Chemical simulation of hypoxia in donor cells improves development of somatic cell nuclear transfer‐derived embryos and increases abundance of transcripts related to glycolysis

Abstract To improve efficiency of somatic cell nuclear transfer (SCNT), it is necessary to modify differentiated donor cells to become more amendable for reprogramming by the oocyte cytoplasm. A key feature that distinguishes somatic/differentiated cells from embryonic/undifferentiated cells is cellular metabolism, with somatic cells using oxidative phosphorylation (OXPHOS) while embryonic cells utilize glycolysis. Inducing metabolic reprogramming in donor cells could improve SCNT efficiency by priming cells to become more embryonic in nature before SCNT hypoxia inducible factor 1‐α (HIF1‐α), a transcription factor that allows for cell survival in low oxygen, promotes a metabolic switch from OXPHOS to glycolysis. We hypothesized that chemically stabilizing HIF1‐α in donor cells by use of the hypoxia mimetic, cobalt chloride (CoCl2), would promote this metabolic switch in donor cells and subsequently improve the development of SCNT embryos. Donor cell treatment with 100 µM CoCl2 for 24 hr preceding SCNT upregulated messenfer RNA abundance of glycolytic enzymes, improved SCNT development to the blastocyst stage and quality, and affected gene expression in the blastocysts. After transferring blastocysts created from CoCl2‐treated donor cells to surrogates, healthy cloned piglets were produced. Therefore, shifting metabolism toward glycolysis in donor cells by CoCl2 treatment is a simple, economical way of improving the in vitro efficiency of SCNT and is capable of producing live animals.


| INTRODUCTION
Since the birth of the first animal cloned with a somatic cell in 1996, somatic cell nuclear transfer (SCNT) has developed into a useful research tool (Wilmut, Schnieke, McWhir, Kind, & Campbell, 1997).
Today SCNT is used for biomedical models, including xenotransplantation, as well as agricultural models that have led to the discovery of novel treatments for human diseases, animals that are disease resistant, and have put animal-to-human organ transplant within reach (Whitworth & Prather, 2017;Prather, Lorson, Ross, Whyte, & Walters, 2013). Even with the current success of SCNTcreated animals, the overall efficiency of SCNT remains low (<5%) with few live births resulting from the SCNT process (Whitworth & Prather, 2010). Due to the lack of authentic embryonic stem cells and induced pluripotent stem cell lines capable of producing live pigs, porcine SCNT is limited to the use of somatic cell types. Since somatic cells have already undergone some degree of differentiation, a possible explanation for poor SCNT efficiency is the inability to successfully remodel somatic nuclei through the SCNT process. A key feature that distinguishes embryonic/undifferentiated cells from somatic/differentiated cells is the metabolism that is used. Differentiated cells utilize mitochondrial oxidative phosphorylation (OXPHOS), while undifferentiated cells use glycolysis. There is mounting evidence to suggest that metabolic reprogramming, or the switch from OXPHOS to glycolysis, is necessary to revert cells back to an undifferentiated state and maintain stemness (Prigione et al., 2014).
HIFs are a class of master transcription factors responsible for the cellular survival response to hypoxic conditions. HIF stabilization promotes the transcription of target genes related to glycolysis, angiogenesis, cell survival and proliferation, cell migration, apoptosis, and erythropoiesis (Hu, Wang, Chodosh, Keith, & Simon, 2003). Hypoxic stress is alleviated by these downstream targets by modifying the need for oxygen for cellular mechanisms, such as energy production, or allowing for greater oxygen delivery. For example, downstream targets related to glucose metabolism, such as the glucose transporters SLC2A1 and SLC2A3, allow for energy production through glycolysis as opposed to mitochondrial OXPHOS, which can only occur in the presence of oxygen (Semenza, 2000).
Previous studies have shown that donor cell culture in hypoxia (1.25% O 2 ) results in an upregulation of genes related to glycolysis in donor cells, as well as increased blastocyst production and in utero survivability following SCNT (Mordhorst et al., 2018(Mordhorst et al., , 2019. However, hypoxic cell culture can be costly and often requires specialized mixed gas tanks to achieve low oxygen tensions. There is also no reliable way to monitor the oxygen tension that the donor cells are being exposed to when cultured in hypoxia, as it requires culture in chambers that must remain sealed. In addition, HIF 1-α, the modulator of the hypoxic response in cells, has a high turnover rate with degradation occurring in 5-8 min once cells are exposed to atmospheric oxygen levels. During the SCNT process, the time between cell collection and cell-oocyte fusion/activation is typically greater than 1 hr. Therefore, the influence of HIF 1-α in these cells may be greatly diminished by the conclusion of the SCNT process. Due to the possible instability of hypoxia inducible factor 1-α (HIF1-α) in hypoxia cultured cells, we proposed a chemical hypoxia mimetic that allows a sustained effect of HIF1-α outside of physiological hypoxia. In normoxia, HIF1-α is hydroxylated by prolyl hydroxylases that require oxygen and iron for their enzymatic activity.
This hydroxylation serves as a docking site for Von Hippel Lindeau protein that marks HIF1-α for degradation by the 26S proteasome. In hypoxic conditions, the oxygen required for the prolyl hydroxylases is not available; and therefore, the cascade of events leading to HIF1-α degradation cannot be initiated. This allows HIF1-α protein to accumulate in the cytoplasm and subsequently translocate to the nucleus to dimerize with HIF1-β and direct transcription of downstream targets (Semenza, 2000). Cobalt chloride (CoCl 2 ) is a known hypoxia mimetic that inhibits the activity of prolyl hydroxylases by replacing the required iron domain of the prolyl hydroxylases with cobalt (Hirsila et al., 2005). This chemical simulation allows stabilization of the volatile HIF1-α, even in the presence of atmospheric oxygen.
Once stabilized, HIF1-α can activate its downstream targets including genes that induce the reprogramming of metabolic processes to favor glycolytic metabolism over OXPHOS.
Therefore, the objective of this study was to determine if treatment of somatic donor cells with the hypoxia mimetic, CoCl 2 , can induce metabolic reprogramming in the donor cells and promote better nuclear reprogramming before SCNT to improve development of SCNT embryos.

| Gene expression in donor cells following CoCl 2 exposure
Real-time quantitative polymerase chain reaction (PCR) was used to analyze differences in message abundance between CoCl 2 treated donor cells, hypoxia treated donor cells, and untreated control cells (Table 1) for HIF1-α and non HIF1-α gene targets (Liu, Shen, Zhoa, & Chen, 2012). Glucose transporters, SLC2A1 and SLC2A3, as well as glycolytic enzymes HK1, HK2, GPI, ALDOC, GAPDH, PGK1, PGAM1, ENO1, PKM2, PDK1, and LDHA were upregulated in the CoCl 2 group compared with the control. The same transcripts, with the exception of SLC2A1, ALDOC, GAPDH, and PGAM1 were also upregulated in the hypoxia group compared with the control. Transcript abundance of the mitophagy-associated gene BNIP3, GPI, and PDK1 were differentially expressed between all treatment groups with the lowest expression present in the control cells and the highest expression in the CoCl 2 cells. Non HIF1-α targets, TALDO1, EPAS1, YWHAG, LDHB, and BCL2 were not differentially expressed between the groups.

| SCNT embryo development and quality
The use of CoCl 2 -treated donor cells for SCNT resulted in an increased rate of development to the blastocyst stage compared with untreated control donor cells (50.3 ± 2.6% vs. 32.6 ± 1.9%; p = .0002;  Table 2).

| Gene expression in SCNT blastocyst stage embryos produced by CoCl 2 donor cells
Genes that were evaluated in donor cells were also analyzed in blastocyst-stage embryos created with CoCl 2 treated donor cells and blastocyst-stage embryos created from untreated control cells (Table 3). Of the genes evaluated, SLC2A1, PGAM1, and LDHA were upregulated in Day 6 blastocyst-stage embryos created from CoCl 2 treated donor cells compared with control donor cells (p < .05).
F I G U R E 2 Cell viability following a 72 hr recovery period after treatment with 0, 50, 100, or 150 µM of CoCl 2 for 24, 48, or 72 hr. Data represented as means ± SEM. Statistical differences represented by different lowercase letters (a, b, c, d). SEM, standard error of mean T A B L E 1 Normalized abundance ± SEM of gene products related to glycolysis and mitophagy. Treatments include a control (cultured at 5% O 2 for 3 days), CoCl 2 treatment (100 µM CoCl 2 for 24 hr), and a hypoxic treatment (cultured at 1% O 2 for 3 days) At 120 days of gestation, the remaining pregnant surrogate farrowed naturally and delivered five piglets. Three of the five piglets were stillborn, and the surviving two piglets were healthy with no signs of abnormalities ( Figure 5). No obvious defects were detected in the stillborn piglets; however, a necropsy was not performed. Birthweights ranged from 0.800 to 1.155 kg, with an average birthweight of 0.955 kg. Weaning weights recorded at 3 weeks were 4.720 and 4.120 kg, for an average weight of 4.420 kg (Table 4). In an aerobic system, once pyruvate has been produced through glycolysis, it is subsequently converted to acetyl coenzyme A (CoA) through the mitochondrial enzyme pyruvate dehydrogenase. However, in glycolytic systems, the production of the enzyme PDK1 results in phosphorylation of pyruvate dehydrogenase which inactivates the complex and directs pyruvate away from the TCA cycle, inhibiting its oxidation. PDK1 has been demonstrated by microarray and chromatin immunoprecipitation to be a direct target of HIF1-α, and is an important player in the switch from aerobic to anaerobic metabolism through its ability to block acetyl CoA production so that pyruvate can be converted to lactate (Kim, Tchernyshyov, Semenza, & Dang, 2006).

| DISCUSSION
Since PDK1 increases availability of pyruvate in the cell, it is then able to be converted to lactate by LDHA. The conversion of pyruvate to lactate is crucial for anaerobic glycolysis. In human pancreatic cancer cells, LDHA is upregulated by hypoxia and is directly activated by HIF1-α. Induced expression of LDHA promotes the proliferation and migration of pancreatic cancer cells, and knocked down expression inhibits cell growth and migration (Cui et al., 2017). This indicates that LDHA and its effect in hypoxic conditions is crucial for cancer cell survival.
Although the majority of gene expression changes found in this study relate to the SCNT donor cells, there were also several genes upregulated in CoCl 2 treated donor cell SCNT blastocyst stage embryos ( Our findings indicate that the use of CoCl 2 as a novel treatment for SCNT donor cells induces the same glycolytic response as culture in 1% oxygen (hypoxia) for 3 days. The use of the hypoxia mimetic allows the cells to be maintained in any oxygen tension, without the need for specialized gas tanks or chambers and eliminates the need for long term culture of donor cells in hypoxic conditions to establish the same effect. The upregulation of genes that are known to be downstream targets of HIF1-α in the CoCl 2 treated and hypoxia treated donor cells, along with the lack of differential expression of non-HIF1-α targets suggests that the transcription factor may be activated through these treatments. Therefore, promoting metabolic reprogramming in donor cells through CoCl 2 treatment improves the efficiency of the SCNT process through alterations in gene expression in donor cells and resultant SCNT blastocysts, improvement in the quality and development rate of SCNT embryos, and production of healthy, cloned piglets. CoCl 2 exposure, the beforementioned conditions were applied to cells plated at equal densities, followed by aspiration of media containing CoCl 2 and replacement with fresh media. The cells were grown for 3 days subsequent to CoCl 2 removal and then trypsinized and subjected to Trypan blue exclusion to determine live and total cell number.

| Ethics statement
For SCNT, fibroblast cells were thawed 4 days before SCNT, counted by Trypan blue exclusion, plated at a density of 7.5 × 10 4 cells/T25 flask and cultured in a humidified incubator with an atmosphere of 5% CO 2 , 5% O 2 , and 90% N 2 at 37.5°C. On Day 3, 24 hr before SCNT, CoCl 2 was added at a 100 µM concentration. The control cells were left untreated. used to aspirate follicles that were 3-6 mm in size and showed Metaphase II oocytes were placed on the stage of an inverted microscope equipped with micromanipulators in drops containing manipulation medium (Lai & Prather, 2004)

| Relative quantitative PCR
Relative quantitative PCR was performed with each sample from cDNA synthesis. Message evaluated included HIF1-α targets associated with glycolysis, autophagy, and pluripotency in fibroblast cells and blastocyst stage embryos (Table 5). Samples from each biological replicate were diluted to 5 ng/µL, and quantitative PCR was run in triplicate to determine differential expression of the selected transcripts with the conditions: 95°C for 3 min, and 40 cycles of 95°C for 10 s, 55°C for 10 s, and 72°C for 30 s. A dissociation curve was generated after amplification to ensure that a single product was amplified. Abundance of each mRNA transcript was calculated relative to a housekeeping gene, β actin, and a pig genome reference sample. The comparative quantification cycle method was used to determine relative mRNA expression for each treatment.

| Surgical embryo transfer
For the embryo transfer experiment, donor cells used for SCNT were a wild-type Ossabaw cell line (RRID NSRRC:0008) that had been proven clonable (Mordhorst et al., 2019). Following SCNT, Day 6 blastocyst-stage embryos created from CoCl 2 -treated donor cells were transferred into recipient surrogates. Briefly, two gilts 4 days post-observed estrus were aseptically prepared for surgery, and the infundibulum was exposed by entry though the lower abdominal wall.
A Tomcat catheter containing 42 blastocyst-stage embryos was inserted into one ampullary-isthmic junction of each surrogate where the blastocysts were deposited. Pregnancy was determined by ultrasound on Day 25 and monitored by biweekly ultrasounds thereafter. After farrowing, birth weights, weaning weights, and phenotypes were recorded.