Controlled spermatozoa–oocyte interaction improves embryo quality in sheep

The current protocols of in vitro fertilization and culture in sheep rely on paradigms established more than 25 years ago, where Metaphase II oocytes are co-incubated with capacitated spermatozoa overnight. While this approach maximizes the number of fertilized oocytes, on the other side it exposes them to high concentration of reactive oxygen species (ROS) generated by active and degenerating spermatozoa, and positively correlates with polyspermy. Here we set up to precisely define the time frame during which spermatozoa effectively penetrates and fertilizes the oocyte, in order to drastically reduce spermatozoa-oocyte interaction. To do that, in vitro matured sheep oocytes co-incubated with spermatozoa in IVF medium were sampled every 30 min (start of incubation time 0) to verify the presence of a fertilizing spermatozoon. Having defined the fertilization time frame (4 h, data from 105 oocytes), we next compared the standard IVF procedures overnight (about 16 h spermatozoa/oocyte exposure, group o/nIVF) with a short one (4 h, group shIVF). A lower polyspermic fertilization (> 2PN) was detected in shIVF (6.5%) compared to o/nIVF (17.8%), P < 0.05. The o/nIVF group resulted in a significantly lower 2-cell stage embryos, than shIVF [34.6% (81/234) vs 50.6% (122/241) respectively, P < 0.001]. Likewise, the development to blastocyst stage confirmed a better quality [29% (70/241) vs 23.5% (55/234), shIVF vs o/nIVF respectively] and an increased Total Cell Number (TCN) in shIVF embryos, compared with o/n ones. The data on ROS have confirmed that its generation is IVF time-dependent, with high levels in the o/nIVF group. Overall, the data suggest that a shorter oocyte-spermatozoa incubation results in an improved embryo production and a better embryo quality, very likely as a consequence of a shorter exposure to the free oxygen radicals and the ensuing oxidative stress imposed by overnight culture.


Scientific Reports
| (2021) 11:22629 | https://doi.org/10.1038/s41598-021-02000-z www.nature.com/scientificreports/ to the acquisition of sperm fertilizing capability, as hyperactivation and phosphorylation involved in sperm capacitation 13 . However, high ROS free radicals react with numerous biological molecules, such as lipids, proteins and DNA, and can trigger rapid chain reactions causing irreversible cell and DNA damage. Thus, the particular genome conformation in early-stage embryos, very open and highly accessible, might render them vulnerable to ROS, thus compromising further development.
Here, having confirmed and quantified the high levels of ROS produced in IVF medium and finely tracked the early fertilization event, we have observed that restricting gamete co-incubation to only 4 h resulted in a significant amelioration in embryo production outcomes, both quantitatively and qualitatively.

Results
Detection of spermatozoa-oocyte interaction. Matured oocytes were co-incubated with capacitated spermatozoa in IVF medium, and aliquots of oocytes (batches of 7 oocytes sampled every 30 min) were monitored for spermatozoa-oocyte interaction ( Fig. 1), previous enzymatic/mechanical cumulus cells removal. The use of Propidium Iodide (PI) allowed us to localize the spermatozoa adhered to, or crossing the ZP, by scanning the entire Z axes in the confocal microscope of individual oocytes. The first spermatozoa bound to ZP were observed 90 min after IVF (54.5%, 6/11) (Fig. 2a), with almost half of the oocyte having ZP-bound spermatozoa by 120 min (45.5%, 5/11) (Fig. 2b). At 180 min, spermatozoa were found in the perivitelline space in the 55.6% of oocytes (5/9) (Fig. 2c). Oocytes monitored 240 min past IVF displayed already a single spermatozoon (66.7%, 4/6), or in fewer cases 2, within the oocyte's cytoplasm (Fig. 2d). This observation, consistently made in 15 replicates, let to us conclude that 4 h of gamete interaction suffices for successful fertilization (Fig. 1). Based on these findings, we next compared fertilization, polyspermy and embryo development potential between short-IVF (shIVF, 4 h co-incubation) and overnight-IVF (o/nIVF, approximately 16 h of co-incubation).
Reduced gamete interaction ameliorates embryo development. IVF Fig. 3a). The beneficial effects of reduced gamete interaction on embryo development were also qualitative. Figure 1. Experimental design. Fertilization window has been detected by localization of sperm within 4 h from sperm-oocyte coincubation. Then, polyspermy, embryo development and blastocysts quality were compared between short and overnight IVF. ROS production from medium only and spermatozoa has been evaluated at t0, after short and overnight incubation. Further details are in the text.  (Fig. 3b,c).
To get a functional glimpse on the obtained blastocysts, the expression level of a panel of genes ( Fig. 4) correlated to embryo development were quantified by reverse transcription quantitative real-time PCR (RT-qPCR). To justify the higher number of cells recorded in the shIVF, we decided to detect the expression of the Proliferating Cells Nuclear Antigen (PCNA) and an antioxidant-related gene, superoxide dismutase (SOD1). Although the two groups (n = 3) did not show significant variation in the relative expression levels of mRNA transcripts for all the genes tested, PCNA and SOD1 showed a decrease in o/nIVF.      To provide an index of oxidation during IVF, ROS production was evaluated in the IVF medium at the beginning (t = 0), and after shIVF time (t = sh) and o/nIVF incubation (t = o/n). To have an index of ROS production from spermatozoa only, we performed the same analysis in PBS.

Discussion
The current practice in laboratory production of sheep embryos foresees an overnight incubation of mature metaphase II oocytes with capacitated spermatozoa. These settings though comply more with logistic-personnel planning, rather than real biological needs. It has been previously demonstrated that a prolonged permanence of thousands of spermatozoa with the oocytes increases polyspermic fertilization 8,14 . Moreover, the presence of thousands of decaying spermatozoa in a small volume of medium exposes the oocytes to high concentration of lytic, acrosomal enzymes and ROS, a situation that drifts considerably from the physiological situation taking place in the oviduct.
In our work, we have first finely monitored the dynamics of oocyte-spermatozoa interaction leading to fertilization, and have established that 4 h is sufficient to achieve an acceptable fertilization rate (77%). This information prompted us to compare embryo production outcomes using a controlled oocyte-spermatozoa interaction, with the standard overnight procedure. The beneficial effects were evident since the first cleavage, significantly higher in the shIVF group, because an important proportion of embryos of the o/nIVF did not enter the first mitosis (more than 60%).
Recently, it has been reported that the IVF media itself could produce a certain amount of ROS within few hours from the incubation 15 . To investigate the source of ROS in IVF system, we measured the level of ROS produced from IVF medium alone as well as those produced by spermatozoa suspended in IVF medium, from the beginning of IVF up to the end of an overnight incubation. We used the fluorescent dye CM-H 2 DCFDA, sensitive to some ROS, that gives an index of general ROS production from media and cells.
Interestingly, we found a peak of oxidation rate at 4 h (shIVF) from sperm incubation in IVF medium, followed by a strong decrease of oxidation rate after an overnight incubation. This behaviour positively correlates with a higher cleavage rate observed after shIVF and could be likely due to the acquisition of capacitation and fertilization capability of spermatozoa, and the ROS production in this early time of IVF, also reflected by the higher number of spermatozoa still attached to the Zona Pellucida at 4 h from IVF. In fact, it was observed, that ROS inducing lipid peroxidation of sperm membrane enhances the binding of spermatozoa to the ZP 16 , a fundamental aspect of sperm-oocytes interaction. Furthermore, at the end of the canonical o/nIVF, there is a strong decrease in the motility of the spermatozoa, this could explain the low levels of ROS in this group.
Polyspermic fertilization shrank dramatically in the shIVF group, thus confirming previous reports 7, 17,18 . The removal of the oocytes from IVF shortly after fertilization positively affected embryonic development, by subtracting them to the high levels of ROS, whose detrimental effects are well established. The positive trend in shIVF group continued till day 7th of culture, with a significantly improvement of the blastocyst development www.nature.com/scientificreports/ comparing to o/nIVF group (29% vs 23% of total incubated oocytes). Avoiding a long ROS exposure ameliorated not only the number of blastocysts, but even their quality. Accordingly, we have detected an enhanced expression of SOD in our shIVF blastocysts, as already reported in bovine embryos 19,20 . Likewise, the mitochondrial form of superoxide dismutase, the manganese-SOD (Mn-SOD), has been found to be significantly higher in bovine blastocysts cultured in vivo compared to those cultured in vitro, thus positively associated with a high-quality blastocyst 21 . Besides the morphological appearance, even the total number of cells was significantly higher in shIVF derived blastocysts. The increased proliferation rate was backed up by the upregulation of the PCNA, a specific marker of S-phase of the cell cycle which correlates with cell proliferation and DNA synthesis 22,23 , detected by us in shIVF blastocyst. The upregulated expression of PCNA has been found in ovine embryos produced in vitro at low oxygen concentration, demonstrating that low ROS level induce PCNA up-regulation 24 .
To summarize, our findings demonstrated that a long exposure high ROS level negatively affects embryo development since the early stages. Thus, the final message that our article conveys is that restricting the gametes interaction to an optimal time frame compatible with a satisfactory fertilization rate, probably to be finely investigated in each animal model, quantitatively and qualitatively improves in vitro embryo production in sheep.

Material and methods
All materials were purchased from Sigma Aldrich, Milan, unless otherwise stated.

Experimental design.
A total of 733 in vitro matured sheep oocytes were co-incubated with spermatozoa in IVF medium. Then, small batches (n = 21; subdivided in 3 replicates) of oocytes were collected every 30 min to check for the presence of a fertilizing spermatozoon. Once the fertilization window has been detected, we performed IVF for a short (shIVF) and conventional overnight (o/nIVF) incubation of spermatozoa-oocytes. IVF outcomes were compared between the two groups for embryo development, polyspermy, blastocysts quality. Furthermore, we evaluated ROS production in IVF medium at time 0 (t0), and after short and overnight incubation (Fig. 1).

In vitro fertilization (IVF).
Twenty-four hours post IVM, COCs were observed under the stereomicroscope and only the COCs that presented an expansion of cumulus cells were selected for IVF. After that, COCS were quickly pipetted in 300 U/ml hyaluronidase solution (dissolved in TCM-199), washed twice in H199 and placed into 50 μl drops (in number of 8 to 10 oocytes/drop) of IVF medium (SOF-with 20% oestrus sheep serum and 16 mM isoproterenol), covered by mineral oil. Single straw of frozen semen, containing 100 × 10 6 spermatozoa, was fast-thawed in 35 °C water and centrifuged in sperm-wash medium (bicarbonate-buffered synthetic oviductal fluid (SOF-) with 0.4% (w:v) fatty-acid free BSA), at 1200 rpm for 5 min. Supernatant was discarded and 5 × 10 6 spermatozoa were added to each drop and incubated in a humidified atmosphere at 38.5 °C, 5% CO 2 , and 7% O 2 .
Detection of sperm entrance. To  Embryo development. Presumptive zygotes from both groups were cultured in 20 μl drops of SOF-aa covered by mineral oil, in a humidified atmosphere at 38.5 °C, 5% CO 2 , and 7% O 2 . The medium was renewed at day 3 (SOF + aa) (SOF-aa supplemented with 0.27 mg/ml glucose (SOF +), 2% EAA, 1% NEAA), on day 5 RT-qPCR. Total RNA was extracted from blastocyst (n = 3) in shIVF and o/nIVF groups (n = 3) using TRIzol (Invitrogen, Carlsbad, CA) and purified with NucleoSpin miRNA kit (Macherey-Nagel, Germany), following the protocol in combination with TRIzol lysis with small and large RNA in one fraction (total RNA) with minor modification. Briefly, 150 μl of TRIzol were added into each pool of three blastocysts, homogenized and incubated 5 min at room temperature. The mixture was then added 30 μl of chloroform, homogenized and incubated 3 min at room temperature. Aqueous phase containing RNA was separated after centrifugation (15, Table 3 reported the previous reported primers used in this study. Primers were designed from specific exon-exon junctions to avoid amplifying genomic DNA. RT-qPCR was performed with three technical replicate each, using 4 μl of diluted cDNA (1:20 Vol.), 5 μl of the Power SYBR Green Master Mix (Applied Biosystems, Carlsbad, California, USA) and 0.5 μl of forward and reverse primers (final concentration 600 nM for PCNA and ACTB and 900 nM for SOD1) with QuantStudio 6 Flex Real-Time PCR Systems (Applied Biosystems, Carlsbad, California, USA). RT-QPCR efficiencies calculated from slope were 1.14, 1.16 and 0.94 for PCNA, SOD1 and ACTB, respectively. Gene relative expressions between shIVF and o/nIVF groups and P-Value (t-test) were calculated by the PCR R package 26 . Measurement of ROS generation. The amount of ROS produced by the IVF medium and spermatozoa has been calculated by the oxidation rate of the fluorescent probe CM-H 2 DCFDA. To estimate the rate of ROS produced by spermatozoa, we incubated spermatozoa both in IVF medium and in PBS. Measurements were taken at three temporal stages: at the beginning (0 h), at short (shIVF) and overnight (o/nIVF) incubation. To avoid photo-bleaching, IVF medium was diluted 1:9 with PBS, immediately before measurement 15 . For each group, a total of 200 μl of cell-free or spermatozoa-containing medium/buffer, were placed in each well of a 96-well plate, and then mixed with CM-H 2 DCFDA dye to a final dye working concentration of 3 μM.
Fluorescence emission measurements were carried out in a multimode ENSPIRE plate reader (Perkin Elmer) at a constant temperature of 38 °C. To avoid condensation, a difference of 0.3 °C between the bottom and the top of the plate was set. The excitation and emission wavelengths were set respectively to 490 nm and 520 nm. 40 repeats were measured for each well with an interval of 1 min for each repeat. For each repeat, 100 flashes were lighted at a height of 8 mm from the top of the plate.
The kinetics of CM-H 2 DCFDA oxidation rate was evaluated from the slope of the fluorescence emission intensity in the initial 10 min of the total recording time. Results are reported as mean ± SD of triplicate measurements.
A calibration curve was set-up by measuring the oxidation rate of the dye as a function of H 2 O 2 concentration in PBS. Dye-free PBS/IVF medium have been used as negative controls.