Follicular extracellular vesicles enhance meiotic resumption of domestic cat vitrified oocytes

Extracellular vesicles (EVs) contain multiple factors that regulate cell and tissue function. However, understanding of their influence on gametes, including communication with the oocyte, remains limited. In the present study, we characterized the proteome of domestic cat (Felis catus) follicular fluid EVs (ffEV). To determine the influence of follicular fluid EVs on gamete cryosurvival and the ability to undergo in vitro maturation, cat oocytes were vitrified using the Cryotop method in the presence or absence of ffEV. Vitrified oocytes were thawed with or without ffEVs, assessed for survival, in vitro cultured for 26 hours and then evaluated for viability and meiotic status. Cat ffEVs had an average size of 129.3 ± 61.7 nm (mean ± SD) and characteristic doughnut shaped circular vesicles in transmission electron microscopy. Proteomic analyses of the ffEVs identified a total of 674 protein groups out of 1,974 proteins, which were classified as being involved in regulation of oxidative phosphorylation, extracellular matrix formation, oocyte meiosis, cholesterol metabolism, glycolysis/gluconeogenesis, and MAPK, PI3K-AKT, HIPPO and calcium signaling pathways. Furthermore, several chaperone proteins associated with the responses to osmotic and thermal stresses were also identified. There were no differences in the oocyte survival among fresh and vitrified oocyte; however, the addition of ffEVs to vitrification and/or thawing media enhanced the ability of frozen-thawed oocytes to resume meiosis. In summary, this study is the first to characterize protein content of cat ffEVs and their potential roles in sustaining meiotic competence of cryopreserved oocytes.


Results and discussion
Follicular fluid EVs characterization. The Total Exosome Isolation Kit (Invitrogen, USA) was used to recover cat ffEVs, as previously utilized for cat oviductal EVs 50 . A combination of Nanoparticle Tracking Analysis (NTA) and Transmission Electron Microscopy (TEM) were used to confirm the presence of, characterize, and quantify cat ffEVs. Zeta View NTA showed the presence of EVs with an average size of 129.3 ± 61.7 nm (Fig. 1a). TEM confirmed the presence of circular vesicles with the characteristic doughnut shape with an average size of 93.2 ± 76.5 nm (12 to 507 nm, Fig. 1a-c). The average size of cat ffEVs observed in the present study is consistent with that reported in the cow (Bos taurus; 128-142 nm) 29,33 . The total ffEVs concentration detected by NTA ranged from 1.4 ×10 10 to 14.0 ×10 10 particles mL −1 , with an average of 6.3 ± 5.8 ×10 10 particles mL −1 .
Notably, it has been shown that the isolation of extracellular vesicles using precipitation methods, as the one used in the present study, can lead to a higher number of non-EV co-precipitates 55 . Following the Minimal Information for Studies of Extracellular Vesicles guidelines (MISEV 2018) 55 , we identified different kinds of apolipoproteins as potential co-precipitates in the ffEVs. However, apolipoproteins also play a significant role in fertilization and embryo development. It is therefore likely that reproductive fluid EVs naturally contain apolipoproteins, unlike other, non-reproductive EVs used in the MISEV guidelines. In this regard, apolipoproteins have been identified in proteomics of female fluids/EVs or produced by embryos in numerous species, including from EVs of the porcine endometrium 46 , sheep conceptus 56 , sheep uterine fluid of pregnant and non-pregnant sheep 44 , cat oviduct 50 , and from human follicular fluid 57 . In the present context, it is therefore difficult to ascertain if apolipoproteins are co-precipitates or are normally present in reproductive EVs.

Follicular fluid EVs protein content and their possible role on oocyte structure and function.
To further characterize domestic cat ffEV proteins, a functional analysis of the 1,974 protein entries was evaluated using the Cytoscape ClueGO plugin 58 . The comparative analysis of GO terms identified a total of 429 GO biological processes, 136 GO molecular functions, and 151 GO cellular components by analyzing the corresponding genes to all identified proteins in the cat ffEVs in reference to the domestic cat (Felis catus) genome (Supplementary data 1). For GO biological processes, 54 terms important for the maintenance of COCs structure and function were identified (Table 1).
Through these analyses, twenty-four GO protein classes were identified to have roles in modulating response to cryopreservation, including oxidoreductases, cytoskeletal proteins, and chaperones (Fig. 2b). Chaperone proteins, such as the heat shock proteins (HSPs) are known to respond to a series of stresses, including sub-or supra physiological temperatures, toxic substances, extreme concentration of ions and other osmolytes 70,71 , conditions which are encountered by the oocyte during cryopreservation. Nine HSPs were abundantly detected in the cat ffEVs (HSPE1, HSPD1, HSP90AB1, HSPB1, HSP90AA1, HSPA8, LOC105260573, HSPA4L and HSPA2). Another class of chaperons, the chaperonins, that are necessary for folding actin, tubulin and newly synthesized proteins, playing a role on establishing functional cytoskeleton 72 , were also detected in the cat ffEVs (CCT8, CCT5, TCP1, CCT7, CCT2, CCT6B, CCT6A, CCT3 and CCT4). Additionally, many identified proteins from ffEVs are known to regulate oocyte maturation processes, including the Germinal Vesicle Breakdown (GVB), spindle migration and rotation, chromosome segregation, polar body extrusion, cumulus cells expansion, cell junctions, and cytoplasm maturation 63,69,[73][74][75][76] (Fig. 3). Taken together, we postulated that the transfer of these ffEV proteins to the oocytes prior to vitrification and/or during thawing could improve immature oocyte cryotolerance. Follicular fluid EVs are taken up by the immature COCs and deliver their membrane protein and lipid contents to the cocs. Extracellular vesicle entry and cargo release into cells has been proposed to occur via endocytosis, phagocytosis, micropinocytosis and/or through direct EV-plasma membrane fusion (reviewed in 77 ). Lipid dyes, such as the fluorescent neutral lipid BODIPY TR ceramide and the carbocyanine dyes DiO and DiL, and amine groups binding dyes, such as the Ghost dye UV (GD), have already been used to investigate the binding and delivery of EVs lipids and proteins, respectively, to cells [78][79][80] . To determine whether cat ffEVs were taken up by COCs, COCs were incubated with BODIPY TR Ceramide labeled ffEVs (1.5 ×10 7 particles/ml) and imaged at 15 minutes, 30 minutes, 1 hour and 18 hours. Uptake of ffEVs by COCs was first detected after 30 minutes incubation and, at 1 hour, all analyzed COCs had the red fluorescence in their cumulus cells layer (Fig. 4a) which was not observed in the no ffEVs controls ( Supplementary Fig. 5). Next, COCs were incubated with DiO and GD labeled ffEVs (1.5 ×10 7 particles/ml) and analyzed at 1 hour and 18 hours for lipid and membrane-bound protein uptake, respectively. Similar to the BODIPY staining, both lipids and membrane-bound proteins were detected in the cumulus cells at 1 hour and maximized at 18 hours (Fig. 4b).  www.nature.com/scientificreports www.nature.com/scientificreports/ However, labeled lipids and membrane-bound proteins were not detected in the oocytes even after an 18 hour incubation period. This finding is consistent with a bovine study in which it was reported that ffEVs were taken up by cumulus cells but were not observable via confocal microscopy in the oocyte after 16 hours incubation 33 . In another study, oviductal EVs were observed to be taken up by domestic dog oocytes after 72 hours incubation 81 . In the cow study as in the present, it was not conclusively determined if the lack of observable labeled proteins/ lipids in the oocyte and/or its plasma membrane was due to their absence, or if the amount delivered was below the threshold of detection by confocal and widefield fluorescence microscopy, respectively. Future studies using higher resolution microscopy, such as the stochastic optical reconstruction microscopy (STORM) could help answer this. Nevertheless, based on these results we selected the 1 h time for pre-incubation of COCs with ffEVs for the vitrification study.
Incubation with ffEVs improves oocyte meiosis resumption after vitrification. We next sought to evaluate the impact of ffEV supplementation to cat oocyte survival immediately and 26 h post-vitrification. There was no significant difference in oocyte viability among groups (Fig. 5). When comparing individual oocyte maturation stages, there also were no significant difference among treatment groups (Fig. 6).
The improved meiotic resumption of vitrified oocytes co-incubated with ffEVs could be explained by the delivery of different proteins, RNAs, and lipids from the ffEVs to the COCs. Potential modulating proteins can be broadly grouped into four categories based on their function: cell-cell communications, meiosis resumption, structural stabilization, and metabolism. The bidirectional communication between the oocyte and the granulosa cells via gap junction permits follicular and oocyte development and is mediated by a network of cellular junctions 76 . In the present study, we demonstrated that proteins regulating the formation of tight junctions (such as PP2A, DLG3, ZONAB, ERM, SYNPO, different actin isoforms, Arp2/3, Integrin, myosin II and TUBA) are also present in the cat ffEVs. Because proteins from tight junctions can also regulate gap junction formation, it is possible that these proteins 84,85 could play roles in maintaining the communication between the granulosa cells and, possibly, between granulosa cells and the oocytes, after cryopreservation.
Follicular fluid EV proteins with potential roles in modulating meiosis resumption include those involved in cumulus cell expansion, nuclear envelope breakdown, and spindle stabilization. Previous studies have shown that ffEVs are taken up by bovine COCs and enhance cumulus cell expansion 33 . This effect is possibly mediated by proteins involved in MAPK and PI3K pathways 76 , many of which were found in cat ffEVs. Cumulus cell expansion is also dependent on cumulus extracellular matrix formation and composition 69 . Here, we identified ECM components and receptors in the cat ffEVs (Fig. 2a), previously demonstrated to be required for COCs matrix formation in bovine granulosa cells 76 . It is also known that MAPK and PI3K pathways are important for the meiotic resumption in the oocyte 86 . In the present study, several proteins involved in the MAPK and PI3K signaling pathways, including SMC1A, SMC3, PP2Rs, CALM3, HSP90B1, HSPA8, HSPB1, EIF4A, MAK, and ICK, were present in cat ffEVs. Likewise, miRNAs regulating the same pathways have been found in bovine, equine and human ffEVs 32,33,37,38,87,88 . Therefore, these proteins and miRNAs could play important roles on oocyte meiotic resumption. For example, the Arp2/3 complex is essential for F-actin shell nucleation and consequent nuclear membrane fragmentation, leading to the nuclear envelope breakdown (NEB) in starfish 65 . Components of the Arp2/3 complex that play roles in actin cytoskeleton formation were abundant in the cat ffEVs and could also contribute to the higher meiosis resumption of vitrified COCs in the presence of ffEVs.
Improved mitochondrial function could also contribute to the higher meiotic resumption of ffEV supplemented vitrified oocytes. When porcine oocytes were incubated with the mitochondrial activity inhibitor Carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP), a significant reduction in the membrane potential and first polar body extrusion was observed 89 . Similarly, FCCP reduced the percentage of oocyte with nuclear maturation, normal spindle formation and chromosome alignment in mice 90 demonstrating the importance of mitochondrial activity for normal oocyte maturation. The cat ffEVs had, at least, 26 proteins that are part of the oxidative phosphorylation and could modulate mitochondria function, improving vitrified oocyte maturation. It is likely that ffEVs also contributed to cell structural recovery and to stress response following vitrification. Vitrification was previously described to increase the number of abnormal spindle in MII oocytes from sow, cow, woman, mouse, horse and cat 18,54,[91][92][93] . In our study, no abnormal MII spindles were observed in the non-vitrified controls or in the vitrified oocytes that were incubated with ffEVs. It may be that ffEVs delivered factors such as F-Actin, Arp2/3, PIR121, VCL, actin, myosin, ERM, PI4P5K, CFN, FN1, UTG, GSN, MLC, among others, that could positively affect the spindle stabilization of vitrified oocytes after vitrification. Likewise, an increased stress induced response through HSPs and chaperonins could also prevent spindle abnormalities. In the mouse, the oocyte has maximized heat shock response during the growth period, which declines when they acquire their full size and is shut off in later stages of follicular maturation, around the GVB, which could explain why mammalian oocytes are sensitive to thermic stress 94,95 . Reduced polar body extrusion and increased abnormal spindle assemblies have been observed when the modulation of the chaperonin TCP1 function was depleted in mouse oocytes (by siRNA silencing of its modulator Txndc9) 72

conclusion
In summary, the present study is the first to characterize protein content of cat ffEVs. Our results showed that ffEVs are enriched in proteins that can play roles in regulating follicle growth, oocyte energy metabolism, oocyte maturation, stress response and cell-cell communication, suggesting that ffEVs may play important roles in the crosstalk that occurs between the somatic and germline follicular components. We also demonstrated, for the first time, that ffEVs improve meiotic resumption of vitrified COCs and can be used as a tool to improve gamete  59 and GO protein class of ffEVs identified proteins. In a, KEGG pathways related to ECM-receptor interaction. Note that proteins present in ffEVs are shown in red. In b, pie chart of the 24 different GO protein classes corresponding to the proteins present in the cat ffEVs. (2020) 10:8619 | https://doi.org/10.1038/s41598-020-65497-w www.nature.com/scientificreports www.nature.com/scientificreports/ cryopreservation. The next steps are to identify mechanisms by which ffEVs regulate follicle and oocyte development. Furthermore, it is well established that the plasma membrane permeability modulates several types of cell injury associated with cryopreservation, including volume changes due to osmotic stress and CPA toxicity 96,97 . Therefore, investigations on the role of ffEVs lipids on COC function and cryopreservation are also required.

Materials and methods
Reagents. All reagents were purchased from Sigma Aldrich (St. Louis), unless otherwise stated.

Follicular fluid EV (ffEV) isolation and oocyte collection. Domestic cat ovaries voided of corpora
hemorrhagica and/or lutea (4 months to 3 years old) were opportunistically collected from local veterinary clinics after routine ovariohysterectomy and transported at 4 °C to the laboratory within 6 hours of excision. No additional permissions were required since these biological materials were designated for disposal via incineration. After being washed three times in Phosphate Buffer Saline solution (PBS, GIBCO, USA), follicular fluid was aspirated from antral-stage follicles (1-8 mm diameter), centrifuged at 2,000 × g at room temperature for 30 minutes to remove cells and debris. The supernatant was then mixed with 500 μL of the Total Exosome Isolation Reagent (Invitrogen, USA) and incubated overnight at 4 °C. The samples were centrifuged at 10,000 × g for 1 hour, and the pellet resuspended in 50 μL of PBS. ffEVs were then aliquoted, and stored at −20 °C until use. Each ffEV aliquot was only thawed once immediately prior to utilization, to avoid multiple freeze/thaw cycles.
Oocytes were collected from spayed domestic cat ovaries (>10 months old). Ovaries were washed in handling medium composed of MEM eagle with 100 U ml −1 penicillin G, 10 mg ml −1 streptomycin sulfate, 100 mM pyruvate, 25 mM HEPES, and 4 mg ml −1 bovine serum albumin. Oocytes were collected via dicing of the ovarian cortex with a scalpel blade. Homogenously dark, circular oocytes with at least two layers of surrounding cumulus cells were selected for use. Follicular fluid EV quantitation. Nanoparticle tracking analysis was done using the ZetaView S/N 17-332 (Particle Metrix, Meerbusch, Germany) and data analyzed using its software (ZetaView 8.04.02) by Alpha Nano Tech (Research Triangle Drive, NC, USA) as previously described 98 . For each replicate (n = 4, pooled frozen-thawed samples from 1-4 individuals, totaling 9 cats), ffEVs sample (1 ml) was diluted 100X in PBS, loaded into the cell, and the instrument measured each sample at 11 different positions throughout the cell, with Proteomic analyses. Follicular fluid EVs (n = 7 cats) were pooled, frozen at −20 °C. Proteins were extracted and prepared via single-pot, solid phase-enhanced sample-preparation (SP3) technology 99 by Bioproximity LLC (Chantilly, VA), and analyzed using ultraperformance liquid chromatography and tandem mass spectrometry (UPLC -Thermo Easy-nLC 1200 fitted with a heated, 25 cm Easy-Spray column -MS/MS -Thermo Q-Exactive HF-X quadrupole-Orbitrap mass spectrometer). The peptide dataset (mzML format) were exported to Mascot generic format (mgf) and searched using X!!Tandem 100 using both the native and k-score scoring algorithms 101 , and by OMSSA 102 . RAW data files were compared with the protein sequence libraries available for the domestic cat (Felis catus, taxa 9685). Label free quantification (MS1-based) was used and peptide peak areas were calculated using OpenMS 103 . Proteins were required to have one or more unique peptides across the analyzed samples with E-value scores of 0.01 or less. For enrichment analysis, the cut off was set to p < 0.05. The Cytoscape 3.5.1 plugin ClueGO 58 was used to visualize interactions of EVs proteins and networks integration, by GO terms "biological processes", "molecular function" and "cellular components" using the Felis catus genome. The evidence was set to "Inferred by Curator (IC)", and the statistical test was set to a right-sided hypergeometrical test with a κ score of 0.7-0.9 using  Freshly collected domestic cat cumulus-oocyte complexes (COCs, n = 3 animals, 100 oocytes) were co-incubated in 50 µl droplets of handling medium, immersed in mineral oil, in the absence or presence of 1.5 ×10 7 particles ml −1 dyed ffEVs. Oocytes (5-7/droplet) were imaged at 10, 20, 40 and 100 × under a fluorescence microscope (EVOS FL auto 2, Invitrogen, USA) after 0, 15, 30, 60, and 1080 min (18 hr) co-incubation at 5% CO 2 and 38 °C.
Oocyte Cryopreservation and Thawing. COCs were cryopreserved using a protocol modified from Comizzoli et al. (2009) 105 and Colombo et al. (2019) 16 . Briefly, COCs were transferred in small groups (7-11 oocytes) via a mouth pipette to an equilibration solution, consisting of handling medium supplemented with 7.5% (v/v) ethylene glycol and 7.5% (v/v) dimethylsulfoxide for 3 minutes on ice. COCs were then briefly exposed to vitrification solution containing 15% ethylene glycol, 15% DMSO, and 0.5 M sucrose in handling medium for 30 seconds before being loaded in minimal volume on a Cryotop (Kitazato, Japan) and plunged into liquid nitrogen according to manufacturer's instructions. COCs were thawed by passing them through a sucrose gradient (1.0, 0.5, 0.25, and 0 M) of warmed (38.5 °C) handling medium with or without 1.5 ×10 7 particles ml −1 ffEV.

In Vitro Maturation.
In vitro maturation of cat oocytes was performed using a method previously described for the domestic cat 106 . Briefly, 50 µl droplets of Quinn's Advantage Protein Plus Blastocyst Medium, supplemented with 1 µg ml −1 follicle stimulating hormone (Folltropin, Vetoquinol, USA), and luteinizing hormone (Lutropin V, Vetoquinol, USA), were equilibrated under mineral oil at 5% CO 2 and 38 °C for at least 2 hours before use. Fresh or frozen-thawed cat COCs were transferred in groups of 14-28 oocytes to each droplet and incubated for 24 hours. Following IVM, oocytes were washed in PBS containing EDTA (0.1 mM), EGTA (0.1 mM), Imidazole (50 mM), 4% Triton-X, PVP (3 mg mL −1 ) and PMSF (24 µM). During this step, oocytes were rapidly pipetted up and down to remove cumulus cells, then fixed overnight in 4% paraformaldehyde in PBS at 4 °C.
Oocyte survival, defined as maintaining normal morphology with spherical shape, intact membranes, clear zone without rupture, and uniform cytoplasm, was evaluated immediately after thawing (0 h) and following in vitro maturation (26 h) via light microscopy 107 . Oocyte Spindle Staining and Imaging. After fixation, oocytes were washed in washing buffer consisting PBS with 2.5% (v/v) normal goat serum and 0.5% triton X, then blocked for 30 min at 38 °C in PBS with 5% normal goat serum and 0.5% Triton X before incubation with rabbit anti-alpha tubulin (ab18251, Abcam) at www.nature.com/scientificreports www.nature.com/scientificreports/ 1:125 dilution in washing buffer for 1 hour at 38 °C. After washing, oocytes were incubated for 1 hour at 38 °C in goat-anti-rabbit IgG (1:200 dilution) and Hoechst 33342 (5 µg mL −1 , Invitrogen) washed and placed into a small drop of ProLong Glass Anti-fade mounting solution (Invitrogen P36980) on a glass slide, and imaging under a fluorescent microscope (EVOS FL auto 2). Oocytes were imaged under 1,000 × magnification using a fluorescence microscope (EVOS FL auto 2, Invitrogen, USA) to observe microtubule organization, and staged as previously described with some modifications 108 . experimental Design. Study 1: Characterization of cat follicular fluid. Follicular fluid was collected and pooled from 1-4 individuals in domestic cat (n = 9). The amount of EVs in pooled follicular fluid was then quantified by Nano tracking analysis and its presence confirmed by transmission electron microscopy. Pooled ffEV samples from 7 of same 9 cats were subjected to proteomics analysis and uptake testing.
Study 2: Effect of follicular fluid EV on oocyte vitrification. Domestic cat COCs (n = 15 cats, 267 oocytes) were co-incubated in the absence (with PBS) or presence of 1.5 ×10 7 particles ml −1 unlabeled ffEV (diluted in PBS) at 5% CO 2 and 38 °C for 1 hour. A subset of oocytes was then in vitro matured as fresh controls (n = 44 oocytes without ffEV, n = 45 with ffEV), and the remaining oocytes were vitrified in the presence or absence of ffEV at the same concentration in equilibration and vitrification solutions. COCs were maintained in liquid nitrogen for at least 30 minutes before being thawed (with or without ffEV) and subjected to in vitro maturation (Fig. 8).
Statistical Analysis. Data are presented as means ± standard deviation (SD). Comparisons in oocyte viability and maturation stage among treatments were evaluated by the Friedman non-parametric analysis of variance (GraphPad PRISM 7, GraphPad Software, USA). Comparisons in oocyte maturation between control and vitrification group and between vitrification with/without ffEV treatment were evaluated via Kaplan-Meier (displayed as log-rank prob > ChiSquare) with oocyte maturation stage coded as a continuous variable (germinal vesicle = 0, germinal vesicle breakdown = 1… Metaphase II = 6) in JMP Pro 12 (SAS Institute Inc., USA). Differences were considered significant at p < 0.05.

Data availability
The authors declare that all data supporting the findings of this study are available within the article, Supplementary Files, or from the corresponding author upon reasonable request. UPLC-MS/MS (mzML) file has been deposited in FIGSHARE database under DOI number: 10.6084/m9.figshare.7837331.