Trophoblast-like human choriocarcinoma cells serve as afrom high density lipoproteins Christian Wadsack1, Andelko Hrzenjak1, Astrid Hammer2, Birgit Hirschmugl1, Sanja Levak-Frank1, Gerno
Trang 1Trophoblast-like human choriocarcinoma cells serve as a
from high density lipoproteins
Christian Wadsack1, Andelko Hrzenjak1, Astrid Hammer2, Birgit Hirschmugl1, Sanja Levak-Frank1,
Gernot Desoye3, Wolfgang Sattler1and Ernst Malle1
1
Institute of Medical Biochemistry and Molecular Biology,2Institute of Histology and Embryology,3Clinic of Obstetrics
and Gynecology, Karl-Franzens University Graz, Austria
As human choriocarcinoma cells display many of the
biochemical and morphological characteristics reported for
in utero invasive trophoblast cells we have studied
cho-lesterol supply from high density lipoproteins (HDL) to
these cells.Binding properties of 125I-labeled HDL
sub-class 3 (HDL3) at 4C were similar for BeWo, JAr, and
Jeg3 choriocarcinoma cell lines while degradation rates at
37C were highest for BeWo.Calculating the selective
cholesteryl ester (CE)-uptake as the difference between
specific cell association of [3H]CE-labeled HDL3 and
holoparticle association of125I-labeled HDL3revealed that
in BeWo cells, the selective CE-uptake was slightly lower
than holoparticle association.However, the pronounced
capacity for specific cell association of [3H]CE-HDL3and
selective [3H]CE-uptake in excess of HDL3–holoparticle
association, and cAMP–mediated enhanced cell
associ-ation of [3H]CE-HDL3 in JAr and Jeg3 suggested the
scavenger receptor class B, type I (SR-BI) to be respon-sible for this pathway.Abundant expression of SR-BI (but not SR-BII, a splice variant of SR-BI) could be observed
in JAr and Jeg3 but not in BeWo cells using RT-PCR, Northern and Western blot analysis, and immunocyto-chemical technique.Adenovirus-mediated overexpression
of SR-BI in all three choriocarcinoma cell lines resulted in
an enhanced capacity for cell association of [3H]CE-HDL3 (20-fold in BeWo; fivefold in JAr and Jeg3).The fact that exogenous HDL3 remarkably increases proliferation in JAr and Jeg3 supports the notion that selective CE-uptake and subsequent intracellular generation of cholesterol is coupled to cellular growth.From our findings we propose that JAr and Jeg3 cells serve as a suitable in vitro model to study selective CE-supply to human placental cells Keywords: BeWo; JAr; Jeg3; placenta; SR-BI
Mammalian fetal nutrition during development is wholly
dependent on the transport of nutrients by the placenta
Cholesterol is not only a major component of cell
membranes but is also necessary for the maintenance of
fetal growth and serves as the initial substrate for steroid
hormone synthesis [1].There are two sources of cholesterol
in the fetus, as in any other tissues.The first source is
endogenous and constitutes cholesterol synthesized within
the fetus.The second source of fetal cholesterol is exogenous [2,3].Exogenous cholesterol could be derived from choles-terol synthesized in the yolk sac and placenta or by transport of lipoprotein-associated cholesterol across the placenta from the maternal circulation [4–8]
The fact that the placenta binds and internalizes maternal lipoproteins both in vivo and in vitro [8] suggests the existence of functionally intact lipoprotein receptors [2–11] The presence of active lipoprotein receptors in the yolk sac, the placenta and placenta-derived cells support internaliza-tion of cholesterol-rich maternal lipoproteins [8].In addiinternaliza-tion
to low-density lipoprotein (LDL) that are taken up by the classical receptor-mediated endocytosis pathway (the
LDL-or apolipoprotein B/E-receptLDL-or), the high clearance rate of high-density lipoprotein (HDL) in rodents suggested specific HDL receptors [8] contributing to sterol metabolism in fetal tissues.Among these gp280 preferentially mediates HDL-holoparticle-uptake [9] while scavenger receptor class, B, type I (SR-BI), the predominant receptor mediating select-ive uptake of cholesteryl esters (CEs) without concomitant holoparticle-uptake [12,13], has been identified in steroido-genic tissue and the liver [12] as well as in human [14] and rodent placental tissues [15,16].Immunohistochemistry revealed strong induction of SR-BI expression in murine giant trophoblast cells that surrounded the developing embryo [16].The fact that murine SR-BI (mSR-BI) is
Correspondence: E.Malle, Karl-Franzens University Graz,
Institute of Medical Biochemistry and Molecular Biology,
Harrachgasse 21, A )8010 Graz, Austria;
Fax: +43 316 380 9615; Tel.: +43 316 380 4208;
E-mail: ernst.malle@kfunigraz.ac.at
Abbreviations: [3H]CE, [1,2,6,7–3H]-cholesteryl palmitate; CE(s),
cholesteryl ester(s); CHO, Chinese hamster ovary; 8-CPTcAMP,
8-(4-chlorophenylthio)adenosine 3¢:5¢-cyclic monophosphate;
DMEM, Dulbecco’s minimum essential medium; FBS, fetal bovine
serum; HBSS, Hank’s balanced salt solution; HDL 3 , high density
lipoprotein subclass 3; LDL, low density lipoprotein; LPDS,
lipo-protein-deficient serum; mSR-BI, murine scavenger receptor class B,
type I; TCA, trichloroacetic acid.
(Received 22 July 2002, revised 10 October 2002,
accepted 26 November 2002)
Trang 2expressed on the side of the tissue that faces the maternal
blood expression supports the notion that this protein could
act as a candidate receptor for supplying cholesterol to
developing embryonic tissues for placental steroid
biosyn-thesis
The process of binding and uptake of maternal
lipopro-teins is carried out by placental trophoblast [17], which lines
the chorionic villi and represents the epithelial layer
separating the maternal and fetal circulation [18].In the
human placenta chorionic villi, a layer of
syncytiotropho-blasts lies over the cytotrophosyncytiotropho-blasts, and surrounds the
internal mesoderm and fetal capillaries [18].Trophoblasts
may also give rise to choriocarcinoma which are malignant
tumors of epithelial origin but have been shown to display
characteristics of invasive trophoblasts
[19].Choricar-cinoma cells are morphologically similar to their cell of
origin, the trophoblast of the normal first trimester placenta
and may serve as a valid and convenient in vitro model
system for studying the cellular activities and regulation of
transplacental transport and uptake mechanisms.To date,
only limited information on binding and
holoparticle-uptake of lipoproteins by choriocarcinoma cell lines is
available [20,21].However, no studies have been performed
to clarify whether choriocarcinoma cell lines may serve a
suitable in vitro model for supplying cholesterol to placental
cells via selective CE-uptake from HDL
We therefore have studied holoparticle association of and
selective CE-uptake from HDL by three human
tropho-blast-like choriocarcinoma cell lines, i.e BeWo, JAr, and
Jeg3.The BeWo cell line which is heterogenous on several
criteria [22], is comprised of cytotrophoblasts with no
differentiation to syncytium [23] under non activated
conditions [24].This is in contrast to primary cultures of
term cytotrophoblasts which aggregate and form syncytia
In contrast to BeWo and Jeg3 cell lines, JAr cells share many
of the characteristics of early placental trophoblasts, such as
synthesis of human chorionic gonadotropin and steroids
[25], and the ability to differentiate into
syncytiotrophoblast-like cells in vitro [19].Jeg3, derived originally from the BeWo
[26], express abundant human chorionic gonadotropin and
placental lactogen – hallmarks of trophoblasts [27] – and
form large, multinucleated syncytia in culture [28] which
resemble that of syncytiotrophoblasts in vivo
Results from the present study demonstrate that the
capacity of selective CE-uptake by all three
choriocar-cinoma cell lines is closely related to the expression level of
SR-BI.Only JAr and Jeg3 could serve as suitable in vitro
models to study selective CE supply to human placental
cells.High level adenoviral-overexpression of SR-BI
resul-ted in an enhanced and comparable capacity for selective
CE-uptake by all three choriocarcinoma cell lines.The lack
of SR-BII protein suggested that this splice variant of SR-BI
[29] does not contribute to selective CE-uptake by BeWo,
JAr, and Jeg3 cells
Experimental procedures
Human lipoproteins
LDL (d¼ 1.035–1.065 gÆmL)1) and HDL subclass 3
(HDL3, lacking apolipoprotein E, a ligand for the
LDL-receptor, d¼ 1.125–1.21 gÆmL)1) was prepared by
discontinuous density ultracentrifugation from plasma obtained from normolipidemic blood donors [30].Purity
of lipoprotein preparations was checked by SDS/PAGE The protein concentration was determined as described [31] using bovine serum albumin as a standard.Oxidation of LDL (ox-LDL) was performed with 1.66 lM of Cu2+in NaCl/Pias described [32]
Lipoprotein labeling procedures Single-labeling (a) Iodination of HDL3 with 125I-Na (NEN, Vienna, Austria) was performed using N-Br-succin-imide as the coupling agent [33].This procedure resulted in specific activities between 300 and 500 d.p.m.Æng)1protein with less than 3% lipid associated activity.All125I-HDL3 -preparations were monitored by SDS/PAGE to ensure that the preparations were free of radiolytic or oxidative damage (b) HDL3was labeled with [cholesteryl-1,2,6,7–3 H]-palmi-tate ([3H]CE, NEN) by cholesteryl ester transfer, protein-catalyzed transfer from donor liposomes exactly as described [34].Subsequently, the labeled HDL3-fractions were reiso-lated in a TLX120 bench-top ultracentrifuge in a TLA100.4 rotor (Beckman, Austria).The HDL3-band was aspirated and dialyzed against 10 mMNaCl/Pi, pH 7 4
Double-labeling Labeling of HDL3 with [cholesteryl-1,2,6,7–3H]-oleate (NEN) was performed by incubation of the lipoprotein with palmityl oleyl phosphatidylcholine-[3H]cholesteryl oleate vesicles exactly as described [35] Following separation from the vesicles by ultracentrifuga-tion, the labeled lipoproteins were subsequently labeled with
125I-Na as described above using N-Br-succinimide
Cell culture studies Choriocarcinoma cells The BeWo cell line (ATCC, Man-assas, VA, USA) was cultured in F12K (Kaighn’s Modi-fication) Nutrient Mixture (Gibco BRL, Vienna, Austria) supplemented with 10% (v/v) heat inactivated fetal bovine serum (FBS, hyClone, Utah, USA) containing 100 UÆmL)1 penicillin/streptomycin and 2 mM L-glutamine at 37C under 5% CO2.The JAr and Jeg3 cell lines (ATCC) were maintained in Dulbecco’s minimum essential medium (DMEM, supplemented with 10% [v/v] FBS and
100 UÆmL)1 penicillin/streptomycin) at 37C under 5%
CO2.To assess influences of exogenous HDL3 on cell proliferation, the cells were seeded into 6-well plates at a starting density of 2.5· 104cells in 2 mL 10% lipoprotein-deficient serum (LPDS) medium for BeWo, 5· 104cells for JAr and 1· 105cells for Jeg3 cells, respectively.After 24 h, media was removed, the cells were washed with Hank’s balanced salt solution (HBSS), trypsin-treated and counted with a cell counter and analyzer system (CASY1, Scha¨rfe, Reutlingen, Germany).Fresh medium containing lipopro-tein-deficient serum or lipoprolipopro-tein-deficient serum plus
100 lgÆmL)1HDL3was added and the cells were incubated
At the indicated time points the cells were washed twice with HBSS, trypsin-treated and counted
Chinese hamster ovary (CHO) cells LdlA7 (clone 7, a LDL receptor-negative CHO cell line) which express minimal levels of SR-BI were cultured in HAM’s
Trang 3F12 medium containing 5% (v/v) FBS, 2 mM glutamine,
50 UÆmL)1 penicillin and 50 lgÆmL)1 streptomycin [36];
ldlA(mSR-BI) (ldlA7 stably transformed with mSR-BI)
were maintained in medium containing 0.5 mgÆmL)1G-418
Both cell lines were kindly provided by Dr M.Krieger
(MIT, Cambridge, MA, USA)
Binding studies
Binding studies of human HDL3to choriocarcinoma cells
were performed at 4C in medium (F12K/DMEM without
FBS) with increasing amounts of125I-labeled HDL3(1, 5, 10,
25, 50, and 100 lg proteinÆmL)1) in the absence (total
binding) or presence of a 20-fold excess of unlabeled HDL3
(nonspecific binding) [37].Following this incubation the
medium was aspirated and the cells were washed twice with
HBSS containing 5% (v/w) bovine serum albumin followed
by two washes in HBSS.Cells were lysed with 0.3MNaOH
(1 mL, 1 h at 4C) to determine bound-radioactivity and cell
protein in the lysate.Protein measurement was performed as
described [31].Specific binding (4C) was calculated as the
difference between total and nonspecific binding [37]
To determine cell-associated (the sum of binding and
internalization) and degraded 125I-labeled HDL3-protein,
the cells were incubated at 37C for 4 h as described above
with the same amounts of labeled and unlabeled HDL3
Subsequently, the medium was aspirated and the cells were
rinsed as described above [34].Degradation of125I-labeled
HDL3by choriocarcinoma cells was estimated by
measur-ing the nontrichloroacetic acid-precipitable radioactivity in
the medium after precipitation of free iodine with AgNO3
exactly as described [38].Briefly, 0.5 mL of medium was
removed, mixed with 100 lL bovine serum albumin
(30 mgÆmL)1) and 1 mL trichloroacetic acid (3M, 4C)
and left at 4C for 30 min.Subsequently 250 lL of AgNO3
(0.7M) was added, mixed, and the samples were
centrifu-gated at 1500· g at 4 C for 15 min.One milliliter of the
supernatant was counted on a c-counter
To determine cell association of tritiated-HDL3, the cells
were incubated with [3H]CE-HDL3 at 37C for 4 h as
described above with the same amounts of labeled and
unlabeled HDL3 [34].After removing the medium and
washing the cells, cell association was estimated by
measur-ing the radioactivity and protein content of the cell lysates,
respectively.Specific cell association was calculated as the
difference between total and nonspecific cell association
In a parallel set of experiments, the cell association of
[3H]CE-HDL3by choriocarcinoma cells was measured by
increasing concentrations of unlabeled competitors (HDL3
and ox-LDL) as well as in the absence or presence of 300 lM
8-(4-Chlorophenylthio)adenosine 3¢,5¢-cylic monophosphate
(8-CPTcAMP, Sigma, Vienna)
Selective CE-uptake
To calculate the selective CE-uptake from HDL3during the
same experiment, choriocarcinoma cells were incubated
with125I-labeled and [3H]CE-HDL3.In case of125I-labeled
HDL3, the cell-associated and nontrichloroacetic
acid-precipitable radioactivity in the medium was counted.The
sum of cell-associated and degraded 125I-labeled HDL3
reflects holoparticle association.For the [3H]CE-HDL
-particles only specific cell-associated radioactivity (the difference between total and nonspecific cell association of [3H]CE-HDL3) was counted.To facilitate quantitative comparison of125I-labeled and [3H]CE association, results are expressed in terms of HDL3-protein equivalents [39] calculated on the basis of the specific activity of the correspondingly labeled HDL preparation used that would
be necessary to deliver the observed amount of tracer.These calculations are performed to compare uptake of125I- and
3H-tracers on the same basis
In a parallel set of experiments, the selective CE-uptake from HDL3was calculated from double-labeled HDL3as the difference between [3H]cholesteryl oleate- and 125 I-labeled lipoprotein [35].The sum of specific degradation (measured as described above) and specific cell association of
125I-labeled HDL3reflects HDL3–holoparticle association
SDS/PAGE and Western blotting Detergent extracts of solubilized membrane protein frac-tions [33] or total cell proteins of choriocarcinoma and CHO cells were separated on 8% SDS/PAGE.Protein transfer to nitrocellulose membranes was performed electrophoreti-cally at 150 mA, 4C, 50 min [34].Immunochemical detection of SR-BI and SR-BII (a splice variant of SR-BI lacking the C-terminal portion [29]) was performed with a sequence-specific rabbit anti-SR-BI-peptide (496–509) and anti-SR-BII peptide (491–506) IgG (dilutions 1 : 1500, Abcam, Cambridge, UK).Immunoreactive bands were visualized with peroxidase-conjugated goat anti-rabbit IgG and ECL development [30]
Reverse-transcriptase-polymerase chain reaction Total RNA from choriocarcinoma cell lines was isolated by using RNeasy kit (Qiagen, Vienna, Austria).Three micro-grams of total RNA were treated with RQ1 RNase-free DNase I (Promega, Mannheim, Germany) for 15 min at
37C and subsequently used as a template for first strand cDNA synthesis in a 30-lL reaction.The PCR condi-tions applied were described in detail elsewhere [40].The following oligonucleotides were used: forward primer (A) 5¢-TCTACCCACCCAACGAAGGCT-3¢ (nucleotides 1007–1027), SR-BI reverse primer (B) 5¢-CCTGAATGGC CTCCTTATCCT-3¢ (1514–1534) and SR-BII reverse primer C-5¢-AGAAGCGGGGTGTAGGGACTGG-3¢ (1655–1676) (MWG Biotech, Ebersberg, Germany).By using the primers A and B, a 527-bp fragment was obtained
By using the primers A and C, two fragments were obtained:
a 669-bp-fragment for SR-BI and a 540-bp-fragment for SR-BII (129 bp shorter than SR-BI).All fragments were subcloned in TOPO-TA cloning vector and sequenced Northern blot analysis
Total RNA was isolated from choriocarcinoma and human liver tissues (used as a positive control) by the RNA-easy kit (Qiagen) exactly as described [40].A 553-bp fragment, amplified by RT-PCR [forward primer 5¢-TCGCTCAT CAAGCAGCAGGT-3¢ (nucleotides 169–188), reverse pri-mer 5¢-GCCCAGAGTCGGAGTTGTTG-3¢ (nucleotides 702–721)] was used as a probe [30]
Trang 4Construction of recombinant human SR-BI adenovirus
The adenoviral plasmid shuttle vector (pAvCvSv) and
pJM17 vector were kindly supplied by L.Chan (Baylor
College of Medicine, Houston, Texas).Human SR-BI
cDNA (kindly provided by H.Hauser, ETH, Zu¨rich,
Switzerland) which was originally inserted into pcEXV-3
vector was partially restricted with EcoR I and the 2 5 kb
band was eluted from the gel.This band was subcloned into
pBluescript using the EcoRI site, amplified, restricted with
Kpn-I and this fragment was finally partially restricted with
BamHI.The plasmid shuttle vector was opened using Kpn-I
and BglII, and Kpn-I/BamHI restricted SR-BI cDNA was
inserted.These modifications were necessary to enable
insertion of SR-BI cDNA under the control of the CMV
promoter in the plasmid shuttle vector.Recombinant
vector (pAvCvSv/human SR-BI) was used to transform
E coliDH5-a competent cells in order to amplify the
recombinant plasmid.Positive clones were confirmed by
restriction analysis and DNA-sequencing.The resulting
recombinant shuttle plasmid (5 lg) was cotransfected with
5 lg supercoiled pJM17 into 293 cells by calcium-phosphate
coprecipitation.Two weeks after transfection, infectious
recombinant adenoviral vector plaques were picked,
pro-pagated, and screened for SR-BI sequences by PCR
Adenoviral vectors containing SR-BI were further amplified
in 293 cells and the expression was determined by Western
blotting analysis.Large-scale production of high-titer
recombinant adenovirus was performed as described [41]
The virus was purified twice by caesium chloride density
gradient centrifugation and dialyzed for 14 h at 4C against
a buffer containing 10 mMTris/HCl, pH 7.5, 1 mMMgCl2,
10% glycerol and stored at)70 C.Virus-titer was
deter-mined by plaque-assay on 293 cells and was found to be
2· 1010 pfuÆmL)1.Control b-galactosidase (b-gal) and
viruses were amplified and purified as described above
Adenovirus infection of choriocarcinoma cells
Choriocarcinoma cells were cultivated in 12-well culture
dishes.At a density of 5· 104cells per cm2they were rinsed
once with NaCl/Piand infected with recombinant
adeno-viruses (MOI¼ 30 pfuÆmL)1) in 1 mL of infection media
(DMEM medium containing 2% FBS, 50 UÆmL)1
penicil-lin, and 50 lgÆmL)1streptomycin) for 2 h (37C, 5% CO2,
humidified atmosphere).After removing infection-media,
the cells were supplied with 2 mL of S12K medium
(containing 5% FBS, 2 mMglutamine, 50 UÆmL)1
penicil-lin and 50 lgÆmL)1streptomycin) and the incubation was
continued for 36 h without rocking.Control cells were
infected with b-gal virus as described for SR-BI infected
cells.The expression rate of SR-BI was determined by
Northern and Western blot techniques as described above
Immunofluorescence and confocal
laser scanning microscopy
Immunofluorescence was performed on choriocarcinoma
cells (24 h in DMEM containing 10% FBS, 1% glutamine
and 1% penicillin/streptomycin) and CHO cells (HAM’s
F12 medium containing 5% (v/v) FBS, 2 mM
gluta-mine, 50 UÆmL)1penicillin and 50 lgÆmL)1streptomycin)
cultured on Laboratory-Tek chamber slides (Nunc, USA) After 24 h the cells were washed with HBSS, air dried (2 h
at 22C), and acetone-fixed (5 min) [42].Fixed chamber-slides were incubated for 30 min with rabbit anti-SR-BI IgG (dilutions of 1 : 1000, NB 400–104, Novus Biologicals, USA) followed by a 30-min incubation with cyanine-3-labeled goat anti-rabbit IgG (dilution 1 : 400, Jackson Dianova).NaCl/Piwas used for washing sections between different incubation steps.Sections were mounted with Moviol (Calbiochem-Novabiochem, La Jolla, CA) and analyzed on a confocal laser scanning microscope (Leica TCS NT, Heidelberg, Germany) equipped with an argon-krypton laser.Signals were detected with a double dichroitic beam splitter (488/568 nm), using a 590-nm long pass filter for cyanine-3 [42]
Control experiments encompassed immunocytochemis-try (a) without primary detection antibodies (b) with polyclonal nonimmune antibodies as primary antibodies (c) without secondary antibodies, or (d) using ldlA(mSR-BI) membrane protein fractions as competing antigen; the competitor (20-fold molar excess) was preincubated with the primary antibody for 20 min before adding to the section Pictures were taken with an Axiophot microscope (Zeiss, Oberkochen, Germany)
Results
Binding of HDL to choriocarcinoma cells Specific binding of HDL3to all three choriocarcinoma cell lines at 4C was saturable at the highest lipoprotein concentrations (100 lg proteinÆmL)1) used.Calculation of binding parameters yielded similar Kdand bmaxvalues for all three cell lines (Table 1).Also the specific cell association of
125I-labeled HDL3(the sum of binding and internalization) measured at 37C was comparable for all three cell lines (Fig.1A–C).The specific degradation of125I-labeled HDL3 determined in a parallel set of experiments was highest for BeWo and decreased in the following order: BeWo > JAr > Jeg3
Selective HDL-CE uptake by choriocarcinoma cells For BeWo cells, the specific cell association of [3 H]CE-HDL3 increased up to 1200 ng HDL3Æmg)1 cell protein (Fig.2A) and is approximately 1.8-fold higher than
125I-labeled HDL3–holoparticle association (the sum of specific cell association and degradation of 125I-labeled
Table 1 Specific binding of 125 I-labeled HDL to choriocarcinoma cell lines at 4 °C Binding constants were calculated by non linear regres-sion analysis ( GRAPHPAD ).Values are given as means ± SD of three independent experiments.
K d
(lg HDL 3 -proteinÆmL)1)
b max
(ng HDL 3 -proteinÆmg)1 cell protein)
BeWo 41.9 ± 10.1 161.3 ± 17.3 JAr 34.1 ± 3.9 145.9 ± 6.9 Jeg3 29.7 ± 7.4 186.9 ± 21.2
Trang 5HDL3).However, calculating the selective CE-uptake from
HDL3as the difference between both tracers revealed that
this pathway is not a preferential routing to supply BeWo
with CEs from HDL3.To confirm the low capacity of
BeWo cells for selective CE-uptake, the HDL3-particle was
double-labeled and values obtained for specific cell
associ-ation of [3H]cholesteryl oleate- and125I-labeled HDL3were
compared with those obtained by single labeling
experi-ments.Comparable results were obtained with both labeling
techniques (data no shown).In contrast to BeWo, JAr cells
showed a pronounced capacity for specific CE tracer uptake
from HDL3in excess of holoparticle association, exceeding
holoparticle association by approximately fivefold
(Fig.2B) In parallel, the selective CE–uptake exceeded
holoparticle association by approx.4-fold (2703 vs.652 ng
lipoproteinÆmg)1cell protein) at the highest HDL3
-concen-trations used (Fig.2B).Although specific cell association
of [3H]CE-HDL3 in Jeg3 cells (Fig.2C) was similar to
BeWo (Fig.2A), the capacity for selective CE-uptake from
HDL3exceeded holoparticle association by approximately
sixfold (1537 vs.267 ng lipoproteinÆmg)1 cell protein) in
these cells
Identification and characterization of SR-BI
Based on the pronounced capacity of JAr and Jeg3 cells
(Fig.2B,C) for selective uptake of HDL3-associated CEs it
was reasonable to assume that this pathway is caused by
high expression levels of SR-BI.Using specific primers for
SR-BI the corresponding 527 bp product (primer A and B,
Fig.3A) and 669 bp product (primer A and C, not shown)
was abundantly present in JAr and Jeg3 cells while only
negligible amounts became apparent in BeWo cells.In
parallel (using primers A and C) a faint 540 bp band
(indicative for SR-BII) was found in all three cells lines (not
shown).Northern blotting experiments confirmed data
obtained by RT-PCR (Fig.3B).Western blot experiments
finally revealed high expression of SR-BI in JAr and Jeg3
cells (Fig.3C).In BeWo cells, marginal SR-BI expression
was observed on RNA level but no SR-BI protein was
detectable under the conditions described (Fig.3).Western
blot experiments of detergent solubilized membrane protein
fractions revealed no expression of SR-BII on all three
choriocarcinoma cell lines (not shown)
Using specific antibodies, pronounced staining for SR-BI could be observed primarily on the plasma membrane of JAr and Jeg3 cells and to a much lesser degree in the cytoplasma of both cell lines (Fig 4B,C).Omission of the primary antibody (data not shown) or replacement of primary antibodies with an IgG isotype control (data not shown) eliminated all staining.Competition experiments of antiSR-BI IgG preabsorbed with SR-BI-enriched mem-brane protein fraction from ldlA(mSR-BI) cells at a molar ratio of 1 : 20 prevented antibody binding, demonstrating that staining was specific for SR-BI.The lack of SR-BI expression in BeWo cells (Fig.3C) could be confirmed by
Fig 1 Specific cell association and degradation of125I-labeled HDL by
choriocarcinoma cells Following preincubation in F12K, [BeWo (A)],
or DMEM [JAr (B) and Jeg3 cells (C) without FBS (12 h)], the cells
were incubated for 4 h (37 C) in the presence of increasing amounts of
125 I-labeled HDL 3 To determine specific cell association of 125
I-labe-led HDL 3 (closed triangles), the cells were incubated in the absence
(total cell association) or presence of a 20-fold excess of unlabeled
HDL 3 (non specific cell association).The cells were washed and lysed
with 0.3 M NaOH to determine the cell-associated fraction (the sum of
bound and internalized radioactivity).To determine specific
degrada-tion (closed circles – the difference between total and nonspecific
degradation) the cells were incubated under the same conditions as
described above.After 4 h the cellular supernatant was collected to
determine the nontrichloroacetic acid-precipitable radioactivity as
described in Materials and methods (degraded).Results are given as
means ± SD of three independent experiments.
Trang 6immunocytochemistry (Fig.4A).To further verify
specifi-city of the primary antibody on choriocarcinoma cells,
CHO cells were used as controls.While faint staining was
observed on ldlA7 cells (expressing minimal levels of
SR-BI, Fig.4E), bright staining could be observed on
ldlA(mSR-BI) cells (Fig.4D)
SR-BI expressed on choriocarcionoma cells contributes
to CE-uptake from HDL
To further confirm that SR-BI accounts for the high rates of
selective CE-uptake by JAr and Jeg3 cells, a series of
competition experiments were performed.While HDL3and ox-LDL (a high affinity ligand for SR-BI [10]) were equally effective to compete for cell association of [3H]CE-HDL3in BeWo cells, HDL3 lowered cell association of [3 H]CE-HDL3 by approx 25% in JAr and Jeg3 cells (Table 2) Further inhibition (up to 60%) was achieved by increasing the competitor concentration of unlabeled HDL3 to
500 lgÆmL)1.The addition of ox-LDL led to a pronounced inhibition in JAr and Jeg3 cells, findings in line with previous results [43] performed on liver cells that express high levels of SR-BI
Next, cell association from [3H]CE-HDL3was studied in cells preincubated with 8-CPTcAMP, a direct activator of SR-BI via cAMP-dependent protein kinase pathway [10,11].In line with findings shown in Fig.2, cell association
of [3H]CE-HDL3is lowest in BeWo compared with JAr and Jeg3 cells (Fig.5).Following 8-CPTcAMP treatment, cell association of [3H]CE-HDL3was unaltered in BeWo but increased in JAr and Jeg3 cells.These changes were paralleled by SR-BI mRNA levels (Fig.5)
Finally, we tested whether transient overexpression of SR-BI would restore the ability of BeWo for cell association of [3H]CE-HDL3.Therefore, choriocarcinoma cell lines were transiently transfected with the human SR-BIgene and expression was followed by Northern blot (data not shown) and Western blot experiments (Fig.6) The SR-BI protein expressed was predominantly localized
at the plasma membrane as determined by immunoblot analysis of plasma membrane preparations.In order to test the functionality of the adenoviral SR-BI construct, the cells were incubated with [3H]CE–HDL3 and cell association of [3H]CE-HDL3was measured.While mock-transfection of all three choriocarcinoma cells did not change the capacity for cell association of [3H]CE-HDL3
compared with wild type cells, adenoviral overexpression increased the capacity for cell association of [3H]CE-HDL3
approximately fivefold (JAr and Jeg3) and 20-fold (BeWo), respectively.From this set of experiments, we conclude that transfection of all three choriocarcinoma cell lines results in functionally active SR-BI protein, and that transfection in BeWo (a cell line with low SR-BI expression comparable to ldlA7 cells [44]) results in cell association of [3H]CE-HDL3to an extent similar as shown with JAr and Jeg3 under the same conditions
Fig 2 Specific cell association of [3H]CE-HDL and selective CE-uptake from HDL by human choriocarcinoma cells Following preincubation in F12K [BeWo (A)], or DMEM [JAr (B) and Jeg3 cells (C) without FBS (12 h)], the cells were incubated for 4 h with increasing amounts of [ 3 H]CE-HDL 3 at 37 C.To determine spe-cific cell association of [ 3 H]CE-HDL 3 (closed circles) the cells were incubated in the absence (total association) or presence of a 20-fold excess (nonspecific association) of unlabeled HDL 3 Subsequently, the cells were washed and lysed in 0.3 M NaOH to measure associated radioactivity.The selective CE-uptake (closed triangles) was calculated as the difference between specific cell association of [ 3 H]CE-HDL (closed circles) and 125 I–labeled HDL holoparticle association (closed squares, the sum of specific cell association and degradation, Fig.2) Results are given as means ± SD of three independent experiments.
Trang 7Effects of exogenous HDL on growth rates
To investigate whether (a) the pronounced cell association
of [3H]CE-HDL3and the high capacity for selective
CE-uptake from HDL3can be shown during culture conditions
and (b) to assess the role of exogenous cholesterol pool on
cell proliferation of choriocarcinoma cells, the
time-depend-ent effect of exogenous HDL3on cellular cholesterol levels
was measured (Fig.7A–C).While HDL3had no significant
effect on cell proliferation in BeWo cells, cell numbers were
significantly (P > 0.05) increased in JAr and Jeg3 cells in
the presence of HDL3 after 48 h.Analysis of the cellular
cholesterol content revealed that HDL3led to a remarkable
increase in cellular cholesterol levels in JAr and Jeg3 cells
compared to cells cultured in the absence of HDL.The
cellular cholesterol content of BeWo cultures was unaffected
by the presence of HDL3; findings which are in line with the low rate of selective CE-uptake from HDL3in these cell lines (Fig.2) demonstrating that exogenous HDL3may not significantly alter CE synthesis in these cell lines [45]
Discussion
During placental development the trophoblast cells develop along a cell lineage forming the villous cytotrophoblast with the overlaying syncytiotrophoblast, both responsible for hormone production and fetomaternal exchange (reviewed
in ref [18]).In addition, another trophoblast population, the extravillous trophoblast of cell island and cell columns, is formed which maintains the ability of proliferation and invasion.Choriocarcinoma is a malignant neoplasm that represents the early trophoblast of the attachment phase or
as later invasive stage [46–48].Thus, in most cases, choriocarcinoma has the appearance of trophoblast, being predominantly syncytiotrophoblastic or cytotrophoblastic Some cytotrophoblastic choriocarcinomas secrete little human chorionic gonadotropin and some choriocarcino-mas also secrete human placental lactogen [22–27].A number of choriocarcinoma cell lines have been established; these replicating trophoblasts, derived from the malignant tumor or produced by viral transformation of normal trophoblasts, are appropriate systems to mimick tropho-blast behaviour in vitro
Exogenous sources for cholesterol supply to fetal tissues involve receptor-mediated uptake of maternal LDL [2,17] Both trophoblasts [17,49] and trophoblast-like cells have been reported to bind LDL [17,20,21] Simpson et al.[21] further reported that HDL3 was taken up and degraded
by BeWo cells in a time- and concentration-dependent fashion, but the rate of degradation was considerably less than was the rate of degradation of LDL.HDL serves as
a physiological cholesterol/CE-carrier during reverse cho-lesterol transport from peripheral tissue to the liver and/or steroidogenic tissues.As HDL may exert biological action
in human trophoblast cells [50] we have investigated binding, cell association, and holoparticle association of HDL3as well as selective CE-uptake from HDL3in three different choriocarcinoma cells lines.Binding of 125 I-labeled HDL3 (at 4C) to BeWo, JAr, and Jeg3 cells studied here revealed Kd and bmax values (Table 1) comparable to those when binding of HDL3 to tropho-blast membrane protein fractions [3,51] or to intact first trimester trophoblasts was investigated [52].However, the pronounced capability for lipid tracer uptake from CE-labeled HDL3 in excess of holoparticle association in JAr and Jeg3 (but not in BeWo) cells was indicative for SR-BI-mediated selective CE-uptake from HDL3in these cell lines.SR-BI has been identified in human placental tissues by Northern blot experiments [14].Initial attempts
to explore a possible role of SR-BI on murine trophoblast cells have involved immunofluorescence microscopy to define the temporal and spatial pattern of SR-BI expres-sion during rodent embryogenesis [16].Here, we provide evidence that SR-BI is highly expressed on human trophoblast-like choriocarcinoma cell lines JAr and Jeg3 and we further propose that HDL3does act as cholesterol delivery vehicle to these cells via the SR-BI-mediated
Fig 3 Identification of SR-BI on mRNA and protein level in human
choriocarcinoma cells (A) RT-PCR analysis.cDNA for SR-BI was
amplified using specific forward and reverse oligonucleotide primers as
described in Materials and methods.The PCR product was separated
on a 1.5% agarose gel: lane 1 (BeWo), lane 2 (JAr), lane 3 (Jeg3), lane 4
(bp standard).(B) Northern blot analysis.Total RNA (15 lg) was
subjected to agarose gel electrophoresis and hybridized using the
553 bp fragment as a probe: lane 1 (BeWo), lane 2 (JAr), lane 3 (Jeg3),
lane 4 (total liver tissue).The blot was then stripped and reprobed by
using a fragment from the human glyceraldehyde 3-phosohate
dehy-drogenase cDNA (Clontech).Conventional Northern blotting did not
distinguish between SR-BI and SR-BII isoforms [27].(C) Western blot
analysis.Immunoblotting experiments of detergent solubilized
mem-brane protein fractions (lane 1–3, 50 lg protein; lane 4, 10 lg protein)
were separated on 8% SDS/PAGE.Immunoreactive bands were
visualized with rabbit anti-(SR-BI peptide) IgG (1 : 1500 dilution),
peroxidase-conjugated goat anti-rabbit IgG, and the ECL-detection
system.Arrow indicates the 84-kDa position of SR-BI: lane 1 (BeWo),
lane 2 (JAr), lane 3 (Jeg3), lane 4 [ldlA(mSR-BI)].
Trang 8pathway.The fact that the C-terminal cytosolic domain
of SR-BI which is a critical domain involved in selective
CE-uptake [29,53] is lacking in SR-BII and the lack of
SR-BII protein in all three choriocarcinoma cell lines
investigated here suggested that this receptor does not play
a specific role for CE-supply during fetal development
SR-BI may probably exert more profound capabilities
than the common LDL-receptor pathway mediating
holo-particle-uptake of apolipoprotein B- or E-containing
lipo-protein particles.This must be seen in light of the fact that
expression of LDL-receptor mRNA decreases from first
trimester to term tissues [54] and thus SR-BI could provide
an alternative pathway for cholesterol/CE-supply during
fetal development.As SR-BI is critical for cholesterol
transport its regulation has been addressed in a number of
placental tissues involved in steroidogenesis [55–57].In
humans, lacking functional LDL-receptor, fetal
develop-ment and cholesterol supply is normal compared to
controls.Also in mice with a targeted deletion of the
LDL-receptor gene, the up-regulation of selective
CE-uptake suggests that SR-BI can compensate for the loss of
LDL-receptor function [58].Recently, the LDL-receptor
related protein has also been addressed to mediate, at least
in part, selective lipid uptake [59].LDL-receptor related
pro-tein was found to be associated with syncytiotrophoblasts
(but not cytotrophoblasts) and BeWo cells [10].However, cAMP-treatment decreases expression of LDL-receptor related protein [10] but increases expression of SR-BI in parallel (Fig.7 [13])
Fig 4 Immunocytochemical evidence for SR-BI on choriocarcinoma cells (A) BeWo, (B) Jar, (C) Jeg3, (D) ldlA(mSR-BI) and (E) ldlA7 cells were cultured on Laboratory-Tek chamber slides as described in Materials and methods.Labeling was peformed with poly-clonal rabbit anti-(SR-BI) IgG (dilution of
1 : 1000) followed by goat anti-rabbit
cyanine-3 secondary antibody (dilution 1 : 500).
Table 2 Effect of HDL and ox–LDL on cell association of [3
H]CE-HDL at 37 °C Choriocarcinoma cells were incubated for 4 h at 37 C
with 10 lgÆmL)1of [3H]CE-labeled HDL 3 in the absence (control) or
presence of 100 lg of proteinÆmL)1of unlabeled competitor (HDL or
ox-LDL) in F12K (BeWo) or DMEM (JAr and Jeg3).The 100% value
for cell association of [3H]CE-HDL was 105 ± 8 ng (BeWo),
255 ± 17 (JAr) and 286 ± 21 ng HDL 3 -protein/mg cell protein
(Jeg3).The cell association is expressed as the percentage of the
radio-activity measured in the absence (100%) or presence of competitor.
Results are given as means ± SD of three independent experiments.
[ 3 H]CE-HDL Control HDL ox-LDL
BeWo 100 43.5 ± 6.6 37.6 ± 5.7
JAr 100 70.5 ± 9.5 21.2 ± 3.2
Jeg3 100 79.4 ± 8.7 28.4 ± 4.3
Fig 5 Effect of 8–CPTcAMP on cell association of [3H]CE-HDL and mRNA SR-BI expression in choriocarcinoma cells Following preincu-bation in F12K and DMEM without FBS (12 h) in the absence (open squares) or presence (full squares) of 0.3 m M 8-CPT-cAMP, chorio-carcinoma cells were incubated for 4 h with 10 lgÆmL)1of [ 3 H]CE-labeled HDL 3 at 37 C.The washed cells were lysed with 0.3 M NaOH
to determine the cell-associated fraction.To determine the in vitro effect of 8-CPT-cAMP on SR-BI mRNA expression, choriocarcinoma cells were incubated as described above.Total RNA was isolated, and Northern blot analysis was performed using radiolabeled SR-BI cDNA probe for each cell line (top panel) in the absence (lane 1, 3, 5)
or presence of 0.3 m M 8-CPT-cAMP (lane 2, 4, 6).Radiolabeled cDNA for GAPDH was used to monitor RNA loading (bottom panel).Results are given as means ± SD of three independent experiments.
Trang 9Different lines of evidence are of support that SR-BI is a
physiologically relevant HDL receptor, studies which are
primarily addressing its role in cholesterol metabolism
[12,13].Further studies demonstrated that SR-BI-deficiency
in rodents leads to defective erythroid maturation and
abnormalities in the female reproductive system [13,60]
Evidence is accumulating that cholesterol must be
consid-ered as an essential agent in embryonic development
Histochemical analysis of ovaries from superovulated
females showed reduced oil red O staining of lipids in the
ovarian copora lutea of SR-BI knock out relative to those of
wild-type animals suggesting reduced CE storage as a source
for steroid hormone production [61].Also plasma
pro-gesterone levels between pseudopregnant controls and
knock out females 6 days after mating were slightly
although not significantly impaired.Finally, the majority
of embryos from SR-BI knock out females at harvesting
showed abnormal, nonrefractile morphology of oocytes and
embryos [61] similar to wild-type females that had been
treated with cholesterol-binding drugs that can perturb
membrane structure.Female mice with a targeted null
mutation of the SR-BI gene are infertile [61], a fact
underscoring the importance of the SR-BI pathway during
embryonic development.As SR-BI is expressed on the
maternal–fetal interface [16], this could indicate defective
cholesterol/CE-supply to the growing embryo
From our studies we propose that some choriocarcinoma
cell lines can be used as a suitable model to mimick
Fig 6 Transfection of choriocarcinoma cells Untransfected (1, 4,
and 7), mock-transfected (b-gal-transfected, 2, 5, and 8), and
SR-BI-transfected (3, 6, and 9) BeWo (1–3), JAr (4–6), and Jeg3 cells (7–9)
were incubated for 4 h with 10 lgÆmL)1protein of [3H]CE-HDL 3 at
37 C as described in Fig.5.Thereafter, the incubation medium was
removed, the cells were washed and analyzed for cell-associated
radioactivity of [3H]CE-HDL 3 Results are given as means ± SD of
three independent experiments.Immunoblot experiments of detergent
solubilized membrane protein fractions (50 lg protein per lane) of
untransfected (1, 4, and 7) and SR-BI-transfected cells (3, 6, and 9)
were performed as described in Fig.3C.
Fig 7 Effect of exogenous HDL on cellular growth rates and choles-terol content Cells were seeded in 6-well trays and incubated in F12K (BeWo) and DMEM (JAr and Jeg3) containing 10% deficient serum (open symbols) or in medium containing lipoprotein-deficient serum and 100 lgÆmL)1 HDL 3 (closed symbols).At the indicated time points the cells were washed with HBBS, trypsin-treated and the cell number was counted.The cellular cholesterol content of lysed cells was analysed with enzymatic cholesterol reagent.Data shown represent means ± SD of three independent experiments.(A) BeWo, (B) JAr, (C) Jeg3.
Trang 10trophoblast–lipoprotein interactions in vitro.In BeWo the
LDL-receptor seems to be the predominant pathway
supplying these cells with cholesterol/CE via
holoparticle-uptake [10].The fact that BeWo, a cell line with many
characteristics of cytotrophoblast, are almost lacking SR-BI
and SR-BI-mediated selective CE-uptake from HDL3,
supports the assumption that JAr and Jeg3 resembling
syncytiotrophoblasts-like properties are suitable in vitro
models to study directed cholesterol/CE-transport in
polar-ized cells.Invasiveness of trophoblast cells in vivo and
choriocarcinoma cells in vitro apparently is linked to
cell-differentiation and proliferation [47] similar to that reported
for other tumors [62,63]
In line with previous findings using various
differenti-ation-modulating agents [47,48] we also observed
pro-nounced differences between BeWo and JAr/Jeg3 cells
regarding proliferation rates.The fact that exogenous
HDL3remarkably increases proliferation in JAr and Jeg3
supports a receptor-mediated specific and selective
CE-uptake and cellular growth in parallel.The results of this
study demonstrate that JAr and Jeg3 cells, replicating
trophoblast cells derived from a malignant tumor are
appropriate cellular systems to mimick cholesterol supply
from maternal lipoproteins to developing embryonic tissues
via SR-BI-mediated CE-uptake from HDL3
Acknowledgements
The authors thank Dr Vardon (local blood bank) providing human
plasma.This work was supported by the Austrian Science Fund (FWF,
15404 to E.M.and SFB 007–716 to W.S.) and the Austrian National
Bank OENB 8840 (to A H.), and 8778 and 9962 (to E M.).
References
1.Woollett, L A.(2001) The origins and roles of cholesterol and fatty
acids in the fetus Curr Opin Lipidol 12, 305–312.
2 Alsat, E , Bouali, Y , Goldstein, S , Malassine, A , Berthelier, M ,
Mondon, F.& Cedard, L.(1984) Low-density lipoprotein binding
sites in the microvillous membranes of human placenta at different
stages of gestation Mol Cell Endocrinol 38, 197–203.
3 Malassine, A , Besse, C , Roche, A , Alsat, E , Rebourcet, R ,
Mondon, F.& Cedard, L.(1987) Ultrastructural visualization of
the internalization of low density lipoprotein by human placental
cells Histochemistry 87, 457–464.
4 Cummings, S.W., Hatley, W., Simpson, E.R & Ohashi, M (1982)
The binding of high and low density lipoproteins to human
placental membrane fractions J Clin Endocrinol Metab 54,
903–908.
5.Woollett, L A.(1996) Origin of cholesterol in the fetal golden
Syrian hamster: contribution of de novo sterol synthesis and
maternal-derived lipoprotein cholesterol J Lipid Res 37,
1246–1257.
6 Farese, R.V Jr, Cases, S., Ruland, S.L., Kayden, H.J., Wong, J.S.,
Young, S G.& Hamilton, R L.(1996) A novel function for
apolipoprotein B: lipoprotein synthesis in the yolk sac is critical
for maternal-fetal lipid transport in mice J Lipid Res 37,
347–360.
7 McConihay, J.A., Honkomp, A.M., Granholm, N.A & Woollett,
L.A (2000) Maternal high density lipoproteins affect fetal mass
and extra-embryonic fetal tissue sterol metabolism in the mouse.
J Lipid Res 41, 424–432.
8 Wyne, K.L & Woollett, L.A (1998) Transport of maternal LDL
and HDL to the fetal membranes and placenta of the Golden
Syrian hamster is mediated by dependent and receptor-independent processes J Lipid Res 39, 518–530.
9 Moestrup, S.K & Verroust, P.J (2001) Megalin- and cubilin-mediated endocytosis of protein-bound vitamins, lipids, and hor-mones in polarized epithelia Annu Rev Nutr 21, 407–428.
10 Gafvels, M.E., Coukos, G., Sayegh, R., Coutifaris, C., Strickland, D.K & Strauss, J.F., 3rd (1992) Regulated expression of the trophoblast alpha 2-macroglobulin receptor/low density lipopro-tein receptor-related prolipopro-tein.Differentiation and cAMP modulate protein and mRNA levels J Biol Chem 267, 21230–21234.
11 Wittmaack, F.M., Gafvels, M.E., Bronner, M., Matsuo, H., McCrae, K R , Tomaszewski, J E , Robinson, S L , Strickland, D.K & Strauss, J.F., 3rd (1995) Localization and regulation of the human very low density lipoprotein/apolipoprotein-E recep-tor: trophoblast expression predicts a role for the receptor in placental lipid transport Endocrinology 136, 340–348.
12.Krieger, M.(1999) Charting the fate of the good cholesterol: identification and characterization of the high-density lipoprotein receptor SR-BI Annu Rev Biochem 68, 523–558.
13.Krieger, M.(2001) Scavenger receptor class B type I is a multi-ligand HDL receptor that influences diverse physiologic systems.
J Clin Invest 108, 793–797.
14 Cao, G , Garcia, C K , Wyne, K L , Schultz, R A , Parker, K L & Hobbs, H.H (1997) Structure and localization of the human gene encoding SR-BI/CLA-1.Evidence for transcriptional control by steroidogenic factor 1 J Biol Chem 272, 33068–33076.
15 Cao, G., Zhao, L., Stangl, H., Hasegawa, T., Richardson, J.A., Parker, K.L.& Hobbs, H.H.(1999) Developmental and hormonal regulation of murine scavenger receptor, class B, type 1 Mol Endocrinol 13, 1460–1473.
16 Hatzopoulos, A K , Rigotti, A , Rosenberg, R D & Krieger, M (1998) Temporal and spatial pattern of expression of the HDL receptor SR-BI during murine embryogenesis J Lipid Res 39, 495–508.
17 Alsat, E.& Malassine, A.(1991) High density lipoprotein inter-action with human placenta: biochemical and ultrastructural characterization of binding to microvillous receptor and lack of internalization Mol Cell Endocrinol 77, 97–108.
18 Benirschke, K.& Kaufmann, P.(2000) Pathology of the Human Placenta (Benirschke, K.& Kaufmann, P., eds), Springer Verlag, New York: USA, 4th edn.
19.Gru¨mmer, R., Hohn, H.P., Mareel, M.M & Denker, H.W (1994) Adhesion and invasion of three human choriocarcinoma cell lines into human endometrium in a three-dimensional organ culture system Placenta 15, 411–429.
20 Simpson, E R , Porter, J C , Milewich, L , Bilheimer, D W & MacDonald, P.C (1978) Regulation by plasma lipoproteins of progesterone biosynthesis and 3-hydroxy-3-methyl glutaryl coen-zyme a reductase activity in cultured human choriocarcinoma cells J Clin Endocrinol Metab 47, 1099–1105.
21 Simpson, E.R., Bilheimer, D.W., MacDonald, P.C & Porter, J.C (1979) Uptake and degradation of plasma lipoproteins by human choriocarcinoma cells in culture Endocrinology 104, 8–16.
22 Aplin, J.D., Sattar, A & Mould, A.P (1992) Variant cho-riocarcinoma (BeWo) cells that differ in adhesion and migration
on fibronectin display conserved patterns of integrin expression.
J Cell Sci 103, 435–444.
23 Pattillo, R.A., Gey, G.O., Delfs, E & Mattingly, R.F (1968) Human hormone production in vitro Science 159, 1467–1469.
24 Speeg, K.V Jr, Azizkhan, J.C & Stromberg, K (1976) The sti-mulation by methotrexate of human chorionic gonadotropin and placental alkaline phosphatase in cultured choriocarcinoma cells Cancer Res 36, 4570–4576.
25 White, T E , Saltzman, R A , Di Sant’Agnese, P A , Keng, P C , Sutherland, R.M.& Miller, R.K.(1988) Human choriocarcinoma (JAR) cells grown as multicellular spheroids Placenta 9, 583–598.