Enhanced proapoptotic gene expression of XAF1, CASP8 and TNFSF10 in the bovine endometrium during early pregnancy is not correlated with augmented apoptosis

Article Outline

• Abstract
• 1. Introduction
• 2. Material and methods
  • 2.1. Pretreatment of animals
  • 2.2. Tissue sampling
  • 2.3. Extraction of total RNA
  • 2.4. Two step quantitative real-time RT-PCR
  • 2.5. Data analysis and statistics
  • 2.6. IFNT bioactivity
  • 2.7. Caspase activity
  • 2.8. TdT-mediated dUTP-biotin nick end labelling (TUNEL)
  • 2.9. Immunohistochemistry
  • 2.10. In vitro co-culture of glandular and stroma cells
• 3. Results
  • 3.1. Expression of intrinsic and extrinsic pathway factors
  • 3.2. Expression and activity of initiator and effector caspases as well as expression of inhibitor of apoptosis (IAPs) and their antagonists
  • 3.3. Bioactivity of IFNT in uterine flushing fluids and expression of IFN type I receptor
  • 3.4. Detection of apoptotic cells by using the TUNEL technique and immunohistochemical localisation of FASLG in bovine endometria and day 18 trophoblasts
  • 3.5. Expression of XAF1, CASP8 and FASLG in an in vitro co-culture system of glandular and stroma cells
• 4. Discussion
• Acknowledgements
• References
• Copyright

Abstract 

Bovine trophoblast cells release interferon-τ (IFNT), a type I IFN, as the pregnancy recognition signal. Since type I IFNs exert growth inhibitory and proapoptotic actions, the effect of the conceptus on components of the apoptosis pathways was determined in the bovine endometrium during the periimplantation period. Uteri of Simmental heifers were flushed post mortem at days 12, 15, and 18 of cycle or pregnancy for the recovery of conceptuses and the sampling of ipsilateral endometrial tissue at slaughter for quantitative RT-PCR, immunohistochemistry, caspase activity and TUNEL assays. Endometrium samples of pregnant animals revealed increased transcript levels for the proapoptotic genes XAF1 (day 15: 2.9-fold; day 18: 15.1-fold; p=0.005) and CASP8 (day 18: 2.4-fold; p=0.007). The mRNA expression increased significantly with the day of the cycle for the proapoptotic genes FASLG, TNFSF10, TNF and TNFSF1A (p=0.004, p=0.006, p=0.001 and p=0.007) and the antiapoptotic gene BIRC4 (p=0.03). We detected high amounts of FASLG transcripts in day 18 conceptuses (16-fold higher than day 18 endometria). This finding was validated at the protein level by immunohistochemistry. To further analyse the endometrial activation of the caspase cascade, the activities of initiator caspase 8 and effector caspases 3/7 were determined luminometrically. No difference between pregnant and cyclic animals was found for either caspase activity. Additionally, a TUNEL assay showed no increase of apoptotic cells in the pregnant endometrium. In conclusion, although the bovine conceptus induces the expression of proapoptotic genes, neither an activation of a caspase cascade nor an increase of apoptotic cells was noticed. These results suggest inhibitory mechanisms preventing endometrial cells from programmed cell death.

Keywords: Bos taurus, Apoptosis, Endometrium, Periimplantation

Back to Article Outline

1. Introduction 

Interferons (IFNs) are a family of cytokines that exert immune regulatory and inflammatory actions [1]. Initially described as trophoblast protein-1, interferon-τ (IFNT), a type I IFN, serves as the primary maternal pregnancy recognition signal by preventing the release of luteolytic PGF_2α pulses by the endometrium [2]. The release of IFNT appears to occur from day 12 to at least day 25 of pregnancy in cattle [3] with a secretion of up to 10⁵ units per hour in culture at day 18 [4], although mRNA can be observed as early as the 16-cell and blastocyst stages. Upon binding of IFNT to the common type I IFN receptor subunits IFNAR1 and IFNAR2 [5], the JAK-STAT signalling pathway is activated and in further downstream reactions the transcription of a variety of IFN stimulated genes (ISGs) is upregulated. Common to all type I IFNs including IFNT are their antiviral, growth inhibitory and proapoptotic activities [6], [7]. Induction of apoptosis by interferon-α, the most extensively studied type I IFN, is initiated by diverse mechanisms and involves both up-regulation of proapoptotic factors [8] as well as suppression of antiapoptotic actions [9].

Apoptosis is a tightly controlled process by which tissue eliminates unwanted cells. Either activation of the intrinsic or the extrinsic pathway finally results in formation of characteristic vesicles also known as apoptotic bodies [10], which are phagocytosed by macrophages without inducing an inflammatory response [11]. The intrinsic or mitochondrial pathway may be induced via multiple stimuli like cellular stress or absence of survival factors. The ratio of pro- and antiapoptotic Bcl-2-family members alters the permeability of mitochondria which results in release of cytochrome c and formation of the apoptosome complex [12]. Extracellular stimuli and the cytokines FAS ligand (FASLG), TNF related apoptosis inducing ligand/TRAIL (TNFSF10) and TNF-α (TNF) initiate the extrinsic, receptor mediated pathway. Through ligation of cell surface death receptors such as the members of the TNF-ligand superfamily FAS, TNF receptor 1 (TNFRSF1A), TNF receptor 2 (TNFRSF1B) and TRAIL receptor 2 (TNFRSF10B) trimerization can occur [13] which causes recruitment of adaptor proteins to the cytoplasmic death domain followed by activation and subsequent autocatalytic cleavage of initiator procaspase 8 (CASP8).

Both apoptosis pathways activate further downstream effector caspases that cleave numerous substrates such as ICAD (DFFA) [14]. The activation can be prevented by inhibitors of apoptosis (IAPs). Otherwise, IAPs e.g. XIAP (BIRC4) or SURVIVIN (BIRC5) can be inactivated again by proapoptotic antagonists such as DIABLO (SMAC) or XIAP associated factor 1 (XAF1). Cell death may be impeded at every level even after DNA fragmentation; there are factors e.g. p53 (TP53) that stop the cell cycle in order to allow mending by DNA-polymerases and -ligases.

In our recent transcriptome studies on endometrial gene expression during the periimplantation period, the increased transcription of specific ISGs related to apoptosis has been shown [15], [16]. The programmed cell death might act as an additional mechanism to generate immune privilege for the protection of the semi-allogenic conceptus from the maternal immune system [17]. Death receptor ligands provided by the conceptus mediate receptor induced killing of death receptor bearing leukocytes [18]. In primates and rodents showing a hemochorial placentation, the regulation of apoptosis has been studied extensively [19]. Luminal epithelial and stromal cells expressing apoptosis markers and exhibiting DNA damage along with an apoptotic appearance were predominantly located at the implantation sites to allow invasion of the trophoblast [20].

Ruminants exhibit a late implantation with a synepitheliochorial type of placentation. An epitheliochorial placenta is also present in pig, and cell death was absent throughout the window of implantation [21]. As the bovine endometrium is exposed to high amounts of IFNT, mechanisms of apoptosis may as well be important for generating immune privilege allowing early embryonic development. Therefore, we analysed whether the induction of proapoptotic factors in the bovine endometrium leads to an activation of initiator and effector caspases eventually triggering cell death events during the periimplantation period. Inhibitor mechanisms preventing apoptosis derived from either endometrium or conceptuses were as well under investigation.

Back to Article Outline

2. Material and methods 

2.1. Pretreatment of animals 

All experiments were performed in accordance with the International Guiding Principles for Biomedical Research Involving Animals, as promulgated by the Society for the Study of Reproduction and with the European Convention on Animal Experimentation. Cyclic Simmental heifers (Bos taurus, Deutsches Fleckvieh) were estrus-synchronized by injecting intramuscularly 500μg of a single dose of the PGF_2α-analog Cloprostenol (Estrumate; Essex Tierarznei, Munich, Germany) at diestrus as described previously [22]. The pregnant groups were inseminated after estrus detection, whereas the cyclic control groups received supernatant of centrifuged sperm from the same bull. At day 12, 15, or 18 after insemination, animals were slaughtered (n=4–7 per group); the uterus was removed and flushed with 100ml phosphate buffered saline (PBS; pH 7.4) for the recovery of embryos as described previously [22]. The flushing fluid was centrifuged at 800g for 10min and the supernatant was stored at −20°C until further usage.

2.2. Tissue sampling 

Intercaruncular endometrial samples were taken from the uterine horn ipsilateral to the corpus luteum bearing ovary as previously described [15]. Whole conceptuses were transferred into vials containing 4ml RNAlater (Ambion, Huntingdon, Cambridgeshire, UK) and further processed as the endometrial samples. For TUNEL assays and immunohistochemistry, tissue samples (uteri and whole conceptuses) were transferred to Bouin's fixation solution prior to embedding in paraffin.

2.3. Extraction of total RNA 

Total RNA from endometrial samples and whole conceptuses were isolated using TRIzol reagent (Invitrogen Corporation, Carlsbad, CA, USA) according to manufacturer's instructions. Quality of RNA was controlled by the Agilent 2100 Bioanalyzer (RNA 6000 Nano Assay Kit). RNA integrity numbers ranged between 7 and 10. Quantity was spectroscopically determined at 260nm by the Nanodrop 1000.

2.4. Two step quantitative real-time RT-PCR 

The quantitative real-time PCR experiments were performed in accordance with the MIQE guidelines [23]. In the present study, quantitative real-time PCRs using the LightCycler DNA Master SYBR Green I protocol (Roche Diagnostics, Mannheim, Germany) were undertaken as described earlier [22]. Sequences of commercially synthesized PCR primer pairs (Eurofins MWG Operon, Ebersberg, Germany), references, mean Cq, annealing temperature (AT), fluorescence acquisition points (FA), melting points (MP) as well as product length [bp] are depicted in Table 1.

Table 1. Forward and reverse primer sequences. Sequences of primer pairs and corresponding references as well as annealing temperatures (AT), fluorescence acquisition (FA), melting points (MP) and the resulting fragment length are shown.

2.5. Data analysis and statistics 

The cycle number (Cq) to attain a definite fluorescence signal was calculated by the second derivative maximum method (LC software 4.05), as the Cq is inversely correlated with the logarithm of the initial template concentration. The Cq values from the target genes were normalized against the geometric mean of three reference genes (UBQ3, H3F3A and 18S rRNA) according to the bestkeeper method [24]. In order to avoid negative digits while allowing an estimation of a relative comparison between two genes, data are presented as means±SEM subtracted from the arbitrary value 30 (ΔCq). Thus, a high ΔCq resembles high transcript abundance [25]. An increase of one ΔCq represents a two-fold increase of mRNA transcripts. For statistical analysis the SAS program package release 9.1.3 (2002; SAS Institute, Inc., Cary, NC, USA) was used. The data comparing endometria from cyclic and pregnant uteri were subjected to least-square analysis of variance using the General Linear Models procedure to determine effects of the day, the status (cyclic vs. pregnant), and their interaction. Significant different days within each status as well as significant different groups at each time point were localized by the differences of least-square means. Graphs were plotted using Sigma-Plot 8.0 (SPSS software).

2.6. IFNT bioactivity 

IFN production was quantified in flushing fluids by a bioassay based on the inhibition of the cytopathic effect of vesicular stomatitis virus (Indiana strain) on Madin–Darby bovine kidney (MDBK) cells [26]. The NIH recombinant human IFN-α₂ reference preparation (No. Gxa01-901-535, NIH-Research Reference Reagent Note No. 31, 1984) was included in each assay. The antiviral activity was shown to be mediated by IFNT, as the effects of supernatant and an appropriate control IFNT preparation were blocked by specific anti-IFN sera (kindly provided by Dr. R.M. Roberts, University of Missouri, Columbia, MO) [27].

2.7. Caspase activity 

For Caspase-Glo^® 3/7 Assay and Caspase-Glo^® 8 Assay (Promega GmbH, Mannheim, Germany) endometrial samples were homogenized and centrifuged through a NucleoSpin Filter L (Macherey-Nagel GmbH & Co KG, Düren, Germany). Protein concentrations were determined using a BCA standard protocol (Sigma–Aldrich Chemie GmbH, Steinheim, Germany). The activity assay was performed according to manufacturer's instructions. Luminescence measurement, lasting 0.1s, was carried out in the Victor Light Luminescence Counter. Results are shown as mean±SEM of RLU (relative light units) per μg total protein (TP).

2.8. TdT-mediated dUTP-biotin nick end labelling (TUNEL) 

For analysis of single strand DNA nicks in uterine tissue sections, the DeadEnd™ Colorimetric TUNEL System (Promega) was performed on sections of three animals per group according to manufacturer's protocol. The negative control was incubated without rTdT enzyme and DNA nicks were induced in the positive control by 10 units per ml of RQ1 RNase-free DNase I (Promega). After blocking of endogenous peroxidases in the presence of 0.3% H₂O₂, streptavidin-horseradish peroxidase (HRP) was introduced. HRP converts H₂O₂ into H₂O and protons which results in oxidation of 3,3′-diaminobenzidine (DAB) to a brown insoluble product. Apoptotic cells in uterine tissue sections were determined by light microscopy using Olympus Cell-F software.

2.9. Immunohistochemistry 

FASLG protein was localized in trophoblasts and endometrial tissue. Endogenous peroxidase activity was blocked with 1% H₂O₂ in methanol. Following blocking with 10% goat serum, samples were incubated overnight at 4°C with the rabbit polyclonal N-20 sc-834 antibody (4.0μg/mL, Santa-Cruz, Biotechnology Inc. Santa-Cruz, CA, USA). Samples were then incubated with secondary antibody anti-rabbit IgG peroxidase-conjugated (2.5μg/mL, Sigma–Aldrich). Binding of antibody was detected by incubating with DAB in the presence of 0.1% H₂O₂ for 5min. Tissues were counterstained using Mayer's Haemalaun (Carl Roth GmbH, Karlsruhe, Germany).

2.10. In vitro co-culture of glandular and stroma cells 

Co-culture of glandular and stroma cells was performed as described earlier [28]. Cells were stimulated with recombinant bovine IFNT (antiviral activity, 4.8×10³U/ml medium; PBL Biomedical Laboratories, Piscataway, NJ) for four hours. RNA extraction and qPCR experiments were carried out as detailed for the endometrial samples.

Back to Article Outline

3. Results 

3.1. Expression of intrinsic and extrinsic pathway factors 

The death receptor ligands investigated increased significantly over time most prominently during pregnancy as the mRNA expression of FASLG (day p=0.004), TNFSF10 (day p=0.006) and TNF (day p=0.001) was 3.6-fold (Fig. 1A), 11.9-fold (Fig. 1B), and 4.8-fold (Fig. 1C) higher in endometria of day 18 than of day 12 pregnant animals, respectively. In addition, we detected a significant day^∗status interaction for TNFSF10 (p=0.048). The mRNA levels of analysed death receptors FAS (Fig. 1D), TNFRSF10B (Fig. 1E), and TNFRSF1B (not shown) exhibited no variation. Solely for TNFRSF1A (Fig. 1F) we detected an increase in expression over time (day p=0.007) in endometria. The expression of FASLG strikingly increased 82-fold in conceptuses between days 15 and 18 (p=0.0001). The mRNA of conceptus-derived TNFSF10 was more than 2,000-fold lower than the endometrial expression, and the transcript abundance of TNF in conceptuses was below the detection limit (Cq>40). Transcript abundance revealed no variation for either endometrial BAX or BCL2L1 (not shown) subjected to the day and to the status. While the expression of BAX was augmented in conceptuses of day 18 vs. 15 (20.4±0.2 ΔCq vs 21.2±0.3 ΔCq, p=0.049), the expression of BCL2L1 dropped more than 2.1-fold (19.9±0.2 ΔCq vs 18.8±0.2 ΔCq, p=0.02).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1 

  Transcript abundances of extrinsic apoptosis factors. Messenger RNA expression of extrinsic TNF family members FASLG (C), TNFSF10 (D) and TNF (E) and their death receptors FAS (F), TNFRSF10B (G) and TNFRSF1B (H) in endometria and conceptuses are shown as means ΔCq±SEM. Normalized Cq (ΔCq) values were subtracted from the arbitrary value 30. The effect of the day within the status is shown by different superscript letters (x–z in cyclic animals and a–c in pregnant animals). Asterisks indicate significant differences (p<0.05). nd indicates non-determinable values.

3.2. Expression and activity of initiator and effector caspases as well as expression of inhibitor of apoptosis (IAPs) and their antagonists 

Transcripts of initiator caspase 8 (CASP8) constantly rose from day 12 to day 18 in control and pregnant animals (1.6-fold and 2.9-fold, for control or pregnant animals (day p=0.007), respectively) (Fig. 2A). Day 18 pregnant animals exhibited (2.4-fold) increased transcript abundance compared to the respective control animals at day 18. CASP8 transcripts in pregnant animals were status dependent enhanced (status p=0.007). We could not detect changes in transcript abundance of pregnant animals of the CASP8 binding site competitor FLIP (CFLAR) (data not shown). The measurement of luminescence in endometrial homogenates revealed a 30% decrease of CASP8 activity from day 12 to day 18 (day p=0.0001) (Fig. 2C) in both pregnant and non-pregnant animals. The mRNA levels of effector caspases 3 (CASP3) (Fig. 2B) and 7 (CASP7) (not shown) were not influenced. CASP7 was only marginally expressed (mean Cq 32.3) and caspase 6 (CASP6) expression was below the detection limit of the assay (Cq>40) (data not shown). CASP8 was expressed to a lesser extent in conceptuses than in endometria (Fig. 2C) whereas effector CASP3 (Fig. 2D) exhibited a more pronounced expression in day 18 conceptuses.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 2 

  Messenger RNA expression of caspases as well as inhibitor of apoptosis (IAPs) and their antagonists. Messenger RNA expression of initiator CASP8 (A) and effector CASP3 (B) as well as inhibitor of apoptosis (IAPs) BIRC4 (E) and BIRC5 (F) in addition to their antagonists XAF1 (G) and DIABLO (H) in endometria and corresponding conceptuses are shown. Messenger RNA expression is shown as means ΔCq±SEM. Caspase activity of CASP8 (D) and CASP3/7 (E) in endometrial homogenates is presented as means±SEM relative light units per μg total protein (RLU/μg TP). The effect of the day within the status is shown by different superscript letters (x–z in cyclic animals and a–c in pregnant animals). Asterisks indicate significant differences (p<0.05). nd indicates non-determinable values.

BIRC4 transcripts (Fig. 2E) rose 2.3-fold and 2.8-fold in cyclic and pregnant animals over the analysed time points (day p=0.02), while BIRC5 (Fig. 2F) did not show marked variation. Strikingly, the transcript abundance of XAF1 (Fig. 2G) was 2.9-fold increased in day 15 pregnant animals when compared to their cyclic counterparts, and 15.1-fold increased at day 18 (day p=0.0008, status p=0.0005, day^∗status p=0.003). No differences were observed for mRNA expression of DIABLO (Fig. 2H). In contrast to pregnant endometria XAF1 was expressed at very low abundance in conceptuses, whereas BIRC5 mRNA was 76-fold more abundant in conceptuses than in endometria of day 18 pregnant animals.

3.3. Bioactivity of IFNT in uterine flushing fluids and expression of IFN type I receptor 

Bioactivity of IFNT increased significantly (p<0.001) with proceeding pregnancy in uterine flushing fluids (Fig. 3A), whereas flushing fluids of non-pregnant control animals showed no antiviral activity (not shown). Gene expression analysis of IFNAR receptor subunits revealed no variation for IFNAR1 and IFNAR2 over the analysed time points (Fig. 3B, C). IFNAR1 and IFNAR2 showed a 5.8-fold and more than 1000-fold lower expression in day 18 conceptuses when compared to endometria of day 18 pregnancy.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 3 

  Bioactivity of IFNT in uterine flushing fluids (A) and messenger RNA expression of IFN type I receptor subunits IFNAR1 (B) and IFNAR2 (C). Results are shown as means±SEM. Asterisks indicate significant differences (p<0.05).

3.4. Detection of apoptotic cells by using the TUNEL technique and immunohistochemical localisation of FASLG in bovine endometria and day 18 trophoblasts 

DNA fragmentation indicated by TUNEL was barely detected in bovine endometria of cyclic and early pregnant cows. Only a few positive cells were located in the subepithelial stroma and the luminal epithelium close to the uterine lumen (Fig. 4B). The number was slightly enhanced in day 15 control animals. Additionally, most TUNEL-positive cells also showed apoptotic morphology and apoptotic bodies.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 4 

  Localisation of apoptotic cells by TUNEL in bovine endometrium and immunohistochemical localisation of FASLG in bovine tissue sections. TUNEL experiments: The negative control was incubated without rTdT enzyme (A) and the positive control with 10 Units DNase/ml (insert). Light microphotograph of an endometrial section showing TUNEL-positive cells (B). Arrows point to apoptotic cells. FASLG immunohistochemistry: Pictures show representative sections of the superficial endometrium (C), the deep endometrium (D), endothelial cells (E) and day 18 trophoblasts (F). Inserts represent negative controls.

Deep endometrial glands showed intense FASLG immune labelling (Fig. 4D), whereas the staining in the luminal epithelial cells was only weak (Fig. 4C). Endothelial cells also showed intense FASLG labelling (Fig. 4E). In uterine sections there were no obvious differences in immunostaining pattern of FASLG protein among cyclic or pregnant animals irrespective of the day. In day 18 trophoblast cells, we found a strong immune staining (Fig. 4F).

3.5. Expression of XAF1, CASP8 and FASLG in an in vitro co-culture system of glandular and stroma cells 

Expression of XAF1 (7.9-fold) and CASP8 (6.2-fold) was significantly elevated following IFNT treatment (p<0.0001) in glandular epithelial cells (Table 2). The increase in transcript abundance was even higher in stroma cells (32.2-fold and 13.4-fold for XAF1 and CASP8, p<0.0001) (Table 2). We did not observe an IFNT dependent increase for FASLG in either cell type analysed (Table 2).

Table 2. Messenger RNA expression of XAF1, CASP8 and FASLG in IFNT treated and untreated glandular and stroma cells in an in vitro co-culture system. Messenger RNA expression is shown as means ΔCq±SEM. The relative increase of transcript abundance following IFNT treatment is also shown.

	glandular epithelial cells mRNA expression				stroma cells mRNA expression
	ΔCq [log2] mean±SEM		fold increase	p-value	ΔCq [log2] mean±SEM		fold increase	p-value
	control	IFNT treatment			control	IFNT treatment
XAF1	25.4±0.2	28.4±0.2	7.9	<0.001	22.1±0.5	27.1±0.2	32.2	<0.001
CASP8	25.7±0.3	28.3±0.4	6.2	<0.001	23.8±0.4	27.5±0.2	13.4	<0.001
FASLG	13.6±0.5	14.1±0.5	1.4	0.5	13.9±0.2	14.3±0.6	1.3	0.9

Back to Article Outline

4. Discussion 

Extensive apoptosis due to the process of implantation has been demonstrated in rodent and primate species as the attachment of the conceptus induces vigorous tissue remodeling during the invasive hemochorial placentation [19]. Implantation in ruminants is sparsely invasive, since only binucleate trophoblast cells invade the maternal epithelium from day 20 onward when IFNT secretion has already reached its maximum [3], [29]. Degenerative changes within the uterine epithelium are described in the caruncular area not earlier than day 22 of pregnancy [29]. A recently published gene expression analysis revealed 96 differentially expressed genes related to cell death in the caruncular areas of bovine day 20 endometria, but 245 genes were as well differentially expressed in intercaruncular tissue [30]. As little is known about apoptotic events in the intercaruncular endometrium when conceptus-derived IFNT acts on maternal tissue, we focused in our analysis on the functional secretory endometrium of the luteal phase.

In ruminants, IFNT is massively released during the periimplantation period [3], and the elevation of proapoptotic gene transcripts in bovine endometrium prior to implantation has been described [7], [15], [16], [31]. Elevated transcript levels of ISGs showing proapoptotic function, including XAF1, CASP8, TNFSF10, and FASLG, are well-known from humans following type I IFN treatment (reviewed in [6]). In the present study, in vivo during preimplantation as well as in glandular epithelial and stroma cells in vitro following IFNT treatment the expression of XAF1 and CASP8, but not FASLG, was enhanced, indicating species-specific differences. The IFN-signalling suppressor interferon regulatory factor 2 (IRF2) has been shown to be present in the luminal and superficial glandular epithelium in the ewe during pregnancy restricting the expression of ISGs to the zona basalis of the endometrium [32], [33], [34]. Interestingly, an in situ hybridization of XAF1 performed earlier [16] showed that XAF1 in bovine (referred to as HSXIAPAF1) was present in the luminal epithelium as well as in superficial and deep endometrial glands in both day 18 cyclic and pregnant animals indicating another species-specific peculiarity. Whether IRF2 in bovine might in general be restricting IFNT related gene expression to the endometrial zona basalis during pregnancy in vivo as found in the ewe needs to be proven. Nonetheless, most probably the presence of IFNT in the bovine uterus during early pregnancy led to enhanced endometrial XAF1 and CASP8 transcripts without concomitant apoptotic events in any region of the endometrium.

The most intense increase of a proapoptotic gene expression was observed for XAF1 in line with rising IFNT during the periimplantation period. The results obtained are in accordance with earlier transcriptome analyses [15], [16]. Therein, four percent of the genes upregulated in endometria of day 18 pregnant animals were associated with apoptotic function of which XAF1 was the most pronounced [15]. The main function of XAF1 is the direct binding of the BIRC4-caspase complex and the subsequent transfer to the nucleus for effective degradation [35] (Fig. 5). XAF1 causes further sensitization to TNF and TNFSF10-induced cell death and additionally is able to promote cytochrome c release from mitochondria through blocking Bcl-2-family proteins [36]. These multiple functions suggest a fundamental role in the endometrium at the window of implantation. Following IFNT stimulation, XAF1 may primarily act on endometrial cells or may be able to elicit selective cell death in lymphocytes in order to protect the semi-allogenic embryo. In the present study, the observed apoptotic cells were primarily located close to the uterine lumen in both pregnant and cyclic animals which might either indicate endometrial cells in natural cell turnover or leukocytes in a stage of apoptosis [37].

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 5 

  Apoptosis can be either elicited by the intrinsic pathway through proapoptotic Bcl-2-family factors (BAX) or the extrinsic pathway via death receptor ligands (FASLG, TNFSF10 and TNF) by binding to respective receptors (FAS, TNFRSF10B, TNFRSF1A,-B). Both pathways cause activation of initiator caspases (CASP8) and effector caspases (CASP3, 6, 7). Subsequent activation of caspase activated DNases result in DNA fragmentation and apoptosis. Proapoptotic components (XAF1, DIABLO) may diminish the effects of antiapoptotic factors (BCL2L1, CFLAR, BIRC4, 5). Although expression of proapoptotic genes XAF1, CASP8 and TNFSF10 was elevated in pregnant animals in the presence of trophoblast IFNT the activity of caspases and the incidence of apoptotic cells did not increase. Inhibitory mechanisms preventing endometrial cells from apoptosis are presumed.

Apoptosis can be either elicited by the intrinsic pathway through proapoptotic Bcl-2-family factors (BAX) or the extrinsic pathway via death receptor ligands (FASLG, TNFSF10 and TNF) by binding to respective receptors (FAS, TNFRSF10B, TNFRSF1A,-B). We did not detect changes in gene expression of the intrinsic apoptosis pathway, but a conspicuous increase of FASLG expression concomitant to an intense immune labelling of trophoblast cells suggests a role for FASLG in ruminants during the periimplantation phase. Studies in rodents and primates showed that immune privilege is generated via the FAS/FASLG system in order to protect the semi-allogenic embryo from activated lymphocytes [38]. The human trophoblast itself produces low levels of FAS and is usually resistant to FASLG mediated apoptosis [39]. In our study, transcripts of FAS were comparably abundant in both the endometrium and the conceptus. Aside from membrane-bound and soluble FASLG, a form secreted via microvesicles also exists in trophoblast cells which may act on maternal FAS-bearing immune cells [17]. Whether this secretion of FASLG also occurs in ruminants is currently unknown. If released, trophoblast-derived FASLG could act on FAS-bearing leukocytes in the luminal epithelium (Fig. 5). In the present study, FASLG protein was found in the deep endometrium which is in contrast to the human endometrium where FASLG protein was primarily localized to glandular epithelial cells of the zona functionalis [40]. Only the deep maternal endometrium expresses ISGs as MHC class I and beta-2 microglobulin in response to IFNT [15], [33] which impedes recognition of conceptus antigens by immune cells. Expression of FASLG in the zona basalis might inhibit the entrance of leukocytes to the endometrium as suggested in human [40] and therefore could act as additional mechanism to generate immune privilege by possibly preventing the maternal immune cells from attacking the conceptus.

A role in developing immune privilege during early pregnancy is also attributed to the death receptor ligand TNFSF10, since TNFSF10 has prominently been found in human syncytiotrophoblast and in Hofbauer cells [41]. TNFSF10 and its receptor TNFRSF10B are also expressed in rodent trophectoderm cells during the periimplantation phase, whereas the expression in bovine conceptus cells was only minor. In cattle, a more important role for maternal TNFSF10 in enabling immune privilege might be suggested. Four receptor subtypes exist of which two (TNFRSF10C/TRAILR3 and TNFRSF10D/TRAILR4) lack a functional death domain and therefore act as decoy receptors [42]. Activation of apoptosis by TNFSF10 depends on the binding of the receptor, and as we could not observe a remarkable increase of apoptosis, decoy receptors could prevent the endometrium from detrimental TNFSF10 effects. TNF represents a pleiotropic cytokine with both proinflammatory and proapoptotic properties. Transcripts rose steadily in cyclic and foremost in pregnant animals. Apart from proapoptotic action transduced by TNFRSF1A and TNFRSF1B, antiapoptotic genes can be activated via the NF-κB pathway [43]. Despite an increased TNF expression we could not demonstrate a pronounced expression of antiapoptotic genes. Our data indicate an increase of death receptor ligand transcripts, whereas the analysed death receptors (apart from TNFRSF1A) appear to be constitutively expressed in the endometrium. Although we found a marked increase of CASP8 transcripts due to the progressing periimplantation phase, the activity decreased over time in both cyclic and pregnant animals to the same extent (Fig. 5). As CASP3 but not CASP7 showed abundant expression, we assumed that CASP3 contributed primarily to the activity in endometrial homogenates. These data are contrary to findings in the human endometrium where activity of CASP8 and CASP3 significantly increased within the secretory phase [44]. We therefore suggest that the proapoptotic signals caused by IFNT are not transduced.

Resistance to IFNT induced cell death might derive from the competition with CFLAR (FLIP) which shares sequence homologies to CASP8 and competes for binding to the death inducing signalling complex. CFLAR exists in multiple splice variants, which differ in their antiapoptotic abilities (reviewed in [13]). We could not demonstrate changes in CFLAR expression; however ongoing investigations will include the differentiation of splice variants possibly involved as well as the analysis of transcript and protein stability which is important for adequate availability of this factor.

Bovine periimplantation conceptuses showed differential expression of apoptosis related factors. The proapoptotic FASLG along with the antiapoptotic factor BIRC5 exhibited marked expression in conceptuses compared to the maternal endometrium, whereas BCL2L1, TNFSF10, TNF, CASP8, CASP7 as well as XAF1 were only marginally expressed in conceptuses. In order to explain whether the differential expression of apoptosis related genes between the endometrium and conceptuses was due to the action of IFNT, the expression of the heterodimeric type I IFN receptor was analysed. A very faint expression in particular for IFNAR2 in conceptuses could be demonstrated. Thus, our data indicate a reduced susceptibility of the conceptus to its own IFN-signalling as described in the ewe [45] and hence a lack of proapoptotic gene expression induction through IFNT.

Conceptus BIRC5 exceeded by far the expression in the endometrium. BIRC5 is not a target of XAF1, it is rather inhibited by degradation, since the XAF1–BIRC4 complex causes its ubiquitination [46]. BIRC5 represents the smallest IAP and exhibits a bifunctional role in the regulation of cell division and in the suppression of effector caspases. In conceptuses, this factor is localized to microtubules and/or kinetochores in order to ensure an accurate distribution of chromosomes [47]. A contribution to the development of bovine embryos has also been described, since BIRC5 was expressed throughout all stages of oocyte development and in in vitro produced bovine conceptuses. Moreover, suppression of BIRC5 resulted in increased apoptosis [48]. Our data indicate a particular relevance for the intrinsic pathway, because BAX along with CASP3 was markedly expressed in conceptuses. On the other hand neither factors from the extrinsic pathway (except for FASLG) nor CASP8 was appreciably present. Apoptosis is a tightly regulated process which also occurs in early bovine conceptuses [49]. Apoptotic events as part of regular development could be suppressed as long as the formation of the conceptuses is untainted, however if cell division failures or insufficient supply of maternal growth factors prevail, the balance could be immediately shifted towards apoptotic cell death.

In the present study the messenger RNA transcripts of proapoptotic factors XAF1, CASP8 and TNFSF10 were increased close to the beginning of trophoblast attachment in pregnant animals, but not concomitantly accompanied by an elevation of apoptosis. The lack of an increased activity of the caspase cascade and the presence of only few apoptotic cells in the early pregnant endometrial tissue may suggest that inhibitory mechanisms exist to protect the bovine endometrium from tissue damage potentially caused by IFNT. In addition, FASLG release by the trophoblast might prevent from the attack of immune cells sensitized to conceptus antigens to support implantation and conceptus development.

Back to Article Outline

Acknowledgements 

We greatly acknowledge the support of the German Research Foundation (UL 350/1-2, FOR 478). The authors sincerely thank Katrin Mitko for excellent help with sampling of uterine tissue, Heike Lang and Romy Renner for helpful assistance conducting the interferon and luminometric assays and Dr. Ales Tichopad for his greatly appreciated support with the statistical analysis.

Back to Article Outline

References 

1. Pontzer CH, Bazer FW, Johnson HM. Antiproliferative activity of a pregnancy recognition hormone, ovine trophoblast protein-1. Cancer Res. 1991;51:5304–5307
   • View In Article
   • MEDLINE
2. Helmer SD, Gross TS, Newton GR, Hansen PJ, Thatcher WW. Bovine trophoblast protein-1 complex alters endometrial protein and prostaglandin secretion and induces an intracellular inhibitor of prostaglandin synthesis in vitro. J Reprod Fertil. 1989;87:421–430
   • View In Article
3. Farin CE, Imakawa K, Hansen TR, McDonnell JJ, Murphy CN, Farin PW, et al. Expression of trophoblastic interferon genes in sheep and cattle. Biol Reprod. 1990;43:210–218
   • View In Article
   • MEDLINE
   • CrossRef
4. Cross JC, Roberts RM. Constitutive and trophoblast-specific expression of a class of bovine interferon genes. Proc Natl Acad Sci U S A. 1991;88:3817–3821
   • View In Article
   • MEDLINE
   • CrossRef
5. Li J, Roberts RM. Interferon-tau and interferon-alpha interact with the same receptors in bovine endometrium. Use of a readily iodinatable form of recombinant interferon-tau for binding studies. J Biol Chem. 1994;269:13544–13550
   • View In Article
   • MEDLINE
6. Chawla-Sarkar M, Lindner DJ, Liu YF, Williams BR, Sen GC, Silverman RH, et al. Apoptosis and interferons: role of interferon-stimulated genes as mediators of apoptosis. Apoptosis. 2003;8:237–249
   • View In Article
   • MEDLINE
   • CrossRef
7. Wang B, Xiao C, Goff AK. Progesterone-modulated induction of apoptosis by interferon-tau in cultured epithelial cells of bovine endometrium. Biol Reprod. 2003;68:673–679
   • View In Article
   • MEDLINE
   • CrossRef
8. Baker PK, Pettitt AR, Slupsky JR, Chen HJ, Glenn MA, Zuzel M, et al. Response of hairy cells to IFN-alpha involves induction of apoptosis through autocrine TNF-alpha and protection by adhesion. Blood. 2002;100:647–653
   • View In Article
   • MEDLINE
   • CrossRef
9. Steiner T, Junker U, Henzgen B, Nuske K, Durum SK, Schubert J. Interferon-alpha suppresses the antiapoptotic effect of NF-kB and sensitizes renal cell carcinoma cells in vitro to chemotherapeutic drugs. Eur Urol. 2001;39:478–483
   • View In Article
   • MEDLINE
   • CrossRef
10. Earnshaw WC, Martins LM, Kaufmann SH. Mammalian caspases: structure, activation, substrates, and functions during apoptosis. Annu Rev Biochem. 1999;68:383–424
    • View In Article
    • MEDLINE
    • CrossRef
11. Abrahams VM, Kim YM, Straszewski SL, Romero R, Mor G. Macrophages and apoptotic cell clearance during pregnancy. Am J Reprod Immunol. 2004;51:275–282
    • View In Article
12. Oltvai ZN, Milliman CL, Korsmeyer SJ. Bcl-2 heterodimerizes in vivo with a conserved homolog, Bax, that accelerates programmed cell death. Cell. 1993;74:609–619
    • View In Article
    • MEDLINE
    • CrossRef
13. Thorburn A. Death receptor-induced cell killing. Cell Signal. 2004;16:139–144
    • View In Article
    • MEDLINE
    • CrossRef
14. Martin SJ, Green DR. Protease activation during apoptosis: death by a thousand cuts?. Cell. 1995;82:349–352
    • View In Article
    • MEDLINE
    • CrossRef
15. Bauersachs S, Ulbrich SE, Gross K, Schmidt SE, Meyer HH, Wenigerkind H, et al. Embryo-induced transcriptome changes in bovine endometrium reveal species-specific and common molecular markers of uterine receptivity. Reproduction. 2006;132:319–331
    • View In Article
    • MEDLINE
16. Klein C, Bauersachs S, Ulbrich SE, Einspanier R, Meyer HH, Schmidt SE, et al. Monozygotic twin model reveals novel embryo-induced transcriptome changes of bovine endometrium in the preattachment period. Biol Reprod. 2006;74:253–264
    • View In Article
    • MEDLINE
    • CrossRef
17. Abrahams VM, Straszewski-Chavez SL, Guller S, Mor G. First trimester trophoblast cells secrete Fas ligand which induces immune cell apoptosis. Mol Hum Reprod. 2004;10:55–63
    • View In Article
    • CrossRef
18. Makrigiannakis A, Karamouti M, Drakakis P, Loutradis D, Antsaklis A. Fetomaternal immunotolerance. Am J Reprod Immunol. 2008;60:482–496
    • View In Article
19. Liu YX, Gao F, Wei P, Chen XL, Gao HJ, Zou RJ, et al. Involvement of molecules related to angiogenesis, proteolysis and apoptosis in implantation in rhesus monkey and mouse. Contraception. 2005;71:249–262
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1393 KB)
    • CrossRef
20. Tassell W, Slater M, Barden JA, Murphy CR. Endometrial cell death during early pregnancy in the rat. Histochem J. 2000;32:373–379
    • View In Article
    • MEDLINE
    • CrossRef
21. Okano A, Ogawa H, Takahashi H, Geshi M. Apoptosis in the porcine uterine endometrium during the estrous cycle, early pregnancy and post partum. J Reprod Dev. 2007;53:923–930
    • View In Article
    • CrossRef
22. Ulbrich SE, Schulke K, Groebner AE, Reichenbach HD, Angioni C, Geisslinger G, et al. Quantitative characterization of prostaglandins in the uterus of early pregnant cattle. Reproduction. 2009;138:371–382
    • View In Article
23. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, et al. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem. 2009;55:611–622
    • View In Article
    • CrossRef
24. Pfaffl MW, Tichopad A, Prgomet C, Neuvians TP. Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper–Excel-based tool using pair-wise correlations. Biotechnol Lett. 2004;26:509–515
    • View In Article
    • MEDLINE
    • CrossRef
25. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods. 2001;25:402–408
    • View In Article
    • MEDLINE
    • CrossRef
26. Rubinstein S, Familletti PC, Pestka S. Convenient assay for interferons. J Virol. 1981;37:755–758
    • View In Article
    • MEDLINE
27. Stojkovic M, Buttner M, Zakhartchenko V, Riedl J, Reichenbach HD, Wenigerkind H, et al. Secretion of interferon-tau by bovine embryos in long-term culture: comparison of in vivo derived, in vitro produced, nuclear transfer and demi-embryos. Anim Reprod Sci. 1999;55:151–162
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (114 KB)
    • CrossRef
28. Ulbrich SE, Frohlich T, Schulke K, Englberger E, Waldschmitt N, Arnold GJ, et al. Evidence for estrogen-dependent uterine serpin (SERPINA14) expression during estrus in the bovine endometrial glandular epithelium and lumen. Biol Reprod. 2009;81:795–805
    • View In Article
    • CrossRef
29. Wathes DC, Wooding FB. An electron microscopic study of implantation in the cow. Am J Anat. 1980;159:285–306
    • View In Article
    • MEDLINE
    • CrossRef
30. Mansouri-Attia N, Aubert J, Reinaud P, Giraud-Delville C, Taghouti G, Galio L, et al. Gene expression profiles of bovine caruncular and intercaruncular endometrium at implantation. Physiol Genomics. 2009;39:14–27
    • View In Article
    • CrossRef
31. Chen Y, Antoniou E, Liu Z, Hearne LB, Roberts RM. A microarray analysis for genes regulated by interferon-tau in ovine luminal epithelial cells. Reproduction. 2007;134:123–135
    • View In Article
32. Choi Y, Johnson GA, Burghardt RC, Berghman LR, Joyce MM, Taylor KM, et al. Interferon regulatory factor-two restricts expression of interferon-stimulated genes to the endometrial stroma and glandular epithelium of the ovine uterus. Biol Reprod. 2001;65:1038–1049
    • View In Article
    • MEDLINE
    • CrossRef
33. Choi Y, Johnson GA, Spencer TE, Bazer FW. Pregnancy and interferon tau regulate major histocompatibility complex class I and beta2-microglobulin expression in the ovine uterus. Biol Reprod. 2003;68:1703–1710
    • View In Article
    • MEDLINE
    • CrossRef
34. Johnson GA, Austin KJ, Collins AM, Murdoch WJ, Hansen TR. Endometrial ISG17 mRNA and a related mRNA are induced by interferon-tau and localized to glandular epithelial and stromal cells from pregnant cows. Endocrine. 1999;10:243–252
    • View In Article
35. Liston P, Fong WG, Kelly NL, Toji S, Miyazaki T, Conte D, et al. Identification of XAF1 as an antagonist of XIAP anti-Caspase activity. Nat Cell Biol. 2001;3:128–133
    • View In Article
    • MEDLINE
    • CrossRef
36. Leaman DW, Chawla-Sarkar M, Vyas K, Reheman M, Tamai K, Toji S, et al. Identification of X-linked inhibitor of apoptosis-associated factor-1 as an interferon-stimulated gene that augments TRAIL Apo2L-induced apoptosis. J Biol Chem. 2002;277:28504–28511
    • View In Article
37. King GJ, Atkinson BA, Robertson HA. Development of the intercaruncular areas during early gestation and establishment of the bovine placenta. J Reprod Fertil. 1981;61:469–474
    • View In Article
    • MEDLINE
    • CrossRef
38. Uckan D, Steele A, Cherry , Wang BY, Chamizo W, Koutsonikolis A, et al. Trophoblasts express Fas ligand: a proposed mechanism for immune privilege in placenta and maternal invasion. Mol Hum Reprod. 1997;3:655–662
    • View In Article
    • MEDLINE
    • CrossRef
39. Aschkenazi S, Straszewski S, Verwer KM, Foellmer H, Rutherford T, Mor G. Differential regulation and function of the Fas/Fas ligand system in human trophoblast cells. Biol Reprod. 2002;66:1853–1861
    • View In Article
    • MEDLINE
    • CrossRef
40. Selam B, Kayisli UA, Mulayim N, Arici A. Regulation of Fas ligand expression by estradiol and progesterone in human endometrium. Biol Reprod. 2001;65:979–985
    • View In Article
    • MEDLINE
    • CrossRef
41. Phillips TA, Ni J, Pan G, Ruben SM, Wei YF, Pace JL, et al. TRAIL (Apo-2L) and TRAIL receptors in human placentas: implications for immune privilege. J Immunol. 1999;162:6053–6059
    • View In Article
    • MEDLINE
42. Kim Y, Seol DW. TRAIL, a mighty apoptosis inducer. Mol Cells. 2003;15:283–293
    • View In Article
    • MEDLINE
43. Stehlik C, de MR, Kumabashiri I, Schmid JA, Binder BR, Lipp J. Nuclear factor (NF)-kappaB-regulated X-chromosome-linked iap gene expression protects endothelial cells from tumor necrosis factor alpha-induced apoptosis. J Exp Med. 1998;188:211–216
    • View In Article
    • MEDLINE
    • CrossRef
44. Abe H, Shibata MA, Otsuki Y. Caspase cascade of Fas-mediated apoptosis in human normal endometrium and endometrial carcinoma cells. Mol Hum Reprod. 2006;12:535–541
    • View In Article
    • MEDLINE
    • CrossRef
45. Han CS, Mathialagan N, Klemann SW, Roberts RM. Molecular cloning of ovine and bovine type I interferon receptor subunits from uteri, and endometrial expression of messenger ribonucleic acid for ovine receptors during the estrous cycle and pregnancy. Endocrinology. 1997;138:4757–4767
    • View In Article
    • MEDLINE
    • CrossRef
46. Arora V, Cheung HH, Plenchette S, Micali OC, Liston P, Korneluk RG. Degradation of survivin by the X-linked inhibitor of apoptosis (XIAP)-XAF1 complex. J Biol Chem. 2007;282:26202–26209
    • View In Article
    • CrossRef
47. Fortugno P, Wall NR, Giodini A, O'Connor DS, Plescia J, Padgett KM, et al. Survivin exists in immunochemically distinct subcellular pools and is involved in spindle microtubule function. J Cell Sci. 2002;115:575–585
    • View In Article
    • MEDLINE
48. Park SY, Kim EY, Jeon K, Cui XS, Lee WD, Kim NH, et al. Survivin acts as anti-apoptotic factor during the development of bovine pre-implantation embryos. Mol Reprod Dev. 2007;74:582–590
    • View In Article
    • MEDLINE
    • CrossRef
49. Gjorret JO, Knijn HM, Dieleman SJ, Avery B, Larsson LI, Maddox-Hyttel P. Chronology of apoptosis in bovine embryos produced in vivo and in vitro. Biol Reprod. 2003;69:1193–1200
    • View In Article
    • MEDLINE
    • CrossRef
50. Kliem H, Berisha B, Meyer HHD, Schams D. Regulatory changes of apoptotic factors in the bovine corpus luteum after induced luteolysis. Mol Reprod Develop. 2009;76:220–230
    • View In Article