Nuclear transfer stem cells hold considerable promise in the field of regenerative medicine and cell-based drug discovery. In this study, a total of 29 oocytes were obtained from three young (20–24 years old) reproductive egg donors who had been successful in previous cycles. These oocytes, deemed by intended parents to be in excess of their reproductive needs, were donated for research without financial compensation by both the egg donor and intended parents after receiving informed consent. All intended parents successfully achieved ongoing pregnancies with the oocytes retained for reproductive purposes. Mature oocytes, obtained within 2 hours following transvaginal aspiration, were enucleated using one of two methods, extrusion or aspiration, after 45 minutes of incubation in cytochalasin B. Rates of oocyte lysis or degeneration did not differ between the two methods. Somatic cell nuclear transfer (SCNT) embryos were constructed using two established adult male fibroblast lines of normal karyotype. High rates of pronuclear formation (66%), early cleavage (47%), and blastocyst (23%) development were observed following incubation in standard in vitro fertilization culture media. One cloned blastocyst was confirmed by DNA and mitochondrial DNA fingerprinting analyses, and DNA fingerprinting of two other cloned blastocysts indicated that they were also generated by SCNT. Blastocysts were also obtained from a limited number of parthenogenetically activated oocytes. This study demonstrates, for the first time, that SCNT can produce human blastocyst-stage embryos using nuclei obtained from differentiated adult cells and provides new information on methods that may be needed for a higher level of efficiency for human nuclear transfer.
Disclosure of potential conflicts of interest is found at the end of this article.
The consistent and efficient generation of both heterologous and autologous stem cells using somatic cell nuclear transfer (SCNT) has the potential to revolutionize our understanding and development of treatments for degenerative diseases [1, , , , , , , , –10]. The recent voluntary withdrawal, amid allegations of scientific impropriety, of publications that claimed successful human nuclear transfer  suggests that a reassessment of the critical pathways to performing successful SCNT with human oocytes is required. Somatic cell nuclear transfer in mammals was first performed in the early 1980s, and numerous approaches have been used to achieve the four major steps involved in the nuclear transfer procedure (reviewed in [12, –14]). These four steps are preparation of recipient cytoplasts (enucleated oocytes), preparation of donor DNA, transfer of the donor DNA into the cytoplast, and resumption of embryonic development by parthenogenetic activation, all of which must be completed without damaging the donor nucleus or the host oocyte/cellular integrity.
Derivation of cloned human embryos from SCNT techniques remains in its infancy, with the only reliable publication describing the generation of a single cloned blastocyst, which did not produce a human embryonic stem cell (hESC) line . Importantly, this involved an undifferentiated hESC donor cell, rather than an adult cell. Lu et al. also reported having cloned human embryos using fetal fibroblasts and granulosa cells, although the reliability of the finding is uncertain since no DNA fingerprinting of the cloned blastocysts was provided . Other preliminary reports have generated only early cleavage stage nuclear transfer embryos [17, 18] using cytoplasts obtained from both in vitro and in vivo maturation and aged human in vitro fertilization (IVF) oocytes that had failed-to-fertilize [18, 19]. Clearly, a number of significant hurdles need to be addressed if SCNT is to lead to the successful application of cell-based therapies using autologous hESC lines. The scarcity of donated mature human metaphase II oocytes available for research is a significant impediment. Using an alternative source of failed-to-fertilize human oocytes obtained at 16–18 hours postinsemination  or 48 hours after retrieval  resulted in cleavage abnormalities and early-embryonic arrest, primarily as a consequence of aneuploidy and spindle defects. These types of oocytes also appear to be more difficult to enucleate and fuse compared with freshly recovered human oocytes . The use of aged oocytes has been reported to support embryonic nuclear remodeling and reprogramming to various degrees in the rabbit , mouse , and bovine . However, failed-to-fertilize human oocytes may show irreversible cytoplasmic aging effects that do not adequately support the effective nuclear remodeling and reprogramming of somatic cells [19, 24]. Similar failed-to-fertilize oocytes also cleave poorly after nuclear transfer (NT) in nonhuman primates .
The aim of the present study was to examine the development of human cloned embryos derived from adult fibroblasts. Restoration of developmental competence to the donor nuclei was determined by evidence of remodeling (pronuclear formation) and reprogramming embryo development (blastocyst formation).
All experiments were approved by an independent review committee (Independent Review Consulting, Inc., San Anselmo, CA, http://www.irb-irc.com) in accordance with the U.S. Department of Health and Human Services Policy for Protection of Human Research Subjects (45CFR46) and the National Academy of Sciences Guidelines for Human Embryonic Stem Cell Research  and appear also to comply with the recently released guidelines from the International Society for Stem Cell Research . All donated oocytes were obtained from three egg donation cycles performed at the Reproductive Sciences Center (RSC), a fertility center located in La Jolla, California, that is fully compliant with Society for Assisted Reproductive Technologies and American Society Reproductive Medicine guidelines. Its laboratory is licensed by the State of California and American Association of Tissue Banks, accredited by the College of American Pathologists, and registered with the Food and Drug Administration. The clinical procedures and operative facility have been inspected and accredited by the American Association of Ambulatory Health Care. All facilities are subject to inspection by the various agencies.
Following operative and clinical procedures, excess-to-reproductive-needs oocytes were donated without direct linkage information, so that the SCNT researcher was unaware of the identity of any oocyte donor. However, the SCNT researcher maintains a record of every gamete, somatic cell, embryo donation, or product of SCNT that has been donated, created, or used. This record is sufficient to allow, if necessary, the provenance and disposition of such materials to be determined.
All chemicals used for the production of SCNT and parthenogenetically activated (PA) embryos were purchased from Sigma-Aldrich (St. Louis, http://www.sigmaaldrich.com) unless otherwise stated. All embryo manipulations were performed at 37°C except for the process of electrical fusion, which was performed at room temperature. All cultureware underwent standard quality control testing prior to use, and culture media were preincubated to 37°C and, when necessary, pre-equilibrated in a humidified atmosphere of 6% CO2, 5% O2, 89% N2 in a dual gas incubator (Innova CO-170; New Brunswick Scientific, Edison, NJ, http://www.nbsc.com).
Preparation of Karyoplasts
Primary adult fibroblast cell lines were isolated from a 2–3-mm skin biopsy sample taken from two healthy, adult male donors (AF1 and AF2) using standard procedures approved in the SCNT research ethics application (described in Human Ethics). Cell lines were established and maintained in α-minimum essential medium (α-MEM) containing GlutaMAX, ribonucleosides, and deoxyribonucleosides supplemented with 100 U/ml penicillin, 100 μg/ml streptomycin, 55 μM 2-mercaptoethanol (Gibco, Carlsbad, CA, http://www.invitrogen.com) and 20% (vol/vol) human serum (pooled from patients undergoing fertility treatment at RSC). All sera used tested negative for the following diseases: Human Immunodeficiency Virus (1 and 2), Human T-lymphotropic virus (Virus type 1 and 2), Syphilis (rapid plasma reagin), Hepatitis C (Hepatitis C antibody and anti-Hepatitis C Core Immunoglobulin G (IgG)), Hepatitis B (Hepatitis B surface antigen and anti-Hepatitis B Core Immunoglobulin M (IgM)), Cytomegalovirus IgG and IgM, Chlamydia and Gonnorhea.
For cytogenetic analysis, donor cell lines were incubated in α-MEM with 0.1 mg/ml of Colcemid (KaryoMAX; Gibco) for 3–4 hours, trypsinized, resuspended in 0.075 M KCl, incubated for 20 minutes at room temperature, and then fixed in 3:1 methanol acetic acid. Chromosome number and size were determined using Giemsa-stained metaphase spreads.
For cell cycle analysis, actively dividing fibroblasts AF1 and AF2 were harvested (TrypLE; Gibco) to a single-cell suspension, washed, and resuspended in Dulbecco's phosphate buffered saline. A total of 1 × 106 cells were resuspended in 0.25 ml of ice-cold GM buffer (6 mM glucose, 0.14 M NaCl, 5.4 mM KCl, 1.1 mM Na2HPO4-12H2O, 1.1 mM KH2PO4, and 0.5 mM EDTA) before adding 0.75 ml of chilled 95% ethanol. Samples were sent on ice to Phoenix Flow Systems, Inc. (San Diego, http://www.phnxflow.com), where cells were stained with 50 μl/ml propidium iodide/RNase A solution, acquired on a Becton Dickinson LSR Flow Cytometer (Becton, Dickinson and Company, Franklin Lakes, NJ, http://www.bd.com), and analyzed with MultiCycle AV DNA analysis software. A control sample was included in the analysis.
Fluorescence In Situ Hybridization Analyses
Fluorescence in situ hybridization (FISH) was performed with probes for chromosomes 13, 18, 21, X, and Y in accordance with the manufacturer's codenaturation and posthybridization instructions accompanying the hybridization kit (Multi-Vysion PGT Fluorescent Probe Kit; Abbott Molecular Inc., Des Plaines, IL, http://www.abbottmolecular.com). Fibroblast cells were analyzed with a fluorescence microscope (BX61TRF; Olympus America Inc., Melville, NY, http://www.olympusamerica.com) equipped with single-band-pass filters for red (chromosome 13), aqua (chromosome 18), green (chromosome 21), blue (X chromosome), and gold (Y chromosome). The images were analyzed with Cytovision software (version 3.6; Genetix Limited, Queensway, New Milton, Hampshire, U.K., http://www.genetix.com).
Actively dividing fibroblasts (passages 3–5; 60%–80% confluent) without serum starvation were used as donor karyoplasts. Single-cell suspensions of the donor cell lines (AF1 and AF2) were prepared with recombinant trypsin (TrypLE Express; Gibco) on the day of SCNT.
Prior to any procedures, oocyte donors and oocyte recipients (intended parents) were independently counseled, and informed consent was obtained for the anonymous donation of those oocytes considered by the intended parents to be in excess of their reproductive needs. Typically, intended parents elected to donate those oocytes in excess of 10–12 per transvaginal aspiration [28, 29].
Oocyte donors, all of whom had had at least one previous successful donation cycle, underwent routine controlled ovarian hyperstimulation involving a combination of follicle-stimulating hormone (FSH) and luteinizing hormone containing gonadotropins using either a standard gonadotropin-releasing hormone (GnRH) agonist long protocol (leuprolide acetate [Lupron; TAP Pharmaceutical Products Inc., Lake Forest, IL, http://www.tap.com]) or a GnRH antagonist protocol (cetrorelix [Cetrotide; EMD Serono, Inc., Rockland, MA, http://www.emdserono.com]) following suppression with an oral contraceptive .
Beginning on day 3 of their treatment cycle, donors received 150–225 IU/day recombinant FSH (Gonal-f; EMD Serono) and 75–150 IU/day hMG (Menopur; Ferring Pharmaceuticals, Inc., Suffern, NY, http://www.ferring.com) for 9–10 days. Dosages were adjusted based on ovarian response as assessed by serum estradiol concentrations and ultrasound imaging of the ovaries. When a GnRH antagonist was used, a flexible start protocol was used, with the antagonist initiated when the largest ovarian follicle present reached a mean diameter of at least 12 mm. When at least two follicles had a mean diameter of 18 mm, ovulation was induced with 5,000 or 10,000 IU of human chorionic gonadotropin (APP Pharmaceuticals, East Schaumburg, IL, http://www.appdrugs.com). Cumulus-oocyte complexes were retrieved by ultrasound-guided transvaginal aspiration 34–36 hours later (day 0) from follicles ranging in size from approximately 14 to 20 mm.
Preparation of Cytoplasts
The cumulus matrix was removed from oocytes (1–2 hours postaspiration [hpa]) by repeated gentle pipetting in 80 U/ml recombinant human PH20 hyaluronidase enzyme per the manufacturer's instructions (Cumulase; Halozyme Therapeutics, Inc., San Diego, http://www.halozyme.com) in human tubal fluid (HTF)-HEPES with EDTA and glutamine medium (InVitroCare, Inc., Frederick, MD, http://www.invitrocare.com) supplemented with 5% (vol/vol) human serum albumin (HSA) (InVitroCare).
Cumulus-free oocytes were then incubated in 5 μg/ml cytochalasin B (CYTB) in HTF-HEPES for 45 minutes, with Hoechst 33342 (6 μg/ml) added to the medium for the last 15 minutes of incubation to enable visualization of nuclear material under ultraviolet (UV) light. Conditions for the enucleation of oocytes using the aspiration method were based on a preliminary study using eight failed-to-fertilize or metaphase I (MI) oocytes, which showed that increasing the incubation time in CYTB decreased the level of oocyte lysis.
At 3–4 hpa, oocytes were placed in HTF-HEPES under paraffin oil (Ovoil; Vitrolife, Inc., Englewood, CO, http://www.vitrolife.com) and enucleated using one of two methods: extrusion or aspiration. Using the first method, the zona pellucida was pierced with a sharp beveled glass pipette to make a slit, and the adjacent cytoplasm containing the metaphase II (MII) chromosomes, and sometimes the first polar body, were extruded through the slit in the zona pellucida by applying pressure to the oocyte with the needle . With the alternative method, the MII chromosomes were aspirated with a small volume of cytoplasm into a sharp beveled glass pipette (optical density, 20–25 μm). The removal of DNA was confirmed by DNA staining and brief visualization under UV light (supplemental online Fig. B).
Somatic Cell Nuclear Transfer
A single fibroblast donor cell 10–15 μm in diameter was inserted under the zona pellucida so that it remained in contact with the cytoplast. This size range selected was based on the work of Boquest et al., who showed that the highest proportion (up to 95%) of cells of this diameter represented the G1 and G0 stages of the cell cycle .
When the first polar body was not removed due to location of the metaphase II plate, the donor fibroblast was inserted in a position that provided at least 90° of separation from the polar body. This allowed the correct parallel alignment of donor cell/cytoplast with the fusion electrodes and avoided interference of the polar body with the fusion process.
NT cytoplast-fibroblast pairs were equilibrated (5 minutes) in fusion medium consisting of 0.3 mol/l sorbitol, 0.1 mmol/l MgSO4, 0.5 μmol/l CaCl2, and 3% (wt/vol) HSA. Fusion was induced in a cell fusion chamber (0.5-mm gap) with two electrical pulses (1.7 kV cm−1 for 15 microseconds for two pulses administered 0.1 second apart) generated from a CF-150B impulse generator (Biological Laboratory Equipment, Maintenance and Service Ltd., Budapest, Hungary).
All embryos were examined morphologically after 20 minutes, and those embryos not fused with the donor cell underwent a second round of electrical fusion. Embryos where the first polar body had not been removed at the time of enucleation were examined carefully to ensure that only fusion of the donor cell had occurred (i.e., the first polar body was still present after fusion).
Reconstructed SCNT and metaphase II (PA) embryos were parthenogenetically activated 2–3 hours after electrical fusion (5 hpa) with calcium ionophore (CI) A-23187 (10 μM, 4 minutes) in HTF-HEPES and washed three times in fresh HTF-HEPES before being transferred either to the low-glucose, phosphate-free formulation of IVC-TWO (InVitroCare) supplemented with 8% (vol/vol) HSA and 2 mM 6-dimethylaminopurine (6-DMAP) or to 10 μg/ml cycloheximide (CHX) plus 2.5 μg/ml cytochalasin D (CYTD) for an additional 3–4 hours of incubation under paraffin oil in a high-humidity atmosphere of 6% CO2, 5% O2, 89% N2 for 4 hours.
Following a 3–4-hour incubation (8–9 hpa) in 6-DMAP or CHX/CYTD, SCNT and PA embryos were washed thoroughly with IVC-TWO supplemented with 8% (vol/vol) HSA and transferred in group culture conditions into one well per treatment of a 4-well plate for In Vitro Fertilization (BD Falcon, Becton Dickinson) containing IVC-TWO with 8% HSA under paraffin oil and maintained at 37°C in a humidified atmosphere of 5% CO2, 6% O2, 89% N2. On day 2, embryos were transferred to IVC-TWO with 9% HAS, and then on late day 3/early day 4 to the phosphate-free, elevated-glucose formulation of IVC-THREE (InVitroCare) with 12% HSA. The rates of pronuclear formation, early cleavage, and blastocyst formation were evaluated stereomicroscopically and photographed throughout in vitro culture.
Total Blastocyst Cell Count
Prior to embryo lysis for microsatellite analysis, SCNT and parthenogenetic embryos were stained with Hoechst 33342 (6 μg/ml) and photographed under UV light to determine total cell number.
Groups of 5–10 cumulus and fibroblast cells from the oocyte donor and male skin donor, respectively, and SCNT and PA blastocysts were sent on ice in 5 μl of lysis buffer to Genesis Genetics Institute (Detroit, MI, http://www.genesisgenetics.org/) in coded vials for independent microsatellite and mitochondrial DNA (mtDNA) analyses. Single-stranded human genomic DNA was extracted using specifically designed and tested single-cell techniques , and 15 autonomous DNA markers and the sex chromosomes were amplified using the PowerPlex 16 System kit (Promega, Madison, WI, http://www.promega.com). Genescan analysis on the amplified fragments was performed for the following microsatellite markers: Penta E (15q), D18S51 (18q21.3), D21S11 (21q11–21q21), TH01 (11p15.5), D3S1358 (3p), FGA (4q28), TPOX (2p23–2pter), D8S1179 (8q), vWA (12p12-pter), Amelogenin (Xp22.1–22.3 and Y), Penta D (21q), CSF1PO (5q33.3–34), D16S539 (16q24-qter), D7S820 (7q11.21–22), D13S317 (13q22-q31), and D5S818 (5q23.3–32). These were run on an ABI 3130 Genetic Analyzer, and allele calls determined by GeneMapper version 4.0 using bin sets optimized for that instrument (Applied Biosystems, Foster City, CA, http://www.appliedbiosystems.com).
Following microsatellite analyses, DNA from the samples submitted in coded vials was used to amplify mtDNA hypervariable region II (HVII). The polymerase chain reaction (PCR) forward primer spans bases 15–34, and the reverse primer spans bases 431–412 for Homo sapiens mitochondria; complete genome (GenBank accession no. NC_001807.4) and reaction conditions were essentially as previously described [33, 34], except that −21M13 and 28M13Rev tails were added to the forward and reverse PCR primers, respectively, for use in subsequent sequencing reactions. Separate sequencing reactions were performed on both strands of each PCR product by priming off the appropriate M13 tail, using the ABI Prism BigDye Terminator version 3.1 cycle sequencing kit (Applied Biosystems). Unincorporated dye-labeled products were removed by spin column purification, using Sephadex G-50 Superfine (GE Healthcare, Uppsala, Sweden, http://www.gehealthcare.com). The eluate was run on an ABI 3130 Genetic Analyzer, and base calls were determined by Sequence Analysis 5.1 software (Applied Biosystems).
Preparation of Donor Karyoplasts
Cytogenetic analysis of established fibroblast cell lines (Fig. 1A, 1B) showed that 95% of AF1 (n = 20) and 100% of AF2 (n = 20) cells examined maintained a diploid 22 chromosomes + XY karyotype. (Fig. 1C [AF1], 1D [AF2]). This was confirmed with FISH analysis of both adult male fibroblast cell lines, which demonstrated a diploid pattern for chromosomes 13, 18, and 21 and the presence of X and Y chromosomes (Fig. 1E [AF1], 1F [AF2]).
Cell Cycle Analyses
The distribution of cycling AF1 and AF2 adult somatic cells existing in the various phases of the cell cycle as determined by flow cytometry is shown in supplemental online Figure A. The prominent DNA peaks (in blue) show that the majority of adult somatic cells at a given point in time were in G1 (+G0), where values of 63.4% of AF1 (supplemental online Fig. Aa) and 53.6% of AF2 (supplemental online Fig. Ab) were determined, compared with values for other phases of the cell cycle, S-phase (light green) and G2+M (green), where 18.7% of AF1 and 30.4% of AF2 were in S-phase and 17.9% of AF1 and 16% of AF2 were in G2+M phase. The presence of cellular debris was determined at 2.7% and 2.2% for AF1 and AF2, respectively. Supplemental online Figure Ac shows control DNA.
From March to September 2006, three oocyte donors (20–24 years old) and their intended parents agreed to participate in the SCNT research project. For this part of the SCNT project, 29 excess-to-reproductive-needs oocytes were anonymously donated within 1–2 hours of transvaginal aspiration.
Preparation of Cytoplasts
The first polar body (PB) was observed in 90% (26 of 29) of oocytes, indicating that most had matured to the MII stage of development (Table 1). Two methods of enucleation were used following incubation in CYTB, extrusion, and aspiration (Fig. 2A2). Additional images of the aspiration process can be seen in the supplemental online data. Supplemental online Figure Ba1–Ba3 shows metaphase II enucleation with the polar body, and supplemental online Figure Bb1, Bb2 shows metaphase II enucleation without removal of the polar body. The PB was removed (12 of 21; 57% of enucleation attempts) only if in close proximity to the metaphase plate. The aspiration method removed the least amount of associated cytoplasm. Enucleation was confirmed with UV fluorescence following incubation of the oocyte in Hoechst 33342 (Fig. 2A; supplemental online Fig. Ba1–Ba3, Bb1, Bb2).
Table Table 1.. SCNT and parthenogenetic activation with freshly donated human oocytes
Of 23 SCNT embryos enucleated, 7 (33%) either lysed after enucleation or following electrical fusion or showed no evidence of pronuclear formation and degenerated after 24 hours of culture (Fig. 2C). There was no difference between the methods of enucleation in terms of rates of degeneration or lysis.
Somatic Cell Nuclear Transfer
Following SCNT, 14 (66%, or 5 of 8 [63%] AF1 and 9 of 13 [69%] AF2) reconstructed embryos showed evidence of DNA remodeling (pronuclear formation) at 6–7 hours after chemical activation (Table 1; Fig. 2C, 2C2, 2C4). On day 3, 10 embryos (47%, or 4 of 8 [50%] AF1 and 6 of 13 [46%] AF2) with five or more cells were observed (Fig. 2C, 2C4), and 5 (50% of multicellular embryos) of those (2 of 8 [25%] AF1 and 3 of 13 [23%] AF2) formed blastocysts of days 5-6 (Fig. 2C, 2C6; supplemental online Fig. Ca–Cc). Blastocysts were also observed in the small number of metaphase II blastocysts (Bl) that underwent parthenogenetic activation (n = 5) from CI/dimethylaminopurine (6-DMAP) (two Bl; n = 4) and CI/CHX/CYTD (one Bl; n = 1) activation groups.
Total Blastocyst Cell Count
Total cell counts were performed on three parthenogenetic and five SCNT blastocysts obtained following nuclear transfer with both AF1 and AF2 donor cells as determined by DNA staining with Hoechst 33342 prior to preparation for microsatellite analysis. Cell numbers for parthenogenetic embryos ranged from 35 to 54 and from 41 to 72 for SCNT blastocysts (supplemental online Fig. Da [parthenogenetic blastocyst; coded reference K1], Db [SCNT blastocyst; coded reference K8]) following nuclear transfer with AF2 (coded references C5 and C6) donor cell to oocyte derived from egg donor 3 (coded references F1 and F2).
A total of 18 coded samples were sent for independent microsatellite analysis and included duplicate samples of cumulus cells from the two egg donors (egg donors 1 [coded references A1 and A2] and 3 [coded references F1 and F2]), fibroblast cells from the two somatic cell donors (AF1 [coded references B3 and B4] and AF2 [coded references C5 and C6]), five SCNT blastocysts, four parthenogenetic embryos, and one blank sample (no DNA) to check for contamination due to handling. Two of the SCNT blastocysts (9F and K7) and two parthenogenetic samples failed to amplify and gave no reliable data, and the blank was negative (data not shown). The microsatellite analyses of the three remaining cloned blastocyst, coded reference 8E and coded references 10G and K8 were consistently matched to the DNA of AF1 (coded references B3 and B4) and AF2 (coded references C5 and C6) adult skin fibroblast donors, respectively. Assuming allele dropout (ADO) of the TPOX  allele, blastocyst 8E from egg donor 1 (coded references A1 and A2) can be explained using alleles from adult fibroblast donor AF1 (coded references B3 and B4) alone; citing ADO of D13S317  allele, blastocyst 10G from egg donor 1 (coded references A1 and A2) can be explained using alleles from fibroblast donor AF2 (coded references C5 and C6) alone; and blastocyst 8K from egg donor 3 (coded references F1 and F2) can be explained using alleles from AF2 (coded references C5 and C6) alone, assuming that alleles THO1 (9.3), D13S317 , and TPOX  peaks are due to unvisualized alleles in the AF2 (coded references C5 and C6) donor cell samples (Fig. 3A, 3B). The original fingerprinting sequence data and allele calls for each of the cloned blastocysts are provided in the supplemental data (supplemental online Figs. F–L). Microsatellite analysis of parthenogenetic blastocyst (coded reference K1) and morula (coded reference K6) can be explained using alleles from egg donor 3 (coded references F1 and F2) alone (supplemental online Fig. E, M, N).
Mitochondrial DNA Analyses
Following successful microsatellite analyses, DNA from the samples submitted in coded vials was used to amplify the mtDNA HVII region. Separate sequencing reactions from both strands of each PCR product are shown in the supplemental data (supplemental online Table A; electropherogram data of mtDNA hypervariable region II; supplemental online Figures Q–S). Reliable data were not obtained from all SCNT blastocysts due to the amount of material available and variations in the amount of amplified DNA. However, SCNT blastocysts (coded reference K8) showed the variations T at 33 base pairs (bp), G at 34 bp, and 8C at 270 bp in forward sequencing of HVII and can be explained by the mitochondrial DNA sequence of egg donor 3 (coded references F1 and F2); their sequence also matched that of the parthenogenetic embryos (coded references K1 and K6) from the same egg donor.
This study demonstrates, for the first time, that SCNT can be used to generate cloned human blastocysts using differentiated adult donor nuclei remodeled and reprogrammed by human oocytes. Evidence of successful SCNT was shown with DNA fingerprinting analyses of three SCNT cloned blastocysts, where embryo genomic DNA was that of the donor fibroblast cell line and not of parthenogenetic origin or as a result of oocyte fragmentation [35, 36]. Examination of mtDNA sequence from the HVII region showed that one SCNT blastocyst (coded reference K8) was matched to egg donor 3 (coded references F1 and F2) mtDNA and was identical to the parthenogenetic embryos (coded references K1 and K6) obtained in the same experimental series.
The study describes a method allowing the derivation of human NT blastocysts using young oocytes and established adult male fibroblast lines of normal karyotype. The rate of SCNT blastocyst formation is similar to the development rates seen in standard IVF cycles (fertilization rate of 70%–80% of oocytes collected, and 40%–60% of fertilized oocytes develop into blastocysts) [37, –39] and to a previous nuclear transfer study with undifferentiated hESC donor cells . Total cell counts in SCNT blastocysts improved with each subsequent experimental series and appear comparable to reports for various grades of human IVF embryos . Blastocysts were also derived from a limited number of parthenogenetically activated oocytes using both of the activation methods used. DNA fingerprint and mtDNA, of the HVII region, analyses of the parthenogenetic embryos (coded references K1 and K6) confirmed that they originated solely from egg donor 3 (coded references F1 and F2). The frequency of blastocyst generation is similar to that reported in the recent work of Revazova et al., where a total of 44 donated oocytes resulted in 23 blastocysts after parthenogenetic activation (52%), of which 11 (25%) had a visible inner cell mass and generated six human parthenogenetic embryonic stem cell lines .
DNA fingerprinting analysis of the adult donor fibroblasts, egg donor cumulus cells, and all blastocysts observed following nuclear transfer was conducted to determine whether the embryonic pathways in the donor nucleus had been reinitiated. Of the five putative SCNT blastocysts generated, successful DNA fingerprints from three SCNT blastocysts were consistent with those of the somatic cell donor used, with no evidence of contamination from the egg donors, indicating that embryonic development was being controlled by the donor cell genome. The DNA fingerprint comparisons were not identical, which was not unexpected given the reported incidence of ADO and unvisualized alleles in single-cell DNA fingerprinting systems [42, 43].
Further amplification of the samples for forward and reverse HVII region mtDNA sequencing showed three alterations in sequencing in one cloned blastocyst, T or C at 33 bp, G or A at 34 bp, and 8C or 9C at 270–271 bp, and indicates that mtDNA in the cloned blastocyst K8 was from egg donor 3 (coded references F1 and F2) and not cell donor AF2 (coded references C5 and C6). It should be noted that the stutter observed after the poly(C) stretch is almost certainly due exclusively to the repetitive nature of the sequence; there are no heterozygotes in the K cell series at the T or C and the G or A loci, with only the T allele and the G allele from egg donor 3 (coded references F1 and F2) clearly visualized. This would indicate that if any mitochondria from fibroblast donor AF2 (coded references C5 and C6) were transferred along with the nuclear material, they were vastly outnumbered by those of egg donor 3 (coded references F1 and F2).
Sequence data in both directions are appropriate until the string of C homopolymer is reached in the forward direction or G homopolymer in the reverse direction. The presence of repetitive sequence, especially in the case of homopolymer and dinucleotide repeats (less so for higher order repeats), causes slippage by DNA polymerase and causes stutter in sequence downstream of that sequence. The presence of poly(C) and poly(G) is especially problematic, and significant sequence degradation can be seen even with fewer than 10 repeats (much more so than poly(A) or poly(T), which can tolerate ∼13–15 repeats). In the case of the sequences seen here, their overall quality gives the expected result, with significant degradation of sequence quality once the critical repeat number is exceeded for the repetitive sequence in question.
Several modifications to the nuclear transfer process were investigated. Using a different method of enucleation (either extrusion or aspiration of the metaphase plate) resulted in no discernible differences in terms of embryo lysis or degeneration or in the developmental competence of SCNT embryos. The longer incubation of oocytes in a cell-permeable mycotoxin, cytochalasin B (45 minutes compared with 15–20 minutes), prevents the formation of contractile microfilaments and disruption of actin filaments and polymerization. Incubation in cytochalasin B using the time intervals reported in previous studies using failed-to-fertilize and MI oocytes resulted in increased oocyte lysis. Increasing the exposure of oocytes to cytochalasin B (45 minutes) decreased oocyte lysis and supported the findings of a previous report . It is not known whether the use of recombinant human hyaluronidase, instead of the commonly used bovine source, to remove the cumulus matrix also contributed to the efficiency of this procedure, but additional studies with this new product are certainly warranted.
Eleven successful enucleation attempts did not remove the first polar body because of untoward location that would not have allowed us to achieve our goal of minimizing the amount of cytoplasm removed during the process. Increasing the amount of cytoplasm removed during the enucleation process has an impact on the development potential of nuclear transfer embryos [44, 45]. Anecdotally, removal of the first polar body with either technique appeared to increase the amount of cytoplasm removed.
Although mice have been generated from the first polar body following NT, that method of transfer required intracytoplasmic injection of the polar body, primarily as a result of apoptotic events associated with the breakdown and degeneration of the first polar body in the mouse . The presence of the first polar body is of little concern because, in practice, it is difficult to initiate successful electrical fusion of the polar body with the enucleated oocyte and because the donor cell is easily distinguished from the polar body during the microscopic examinations required as part of the electrical fusion. The presence of two pronuclei during the assessment of remodeling, which would have indicated fusion of the first polar body, was not detected. In addition, DNA fingerprinting analysis did not confirm the presence of maternal DNA in the SCNT blastocysts.
In this study, parthenogenetic activation using either CI in combination with protein synthesis (6-DMAP) or protein kinase inhibitors (CHX/CYTD) artificially activated fresh oocytes that subsequently underwent pronuclear formation and cleavage division in a manner consistent with previous reports [24, 47, –49]. Prevention of second polar body extrusion by the inhibition of either protein synthesis pathways (6-DMAP) or protein kinase pathways (CHX/CYTD) can alter the developmental competence of SCNT embryos . However, in our assessment, there appeared to be no difference between the two activation treatments in terms of pronuclei and blastocyst formation, although the limited number of activated human oocytes used makes it difficult to draw conclusions. Of note, parthenogenetically activated blastocysts observed were of lower embryonic grades due to a slower development rate and lower total cell number.
Additional modifications to the activation method, as suggested by a number of recent studies, may improve efficacy. For example, cytosolic extracts from sperm have been shown to cause Ca2+ oscillations in a range of different mammalian oocytes, including humans . In a further refinement, Saunders et al.  isolated the specific isoform of phospholipase C (PLC)  thought to be responsible for generating Ca2+ release and inducing inositol trisphosphate production. Microinjection of cRNA encoding human PLCζ induced a response in aged human oocytes that closely mimicked the repetitive calcium Ca2+ oscillations stimulus provided by the sperm during human fertilization and induced parthenogenesis and development to the blastocyst stage .
The protocols produced high rates of pronuclear formation (66%), early cleavage (47%), and blastocyst development (23%) following incubation in standard IVF culture media. As would be expected, a range of blastocyst grades were observed following both parthenogenetic activation and SCNT; however, all showed the presence of a defined inner cell mass. The observation of morphologically normal SCNT blastocysts is the prelude to obtaining embryonic stem cell lines, which will now be the focus of this research. This study examined the reinitiation of the embryonic genome in cloned blastocysts using the methods of DNA fingerprinting and mtDNA sequencing. However, future studies to support the total cell number findings will examine the quality of the SCNT blastocyst and examine chromosome number, ICM/trophectoderm ratio, and expression of critical embryonic genes at the time of blastocyst formation (Oct4, Cdx2) [55, 56]. The process of preparing the samples for DNA fingerprinting, however, precluded the application of these methods.
Other factors that may have contributed to the success of SCNT procedures include the quality of the oocytes obtained from young egg donors. The best donor candidates are likely to be younger women, without a history of infertility, who respond well to ovarian stimulation. A major impediment to conducting studies of this type is the difficulty of obtaining suitable donor oocytes, something that is influenced by both ethical and legal considerations. Models for donation include a range of potential options, including altruistic oocyte donation, monetary inducement, or compensation in the form of reduced treatment fees. After obtaining the appropriate informed consent from the egg donor and intended parents, this study involved using oocytes that were excess to reproductive needs and that were donated without compensation. All three intended parents were successful in achieving ongoing pregnancies with the oocytes they retained for reproductive purposes. The average in vitro development rates of assisted reproductive technologies oocytes following either in vitro fertilization and/or intracytoplasmic sperm injection was 76% for normal fertilization (two pronuclei) and 92% for cleavage to good-quality embryos (grades I and II) determined on days 2–4 (data not shown). Since a significant percentage of couples undergoing fertility treatments appear willing to participate in this type of research [28, 29], we believe that the method described to obtain donated oocytes is a viable and ethically acceptable strategy that can enable optimization of the nuclear transfer stem cell method to produce embryonic stem cells . Although it allows the acquisition of only limited numbers of oocytes and thus small groups of embryos from an individual egg donor, it clearly is a more efficacious process than those in previous studies, which obtained oocytes predominantly from patients of increasing maternal ages undergoing a variety of IVF treatments [15, 24].
We are currently pursuing the generation of embryonic stem cell lines from SCNT embryos generated using these protocols. The recent findings with mouse  and human  embryos showing the generation of ESC lines from individual blastomeres in human from arrested embryos  and primate embryonic stem cells following somatic cell nuclear transfer  may hasten the attainment of this goal.
In conclusion, it has been demonstrated that heterologous SCNT blastocysts can be successfully and reliably generated from two male adult fibroblast lines using oocytes from young oocyte donors obtained shortly after an oocyte retrieval that followed a standard oocyte donation cycle. Following SCNT, the somatic cell nuclei underwent remodeling, as evidenced by pronuclear formation. Of the five putative cloned blastocysts produced, only one cloned blastocyst was confirmed by DNA and mtDNA fingerprinting analyses; however, DNA fingerprinting of two other cloned blastocysts indicated that they were generated by SCNT, although we were unable to conclusively support these finding with mitochondrial DNA analyses.
Disclosure of Potential Conflicts of Interest
The authors indicate no potential conflicts of interest.
We thank the clinical and laboratory staff of the Reproductive Sciences Center for valuable assistance in the treatment and collection of anonymously donated oocytes from egg donors. Our gratitude also goes to Kevin Becker at Phoenix Flow Systems, Inc., for the cell cycle analyses; W. Umansky, M.D., for performing the skin biopsies required to generate the donor fibroblast line; and A.O. Trounson for valuable assistance in critical review and preparation of the manuscript.