Human Reproduction Vol.21, No.1 pp. 202–209, 2006
doi:10.1093/humrep/dei286
Advance Access publication September 30, 2005.
Adverse effects of retinoic acid on embryo development and
the selective expression of retinoic acid receptors in mouse
blastocysts
Fu-Jen Huang1,2,4, Yan-Der Hsuuw1,3, Kou-Chung Lan1, Hong-Yo Kang1,2,
Shiuh-Young Chang1,2, Yu-Cheng Hsu1 and Ko-En Huang1,2
1
Department of Obstetrics and Gynecology, Chang Gung Memorial Hospital, Kaohsiung, Taiwan, 2Chang Gung University School of
Medicine and 3Animal Science, National Ping-Tung University of Science Technology
4
To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, Chang Gung Memorial Hospital, 123, TaPei Rd., Niao-Sung Hsiang, Kaohsiung County, Taiwan. E-mail: [email protected]
BACKGROUND: All-trans retinoic acid (RA), the oxidative metabolite of vitamin A, is essential for normal development. In addition, high levels of RA are teratogenic in many species. We have previously shown that excess RA results in
immediate effects on the preimplantation embryo and on blastocyst development. This study was conducted to clarify
the long-term survival of mouse blastocyst and the effect of RA on gene expression. METHODS AND RESULTS: Using
an in vitro model, we identified the immediate adverse impact of RA on mouse blastocyst development. This involved an
inhibition of cell proliferation and growth retardation. Using an in vivo model, we also identified the resorption of postimplanted blastocysts that had been treated with excess RA. Analysis of RA-mediated gene induction was also included.
The retinoic acid receptors RAR␣ and RAR␥ were constitutively expressed in the blastocyst and the inner cell mass,
whereas RAR␤ was induced upon RA treatment. CONCLUSIONS: This is the first evidence to show the impacts of RA
on mouse blastocysts in vitro and any carry-over effects in the uterus. There is a retardation of early postimplantation
blastocyst development and then subsequent blastocyst death. Our findings also show that there is some degree of selective induction of retinoic acid receptors when excess RA is administered to the blastocysts.
Key words: blastocyst/embryo development/retinoic acid/retinoic acid receptor
Introduction
Vitamin A and its physiological metabolites, retinoic acid (RA)
and other retinoids, exhibit a striking effect on pattern formation
and may be one of the morphogens that control embryonic
development (Dencker et al., 1987; Eichele et al., 1989; Maden
et al., 1982; Gudas et al., 1992; Hofmann and Eichele, 1994).
While normal embryo development requires retinoids, too much
or too little, at the wrong stage or at the wrong time, can
adversely affect the developing embryo (Sporn and Roberts,
1991). Embryos exposed to low concentrations of RA (0.0001–
0.1 μmol/l) develop normally, whereas those exposed to higher
concentrations (0.5–100 μmol/l) develop characteristic dosedependent defects (Vandersea, 1998). Embryonic exposure to
RA causes a wide spectrum of severe malformation in the offspring of humans (Lammer et al., 1985), rodents (Shenefelt,
1972; Kessel and Gruss, 1991), chickens (Thaller and Eichele,
1987) and Xenopus (Durston et al., 1989; Sive et al., 1990), and
these include neural tube and central nervous system defects,
skeletal defects, cleft palate, ear, and other craniofacial malformations, defects of the heart, thymus, and urogenital system, and
limb and digit reduction or duplication.
Several mouse studies have demonstrated a potential
adverse effect of RA when administered during mid- and late
pregnancy in mice. Administration of RA on day 7 or day 8 of
gestation causes retardation of general development, abnormal
differentiation of the cranial neural plate and abnormal development of the hindbrain (Morriss-Kay et al., 1991). Administration of RA on day 9 of gestation induces dysmorphogenesis
of the inner ear in mice (Frenz et al., 1996). Abnormalities of
limb and neural plate development have been induced when
RA is administered between day 10 and day 16 of gestation
(Kochhar, 1973; Kochhar et al., 1984; Stafford et al., 1995). In
our previous studies (Huang et al., 2001, 2003), the dosedependent effects of RA on blastocysts have been evaluated in
the presence of 0, 0.001, 0.1 and 10 μmol/l RA. An adverse
effect of RA was detected in the presence of 10 μmol/l RA. It
has previously been demonstrated that in utero exposure to
excess RA during the preimplantation period (early cleavage
stage) does not result in any adverse effect on preimplantation
embryo development (blastocyst formation) (Huang and Lin,
2001a). However, in vitro exposure to excess RA during the
peri-implantation (blastocyst stage) and early postimplantation
202 © The Author 2005. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: [email protected]
Adverse effects of retinoic acid on mouse blastocysts
period does result in early growth retardation and later embryo
death in postimplantation (comparable to day 3 to day 8 of
gestation) embryos (Huang et al., 2001). Cellular responses to
RA in vitro range from cell death to cell differentiation. Some
teratogenic effects of retinoids may be due to their ability to
induce apoptosis. RA-induced cell death with the characteristics
of apoptosis has been observed during our blastocyst studies
(Huang et al., 2003, 2005). The expression of mRNA encoding
all three RA receptors (RAR) -α, -β and -γ has been found in
mouse blastocysts and in early postimplantation mouse
embryos (Wu et al., 1992a). These nuclear receptors regulate
gene transcription in a ligand-dependent fashion through binding to specific DNA sequences, generally located upstream of
the promoter of the target gene (Umesono et al., 1988; de The et
al., 1989, 1990). Furthermore, these nuclear RA receptors act as
RA-inducible enhancer factors (Evans, 1988; Green and Chambon, 1988; Vasios et al., 1989) in mouse and human. These
findings suggest that RA may be involved in the developmental
process of early mammalian embryos through interaction with
the RAR and the marked effects of retinoic acid on morphogenesis result from the activation of development control genes.
Thus, the differential expression of RAR seems to be an important step in RA signalling of the embryonic cells.
Although the short-term effects of RA on blastocysts (around
day 8–9 of gestation) have been demonstrated in in vitro studies
(Huang et al., 2001), maternal compensation for RA-treated blastocyst has not been demonstrated. In order to investigate the longterm effects of RA on blastocysts, we examined whether incubating mouse embryos with RA during the blastocyst stage resulted
in an immediate effect of RA on their developmental ability in
vitro, and whether there is a long-term effect of RA in vivo. We
also investigated the expression of RARα, RARβ or RARγ in the
system to detect the RA signalling in the blastocysts.
Materials and methods
Blastocyst collection and culture with RA
ICR mice (6 ± 8 weeks old) were induced to superovulate by injecting
5 IU pregnant mare’s serum gonadotrophin (PMSG; Sigma Chemical
Co., St Louis, MO, USA) and this was followed 48 h later by the
injection of 5 IU HCG (Serono, NV Organon Oss, The Netherlands).
They were then mated overnight with a single fertile male of the same
strain. Mating was confirmed by the presence of a copulatory plug the
following day. The ICR virgin albino mice, male mice and pregnant
mice were maintained on breeder chow and kept under a 12 h day/12 h
night regimen, with food and water available ad libitum. All animals
received humane animal care as outlined in the Guidelines for Care
and Use of Experimental Animals (Canadian Council on Animal Care,
1984). The morning after an overnight mating period, female mice
with vaginal plugs were separated and used experimentally. The day a
vaginal plug was found was defined as day 1 of pregnancy. Embryos
were obtained by flushing the uterine horn and Fallopian tubes on the
afternoon of day 4 of gestation with CMRL-1066 culture medium
(Gibco Life Technologies, Grand Island, NY, USA) containing 1
mmol/l glutamine and 1 mmol/l sodium pyruvate (Sigma). The
embryos were collected in uncoated plastic 35 mm Falcon culture
dishes and washed a minimum of three times. Expended blastocysts
from different females were pooled and randomly selected for the
various experiments. Expended blastocysts were exposed to 0, 0.1 or
10 μmol/l all-trans RA without serum supplementation for 24 h. This
covers the concentrations expected in embryos after oral administration (Kraft et al., 1989) and forms the treated group. They were then
washed in RA-free medium and further cultured in CMRL-1066
medium in order to assess their further development as outlined
above. All-trans RA (Sigma) was prepared in an aqueous solution of
0.01% dimethylsulphoxide (DMSO) and a similar solution of DMSO
alone was used as the control.
Development of blastocyst in vitro
Outgrowing embryo culture
Blastocysts from ICR mice were collected on day 4 of pregnancy and
incubated in the presence or absence of RA (10 μmol/l) as described
in a previous study (Huang et al., 2001). After 24 h of incubation,
blastocysts were transferred to fibronectin-coated culture wells and
cultured individually in the absence of RA for 72 h in CMRL-1066
medium supplemented with 20% FBS (CMRL–FBS) (Armant et al.,
1986). Under these culture conditions, the hatched blastocysts will
attach onto the fibronectin and outgrow to form a cluster of inner cell
mass (ICM) cells over the trophoblastic layer [trophectoderm (TE)
outgrowth].
Morphology of outgrowing embryos
The morphological scores for outgrowth were assessed after a total of
96 h of incubation. The outgrowing embryos were classified as
attached or as outgrowth, in which they develop a cluster of ICM over
the trophoblastic layer. The scores of ICM clusters were given according to the shape of the ICM clusters from compact-rounded ICM
(+++) to a few scattered cells (+) over the trophoblastic layer as
described previously (Pampfer et al., 1994). The surface area of the
trophoblastic layer (trophectoderm outgrowth) was photographed and
measured by weighing the cut-out images, and then indirectly evaluating the surface of the images from a standard curve of weight versus
area. The morphological distribution was expressed as a percentage
and compared by χ2-analysis.
Cell proliferation of outgrowths
The proliferation of outgrowths was estimated by counting the nuclei
directly on the dish using Giemsa stain. This method was based on
nuclear labelling by Giemsa stain (Wuu et al., 1999) and the cell
spreading technique used for rat implanting embryos (Pampfer et al.,
1994).
The culture medium was carefully removed and replaced with
0.05% hypotonic sodium citrate (30 μl/well) at room temperature for
5 min. This solution was evaporated under partial vacuum (200 torr)
at 50–55°C for 60–90 min. Outgrowths were fixed with 50 μl per well
of FixDenat fixative at room temperature for 30 min. The total
number of nuclei in the outgrowths was examined by staining with a
4% Giemsa solution at room temperature for 30 min. The proliferation
of an individual outgrowth was expressed as the number of Giemsa
stained nuclei. The dead cells of outgrowths appeared either as fragmented nuclei (karyorrhexis) detected by bisbenzimide stain or as
DNA strand breaks (karyolysis) detected by the TdT (terminal deoxynucleotidyl transferase)-mediated dUDP nick-end labelling (TUNEL)
method, as described in the previous studies (Huang et al., 2003).
Data were represented as a mean ± SEM, and the difference between
the control and RA-treated groups was analysed by one-way analysis
of variance (ANOVA).
Development of blastocysts in vivo
Embryo transfer
The ability of expanded blastocysts to implant and develop in vivo
was evaluated by transferring the embryos to a recipient mouse. The
203
Fu-Jen Huang et al.
ICR females were mated with vasectomized males (C57BL/6J) to produce pseudopregnant mothers as recipients for the embryo transfer
experiments. The impact of RA on postimplantation growth in vivo
was addressed by exposing blastocysts to 0 or 10 μmol/l of RA for 24
h and transferring them (five to eight embryos) in parallel into pairs of
uterine horns in day 4 pseudopregnant mice. The surrogate mouse was
killed on day 18 of the pregnancy. The frequency of implantation was
calculated as the number of implantation sites per number of embryos
transferred. The frequency of resorption or surviving fetuses was calculated as the number of resorptions or of surviving fetuses per
number of implantations. The weight of the surviving fetuses and placentas was measured immediately after dissection. The results were
represented as mean ± SEM and the difference between control and
RA-treated groups was analysed by one-way analysis of variance
(ANOVA).
Quantitative expression RA receptor ␣, ␤, ␥ in the blastocyst
and ICM
Collection of ICM and in vitro culture from the blastocysts
The ICM clump was isolated by immunosurgery as described for rat
blastocysts (Pampfer et al., 1990) and mouse blastocysts (Huang
et al., 2005). Using the hanging drop culture, the ICM were isolated
from the blastocysts in aliquots of 20 μl Earle’s balanced salt solution
containing one ICM, in the presence of 0, 0.1, 10 μmol/l of all-trans
RA and placed on the lids of Petri dishes. The lids were inverted,
placed onto sterile water-filled dishes and incubated at 37°C for 24 h.
In hanging drop culture, the ICM moves to the bottom of the drop and
does not attach to the culture dish. After the 24 h culture, the ICM is
washed and collected into the individual test groups for treatment over
another 24 h culture.
Primers and PCR consumables
Four sets of primers were designed from the available mouse RARα,
RARβ and RARγ sequences and the β-actin sequence found in GenBank. The primers were chosen with the assistance of two computer
programs, Oligo 4.0 (National Biosciences, Plymouth, MN, USA) and
Primer Express (Perkin–Elmer Applied Biosystems, Foster City, CA,
USA). The nucleotide sequences of the primers are shown in Table I.
RNA extraction. Blastocyts were exposed to 0, 0.1, 10 μmol/l of RA
in serum supplement for 24 h. ICM were also exposed to the same
dose of RA in a separate experiment. Following the treatments, 40
blastocysts and 40 ICM in each group for one experiment were lysed
in Trizol Reagent (Invitrogen, Life Technologies, Inc., Gaithersburg,
MD, USA) by repetitive pipetting. At the end of a series of procedures, the RNA pellet was briefly dried and then finally the RNA was
dissolved in 15 μl of diethyl pyrocarbonat (DEPC) water by passing
the solution a few times through a pipette tip followed by incubation
Table I. Primers used for amplification of the retinoic acid receptors (RAR)
and β-actin cDNA
RARα
Forward
Reverse
RARβ
Forward
Reverse
RARγ
Forward
Reverse
β-Actin
Forward
Reverse
204
5′-ACCATCGAGACCCAGAGCA-3′
3′-CCTTACAGCCCTCACAGGC-5′
5′-GGGCTGCAAGGGTTTCTTC-3′
3′-CTCCCTTTTCATGCCAGTGG-5′
5′-TGGGCAAGTACACCACGAACTC-3′
3′-CGCTTCGCAAACTCCACAAT-5
5′-CGTACCACAGGCATTGTGATG-3′
3′-CACGCTCGGTCAGGATCTTC-5′
for 10 min at 60°C. The quality of the RNA samples was determined
by agarose gel electrophoresis with ethidium bromide staining. The
18S and 28S RNA bands were easily visualized under UV light.
cDNA synthesis. RNA was reverse-transcribed in a final volume of
25 μl containing 5×reverse transcriptase buffer, 10 mmol/l each
dNTP, 15 mmol/l MgCl2, 375 mmol/l KCl, and 250 mmol/l Tris–HCl
(pH 8.30), 25 IU of Rnasin Rnase inhibitor (Promega, Madison, WI,
USA), 50 mmol/l dithiothreitol, 200 IU of M-MLV (Moloney Murine
Leukemia Virus) Reverse Transcriptase) (Promega, Madison, WI,
USA), 0.5 μg oligo-dT primer (Promega), and 2 μg total RNA.
PCR amplification. All of the PCR reactions were performed using
an ABI Prism 7700 Sequence Detection System (Perkin–Elmer
Applied Biosystems). PCR was performed using the SYBR Green
PCR Core Reagents kit (Perkin–Elmer Applied Biosystems). Specific
PCR amplification products were detected using the fluorescent
double-stranded DNA-binding dye, SYBR Green (Schmittgen et al.,
2000). Experiments were performed in triplicate for each data point.
All of the samples with a coefficient of variation for the Ct value >1%
were retested.
Theoretical basis of real-time RT–PCR. Quantitative values are
obtained from the threshold cycle (Ct) number at which the increase in
signal associated with an exponential growth of PCR products starts to
be detected (using Perkin–Elmer Biosystems analysis software),
according to the manufacturer’s manual. The precise amount of total
RNA added to each reaction (based on absorbance) and its quality (i.e.
lack of extensive degradation) are both difficult to assess. We therefore also quantified transcripts of the gene β-actin as an endogenous
RNA control, and each sample was normalized on the basis of the βactin content. Final results, expressed as logtarget with target gene
expression relative to the β-actin gene, are termed ‘logtarget’, and
were determined as follows: log target = log10 (105 × 2ΔCt sample), where
ΔCt values of the sample are determined by subtracting the average Ct
value of the target gene from the average Ct value of the β-actin gene
in each sample.
Results
Development in vitro
In order to determine whether a 24 h exposure to RA would
impair their subsequent growth in vitro, blastocysts were incubated with 0, 0.1 or 10 μmol/l RA and then transferred (in the
absence of RA) onto fibronectin-coated dishes for 72 h. At the
end of the incubation period, 96.5% of the control blastocysts
had attached and outgrown compared with 92.1% of the RApretreated group. Compared with control outgrowths, fewer
implants derived from RA-pretreated blastocysts (10 μmol/l of
RA) featured a compact and structured ICM at the centre of the
trophectoderm (TE) layer, and more of them showed an apparently normal TE expansion surrounding a limited cluster of
ICM cells (P < 0.05, Table II). Further analysis of the outgrowths showed that the number of nuclei per outgrowth
decreased significantly after RA pretreatment (143 ± 20 versus
118 ± 24 versus 108 ± 19; control versus 0.1 versus 10 μmol/l,
Figure 1). To know whether this decrease of nuclei in outgrowth was associated with cell death, we applied the TUNEL
method and bisbenzimide stain to analyse the index of nuclear
labelling. Results showed that neither the number of fragmented nuclei cells nor the number of TUNEL-positive cells in
outgrowths was increased by RA pretreatment (not significant,
Table III).
Adverse effects of retinoic acid on mouse blastocysts
Table II. Morphological effects of retinoic acid (RA)-pretreated blastocysts after 72 h incubation in vitro
Blastocyst outgrowth to:
RA dose (μmol/l)
P
0
Attached
ICM+
ICM++
ICM+++
0.1
6/173 (3.5)
62/173 (35.8)a
47/173 (27.2)
58/173 (33.5)a
10
14/177 (7.9)
63/177 (35.6)a
40/177 (22.6)
60/177 (33.9)a
15/190 (7.9)
108/190 (56.9)b
39/190 (20.5)
28/190 (14.7)b
0.149
< 0.001
0.492
< 0.001
Values in parentheses are percentages.
Within columns, the proportion of embryos with different superscripts (a, b) are significantly different from each treatment group (P < 0.05; χ2-test).
ICM = inner cell mass.
Figure 1. Total number of cells per blastocyst cultured in the presence or absence of retinoic acid. Results (pooled from at least five reproducible
experiments) are given as means ± SEM and based on 50, 52, 50 values in each group (0, 0.1 and 10 μmol/L respectively). For all graphs, **p <
0.01, ***p < 0.001 are measured relative to the control value by one-way analysis of variance (ANOVA).
Table III. Nuclear labelling index in outgrowths of retinoic acid (RA)-pretreated blastocysts at 72 h
Fragmented nuclei cells
TUNEL-positive cells
Control blastocysts
n
45
45
ICM
2.0 ± 1.1
6.8 ± 3.8
TE
1.1 ± 0.8
1.8 ± 1.5
RA-pretreated blastocysts
n
45
45
ICM
2.4 ± 1.4
6.0 ± 3.3
TE
1.0 ± 0.8
1.3 ± 1.0
Surviving fetuses
No.
49
26
(%)b
54.0 ± 22.4
31.3 ± 17.3*
Results are given as the mean of labelled nuclei per outgrowth ± SEM.
n = number of outgrowths in each group.
ICM = inner cell mass; TE = trophectoderm.
Table IV. Percentages of implantations, resorptions and surviving fetuses in vivo from retinoic acid (RA)-pretreated blastocysts
RA dose
(μmol/l)
0
10
n
15
15
Embryos transferred
No.
129
129
Implantations
No.
88
76
(%)a
70.8 ± 22.1
60.2 ± 23.4
Resorptions
No.
39
50
(%)b
46.0 ± 22.4
68.7 ± 17.3*
The proportions of embryos are given as means ± SEM.
Number of implantations per number of embryos transferred ×100.
b
Number of resorptions or of surviving fetuses per number of implantations ×100.
*P < 0.05 compared with the corresponding control value (ANOVA).
a
Development in vivo
Assessment of the uterine content was performed 14 days
after transfer. Control blastocysts (88 of 129 embryos in 15
recipients) implanted at a rate of ∼70.8% (Table IV), and RA
pretreatment induced a 10% decrease in that proportion (76
of 129 embryos) (P = 0.314). Early resorption resembled
small dark moles without distinct structure. Late resorption
resembled placenta without fetal structure. Late resorption
was found in the RA-pretreated group. The proportion of
implanted embryos that failed to develop normally (early and
205
Fu-Jen Huang et al.
late resorptions) was 68.7% higher in the RA-pretreated
group (50 of 76 implants) in comparison to the control group
(39 of 88 implants) (P = 0.017). The external examination of
surviving fetuses did not reveal any gross morphological
alterations in both groups. The external examination of
organs (including nervous system, heart, lung, thymus, gastrointestinal system and urological system) also showed no
gross morphological alterations. Although a comparison of
fetal weight did not show a significant difference between the
two groups (528 ± 92 versus 511 ± 71, control versus RAtreated), there was a 28% decrease in the fetuses that were
>600 mg in weight among the RA-pretreated blastocysts (P =
0.043, Figure 2). No difference was found in placental
weights between the two groups (138 ± 27 versus 126 ± 25,
control versus RA-treated).
Effect of RA on the expression of RARα, RARβ, RARγ in
the blastocyst and ICM
The data in each group were averaged from three experiments.
Forty blastocysts or ICM in each group were pooled to analyse
gene expression. The expression levels of RARα, RARβ, and
RARγ were quantified in the three groups of blastocysts (Figure 3). The expression levels were determined as described in
Materials and methods. Comparison of the various RAR and
the reference β-actin gene was used to correct for variations in
the amounts of RNA. The average expression levels of RARα
did not vary significantly across the three groups (1.72 ± 0.09,
10 mmol/l of RA versus 1.67 ± 0.03, 0.1 mmol/l of RA versus
1.63 ± 0.06, control; not significant). The average expression
level of RARγ also did not vary significantly (1.67 ± 0.11, 10
mmol/l of RA versus 1.86 ± 0.05, 0.1 mmol/l of RA versus
1.61 ± 0.11, control). However, the expression levels of RARβ
Figure 2. Distribution of the weight of surviving fetuses on day 18 of
pregnancy. Surviving fetuses were obtained from the control and RA
pre-treated blastocysts following embryo transfer as described in
Materials and Methods. *p < 0.05.
206
in the group treated with RA were significantly higher than for
the control (1.76 ± 0.02, 10 mmol/l of RA versus 0.92 ± 0.03,
control, P < 0.001; 1.68 ± 0.08, 0.1 mmol/l of RA versus 0.92
± 0.03, control, P < 0.001). The expression levels of RARα,
RARβ and RARγ were also quantified in the three groups of
ICM (Figure 4). The average expression levels of RARα did
not differ significantly between the three groups (2.84 ± 0.08,
10 mmol/l of RA versus 2.69 ± 0.01, 0.1 mmol/l of RA versus
2.75 ± 0.23, control). The average expression levels of RARγ
were also not significantly different (2.04 ± 0.06, 10 mmol/l of
RA versus 2.03 ± 0.05, 0.1 mmol/l of RA versus 2.04 ± 0.23,
control). However, the expression levels of RARβ in the group
with RA-treatment were significantly higher than for the control (2.78 ± 0.10, 10 mmol/l of RA versus 1.78 ± 0.12, control,
P < 0.001; 2.60 ± 0.02, 0.1 mmol/l of RA versus 1.78 ± 0.12,
control, P < 0.001).
Discussion
Embryonic levels of RA and the response of cells to RA are
critical for normal development in vertebrates. However, the
administration of pharmacological doses of RA and its analogues (retinoids) during pregnancy results in a wide range of
cranio-facial and limb malformations (Sulik et al., 1988; Alles
and Sulik, 1989; Satre and Kochhar, 1989), as well as defects
in brain development (Langman and Welch, 1967; Lammer
et al., 1985). This provides evidence, albeit indirect, of the role
of RA in mammalian embryogenesis. Maternal exposure to RA
(20 mg/kg) on day 9 of gestation has been found to induce dysmorphogenesis of the inner ear in mouse embryos (Frenz et al.,
1996). Similarly, a congenital limb anomaly can be induced
following exposure of pregnant Swiss–Webster mice to nonphysiological levels of RA (120 mg/kg) on day 10 and day 11
of gestation (Kochhar, 1973; Kochhar et al., 1984). Additionally, RA (30 mg/kg) adversely affects early postimplantation
embryogenesis on day 3 or day 4 of gestation (Huang et al.,
unpublished results). However, 50 mg/kg of RA, and even
higher doses of RA (100 mg/kg), do not affect the development
of early or late preimplantation embryos adversely in vivo
(Huang and Lin, 2001). According to general pharmacokinetics,
it seems to be higher in oviduct or uterine fluids if mice are
exposed to high level of RA in vivo. But there are still some
factors to counteract the possibility. For example, the fluid in
oviduct or uterine cavity is hydrophilic but the RA is hydrophobic. It will determine the final concentration of RA in the
oviduct or uterine cavity. Another possible factor is the cells
and zona pellucida surrounding the embryo. They may protect
the preimplantation embryos in the oviduct or uterine cavity.
However that may be, it i needs to be confirmed by measuring
the concentration of RA in oviduct and uterine fluids of pregnant mice. The maternal compensation mechanism for the RAdamaged preimplantation embryo could be explained in a further
study. However, RA in vitro (10 μmol/l) has an adverse effect on
germ layer and subsequent neurula development from day 3 to
day 8 of gestation (Huang et al., 2001). These findings suggest
that embryos at different stages have different sensitivities to
an increase in RA concentration and the adverse effects of RA
are different at different stages of embryo development.
Adverse effects of retinoic acid on mouse blastocysts
Figure 3. The effect of various doses of RA on RAR-α, RAR-β and RAR-γ mRNA levels in the blastocysts. Blastocysts were given 0, 0.1, or
10 μmol/l of RA. Twenty-four hours after treatment the blastocysts were collected to isolate the total RNA. Total RNA was isolated from 40
blastocysts and pooled for each group for each RAR analysis. The level of RAR mRNA was determined by real-time PCR and quantified as
described in Materials and Methods. Expression level was expressed as log10. Data are given as means ±SEM and measured from three separate
experiments. For all graphs, ***p < 0.001 are measured relative to the control value by one-way analysis of variance (ANOVA).
Figure 4. The effect of various doses of RA on RAR-α, RAR-β and RAR-γ mRNA levels in the ICMs. ICMs were given 0, 0.1, or 10 μmol/l of
RA. Twenty-four hours after treatment the ICMs were collected to isolate the total RNA. Total RNA was isolated from 40 ICMs and pooled for
each group for each RAR analysis. The level of RAR mRNA was determined by real-time PCR and quantified as described in Materials and
Methods. Expression level was expressed as log10. Data are given as means ±SEM and measured from three separate experiments. For all graphs,
**p < 0.01, ***p < 0.001 are measured relative to the control value by one-way analysis of variance (ANOVA).
In previous experiments, blastocysts with an ICM population reduced to an average of 70% of the normal cell population gave a high incidence of fetal loss, in spite of a normal TE
population and a normal implantation rate (Tam et al., 1988). It
has also been suggested that a sufficient number of ICM cells
is essential to the process of implantation and an excessive
reduction in this cell lineage may impair embryo viability
(Pampfer et al., 1990; Kelly et al., 1992). Morphogenesis of
the postimplantation mouse embryo involves the transformation of the solid ICM of the late blastocyst into a hollow structure known as the ‘egg cylinder’ (Huang et al., 2001). It is
possible that the selective effect of RA on the ICM cell population may have an impact on subsequent embryo development,
implantation and embryogenesis. The effects of RA on the
blastocyst, especially the ICM, have been demonstrated in our
previous studies (Huang et al., 2001, 2003, 2005) and its
effects are dependent on the concentration of RA. However,
maternal compensation for RA-pretreated blastocysts in utero
was not demonstrated in these studies. To clarify this question,
we aimed to collect adequate data in the present study. Here,
RA-pretreated blastocysts were shown to have an apparent
reduction in cell growth at 72 h of culture. If the blastocysts
were transferred into the maternal uterus, damaged blastocysts
could not be rescued in the maternal environment. Although
implantation could be reached, the postimplantation embryo
development was highly impacted and this led to a higher incidence of embryo resorption in the RA-treated blastocysts. It
has been shown that the impact of RA on the ICM is deeper
207
Fu-Jen Huang et al.
than only on the trophectoderm of blastocysts (Huang et al.,
2003, 2005). In the present study, we observe that the pretreatment of blastocyst with RA reduced its ability to develop into
outgrowth with a well-structured ICM cluster. Those outgrowths derived from RA-pretreated blastocysts contained
fewer nuclei, suggesting that the pretreatment of blastocysts
with RA had a long-lasting effect on embryo proliferation in
vitro. In contrast, a previous study based on morphological
observation showed that a long period of exposure of outgrowing embryos to RA alter the embryo development in vitro,
regardless of the aspect of the ICM cluster over the outgrowth
(Huang et al., 2001). However, TE as well as trophoblast cells
have been shown to be resistant to RA in the study, which
might explain that the effect of RA on TE outgrowth is small or
absent. Because these outgrowths did not present signs of persisting occurrence of cell death in the study, it was assumed
that the cell number difference was secondary to the irreversible loss of cells detected immediately after RA treatment
(Huang et al., 2003). Considering our previous data that RA
has a major inhibition on the ICM cell population (Huang
et al., 2003, 2005), we may believe that the high rate of resorption is due to a RA-mediated deficiency of ICM cells in
embryos. Although the rate of implantation was not affected by
RA pretreatment, our data suggest that the deleterious effect of
RA on ICM cells can result in a potential risk of fetal death and
embryogenesis arrest in spite of a normal TE invasion.
RA signalling is mediated by two distinct classes of nuclear
receptors, RAR and retinoid X receptors (RXR). These receptors are members of the nuclear receptor family and function
by directly activating transcription via binding to promoter elements of the target genes (Petkovich et al., 1987). This process
influences several developmental processes by interacting with
different types of nuclear receptors. Each family is composed
of three genes, namely, RARα, RARβ and RARγ, and RXRα,
RXRβ and RXRγ. There is both convergence and divergence
in the signal transduction pathways between the RAR and
RXR families and between the different members of each family. For example, RARα, RARβ and RARγ bind to elements
that activate transcription in response to all-trans-RA and 9cis-RA. In contrast, the retinoid X receptors (RXRα, RXRβ
and RXRγ) bind and activate transcription in response to 9-cisRA (Heyman et al., 1992; Levin et al., 1992; Zhang et al.,
1992; Allenby et al., 1993). These nuclear receptors regulate
gene transcription in a ligand-dependent fashion through binding to specific DNA sequences, generally located upstream of
the promoter of the target gene (Umesono et al., 1988; de The
et al., 1989, 1990). The discovery of specific nuclear receptors
for retinoic acid provides strong evidence that the marked
effects of retinoic acid on morphogenesis result from the activation of development control genes. These nuclear RA receptors act as RA-inducible enhancer factors (Evans et al., 1988;
Green et al., 1988; Vasios et al., 1989) in mouse and human
and provide a basis for understanding how RA signals are
transduced at the level of gene expression. However, they have
been shown to result in differential expression, depending upon
the RA concentration in different systems. For example, RA
differentially regulates the expression of different transcripts
for RARα, RARβ and RARγ depending upon the duration and
208
dose of the treatment in F9 teratocarcinoma cells (Hu
et al.,1990; Martins et al., 1990; Wu et al., 1992b). In Harnish’s studies, RARβ mRNA levels in 9 day gestation mouse
conceptuses are increased as early as 3 h after administration of
a completely teratogenic dose of retinoic acid (100 mg/kg body
weight).
Given the limited information available concerning RAmediated RAR expression, particularly in the blastocyst or the
ICM other than embryonic stem cells, this study was performed
to gain an insight into the possible role of RARα, RARβ, RARγ
expression in the blastocyst and the ICM using a quantitative
real-time RT–PCR assay. Our results demonstrate that an in
vitro teratogenic dose (10 μmol/l) of retinoic acid increases the
level of RARβ mRNA, but not RARα or RARγ mRNA in the
blastocysts and the ICM. This is consistent with a report on
RAR expression in hepatoma cells in vitro (de The et al., 1990)
and for 9 day gestation mouse conceptuses (Harnish et al.,
1990). Thus RARα, -β or -γ mRNA levels in the blastocyst,
especially in the ICM, are influenced differentially by RA treatment in vitro and these are closely related to over-expression of
the RARβ gene. However, the present data could not explain by
which mechanism RA exerts a deleterious effect. It seems to be
an intranuclear signalling in the embryonic cell, which RAR-β
is increased in the presence of high concentrations of RA.
In conclusion, excess RA resulted in adverse impact on the
blastocysts, especially on the ICM in vitro and this led to the
resorption of postimplanted blastocysts or retardation of the
surviving fetus in vivo. As part of the RA signalling in the
embryonic cells, the differential expression of RARα, RARβ
and RARγ was noted in RA-treated blastocysts and the ICM of
blastocysts. These results further confirm the concept that the
ICM is the major target of RA impact and plays a very important role in postimplantation development. This agrees with our
previous findings that excess RA affects the postimplantation
development of blastocysts adversely. The data reported here,
combined with previously reported results, re-emphasize the
role of RA in the ICM of blastocysts and its effects on early
embryogenesis. RA can have profound effects on the mouse
ICM making it important to define how RA receptor expression changes in a RA-induced ICM.
Acknowledgement
This work was supported by grant NSC-91-2314-B-182A-150 and
NSC-92-2314-B-182A-176 from the National Science Council of Taiwan.
References
Allenby G, Bocquel MT, Saunders M, Kazmer S, Speck J, Rosenberger M,
Lovey A, Kastner P, Grippo JF, Chambon P et al (1993) Retinoic acid receptors and retinoid X receptors: interactions with endogenous retinoic acids.
Proc Natl Acad Sci USA 90,30–34.
Alles AJ and Sulik KK (1989) Retinoic-acid-induced limb-reduction defects:
perturbation of zones of programmed cell death as a pathogenetic mechanism. Teratology 40,163–171.
Armant DR, Kaplan HA and Lennarz WJ (1996) Fibronectin and laminin
promote in vitro attachment and outgrowth of mouse blastocytes. Dev Biol
116,519–523.
Dencker L, d’Argy R, Danielsson BRG, Ghantous H and Sperber O (1987)
Saturable accumulation of retinoic acid in neural and neural crest derived
cells in early embryonic development. Dev Pharmacol Ther 10,212–223.
Adverse effects of retinoic acid on mouse blastocysts
Durston AJ, Timmermans JP, Hage WJ, Hendriks HF, de Vries NJ, Heideveld
M and Nieuwkoop PD (1989) Retinoic acid causes an anteroposterior transformation in the developing central nervous system. Nature 340,140–144.
de The H, Marchio A, Tiollais P and Dejean A (1989) Differential expression
and ligand regulation of the retinoic acid receptor α and β genes. EMBO J
8,429–433.
de The H, Tiollais P and Dejean A (1990) The retinoic acid receptors. Nouv
Rev Fr Hematol 32,30–32.
Eichele G (1989) Retinoids and vertebrate limb pattern formation. Trends
Genet 5,246–251.
Evans RM (1988) The steroid and thyroid hormone receptor superfamily. Science 240,889–895.
Frenz DA, Liu W, Galinovic-Schwartz V and Van De Water TR (1996) Retinoic acid-induced embryopathy of the mouse inner ear. Teratology 53,292–
303.
Green S and Chambon P (1988) Nuclear receptors enhance our understanding
of transcription regulation. Trends Genet 4,309–314.
Gudas LJ (1992) Retinoids and vertebrate development. J Biol Chem
269,15399–15402.
Harnish DC, Barua AB, Soprano KJ and Soprano DR (1990) Induction of βretnoic acid receptor mRNA by teratogenic doses of retinoids in murine
fetuses. Differentiation 45,103–108.
Heyman RA, Mangelsdorf DJ, Dyck JA, Stein RB, Eichele G, Evans RM and
Thaller C (1992) 9-cis retinoic acid is a high affinity ligand for the retinoid
X receptor. Cell 68,397–406.
Hofmann C and Eichele G (1994) Retinoids in development. In Sporn MB,
Roberts AB and Goodman DS (eds) The Retinoids, Biology, Chemistry, and
Medicine. 2nd edn, Raven Press, New York, pp 387–441.
Hu L and Gudas LJ (1990) Cyclic AMP analogs and retinoic acid influence the
expression of retinoic acid receptor alpha, beta, and gamma mRNAs in F9
teratocarcinoma cells. Mol Cell Biol 10,391–396.
Huang FJ and Lin YL (2001) Effects of retinoic acid on pre-implantation
embryo development in mice. Chang Gung Med J 24,681–687.
Huang FJ, Wu TC and Tsai MY (2001) Effect of retinoic acid on implantation
and post-implantation development of mouse embryos in vitro. Hum Reprod
16,2171–2176.
Huang FJ, Shen CC, Chang SY, Wu TC and Hsuuw YD (2003) Retinoic acid
decreases the viability of mouse blastocysts in vitro. Hum Reprod 18,130–136.
Huang, FJ, Hsu, YC, Kang, HY, Chang, SY, Hsuuw, YD and Huang, KE
(2005) Effects of retinoic acid on the inner cell mass in mouse blastocysts.
Fertil Steril, 83, 238–242.
Kelly SM, Robaire B and Hales BF (1992) Paternal cyclophosphamide treatment causes postimplantation loss via inner cell mass-specific cell death.
Teratology 45,313–318.
Kessel M and Gruss P (1991) Homeotic transformations of murine vertebrae and
concomitant alteration of Hox codes induced by retinoic acid. Cell 67,89–104.
Kochhar DM (1973) Limb development in mouse embryos. I. Analysis of teratogenic effects of retinoic acid. Teratology 7,289–295.
Kochhar DM, Penner J and Tellone C (1984) Comparative teratogeneic activities of two retinoids: effects on palate and limb development. Teratogen
Carcin Mut 4,377–387.
Kraft JC, Lofberg B, Chahoud I, Bochert G and Nau H (1989) Teratogenicity
and placental transfer of all-trans-, 13-cis-, 4-oxo-all-trans-, and 4-oxo-13cis-retinoic acid after administration of a low oral dose during organogenesis in mice. Toxicol Appl Pharm 100,162–176.
Lammer EJ, Chen DT, Hoar RM, Agnish ND, Benke PJ, Braun JT, Curry CJ,
Fernhoff PM, Grix AW Jr, Lott IT et al (1985) Retinoic acid embryopathy.
New Engl J Med 313,837–841.
Langman J and Welch GW (1967) Excess vitamin A and development of the
cerebral cortex. J Comp Neurol 131,15–26.
Levin AA, Sturzenbecker LJ, Kazmer S, Bosakowski T, Huselton C, Allenby
G, Speck J, Kratzeisen C, Rosenberger M, Lovey A et al (1992) 9-cis retinoic acid stereoisomer binds and activates the nuclear receptor RXR alpha.
Nature 355,359–361.
Maden M (1982) Vitamin A and pattern formation in the regenerating limb.
Nature 295,672–675.
Martins CA, Ziegler LM and Napoli JL (1990) Retinoic acid, dibutyryl-cAMP
and differentiations affect the expression of retinoic acid receptors in F9
cells. Proc Natl Acad Sci USA 87,4804–4808.
Morriss-Kay GM, Murphy P, Hill RE and Davidson DR (1991) Effects of
retinoic acid excess on expression of Hox-29 and Krox-20 and on morphological segmentation in the hindbrain of mouse embryos. EMBO J 10,2985–
2995.
Pampfer S, De Hertogh R, Vanderheyden I, Michiels B and Vercheval M
(1990) Decreased inner cell mass proportion in blastocysts from diabetic
rats. Diabetes 39,471–476.
Pampfer S, Wuu YD, Vanderheyden I and De Hertogh R (1994) In vitro study
of the carry-over effect associated with early diabetic embryopathy in the
rat. Diabetologia 37,855–862.
Petkovich M, Brand NJ, Krust A and Chambon P (1987) A human retinoic
acid receptor which belongs to the family of nuclear receptors. Nature
330,444–450.
Satre MA and Kochhar DM (1989) Elevations in the endogenous levels of the
putative morphogen retinoic acid in embryonic mouse limb-buds associated
with limb dysmorphogenesis. Dev Biol 133,529–536.
Schmittgen TD, Zakrajsek BA, Mills AG, Gorn V, Singer MJ and Reed MW
(2000) Quantitative reverse transcription–polymerase chain reaction to
study mRNA decay: comparison of endpoint and real-time methods. Anal
Biochem 285,194–204.
Shenefelt RE (1972) Gross congenital malformations. Animal model: treatment of various species with a large dose of vitamin A at known stages in
pregnancy. Am J Pathol 66,589–592.
Sive HL, Draper BW, Harland RM and Weintraub H (1990) Identification of a
retinoic acid-sensitive period during primary axis formation in Xenopus laevis. Genes Dev 4,932–942.
Sporn MB and Roberts AB (1991) Interactions of retinoids and transforming
growth factor-beta in regulation of cell differentiation and proliferation. Mol
Endocrinol 5,3–7.
Stafford DL, Lussier MR, Sank AC and Shuler CF (1995) In vitro model of
syndactyly replicates the morphologic features observed in vivo. Plast
Reconstr Surg 96,1169–1176.
Sulik KK, Cook CS and Webster WS (1988) Teratogens and craniofacial malformations: relationships to cell death. Development 103(Suppl),213–231.
Tam PP (1988) Postimplantation development of mitomycin C-treated mouse
blastocysts. Teratology 37,205–212.
Thaller C and Eichele G (1987) Identification and spatial distribution of retinoids in the developing chick limb bud. Nature 327,625–628.
Umesono K, Giguere V, Glass CK, Rosenfeld MG and Evans RM (1988)
Retinoic acid and thyroid hormone induce gene expression through a common responsive element. Nature 336,262–265.
Vandersea MW, McCarthy RA, Fleming P and Smith D (1998) Exogenous
retinoic acid during gastrulation induces cartilaginous and other craniofacial
defects in Fundulus heteroclitus. Biol Bull 194,281–296.
Vasios GW, Gold JD, Petkovich M, Chambon P and Gudas LJ (1989) A retinoic acid-responsive element is present in the 5′ flanking region of the laminin B1 gene. Proc Natl Acad Sci USA 86,9099–9103.
Wu TC, Wang L and Wan YJ (1992a) Differential expression of retinoic acid
receptor mRNA during mouse embryogenesis. Dev Growth Differ 34,685–
691.
Wu TC, Wang L and Wan YJ (1992b) Retinoic acid regulates gene expression
of retinoic acid receptors α, β, and γ in F9 mouse teratocarcinoma cells. Differentiation 51,219–224.
Wuu YD, Pampfer S, Becquet P, Vanderheyden I, Lee KH and De Hertogh R
(1999) Tumor necrosis factor alpha decreases the viability of mouse blastocysts in vitro and in vivo. Biol Reprod 60,479–483.
Zhang XK, Lehmann J, Hoffmann B, Dawson MI, Cameron J, Graupner G,
Hermann T, Tran P and Pfahl M (1992) Homodimer formation of retinoid X
receptor induced by 9-cis retinoic acid. Nature 358,587–591.
209