Immunochemical localization of vitamin D-dependent calcium-binding protein in mouse placenta and yolk sac.

THE ANATOMICAL RECORD 213514-517 (1985)
lmmunochemical Localization of Vitamin DDependent Calcium-Binding Protein in Mouse
Placenta and Yo1k Sac
M. ELIZABETH H. BRUNS, ELLEN KLEEMAN, STACEY E. MILLS, DAVID E. BRUNS,
AND JOHN C. HERR
Departments of Pathology (M.E.H. B., E.K., S.E. M., D.E.B.), Biochemistry (M.E. H. R.), and
An,atomy (J.C.H.), University of Virginia Medical Center, Charlottesuille, VA 22908
ABSTRACT
A 10,000-dalton calcium-binding protein (10-kd CaBP) has been
described in the placentae and yolk sacs of rats and mice. This protein is similar or
identical to vitamin D-dependent intestinal CaBP and these proteins have been
implicated in the molecular mechanisms of intestinal calcium absorption and transplacental calcium transport. Using a n antiserum to the purified 10-kd rat intestinal
CaBP, the localization of CaBP in the 16-17-day mouse yolk sac and placenta was
studied immunocytochemically with peroxidase-antiperoxidase labelling and quantified by radial immunodifussion. A high concentration of immunolabelling was
observed in the endodermal cells of the intraplacental yolk sac lining the sinuses of
Duval. The columnar endodermal cells lining one side of the endodermal sinuses
adjacent to fetal vessels contained most of the immunoreactive label. Quantitation
by radial immunodiffusion demonstrated a 5.5-fold higher concentration of CaBP in
the portion of the placenta containing most of the intraplacental yolk sac than in
the maternal half of the placenta. This demonstration of a 10-kd CaBP within the
intraplacental yolk sac suggests this protein functions to facilitate placental calcium
transport and is the first study which directly supports the hypothesis of a functional
role for the sinuses of Duval in calcium transport.
The placentae and yolk sacs of mice and rats contain
high concentrations of a 10-kilodalton (kd) vitamin Ddependent calcium-binding protein (CaBP) (Bruns et al.,
1978, 1981; Marche et al., 1978; Delorme et al., 1979,
1982). The steady-state concentration of the vitamin Ddependent CaBP has proven to be a consistent marker
of vitamin D metabolite action on calcium transport in
the small intestine (Wasserman and Fullmer, 1982). A
similar role of CaBP in placental calcium transfer is
suggested by several findings, including evidence that
rat yolk sac and placenta contain 1,25-dihydroxycholecalciferol (Stumpf et al., 1983) as well as its receptors
(Pike et al., 1980; Danan et al., 1981). A role for CaBP
in placental calcium transfer is suggested by the parallel gestational changes in placental CaBP and fetal
growth (Bruns et al., 1978), which in turn reflect the
fetal accumulation of calcium (Comar, 1956).
The present studies were undertaken to investigate
the cellular localization of CaBP in placenta in order to
understand how this protein may function in the transfer of calcium from mother to fetus. As reported by
Delorme et al. (19831, CaBP in the yolk sac is localized
to columnar epithelial cells which resemble absorptive
intestinal cells. We have found CaBP to be concentrated
additionally in columnar cells of visceral yolk sac endoderm which penetrate the placental labyrinth in the
region bordering the attachment of the umbilical cord.
As detailed below, these cells line the endodermal sinuses or pits, or sinuses of Duval (Mossman, 1937; Amo0 1985 ALAN R. LISS, INC.
roso, 1952).These findings are of anatomical importance
because CaBP provides a unique marker for intraplacental endodermal cells that will be useful in examining
their distribution and appearance during gestation.
Moreover, the finding of CaBP within the intraplacental
yolk sac cells suggests that this region is a site for active
calcium transport from mother to fetus.
MATERIALS AND METHODS
Maferials
Glutaraldehyde and 3,3-diaminobenzidine tetrahydrochloride were from Sigma. Antirabbit IgG and peroxidase were obtained from Cappell or Miles. Acrolein
was purchased from Polysciences Inc.
Animals
Pregnant ICR mice were obtained from Charles River.
The day of mating was considered day 0 of gestation.
Term was observed at 19 days of pregnancy. The animals
were fed normal rat chow ad libitum and studied on
days 16-17.
CaBP and Antiserum
The rat intestinal CaBP was purified as described
(Bruns et al., 1977) with a final preparative slab gel
Received February 28, 1985; accepted June 4, 1985.
CABPIN MOUSE PLACENTA AND YOLK SAC
5 15
electrophoresis using 0.1 M EDTA added to buffer. The ta’s fetal surface surrounding the umbilical attachment
production and validation of rabbit antisera which cross- and 2) the placenta minus intercept (180-290 mg wet
react with both rat and mouse CaBP were performed as weight) which was comprised of peripheral portions of
the labyrinth. All tissues were weighed and then homogoutlined (Bruns et al., 1978).
enized, using a Tissuemizer, in a n equal amount (1 ml
Tissue Fixation
per gram of tissue) of buffer containing 10 mM Tris-HC1,
A freeze-substitution method for tissue fixation (Tay- 5 mM benzamidine, 1 mM 2-mercaptoethanol, pH 7.2,
lor, 1981)was used first for immumohistochemical stain- plus 2 mM phenylmethylsulfonyl fluoride and 0.2 tryping of placental calcium binding protein by the per- sin inhibitor unit/ml of aprotinin. The homogenate was
oxidase-antiperioxidase technique (PAP) (Sternberger et centrifuged a t 40,OOOg for 20 minutes; the amount of
al., 1970). Taylor (1981) has shown that freeze substitu- immunoreactive CaBP in the supernatant was meation provides accurate localization of CaBP in the chick sured by radial immunodiffusion as described (Bruns et
intestine. Using this fixation method with mouse intes- al., 1978).
tine, we observed the expected localization of CaBP in
(Color Figures 2 and 3 were prepared and appear elsethe absorptive cells of intestinal villi, with acceptable where in this issue. Please see pp. 532-535 for these
preservation of morphology. In placenta, we found stain- figures and their accompanying legends.)
ing in the placental hilum, in the region of the sinuses
RESULTS
of Duval. Controls using absorbed antiserum were negTo better explain the results of the immunochemical
ative. However, the cellular structures were poorly preserved and positive identification of stained types was staining of the 10-kd CaBP in placenta and yolk sac, a
not possible. To achieve better preservation of placental drawing depicting the relationships of the fetal membranes to the mouse fetus and placenta at 16-17 days
cell morphology, a perfusion method was used.
The pregnant mouse was anesthetized using ether, a gestation is shown in Figure 1; further details may be
laparotomy was performed, and the heart was exposed. found in the references (Mossman, 1937; Amoroso, 1952;
A needle was placed in the left ventricle and 10 ml of Wislocki and Padykula, 1953; Brambell and Halliday,
buffered saline was perfused a t a rate of 8 ml/minute. 1956; Padykula and Richardson, 1963). Padykula and
The intercostal arteries cut during exposure of the heart Richardson divided the rat yolk sac into three distinct
served as the outflow for the perfusate. After the liver parts: 1) the outer parietal wall (bilaminar omphalocleared, the femoral artery was cut and 50 ml of fixative pleure), which we divided into a) the juxtauterine yolk
was perfused (2% acrolein in 0.1 M Na2HP04 buffer, pH sac and b) the periplacental yolk sac, which is the contin7.4, containing 0.25% glutaraldehyde). Tissues were col- uation of bilaminar omphalopleure over the fetal surface
lected and immersed in the same fixative for 2-4 hours of the placenta; 2) a n inner visceral wall (vascular splanand rinsed five times in phosphate-buffered saline. The chonopleure) which may be subdivided into a) villous
samples were dehydrated with graded ethanol series yolk sac near the placenta and b) a smooth visceral
and cleared in toluene for 2-4 hours and embedded in surface over much of the remaining yolk sac; and 3) the
placenta’s endodermal sinuses (sinuses of Duval), which
paraffin after infiltration for 1hour.
we term intraplacentd yoZk sac.The intraplacental yolk
lmmunoperoxidase Staining
sac endoderm is polarized, consisting of a high to low
The peroxidase-antiperoxidase technique of Sternber- columnar epithelium adjacent to fetal vessels and a
ger was used (Sternberger et al., 1970). The tissue ob- squamous epithelium adjacent to maternal blood spaces.
For the study of CaBP localization in mouse placenta,
tained by perfusion was deparafinized using toluene and
a graded ethanol series. Nonspecific endogenous peroxi- the complete fetal unit was processed en bloc, including
dase was removed using methanol with 0.3% HzOz for placenta, yolk sac, uterus, and fetus, through fixation,
30 minutes. Background staining was reduced by incu- embedding, and immunoperoxidase staining. Fixing the
bating sections in 1%cold sheep serum in 0.05 M Tris- whole unit together proved to be a surprisingly valuable
buffered saline (TBS), pH 7.6, for 30 minutes. The pri- maneuver. Within the 16-day placenta, immunoreactive
mary antiserum (anti-CaBPto rat intestinal CaBP, 150- CaBP was localized to the endodermal cells of both the
1:500 in TBS) was applied for 30 minutes a t room tem- villous yolk sac and the intraplacental yolk sac (Fig. 2).
perature. The slides were rinsed in 1% sheep serum; Immunoreactive CaBP is present in the endodermal cells
sheep antirabbit IgG (heavy and light chains, Cappell of the villous portion of the yolk sac as indicated by
Labs), 1:200 in TBS, was applied for 15 minutes. The moderate brown staining. An abrupt transition from
slides were rinsed in 1% sheep serum; PAP (rabbit, Cap- moderate to intense staining was readily observed a t
pell, in TBS 1:200) was applied for 30 minutes. The the junction of the villous and intraplacental portions of
slides were rinsed in TBS and 3,3-diaminobenzidene was the yolk sac (Fig. 2). The immunoreactive intraplacental
applied for 15 minutes to yield a brown reaction product. yolk sac cells ranged from high to low columnar and
The slides were rinsed in TBS, counterstained in hema- could readily be distinguished from the squamous entoxylin, and dehydrated in a graded alcohol series and dodermal cells of both the periplacental yolk sac and the
endodermal sinuses. No specific staining, except a n octoluene.
casional island of yolk sac endoderm which was cut
Tissue Preparation for Radial lrnrnunodiffusion Assay
obliquely, was observed in other parts of the placenta
Tissue samples of placenta and yolk sac were obtained including the chorioallantoic labyrinth. Figure 2B shows
from 16-17-day pregnant mice under ether anesthesia. a control, using antiserum that was absorbed with
The placenta was dissected into two regions: 1)the pla- highly purified CaBP. This region of intraplacental yolk
cental intercept which consisted of a disk-shaped zone sac showed no immunoreactivity using the absorbed
(36-78 mg wet weight) cut from the center of the placen- antiserum.
M.E.H. BKUNS El' AL
5 16
n
lntraplacental Bilaminar Omphalopleure
(Yo1k Sac
Periplacental Bilaminar
Omphalopleure (Yolk Sac)
(Vascular Splanchnopleure)
Smooth Yolk Sac
(Vascular Splanchnopleure)
Juxtauterine Bilaminar Omphalopleure
(Yolk Sac)
Fig 1 A drawing depicting the relationships of the murine yolk sac membranes, placenta,
uterus, and fetus in late gestation For details see the text The square outlines a region
typically containing intraplacental yolk sac and is fox orientation in interpreting Figure 2 on
p 532 of this issue
Figure 3 demonstrates staining of 16-day intraplacen- mous endodermal cells on the opposite side of the sinus
tal yolk sac and the relationship of several fetal and makes determination of their immunoreactivity for
maternal vessels to it. This tissue was perfused through CaBP difficult at the light microscopic level; however,
the maternal vessels; maternal blood spaces contain no immunolocalization was detected in these cells.
In order to confirm the immunohistochemical localizamany fewer blood elements than do fetal vessels. The
intensely stained columnar intraplacental endodermal tion of CaBP in endodermal cells of the intraplacental
cells are often found in close proximity to the fetal ves- yolk sac, quantitative CaBP measurements were made
sels. The attenuated cytoplasm of the cuboidal-to-squa- by a radial immunodiffusion assay. The 17-day placenta
was dissected into two parts: the intraplacental yolk sac
region was termed the placental intercept and the rest of
the placenta was termed placenta minus intercept. Figure 4 indicates the CaBP concentration (pg CaBP/mg
T
protein) in these tissue and in extraplacental yolk sac.
The placental intercept, which contained the intraplacental yolk sac but no villous yolk sac, showed by far
.-c 6 the
highest concentrations of immunoreactive CaBP. A
al
c
greater amount of immunochemical CaBP in the fetal
2
side of the placenta has also been observed in rats (Dea 5lorme et al., 1979).
The visual difference in the staining intensities of
\
villous yolk sac and intraplacental yolk sac (Fig. 2) was
2c 4 matched by the quantitative CaBP determinations-2.6
al
pg CaBP/mg protein versus 6.8 pg CaBP/mg protein for
5
the villous and intraplacental portions, respectively. The
'3 3U
placenta minus the intercept, although low in CaBP
0)
concentration as compared to placenta plus intercept,
had measureable amounts of CaBP--1.2 pg CaBP/mg
protein. The placenta minus the intercept was stained
using the immunohistochemical method; no specific placental staining was observed, except for intense staining
in islands of yolk sac cells that happened to have been
outside the intercept dissection. Thus the low concentraPlacenta
Placental
Splanchnopleuric
tion of CaBP, which was quantitated in the placenta
minus
intercept
Yolk Sac
minus the intercept, may be contained in those cells of
intercept
the sinuses of Duval which deeply evaginate into the
placental labyrinth.
'i
F
Fig. 4. Immunoreactive CaBP in placental intercept, placenta minus
intercept, and splanchnopleuric yolk sac in the 17-daymouse placenta.
The placenta intercept consisted of a disk-shaped zone (36-78 mg wet
weight) cut from the center of the placenta's fetal surface surrounding
the umbilical attachment.
DISCUSSION
The results described here indicate that the 10-kd
CaBP is located not only within the extraplacental yolk
CABPIN MOUSE PLACENTA AND YOLK SAC
sac but is to be found as well within intraplacental yolk
sac endoderm. This has been shown by immunohistochemical staining using a PAP technique and quantitated by radial immunodiffusion assay. The placental
CaBP immunoreactivity is localized to columnar yolk
sac cells lining the sinuses of Duval in close proximity
to fetal vessels. Duval, in 1892, first described outpocketings of yolk sac that penetrate the fetal surface of the
chorioallantoic placenta, intertwining among the branching roots of the allantoic blood vessels. These outpocketings of the yolk sac endoderm have been referred to as
Duval’s entodermal sinuses (Amoroso, 19521, or endodermal sinuses (Padykula and Richardson, 19631, and
have been postulated to play a role in the transport of
nutrients and antibodies from fetus to mother (Amoroso,
1952; Brambell and Halliday, 1956; Padykula and Richardson, 1963). Although the precise gestational time for
the initiation of these evaginations is not fully established, they have been observed to be present on the
15th day of gestation and to develop until term (Padykula and Richardson, 1963). Thus far, they have been
observed only in genera which are members of the family Muridae (Mossman, personal communication). Mossman refers to endodermal sinuses as “placental pits”
and describes them as “intraplacental diverticula of the
periplacental yolk sac wall.” Endodermal sinuses are
lined by two histologically distinct endodermal cell types,
described by Wislocki and Padykula (1953) as visceral
and parietal endodermal cells. In the present study, we
refer to the different cells in the intraplacental yolk sac
a s cuboidal to squamous yolk sac (parietal) and columnar yolk sac (visceral) cells.
The availability of immunoreactive CaBP as a marker
for intraplacental endoderm should facilitate anatomic
studies of the three-dimensional structure of the sinuses.
We anticipate temporal studies of the appearance of
CaBP-containing cells in the placenta. By analogy to the
intestine, the high concentration of CaBP in these cells
suggests that active transcellular transport of calcium
is a n important function of the intraplacental portion of
the yolk sac in Muridea. This hypothesis proposes a
specific function for the intraplacental yolk sac and carries important implications for future studies of the
mechanisms of placental calcium transfer.
ACKNOWLEDGMENTS
This research was supported by United States Public
Health Service grant HD 123345. The authors wish to
thank Mrs. Grace Kohler and Mrs. Betty Roberts for
their excellent secretarial assistance. Also we wish to
thank Dr. William B. Rhoten for his valuable assistance
in the immunoperoxidase methodology. The suggestions
of Dr. Harland Mossman are greatly appreciated.
517
LITERATURE CITED
Amoroso, E.C. (1952)Placentation. In: Marshall’s Physiology of Reproduction, 3rd Ed. AS. Parkes, ed. Longman, Green, New York, Vol.
2, pp. 127-311.
Brambell, F.W.R., and R. Halliday (1956) The route by which passive
immunity is transmitted from mother to fetus in the rat. Proc. Roy.
Sac. Land. [Biol.], 145t170-178.
Bruns, M.E.H., E.B. Fliesher, and L.V. Avioli (1977)Control of vitamin
D-dependent calcium-binding protein in rat intestine by growth
and fasting. J. Biol. Chem., 2524145-4150.
Bruns, M.E.H., A. Fausto, and L.V. Avioli (1978) Placental calcium
binding protein in rats. J. Biol. Chem., 253:3186-3190.
Bruns, M.E.H., S. Vollmer, V. Wallshein, and D.E. Bruns (1981) Vitamin D-dependent calcium binding protein: Immunochemical studies and synthesis by placental tissue in uitro. J. Biol. Chem.,
256:4649-4653.
Comar, C.L. (1956)Radiocalcium studies in pregnancy. Ann. N.Y. Acad.
Sci., 64t281-298.
Danan, J.L., A.C. Delorme, and P. Cuisinier-Gleizes(1981)Biochemical
evidence for a cytoplasmic 1,25-dihydroxyvitaminDs receptor-like
protein in rat yolk sac. J. Biol. Chem., 256t4847-4850.
Delorme, A.C., P. Marche, and J.M. Garel (1979) Vitamin D-dependent
calcium-binding protein. Changes during gestation, prenatal and
postnatal development in rats. J. Dev. Physiol., 1:181-194.
Delorme, A.C., J.L. Danan, M.A. Ripoche, and H. Mathieu (1982)
Biochemical characterization of mouse vitamin D-dependent calcium-binding protein. Biochem. J., 205t49-57.
Delorme, A.C., P. Cassier, B. Geny, and H. Matheiu (1983) Immunocytochemical localization of vitamin D-dependent calcium-binding
protein in the yolk sac of the rat. Placenta, 4263-270.
Duval, M. (1892) La Placenta des Rongeurs. F. Alcan, ed.
Paris:Ancienne Libraire Germer Bailier et Cic. 2.
Marche P.A., A. Delorme, and P. Cuisinier-Gleizes (1978) Intestinal
and placental calcium-binding proteins in vitamin D-deprived or supplemented rats. Life Sci., 232555-2562.
Mossman, H.W. (1937) Comparative morphogencsis of the fetal membranes and accessory uterine structures. Carnegie Contrib. Embryol., 26t129-246.
Padykula, H.A., and D. Richardson (1963) A correlated histochemical
and biochemical study of glycogen storate in the rat placenta. Am.
J. Anat., 112.215-229.
Pike, J.W., L.J. Gooze, and M.R. Haussler (1980) Biochemical evidence
for 1;25-dihydroxyvitamin D receptor macromolecules in parathyroid, pancreatic, pituitary, and placental tissues. Life Sci., 26:407414.
Sternberger, L.A., P.H. Hardy, J.J.Cuculis, and H.G. Meycr (1970)The
unlabeled antibody enzyme method of immunohistochemistry:
Preparation and properties of soluble antigen-antibody complex
(horseradish peroxidase-antihorseradish peroxidase) and its use in
identification of spirochetes. J. Histochem. Cytochem., I8t315-333.
Stumpf, W.E., M. Sar, R. Narbaitz, S. Huang, and H.F. DeLuca (1983)
Autoradiographic localization of 1,25 dihydroxyvitamin D3 in rat
placenta and yolk sac. Harm. Res., 18t215-220.
Taylor, A.N., (1981) Immunocytochemical localization of the vitamin
D-induced calcium-binding protein: Relocation of antigen during
frozen section processing. J. Histochem. Cytochem., 29:65-73.
Wasserman, R.H., and C.S. Fullmer (1982)Vitamin D-induced calciumbinding proteins. In: Calcium and Cell Function. W.Y. Cheung, ed.
Academic Press, New York. Val. 2 , pp. 175-216.
Wislocki, G.B., and H.A. Padykula (1953) Reichert’s membrane and
the yolk sac of the rat investigated by histochemical means. Am.
J. Anat.. 92:117-151.