Content uploaded by Eduardo L. Gastal
Author content
All content in this area was uploaded by Eduardo L. Gastal on Dec 24, 2018
Content may be subject to copyright.
Animal Reproduction Science 90 (2005) 273–285
Changes in steady-state concentrations of messenger
ribonucleic acids in luteal tissue during
prostaglandin F2␣induced luteolysis in mares
M.A. Bega,c, E.L. Gastala,c, M.O. Gastala,c,
S. Jib, M.C. Wiltbankb, O.J. Ginthera,c,∗
aDepartment of Animal Health and Biomedical Sciences, 1656 Linden Drive,
University of Wisconsin-Madison, WI 53706, USA
bDepartment of Dairy Science, University of Wisconsin-Madison, WI 53706,USA
cThe Eutheria Foundation, Cross Plains, WI 53528, USA
Received 18 January 2005; received in revised form 16 February 2005; accepted 28 February 2005
Available online 11 April 2005
Abstract
Transvaginal ultrasound-guided luteal biopsy was used to evaluate the effects of prostaglandin
(PG)F2␣onsteady-stateconcentrationsofmRNAforspecificgenes that may be involvedinregression
of the corpus luteum (CL). Eight days after ovulation (Hour 0), mares (n= 8/group) were randomized
into three groups: control (no treatment or biopsy), saline+ biopsy (saline treatment at Hour 0 and
luteal biopsy at Hour 12), or PGF2␣+ biopsy (5mg PGF2␣at Hour 0 and luteal biopsy at Hour 12).
The effects of biopsy on CL were compared between the controls (no biopsy) and saline+biopsy
group. At Hour 24 (12h after biopsy) there was a decrease in circulating progesterone in saline
group to 56% of pre-biopsy values, indicating an effect of biopsy on luteal function. Mean plasma
progesterone concentrations were lower (P<0.001) at Hour 12 in the PG group compared to the
other two groups. The relative concentrations of mRNA for different genes in luteal tissue at Hour 12
was quantified by real time PCR. Compared to saline-treated mares, treatment with PGF2␣increased
mRNA for cyclooxygenase-2 (Cox-2, 310%, P< 0.006), but decreased mRNA for LH receptor to 44%
(P< 0.05), steroidogenic acute regulatory protein to 22% (P<0.001), and aromatase to 43% (P<0.1)
of controls. There was no difference in mRNA levels for PGF2␣receptor between PG and saline-
treated groups. Results indicated that luteal biopsy alters subsequent luteal function. However, the
∗Corresponding author. Tel.: +1 608 262 107; fax: +1 608 262 7420.
E-mail address: ojg@ahabs.wisc.edu (O.J. Ginther).
0378-4320/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.anireprosci.2005.02.008
274 M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285
biopsy approach was effective for collecting CL tissue for demonstrating dynamic changes in steady-
state levels of mRNAs during PGF2␣-induced luteolysis. Increased Cox-2 mRNA concentrations
suggested that exogenous PGF2␣induced the synthesis of intraluteal PGF2␣. Thus, the findings are
consistent with the concept that an intraluteal autocrine loop augments the luteolytic effect of uterine
PGF2␣in mares.
© 2005 Elsevier B.V. All rights reserved.
Keywords: Corpus luteum; Gene regulation; Mare; Progesterone
1. Introduction
Prostaglandin (PG)F2␣is considered to be the luteolysin in most mammals (reviewed
in Arosh et al., 2004). Surgical removal of the uterus prolongs the life span of the cor-
pus luteum (CL) in sheep, cattle, pigs, horses and some laboratory animals, indicating that
luteolytic PGF2␣is of uterine/endometrial origin in these species (Wiltbank and Casida,
1956; Anderson et al., 1961; Moor and Rowson, 1966; Ginther and First, 1971). Endome-
trial PGF2␣reaches the CL by local, systemic, or a combination of both routes depending
on the species (Bonnin et al., 1999). In mares, several experimental approaches have in-
dicated that the pathway from uterus to ovaries for uterine-induced luteolysis is systemic
(reviewed in Ginther, 1998). In contrast, in primates including women uterine PGF2␣is
not the initiator of luteolysis as hysterectomy does not prolong the lifespan of CL (Neill
et al., 1969; Beling et al., 1970), and it has been postulated that autocrine and paracrine
actions of intraluteal PGF2␣may be involved in initiation of luteolysis (Arosh et al., 2004).
The CL of primates has the capacity to synthesize PGs, and luteal PG synthesis increases
near the time of normal luteolysis (reviewed in Wiltbank and Ottobre, 2003). Luteal PG
production also has been demonstrated in a number of species in which uterine PGF2␣
has a primary role in luteolysis (Wiltbank and Ottobre, 2003). In this regard, production
of PGF2␣by luteal tissue has been reported in cattle (Milvae and Hansel, 1983), sheep
(Rexroad and Guthrie, 1979), pigs (Guthrie et al., 1978) and horses (Watson and Sertich,
1990). Luteal PGF2␣increases after treatment with PGF2␣(Rexroad and Guthrie, 1979;
Guthrie and Rexroad, 1980; Diaz et al., 2000) consistent with a role for luteal PGF2␣in
regression of the CL in cattle, sheep and pigs but apparently has not been studied in horse.
The biochemical events occurring during PGF2␣-induced luteolysis have not been fully
described. One of the hallmarks of luteolysis is a decrease in luteal progesterone production
with an associated decrease in expression of steroidogenic acute regulatory protein (StAR).
StAR facilitates the rate-limiting step in steroidogenesis, transport of cholesterol from the
outer to inner mitochondrial membrane (Watson et al., 2000). After PGF2␣treatment there
is a decrease in mRNA and protein for StAR that corresponds closely to the PGF2␣-induced
decreaseincirculatingprogesterone(Juengel et al., 1995; Tsai et al., 2001). In contrast, treat-
ment with PGF2␣increases a key enzyme in the PG biosynthesis pathway, cyclooxygenase
(Cox) (Tsai and Wiltbank, 1997, 1998). The Cox enzymes catalyze the conversion of arachi-
donic acid to PGH2 which is the first committed step in PG synthesis (reviewed in Wiltbank
and Ottobre, 2003). Attention has focused on the Cox-2 isoform of the enzyme because it
M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285 275
is highly regulated by cytokines and other growth factors (Herschman, 1994). Treatment of
sheep (Rexroad and Guthrie, 1979) and pigs (Diaz et al., 2000) with PGF2␣analogue in-
creased the in vitro production of PGF2␣by luteal tissue and dramatically increased Cox-2
expression. It has been postulated that stimulation of intraluteal Cox-2 mRNA and subse-
quentlythe Cox-2 protein could potentially initiatemechanisms within the CL for intraluteal
amplificationofthePGF2␣signalby producing considerable local concentrations of PGF2␣
(Tsai and Wiltbank, 1997). Thus, two distinct biochemical responses to PGF2␣result in
increased luteal PGF2␣production and decreased luteal progesterone production. In addi-
tion, luteal expression of aromatase was either dramatically increased (pig CL; Diaz and
Wiltbank,2004)ordecreased(miceCL; Stocco, 2004) in response to PGF2␣suggesting that
luteal production of estradiol may also be regulated by PGF2␣. The biochemical and molec-
ular events occurring in the CL during PGF2␣-induced luteolysis have not been studied in
mares.
Transvaginal ultrasound-guided biopsy has been used to collect CL tissue in cat-
tle (Kot et al., 1999; Tsai et al., 2001), but has not been reported in mares. The
collection of luteal samples in mares by transvaginal ultrasound-guided biopsy with
minimal disturbance to luteal function would be useful for clinical and experimental
purposes.
The objective of the present experiment was to characterize the changes in the steady-
state concentrations of five different mRNAs during PGF2␣-induced luteolysis in mares.
The mRNA corresponding to the enzymes involved in the rate-limiting steps in production
of progesterone (StAR), PGF2␣(Cox-2), and estradiol (aromatase) were chosen for anal-
ysis. In addition, the two classical receptors involved in regulating luteal function, PGF2␣
receptor and LH receptor (LHr), were also evaluated. It was hypothesized that in mares an
intraluteal autocrine loop augments the luteolytic effect of exogenous PGF2␣based on the
steady state mRNA concentrations of Cox-2. The effects of collection of luteal tissue by
transvaginal ultrasound-guided biopsy on subsequent progesterone production and size of
CL were also considered.
2. Materials and methods
2.1. Animals and groups
Animals were handled in accordance with the United States Department of Agriculture
Guide for Care and Use of Agricultural Animals in Agricultural Research. Twenty four
mares were used during the ovulatory season (May–September; Northern Hemisphere).
The mares were mixed breeds of ponies, 10–17 years of age, and weighed 300–500 kg. The
feeding program and the technique of transrectal and transvaginal ultrasound scanning have
been described (Ginther, 1995; Gastal et al., 1997). The mares were scanned by ultrasound
daily to record the day of ovulation as determined by the disappearance of a preovulatory
follicle. The treatments were given eight days after ovulation between 20:00 and 21:00h
and time of treatment was designated as Hour 0. The mares were randomized into three
groups: (1) control (no treatment); (2) saline (0.9% NaCl, 1 ml, i.m.); and (3) PGF2␣(1ml
Lutalyse; 5mg/ml, i.m., Pfizer Animal Health, Kalamazoo, MI, USA). Luteal biopsies
276 M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285
were taken from saline and PG-treated mares at Hour 12 but not from the controls, using
a transvaginal ultrasound technique. The diameter and the area of the CL were recorded
in all mares at Hours 0 and 12 and then every 24h until the CL was not detectable. Blood
samples in all mares were collected from Hour 0 to Hour 36 at 12h intervals and then
every 24h until ovulation. The day of ovulation and length of interovulatory interval were
recorded.
2.2. Ultrasound-guided biopsy of CL
The CL biopsies were taken using a transvaginal ultrasound-guided technique as de-
scribed previously for cattle (Kot et al., 1999; Tsai et al., 2001). The biopsy instrument was
a 60 cm long automated spring-loaded device (US Biopsy, Franklin, IN, USA) with an inner
trocar point plunger having a 15mm notch covered by an outer 12ga cutting needle. The
automated spring loaded biopsy instrument was set to ready position so that the specimen
notch was covered by the outer cutting needle. The ovary containing the target CL was
positioned transrectally against the vaginal wall and over the face of a 5.0MHz transvagi-
nal transducer so that the CL was transected by a built-in line on the ultrasound monitor
representing the projected needle path. The biopsy instrument was introduced through the
needle channel of the transvaginal transducer. When the ovary with the CL was in position
against the vaginal wall and transducer face, the biopsy needle was inserted into the CL at
its maximum diameter. The inner plunger was advanced into the CL so that the specimen
notch was within the CL as indicated by the ultrasound image. The biopsy instrument was
fired and withdrawn. The biopsied tissue in the specimen notch was exposed by pressing
the inner plunger. Multiple biopsy attempts were done per mare during each session until
an adequate quantity of luteal tissue was obtained. Each biopsy tract was viewed on the
ultrasound image and was avoided during subsequent biopsies. A successful biopsy was
defined as a partial to full notch of CL tissue. The total mean weight of CL tissue removed
from each mare ranged from 30 to 183mg. The biopsy tissue was frozen in a tube placed
in dry ice and taken to the laboratory for weighing and processing for extraction of mRNA.
2.3. Isolation of mRNA from the CL biopsies
The detailed procedure of mRNA isolation from CL tissue using Magnetight oligo(dT)
beads (Novagen, Madison, WI, USA) has been described (Tsai and Wiltbank, 1998; Tsai
et al., 2001). Briefly, approximately 20mg of luteal tissue was homogenized in 400l lysis
buffer (4M guanidium isothiocyanate, 0.5% (w/v) sarcosyl, 100mM Tris–HCl, pH 8.0 and
1% (w/v) dithiothreitol) using a glass homogenizer. Chromosomal DNA was sheared by
passing the homogenate through a 25-gauge needle 10 times. Two volumes of binding
buffer (100mM Tris–HCl, pH 8.0, 20mM EDTA and 400mM NaCl) were added to the
homogenate. Samples were centrifuged at 16,000 ×gfor 5 min at 4 ◦C to pellet cellular
debris. Supernatant was transferred to a tube containing oligo(dT) beads and allowed to
hybridize for 10 min. Beads were captured on a magnetic stand and washed four times with
500l wash buffer (150 mM NaCl, 10 mM Tris–HCl, pH 8.0 and 1 mM EDTA), and mRNA
was eluted with 30l elution buffer (2mM EDTA) after heating to 65◦C for 5min. The
mRNA was stored at −80◦C until used.
M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285 277
2.4. Quantitative real-time PCR
The mRNA was quantitated using real-time reverse transcription polymerase chain re-
action (RT-PCR). The gene-specific primers were designed according to cDNA sequences
from GenBank using the Primer ExpressTM software (PE Applied Biosystems, Foster City,
CA, USA) (Table 1). The mRNA was reverse transcribed at 42 ◦C for 1h followed by heat-
ing to 95 ◦C for 10min and quick chilling to 4 ◦C as described (Tsai et al., 2001). Real-time
PCR was carried out in a GeneAmp 5700 Sequence Detection System (PE Applied Biosys-
tems, Foster City, CA, USA) using the double-strand DNA-binding dye, Syber Green I
(Molecular Probes, Eugene, OR, USA) for the detection of PCR products. Two microliters
of RT product in the presence of PCR reaction mixture (1×thermophilic buffer, 1.5mM
MgCl2, 0.2mM dNTPs, 0.4 M each of forward and reverse primers, and 1 U Taq DNA
polymerase) were subjected to 40 cycles of amplification by real-time PCR (30s denat-
uration at 95 ◦C, 30 s annealing at 57 ◦C, and 30 s extension at 72◦C). The fluorescence
intensity of the Syber Green I was read after the end of each extension step. The threshold
cycle number (Ct) for each gene was generated by real-time PCR and used to quantify the
relative abundance of each mRNA. Control reactions lacking reverse transcriptase enzyme
were set up to assess any false positives but none were found. A dissociation curve was
created using the built-in melting curve program of the GeneAmp 5700 Sequence Detection
System to confirm the presence of a single PCR product. In some experiments, the real-time
PCR products were also electrophoresed on a 5% polyacrylamide gel and were confirmed
to have a single PCR product of expected size. Messenger RNA for glyceraldehyde-3 phos-
phate dehydrogenase (G3PDH) was also measured in each sample as an internal control
and to normalize the threshold cycle for each sample. The normalized threshold cycle
number, Ct, was calculated as Ct[gene]-Ct[G3PGH]. The relative abundance of the gene
of interest was then evaluated as fold change using the expression 2−Ct, where the
Table 1
Sequences of primer sets used for real time RT-PCR and accession numbers of GeneBank cDNA sequences
Gene Primer sequence Species Size (bp) Accession no.
StAR For TCAACCAGGTCCTTTCGCA Mare 104 AF031696
Rev GCAAGTTGGTCTTTAACACC
Cox-2 For ATCTACCCGCCTCATATTCCT Mare 101 AB041771
Rev CGCAGCCAAATCGTGGCATAC
Aromatase For GTGCCCGAAGTCATGCCTGTC Mare 148 AJ012610
Rev GGAACCGGAGGTGGGAAATGA
PGF2␣receptor For CTTCGAATGGCAACATGGAAT Bovine 105 NM 181025
Rev TCCACAACAGCGTCTGGTACA
LH receptor For TTGCCACATCATCCTATTCTC Mare 122 AY271258
Rev TTCTTTTGTTGGCAAGTTTCT
G3PDH For ACCACTTTGGCATCGTGGAG Mare 76 AF157626
Rev GGGCCATCCACGGTCTTCTG
StAR, steroidogenic acute regulatory protein; Cox-2, cyclooxygenase-2; G3PDH, glyceraldehyde-3 phosphate
dehydrogenase. For, forward; Rev, reverse; and bp, base pairs. All primer sequences are from 5to 3.
278 M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285
value of Ctwas obtained by subtracting the Ctof the lowest control sample from
each Ct.
2.5. Plasma progesterone analysis
Plasma samples were centrifuged (500 ×gfor 20 min), decanted and stored (−20 ◦C)
until assay. Progesterone concentrations in plasma samples were determined using a com-
petitive ELISA that has been described for use in cattle (Rasmussen et al., 1996) in which
the color intensity of the enzyme substrate was inversely proportional to the concentration.
The assay was validated for use with mare plasma in our laboratory. The unknown plasma
samples (200l) were ether extracted and the dried extract was dissolved in 500lof
ELISA assay buffer. Progesterone concentrations were determined in 100l of dissolved
ether extract in duplicate wells for each sample. Serial volumes of a pool of diestrous mare
plasma (100–400l) processed similarly as unknown samples resulted in a displacement
curve that was similar to the standard curve. The intraassay and interassay CVs were 7.0
and 6.1%, respectively, and the sensitivity was 0.03ng/ml.
2.6. Statistical analyses
The data for CL diameter and area and progesterone concentrations were tested for
normality with the Kolmogorov–Smirnov test; when the normality test was significant
(P< 0.05), the data were transformed by natural logarithm. The end points were analyzed
by SAS Mixed procedure (8.2 version; SAS Institute, Cary, NC, USA) with a repeated
measures statement and using mare within treatment as a random effect. Main effects
of group and hour and their interaction were determined. The differences in the relative
expression of mRNA for different genes between PG and saline groups were tested using
unpaired t-tests. A probability of P≤0.05 indicated that the difference was significant and
probabilities between P> 0.05, ≤0.1 indicated that the difference approached significance.
The results are presented as mean±S.E.M.
3. Results
One mare in the saline+ biopsy group was omitted because of a persistant CL for more
than 30 days. Six to 10 biopsy attempts were made per mare during each session and two
to eight attempts were successful. Mean number of biopsy attempts between PG and saline
Fig.1. Mean(±S.E.M.)changesin corpus luteum (CL)diameterandarea,and plasma progesterone concentrations
in mares given no treatment or biopsy (control; n=8), saline+biopsy (n= 7) or PGF2␣+biopsy (n=8). For CL
diameter, there was a main effect of hour (P<0.0001) and a main effect of group that approached significance
(P<0.1). For CL area, there was a main effect of hour (P<0.0001) and an interaction of group by hour that
approached significance (P< 0.08). For plasma progesterone concentrations, there were main effects of group
(P<0.0001) and hour (P<0.0001) and an interaction of group by hour (P<0.0001). An asterisk (*) indicates
first difference between the PGF2␣+biopsy group and control and saline+biopsy groups, and a pound mark (#)
indicates first difference between control and saline+biopsy groups.
M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285 279
280 M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285
groups (7.6 and 8.4 per mare, respectively) and successful biopsy attempts (5.4 and 5.6
per mare, respectively) were not significantly different between groups. The overall suc-
cess rate was 68%. The weight of the individual CL biopsies ranged from 3 to 46mg with
a total mean weight per mare of 30–183mg. Mean changes in the CL diameter and area
and plasma progesterone concentrations and results of the statistical analyses are shown
(Fig. 1). There was a main effect of hour for both CL diameter and area. The CL diameter
and area were similar between saline+ biopsy and control groups throughout the experi-
ment and among the three groups at the time of treatment (Hour 0) and biopsy (Hour 12).
Although the group effect or interaction only approached significance, the CL diameter at
Hour 36 (P< 0.04) and area at Hour 60 (P< 0.01) were smaller in PG +biopsy group than in
the control and saline+ biopsy groups. For plasma progesterone concentrations there were
significant effects of group and hour and an interaction of group by hour. Mean concen-
trations were similar among the three groups at Hour 0. The concentrations were lower
(P< 0.001) at Hour 12 in the PG group than in the other two groups. The progesterone con-
centrations were lower (P<0.009) in saline+ biopsy group at Hour 24 (12 h after biopsy)
than the controls (no biopsy) and continued to remain intermediate in saline +biopsy group
compared to control and PG+ biopsy groups. Progesterone data of individual mares in the
saline group were compared between samples collected prebiopsy (average of Hour 0 and
12 progesterone= 100% for each individual mare) and post-biopsy (saline + biopsy: Hour
24= 56.7 ±5.2%; Hour 36 =51.3±9.2%) and for control mares (Hour 24=98.3 ±3.8%;
Hour 36= 105.3±9.8%); the greatest progesterone decrease in an individual control mare
was 12% compared to decreases of 20–66% in the saline + biopsy group. The interovulatory
interval in three of the biopsied mares (18 days for each mare) was shorter than for all other
mares in the saline+ biopsy group (19–27 days). The interovulatory interval was reduced
(P<0.02) in both PG +biopsy and saline+biopsy groups (18.1±0.6 and 20.1 ±1.2 days,
respectively) compared to control group (23.1±0.8 days).
The relative abundance of mRNA for Cox-2, StAR, LHr, PGF2␣receptor and aromatase
are presented in Table 2. Compared to saline-treated mares, treatment with PGF2␣induced
a significant increase in mRNA for Cox-2, but a significant decrease in mRNA for StAR and
LHr, whereas a decrease in mRNA for aromatase approached significance. There was no
differenceinmRNAexpressionofPGF2␣receptorbetweenPG+biopsyandsaline+ biopsy
groups.
Table 2
Mean±S.E.M. relative expression of mRNAs for different genes in luteal tissue of mares after treatment with
saline or PGF2␣
mRNA Groups
Saline PGF2␣P-value
Cox-2 3.09 ±0.82 9.58 ±1.97 <0.006
StAR 50.04 ±10.82 11.21 ±2.95 <0.001
LH receptor 2.21 ±0.45 0.98 ±0.47 <0.04
PGF2␣receptor 1.52 ±0.79 2.08 ±0.87 NS
Aromatase 3.74 ±1.34 1.60 ±1.04 <0.10
Cox-2, cyclooxygenase-2; StAR, steroidogenic acute regulatory protein; and NS, non significant. All results were
calculated relative to the lowest value in the dataset for each mRNA (fold difference from the lowest value).
M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285 281
4. Discussion
An ultrasound-guided biopsy technique previously used in cattle (Kot et al., 1999; Tsai
et al., 2001) was developed and used in this study to collect CL biopsies in mares. A
mean of eight biopsy procedures were done per mare. The technique was successful in
68% of the attempts. There was no detectable post-biopsy effect on CL diameter or area;
however, the plasma progesterone concentrations in saline-treated mares decreased after
biopsy compared to controls (no biopsy). The progesterone decrease in each mare in the
saline+ biopsy group was greater than for any mare in the control group indicating a uni-
form negative effect of the biopsy technique on progesterone concentrations. In addition,
biopsy alone (saline +biopsy group) appeared to cause a shorter interovulatory interval (20
days); 3 of 7 mares had particularly short interovulatory intervals (18 days) with indications
of premature luteal regression. In cattle (Kot et al., 1999), no detectable effect of luteal
biopsy was observed either on CL diameter or plasma progesterone concentrations and
no premature luteal regressions were observed. The use of a 12-ga biopsy instrument and
more biopsies attempts during a single session in the present study compared to an 18-ga
needle and less biopsy attempts in cattle were probably the causes of the effect of biopsy
on luteal function in the present study. These results indicated that the collection of luteal
tissue by this biopsy method would not be satisfactory for the study of the CL after biopsy.
However, as expected, at the time of biopsy (Hour 12) there was no difference in diameter
of CL among the three groups or plasma progesterone concentrations between control and
saline-treated groups.
Studies in nonequine species indicate that the first committed step in PG synthesis is
catalyzed by Cox enzyme, which converts arachidonic acid to PGH2 (Tsai and Wiltbank,
1997). This is generally considered a rate-limiting step in PG production. The CL has a
rich source of arachidonic acid stored in membrane phospholipids, and this is the primary
precursor of all PGs (reviewed in Arosh et al., 2004). Arachidonic acid is liberated from
phospholipids by cytosolic phospholipase A2 (cPLA2) and Cox-1 and 2 convert arachidonic
acid into PGH2, the common intermediate metabolite for various forms of PGs including
PGF2␣. The downstream enzymes (e.g., PGE and PGF synthases) catalyze the conversion
of PGH2 to PGE2 and PGF2␣(Arosh et al., 2004). The Cox-2 isoform is generally present
in low concentrations under normal physiological conditions (O’Neill and Ford, 1993)but
is stimulated by hormones and growth factors in a variety of cell types (Herschman, 1994).
A variety of these hormonal signals also increase activity of cPLA2 and production of
endogenous arachidonic acid but conversion of this endogenous arachidonic acid to PGs
appears to require Cox-2 enzyme (Karim et al., 1996). Treatment with PGF2␣dramatically
inducesCox-2mRNAinsheepCL and cattle CL and in cultured sheeplargelutealcells(Tsai
and Wiltbank, 1997; Tsai et al., 2001). Induction of intraluteal PGF2␣synthesis in response
to extraluteal PGF2␣may be critical in completion of the luteolytic process. The present
study provides previously unavailable information on the mechanism of PGF2␣action
during luteolysis in mares. Exogenous treatment with PGF2␣induced more than a three-
fold increase in luteal Cox-2 mRNA. This finding indicates that mares are similar to ovine
and bovine species in that exogenous PGF2␣, and presumably uterine PGF2␣, induces the
synthesis of intraluteal PGF2␣. Thus, the hypothesis of presence of an intraluteal autocrine
loop in mares was supported. Studies in other species (Tsai and Wiltbank, 1997) indicate
282 M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285
that the intraluteal PGF2␣is involved in an autocrine loop to amplify and augment the
luteolytic effect of PGF2␣.
TheenzymeStAR has been identified in several species including mares as a rate-limiting
step in steroidogenesis (Watson et al., 2000). This protein is believed to be responsible
for transport of cholesterol from the outer to inner mitochondrial membrane. In cattle CL,
mRNA for StAR was decreased after PGF2␣treatment (Tsai et al., 2001). Similar decreases
in the StAR mRNA have been demonstrated in a number of species, but not in horses, during
natural or induced luteolysis (Juengel et al., 1995; Stocco and Clark, 1996; Townson et al.,
1996; Tsai et al., 2001). Treatment with PGF2␣in mares in the present study resulted in a
decrease in StAR mRNA in luteal tissue. The decrease in StAR mRNA was also reflected
in the decrease in concentrations of circulating progesterone after PGF2␣treatment. In
addition, the decrease in StAR mRNA was reflected in corresponding decreases in the
mRNA for LHr. A decrease in LHr mRNA in response to PGF2␣has been previously
reported in cattle (Tsai et al., 2001) and sheep (Guy et al., 1995), and this decrease may
disrupt LH-stimulated steroidogenesis during luteolysis. In mares, it has been reported that
the number of LH receptors decreased concomitantly with a decrease in serum and luteal
progesterone concentrations during natural (Roser and Evans, 1983) and PGF2␣-induced
(Roser et al., 1982) luteolysis.
In vivo studies showed that PGF2␣treatment decreased the PGF2␣receptor mRNA
in mid-cycle sheep CL (Juengel et al., 1996) and in early and mid-cycle cattle CL (Tsai
and Wiltbank, 1998; Tsai et al., 2001). During physiologic luteal regression, there is also
a dramatic decrease in the PGF2␣receptor mRNA concentration in sheep (reviewed in
Anderson et al., 2001). Similarly, treatment with PGF2␣in vitro decreased PGF2␣receptor
mRNA in ovine large luteal cells (Tsai et al., 1998). In contrast, PGF2␣treatment increased
PGF2␣receptor mRNA expression in rat CL (Olofsson et al., 1996), suggesting species
variation in regulation of PGF2␣receptors. However, there was no difference in expression
of PGF2␣receptor in the present study between saline+ biopsy and PG + biopsy treated
mares. The physiological importance of any changes in PGF2␣receptor expression during
luteolysis and reasons for the differences in regulation among species are not known.
In addition to progesterone, the CL produces estradiol in some species, including pigs,
mares, rodents, possums and primates (Amri et al., 1993; Duffy et al., 2000; Whale et al.,
2003; Diaz and Wiltbank, 2004; Stocco, 2004). In vivo treatment of pigs with PGF2␣in-
creased estradiol production by luteal tissue and this increase was closely associated with
a large increase in aromatase mRNA (Diaz and Wiltbank, 2004). The physiological role of
estradiol production in CL is not clearly understood; however, estradiol has both luteolytic
(Auletta et al., 1976; Duffy et al., 2000) and luteotropic effects (Ford and Christenson, 1991)
in pigs and other species. In rabbits, follicular estradiol is critical for luteal progesterone
production (Bill and Keys, 1983). In pigs, estradiol implants promote greater CL develop-
ment than non-implanted CL (Conley and Ford, 1989). The finding of elevated estradiol
production and increased estradiol receptor expression in regressing CL suggests a role
for estradiol in luteolysis in pigs (Diaz and Wiltbank, 2004). The failure to upregulate these
pathways in porcine Day 9 CL may be responsible, in part, for the inability of PGF2␣to
induce luteolysis at this time in pigs. In mares, an effect of estradiol on the luteal life span
has not been adequately studied. In contrast, aromatase mRNA was reduced by PGF2␣
treatment of mouse CL (Stocco, 2004) suggesting species variation in regulation of luteal
M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285 283
aromatase expression. In the present study in mares, treatment with PGF2␣also decreased
luteal aromatase mRNA; the difference however only approached significance. Previous
studies using immunocytochemistry and Northern blots reported no detectable change in
aromatase protein and mRNA between mid-diestrous and pregnancy in mares (Albrecht et
al., 1997, 2001). Thus, the findings of the present study indicate differences between mares
and pigs but similarities to the mouse in patterns of mRNA expression for aromatase in
CL tissue during luteolysis. The reasons for this species difference are not known; further
studies are indicated in mares.
In summary, mares were treated with PGF2␣eight days after ovulation (Hour 0), and
luteal samples were collected by transvaginal ultrasound-guided biopsy at Hour 12 to
study steady-state concentrations of mRNAs during PGF2␣-induced luteolysis. Exogenous
PGF2␣caused a precipitous decrease in plasma progesterone concentrations by Hour 12 in
all mares. Study of the biopsies at Hour 12 indicated that treatment with PGF2␣decreased
mRNA for StAR and LH receptor and increased mRNA for Cox-2 enzyme; a decrease in
aromatase mRNA approached significance. The results indicated that in mares exogenous
PGF2␣, and presumably uterine PGF2␣, induces the synthesis of intraluteal PGF2␣and are
compatible with the establishment of an intraluteal autocrine loop to amplify and augment
the luteolytic effect of PGF2␣.
Acknowledgements
This work was supported by the Eutheria Foundation (Cross Plains, WI). Project (P1-
MAB-03). The authors thank Pfizer Animal Health for a gift of Lutalyse, US Biopsy for
biopsy needles and Susan Jensen for technical assistance.
References
Albrecht, B.A., Macleod, J.N., Daels, P.F., 1997. Differential transcription of steroidogenic enzymes in the equine
primary corpus luteum during diestrus and early pregnancy. Biol. Reprod. 56, 821–829.
Albrecht, B.A., Macleod, J.N., Daels, P.F., 2001. Expression of 3beta-hydroxysteroid dehydrogenase, cytochrome
p450 17alpha-hydroxylase/17, 20-lyase and cytochrome p450 aromatase enzymes in corpora lutea of diestrous
and early pregnant mares. Theriogenology 55, 551–561.
Amri, H., Gaillard, J.L., Altimimi, I., Silberzahn, P., 1993. Equine ovarian aromatase: evidence for a species
specificity. Biochem. Cell Biol. 71, 296–302.
Anderson, L.E., Wu, Y.L., Tsai, S.-J., Wiltbank, M.C., 2001. Prostaglandin F2␣receptor in the corpus lu-
teum: recent information on gene, messenger ribonucleic acid, and protein. Biol. Reprod. 64, 1041–
1047.
Anderson, L.L., Butcher, R.L., Melampy, R.M., 1961. Subtotal hysterectomy and ovarian function in gilts. En-
docrinology 69, 571–580.
Arosh, J.A., Banu, S.K., Chapdelaine, P., Madore, E., Sirois, J., Fortier, M.A., 2004. Prostaglandin biosynthesis,
transport, and signaling in corpus luteum: a basis for autoregulation of luteal function. Endocrinology 145,
2551–2560.
Auletta, F.J., Caldwell, B.V., Speroff, L., 1976. Estrogen-induced luteolysis in the rhesus monkey: reversal with
indomethacin. Prostaglandins 11, 745–752.
Beling, C.G., Marcus, S.L., Markham, S.M., 1970. Functional activity of corpus luteum following hysterectomy.
J. Clin. Endocr. Metab. 30, 30–39.
284 M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285
Bill, C.H., Keys, P.L., 1983. 17 beta-estradiol maintains normal function of corpora lutea throughout pseudo
pregnancy in hypophysectomized rabbits. Biol. Reprod. 28, 608–617.
Bonnin, P., Huynh, L., Haridon, R.L., Chene, N., Martal, J., 1999. Transport of uterine PGF2␣to the ovaries by
systemic circulation and local lymphovenous-arterial diffusion during luteolysis in sheep. J. Reprod. Fertil.
116, 199–210.
Conley, A.J., Ford, S.P., 1989. Direct luteotrophic effect of oestradiol-17 beta on pig corpora lutea. J. Reprod.
Fertil. 87, 125–131.
Diaz, F.J., Wiltbank, M.C., 2004. Acquisition of luteolytic capacity: changes in prostaglandin F2␣regula-
tion of steroid hormone receptors and estradiol biosynthesis in pig corpora lutea. Biol. Reprod. 70, 1333–
1339.
Diaz, F.J., Crenshaw, T.D., Wiltbank, M.C., 2000. Prostaglandin F2␣induces distinct physiological responses in
porcine corpora lutea after acquisition of luteolytic capacity. Biol. Reprod. 63, 1504–1512.
Duffy, D.M., Chaffin, C.L., Stouffer, R.L., 2000. Expression of estrogen receptor alpha and beta in rhesus monkey
corpus luteum during menstrual cycle: regulation by luteinizing hormone and progesterone. Endocrinology
141, 1711–1717.
Ford, S.P., Christenson, L.K., 1991. Direct effects of oestradiol-17 beta and prostaglandin E-2 in protecting pig
corpora lutea from luteolytic dose of prostaglandin F2␣. J. Reprod. Fertil. 93, 203–209.
Gastal, E.L., Gastal, M.O., Bergfelt, D.R., Ginther, O.J., 1997. Role of diameter differences among follicles in
selection of future dominant follicle in mares. Biol. Reprod. 57, 1320–1327.
Ginther, O.J., 1995. Ultrasonic Imaging and Animal Reproduction: Book 2 Horses. Equiservices Publishing Cross
Plains, WI, USA, pp. 23–88.
Ginther, O.J., 1998. Equine pregnancy: physical interactions between the uterus and conceptus. Proc. Am. Assoc.
Equine Pract. 44, 73–104.
Ginther, O.J., First, N.L., 1971. Maintenance of corpus luteum in hysterectomized mares. Am. J. Vet. Res. 32,
1687–1691.
Guthrie, H.D., Rexroad Jr., C.E., 1980. Blockade of luteal prostaglandin F release in vitro during cloprostenol-
induced luteolysis in pig. Biol. Reprod. 23, 358–362.
Guthrie, H.D., Rexroad Jr., C.E., Bolt, D.J., 1978. In vitro synthesis of progesterone and prostaglandin F by luteal
tissue and prostaglandin F by endometrial tissue from pig. Prostaglandins 16, 433–440.
Guy, M.K., Juengel, J.L., Tandeski, T.R., Niswender, G.D., 1995. Steady state concentrations of mRNA encoding
the receptor for luteinizing hormone during estrous cycle and following prostaglandin F2␣treatment of ewes.
Endocrine 3, 585–589.
Herschman, H.R., 1994. Regulation of prostaglandin synthase-1 and prostaglandin synthase-2. Cancer Metastasis
Rev. 13, 241–256.
Juengel,J.L.,Meberg,B.M., Turzillo, A.M.,Nett,T.M., Niswender, G.D.,1995.Hormonalregulationofmessenger
ribonucleic acid encoding steroidogenic acute regulatory protein in ovine corpora lutea. Endocrinology 136,
5423–5429.
Juengel, J.L., Wiltbank, M.C., Meberg, B.M., Niswender, G.D., 1996. Regulation of steady-state concentrations
of messenger ribonucleic acid encoding prostaglandin F2␣receptor in ovine corpus luteum. Biol. Reprod. 54,
1096–1102.
Karim, S., Habib, A., Levy-Toledano, S., Maclouf, J., 1996. Cyclooxygenase-1 and -2 of endothelial cells utilize
exogenous or endogenous arachidonic acid for transcellular production of thromboxane. J. Biol. Chem. 271,
12042–12048.
Kot, K., Anderson, L.E., Tsai, S.-J., Wiltbank, M.C., Ginther, O.J., 1999. Transvaginal ultrasound-guided biopsy
of the corpus luteum in cattle. Theriogenology 52, 987–993.
Milvae, R.A., Hansel, W., 1983. Prostacyclin, prostaglandin F2␣and progesterone production by bovine luteal
cells during estrous cycle. Biol. Reprod. 29, 1063–1068.
Moor, R.M., Rowson, L.E.A., 1966. Local uterine mechanisms affecting luteal function in the sheep. J. Reprod.
Fertil. 11, 307–310.
Neill, J.D., Johansson, E.D.B., Knobil, E., 1969. Failure of hysterectomy to influence the normal pattern of cyclic
progesterone secretion in the rhesus monkey. Endocrinology 84, 464–465.
O’Neill, G.P., Ford, H.A., 1993. Expression of mRNA for cyclooxygenase-1 and cyclooxygenase-2 in human
tissues. FEBS Lett. 330, 156–160.
M.A. Beg et al. / Animal Reproduction Science 90 (2005) 273–285 285
Olofsson, J.I., Leung, C.H., Bjurulf, E., Ohno, T., Selstam, G., Peng, C., Leung, P.C., 1996. Characterization
and regulation of a mRNA encoding the prostaglandin F2␣receptor in rat ovary. Mol. Cell. Endocrinol. 123,
45–52.
Rasmussen, F.E., Wiltbank, M.C., Christensen, J.O., Grummer, R.R., 1996. Effects of fenprostalene and oestradiol
17benzoate on parturition and retained placenta in dairy cows and heifers. J. Dairy Sci. 79, 227–234.
Rexroad Jr., C.E., Guthrie, H.D., 1979. Prostaglandin F2 alpha and progesterone release in vitro by ovine luteal
tissue during induced luteolysis. Adv. Exp. Med. Biol. 112, 639–644.
Roser, J.F., Evans, J.W., 1983. Luteal luteinizing hormone receptors during postovulatory period in the mare. Biol.
Reprod. 29, 499–510.
Roser, J.F., Evans, J.W., Mikuckis, G.M., Adams, T.E., Hughes, J.P., 1982. Effect of PGF2␣on LH receptors in
the equine corpus luteum. J. Reprod. Fertil. Suppl. 32, 235–245.
Stocco, C., 2004. In vivo and in vitro inhibition of cyp19 gene expression by prostaglandin F2␣in murine luteal
cells: implication of GATA-4. Endocrinology 145, 4957–4966.
Stocco, D.M., Clark, B.J., 1996. Regulation of acute production of steroids in steroidogenic cells. Endocr. Rev.
17, 221–244.
Townson, D.H., Wang, X.J., Keyes, P.L., Kostyo, J.L., Stocco, D.M., 1996. Expression of steroidogenic acute
regulatory protein in corpus luteum of rabbit: dependence upon luteotropic hormone, estradiol-17 beta. Biol.
Reprod. 55, 868–874.
Tsai, S.-J., Wiltbank, M.C., 1997. Prostaglandin F2␣induces expression of prostaglandin G/H synthase-2 in the
ovine corpus luteum: a potential positive feedback loop during luteolysis. Biol. Reprod. 57, 1016–1022.
Tsai, S.-J., Wiltbank, M.C., 1998. Prostaglandin F2␣regulates distinct physiological changes in early and mid-
cycle bovine corpora lutea. Biol. Reprod. 58, 346–352.
Tsai, S.-J., Anderson, L.E., Juengel, J., Niswender, G.D., Wiltbank, M.C., 1998. Regulation of prostaglandin F2␣
and E receptor mRNA by prostaglandin F2␣in ovine corpora lutea. J. Reprod. Fertil. 114, 69–75.
Tsai, S.-J., Kot, K., Ginther, O.J., Wiltbank, M.C., 2001. Temporal gene expression in bovine corpora lutea after
treatment with PGF2␣based on serial biopsies in vivo. Reproduction 121, 905–913.
Watson, E.D., Sertich, P.L., 1990. Secretion of prostaglandins and progesterone by cells from corpora lutea of
mares. J. Reprod. Fertil. 88, 223–229.
Watson, E.D., Thomson, S.R.M., Howie, A.F., 2000. Detection of steroidogenic acute regulatory protein in equine
ovaries. J. Reprod. Fertil. 119, 187–192.
Whale, L.J., Eckery, D.C., Juengel, J.L., 2003. Determination of steroidogenic potential of ovarian cells of the
brushtail possum (Trichosurus vulpecula). Biol. Reprod. 69, 947–958.
Wiltbank, J.N., Casida, L.E., 1956. Alterations of ovarian activity by hysterectomy. J. Anim. Sci. 15, 134–140.
Wiltbank, M.C., Ottobre, J.S., 2003. Regulation of intraluteal production of prostaglandins. Reprod. Biol. En-
docrinol. 1, 91–102.
