Parthenogenesis refers to maternally uniparental reproduction, which is a reproduction strategy through the embryonic development of oocytes without fertilization (Daughtry and Mitalipov, 2014; Jung et al., 2020). Parthenogenesis is not a natural form of reproduction in mammals. The birth of offspring from parthenogenetic embryos is not possible because of the lack of expression of imprinted paternal genes, which are required for the generation of the functional placenta (McGrath and Solter, 1984). This characteristic allows the application of parthenogenetic embryonic stem cells (pESCs), avoiding the ethical issues with biparental ESCs. Further, pESCs show self-renewal and pluripotent differentiation ability, which are similar characteristics to biparental ESCs (Didi? et al., 2013).

In the experiments for the artificial activation of mature oocytes, the activated oocytes, inhibited the extrusion of the second polar body (the 2^nd PB), could generate pseudodiploid embryos (Lee and Kang, 2019). These parthenogenetic oocytes can harbor unseparated sister chromatids, enter mitosis, and produce partially heterozygous embryos due to chromosome crossover. These embryos showed a similar developmental rate into blastocysts compared to fertilized embryos (Mitalipov et al., 2001). When the activated oocytes extrude the 2^nd PB similar to natural activation, these parthenogenetic oocytes harbor the separated sister-chromatid and finally produce haploid embryos with homozygosity (Lee and Kang, 2019). These embryos can be established to haploid ESCs, which could be a valuable source for genetic screening and cell therapy (Wutz, 2014; Ding et al., 2015). For the genetic screening, diploid cells could have heterozygous mutations, which could show few or no phenotypic changes in recessive mutations (Wutz, 2014). However, the mutations that existed in a haploid genome, a hemizygous state, can result in the complete exposure of phenotypes in the cells. Haploid cells show homozygosity for the human leukocyte antigen (HLA) type, which could be a useful source for cell therapy by preventing immune rejection in allografts (Ding et al., 2015).

The pESCs are considered a valuable source to study the influence of paternal imprinting. There is a limitation to the full-term development of parthenogenetic embryos due to their genomic imprinting status (McGrath and Solter, 1984). A recent study demonstrated that the modification of imprinted genes in oocytes could produce viable offspring from the parthenogenetic embryos (Wei et al., 2022).

Previous studies reported that the haploid embryos showed impaired development in preimplantation embryos compared to diploid embryos (Witkowska, 1973; Kaufman and Sachs, 1976; Latham et al., 2002). To enhance the development of haploid embryos, we hypothesized that the electronic fusion of blastomeres in a haploid embryo could induce diploid, which can result in an increased rate of embryo development and ESC establishment.

In this study, we analyzed the embryo development of the parthenogenetic haploid embryos in mice and investigated that electronic fusion of the haploid embryo could increase the rate of embryo development and homozygous ESC establishment.

Animals

B6D2F1 (C57BL/6N female × DBA2 male) female (8 to 9-week-old, Charles River) were used for the experiment. All animal maintenance and experimental procedures were performed following the guidelines of CHA University. The animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC No.210035) of CHA University.

Metaphase II (MII) oocytes collection

The experiment related to MII oocyte collection was performed according to the previously described methods with minor modification (Kang et al., 2014; Kim and Lee, 2019; So et al., 2020). Female mice were super-ovulated with 5 international units (IU) pregnant mare’s serum gonadotropin (PMSG), followed by 5 IU human chorionic gonadotropin (hCG) 48 h later. MII oocytes were collected 16 h after hCG injection. The denuded oocytes by 0.1% hyaluronidase solution (Sigma) were kept in KSOM (Merck) at 37℃ in a humidified incubator with 5% CO₂.

Activation of oocytes and embryo culture

Oocytes were incubated in an activation medium supplemented with 10 mM SrCl₂ (Sigma) in the presence or absence of 5 ug/mL cytochalasin B (CB, Sigma) for six hours at 37℃ in a humidified incubator with 5% CO₂. After activation, the oocytes were examined for the formation of pronucleus (PN). Activated oocytes were cultured for four to five days to the blastocysts in KSOM (Merck) at 37℃ in a humidified incubator with 5% CO₂ (Lee and Kang, 2021).

Electronic fusion

2-cell embryos are placed between the electrodes of a 100 ?m gap electrode chamber (BLS Ltd.) containing 0.28 M Mannitol (Sigma), 0.1 mM magnesium sulfate hydrate (Sigma), 50 ?M calcium chloride (Sigma), and 0.3% bovine serum albumin (Sigma) and fused by Electro Cell Manipulator ECM 2001 (BTX). An initial electrical field of 2V is applied to the embryos followed by one peak pulse of 50V for 35 ?s. After fusion, the embryos were transferred to KSOM and observed for fusion.

Establishment and culture of mESCs

Denuded blastocysts were placed onto mitomycin C (Sigma)-treated mouse embryonic fibroblast (MEF) feeder in mESC medium: KODMEM (Gibco) containing 20% KOSR (Gibco), 1 mM L-glutamine (Gibco), 100 units/mL penicillin (Hyclone), 100 ?g/mL streptomycin (Hyclone), 100 ?M β-mercaptoethanol (Sigma), 100 ?M nonessential, and 1,000 units/m LIF (Stemgent) under 5% CO₂ at 37℃ in a humidified incubator. The cell outgrowth was replaced on the new MEF feeder. mESCs were passaged every 3 to 4 days for further experiments.

Cell cycle analysis

Cells were fixed with 70% ethanol at 4℃ for 15 min and stained with 50 ?g/mL propidium iodide (Sigma) solution containing 100 ?g/mL RNase A at 4℃ for 40 min. The cell cycle was measured by a FACS Calibur (BD Biosciences) and the data were analyzed by FlowJo v10.5.3 (FlowJo LLC).

Karyotyping

KaryoMAX Colcemide (Gibco) with 150 ng/mL concentration was applied to the cultured ESCs for 1.5 h at 37℃. The treated cells were detached by trypsin/EDTA. The cells treated with 0.075 M KCL hypotonic solution for 30 min were fixed with methanol: acetic acid (3:1 v/v) and dropped onto a slide. The slides were mounted in Prolong Diamond Antifade Mountant with DAPI (Invitrogen).

Statistical analyses

Data were analyzed with Fisher’s exact test for two groups and ANOVA with Tukey analysis for multiple comparisons. Statistical analyses were performed using GraphPad Prism (version 6). Data are expressed as mean ± standard error of the mean (S.E.M), and p < 0.05 was considered significant. More than three replicates were performed for each experimental group.

To produce diploid parthenogenic embryos by activation, MII oocytes were activated with CB, resulting in that the zygotes could show two pronuclei (2 PNs) and no extrusion of the 2^nd PB after activation (Bałakier and Tarkowski, 1976) (SrCl₂+CB group in Fig. 1A). While, for the generation of haploid embryos, MII oocytes were activated without CB, and the zygote could show one pronucleus (1 PN) and the extrusion of the 2^nd PB (SrCl₂-CB group in Fig. 1A). Some haploid blastomeres on the 2-cell stage were performed electronic fusion to induce diploidy (SrCl₂-CB with 2-cell fusion group in Fig. 1A).

Figure 1.The development of parthenogenetic embryos. (A) Experimental design for the activation of oocytes and the fusion of 2-cells. MII oocytes were activated with or without Cytochalasin B (CB) (SrCl₂+CB and SrCl₂-CB group, respectively). The 2-cells in the SrCl₂-CB group were performed electronic fusion (SrCl₂-CB with 2-cell fusion group). (B) The rate of embryo development. The blastocyst rate was significantly lower in the SrCl₂-CB group than in the SrCl₂+CB group. However, the SrCl₂-CB with 2-cell fusion group showed a significantly higher blastocyst rate than the SrCl₂-CB group. N, the number of embryos; N in PN formation bars, the number of zygotes formed PN / the number of MII oocytes; N in 2-cell bar of SrCl₂-CB with 2-cell fusion group, the number of 2-cells / the number of the fused embryos. Mean ± S.E.M. *,**, ***, p < 0.05 by ANOVA with Tukey analysis (C) The morphologies of embryos at different time points. Scale bar: 100 μm. (D) The distribution of embryos at different time points. The distribution of 4-cells, morula, and blastocyst was significantly lower in the SrCl₂-CB group compared to the SrCl₂+CB group. SrCl₂-CB with a 2-cell fusion group improved the developmental delay at the morula and blastocyst stage. *,**, ***, p < 0.05 by Fisher’s exact test.

After activation, the PN formation rate in the SrCl₂+CB and SrCl₂-CB group was comparable (100% vs. 97%) (Fig. 1B). Both groups showed the embryo development to blastocyst (Fig. 1C). The embryo development from 2-cell to morula stage was similar between the SrCl₂+CB and SrCl₂-CB group (100% vs. 96% for 2-cell; 100% vs. 89% for 4-cell; 100% vs. 91% for morula). However, the blastocyst rate was significantly lower in an SrCl₂-CB group than in the SrCl₂+CB group (37% vs. 100%) (Fig. 1B).

When the 2-cell embryos in the SrCl₂-CB group were fused electronically, the embryos were changed their morphology to one cell, cleaved again on the next day, and developed into blastocysts (Fig. 1C). The embryo development from 2-cell to morula stage was 86% for 2-cell, 94% for 4-cell, and 91% for morula in SrCl₂-CB with 2-cell fusion group, which was comparable to SrCl₂+CB and SrCl₂-CB group (Fig. 1B). However, the blastocyst rate was significantly higher in the SrCl₂-CB with 2-cell fusion group than in the SrCl₂-CB group (55% vs. 37%), which was significantly lower than in the SrCl₂+CB group (100%) (Fig. 1B).

During the embryo development, the blastocysts in the SrCl₂+CB group were formed on day 4 and hatched on day 5 (Fig. 1C). While, in the SrCl₂-CB group, the blastocysts were started to appear on day 5 and the quality of blastocyst was worse compared to the SrCl₂+CB group (Fig. 1C). Based on this result, we examined what embryo stage was delayed during the embryo culture in the SrCl₂-CB group. We compared the distribution of 2-cell at 32 hours, 4-cells at 56 hours, morula at 80 hours, and blastocyst at 120 and 128 hours in the cultured embryos of SrCl₂+CB group, SrCl₂-CB group, and SrCl₂-CB with 2-cell fusion group (Fig. 1D). There was no significant difference in 2-cell distribution at 32 hours between the SrCl₂+CB group and SrCl₂-CB group. However, the distribution of 4-cells at 56 hours, morula at 80 hours, and blastocyst at 120 and 128 hours was significantly lower in the SrCl₂-CB group compared to the SrCl₂+CB group. In SrCl₂-CB with 2-cell fusion group, the distribution of morula at 80 hours, and blastocyst at 120 and 128 hours was significantly higher than SrCl₂-CB group but lower than SrCl₂+CB group. Based on results, the SrCl₂-CB group showed developmental delay from 4-cell to blastocyst stage while SrCl₂-CB with 2-cell fusion group displayed the improved development delay.

ESCs were established from the blastocysts in each group and the cell cycle analysis was performed (Fig. 2A and 2B). The mouse fibroblasts were used as a diploid control. ESCs in SrCl₂+CB and SrCl₂-CB with 2-cell fusion group were confirmed the diploid (Fig. 2B). While ESCs in the SrCl₂-CB group displayed haploid. We confirmed the haploid in the SrCl₂-CB group and diploid in SrCl₂-CB with the 2-cell fusion group by karyotyping (Fig. 2C). The rate of ESC establishment from the blastocysts was significantly lower in the SrCl₂-CB group compared to SrCl₂+CB and SrCl₂-CB with 2-cell fusion group (5.3% vs. 78.3% and 51.5%) (Fig. 2D). The rate in SrCl₂-CB with 2-cell fusion group was lower than SrCl₂+CB group but was not a significant difference. When the rate of ESC establishment was calculated based on the number of MII oocytes, the SrCl₂-CB group showed a significantly lower rate than SrCl₂+CB and SrCl₂-CB with 2-cell fusion group (1.4% vs. 78.3% and 20%) (Fig. 2D). The rate in SrCl₂-CB with the 2-cell fusion group was significantly lower than in the SrCl₂+CB group. These results suggested that electronic fusion could increase the rate of homozygous ESC establishment.

Figure 2.ESCs establishment derived from the parthenogenetic embryos. (A) The morphology of ESCs in each experimental group. Scale bar: 200 μm (B) Cell cycle analysis of the ESCs. ESCs in SrCl₂+CB and SrCl₂-CB with 2-cell fusion group showed the diploid, while ESCs in the SrCl₂-CB group displayed haploid. The mouse fibroblasts were used as a diploid control. (C) The karyotyping of the ESCs in SrCl₂-CB group and SrCl₂-CB with 2-cell fusion group. The SrCl₂-CB group showed haploid and SrCl₂-CB with 2-cell fusion group displayed diploid. (D) The rate of ESC establishment from blastocysts and MII oocytes. The rate in SrCl₂-CB with the 2-cell fusion group was significantly higher than in the SrCl₂-CB group. N, the number of ESCs / the number of blastocysts for the left panel; the number of ESCs / the number of MII oocytes for the right panel. Mean ± S.E.M. *,**, ***, p < 0.05 by ANOVA with Tukey analysis.

The present study demonstrated whether the electronic fusion of haploid embryos could induce the increase of homozygous ESCs establishment. Electronic fusion of haploid embryos induced significantly higher blastocyst generation and ESC establishment compared to non-fused ones.

In mice, parthenogenetic embryos usually maintained haploidy at the blastocyst stage (Tarkowski et al., 1970; Kaufman et al., 1983), however, post-implantation haploid embryos tended to become diploid in vivo (Kaufman, 1978). Spontaneous diploidization also occurred in haploid pESCs in vitro culture (Kaufman et al., 1983). Some studies performed the cell sorting process to maintain the haploidy in ESCs (Sagi et al., 2016). However, diploidization is required for the application of cell therapy in humans and the diploid ESCs derived from haploid one are also useful for genetic screening because of their homozygosity.

Previous studies demonstrated that haploid embryos showed a lower developmental rate in preimplantation embryos than the diploid embryos (Witkowska, 1973; Kaufman and Sachs, 1976; Latham et al., 2002), which was also observed in our study. Further, we demonstrated that the developmental delay of haploid embryos occurred in 4-cell to the blastocyst stage. Electronic fusion was performed in the haploid 2-cell, resulting in the blastocyst rate being significantly increased compared to non-fused embryos. Further, the rate of ESCs establishment was higher in fused embryos than non-fused one, suggesting more homozygous ESCs could be generated due to the higher efficiency of blastocyst and ESC generation.

Electronic fusion of 2-cell blastomeres is often used for tetraploid complementation (Gertsenstein, 2015). When we applied the same method in haploid 2-cells, the embryo development of 2-cell to morula was comparable to the non-fused haploid embryo, further the blastocyst rate was significantly higher, resulting in the electronic fusion did not damage the embryo development and quality in haploid embryos.

There is another method to produce parthenogenic haploid embryos, which involves removing the male PN from the fertilized zygote (Bai et al., 2016). However, this method requires complex processes, such as fertilization and removing PN by manipulation, and is hard to apply in humans due to the ethical issue of destroying the fertilized embryos. Therefore, the activation of mature oocytes could be a simple and efficient method to produce a parthenogenetic haploid embryo. Further, the electronic fusion of the haploid embryo could induce a higher rate of blastocyst development, which could lead to higher efficiency for the generation of ESCs with homozygosity.

Haploidization of the diploid somatic genome was tried using mouse oocytes (Tateno et al., 2003; Lee et al., 2022). The method described in the current article could be applied in somatic haploid embryos to improve the establishment of somatic homozygous ESCs.

Haploid embryos showed a lower blastocyst rate and the developmental delay from 4-cell to blastocyst stage compared to diploid embryos. When the electronic fusion was performed in haploid embryos, the developmental delay was improved. The rates of blastocyst development and ESC establishment were significantly increased compared to the non-fused group. Based on the result, we concluded that the electronic fusion of haploid embryos could be an efficient method to generate homozygous ESCs.

None.

Conceptualization, Y.L., E.K.; data curation, B.K., Y.L.; formal analysis, B.K., S.S., J.C., Y.L.; funding acquisition, Y.L., E.K.; investigation, B.K., S.S., J.C., Y.L.; methodology, B.K., S.S., J.C., Y.L.; project administration, Y.L., E.K.; resources, B.K., Y.L.; software, B.K., J.C., Y.L.; supervision, Y.L., E.K.; validation, B.K., S.S., J.C., Y.L.; visualization, B.K., Y.L.; writing - original draft, B.K.; writing - review & editing, Y.L., E.K.

This work has been supported by the National Research Foundation of Korea (grant nos. NRF-2018R1A2B3001244 and NRF-2021R1I1A1A01049705).

The animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC No.210035) of CHA University.

Not applicable.

Not applicable.

Not applicable.

No potential conflict of interest relevant to this article was reported.