Cell‐free DNA content in follicular fluid: A marker for the developmental ability of porcine oocytes

Abstract Purpose The present study examined the relationships among the amount of cell‐free‐DNA (cfDNA) in porcine follicular fluid (FF), the developmental ability of enclosed oocytes, and characteristics of granulosa cells and examined the effect of cfDNA content in maturation medium on the developmental ability of the oocytes. Methods Oocytes and FF were collected from individual gilts, and the gilts were rated based on the ability of their oocytes to develop to the blastocyst stage and the amount of cfDNA in the FF. The copy numbers of mitochondrial DNA (Mt‐DNA) and nuclear DNA (N‐DNA) were measured by real‐time PCR and the DNA sequence. FF or cfDNA was added to the maturation medium, and the developmental ability of the oocytes was examined. Results The amount of cfDNA was associated with apoptosis of the granulosa cells, and high‐cfDNA content in FF was associated with low developmental ability of oocytes. Supplementation of the maturation medium with FF containing high cf‐Mt‐DNA or with DNA extracted from the FF did not affect oocyte developmental competence. Conclusions Cell‐free DNA content in FF is a marker for oocyte competence, but cfDNA in the oocyte maturation environment did not affect oocyte developmental ability.

nuclear or mitochondrial. 7 A previous study showed that both cf-N-DNA and cf-Mt-DNA, analyzed by real-time PCR, are closely related to cfDNA content (ng) in FF, and living granulosa cells actively secreted cf-Mt-DNA into the medium in response to mitochondrial dysfunction. 7 However, other factors associated with this cfDNA in the follicles remain unclear. Furthermore, whether the amount of cf-N-and/ or cf-Mt-DNA reflects the developmental competence of oocytes remains unclear. For example, it has been reported that high developmental ability of oocytes is related to low cf-Mt-DNA content in FF, but not low cf-N-DNA. 8 Another question that remains unanswered is whether the cfDNA content of FF is a causal factor for poor oocytes, or a consequence of poor follicle development. Guan et al 9 reported that when DNA is added to the culture medium, it causes granulosa cell apoptosis. However, in that study, extensively high concentrations of DNA were directly added, and cfDNA is believed to be contained in extracellular vesicles or present and free floating. 7,10 Therefore, additional studies are needed to evaluate the significance of cfDNA in FF.
In the present study, we collected the ovaries of gilts which were kept in identical environments, and examined the relationships among characteristics of granulosa cells, developmental competence of oocytes, and the cf-N-and cf-Mt-DNA contents in the corresponding FF. Furthermore, we investigated the effect of supplementation of maturation medium with FF containing high-or low-cfDNA content, or with cfDNA extracted from the FF, on the ability of oocytes to develop to the blastocyst stage.
FF was collected from the antrum follicles (3-5 mm in diameter) of 100 gilts, centrifuged (10 000 × g for 5 minutes), and stored at −20°C until use. Porcine zygote medium 3 was used for in vitro culturing (IVC) of embryos. 12 IVM was performed under atmospheric conditions of 5% CO₂ and 95% air at 38.5°C, and IVC was performed under atmospheric conditions of 5% O 2 , 5% CO 2 , and 90% N 2 at 38.5°C.

| Cumulus cell-oocyte complex collection, IVM, activation, and IVC
Ovaries were collected from gilts obtained from a slaughterhouse and transferred to the laboratory within 30 minutes in phosphatebuffered saline containing antibiotics at 37°C. Cumulus cell-oocyte complexes (COCs) were collected from the antral follicles (AFs, 3-6 mm in diameter) of ovaries and cultured in 100 µL droplets of IVM medium for 44 hours. After IVM, the oocytes were denuded from the surrounding GCs, parthenogenetically activated by a single electrical pulse of 60 V for 0.1 ms using a NEPA21 (NepaGene Co. Ltd), followed by incubation in PZM3 containing 10 µg/mL cytochalasin B and 10 µg/mL cycloheximide for 5 hours, and then cultured in PZM3 for 7 days to determine the rate of blastulation. The total cell number of the blastocysts was counted by Hoechst 33342 staining under a fluorescence microscope (Olympus).

| DNA extraction from FF
Prior to DNA extraction, FF was diluted 400 times with water, and the diluted FF was mixed with an equal amount of DNA extraction buffer (Tris-HCl, 20 mmol/L; Nonidet-40; Tween 20, 0.9%; and proteinase K, 0.4 mg/mL). The buffer was then heated at 55°C for 30 minutes, followed by 98°C for 5 minutes, to obtain DNA. In preliminary experiments, it was confirmed that high concentrations of FF hamper the PCR, but the 800 times dilution of FF did not affect PCR efficiency.

| Evaluation of cf-N-and cf-Mt-DNA copy numbers in FF
The N-DNA and Mt-DNA copy numbers in FF were determined using real-time PCR targeting the porcine mitochondrial genome and a single-copy nuclear gene. PCR was performed using a CFX Connect™ real-time PCR detection system (Bio-Rad) with primers targeting a single-copy gene or mitochondrial genome and Ssofast-TM EvaGreen Supermix (Bio-Rad). Primers used for detection of one-copy gene were 5′-agcagaatcaacaccatcggt-3′ and 5′-tggctccacccatagaatgc-3′ (154 bp), and those for detection of mitochondrial genome were 5′-atccaagcactatccatcacca-3′ and 5′-ccgatgattacgtgcaaccc-3′ (155bp). Primers were designed using Primer3Plus (http://sourc eforge.net/proje cts/prime r3/) and the NCBI database (GCG glucagon, NC_010457.4 and Sus scrofa mitochondrion, complete genome NC_000845.1). PCR was performed with an initial denaturation at 95°C for 1 minutes, followed by 40 cycles at 98°C for 5 seconds and 60°C for 10 seconds. A standard curve was generated for each run using 10-fold serial dilutions of the representative copies of the external standard. The external standard was the PCR product of the corresponding gene sequence, cloned into a vector using a Zero Blunt TOPO PCR cloning kit (Invitrogen). The PCR product was sequenced for confirmation prior to use. Amplification efficiencies of all assays were > 1.98.

| Validation of measurement for cfDNA content in FF using DNA seq
Cell-free DNA contained in FF has a wide variety of origins, and real-time PCR targeting of one or two sequences in the nuclear and mitochondrial genome was validated by DNA seq. FF was collected from antral follicles (3-6 mm in diameter) of 6 differ-

| Preparation and rating of FF based on the developmental ability of enclosed oocytes
A design of the preparation of FF rated on the developmental ability of oocytes is depicted in Figure 1A. Ovaries were collected from 40 gilts (a food lot of a farm), and follicular contents were aspirated from at least thirty AFs (3-6mm in diameter) of each gilt ovary. COCs were extracted from the follicular content under a stereo microscope, after which the follicular content was centrifuged (3000 × g) for 10 minutes to obtain FF. Thirty randomly selected COCs were selected from each gilt and subjected to IVM followed by activation and subsequent IVC for 7 days, to determine the developmental ability of the oocytes.

| Introduction of cfDNA into COCs or granulosa cells
Extracted DNA (see Section 2.3, final concentration 10 ng/μL) and Lipofectamine were used for the introduction, according to the protocol provided by the manufactures. Validation of the cfDNA introduction was determined by immunostaining against dsDNA.

Granulosa cells were cultured on a glass chamber (Millicell, Merck
Millipore) in 199 medium (Thermo Fisher) containing 5% FCS and the cfDNA-Lipofectamine mix for 24 h, then subjected to immunostaining. Immunostaining was conducted as previously described. 13 Antibodies used for the immunostaining were mouse anti-dsDNA

| Statistical analysis
Student's t test was used for analyzing the data of two groups, and the data (blastulation rate) from three or more groups were analyzed using analysis of variance (ANOVA), followed by Tukey's post hoc test. Percentages were arcsine transformed prior to analyses.
The results mean that the cfDNA content in FF determined using at least 15 AFs reflects the character of FF from each ovary. Therefore, the cfDNA content in FF from 30 AFs used in the following experiments reflects the character of the whole FF from each gilt.

| cfDNA content in FF relates to the apoptotic rate of corresponding granulosa cells
The next experiment addressed the causal factors of cfDNA secretion in FF. As described above (Section 2.9), the dead cell rate, apoptotic cell rate, and copy number of cfDNA in the FF were compared. As summarized in Table 2, the dead cell rate did not correlate with the rate of apoptosis in granulosa cells. In addition, the rate of dead cells correlated with neither cf-N-nor cf-Mt-cfDNA copy number in the corresponding FF. On the contrary, the rate of apoptosis in granulosa cells was significantly correlated with the cf-N-DNA copy number in FF (r = .53, P < .05) and tended to correlate with the cf-Mt-DNA in FF (r = .44, P = .09).

| High developmental ability of oocytes is related to low cfDNA content in the corresponding FF
Then, we prepared 4 High-and Low-FF batches from 40 gilts based on the developmental ability of the enclosed oocytes (Section 2.6).
Both cf-N-and cf-Mt-DNA contents were higher for Low-FF than for High-FF (P < .05, Figure 4)

| cfDNA content in the maturation medium did not affect oocyte developmental competence
We found that the origins of FF significantly affected oocyte development. We therefore addressed the question of whether cfDNA content in the maturation medium affects developmental ability to the blastocyst stage. FF containing high cfDNA or low cfDNA (Section 2.7, High-cfDNA-FF and Low-cfDNA-FF) was added to the maturation medium at a the concentration of 10 or 30%, and we found that neither the FF concentrations nor the cf-Mt-DNA content affected the rate of development of the oocytes to the blastocyst stage (High-cfDNA-FF 30%, 4.8 ± 0.7 vs Low-cfDNA-FF 30%, 4.6 ± 1.2, and High-cfDNA-FF 10%, 6.0 ± 1.0 vs Low-cfDNA-FF 10%, 3.9 ± 1.0; Figure 5A). Furthermore, when cfDNA was extracted from FF (Section 2.8) was added to the IVM medium at concentrations of 0 or 10 ng/μL, the developmental rate of the oocytes to the blastocyst stage was similar between the two DNA concentrations ( Figure 5B). Moreover, we examined the effect of introducing cfDNA into the granulosa cells during in vitro maturation on oocyte developmental competence. As seen in Figure 6, treatment of cells with cfDNA and lipofection increased dsDNA content in the cytoplasm, but did not affect developmental ability of the oocytes to the blastocyst stage.

| D ISCUSS I ON
The present study showed that cfDNA is associated with apopto- Clinical data have shown that cf-DNA in FF is a noninvasive marker of oocytes. 5,6,9,15 An in vitro model of the relationship between cfDNA in culture medium and competence of oocytes has been reported by Munakata et al, 16 such that when oocyte-granulosa cell complexes (OGCs) from early antral follicles were cultured in vitro, the quality of in vitro grown oocytes was associated with low cf-Mt-DNA content in the spent culture medium. The present study is the first to confirm that both cf-N-and cf-Mt-cfDNA contents in FF were low for follicles containing competent oocytes using young, large animals kept in similar conditions.
Cell-free-DNA is reported to be derived from apoptotic granulosa cells, 9 but others suggest that cfDNA does not reflect apoptosis In in vitro studies, it has been reported that culture conditions and culture substrates affect the release of cfDNA. 16,17 In addition, Interestingly, in the present study, we found that FF collected from ovaries containing highly developmentally competent oocytes (High-FFs) had an increased ability to support oocyte maturation, and subsequent embryonic development, compared with Low-FF (FF collected from ovaries having poorly developmentally competent oocytes). These results remind us that FF itself decides the oocyte quality, and certain adverse factors such as high cfDNA content in Low-FF may aggravate the oocyte quality. Mitochondrial DNA has been reported to be a causal factor for cellular stress and apoptosis. 19 F I G U R E 5 Effect of supplementation of maturation medium with High-cfDNA-FF and Low-cfDNA-FF on blastulation rate of the oocytes. A, Thirty COCs were cultured in IVM medium containing High-cfDNA-FF and Low-cfDNA-FF (10 or 30%), and developmental rate to the blastocyst stage was examined following activation. Experiments were repeated 6 times. B, Thirty COCs were cultured in IVM medium supplemented with 0 or 10 ng/μL of cfDNA extracted from FF. The oocytes were examined for their developmental rate to the blastocyst stage following activation. Experiment was repeated 4 times; Bar represents SEM F I G U R E 6 Effect of introducing cfDNA into COCs on the blastulation rate of the oocytes. A, Representative pictures of granulosa cells cultured with 0 or 10 ng/μL cfDNA extracted from FF for 24 hours followed by immunostaining against dsDNA. Cytoplasm of granulosa cells cultured with dsDNA had positive signals (arrow). B, Effect of coincubation of COCs with 0 or 10 ng/μL cfDNA extracted from FF on the rate of blastulation 7 days after activation. Bar represents SEM Induction of mitochondrial dysfunction such as by CCCP treatment increased cytoplasmic DNA content, which induced cellular inflammation. 20 Therefore, it is hypothesized that high-cfDNA content induces high cellular stress, which impairs oocyte competence. In testing this hypothesis, Kostyuk et al 21 have reported that cfDNA induced ROS generation in human mesenchymal stem cells and addition of cfDNA to the culture medium induced apoptosis of human granulosa cells. 9 However, in that study, 4 mg/mL cfDNA was added to the culture medium, whereas the cfDNA content in porcine FF is roughly 10 µg/mL. 7 Furthermore, in the FF and plasma, cfDNA is contained in exosomes and extracellular vesicles, 6

ACK N OWLED G M ENTS
This study was supported by the Science Research Promotion Fund from the Promotion and Mutual Aid Corporation for Private Schools of Japan.

CO N FLI C T O F I NTE R E S T
Kana Ichikawa, Hidenori Shibahara, Komei Shirasuna, Takehito Kuwayama, and Hisataka Iwata declare that they have no conflict of interest.

H U M A N R I G HT S S TATE M E NT S A N D I N FO R M E D CO N S E NT
This article does not contain any studies with human subjects.

A N I M A L S TU DY
In this study, porcine ovaries were collected from a slaughterhouse. The ovaries were discarded without any use for edible meat, and thus, this study was approved by the Ethical Committee for Animal Experiment of Tokyo University of Agriculture. All institutional and national guidelines for the care and use of laboratory animals were followed.