The Journal of Clinical Endocrinology & Metabolism Vol. 82, No. 11 3603-3611
Copyright © 1997 by The Endocrine Society
A Therapeutic Role of Prolactin Supplementation in Ovarian Stimulation for in Vitro Fertilization: The Bromocriptine-Rebound Method
Masao Jinno,
Yuuko Katsumata,
Toshihisa Hoshiai,
Yukio Nakamura,
Kazuya Matsumoto and
Yasunori Yoshimura
Department of Obstetrics and Gynecology (M.J., Y.K., T.H., Y.N.),
Kyorin University School of Medicine, Tokyo; the Faculty of
Biology-Oriented Science and Technology (K.M.), Kinki University,
Wakayama; the Department of Obstetrics and Gynecology (Y.Y.), Keio
University School of Medicine, Tokyo, Japan
Address all correspondence and requests for reprints to: Masao Jinno, M.D., Department of Obstetrics and Gynecology, Kyorin University School of Medicine, 6–20-2 Shinkawa, Mitaka City, Tokyo 181, Japan.
 |
Abstract
|
|---|
In a prospective randomized study, we examined whether a novel method
of ovarian stimulation, the bromocriptine-rebound method, improves
in vitro fertilization (IVF) outcomes compared with the
conventional long protocol using GnRH agonist and human menopausal
gonadotropin (hMG). Ovulatory women with previous failed IVF-embryo
transfer using the long protocol were prospectively assigned to either
the bromocriptine-rebound method (group 1, 82 cycles) or the long
protocol (group 2, 80 cycles). The bromocriptine-rebound method was the
same as the long protocol, except that bromocriptine was administered
daily from day 4 of the preceding cycle until 7 days before hMG
stimulation. The numbers of follicles, fertilized oocytes, and embryos
with superior morphology were higher in group 1 than in group 2. The
rates of clinical pregnancy and live birth delivery per cycle were
significantly higher in group 1 (38% and 33%, respectively) than in
group 2 (21% and 19%, respectively). The mean concentration of serum
PRL during hMG administration was significantly higher in group 1 than
group 2. A significant correlation between the number of superior
embryos and PRL concentrations was observed in group 1, but not in
group 2.
Next, we performed a retrospective study to investigate how the
bromocriptine-rebound method exerts its beneficial effects. In the
initial IVF with the long protocol, the mean concentration of serum PRL
during hMG administration and the expression of PRL receptor (PRLr)
messenger ribonucleic acid (mRNA) in granulosa cells were significantly
higher in nonpregnant patients than in pregnant ones. When IVF was
repeated with the bromocriptine-rebound method in the nonpregnant
patients, the expression of PRLr mRNA decreased significantly. In
conclusion, the bromocriptine-rebound method enhances embryonic
development and the rate of live birth delivery in patients with
previous failed IVF using the long protocol. We hypothesize that in the
nonpregnant patients using the long protocol, the serum PRL
concentration and PRLr mRNA expression are increased to compensate for
poor postreceptor responsiveness of granulosa cells to PRL during
oocyte maturation. The bromocriptine-rebound method may improve oocyte
maturation in such patients by restoring postreceptor
responsiveness of granulosa cells to PRL during the hypoprolactinemic
period and increasing the PRL concentration by a rebound phenomenon
after discontinuation of bromocriptine.
|
Introduction
|
|---|
ONE OF THE most crucial factors in
achieving pregnancy in humans by in vitro fertilization
(IVF) is the ability to obtain fully matured oocytes. For this purpose,
various methods of ovarian stimulation have been developed, using
clomiphene citrate (1), human menopausal gonadotropin (hMG) (2), and
GnRH agonists (GnRH-a) (3). Consequently, an adequate pregnancy rate is
achieved for many patients, although there are still patients who do
not respond well to conventional methods. The standard methods of
ovarian stimulation stimulate follicular development almost exclusively
by increasing gonadotropin concentrations through modulation of
endogenous gonadotropin secretion and/or administration of exogenous
gonadotropins. In natural menstrual cycles, however, normal growth and
maturation of oocytes are accomplished by the orchestrated action of
pituitary gonadotropins as primary endocrine regulators as well as many
other endocrine/paracrine/autocrine factors, such as steroids, growth
factors, inhibin/activin/follistatin, and cytokines (4). A
GH/insulin-like growth factor system is viewed as one such regulatory
factor (4); adjuvant GH therapy with gonadotropin stimulation has
recently been attempted, but has shown only minor beneficial effects
(5).
PRL is also known to play a significant role in regulating ovarian
functions, including folliculogenesis, steroidogenesis, ovulation, and
corpus luteum function (6). The PRL and GH receptors as well as the
hormones themselves are homologous, forming a protein family encoded by
related genes (7). Because several clinical and animal investigations
support a stimulatory role for PRL in growth and maturation of oocytes
(8, 9, 10, 11), we have devised a new method of ovarian stimulation, the
bromocriptine-rebound method, in which the endogenous PRL concentration
during gonadotropin stimulation is increased by a rebound phenomenon
after discontinuation of bromocriptine administration (12). Our
clinical trial (12) showed that the bromocriptine-rebound method
significantly improves embryonic development and the pregnancy rate
compared with the conventional long protocol using GnRH-a and hMG,
currently the most widely used method in IVF programs. In this report,
we first conducted a prospective randomized study (study I) to confirm
that the bromocriptine-rebound method improves IVF outcomes compared
with the conventional long protocol. Next, retrospective analyses
(study II) were performed to determine how the bromocriptine-rebound
method exerts its beneficial effects and which patients should respond
to this method.
|
Materials and Methods
|
|---|
Patients and study design
Endocrinologically normal ovulatory women under 40 yr of age
with at least one unsuccessful IVF attempt using the conventional long
protocol of the GnRH-a and hMG regimen, were selected for study I.
Normal concentrations of serum PRL were confirmed at least twice at
intervals more than 3 weeks apart. Patients whose partners had abnormal
semen analyses according to the WHO criteria (13) were excluded.
Patients were prospectively assigned at random to receive either the
bromocriptine-rebound method or the long protocol of the GnRH-a and hMG
regimen for ovarian stimulation. Randomization was performed as
follows: 200 sealed envelopes indicating either method on inside cards
(100 for each method) were shuffled, and 1 envelope was drawn on day 4
of the follicular phase in the preceding cycle. Consequently, 82 and 80
cycles of IVF in 68 and 70 patients were performed using the
bromocriptine-rebound method and long protocol, respectively. The IVF
outcomes and hormonal data were compared between the 2 groups to
examine clinical efficacy and endocrinological effects of the
bromocriptine-rebound method. No significant difference was
observed between the bromocriptine-rebound method and long protocol
groups with respect to mean age (mean ± SEM,
34.1 ± 0.3 and 34.1 ± 0.4 yr, respectively), the number of
previous failed IVF attempts (1.6 ± 0.1 and 1.4 ± 0.1,
respectively), and the distribution of the causes of infertility, which
consisted of tubal infertility (71% and 70%, respectively),
endometriosis (6.1% and 8.8%, respectively), and unexplained
infertility (23% and 21%, respectively).
In study II we retrospectively analyzed data from a total of 149 cycles
of initial IVF attempts using the long protocol of the GnRH-a and hMG
regimen for ovarian stimulation. The study included only
endocrinologically normal ovulatory women younger than 40 yr of age,
whose partners had normal semen analyses. The mean age was 33.0 ±
0.3 yr, and the causes of infertility consisted of tubal infertility,
endometriosis, and unexplained infertility (71.8%, 9.4%, and 18.8%,
respectively). First, hormonal data were compared between pregnant (79
cycles) and nonpregnant (70 cycles) initial IVF attempts. Next, serum
PRL concentrations were compared among 4 groups as follows; group A, in
which initial IVF with the long protocol resulted in an ongoing
pregnancy (47 cycles); group B, in which initial IVF with the long
protocol failed, but repeated IVF with the long protocol resulted in an
ongoing pregnancy (4 cycles); group C, in which initial IVF with the
long protocol failed, but repeated IVF with the bromocriptine-rebound
method resulted in an ongoing pregnancy (15 cycles); and group D, in
which initial IVF with the long protocol and repeated IVF with the long
protocol and/or the bromocriptine-rebound method did not result in an
ongoing pregnancy (28 cycles). The remaining 55 cycles were excluded
from this analysis because they did not repeat IVF. Finally, some
patients were selected at random from the patient population in studies
I and II, and the TRH test and measurements of the expression of PRL
receptor (PRLr) messenger ribonucleic acid (mRNA) in granulosa cells
were performed. The results were compared among 3 groups of patients,
consisting of pregnant and nonpregnant initial IVF with the long
protocol and repeated IVF with the bromocriptine-rebound method (the
TRH test in 7, 9, and 7 patients, respectively, and PRLr mRNA
measurements in 3, 7, and 6 patients, respectively).
Both studies I and II were approved by the Kyorin University ethical
committee.
Ovarian stimulation regimens
In the bromocriptine-rebound method (Fig. 1
), bromocriptine (Parlodel, Sandoz,
Tokyo, Japan) was administered orally at a dose of 1.25 mg/day from
days 4–6 of the follicular phase in the preceding cycle, and
thereafter at 2.5 mg/day. The administration of bromocriptine was
discontinued 7 days before the beginning of hMG administration. Nasal
administration of buserelin acetate (Suprecur, Hoechst, Tokyo, Japan)
(900 µg/day) was started on day 4 of the luteal phase in the
preceding cycle and continued at the same dose until administration of
hCG. Ovarian suppression was confirmed by a serum 17ß-estradiol
concentration of less than 20 pg/mL [conversion factor to Systeme
International (SI) unit, 3.671] and the absence of follicle greater
than 12 mm in diameter on transvaginal ultrasonography. Daily
administration of three ampules of hMG (Humegon, Organon, Tokyo, Japan;
75 IU LH plus 75 IU FSH in each ampule) was begun between days 3 and 10
of the follicular phase in the IVF cycle. Serum concentrations of
17ß-estradiol, PRL, LH, and GH were measured every morning until the
day of oocyte retrieval. When the serum 17ß-estradiol concentration
exceeded 200 pg/mL (conversion factor to SI unit, 3.671), the dosage of
hMG was reduced to two ampules per day. hCG (10,000 IU) was
administered when one or more follicles were greater than 17 mm in
diameter and the serum 17ß-estradiol concentration was above 400
pg/mL (conversion factor to SI unit, 3.671). Oocytes were collected
transvaginally 36 h after hCG administration. The long protocol of
the GnRH-a and hMG regimen (Fig. 1) was similar to the
bromocriptine-rebound method, except that bromocriptine was not
administered.
IVF-embryo transfer (ET) procedures
Oocyte maturity at retrieval was assessed according to the
morphology of the cumulus mass and corona radiata (14). Semen was
washed twice by centrifugation, and motile spermatozoa were collected
by a swim-up technique. Oocytes were inseminated 2–6 h after their
retrieval at a concentration of 100,000 progressively motile
spermatozoa/mL. Human tubal fluid medium (no. 9962, Irvine Scientific,
Santa Ana, CA) with 10% patient serum was used for IVF and embryo
culture as previously reported (15). Oocytes were considered to be
fertilized when two pronuclei were observed 16–18 h after
insemination. Embryos were transferred 38–48 h after insemination.
Morphologically superior embryos were defined as grades 1 and 2
according to Veeck’s criteria (16). A 25-mg dose of progesterone was
administered daily throughout the luteal phase. Clinical pregnancy was
defined as the presence of a gestational sac on an ultrasonogram, and
an ongoing pregnancy was a normal gestation beyond 16 weeks, including
already delivered pregnancies.
Hormone determinations and TRH test
The serum concentrations of 17ß-estradiol, PRL, and GH were
measured by RIA [estradiol kit, Diagnostic Products Corp. (Los
Angeles, CA); Spack-S PRL, Daiichi Radioisotope (Tokyo, Japan); and GH
kit, Daiichi Radioisotope]. The sensitivities of the 17ß-estradiol,
PRL, and GH assays were 10 pg/mL, 1.0 ng/mL, and 0.01 ng/mL (conversion
factors to SI units: 17ß-estradiol, 3.671; PRL, 32.5; GH, 2.000),
respectively. The intra- and interassay coefficients of variation were
5.6% and 6.8% for 17ß-estradiol, 6.3% and 6.9% for PRL, and 5.3%
and 6.9% for GH, respectively. The serum concentrations of LH and hCG
were measured by enzyme immunoassay (IMx LH and IMx hCG, Dainabot,
Tokyo, Japan). The sensitivities of the LH and hCG assays were 0.5
and 0.7 IU/L, respectively. The intra- and interassay coefficients of
variation were 4.7% and 4.7% for LH and 4.2% and 4.7% for hCG,
respectively.
The TRH test was performed 3 days before the beginning of hMG
administration. Blood samples for PRL measurements were drawn just
before and 15, 30, 60, and 120 min after the iv administration of 500
µg TRH (Hirtonin, Takeda, Tokyo, Japan). The maximal PRL levels and
relative increment, i.e. maximal PRL/basal PRL, were
analyzed.
Granulosa cell preparation
Granulosa cells were harvested from follicles larger than 16 mm
in diameter at oocyte retrieval. Follicular aspirates were centrifuged
at 1500 rpm for 10 min, and the supernatants were decanted. Cells were
resuspended in Dulbecco’s phosphate-buffered saline (PBS; 14287–080,
Life Technologies, Tokyo, Japan) and centrifuged at 1500 rpm for 10
min. Cells were resuspended in 4 mL PBS, layered on 4 mL 50% Percoll,
and centrifuged at 1500 rpm for 30 min to pellet red blood cells. A
purified granulosa cell preparation was aspirated from the interface
and washed three times in PBS by centrifugation at 1500 rpm for 10 min.
The cell pellet was plunged into liquid nitrogen and stored until later
RNA analysis.
Quantitative RNA analysis
In humans, two different PRLr isoforms have been identified
(17). For detection of human PRLr mRNA by reverse transcription-PCR
(RT-PCR), two oligonucleotide primers (forward primer,
5'-ACTATGAGGACTTGCTGGTGGAGTATTT-3'; reverse primer,
5'-CACTTGCTTGATGTTGCAAGTGAAGTT-3') were used as described previously
(17).
Total RNA was isolated from granulosa cells using acid guanidium
thiocyanate-phenol-chloroform (18), then digested with
ribonuclease-free pancreatic deoxyribonuclease I (Takara Shuzo Co.,
Kyoto, Japan) at 37 C for 10 min. Subsequently, total RNA was extracted
with phenol-chloroform, precipitated, and resuspended. One microgram of
total RNA was reverse transcribed at 37 C using random 9-mer (Takara
Shuzo Co.) and avian myeloblastosis virus reverse transcriptase XL
(Takara Shuzo Co.) in a 40-µL reaction solution supplemented with 40
U ribonuclease inhibitor (Takara Shuzo Co.). One microliter of RT
products was amplified by PCR. The reaction was carried out in Takara
Ex Taq buffer (Takara Shuzo Co.), 200 µmol/L deoxy-NTPs, 50 µmol/L
forward and reverse primers, and 1.25 U Takara Ex Taq polymerase
(Takara Shuzo Co.) in a total volume of 50 µL overlaid with light
mineral oil. The amplification profile was 1 min each of denaturation
at 94 C, annealing at 55 C, and extension at 72 C after initial
denaturation at 94 C for 5 min.
The procedures for quantitative PCR were essentially based on those
reported by Martini et al. (19) with some modifications. In
preliminary experiments, we tested various amplification cycles using 1
µg total RNA from human granulosa cell samples to determine
appropriate conditions for quantitative PCR (data not shown). The
exponential phase of the reaction lasted up to 36 cycles, followed by a
plateau phase. Thus, PCR was carried out for 33 cycles throughout this
study. For quantitative analysis of PCR products, 10 µL PCR products
were separated on 2% agarose gels and stained with ethidium bromide.
The pixel count of stained DNA bands was quantified on a BioImage
analyzer (Millipore Corp., Ann Arbor, MI).
Statistical analysis
Data were analyzed using the 
2 square
test, Fisher’s exact test, ANOVA, Fisher’s protected least
significant difference (PLSD) test, Student’s t test, or
Pearson analysis as appropriate. P < 0.05 was
considered significant. Results are presented as the mean ±
SEM unless otherwise stated.
|
Results
|
|---|
Study I
Significantly more follicles developed with the
bromocriptine-rebound method than with the long protocol (Table 1). The serum concentration of
17ß-estradiol on the day of hCG administration was significantly
higher with the bromocriptine-rebound method than with the long
protocol. The numbers of fertilized oocytes and cleaved embryos were
significantly higher with the bromocriptine-rebound method than with
the long protocol (Table 2).
Significantly more embryos had superior morphology consistent with
grades 1 and 2 of Veeck’s classification (14) with the
bromocriptine-rebound method than with the long protocol. Consequently,
the rates of clinical pregnancy and live birth delivery were
significantly higher with the bromocriptine-rebound method than with
the long protocol. There were no significant differences between
bromocriptine-rebound method and long protocol groups in the quality of
semen used for IVF: volume, 3.7 ± 0.2 vs. 3.4 ±
0.2 mL; sperm concentration, 7.4 ± 0.4 vs. 7.1 ±
0.3 x 107/mL; and motility, 74.1 ± 1.8
vs. 73.2 ± 2.2%, respectively.
View this table:
[in this window]
[in a new window]
|
Table 1. Follicular development and hMG requirement with the
bromocriptine-rebound method and the long protocol of GnRH-a and hMG
regimen
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. IVF outcomes with the bromocriptine-rebound method
and the long protocol of GnRH-a and hMG regimen for ovarian stimulation
|
|
The concentration of PRL decreased gradually from the luteal phase of
the preceding cycle until day -4 (P < 0.0001, by
paired t test) and then increased rapidly until day +2
(P < 0.0001, by paired t test) in the long
protocol (Fig. 2). In the
bromocriptine-rebound method, the concentration of PRL significantly
decreased with bromocriptine treatment, but increased significantly
beyond the pretreatment value after discontinuation of bromocriptine.
The concentration of PRL then gradually decreased between the initial
day of hMG administration and day -5 (P < 0.0001, by
paired t test) and increased rapidly between days -3 and +2
(P < 0.0001, by paired t test). All PRL
values from 4 days after the discontinuation of bromocriptine until day
+2 were significantly elevated compared with that before bromocriptine
treatment. The concentration of PRL on any day from the initial day of
hMG administration until day 0 was significantly higher in the
bromocriptine-rebound method than in the long protocol. The mean
concentration of PRL during administration of hMG (from the initial day
of hMG until day 0) was significantly higher in the
bromocriptine-rebound method than in the long protocol (13.0 ±
1.0 and 8.4 ± 0.5, respectively; P < 0.0001, by
unpaired t test).
View larger version (18K):
[in this window]
[in a new window]
|
Figure 2. Serum concentrations of PRL (top
panel), GH (middle panel), and 17ß-estradiol
(bottom panel) during ovarian stimulation with the
bromocriptine-rebound method (•—•; 82 cycles) and the long
protocol (x----x; 80 cycles). In the
bromocriptine-rebound method, bromocriptine was administered from the
follicular phase in the preceding cycle (Pre F) until 7 days before the
beginning of hMG treatment (Last Bro). In both methods, buserelin was
administered from the luteal phase in the preceding cycle (Pre L) until
the day of hCG (0 d). *, P < 0.05 between two
methods (by unpaired t test); #, P
< 0.0001 vs. any other PRL data in
bromocriptine-rebound method (by paired t test); §,
P < 0.0001 vs. PRL value on Last
Bro in bromocriptine-rebound method (by paired t test).
To convert values for 17ß-estradiol to picomoles per L, multiply by
3.671.
|
|
With the bromocriptine-rebound method and the long protocol, the serum
concentration of GH decreased from the luteal phase of the preceding
cycle until the initial day of hMG administration (P <
0.0001, by paired t test), increased from days -6 and -4,
respectively, until day 0 (P < 0.0001, by paired
t test), and dropped rapidly on day +1 (P <
0.0001, by paired t test) (Fig. 2
). Between the
bromocriptine-rebound method and long protocol groups, however, there
was no significant difference in the GH concentration on any day of
observation (P > 0.05, by unpaired t test).
The concentrations of 17ß-estradiol in the luteal phase of the
preceding cycle as well as on any day from day -6 until day 0 were
significantly higher with the bromocriptine-rebound method than with
the long protocol (Fig. 2). The serum concentration of LH was not
significantly different between the two methods on any day of
observation (P > 0.05, by unpaired t test;
data not presented).
A significant correlation between the serum concentration of PRL and
the number of morphologically superior embryos was observed the day
before, the day of, and the day after hCG administration with the
bromocriptine-rebound method (r = 0.27, 0.24, and 0.36,
respectively; P < 0.05, by Pearson analysis), but not
on any day of PRL measurement with the long protocol (Fig. 3).
View larger version (24K):
[in this window]
[in a new window]
|
Figure 3. Correlations between the number of
morphologically superior embryos and the serum concentration of PRL the
day after hCG administration in the bromocriptine-rebound method
(top panel) and the long protocol (bottom
panel); the regression line and 95% confidence interval are
shown.
|
|
Study II
The serum concentrations of PRL, GH, and 17ß-estradiol were
compared retrospectively between pregnant and nonpregnant initial IVF
attempts using the long protocol (Table 3). The mean concentration of PRL during
hMG administration was significantly higher in nonpregnant cycles than
in pregnant ones, whereas there was no significant difference in the
mean concentrations of GH during hMG administration.
View this table:
[in this window]
[in a new window]
|
Table 3. Comparison of serum concentrations of PRL, GH, and
17ß-estradiol between pregnant and nonpregnant initial IVF attempts
using the long protocol of GnRH-a and hMG regimen
|
|
The serum concentrations of PRL in initial IVF attempts with the long
protocol were compared among four groups, into which patients were
classified according to which of methods, bromocriptine-rebound method
or long protocol, resulted in an ongoing pregnancy in the initial and
repeated IVF attempts. The PRL concentration 3 days after the onset of
menstruation in the IVF cycle was significantly higher in group C, in
which the bromocriptine-rebound method was effective, than in groups A
and B, in which the long protocol was effective (Fig. 4).
View larger version (17K):
[in this window]
[in a new window]
|
Figure 4. Comparison of the serum PRL concentration 3
days after the onset of menstruation in initial IVF cycle with long
protocol. An ongoing pregnancy was achieved with the initial long
protocol (group A), repeated long protocol (group B), and repeated
bromocriptine-rebound method (group C), but with neither the long
protocol nor the bromocriptine-rebound method (group D). Data were
analyzed by ANOVA (P = 0.01) and then Fisher’s
PLSD test (*, P < 0.05 vs. group A
or B; **, P < 0.05 vs. group A).
|
|
There were no significant differences among three groups of pregnant
and nonpregnant initial IVF with the long protocol and repeated IVF
with the bromocriptine-rebound method in the maximal PRL level
(39.7 ± 4.4, 43.7 ± 4.0, and 47.1 ± 5.5 ng/mL,
respectively) and the relative increment (6.4 ± 1.3, 8.2 ±
1.1 and 4.8 ± 0.4, respectively) of the TRH test.
This study used oligonucleotide primers matching the sequences of the
intracytoplasmic domains of both the long and intermediate isoforms of
human PRLr. The expression of long PRLr mRNA was detected in human
granulosa cells, whereas the intermediate PRLr mRNA transcript was not
(Fig. 5). The level of expression of long
PRLr mRNA differed among patients (Fig. 5), and further quantitative
analysis was performed (Fig. 6). In the
initial IVF attempts with the long protocol, the expression of long
PRLr mRNA was significantly higher in nonpregnant patients than in
pregnant ones. When IVF was repeated with the bromocriptine-rebound
method in nonpregnant patients with the long protocol, expression of
long PRLr mRNA decreased significantly.
View larger version (46K):
[in this window]
[in a new window]
|
Figure 5. RT-PCR analysis of human PRLr mRNA in
granulosa cells harvested at oocyte retrievals in pregnant and
nonpregnant initial IVF patients with the long protocol (no. 4 and 11
and no. 3, 6, 7, 9, and 12, respectively) and repeated IVF patients
with the bromocriptine-rebound method after unsuccessful initial IVF
with the long protocol (no. 1, 2, 5, 8, 10, and 13). mRNA of long human
PRLr (893 bp) (17) was detected, but intermediate human PRLr mRNA (320
bp) (17) was not. Human ß-actin mRNA transcripts in granulosa cells
of each patient are also shown in the lower panel.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Figure 6. Comparisons of the expression of human long
PRLr mRNA by quantitative RT-PCR analysis in granulosa cells, which
were harvested at the time of oocyte retrieval in pregnant and
nonpregnant initial IVF patients with the long protocol (LP-p and LP-n,
respectively) and in repeated IVF patients with the
bromocriptine-rebound method after unsuccessful initial IVF with the
long protocol (BR). Data were analyzed by ANOVA (P
= 0.006) and then Fisher’s PLSD (*, P < 0.01
vs. LP-p or BR).
|
|
|
Discussion
|
|---|
This study demonstrated that in patients who previously failed in
IVF-ET using the long protocol, the bromocriptine-rebound method
improves folliculogenesis and subsequent embryonic development,
resulting in a significant increase in the rate of live birth delivery.
Nonpregnant patients with the long protocol are characterized by
increased serum PRL concentrations and enhanced PRLr mRNA expression in
granulosa cells. The bromocriptine-rebound method increases the serum
PRL concentrations further and decreases the expression of PRLr mRNA in
granulosa cells, and these changes appear to be associated with the
beneficial effects of the rebound method.
A midcycle increase in serum PRL concentrations and PRL receptors in
granulosa cells has been demonstrated in humans (6), and abnormal
follicular development was observed in a woman with isolated PRL
deficiency (20). These observations suggest that PRL plays a
physiological role in human folliculogenesis. In human IVF, high
concentrations of PRL in follicular fluid are associated with
maturation of the oocyte-cumulus complex, successful fertilization, and
pregnancy (8). Hypoprolactinemia, induced with continuous
administration of bromocriptine, is associated with lower rates of
fertilization (10, 11) and embryonic cleavage (11). Direct
supplementation of PRL into the medium for in vitro
maturation of rabbit oocytes enhances subsequent embryonic development
in a dose-dependent manner (9). Therefore, PRL appears to play a
stimulatory role in the growth and maturation of oocytes, promoting
fertilization and embryonic development.
Oocyte maturation consists of nuclear, cytoplasmic, and membranous
changes; the completion of these maturational changes confers upon the
oocyte the capacity for normal fertilization and embryonic development
(21). Nuclear maturation can be assessed to some extent by the
observation of a first polar body, whereas no reliable, noninvasive
markers of cytoplasmic and membrane maturation have been established to
date, other than observation of fertilization and embryonic
development. In our previous study (12), the number of oocytes with a
first polar body was similar between the bromocriptine-rebound method
and the long protocol. It is often observed in humans as well as in
other mammals that inadequate maturation of cytoplasm and membrane
leads to fertilization failure and impaired embryo viability, even with
completion of nuclear maturation (21). The bromocriptine-rebound method
probably improves cytoplasmic and membrane maturation, resulting in an
increase in embryo viability. In this study, assessment of oocyte
maturity according to the morphology of the cumulus mass and corona
radiata does not appear to be sensitive enough to detect an improvement
in oocyte maturation due to the bromocriptine-rebound method.
To explain the results of the present study, we propose the hypothesis
illustrated in Fig. 7. Patients with poor
IVF outcomes using the long protocol may have low responsiveness of
granulosa cells to PRL, possibly in the postreceptor pathway, causing
compensatory increases in the levels of serum PRL, PRLr mRNA, and PRLr.
It is likely, however, that the compensation is inadequate, so that
oocyte maturation is impaired. In such patients, the
bromocriptine-rebound method probably improves oocyte maturation by
restoring postreceptor responsiveness of granulosa cells to PRL during
the hypoprolactinemic period and increasing the serum PRL concentration
by a rebound phenomenon after discontinuation of bromocriptine. A
decrease in the expression of PRLr mRNA appears to reflect the recovery
of postreceptor responsiveness to PRL. Testing this hypothesis clearly
requires a direct analysis of PRLr levels. In preliminary experiments,
however, we have failed to detect the amount of PRLr in human granulosa
cells by means of autoradiography using [125I]PRL as well
as by flow cytofluorometry and immunocytochemistry using a monoclonal
antirat PRLr antibody, U5, which has been shown to cross-react with
human PRLr (22). It is not clear why we could not detect PRLr, but
there are some possible reasons other than methodological ones. A
single class of PRLr with high binding affinity has been demonstrated
by Scatchard analysis in human granulosa cells, but the number of
binding sites is limited (23). [125I]PRL binding to
granulosa is markedly reduced at the time of and after the gonadotropin
surge in hamsters (24). Thus, the timing of granulosa cell retrieval in
the present study may not have been appropriate for PRLr measurements
because it was at the nadir of PRLr expression. Because changes in PRLr
mRNA expression correspond roughly to those of PRLr (25), and RT-PCR
quantitation of mRNA is much more sensitive than methods to measure
receptors, we analyzed PRLr mRNA expression in this study.
View larger version (30K):
[in this window]
[in a new window]
|
Figure 7. A hypothesis for the mechanism of action of
the bromocriptine-rebound method to improve IVF outcome.
|
|
Long, intermediate, and short isoforms of PRLr have been identified in
the rat, which differ in the lengths of their intracytoplasmic domains
(7). In humans, originally only a long PRLr isoform was identified (7),
but recently an intermediate PRLr isoform was demonstrated in human
breast carcinoma and lymphocytes (17). In this study, however, long
PRLr mRNA was detected in human granulosa cells, whereas the
intermediate PRLr mRNA was not. This result agrees with the observation
that in rat granulosa cells, the expression of long PRLr mRNA is 6- to
10-fold higher than that of short PRLr mRNA (26).
The subjects in study I were not a special small subgroup such as poor
IVF responders, but were a large population of patients whose previous
IVF attempts using the long protocol had not achieved a viable
pregnancy. Thus, PRL hyporesponders may be common and have not been
identified simply because of the lack of a method to detect this
disorder. However, it is more likely that PRL hyporesponsiveness in
these patients is latent, i.e. compensated, in natural
menstrual cycles, and that some factors in ovarian stimulation
aggravate this condition to make it overt. Estrogens have been shown to
stimulate PRL synthesis, storage, and secretion at the pituitary level
(27) as well as to induce PRL receptors on effector cells (6, 7, 27).
It has been shown that PRL concentrations increase proportionally with
the preovulatory increase in the LH pulse frequency before a
significant increase in estradiol secretion (6). This suggests a
stimulatory mechanism affecting PRL secretion that is more directly
related to LH release (6, 27). Therefore, it is conceivable that
hypogonadotropic hypogonadism by pituitary desensitization in the long
protocol causes a decrease in PRL hormone and/or receptor levels,
aggravating PRL hyporesponsiveness. However, the significance of overt
PRL hyporesponsiveness as a cause of infertility in the general
population remains to be seen.
This study has demonstrated excellent effectiveness of a new method of
ovarian stimulation, the bromocriptine-rebound method, in a
prospective randomized trial in patients who previously failed in
IVF-ET using the long protocol. The results are profoundly significant
in offering a new approach to more optimal stimulation of
folliculogenesis by coordinated modulation of gonadotropin and other
endocrine hormones. In addition, the significance of follicular
hyporesponsiveness to PRL as a new cause of infertility has been
suggested and awaits further investigation.
Received March 6, 1997.
Revised July 16, 1997.
Accepted July 21, 1997.
|
References
|
|---|
-
Trounson AO, Leeton JF, Wood C, Webb J, Wood
J. 1981 Pregnancies in humans by fertilization in vitro
and embryo transfer in the controlled ovulatory cycle. Science. 212:681–682.[Abstract/Free Full Text]
-
Jones Jr HW, Jones GS, Andrews MC, et al. 1982 The
program for in vitro fertilization at Norfolk. Fertil
Steril. 38:14–21.[Medline]
-
Porter RN, Smith W, Craft IL, Abdulwahid NA, Jacobs
HS. 1984 Induction of ovulation for in-vitro fertilisation using
buserelin and gonadotropins. Lancet. 2:1284–1285.
-
Gougeon A. 1996 Regulation of ovarian follicular
development in primates: facts and hypotheses. Endocr Rev. 17:121–155.[CrossRef][Medline]
-
Shulman A, Maymon R, Bahary C, Ben-Nun I, Friedler C,
Dor J. 1993 Growth hormone: non multum sed multa (quality, not
quantity). Int J Fertil. 38:289–295.
-
McNeilly AS. 1984 Prolactin and ovarian function.
In: Muller EE, MacLeod RM, eds. Neuroendocrine perspectives. Amsterdam:
Elsevier; vol 3:279–316.
-
Kelly PA, Djiane J, Postel-Vinay M, Edery M. 1991 The prolactin/growth hormone receptor family. Endocr Rev. 12:235–251.[CrossRef][Medline]
-
Laufer N, Botero-Ruiz W, DeCherney AH, Haseltine F,
Polan ML, Behrman HR. 1984 Gonadotropin and prolactin levels in
follicular fluid of human ova successfully fertilized in
vitro. J Clin Endocrinol Metab. 58:430–434.[Abstract]
-
Yoshimura Y, Hosoi Y, Iritani A, Nakamura Y, Atlas SJ,
Wallach EE. 1989 Developmental potential of rabbit oocyte matured
in vitro: the possible contribution of prolactin. Biol
Reprod. 40:26–33.
-
Gonen Y, Casper RF. 1989 The influence of transient
hyperprolactinemia on hormonal parameters, oocyte recovery, and
fertilization rates in in vitro fertilization. J In
Vitro Fertil Embryo Transf. 6:155–159.
-
Oda T, Yoshimura Y, Takehara Y, et al. 1991 Effects
of prolactin on fertilization and cleavage of human oocytes. Horm Res. 35:33–38.
-
Jinno M, Yoshimura Y, Ubukata Y, Nakamura Y. 1996 A
novel method of ovarian stimulation for in vitro
fertilization: bromocriptine-rebound method. Fertil Steril. 66:271–274.[Medline]
-
World Health Organization. 1987 WHO laboratory
manual for examination of human semen and semen-cervical mucus
interaction. Cambridge: Cambridge University Press; 1–64.
-
Hill GA, Freeman M, Bastias MC, et al. 1989 The
influence of oocyte maturity and embryo quality on pregnancy rate in a
program for in vitro fertilization-embryo transfer. Fertil
Steril. 52:801–806.[Medline]
-
Jinno M. 1986 Comparison of media used for human
in vitro fertilization and embryo transfer programs: a new
method of serum preparation. Acta Obstet Gynaecol Jpn. 38:102–110.
-
Scott RT, Hofmann GE, Veeck LL, Jones Jr HW, Muasher
SJ. 1991 Embryo quality and pregnancy rates in patients attempting
pregnancy through in vitro fertilization. Fertil Steril. 55:426–428.[Medline]
-
Clevenger CV, Chang W, Ngo W, Pasha TLM, Montone KT,
Tomaszewski JE. 1995 Expression of prolactin and prolactin
receptor in human breast carcinoma. Am J Pathol. 146:695–705.[Abstract]
-
Chomoczyski P, Sacchi N. 1987 Single-step method of
RNA isolation by acid guanidium thiocyanate-phenol-chloroform
extraction. Anal Biochem. 162:156–159.[Medline]
-
Martini J, Villares SM, Nagano M, et al. 1995 Quantitative analysis by polymerase chain reaction of growth hormone
receptor gene expression in human liver and muscle. Endocrinology. 136:1355–1360.[Abstract]
-
Kauppila A, Chatelain P, Kirkinen P, Kivinen S, Ruokonen
A. 1987 Isolated prolactin deficiency in a woman with puerperal
alactogenesis. J Clin Endocrinol Metab. 64:309–312.[Abstract]
-
Plachot M, Mandelbaum J. 1990 Oocyte maturation,
fertilization and embryonic growth in vitro. Br Med Bull. 46:675–694.[Abstract/Free Full Text]
-
Dardenne M, Leite de Moraes M, Kelly PA, Gagnerault
M. 1994 Prolactin receptor expression in human hematopoietic
tissues analyzed by flow cytofluorometry. Endocrinology. 134:2108–2114.[Abstract]
-
Ben-David M, Schenker JG. 1982 Human ovarian
receptors to human prolactin: implications in infertility. Fertil
Steril. 38:182–186.[Medline]
-
Oxberry BA, Greenwald GS. 1982 An autoradiographic
study of the binding of 125I-labeled follicle-stimulating
hormone, human chorionic gonadotropin and prolactin to the hamster
ovary throughout the estrous cycle. Biol Reprod. 27:505–516.[CrossRef][Medline]
-
Jolicoeur C, Boutin J, Okamura H, Raguet S, Djiane J,
Kelly PA. 1989 Multiple regulation of prolactin receptor gene
expression in rat liver. Mol Endocrinol. 3:895–900.[CrossRef][Medline]
-
Clarke DL, Arey BJ, Linzer DIH. 1993 Prolactin
receptor messenger ribonucleic acid expression in the ovary during the
rat estrus cycle. Endocrinology. 133:2594–2603.[Abstract]
-
Lamberts SWJ, Macleod RM. 1990 Regulation of
prolactin secretion at the level of the lactotroph. Physiol Rev. 70:279–318.[Free Full Text]
Top
Abstract
Introduction
