The aromatase inhibitor anastrozole is associated

with favorable embryo development and implantation

markers in mice ovarian stimulation cycles
Öznur Karaer, M.D.,a H. Seda Vatansever, M.D.,c Semra Oruç, M.D.,b Kemal Özbilgin, M.D.,c
Serap Cilaker, M.D.,c and M. Faik Koyuncu, M.D.b
a
Free-attending Obstetrics and Gynecology Specialist, Mersin, Turkey; and b Departments of Obstetrics and Gynecology and
c
Histology and Embryology, School of Medicine, Celal Bayar University, Manisa, Turkey

Objective: To investigate the embryonic and endometrial effects of anastrozole in preimplantation and implan-
tation phases in FSH-induced cycles in mice.
Design: Blind randomized study.
Setting: University research laboratory.
Animal(s): Twenty-seven mature female mice.
Intervention(s): Single-dose anastrozole (25 mg/kg [0.75 mg]), recombinant FSH (5 IU/mL), and hCG (5 IU/mL)
(n ⫽ 9); recombinant FSH (5 IU/mL) and hCG (5 IU/mL) (n ⫽ 9); or sterile saline (1 mL) (n ⫽ 9). The morning
of finding the vaginal plug was designated as day 1 of embryonic development (E1). Three mice from each group
were sacrificed on E1 and embryos aspirated from uterine tubes. The rest of the mice were sacrificed on E2.5–3
and uteruses removed.
Main Outcome Measure(s): Embryo quality, endometrial histologic evaluation, and immunohistochemical
analysis of tumor necrosis factor-␣, leukemia inhibitory factor, laminin, and collagen IV staining.
Result(s): Anastrozole use in FSH-induced cycles not only caused an increase in preimplantation receptivity and
implantation but also supported release of implantation markers. The enhanced embryo development seen in this
study would explain the higher implantation because embryo development is synchronized with endometrial
development.
Conclusion(s): In mice, the use of anastrozole in FSH-induced cycles has a positive effect on embryo quality and
implantation. This effect might be species dependent, and human studies are needed. (Fertil Steril威 2005;83:
1797– 806. ©2005 by American Society for Reproductive Medicine.)
Key Words: Anastrozole, aromatase inhibitor, ovulation induction, implantation, mouse embryo

In infertile women, ovarian stimulation can achieve a
fertility rate that is approximately equal to or in some
anovulation types more than that for normal populations.
Ovarian stimulation is the first choice for gaining ovula-
tory cycles in anovulatory conditions like polycystic
ovary syndrome (1, 2). It is also used in ovulatory infer-
tility conditions, such as endometriosis (3) and unex-
plained infertility (4, 5), to augment ovulation. Ovulation
stimulation is also used in conjunction with assisted re-
productive technologies to produce several mature folli-
cles (controlled ovarian stimulation) (6).
Clomiphene citrate (CC) is the first-line therapy in ovula-
tion induction. In CC failures, gonadotropins (predominantly
recombinant FSH) have been used for ovarian hyperstimu-
lation (7). Multiple follicular development, resulting in ovar-
ian hyperstimulation syndrome and multiple pregnancies,
and high treatment costs are major drawbacks of gonadotro-
pin treatment (8 –11).
Received July 23, 2004; revised and accepted January 29, 2005.
Reprint requests: Dr. Öznur Karaer, Guvenevler M. 1905 Sok. Cenk
Apt. 5/14, Pozcu/Mersin, Turkey (FAX: 90-324-237 09 32; E-mail:
oznurkaraer@yahoo.com).
Aromatase (a product of the CYP 19 gene) is a cyto-
chrome P450 hemoprotein-containing enzyme complex
that catalyzes the rate-limiting step in the production of
estrogens: the conversion of androstenedione and T
through three hydroxylation steps to estrone and E2 (12).
Aromatase is a good target for selective inhibition of
estrogen (E) biosynthesis because it is the terminal step in
biosynthetic sequence. Aromatase inhibitors have been
used in clinical applications over the last 20 years (13).
The lack of specificity and the unfavorable toxicity profile
of first-generation aromatase inhibitors have led to re-
search for more selective aromatase inhibitors. Third-
generation aromatase inhibitors, including anastrozole
and letrozole, have been approved for the treatment of
postmenopausal breast cancer (14). These aromatase in-
hibitors have a short half-life (approximately 45 hours),
few side effects, and more potency (15).
Mitwally and Casper (16, 17) have suggested aromatase
inhibitor use in ovulation stimulation as an alternative to
CC treatment. They have hypothesized that in ovarian
stimulation, administration of aromatase inhibitors, which
have a short half-life and do not possess the adverse
antiestrogenic effects of CC in the early part of the

0015-0282/05/$30.00 Fertility and Sterility姞 Vol. 83, No. 6, June 2005 1797
doi:10.1016/j.fertnstert.2005.01.100 Copyright ©2005 American Society for Reproductive Medicine, Published by Elsevier Inc.
menstrual cycle, might suppress E production, release the
hypothalamic–pituitary axis from E-negative feedback,
and increase gonadotropin secretion, resulting in stimula-
tion of ovarian follicles.
Despite progress in assisted reproduction technologies,
the lack of control of implantation remains a major ob-
stacle to obtaining successful pregnancies. Embryo im-
plantation is a complex event involving apposition, fol-
lowed by the adhesion of the blastocyst to the maternal
endometrium, and finally the invasion of this endome-
trium (18).
Tumor necrosis factor-␣ (TNF-␣) is a marker that is
expressed from different layers of trophoblasts and is
classified as one of the most important cytokines taking
part in implantation (19). The expression of leukemia
inhibitory factor (LIF) and LIF receptor (LIFR) proteins
and pinopodes in endometrial samples from healthy
women suggests that both molecular and structural cell
changes are important in the initiation of human blasto-
cyst implantation (20). Laminin-1, a multifunctional gly-
coprotein of the basement membrane, is thought to be
important in embryogenesis, embryonic implantation, and
placentation (21). Collagen IV and laminin reactivity in-
crease in the basal lamina and underlying subepithelial
stroma as pregnancy proceeds (22).
The present study was undertaken to determine the em-
bryonic and endometrial effects of aromatase inhibitors in
preimplantation and implantation phases of FSH-induced
cycles in mice.
MATERIALS AND METHODS
Animals
All experiments were performed according to the institu-
tional guidelines for animal experimentation at Celal Bayar
University Faculty of Medicine (Manisa, Turkey). We ob-
tained approval from the research ethics board of The Uni-
versity of Celal Bayar for the study.
Twenty-seven sexually mature female mus musculus al-
bino (C/C) mice, weighing 30 g and 6 weeks of age, were
used. The mice were kept in our laboratory for at least 3
weeks before the experiments under controlled temperature
conditions and a 12-hour light/dark regimen (lights on from
7:00 AM to 7:00 PM).
The 27 mice were divided into three groups of 9 mice
each. Vaginal smears were obtained from all mice to confirm
that they had regular cycles for three cycles before drug
administration. Mice were weighed daily. Group I received
single-dose (25 mg/kg [0.75 mg]) anastrozole (Arimidex;
Astra Zeneca, Istanbul, Turkey) in the proestrous phase
(5:00 PM) by lavage. The following day (1:00 PM), a single
dose (5 IU/mL SC) of recombinant FSH (Puregon; Organon,
Istanbul, Turkey, or Gonal-F; Serono, Istanbul, Turkey) was
given, and 48 hours after FSH, hCG (5 IU/mL SC) (Pregnyl;
1798 Karaer et al. Effect of anastrozole on implantation
Organon or Profasi-HP; Serono) was given. Group II was not
given anastrozole but received FSH (in estrous phase) and
hCG as in group I. Group III (control group) received sterile
saline (1 mL SC). Mice were caged one female to two fertile
males on the night of hCG administration and mated. The
morning of finding the vaginal plug was designated as day 1
of embryonic development (E1). Three mice from each
group were sacrificed on E1, and 42 embryos were aspirated
from uterine tubes with the micropipette flushing method.
The rest of the mice were sacrificed on E2.5–3 (implantation
period). Implantation areas were detected in uteruses of
groups II and III, and six to eight embryonic sacs were
macroscopically identified in group I animals. The uteruses
were cut into 3– 4-mm pieces so the identified implantation
sites and embryos would stay in the center of each piece.
Uteruses with embryo were removed and fixed in 10%
formalin for 48 hours.
Samples were then washed with tap water and soaked in a
series of 50%, 60%, 70%, 80%, and 90% ethanol for 30
minutes, then in 95% and 100% ethanol for 1 hour. They
were held in a solution of 100% ethanol and xylene (1:1
ratio) for 30 minutes, then embedded in paraffin and held at
60°C for 1 hour to make paraffin blocks. Transverse sections
(5 ␮m) were taken from the blocks and prepared for both
histochemical and immunohistochemical staining.
Histochemical Observation
Sections dewaxed at 60°C overnight were immersed in xy-
lene for 1 hour, then rehydrated through a graded series of
ethanol (100%, 95%, 80%, 70%, and 60%) for 2 minutes in
each concentration, then washed in tap water. Sections were
stained by hematoxylin-eosin according to routine protocols.
Slides were mounted with Entellan UN 1866 (Merck, Darm-
stadt, Germany), covered with glass cover slips before view-
ing, and photographed under light microscopy (Olympus
BX-40; Tokyo, Japan).
Immunohistochemistry
After deparaffination at 60°C overnight, sections were
held in xylene for 1 hour. After washing with serial
concentrations of ethanol (95%, 80%, 70%, and 60% for
2 minutes each), the sections were washed with distilled
water and phosphate-buffered saline (PBS) for 10 min-
utes. They were held in 2% trypsin in Tris buffer at 37°C
for 15 minutes, then washed in PBS (three 5-minute
washes). The limits of sections were drawn with a Dako
pen (S-2002; Dako, Carpinteria, CA) and incubated in 3%
hydrogen peroxidase for 15 minutes to inhibit the endog-
enous peroxidase activity. The tissues were then given
three 5-minute washes in PBS. The primary antibodies—
monoclonal anti-TNF-␣ in a 1/100 dilution (1E8-G6;
Santa Cruz Biotechnology, Santa Cruz, CA), polyclonal
anti-LIF in a 1/100 dilution (SC-1336; Santa Cruz Bio-
technology), polyclonal anti-laminin in a 1/100 dilution
(L-9393; Sigma Chemical, St. Louis, MO), and monoclo-
Vol. 83, No. 6, June 2005
 FIGURE 1
Embryos obtained from the three study groups in the preimplantation period. (A) In group I, embryos were in
the eight-cell stage; (B) in group II, embryos were in the two-cell stage; (C) no embryo was observed in
group III. Original magnification, ⫻400.
Karaer. Effect of anastrozole on implantation. Fertil Steril 2005.
nal anti-collagen IV in a 1/200 dilution (C-1926; Sigma
Chemical)—were incubated for 18 hours. They were then
given an additional three 5-minute washes in PBS, fol-
lowed by incubation with biotinylated anti-rabbit IgG and
administration of streptavidin peroxidase (Histostain Plus
kit Zymed 87-9999; Zymed, San Francisco, CA). After
washing the secondary antibody with PBS three times for
5 minutes, the sections were washed for 5 minutes in
Dako DAB Substrate system containing diaminobenzidine
to detect the immunoreactivity, then stained with Mayer’s
hematoxylin. They were covered with mounting medium
(catalogue no. 1012; Signet Laboratories, Dedham, MA)
and observed with light microscopy (Olympus BX-40).
Color photographs were taken with 100 ASA color film
(Fujifilm, Tokyo, Japan). Control samples were processed
in an identical manner, but in the absence of the primary
antibody. Two observers, blind to the clinical information
of the samples, independently evaluated the staining
scores. Staining intensity was assigned according to a
semiquantitative immunohistochemical scoring system as
follows: absent (⫺), very mild (⫹/⫺), mild (⫹), moderate
(⫹⫹), and strong (⫹⫹⫹).
RESULTS
Preimplantation Stage
With a micropipetting flushing method, embryos were
collected from uterine tubes. The embryos obtained from
group I in the preimplantation period were in the eight-
cell stage (Fig. 1A), whereas group II embryos were in the
two-cell stage (Fig. 1B). No embryos were observed in
group III (Fig. 1C).
Postimplantation Stage
When uteruses were collected at the E2.5–3 stage, six to
eight embryonic sacs were detected macroscopically in each
Fertility and Sterility姞
uterine horn in group I. Although implantation areas were
observed in each uterine horn in group II, an embryonic sac
was not detected in any uterine horn in this group. In group
III, neither implantation areas nor embryonic sacs were
observed.
After histologic evaluation, embryos were implanted in
group I, and polygonal decidual cells containing abundant
glycogen and lipid were noted (Fig. 2A). Although we
collected the embryos at the E2.5–3 stage, they were sur-
prisingly consistent with E5–5.5-stage embryos. In group II,
implantation did not take place; however, a clump of em-
bryonic cells were in the horn cavity (Fig. 2B). The appear-
ance of endometrial glands containing glycogen and tortuous
coiling was characteristic of the secretory phase (Fig. 2B). In
group III, the properties of the endometrium were consistent
with the late proliferative phase, with thick endometrial
epithelium and tubular glands (Fig. 2C).
Staining intensities of TNF-␣ and LIF were detected in
embryonic, decidual, and endometrial cells in group I (Table
1) (Fig. 3). In group I, the staining intensity of TNF-␣ was
strong in all compartments (Fig. 3A, B), whereas mild and
weak staining of LIF were detected in trophoectoderm and
decidual cells, respectively (Fig. 3C, D). Immunoreactivity
of laminin and collagen was seen in the embryonal and
maternal parts of the implantation region; these immunore-
activities were strongly detected in the trophoectoderm and
basement membrane of the endometrial epithelium (Fig.
3E–H).
In group II, weak staining in the embryo and uterine
epithelium and mild staining in the stroma were detected for
TNF-␣ and LIF (Fig. 4A–D). In addition, a clump of em-
bryonic cells were stained with laminin and collagen IV,
which are markers for trophoectoderm cells (Fig. 4E–H).
Laminin staining was detected in both epithelial and stromal
parts of the endometrium, and the intensity of laminin was
1799
 FIGURE 2
Histologic evaluation of endometrium in the three groups in the implantation stage by hematoxylin-eosin
staining. (A) In group I, embryos were implanted, and polygonal decidual cells containing abundant glycogen
and lipid were noted. Although we collected the embryos at the E2.5–3 stage, they were surprisingly
consistent with stage E5–5.5 embryos. (B) In group II, although implantation did not take place, a clump of
embryonic cells were detected in the horn cavity. Endometrial glands containing glycogen and tortuous
coiling were characteristic of the secretory phase. (C) In group III, properties of endometrium were
consistent with the late proliferative phase, with thick endometrial epithelium and tubular glands. Original
magnification, ⫻100.

Karaer. Effect of anastrozole on implantation. Fertil Steril 2005.

strong in the stroma (Fig. 4E, F). Immunoreactivity of col- In group III, the uterine epithelium and stroma were
lagen IV was also seen in both the epithelial and stromal stained with all antibodies (Fig. 5), and strong immunoreac-
parts of the endometrium, but this immunoreactivity was tivities were observed for LIF and collagen IV (Fig. 5C, D,
lesser than the laminin staining (Fig. 4G, H). G, H).

TABLE 1
Staining intensities of TNF-␣, LIF, Laminin, and Collagen IV using a semiquantitative
immunohystochemical scoring system.

Group I Group II Group III

T E D UE S E UE S UE S

TNF-␣ ⫹⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹⫹ ⫹ ⫹ ⫹⫹⫹ ⫹ ⫹⫹

LIF ⫹⫹ ⫹ ⫹ ⫹ ⫹⫹ ⫹/⫺ ⫹/⫺ ⫹ ⫹ ⫹⫹⫹
Laminin ⫹⫹⫹ ⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹⫹ ⫹ ⫹/⫺
Collagen IV ⫹⫹ ⫹ ⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹ ⫹⫹ ⫹ ⫹⫹⫹
Note: T ⫽ trophoectoderm; E ⫽ embryo; D ⫽ decidua; UE ⫽ uterine epithelium; S ⫽ stroma; TNF-␣ ⫽ tumor necrosis
factor-␣; LIF ⫽ leukemia inhibitory factor; ⫺ ⫽ absent; ⫹/⫺ ⫽ very mild; ⫹ ⫽ mild; ⫹⫹ ⫽ moderate; ⫹⫹⫹ ⫽ strong.
Karaer. Effect of anastrozole on implantation. Fertil Steril 2005.

1800 Karaer et al. Effect of anastrozole on implantation Vol. 83, No. 6, June 2005
 FIGURE 3

Immunohistochemistry of TNF-␣, LIF, laminin, and collagen IV in embryonic, decidual, and endometrial cells
in group I. In group I, although the staining intensity of TNF-␣ was strong in all compartments (A, B), mild
and weak staining of LIF were detected in trophoectoderm and desidual cells, respectively (C, D).
Immunoreactivity of laminin and collagen was seen in the embryonal and maternal parts of the implantation
region; these immunoreactivities were strongly detected in trophoectoderm and basement membrane of the
endometrial epithelium (E–H). Original magnification, ⫻ 100 (A, C, E, G) and ⫻400 (B, D, F, H).

Karaer. Effect of anastrozole on implantation. Fertil Steril 2005.

Fertility and Sterility姞 1801

FIGURE 4
Immunohistochemistry of TNF-␣, LIF, laminin, and collagen IV in group II. (A–D) Weak staining in embryo
and uterine epithelium and mild staining in stroma for TNF-␣ and LIF were detected. (E–H) Clumps of
embryonic cells were stained with laminin and collagen IV, which are markers for trophoectoderm cells.
Laminin staining was detected in both epithelial and stromal parts of the endometrium, and the intensity of
laminin was strong in stroma (E, F). Immunoreactivity of collagen IV was also seen in both the epithelial and
stromal parts of the endometrium, and was less than laminin staining (G, H). Original magnification, ⫻100
(A, C, E, G) and ⫻400 (B, D, F, H).

Karaer. Effect of anastrozole on implantation. Fertil Steril 2005.

1802 Karaer et al. Effect of anastrozole on implantation Vol. 83, No. 6, June 2005
 FIGURE 5

Immunohistochemistry of (A, B) TNF␣, (C, D) LIF, (E, F) laminin, and (G, H) collagen IV in group III. Uterine
epithelium and stroma were stained with all antibodies. Strong immunoreactivities were observed for LIF
and collagen IV (C, D, G, H). Original magnification, ⫻100 (A, C, E, G) and ⫻400 (B, D, F, H).

Karaer. Effect of anastrozole on implantation. Fertil Steril 2005.

In the control group, no embryos or implantation signs DISCUSSION

were detected. Immunohistochemical analysis showed that In recent decades, great improvement has been gained in
in the aromatase inhibitor–added group (group I), decidual infertility treatment, especially in ovulation induction. It has
cells developed; LIF especially had more positive staining been observed that although the ovulation rate is high with
compared with other implantation markers. CC (57%–91%), the pregnancy rate is lower (approximately

Fertility and Sterility姞 1803

20%– 40%), and the incidence of miscarriage is higher in
pregnancies conceived after CC treatment (23, 24).
It has been shown that gonadotropins used for controlled
ovarian stimulation might create a hostile uterine environ-
ment that is not conducive for successful implantation (25).
It is possible that supraphysiologic levels of E, attained
during ovarian stimulation, might explain the adverse effects
of ovarian stimulation on the outcome of infertility treatment
(26). Although the actual mechanisms of an adverse effect of
high levels of E on reproductive outcomes are unknown,
possible mechanisms include deleterious effects of E on the
endometrium (27), the embryo (28), the coagulation system
(29), and the oviduct (30).
Embryonic implantation is a process involving the em-
bryo and maternal endometrium. There seems to be a
temporal and spatially unique period called the “window
of implantation” in women, during which implantation
might occur. Analysis of the window of implantation has
revealed numerous processes to be occurring simulta-
neously or sequentially (31). It seems that embryo im-
plantation outside of this “window” is either outright
prevented or, at best, inhibited (32).
Mirkin et al. (33) conducted a prospective randomized
study to investigate gene expression profiles of the human
endometrium during the window of implantation. They com-
pared gonadotropin-stimulated cycles with temporally
matched natural cycles. Their results corroborated that go-
nadotropin-stimulated cycles showed advanced pinopode ap-
pearance, histological features, and steroid receptor down-
regulation when compared with natural cycles (33).
Tavaniotou et al. (34) compared integrin expression in
gonadotropin-stimulated and natural cycles. Endometrial
features, in terms of dating and integrin expression, were
advanced in the early luteal phase of the stimulated cycles.
They found reduced ␣1, ␣4, and ␣v␤3 integrin expression
during the midluteal phase in stimulated cycles and con-
cluded that these altered endometrial characteristics might
affect endometrial– embryonic synchronization and the im-
plantation process (34).
In an animal experiment, Guo et al. (35) gave daily
injections of anastrozole to mice and induced ovulation
induction with pregnant mare serum gonadotropin (PMSG)
to determine whether E is required for follicular and embry-
onic development. They compared an anastrozole-plus-
PMSG group with a PMSG-only group and measured E and
P levels. They found that the E level in the anastrozole group
was significantly lower compared with the control group, but
they did not study implantation markers (35).
Years ago, an embryo loss of 44% before implantation in
mice that had been superovulated was described by Beau-
mont and Smith (36). Recently, Van der Auwera and
D’Hooghe (37) studied the effect of gonadotropin treatment
on embryo in a mouse model for IVF and reported that
ovarian stimulation itself caused a delay in embryonic de-
1804 Karaer et al. Effect of anastrozole on implantation
velopment both in vitro and in vivo, an increase in abnormal
blastocyst formation, a fetal growth retardation of 22%, and
an increase in the number of resorption sites.
In experimental research, Dursun et al. (38) showed that
exogenous administration of gonadotropins significantly af-
fected the morphology of the endometrium and the mitotic
index at the implantation period of the embryo in rat. The
morphologic effects became more pronounced as the admin-
istered dose of exogenous gonadotropins was increased. The
investigators concluded that low-dose stimulation should be
preferred compared with high-dose stimulation to minimize
the unfavorable effects of gonadotropins on the endome-
trium (38).
After the introduction of aromatase inhibitors in ovarian
stimulation cycles by Mitwally and Casper (16, 17), we
aimed to study the effect of anastrozole on embryo devel-
opment and implantation markers in FSH-stimulated cycles
and to determine whether retreating from either the anties-
trogenic effects of CC or supraphysiologic levels of gonad-
otropin induction makes a difference in implantation, a mul-
tifactorial issue that has not yet been solved by scientists. We
performed an animal experiment; we could not get approval
from our institutional ethics committee for an experiment in
humans because the use of aromatase inhibitors still repre-
sents a novel method of ovulation induction. This is the first
immunohistochemical study of the effect of aromatase in-
hibitors on implantation and embryos in FSH-induced ovu-
lation cycles in mice.
The implantation markers, including TNF-␣ and LIF,
have a significant role in the dynamic developmental events
of human pregnancy. The presence of TNF-␣ receptors at the
cell surface of mouse blastocysts is demonstrated by the
indirect immunofluorescence technique, which indicates that
the expression of TNF-␣ receptors is developmentally reg-
ulated before implantation and that preimplantation embryos
are responsive to TNF-␣ (39). Leukemia inhibitory factor
not only affects trophoblast differentiation but also controls
the phenotype of adhesion in embryos. During implantation,
LIF also causes secretion of paracrine signals from embry-
onic tissues and uterine epithelium (40).
Laminin and collagen IV form 70%– 80% of basal mem-
brane proteins. It has been shown that the continuous struc-
ture of uterine lumen epithelium is destroyed during blasto-
cyst implantation in mice. This destruction is related to
decidual cell differentiation and invasion of trophoblast
cells. The decrease in laminin and collagen IV intensity is
first detected around trophoblast on the sixth day of mouse
pregnancy (41). With the loss of the basal membrane in the
lumen epithelium, an increase in laminin and collagen IV in
decidual cells near the stroma is detected. They are also
shown to localize in trophoblast basement membrane (42).
These four markers were chosen to evaluate receptivity
because TNF-␣ and LIF are endometrial receptivity markers,
and in addition to secretion of laminin and collagen IV from
endometrium, they are also markers of embryo.
Vol. 83, No. 6, June 2005
 In our results, macroscopic observation in group III
showed that the endometrium was in the late proliferative
phase, and preparation for implantation was not detected. In
group II, endometrium, which was being affected by FSH
and hCG, was in the secretory phase, and embryos had not
yet implanted. These embryos gained in the preimplantation
period were consistent with their embryonic day. In group I,
which received aromatase inhibitor before FSH induction,
embryos and endometrium in either the preimplantation or
implantation phases were affected. According to our obser-
vation, embryos in the preimplantation phase were not con-
sistent with their embryonic day; their differentiation was
quicker. All animals in this group were implanted, and
endometrial stromal cells were differentiated to decidual
cells.
In histochemical analysis, we detected TNF-␣ and LIF
immunoreactivity in all groups, but more immunoreactivity
of TNF in group II suggested that embryo existing in lumen
had not implanted and that endometrial implantation markers
were preparing the endometrium. In group I, immunoreac-
tivity of TNF-␣ was the same as in group II, but immuno-
reactivity of LIF proved that invasion was still occurring.
Extracellular matrix proteins laminin and collagen IV
were detected in both embryonal and uterine regions in
group I. Especially their positive staining in trophoectoder-
mal cells showed that these cells were differentiating. These
immunoreactivities were similar in the other groups. In
group II, the immunoreactivities for laminin and collagen IV
were positive in clumps of embryonic cells in lumen, which
proved that they were trophoectodermal.
We showed that anastrozole has a positive effect on em-
bryo and implantation in gonadotropin-induced cycles.
Anastrozole use in FSH-induced cycles not only was asso-
ciated with an increase in preimplantation receptivity and
implantation but also supported release of implantation
markers.
Hu et al. (43) tried to determine the effects of anastrozole
on in vitro follicle and oocyte development in preantral
mouse follicle culture. The results of their study demon-
strated that although they keep follicles under a constant
gonadotropin tonus, drastic changes in intrafollicular estro-
gen/androgen balance exert no major negative impact on
folliculogenesis, oogenesis, and meiosis. Although the fer-
tilization rate was impaired in oocytes cultured under anas-
trozole, their further preimplantation embryo development
seemed unaffected. The findings of Hu et al. (43) might help
explain the findings of our study, particularly the enhanced
embryo development.
We believe that the enhanced embryo development seen
in this study would explain the higher implantation because
embryo development is synchronized with endometrial de-
velopment.
In this study, the positive results might be attributed to the
E-lowering effect of anastrozole, but the research of Moud-
Fertility and Sterility姞
gal et al. (44) proved that the role of E in embryo develop-
ment and implantation is species dependent. This might be a
limitation in extrapolating these data to clinical practice in
humans, and more clinical studies in human tissues are
needed to prove this fact thoroughly.
Acknowledgments: The authors thank İsmail Zonguldak, technician, for
assistance in the care of mice.
REFERENCES
1. Lunenfeld B, Insler V. Classification of amenorrhoeic states and their
treatment by ovulation induction. Clin Endocrinol (Oxf) 1974;3:223–
37.
2. Insler V, Melmed H, Mashiah S, Monselise M, Lunenfeld B, Rabau E.
Functional classification of patients selected for gonodotropic therapy.
Obstet Gynecol 1968;32:620 – 6.
3. Buyalos RP, Agarwal SK. Endometriosis-associated infertility. Curr
Opin Obstet Gynecol 2000;12:377– 81.
4. Aboulghar MA, Mansour RT, Serour GI, Amin Y, Abbas AM, Salah
IM. Ovarian superstimulation and intrauterine insemination for the
treatment of unexplained infertility. Fertil Steril 1993;60:303– 6.
5. Fisch P, Casper RF, Brown SE, Wrixon W, Collins JA, Reid RL, et al.
Unexplained infertility: evaluation of treatment with clomiphene citrate
and human chorionic gonadotropin. Fertil Steril 1989;51:828 –33.
6. Laufer N, DeCherney AH, Haseltine FP, Polan ML, Mezer HC, Dlugi
AM, et al. The use of high-dose human menopausal gonadotropin in an
in vitro fertilization program. Fertil Steril 1983;40:734 – 41.
7. Hornnes P, Giroud D, Howles C, Loumaye E. Recombinant human
follicle-stimulating hormone treatment leads to normal follicular
growth, estradiol secretion, and pregnancy in a World Health Organi-
zation group II anovulatory woman. Fertil Steril 1993;60:724 – 6.
8. Ellis JD, Williamson JG. Factors influencing the pregnancy and com-
plication rates with human menopausal gonadotropin therapy. Br J
Obstet Gynaecol 1975;82:52–7.
9. Neyro JL, Barrenetxea G, Montoya F, Rodriguez-Escudero FJ. Pure
FSH for ovulation induction in patients with polycystic ovary syndrome
and resistant to clomiphene citrate therapy. Hum Reprod 1991;6:218 –
21.
10. Muasher SJ, Garcia JE, Rosenwaks Z. The combination of follicle-
stimulating hormone and human menopausal gonadotropin for the
induction of multiple follicular maturation for in vitro fertilization.
Fertil Steril 1985;44:62–9.
11. Gleicher N, Vietzke M, Vidali A. Bye-bye urinary gonadotropins?
Recombinant FSH: a real progress in ovulation induction and IVF?
Hum Reprod 2003;18:476 – 82.
12. Steinkampf MP, Mendelson CR, Simpson ER. Regulation by follicle-
stimulating hormone of the synthesis of aromatase cytochrome P-450 in
human granulosa cells. Mol Endocrinol 1987;1:465–71.
13. Karaer Ö, Oruç S, Koyuncu FM. Aromatase inhibitors: possible future
applications. Acta Obstet Gynecol Scand 2004;83:699 –706.
14. Buzdar A, Howell A. Advances in aromatase inhibition: clinical effi-
cacy and tolerability in the treatment of breast cancer. Clin Cancer Res
2001;7:2620 –35.
15. Bhatnagar AS, Hausler A, Schieweck K, Lang M, Bowman R. Highly
selective inhibition of estrogen biosynthesis by CGS 20267, a new
non-steroidal aromatase inhibitor. J Steroid Biochem Mol Biol 1990;
37:1021–7.
16. Mitwally MF, Casper RF. Aromatase inhibition improves ovarian re-
sponse to follicle-stimulating hormone in poor responders. Fertil Steril
2002;77:776 – 80.
17. Mitwally MF, Casper RF. Aromatase inhibition reduces gonadotropin
dose required for controlled ovarian stimulation in women with unex-
plained infertility. Hum Reprod 2003;18:1588 –97.
18. Perrier d’Hauterive S. Implantation: the first maternal-embryo
crosstalk. J Gynecol Obstet Biol Reprod 2004;33:S5– 8.
19. Chen HL, Yang YP, Hu XL, Yelavarthi KK, Fishback JL, Hunt JS.
1805
 Tumor necrosis factor alpha mRNA and protein are present in human
placental and uterine cells at early and late stages of gestation. Am J
Pathol 1991;139:327–35.
20. Aghajanova L, Stavreus-Evers A, Nikas Y, Hovatta O, Landgren BM.
Coexpression of pinopodes and leukemia inhibitory factor, as well as its
receptor, in human endometrium. Fertil Steril 2003;79(Suppl 1):808 –
14.
21. Inagaki J, Sugiura-Ogasawara M, Nomizu M, Nakatsuka M, Ikuta K,
Suzuki N, et al. An association of IgG anti-laminin-1 autoantibodies
with endometriosis in infertile patients. Hum Reprod 2003;18:544 –9.
22. MacIntyre DM, Lim HC, Ryan K, Kimmins S, Small JA, MacLaren
LA. Implantation-associated changes in bovine uterine expression of
integrins and extracellular matrix. Biol Reprod 2002;66:1430 – 6.
23. Hull MGR. The causes of infertility and relative effectiveness of
treatment. In: Templeton AA, Drife JO, eds. Infertility. London:
Springer, 1992:33– 62.
24. Hack M, Brish M, Serr DM, Insler V, Salomy M, Lunenfeld B.
Outcome of pregnancy after induced ovulation. Follow-up of pregnan-
cies and children born after clomiphene therapy. JAMA 1972;220:
1329 –33.
25. Check JH, Choe JK, Nazari A, Fox F, Swenson K. Fresh embryo
transfer is more effective than frozen for donor oocyte recipients but not
for donors. Hum Reprod 2001;16:1403– 8.
26. Pellicer A, Valbuena D, Cano F, Remohi J, Simon C. Lower implan-
tation rates in high responders: evidence for an altered endocrine milieu
during the preimplantation period. Fertil Steril 1996;65:1190 –5.
27. Benadiva CA, Metzger DA. Superovulation with human menopausal
gonadotropins is associated with endometrial gland-stroma dyssyn-
chrony. Fertil Steril 1994;61:700 – 4.
28. Warner CM, Cao W, Exley GE, McElhinny AS, Alikani M, Cohen J, et
al. Genetic regulation of egg and embryo survival. Hum Reprod 1998;
13(Suppl 3):178 –90.
29. Rice VC, Richard-Davis G, Saleh AA, Ginsburg KA, Mammen EF,
Moghissi K, et al. Fibrinolytic parameters in women undergoing ovu-
lation induction. Am J Obstet Gynecol 1993;169:1549 –53.
30. Van der Auwera I, Pijnenborg R, Koninckx PR. The influence of
in-vitro culture versus stimulated and untreated oviductal environment
on mouse embryo development and implantation. Hum Reprod 1999;
14:2570 – 4.
31. Giudice LC. Microarray expression profiling reveals candidate genes
for human uterine receptivity. Am J Pharmacogenomics 2004;4:299 –
312.
1806 Karaer et al. Effect of anastrozole on implantation
32. Navot D, Scott RT, Droesch K, Veeck LL, Liu HC, Rosenwaks Z. The
window of embryo transfer and the efficiency of human conception in
vitro. Fertil Steril 1991;55:114 – 8.
33. Mirkin S, Nikas G, Hsiu JG, Diaz J, Oehninger S. Gene expression
profiles and structural/functional features of the peri-implantation en-
dometrium in natural and gonadotropin-stimulated cycles. J Clin En-
docrinol Metab 2004;89:5742–52.
34. Tavaniotou A, Bourgain C, Albano C, Platteau P, Smitz J, Devroey P.
Endometrial integrin expression in the early luteal phase in natural and
stimulated cycles for in vitro fertilization. Eur J Obstet Gynecol Reprod
Biol 2003;108:67–71.
35. Guo Y, Guo KJ, Huang L, Tong XG, Li X. Effect of estrogen depri-
vation on follicle/oocyte maturation and embryo development in mice.
Chin Med J 2004;117:498 –502.
36. Beaumont HM, Smith AF. Embryonic mortality during the pre- and
post-implantation periods of pregnancy in mature mice after superovu-
lation. J Reprod Fertil 1975;45:437– 48.
37. Van der Auwera I, D’Hooghe T. Superovulation of female mice delays
embryonic and fetal development. Hum Reprod 2001;16:1237– 43.
38. Dursun A, Sendag F, Terek MC, Yilmaz H, Oztekin K, Baka M, et al.
Morphometric changes in the endometrium and serum leptin levels
during the implantation period of the embryo in the rat in response to
exogenous ovarian stimulation. Fertil Steril 2004;82(Suppl 3):1121– 6.
39. Pampfer S, Wuu YD, Vanderheyden I, De Hertogh R. Expression of
tumor necrosis factor-alpha (TNF alpha) receptors and selective effect
of TNF alpha on the inner cell mass in mouse blastocysts. Endocrinol-
ogy 1994;134:206 –12.
40. Stewart CL. The role of leukemia inhibitory factor (LIF) and other
cytokines in regulating implantation in mammals. Ann N Y Acad Sci
1994;734:157– 65.
41. Thomas T, Dziadek M. Expression of laminin and nidogen genes
during the postimplantation development of the mouse placenta. Biol
Reprod 1993;49:1251–9.
42. Enders AC. Morphological manifestations of maturation of the blasto-
cyst. Prog Clin Biol Res 1989;294:211–23.
43. Hu Y, Cortvrindt R, Smitz J. Effects of aromatase inhibition on in vitro
follicle and oocyte development analyzed by early preantral mouse
follicle culture. Mol Reprod Dev 2002;61:549 –59.
44. Moudgal NR, Shetty G, Selvaraj N, Bhatnagar AS. Use of a specific
aromatase inhibitor for determining whether there is a role for oestro-
gen in follicle/oocyte maturation, ovulation and preimplantation em-
bryo development. J Reprod Fertil Suppl 1996;50:69 – 81.
Vol. 83, No. 6, June 2005