Fibroblast Growth Factor 2 Promotes Primitive Endoderm ...

BIOLOGY OF REPRODUCTION 85, 946–953 (2011) Published online before print 20 July 2011. DOI 10.1095/biolreprod.111.093203

Fibroblast Growth Factor 2 Promotes Primitive Endoderm Development in Bovine Blastocyst Outgrowths1 Qi En Yang, Sarah D. Fields, Kun Zhang, Manabu Ozawa, Sally E. Johnson, and Alan D. Ealy2 Department of Animal Sciences, University of Florida, Gainesville, Florida

Primitive endoderm (PE) is the second extraembryonic tissue to form during embryogenesis in mammals. The PE develops from pluripotent cells of the blastocyst inner cell mass. Experimental results described herein provide evidence that FGF2 stimulates PE development during bovine blastocyst development in vitro. Bovine blastocysts were cultured individually on a feeder layer-free, Matrigel-coated surface in the presence or absence of FGF2. A majority of blastocysts cultures formed outgrowths (76.8%) and the rate of outgrowth formation was not affected by FGF2 supplementation. However, supplementation with FGF2 increased the incidence of PE outgrowths on Days 13 and 15 after in vitro fertilization. Presumptive PE cultures contained cells with a phenotype distinct from trophectoderm (TE). Cell identity was validated by expression of GATA4 and GATA6 mRNA and transferrin protein, all markers of the PE lineage. Expression of GATA4 occurred coincident with blastocyst expansion and hatching. These cells did not express IFNT and CDX2 (TE lineage markers). Profiles of FGF receptor (FGFR) isoforms were distinct between PE and TE cultures. Specifically, FGFR1b and FGFR1c were the predominant FGFR transcripts in PE whereas FGFR2b transcripts were abundant in TE. Supplementation with FGF2 increased the mitotic index of PE but not TE. Moreover, FGF signaling appears important for initiation of PE formation in blastocysts, presumably by lineage committal from NANOG-positive epiblast cells, because chemical disruption of FGFR kinase activity with PD173074 reduces GATA4 expression and increases NANOG expression. Collectively, these results indicate that FGF2 and potentially other FGFs specify PE formation and mediate PE proliferation during early pregnancy in cattle. blastocyst, developmental biology, embryo, endoderm, FGF2, GATA4, gene expression, hormone action, lineage specification, primitive endoderm, trophoblast

INTRODUCTION Early conceptus development in ruminants is different from rodents, primates, and many other species because of their prolonged pre- and peri-implantation development. Bovine blastocysts hatch from the zona pellucida between Days 7 and 1 Supported by National Research Initiative Competitive Grant no. 2008-35203-19106 from the U.S. Department of Agriculture’s National Institute of Food and Agriculture. 2 Correspondence: Alan D. Ealy, Department of Animal Sciences, University of Florida, P.O. Box 110910, Gainesville, FL 32611-0910. FAX: 352 392 5595; e-mail: [email protected]

Received: 5 May 2011. First decision: 13 June 2011. Accepted: 11 July 2011. Ó 2011 by the Society for the Study of Reproduction, Inc. eISSN: 1529-7268 http://www.biolreprod.org ISSN: 0006-3363

946

Downloaded from www.biolreprod.org.

10 postfertilization and remain free-floating for another week or more before adhering to the uterine lining on or after Day 19 of pregnancy [1–3]. Conceptus elongation begins during the second and third weeks of pregnancy as a result of the reorganization and proliferation of trophectoderm (TE) [3, 4]. Rapid TE expansion maximizes early placental surface interactions with the uterus and ensures the production of sufficient amounts of interferon-tau (IFNT), the signal of maternal pregnancy [5, 6]. After blastocyst formation and specification of TE has occurred, the inner cell mass (ICM) develops into distinct hypoblast and epiblast (EPI) layers. The EPI undergoes rearrangement and gastrulation to form the three primary germ layers of the developing embryo. The hypoblast is made up of primitive endoderm (PE). These cells are first evident between Days 8 and 10 postfertilization in cattle [7–9] and form an extraembryonic layer on the internal surface of the TE and ICM that eventually forms the inner layer of the yolk sac [10, 11]. In mice, determination of EP and EPI fates is not defined by their location within the ICM but rather is controlled by differential expression of specific transcription factors. Specifically, ICM cells that produce primarily NANOG become EPI whereas cells that produce greater amounts of GATA4 and GATA6 form PE [12–14]. Targeted disruption of GATA4 or GATA6 expression blocks PE formation in embryonic stem (ES) cells [15, 16] while ectopic expression of either factor in ES cells induces PE formation [17, 18]. In cattle, select ICM cells are immunopositive for GATA6, but it remains unclear if they are fated to become PE [19]. Indeed beyond the establishment and propagation of bovine PE cultures from blastocyst outgrowths [20, 21], little information exists regarding growth factor and morphogen control of PE lineage specification in cattle. There is good evidence that fibroblast growth factor (FGF) signaling and MAPK3/1 phosphorylation play key roles in initiating PE formation in mouse embryos. Embryos that lack ICM-derived FGF4, its cognate receptor FGFR2, or the receptor adapter protein GRB2 do not form PE [22–25]. Overexpression of a dominant-negative FGFR in ES cells also prevents PE formation [26]. FGF signaling through GRB2 likely induces PE formation via GATA factor transactivation. Mouse embryos null for Grb2 do not produce GATA6, and all the ICM cells become NANOG-positive EPI [12]. Pharmacological inhibition of FGFR or MAPK3/1 activity prevents Gata6 expression and PE formation in mouse blastocysts whereas supplementation with FGF4 increases the number of GATA6-positive PE cells in the ICM [27]. In cattle, FGF2 is produced by endometrial epithelium and secreted into the uterine lumen throughout the pre- and periimplantation period [28, 29]. Transcripts for FGF2 and FGF4 are detected in bovine blastocysts [30–32] suggesting that FGF signaling is vital to bovine embryogenesis. The results presented herein establish a linkage between FGF2 signals and PE formation in cattle.

ABSTRACT

FGF2 SUPPORTS ENDODERM DEVELOPMENT IN BLASTOCYSTS

947

MATERIALS AND METHODS

Quantitative RT-PCR

Materials

The concentration and quality of tcRNA (A260/A280 ratio  1.8) was determined using a NanoDrop 2000 Spectrophotometer (Thermo Scientific). Samples were incubated with RNase-free DNase for 30 min at 378C. After heat inactivating the DNase (758C for 10 min), RNA was reverse transcribed using the High Capacity cDNA Reverse Transcription kit using random hexamers. The SybrGreen Detection System was used in combination with primer pairs (200 nM) (see Supplemental Table S1, available online at www.biolreprod. org). A 7300 Real Time PCR System (Applied Biosystems) was used to quantify the relative abundance of specific transcripts. GAPDH mRNA level was used to normalize values. GAPDH mRNA abundance did not change based on cell type or treatment in these studies (data not shown). Also, Purcell et al. [38] have shown that GAPDH mRNA concentrations remain constant throughout pre- and peri-attachment conceptus development in sheep (i.e., as elongation occurs). A dissociation curve analysis (608C–958C) was used to verify the amplification of a single product, and each sample was performed in triplicate. A fourth reaction lacking reverse transcriptase was included to control for genomic DNA contamination. The comparative threshold cycle (CT) method was used to quantify mRNA abundance [32].

In Vitro Production of Bovine Embryos All the animal research was completed in accordance with federal, state, and local laws and regulations and with the approval of the University of Florida Institutional Animal Care and Use Committee. In vitro produced (IVP) bovine blastocysts were generated using maturation, fertilization, and culture procedures described previously [33, 34] using bovine ovaries collected from a local slaughterhouse and transported in 0.9% (w/v) NaCl at room temperature. Matured oocytes were fertilized with 1 3 106 Percoll-purified spermatozoa from frozen-thawed semen collected from three bulls. Different bulls were used throughout the studies. After in vitro fertilization (IVF), putative zygotes were denuded by vortexing and placed in groups of 20–30 in 50 ll of SOF at 38.58C in a humidified atmosphere containing 5% (v/v) O2, 5% (v/v) CO2, and 90% (v/v) N2. For one study, blastocysts were collected and used to generate outgrowth cultures (see below). In another study, embryos were collected at various stages for RNA extraction using the PicoPure RNA Isolation Kit. In a third study, regular and expanded blastocysts collected at Day 7 post-IVF (n ¼ 10/well) were exposed to 50 ng/ml FGF2 or 1 lM FGFR inhibitor (PD173074) [35, 36] in DMEM containing 5% FBS, 100 lM nonessential amino acids, 55 lM b-mercaptoethanol, and 250 units/ml antibiotic-antimycotic mix at 38.58C in a 5% oxygen environment as described previously [37]. After 24 h, tcRNA (total cellular RNA) was isolated using the PicoPure RNA Isolation Kit.

Blastocyst Outgrowth Culture On Day 8 postfertilization, regular and expanded blastocysts were placed individually into Matrigel-coated wells (0.95 cm2) in 500 ll DMEM containing 5% FBS and other supplements at 38.58C in a 5% oxygen environment as described previously [37]. Serum was included based on previous work showing that serum supplementation was required to limit blastocyst degeneration during extended culture [37]. Different concentrations of FGF2 (0.5, 5, and 50 ng/ml in DMEM containing 1% [w/v] bovine serum albumin) or controls (carrier only) was provided at the beginning of culture on Day 8 postfertilization. The medium was replaced with fresh medium lacking FGF2 on Days 13 and 15 post-IVF without disrupting blastocysts that had attached to the matrix. Approximately one-half the medium was exchanged in embryos that had not attached on Day 13 or 15. Each blastocyst was assessed under phase contrast microscopy for its attachment and viability status (floating/unattached; attached; outgrowth formation). In follow-up work, some TE and PE outgrowths were maintained to Day 20 post-IVF (n ¼ 6) when RNA was isolated using the PicoPure RNA Isolation Kit. Additional outgrowths of each line (n ¼ 6 for each) were passaged to establish PE and TE lines. Outgrowths containing PE were passaged by exposure to 0.05% (w/v) trypsin and 0.43 mM ethylenediaminetetraacetic acid. These cells could be maintained in the absence of Matrigel basement membrane. TE outgrowths were passaged by mechanical disruption onto Matrigel-coated plates as described previously [28].

Immunofluorescence and Confocal Microscopy Primary TE and PE cultures were fixed with 4% (w/v) paraformaldehyde for 15 min, permeabilized with 0.5% (v/v) Triton X-100 in 0.01 M PBS, pH 7.4, for 30 min, and blocked with 10% (v/v) goat serum for 60 min. Primary antibodies used included rabbit anti-GATA4 (1:200 dilution) and mouse antiCDX2 (1:500 dilution). Secondary antibodies were conjugated to FITC (goat anti-rabbit) or Alexa Fluor 555 (goat anti-mouse). Nuclei were labeled with DAPI. Immunoreactive complexes were visualized with an Eclipse TE2000-U inverted microscope (Nikon) equipped with an X-Cite 120 epifluorescence illumination system (EXFO). Images were captured with a Nikon DXM-1200F digital camera and assembled with NIS-Elements Software (Nikon). For confocal microscopy, immunoreactive signals were visualized with an Olympus IX81-DSU Spinning Disk Confocal Microscope. Images were captured with a Hamamatsu C4742-80-12AG Monochrome CCD Camera (Hamamatsu Corporation) and assembled with SlideBook software (Intelligent Imaging Innovations).

Western Blot Analyses The PE and TE outgrowths (n ¼ 3 for each cell type) were seeded onto Matrigel-coated plates (25 mm diameter). After 3 days, cellular protein was extracted, and SDS-PAGE and Western blot analysis was completed as described previously [39]. Protein (20 lg) was loaded and separated on 10% (w/v) SDS-PAGE gels and electrotransferred onto 0.45 lm PVDF. Membranes were blocked with 5% (w/v) nonfat dry milk in 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, and 0.1% Tween-20 and incubated with anti-transferrin antiserum as described previously [40]. Horseradish peroxidase-conjugated anti-rabbit IgG was used in combination with ECL to visualize immunoreactive bands after exposure to BioMax film. After detection, membranes were stripped, and atubulin was detected to serve as a loading control. Three independent Western blots generated from independent PE or TE cultures were completed.

Proliferation Assay Cells were seeded on Matrigel-coated plates (25 mm diameter). Upon achieving 40%–50% confluency, cells were serum-starved for 12 h after which FGF2 was added (0, 5, or 50 ng/ml). After 11 h, 20 lM 5-ethynyl-2 0 deoxyuridine (EdU) reagent was added, and the cells were incubated for 60 min before fixation in 4% paraformaldehyde in PBS for 15 min. EdU-positive cells were determined by reaction with Alexa Fluor 488 azide. All the nuclei were visualized after staining with 0.1 lM Hoechst 33342. Total and EdU-positive nuclei were counted in four representative fields in each well using NISElements Software (Nikon) after capture on an Eclipse TE2000-U inverted microscope.

Statistical Analyses All the analyses were performed by least-squares ANOVA using the general linear model of the Statistical Analysis System (SAS Institute Inc.). The percentage data were arcsine transformed before analysis. Differences between individual means were compared using pairwise comparisons (PDIFF [probability of difference] analysis in SAS). For qRT-PCR, CT values were expressed relative to that of the reference control (GAPDH) (2CT (transcript of interest)/2CT (GAPDH)) and analyzed after log transformation

Downloaded from www.biolreprod.org.

All the cell culture reagents, including Dulbecco modified essential medium containing high glucose (DMEM) and fetal bovine serum (FBS), as well as PCR primers, 4 0 ,6-diamidino-2-phenylindole (DAPI), and EdU Cell Proliferation Assay were purchased from Invitrogen Corp. Synthetic Oviduct Fluid (SOF) medium was purchased as a custom formulation from the Specialty Media Division of Millipore Corp. Bovine recombinant FGF2 was purchased from R&D Systems. Matrigel Basement Membrane Matrix was purchased from BD Biosciences. The PicoPure RNA Isolation Kit was purchased from MDS Analytical Technologies. RNase-free DNase was purchased from New England Biolabs. The High-Capacity cDNA Reverse Transcription Kit and SybrGreen Detector System were purchased from Applied Biosystems Inc. The pharmacological inhibitor for FGFR (PD173074) was purchased from EMD Chemicals. The proteinase and phosphatase inhibitors cocktails, BCA Protein Assay, and BioMax film were purchased from Thermo Fisher Scientific. Polyvinylidene difluoride (PVDF) (Immobilon-P) membrane was purchased from Millipore Co. Monoclonal anti-CDX2 antibody was purchased from BioGenex. Polyclonal anti-GATA4 antibody was purchased from Santa Cruz Biotechnology. Fluorescein isothiocyanate (FITC)-labeled anti-mouse immunoglobulin G (IgG) and anti-tubulin were purchased from Abcam. Alexa Fluor 555-labeled anti-rabbit IgG and horseradish peroxidase-labeled anti-rabbit IgG were purchased from Cell Signaling Technology. Anti-transferrin antiserum was purchased from Sigma-Aldrich, Inc. The enhanced chemiluminescence (ECL) Western blot detection system was purchased from GE Healthcare.

948

YANG ET AL. TABLE 1. The incidence of blastocyst attachment, outgrowth formation, and type of outgrowth formation during extended culture.* Outgrowth type (%)§

Attached FGF2 (ng/ml) Day 13 0.0 0.5 5.0 50.0 Day 15 0.0 0.5 5.0 50.0

No outgrowth (%) 

Outgrowth formation (%)z

TE

PE

20.22 17.67 13.92 20.37

6 6 6 6

5.45 3.11 4.20 3.45

61.56 62.65 71.98 63.52

6 6 6 6

2.97 4.63 5.18 6.17

98.48 90.24 76.85 66.36

6 6 6 6

3.71ab 6.43ab 3.52c 3.65c

1.51 9.76 23.15 32.12

6 6 6 6

1.51a 6.44ab 3.52c 3.74c

10.82 16.32 10.73 12.02

6 6 6 6

3.48 2.49 3.22 1.96

78.52 73.27 80.91 74.63

6 6 6 6

6.06 4.70 4.66 3.74

98.81 88.51 76.40 69.68

6 6 6 6

1.19a 7.50b 2.82c 4.33c

1.19 11.48 23.59 30.32

6 6 6 6

1.19a 7.50b 2.82c 4.33c

* n ¼ 76–79 blastocysts (8 to 15 blastocysts/replicate; 6 replicates).   An overall decrease (P ¼ 0.04) in attached blastocysts that lacked outgrowth formation was observed at Day 15 when compared with Day 13. The main effect of FGF2 treatment and the interaction between FGF2 treatment and day were not significant. z An overall increase (P ¼ 0.003) in outgrowth formation was detected at Day 15 when compared with Day 13. The main effect of FGF2 treatment and the interaction between FGF2 treatment and day were not significant. § A main effect of FGF2 treatment on the incidence of TE and PE outgrowth formation was detected (P ¼ 0.0001). a–c Different superscripts within each column indicate differences between individual pairs (P , 0.05).

RESULTS Effect of FGF2 on Blastocyst Outgrowth Formation Blastocysts attach to and develop outgrowths when cultivated on complex extracellular matrices (e.g., Matrigel). Outgrowths were detected beginning at Day 12 post-IVF (4 days in outgrowth culture conditions), and the overall percentage of outgrowth formation averaged 64.6% 6 2.4% at Day 13 and increased (P ¼ 0.003) to 76.8% 6 2.3% at Day 15 post-IVF. The percentage of attached blastocysts that had not formed outgrowths was greater (P ¼ 0.04) at Day 13 than Day 15 (18.0% 6 2.0% vs. 12.5% 6 1.4%). Similarly, the

FIG. 1. Formation of TE and PE outgrowths from bovine blastocyst cultures. Bovine blastocysts (Day 8 post-IVF) were cultured individually in a 48-well plate coated with Matrigel. After 5 days, representative outgrowths for TE (A) and PE (B) were taken under phase contrast microscopy (original magnification 3200). Outgrowths were cultured for another 5 days, and cell lysates were combined into three pools and electrophoresed. Sequential Western blot analysis for transferrin and atubulin was then completed (C).

percentage of nonattached blastocysts was greater (P ¼ 0.05) at Day 13 than Day 15 (17.3% 6 3.2% vs. 10.7% 6 1.9%). No additional outgrowths were observed after Day 15. Effect of FGF2 on TE and PE Outgrowth Formation Supplementation with FGF2 did not affect the percentage of attached blastocysts or the percentage of blastocysts that formed outgrowths on Days 13 and 15 (Table 1). However, FGF2 influenced the types of cells observed in these outgrowths. Small, cobblestone-like cells containing prominent nuclei and small cytoplasmic areas growing in well-defined colonies form in the absence of FGF2. These cells represented TE (Fig. 1A). A single outgrowth from control embryos also contained cells that were larger than TE that had a polyhedral shape and were less densely arranged than TE (Fig. 1B). The morphology of these cells suggested they were PE [21, 40]. Treatment with 0.5 ng/ml FGF2 on Day 8 post-IVF increased (P , 0.05) the proportion of PE outgrowths observed at Day 15. Treatment with 5 or 50 ng/ml FGF2 produced more profound effects on the percentage of PE outgrowths, and these concentrations increased (P , 0.05) the percentage of PE outgrowths when compared with control and low-dose FGF2 treatments at Days 13 and 15. Many of these PE-containing outgrowths also contained TE at Day 13, but PE rapidly overran these cultures; on Day 15 and thereafter a majority of these outgrowths contained primarily PE (data not shown). Concurrent with observing more PE outgrowths, reductions in the percentage of outgrowths containing only TE (i.e., outgrowths devoid of PE) were noted in FGF2-treated blastocysts (Table 1). Reductions (P , 0.05) in the percentage of TE-only outgrowths were evident at Day 13 in cultures treated with 5 or 50 ng/ml FGF2 and at Day 15 in cultures containing 0.5, 5, or 50 ng/ml FGF2. The identity of the presumptive PE cells was confirmed by biochemical evaluation of endoderm protein expression. Pools of morphologically defined PE derived from FGF2-treated cultures and TE outgrowths from nontreated cultures (n ¼ 5–10 outgrowths/group; three separate groups) were processed for Western blot assessment of transferrin production, a PEspecific product [20]. Immunoreactive transferrin was detected in each of the three groups of PE outgrowths and was absent in

Downloaded from www.biolreprod.org.

[28, 32]. Differences between individual means were contrasted with pair-wise analysis (PDIFF analysis in SAS). The results are presented as arithmetic means 6 SEM. A P-value 0.05 was considered significant.

FGF2 SUPPORTS ENDODERM DEVELOPMENT IN BLASTOCYSTS

949

FIG. 2. Expression of selective transcripts and proteins in PE and TE. Outgrowths defined as PE and TE by visual assessment under phase contrast microscopy were either processed for qRT-PCR (A) or immunostaining (B). A) The relative abundance of transcripts for several putative PE- and TEspecifying factors was determined (n ¼ 6 of each; two outgrowths collected on three separate occasions). Data are presented relative to GAPDH mRNA abundance as means 6 SEM. The asterisk denotes differences (P , 0.05) in relative mRNA abundance between cell types. B) Representative pictures are provided to depict TE and PE stained with DAPI or immunostained with antisera for CDX2 or GATA4 and Alexa Fluor-labeled secondary antibodies. After staining, the cells were examined by using epifluorescence microscopy (original magnification 3200). Color images were converted to black and white.

Profiling EPI, PE, and TE Lineage Marker Expression in Bovine Embryos

Profiling of Lineage Markers in PE and TE Outgrowths

The spatio-temporal expression profile of NANOG, CDX2, and GATA4 was examined in IVP bovine embryos to understand the timing of PE formation in these embryos. Total RNA was collected from embryos at representative developmental stages between Days 6 and 8 post-IVF and analyzed by qRT-PCR (Fig. 3A). Transcripts for NANOG were detected at the morula stage, and its relative abundance increased (P , 0.05) in blastocysts at regular, expanded, and hatched stages. Similarly, transcripts for CDX2 were detected at the morula and increased (P , 0.05) in abundance in the various blastocyst stages. By contrast, GATA4 mRNA was not detected at the morula stage and low amounts of GATA4 mRNA were detected in regular blastocysts. The abundance of GATA4 mRNA increased (P , 0.05) in expanded and hatched blastocysts. Importantly, GATA4 protein localized within cells of the ICM and were absent from the outer TE, which immunoreacted with CDX2 antisera (Fig. 3B).

Several transcriptional regulators that serve as lineage markers for PE, TE, and pluripotent cells (i.e., EPI) in rodents and human were examined to describe similarities and differences in lineage marker expression in the bovid and to further verify PE and TE phenotypes in these cultures. In the first study, a subset of PE and TE outgrowths (n ¼ 6 of each) were cultured until Day 20 post-IVF, then tcRNA was extracted and qRT-PCR was completed to examine the relative abundance of PE and TE transcripts (Fig. 2A). Transcript abundance for GATA4 and GATA6, PE-specifying factors in mice and humans, was greatest (P , 0.05) in bovine PE outgrowths. GATA4 mRNA was only detected in PE whereas GATA6 mRNA was identified in both PE and TE. Genetic markers of pluripotency (NANOG, POU5F1) and markers of TE (CDX2, IFNT) also were examined from these samples (Fig. 2A). Neither the PE nor the TE outgrowths contained detectable amounts of NANOG whereas both contained equivalent trace amounts of POU5F1 mRNA. Both CDX2 and IFNT mRNA were very abundant in TE. CDX2 mRNA was not detected in any PE outgrowths while trace amounts of IFNT mRNA could be detected in some of the PE outgrowths. This likely reflects low-level TE contamination in some PE outgrowths at the time of cell collection. Immunofluorescence microscopy confirmed the exclusive production of GATA4 and CDX2 to PE and TE, respectively (Fig. 2B).

Expression of FGFRs in PE and TE Cultures The relative expression of FGFR isotypes in PE and TE were determined as a first step in defining the mechanism by which FGF2 differentially regulates these lineages in cattle. Each of the four major tyrosine kinase FGFRs (R1–4) were examined (Fig. 4A). Transcripts for FGFR1 and FGFR2 were detected in PE and TE. The relative abundance of FGFR1 mRNA was greater (P , 0.05) in PE than TE whereas FGFR2 mRNA was greater (P , 0.05) in TE than PE. FGFR3 mRNA

Downloaded from www.biolreprod.org.

each of the protein lysates from TE outgrowths that lacked PE (Fig. 1C).

950

YANG ET AL.

FIG. 3. Lineage marker expression in bovine embryos. A) Embryos were collected at the morula (M), early/regular blastocyst (B), expanded blastocyst (EB), or hatched blastocyst (HB) stages on Days 6 (M), 7 (M, B), or 8 (EB, HB) after IVF (n ¼ 12–18 embryos/stage; three replicate studies). Total cellular RNA was isolated, and qRT-CPR was completed. Data are presented relative to GAPDH mRNA abundance as means 6 SEM. Different superscripts denote differences (P , 0.05) in relative mRNA abundance for each transcript across embryo stages. B) Representative pictures are provided for hatched blastocysts immunostained for DAPI, CDX2, or GATA4. Alexa Fluor-labeled secondary antibodies were used for detection, and the outcomes were examined by using epifluorescence microscopy (original magnification 3200). Color images were converted to black and white.

Divergent Actions of FGF Signaling During Bovine Embryogenesis A definitive role for FGF2 in regulating PE colony expansion was described. Outgrowths of TE and PE were established and treated with 5 or 50 ng/ml of FGF2. After 24 h, the mitotic index of TE and PE outgrowths was measured by EdU, a thymidine analog, incorporation into DNA (Fig. 4C). In TE, neither concentration of FGF2 affected the percentage of EdU-positive cells. By contrast, both concentrations of FGF2 increased (P , 0.05) the percentage of EdU-positive PE cells when compared with the control. The importance of the FGF ligand and receptor signaling during bovine blastocyst development was examined in IVP embryos. Regular and expanded blastocysts collected on Day 7 post-IVF were treated for 24 h with 1 lM PD173074, a paninhibitor of FGFR kinase activity [35, 36], an equivalent amount of dimethyl sulfoxide, or 50 ng/ml FGF2. The concentration of inhibitor was chosen because it was effective at blocking FGF2-induced effects in bovine trophoblast cultures (Ozawa and Ealy, unpublished results). The PD173074 compound acted in the same manner as another FGFR kinase inhibitor, termed SU5402 [41]; PD173074 was used in place of the other inhibitor because it contained a lower

effective concentration (1 vs. 25 lM) and because it was more readily available through a U.S. vendor. Changes in the proportions of EPI, TE, and PE were measured indirectly by quantification of NANOG, CDX2, and GATA4 transcripts, respectively (Fig. 5). Treatment of blastocysts with 50 ng/ml FGF2, a concentration effective at promoting PE outgrowths in extended culture (see Table 1), did not alter transcript abundance for any of the lineage markers examined. However, exposure to the FGFR inhibitor increased (P , 0.05) the abundance of NANOG mRNA, did not affect CDX2 mRNA concentrations, and decreased (P , 0.05) the abundance of GATA4 mRNA. These results support a role for FGFR-initiated signals for the specification of hypoblast cells that likely are mediated by an endogenous FGF ligand. DISCUSSION The roles that FGFs play in regulating early conceptus development in ruminants are just beginning to be realized. This work describes a finding that links FGF2 with PE outgrowth formation in bovine blastocysts. Outgrowths containing TE are common in cattle [42–44], and providing FGF2 did not influence the overall incidence of outgrowths that initially contained TE in this work. The generation of primitive PE from bovine blastocyst outgrowths also has been observed previously in studies utilizing feeder cell systems [20, 21]. In this feeder cell-free system, a single PE-containing outgrowth was observed under control conditions, but substantially more PE-containing cultures were evident when blastocysts were cultured in medium containing FGF2. Verifying that these cells were PE was accomplished in several ways. Protein lysates from presumptive PE outgrowths were transferrin positive, a marker of PE [40], and PE lines generated from these outgrowths contained GATA4 and GATA6 transcripts. GATA4 mRNA was not detected in TE outgrowths and small amounts of GATA6 mRNA were detected in TE. Also, immunoreactive

Downloaded from www.biolreprod.org.

was detected in TE but not PE. Transcripts for FGFR4 were not detected in TE or PE. To further define FGFR isotype profiles in PE and TE, a subsequent study examined the expression profile of the predominant splice variant forms of FGFR1 and FGFR2 (IIIb and IIIc variants, referred to herein as R1b, R1c, R2b, and R2c) (Fig. 4B). Transcripts for R1b and R1c were greater (P , 0.05) in PE than TE whereas R2b mRNA was more abundant (P , 0.05) in TE than PE. The R2c isotype was in low abundance in both cells, and its relative abundance was not different between cell types.

FGF2 SUPPORTS ENDODERM DEVELOPMENT IN BLASTOCYSTS

951

FIG. 4. Expression profiles for FGFRs and regulation of cell proliferation by FGF2 in cells. A, B) Total cellular RNA was extracted from TE and PE outgrowths, and the relative abundance for FGFR1–3 (A) or specific alternatively spliced variants of FGFR1 and FGFR2 (B) was determined with qRT-PCR (n ¼ 6 of each; two outgrowths collected on three separate occasions). Data are presented relative to GAPDH mRNA abundance as means 6 SEM. The asterisk denotes differences (P , 0.05) in relative mRNA abundance between cell types. C) Extended cultures established from TE and PE outgrowths (n ¼ 3) were used to determine if FGF2 supplementation affects cell proliferation. Cells were serum-starved for

3 12 h before supplementation with FGF2 (or carrier only). After 11 h, 20 lM EdU was added, and the cells were incubated for 60 min before fixation. Alexa Fluor-488 azide was used to visualize EdU-positive nuclei. All the nuclei were visualized after staining with Hoechst 33342 (0.1 lM). Total and EdU-positive nuclei were counted (six representative fields/well; three wells/treatment). Data are presented as means 6 SEM. Different superscripts denote differences (P , 0.05) between treatments within each cell type.

Downloaded from www.biolreprod.org.

GATA4 protein localized to the nucleus of PE cells and was absent in TE. The PE lines also were devoid of CDX2 mRNA and protein. Expression profiling of PE and TE cell lines also helped to validate their lineage types. A low amount of IFNT mRNA was detected in some but not all PE outgrowths, likely because these initial outgrowths contained a small amount of TE cells. Expression of this factor was transient, and no IFNT mRNA could be detected in serial passaged PE (data not shown). Bovine TE cells are sensitive to trypsin treatment [20, 43], and using this agent to passage the PE lines likely limited TE passage rates. Both PE and TE were devoid of NANOG, which is consistent with the notion that these cells were committed to specific cell fates and were no longer in a ground, or noncommitted, state. Interestingly, both cell types contained small amounts of POU5F1 mRNA. POU5F1 is a pluripotency marker in mouse and human embryonic cells [45, 46] but not in cattle, sheep, and goats. POU5F1 mRNA and protein can be detected in TE for several days after their formation in blastocysts [19, 47, 48]. This work provides evidence that POU5F1 also is expressed in bovine PE outgrowths. Identifying that GATA4 expression was tightly linked with PE lineage specification permitted us to describe the ontogeny of PE formation in bovine embryos. The emergence of GATA4-producing PE was evident within the ICM after the divergence of TE and ICM. The ontogeny of this GATA4 profile is consistent with observations made by others describing the ontogeny of PE formation at Days 8 to 10 postfertilization [7–9]. Attempts were made to use dualimmunofluorescence to describe noncommitted EPI and PEcommitted cells within bovine ICMs. This technology was used to describe PE lineage emergence in mouse embryos [49, 50]. Unfortunately, clear indications of cell-specific staining patterns could not be completed in bovine blastocysts. An antiserum that reacted with bovine GATA4 was identified and reacted with a subset of cells within the ICM (see Fig. 3), but a suitable NANOG antiserum could not be found to identify noncommitted EPI. Also, the bovine ICM is substantially smaller than that of the mouse, and this prevented us from reliably counting the proportion of ICM cells that were GATA4 positive. Because of these issues, estimates of the relative portion of EPI and PE cell types within blastocysts were examined by determining changes in the relative abundance of NANOG and GATA4 transcripts over time or following exposure to specific treatments. Two primary studies were completed to understand how FGF2 promotes PE outgrowth formation in extended blastocyst cultures. The first study determined that FGF2 increases PE proliferation. Mitotic index rates were increased in PE but not TE cultures. Interestingly, supplementing blastocysts with FGF2 did not alter the GATA4 and NANOG transcript abundance, suggesting that the proportion of PE and EPI cells were not altered with this treatment. However, providing an FGFR inhibitor to block actions of endogenous FGFs did modify these transcript profiles. Observing an increase in NANOG and a corresponding decrease in GATA4 transcript

952

YANG ET AL.

FIG. 5. Changes in NANOG and GATA4 transcript abundance in blastocysts exposed to an FGFR inhibitor. At Day 7 post-IVF, blastocysts were exposed to 50 ng/ml FGF2, 1 lM PD173074 (FGFR kinase inhibitor), or carrier only (Control) for 24 h (n ¼ 10–12/ treatment group; three replicate studies). Total cellular RNA was extracted and qRTPCR was completed. Data are presented relative to GAPDH mRNA abundance as means 6 SEM. Different superscripts denotes treatment-dependent differences (P , 0.05) in relative mRNA abundance for each transcript.

development [23–25]. FGFR1 also is required for normal embryogenesis. Disruption of FGFR1 leads to an embryonic lethal phenotype characterized by failed gastrulation and mesoderm formation [54, 55]. Collectively, these observations implicate FGF2 and potentially other FGFs as mediators of PE development in bovine embryos. The PE lineage emerges after ICM and TE formation in bovine embryos, and GATA4 is a useful marker for PE determination during early embryogenesis. Embryoderived FGFs mediate PE lineage formation. Providing exogenous FGF2 promotes the establishment of PE outgrowths derived from bovine blastocysts, and this event likely occurs because FGF2 promotes PE proliferation. These outcomes are consistent with cell fate specification studies completed in mice and provide evidence that the molecular events controlling PE specification are similar among mammals that contain marked differences in other aspects of conceptus development. ACKNOWLEDGMENTS The authors thank William Rembert for his assistance with collecting ovaries and the personnel at Central Beef Packing Co. for their generosity in supplying ovaries for the research. The authors also thank Dr. Peter J. Hansen and his laboratory for their cooperation with producing embryos in vitro.

REFERENCES 1. Curran S, Pierson RA, Ginther OJ. Ultrasonographic appearance of the bovine conceptus from days 10 through 20. J Am Vet Med Assoc 1986; 189:1289–1294. 2. Blomberg L, Hashizume K, Viebahn C. Blastocyst elongation, trophoblastic differentiation, and embryonic pattern formation. Reproduction 2008; 135:181–195. 3. Wooding FB. Current topic: the synepitheliochorial placenta of ruminants: binucleate cell fusions and hormone production. Placenta 1992; 13:101– 113. 4. Bazer FW, Spencer TE, Johnson GA, Burghardt RC, Wu G. Comparative aspects of implantation. Reproduction 2009; 138:195–209. 5. Ealy AD, Yang QE. Control of interferon-tau expression during early pregnancy in ruminants. Am J Reprod Immunol 2009; 61:95–106. 6. Thatcher WW, Guzeloglu A, Meikle A, Kamimura S, Bilby T, Kowalski AA, Badinga L, Pershing R, Bartolome J, Santos JE. Regulation of embryo survival in cattle. Reprod Suppl 2003; 61:253–266. 7. Maddox-Hyttel P, Alexopoulos NI, Vajta G, Lewis I, Rogers P, Cann L, Callesen H, Tveden-Nyborg P, Trounson A. Immunohistochemical and ultrastructural characterization of the initial post-hatching development of bovine embryos. Reproduction 2003; 125:607–623. 8. Vejlsted M, Avery B, Schmidt M, Greve T, Alexopoulos N, MaddoxHyttel P. Ultrastructural and immunohistochemical characterization of the bovine epiblast. Biol Reprod 2005; 72:678–686. 9. Hue I, Renard JP, Viebahn C. Brachyury is expressed in gastrulating bovine embryos well ahead of implantation. Dev Genes Evol 2001; 211:157–159. 10. Zernicka-Goetz M, Morris SA, Bruce AW. Making a firm decision:

Downloaded from www.biolreprod.org.

abundance implicates endogenous FGFs as mediators of PE specification. During early mouse embryogenesis, FGFmediated activation of MAPK is essential for Gata4 and Gata6 expression and PE lineage specification [12, 23–26]. Previous work by this laboratory and others [30, 32] detected FGF1, FGF2, and FGF4 mRNA in bovine blastocysts. Based on these observations, it appears that these and potentially other embryo-derived FGFs are sufficient to induce PE specification in cultured bovine embryos. Also, since providing additional FGF2 does not alter the rates of lineage committal in bovine blastocysts, supplementation with FGF2 likely promotes PE proliferation rates after lineage specification occurs. It is interesting to note that FGF2 is not required to sustain PE cultures once they are established. These studies also showed that FGF2 does not affect the incidence of TE outgrowth formation and TE proliferation. TE outgrowths were unaffected by FGF2 supplementation, and CDX2 mRNA concentrations were unchanged in blastocysts provided FGF2. This lack of effect in TE cells is consistent with previous work [28, 32, 41]. However, FGF2 and other FGFs affect TE in other ways. Over the past few years, FGF2 was found to stimulate IFNT expression and increase TE migratory/chemotactic activity [39, 51]. FGF2 was chosen as the FGF isoform to study for several reasons. It is expressed by both the uterus and pre- and periattachment conceptus during early pregnancy in cattle and sheep, and ample FGF2 exists in the uterine secretory milieu as the endoderm lineage emerges in these species [28, 29, 32]. Because this uterine source of FGF2 is lacking in cultured embryos, the initial experiments were developed to examine whether an exogenous source of this factor is important for normal embryogenesis. FGF2 was chosen because recombinant forms of bovine FGF2 are available commercially whereas recombinant forms of many other FGFs, including FGF4, are not available commercially. There were notable differences in FGFR expression profiles between PE and TE. Four genes encode tyrosine kinase receptors that recognize FGFs (abbreviated FGFR1–4). Also, numerous posttranscriptional modifications, including alternative splicing events, generate several FGFR isotypes for FGFR1–3 but not FGFR4 [52, 53]. Selective removal of sequences within the third immunoglobulin-like domain generates the primary alternatively spliced variants studied herein (b and c isotypes). The FGFRs detected in PE included R1b and R1c, both of which are high-affinity partners for FGF2 [52, 53]. The TE contained primarily FGFR2 transcripts, and specifically R2b. Both FGFR1 and R2 are linked with events of early embryogenesis in mice. FGFR2 is best known for its role in maintaining a TE stem cell population in mouse embryos, but it also is required for normal PE formation and

FGF2 SUPPORTS ENDODERM DEVELOPMENT IN BLASTOCYSTS

11. 12. 13. 14. 15.

16. 17. 18.

20.

21.

22.

23. 24.

25. 26. 27. 28.

29.

30. 31. 32.

33. 34.

35.

36. 37. 38. 39. 40.

41. 42. 43.

44.

45.

46.

47. 48.

49. 50. 51. 52. 53. 54. 55.

ment and regulate interferon-tau expression from trophectoderm. Reproduction 2009; 137:259–269. Jousan FD, Hansen PJ. Insulin-like growth factor-I as a survival factor for the bovine preimplantation embryo exposed to heat shock. Biol Reprod 2004; 71:1665–1670. Loureiro B, Bonilla L, Block J, Fear JM, Bonilla AQ, Hansen PJ. Colonystimulating factor 2 (CSF-2) improves development and posttransfer survival of bovine embryos produced in vitro. Endocrinology 2009; 150:5046–5054. Mohammadi M, Froum S, Hamby JM, Schroeder MC, Panek RL, Lu GH, Eliseenkova AV, Green D, Schlessinger J, Hubbard SR. Crystal structure of an angiogenesis inhibitor bound to the FGF receptor tyrosine kinase domain. EMBO J 1998; 17:5896–5904. Turner N, Grose R. Fibroblast growth factor signaling: from development to cancer. Nat Rev Cancer 2010; 10:116–129. Rodina TM, Cooke FN, Hansen PJ, Ealy AD. Oxygen tension and medium type actions on blastocyst development and interferon-tau secretion in cattle. Anim Reprod Sci 2009; 111:173–188. Purcell SH, Cantlon JD, Wright CD, Henkes LE, Seidel GE Jr, Anthony RV. The involvement of proline-rich 15 in early conceptus development in sheep. Biol Reprod 2009; 81:1112–1121. Yang QE, Johnson SE, Ealy AD. Protein kinase C delta mediates fibroblast growth factor-2-induced interferon-tau expression in bovine trophoblast. Biol Reprod 2011; 84:933–943. Talbot NC, Caperna TJ, Edwards JL, Garrett W, Wells KD, Ealy AD. Bovine blastocyst-derived trophectoderm and endoderm cell cultures: interferon tau and transferrin expression as respective in vitro markers. Biol Reprod 2000; 62:235–247. Fields SD, Hansen PJ, Ealy AD. Fibroblast growth factor requirements for in vitro development of bovine embryos. Theriogenology 2011; 75:1466– 1475. Kurosaka S, Eckardt S, Ealy AD, McLaughlin KJ. Regulation of blastocyst stage gene expression and outgrowth interferon tau activity of somatic cell clone aggregates. Cloning Stem Cells 2007; 9:630–641. Talbot NC, Powell AM, Camp M, Ealy AD. Establishment of a bovine blastocyst-derived cell line collection for the comparative analysis of embryos created in vivo and by in vitro fertilization, somatic cell nuclear transfer, or parthenogenetic activation. In Vitro Cell Dev Biol Anim 2007; 43:59–71. Talbot NC, Powell AM, Ocon OM, Caperna TJ, Camp M, Garrett WM, Ealy AD. Comparison of the interferon-tau expression from primary trophectoderm outgrowths derived from IVP, NT, and parthenogenote bovine blastocysts. Mol Reprod Dev 2008; 75:299–308. Babaie Y, Herwig R, Greber B, Brink TC, Wruck W, Groth D, Lehrach H, Burdon T, Adjaye J. Analysis of Oct4-dependent transcriptional networks regulating self-renewal and pluripotency in human embryonic stem cells. Stem Cells 2007; 25:500–510. Strumpf D, Mao CA, Yamanaka Y, Ralston A, Chawengsaksophak K, Beck F, Rossant J. Cdx2 is required for correct cell fate specification and differentiation of trophectoderm in the mouse blastocyst. Development 2005; 132:2093–2102. Ezashi T, Ghosh D, Roberts RM. Repression of Ets-2-induced transactivation of the tau interferon promoter by Oct-4. Mol Cell Biol 2001; 21:7883–7891. He S, Pant D, Schiffmacher A, Bischoff S, Melican D, Gavin W, Keefer C. Developmental expression of pluripotency determining factors in caprine embryos: novel pattern of NANOG protein localization in the nucleolus. Mol Reprod Dev 2006; 73:1512–1522. Gasperowicz M, Natale DR. Establishing three blastocyst lineages—then what? Biol Reprod 2011; 84:621–630. Rossant J. Stem cells and early lineage development. Cell 2008; 132:527– 531. Yang QE, Giassetti MI, Ealy AD. Fibroblast growth factors activate mitogen-activated protein kinase pathways to promote migration in ovine trophoblast cells. Reproduction 2011; 141:707–714. Bottcher RT, Niehrs C. Fibroblast growth factor signaling during early vertebrate development. Endocr Rev 2005; 26:63–77. Powers CJ, McLeskey SW, Wellstein A. Fibroblast growth factors, their receptors and signaling. Endocr Relat Cancer 2000; 7:165–197. Yamaguchi TP, Harpal K, Henkemeyer M, Rossant J. Fgfr-1 is required for embryonic growth and mesodermal patterning during mouse gastrulation. Genes Dev 1994; 8:3032–3044. Deng CX, Wynshaw-Boris A, Shen MM, Daugherty C, Ornitz DM, Leder P. Murine FGFR-1 is required for early postimplantation growth and axial organization. Genes Dev 1994; 8:3045–3057.

Downloaded from www.biolreprod.org.

19.

multifaceted regulation of cell fate in the early mouse embryo. Nat Rev Genet 2009; 10:467–477. Vejlsted M, Avery B, Schmidt M, Greve T, Alexopoulos N, MaddoxHyttel P. Ultrastructural and immunohistochemical characterization of the bovine epiblast. Biol Reprod 2005; 72:678–686. Chazaud C, Yamanaka Y, Pawson T, Rossant J. Early lineage segregation between epiblast and primitive endoderm in mouse blastocysts through the Grb2-MAPK pathway. Dev Cell 2006; 10:615–624. Li L, Arman E, Ekblom P, Edgar D, Murray P, Lonai P. Distinct GATA6and laminin-dependent mechanisms regulate endodermal and ectodermal embryonic stem cell fates. Development 2004; 131:5277–5286. Cai KQ, Capo-Chichi CD, Rula ME, Yang DH, Xu XX. Dynamic GATA6 expression in primitive endoderm formation and maturation in early mouse embryogenesis. Dev Dyn 2008; 237:2820–2829. Soudais C, Bielinska M, Heikinheimo M, MacArthur CA, Narita N, Saffitz JE, Simon MC, Leiden JM, Wilson DB. Targeted mutagenesis of the transcription factor GATA-4 gene in mouse embryonic stem cells disrupts visceral endoderm differentiation in vitro. Development 1995; 121:3877–3888. Morrisey EE, Tang Z, Sigrist K, Lu MM, Jiang F, Ip HS, Parmacek MS. GATA6 regulates HNF4 and is required for differentiation of visceral endoderm in the mouse embryo. Genes Dev 1998; 12:3579–3590. Fujikura J, Yamato E, Yonemura S, Hosoda K, Masui S, Nakao K, Miyazaki JJ, Niwa H. Differentiation of embryonic stem cells is induced by GATA factors. Genes Dev 2002; 16:784–789. Shimosato D, Shiki M, Niwa H. Extra-embryonic endoderm cells derived from ES cells induced by GATA factors acquire the character of XEN cells. BMC Dev Biol 2007; 7:80. Kuijk EW, Du Puy L, Van Tol HT, Oei CH, Haagsman HP, Colenbrander B, Roelen BA. Differences in early lineage segregation between mammals. Dev Dyn 2008; 237:918–927. Talbot NC, Caperna TJ, Edwards JL, Garrett W, Wells KD, Ealy AD. Bovine blastocyst-derived trophectoderm and endoderm cell cultures: interferon tau and transferrin expression as respective in vitro markers. Biol Reprod 2000; 62:235–247. Talbot NC, Caperna TJ, Powell AM, Ealy AD, Blomberg LA, Garrett WM. Isolation and characterization of a bovine visceral endoderm cell line derived from a parthenogenetic blastocyst. In Vitro Cell Dev Biol Anim 2005; 41:130–141. Cheng AM, Saxton TM, Sakai R, Kulkarni S, Mbamalu G, Vogel W, Tortorice CG, Cardiff RD, Cross JC, Muller WJ, Pawson T. Mammalian Grb2 regulates multiple steps in embryonic development and malignant transformation. Cell 1998; 95:793–803. Feldman B, Poueymirou W, Papaioannou VE, DeChiara TM, Goldfarb M. Requirement of FGF-4 for postimplantation mouse development. Science 1995; 267:246–249. Arman E, Haffner-Krausz R, Chen Y, Heath JK, Lonai P. Targeted disruption of fibroblast growth factor (FGF) receptor 2 suggests a role for FGF signaling in pregastrulation mammalian development. Proc Natl Acad Sci U S A 1998; 95:5082–5087. Goldin SN, Papaioannou VE. Paracrine action of FGF4 during periimplantation development maintains trophectoderm and primitive endoderm. Genesis 2003; 36:40–47. Chen Y, Li X, Eswarakumar VP, Seger R, Lonai P. Fibroblast growth factor (FGF) signaling through PI 3-kinase and Akt/PKB is required for embryoid body differentiation. Oncogene 2000; 19:3750–3756. Yamanaka Y, Lanner F, Rossant J. FGF signal-dependent segregation of primitive endoderm and epiblast in the mouse blastocyst. Development 2010; 137:715–724. Michael DD, Alvarez IM, Ocon OM, Powell AM, Talbot NC, Johnson SE, Ealy AD. Fibroblast growth factor-2 is expressed by the bovine uterus and stimulates interferon-tau production in bovine trophectoderm. Endocrinology 2006; 147:3571–3579. Ocon-Grove OM, Cooke FN, Alvarez IM, Johnson SE, Ott TL, Ealy AD. Ovine endometrial expression of fibroblast growth factor (FGF) 2 and conceptus expression of FGF receptors during early pregnancy. Domest Anim Endocrinol 2008; 34:135–145. Daniels R, Hall V, Trounson AO. Analysis of gene transcription in bovine nuclear transfer embryos reconstructed with granulosa cell nuclei. Biol Reprod 2000; 63:1034–1040. Daniels R, Hall VJ, French AJ, Korfiatis NA, Trounson AO. Comparison of gene transcription in cloned bovine embryos produced by different nuclear transfer techniques. Mol Reprod Dev 2001; 60:281–288. Cooke FN, Pennington KA, Yang Q, Ealy AD. Several fibroblast growth factors are expressed during pre-attachment bovine conceptus develop-

953