98 In vitro fertilisation with sperm prebound to oviduct glycans immobilised on a coverslip

In vitro fertilisation (IVF) bypasses sperm selection and transport through the oviduct. Oviduct epithelial cells have specific glycan motifs that bind sperm and may select the fittest sperm that yield the highest frequency of normal development. The aim of our study was to develop an IVF system using immobilised oviduct glycans that replicates sperm-oviduct adhesion in vivo. Porcine cumulus-oocytes complexes (COCs) were cultured with sperm bound to a coverslip coated with 3-O-sulfated Lewis X trisaccharide (suLe^X). Briefly, biotinylated suLe^X was coupled to a streptavidin-coated coverslip (PolyAn). Washed ejaculated sperm were allowed to bind immobilised suLe^X in droplets. Free-swimming sperm were removed by replacing the medium. Mature COCs were transferred to the droplets and cultured under conventional IVF conditions. First, sperm binding to suLe^X was assessed, compared to binding immobilised fibronectin (positive control), N-acetyl-d-lactosamine (LacNAc, control glycan) and a non-coated coverslip (control). After 30 min of culture with 1 × 10⁶ spermatozoa mL⁻¹, the number of sperm bound to the coverslip was counted (mean of five random areas). The sperm binding over time (30, 60, and 90 min) and with increasing sperm concentrations (0.25, 1, and 5 × 10⁶ spermatozoa mL⁻¹) was also tested. Last, sperm ability to fertilise oocytes was evaluated. Following a 30-min incubation of 0.25 × 10⁶ spermatozoa mL⁻¹, bound sperm were cultured with COCs, dead COCs (heat-stressed), and mTBM medium only (control). After 5 h, the remaining bound sperm were counted and the COCs were cultured for 7 h in PZM-5. Pronuclear formation was assessed with Hoechst 33342. Conventional IVF (5 h with 1 × 10⁵ spermatozoa mL⁻¹) was performed concurrently. Three independent replicates were performed of each experiment with duplicate droplets. Data were analysed by two-way ANOVA followed by Tukey’s post hoc test. Sperm bound to suLe^X and fibronectin-coated coverslips, but not to LacNAc or control (5468.6 ± 897.1, 4205.5 ± 623.5, 56.3 ± 12.0, and 76.1 ± 22.9 sperm mm⁻², respectively). No difference was observed in sperm binding to suLe^X over 30 to 90 min. There was a proportional increase of bound sperm with increasing sperm concentrations (1241.5 ± 127.5, 5904.4 ± 361.1, and 22 580.7 ± 2540.6 sperm mm⁻² for 0.25, 1, and 5 × 10⁶ spermatozoa mL⁻¹; P < 0.05). After COC culture, there were fewer bound sperm than when medium was added (424.4 ± 73.7 and 940.2 ± 85.8 sperm mm⁻²; P < 0.05), suggesting sperm release by COCs. Yet, dead COCs also decreased bound sperm (413.4 ± 44.4 sperm mm⁻²; P < 0.05). This may be due to sperm interaction with hyaluronan-rich cumulus, which are together removed with medium replacement. Last, using the glycan-coverslip produced similar rates of fertilised oocytes (75.1 ± 2.2%) to control IVF (72.0 ± 3.2%), although the results are not experimentally comparable, as the sperm concentration was not adjusted to have similar free-swimming sperm. This study is a proof of principle that glycan-coverslips can be used to retain sperm that fertilise porcine COCs. Future studies will optimise this method, which could improve IVF success by mimicking sperm selection by the oviduct.

This research was supported by NIH RO1HD095941.