Cell cycle synchronization of leukemia inhibitory factor (LIF)-dependent porcine-induced pluripotent stem cells and the generation of cloned embryos

Abstract

Nuclear transfer (NT) from porcine iPSC to create cloned piglets is unusually inefficient. Here we examined whether such failure might be related to the cell cycle stage of donor nuclei. Porcine iPSC, derived here from the inner cell mass of blastocysts, have a prolonged S phase and are highly sensitive to drugs normally used for synchronization. However, a double-blocking procedure with 0.3 μM aphidicolin for 10 h followed by 20 ng/ml nocodazole for 4 h arrested 94.3% of the cells at G₂/M and, after release from the block, provided 70.1% cells in the subsequent G₁ phase without causing any significant loss of cell viability or pluripotent phenotype. Nuclei from different cell cycle stages were used as donors for NT to in vitro-matured metaphase II oocytes. G₂/M nuclei were more efficient than either G₁ and S stage nuclei in undergoing first cleavage and in producing blastocysts, but all groups had a high incidence of chromosomal/nuclear abnormalities at 2 h and 6 h compared with non-synchronized NT controls from fetal fibroblasts. Many G₂ embryos extruded a pseudo-second polar body soon after NT and, at blastocyst, tended to be either polyploid or diploid. By contrast, the few G₁ blastocysts that developed were usually mosaic or aneuploid. The poor developmental potential of G₁ nuclei may relate to lack of a G₁/S check point, as the cells become active in DNA synthesis shortly after exit from mitosis. Together, these data provide at least a partial explanation for the almost complete failure to produce cloned piglets from piPSC.

Introduction

The generation of genetically modified pigs has many potential applications in both the livestock industry and biomedical research.^1,2 While the ability to introduce genetic modifications into embryonic stem cells (ESC) competent to contribute to the germ line of chimeric animals revolutionized mouse developmental genetics,^3,4 there has been no equivalent progress over the past 25 y in large farm species, including the pig, in large part because no suitable embryonic stem cell lines capable of a germ cell contribution appear to exist. Instead, the introduction of genetic change through use of ESC has been abandoned in favor a dual technology that employs genetic modification of somatic cells followed by nuclear transfer (NT) to create heterozygous cloned animals that can be inter-bred to yield the desired homozygote. Recent development and application of CRISPR,⁵ TALENs,⁶ and zinc finger nuclease technologies⁷ have enabled the production of knockout animals with higher efficiency than standard homologous recombination. However, somatic cells have a poor ability to proliferate continuously in culture relative to pluripotent stem cells, which can be maintained through large numbers of doublings without senescence, thereby reducing the potential to select for particular genetic changes.^8-11 In addition, somatic cells are thought to be more difficult to reprogram during NT than ESC.¹² The persistence of donor-cell-specific gene expression in NT embryos may also contribute to abnormalities in offspring born as a result of somatic cell NT.^13,14 These characteristics of somatic cells have probably limited the efficiency of NT to create genetically modified pigs.

Although there is a paucity of authenticated porcine ESC lines, the production of porcine induced pluripotent stem cells (piPSC) technology has been quite straightforward through use of defined cocktails of transcription factors,¹⁵ especially through transduction with reprogramming factors, usually the classic combination of POU5F1, SOX2, MYC, and KLF4.^16-19 All the initial piPSC lines generated were FGF2-dependent and of the primed/epiblast type, with colony morphologies resembling those of human ESC and iPSC. More recently, our laboratory generated an additional piPSC line from the inner cell mass (ICM) of porcine blastocysts (pICM-iPSC) by upregulating the expression of KLF4 and POU5F1 with a Tet-inducible lentiviral vector system.²⁰ These cells are leukemia inhibitory factor (LIF)-dependent and of the so-called naïve type with a colony morphology similar to that of mouse ESC. Compared with primed/epiblast type stem cells, the naïve type stem cells proliferate rapidly, appear immortal, and can be dissociated into single cells by trypsin-like proteinases for routine sub-culture without inducing apoptosis. Accordingly, we predicted that such cells may be superior to the primed/epiblast type stem cells for cryopreservation and transfection, as well as a source of donor nuclei during NT. Moreover, as pICM-iPSC were derived from the undifferentiated porcine ICM, we anticipated that they would lack the “epigenetic memory” of somatic cell types and, hence, be more readily reprogrammed within the cytoplasm of the oocyte after NT, thereby providing more efficient cloning and fewer abnormalities in offspring born.

In order for NT to work well, it is important to ensure cell cycle coordination between the nuclear donor and recipient cytoplasm of the oocyte. For example, experiments with mice indicate that it is probably best to transfer diploid nuclei from the G₀/G₁ phase of the cell cycle when using metaphase II stage oocytes as recipient cytoplasts^21,22 and avoid cells that are in S or G₂. However, even in mice, only 15% of reconstructed embryos derived from ESC developed to blastocysts, while the success rate from differentiated ovarian cumulus cells and tail-tip cells was much greater (50–60% blastocyst formation).^23-25 Despite this apparent lack of efficiency, as donors in embryo transfer, the potential of a cloned blastocyst, once formed, to provide a viable pup was higher if the original donor nucleus had been from an ESC rather than from a somatic cell.^12,23,26,27 Thus, initial pre-implantation development of a reconstructed embryo may depend upon cell cycle stage of the donor nucleus, whereas post-implantation development is strongly influenced by the epigenetic status of the donor nucleus. In the manuscript that follows, we have sought to develop a cell cycle synchronization protocol to provide nuclei from pICM-iPSC that are the most suitable donors in NT.

Results

Preimplantation development of NT embryos and cell cycle distribution of pICM-iPSC and porcine fetal fibroblasts (PFF)

All experiments were performed with in vitro-matured oocytes. Importantly, for evaluating the experiments that follow, depending upon the batch, only about 25–35% of such oocytes when fertilized in vitro and cultured under optimal conditions provide blastocysts within 6 d.

First, we compared the preimplantation development of NT embryos from pICM-iPSC and PFF. The pICM-iPSC provided lower initial cleavage assessed at 24 h post-NT and fewer blastocysts at 6 d than the PFF (Fig. 1A). Cell numbers, however, did not differ between blastocysts derived from the 2 different cell types.

Figure 1. Preimplantation development of NT embryos (A) and cell cycle distribution of unsynchronized pICM-iPSC and PFF (B). (A) The data of preimplantation development are from 4 experimental replicates employing a total of 419 reconstructed embryos. Values are means ± SEM. Values with different lowercase letters are significantly different (P < 0.05). (B) The distribution of cell stages in unsynchronized pICM-iPSC and PFF populations as measured by flow cytometry. Three independent cell samples from each cell type group were analyzed. Values are means ± SEM.

Next, we examined the distribution of cell cycle stages across pICM-iPSC and PFF by flow cytometry. Both types of cell were in logarithmic growth and collected as single cell suspensions at day 3 after routine passage. Results were highly consistent between experiments. While over two-thirds of the PFF were in G₁ (67.3 ± 2.1%), less than one-third of the faster dividing pICM-iPSC at that stage of the cycle (30.8 ± 1.1%), while the remainder were in either G₂ (29.7 ± 1.1%) or S (39.4 ± 1.5%) (Fig. 1B).

Dose response experiments of lovastatin, aphidicolin (APN), and nocodazole (NCDZ)

To test the ability of 3 potential cell cycle inhibitors, lovastatin, APN, and NCDZ, each drug was tested singly over a range of concentrations to determine its toxicity (as judged by trypan blue exclusion) and ability to block the cells at an anticipated stage of the cell cycle (as measured by flow cytometry) (Fig. 2A–C). Since the cell doubling time for the untreated pICM-iPSC was around 9.5 h, the cells were exposed to the inhibitors for 10 h in order to encompass all cell cycle phases. Overall, these pluripotent cells were much more sensitive to the drugs than previously tested somatic cells. For example, 10 h exposures to 5 μM lovastatin, 10 ng/ml NCDZ, or 0.6 μM APN caused significantly increased cell death compared with untreated controls (Fig. 2A–C), whereas somatic cell types can withstand concentrations a magnitude higher than these.^28-30 Of the 3 drugs, lovastatin, which was expected to pause the cell cycle in G₁, had no ability to mediate a significant cell cycle blockage. By contrast, 5 ng/ml NCDZ provided 49.5 ± 1.8% of the cells at G₂/M, while 0.3 μM APN (10 h) arrested 74.3 ± 3.2% of the cells in S phase, with neither compound causing significant increase in cell death at those concentrations (Fig. 2A–C).

Figure 2. Effects of increasing concentrations of lovastatin (A), NCDZ (B), and APN (C) on cell viability and cell cycle parameters of pICM-iPSC. Values for cell viability marked with asterisks are significantly different from controls (P < 0.05). DMSO vehicle-treated control was only tested at the highest concentration. There was no difference in cell viability between the DMSO vehicle-treated control and the non-treated control. (D) The relatively poor ability of 0.3 μM APN to synchronize the pICM-iPSC subsequent to its block at the S phase. Cell cycle distribution was analyzed every 60 min after release from 0.3 μM APN treatment.

APN was clearly more effective in blocking cell cycle progression of pICM-iPSC than either lovastatin or NCDZ. In an attempt to collect G₁-phase pICM-iPSC, cells were treated with 0.3 μM APN for 10 h and then released from inhibition to allow the synchronized cells to progress to the next G₁ phase. However, by 6 h following release from APN inhibition, only 52.1 ± 0.9% of the cells had reached the subsequent G₁ phase. It would appear that 10 h APN treatment could not sustain synchronization beyond a single cell cycle (Fig. 2D).

Development of the double-blocking procedure

Since no single drug could effectively synchronize the pICM-iPSC population beyond the cell cycle in which it was administered, a double-blocking procedure was developed. The cells were first arrested at S phase by APN (0.3 μM, 10 h), provided medium without any inhibitor for 1 h (as a “wash out” period and to release the block), and then treated with NCDZ to pause the cells at G₂/M. Four different concentrations and 3 different times of exposure to NCDZ were tested in order to provide the most efficient protocol. With a 4 h treatment (Fig. 3A), 40 ng/ml of NCDZ caused the majority of cells to halt in S phase, emphasizing its likely toxicity, whereas 5 ng/ml was too low a concentration to be effective (Fig. 3A). On the other hand, both 10 and 20 ng/ml NCDZ could efficiently arrest the cells at G₂/M. The time of NCDZ exposure with 10 ng/ml and 20 ng/ml NCDZ, however, was not particularly critical (Fig. 3B). In all further synchronization experiments with the double-blocking procedure we, therefore, employed 20 ng/ml NCDZ for 4 h, because this combination of concentration and exposure appeared to provide the highest number of cells at G₂/M (Fig. 3B).

Figure 3. Development of the double-blocking procedure. (A) Effect of increasing NCDZ concentrations on cell cycle distribution of pICM-iPSC measured by flow cytometry. pICM-iPSC were first arrested at S phase by APN (0.3 μM, 10 h), provided medium without any inhibitor for 1 h to release the block, and then treated with NCDZ at different concentrations for 4 h. (B) Effect of NCDZ treatment time on the cell cycle distribution of pICM-iPSC measured by flow cytometry. pICM-iPSC were treated with 10 or 20 ng/ml NCDZ for 2, 3, or 4 h, either after no prior treatment with APN (2 left panels) or following APN treatment (0.3 μM, 10 h), i.e., they received the double block (2 right panels). (C) The ability of the selected double-blocking procedure to synchronize the pICM-iPSC subsequent to the release of the cells from the block at the G₂/M phase. Cell cycle distributions were analyzed by flow cytometry every 60 min.

An initial blockage by APN and the second treatment by NCDZ were necessary to achieve efficient synchronization, because NCDZ alone was relatively ineffective (Fig. 3B). The double-blocking procedure also had no detrimental effects on either cell viability (Fig. 4A) or cell doubling times (Fig. 4B) assessed 0, 12, 24, and 36 h after treatment. The increased cell doubling time of all cells at 24 h and 36 h relative to 12 h after release from the second block probably reflected loss of critical growth factors from the medium as the cells continued to proliferate. This phenomenon is particularly evident for the control cells at 36 h, which had an approximate 14 h “lead” over the synchronized cells. Expression of pluripotent markers POU5F1, SOX2, and SSEA1 appeared not to be affected by the drug exposure (Fig. 4C).

Figure 4. Effects of the double-blocking procedure on cell viability (A), cell doubling time (B), and features of a pluripotent phenotype (C). The pICM-iPSC were treated with 20 ng/ml NCDZ for either 2 or 4 h to provide the second blockage at G₂/M. Cell viability subsequent to release from the block was measured by the trypan blue exclusion assay. Cell doubling time was calculated based on the cell numbers at 0, 12, 24, and 36 h after release from the double-blocking procedure. Expression of POU5F1, SOX2, and SSEA1 at 12 h after release from the block was examined by immunofluorescence. pICM-iPSC, cultured without any inhibitor, were used as the control group. Scale bar = 100 μm.

After release from NCDZ inhibition, the majority of cells divided and entered G₁ within 2–3 h, when 70.1 ± 2.2% and 71.8 ± 0.6%, respectively, of the cells were diploid. By 11 h, 68.2 ± 2.2% of the cells had progressed to the next G₂/M phase. Thus, reasonably good cell cycle synchronization could be maintained for more than one cell cycle by using the double-blocking procedure (Fig. 3C).

NT embryo development of synchronized pICM-iPSC at different cell cycle stages

As cells in G₁ were anticipated to provide the best source of nuclei for preimplantation development, G₁ phase pICM-iPSC were harvested at 2 h after release from the double block. Following NT, there were no significant differences between the G₁ pICM-iPSC and unsynchronized control cells in total cleavage, blastocyst formation, and blastocyst cell numbers (Fig. 5A). Therefore, synchronization of pICM-iPSC to the G₁ phase of the cell cycle did not improve preimplantation development of the NT embryos relative to controls.

Figure 5. Embryo development after performing NT with synchronized pICM-iPSC. (A) Comparison of preimplantation development of NT embryos derived from G₁ pICM-iPSC and PFF. This experiment was replicated 5 times, with a total of 575 reconstructed embryos. Values are means ± SEM. Values with different lowercase letters are significantly different (P < 0.05). (B) Comparison of preimplantation development of NT embryos derived from pICM-iPSC at different cell cycle stages. This experiment was replicated 4 times, with a total of 682 reconstructed embryos. Values are means ± SEM. Values with different lowercase letters are significantly different (P < 0.05).

We then investigated whether other phases of the cell cycle provided more efficient NT than G₁. To obtain populations in which the majority of cells were in S, pICM-iPSC were harvested immediately after release from APN inhibition (0.3 μM, 10 h). By contrast, G₂ phase cells were harvested 0 h after the double-blocking procedure. Unsynchronized pICM-iPSC were used as the control group. The developmental competence of NT embryos derived from S-phase pICM-iPSC was very low. Few cleaved within 24 h, and only about 5% reached blastocyst (Fig. 5B). G₂ phase nuclei, however, provided a rate of cleavage and development to blastocyst comparable to the control nuclei and were superior in this regard to nuclei provided by both the G₁ and S groups.

Assessment of DNA status at 2, 6, 24 h after NT

Nuclear morphology was then examined at increasing times after NT for the different groups of synchronized cells, with NT embryos from unsynchronized PFF acting as the controls. During normal NT embryo development, the donor nucleus undergoes nuclear envelope breakdown and premature chromosome condensation within a few hours after NT. Then, a single pseudo-pronucleus begins to form, and embryos enter the so-called pronuclear stage around 6 h and initiate DNA synthesis. Extrusion of a pseudo-second polar body extrusion may also occur at this time in a number of NT embryos. Around 24 h, the transferred nucleus completes first mitosis, and the embryo cleaves to form 2 blastomeres.^21,31 Accordingly, we collected NT embryos at 2, 6, and 24 h to examine whether this predicted series of events ensued, and whether there were accompanying chromosomal abnormalities and cleavage failures. All transferred nuclei underwent nuclear envelope breakdown and premature nuclear condensation by 2 h after NT (Fig. 6A–D). In many cases, the nuclei had not only condensed, but had formed an organized metaphase plate (Fig. 6A and B). In others, chromosomal organization appeared disrupted (Fig. 6C and D). As shown in Figure 7A, S-phase-derived embryos provided significantly higher numbers of observable chromosomal abnormalities than those from G₁, G₂, and control cells 2 h after NT. By 6 h, the percentage of embryos forming a single, normal-appearing, pseudo-pronucleus (see Fig. 6E and F) was higher (~80%) in the control (PFF) than in any of the pICM-iPSC groups (Fig. 7B). Conversely, numbers of embryos with condensed DNA, and otherwise abnormal chromosomal features were significantly lower in the controls than in the 3 other NT groups, while embryos with multiple pseudo-pronuclei were relatively common in all groups (Fig. 7B). By 24 h, approximately 40% of the PFF group had cleaved (Fig. 7C). The G₂ group had also cleaved relatively efficiently by 24 h (see Fig. 6N). Moreover, G₁- and S-phase nuclei provided higher percentages of embryos with multiple pseudo-pronuclei (see Fig. 6O and P) than either G₂ or PFF nuclei (Fig. 7C). Finally, at 24 h, embryos derived from G₂ and G₁ nuclei provided higher numbers (P < 0.05) of embryos with extruded polar bodies than S-phase nuclei (Fig. 7D). Polar body extrusion for the G₂-phase nuclei averaged approximately 40% and approached but did not reach significance when compared with the G₁ and PFF groups.

Figure 6. Representative nuclear morphologies of pICM-iPSC-derived embryos. White arrows are directed toward nuclear material. Black arrows indicate pseudo-second polar bodies. From left to right, pictures in each column represent nuclear morphologies of the NT embryos derived from unsynchronized PFF, and pICM-iPSC synchronized at G₂, G₁, and S phases, respectively. From top to bottom, images in each row represent nuclear morphologies of NT embryos at 2, 6 and 24 h after NT, respectively. (A and B) Embryos with condensed chromosomes; (C, D, and H) abnormal chromosomes with disrupted organization; (E) embryo with a single pseudo-pronucleus; (F) embryo with a single pseudo-pronucleus and with a pseudo-second polar body; (G) embryo with several pseudo-pronuclei; (I) embryo with metaphase chromosomes at the spindle equator; (J) embryo with metaphase chromosomes at the spindle equator and a pseudo-second polar body; (K and L) embryos with multiple pseudo-pronuclei; (M) two-cell stage embryo demonstrating apparently normal cleavage with 1 nucleus in each blastomeres; (N) two-cell stage embryo with normal cleavage and a pseudo-second polar body; (O and P) two-cell stage embryo after abnormal cleavage, with multiple nuclei appearing in 1 blastomere.

Figure 7. Assessment of chromosomal status of NT embryos at 2, 6, and 24 h after NT. (A) Two hours after NT. This experiment was replicated 3 times with a total of 122 reconstructed embryos. Embryos with condensed chromosomes were considered normal, while embryos with a disrupted chromosomal organization were classified as abnormal. (B) Six hours after NT. This experiment was replicated 3 times with a total of 125 reconstructed embryos. Embryos were classified into 4 types: ones with condensed chromosomes; embryos with a single pseudo-pronucleus (1 PPN); embryos with multiple pseudo-pronuclei (multi-PPN); embryos with disrupted chromosomes (abnormal chromosome). (C) Nuclear status at 24 h after NT. This experiment was replicated 3 times with a total of 166 reconstructed embryos. Embryos were classified into 6 types: embryos with condensed or disrupted chromosomes (condensed/abnormal chromosome); embryos with a single pseudo-pronucleus (1 PPN); embryos with multiple pseudo-pronuclei (multi-PPN); embryos with chromosomes aligned at the spindle equator (chromosome segregation); cleaved embryos with 1 nucleus in each blastomere (cleavage); cleaved embryos with multiple nuclei in at least 1 blastomere at the 2-cell stage (abnormal cleavage). (D) The percentages of embryos with pseudo-second polar body extrusion at 24 h after NT. Embryos with condensed or abnormal chromosomes were excluded from this calculation. In all figures, values are means ± SEM. Values within the same clusters with different lowercase letters are significantly different (P < 0.05).

Nuclear ploidy of NT blastocysts from pICM-iPSC at different cell cycle stage

The chromosomal spreads were analyzed and classified as haploid (Fig. 8A), diploid (Fig. 8B), or polyploid (Fig. 8C and D), with approximately 19, 38, or more than 57 chromosomes, respectively. The mosaic embryos were those with a mixture of diploid and aneuploid cells. The “aneuploid pooled” category is the sum of polyploid and mosaic embryos. The percentage of normal diploid blastocysts was higher in the PFF group (72.7%) and the in vitro fertilization (IVF) control group (69.0%) than in the G₁ (36.4%) and G₂ (25.0%) groups. The G₁ pICM-iPSC group had proportionately more mosaic blastocysts than the others. Whereas the majority of G₂ pICM-iPSC-derived blastocysts were polyploid, a contingent of diploid embryos was also evident (Fig. 8E).

Figure 8. Blastocyst ploidy assessment and representative images of chromosomal spreads. (A) Haploid nucleus with 19 chromosomes; (B) diploid nucleus with 38 chromosomes; (C) triploid nucleus with 57 chromosomes; (D) tetraploid nucleus with 76 chromosomes. (E) Nuclear ploidy of NT blastocysts. A total of 75 NT blastocysts were processed for analysis. Overall, it was possible to obtain analyzable chromosomal spreads from 36 of the fixed NT blastocysts. Blastocysts in which all analyzed nuclei were diploid cells were scored as diploid. Those in which all nuclei were polyploid were classified as polyploid. Blastocysts containing both diploid cells and cells with another chromosomal complement were considered mosaic. The “aneuploid pooled” category is the sum of polyploid and mosaic embryos.

Analysis of DNA synthesis after double-blocking procedure

To assess the time of onset of DNA synthesis following mitosis and achieving the diploid state, pICM-iPSC were provided with bromodeoxyuridine (BrdU) for 1 h at 5 time points (1 h prior to release from the block, at the time of release from the block, and at intervals of 1 hour thereafter). Incorporation was then assessed for each time point, i.e., at the termination of the block when they were about to enter G₁, and at 4 subsequent hourly time points. Two hours after release from the block, cells with incorporated label were detectable (Fig. 9A and B), but the signal had risen significantly above background by 3 h (Fig. 9B), when it appeared that the majority of, if not all, cells in the colonies had incorporated label (Fig. 9A). Intensity of the nuclear signal continued to increase as the cells began to enter the predicted S-phase of the cell cycle 4 h after release from the block (Fig. 3C).

Figure 9. Assessment of DNA synthesis in the hours subsequent to the metaphase block. The pICM-iPSC were arrested at G₂/M by the double-blocking procedure and then released from inhibition to allow the synchronized cells to progress to the next G₁ phase. BrdU was added to the culture medium for 1 hour at 0 h, 1 h, 2 h, 3 h, or 4 h after release from the block. (A) Representative images after immunostaining for BrdU incorporation at each time point. Scale bar, 100 μm. (B) Quantitative assessment of BrdU nuclear labeling at different times after release from mitosis. The BrdU labeling index was calculated as the intensity of BrdU staining normalized to the intensity of DAPI staining on the same image. Values are means ± SEM. Values with different lowercase letters are significantly different (P < 0.05).

Discussion

Pluripotent cells are characterized by a high rate of proliferation, a short G₁ phase, and an extended S phase.³² These characteristics probably place limitations on the ability of pluripotent stem cells to act as nuclear donors in embryo reconstitution experiments. As LIF-dependent pICM-iPSC proliferate rapidly, yet have properties that might make them useful for genetic modification of swine, we sought to develop a synchronization method to provide an enrichment of G₁ cells relative to the other 2 phases of the cell cycle. Three cell cycle inhibitors, lovastatin, APN and NCDZ, were first tested individually. Lovastatin is believed to arrest cells in early G₁ by interfering with the proteasome pathway and inhibition of CDK2 activity,^28,33,34 while APN is a reversible inhibitor of DNA polymerases and blocks the cell cycle at the early S phase.³⁵ NCDZ, like colcemid, interferes with mitotic spindle formation to arrest cells at prometaphase with a G₂/M-phase DNA content.³⁶ Why lovastatin failed to provide even an inefficient block in G₁ for pICM-iPSC is unclear. APN alone was somewhat more efficient. At a concentration of 0.3 μM for 10 h, the proportion of S-phase cells in the population increased from 39.4% to 74.3% without causing significant cell death (Fig. 2C). Nonetheless, synchronization was not efficient (Fig. 2D). We speculate that the high number of cells in the extended S phase were blocked asynchronously and provided a non-uniform population when the block was removed. NCDZ, the third reagent tested, is generally employed between 40 to 200 ng/ml,²⁹ but was toxic to pICM-iPSC well below these concentrations, such that 10 ng/ml for 10 h only provided an approximate 60% accumulation of cells at G₂/M phase, but also significant cell death (Fig. 2B). To overcome these difficulties, we devised a double-blocking procedure that combined an initial 10 h exposure to 0.3 µM APN followed 1 h later by a relatively high concentration (20 ng/ml) of NCDZ for a shorter period (4 h) than when it was tested alone. This protocol produced over 90% of the cells at G₂/M, and, after the block was removed, about 75% in G₁ 2–4 h later without causing significant loss of either viability (Fig. 4A) or loss of pluripotency markers (Fig. 4C). Clearly such a double-blocking procedure might be useful with other rapidly growing cell types that are also abnormally sensitive to the drugs used for synchronization, although concentrations and treatment times would need to be adjusted for each cell type.

However, such G₁-phase nuclei from synchronized G₁ pICM-iPSC obtained at 2 h after double-blocking procedure did not improve performance in NT experiments relative to unsynchronized cells (Fig. 5A). Not unexpectedly, the S-phase pICM-iPSC had the least ability to support the development of NT embryos (Fig. 5B). Perhaps most surprising was the better functioning of G₂-phase nuclei in providing embryo development to blastocyst than nuclei from G₁-phase cells (Fig. 5B).

One likely reason for the poor performance of the G₁ nuclei for NT from ESC of various types is the lack of a G₁ checkpoint ^37-39 within an abbreviated G₁ phase.⁴⁰ As shown in Figure 9, the pICM-iPSC began to replicate their DNA very shortly after exit from mitosis, yet were still technically in G₁ as defined by flow cytometry (Fig. 3C). S-phase nuclei are notoriously poorer NT donors than G₁ nuclei ⁴¹ and give rise to poorly organized and disrupted metaphase plates prior to first cleavage.⁴² We observed similar aberrations with both the S and G₁ nuclei from pICM-iPSC (Fig. 6C and D), although they appeared to occur less frequently with the latter (Fig. 7A). In addition, both G₁ and S nuclei gave rise to a high percentage of multiple pseudo-pronuclei 24 h after NT (Figs. 6K and L and 7C), possibly as a result of nuclear envelope material enclosing the disorganized and dispersed chromosomes noted at the earlier 2 h time point. A correlation between chromosome “scattering” and multiple pseudo-pronuclei after NT has been noted previously.⁴³ DNA abnormalities clearly persisted in the cloned S and G₁ embryos that cleaved (Fig. 7C) and likely contribute to their poor development from the 2-cell stage to blastocyst. Indeed, of those blastocysts that did form from G₁ nuclei, about two-thirds had abnormal chromosome complements, presumably related to the disorganized nuclear morphologies and multiple pseudo-pronuclei associated with these embryos before they were first cleaved. By contrast the control, unsynchronized PFF nuclei and a second control group from non-manipulated embryos generated by IVF produced blastocysts that were predominantly diploid (Fig. 8E). Together, the data show that G₁ nuclei from pICM-iPSC are not suitable for NT into pig MII-oocytes, most likely because they are not in a stable diploid state.

A higher proportion of G₂ reconstituted embryos divided and progressed to blastocyst than G₁-derived embryos. These embryos also tended to extrude a pseudo-second polar body at first mitosis, whereas the other groups did so less frequently (Fig. 7D), and, thus, they might be “correcting” their ploidy from 4C to 2C at first cleavage. However, 58% (7/12) of the blastocysts derived from G₂ pICM-iPSC were still polyploid and only 25% (3 of 12) diploid. Thus, the ability of these embryos to correct their tetraploid status was limited.

Recently, Fan et al.⁴⁴ concluded that it was not possible to produce cloned pigs from unsynchronized iPSC. Although this lack of success was attributed to the continued ectopic expression of the transgenes used in reprogramming, this hypothesis was not tested, e.g., by using cells reprogrammed with non-integrating vectors. We have also had no success in 4 attempts to produce cloned piglets from pICM-iPSC (data not shown), but suggest that the reason for failure is an inability to retain a stable diploid complement of chromosomes in embryos reconstituted from such cells. If our explanation is correct, we must ask why it has been possible to generate cloned pups from mouse ESC^23,26,27 and iPSC.^45,46 One observation consistent with ours was the unexpectedly low efficiency achieved with G₁-stage mouse iPSC to give rise to blastocysts.⁴⁵ A second was the relatively high efficiency of G₂/M-phase nuclei. Conceivably, the extrusion of a pseudo-second polar body after NT with G₂/M cells^27,45,47 may be the key to cloning success. Pig MII oocytes may be less efficient than mouse oocytes in permitting the extrusion of “extra” DNA. Nevertheless, if cloning is to be performed successfully with pig iPSC, it might be best attempted with cells synchronized to the G₂ phase of the cell cycle.

Materials and Methods

Chemicals

All chemicals were purchased from Sigma Aldrich unless specified otherwise.

Cell culture

Procedures for pICM-iPSC line culture were performed as previously described with slight modification.²⁰ Briefly, cells were cultured on a laminin-coated substratum with N2B27-3i medium (1:1 ratio of DMEM/F12 and Neurobasal medium [Life Technologies], supplemented with N2B27 [Life Technologies] and 3 inhibitors, CHIR99021 [GSK3 kinase inhibitor, Stemgent], PD032591 [MEK inhibitor, Stemgent], and PD173074 [FGF tyrosine kinase receptor inhibitor]) supplemented with 2 μg/ml doxycycline (Stemgent), and 1000 unit/ml human LIF (Millipore). Cells were cultured at 38.5 °C in a 5% O₂, 5% CO₂, and 90% N₂ atmosphere with daily exchange of fresh medium and passaged every 3–4 d by using Accutase (Millipore).

Population doubling time and cell viability assay

Population doubling times for pICM-iPSC were estimated as previously described.^18,20 pICM-iPSC were seeded at 2 × 10⁴ cells/well on 12-well plates coated with laminin under the culture conditions described above. The number of cells in each well was counted at 0, 12, 24, and 36 h after the cell cycle inhibition. Cells were stained with 0.4% trypan blue dye (Bio-Rad) prior to cell count. The cell number and cell viability at each time point were determined by TC10 automated cell counter (Bio-Rad) on 3 independent cell samples from each treatment group. Cell doubling time was calculated at 12, 24, and 36 h after the cell cycle inhibition using the formula T_d = (t₂ – t₁) × log (2)/log (q₂/q₁), in which T_d is the doubling time and q₁ and q₂ are the cell numbers at time t₁ and t₂.

Flow cytometry analysis of cell cycle distribution

Cells were harvested and resuspended in phosphate buffered saline (PBS) and then fixed in cooled 70% ethanol at 4 °C for 30 min. The cells were incubated in a solution containing 1 mg/ml RNase A, 20 μg/ml propidium iodide (Fisher), and 0.05% Triton X-100 (Fisher) at 37 °C for 40 min. Subsequently, the cells were resuspended in 500 μl PBS. Cell cycle analysis was performed on at least 3 independent cell samples. For each cell population, at least 10 000 cells were analyzed in the Accuri C6 Flow Cytometer (BD Biosciences), and the proportion in G₀/G₁, S, and G₂/M phases was estimated using the FlowJo7.6.5 software (Tree Star Inc).

Immunohistochemistry

Cells grown on coverslips were fixed in 4% paraformaldehyde in PBS for 15 min at room temperature, washed, and exposed to either 5% goat serum or 5% donkey serum, 1% bovine serum albumin (BSA), and 0.1% Triton X-100 (Fisher) in PBS for 30 min. The fixed specimens were then incubated with primary antibody at 4 °C overnight. After washing, they were incubated with secondary antibody. A secondary antibody-only staining of the coverslips was used as a control. VECTASHIELD mounting medium with DAPI (Vector Laboratories) was used to mount the coverslips. Primary antibodies were POU5F1 (1:100), SOX2 (1:1000; Millipore), SSEA-1 (1:50; Developmental Studies Hybridoma Bank [DSHB]), and BrdU (1:50; DSHB). Secondary antibodies were Alexa Fluor 546-conjugated goat anti-rabbit IgG (POU5F1), Alexa Fluor 546-conjugated goat anti-mouse IgG (SOX2), Alexa Fluor 488-conjugated goat anti-mouse IgM (SSEA-1), and Alexa Fluor 568-conjugated goat anti-mouse IgG (BrdU) (all from Life Technologies).

NT and in vitro embryo culture

NT was performed as previously described, with slight modification.^48,49 Sow oocytes were purchased from ART and shipped overnight to Columbia, MI, in TCM 199 maturation medium. After 40–42 h of maturation, cumulus cells were removed by vortexing for 3 min in the presence of 0.01% hyaluronidase. Oocytes with a visible polar body were placed in manipulation medium (TCM199 with 0.6 mM NaHCO₃, 2.9 mM Hepes, pH 7.3, 30 mM NaCl, 10 ng/ml gentamicin, and 3 mg/ml BSA, 7.0 mg/ml cytochalasin B; at a final osmolarity of 305). The polar body along with a portion of the adjacent cytoplasm was removed, and a donor cell was placed in the perivitelline space. The reconstructed embryos were then fused in fusion medium (0.3 M mannitol, 0.1 mM CaCl₂, 0.1 mM MgCl₂, and 0.5 mM Hepes buffer) by 2 direct-current (DC) pulses (1 s interval) at 1.2 kV/cm for 30 μsec. After fusion, embryos were chemically activated with 200 mM thimerosal for 10 min in the dark and 8 mM dithiothreitol for 30 min.⁵⁰ Embryos were incubated in porcine zygote medium 3(PZM3)⁵¹ with 0.5 mM Scriptaid for 14–16 h and then cultured in PZM3 at 38.5 °C in 5% O₂, 5% CO₂, 90% N₂ for 6 d. Percent of cleaved embryos was evaluated at 24 h (d 1) and total cleavage at 48 h after activation. Blastocyst formation was evaluated at day 6 of culture. Day 6 blastocysts were stained with 0.01 mg/ml Hoechst 33342 to determine total cell number under a fluorescence microscope.

Assessment of chromosomal status after NT

Reconstructed embryos at 2, 6, and 24 h after NT were mounted onto glass slides and fixed for 48 h in ethanol-acetic acid (3:1) at 4 °C. Fixed embryos were then stained with 1% (w/v) orcein in 45% (v/v) acetic acid. The morphology of chromosomes was assessed under bright-field microscopy (magnification 400×). Reconstructed embryos (n = 413) in 3 replicates were evaluated in this experiment, with around 10–20 reconstructed embryos from each group examined at each time point.

Blastocyst ploidy assessment by cytogenetic analysis

Cytogenetic analysis was performed with modification of a previously described procedure.^52,53 Briefly, day 6 blastocysts were incubated in culture medium with 0.2 μg/ml colcemid for 3 h at 38.5 °C in 5% O₂, 5% CO₂, 90% N₂. They were then removed from the medium, and the zona pellucida removed by treatment with 0.005% (w/v) pronase in PBS. The denuded blastocysts were then immersed in 0.8% (w/v) sodium citrate for 3 min, followed by 75 mM KCl for 2 min. Blastocysts were fixed overnight in ethanol-acetic acid (3:1) and placed on clean glass slides to allow the chromosomes to spread. Finally they were stained with 5% (v/v) Giemsa dye for 10 min, rinsed in distilled H₂O, and mounted with Permount mounting medium (Fisher) beneath a glass coverslip. Chromosomal spreads were examined under oil-immersion, bright-field microscopy (magnification 1000×) to determine chromosome number. Nuclei with distinguishable chromosomes were classified on the basis of the chromosome numbers: haploid, diploid, and polyploid nuclei were those with approximately 19, 38, and 57–76 chromosomes, respectively. Embryos in which all analyzed nuclei were diploid were scored as diploid embryos. Those in which all nuclei were polyploid were considered polyploid embryos. Embryos containing a mixture of diploid cells and cells with haploid or polyploid cells were considered mosaic. Two replicates with 75 NT blastocysts were analyzed.

Analysis of DNA synthesis by BrdU labeling

BrdU labeling was performed by adding 20 μg/ml BrdU to the culture medium 1 h before sample fixation. Cells were then fixed in 4% paraformaldehyde in PBS for 15 min, and incubated with 1.5 M HCl for 30 min at room temperature before proceeding with immunohistochemistry as described above. All images were taken under the same settings on an Olympus Provis AX-70 model inverted microscope equipped with a cooled, charge-coupled device camera (Photometrics CoolSnap-ES). Intensity of the florescence on each image was measured by ImageJ software.⁵⁴ The BrdU labeling index was calculated by the intensity of BrdU staining normalized to the intensity of DAPI staining on the same image. Three independent cell samples were collected at each time point.

Statistical analysis

Data were analyzed by using one-way ANOVA, with treatment included in the model as a fixed factor. Percentage data were arcsine transformed. Differences were determined by Fisher LSD multiple comparison test when treatment was significant. Significance was categorized as P < 0.05. Data are reported as means ± SEM.

Glossary

Abbreviations:

ESC

embryonic stem cells

NT

nuclear transfer

piPSC

porcine induced pluripotent stem cells

ICM

inner cell mass

pICM-iPSC

piPSC line from ICM of porcine blastocysts

LIF

leukemia inhibitory factor

PFF

porcine fetal fibroblasts

APN

aphidicolin

NCDZ

nocodazole

IVF

in vitro fertilization

BrdU

bromodeoxyuridine

PBS

phosphate buffered saline

BSA

bovine serum albumin

PZM3

porcine zygote medium 3