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