An air-liquid interphase approach for modeling the early embryo-maternal contact zone

We developed an air-liquid interphase culture procedure for mammalian oviduct epithelial cells leading to the formation of functional epithelial tissues, which generate oviduct fluid surrogates. These in vitro oviduct epithelia can be co-cultured with living zygotes and enable embryonic development up to the blastocyst stage without addition of embryo culture medium. The described strategy is broadly applicable to analyze early embryo-maternal interactions under standardized in vitro conditions.

SCIeNtIFIC REPORtS | 7:42298 | DOI: 10.1038/srep42298 To prove reproducibility of OFS formation, we tested its osmolality, its protein pattern in SDS-PAGE, and secretion of the marker protein oviductin (OVGP1) in five biological replicates per species. The OFS of all three species exhibited an in vivo-like osmolality (ALI-MOEC: 341 ± 4.4 mOs/kg, ALI-POEC: 348 ± 8.3 mOs/kg, ALI-BOEC: 340 ± 3.5 mOs/kg) with low variability. SDS-PAGE showed a reproducible species specific protein pattern (Fig. 4A). Oviductin is a marker-mucin almost exclusively expressed in oviduct epithelium. It is involved in zona pellucida hardening as well as sperm-oocyte interaction. Under standard in vitro conditions its expression was rapidly down-regulated 4,5,12 . Under ALI conditions, however, it was abundant in OFS from all three species and secreted in different glycosylated (and therefore potentially functional) forms as shown by N-glycosidase F (PNGaseF; cleaves complete N-glycans) and sialidase (releases sialic acids) treatment (Fig. 4B,C).
To characterize the protein content of the OFS we performed proteome analysis by mass spectrometry. Murine, bovine and porcine OFS contained proteins previously reported to be highly abundant in oviductal fluid in vivo 13 (table of most abundant OFS proteins: Suppl. Table 1; complete proteome data: Suppl. Table 2). Along with OVGP1 many other proteins, which are proven to influence fertilization and/or embryonic development (e.g. osteopontin, components of the plasminogen/plasmin system, heat shock proteins, lactoferrin) are present in the oviductal fluid surrogate (OFS) formed by the ALI-OEC systems. In total 1756, 2979 and 3094 proteins were identified in murine, porcine and bovine OFS, respectively. The only fully available, comprehensive proteome study of in vivo oviductal fluid was conducted in sheep and was very recently published 13 . Taking this data set as reference ~97% (murine: 96.9%; porcine: 96.3%; bovine: 96.8%) of the proteins detected in the OFS are also abundant in the sheep oviductal fluid in vivo.

Figure 2. Oviduct epithelial cells of different mammalian species grown at the air-liquid interphase (ALI).
Immunodetection of epithelial markers in murine, porcine and bovine oviduct tissue (from left to right; upper pictures) and respective ALI-OEC (lower pictures) after 21 d (ALI-MOEC and -POEC) or 28 d (ALI-BOEC) of culture. Red: nuclei; green: beta-catenin (cell-cell adhesions); blue: acetylated tubulin (cilia); bar = 10 μ m. To finally prove the functional integrity of the cell culture system, we tested the capability of the in vitro formed epithelial tissue to support embryo development in long-term co-culture. Zygotes were either produced in vitro (IVM/IVF; bovine) or in vivo (mouse and pig). After IVF or flushing from the in vivo oviducts, potential zygotes were briefly washed in PBS and placed in the OFS on top of the ALI-OEC in groups of 10-30 (Suppl. Video 1). Two experimental setups were conducted in each species. First, co-cultures were terminated on d 2 to determine cleavage rate. Second, we prolonged embryo co-culture to test whether the milieu of the OFS provided by the ALI-OEC supports further embryonic development (mouse: 4.5 d, pig 7 d and cattle 8 d). In all three species cleavage as well as blastocyst formation could be observed without supplementation of any embryo culture medium ( Table 1, Fig. 5).
In sum, we developed a culture procedure for the formation of an in vivo-like oviduct tissue substitute from primary oviduct epithelial cells. We demonstrated that the formed tissue is fully functional in terms of morphological differentiation (polarization, columnar shape, ciliary activity) and in terms of oviductal fluid surrogate formation supporting embryo development in vitro without additional embryo culture medium supply.
The blastocyst rates in co-culture could not yet match the outcome of optimized standard IVEP procedures. Therefore the model could be further improved by a) simulation of the hormonal changes taking place during the periconceptional period and b) development of a sequential culture system using oviductal as well as uterine epithelial cells. This might increase the efficiency of the system both quantitatively and qualitatively. Further experiments including in vivo embryo transfer then have to be conducted to assess the quality of ALI-produced blastocysts.
The presented culture strategy is broadly applicable to analyze early embryo-maternal interactions under standardized in vitro conditions. It can prospectively serve as a tool to advance IVEP procedures by analyzing specific components of the dynamic oviduct milieu regarding their impact on the early embryo. This might facilitate the development of new strategies (e.g. media supplements) for livestock as well as for human ARTs.  Media used in this culture procedure are modifications of a protocol reported for mouse tracheal cells [10]. The basic medium consisted of DMEM/Ham's F12, 15 mM HEPES, 100 U/ml penicillin, 100 μ g/ml streptomycin and 0.25 μ g/ml amphotericin B. The proliferative medium (M1) for d 0-d 7 was basic medium supplemented with 5% FBS, 10 μ g/ml insulin, 5 μ g/ml transferrin, 0.1 μ g/ml cholera toxin, 25 ng/ml epidermal growth factor, 30 μ g/ml bovine pituitary extract. Differentiation medium (M2, from d 7 onwards) for ALI-MOEC and -POEC (M2a) was serum free: basic medium supplemented with 1 mg/ml BSA, 5 μ g/ml insulin, 5 μ g/ml transferrin, 0.025 μ g/ml cholera toxin, 5 ng/ml epidermal growth factor, 30 μ g/ml bovine pituitary extract. For ALI-BOEC the differentiation medium (M2b) consisted of basic medium with 3% FBS and 2% Nuserum (Corning). All culture media were freshly added with 0.05 μ M retinoic acid directly before use.  Animals. The FBN mice strain Fzt:DU 14 was included in this study. Porcine and bovine tissue samples were slaughterhouse by-products and collected in local abattoirs. All animal procedures were done in accordance to national and international guidelines and approved by the institutional Animal Protection Board at FBN.

Isolation and culture of MOEC.
For the isolation of MOEC 6-8 weeks old female mice were killed in oestrus (detected by vaginal cytology). Both oviductal tubes were resected from the reproductive tract, washed in cold basic medium, combined and sliced into small pieces and incubated in basic medium with 0.15% Pronase at 4 °C overnight. The next day the sample was centrifuged at 250 × g for 8 min and the pellet was subsequently digested in 300 μ l of 0.5 mg/ml DNase I for 10 min. After passing through a 100 μ m cell strainer, the dissociated cells were centrifuged and re-suspended in M1 medium. The cell yield from each mouse was approximately 1-3 × 10 5 epithelial cells. A schematic diagram for the insert-supported culture is shown in Fig. 1A. 24-well inserts with a pore size of 1.0 μ m (Merck Millipore) were coated with 100 μ l/insert human placental collagen (1 mg/ml) overnight and washed three times with PBS afterwards. 1-1.5 × 10 5 cells isolated from single mice were seeded per insert. From d 0 to d 6 (proliferation phase), cells were held submerged in medium with 200 μ l of M1 medium in the apical compartment and 1 ml of M1 in the basal compartment. From d 7 onwards (differentiation phase) cells were switched to air-liquid interphase (ALI), with access to M2a medium only in the basal compartment. Cultures were maintained at 37 °C, 5% CO 2 , with medium refreshment twice per week. Apical fluid was removed from the upper compartment during each medium change.

Isolation and culture of POEC and BOEC. Isolation of oviductal cells from large farm animals as pig
and cattle was performed as previously reported by our group 6,11 . Briefly, oviducts were collected from a local slaughterhouse and transported immediately on ice for laboratory processing. After washing in PBS, each oviductal tube was filled up with 1 mg/ml collagenase 1 A and incubated for 1 h at 37 °C. Big epithelial clusters were collected with a cell strainer and then further digested in accutase (Life technologies) for 10 min. Thereafter cells were centrifuged and re-suspended in M1 medium for seeding.
While the culture of POEC was performed as described for MOEC (see above), for BOEC the procedure had to be modified: 1. Cells were maintained on 0.4 μ m-pore-size 24 well inserts (no collagen coating); 2. Medium for the differentiation phase was M2b.

Characterization of oviduct fluid surrogates (OFS). SDS PAGE, Western blot, immunodetection of ovi
Protein identification by LC-MS/MS. The OFS recovered from ALI-MOEC, -POEC and -BOEC (N = 1 per species) were run on a SDS-PAGE gel and cut into 13 equal-sized Coomassie-stained bands. In-gel protein digestion was performed as previously described 16 . Digested samples were re-dissolved in 0.1% TFA and 5% acetonitrile, and peptides were analyzed by a reversed-phase capillary liquid chromatography system (Ultimate 3000 nanoLC system, Thermo Scientific) connected to an Orbitrap Fusion mass spectrometer (Thermo Scientific). LC separation was performed on an in-house packed 75 μ m inner diameter PicoTip column (25 cm) packed with ReproSil-Pur C18AQ particles, 3 μ m, 120 Å (Dr. Maisch). The flow rate was 200 nL/min using a gradient of 3− 30% B in 60 min. Mobile phase solvent A contained 0.1% formic acid in water and mobile-phase solvent B contained 0.1% formic acid in acetonitrile. For MS/MS measurements, FT survey scans were acquired with a resolution of 120.000. The data-dependent acquisition (DDA) mode and monoisotopic precursor ions with charge states 2 and 3 were selected. HCD MS/MS spectra were acquired in the linear ion trap using a quadrupole isolation window of 1.6 Da.
Protein identification was performed using Mascot Distiller (version 2.5.1.0). Processed data were searched against a SwissProt database (version 2014_12; 547,085 sequences). The mass tolerance of precursor and sequence ions was set to 10 ppm and 0.35 Da, respectively. A maximum of two missed cleavages was allowed. Methionine oxidation and the acrylamide modification of cysteine were used as variable modifications. Scaffold (version 2.01; Proteome Software Inc.) was used to validate MS/MS based peptide and protein identifications. Peptide identifications were accepted if their probability was established at > 70%, as specified by the Peptide Prophet algorithm. Protein identifications were accepted if their probability was established at > 90% and if they contained at least two identified tryptic peptides.
Osmolality of OFS. The osmolality of apical fluids collected from ALI-MOEC, -POEC and -BOEC, was measured using the cryoscopic osmometer (Gonotec) according to the manufacturer´s instructions. Measurement was performed with 15 μ l of apical fluid (N = 5 inserts, 3-5 different donor animals per species).
Zygote production. Mice. 20 virgin females of the mouse line Fzt:DU (10-12 weeks old) were randomly selected. Potential zygotes were collected from the oviducts of naturally mated female mice approx. 12 h post conception (vaginal plug).
Pig. For collection of in-vivo derived porcine embryos, in total 10 sows were slaughtered and potential zygotes and 2-cell embryos were recovered approximately 12 h after artificial insemination.
Cattle. Bovine ovaries were collected from a local slaughterhouse and transported in DPBS with 1% penicillin to the laboratory within 3 h after slaughter. Upon arrival, cumulus oocyte complexes (COCs) were recovered using DPBS supplemented with 0.2% BSA. The COCs were matured in Medium 199 supplemented with 5% estrous cow serum, 0.07 IE/ml FSH, 0.03 IE/ml HCG, 1 μ g/ml estradiol, 2 mM L-glutamine and 1% penicillin/ streptomycin at 38.5 °C (5% CO 2 ) for 24 h. After maturation, 10 COCs were transferred to each 100 μ l droplet of fertilization medium. Motile sperm were recovered by swim-up separation of frozen-thawed semen. 3 × 10 4 sperm were added to each fertilization droplet (Tyrode's albumin lactate pyruvate medium) and incubated with the matured COCs for 18 h.
Co-Culture. On d 24 (ALI-MOEC), d 26 (ALI-POEC) or d 33 (ALI-BOEC) of culture potential zygotes of all three species were transferred to the apical side of the respective OEC cultures in groups of 10-30.
As exact evaluation of cleavage is difficult on the inserts, three experimental trials were performed: in trial 1 and 2 the experiment was terminated on d 2 of co-culture to determine the percentage of cleaved embryos. In trial 3 embryos were further cultured until d 4.5 (mice), 7 (pig) or 8 (cattle) to evaluate if embryonic development was further supported by ALI-OEC. Blastocysts were fixed in buffered formol saline (4%) and stained with Hoechst 33258.