Bovine ovaries were recovered at a local abattoir (IT 2270M CE; Inalca S.p.A., Ospedaletto Lodigiano, LO, Italy) from Holstein cows subjected to routine veterinary inspection and according to the specific health requirements. No animals were raised, nor euthanized for conducting these analyses. Instead, ovaries were collected from animals culled for human meat consumption, in fulfillment with the 3Rs principle: Replacement, Reduction, and Refinement of animals used for experimentation established by EU Directive 2010/63/EU and subsequent amendments. The leftover tissues were disposed according to the regulation (EC) No 1069/2009 of the European Parliament and of the council of 21 October 2009.

Ovaries with more than 10 medium antral follicles visible on the surface were collected in sterile saline (NaCl, 9 g/L), supplemented with penicillin 100 U/mL and streptomycin 0.1 mg/mL (pen/strep) at 26-28°C and transported to the laboratory within 3 hrs and processed as previously described with minor modifications (Modina et al., 2007). The following collection, selection and culture procedures were conducted as previously described (Luciano et al., 2013). Briefly, medium antral follicles (diameter 2–6 mm) were punctured with a 16-gauge needle connected to an aspiration pump (COOK-IVF, Brisbane, QLD, Australia) to collect the follicular fluid containing the cumulus-enclosed oocytes (CEOs) in a 20 mL tube prefilled with TCM-199 supplemented with HEPES 20 mM, 1790 U/l heparin, and 0.4% of bovine serum albumin (BSA) (H-M199). The sediment, composed of CEOs and other follicular cells, was examined under a stereomicroscope to retrieve CEOs medium brown in color, with five or more complete layers of cumulus cells and oocytes with finely granulated, homogenous ooplasm. The selected CEOs were washed twice in H-M199 and in vitro matured in groups of 25–30 for 24 hrs in 500 µL of TCM-199 modified to contain 0.68 mM L-glutamine, 25 mM NaHCO₃, 0.4% fatty acid-free BSA, 0.2 mM sodium pyruvate, 0.1 mM cysteamine, 50 µg/mL kanamycin, and 0.1 IU/mL recombinant human FSH (Gonal-F, Merck-Serono) in humidified air under 5% CO₂ at 38.5°C. The 4-well dishes containing the IVM medium were pre-equilibrated in the incubator for at least 4 hrs before starting the culture.

At the end of the IVM, the CEOs were moved to the in vitro fertilization (IVF) dish.

X-sorted semen was processed either following the previously described protocol for unsorted semen in use in our laboratory (Lodde et al., 2021), or modified as detailed below.

Briefly, starting from 1–2 hrs before the end of IVM, X-sorted, commercially available, cryopreserved semen of a bull of proven fertility, previously used in AI schemes (Frau et al., 2024), was thawed by submerging the straw(s) in a 35 °C water bath for 1 min and the content transferred in sterile 2 mL tubes. Upon thawing, a 10 µL drop was visually assessed for progressive motility, expressed as the percentage of spermatozoa showing progressive and straight motion on the total spermatozoa visible in a given microscope field (Srivastava et al., 2013). The content of the straw was then gently layered on discontinuous density gradients (DDGs) prepared by mixing Percoll, a commercial PVP-coated colloidal silica solution (Merck Life Science S.r.l., Milano, Italy), and Ca²⁺-free, HEPES-buffered Tyrode Albumin Lactate Pyruvate (Ca²⁺-free TALP) in different proportions and centrifuged, as detailed in Table 1. When different protocols for semen preparation were tested, the content of multiple straws was pulled after the initial motility check, gently mixed, and equally divided into aliquots that underwent parallel downstream procedures.

Upon DDG centrifugation, the separated fraction of spermatozoa was recovered in a fixed volume of 100 µL - corresponding to the pellet - and was washed twice by centrifuging in Ca²⁺-free TALP. Wash volume, centrifugation speed, and time were adjusted according to the size of the DDG, as detailed in Table 1. A 10 µL drop of spermatozoa suspension was used to visually assess the progressive motility, as described above, by 3 operators, 2 of whom were blinded. Spermatozoa concentration was calculated using a Neubauer chamber after diluting 1:4 in HCl 1M and adjusted to 220.000 spermatozoa/mL. The appropriate volume of spermatozoa suspension was then added in 100 µL drops of TALP supplemented with 0.6% (w/v) BSA fatty acid free, 10 µg/mL heparin, 20 µM penicillamine, 1 µM epinephrine, and 100 µM hypotaurine (TALP IVF), containing the CEOs. Spermatozoa and CEOs were co-incubated at 38.5°C under 5% CO₂ in humidified air for 8–18 hrs, according to the experimental design. TALP IVF drops, covered with paraffin oil (Vitrolife, Göteborg, Sweden), were pre-equilibrated in the incubator starting from the night before the beginning of IVF. At the end of the co-incubation, the presumptive zygotes were either fixed to assess the fertilization rate or cultured for 8 days as detailed below.

At the end of the IVF culture, residual cumulus cells and spermatozoa were removed by vortexing for 1–2 min at 35 Htz in 500 μL of synthetic oviduct fluid (SOF) buffered with 10 mM of HEPES and 5 mM of NaHCO₃ (SOF wash), in a 5 mL round bottom tube. The inner wall of the tube was washed with an additional 500 μL SOF wash and the presumptive zygotes were allowed to sink for 30–60 sec. The medium was recovered from the bottom of the tube and examined under a stereomicroscope to retrieve the presumptive zygotes that, after 2 washes in SOF wash, were transferred in SOF buffered with 25 mM of NaHCO₃ and supplemented with MEM essential and nonessential amino acids, 0.72 mM sodium pyruvate, 2.74 mM myo-inositol, 0.34 mM sodium citrate, and 5% calf serum (SOF-IVC). Incubation was performed at 38.5 °C with a humidified gas mixture composed of 5% CO₂, 5% O₂, and 90% N₂ for 8 days. The 4-well dishes, containing 500 µL/well of SOF-IVC and 2 mL of water in the space between the wells, were pre-equilibrated in the incubator for at least 8 hrs before the start of IVC.

At the end of the culture period, the formed blastocysts were counted using a stereomicroscope and the blastocyst rate was calculated on the total number of initial structures. Furthermore, the blastocysts were classified as early, expanded, and hatching/hatched based on the observation of an initial or well-developed blastocoel (early or expanded morphology, respectively), and emergence - initial or complete - from the zona pellucida (hatching/hatched morphology).

All the structures obtained at the end of the culture were fixed in a mixture of 60% methanol and 40% Dulbecco’s phosphate buffered saline at 4°C, stained with 0.5 mg/mL propidium iodide or 1 µg/mL 4’,6-diamidino-2-phenylindole (DAPI) and examined under a fluorescence microscope to confirm the morphological observations.

Some embryos were fixed, stained, and imaged at the end of the IVF culture to monitor the formation of pronuclei.

Statistical analyses were conducted using GraphPad Prism version 8.4.3 (GraphPad Software, Boston, Massachusetts USA, www.graphpad.com). Continuous data were first assessed for normal distribution by the Shapiro-Wilk test and then analyzed either by t-test if normally distributed or using a non-parametric test (Wilcoxon test) if the distribution was not normal. Categorical data are presented as percentages and analyzed by Fisher’s exact test. P<0.05 was considered statistically significant.

Three series of experiments were conducted that led us to progressively refine the IVF procedure with X-sorted spermatozoa. All the experiments were performed at least 3 times. A graphical representation of the experimental plan is provided in Figure 1.

In Experiment 1 (Figure 1A), three DDGs (namely DDG-A, DDG-B, and DDG-C) were compared to test the efficiency of retrieval of live spermatozoa. The biological endpoints of this set of experiments were spermatozoa motility and concentration. Statistical significance was analyzed by two-tailed paired t-test (motility) or two-tailed Wilcoxon matched-pairs signed rank test (concentration), according to the results of the normality test.

In Experiment 2 (Figure 1B), DDG-A and DDG-C were compared to test the fertilization ability. The biological endpoints of this set of experiment were PN formation and data were analyzed by two-tailed Fisher’s exact test.

In Experiment 3 (Figure 1C), different co-incubation times of DDG-C processed spermatozoa and COCs were tested to assess the effect on developmental competence. The biological endpoints of this set of experiment were blastocyst rate and morphology and data were analyzed by two-tailed Fisher’s exact test.

In the first series of experiments we tested if decreasing the time required for the procedures and modifying the DDG composition might affect the retrieval of viable spermatozoa. This hypothesis was tested by comparing the motility and the concentration of X-sorted spermatozoa obtained using 3 different DDG protocols. Specifically, DDG-A, representing the standard for unsorted semen in our lab, was composed of 2 mL of 45% Percoll layered on top of 2 mL of 90% Percoll. With DDG-B, the volume was reduced to a quarter of DDG-A and the time of processing was decreased, but the composition (45/90) remained unchanged. DDG-C was also small and required a shorter processing time, but compared with the previous two it was modified to have a higher density on top, intermediate between 45 and 90% (67.5%), and a lower density at the bottom, intermediate between 67.5 and 90% (78.75%).

After centrifugation and wash, the obtained sperm suspension with small-volume DDGs (B and C) had a cloudy appearance, likely due to persistent Percoll residues. The opacity made the counting difficult in the Neubauer chamber and posed a risk of Percoll contamination of the IVF drop. An additional wash was therefore included to obtain a clear sperm suspension. The additional wash was performed also with DDG-A to avoid the introduction of confounding factors.

Since DDG-B underperformed in preliminary trials, giving a remarkably low motility (10%) and a low spermatozoa concentration (950.000 sperm/mL), it was immediately discarded and not carried over in the following experiments. When comparing the remaining two gradients, spermatozoa concentration was not significantly changed (P = 0.375, Wilcoxon test, matched pairs, N = 4), but the motility was superior when using DDG-C compared to DDG-A (P = 0.0462, two-tailed paired t-test, N = 4, Figure 2).

The fertilization rate of X-sorted spermatozoa processed with DDG-A and DDG-C was investigated on 65 presumptive zygotes fixed at the end of the IVF culture. Notably, 87% of the oocytes co-incubated with spermatozoa prepared with DDG-C showed 2 pronuclei (2PN) and were therefore considered successfully fertilized. The remaining 13% were either metaphase I (MI) and therefore considered not matured oocytes, metaphase II (MII) and therefore considered unfertilized oocytes, or polyspermic zygotes (Figures 3A, B). The pronuclei rate with DDG-A was instead significantly lower (P<0.0001, Fisher’s exact test, Two-tailed), only 26%, with the majority of the oocytes failing to be fertilized and being at the MII stage (Figures 2A, B).

Despite a considerably high 2PN rate obtained with DDG-C, the blastocyst rate was rather low. Since premature capacitation has been reported for X-sorted semen (Maxwell and Johnson, 1999; Umezu et al., 2017), we reasoned that 18 hrs co-incubation might be too long and that formed zygotes might benefit of an earlier switch to embryo culture conditions. This hypothesis was tested by comparing the blastocyst rate obtained by switching the presumptive zygotes from IVF to IVC conditions after 10 or 18 hrs and further culture for 8 days. We observed that reducing the co-incubation time to 10 hrs significantly improved the overall blastocyst rate (Figure 4A). The morphological observation did not indicate differences in the quality of the blastocyst, as the rate of early, expanded, and hatched blastocyst was not different (Figure 4B).

For practical reasons, a pilot experiment was conducted to further reduce the co-incubation time to 8 hrs and the IVM time to 22 hrs. However, an additional beneficial effect was not observed (not shown).