BIOLOGY OF REPRODUCTION 50, 561-571 (1994)
CaBP 9K Levels during the Luteal and Follicular Phases of the Estrous Cycle
in the Bovine Uterus'
N. INPANBUTR, 2 E.K. MILLER,3 B.K. PETROFF, and A.M. IACOPINO3
Department of Veterinary Anatomy and CellularBiology
Ohio State University College of Veterinary Medicine, Columbus, Ohio 43210
ABSTRACT
The expression of calbindin-DgK (CaBP9K) and calbindin-D2 8K (CaBPsK) genes in the reproductive system is well established
for rodent and avian species, but not for domestic livestock. This investigation expanded the study of these proteins to include
the bovine uterus and examined the levels of CaBP9K and CaBP9K mRNA in the nonpregnant bovine uterus during the estrous
cycle. Immunohistochemical studies revealed that CaBPgK was present in all uterine glandular and luminal epithelial cells. In
contrast, the closely related calcium binding protein CaBPz8 Kwas present in only one to two glandular cells in all the samples
examined. Neither protein was localized in the myometrium or in the stromal cells of the endometrium. RIA and dot blot
hybridization were used to quantify the amount of CaBP9K and CaBPgK mRNA. The levels of both the protein and its mRNA were
threefold higher during the luteal phase than during the follicular phase. RIA was also used to determine bovine uterine levels
of 17-estradiol and progesterone. Progesterone levels were higher during the luteal phase than during the follicular phase, while
17p-estradiol levels were higher during the follicular phase. This investigation represents the first characterization of CaBP9K
gene expression in the bovine uterus. It demonstrated that the expression of CaBPgK and CaBPgK mRNA was greatest during the
progesterone-dominated luteal phase of the bovine estrous cycle. These results indicated that CaBPgK may be involved in uterine
glandular function during the luteal phase.
INTRODUCTION
Cellular response to a variety of stimuli involves the
modulation of the cytoplasmic calcium content. Reproductive endocrine tissue and its associated musculature function through mechanisms involving fluctuations in intracellular calcium concentration, whether through the influx
of extracellular calcium or in the release of intracellular
stores. The calcium binding proteins calbindin-D9K (CaBP9 K)
and calbindin-D2 8K (CaBP28K) were first implicated as mediators of vitamin D-dependent transcellular calcium transport in chick and rat intestine [1, 2]. Recent investigations
reported that these proteins were present in both male and
female reproductive systems of domestic animals [3-12].
Initial studies demonstrated that CaBP 2 8K was present in
the shell gland of chickens [3]. CaBP 9K was later localized
in the glandular epithelium, myometrium, and endometrial
stroma of pregnant rat uterus [4]. Subsequent investigations
demonstrated that in pregnant rat uterus and placenta, increases in CaBP9K gene expression coincided with periods
of rapid fetal bone growth and myometrial contractility [5].
In uterus, the expression of these calbindins appears to
be regulated primarily through the steroids estrogen and
Accepted October 25, 1993.
Received August 20, 1993.
'This work was supported by The Ohio State University Seed Grant Program
(N.I.) and Baylor College of Dentistry Intramural Funding (A.M.I). A preliminary
report of this work was presented at the 20th annual meeting of the Society for the
Study of Reproduction, Fort Collins, CO, 1993.
2
Correspondence: Dr. Nongnuch Inpanbutr, Department of Veterinary Anatomy
and Cellular Biology, Ohio State University College of Veterinary Medicine, 1900
Coffey Road, Columbus, OH 43210. FAX: (614) 292-7599.
3Current address: Department of Biomedical Sciences, Molecular Biology Laboratory, Baylor College of Dentistry, 3302 Gaston Avenue, Dallas, TX 75246.
progesterone, rather than via vitamin D [4, 7, 8, 10, 12, 13].
In the rat uterus, progesterone inhibits estrogen-induced
CaBP9K gene expression during the estrous cycle. This effect occurs more than 24 h after progesterone administration [12].
To date, CaBP9K and CaBP 2 8K expression in reproductive
organs has been reported in the uterus of rats and mice
[4, 5, 7, 8, 10,12], which have estrous cycles of 4-5 days, and
in the shell gland of chickens [3,13]. The purpose of our
study was to measure the levels of CaBP 9K or CaBP 2 8K in the
uterus of a species with long estrous cycles with distinct
luteal and follicular phases such as cattle. In the pig, which
has a long estrous cycle, CaBP9 k has recently been cloned
and its expression demonstrated in the immature and mature uterus [14]. We determined that CaBP28K was not present to any significant degree in the nonpregnant bovine
uterus, in contrast to that of the rodent and avian species.
The results of our study show different levels of CaBP9 K
during luteal and follicular phases in the nonpregnant uterus.
To our knowledge, this study represents the first characterization of CaBP9 K gene expression in the bovine uterus.
MATERIALS AND METHODS
Tissue Procurementand Preparation
Uteri from nonpregnant cows were obtained from an
abattoir within 15 min of slaughter. Uterine samples designated for immunohistochemistry were immediately immersed for 20 sec in isopentane cooled in liquid nitrogen,
transferred to 0.5% glutaraldehyde in absolute ethanol for
fixation by freeze substitution, and processed as detailed
previously [15]. Briefly, samples were maintained in the fix-
561
562
INPANBUTR ET AL.
ative for 14 days at -75 0C. The fixed tissues were cleared
in acetone and embedded in paraffin, and sections were
cut at 6 pLm for immunostaining.
Samples designated for steroid, mRNA, and protein analyses were dissected by RNase-free techniques. Immediately
after dissection, tissues destined for isolation of total RNA,
isolation of total protein, or steroid extraction were wrapped
in sterile foil and frozen in liquid nitrogen.
Uterine samples were classified as luteal (N = 11) or
follicular (nonluteal) phase (N = 7) samples according to
modifications of the protocol of Ireland et al. [16]. Changes
in the gross appearance of the corpus luteum and follicles
on the ovaries determined whether animals were in the
luteal phase (Day 5-Day 17) or follicular phase (Day 18-Day
4). Sample classifications were confirmed by the levels of
tissue steroids.
Immunohistochemstry
CaBPgK and CaBP2 K,,were immunolocalized by the peroxidase-antiperoxidase method of Sternberger et al. [17].
Sections were deparaffinized in xylene, cleared in absolute
ethanol, and rehydrated sequentially in 95% ethanol, distilled H 20, and PBS. Samples were pretreated in 3% H20 2
in absolute methanol for 30 min before incubation. The
uterine sections to be examined for CaBP9K were incubated
with 3% normal goat serum (NGS), rabbit anti-bovine CaBP9K
(1:500 in 1% NGS), sheep anti-rabbit IgG (1:20 in PBS), and
rabbit peroxidase-antiperoxidase (1:100 in 1% NGS) for 30
min each. Sections were incubated with diaminobenzidineHCI (0.05%) and H20 2 (0.01%) for 5 min. Immunocontrols
consisted of rabbit nonimmune serum and preadsorbed
antibody (purified CaBP9 K was preincubated with rabbit antiCaBP 9K for 2 h prior to use; 100 Ixg CaBPgK/ml of 1:500 antiCaBP9K) exposed to adjacent sections.
CaBP 2 8K was localized by the same technique using sheep
anti-chick CaBP2 8K (a gift from Dr. AN. Taylor, Baylor College of Dentistry, Dallas, TX) as previously described [18].
Hematoxylin-and-eosin-stained sections were used to verify
normal morphology in all samples.
Isolation of Total RNA
Total RNA was extracted from bovine uterus by a modification [19] of the guanidinium hydrochloride procedure
of Ilaria et al. [20]. Tissues were homogenized with a Tissumizer (Tekmar SDT 1810, Tekmar, Cincinnati, OH) and
homogenates were aliquoted into several equal volumes for
independent RNA isolations from each tissue sample. The
concentration of total RNA was determined by UV spectrophotometry. The integrity of the RNA isolates was determined by agarose gel electrophoresis [20]. RNA isolates were
considered intact if the fluorescence of the 28S rRNA band
was twice as great as that of the 18S rRNA band and if no
other fluorescence was detected below the 18S mRNA band.
Oligonucleotide Probes and [32p] T-4 Kinase End Labeling
An oligonucleotide probe corresponding to nucleotides
443-471 of the rat CaBP9K gene sequence was used for these
studies [211. The sequence of these 29 nucleotides contained only one mismatch to the recently published bovine
sequence and would allow hybridization to CaBP 9K mRNA
[22]. A human 28S mRNA oligodeoxynucleotide probe was
obtained commercially (Oncogene Science ON135; Oncogene Science, Uniondale, NY).
Oligodeoxynucleotide probes were 5' end labeled with
32
[y p] ATP (NEG 002H; New England Nuclear Corp., Boston,
MA) and polynucleotide T-4 kinase [23]. Unincorporated
nucleotides were removed by a modification [19] of cetylpyridinium bromide precipitation [24]. The CaBP 9K oligodeoxynucleotide probe was labeled to a specific activity of
2-4 x 109 cpm/lg. The 28S rRNA oligodeoxynucleotide
probe was labeled to a specific activity of 1-2 x 106 cpm/
'xg.
Northern Blot Hybridizationsand Dot Blot
Hybridizations
Northern blot analyses were used to determine the size
of the CaBP9K mRNA and the specificity of the CaBP9K probe
for this mRNA. Ten micrograms of each total RNA sample
was prepared for denaturing gel electrophoresis as previously described [19]. RNA was transferred from the Northern gels to Zeta Probe membrane (Bio-Rad Labs., Richmond, CA) with a TransBlot Electrophoresis Tank (Bio-Rad).
The manufacturer's protocol was used with the following
adaptations: 1) total RNA was transferred in an 18-h run
using a constant current of 0.2 amps (40 V), and 2) the
temperature of the transfer buffer was maintained at 4°C
using a metal cooling coil (Bio-Rad) connected to a circulating refrigeration bath (Lauda RC20; Brinkman Inst.,
0
Westbury, NY). The membranes were dried in vacuo at 80 C
for 1.5 h. The positions of the 28S rRNA and the 18S rRNA
were marked on the membrane.
For dot blot hybridization and quantification, 4-, 8-, and
16-pxg inputs of total RNA were processed as previously described [19]. The samples were aspirated onto Zeta Probe
membrane and equilibrated in 10-strength saline sodium
citrate by use of an 8 x 12 dot matrix minifold dot blot
apparatus (Canberra S962960; Packard Inst., Downer's Grove,
IL). Each dot blot also contained 4-, 8-, and 16-lg amounts
of rabbit tRNA as an internal control for nonspecific hybridization. The membranes were dried as described for
Northern blots.
Northern blot and dot blot membranes were washed for
0
1 h in 0.1-strength SSC (pH = 7.0) and 0.5% SDS at 65 C.
All membranes were prehybridized for 1-2 h at 45°C in 5strength SSC, 20 mM NaH2PO 4, 10-strength Denhardt's solution, 7% SDS, and 100 ixg/ml denatured herring sperm
DNA 123]. To this hybridization solution was added 2 x 106
cpm/ml of [3 2P]-labeled CaBP 9K probe, and the membranes
CaBP9K LEVELS IN BOVINE UTERUS
563
FIG. 1. Western blot for CaBP9K in bovine uterus. Lane 1, molecular weight standards (strip cut from SDS gel), lane 2, bovine liver (negative control),
lane 3, purified CaBP9g,
lane 4, bovine uterus. Lanes 2 and 4 were loaded with 400 g of total protein; lane 3 contained 20 Itg of purified CaBPSK. Arrowheads
mark position of CaBPK band (9K) and positions of standards (12.5 K cytochrome C, CC; and 6.5 K trypsin inhibitor, TI).
were hybridized to the probe for a minimum of 4 h at 470C.
For 28S rRNA hybridizations, 2 x 10 4 cpm/ml of [32P]-labeled 28S rRNA probe was used. For hybridization to both
probes, membranes were washed in 3-strength SSC, 3% SDS,
and 25 mM NaH2PO4 at room temperature; washed in single-strength SSC and 1% SDS at 47°C; and rinsed twice in
single-strength SSC prewarmed to 60°C. Membranes were
exposed to x-ray film (Kodak XAR-5, Rochester, NY) in cassettes between two intensifying screens at -80 0C. Membranes were stripped of the signal by being washed 1 h in
0.1-strength SSC, 0.05% PPi, and 35% formamide at 100 0C.
Subsequent autoradiography confirmed signal removal.
The ratio between signals for the 4-, 8-, and 16-p.g inputs
determined linear autoradiographic exposures. Autoradiographs in the linear range were scanned with a dual wave
length scanning densitometer with signal integration software (Shimadzu CS-9000 U Dual-wavelength Flying Spot
Scanner; Shimadzu, Kyoto, Japan). The areas of the peaks
were used for quantifying the levels of CaBP9K mRNA or 28S
rRNA. A second method was also used to quantify the levels
of CaBP9K mRNA. The counts per minute for the CaBP9K probe
or the 28S rRNA probe present on the membranes were
counted directly by use of a Matrix 96 Direct Beta Counter
(Packard Inst.). These two complementary methods were
used to confirm the quantitative results.
CaBP9 K Antisera
Antisera to bovine CaBP 9K were raised in adult New Zealand white rabbits via commonly used protocols [25,26].
Rabbits were inoculated subdermally in multiple sites with
500 pl of an emulsion containing 100 ,ig of bovine intestinal CaBP9K (C1540; Sigma Chemical Co., St. Louis, MO)
dissolved in sterile PBS (pH = 7.4) and Freund's Complete
Adjuvant (Pierce Biochemicals, Rockford, IL). Booster injections (an emulsion containing 50 ,ug of CaBP 9Kdissolved
in Freund's Incomplete Adjuvant [Pierce Biochemicals] and
sterile PBS) and bleeds for antiserum collection were performed regularly at 4-6 wk intervals.
Antisera were prepared by clotting the blood in sterile
polypropylene tubes at room temperature for 1 h and then
at 4°C overnight. Antisera were removed and centrifuged
at 1500 x g at room temperature. The antisera were determined, to be specific for CaBP9 K by Western blot analysis
(see Fig. 1).
Protein Isolation and Determination
Tissue samples were homogenized (20% w/v) in cold
sterile PBS (pH = 7.4). Homogenates were centrifuged at
1500 x g for 30 min at 20 0C. Clarified supernatants were
removed and stored at -20°C. The amount of total protein
was determined by the method of Lowry et al. [27].
564
INPANBUTR ET AL.
CaBP9K LEVELS IN BOVINE UTERUS
565
FIG. 3. CaBPgK immunoreactivity in luminal epithelium of bovine uterus. A) Arrows indicate luminal epithelium. x25. B) CaBPSK immunoreactivity
localized within individual glandular cells (arrow). x240.
Western Blot (Immunoblot)
Standard Western blot procedures were used in conjunction with the rabbit polyclonal antisera against bovine
intestinal CaBP9K. Briefly, proteins were separated by SDSPAGE (20% acrylamide). Proteins from the polyacrylamide
gels were transferred onto nitrocellulose paper (Bio-Rad) via
a Hoeffer (San Francisco, CA) transfer unit. Before transfer, a
lane containing molecular weight standards was cut from the
gel for staining with Coomassie blue dye. An absence of protein staining in the slab gel after electrophoretic transfer indicated that the transfer of the proteins to nitrocellulose paper was complete. The nitrocellulose paper was then
sequentially incubated at 4°C with 3% BSA-PBS overnight, 20%
normal rabbit serum in 3% BSA-PBS for 10 min, and 0.4%
FIG. 2. CaBPg immunoreactivity in glandular epithelium of bovine
uterus. A mmunohistochemical labeling of CaBP9k in nonpregnant luteal
phase uterus. Reactive cells are located in uterine glands as indicated by
dark staining. x25. B) Immunocontrol of adjacent section stained with antibody preincubated with antigen. x25. C) Higher magnification of A showing the exclusive localization of CaBPk in the uterine gland. Both endometrial stroma and myometrium are devoid of the protein. x60. D}
Myometrium and endometrium of nonpregnant follicular phase bovine
uterus. CaBPgk-positive uterine gland cells are less intensely reactive as
compared to the gland cells from the luteal phase in C. x60. L, lumen; M,
myometrium; S, stromal cells of endometrium; Arrowhead, CaBPk-positive
uterine gland cells.
bovine intestinal antiserum in 3% BSA-PBS overnight. After
each step, three 5-min washes were performed with 3% BSAPBS. The protein-antiserum complex was visualized by reaction with 251-labeled protein A (New England Nuclear Corp.).
Nitrocellulose filters were sealed in a bag containing 25 ml 25I
protein A (15 000 cpm/100 1 l) in 3% BSA-PBS with 0.5%
Tween-20 and incubated for 2 h at 4C. Filters were then
washed (3 times, 5 min with 50 ml 3%BSA-PBS/0.5% Tween
20; 2 times, 5 min with 50 ml PBS), air-dried, and exposed to
Kodak XAR-5 film at -80°C for 2 h using intensifying screens.
RIA for CaBPgK
The amount of CaBPK present in bovine uterine samples
was determined by RIA [28], with use of the CaBPgK antisera
raised against bovine intestinal CaBPK and '25I-labeled CaBPK.
Iodination of CaBP 9K was performed via the iodagen procedure [291. Briefly, the reaction tube was coated with Iodagen (Pierce Biochemicals). Ten micrograms of CaBP 9Kin
15 IxI of sterile H 2 0 and 20 Al of 0.5 M NaPO4 were added.
The iodination reaction was initiated with the addition of
1 mCi of 125I (NEZ 033H; New England Nuclear Corp.). The
reaction was gently shaken every 5 min for a total of 20
min. A Sephadex G-75 column was prepared in a 3-mm
syringe. The beads were washed thoroughly with several
566
INPANBUTR ET AL.
Uterine Calbindin-D9k Levels
10
0
8
,t
O
6
I-
P
[
-
RIA for 1 713-Estradiol and Progesterone
E
4
a.
1l of goat anti-rabbit y-globulin at 4°C for 12 h. Immunoprecipitates were centrifuged at 1500 x g for 30 min at 40C.
Supernatants were decanted, and the pellets were counted.
The counts per minute for 125I CaBP 9 K bound in the absence
of antisera determined the nonspecific binding (background) and were subtracted from the counts per minute
for total binding. Seven uteri from luteal phase and seven
from follicular phase were used in this study (each sample
was measured in triplicate). Results are expressed as g
CaBP 9k/mg total protein. Student's t-test was used to measure significance.
2
0
0
Phase
Follicular
Luteal
FIG. 4. CaBP levels in bovine uterus during follicular and luteal phases
of estrous cycle. Mean amounts of CaBP9K in Lg SEM per mg of total
soluble protein in follicular (N = 7, n = 21) and luteal phases (N = 7, n =
21). * Significant difference (p < 0.001, t-test).
volumes of 1% BSA-PBS and one volume of 0.1 M NaPO4.
The reaction mixture was applied to the column and eluted
with 0.1 M NaPO 4. Fourteen 800-pl fractions were collected,
and 10 pl from each fraction were counted (Packard Auto
Gamma Counter). The 125I CaBP9 Kelution peak and the fractions on either side of the peak were pooled and used as
the competing antigen in the RIA.
A CaBP9K standard reference solution was prepared from
purified bovine intestinal CaBP9K in RIA buffer containing
0.01 M NaPO 4, 0.15 M NaCl, and 0.1% BSA (pH = 7.4). Each
RIA reaction mixture contained 1) 5.0 x 104 cpm of 1251labeled CaBP9K in 200 Rl RIA buffer; 2) 200 pl of CaBP9K
antisera diluted 1:200 with 1% normal rabbit serum and
0.05 M EDTA-PBS (pH = 7.4); and 3) samples containing
either known amounts of unlabeled CaBP9Kfor standard curve
dilutions or bovine uterine samples with an unknown
amount of CaBP9K in 400 p1 RIA buffer. The reaction was
incubated at 40C for 48 h and immunoprecipitated with 100
Progesterone and estrogen were extracted from uterine
tissue by a modification of the technique of Tanabe et al.
[30]. Uterine samples were homogenized in 10% NH 4CI, titrated to neutrality by use of 1 N HCI, and then extracted
with diethyl ether. Ether extract was reconstituted in bovine
gamma globulin saline (500 p1). Efficiency of extraction, as
measured by labeled steroid extraction, was 79.5% (progesterone) and 93.3% (estrogen).
RIAs were performed to determine concentrations of
progesterone and 1713-estradiol in ether extracts of uterine
samples. Antibodies specific for progesterone (obtained from
Dr. Gordon Niswender, Colorado State University, Fort Collins, CO) and 173-estradiol (obtained from Dr. Roy Butcher,
West Virginia State University, Morgantown, WV) were used.
Radiolabeled steroids were obtained commercially (New
England Nuclear Corp., Wilmington, DE). Standard curves
with slopes of approximately -2.2 were generated for antigen-antibody binding by means of log/logit transformation. The amount of steroid present in uterine extracts was
calculated from the standard curve. The progesterone assay
was standardized for 10-1 extract volumes with use of standard curve values of 5-, 10-, 25-, 50-, 100-, 250-, and 500-pg
amounts. Progesterone recovery within these parameters
was 100 + 0.20%. The 1713-estradiol assay was performed
using sample volumes of 5, 10, and 25 pl and standard curve
values of 5, 10, 20, 50, 100, and 200 pg. 17[-Estradiol recoveries for these standard curve values were 81
09%88 + 18%. All samples used for determination of 1713-estradiol and progesterone levels were run in the same assay.
Intraassay coefficients of variation were less than 5% for
both the progesterone and 1713-estradiol RIA.
RESULTS
Immunohistochemistry
CaBP9 K immunoreactivity was observed in the uterine
glandular epithelium for all the uterine samples examined
(Fig. 2). Staining intensity of the glandular epithelium varied between animals and within the same gland. Typically,
in the glandular epithelium, immunoreactivity for CaBP9 K
was lowest at the neck of the gland near the uterine lumen,
CaBP9 K LEVELS IN BOVINE UTERUS
567
FIG. 5. Northern blot hybridization to the CaBPs, oligodeoxynucleotide probe. Total RNA from follicular and luteal
phases of bovine uterus hybridized to the CaBPg% probe. Arrowheads indicate position of CaBPg mRNA band 1-0.7
kb) as well as the 28s (-5.1 kb) and 18s (-1.9 kb) rRNA bands. Lane 1, total RNA from the luteal phase, lane 2, total
RNA from the follicular phase, lane 3, total RNA from bovine kidney (positive control. 9K, CaBPa< mRNA
whereas staining intensity increased deeper within the endometrium (Fig. 2A). The specificity of CaBP9K immunoreactivity in the uterus was confirmed by negative labeling
with preadsorbed antiserum (equivalent to nonimmune
serum) in place of the specific antiserum (Fig. 2B). Comparison of the staining intensities indicated that the immunoreactivity for CaBP9 K was more intense for the luteal
phase than for the follicular phase (Fig. 2, C and D). No
CaBP9 K immunoreactivity was detected in either the myometrium or in the endometrial stroma (Fig. 2, A, C, and D).
CaBP 9 K immunoreactivity was also observed in the luminal
epithelium (Fig. 3A). In the luminal epithelium, cells reacting strongly for CaBP9 K were often adjacent to morphologically identical cells that were negative for the protein.
CaBP 9K was localized in the cytoplasm and in the nucleus
of positive cells (Fig. 3B).
Concomitant studies using CaBP 28K-specific antisera revealed that only one or two glandular epithelial cells in
thousands of cells of all samples examined showed a positive immunoreactivity for CaBP28 K (data not shown). This
demonstrated that CaBP,,8
K was not expressed to any significant degree in the nonpregnant bovine uterus.
CaBP9K Protein Levels
CaBP9K levels were determined by RIA and compared
between the luteal and follicular phases of the estrous cycle
(Fig. 4). Homogenates of total soluble protein from the luteal phase demonstrated nearly three times more CaBP9K
than those from the follicular phase. The luteal phase samples contained 8.9
0.3 i-g CaBP 9K per milligram of total
soluble protein as compared to the 3.3 + 0.3 g CaBP9 K
per milligram of total soluble protein determined for the
follicular phase samples.
CaBP9 K mRNA
Northern blot hybridization was used to determine the
size of the CaBP 9K mRNA and to characterize the specificity
of the CaBP9 K oligodeoxynucleotide probe. The relative levels of CaBP 9K mRNA were determined by dot blot hybridization. Figure 5 shows a representative Northern blot of
568
INPANBUTR ET AL.
FIG. 6. Representative autoradiograph from dot blot hybridization of total bovine uterine RNA to the CaBPsK oligodeoxynucleotide probe. Inputs of 4,
8, and 16 ,Ig from three different animals in each group are indicated. Blots were stripped and reprobed with a 28s rRNA oligoprobe to normalize any
loading variation. F, follicular phase; L, luteal phase.
the total RNA hybridized to the CaBP9K probe. The CaBP 9 K
probe hybridized to a 0.7-kb mRNA species in bovine uterine samples (lanes 1 and 2). In the bovine kidney (positive
control), the CaBP9 Kprobe identified an identical band. In
the bovine liver (negative control), no CaBP9K bands were
detected. The Northern blot also indicated that luteal phase
levels of CaBP9 K (lane 1) were higher than follicular phase
levels (lane 2). The positive control of total RNA from bovine kidney was shown (lane 3).
Dot blot hybridization was used to determine the levels
of CaBP 9K mRNA relative to the levels of 28S rRNA present
in uterine samples from the luteal and follicular phases.
Figure 6 depicts a representative autoradiograph of the dot
blot hybridization to the CaBP 9K probe. Figure 7 represents
a quantitative comparison of the relative levels of CaBP9 K
mRNA in the bovine uterus during the luteal and follicular
phases. The relative levels of CaBP9K mRNA were determined by two complementary methods. Linear autoradiographs were scanned with a scanning densitometer, and the
area for each peak was integrated. The relative levels of
CaBP 9K mRNA were also determined by direct beta counting. The data from both methods demonstrated that the levels of CaBP 9K mRNA were elevated nearly threefold during
the luteal phase. In the areas of the peaks integrated by the
scanning densitometer, the luteal phase contained 16 383.8
+ 1213.9 units compared to 5325.9
409.4 units determined for the follicular phase. These results confirmed that
the differences seen in the Northern blot CaBP9K band intensities demonstrated a significant difference between the
amounts of CaBP9K mRNA present in total RNA isolated from
luteal and follicular phase samples.
Progesteroneand 1 73-Estradiol Levels
Progesterone concentrations as measured by RIA (Fig. 8)
were nearly five times greater (p < 0.01) in luteal phase
uterine samples (5.5 ± 0.5 pg/mg) than in follicular phase
samples (1.0 + 0.6), reflecting the presence of a functional
corpus luteum. Uterine 1713-estradiol levels (Fig. 9) were
significantly elevated (p < 0.01) in follicular phase uterine
samples (312.7 ± 40.2 pg/g) as compared to luteal phase
uterine samples (187.4 + 32.1 pg/g).
DISCUSSION
In this study, we examined the expression of CaBP9 K and
CaBP28K in the bovine uterus and its relationship to the ste-
roid hormones progesterone and 17-estradiol during the
luteal and follicular (nonluteal) phases of the estrous cycle.
CaBP 28K-positive cells were observed in only one to two
glandular epithelial cells among the thousands of cells in
all samples examined. CaBP9K was localized predominantly
in the glandular epithelium, but it was also present in the
luminal epithelium of the uterine tissues. CaBP 9K and its
mRNA were significantly higher during the progesterone-
569
CaBP9K LEVELS IN BOVINE UTERUS
Uterine Calbindin-D9k mRNA Level
Uterine Progesterone Levels
7
0
0
20
6
0
T
16
4
W
s-
2
T
0
5
4
12
o0
0
3
o)
(fl
0,
L
2
8
T
1
4
n
V
_
a.
0
0
Phase of Estrous Cycle
_ Follicular
0
Luteal
m
FIG. 8. Progesterone levels (measured by RIA) in bovine uterine tissue
collected during follicular and luteal phases of estrous cycle.
Phase
Follicular
M Luteal
FIG. 7. Differences in relative levels of CaBP9K mRNA between follicular and luteal phases of bovine uterus determined by dot blot hybridization. Three to five independent RNA isolations were used for each uterine
sample. For each independent total RNA isolation, three input amounts of
total RNA were used for dot blot hybridizations (4, 8, 16 I.g). Levels of CaBPK
mRNA are shown as mean values of the area ( SEM) of the signals scanned
for samples from follicular (N = 8, n = 25) and luteal (N = 13, n = 58)
phases. Levels of mRNA are given in arbitrary units. * Significant difference
(p < 0.001, t-test).
Uterine Estrogen Levels
400
350 _
5)
I
300
.,
250
dominated phase (luteal phase) than during the estrogendominated phase (follicular phase). This suggests that the
CaBP9 Kgene expression may be under the influence of progesterone rather than 173-estradiol, as reported for other
mammalian uterine tissues [4, 12, 31, 32] and for the chick
shell gland [13]. In the pregnant rat uterus, CaBP 9 K is localized in the myometrium, the endometrial stromal cells,
and the luminal epithelium [4]. The localization of CaBP9K
in the myometrial and endometrial stromal cells, but not
in the luminal epithelium, was reproducible in the nonpregnant uterus by administration of tamoxifen or estrogen
[4]. These authors suggested that CaBPgK may play an important role in the transport of calcium during pregnancy.
In the rat uterus, CaBP 9K mRNA levels were highest at
proestrus, dropped 10-fold at estrus, and were undetectable at diestrus [31]. It was concluded that the CaBP 9K gene
o
0
T
200
150
v,
'-
100
50
f
v
Phase of Estrous Cycle
Follicular
E
Luteal
FIG. 9. 173-Estradiol levels (measured by RIA) in bovine uterine tissue
collected during follicular and luteal phases of estrous cycle.
570
INPANBUTR ET AL.
expression in the rat uterus is regulated by 17-estradiol
during the estrous cycle [31]. Interestingly, despite the higher
levels of 1713-estradiol during diestrus than during estrus,
there was a lag period for CaBP9K induction. Two possible
explanations are proposed for the occurrence of this lag
period: 1) elevated levels of 17-estradiol did not induce
the CaBP9K gene expression; and 2) threshold levels of 173estradiol, in combination with decreased serum progesterone levels, are necessary for CaBP9K induction. However, at
the end of proestrus, serum levels of progesterone are increased in the rat [33]. It is noteworthy that in the rat uterus,
the number of progesterone receptors is under the control
of both estrogen and progesterone; estrogen increases progesterone receptor concentration while progesterone decreases receptor concentration [34-36].
Recent studies show that progesterone antagonizes estrogen-induced CaBP9 K gene expression in the rat uterus
[12]. Progesterone was shown to down-regulate estrogenstimulated CaBP 9K mRNA at the end of estrus and during
diestrus. However, progesterone did not appear to be involved in the rapid decrease of CaBP9K mRNA during estrus.
After progesterone injection, changes in CaBP 9K mRNA are
detected only after a lag time of 24 h, suggesting that progesterone may act indirectly [12]. In fact, vitamin D administration has been shown to stimulate the synthesis of CaBP9 k
mRNA and CaBP 28k mRNA at the transcriptional level at 612 h and the proteins at the post-transcriptional level at 3048 h in mammalian intestine and kidney [37,38]. Thus, the
effect of estrogen and progesterone should be similar to
that of vitamin D, which is considered to be steroid hormone-type regulation. Additionally, there may be an intermediate step between the binding of progesterone to its
receptor and its subsequent action on the CaBP9Kgene. Even
though progesterone down-regulates estrogen-induced
CaBP9 Kgene expression, an increase in estrogen level was
necessary for the biological effect of progesterone [12]. It
is generally accepted that progesterone action is complex:
it is cell- and tissue-specific and depends both on the time
and the dose administered. Studies imply that estrogen and
progesterone interaction is required in order to produce
effects on CaBP 9K gene expression [12,31]. Examination of
progesterone and estrogen receptors in these tissues during the different periods of the estrous cycle would help
to elucidate this interaction and the influence of these receptors on CaBP9K gene expression.
Our results showed that CaBP 9K gene expression was
greatest during the progesterone-dominated (luteal) phase.
CaBP 9K was localized primarily in the glandular epithelial
cells. During the luteal phase, the bovine uterus is under
the influence of progesterone, and the glandular epithelial
activity is increased. CaBP 9K was significantly lower during
the estrogen-dominated (follicular) phase, when the glandular epithelium is less active. This suggested that in the
bovine uterus, CaBP 9Kcould play a role in the processes of
calcium transport or homeostasis in glandular epithelium.
It is also possible that CaBP 9K is involved in the exocytosis
of uterine glandular products. In support of such a role, rat
intestinal CaBP9K is shown to stimulate Ca2 +-Mg 2+ ATPase,
while Ca2 + pump activity increases when rat duodenal basolateral membranes are exposed to CaBPK [39,40].
It is noteworthy that estrogen levels are appreciably higher
throughout the luteal phase of the bovine estrous cycle than
in the rodent estrous cycle. Basal levels of estrogen may be
sufficient to initiate the interaction between estrogen and
progesterone and could be involved in the control of CaBP9K
gene expression for the bovine uterus. It should be noted
that in the rat uterus, CaBP9K expression occurs during the
estrogen-dominated phase, and that the protein in this species is localized in the myometrium and stroma. In contrast,
the highest levels of CaBP 9Kgene expression in the bovine
uterus were associated with the progesterone-dominated
luteal phase and were localized in the glandular epithelium. This suggests that CaBP9K may play a different role in
each species. Indeed, preliminary data from human myometrial tissues have also demonstrated elevated CaBP9K in
the luteal phase [Miller and Jacopino, unpublished data].
Peak CaBP9K levels corresponded to peak progesterone levels during the menstrual cycle (Day 21), and elevated CaBP9 K
levels were found to be maintained during pregnancy until
the onset of labor. These data would also support differential steroid regulation of CaBP9K in higher vertebrates. In
conclusion, these studies demonstrated that CaBP 9K gene
expression in the bovine uterus occurred mainly in the
glandular epithelium and was significantly higher in the luteal phase (progesterone-dominated) than in the follicular
phase (estrogen-dominated). These studies were the first
characterization of CaBP 9K in the bovine uterus and suggested that CaBP9 K may have a role in uterine gland function during the luteal phase of the bovine estrous cycle.
ACKNOWLEDGMENTS
We thank Dr. Dennis Miller of the University of Texas, Dallas, Texas, for assistance in synthesizing the CaBPgK oligodeoxynucleotide probe, and Dr. Alan N. Taylor
of the Baylor College of Dentistry, Dallas, Texas, for the gift of antiserum to the
CaBP2S. We thank Tanya Franklin, Damiane de Wit, and Jacqueline Rivas for technical
assistance during the experimental assays.
REFERENCES
1. Wasserman RH, Taylor AN. Vitamin D3 induced calcium binding protein in chick
intestinal mucosa. Science 1966; 152:791-793.
2. Kallfelz FA, Taylor AN, Wasserman RH. Vitamin D-induced calcium binding factor
in rat intestinal mucosa. Proc Soc Exp Biol Med 1967; 125:54-58.
3. Corradino RA, Wasserman RH, Pubols MH, Chang SI. Vitamin D3 induction of a
calcium-binding protein in the uterus of the laying hen. Arch Biochem Biophys
1968; 125:378-380.
4. Bruns ME, Overpeck JG, Smith GC, Hirsh GN, Mills SE, Bruns DE. Vitamin D
dependent calcium binding protein in rat uterus: differential effects of estrogen,
tamoxifen, progesterone, and pregnancy on accumulation and cellular localization. Endocrinology 1988; 122:2371-2378.
5. Mathieu CL, Burnett SH, Mills SE, OverpeckJG, Bruns DE, Bruns ME. Gestational
changes in calbindin-D9K in rat uterus, yolk sac, and placenta: implications for
maternal-fetal calcium transport and uterine muscle function. Proc Natl Acad Sci
USA 1989; 86:3433-3437.
CaBP9K LEVELS IN BOVINE UTERUS
6. Kagi U, Chafouleas JG, Norman AW, Heizmann CW. Developmental appearance
of the Ca2+-binding proteins parvalbumin, calbindin D28K, S-100 proteins and calmodulin during testicular development in the rat. Cell Tissue Res 1988; 252:359365.
7. L'Horset F, Perret C, Brehier A, Thomasset M. 171-Estradiol stimulates the calbindin-DgK (CaBP9K) gene expression at the transcriptional and posttranscriptional levels in the rat uterus. Endocrinology 1990; 127:2891-2897.
8. Darwish H, Krisinger J, Furlow JD, Smith C, Murdock FE, DeLucas HF. An estrogen-responsive element mediates the transcriptional regulation of calbindin-DK
gene in rat uterus. J Biol Chem 1991; 266:551-558.
9. Inpanbutr N, Taylor AN. Expression of calbindin-D 2K in developing and growing
chick testes. Histochemistry 1992; 97:335-339.
10. Opperman LA, Saunders TJ, Bruns DE, Boyd JC, Mills SE, Bruns E. Estrogen inhibits calbindin-D28K expression in mouse uterus. Endocrinology 1992; 130:17281735.
11. Inpanbutr N, Taylor AN. Expression Calbindin-D28K in developing and growing
ovaries of chicken embryos. Am J Vet Res 1993; 54:514-519.
12. L'Horset F, Blin C, Brehier A, Thomasset M, Perret C. Estrogen-induced calbindinD 9k gene expression in the rat uterus during the estrous cycle: late antagonistic
effect of progesterone. Endocrinology 1993; 132:489-495.
13. Corradino RA, Smith CA, Krook LP, Fullmer CS. Tissue-specific regulation of shell
gland calbindin D28K biosynthesis by estradiol in precociously matured, vitamin
D-depleted chicks. Endocrinology 1993; 132:193-198.
14. Jeung E, Krisinger J, Dann JL, Leung PCK. Cloning of the porcine calbindin-D9K
complementary deoxyribonucleic acid by anchored polymerase chain reaction
technique. Biol Reprod 1992; 47:503-508.
15. Inpanbutr N, Slemons RD. Immunocytochemical localization of type A influenza
virus nucleoprotein in chicken kidney, using freeze substitution technique for
tissue fixation. Am J Vet Res 1993; 54:525-528.
16. Ireland J, Murphee RL, Coulson PB. Accuracy of predicting stages of bovine estrous cycle by gross appearance of the corpus luteum. J Dairy Sci 1988; 63:155160.
17. Sternberger LA, Hardy PH, CuculisJ, Meyer HG. The unlabeled antibody-enzyme
method of immunohistochemistry: preparation and properties of soluble antigen
antibody complex (horseradish peroxidase-antiperoxidase) and its use in identification of spirochetes. J Histochem Cytochem 1977; 18:315-333.
18. Inpanbutr N, Taylor AN. Expression of calbindin D28Kin developing and growing
chick ovary. Am J Vet Res 1993; 54:514-519.
19. Miller EK, Raese JD, Morrison-Bogorad MR. Expression of heat shock protein 70
and heat shock cognate 70 messenger RNA in rat cortex and cerebellum after
heat shock and amphetamine treatment. J Neurochem 1991; 56:2060-2071.
20. Ilaria R, Wines D, Pardue S, Jamison S, Ojeda SR, SniderJ, Morrison MR A rapid
microprocedure for isolating RNA from multiple samples of human and rat brain.
J Neurosci Methods 1985; 15:165-174.
21. Peret C, Lomri N, Gouhier N, Auffray C, Thomasset M. The rat vitamin D-dependent calcium binding protein (9-KDa CaBP) gene: complete nucleotide sequence
and structural organization. Eur J Biochem 1988; 172:43-46.
22. Kumar R, Wieben E, Beecher SJ. The molecular cloning of the complementary
deoxyribonucleic acid for vitamin D-dependent calcium-binding protein: structure of the full-length protein and evidence for homologies with other calcium-
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
571
binding proteins of the Troponin-C superfamily of proteins. Mol Endocrinol 1989;
3:427-432.
Maniatis T, Fritsch EF, Sambrook J. Molecular cloning: a laboratory manual. Cold
Spring Harbor, New York: Cold Spring Harbor Laboratory; 1982.
Geck P, Nasz I. Concentrated, digestible DNA after hydroxylapatite chromatography with cetylpyridinium bromide precipitation. Anal Biochem 1983; 135:264268.
Lin S-C, Morrison-Bogorad M. Developmental expression of mRNAs encoding
thymosin 10 in rat brain and other tissues. J Mol Neurosci 1991; 2:35-44.
Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: A Laboratory Manual, Second Edition. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory;
1989.
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the
folin phenol reagent. J Biol Chem 1951; 193 :26 5 -275 .
Sonnenberg J, Pansini AR, Christakos S. Vitamin D-dependent rat renal calcium
binding protein: development of a radioimmunoassay, tissue distribution, and
immunologic identification. Endocrinology 1984; 115:640-648.
Markwell MAK Iodination of proteins using the iodagen catalyst. Anal Biochem
1982; 125:427-431.
Tanabe Y, Yano T, Nakamura T. Steroid hormones synthesis and secretion by
testes, ovary and adrenal of embryonic and postembryonic ducks. Gen Comp
Endocrinol 1983; 49:144-153.
Krisinger J, Dann JL, Currie WD, Jeung EB, Leung PCK. Calbindin-D9K mRNA is
tightly regulated during the estrous cycle in the rat uterus. Mol Cell Endocrinol
1992; 86:119-123.
Krisinger J, DannJL, Jeung EB, Leung PCK. Calbindin-D9K gene expression during
pregnancy and lactation in the rat. Mol Cell Endocrinol 1992; 88:119-128.
Sodersten P, Eneroth P. Serum levels of estradiol-1713 and progesterone in relation to sexual receptivity in intact and ovariectomized rats. J Endocrinol 1981;
89:45-54.
Ekka E, Vanderheyden I, Glorieux B, De Hertoch R Estradiol-induced progesterone receptor synthesis in normal and diabetic ovariectomized rat uterus. J
Steroid Biochem 1987; 28:61-64.
Milogrom E, Luu Thi M, Atger M, Baulieu EE. Mechanisms regulating the concentration and the conformation of progesterone receptor(s) in the uterus. J Biol
Chem 1973; 248:6366-6374.
Leavitt WW, Chen TJ, Allen TC. Regulation of progesterone receptor formation
by estrogen action. Ann NY Acad Sci 1977; 286:210-216.
Dupret J, Perret C, Lomri N, Thomasset M, Cuisiner-Gleizes P. Transcriptional
and posttranscriptional regulation of vitamin D-dependent calcium binding protein in rat duodenum by 1,25 dihydroxycholecalciferol. J Biol Chem 1987;
262:16553-16559.
Varghese S, Deaven LL,Huang YC, Gill RK, Iacopino AM, Christakos S. Transcriptional Regulation and chromosomal assignment of the mammalian Calbindin-D 28k
gene. Mol Endocrinol 1989; 3:495-502.
Freund TS. Vitamin D-dependent intestinal calcium binding protein as an enzyme modulator. In: Norman AW, Schaefer K, Herrath DV, Grigoleit HG (eds.),
Vitamin D. Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism. Berlin: de Gruyter; 1982: 249-251.
Walters JR Calbindin-D9k stimulates the calcium pump in rat enterocyte basolateral membranes. Am J Physiol 256 (Gastrointest Liver Physiol 19) 1989; G124G128.