Placenta (2001), 22, 591–596

doi:10.1053/plac.2001.0688, available online at http://www.idealibrary.com on

Placental Calcification: A Metastatic Process?

S. H. Poggia,d, K. I. Bostromb, L. L. Demerb, H. C. Skinnerc and B. J. Koosa

a
Nicholas S. Assali Perinatal Research Laboratory, Departments of Obstetrics and Gynecology, University of California,
Los Angeles, b Department of Medicine, University of California, Los Angeles and c Department of Geology and Geophysics,
Yale University, New Haven, CT 065113, USA
Paper accepted 7 March 2001

Placental calcification commonly increases with gestational age. The mechanism of apatite mineralization probably involves one of
three known mechanisms of tissue calcification: physiological (like bone), dystrophic (ischaemia-related) or metastatic (mineral-
ization in a supersaturated environment). This study was designed to determine the mechanism of calcification by examining
(1) the mineral content of placental calcifications in comparison to other physiological and pathological apatites, and (2) the
expression of bone morphogenetic proteins (BMPs), which are important in physiological calcification, across gestational age. By
energy-dispersive x-ray analysis (EDXA), the Ca/P weight ratio for apatitic mineral from mature calcifications was 2.000.05
(s.e.), which is similar to that for stones formed in a metastatic, supersaturated environment and lower than that observed in
physiological calcification. Biologically active BMP, which was determined by bioassay, was demonstrated in mature and
postmature placentae. The BMPs PLAB, PDF and related protein INSL-4 were identified by semiquantitative reverse
transcriptase polymerase chain reaction (RT-PCR), but their mRNA expression was independent of gestational age (7–41 weeks
of gestation). We conclude that (1) the identified BMPs were not related directly to placental calcification, which argues against
physiological calcification, and (2) the chemical composition of the apatitic mineral was suggestive of rapid formation in a
supersaturated environment, which is consistent with a metastatic mechanism of calcification.
Placenta (2001), 22, 591–596  2001 Harcourt Publishers Ltd

INTRODUCTION Three processes may lead to tissue calcification (Anderson,

The calcium content of placentas normally increases with
gestational age, through a mineralization process that may be
accelerated by some disorders of pregnancy (Jeacock et al.,
1963). For example, calcifications are clearly visible in most
post-term (>41 weeks gestation) placentae and are often
observed in placentas associated with pre-eclampsia and/or
fetal growth restriction (Jeacock et al., 1963; Quinlan et al.,
1982). The mineral is found in the basal plate of the immature
placenta in association with extracellular fibrinoid (Benirschke
and Kaufman, 2000) and, later in gestation, in the fibrinoid
surrounding the basement membranes of villous endothelial
cells and in placental septae (Avery and Aterman, 1970; Varma
and Kim, 1985). Calcification occurs within extracellular
matrix vesicles of the placenta (Al-Zuhair et al., 1984; Varma
and Kim, 1985), and in that respect the calcification process
resembles the extracellular deposition associated with bone
formation. This provocative finding suggests that placenta and
bone may involve a similar mechanism of calcification.
d
To whom correspondence should be addressed at: 3PHC,
Department of Obstetrics and Gynecology, Georgetown University
School of Medicine, 3800 Reservoir Road, Washington, DC, 20007.
Tel.: +1 202 687 8801; Fax: +1 202 687 1757; E-mail:
poggis@gunsun.georgetown.edu
0143–4004/01/060591+06 $35.00/0
1983). In physiological calcification, as seen in bone and teeth,
osteoblasts produce osteoid matrix, which permits hydroxy-
apatite (HA) formation in and on collagen fibres (Anderson,
1989; Skinner, 1988). Secondly, dystrophic calcification occurs
in tissue necrosis. In this process, injured or dysfunctional
cellular membranes allow extracellular calcium to enter cells
and combine with phosphate, producing apatitic mineral
(Anderson, 1983). Finally, metastatic calcification develops in
an environment that is supersaturated with calcium and
phosphate or oxalate, such as occurs with urolith formation
(Anderson, 1983).
The calcium/phosphate (Ca/P) weight ratio of the mineral
varies according to mechanism of calcification. For example,
the Ca/P weight ratio is greater in mature bone than in the less
organized apatites that are associated with metastatic calcifica-
tion. Thus, the chemical composition of the mineral may help
distinguish a physiological mechanism (e.g. bone formation)
from a metastatic process.
Physiological and non-physiological calcification may also be
distinguished by determining the association of bone morpho-
genetic proteins (BMPs) and mineral. The formation and
maintenance of bone tissues involves BMPs, a class of growth
factors that are defined by their ability to induce ectopic bone
formation via the induction of osteogenic cell differentiation
 2001 Harcourt Publishers Ltd
592 Placenta (2001), Vol. 22

and the subsequent formation of lamellar bone and marrow

(Urist, 1965). In particular, BMP-2 appears to be essential for
the normal formation of bone and teeth (Riley et al., 1996). In
addition to bone, this physiological process involving BMPs
has been observed in atherosclerotic plaque (Bostrom et al.,
1993).
Several BMPs have been detected in human placentae,
including prostate-derived factor (PDF) (Hromas et al., 1997)
and placenta BMP (PLAB) (Paralkar et al., 1998). Insulin-like
growth factor (INSL-4), an inductive protein that is important
in bone and cartilage formation, has been also been detected in
immature placentae (Laurent et al., 1998). These proteins have
been implicated in embryogenesis, but they may also be
involved in other processes, including placental calcification. If
these BMPs are involved in placental calcification, their
expression might be expected to increase with gestational age
as the placentae become more calcified.
This study was designed to determine (1) the mineral
content of placental calcifications, and (2) the expression of
bone morphogenetic proteins across gestational age. The
results provide information on whether placental calcification
is formed passively in a metastatic or dystrophic environment Figure 1. Scanning photomicrograph showing calcified villi from a post-date
or actively propagated like the physiological process involved placenta. The apatitic mineral (white material) consists of a porous aggregate
in the formation of bone, teeth or atherosclerotic plaque. of tiny (1–5
) lamellae (magnification bar=200
m).

METHODS
The following procedures were approved by the UCLA
Medical Institutional Board and the Chancellor’s Animal
Research Committee.
Chemical analysis of placental mineral
Five heavily calcified term (37–41 weeks gestation) and post-
dates placentas were fixed in 10 per cent zinc formalin solution,
dehydrated using serial ethanol solutions, cleared with xylene,
and infiltrated with paraffin. Sections (6
) were mounted on
Thermanox coverslips (Nunc, Naperville, IL, USA) to avoid
any chemical artifacts, especially silica (Poggi et al., 1998). The
samples on the coverslips were treated with xylene to extract
paraffin. After carbon-coating, the energy dispersive JEOL
JXA-8600 electron microprobe was used to make the chemical
analyses using settings that have been previously described
(Poggi et al., 1998). Five to 10 replicate analyses were per-
formed on each of five placentae. The mineral weight ratios are
expressed as meanss.e.
Detection of BMP activity in vivo
BMP was isolated from heavily calcified term and postdate
placentas using guanidine hydrochloride extraction followed
by lyophilization (Kawamura and Urist, 1988). Gelatin cap-
sules containing the isolate (25–100 mg) were implanted in the
left and right vastus lateralis muscle of two nude mice and in
the left and right rectus abdominus muscle of three Sprague-
Dawley rats. Empty capsules were placed in two animals (one
mouse and one rat) as a control. The animals were euthanized
when radiographs showed radioopacities consistent with calci-
fication. The tissue was fixed with 4 per cent paraformalde-
hyde and stained with haematoxylin and eosin. Histologic
examination was performed using light microscopy (100 and
400 magnification). A pathologist, who was blinded to the
origin of the material, examined the slides for the presence
of bone.
RT-PCR
Placentae from 7–42 weeks’ gestation were snap-frozen
(
80C) immediately after delivery. Total ribonucleic acid
(RNA) was extracted using an RNA isolation kit (Statagene,
La Jolla, CA, USA). Reverse transcription of 3 mg total RNA
was carried out for 90 min at 37C in 50
T buffer, supple-
mented with 0.5  of each d NTP (Pharmacia, Piscataway, NJ,
USA), 50–80 units RNase Block (Stratagene), and 50 U of
M-MuLV reverse transcriptase (Stratagene).
Polymerase chain reaction (PCR) was performed using
specific primers for BMP-2, BMP-4, PLAB, PDF, and
INSL-4 (Gibco, Grand Island, NY, USA) in a volume of 10
,
using Pfu buffer, 7 U of native Pfu polymerase (Stratagene),
0.1
l [P32]--dCTP (Amersham), 25 ng of each primer, and
1
l of template (from a 50-
l RT reaction). Thermal cycling
was performed as follows: (1) initial denaturation at 96C for
Poggi et al.: Placental Calcification
Figure 2. Photomicrograph of a section of rat rectus muscle 42 days after implantation of placental extract. Osteoid, lacunae, osteoblasts (OB) and osteoclasts
(OC), which are consistent with immature bone, are visible next to normal muscle (M) (magnification bar=10
m).
one min, (2) cycling for cDNA-specific number of cycles
between 96C for 1 min, at 65C for 2 min and at 72C for
1 min, (3) final extension at 72C for 5 min. For semiquanti-
tative PCR, multiple PCR cycles were performed and the
number of cycles was chosen so that amplification remained
well within the linear range, as assessed by densitometry (NIH
Image J, version 1.08I, public domain program, http://
rsb.info.nih.gov/nih-image). An equal volume from each PCR
reaction was analyzed by 6 per cent non-denaturing poly-
acrylamide gel electrophoresis and dried gels were examined
using autoradiography.
RT-PCR trends were confirmed by Northern analysis,
which was performed as generally described by Tintut et al.
(1998). Ten
g of total RNA from placental tissues was used
for analysis that involved the use of the 5 -ended-labelled PCR
primers (non-coding sequences) as probes. Hybridization was
performed overnight at 42C, and the membranes were washed
twice at RT for 10 min with 5SSPE and once at 37C
for 10 min with 5SSPE with 0.1 per cent SDS before
autoradiography.
RESULTS
Energy dispersive spectroscopic analysis detected calcium and
phosphate as the only significant elements in the mineral of the
593
placentae. The mineral, which was very porous in appearance
(Figure 1), consisted of small (1–5
in diameter) lamellae that
formed larger mineral composites that ranged to over 200
in
diameter. In five term and post-term placentae, the Ca/P
weight ratio was 2.000.05. Mg +2, Na + , K + , and Cl
were
present, but in very low concentrations (0.140.04,
0.460.17, 0.040.01, 0.030.01 weight per cents,
respectively).
The presence of biologically active BMP in placental tissue
was determined by bioassay that involved placing encapsulated
GuHCl-extracted placental fragments in the musculature of
rodents. Subsequent microscopic examination of the tissue
(35–42 days after implantation) surrounding the site of implan-
tation showed osteoid, lacunae, osteoblasts and osteoclasts
(Figure 2). These findings, which are consistent with early
bone formation, indicate that biologically active BMPs are
present in term and postdate placentas. Implantation of gelatin
capsules alone failed to induce calcification.
RT-PCR determined whether the biologically-active BMPs
corresponded to the expression of previously described placen-
tal BMPs. The results show that PDF and PLAB mRNA is
expressed in all placentae analysed from 7 to 41 weeks of
gestation. The expression of these BMPs appeared to be
relatively constant over this broad range of gestational ages
(Figure 3). Genetic expression of INSL-4 tended to increase
594
Figure 3. Upper panel. Placental expression of PDF, PLAB, and INSL-4
after indicated number of weeks. Total RNA and cDNA was prepared and
PCR was performed using specific primers for each growth factor. GAPDH is
shown for comparison. Lower panel. Relative signal intensity of the three
growth factors after normalization to GAPDH. The results are expressed as
fold increases of expression. The signal corresponding to 7 weeks was in each
case regarded as one-fold.
Placenta (2001), Vol. 22
with gestational age. BMP-2 and the closely related BMP-4
were not detected in embryonic, preterm, term or post-term
placentas. The relatively constant expression of PLAB and
PDF, as well as the increasing trend in INSL-4, was confirmed
by Northern blot analysis.
DISCUSSION
The Ca/P weight ratio of 2.000.05 observed in human
placental calcification resembles that seen in human stones
(uroliths, sialoliths) that are formed in supersaturated environ-
ments (Table 1). The very rapid crystallization that occurs
under these conditions results in a less organized, porous
apatitic lattice consisting of small lamellar crystals with a lower
Ca/P weight ratio. In contrast, bone and atherosclerotic lesions
typically have Ca/P weight ratios that approximate 2.15, the
value for stoichiometrically ideal hydroxyapatite (Table 1).
The physiological calcification process, in which osteoid
mineralizes about 10 days after its initial secretion, allows the
crystal lattice to be propagated in an organized manner,
allowing formation of mineral in equilibrium (Skinner, 1988).
This finding contrasts with metastatic calcifications, where
very rapid mineralization results in a less organized and less
mature mineral. Less is known about the ‘typical’ calcification
in dystrophic calcifications, which appear to have a Ca/P
weight ratio above 2.15, which suggests that another phase
besides apatite may be present (Poggi et al., 1998). The lower
Ca/P ratios and porous, lamellar aggregates observed support
a non-physiological mechanism of placental calcification that
most closely resembles metastatic calcification.
Bone morphogenetic proteins, which are biologically active
even after extraction with the strong denaturant GuHCl, were
detected in heavily calcified, term placentae using the classic
in-vivo assay. RT-PCR and Northern analysis showed PLAB
and PDF, as well as INSL-4 expression of the previously
described placental BMPs. The relative mRNA expression of
the placental BMPs occurred virtually independent of gesta-
tional age, but BMP-2 and BMP-4, which have been suggested
to be involved in both bone and atherosclerotic plaque calcifi-
cation, were not detected. INSL-4 showed a slight trend
towards increased expression. These results do not support
the hypothesis that the increased placental calcification with
gestational age depended critically on greater expression of
BMPs. Thus, placental calcification does not appear to be a
physiological process.
These results do not exclude the involvement of a dys-
trophic process. Ca/P weight ratios in dystrophic calcification
have been reported only for calcified necrotic breast carcinoma.
Thus, the limited information concerning the mineral com-
position of dystrophic calcifications prevents firm comparative
conclusions.
The results must be interpreted with caution for other
reasons. For example, non-identified BMPs may be present
in placenta which promote calcification. In addition, BMP
inhibitors which have been previously described for other
Poggi et al.: Placental Calcification
Table 1. Calcium/Phosphate (Ca/P) weight ratios associated with tissue calcification
Tissue Mature Bonea Arteryb
Ca/P ratio 2.15 2.16
Mechanism Physiological Physiological
a
Skinner, 1988; b Schmid et al., 1980; c Poggi et al., 1998; d Kodaka et al., 1994.
BMPs (e.g., chordin, noggin and Gremlin) may be present
(Piccolo et al., 1996; Brunet et al., 1998; Hsu et al., 1998).
Altered expression of similar inhibitors to PLAB or PDF
would alter their biological activity, but would be removed by
the strong denaturant used prior to the in-vivo assay. Thus,
the results of the in-vivo assay may not reflect the true
biological activity of placental BMPs.
Calcium is actively transported across the placenta through-
out gestation, making the fetus relatively hypercalcemic
(Husain and Mughal, 1992; Pitkin, 1985). This action, which
supports the growing fetal skeleton, is mediated primarily by
fetal parathyroid related protein, which increases greatly
through pregnancy, rather than maternal parathyroid hormone
(Hosking, 1996). Calcium binding proteins are thought to
buffer the process, preventing disruption of intracellular pro-
cesses (Tuan, 1985). Growth-restricted fetuses often have
abnormalities in mineralization of bone, which may be
ACKNOWLEDGEMENTS
We wish to thank Dr Alan DeCherney for his generous support and Dr Scott Nelson for his review of the histology. This work was supported in part by National
Institutes of Health Grants HD-18478 and HL-30568.
REFERENCES
Al-Zuhair AG, Ibrahim ME & Mighal SE (1984) Calcium deposition on
the maternal surface of the human placenta: a scanning electron microscope
study. Arch Gynecol, 234, 167–172.
Anderson HC (1983) Calcific diseases: a concept. Arch Pathol Lab Med, 107,
341–348.
Anderson HC (1989) Mechanism of mineral formation in bone. Lab Invest,
60, 320–330.
Avery C & Aterman K (1970) Calcification of the basement membrane of
placental villi. J Path, 103, 199–200.
Bernischke K & Kaufman P (2000) Pathology of the human placenta.
New York: Springer-Verlag.
Bostrom K, Watson K, Horn S, Wortham IM & Demer L (1993) Bone
morphogenetic protein expression in human atherosclerotic lesions. J Clin
Invest, 91, 1800–1809.
Brunet LJ, McMahon AP & Harland RM (1988) Noggin, cartilage
morphogenesis and joint formation in the mammalian skeleton. Science, 28,
1455–1457.
Cho KW & Blizt IL (1998) BMPs, Smads and metalloproteases: extracellular
and intracellular modes of negative regulation. Curr Opin Genet Dev, 8,
443–449.
Hosking D (1996) Calcium homeostasis in pregnancy. Clin Endocrinol (Oxf),
45, 1–6.
Hromas S, Hufford M, Sutton J, Xu D, Li Y & Lu L (1997) PLAB a novel
placental bone morphogenetic protein. Biochim Biophys Acta, 1354, 40–44.
Hsu DR, Economides AN, Wang S, Eimon PM & Harland R (1998) The
Xenopus dorsalizing factor Gremlin identifies a novel family of secreted
proteins that antagonize BMP activities. Mol Cell, 1, 673–683.
Husain SM & Mughal MZ (1992) Mineral transport across the placenta.
Arch Dis Child, 67, 874–878.
595
Necrotic Tumourc Urolithd Placenta
2.440.08 1.980.04 2.000.05
Dystrophic Metastatic Metastatic?
observed in delayed appearance of ossification sites (Zilanti
et al., 1987). In such fetuses, calcium, which would otherwise
be used for bone formation, would likely pass back to the
maternal circulation via passive placental diffusion. Such a
blunting of net calcium flow to the fetus may create the
calcium-enriched environment that promotes the precipitation
of calcium and phosphate to form highly insoluble apatite in
the ‘placental sink’. An alteration in fetal calcium utilization
may also contribute to calcium deposition in mature placentae,
although it does not exclude other possibilities.
In summary, the mechanism of placental calcification
appears to be most consistent with a process that occurs in an
environment that is supersaturated with calcium and phos-
phate. A physiological, bone-like mechanism and a dystrophic
process appear less likely to be involved.
Jeacock M, Scott J & Plester J (1963) Calcium content of the human
placenta. Am J Obstet Gynecol, 87, 34–40.
Kawamura M & Urist MR (1988) Growth factors, mitogens, cytokines and
bone morphogenetic protein in induced chondrogenesis in tissue cultures.
Dev Bio, 130, 435–42.
Kodaka T, Debari K, Sano T & Yamado M (1994) Scanning electron
microscopy and energy-dispersive x-ray microanalysis studies of several
human calculi containing calcium phosphate crystals. Scanning Microscopy
International, 8, 241–257.
Laurent A, Rouillac C, Delezoide AL, Giovangrandi Y, Vekemans M,
Bellet D, Abitbol M & Vidaud M (1998) Insulin-like 4 gene expression
in human embryogenic and trophoblastic tissues. Mol Reprod Dev, 51,
123–129.
LeGeros R & Le Geros J (1984) Phosphate minerals in human tissues. In
Phosphate Minerals (Eds) Nriagu J & Moore P, pp. 351–385. Berlin:
Springer Verlag.
Paralkar M, Vail A, Grasser WA, Brown TA, Xu H, Vukicenvic S, Ketlz
H, Owen TA & Thompson DD (1998) Cloning and characterization of a
novel member of the transforming growth factor- /bone morphogenetic
protein family. J Biol Chem, 273, 13760–13767.
Piccolo S, Saisai Y, Lu B & De Robertis EM (1996) Dorsoventral
patterning in Xenopus: inhibition of ventral signals by direct binding of
chordin to BMP-4. Cell, 86, 589–598.
Pitkin R (1985) Calcium metabolism in pregnancy and the perinatal period: a
review. Am J Obstet Gynecol, 151, 99–109.
Poggi S, Skinner C, Ague J & Carter D (1998) Using scanning
electron microscopy to study mineral deposits in breast tissues. American
Mineralogist, 83, 1122–1126.
Quinlan R, Cruz A, Buhi & Martin M (1982) Changes in placental
ultrasonic appearance. I. Incidence of grade III changes in the placenta in
correlation to fetal pulmonary maturity. Am J Obstet Gynecol, 144, 471–473.
596
Riley E, Lane J, Urist M, Lyons K & Lieberman J (1996) Bone
morphogenetic protein-2. Clinical Orthopeaedics and related research, 324,
39–44.
Schmid K, McSharry W, Pameijer C & Binette J (1980) Chemical and
physiochemical studies on the mineral deposits of the human atherosclerotic
aorta. Atherosclerosis, 37, 199–210.
Skinner HC (1988) Mineral composition of bone. In The Scientific Basis of
Orthopedics (Eds) Albright J & Brand R. New York: Appleton and Lange.
Tintut Y, Parhami F, Boström K, Jackson SM & Demer LL (1998)
cAMP stimulates osteoblast-like differentiation of calcifying vascular cells.
J Biol Chem, 273, 7547–7553.
Placenta (2001), Vol. 22
Tuan RS (1985) Calcium-binding protein of the human placenta, charac-
terization, immunohistochemical localization and functional involvement in
Ca2+ transport. Biochem J, 227, 317–326.
Urist M (1965) Bone: Formation by autoinduction. Science, 150, 893–898.
Varma V & Kim K (1985) Placental calcification: ultrastructural and x-ray
microanalytic studies. Scanning Microsc, IV, 1567–1572.
Zilanti, M, Fernandez, S, Azuega, A, Jorgez, J, Severi, FM & Colosi, E
(1987) Ultrasound evaluation of the distal femoral ephiphyseal ossification
center as a screening test for intrauterine growth retardation. Obstet Gynecol,
70, 361–364.