Download Prostate-Specific Antigen Synthesis and Secretion by Human Placenta

0021-972X/00/$03.00/0
The Journal of Clinical Endocrinology & Metabolism
Copyright © 2000 by The Endocrine Society
Vol. 85, No. 1
Printed in U.S.A.
Prostate-Specific Antigen Synthesis and Secretion by
Human Placenta: A Physiological Kallikrein Source
during Pregnancy*
MANUELA MALATESTA, FERDINANDO MANNELLO, FRANCESCA LUCHETTI,
FRANCESCO MARCHEGGIANI, LEONE CONDEMI, STEFANO PAPA, AND
GIANCARLO GAZZANELLI
Istituto di Istologia and Analisi di Laboratorio (M.M., F.M., F.M., G.G.) and Istituto di Scienze
Morfologiche (F.L., S.P.), Facoltà di Scienze MFN, Università degli Studi di Urbino, and Divisione di
Ginecologia and Ostetricia, Ospedale Civile (L.C.), 61029 Urbino, Italy
ABSTRACT
Prostate-specific antigen (PSA), a kallikrein-like serine protease
until recently thought to be prostate specific, has been demonstrated
in various nonprostatic tissues and body fluids. PSA has been also
found in human endometrium and amniotic fluids, even if the significance of this novel expression is unclear. In this study, we have
demonstrated by multiple techniques that human placental tissue,
obtained at delivery from normal full-term pregnancies, synthesizes
and secretes PSA. RT-PCR showed the presence of PSA messenger
ribonucleic acid; biochemical, chromatographic, and immunological
studies revealed the expression of both free and complexed PSA forms;
immunoelectron microscopy indicated the syncytiotrophoblast as the
site of PSA synthesis and secretion. Moreover, in vitro experiments
demonstrated that PSA production and secretion are up-regulated by
17b-estradiol, a pregnancy-related steroid hormone. These results
suggest that human placenta is a source of the PSA present in amniotic fluid and maternal serum during pregnancy. (J Clin Endocrinol
Metab 85: 317–321, 2000)
T
HE KALLIKREINS (KLK) are a family of serine proteases involved in the posttranslational processing of
polypeptides to their bioactive or inactive forms. They are
encoded by a multigene family, of which only three members
have been characterized to date in the human: KLK 1, encoding the true glandular kallikrein; and KLK 2 and KLK 3,
two genes primarily expressed in the prostate that encode for
prostate-specific antigen (PSA) (1). PSA is a serine protease
with chymotryptic-like activity (2) that until recently has
been thought to be exclusively produced by epithelial cells
of the prostate gland and then used as a marker for the
diagnosis and management of prostate cancer (3). However,
several studies have recently demonstrated the widespread
distribution of PSA in a variety of human normal and tumoral tissues, cell lines, and biological fluids (4). Although
the physiological role and the biological significance of extraprostatic PSA are currently unknown, it has been suggested that this serine protease can be regarded as a growth
factor regulator produced by cells bearing steroid hormone
receptors (5).
PSA immunoreactivity has been also revealed in normal
and pathological amniotic fluids, with varying content in
relation to gestational age: biologically and immunologically, PSA found in amniotic fluids was identical to the pros-
tate KLK-like serine protease, but its physiological function
and biological origin have not been yet clarified (6 – 8). A
study has also demonstrated that the PSA gene is expressed
in human normal cycling endometrium (9), suggesting the
presence of a local KLK-kinin system in this hormone-responsive tissue. Moreover, preliminary results showing the
presence of PSA protein in human at term placenta have been
recently reported (10).
The present study provides biochemical, molecular, and
immunocytochemical evidence for the synthesis and secretion of PSA by human placental tissue.
Materials and Methods
Samples
Seven fresh human placentas were collected from women (aged 25–38
yr) undergoing normal, full-term pregnancies (40 6 2 weeks) immediately after delivery. After the membranes were stripped, each placenta
was immediately processed for the different analyses. Blood was also
drawn from healthy control women (n 5 15) and pregnant women (n 5
7), and after blood clotting, the samples were centrifuged at 500 3 g for
10 min, and sera were stored at 230 C until assay (,2 weeks). The
subjects gave informed consent to the study, which was performed in
accordance with the ethical standards of Helsinki Declaration of 1975,
as revised in 1983.
PSA and protein measurements
Received January 6, 1999. Revision received July 6, 1999. Rerevision
received September 3, 1999. Accepted September 14, 1999.
Address all correspondence and requests for reprints to: Dr. Ferdinando Mannello, Istituto di Istologia ed Analisi di Laboratorio, Università degli Studi, Facoltà di Scienze MFN, Via E. Zeppi, 61029 Urbino
(PU), Italy. E-mail: [email protected].
* This work was supported in part by a grant from the Assessorato
alla Sanità, Regione Marche, Italy.
Fragments of placenta were homogenized and then sonified on ice
according to the procedure described previously (10). The lysates were
centrifuged at 9150 3 g at 4 C for 30 min, after which the supernatants
were immediately stored at 280 C until analysis (,2 weeks). The total
protein content was determined with the bicinchoninic method, using
a commercially available kit (Bio-Rad Laboratories, Inc., Hercules, CA).
Free and total PSA concentrations were determined in serum and cytosolic extracts of placenta with the AxSYM PSA assay (Abbott Labo-
317
318
JCE & M • 2000
Vol 85 • No 1
MALATESTA ET AL.
ratories, Abbott Park, IL) (11, 12). PSA immunoreactivity, determined for
a minimum of three concentrations at least in triplicate, was expressed
as micrograms per L. The detection limits of the test were 0.02 and 0.01
mg/L for total and free PSA, respectively. Placental extracts were serially
diluted in PSA-negative female serum and reanalyzed for the response
linearity to exclude the possibility that the detection of nonprostatic PSA
was due to a matrix effect. Analytical recovery of two concentrations (6.5
and 13 mg/L) of purified PSA (Sigma, St. Louis, MO) was performed as
previously detailed (13).
Immunogram and Western blotting
Sample components were separated on a 600 3 9-mm column of
Sephacryl S-300 (Pharmacia Biotech, Uppsala, Sweden). The samples
were applied to the column and eluted with 0.05 mol/L Tris-HCl buffer,
pH 7.5. Fractions of 0.5 mL each were collected and analyzed for PSA
content. Our Western blotting protocol was followed throughout, using
an antihuman PSA monoclonal mouse antibody (DAKO Corp., Milan,
Italy) (11, 12). PSA from culture supernatant of LNCaP, a human prostate
carcinoma cell line that constitutively secretes PSA (1), was used as a
positive control.
Immunoelectron microscopy
Immediately after labor, fragments of placental tissue were fixed by
immersion in a mixture of 4% paraformaldehyde and 0.5% glutaraldehyde in 0.1 mol/L Sörensen phosphate buffer, pH 7.4, at 4 C for 2 h and
then dehydrated and embedded in LRWhite resin (10). Ultrathin sections were processed for immunocytochemistry using a rabbit polyclonal antihuman PSA antibody (Biomeda, Foster City, CA) and a secondary gold-conjugated antibody (Jackson ImmunoResearch
Laboratories, Inc., West Grove, PA) following our previously reported
protocol (12). As controls, some sections were treated in the absence of
anti-PSA antibody.
Extraction of ribonucleic acid (RNA) and RT-PCR
Total RNA from placental tissue, collected immediately after labor,
and from LNCaP cells was extracted using a commercial reagent, RNAFast (Promega Corp., Madison, WI), according to the manufacturer’s
recommendations. Total RNA (5 mg) underwent RT for synthesis of the
first strand of complementary DNA (cDNA), using 1 mmol/L deoxynucleoside triphosphates, 10 mmol/L dithiothreitol, and 200 U SuperScript II reverse transcriptase (Life Technologies, Inc., Gaithersburg,
MD). The reaction was performed at 42 C for 1 h, followed by a denaturation step for 5 min at 95 C. Amplification of the cDNA was performed as previously described (12). An initial denaturation step (95 C
for 2 min) was followed by 40 cycles (94 C for 50 s, 61 C for 50 s, and
72 C for 90 s) and a final extension for 10 min. The PSA was amplified
in 45 mL of a PCR mixture containing 1 3 PCR buffer, 1.5 mmol/L
magnesium chloride, 200 nmol/L of each primer, 200 mmol/L deoxynucleoside triphosphates, and 2.5 U AmpliTaq DNA polymerase
(Promega Corp.). Ten microliters of each PCR reaction were electrophoresed on 1.5% agarose gels and visualized by ethidium bromide
staining under a UV light source. The new PSA primer sequences,
designed on the basis of sequence data obtained from the European
Molecular Biology Gene Bank and used to avoid amplification of the
highly homologous human glandular kallikrein gene (14), were as follows: PSA E-S, 59-CTCTCGTGGCAGGGCAGT-39 (exon 2); and PSA
AE-S, 59-CCCCTGTCCAGCGTCCAG-39 (exon 4). The predicted PSA
primer-amplified product was 485 bp in size. For placental tissue, LNCaP cells, and negative control samples, messenger RNA extraction and
cDNA amplification were carried out one sample at a time to avoid
cross-contamination.
were used, because phenol red has weak estrogenic activity (15),
whereas charcoal-stripped FBS is devoid of any steroid hormones. The
cells were also grown in serum-free medium in the presence of 17bestradiol (Schering AG, Berlin, Germany). Stimulation was initiated by
adding 1027 mol/L steroidal compound dissolved in absolute ethanol
and incubating the explants for up to 7 days. Tissue culture supernatants
were removed for PSA analysis on days 1, 3, 5, and 7. The secretion index,
defined as the secreted PSA divided by the cell-associated PSA, was
expressed as a percentage. The media removed at the end of each day
were centrifuged at 3000 3 g for 15 min at 4 C and stored at 230 C until
assay. At the end of the culture period, tissue explants were homogenized in lysis buffer, as previously described (11). LNCaP (American
Type Culture Collection, Manassas, VA) were grown in RPMI 1640
containing 10% FCS, 2 U/L penicillin, 200 mg/L streptomycin, and 5
mmol/L glutamine.
Statistical analyses
Statistical analysis of results, reported as the mean 6 se of at least
three independent experiments, was performed with the StatView 4
package (Abacus Concepts, Berkeley, CA), using a Macintosh PB (Apple
Computer, Cupertino, CA).
Results
The average serum PSA content of the healthy control
women examined (n 5 15) was 0.03 6 0.01, vs. 0.15 6 0.05
mg/L in serum from pregnant women (n 5 7; P , 0.0008).
The mean concentration of total PSA in placental tissues (n 5
7) was 57 6 9 mg/L, with about 30% in the free noncomplexed
form (17.31 6 2.64 mg/L). The dilution studies revealed a
good linearity (n 5 7; r2 5 0.98), demonstrating that placental
matrix (i.e. lipids, hemoglobin, hormones, and proteins) did
not affect the performance of PSA assay specific for serum
samples. The immunoenzymometric tests revealed that more
than 70% of the total PSA in placenta was in a bound form,
and the remainder was free uncomplexed protease; these
data were also confirmed by the elution chromatographic
Culture of explants
Placental tissue was dissected, rinsed with Earle’s Balanced Salt Solution, cultured in 25-cm2 tissue culture flasks, and maintained up to 7
days at 37 C in a 5% CO2 incubator with RPMI 1640 phenol red-free
medium containing 20 mmol/L HEPES, 10% charcoal-stripped FBS, 2
mmol/L l-glutamine, nonessential amino acids, 1% antibiotic-antimycotic solution, and 0.075% NaHCO3 (Sigma). The phenol red-free media
FIG. 1. Elution profile of placental PSA immunogram from the
Sephacryl S-300 column. The positions of the molecular mass markers
are indicated at the top: IgG (158 kDa), BSA (66.2 kDa), and bovine
carbonic anhydrase (31 kDa). Western blot analysis of PSA in placenta (8 ng) is reported in the inset.
PLACENTA AS SOURCE OF PSA DURING PREGNANCY
profile, revealing the highest immunoreactivity in the fraction where the PSA complex with a1-antichymotrypsin
(;100 kDa) was expected (Fig. 1). The immunoreactivity of
the free PSA form was also found in fractions 56 – 67 (molecular mass, 35– 40 kDa); these data were confirmed by
Western blotting analysis performed with a specific monoclonal antibody (Fig. 1).
The electron microscopic examination of immunolabeled
samples of placenta revealed that most PSA labeling occurred in epithelial cells (syncytiotrophoblast) coating the
villous surface (Fig. 2). These cells showed a diffuse cytoplasmic signal, frequently occurring on rough endoplasmic
reticulum cisternae (Fig. 2a); moreover, the labeling appeared concentrated in the apical region, in particular in the
microvilli and the cytoplasmic layer just beneath them (Fig.
2b). Some labeling was also observed in the basal region of
these cells, in the numerous cytoplasmic protrusions spreading out in the connective matrix (Fig. 2c). This matrix as well
as the fibroblasts distributed therein showed a weak signal
(not shown). No significant labeling was observed in cell
nuclei. Control samples were virtually unlabeled.
As shown in Fig. 3, ethidium bromide-stained agarose gel
electrophoresis demonstrated that placental tissue as well as
319
LNCaP cells, which are both positive for PSA immunoreactivity, produced the expected 485-bp transcript.
The cellular extracts of placental explants cultured in vitro
showed a total PSA content of 5.47 6 0.07 mg/mg protein,
1.53 6 0.06 mg of which was in the free PSA form. The
FIG. 3. Ethidium bromide-stained agarose gel of RT-PCR products of
PSA messenger RNA isolated from normal human placental tissue
and LNCaP cells. MW, Molecular weight markers expressed in base
pairs. Lane 1, LNCaP cells (as positive control); lane 2, placental
tissue; lane 3, negative control.
FIG. 2. Immunocytochemical localization of PSA in the syncytiotrophoblast. a, PSA labeling is located in the RER cisternae (arrowheads) as
well as free in the cytoplasm (arrows). The nucleus (N) is almost devoid of labeling. b, In the apical region of the syncytiotrophoblast, strong
labeling is present in microvilli (arrows). c, The basal region of the syncytiotrophoblast shows specific labeling in the cytoplasmic protrusions
(arrows) spreading out in the connective tissue (C). The connective matrix displays a weak signal (arrowhead). The bars represent 0.25 mm.
320
JCE & M • 2000
Vol 85 • No 1
MALATESTA ET AL.
explants secreted in culture medium an immunoreactive
PSA protein with a steady increase in rate; the average daily
secretion in medium was 0.10 6 0.01 mg/L, and the total
amount of PSA released over the 7-day period represented
about 15% of the amount in the tissue explant (Table 1).
Addition of 17b-estradiol to the cultured placental explants
resulted in a significant stimulation of PSA production (P ,
0.001; Fig. 4).
Discussion
PSA, until recently considered a prostate-specific serine
protease (1), has been demonstrated to be a widespread
biochemical marker, regulated by several steroidal compounds (5). In particular, the PSA gene may be up-regulated
in physiological conditions associated with a steroid hormone overproduction (e.g. during pregnancy and in the endometrial cycle), suggesting new biological roles of PSA both
in fetal growth/development (6 – 8) and as a potential regulator of uterine function (9). However, the source of PSA
found in amniotic fluids and maternal serum has not yet been
clarified.
In a preliminary study, we detected PSA in human normal
placental tissue (10). In this report we demonstrated that
human normal placenta synthesizes and secretes PSA. The
molecular approach showed that the PSA gene is present and
functionally active, the biochemical analyses showed that
placental PSA occurs as both free and complexed molecules,
and immunocytochemistry revealed the syncytiotrophoblast
as the main responsible for placental PSA biosynthesis and
secretion. In particular, the ultrastructural results suggest
that the syncytiotrophoblast is a bipolar structure. PSA is
synthesized in the rough endoplasmic reticulum cisternae,
then transported and secreted as free/complexed molecules
mainly in the apical region and in a much lower amount in
the basal region. Moreover, our in vitro experiments not only
confirmed that placental cells actively produce and secrete
PSA, but also demonstrated that 17b-estradiol, a steroid hormone related to pregnancy, is able to up-regulate PSA secretion. A similar phenomenon has been observed in breast
cancer cell lines, when treated with several steroid compounds (15).
The expression of PSA in nonprostatic sources and, in
particular, in female tissues and fluids suggests new important biological roles of this serine protease, i.e. as a potential
sensitive biochemical/molecular marker of hormone responsiveness (5, 11, 15). The concomitant presence of the steroid
hormones and receptors in human placenta (16) and the
significant increase in PSA production under hormone stimulation (present study) suggest the possibility of placental
PSA modulation by steroid hormones (17).
Our results strongly support the hypothesis that placental
TABLE 1. PSA in culture media and decidua
Days
PSA in media
(mg/L)
Cell-associated
PSA/mg protein
Secretion index (%)
1
3
5
7
0.125 6 0.004
0.177 6 0.005
0.214 6 0.009
0.261 6 0.008
2.155 6 0.031
1.864 6 0.095
1.757 6 0.053
1.657 6 0.048
5.80
9.49
12.18
15.69
FIG. 4. Time course of PSA release by placental explants into culture
medium after 17b-estradiol stimulation (F). The negative control (E)
was treated only with absolute ethanol.
tissue represents a source of the PSA found in both maternal
serum and amniotic fluid during pregnancy (6 – 8). PSA in
these body fluids may play a role as a growth factor modulator and/or as a translational/posttranscriptional protein
regulator. In fact, PSA hydrolyzes the insulin chains and
interleukin-2 (2), enzymatically digests insulin-like growth
factor-binding proteins (18), activates latent transforming
growth factor (19), inactivates protein C inhibitors (20, 21),
and regulates the hormonal bioactivity of PTH-related protein (22, 23). The proteolytic activity of PSA on these different
biological substrates, all detected in placenta (24), could explain in part the novel potential role of PSA in this tissue, not
only as a sensitive molecular marker implicated in hormone
responsiveness but also as an initiator of the protease cascade, an important biological mechanism for tissue remodeling in the uterus (25, 26).
References
1. McCormack RT, Wang TJ, Rittenhouse HG, Wolfert RL, Finlay JA, Sokoloff
RL. 1995 Molecular forms of prostate-specific antigen and the human kallikrein
gene family: a new era. Urology. 45:729 –744.
2. Watt KWK, Lee PJ M, ’Timkulu T, Chan WP, Loor R. 1986 Human prostatespecific antigen: structural and functional similarity with serine proteases. Proc
Natl Acad Sci USA. 83:3166 –3170.
3. Chu MT. 1997 Prostate-specific antigen and early detection of prostate cancer.
Tumor Biol. 18:123–134.
4. Diamandis EP, Yu H. 1997 Nonprostatic sources of prostate-specific antigen.
Urol Clin North Am. 24:275–282.
5. Diamandis EP, Yu H. 1995 New biological functions of prostate-specific antigen. J Clin Endocrinol Metab. 80:1515–1517.
6. Yu H, Diamandis EP. 1995 Prostate-specific antigen immunoreactivity in
amniotic fluid. Clin Chem. 41:204 –210.
7. Melegos DN, Yu H, Allen LC, Diamandis EP. 1996 Prostate-specific antigen
in amniotic fluid of normal and abnormal pregnancies. Clin Biochem.
29:555–562.
8. Filella X, Molina R, Alcover J, Carretero P, Ballesta AM. 1996 Detection of
nonprostatic PSA in serum and nonserum samples from women. Int J Cancer.
68:424 – 427.
9. Clements J, Mukhtar A. 1994 Glandular kallikreins and prostate-specific antigen are expressed in the human endometrium. J Clin Endocrinol Metab.
78:1536 –1539.
10. Mannello F, Malatesta M, Fusco E, Bianchi G, Cardinali A, Gazzanelli G.
1998 Biochemical characterisation and immunolocalization of prostate-specific
antigen in human term placenta. Clin Chem. 44:1735–1738.
PLACENTA AS SOURCE OF PSA DURING PREGNANCY
11. Mannello F, Sebastiani M, Amati S, Gazzanelli G. 1997 Prostate-specific
antigen expression in a case of intracystic carcinoma of the breast: characterisation of immunoreactive protein and literature surveys. Clin Chem.
43:1448 –1454.
12. Mannello F, Malatesta M, Luchetti F, Papa S, Battistelli S, Gazzanelli G. 1999
Immunoreactivity, ultrastructural localization, and transcript expression of
prostate-specific antigen in human neuroblastoma cell lines. Clin Chem.
45:78 – 84.
13. Mannello F, Miragoli G, Bianchi G, Gazzanelli G. 1997 Immunoreactive
prostate-specific antigen in pleural effusions. Clin Chem. 43:847– 848.
14. Corey E, Arfman EW, Liu AY, Vessella RL. 1997 Improved reverse transcriptase-polymerase chain reaction protocol with exogenous internal competitive control for prostate-specific antigen mRNA in blood and bone marrow.
Clin Chem. 43:443– 452.
15. Zarghami N, Grass L, Diamandis EP. 1997 Steroid hormone regulation of
prostate-specific antigen gene expression in breast cancer. Br J Cancer.
75:579 –588.
16. Strauss JF, Martinez F, Kiriakidon M. 1996 Placental steroid hormone synthesis: unique features and unanswered questions. Biol Reprod. 54:303–311.
17. Petraglia F, de Micheroux AA, Florio P, et al. 1995 Steroid-protein interaction
in human placenta. J Steroid Biochem Mol Biol. 53:227–231.
18. Cohen P, Graves HCB, Peehl DM, Kamarei M, Giudice LC, Rosenfeld RC.
1992 Prostate-specific antigen is an insuline-like growth factor binding protein-3 protease found in seminal plasma. J Clin Endocrinol Metab.
75:1046 –1053.
321
19. Killian CS, Corral DA, Kawinsky E, Constantine RI. 1993 Mitogenic response
of osteoblast cells to prostate-specific antigen suggests an activation of latent
TGF-b and a proteolytic modulation of cell adhesion receptors. Biochem Biophys Res Commun. 92:940 –947.
20. Espana F, Gilabert J, Estelles A, Romeu A, Asmar J, Cabo A. 1991 Functionally
active protein C inhibitor/plasminogen activator inhibitor-3 is secreted in
seminal plasma and complexes with prostate-specific antigen. Thromb Res.
64:309 –320.
21. Kise H, Nishioka J, Kawamura L, Suzuki K. 1996 Characterization of semenogelins and its molecular interaction with prostate-specific antigen and
protein C inhibitor. Eur J Biochem. 238:88 –96.
22. Iwamura M, Hellman J, Cockett AT, Lilja H, Gershagen S. 1996 Alteration
of the hormonal bioactivity of parathyroid hormone-related protein as a results
of limited proteolysis by prostate-specific antigen. Urology. 48:317–325.
23. Cramer SD, Chen Z, Peehl DM. 1996 Prostate-specific antigen cleaves parathyroid hormone-related protein in the PTH-like domain: inactivation of
PTHrP-stimulated cAMP accumulation in mouse osteoblasts. J Urol.
156:526 –531.
24. Lala PK, Hamilton GS. 1996 Growth factors, proteases and protease inhibitors
in the maternal-fetal dialogue. Placenta. 17:545–555.
25. Salamonsen LA. 1994 Matrix metalloproteinases and endometrial remodelling. Cell Biol Int. 18:1139 –1144.
26. Clements J, Mukhtar A, Yan S, Holland A. 1997 Kallikreins and kinins in
physiologic events in the reproductive tract. Pharmacol Res. 35:537–540.