. The low catalytic activity of the enzyme at physiological pH,...
493KB Sizes 0 Downloads 0 Views