Exploration of the Cytoplasmic Function of Abnormally Fertilized Embryos via Novel Pronuclear-Stage Cytoplasmic Transfer

In regular IVF, a portion of oocytes exhibit abnormal numbers of pronuclei (PN) that is considered as abnormal fertilization, and they are routinely discarded. However, it is known that abnormal ploidy still does not completely abandon embryo development and implantation. To explore the potential of cytoplasm from those abnormally fertilized oocytes, we developed a novel technique for the transfer of large cytoplasm between pronuclear-stage mouse embryos, and assessed its impact. A large volume of cytoplast could be efficiently transferred in the PN stage using a novel two-step method of pronuclear-stage cytoplasmic transfer (PNCT). PNCT revealed the difference in the cytoplasmic function among abnormally fertilized embryos where the cytoplasm of 3PN was developmentally more competent than 1PN, and the supplementing of fresh 3PN cytoplasm restored the impaired developmental potential of postovulatory “aged” oocytes. PNCT-derived embryos harbored significantly higher mitochondrial DNA copies, ATP content, oxygen consumption rate, and total cells. The difference in cytoplasmic function between 3PN and 1PN mouse oocytes probably attributed to the proper activation via sperm and may impact subsequent epigenetic events. These results imply that PNCT may serve as a potential alternative treatment to whole egg donation for patients with age-related recurrent IVF failure.

Germline nuclear transfer in mice may rescue poor embryo development associated with advanced maternal age and early embryo arrest

STUDY QUESTION

Can pronuclear transfer (PNT) or maternal spindle transfer (ST) be applied to overcome poor embryo development associated with advanced maternal age or early embryo arrest in a mouse model?

SUMMARY ANSWER

Both PNT and ST may have the potential to restore embryonic developmental potential in a mouse model of reproductive ageing and embryonic developmental arrest.

WHAT IS KNOWN ALREADY

Germline nuclear transfer (NT) techniques, such as PNT and ST, are currently being applied in humans to prevent the transmission of mitochondrial diseases. Yet, there is also growing interest in the translational use of NT for treating infertility and improving IVF outcomes. Nevertheless, direct scientific evidence to support such applications is currently lacking. Moreover, it remains unclear which infertility indications may benefit from these novel assisted reproductive technologies.

STUDY DESIGN, SIZE, DURATION

We applied two mouse models to investigate the potential of germline NT for overcoming infertility. Firstly, we used a model of female reproductive ageing (B6D2F1 mice, n = 155), with ages ranging from 6 to 8 weeks (young), 56 (aged) to 70 weeks (very-aged), corresponding to a maternal age of <30, ∼36 and ∼45 years in humans, respectively. Secondly, we used NZB/OlaHsd female mice (7-14 weeks, n = 107), as a model of early embryo arrest. This mouse strain exhibits a high degree of two-cell block. Metaphase II (MII) oocytes and zygotes were retrieved following superovulation.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Ovarian reserve was assessed by histological analysis in the reproductive-aged mice. Mitochondrial membrane potential (△Ψm) was measured by JC-1 staining in MII oocytes, while spindle-chromosomal morphology was examined by confocal microscopy. Reciprocal ST and PNT were performed by transferring the meiotic spindle or pronuclei (PN) from unfertilised or fertilised oocytes (after ICSI) to enucleated oocytes or zygotes between aged or very-aged and young mice. Similarly, NT was also conducted between NZB/OlaHsd (embryo arrest) and B6D2F1 (non-arrest control) mice. Finally, the effect of cytoplasmic transfer (CT) was examined by injecting a small volume (∼5%) of cytoplasm from the oocytes/zygotes of young (B6D2F1) mice to the oocytes/zygotes of aged or very-aged mice or embryo-arrest mice. Overall, embryonic developmental rates of the reconstituted PNT (n = 572), ST (n = 633) and CT (n = 336) embryos were assessed to evaluate the efficiency of these techniques. Finally, chromosomal profiles of individual NT-generated blastocysts were evaluated using next generation sequencing.

MAIN RESULTS AND THE ROLE OF CHANCE

Compared to young mice, the ovarian reserve in aged and very-aged mice was severely diminished, reflected by a lower number of ovarian follicles and a reduced number of ovulated oocytes (P < 0.001). Furthermore, we reveal that the average △Ψm in both aged and very-aged mouse oocytes was significantly reduced compared to young mouse oocytes (P < 0.001). In contrast, the average △Ψm in ST-reconstructed oocytes (very-aged spindle and young cytoplast) was improved in comparison to very-aged mouse oocytes (P < 0.001). In addition, MII oocytes from aged and very-aged mice exhibited a higher rate of abnormalities in spindle assembly (P < 0.05), and significantly lower fertilisation (60.7% and 45.3%) and blastocyst formation rates (51.4% and 38.5%) following ICSI compared to young mouse oocytes (89.7% and 87.3%) (P < 0.001). Remarkably, PNT from zygotes obtained from aged or very-aged mice to young counterparts significantly improved blastocyst formation rates (74.6% and 69.2%, respectively) (P < 0.05). Similarly, both fertilisation and blastocyst rates were significantly increased after ST between aged and young mice followed by ICSI (P < 0.05). However, we observed no improvement in embryo development rates when performing ST from very-aged to young mouse oocytes following ICSI (P > 0.05). In the second series of experiments, we primarily confirmed that the majority (61.8%) of in vivo zygotes obtained from NZB/OlaHsd mice displayed two-cell block during in vitro culture, coinciding with a significantly reduced blastocyst formation rate compared to the B6D2F1 mice (13.5% vs. 90.7%; P < 0.001). Notably, following the transfer of PN from the embryo-arrest (NZB/OlaHsd) zygotes to enucleated non-arrest (B6D2F1) counterparts, most reconstructed zygotes developed beyond the two-cell stage, leading to a significantly increased blastocyst formation rate (89.7%) (P < 0.001). Similar findings were obtained after implementing ST between NZB/OlaHsd and B6D2F1 mice, followed by ICSI. Conversely, the use of CT did not improve embryo development in reproductive-age mice nor in the embryo-arrest mouse model (P > 0.05). Surprisingly, chromosomal analysis revealed that euploidy rates in PNT and ST blastocysts generated following the transfer of very-aged PN to young cytoplasts and very-aged spindles to young cytoplasts were comparable to ICSI controls (with young mouse oocytes). A high euploidy rate was also observed in the blastocysts obtained from either PNT or ST between young mice. Conversely, the transfer of young PN and young spindles into very-aged cytoplasts led to a higher rate of chromosomal abnormalities in both PNT and ST blastocysts.

LARGE SCALE DATA

N/A.

LIMITATIONS, REASONS FOR CAUTION

The limited number of blastocysts analysed warrants careful interpretation. Furthermore, our observations should be cautiously extrapolated to humans given the inherent differences between mice and women in regards to various biological processes, including centrosome inheritance. The findings suggest that ST or PNT procedures may be able to avoid aneuploidies generated during embryo development, but they are not likely to correct aneuploidies already present in some aged MII oocytes.

WIDER IMPLICATIONS OF THE FINDINGS

To our knowledge, this is the first study to evaluate the potential of PNT and ST in the context of advanced maternal age and embryonic developmental arrest in a mouse model. Our data suggest that PNT, and to a lesser extent ST, may represent a novel reproductive strategy to restore embryo development for these indications.

STUDY FUNDING/COMPETING INTEREST(S)

M.T. is supported by grants from the China Scholarship Council (CSC) (Grant no. 201506160059) and the Special Research Fund from Ghent University (Bijzonder Onderzoeksfonds, BOF) (Grant no. 01SC2916 and no. 01SC9518). This research is also supported by the FWO-Vlaanderen (Flemish fund for scientific research, Grant no. G051017N, G051516N and G1507816N). The authors declare no competing interests.

TRIAL REGISTRATION NUMBER

N/A.

Evidence of Selection Against Damaged Mitochondria During Early Embryogenesis in the Mouse

There is evidence of a purifying filter acting in the female germline to prevent the expansion of deleterious mutations in the mitochondrial DNA (mtDNA). Given our poor understanding of this filter, here we investigate the competence of the mouse embryo to eliminate dysfunctional mitochondria. Toward that, mitochondria were damaged by photoirradiation of NZB/BINJ zygotes loaded with chloromethyl-X-rosamine (CMXRos). The resultant cytoplasm was then injected into C57BL/6J zygotes to track the levels of NZB/BINJ mtDNA during the preimplantation development. About 30% of NZB/BINJ mtDNA was present after injection, regardless of using photoirradiated or non-photoirradiated cytoplasmic donors. Moreover, injection of photoirradiated-derived cytoplasm did not impact development into blastocysts. However, lower levels of NZB/BINJ mtDNA were present in blastocysts when comparing injection of photoirradiated (24.7% ± 1.43) versus non-photoirradiated (31.4% ± 1.43) cytoplasm. Given that total mtDNA content remained stable between stages (zygotes vs. blastocysts) and treatments (photoirradiated vs. non-photoirradiated), these results indicate that the photoirradiated-derived mtDNA was replaced by recipient mtDNA in blastocysts. Unexpectedly, treatment with rapamycin prevented the drop in NZB/BINJ mtDNA levels associated with injection of photoirradiated cytoplasm. Additionally, analysis of mitochondria-autophagosome colocalization provided no evidence that photoirradiated mitochondria were eliminated by autophagy. In conclusion, our findings give evidence that the mouse embryo is competent to mitigate the levels of damaged mitochondria, which might have implications to the transmission of mtDNA-encoded disease.

Macrophage-Dependent Cytoplasmic Transfer during Melanoma Invasion In Vivo

Interactions between tumor cells and tumor-associated macrophages play critical roles in the initiation of tumor cell motility. To capture the cellular interactions of the tumor microenvironment with high-resolution imaging, we directly visualized tumor cells and their interactions with macrophages in zebrafish. Live imaging in zebrafish revealed that macrophages are dynamic, yet maintain sustained contact with tumor cells. In addition, the recruitment of macrophages to tumor cells promotes tumor cell dissemination. Using a Cre/LoxP strategy, we found that macrophages transfer cytoplasm to tumor cells in zebrafish and mouse models. Remarkably, macrophage cytoplasmic transfer correlated with melanoma cell dissemination. We further found that macrophages transfer cytoplasm to tumor cells upon cell contact in vitro. Thus, we present a model in which macrophage/tumor cell contact allows for the transfer of cytoplasmic molecules from macrophages to tumor cells corresponding to increased tumor cell motility and dissemination.

A limited survey-based uncontrolled follow-up study of children born after ooplasmic transplantation in a single centre

Experimental ooplasmic transplantation from donor to recipient oocyte took place between 1996 and 2001 at Saint Barnabas Medical Center, USA. Indication for 33 patients was repeated implantation failure. Thirteen couples had 17 babies. One patient delivered twins from mixed ooplasmic and donor egg embryos. A limited survey-based follow-up study on the children is reported: 12 out of 13 parents completed a questionnaire on pregnancy, birth, health, academic performance and disclosure. Parents of a quadruplet did not participate. Prenatal development and delivery were uneventful. School grades ranged from good to excellent. Children were of good health. Body mass index (BMI) was normal in 12 out of 13 children. One child had chronic migraine headaches, two mild asthma, three minor vision and three minor skin problems. One boy from a boy/girl twin was diagnosed with borderline pervasive developmental disorder - not otherwise specified at age 18 months, but with no later symptoms. One couple disclosed the use of egg donor to their child. One reported intention to disclose; six were undecided and four reported they would not disclose. This limited follow-up strategy presents a high risk of bias. Parents may not assent to standardized clinical analysis owing to lack of disclosure to their children.

Effects of ROCK inhibitor Y-27632 on cell fusion through a microslit

We previously reported a direct cytoplasmic transfer method using a microfluidic device, in which cell fusion was induced through a microslit (slit-through-fusion) by the Sendai virus envelope (HVJ-E) to prevent nuclear mixing. However, the method was impractical due to low efficiency of slit-through-fusion formation and insufficient prevention of nuclear mixing. The purpose of this study was to establish an efficient method for inducing slit-through-fusion without nuclear mixing. We hypothesized that modulation of cytoskeletal component can decrease nuclear migration through the microslit considering its functions. Here we report that supplementation with Y-27632, a specific ROCK inhibitor, significantly enhances cell fusion induction and prevention of nuclear mixing. Supplementation with Y-27632 increased the formation of slit-through-fusion efficiency by more than twofold. Disruption of F-actin by Y-27632 prevented nuclear migration between fused cells through the microslit. These two effects of Y-27632 led to promotion of the slit-through-fusion without nuclear mixing with a 16.5-fold higher frequency compared to our previous method (i.e., cell fusion induction by HVJ-E without supplementation with Y-27632). We also confirmed that mitochondria were successfully transferred to the fusion partner under conditions of Y-27632 supplementation. These findings demonstrate the practicality of our cell fusion system in producing direct cytoplasmic transfer between live cells.

Mitochondrial replacement: from basic research to assisted reproductive technology portfolio tool-technicalities and possible risks

Mitochondrial DNA (mtDNA) mutations are a relatively common cause of progressive disorders that can be severe or even life-threatening. There is currently no cure for these disorders; therefore recent research has been focused on attempting to prevent the transmission of these maternally inherited mutations. Here we highlight the challenges of understanding the transmission of mtDNA diseases, discuss current genetic management options and explore the use of germ-line reconstruction technologies to prevent mtDNA diseases. In particular we discuss their potential, indications, limitations and possible safety concerns.

Donor oocyte cytoplasmic transfer did not enhance implantation of embryos of women with poor ovarian reserve

PURPOSE

To determine whether donor oocyte cytoplasm transferred into the oocytes of women < or = 40 years or with diminished ovarian reserve would enhance embryo quality, implantation, or pregnancy rates.

METHODS

Study subjects included women > or = 40 years (15) or with abnormal FSH levels (3). Healthy volunteers (18) produced oocytes for cryopreservation. Donor oocytes were thawed and cytoplasm from surviving oocytes was injected with a single sperm into the cytoplasm of recipient oocytes. Outcome measures included embryo quality scores, implantation, and pregnancy rates.

RESULTS

Eighteen donors produced 213 oocytes for cryopreservation and 39/171 (22.8%) survived thawing. Eighteen recipients initiated 25 IVF cycles with embryo transfer in 20 cycles after cytoplasmic transfer (CT). Four cycles resulted in three biochemical losses and one aneuploid clinical loss. Embryo quality did not improve with CT compared to pre-CT IVF cycles in six recipients.

CONCLUSIONS

CT with cryopreserved donor oocyte cytoplasm did not enhance success in women with advanced reproductive age or low ovarian reserve.