| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
Departments of Obstetrics and Gynecology,3 and Dermatology and Plastic Surgery,4 Akita University School of Medicine, Akita, 010-8543 Japan
ABSTRACT
Survival and apoptosis of cells in preimplantation embryos are fundamental for successful pregnancy. Relevant to these processes, tumor necrosis factor (TNF) and transforming growth factor alpha (TGFA) are produced by mammalian oviducts and uteri. In early embryos, TNF induces apoptosis, whereas TGFA could act as a survival factor. Here we investigated the TNF regulation of apoptosis in early mouse embryos and its antagonism by TGFA. TNF receptor superfamily, member 1a mRNA was detectable throughout early embryonic stages, with an increase after the early blastocyst stage, whereas the expression of TNF receptor superfamily, member 1b transcripts were detected only at the expanded blastocyst stage. Although pregnant uteri produced TNF, physiologic levels were low during the preimplantation period. Treatment with TNF inhibited the development of two-cell stage embryos to blastocysts showing decreased proliferation and increased apoptosis both in vitro and in vivo. These detrimental effects of TNF on early embryo development and survival were blocked by a neutralizing anti-TNF antibody. In addition to the death receptor-mediated pathway, TNF-induced apoptosis was further mediated by disruption of mitochondrial functions, characterized by release of cytochrome c and activation of caspase 9. The proapoptotic effects of TNF in blastocysts were counteracted by cotreatment with TGFA. The antagonistic effect of TGFA on TNF-induced apoptosis was blocked by phosphatidylionsitol 3-kinase (PI3K) inhibitors. The present findings demonstrate the stage-selective susceptibility to the apoptosis-inducing effect of TNF in mouse preimplantation embryos and that the TGFA/PI3K signaling system has an important role in the control of TNF-induced apoptosis in blastocysts.
apoptosis, embryo, gene regulation, growth factors, signal transduction
Tumor necrosis factor (TNF) is a proinflammatory cytokine that elicits diverse biologic responses, including apoptosis in many cell types [1]. TNF exerts its biologic activity by binding two distinct cell surface receptors, TNF receptor superfamily, member 1a (TNFRSF1A; also known as TNFR1) and TNF receptor superfamily, member 1b (TNFRSF1B; also known as TNFR2), and activating several intracellular signaling pathways [2]. Both TNFR1 and TNFR2 are type I transmembrane receptors; TNFR1 has a common death domain (DD) in its cytoplasmic tail, whereas TNFR2 lacks DD. Activation of TNFR1 by TNF leads to the recruitment of DD for activation of a cascade of cysteine proteases called caspases and results in the induction of apoptosis [3]. Due to a lack of DD, TNFR2 mainly mediates cell survival signals through activation of the nuclear factor kappa B pathway. However, activation of TNFR2 also could potentiate death signals mediated by TNFR1 [2]. The presence of TNFR1 has been demonstrated in mammalian preimplantation embryos, including mouse and human, and the TNFR1 expression is detected in both trophectoderm (TE) and inner cell mass (ICM) cells at the blastocyst stage [4–7]. In female reproductive tracts, TNF is produced throughout the preimplantation period [8]. Therefore, TNF has the potential to regulate the apoptosis of preimplantation embryos through a paracrine pathway.
Treatment with TNF inhibits cell proliferation in the ICM lineage of rodent blastocysts in vitro [4, 5, 9], and this effect is reversed by addition to the culture medium of antisense oligonucleotides directed against Tnfr1 in rat blastocysts [5]. Furthermore, TNF has been shown to induce apoptosis in rodent and bovine embryos, and to have deleterious effects on early embryonic development [4, 5, 10]. TNF-pretreated blastocysts show a decreased ability to maintain a structured ICM cluster following TE spreading in vitro and a higher rate of fetal resorption after transfer to recipient mice [9]. In diabetic rats, upregulation of TNF production in the uterus is associated with an increase of apoptosis in blastocysts [11]. These findings indicate that TNF plays a key role in regulating embryo development and apoptosis during preimplantation and peri-implantation periods.
Previous studies have shown that treatment with several growth factors, such as epidermal growth factor (EGF), transforming growth factor alpha (TGFA), and insulinlike growth factors I and II promote development of preimplantation embryos in mammals [8, 12–17]. Both EGF and TGFA bind to the EGF receptor (EGFR) [18], and TGFA has been shown to inhibit apoptosis in mouse blastocysts [19–21]. Our previous findings demonstrated that the mechanism by which TGFA inhibits apoptosis in blastocysts involves the phosphatidylinositol 3-kinase (PI3K) pathway [21], which is a well-known mediator of cell growth, proliferation, and survival signals. Although treatment with EGF has been reported to antagonize the inhibition of preimplantation embryo development induced by TNF [22], the effect of TGFA on TNF-induced apoptosis in early embryos is unknown.
The objective of this study was to investigate the TNF regulation of apoptosis in mouse preimplantation embryos and its antagonism by TGFA. We sought to determine the temporal expression of TNF receptors and characterize the TNF-induced apoptotic signaling in early embryos. In mouse blastocysts we further examined the antagonistic action of TGFA in the proapoptotic effect of TNF and involvement of the PI3K pathway in the control of apoptosis regulated by TNF and TGFA.
Animals and Embryo Culture Media
To obtain preimplantation embryos, female B6D2F1 mice at 25 days of age (CLEA Japan Inc., Tokyo, Japan) were superovulated by a single i.p. injection of 7 IU human menopausal gonadotropin (HMG; ASKA Pharmaceutical. Co. Ltd., Tokyo, Japan) followed 48 h later with 10 IU human chorionic gonadotropin (hCG; ASKA Pharmaceutical. Co. Ltd.). Immediately after hCG injection, animals were allowed to mate overnight with fertile mature males of the same strain. The care and use of animals were approved by the Animal Research Committee of the Akita University School of Medicine (Akita City, Japan). M2 medium (MR-015-D; Chemicon Inc., Temecula, CA) or modified M16 medium (MR-010-D; Chemicon Inc.) without serum were used in all experiments.
Collection of Mouse Oocytes, Preimplantation Embryos and Uteri
Cumulus-oocyte complexes and zygotes were obtained following dissection of the oviducts of unmated and mated mice at 12–13 h and 22–23 h after hCG injection, respectively. Following treatment with hyaluronidase (500 U/ml; Chemicon Inc.) for 1–2 min, oocytes and zygotes were separated from cumulus cells using a small-bore pipette under Hoffman modulation contrast microscopy (Nikon Inc., Tokyo, Japan) and were washed three times with M2 medium. Two-cell stage embryos were obtained by flushing the oviducts of the mated mice at 46–47 h after hCG injection and were cultured as described previously [23]. For quantitative real-time RT-PCR, groups of 10–15 embryos were allowed to continue development to different stages in 50-µl drops of modified M16 medium covered with mineral oil (Irvine Scientific, Santa Ana, CA) and were then collected after culturing in individual microdrops at 50–52 h (four-cell), 59–60 h (eight-cell), 70–72 h (morula), 94–96 h (early blastocyst), 119–120 h (expanded blastocyst), and 142–144 h (hatched blastocyst) after hCG injection.
Uteri were obtained from immature untreated mice, mice treated with HMG (7 IU), mice treated with HMG followed 12 h later by treatment with hCG, and animals at 2 and 4 days after mating (corresponding to the migration of two-cell stage and expanded blastocyst stage embryos into the uterus, respectively).
Quantitative real-time RT-PCR of transcript levels in oocytes, preimplantation embryos, and uteri was performed using a SmartCycler (Takara, Tokyo, Japan) as described previously [23–25]. Primers and hybridization probes for Tnfr1, Tnfr2, and Tnf were based on GenBank association numbers NM_011609, NM_011610, and NM_013693, respectively. Tnfr1 primers: forward 5'-ACACGGTGTGTGGCTGTAAG-3', reverse 5'-AGTCCACGCACTGGAAGTGT-3'; Tnfr2 primers: forward 5'-TTGAGCTGCAGTTCTTCCTG-3', reverse 5'-CACACACTCGGTTCTGCTGT-3'; Tnf primers: forward 5'-ATGCACCACCATCAAGGACT-3', reverse 5'-GAGGCAACCTGACCACTCTC-3'; Tnfr1 probe: 5'-6-carboxy-fluorescein (FAM)-CCAGTTCCAACGCTACCTGAGTGAG-6-carboxy-tetramethyl-rhodamine (TAMRA)-3'; Tnfr2 probe: 5'-FAM-TACCACTGACCAGGTGGAGATCCG-TAMRA-3'; and Tnf probe: FAM-AGCCTCGAATGTCCATTCCTGAGTT-TAMRA-3'. The primers and probe for histone H2a were as described previously [23–25]. To determine the absolute copy number of target transcripts, cloned plasmid cDNAs for individual genes were used to generate a calibration curve. Purified plasmid cDNA templates were measured, and copy numbers were calculated from absorbance at 260 nm. A calibration curve was created by plotting the threshold cycle against the known copy number for each plasmid template diluted in log steps from 105 to 101 copies. Each run included standards of diluted plasmids to generate a calibration curve, a negative control without a template, and samples with unknown mRNA concentrations. Data were normalized based on histone H2a transcript levels [26].
For ELISA, mouse uteri were homogenized in a buffer containing 137 mM NaCl, 20 mM Tris-HCl, 1% Nonidet P40, 10% glycerol, and a protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN) before centrifugation at 8000 x g for 5 min at 4°C. Quantification of TNF in the supernatant was performed using mouse TNF ELISA MAX (Biolegend, San Diego, CA). The supernatant was diluted 1:4 in PBS for protein concentration measurement by the DC Protein Assay kit (Bio-Rad, Hercules, CA). The results were normalized by protein concentrations and expressed as nanograms of TNF per gram of uterus.
Two-cell stage embryos were obtained, and groups of 10–15 embryos were placed in 50-µl drops of modified M16 medium with or without increasing concentrations of TNF (R&D Systems, Minneapolis, MN) and covered with mineral oil. Embryos were cultured for 96 h up to the hatched blastocyst stage at 37°C in air with 5% CO2 with fresh media replacement every 24 h. Some embryos were cultured with 0.1 pM TGFA [21] with or without 50 ng/ml TNF. To examine the effects of TNF treatment, groups of 10–15 two-cell stage embryos were cultured with 5 µg/ml neutralizing chimeric monoclonal anti-TNF antibody, which has been developed and approved for the treatment of rheumatoid arthritis and Crohn disease in humans [27] (infliximab; Centocor Inc., Malvern, PA), with or without 50 ng/ml TNF. To investigate involvement of the PI3K pathway as a downstream mediator of the antagonistic effect of TGFA on TNF-induced apoptosis in blastocysts, early blastocysts were cultured for 24 h with TNF at 50 ng/ml with or without 0.1 pM TGFA in the absence or presence of a PI3K inhibitor, either LY294002 (5 µM) or wortmannin (50 nM; both from Sigma, St. Louis, MO). The concentrations of the PI3K inhibitors were based on previous studies of germ cells and early embryos [21, 28]. For controls, embryos were cultured in modified M16 medium alone. Embryonic development was monitored after 24, 48, 72, and 96 h of culture to determine the proportion of embryos at the morula, early blastocyst, expanded blastocyst, and hatched blastocyst stages, respectively. Expanded blastocyst stage embryos in each treatment group were subjected to caspase 3 assay to detect apoptosis. Some embryos were used for determination of cell numbers in expanded blastocysts.
To assess the effect of TNF when combined with a caspase inhibitor, groups of 10–15 two-cell stage embryos were cultured with or without TNF at 50 ng/ml with or without the following caspase inhibitors at 1 µM: z-FA-fmk (caspase inhibitor negative control), z-DEVD-fmk (caspase 3 inhibitor), z-IETD-fmk (caspase 8 inhibitor), or z-LEHD-fmk (caspase 9 inhibitor; all from MBL, Nagoya, Japan). After 72 h of culture, all embryos were assayed for caspase 3 activity.
To examine the roles of TNF on development and apoptosis in preimplantation embryos in vivo, female B6D2F1 mice at 25 days of age were treated with HMG (7 IU), followed 48 h later with 10 IU hCG. Immediately after hCG injection, animals were allowed to mate with fertile males. Because in vitro studies indicated minimal effects of TNF on early embryo development and apoptosis up to the morula stage, TNF was administered twice (i.p.; 1 µg x 2) at 72 and 84 h after hCG injection, corresponding to the development of morula to blastocyst stages in vivo, respectively. At 90–92 h after hCG injection, expanded blastocysts were obtained from the uterine horns. After evaluation of the proportion of embryos developed to the expanded blastocyst stage, embryos were subjected to the determination of cell number and the caspase 3 assay. To verify specific effects of TNF, the neutralizing anti-TNF antibody was administered before TNF injection (i.v.; 50 µg x 2). Due to the high molecular weight of the antibody, administration was performed through tail vein injection. For negative controls, vehicle (0.1% PBS-BSA) was administered.
Blastocyst Cell Number Determination
The number of cells in blastocysts was counted as described previously [25]. Blastocysts were fixed in 4% paraformaldehyde (PFA) at the end of culture and stained with 5 µg/ml of Hoechst 33342 dye (Invitrogen Corp., Carlsbad, CA) for 30 min at room temperature. After three washes in PBS, individual blastocysts were covered in a drop of SlowFade Light Antifade (Invitrogen Corp.) before the total number of nuclei in blastocysts was counted under an epifluorescent microscope.
Caspase 3 Assay for Detection of Apoptosis
To investigate apoptosis in cultured preimplantation embryos, activated caspase 3 was detected using a PhiPhilux G1D2 kit (OncoImmunin Inc., College Park, MD) as described previously [21]. In each experiment, embryos treated with 50 µM staurosporine for 24 h before being incubated with caspase 3 substrate and embryos incubated without caspase 3 substrate were included as positive and negative controls, respectively. Based on the mean number of caspase 3-positive cells in blastocysts recovered from the uteri of pregnant mice at Day 4 after mating, 1.68 ± 0.34 (mean ± SEM), blastocysts with more than three caspase 3-positive cells were defined as caspase 3-positive embryos.
Mitochondrial Function Assay for Detection of Apoptosis
Changes in the membrane potential of mitochondria induced by apoptosis were examined as described previously [23]. Embryos were incubated in 250 nM Mitotracker Orange CMTMRos (Invitrogen Corp.) for 30 min at 37°C in air with 5% CO2. After three washes in prewarmed modified M16 medium, embryos were fixed with 4% PFA for 15 min at 37°C in the dark. After three washes in PBS, embryos were transferred onto a slide and analyzed under an epifluorescent microscope.
The distribution of cytochrome c in embryos was determined by immunohistochemistry as described previously [23]. Embryos were fixed with 4% PFA for 15 min at room temperature and were then permeabilized in PBS with 0.2% Triton X-100 (Sigma) for 5 min. After blocking with Image iT FX Signal Enhancer (Invitrogen Corp.) for 30 min, embryos were incubated with 50 µg/ml mouse anti-cytochrome c monoclonal antibody conjugated with fluorescein isothiocyanate (FITC; eBioscience, San Diego, CA) overnight at 4°C. After three washes, embryos were transferred onto a slide and analyzed under an epifluorescent microscope. For negative controls, sections were subjected to the same method, except antibodies were replaced by the same concentration of mouse nonimmune IgG1 (DAKO Corp., Kyoto, Japan) conjugated with FITC using an EZ-label FITC protein labeling kit (Pierce Biotechnology Inc., Rockford, IL).
One-way ANOVA was used to evaluate differences in all experiments. For multiple comparisons, Fisher protected least significant difference test was adopted as a post-hoc test. Differences were considered to be statistically significant at P < 0.05. Results are presented as means ± SEM of at least three separate experiments.
Temporal Expression of Tnfr1, Tnfr2, and Tnf in Oocytes and Preimplantation Embryos
Quantitative real-time RT-PCR was performed to measure the levels of Tnfr1, Tnfr2, and Tnf mRNA in oocytes and the different developmental stages of early embryos (zygote, two-cell, four-cell, eight-cell, morula, early blastocyst, and expanded blastocyst). The temporal expression of transcripts was studied in embryos cultured in vitro. Based on the calibration curve, the sensitivity of the assay was equal in genes examined (105 to 101 copies/µl of template cDNAs). As shown in Figure 1A, levels of Tnfr1 mRNA were high in the two- and four-cell stage embryos (P < 0.05), decreased as the embryo developed to the eight-cell stage, increased after the morula stage (P < 0.05), and reached the highest levels at the expanded blastocyst stage. In contrast, Tnfr2 mRNA was not expressed in the oocytes and early embryos up to the blastocyst stage, and low levels of Tnfr2 mRNA were detected only at the expanded blastocyst stage (Fig. 1A). Although mouse oocytes and preimplantation embryos expressed Tnf, their levels were quite low as compared with those of Tnfr1 (Fig. 1A).
|
Expression of TNF in Preimplantation Uteri
We examined the expression of TNF in mouse uteri during the preimplantation period. Quantitative real-time RT-PCR analyses indicated that uterine levels of Tnf mRNA were high in immature mice (P < 0.05; Fig. 1B, HMG 0 h), decreased after HMG treatment, and were maintained at a low level until 12 h after hCG treatment, corresponding to the time of ovulation (Fig. 1B). In pregnant mice on Days 2 and 4, the levels of Tnf mRNA were comparable to those in uteri treated with gonadotropin (Fig. 1B). In addition to demonstrating decreases in Tnf transcripts after HMG treatment, similar changes of uterine TNF proteins were detected based on ELISA (Fig. 1C).
In Vitro Effects of TNF on Embryonic Development and Apoptosis
Because preimplantation embryos express different transcript levels of Tnfr1 during development, we tested the effect of TNF treatment on cultured embryos at different developmental stages. Two-cell stage embryos were cultured with or without TNF, and embryonic development in vitro was assessed. Although TNF treatment did not affect the development of two-cell stage embryos to the morula and early blastocyst stages (data not shown), TNF treatment inhibited the development of embryos to the expanded and hatched blastocyst stages evaluated after 72 and 96 h of culture, respectively (P < 0.05; Fig. 2, A and B). We also determined the effect of TNF on cell proliferation in cultured blastocysts. The number of cells in the blastocyst were counted and found to be decreased by TNF treatment (P < 0.05; Fig. 2C).
|
To determine whether TNF acts as a proapoptotic factor in preimplantation embryos, we evaluated apoptosis in embryos treated with TNF (Fig. 2D). Because hatched blastocysts are difficult to retrieve due to extreme stickiness, two-cell stage embryos were cultured for 72 h to the expanded blastocyst stage before assaying for apoptosis. TNF treatment increased the proportion of caspase 3-positive embryos in a dose-dependent manner (P < 0.05; Fig. 2D), thus reflecting apoptosis. In positive control embryos treated with staurosporine, most cells showed activated caspase 3 signals, whereas no caspase 3-positive cells were observed in the blastocysts for negative controls (data not shown).
The specificity of the effects of TNF on preimplantation embryos was examined by using the neutralizing anti-TNF antibody. The effects of TNF on the development of preimplantation embryos, the number of cells in expanded blastocysts, and the induction of apoptosis in the blastocyst were blocked by cotreatment with the neutralizing antibody (Fig. 2, A–D). Because treatment with the neutralizing antibody alone did not affect the embryo development or apoptosis without TNF treatment (Fig. 2, A–D), these data suggest that there was a minimal effect of endogenous TNF in the present experimental model system.
In Vivo Effects of TNF on Embryonic Development and Apoptosis
By using the neutralizing anti-TNF antibody and TNF, we further examined the roles of TNF during early embryonic development and apoptosis in vivo. Immature mice were treated with HMG followed by hCG before mating. Some mated animals were treated with TNF at 72 and 84 h after hCG injection (corresponding to the morula stage and blastocyst stage embryos). This was followed by the evaluation of embryonic development and apoptosis. Similar to in vitro assays, the effects of TNF were not evident in the embryos before the expanded blastocyst stage (data not shown). In vivo treatment with TNF inhibited the proportion of expanded blastocysts without affecting the total number of embryos found (vehicle, 49.5 ± 9.5; TNF, 47.8 ± 8.3; TNF+Ab, 41.8 ± 3.8; Ab, 46.5 ± 4.7; Fig. 3A). Assessment of cell numbers indicated that treatment with TNF also decreased the cell number in the embryos compared with the controls (P < 0.05; Fig. 3B). Furthermore, the recovered embryos from mice treated with TNF showed an increase in apoptosis as judged by caspase 3 activation (P < 0.05; Fig. 3C). The specificity of the in vivo effects of TNF on embryo development and apoptosis were examined using the neutralizing anti-TNF antibody. The administered antibody blocked the effects of TNF, and neutralizing antibody treatment alone showed minimal effects on all three parameters tested (Fig. 3, A–C).
|
TGFA Counteracts the Proapoptotic Effects of TNF on Blastocysts Through the PI3K Pathway
To examine whether TGFA antagonizes the effects of TNF on embryonic apoptosis through paracrine signaling, two-cell stage embryos were cultured with TGFA with or without TNF before evaluation of embryonic development and apoptosis in vitro. TGFA treatment counteracted the TNF-induced apoptosis in blastocysts as judged by caspase 3 activation (P < 0.05; Fig. 4A). Similar to previous studies [21], TGFA treatment alone decreased the proportion of caspase 3-positive embryos relative to the controls (P < 0.05; Fig. 4A).
|
Because induction of apoptosis by TNF treatment was not observed before the early blastocyst stage, early blastocyst stage embryos were cultured for 24 h to examine involvement of the PI3K pathway in the protective action of TGFA on the proapoptotic effects of TNF. When TNF treatment was initiated in early blastocyst stage embryos, the proportion of apoptosis in embryos was similar to that in embryos in which the treatment was started at the two-cell stage (Fig. 4B). The ability of TGFA to counteract the proapoptotic effects of TNF was suppressed by cotreatment with either of the two PI3K inhibitors (LY294002 and wortmannin) with similar efficiency (Fig. 4B). In the absence of TGFA treatment, treatment with the PI3K inhibitors did not affect TNF-induced apoptosis in blastocysts (Fig. 4B).
Effects of Caspase Inhibitors on Apoptosis in Blastocysts Treated with TNF
Induction of apoptosis results in the activation of a caspase cascade from either the extrinsic pathway, which is initiated by activation of membrane-bound death receptors, leading to cleavage of caspase 8, or the intrinsic pathway, which is characterized by mitochondrial dysfunction, release of cytochrome c, and subsequent activation of caspases 9 and 3 [29]. To determine the molecular pathways underlying TNF-induced apoptosis in blastocysts, two-cell stage embryos were cultured with TNF in combination with diverse caspase inhibitors. Treatment with TNF in combination with a specific caspase 3, 8, or 9 inhibitor showed a decrease in the proportion of caspase 3-positive embryos (P < 0.05; Fig. 5), suggesting involvement of both extrinsic and intrinsic pathways in TNF-induced apoptosis in blastocysts. In contrast, embryos treated with TNF and the caspase inhibitor negative control had no effect on embryonic apoptosis (Fig. 5).
|
TNF Induces Loss of Mitochondria Function and Release of Cytochrome c in Blastocysts
To further characterize the TNF-induced apoptosis in blastocysts, we examined the loss of mitochondrial function using a mitochondrial membrane potential-sensitive dye, Mitotracker Orange CMTMRos. The dye allowed us to assess the status of mitochondrial membrane potential during apoptosis [30]. Functional mitochondria take in the dye and show a punctate staining pattern with orange fluorescence, whereas mitochondria in apoptotic cells are not stained. We found that the Mitotracker dye staining in the expanded blastocysts appeared punctate, as expected for their mitochondrial localization (Fig. 6Aa). Treatment with TNF reduced Mitotracker dye staining in expanded blastocysts (Fig. 6Ab).
|
Using the anti-cytochrome c monoclonal antibody characterized previously [31–33], the distribution of cytochrome c was visualized by immunofluorescence microscopy. Without TNF treatment, cytochrome c was localized exclusively to the mitochondria (Fig. 6Ba). After TNF-induced apoptosis, however, cytochrome c was discharged from the mitochondria and showed a diffused distribution pattern throughout the cytoplasm in the blastocysts (Fig. 6Bb). Specificity of the cytochrome c immunostaining was demonstrated using blastocysts incubated with nonimmune IgG1 conjugated with FITC (Fig. 6Bc).
The present study demonstrates the TNF regulation of apoptosis in mouse preimplantation embryos and the antagonistic effects of TGFA on TNF-induced apoptosis in blastocysts. We showed temporal expression of TNF receptors in preimplantation embryos and of TNF in pregnant uteri in mice. The high expression of Tnfr1 in expanded blastocysts is associated with the ability of TNF to inhibit early embryonic development and to induce blastocyst apoptosis. Furthermore, we showed the in vivo effects of TNF on early embryonic apoptosis. In addition to death receptor-mediated extrinsic pathway, we demonstrated an implication of mitochondrial death pathway in TNF-induced apoptosis in blastocysts. Using PI3K inhibitors, we further demonstrated involvement of the PI3K pathway in the counteraction of TGFA in the proapoptotic effects of TNF.
Treatment with TNF inhibited development of the preimplantation embryos beyond the expanded blastocyst stage, and the observed effects were blocked by the neutralizing anti-TNF antibody. The effect of TNF in inhibiting blastocyst development is associated with the ability of TNF to induce apoptosis, showing a decrease in total cell numbers in cultured blastocysts. The stage-specific effects of TNF on embryo development and apoptosis are consistent with our postulation that sensitivity to the TNF stimulus increases after the early blastocyst stage. The lack of detrimental effects of TNF in early stages of embryos was not caused by short exposure to TNF, because inhibition of development and induction of apoptosis also were observed in embryos for which the treatment of TNF was delayed for 48 h (data not shown). However, one cannot rule out the possibility that the downstream apoptotic pathways are simply not present, intact, or operational before blastocyst stage.
Increased TNF synthesis in the uteri of pregnant diabetic rats is associated with an increase of apoptosis in the blastocysts [11]. In addition, enhanced TNF production has been demonstrated during stress-triggered [34] and spontaneous [35] abortions in mice. These results raise the question of whether a decrease in elevated TNF levels improves embryo quality and results in successful implantation and fetal growth. Our in vivo studies demonstrated that the administration of TNF inhibited early embryo development and induced apoptosis in blastocysts, and that these deleterious effects of TNF could be blocked by the neutralizing anti-TNF antibody. Although it is unknown whether the administered TNF acts directly and/or indirectly on the preimplantation embryos, our results suggest that an in vivo application of the neutralizing anti-TNF antibody improves the quality of the impaired embryos induced by TNF.
Because some growth factors secreted from female reproductive tracts regulate preimplantation embryo development [8, 12], such survival factors may protect embryos from TNF-induced apoptosis in vivo. Of the survival factors, TGFA is produced by mouse uteri [36, 37] and inhibits apoptosis in the blastocysts through both paracrine and autocrine signaling [19–21]. We found that TGFA counteracted the proapoptotic effects of TNF in blastocysts via the PI3K pathway. Previous studies demonstrated that PI3K is present in all stages of preimplantation embryo development [21, 38], and it is involved in the TGFA-mediated inhibition of apoptosis in blastocysts [21]. Using specific caspase inhibitors, we demonstrated that TNF treatment induced apoptosis in blastocysts via extrinsic and intrinsic pathways, and involvement of the intrinsic pathway was further confirmed by a loss of mitochondrial function and the release of cytochrome c into the cytoplasm. CASP8 and FADD-like apoptosis regulator (CFLAR, also known as cFLIP) acts as a competitive inhibitor of caspase 8 in the formation of the death-inducing signal complex, which blocks the extrinsic pathway [39]. In human tumor cells, cFLIP expression is increased by lysophosphatidic acid treatment via activation of PI3K signaling [40]. Furthermore, activation of PI3K signaling by TGFA leads to inhibition of apoptosis in blastocysts through upregulation of baculoviral IAP repeat-containing 5 (BIRC5, also known as survivin) [21], which is known to exhibit its antiapoptotic activities through the intrinsic pathway [41, 42]. Thus, cFLIP and survivin likely contribute to the antagonistic activities of TGFA in TNF-induced apoptosis in blastocysts through the extrinsic and intrinsic pathways, respectively. In addition, PI3K inhibits apoptosis by regulating glucose uptake and metabolism in blastocysts [43] and the activity of a proapoptotic Bcl2 family member, BAD, in diverse tumor cells [44]. These findings suggest that TGFA counteracts the proapoptotic effects of TNF in blastocysts via the PI3K pathway at several levels, in both transcription-dependent and independent manners.
Although it is obvious that high levels of TNF induce apoptosis in blastocysts in vitro and in vivo, the physiologic role of TNF in the development of the preimplantation embryo remains to be determined. The expression of TNFR2 was hardly detectable in preimplantation embryos, and the absence of TNFR2 transcripts also is reported in rat blastocysts [5]. Thus, TNFR2 signaling does not likely regulate embryo development. In addition to the low expression of TNF in preimplantation embryos, a blockade of endogenous TNF production from the embryos did not affect their development and apoptosis, suggesting a less potent autocrine action for TNF in the regulation of embryo development. Although TNF was produced by mouse uteri throughout the period examined, our study and previous studies [45] show a low expression of Tnf mRNA in pregnant uteri. It is difficult to measure the concentration of TNF in mouse reproductive tracts due to low yields of oviductal and uterine fluid. However, the concentration of TNF in human uterine flushing fluid is reported to be 8–22 pg/ml [46]. Thus, the physiologic levels of TNF in reproductive tracts may be far lower than those that exhibit deleterious effects on early embryonic development in vitro. The phenotypes of TNFR1 null mice exhibited in normal pregnancy and birth further support the lack of important roles for endogenous TNF
in early embryonic development on physiologic levels [47]. However, the presence of TNFR1 in preimplantation embryos suggests that TNF may play a physiologic role under certain circumstances in embryo demise if it is advantageous or necessary to eliminate developing embryos.
In conclusion, we have shown stage-selective susceptibility to the detrimental effects of TNF in mouse preimplantation embryos both in vitro and in vivo, an implication of mitochondrial death pathway in TNF-induced apoptosis in blastocysts, and the ability of TGFA to antagonize the proapoptotic effects of TNF. The present demonstration of an antagonizing role of the TGFA/PI3K signaling pathway in TNF-induced apoptosis underscores the importance of diverse paracrine systems for early embryonic development. A complete understanding of these intercellular communication networks could provide new approaches for the treatment of infertility.
ACKNOWLEDGMENTS
We thank Dr. Aaron J. W. Hsueh (Stanford University School of Medicine, Stanford, CA) for critical review of the manuscript and C. Spencer for editorial assistance.
FOOTNOTES
1Supported by a Grant-in-Aid for Scientific Research (Grant-in-Aid for Young Scientists B: 177911010 [K.K.] and Grant-in-Aid for Scientific Research B: 18390444 [T.T.]) and the research funds from the Yamaguchi Endocrine Research Association (K.K.), the Kanae Foundation for Life and Socio-medical Science (K.K.), and the Kanzawa Medical Research Foundation (K.K.). 
Correspondence: 2Kazuhiro Kawamura, Department of Obstetrics and Gynecology, Akita University School of Medicine, Hondo 1-1-1, Akita 010-8543, Japan. FAX: 81 18 884 6447; e-mail: kawamura{at}yf7.so-net.ne.jp
Received: 11 October 2006.
First decision: 10 November 2006.
Accepted: 19 December 2006.
REFERENCES
precursors in the mouse uterus during the preimplantation period and after steroid hormone treatments. Biol Reprod 1994; 50:481–491[Abstract]This article has been cited by other articles:
|
|
 |
|
  Y. Wang, P. R. Crisostomo, M. Wang, T. A. Markel, N. M. Novotny, and D. R. Meldrum TGF-{alpha} increases human mesenchymal stem cell-secreted VEGF by MEK- and PI3-K- but not JNK- or ERK-dependent mechanisms Am J Physiol Regulatory Integrative Comp Physiol, October 1, 2008; 295(4): R1115 - R1123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. O'Neill Phosphatidylinositol 3-kinase signaling in mammalian preimplantation embryo development Reproduction, August 1, 2008; 136(2): 147 - 156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Cackovic, C. S. Buhimschi, G. Zhao, E. F. Funai, E. R. Norwitz, E. Kuczynski, C. J. Lockwood, and I. A. Buhimschi Fractional Excretion of Tumor Necrosis Factor-{alpha} in Women With Severe Preeclampsia Obstet. Gynecol., July 1, 2008; 112(1): 93 - 100. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. O'Neill The potential roles for embryotrophic ligands in preimplantation embryo development Hum. Reprod. Update, May 1, 2008; 14(3): 275 - 288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Li, V. Chandrakanthan, M. L Day, and C. O'Neill Direct Evidence for the Action of Phosphatidylinositol (3,4,5)-Trisphosphate-Mediated Signal Transduction in the 2-Cell Mouse Embryo Biol Reprod, November 1, 2007; 77(5): 813 - 821. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
