Granulocyte-Macrophage Colony-Stimulating Factor Alleviates Adverse Consequences of Embryo Culture on Fetal Growth Trajectory and Placental
http://www.100md.com
内分泌学杂志 2005年第5期
Fertilitetscentrum AB and University of G?teborg (C.S., M.W.), S-41345 G?teborg, Sweden; and Department of Obstetrics and Gynaecology (C.T.R., S.A.R.), Research Centre for Reproductive Health, University of Adelaide, Adelaide, South Australia 5005, Australia
Address all correspondence and requests for reprints to: Sarah A. Robertson, Ph.D., Research Centre for Reproductive Health, Department of Obstetrics and Gynaecology, University of Adelaide, Adelaide, South Australia 5005, Australia. E-mail: sarah.robertson@adelaide.edu.au.
Abstract
Growth factors secreted by the female reproductive tract promote development of the preimplantation embryo and potentially act as epigenetic determinants of postimplantation developmental competence and pregnancy outcome. In a comprehensive embryo transfer study in mice, we examined the late gestational and postnatal effects of embryo exposure to the cytokine granulocyte-macrophage colony-stimulating factor (GM-CSF), identified as a key physiological regulator of cell number and viability in mouse and human blastocysts. Embryo development in culture in the absence of GM-CSF restricted fetal growth, accelerated postnatal growth, and increased adult body mass and adiposity in offspring compared with in vivo-grown embryos, especially in males. Addition of GM-CSF to embryo culture medium increased the proportion of transferred embryos that generated viable progeny and alleviated the effects of in vitro culture on fetal and postnatal growth trajectory but did not prevent programing of adult obesity. Placental morphogenesis was modified by embryo culture, which inhibited development of labyrinthine exchange tissue and adversely altered some structural correlates of placental transfer function. GM-CSF reversed the effect of culture on labyrinthine growth and increased the surface area of placental trophoblast available for nutrient exchange. These findings indicate that the detrimental influence of embryo culture on fetal viability and growth may be largely mediated through altered placental morphogenesis and can be alleviated by GM-CSF. This demonstrates that embryonic exposure to GM-CSF is essential for normal placental development and fetal growth.
Introduction
A HEALTHY FEMALE reproductive tract provides the optimal milieu for the developing embryo. As it traverses the oviduct and uterus before implantation, the embryo encounters an array of cytokines and growth factors that influence proliferation, viability, and differentiation of blastomeres into trophectoderm and inner cell mass lineages (1, 2, 3). Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a cytokine identified as essential for normal blastocyst development and subsequent viability (4, 5). GM-CSF is synthesized during early pregnancy in epithelial cells lining the oviduct and uterus in rodents (6), women (7), and other mammalian species including sheep, cattle, and pigs (8, 9, 10). Blastocyst development is impaired in GM-CSF null mutant mice and is linked with altered placental structure and decreased fetal size, increased fetal loss in late gestation, and mortality during early postnatal life (4, 5). In the mouse, ovarian steroid hormones and factors in seminal plasma act synergistically to induce GM-CSF expression at the time of conception (11, 12). GM-CSF targets the preimplantation embryo to promote blastocyst formation, increasing the number of viable blastomeres through inhibiting apoptosis and facilitating glucose uptake (5). Human embryos cultured in GM-CSF are twice as likely to reach the blastocyst stage of development, blastulate earlier, and have increased cell numbers both in the inner cell mass and trophectoderm (13) with reduced apoptosis (14). In both mouse and human embryos, the effects of GM-CSF are mediated through binding to the -subunit of the GM-CSF receptor independently of the ?-subunit of the GM-CSF receptor, ?c (5, 14).
The significance of depriving the embryo of growth factors for long-term health and viability of the fetus has not been explored. With the advent of assisted reproductive technologies, some 2% of children in western countries are now born after in vitro fertilization (IVF) involving culture of the preimplantation embryo in media that do not contain growth factors and so only partially mimic the physiological environment. Critical evaluation of the deficiencies of in vitro culture is now essential in view of recent epidemiological studies that demonstrate growth impairment (15, 16) and an increased likelihood of major birth defects (17) in children born after IVF. Furthermore there is emerging information to link IVF and intracytoplasmic sperm injection techniques with increased incidence of defects in genomic imprinting, the process by which the genome undergoes epigenetic reprogramming in the developing preimplantation embryo (18, 19, 20, 21).
Animal experiments offer compelling evidence that in vitro culture provides a less than ideal environment, with adverse effects on the embryo reflected in compromised fetal growth and development. In rodents, embryo culture in media free of growth factors decreases postimplantation viability and fetal growth rate (22, 23), and effects are exacerbated in media supplemented with serum (24). In ruminants, abnormally large offspring with reports of 20–30% increases in birth weight can result from embryos developed in vitro (25, 26). Aberrant genomic imprinting has been implicated in mediating the effects of culture-induced embryo stress associated with suboptimal culture media (27, 28, 29), supplementation with serum (24), nutrient deprivation (30), or accumulation of toxic metabolites (31). Modified gene methylation patterns are also evident in children conceived by assisted reproductive technologies diagnosed with imprinting disorders (19, 20, 21). The consequences of growth impairment in utero are long-lived, with low birth weight being associated with permanent alterations in metabolic parameters leading to increased susceptibility to chronic disease in adult life (32, 33).
We have now investigated the physiological significance of exposure to GM-CSF during early preimplantation embryo development for subsequent fetal and postnatal growth and body composition. Embryo culture during the preimplantation period is found to have a detrimental influence on fetal and postnatal development in association with altered placental morphogenesis. Addition of GM-CSF to culture medium improves implantation rate, corrects deficiencies in placental structure and fetal growth trajectory, and partly alleviates the long-term adverse effects of embryo culture on postnatal growth in adult mice. This identifies GM-CSF as essential for optimal preimplantation development, as able to ensure normal fetal development in the absence of embryonic exposure to other growth factors, and as a candidate for use in reproductive technologies in humans and other species.
Materials and Methods
Animals, embryo collection, and culture
Ethics approval for this study was obtained from the University of G?teborg (G?teborg) (167:2000), and all experimentation was conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (1996). Mice were housed in an specific pathogen-free unit on a 12-h light, 12-h dark cycle, in cages supplied individually with filtered air, and were provided with food and water ad libitum. After embryo transfer, recipient female mice were housed with not more than four females in each cage, and from d 17 of pregnancy, pregnant females were caged singly. The mother and litter were held together until weaning at 21 d of age, when litters were separated into male and female offspring. From weaning, male offspring were housed with littermates till the age of 7 wk and then caged individually to prevent fighting. Female offspring were housed with littermates with a maximum of five mice per cage.
For embryo transfer, two cell embryos collected from naturally ovulating 10-wk-old C57BL/6 x CBA F1 (B6 F1) female mice mated with 8- to 12-wk-old NMRI males were flushed from the oviduct at 0800 h on d 2 of pregnancy and cultured at a density of 10 embryos in 20-μl drops of protein-free mouse embryo culture medium Scandinavia Quality Control (Vitrolife, G?teborg, Sweden) overlaid with paraffin oil at 37 C in 5% CO2 in air. Two-cell embryos were randomly allocated into two groups; medium alone (including carrier 2 ng/ml BSA, Sigma Chemical Co., St. Louis, MO), or medium supplemented with 2 ng/ml recombinant murine (rm) GM-CSF (R&D Systems, Abingdon, UK). A third group of in vivo-developed blastocysts were flushed from the uterus at 1200 h on d 4 of pregnancy and were held in 20-μl drops of SQC as above for approximately 2 h before embryo transfer.
SQC medium is a protein-free modification of the original G2 formulation (34, 35) with the principal changes comprising deletion of albumin, decreased phosphate ion content (0.2 mM), and increased concentrations of carbohydrates (28.5 mM sodium lactate, 0.87 mM sodium pyruvate, and 5.0 mM glucose). SQC is further supplemented with 1x MEM essential amino acids (from 100x stock, Life Technologies, Inc./Invitrogen, Carlsbad, CA) and 1x MEM nonessential amino acids (from 50x stock, Life Technologies, Inc./Invitrogen), 1x MEM Vitamin Solution (from 100x stock, Sigma), estradiol (3.7x 10–5 M, Sigma), and progesterone (3.2 x 10–6 M; Sigma). All embryo culture was performed in the mouse embryo testing facility at Scandinavian QC Laboratories AB (G?teborg, Sweden), in a controlled air and certified pathogen-free environment, with all materials and disposables tested by mouse embryo assay using test systems and laboratory protocols accredited according to ISO 17025.
Embryo transfer
Embryo recipients were naturally ovulating 10-wk-old B6 F1 females mated with vasectomized 8- to 12-wk-old NMRI males. Blastocysts were transferred at 1400–1600 h on the 4th d of culture (78–80 h after initiation of culture) or on the day of recovery from the tract for in vivo-generated embryos. To avoid selection bias, embryo culture dishes were coded, and all embryos reaching the blastocyst stage were transferred. Embryos were transferred to female recipients on d 3 of pseudopregnancy, and transfers to at least two recipients per treatment group were undertaken on a given day. Transfer was performed under general anesthesia by ip injection of ketamine (Ketalar, 75 μg/g mouse, Parke Davis, Solna, Sweden) and medetomidine (Dormitor, 1 μg/g mouse, Orion Corp., Espoo, Finland), via a dorsal incision as previously described (36). Five blastocysts were transferred to each uterine horn (10 blastocysts/recipient) with a glass embryo transfer pipette (Swemed AB, G?teborg, Sweden). After wound closure, the mouse was resuscitated with ip injection of Antisedan (1 μg/g mouse, Orion Corp.).
Analysis at d 18 of pregnancy
Pregnant recipient mice were killed at 1600 h on d 18 of pregnancy (when day of embryo transfer is defined as d 3 of pregnancy). The number of viable and resorbing implantation sites and position within each uterine horn were recorded, and fetuses and placentas were dissected and weighed.
Birth and postnatal analysis
Additional groups of mice were taken to term to allow assessment in offspring of postnatal growth trajectory, organ weights, and reproductive function. Pregnant recipients selected on the basis of their showing clear outward signs of pregnancy were placed in individual cages on d 17 of pregnancy, and cages were checked for litters every 6–12 h until birth, when the time of parturition was noted. Pups were weighed individually on d 8 postpartum and again at 3, 5, 9, and 11 wk of age after weaning on d 21 when each animal was numbered to enable tracking of individual growth trajectories. Body composition of adult progeny was assessed at 12 wk of age, when groups of 15 animals of each gender from each treatment group, randomly selected from eight litters, were killed by cervical dislocation, and internal organs and tissues were dissected and weighed individually. Organ and tissue weights were calculated relative to total body mass (for fat) or relative to lean body mass (= total body mass – total central fat mass) for all other organs and tissues. Total central fat mass was calculated as the sum of fat mass from abdominal, retroperitoneal, and parametrial deposits.
To investigate fertility of male and female progeny and the normality of second generation fetuses gestated by female progeny, naturally ovulating female progeny were mated at 12–13 wk of age with 8- to 12-wk-old NMRI males, and male progeny were mated at 10–12 wk of age with naturally ovulating 10-wk-old B6 F1 females. Implantation parameters were recorded in females mated with male progeny on d 10 of pregnancy. Female progeny mated with NRMI males were killed on d 18 of pregnancy. The number of implantation sites, viable and resorbing fetuses, and position in each uterine horn were recorded, and fetuses and placentae were dissected and weighed.
Placental histology
Two to three placentas from each of six mothers in each of the three transfer groups (n = 17 in total for each group) were dissected within 5 min of kill and fixed in 4% paraformaldehyde (Apoteksbolaget, Stockholm, Sweden) overnight. Next morning they were bisected in the midline from maternal to fetal sides to yield midsagittal faces and fixed for an additional 3 h. Placentas were then washed three times in PBS (Dulbecco’s Mg2+ and Ca2+ free, Cambrex Bio Science Verviers S.p.r.l., Verviers, Belgium) and stored at 4 C before tissue processing and paraffin wax embedding. Thirty serial midsagittal 7-μm sections were cut, placed on 10 3-aminopropyltriethoxysilane-coated slides, and the first slide with a full placental face was stained with Masson’s trichrome by a standard protocol (37). The total areas of junctional and labyrinthine zones were measured by video image analysis with Video Pro software (Leading Edge Software, Adelaide, Australia) and an Olympus BH2 light microscope equipped with a 2x objective lens and 2.5x ocular lens. Areas were expressed as proportion (percentage) of total midsagittal area comprised by junctional and labyrinthine zones.
Placental labyrinthine morphometry
To distinguish cell types within the placental labyrinth, we employed a double immunolabeling protocol originally developed for the guinea pig placental labyrinth and adapted for the mouse (38). Placental sections were double immunolabeled sequentially with rat antimouse monoclonal antibodies MTS12 (39) (kind gift of Richard Boyd, Department of Pathology and Immunology, Monash University, Melbourne, Australia) to label fetal endothelial cells and rabbit antihuman pan-cytokeratin (Zymed, San Francisco, CA) to label trophoblasts as previously described (38). Sections were deparaffinized and brought to water. For antigen retrieval, sections were incubated at 37 C for 15 min in 0.03% protease (Sigma). Endogenous peroxidase activity was quenched by incubating with 3% hydrogen peroxide for 30 min. Sections were then washed in three changes of PBS for 5 min each and blocked for nonspecific binding with serum-free protein block (DakoCytomation Denmark A/S, Glostrup, Denmark) for 10 min without washing. MTS-12 supernatant with 10% normal mouse serum and 1% BSA were applied first and incubated overnight in a humidified chamber at room temperature. Sections were washed as above, and biotinylated rabbit antirat IgG secondary antibodies (DakoCytomation Denmark A/S) were applied for 45 min, followed by washing. Streptavidin-horse radish-peroxidase (Zymed) was applied for 40 min, and sections were washed as above. MTS-12 binding was visualized by incubating with diaminobenzidine with 2% ammonium nickel (II) sulfate (Sigma) to form a black precipitate. The process was then repeated for the second primary antibody (rabbit antihuman cytokeratin, Zymed) diluted 1:50 with PBS, 10% normal mouse serum, and 1% BSA, but nickel was omitted from the chromogen, leaving a brown precipitate. Negative controls used irrelevant mouse IgG in place of the primary antibodies or the primary antibody diluent on its own.
Double-labeled sections were examined on the same imaging system as above with a 20x objective lens and a 2.5x ocular lens. Volume densities (proportions) of trophoblasts, fetal capillaries, and maternal blood space in the placental labyrinth were quantified by point counting with an isotropic L-36 Merz transparent grid placed on the monitor screen (40). Ten fields (360 points) were counted in one randomly selected section for each placenta, with the first field location chosen at random and subsequent adjacent sections systematically selected 0.5 mm apart with the aid of the stage micrometer. The volume density of each labyrinthine component was calculated using the formula: volume density, Vd = Pa/PT, where Pa is the total number of points falling on that component, and PT is the total number of points applied to the section (38, 40). The surface density (surface area per gram of placenta) of trophoblast was measured by line intercept counting on the same grid on the same fields and calculated taking into account the total magnification on the monitor using the formula: surface density, Sv = 2 x Ia/LT, where Ia is the number of intercepts with the line, and LT is the total length of the lines applied (40). The within-assay variation was <5% as determined by repeated measurement of the same placental section.
The arithmetic mean barrier to diffusion was calculated using the formula: barrier thickness, BT = Vd/Sv, where Vd is the volume density of trophoblast, and Sv is the surface density of trophoblast (40). The barrier thickness, then, is the thickness of syncytiotrophoblast. Mass of labyrinthine tissue (ML) was calculated using the formula ML = proportion (percentage) labyrinth x placental weight (in grams). The surface area (SA) of trophoblast was then calculated according to the formula SA = Sv x ML cm2.
Statistical analysis
Parameters of pregnancy and postnatal growth and placental morphometry were analyzed with SPSS 11.5 software (SPSS Inc., Chicago, IL) by one-way ANOVA followed by Bonferroni t test, except when Shapiro-Wilk test of Q-Q plots showed data were not normally distributed, when data were compared by Kruskal Wallis and Mann Whitney U tests. Adjustment for litter size in comparisons between groups was achieved with Mantel’s technique of pooling (41). Data expressed as proportions were compared by 2 test. Fat deposition was analyzed by repeated measured ANOVA with fat deposit as measure, before analysis by one-way ANOVA for individual tissues and gender of progeny. Differences were considered to be significant when P < 0.05.
Results
Effect of GM-CSF on blastocyst development and pregnancy rate
Of the two-cell embryos cultured in medium without rmGM-CSF, 415/438 (95%) reached the blastocyst stage by the time of transfer to 42 recipients (medium group). An increased proportion of the embryos cultured in media supplemented with 2 ng/ml rmGM-CSF (434 of 437; 99%, P < 0.001) (Table 1) reached the blastocyst stage by the time of transfer to 44 recipients (GM-CSF group). In addition, 383 blastocysts flushed from the uterus on d 4 of pregnancy were transferred to 40 recipients (in vivo group). Early embryo environment did not influence the likelihood of recipients becoming pregnant. Overall, 91% of recipients receiving embryos cultured in media alone, 91% of recipients receiving embryos cultured in GM-CSF, and 100% of recipients receiving embryos flushed from the tract after in vivo development carried live fetuses at d 18 of pregnancy or delivered live pups at term (Table 1). To compare overall viable implantation rates, the proportion of embryos generating viable implantation sites were combined with the proportion of embryos resulting in live offspring at term. Of the embryos transferred after culture in media alone, 72% generated viable progeny, compared with the embryos cultured in GM-CSF, of which 77% generated viable progeny (P = 0.03) (Table 1). Significantly fewer embryos flushed from the tract after in vivo development resulted in viable progeny (64%) compared with either group of in vitro-grown embryos (P < 0.001). At the time of embryo transfer, the development of in vivo-grown embryos is slightly advanced over the d 4 cultured blastocysts (data not shown), so this moderately reduced implantation rate is attributable to the greater extent of asynchrony in development between donor blastocysts and recipient uteri in this group.
TABLE 1. The effect of treatment group on blastocyst development and embryo transfer outcomes
Effect of GM-CSF on fetal and placental growth in utero
To determine the effect of culture with and without GM-CSF during the preimplantation period on subsequent survival and growth of the conceptus during pregnancy, 29–32 recipients in each group were killed approximately 30 h before birth on d 18 of pregnancy, allowing study of 218, 214, and 190 implantation sites in the medium alone, GM-CSF and in vivo groups, respectively. Although cultured embryos achieved higher implantation rates than in vivo-developed embryos, there was no effect of addition of GM-CSF on implantation success in this cohort (Table 2). The proportion of resorbing implantation sites and dead fetuses were not affected by embryo culture.
TABLE 2. The effect of in vitro culture and GM-CSF on d 18 pregnancy parameters
In vitro culture in medium alone was associated with a 10% reduction in fetal weight compared with in vivo-developed blastocysts (P < 0.001; Table 2). This was partially alleviated by addition of GM-CSF to culture medium, which resulted in fetuses that were 4% heavier than those from embryos cultured in media alone (P = 0.005), although still lighter than fetuses from in vivo-formed embryos (P < 0.001). There was no difference in placental weights between groups, but in vitro culture caused an 11% decrease in the mean fetal to placental weight ratio, a measure of placental capacity to supply fetal growth, compared with the in vivo group (P < 0.001). Culture in GM-CSF partially normalized this change, with fetal to placental weight ratio reduced by only 8% (P = 0.05; Table 2).
Effect of GM-CSF on term outcomes and postnatal growth
To investigate the effect of in vitro exposure to GM-CSF on postnatal growth trajectory and subsequent fertility and body composition in adult life, recipients from the transfer experiments were taken to term, allowing study of 74 progeny from 10 recipients in the media-alone group, 105 progeny from 12 recipients in the GM-CSF group, and 74 progeny from 11 recipients in the in vivo group (Table 3). The litter sizes at term and, thus, the proportion of embryos generating viable progeny, were larger in the group of embryos cultured with GM-CSF than in either of the other groups (P < 0.05; Table 3). There was no effect of treatment group on success of parturition or gestation length (Table 3).
TABLE 3. The effect of in vitro culture and GM-CSF on pregnancy parameters at term
There was no effect of treatment group on growth trajectory through the early postnatal period and at the time of weaning at 3 wk of age (Fig. 1, A and B). After weaning, an overcompensation in growth was seen in female progeny resulting from embryos cultured in media alone, with faster weight gain evident at 5 and 7 wk of age compared with progeny from embryos developed in vivo (Fig. 1A), before equivalent weights were reached at 9 and 11 wk. When weight differences were maximal in females at 5 wk of age, mean ± SEM weights for progeny from embryos cultured in medium alone were 22.4 ± 0.3, for the GM-CSF group were 21.5 ± 0.2, and for the in vivo group were 21.5 ± 0.3 g. Early environment had a stronger influence on growth trajectory in males, where greater weight gain in progeny derived from embryos cultured in media alone was evident from 7 wk of age and persisted until the mice were killed after 11 wk (Fig. 1B). In contrast, the presence of GM-CSF in culture medium neutralized the effect of in vitro culture on growth trajectories, and there was no difference in postnatal body size at any time point in either gender compared with progeny derived from blastocysts developed in vivo (Fig. 1). When weight differences were maximal in males at 9 wk of age, mean ± SEM weights for progeny from embryos cultured in medium alone were 39.0 ± 0.5, for the GM-CSF group were 37.6 ± 0.4, and for the in vivo group were 37.6 ± 0.5 g.
FIG. 1. The effect of GM-CSF on postnatal growth in female (A) and male (B) progeny of transfer experiments (mean ± SEM). Values shown at 1 wk after birth are combined male and female weights. The numbers of animals (n) in each group are shown. *, Statistical significance between in vivo and medium groups; P < 0.05. #, Statistical significance between medium and GM-CSF groups; P < 0.05.
The effect of in vitro culture with or without GM-CSF compared with in vivo development on body composition of adult progeny was assessed at 12 wk of age in 15 males and 15 females from each treatment group. There were substantial effects of early environment on body composition, particularly the amount of central fat in adult progeny. Embryo culture increased the absolute mass of central fat tissue in progeny by 30% (P = 0.009) and increased central fat relative to total body mass by 20% (P = 0.004). When progeny were analyzed without regard to gender, GM-CSF was seen to prevent this, and there was no difference in either measure of adiposity in adult progeny resulting from embryos cultured in the presence of cytokine, compared with progeny of embryos developed in vivo. When genders were considered independently, male adults were more prone than females to the development of increased mass of adipose tissues after embryo culture. In vitro culture increased body weight of adult male progeny by 11% compared with progeny of in vivo-developed embryos (P = 0.008; Table 4), and increased their absolute quantity of central fat by 39% (P = 0.004), due to a 36% increase in abdominal fat mass and 47% increase in retroperitoneal fat mass. Within males, increased central fat was seen in cultured embryos regardless of the presence of GM-CSF in culture medium (Table 4). When females were considered separately, the body composition and adiposity of female progeny was not affected by in vitro culture (Table 4). Early environment did not affect the absolute mass of heart, lungs, kidney, liver, spleen, thymus, or quadriceps muscle in male or female progeny (data not shown). However, when organ weights are expressed relative to lean body mass, embryo culture in media alone is associated with decreased relative brain size in males (mean ± SD = 1.25 ± 0.02% vs. 1.37 ± 0.03% body mass in in vivo controls, P = 0.03). Culture in GM-CSF restored this deficit such that relative brain mass (1.31 ± 0.02% lean body mass) was not different from values from in vivo-generated embryos (Table 4). The mass of other organs and tissues relative to lean body mass were not affected by embryo culture or exposure to GM-CSF (data not shown).
TABLE 4. The effect of embryo culture and GM-CSF on fat mass in progeny at 12 wk of age
Effect of GM-CSF on fertility in progeny
To study the fertility of progeny from the three treatment groups, groups of 10 male and 12 female progeny were mated with normal partners, and second generation fetal parameters were measured. When male progeny were mated with B6 F1 females, there were no subsequent effects of early environment on the capacity of males to sire viable pregnancies, as measured by ability to mate, plugging interval, pregnancy rate, and the viability of implantation sites on d 10 of pregnancy (Table 5).
TABLE 5. The effect of in vitro culture and GM-CSF on second generation fertility in male progeny
Naturally ovulating female progeny were mated with NMRI males and killed on d 18 of pregnancy to evaluate their fertility and fecundity. The later time point was chosen to enable differences in female tract receptivity and ability to support optimal fetal and placental growth to be evaluated. There was no difference between groups in the interval between placing with males and detection of a vaginal plug (an index of estrous cycle normality), the proportion of mated mice achieving pregnancy, or the number or viability of implantation sites. Although there was no effect of early environment on fetal weights in the second generation, mothers derived from embryos cultured in vitro were found to gestate fetuses with 19% larger placentae (P < 0.001; Table 6), causing a 14% reduction in the fetal to placental weight ratio (P < 0.001), compared with mothers from in vivo-developed embryos. This effect was alleviated by embryo culture in GM-CSF, such that placental weights and fetal to placental weight ratio values in mothers from this group were comparable with those from mothers derived from in vivo-formed embryos (Table 6).
TABLE 6. The effect of in vitro culture and GM-CSF on second generation fertility in female progeny
Effect of GM-CSF on placental histology and labyrinthine morphometry
To further examine the potential contribution of the placenta in mediating fetal growth impairment after embryo culture, structural correlates of transfer function were evaluated in placental tissue recovered at d 18 of pregnancy. There were substantial differences in placentae derived from embryos cultured in media alone compared with placentae from embryos developed in vivo. In vitro culture altered the proportions of midsagittal cross-sectional areas comprised by junctional zone and labyrinthine regions, increasing and decreasing these both by 5%, respectively (both P = 0.04; Table 7), and consequently decreasing the labyrinth to junctional zone ratio by 9% (P = 0.02; Table 7).
TABLE 7. The effect of in vitro culture and GM-CSF on placental structure at d 18 of pregnancy
The embryo culture environment also affected the proportions of cell compartments comprising the placental labyrinth (Fig. 2C). Embryo culture did not alter the volume density (proportion) of trophoblast cells but increased the volume density of the maternal blood space by 32% (P = 0.001), and decreased the volume density of fetal capillaries by 11% (P = 0.05; Fig. 3A). Furthermore, embryo culture reduced the thickness of trophoblast through which exchange occurs by 20% (P = 0.007; Fig. 3B) and increased the surface density (surface area per gram of labyrinth) of trophoblast apical membrane available for exchange in the placental labyrinth by 14% (P = 0.004; Fig. 3C). Taking into account the mass of labyrinth associated with each placenta, the total surface area for exchange was similar in placentae derived from embryos cultured in media alone and those developed in vivo (Fig. 3D).
FIG. 2. The effect of in vitro culture in medium alone and with GM-CSF on placental structure. Representative midsagittal sections of d 18 placental tissue from implantations of embryos cultured in medium alone (A) or with addition of GM-CSF (B), stained with Masson’s trichrome to distinguish junctional zone and labyrinth layers. Jz, Junctional zone; La, labyrinthine placenta. Scale bar, 300 μm. C, Representative midsagittal section of d 18 placentae double-immunolabeled to reveal structural components of the placental labyrinth, including fetal trophoblast (arrowheads), fetal capillaries (broad arrows), and maternal blood spaces (thin arrows). Scale bar, 20 μm.
FIG. 3. The effect of in vitro culture with and without GM-CSF on structural correlates of placental function. A, Volume densities (proportions) of trophoblast, maternal blood space, and fetal capillaries in the placental labyrinth (exchange region). B, Trophoblast barrier thickness. C, Surface density (surface area/g labyrinth) of trophoblast through which all exchange takes place. D, Total surface area of trophoblast for exchange. Data are means ± SEM. Different superscripted letters denote statistical significance between groups; P < 0.05.
Addition of GM-CSF to culture medium further altered placental structure, restoring the midsagittal cross-sectional areas of junctional zone and labyrinthine regions (both P = 0.03 compared with medium alone; Fig. 2, A and B) to that of placentae from embryos developed in vivo (Table 7). Early exposure to GM-CSF partially normalized the relative volume densities of cell lineages such that values were not different from those from in vivo-derived embryos, although the effect on volume density of fetal capillaries was marginal, with the relative proportion remaining 10% less than control values (P = 0.058; Fig. 3A). GM-CSF had no impact on trophoblast thickness or surface density (surface area per gram of labyrinth) of trophoblast apical membrane, which remained decreased by 14% (P = 0.052) and increased by 12% (P = 0.006), respectively, compared with placentas from in vivo-derived embryos (Fig. 3, B and C). However, because of the larger area comprised by placental labyrinth, the total surface area for exchange was 18% greater than in placentas from embryos cultured in medium without added cytokine (P = 0.05) and nearly 17% greater than in placentas from in vivo-derived controls (P = 0.03; Fig. 3D).
Discussion
GM-CSF is one of several cytokines synthesized in the female reproductive tract that act physiologically to promote growth and development of the preimplantation embryo. The current study provides evidence that exposure to GM-CSF during preimplantation embryonic life can program developmental events with consequences that manifest after implantation, resulting in increased perinatal viability. The presence of this cytokine in the early environment is identified as a key determinant of implantation success and subsequent growth and development of the fetus, with long-term consequences for growth and adiposity in adult progeny. Morphogenesis of the placental labyrinthine tissue is implicated in mediating the adverse effects of embryo culture on fetal growth and is shown to be responsive to the presence of GM-CSF in the early environment, adapting in response to this cytokine to compensate for preimplantation culture-induced stress.
This study confirms and extends previous observations showing that embryo culture is associated with a slower rate of subsequent fetal growth in utero. A similar extent of fetal growth restriction in late gestation has been reported after transfer of cultured embryos (22, 23, 24, 42). Growth impairment was found to occur despite the high rate of blastocyst development and implantation rate after embryo transfer. We attribute the high rates of pre- and postimplantation embryo development to using an embryo culture medium formulated for optimal mouse embryo development and avoiding the potential confounding influences of ovarian hyperstimulation protocols in compromising oocyte competence and eliciting developmental abnormalities (43).
When postnatal growth parameters were evaluated, impaired growth in utero was seen to be associated with overcompensation or catch-up growth after birth. Female progeny and particularly male progeny exhibited differences in growth kinetics after embryo culture, with increased whole-body weight and fat mass in adulthood, as well as decreased relative brain weight in males. This finding is reminiscent of accelerated postnatal weight gain and abnormal organ to body weight ratios reported in dietary models of fetal growth restriction, where nutritional deprivation in pregnancy permanently program metabolic parameters associated with a thrifty phenotype in the neonate (33). Growth retardation at birth after dietary restriction during pregnancy leads to elevated blood pressure, impaired glucose tolerance, and altered cholesterol metabolism in adult rats and guinea pigs (44, 45, 46, 47). This growth restriction is associated with disproportionate growth of several tissues, particularly abdominal fat deposits (48). The preimplantation period is particularly susceptible to nutritional perturbation, with dietary restriction only for the first 4 d of pregnancy resulting in fetal growth retardation followed by overcompensation in growth rate after birth (45). Notably, this pattern is consistent with that seen in humans, where catch-up growth in infancy after intrauterine growth retardation is common and independently predicts obesity, insulin resistance, and increased cardiovascular disease risk in adults (49, 50, 51, 52). As in previous studies (44, 46), we observed the effects of programing to be gender specific, with male progeny being more likely to exhibit disproportionate body structure in adult life. The first rounds of cleavage occur faster in male embryos (53), suggesting that male embryos are more responsive to growth-promoting influences and thereby more susceptible to disturbances in their supply.
Mating experiments in the progeny born after embryo transfer indicated that the effects of early environment are perpetuated into subsequent generations. Although no influence was seen in terms of fertility or fecundity, embryo culture was associated with a significant reduction in fetal to placental weight ratio in the pregnancies of females born after embryo culture. Second generation effects of maternal undernutrition have been observed in women, where reduced birth weight in women born during the Dutch Famine of 1944–1945 was perpetuated in their offspring (54). The underlying mechanisms might involve aberrant programing of metabolic status or disturbances in ovarian or uterine morphogenesis, exerted during fetal development. Spermatogenesis is presumably more resilient to perturbations in early life-reproductive function in male progeny and, at least in terms of capacity to mate and fertilize oocytes, was unimpaired.
Importantly, we provide compelling evidence that impaired fetal growth is the result of altered placental morphogenesis. Late in gestation the mouse placenta retains two distinct regions, the junctional zone that produces hormones and is analogous to the human cytotrophoblastic shell and basal plate and the placental labyrinth where all substrate exchange occurs (55, 56). The size and architecture of the placenta has been shown in previous studies to be susceptible to immunological and nutritional perturbations linked with altered fetal growth trajectory (4, 38, 57, 58). In this study, we show that embryo culture also leads to differences in the structure of the mature placenta, which are consistent with the compromised capacity to meet fetal nutrient demand as initially suggested by reduced fetal to placental weight ratio.
Embryo culture led to a reduction in the proportion of the placenta comprised by labyrinthine tissue and increase in junctional zone tissue. Furthermore, the density of the fetal capillary network within the labyrinth was reduced after culture, with a corresponding increase in maternal blood space. Rather unexpectedly, embryo culture reduced the thickness and increased the surface density (surface area per gram of labyrinth) of the trophoblast barrier through which exchange between maternal and fetal blood takes place. Because labyrinth development is driven by fetal capillary infiltration of the junctional zone tissue after fusion of the allantoic mesoderm and chorionic plate early in placental development (59), it is reasonable to speculate that impaired labyrinthine tissue morphogenesis is a consequence of restricted vascular intrusion, and because allantoic signaling molecules target trophoblast cells, changes in the barrier thickness and convolution may be secondary to this.
When embryos were exposed to GM-CSF during the culture period, the differences in developmental parameters compared with in vivo-generated embryos were partially removed. GM-CSF was seen, as expected on the basis of previous experiments (5), to improve the rate of embryo development to blastocyst stage and to enhance the proportion of transferred embryos developing into viable progeny. Embryo exposure to GM-CSF resulted in a fetal growth trajectory similar to that of in vivo-developed embryos. In addition, the effect of culture on fetal to placental weight ratios in second generation pregnancies was not evident, when mice were derived from embryos cultured with GM-CSF. However, increased adiposity induced in adult progeny by embryo culture was not prevented by GM-CSF, indicating that this growth factor on its own does not completely compensate for the detrimental effects of the culture environment, particularly in males.
Partial reversal of the adverse effects of culture with addition of GM-CSF was associated with effects on placental development. Placentae from the GM-CSF group were structurally more like those derived from embryos developed in vivo than those derived from embryos cultured in media alone. Early exposure to GM-CSF alleviated the culture-induced reduction in labyrinthine tissue and increase in junctional zone tissue. In addition, exogenous GM-CSF restored the proportions of the labyrinth comprised of trophoblast and fetal capillaries to approximate those of in vivo controls, and barrier thickness in placentae from embryos cultured in the presence of GM-CSF was intermediate between values for the media-alone and in vivo-derived groups. Although the extent of trophoblast convolution as measured by surface density was not altered by GM-CSF, the absolute trophoblast surface area for exchange was increased as a consequence of the enlarged placental labyrinth, allowing a greater total area for nutrient exchange even compared with that of control placentae from in vivo-derived embryos. These observations indicate that early exposure to GM-CSF in part determines the structural conformation and function of the placenta in late gestation, facilitating adaptations during placental morphogenesis that largely compensate for deficiencies in placental function resulting from a less than optimal preimplantation environment.
These findings are consistent with previous observations of compromised pregnancy outcomes in mice with a null mutation in GM-CSF, where fetal growth retardation was linked with altered placental structure including a similar decrease in the area of the labyrinthine tissue (4). Although compromised postimplantation development was associated in null mutant mice with retarded blastocyst development, an influence of GM-CSF deficiency in the maternal immune axis or in placental morphogenesis could not be excluded as causal in that model. The similar outcomes now obtained when GM-CSF deprivation is limited to preimplantation development show that the reproductive phenotype in null mutant mice is largely the consequence of early events. The more extreme growth retardation seen in genetic GM-CSF deficiency might be explained by a further role for GM-CSF in directly targeting trophoblast cell proliferation and differentiation in the developing placenta after implantation or in fetal growth itself (60), raising the question of whether GM-CSF or other growth factors present in the postimplantation environment modulate the impact of preimplantation perturbation.
This study provides compelling evidence that the growth factor milieu experienced by the preimplantation embryo is an important mechanistic determinant of fetal programing. The finding has significant implications for our understanding of the causal pathways underlying the fetal origins of adult disease. Despite GM-CSF restoring or enhancing placental structural determinants of function and fetal growth, adult obesity in males persisted, suggesting that more than one mechanistic pathway contributes to the culture-induced phenotype. One potential explanation is that cell lineages in the trophectoderm and inner cell mass are differentially responsive to growth factor modulation of culture-induced stress. The potential for early events to perturb programing is indicated by previous observations, showing that alterations in the abundance and proportion of inner cell mass cell and trophectoderm cell lineages in the blastocyst at implantation can influence subsequent embryonic development. In particular, a reduction in inner cell mass cells resulting from manipulation of maternal nutrition (45), culture environment (42), or chemical reduction in the number of blastomeres (61) has been linked with retardation of fetal growth. In the case of nutritional perturbation during the preimplantation period in rats (45), which gives rise to slower developing blastocysts having fewer inner cell mass cells, effects in the fetus were carried forward into adult growth trajectory and were comparable in nature and magnitude to the consequences of embryo culture seen in the current experiment. A similar linkage is seen in diabetic rats, where blastocysts with a decreased proportion of cells allocated to the inner cell mass are associated with subsequent inhibition of fetal growth (62). A difficulty with these experiments, and indeed with any experiment involving endogenous manipulation, is that a direct relationship between blastocyst parameters and fetal development cannot be proven because of the possibility of indirect confounding influences on other determinants of fetal growth, for example endometrial receptivity. The current experimental strategy employing ex vivo manipulation and embryo transfer overcomes this confounding issue.
Previous studies investigating cytokines and growth factors expressed in the reproductive tract have linked their impact on blastocyst development with implantation success. Epidermal growth factor is a key autocrine growth factor in preimplantation embryos (63), and its supplementation in culture media increases implantation rate after transfer in mice (64). Insulin stimulates cell proliferation in the inner cell mass in vitro in the rat (65) and increases the posttransfer rates of implantation, fetal survival, and weight at birth both in rats (66) and in mice (23). Similarly, platelet-activating factor exerts embryotrophic effects in vitro and increases the proportion of embryos that develop normally after transfer in the mouse (67). Other growth factors thus potentially act in synergy with GM-CSF and together might more adequately protect embryos from the deficit in postnatal outcomes imposed by culture.
The mechanism by which GM-CSF exposure during the preimplantation period determines placental structure and fetal growth is not clear but presumably stems from the same molecular pathways by which GM-CSF promotes blastocyst viability. We have shown that GM-CSF can inhibit apoptosis and promote viability of blastomeres, particularly in the inner cell mass (5, 14). Because GM-CSF promotes glucose transport in embryos, it is possible that its effects are mediated through enhancing metabolic activity. Altered embryo access to glucose or other metabolic substrates is proposed to mediate the consequences of maternal nutritional deprivation (45); thus, disruptions in the metabolic status of the preimplantation embryo could explain the converging effects of growth factors and nutrition on blastocyst development. An alternative interpretation more consistent with evidence that low metabolic activity favors optimal embryo development (68) is that growth factors such as GM-CSF function simply as cell survival signals to prevent activation of the apoptotic cascade in blastomeres.
The means by which memory of metabolic environment in early life may be perpetuated into later fetal and adult life remains ill-defined. It has been speculated that preimplantation stress might result in inappropriate stem cell allocation for normal growth (69), but emerging studies favor mechanisms involving aberrant genomic imprinting, an alternative more easily reconciled with disrupted placental development. Defects are evident in imprinted genes Igf2, H19, Grb10, and Grb7 in mouse embryos after culture in the presence of serum (24) or with suboptimal culture media (27). Aberrant fetal growth and development in cattle and sheep after embryo culture are attributed to imprinting errors in Igf2R after metabolic stress (29, 70). It is known that a number of imprinted genes have important roles in placental development, and placental cell lineages may be relatively more susceptible to persistent imprinting defects after implantation (71). One such gene is Mash-2 (72), which acts in the placenta to promote development of the spongiotrophoblast cell layer within the junctional zone of the placenta. Imprinting aberration in Mash-2 is a candidate mechanism for the placental abnormalities related to large offspring syndrome (73, 74). Whether and how cytokine regulation of viability and metabolic status in blastomeres is linked with the process of imprinting remains to be determined. Because expression of methyltransferases and other methylation machinery are cell cycle-regulated, disruption in their function is proposed to occur after the timing of embryo development is slowed in culture (75), whereas embryotrophic growth factors that accelerate blastomere cell division would oppose this effect. Alternatively, GM-CSF may directly influence epigenetic modification by modulating histone acetylation, specifically through phosphorylation of p44 MAP kinase (76), which activates CREB binding protein histone acetyl transferase (77).
In vitro conditions for culture of human embryos are generally considered to be suboptimal and are believed to compromise embryo quality and contribute to the high rates of implantation failure and impaired fetal outcomes seen in human IVF and related reproductive technologies (15, 16, 17, 18, 19). Inclusion of GM-CSF in human embryo culture media may provide a preimplantation environment more closely approximating the physiological environment and might be particularly beneficial in growing embryos to the blastocyst stage before embryo transfer, the benefits of which include higher implantation rates and better prospects for single embryo transfer to prevent multiple pregnancy (78, 79). Although the current findings provide a path forward for potentially improving perinatal health after IVF, residual adverse programing outcomes indicate the ongoing necessity for a better understanding of the significance of the early environment and its molecular legacy in the embryo to garner further improvements in embryo culture conditions. Thus, the discovery of growth factors that optimize early embryo development has considerable clinical significance and improves the likelihood of eventually defining an embryo growth medium that truly replicates the in vivo milieu.
Acknowledgments
We thank Sofie Nilsson, Aishling O’Connell, and Rebecca Skinner for technical assistance and Professors Julie Owens and Jeffrey Robinson for their critical appraisal of the manuscript and helpful comments.
References
Pampfer S, Arceci RJ, Pollard JW 1991 Role of colony stimulating factor-1 (CSF-1) and other lympho-hematopoietic growth factors in mouse pre-implantation development. Bioessays 13:535–540
Kaye PL, Harvey MB 1995 The role of growth factors in preimplantation development. Prog Growth Factor Res 6:1–24
Kane MT, Morgan PM, Coonan C 1997 Peptide growth factors and preimplantation development. Hum Reprod Update 3:137–157
Robertson SA, Roberts CT, Farr KL, Dunn AR, Seamark RF 1999 Fertility impairment in granulocyte-macrophage colony-stimulating factor-deficient mice. Biol Reprod 60:251–261
Robertson SA, Sjoblom C, Jasper MJ, Norman RJ, Seamark RF 2001 Granulocyte-macrophage colony-stimulating factor promotes glucose transport and blastomere viability in murine preimplantation embryos. Biol Reprod 64:1206–1215
Robertson SA, Mayrhofer G, Seamark RF 1992 Uterine epithelial cells synthesize granulocyte-macrophage colony-stimulating factor and interleukin-6 in pregnant and nonpregnant mice. Biol Reprod 46:1069–1079
Zhao Y, Chegini N 1994 Human fallopian tube expresses granulocyte-macrophage colony stimulating factor (GM-CSF) and GM-CSF and ? receptors and contain immunoreactive GM-CSF protein. J Clin Endocrinol Metab 79:662–665
Imakawa K, Helmer SD, Nephew KP, Meka CS, Christenson RK 1993 A novel role for GM-CSF: enhancement of pregnancy specific interferon production, ovine trophoblast protein-1. Endocrinology 132:1869–1871
Emond V, MacLaren LA, Kimmins S, Arosh JA, Fortier MA, Lambert RD 2004 Expression of cyclooxygenase-2 and granulocyte-macrophage colony-stimulating factor in the endometrial epithelium of the cow is up-regulated during early pregnancy and in response to intrauterine infusions of interferon-. Biol Reprod 70:54–64
O‘Leary S, Jasper MJ, Warnes GM, Armstrong DT, Robertson SA 2004 Seminal plasma regulates endometrial cytokine expression, leukocyte recruitment and embryo development in the pig. Reproduction 128:237–247
Robertson SA, Mayrhofer G, Seamark RF 1996 Ovarian steroid hormones regulate granulocyte-macrophage colony-stimulating factor synthesis by uterine epithelial cells in the mouse. Biol Reprod 54:183–196
Tremellen KP, Seamark RF, Robertson SA 1998 Seminal transforming growth factor ?1 stimulates granulocyte-macrophage colony-stimulating factor production and inflammatory cell recruitment in the murine uterus. Biol Reprod 58:1217–1225
Sjoblom C, Wikland M, Robertson SA 1999 Granulocyte-macrophage colony-stimulating factor promotes human blastocyst development in vitro. Hum Reprod 14:3069–3076
Sjoblom C, Wikland M, Robertson SA 2002 Granulocyte-macrophage colony-stimulating factor (GM-CSF) acts independently of the ? common subunit of the GM-CSF receptor to prevent inner cell mass apoptosis in human embryos. Biol Reprod 67:1817–1823
Schieve LA, Meikle SF, Ferre C, Peterson HB, Jeng G, Wilcox LS 2002 Low and very low birth weight in infants conceived with use of assisted reproductive technology. N Engl J Med 346:731–737
Jackson RA, Gibson KA, Wu YW, Croughan MS 2004 Perinatal outcomes in singletons following in vitro fertilization: a meta-analysis. Obstet Gynecol 103:551–563
Hansen M, Kurinczuk JJ, Bower C, Webb S 2002 The risk of major birth defects after intracytoplasmic sperm injection and in vitro fertilization. N Engl J Med 346:725–730
Maher ER, Afnan M, Barratt CL 2003 Epigenetic risks related to assisted reproductive technologies: epigenetics, imprinting, ART and icebergs? Hum Reprod 18:2508–2511
Gosden R, Trasler J, Lucifero D, Faddy M 2003 Rare congenital disorders, imprinted genes, and assisted reproductive technology. Lancet 361:1975–1977
Cox GF, Burger J, Lip V, Mau UA, Sperling K, Wu BL, Horsthemke B 2002 Intracytoplasmic sperm injection may increase the risk of imprinting defects. Am J Hum Genet 71:162–164
DeBaun MR, Niemitz EL, Feinberg AP 2003 Association of in vitro fertilization with Beckwith-Wiedemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet 72:156–160
Bowman P, McLaren A 1970 Viability and growth of mouse embryos after in vitro culture and fusion. J Embryol Exp Morphol 23:693–704
Kaye PL, Gardner HG 1999 Preimplantation access to maternal insulin and albumin increases fetal growth rate in mice. Hum Reprod 14:3052–3059
Khosla S, Dean W, Brown D, Reik W, Feil R 2001 Culture of preimplantation mouse embryos affects fetal development and the expression of imprinted genes. Biol Reprod 64:918–926
Thompson JG, Gardner DK, Pugh PA, McMillan WH, Tervit HR 1995 Lamb birth weight is affected by culture system utilized during in vitro pre-elongation development of ovine embryos. Biol Reprod 53:1385–1391
Sinclair KD, Young LE, Wilmut I, McEvoy TG 2000 In-utero overgrowth in ruminants following embryo culture: lessons from mice and a warning to men. Hum Reprod 15(Suppl 5):68–86
Doherty AS, Mann MR, Tremblay KD, Bartolomei MS, Schultz RM 2000 Differential effects of culture on imprinted H19 expression in the preimplantation mouse embryo. Biol Reprod 62:1526–1535
Stojanov T, O’Neill C 2001 In vitro fertilization causes epigenetic modifications to the onset of gene expression from the zygotic genome in mice. Biol Reprod 64:696–705
Young LE, Fernandes K, McEvoy TG, Butterwith SC, Gutierrez CG, Carolan C, Broadbent PJ, Robinson JJ, Wilmut I, Sinclair KD 2001 Epigenetic change in IGF2R is associated with fetal overgrowth after sheep embryo culture. Nat Genet 27:153–154
Waterland RA, Jirtle RL 2004 Early nutrition, epigenetic changes at transposons and imprinted genes, and enhanced susceptibility to adult chronic diseases. Nutrition 20:63–68
Lane M, Gardner DK 2003 Ammonium induces aberrant blastocyst differentiation, metabolism, pH regulation, gene expression and subsequently alters fetal development in the mouse. Biol Reprod 69:1109–1117
Barker DJ, Clark PM 1997 Fetal undernutrition and disease in later life. Rev Reprod 2:105–112
Hales CN, Ozanne SE 2003 The dangerous road of catch-up growth. J Physiol 547:5–10
Barnes FL, Crombie A, Gardner DK, Kausche A, Lacham-Kaplan O, Suikkari AM, Tiglias J, Wood C, Trounson AO 1995 Blastocyst development and birth after in-vitro maturation of human primary oocytes, intracytoplasmic sperm injection and assisted hatching. Hum Reprod 10:3243–3247
Gardner DK, Lane M 1997 Culture and selection of viable blastocysts: a feasible proposition for human IVF? Hum Reprod Update 3:367–382
Hogan B, Constantini F, Lacy E 1986 Manipulating the mouse embryo. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory
Drury RAB, Wallington EA 1980 Carleton‘s histological technique. 5th ed. Oxford, UK: Oxford University Press
Roberts CT, Sohlstrom A, Kind KL, Earl RA, Khong TY, Robinson JS, Owens PC, Owens JA 2001 Maternal food restriction reduces the exchange surface area and increases the barrier thickness of the placenta in the guinea-pig. Placenta 22:177–185
Godfrey DI, Izon DJ, Tucek CL, Wilson TJ, Boyd RL 1990 The phenotypic heterogeneity of mouse thymic stromal cells. Immunol 70:66–74
Weibel ER 1979 Morphometry of the human lung: the state of the art after two decades. Bull Eur Physiopathol Respir 15:999–1013
Mantel N 1963 Chi-square test with one degree of freedom. Extension of the Mantel-Haenszel procedure. J Am Stat Assoc 59:690–700
Lane M, Gardner DK 1997 Differential regulation of mouse embryo development and viability by amino acids. J Reprod Fertil 109:153–164
Shi W, Haaf T 2002 Aberrant methylation patterns at the two-cell stage as an indicator of early developmental failure. Mol Reprod Dev 63:329–334
Kind KL, Clifton PM, Katsman AI, Tsiounis M, Robinson JS, Owens JA 1999 Restricted fetal growth and the response to dietary cholesterol in the guinea pig. Am J Physiol 277:R1675–R1682
Kwong WY, Wild AE, Roberts P, Willis AC, Fleming TP 2000 Maternal undernutrition during the preimplantation period of rat development causes blastocyst abnormalities and programming of postnatal hypertension. Development 127:4195–4202
Ozaki T, Nishina H, Hanson MA, Poston L 2001 Dietary restriction in pregnant rats causes gender-related hypertension and vascular dysfunction in offspring. J Physiol 530:141–152
Kind KL, Simonetta G, Clifton PM, Robinson JS, Owens JA 2002 Effect of maternal feed restriction on blood pressure in the adult guinea pig. Exp Physiol 87:469–477
Vickers MH, Breier BH, Cutfield WS, Hofman PL, Gluckman PD 2000 Fetal origins of hyperphagia, obesity, and hypertension and postnatal amplification by hypercaloric nutrition. Am J Physiol Endocrinol Metab 279:E83–E87
Eriksson JG, Forsen T, Tuomilehto J, Winter PD, Osmond C, Barker DJ 1999 Catch-up growth in childhood and death from coronary heart disease: longitudinal study. BMJ 318:427–431
Ong KK, Ahmed ML, Emmett PM, Preece MA, Dunger DB 2000 Association between postnatal catch-up growth and obesity in childhood: prospective cohort study. BMJ 320:967–971
Veening MA, Van Weissenbruch MM, Delemarre-Van De Waal HA 2002 Glucose tolerance, insulin sensitivity, and insulin secretion in children born small for gestational age. J Clin Endocrinol Metab 87:4657–4661
Parker L, Lamont DW, Unwin N, Pearce MS, Bennett SM, Dickinson HO, White M, Mathers JC, Alberti KG, Craft AW 2003 A lifecourse study of risk for hyperinsulinaemia, dyslipidaemia and obesity (the central metabolic syndrome) at age 49–51 years. Diabet Med 20:406–415
Mittwoch U 1993 Blastocysts prepare for the race to be male. Hum Reprod 8:1550–1555
Stein AD, Lumey LH 2000 The relationship between maternal and offspring birth weights after maternal prenatal famine exposure: the Dutch Famine Birth Cohort Study. Hum Biol 72:641–654
Redline RW, Chernicky CL, Tan HQ, Ilan J 1993 Differential expression of insulin-like growth factor-II in specific regions of the late (post day 9.5) murine placenta. Mol Reprod Dev 36:121–129
Georgiades P, Ferguson-Smith AC, Burton GJ 2002 Comparative developmental anatomy of the murine and human definitive placentae. Placenta 23:3–19
Roberts CT, White CA, Wiemer NG, Ramsay A, Robertson SA 2003 Altered placental development in interleukin-10 null mutant mice. Placenta 24(Suppl A):S94–S99
White CA, Johansson M, Roberts CT, Ramsay AJ, Robertson SA 2004 Effect of interleukin-10 null mutation on maternal immune response and reproductive outcome in mice. Biol Reprod 70:123–131
Cross JC 2000 Genetic insights into trophoblast differentiation and placental morphogenesis. Semin Cell Dev Biol 11:105–113
Garcia Lloret MI, Morrish DW, Wegmann TG, Honore L, Turner AR, Guilbert LJ 1994 Demonstration of functional cytokine-placental interactions: CSF-1 and GM-CSF stimulate human cytotrophoblast differentiation and peptide hormone secretion. Exp Cell Res 214:46–54
Tam PP 1988 Postimplantation development of mitomycin C-treated mouse blastocysts. Teratology 37:205–212
Pampfer S, Wuu Y-D, Vanderheyden I, De Hertogh R 1994 Expression of tumor necrosis factor- receptors and selective effect of TNF on the inner cell mass in mouse blastocysts. Endocrinology 134:206–212
Paria BC, Dey SK 1990 Preimplantation embryo development in vitro: cooperative interactions among embryos and role of growth factors. Proc Natl Acad Sci USA 87:4756–4760
Morita Y, Tsutsumi O, Taketani Y 1994 In vitro treatment of embryos with epidermal growth factor improves viability and increases the implantation rate of blastocysts transferred to recipient mice. Am J Obstet Gynecol 171:406–409
De Hertogh R, Vanderheyden I, Pampfer S, Robin D, Dufrasne E, Delcourt J 1991 Stimulatory and inhibitory effects of glucose and insulin on rat blastocyst development in vitro. Diabetes 40:641–647
Zhang X, Armstrong DT 1990 Presence of amino acids and insulin in a chemically defined medium improves development of 8-cell rat embryos in vitro and subsequent implantation in vivo. Biol Reprod 42:662–668
Ryan JP, O’Neill C, Ammit AJ, Roberts CG 1992 Metabolic and developmental responses of preimplantation embryos to platelet activating factor (PAF). Reprod Fertil Dev 4:387–398
Leese HJ 2002 Quiet please, do not disturb: a hypothesis of embryo metabolism and viability. BioEssays 24:845–849
Fleming TP, Ghassemifar MR, Sheth B 2000 Junctional complexes in the early mammalian embryo. Semin Reprod Med 18:185–193
Young LE, Fairburn HR 2000 Improving the safety of embryo technologies: possible role of genomic imprinting. Theriogenology 53:627–648
Mann MR, Lee SS, Doherty AS, Verona RI, Nolen LD, Schultz RM, Bartolomei MS 2004 Selective loss of imprinting in the placenta following preimplantation development in culture. Development 131:3727–3735
Caspary T, Cleary MA, Baker CC, Guan XJ, Tilghman SM 1998 Multiple mechanisms regulate imprinting of the mouse distal chromosome 7 gene cluster. Mol Cell Biol 18:3466–3474
McLaughlin KJ, Szabo P, Haegel H, Mann JR 1996 Mouse embryos with paternal duplication of an imprinted chromosome 7 region die at midgestation and lack placental spongiotrophoblast. Development 122:265–270
Tanaka M, Gertsenstein M, Rossant J, Nagy A 1997 Mash2 acts cell autonomously in mouse spongiotrophoblast development. Dev Biol 190:55–65
De Rycke M, Liebaers I, Van Steirteghem A 2002 Epigenetic risks related to assisted reproductive technologies: risk analysis and epigenetic inheritance. Hum Reprod 17:2487–2494
Welham MJ, Duronio V, Sanghera JS, Pelech SL, Schrader JW 1992 Multiple hemopoietic growth factors stimulate activation of mitogen-activated protein kinase family members. J Immunol 149:1683–1693
Ait-Si-Ali S, Carlisi D, Ramirez S, Upegui-Gonzalez LC, Duquet A, Robin P, Rudkin B, Harel-Bellan A, Trouche D 1999 Phosphorylation by p44 MAP Kinase/ERK1 stimulates CBP histone acetyl transferase activity in vitro. Biochem Biophys Res Commun 262:157–162
Gardner DK, Vella P, Lane M, Wagley L, Schlenker T, Schoolcraft WB 1998 Culture and transfer of human blastocysts increases implantation rates and reduces the need for multiple embryo transfers. Fertil Steril 69:84–88
Jones GM, Trounson AO 1999 Blastocyst stage transfer: pitfalls and benefits. The benefits of extended culture. Hum Reprod 14:1405–1408(Cecilia Sj?blom, Claire T)
Address all correspondence and requests for reprints to: Sarah A. Robertson, Ph.D., Research Centre for Reproductive Health, Department of Obstetrics and Gynaecology, University of Adelaide, Adelaide, South Australia 5005, Australia. E-mail: sarah.robertson@adelaide.edu.au.
Abstract
Growth factors secreted by the female reproductive tract promote development of the preimplantation embryo and potentially act as epigenetic determinants of postimplantation developmental competence and pregnancy outcome. In a comprehensive embryo transfer study in mice, we examined the late gestational and postnatal effects of embryo exposure to the cytokine granulocyte-macrophage colony-stimulating factor (GM-CSF), identified as a key physiological regulator of cell number and viability in mouse and human blastocysts. Embryo development in culture in the absence of GM-CSF restricted fetal growth, accelerated postnatal growth, and increased adult body mass and adiposity in offspring compared with in vivo-grown embryos, especially in males. Addition of GM-CSF to embryo culture medium increased the proportion of transferred embryos that generated viable progeny and alleviated the effects of in vitro culture on fetal and postnatal growth trajectory but did not prevent programing of adult obesity. Placental morphogenesis was modified by embryo culture, which inhibited development of labyrinthine exchange tissue and adversely altered some structural correlates of placental transfer function. GM-CSF reversed the effect of culture on labyrinthine growth and increased the surface area of placental trophoblast available for nutrient exchange. These findings indicate that the detrimental influence of embryo culture on fetal viability and growth may be largely mediated through altered placental morphogenesis and can be alleviated by GM-CSF. This demonstrates that embryonic exposure to GM-CSF is essential for normal placental development and fetal growth.
Introduction
A HEALTHY FEMALE reproductive tract provides the optimal milieu for the developing embryo. As it traverses the oviduct and uterus before implantation, the embryo encounters an array of cytokines and growth factors that influence proliferation, viability, and differentiation of blastomeres into trophectoderm and inner cell mass lineages (1, 2, 3). Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a cytokine identified as essential for normal blastocyst development and subsequent viability (4, 5). GM-CSF is synthesized during early pregnancy in epithelial cells lining the oviduct and uterus in rodents (6), women (7), and other mammalian species including sheep, cattle, and pigs (8, 9, 10). Blastocyst development is impaired in GM-CSF null mutant mice and is linked with altered placental structure and decreased fetal size, increased fetal loss in late gestation, and mortality during early postnatal life (4, 5). In the mouse, ovarian steroid hormones and factors in seminal plasma act synergistically to induce GM-CSF expression at the time of conception (11, 12). GM-CSF targets the preimplantation embryo to promote blastocyst formation, increasing the number of viable blastomeres through inhibiting apoptosis and facilitating glucose uptake (5). Human embryos cultured in GM-CSF are twice as likely to reach the blastocyst stage of development, blastulate earlier, and have increased cell numbers both in the inner cell mass and trophectoderm (13) with reduced apoptosis (14). In both mouse and human embryos, the effects of GM-CSF are mediated through binding to the -subunit of the GM-CSF receptor independently of the ?-subunit of the GM-CSF receptor, ?c (5, 14).
The significance of depriving the embryo of growth factors for long-term health and viability of the fetus has not been explored. With the advent of assisted reproductive technologies, some 2% of children in western countries are now born after in vitro fertilization (IVF) involving culture of the preimplantation embryo in media that do not contain growth factors and so only partially mimic the physiological environment. Critical evaluation of the deficiencies of in vitro culture is now essential in view of recent epidemiological studies that demonstrate growth impairment (15, 16) and an increased likelihood of major birth defects (17) in children born after IVF. Furthermore there is emerging information to link IVF and intracytoplasmic sperm injection techniques with increased incidence of defects in genomic imprinting, the process by which the genome undergoes epigenetic reprogramming in the developing preimplantation embryo (18, 19, 20, 21).
Animal experiments offer compelling evidence that in vitro culture provides a less than ideal environment, with adverse effects on the embryo reflected in compromised fetal growth and development. In rodents, embryo culture in media free of growth factors decreases postimplantation viability and fetal growth rate (22, 23), and effects are exacerbated in media supplemented with serum (24). In ruminants, abnormally large offspring with reports of 20–30% increases in birth weight can result from embryos developed in vitro (25, 26). Aberrant genomic imprinting has been implicated in mediating the effects of culture-induced embryo stress associated with suboptimal culture media (27, 28, 29), supplementation with serum (24), nutrient deprivation (30), or accumulation of toxic metabolites (31). Modified gene methylation patterns are also evident in children conceived by assisted reproductive technologies diagnosed with imprinting disorders (19, 20, 21). The consequences of growth impairment in utero are long-lived, with low birth weight being associated with permanent alterations in metabolic parameters leading to increased susceptibility to chronic disease in adult life (32, 33).
We have now investigated the physiological significance of exposure to GM-CSF during early preimplantation embryo development for subsequent fetal and postnatal growth and body composition. Embryo culture during the preimplantation period is found to have a detrimental influence on fetal and postnatal development in association with altered placental morphogenesis. Addition of GM-CSF to culture medium improves implantation rate, corrects deficiencies in placental structure and fetal growth trajectory, and partly alleviates the long-term adverse effects of embryo culture on postnatal growth in adult mice. This identifies GM-CSF as essential for optimal preimplantation development, as able to ensure normal fetal development in the absence of embryonic exposure to other growth factors, and as a candidate for use in reproductive technologies in humans and other species.
Materials and Methods
Animals, embryo collection, and culture
Ethics approval for this study was obtained from the University of G?teborg (G?teborg) (167:2000), and all experimentation was conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (1996). Mice were housed in an specific pathogen-free unit on a 12-h light, 12-h dark cycle, in cages supplied individually with filtered air, and were provided with food and water ad libitum. After embryo transfer, recipient female mice were housed with not more than four females in each cage, and from d 17 of pregnancy, pregnant females were caged singly. The mother and litter were held together until weaning at 21 d of age, when litters were separated into male and female offspring. From weaning, male offspring were housed with littermates till the age of 7 wk and then caged individually to prevent fighting. Female offspring were housed with littermates with a maximum of five mice per cage.
For embryo transfer, two cell embryos collected from naturally ovulating 10-wk-old C57BL/6 x CBA F1 (B6 F1) female mice mated with 8- to 12-wk-old NMRI males were flushed from the oviduct at 0800 h on d 2 of pregnancy and cultured at a density of 10 embryos in 20-μl drops of protein-free mouse embryo culture medium Scandinavia Quality Control (Vitrolife, G?teborg, Sweden) overlaid with paraffin oil at 37 C in 5% CO2 in air. Two-cell embryos were randomly allocated into two groups; medium alone (including carrier 2 ng/ml BSA, Sigma Chemical Co., St. Louis, MO), or medium supplemented with 2 ng/ml recombinant murine (rm) GM-CSF (R&D Systems, Abingdon, UK). A third group of in vivo-developed blastocysts were flushed from the uterus at 1200 h on d 4 of pregnancy and were held in 20-μl drops of SQC as above for approximately 2 h before embryo transfer.
SQC medium is a protein-free modification of the original G2 formulation (34, 35) with the principal changes comprising deletion of albumin, decreased phosphate ion content (0.2 mM), and increased concentrations of carbohydrates (28.5 mM sodium lactate, 0.87 mM sodium pyruvate, and 5.0 mM glucose). SQC is further supplemented with 1x MEM essential amino acids (from 100x stock, Life Technologies, Inc./Invitrogen, Carlsbad, CA) and 1x MEM nonessential amino acids (from 50x stock, Life Technologies, Inc./Invitrogen), 1x MEM Vitamin Solution (from 100x stock, Sigma), estradiol (3.7x 10–5 M, Sigma), and progesterone (3.2 x 10–6 M; Sigma). All embryo culture was performed in the mouse embryo testing facility at Scandinavian QC Laboratories AB (G?teborg, Sweden), in a controlled air and certified pathogen-free environment, with all materials and disposables tested by mouse embryo assay using test systems and laboratory protocols accredited according to ISO 17025.
Embryo transfer
Embryo recipients were naturally ovulating 10-wk-old B6 F1 females mated with vasectomized 8- to 12-wk-old NMRI males. Blastocysts were transferred at 1400–1600 h on the 4th d of culture (78–80 h after initiation of culture) or on the day of recovery from the tract for in vivo-generated embryos. To avoid selection bias, embryo culture dishes were coded, and all embryos reaching the blastocyst stage were transferred. Embryos were transferred to female recipients on d 3 of pseudopregnancy, and transfers to at least two recipients per treatment group were undertaken on a given day. Transfer was performed under general anesthesia by ip injection of ketamine (Ketalar, 75 μg/g mouse, Parke Davis, Solna, Sweden) and medetomidine (Dormitor, 1 μg/g mouse, Orion Corp., Espoo, Finland), via a dorsal incision as previously described (36). Five blastocysts were transferred to each uterine horn (10 blastocysts/recipient) with a glass embryo transfer pipette (Swemed AB, G?teborg, Sweden). After wound closure, the mouse was resuscitated with ip injection of Antisedan (1 μg/g mouse, Orion Corp.).
Analysis at d 18 of pregnancy
Pregnant recipient mice were killed at 1600 h on d 18 of pregnancy (when day of embryo transfer is defined as d 3 of pregnancy). The number of viable and resorbing implantation sites and position within each uterine horn were recorded, and fetuses and placentas were dissected and weighed.
Birth and postnatal analysis
Additional groups of mice were taken to term to allow assessment in offspring of postnatal growth trajectory, organ weights, and reproductive function. Pregnant recipients selected on the basis of their showing clear outward signs of pregnancy were placed in individual cages on d 17 of pregnancy, and cages were checked for litters every 6–12 h until birth, when the time of parturition was noted. Pups were weighed individually on d 8 postpartum and again at 3, 5, 9, and 11 wk of age after weaning on d 21 when each animal was numbered to enable tracking of individual growth trajectories. Body composition of adult progeny was assessed at 12 wk of age, when groups of 15 animals of each gender from each treatment group, randomly selected from eight litters, were killed by cervical dislocation, and internal organs and tissues were dissected and weighed individually. Organ and tissue weights were calculated relative to total body mass (for fat) or relative to lean body mass (= total body mass – total central fat mass) for all other organs and tissues. Total central fat mass was calculated as the sum of fat mass from abdominal, retroperitoneal, and parametrial deposits.
To investigate fertility of male and female progeny and the normality of second generation fetuses gestated by female progeny, naturally ovulating female progeny were mated at 12–13 wk of age with 8- to 12-wk-old NMRI males, and male progeny were mated at 10–12 wk of age with naturally ovulating 10-wk-old B6 F1 females. Implantation parameters were recorded in females mated with male progeny on d 10 of pregnancy. Female progeny mated with NRMI males were killed on d 18 of pregnancy. The number of implantation sites, viable and resorbing fetuses, and position in each uterine horn were recorded, and fetuses and placentae were dissected and weighed.
Placental histology
Two to three placentas from each of six mothers in each of the three transfer groups (n = 17 in total for each group) were dissected within 5 min of kill and fixed in 4% paraformaldehyde (Apoteksbolaget, Stockholm, Sweden) overnight. Next morning they were bisected in the midline from maternal to fetal sides to yield midsagittal faces and fixed for an additional 3 h. Placentas were then washed three times in PBS (Dulbecco’s Mg2+ and Ca2+ free, Cambrex Bio Science Verviers S.p.r.l., Verviers, Belgium) and stored at 4 C before tissue processing and paraffin wax embedding. Thirty serial midsagittal 7-μm sections were cut, placed on 10 3-aminopropyltriethoxysilane-coated slides, and the first slide with a full placental face was stained with Masson’s trichrome by a standard protocol (37). The total areas of junctional and labyrinthine zones were measured by video image analysis with Video Pro software (Leading Edge Software, Adelaide, Australia) and an Olympus BH2 light microscope equipped with a 2x objective lens and 2.5x ocular lens. Areas were expressed as proportion (percentage) of total midsagittal area comprised by junctional and labyrinthine zones.
Placental labyrinthine morphometry
To distinguish cell types within the placental labyrinth, we employed a double immunolabeling protocol originally developed for the guinea pig placental labyrinth and adapted for the mouse (38). Placental sections were double immunolabeled sequentially with rat antimouse monoclonal antibodies MTS12 (39) (kind gift of Richard Boyd, Department of Pathology and Immunology, Monash University, Melbourne, Australia) to label fetal endothelial cells and rabbit antihuman pan-cytokeratin (Zymed, San Francisco, CA) to label trophoblasts as previously described (38). Sections were deparaffinized and brought to water. For antigen retrieval, sections were incubated at 37 C for 15 min in 0.03% protease (Sigma). Endogenous peroxidase activity was quenched by incubating with 3% hydrogen peroxide for 30 min. Sections were then washed in three changes of PBS for 5 min each and blocked for nonspecific binding with serum-free protein block (DakoCytomation Denmark A/S, Glostrup, Denmark) for 10 min without washing. MTS-12 supernatant with 10% normal mouse serum and 1% BSA were applied first and incubated overnight in a humidified chamber at room temperature. Sections were washed as above, and biotinylated rabbit antirat IgG secondary antibodies (DakoCytomation Denmark A/S) were applied for 45 min, followed by washing. Streptavidin-horse radish-peroxidase (Zymed) was applied for 40 min, and sections were washed as above. MTS-12 binding was visualized by incubating with diaminobenzidine with 2% ammonium nickel (II) sulfate (Sigma) to form a black precipitate. The process was then repeated for the second primary antibody (rabbit antihuman cytokeratin, Zymed) diluted 1:50 with PBS, 10% normal mouse serum, and 1% BSA, but nickel was omitted from the chromogen, leaving a brown precipitate. Negative controls used irrelevant mouse IgG in place of the primary antibodies or the primary antibody diluent on its own.
Double-labeled sections were examined on the same imaging system as above with a 20x objective lens and a 2.5x ocular lens. Volume densities (proportions) of trophoblasts, fetal capillaries, and maternal blood space in the placental labyrinth were quantified by point counting with an isotropic L-36 Merz transparent grid placed on the monitor screen (40). Ten fields (360 points) were counted in one randomly selected section for each placenta, with the first field location chosen at random and subsequent adjacent sections systematically selected 0.5 mm apart with the aid of the stage micrometer. The volume density of each labyrinthine component was calculated using the formula: volume density, Vd = Pa/PT, where Pa is the total number of points falling on that component, and PT is the total number of points applied to the section (38, 40). The surface density (surface area per gram of placenta) of trophoblast was measured by line intercept counting on the same grid on the same fields and calculated taking into account the total magnification on the monitor using the formula: surface density, Sv = 2 x Ia/LT, where Ia is the number of intercepts with the line, and LT is the total length of the lines applied (40). The within-assay variation was <5% as determined by repeated measurement of the same placental section.
The arithmetic mean barrier to diffusion was calculated using the formula: barrier thickness, BT = Vd/Sv, where Vd is the volume density of trophoblast, and Sv is the surface density of trophoblast (40). The barrier thickness, then, is the thickness of syncytiotrophoblast. Mass of labyrinthine tissue (ML) was calculated using the formula ML = proportion (percentage) labyrinth x placental weight (in grams). The surface area (SA) of trophoblast was then calculated according to the formula SA = Sv x ML cm2.
Statistical analysis
Parameters of pregnancy and postnatal growth and placental morphometry were analyzed with SPSS 11.5 software (SPSS Inc., Chicago, IL) by one-way ANOVA followed by Bonferroni t test, except when Shapiro-Wilk test of Q-Q plots showed data were not normally distributed, when data were compared by Kruskal Wallis and Mann Whitney U tests. Adjustment for litter size in comparisons between groups was achieved with Mantel’s technique of pooling (41). Data expressed as proportions were compared by 2 test. Fat deposition was analyzed by repeated measured ANOVA with fat deposit as measure, before analysis by one-way ANOVA for individual tissues and gender of progeny. Differences were considered to be significant when P < 0.05.
Results
Effect of GM-CSF on blastocyst development and pregnancy rate
Of the two-cell embryos cultured in medium without rmGM-CSF, 415/438 (95%) reached the blastocyst stage by the time of transfer to 42 recipients (medium group). An increased proportion of the embryos cultured in media supplemented with 2 ng/ml rmGM-CSF (434 of 437; 99%, P < 0.001) (Table 1) reached the blastocyst stage by the time of transfer to 44 recipients (GM-CSF group). In addition, 383 blastocysts flushed from the uterus on d 4 of pregnancy were transferred to 40 recipients (in vivo group). Early embryo environment did not influence the likelihood of recipients becoming pregnant. Overall, 91% of recipients receiving embryos cultured in media alone, 91% of recipients receiving embryos cultured in GM-CSF, and 100% of recipients receiving embryos flushed from the tract after in vivo development carried live fetuses at d 18 of pregnancy or delivered live pups at term (Table 1). To compare overall viable implantation rates, the proportion of embryos generating viable implantation sites were combined with the proportion of embryos resulting in live offspring at term. Of the embryos transferred after culture in media alone, 72% generated viable progeny, compared with the embryos cultured in GM-CSF, of which 77% generated viable progeny (P = 0.03) (Table 1). Significantly fewer embryos flushed from the tract after in vivo development resulted in viable progeny (64%) compared with either group of in vitro-grown embryos (P < 0.001). At the time of embryo transfer, the development of in vivo-grown embryos is slightly advanced over the d 4 cultured blastocysts (data not shown), so this moderately reduced implantation rate is attributable to the greater extent of asynchrony in development between donor blastocysts and recipient uteri in this group.
TABLE 1. The effect of treatment group on blastocyst development and embryo transfer outcomes
Effect of GM-CSF on fetal and placental growth in utero
To determine the effect of culture with and without GM-CSF during the preimplantation period on subsequent survival and growth of the conceptus during pregnancy, 29–32 recipients in each group were killed approximately 30 h before birth on d 18 of pregnancy, allowing study of 218, 214, and 190 implantation sites in the medium alone, GM-CSF and in vivo groups, respectively. Although cultured embryos achieved higher implantation rates than in vivo-developed embryos, there was no effect of addition of GM-CSF on implantation success in this cohort (Table 2). The proportion of resorbing implantation sites and dead fetuses were not affected by embryo culture.
TABLE 2. The effect of in vitro culture and GM-CSF on d 18 pregnancy parameters
In vitro culture in medium alone was associated with a 10% reduction in fetal weight compared with in vivo-developed blastocysts (P < 0.001; Table 2). This was partially alleviated by addition of GM-CSF to culture medium, which resulted in fetuses that were 4% heavier than those from embryos cultured in media alone (P = 0.005), although still lighter than fetuses from in vivo-formed embryos (P < 0.001). There was no difference in placental weights between groups, but in vitro culture caused an 11% decrease in the mean fetal to placental weight ratio, a measure of placental capacity to supply fetal growth, compared with the in vivo group (P < 0.001). Culture in GM-CSF partially normalized this change, with fetal to placental weight ratio reduced by only 8% (P = 0.05; Table 2).
Effect of GM-CSF on term outcomes and postnatal growth
To investigate the effect of in vitro exposure to GM-CSF on postnatal growth trajectory and subsequent fertility and body composition in adult life, recipients from the transfer experiments were taken to term, allowing study of 74 progeny from 10 recipients in the media-alone group, 105 progeny from 12 recipients in the GM-CSF group, and 74 progeny from 11 recipients in the in vivo group (Table 3). The litter sizes at term and, thus, the proportion of embryos generating viable progeny, were larger in the group of embryos cultured with GM-CSF than in either of the other groups (P < 0.05; Table 3). There was no effect of treatment group on success of parturition or gestation length (Table 3).
TABLE 3. The effect of in vitro culture and GM-CSF on pregnancy parameters at term
There was no effect of treatment group on growth trajectory through the early postnatal period and at the time of weaning at 3 wk of age (Fig. 1, A and B). After weaning, an overcompensation in growth was seen in female progeny resulting from embryos cultured in media alone, with faster weight gain evident at 5 and 7 wk of age compared with progeny from embryos developed in vivo (Fig. 1A), before equivalent weights were reached at 9 and 11 wk. When weight differences were maximal in females at 5 wk of age, mean ± SEM weights for progeny from embryos cultured in medium alone were 22.4 ± 0.3, for the GM-CSF group were 21.5 ± 0.2, and for the in vivo group were 21.5 ± 0.3 g. Early environment had a stronger influence on growth trajectory in males, where greater weight gain in progeny derived from embryos cultured in media alone was evident from 7 wk of age and persisted until the mice were killed after 11 wk (Fig. 1B). In contrast, the presence of GM-CSF in culture medium neutralized the effect of in vitro culture on growth trajectories, and there was no difference in postnatal body size at any time point in either gender compared with progeny derived from blastocysts developed in vivo (Fig. 1). When weight differences were maximal in males at 9 wk of age, mean ± SEM weights for progeny from embryos cultured in medium alone were 39.0 ± 0.5, for the GM-CSF group were 37.6 ± 0.4, and for the in vivo group were 37.6 ± 0.5 g.
FIG. 1. The effect of GM-CSF on postnatal growth in female (A) and male (B) progeny of transfer experiments (mean ± SEM). Values shown at 1 wk after birth are combined male and female weights. The numbers of animals (n) in each group are shown. *, Statistical significance between in vivo and medium groups; P < 0.05. #, Statistical significance between medium and GM-CSF groups; P < 0.05.
The effect of in vitro culture with or without GM-CSF compared with in vivo development on body composition of adult progeny was assessed at 12 wk of age in 15 males and 15 females from each treatment group. There were substantial effects of early environment on body composition, particularly the amount of central fat in adult progeny. Embryo culture increased the absolute mass of central fat tissue in progeny by 30% (P = 0.009) and increased central fat relative to total body mass by 20% (P = 0.004). When progeny were analyzed without regard to gender, GM-CSF was seen to prevent this, and there was no difference in either measure of adiposity in adult progeny resulting from embryos cultured in the presence of cytokine, compared with progeny of embryos developed in vivo. When genders were considered independently, male adults were more prone than females to the development of increased mass of adipose tissues after embryo culture. In vitro culture increased body weight of adult male progeny by 11% compared with progeny of in vivo-developed embryos (P = 0.008; Table 4), and increased their absolute quantity of central fat by 39% (P = 0.004), due to a 36% increase in abdominal fat mass and 47% increase in retroperitoneal fat mass. Within males, increased central fat was seen in cultured embryos regardless of the presence of GM-CSF in culture medium (Table 4). When females were considered separately, the body composition and adiposity of female progeny was not affected by in vitro culture (Table 4). Early environment did not affect the absolute mass of heart, lungs, kidney, liver, spleen, thymus, or quadriceps muscle in male or female progeny (data not shown). However, when organ weights are expressed relative to lean body mass, embryo culture in media alone is associated with decreased relative brain size in males (mean ± SD = 1.25 ± 0.02% vs. 1.37 ± 0.03% body mass in in vivo controls, P = 0.03). Culture in GM-CSF restored this deficit such that relative brain mass (1.31 ± 0.02% lean body mass) was not different from values from in vivo-generated embryos (Table 4). The mass of other organs and tissues relative to lean body mass were not affected by embryo culture or exposure to GM-CSF (data not shown).
TABLE 4. The effect of embryo culture and GM-CSF on fat mass in progeny at 12 wk of age
Effect of GM-CSF on fertility in progeny
To study the fertility of progeny from the three treatment groups, groups of 10 male and 12 female progeny were mated with normal partners, and second generation fetal parameters were measured. When male progeny were mated with B6 F1 females, there were no subsequent effects of early environment on the capacity of males to sire viable pregnancies, as measured by ability to mate, plugging interval, pregnancy rate, and the viability of implantation sites on d 10 of pregnancy (Table 5).
TABLE 5. The effect of in vitro culture and GM-CSF on second generation fertility in male progeny
Naturally ovulating female progeny were mated with NMRI males and killed on d 18 of pregnancy to evaluate their fertility and fecundity. The later time point was chosen to enable differences in female tract receptivity and ability to support optimal fetal and placental growth to be evaluated. There was no difference between groups in the interval between placing with males and detection of a vaginal plug (an index of estrous cycle normality), the proportion of mated mice achieving pregnancy, or the number or viability of implantation sites. Although there was no effect of early environment on fetal weights in the second generation, mothers derived from embryos cultured in vitro were found to gestate fetuses with 19% larger placentae (P < 0.001; Table 6), causing a 14% reduction in the fetal to placental weight ratio (P < 0.001), compared with mothers from in vivo-developed embryos. This effect was alleviated by embryo culture in GM-CSF, such that placental weights and fetal to placental weight ratio values in mothers from this group were comparable with those from mothers derived from in vivo-formed embryos (Table 6).
TABLE 6. The effect of in vitro culture and GM-CSF on second generation fertility in female progeny
Effect of GM-CSF on placental histology and labyrinthine morphometry
To further examine the potential contribution of the placenta in mediating fetal growth impairment after embryo culture, structural correlates of transfer function were evaluated in placental tissue recovered at d 18 of pregnancy. There were substantial differences in placentae derived from embryos cultured in media alone compared with placentae from embryos developed in vivo. In vitro culture altered the proportions of midsagittal cross-sectional areas comprised by junctional zone and labyrinthine regions, increasing and decreasing these both by 5%, respectively (both P = 0.04; Table 7), and consequently decreasing the labyrinth to junctional zone ratio by 9% (P = 0.02; Table 7).
TABLE 7. The effect of in vitro culture and GM-CSF on placental structure at d 18 of pregnancy
The embryo culture environment also affected the proportions of cell compartments comprising the placental labyrinth (Fig. 2C). Embryo culture did not alter the volume density (proportion) of trophoblast cells but increased the volume density of the maternal blood space by 32% (P = 0.001), and decreased the volume density of fetal capillaries by 11% (P = 0.05; Fig. 3A). Furthermore, embryo culture reduced the thickness of trophoblast through which exchange occurs by 20% (P = 0.007; Fig. 3B) and increased the surface density (surface area per gram of labyrinth) of trophoblast apical membrane available for exchange in the placental labyrinth by 14% (P = 0.004; Fig. 3C). Taking into account the mass of labyrinth associated with each placenta, the total surface area for exchange was similar in placentae derived from embryos cultured in media alone and those developed in vivo (Fig. 3D).
FIG. 2. The effect of in vitro culture in medium alone and with GM-CSF on placental structure. Representative midsagittal sections of d 18 placental tissue from implantations of embryos cultured in medium alone (A) or with addition of GM-CSF (B), stained with Masson’s trichrome to distinguish junctional zone and labyrinth layers. Jz, Junctional zone; La, labyrinthine placenta. Scale bar, 300 μm. C, Representative midsagittal section of d 18 placentae double-immunolabeled to reveal structural components of the placental labyrinth, including fetal trophoblast (arrowheads), fetal capillaries (broad arrows), and maternal blood spaces (thin arrows). Scale bar, 20 μm.
FIG. 3. The effect of in vitro culture with and without GM-CSF on structural correlates of placental function. A, Volume densities (proportions) of trophoblast, maternal blood space, and fetal capillaries in the placental labyrinth (exchange region). B, Trophoblast barrier thickness. C, Surface density (surface area/g labyrinth) of trophoblast through which all exchange takes place. D, Total surface area of trophoblast for exchange. Data are means ± SEM. Different superscripted letters denote statistical significance between groups; P < 0.05.
Addition of GM-CSF to culture medium further altered placental structure, restoring the midsagittal cross-sectional areas of junctional zone and labyrinthine regions (both P = 0.03 compared with medium alone; Fig. 2, A and B) to that of placentae from embryos developed in vivo (Table 7). Early exposure to GM-CSF partially normalized the relative volume densities of cell lineages such that values were not different from those from in vivo-derived embryos, although the effect on volume density of fetal capillaries was marginal, with the relative proportion remaining 10% less than control values (P = 0.058; Fig. 3A). GM-CSF had no impact on trophoblast thickness or surface density (surface area per gram of labyrinth) of trophoblast apical membrane, which remained decreased by 14% (P = 0.052) and increased by 12% (P = 0.006), respectively, compared with placentas from in vivo-derived embryos (Fig. 3, B and C). However, because of the larger area comprised by placental labyrinth, the total surface area for exchange was 18% greater than in placentas from embryos cultured in medium without added cytokine (P = 0.05) and nearly 17% greater than in placentas from in vivo-derived controls (P = 0.03; Fig. 3D).
Discussion
GM-CSF is one of several cytokines synthesized in the female reproductive tract that act physiologically to promote growth and development of the preimplantation embryo. The current study provides evidence that exposure to GM-CSF during preimplantation embryonic life can program developmental events with consequences that manifest after implantation, resulting in increased perinatal viability. The presence of this cytokine in the early environment is identified as a key determinant of implantation success and subsequent growth and development of the fetus, with long-term consequences for growth and adiposity in adult progeny. Morphogenesis of the placental labyrinthine tissue is implicated in mediating the adverse effects of embryo culture on fetal growth and is shown to be responsive to the presence of GM-CSF in the early environment, adapting in response to this cytokine to compensate for preimplantation culture-induced stress.
This study confirms and extends previous observations showing that embryo culture is associated with a slower rate of subsequent fetal growth in utero. A similar extent of fetal growth restriction in late gestation has been reported after transfer of cultured embryos (22, 23, 24, 42). Growth impairment was found to occur despite the high rate of blastocyst development and implantation rate after embryo transfer. We attribute the high rates of pre- and postimplantation embryo development to using an embryo culture medium formulated for optimal mouse embryo development and avoiding the potential confounding influences of ovarian hyperstimulation protocols in compromising oocyte competence and eliciting developmental abnormalities (43).
When postnatal growth parameters were evaluated, impaired growth in utero was seen to be associated with overcompensation or catch-up growth after birth. Female progeny and particularly male progeny exhibited differences in growth kinetics after embryo culture, with increased whole-body weight and fat mass in adulthood, as well as decreased relative brain weight in males. This finding is reminiscent of accelerated postnatal weight gain and abnormal organ to body weight ratios reported in dietary models of fetal growth restriction, where nutritional deprivation in pregnancy permanently program metabolic parameters associated with a thrifty phenotype in the neonate (33). Growth retardation at birth after dietary restriction during pregnancy leads to elevated blood pressure, impaired glucose tolerance, and altered cholesterol metabolism in adult rats and guinea pigs (44, 45, 46, 47). This growth restriction is associated with disproportionate growth of several tissues, particularly abdominal fat deposits (48). The preimplantation period is particularly susceptible to nutritional perturbation, with dietary restriction only for the first 4 d of pregnancy resulting in fetal growth retardation followed by overcompensation in growth rate after birth (45). Notably, this pattern is consistent with that seen in humans, where catch-up growth in infancy after intrauterine growth retardation is common and independently predicts obesity, insulin resistance, and increased cardiovascular disease risk in adults (49, 50, 51, 52). As in previous studies (44, 46), we observed the effects of programing to be gender specific, with male progeny being more likely to exhibit disproportionate body structure in adult life. The first rounds of cleavage occur faster in male embryos (53), suggesting that male embryos are more responsive to growth-promoting influences and thereby more susceptible to disturbances in their supply.
Mating experiments in the progeny born after embryo transfer indicated that the effects of early environment are perpetuated into subsequent generations. Although no influence was seen in terms of fertility or fecundity, embryo culture was associated with a significant reduction in fetal to placental weight ratio in the pregnancies of females born after embryo culture. Second generation effects of maternal undernutrition have been observed in women, where reduced birth weight in women born during the Dutch Famine of 1944–1945 was perpetuated in their offspring (54). The underlying mechanisms might involve aberrant programing of metabolic status or disturbances in ovarian or uterine morphogenesis, exerted during fetal development. Spermatogenesis is presumably more resilient to perturbations in early life-reproductive function in male progeny and, at least in terms of capacity to mate and fertilize oocytes, was unimpaired.
Importantly, we provide compelling evidence that impaired fetal growth is the result of altered placental morphogenesis. Late in gestation the mouse placenta retains two distinct regions, the junctional zone that produces hormones and is analogous to the human cytotrophoblastic shell and basal plate and the placental labyrinth where all substrate exchange occurs (55, 56). The size and architecture of the placenta has been shown in previous studies to be susceptible to immunological and nutritional perturbations linked with altered fetal growth trajectory (4, 38, 57, 58). In this study, we show that embryo culture also leads to differences in the structure of the mature placenta, which are consistent with the compromised capacity to meet fetal nutrient demand as initially suggested by reduced fetal to placental weight ratio.
Embryo culture led to a reduction in the proportion of the placenta comprised by labyrinthine tissue and increase in junctional zone tissue. Furthermore, the density of the fetal capillary network within the labyrinth was reduced after culture, with a corresponding increase in maternal blood space. Rather unexpectedly, embryo culture reduced the thickness and increased the surface density (surface area per gram of labyrinth) of the trophoblast barrier through which exchange between maternal and fetal blood takes place. Because labyrinth development is driven by fetal capillary infiltration of the junctional zone tissue after fusion of the allantoic mesoderm and chorionic plate early in placental development (59), it is reasonable to speculate that impaired labyrinthine tissue morphogenesis is a consequence of restricted vascular intrusion, and because allantoic signaling molecules target trophoblast cells, changes in the barrier thickness and convolution may be secondary to this.
When embryos were exposed to GM-CSF during the culture period, the differences in developmental parameters compared with in vivo-generated embryos were partially removed. GM-CSF was seen, as expected on the basis of previous experiments (5), to improve the rate of embryo development to blastocyst stage and to enhance the proportion of transferred embryos developing into viable progeny. Embryo exposure to GM-CSF resulted in a fetal growth trajectory similar to that of in vivo-developed embryos. In addition, the effect of culture on fetal to placental weight ratios in second generation pregnancies was not evident, when mice were derived from embryos cultured with GM-CSF. However, increased adiposity induced in adult progeny by embryo culture was not prevented by GM-CSF, indicating that this growth factor on its own does not completely compensate for the detrimental effects of the culture environment, particularly in males.
Partial reversal of the adverse effects of culture with addition of GM-CSF was associated with effects on placental development. Placentae from the GM-CSF group were structurally more like those derived from embryos developed in vivo than those derived from embryos cultured in media alone. Early exposure to GM-CSF alleviated the culture-induced reduction in labyrinthine tissue and increase in junctional zone tissue. In addition, exogenous GM-CSF restored the proportions of the labyrinth comprised of trophoblast and fetal capillaries to approximate those of in vivo controls, and barrier thickness in placentae from embryos cultured in the presence of GM-CSF was intermediate between values for the media-alone and in vivo-derived groups. Although the extent of trophoblast convolution as measured by surface density was not altered by GM-CSF, the absolute trophoblast surface area for exchange was increased as a consequence of the enlarged placental labyrinth, allowing a greater total area for nutrient exchange even compared with that of control placentae from in vivo-derived embryos. These observations indicate that early exposure to GM-CSF in part determines the structural conformation and function of the placenta in late gestation, facilitating adaptations during placental morphogenesis that largely compensate for deficiencies in placental function resulting from a less than optimal preimplantation environment.
These findings are consistent with previous observations of compromised pregnancy outcomes in mice with a null mutation in GM-CSF, where fetal growth retardation was linked with altered placental structure including a similar decrease in the area of the labyrinthine tissue (4). Although compromised postimplantation development was associated in null mutant mice with retarded blastocyst development, an influence of GM-CSF deficiency in the maternal immune axis or in placental morphogenesis could not be excluded as causal in that model. The similar outcomes now obtained when GM-CSF deprivation is limited to preimplantation development show that the reproductive phenotype in null mutant mice is largely the consequence of early events. The more extreme growth retardation seen in genetic GM-CSF deficiency might be explained by a further role for GM-CSF in directly targeting trophoblast cell proliferation and differentiation in the developing placenta after implantation or in fetal growth itself (60), raising the question of whether GM-CSF or other growth factors present in the postimplantation environment modulate the impact of preimplantation perturbation.
This study provides compelling evidence that the growth factor milieu experienced by the preimplantation embryo is an important mechanistic determinant of fetal programing. The finding has significant implications for our understanding of the causal pathways underlying the fetal origins of adult disease. Despite GM-CSF restoring or enhancing placental structural determinants of function and fetal growth, adult obesity in males persisted, suggesting that more than one mechanistic pathway contributes to the culture-induced phenotype. One potential explanation is that cell lineages in the trophectoderm and inner cell mass are differentially responsive to growth factor modulation of culture-induced stress. The potential for early events to perturb programing is indicated by previous observations, showing that alterations in the abundance and proportion of inner cell mass cell and trophectoderm cell lineages in the blastocyst at implantation can influence subsequent embryonic development. In particular, a reduction in inner cell mass cells resulting from manipulation of maternal nutrition (45), culture environment (42), or chemical reduction in the number of blastomeres (61) has been linked with retardation of fetal growth. In the case of nutritional perturbation during the preimplantation period in rats (45), which gives rise to slower developing blastocysts having fewer inner cell mass cells, effects in the fetus were carried forward into adult growth trajectory and were comparable in nature and magnitude to the consequences of embryo culture seen in the current experiment. A similar linkage is seen in diabetic rats, where blastocysts with a decreased proportion of cells allocated to the inner cell mass are associated with subsequent inhibition of fetal growth (62). A difficulty with these experiments, and indeed with any experiment involving endogenous manipulation, is that a direct relationship between blastocyst parameters and fetal development cannot be proven because of the possibility of indirect confounding influences on other determinants of fetal growth, for example endometrial receptivity. The current experimental strategy employing ex vivo manipulation and embryo transfer overcomes this confounding issue.
Previous studies investigating cytokines and growth factors expressed in the reproductive tract have linked their impact on blastocyst development with implantation success. Epidermal growth factor is a key autocrine growth factor in preimplantation embryos (63), and its supplementation in culture media increases implantation rate after transfer in mice (64). Insulin stimulates cell proliferation in the inner cell mass in vitro in the rat (65) and increases the posttransfer rates of implantation, fetal survival, and weight at birth both in rats (66) and in mice (23). Similarly, platelet-activating factor exerts embryotrophic effects in vitro and increases the proportion of embryos that develop normally after transfer in the mouse (67). Other growth factors thus potentially act in synergy with GM-CSF and together might more adequately protect embryos from the deficit in postnatal outcomes imposed by culture.
The mechanism by which GM-CSF exposure during the preimplantation period determines placental structure and fetal growth is not clear but presumably stems from the same molecular pathways by which GM-CSF promotes blastocyst viability. We have shown that GM-CSF can inhibit apoptosis and promote viability of blastomeres, particularly in the inner cell mass (5, 14). Because GM-CSF promotes glucose transport in embryos, it is possible that its effects are mediated through enhancing metabolic activity. Altered embryo access to glucose or other metabolic substrates is proposed to mediate the consequences of maternal nutritional deprivation (45); thus, disruptions in the metabolic status of the preimplantation embryo could explain the converging effects of growth factors and nutrition on blastocyst development. An alternative interpretation more consistent with evidence that low metabolic activity favors optimal embryo development (68) is that growth factors such as GM-CSF function simply as cell survival signals to prevent activation of the apoptotic cascade in blastomeres.
The means by which memory of metabolic environment in early life may be perpetuated into later fetal and adult life remains ill-defined. It has been speculated that preimplantation stress might result in inappropriate stem cell allocation for normal growth (69), but emerging studies favor mechanisms involving aberrant genomic imprinting, an alternative more easily reconciled with disrupted placental development. Defects are evident in imprinted genes Igf2, H19, Grb10, and Grb7 in mouse embryos after culture in the presence of serum (24) or with suboptimal culture media (27). Aberrant fetal growth and development in cattle and sheep after embryo culture are attributed to imprinting errors in Igf2R after metabolic stress (29, 70). It is known that a number of imprinted genes have important roles in placental development, and placental cell lineages may be relatively more susceptible to persistent imprinting defects after implantation (71). One such gene is Mash-2 (72), which acts in the placenta to promote development of the spongiotrophoblast cell layer within the junctional zone of the placenta. Imprinting aberration in Mash-2 is a candidate mechanism for the placental abnormalities related to large offspring syndrome (73, 74). Whether and how cytokine regulation of viability and metabolic status in blastomeres is linked with the process of imprinting remains to be determined. Because expression of methyltransferases and other methylation machinery are cell cycle-regulated, disruption in their function is proposed to occur after the timing of embryo development is slowed in culture (75), whereas embryotrophic growth factors that accelerate blastomere cell division would oppose this effect. Alternatively, GM-CSF may directly influence epigenetic modification by modulating histone acetylation, specifically through phosphorylation of p44 MAP kinase (76), which activates CREB binding protein histone acetyl transferase (77).
In vitro conditions for culture of human embryos are generally considered to be suboptimal and are believed to compromise embryo quality and contribute to the high rates of implantation failure and impaired fetal outcomes seen in human IVF and related reproductive technologies (15, 16, 17, 18, 19). Inclusion of GM-CSF in human embryo culture media may provide a preimplantation environment more closely approximating the physiological environment and might be particularly beneficial in growing embryos to the blastocyst stage before embryo transfer, the benefits of which include higher implantation rates and better prospects for single embryo transfer to prevent multiple pregnancy (78, 79). Although the current findings provide a path forward for potentially improving perinatal health after IVF, residual adverse programing outcomes indicate the ongoing necessity for a better understanding of the significance of the early environment and its molecular legacy in the embryo to garner further improvements in embryo culture conditions. Thus, the discovery of growth factors that optimize early embryo development has considerable clinical significance and improves the likelihood of eventually defining an embryo growth medium that truly replicates the in vivo milieu.
Acknowledgments
We thank Sofie Nilsson, Aishling O’Connell, and Rebecca Skinner for technical assistance and Professors Julie Owens and Jeffrey Robinson for their critical appraisal of the manuscript and helpful comments.
References
Pampfer S, Arceci RJ, Pollard JW 1991 Role of colony stimulating factor-1 (CSF-1) and other lympho-hematopoietic growth factors in mouse pre-implantation development. Bioessays 13:535–540
Kaye PL, Harvey MB 1995 The role of growth factors in preimplantation development. Prog Growth Factor Res 6:1–24
Kane MT, Morgan PM, Coonan C 1997 Peptide growth factors and preimplantation development. Hum Reprod Update 3:137–157
Robertson SA, Roberts CT, Farr KL, Dunn AR, Seamark RF 1999 Fertility impairment in granulocyte-macrophage colony-stimulating factor-deficient mice. Biol Reprod 60:251–261
Robertson SA, Sjoblom C, Jasper MJ, Norman RJ, Seamark RF 2001 Granulocyte-macrophage colony-stimulating factor promotes glucose transport and blastomere viability in murine preimplantation embryos. Biol Reprod 64:1206–1215
Robertson SA, Mayrhofer G, Seamark RF 1992 Uterine epithelial cells synthesize granulocyte-macrophage colony-stimulating factor and interleukin-6 in pregnant and nonpregnant mice. Biol Reprod 46:1069–1079
Zhao Y, Chegini N 1994 Human fallopian tube expresses granulocyte-macrophage colony stimulating factor (GM-CSF) and GM-CSF and ? receptors and contain immunoreactive GM-CSF protein. J Clin Endocrinol Metab 79:662–665
Imakawa K, Helmer SD, Nephew KP, Meka CS, Christenson RK 1993 A novel role for GM-CSF: enhancement of pregnancy specific interferon production, ovine trophoblast protein-1. Endocrinology 132:1869–1871
Emond V, MacLaren LA, Kimmins S, Arosh JA, Fortier MA, Lambert RD 2004 Expression of cyclooxygenase-2 and granulocyte-macrophage colony-stimulating factor in the endometrial epithelium of the cow is up-regulated during early pregnancy and in response to intrauterine infusions of interferon-. Biol Reprod 70:54–64
O‘Leary S, Jasper MJ, Warnes GM, Armstrong DT, Robertson SA 2004 Seminal plasma regulates endometrial cytokine expression, leukocyte recruitment and embryo development in the pig. Reproduction 128:237–247
Robertson SA, Mayrhofer G, Seamark RF 1996 Ovarian steroid hormones regulate granulocyte-macrophage colony-stimulating factor synthesis by uterine epithelial cells in the mouse. Biol Reprod 54:183–196
Tremellen KP, Seamark RF, Robertson SA 1998 Seminal transforming growth factor ?1 stimulates granulocyte-macrophage colony-stimulating factor production and inflammatory cell recruitment in the murine uterus. Biol Reprod 58:1217–1225
Sjoblom C, Wikland M, Robertson SA 1999 Granulocyte-macrophage colony-stimulating factor promotes human blastocyst development in vitro. Hum Reprod 14:3069–3076
Sjoblom C, Wikland M, Robertson SA 2002 Granulocyte-macrophage colony-stimulating factor (GM-CSF) acts independently of the ? common subunit of the GM-CSF receptor to prevent inner cell mass apoptosis in human embryos. Biol Reprod 67:1817–1823
Schieve LA, Meikle SF, Ferre C, Peterson HB, Jeng G, Wilcox LS 2002 Low and very low birth weight in infants conceived with use of assisted reproductive technology. N Engl J Med 346:731–737
Jackson RA, Gibson KA, Wu YW, Croughan MS 2004 Perinatal outcomes in singletons following in vitro fertilization: a meta-analysis. Obstet Gynecol 103:551–563
Hansen M, Kurinczuk JJ, Bower C, Webb S 2002 The risk of major birth defects after intracytoplasmic sperm injection and in vitro fertilization. N Engl J Med 346:725–730
Maher ER, Afnan M, Barratt CL 2003 Epigenetic risks related to assisted reproductive technologies: epigenetics, imprinting, ART and icebergs? Hum Reprod 18:2508–2511
Gosden R, Trasler J, Lucifero D, Faddy M 2003 Rare congenital disorders, imprinted genes, and assisted reproductive technology. Lancet 361:1975–1977
Cox GF, Burger J, Lip V, Mau UA, Sperling K, Wu BL, Horsthemke B 2002 Intracytoplasmic sperm injection may increase the risk of imprinting defects. Am J Hum Genet 71:162–164
DeBaun MR, Niemitz EL, Feinberg AP 2003 Association of in vitro fertilization with Beckwith-Wiedemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet 72:156–160
Bowman P, McLaren A 1970 Viability and growth of mouse embryos after in vitro culture and fusion. J Embryol Exp Morphol 23:693–704
Kaye PL, Gardner HG 1999 Preimplantation access to maternal insulin and albumin increases fetal growth rate in mice. Hum Reprod 14:3052–3059
Khosla S, Dean W, Brown D, Reik W, Feil R 2001 Culture of preimplantation mouse embryos affects fetal development and the expression of imprinted genes. Biol Reprod 64:918–926
Thompson JG, Gardner DK, Pugh PA, McMillan WH, Tervit HR 1995 Lamb birth weight is affected by culture system utilized during in vitro pre-elongation development of ovine embryos. Biol Reprod 53:1385–1391
Sinclair KD, Young LE, Wilmut I, McEvoy TG 2000 In-utero overgrowth in ruminants following embryo culture: lessons from mice and a warning to men. Hum Reprod 15(Suppl 5):68–86
Doherty AS, Mann MR, Tremblay KD, Bartolomei MS, Schultz RM 2000 Differential effects of culture on imprinted H19 expression in the preimplantation mouse embryo. Biol Reprod 62:1526–1535
Stojanov T, O’Neill C 2001 In vitro fertilization causes epigenetic modifications to the onset of gene expression from the zygotic genome in mice. Biol Reprod 64:696–705
Young LE, Fernandes K, McEvoy TG, Butterwith SC, Gutierrez CG, Carolan C, Broadbent PJ, Robinson JJ, Wilmut I, Sinclair KD 2001 Epigenetic change in IGF2R is associated with fetal overgrowth after sheep embryo culture. Nat Genet 27:153–154
Waterland RA, Jirtle RL 2004 Early nutrition, epigenetic changes at transposons and imprinted genes, and enhanced susceptibility to adult chronic diseases. Nutrition 20:63–68
Lane M, Gardner DK 2003 Ammonium induces aberrant blastocyst differentiation, metabolism, pH regulation, gene expression and subsequently alters fetal development in the mouse. Biol Reprod 69:1109–1117
Barker DJ, Clark PM 1997 Fetal undernutrition and disease in later life. Rev Reprod 2:105–112
Hales CN, Ozanne SE 2003 The dangerous road of catch-up growth. J Physiol 547:5–10
Barnes FL, Crombie A, Gardner DK, Kausche A, Lacham-Kaplan O, Suikkari AM, Tiglias J, Wood C, Trounson AO 1995 Blastocyst development and birth after in-vitro maturation of human primary oocytes, intracytoplasmic sperm injection and assisted hatching. Hum Reprod 10:3243–3247
Gardner DK, Lane M 1997 Culture and selection of viable blastocysts: a feasible proposition for human IVF? Hum Reprod Update 3:367–382
Hogan B, Constantini F, Lacy E 1986 Manipulating the mouse embryo. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory
Drury RAB, Wallington EA 1980 Carleton‘s histological technique. 5th ed. Oxford, UK: Oxford University Press
Roberts CT, Sohlstrom A, Kind KL, Earl RA, Khong TY, Robinson JS, Owens PC, Owens JA 2001 Maternal food restriction reduces the exchange surface area and increases the barrier thickness of the placenta in the guinea-pig. Placenta 22:177–185
Godfrey DI, Izon DJ, Tucek CL, Wilson TJ, Boyd RL 1990 The phenotypic heterogeneity of mouse thymic stromal cells. Immunol 70:66–74
Weibel ER 1979 Morphometry of the human lung: the state of the art after two decades. Bull Eur Physiopathol Respir 15:999–1013
Mantel N 1963 Chi-square test with one degree of freedom. Extension of the Mantel-Haenszel procedure. J Am Stat Assoc 59:690–700
Lane M, Gardner DK 1997 Differential regulation of mouse embryo development and viability by amino acids. J Reprod Fertil 109:153–164
Shi W, Haaf T 2002 Aberrant methylation patterns at the two-cell stage as an indicator of early developmental failure. Mol Reprod Dev 63:329–334
Kind KL, Clifton PM, Katsman AI, Tsiounis M, Robinson JS, Owens JA 1999 Restricted fetal growth and the response to dietary cholesterol in the guinea pig. Am J Physiol 277:R1675–R1682
Kwong WY, Wild AE, Roberts P, Willis AC, Fleming TP 2000 Maternal undernutrition during the preimplantation period of rat development causes blastocyst abnormalities and programming of postnatal hypertension. Development 127:4195–4202
Ozaki T, Nishina H, Hanson MA, Poston L 2001 Dietary restriction in pregnant rats causes gender-related hypertension and vascular dysfunction in offspring. J Physiol 530:141–152
Kind KL, Simonetta G, Clifton PM, Robinson JS, Owens JA 2002 Effect of maternal feed restriction on blood pressure in the adult guinea pig. Exp Physiol 87:469–477
Vickers MH, Breier BH, Cutfield WS, Hofman PL, Gluckman PD 2000 Fetal origins of hyperphagia, obesity, and hypertension and postnatal amplification by hypercaloric nutrition. Am J Physiol Endocrinol Metab 279:E83–E87
Eriksson JG, Forsen T, Tuomilehto J, Winter PD, Osmond C, Barker DJ 1999 Catch-up growth in childhood and death from coronary heart disease: longitudinal study. BMJ 318:427–431
Ong KK, Ahmed ML, Emmett PM, Preece MA, Dunger DB 2000 Association between postnatal catch-up growth and obesity in childhood: prospective cohort study. BMJ 320:967–971
Veening MA, Van Weissenbruch MM, Delemarre-Van De Waal HA 2002 Glucose tolerance, insulin sensitivity, and insulin secretion in children born small for gestational age. J Clin Endocrinol Metab 87:4657–4661
Parker L, Lamont DW, Unwin N, Pearce MS, Bennett SM, Dickinson HO, White M, Mathers JC, Alberti KG, Craft AW 2003 A lifecourse study of risk for hyperinsulinaemia, dyslipidaemia and obesity (the central metabolic syndrome) at age 49–51 years. Diabet Med 20:406–415
Mittwoch U 1993 Blastocysts prepare for the race to be male. Hum Reprod 8:1550–1555
Stein AD, Lumey LH 2000 The relationship between maternal and offspring birth weights after maternal prenatal famine exposure: the Dutch Famine Birth Cohort Study. Hum Biol 72:641–654
Redline RW, Chernicky CL, Tan HQ, Ilan J 1993 Differential expression of insulin-like growth factor-II in specific regions of the late (post day 9.5) murine placenta. Mol Reprod Dev 36:121–129
Georgiades P, Ferguson-Smith AC, Burton GJ 2002 Comparative developmental anatomy of the murine and human definitive placentae. Placenta 23:3–19
Roberts CT, White CA, Wiemer NG, Ramsay A, Robertson SA 2003 Altered placental development in interleukin-10 null mutant mice. Placenta 24(Suppl A):S94–S99
White CA, Johansson M, Roberts CT, Ramsay AJ, Robertson SA 2004 Effect of interleukin-10 null mutation on maternal immune response and reproductive outcome in mice. Biol Reprod 70:123–131
Cross JC 2000 Genetic insights into trophoblast differentiation and placental morphogenesis. Semin Cell Dev Biol 11:105–113
Garcia Lloret MI, Morrish DW, Wegmann TG, Honore L, Turner AR, Guilbert LJ 1994 Demonstration of functional cytokine-placental interactions: CSF-1 and GM-CSF stimulate human cytotrophoblast differentiation and peptide hormone secretion. Exp Cell Res 214:46–54
Tam PP 1988 Postimplantation development of mitomycin C-treated mouse blastocysts. Teratology 37:205–212
Pampfer S, Wuu Y-D, Vanderheyden I, De Hertogh R 1994 Expression of tumor necrosis factor- receptors and selective effect of TNF on the inner cell mass in mouse blastocysts. Endocrinology 134:206–212
Paria BC, Dey SK 1990 Preimplantation embryo development in vitro: cooperative interactions among embryos and role of growth factors. Proc Natl Acad Sci USA 87:4756–4760
Morita Y, Tsutsumi O, Taketani Y 1994 In vitro treatment of embryos with epidermal growth factor improves viability and increases the implantation rate of blastocysts transferred to recipient mice. Am J Obstet Gynecol 171:406–409
De Hertogh R, Vanderheyden I, Pampfer S, Robin D, Dufrasne E, Delcourt J 1991 Stimulatory and inhibitory effects of glucose and insulin on rat blastocyst development in vitro. Diabetes 40:641–647
Zhang X, Armstrong DT 1990 Presence of amino acids and insulin in a chemically defined medium improves development of 8-cell rat embryos in vitro and subsequent implantation in vivo. Biol Reprod 42:662–668
Ryan JP, O’Neill C, Ammit AJ, Roberts CG 1992 Metabolic and developmental responses of preimplantation embryos to platelet activating factor (PAF). Reprod Fertil Dev 4:387–398
Leese HJ 2002 Quiet please, do not disturb: a hypothesis of embryo metabolism and viability. BioEssays 24:845–849
Fleming TP, Ghassemifar MR, Sheth B 2000 Junctional complexes in the early mammalian embryo. Semin Reprod Med 18:185–193
Young LE, Fairburn HR 2000 Improving the safety of embryo technologies: possible role of genomic imprinting. Theriogenology 53:627–648
Mann MR, Lee SS, Doherty AS, Verona RI, Nolen LD, Schultz RM, Bartolomei MS 2004 Selective loss of imprinting in the placenta following preimplantation development in culture. Development 131:3727–3735
Caspary T, Cleary MA, Baker CC, Guan XJ, Tilghman SM 1998 Multiple mechanisms regulate imprinting of the mouse distal chromosome 7 gene cluster. Mol Cell Biol 18:3466–3474
McLaughlin KJ, Szabo P, Haegel H, Mann JR 1996 Mouse embryos with paternal duplication of an imprinted chromosome 7 region die at midgestation and lack placental spongiotrophoblast. Development 122:265–270
Tanaka M, Gertsenstein M, Rossant J, Nagy A 1997 Mash2 acts cell autonomously in mouse spongiotrophoblast development. Dev Biol 190:55–65
De Rycke M, Liebaers I, Van Steirteghem A 2002 Epigenetic risks related to assisted reproductive technologies: risk analysis and epigenetic inheritance. Hum Reprod 17:2487–2494
Welham MJ, Duronio V, Sanghera JS, Pelech SL, Schrader JW 1992 Multiple hemopoietic growth factors stimulate activation of mitogen-activated protein kinase family members. J Immunol 149:1683–1693
Ait-Si-Ali S, Carlisi D, Ramirez S, Upegui-Gonzalez LC, Duquet A, Robin P, Rudkin B, Harel-Bellan A, Trouche D 1999 Phosphorylation by p44 MAP Kinase/ERK1 stimulates CBP histone acetyl transferase activity in vitro. Biochem Biophys Res Commun 262:157–162
Gardner DK, Vella P, Lane M, Wagley L, Schlenker T, Schoolcraft WB 1998 Culture and transfer of human blastocysts increases implantation rates and reduces the need for multiple embryo transfers. Fertil Steril 69:84–88
Jones GM, Trounson AO 1999 Blastocyst stage transfer: pitfalls and benefits. The benefits of extended culture. Hum Reprod 14:1405–1408(Cecilia Sj?blom, Claire T)