|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Reproductive Technology |
Biotechnology and Germplasm Laboratory,2 Agricultural Research Service, USDA, Beltsville, Maryland 20705
Revivicor Inc.,3 Blacksburg, Virgina 24060
Department of Animal Sciences,4 University of Vermont, Burlington, Vermont 05405
| ABSTRACT |
|---|
|
|
|---|
early development, embryo
| INTRODUCTION |
|---|
|
|
|---|
Apparently, the key to successful somatic cell nuclear transfer is proper coordination of ooplasm and donor cell cycle [6]. Though there has been some dispute in the literature regarding which donor cell's cell cycle stages are optimum [2], there has been little disagreement that various combinations of donor cell stage and oocyte age result in live healthy offspring [5, 7, 8].
The vast majority of bovine somatic cell nuclear transfer studies have been initiated with oocytes derived from ovaries collected at slaughter, though in vivo-derived oocytes have also served as starting material [9]. In vitro maturation of oocytes, well established before the advent of somatic cell nuclear transfer, has been adopted with little change [1013]. Most attempts to improve the cloning process in cattle have focused on methods of reconstitution [10], donor cell types [10, 1418], activation of couplets [19, 20], and, to a lesser extent, embryo culture [2123].
Recently, several strategies to increase speed of embryo manipulation and to decrease required skill level have been reported [2426]. All three of these approaches reduce specialized equipment needs, in one case, dramatically [26].
The potential of various cell types to serve as nuclear donors, such as cumulus [27], embryonic stem (ES) cells [28], fetal and adult fibroblasts [14], granulosa [29], myoblast [30], neurons [31], and Sertoli cells [32] has been evaluated. To date, no particular cell type has an overwhelming advantage over another. When fetal and adult fibroblasts have been compared, fetal fibroblasts make superior donor cells. Lower passage cells have an advantage over higher passages cells [15, 3335]. Thus, the number of divisions after the initial mitosis appears to be an important factor in determining a cell's usefulness as a nuclear donor [36].
Beyond evaluating cell types and manipulating serum concentrations in an effort to control the stage of the cell cycle, little attention has been focused on optimizing culture conditions of donor cells. In the study reported here, fibroblasts from six fetuses served as nuclear donor cells. Their clonal growth under G418 selection and ability to generate blastocysts and produce calves was assessed. Fetal fibroblasts from lung and muscle tissue were compared. In addition, during clonal growth, the influences of culture media and serum and oxygen concentrations were evaluated. Of the parameters studied, the fetus from which the fibroblasts were harvested had the greatest influence on efficiency of producing calves.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Jersey females were inseminated with semen from one of three bulls to produce six fetuses (fetuses 7, 8, and 9 from bull A, fetus 10 from bull B, and fetuses 11 and 12 from bull C semen). Fetuses were aseptically collected at slaughter and ranged in gestational age from 62 to 107 days (fetus 7, 101 days; fetus 8, 62 days; fetus 9, 68 days; fetus 10, 95 days; fetus 11, 107 days; and fetus 12, 99 days). After removal of the skin, approximately 1 g of muscle from the biceps femoris was minced in 5 ml of 0.05% trypsin with 0.53 mM EDTA and incubated for 1 h in 5% CO2 in air at 38.5°C. An equal volume of high glucose Dulbecco modified Eagle medium (DMEM) with 10% fetal calf serum (FCS), 4 mM glutamine, 50 U/ml of penicillin and 50 µg/ml streptomycin was added and the cells were centrifuged at 800 x g for 10 min. Pellets were resuspended in fresh DMEM and centrifuged again, after which the cells were transferred into three T75 flasks and cultured at 5% CO2 in air at 38.5°C. Primary muscle fibroblast cultures were passaged either once or twice before being frozen in 92% FCS and 8% dimethyl sulfoxide (DMSO). Cells were passaged when the cells reached 80% to 90% confluence, usually taking 34 days. There were minor growth rate differences between fetuses. Bovine fetal fibroblasts from fetuses 7 and 10 were frozen at passage 1 and BFF from the remaining fetuses were frozen after two passages. Lung fibroblasts were also harvested from fetuses 7, 8, and 10. Lung BFF from fetus 7 was frozen at passage 1 and those from fetuses 8 and 10 were frozen at passage 2.
Bovine Fibroblast Transfection and Selection
Prior to use as nuclear donor cells, BFF were thawed and cultured for 25 days in DMEM with 10% FCS. An aliquot (400 µl) of single-cell suspension containing approximately 1 x 107 cells/ml (4 x 106 cells) was transfected with 10 µg of a transgene construct containing genes for neomycin resistance (neo), green fluorescent protein (GFP), and a peptidoglycan hydrolase (lysostaphin) with expression directed to mammary gland secretory epithelium [37]. The neo coding sequence, driven by the SV40 large T antigen regulatory element, was derived from pEGFP-N1 (Clontech, Palo Alto, CA) for expression in BFF during cell culture. The human elongation factor 1a promoter was used to express nuclear localized, enhanced GFP gene during the blastocyst stage of development [14, 38]. Transfection was achieved in an electroporation cuvette with 4-mm gap width (400 V, 500 µF; Bio-Rad Pulsar II, Hercules, CA). After transfection cells, were cultured for 48 h in DMEM in 100-mm tissue culture dishes. Selection for stable integrants was initiated by addition of G418 (400 µg/ ml). Cells remained under selection and were refed every 7 days with G418-containing medium. Colonies in four or five replicate plates were counted and examined for GFP fluorescence on the 19th day of culture, following initiation of G418 selection (21 days after transfection). Selected GFP-expressing colonies were isolated with an 8-mm cloning cylinder. The cells intended as nuclear donors were harvested by washing twice with Ca2+- and Mg2+-free PBS (plus 0.05% trypsin). Isolated cells were diluted in TL-HEPES with 10% FCS, removed from cloning cylinders, and centrifuged at 800 x g for 10 min. Pellets were resuspended in TL-HEPES and transferred to manipulation drops for nuclear transfer.
Three experiments were designed to compare the effectiveness of BFF isolated from muscle and lung in somatic cell nuclear transfer and to assess the influence of culture media, serum concentration, and oxygen concentration.
Experiment 1 Bovine fetal fibroblasts were isolated from hip or lung tissues and cultured in either minimum essential medium (MEM) or DMEM in a 2 x 2 factorial design. Transfected hip and lung BFF were selected with G418 (400 µg/ml) in either DMEM with 10% FCS in 5% CO2 in air or MEM with 10% FCS in 5% CO2 in air for 1820 days until they were used as nuclear donors. The usefulness of these BFF as nuclear donors in the production of blastocysts was evaluated. Grade 1 and 2 blastocysts were transferred to available recipients.
Experiment 2 The influence of serum starvation during G418 selection was tested. Transfected hip BFF were cultured in MEM or DMEM with 10% FCS and 5% CO2 in air during the first 14 days of selection. On Day 14, one half of the dishes were fed as usual with 10% FCS in appropriate media and the other half were fed with 0.5% FCS. G418 selection was continued in 5% CO2 in air for an additional 46 days, until the cells were used for nuclear transfer. Only BFF grown in MEM were used for nuclear transfer (NT). Blastocyst development was assessed, and grade 1 and 2 blastocysts were transferred to available recipients.
Experiment 3 Transfected BFF were subjected to oxygen concentrations of 5% or 20% (air) during G418 selection in a modified crossover experimental design. Hip BFF were grown in MEM with 10% FCS in either 5% CO2 in air (high O2) or 5% CO2 + 5% O2 + 90% N2 (low O2) before and after transfection (high O2 before and after transfection; low O2, before and after transfection). Hip BFF in dishes were also cultured in high or low O2 concentrations and then switched to the other concentration after transfection (high O2 before, low O2 after transfection; low O2 before, high O2 after transfection). During the last 5 days of selection, serum concentration was reduced to 0.5% for all groups.
Clonal growth was assessed in all three experiments. Dishes were stained with R-250 Coomassie blue stain (1.25 mg/ml) + 10% glacial acetic acid + 50% methanol for 10 min on Day 28 posttransfection. After two washes, dishes were air dried and colonies were counted.
Approximately half of the blastocysts generated in these experiments, along with 120 additional blastocysts were transferred to recipients in an attempt to produce calves. The additional embryos were generated as described in these experiments but were not included in the analyses because they did not have a comparison group for a variety of technical reasons (not enough oocytes on a given day, contamination of a treatment group, not enough recipients, etc.).
Oocyte Harvest, Enucleation, and Nuclear Transfer
Oocytes were purchased from a commercial supplier (BoMed; Madison, WI) and matured in transit in M199 + LH and FSH. Additionally, ovaries were purchased from a local slaughterhouse (MoPAC, Allentown, PA) and matured in the laboratory using a protocol similar to that previously described [39]. Oocytes from BoMed were only used for in vitro studies, while oocytes from MoPAC were used for both in vitro and in vivo studies. Briefly, ovaries were washed in 1% Nolvasan in 0.9% saline and transported to the laboratory at room temperature within 35 h after slaughter. Four- to 8-mm follicles were sliced [40] with scalpel blades and the ovaries were shaken in beakers containing 150 ml of HEPES-buffered Tyrodes lactate solution (TL-HEPES, BioWhittaker, Inc., Walkerville, MD). The oocytes were washed in Em-Con filters (model #04135; ImmunoSystems, Inc., Scarborough, ME). Cumulus-oocyte complexes were placed in culture wells, pipetted to partially remove cumulus, which was left in the well, then matured in Ham F-10 with 10% FCS, penicillin, streptomycin, and LH [11]. Seventeen hours after the initiation of maturation, oocytes were vortexed in 1 ml of 0.9% saline containing 1% hyaluronidase and 10% PVP for 4 min, washed in Dulbecco PBS (D4031; Sigma-Aldrich, St. Louis, MO) with 1% BSA (A3311; Sigma) and evaluated for polar bodies. Oocytes were cultured further in Ham F-10 with 10% FCS until enucleated 1820 h after initiation of maturation.
Just before enucleation, oocytes were exposed to cytochalasin B and Hoechst 33342 DNA stain in TL-HEPES for 20 min and then transferred to TL-HEPES with 1% BSA. Oocytes were manipulated at room temperature and removal of the metaphase plate was confirmed by briefly exposing the enucleation needle to fluorescent light (350-nm excitation, 450-nm bandpass filters). Enucleated oocytes (cytoplasts) were returned to maturation medium. Within several hours after enucleation, fibroblasts were inserted into the perivitelline space of cytoplasts to form couplets. Once 1015 couplets were prepared, they were immediately transferred to electrofusion solution (270 mM mannitol, 50 µM MgCl2). After a short equilibration (less than 10 min), the couplets were transferred to a fusion chamber with a 1.0-mm gap between electrodes, mechanically aligned, and fused with a single DC pulse of 115-V magnitude and 42-µsec duration (BTX 200, Inc., Hawthorne, NY).
Embryo Culture and Transfer
Couplets were then transferred to CR1aa media [41]. One to 4 h after fusion, couplets were activated in 5 nM Ionomycin (Sigma-Aldrich) in CR1aa for 4 min followed by a 4-h exposure to 1.9 mM 4-dimethylaminopyridine (Sigma-Aldrich) in CR1aa. After activation, the fused couplets were washed in CR1aa and cultured for approximately 16 h before being transferred to BARC-1 [38] in 5% CO2 + 5% O2 + 90% N at 38.5°C for 7 days. Blastocysts from each treatment were counted and GFP expression evaluated before embryo transfer. Estrus-synchronized 14- to 17-mo-old Holstein heifers received two grade 1 or 2 blastocysts on Day 7 or 8 of their estrous cycle. Parthenogenotes served as activation controls for each day and in vitro-fertilized embryos served as media controls. Return to estrus was monitored with the aid of the Heat Watch system (DDX, Inc., Denver, CO). Ongoing pregnancies were observed by ultrasound on Days 39, 56, and 69 posttransfer. Fluid detected in the uterus was taken as an initial indicator of pregnancy. Heartbeats were observed by Day 56.
All procedures involving live animals were conducted in accordance with the Guide for the Care and Use of Agricultural Animals in Agricultural Research and Teaching and approved in advance by the Beltsville Agricultural Research Center's Animal Care and Use Committee.
Data Analysis
All statistical analyses were performed with SPSS version 10 software (SPSS, Inc., Chicago, IL). Least-square means and their standard errors are reported. Fused couplets served as the denominator in calculating blastocyst development rates and as a weighting factor in some analyses. Blastocyst percentages were first arcsine transformed before being subjected to analysis of variance. Fetuses from which fibroblasts were derived were always a factor in the statistical model. Other factors such as source tissue, fibroblast culture media, oxygen, and serum concentration were included when appropriate. Pregnancy rates were compared by chi-square analysis.
| RESULTS |
|---|
|
|
|---|
Blastocyst development was evaluated in a 2 x 3 x 2 factorial experimental design. The ability of fibroblasts isolated from hip muscle or lung tissue of four fetal fibroblast sources cultured in MEM or DMEM were compared for their ability to form colonies, express GFP, and produce blastocysts when used as nuclear donor cells.
Lung tissue from fetuses 7, 8, and 10 produced more colonies than hip-derived fibroblasts when enumerated on Day 28 of culture (8.3 ± 0.3 vs. 3.1 ± 0.2 colonies/plate), but hip fibroblasts from fetus 9 were more proliferative than lung fibroblasts (1.2 ± 0.07 vs. 5.3 ± 0.7 colonies, fetus x tissue interaction, P < 0.001). Fetuses 7, 8, and 9 generated a higher proportion of GFP-expressing colonies from hip fibroblasts than from lung fibroblasts (59% ± 2% vs. 29% ± 3% green colonies/plate), while lung tissue from fetus 10 generated a higher percentage of green colonies than did hip fibroblasts (49 ± 3% vs. 67 ± 5%, fetus x tissue interaction, P < 0.001). An interaction between source tissue type and culture media on proportion of GFP-expressing colonies was also observed; lung fibroblasts expressed GFP more frequently when cultured in DMEM while hip-derived fibroblasts produced a higher proportion of GFP colonies when cultured in MEM.
Couplet fusion rate was only moderately influenced by the factors in this study. Fibroblasts derived from the hip tissue of fetus 10 tended to fuse more effectively than fibroblasts from lung (84% vs. 73% fusion, P = 0.048). No other influence on fusion rate by fetus, tissue type, or media was observed (data not shown).
When development to blastocysts was considered, fibroblasts derived from hip muscle had a clear advantage. On average, hip muscle fibroblasts resulted in 50% more blastocysts than did fibroblasts derived from lung tissue (26.1% ± 0.5%, n = 978 vs. 17.0% ± 0.6%, n = 901, P < 0.001). The ability of hip and lung fibroblasts to produce blastocysts were not influenced by the fetus from which they were derived (P = 0.500).
Fibroblasts from fetus 8 produced a higher proportion of blastocysts when cultured in MEM than when cultured in DMEM, whereas culture media did not influence blastocyst development from fetus 7 or 10 fibroblasts (Table 1).
|
An interaction was also observed between culture media and tissue source. Fibroblasts derived from hip muscle produced blastocysts at a higher rate if the fibroblasts were cultured in MEM. Alternatively, lung-derived fibroblasts produced over 30% more blastocysts if they were cultured in DMEM (Table 2).
|
Experiment 2. Influence of Serum Concentration, Culture Media, and Fetal Fibroblast Source
The debate in the scientific literature as to whether donor cell serum starvation is advisable, coupled with our observation that serum starvation appears to compromise the health of fibroblasts, led us to compare colonial growth and transgene expression at two serum concentrations in two fibroblast culture media. Blastocyst development from hip fibroblasts from three fetuses served as nuclear donor cells and MEM was used as the fibroblast culture medium.
In this second experiment, fibroblasts from fetus 11 produced the most colonies/plate (7.9 ± 0.5) and fibroblasts from fetus 7 the fewest (3.2 ± 0.5, P < 0.001). A statistically significant interaction was observed on colonial growth between serum concentration and medium. Cells grown in DMEM with low serum proliferated poorly and somewhat unpredictably (1.0 ± 1.0 colonies/plate), whereas fibroblasts grown in MEM with 10% serum generated the most colonies per plate (4.0 ± 0.4, serum x media interaction, P = 0.017). None of the treatment conditions influenced the proportion of fibroblast colonies that expressed GFP.
As in the first experiment, fusion rate was not influenced by treatment parameters studied. Fusion rates of couplets for fibroblasts incubated in 0.5% or 10% serum were identical at 74%.
Fetal fibroblast cell source affected the influence of serum (Table 3). Serum concentration for fibroblasts from fetus 7 had no effect on blastocyst development. In contrast, blastocyst rate was highest when fibroblasts were incubated in 10% serum for fetus 10 and in 0.5% serum for fetus 11 during the final week of culture.
|
Experiment 3. Influence of Oxygen Tension
In the final series of experiments, the influence of oxygen tension on colony formation and transgene expression of fibroblasts was tested in a crossover experimental design. Subsequently, blastocyst development was evaluated in a factorial design with fetus and O2 tension (atmospheric and low) as main effects.
Colony propagation was greatest when fibroblasts were cultured in 20% O2 and then switched to 5% O2 following transfection (5.8 ± 0.4 colonies/plate, P < 0.001). The other three treatments produced fewer colonies and did not differ from one another.
Fibroblasts from fetus 10 produced almost twice as many colonies (5.6 ± 0.4 colonies/plate) as did fibroblasts from fetuses 7 (3.3 ± 0.4 colonies/plate) and 12 (3.2 ± 0.4 colonies/plate, P < 0.001). The highest proportion of GFP-expressing colonies per plate resulted from fetus 7 cells (75% ± 6%; fetus 10, 57% ± 8%; fetus 12, 41% ± 5%, P < 0.001). The four oxygen tension treatments did not influence the proportion of GFP-expressing colonies.
In an initial pilot study (n = 152 fused couplets), embryos produced from fibroblasts cultured in low or normal (air) oxygen tension were either cultured in BARC-1 or G1/ G2 (Vitrolife, Denver, CO) media for 7 days. No differences in blastocyst development were attributable to embryo culture media (P = 0.620). Therefore, in all subsequent trials, the embryo culture media comparison was dropped and embryos were only cultured in the G1/G2 system.
Oxygen tension during fibroblast culture influenced both rates of couplet fusion and blastocyst development in a fetal fibroblast cell source-dependent manner. Fibroblasts from fetus 7 exhibited a higher fusion rate when grown in low oxygen tension (Table 4). However, the subsequent development of those embryos was extremely poor. While the oxygen tension in culture of donor fibroblasts had no influence on blastocyst development for fetus 10 fibroblasts, higher oxygen concentration was beneficial for blastocyst development for embryos derived from fetus 12 fibroblasts.
|
Embryo Transfer
From 4007 couplets, 920 good-quality blastocysts were generated, including 120 not produced in the experiments described above. Overall blastocyst production rate was lowest for embryos generated from fetus 8 fibroblasts and highest from fetus 11 fibroblasts (Table 5, P < 0.001). The eight live calves produced arose from BFF7 and BFF10 fibroblasts.
|
Eighty-two percent of potential pregnancies were lost within 56 days after transfer (Fig. 1). At that time, the pregnancy rate of recipients receiving embryos generated from BFF7 and BFF10 cells could be distinguished from pregnancy rates of the other recipients. Average pregnancy rate, at 8 wk, for embryos derived from BFF7 and BFF10 cells was higher (24%) than the mean pregnancy rate of recipients receiving embryos produced with cells from the other four fetuses (7%, P < 0.001).
|
The overall efficiency of producing calves, 1% of embryos transferred, was too low to provide adequate statistical power to assess the influence of the various fibroblast and embryo culture conditions tested in the three experiments. However, enough embryos were transferred to demonstrate that fibroblasts from different fetuses affect the calving rate (Table 5). The success rate of blastocyst production seemed to be unrelated to the ability of those embryos to survive to term in utero.
The superior in vitro performance of fibroblasts from fetuses sired by bull C (BFF 11 and 12) was not reflected in in vivo development rates. On the contrary, embryos derived from fetuses sired by bull A (BFF 7, 8, and 9) had substantially higher pregnancy rates than did embryos derived from bull C fetuses. All of the calves born alive were derived from fetuses sired by bulls A and B. As a result, there was a tendency for mean gestation length of recipients receiving embryos from fetal fibroblasts sired by bulls A and B to be longer than for those sired by bull C (sire A, 49.801 ± 4.256; sire B, 58.979 ± 5.814; sire C, 35.058 ± 7.941; P = 0.053). At 8 wk postestrus, the proportion of bull A recipients (recipients receiving embryos derived from fetuses sired by bull A) was 20% whereas bull C recipients was 2%. Bull B (fetus 10) recipient pregnancy rate was similar to bull A at 22%.
| DISCUSSION |
|---|
|
|
|---|
In the first experiment, the influence of fibroblast culture media and fibroblast tissue source on in vitro development was assessed. Many somatic cell types such as cumulus or granulosa, oviductal, uterine, skin fibroblasts, liver cells, thymocytes, spleen cells, and macrophages have been used as nuclear donor sources in bovine nuclear transfer with no clear advantage except that cells from adults do not support in vitro or in vivo development as well as cells from neonates or fetuses [10, 27, 42]. For purposes of producing genetically engineered cattle, fetal fibroblasts are an appealing nuclear donor source because they readily survive in vitro culture. Fibroblasts are most commonly derived from lungs, muscle, and skin. Of the various sources of fibroblasts, lung fibroblasts are among the most prolific [43]. The use of lung fibroblasts as nuclear donors has been previously reported [44] but has not been directly compared with other fibroblast sources. In our hands, lung fibroblasts were demonstrably less efficient as nuclear donor sources for blastocyst production than muscle fibroblasts and were more difficult to manipulate than those derived from fetal muscle.
DMEM appears to be the most commonly used culture medium for somatic donor cells [14, 21, 45], though MEM is also a widely used synthetic cell culture medium [33, 46] and was used in a modified form, GMEM, in the first reported somatic cell nuclear transfer study [1, 33, 46]. Tissue culture medium 199 has also been used to a lesser extent [33, 4649]. In our direct comparison of MEM and DMEM, both media performed similarly across fetal fibroblast sources, but MEM was superior for culturing hip muscle fibroblast donor cells for blastocyst production (Tables 1 and 2). Therefore, MEM is now used as our standard media for fibroblast culture.
Manipulating serum concentration of fibroblast cultures before nuclear transfer was adopted as a means of synchronizing the cell cycle of the donor cell and cytoplast [50]. Blastocyst production rate was enhanced when fetal bovine cells were serum starved, but the advantage of serum starvation was not apparent for fibroblasts harvested from adults [16]. Others have found no benefit from serum starvation for enhancing cleavage or blastocyst production in the bovine [29] or porcine NT embryo development [51]. In the studies reported here, culturing fibroblasts in either 0.5% serum or 10% serum for the final week before use as nuclear donors had differential effects on blastocyst development depending on the fetal fibroblast source (Table 3). Fibroblasts from the three fetuses tested all responded differently to serum concentration. Performance of fibroblasts from the two fetuses that produced live young either exhibited no preference or produced more blastocysts when cultured in 10% serum. Because the three fetuses tested were sired by three different bulls, it is not possible to distinguish between a potential genetic response to serum concentration and differences between fetuses.
Though the literature and our study is equivocal regarding the benefits of serum concentration in fibroblast cultures before NT, the advantage of 10% serum has been clearly demonstrated in the rate of calf production from transfected fibroblasts [5]. In that study, transfected fibroblasts cultured in 10% serum and transferred to recipients resulted in a calving rate of 29%, as compared with 4% if the transfected fibroblasts were serum starved (0.5% serum) before use as nuclear donor cells [5]. It was also shown that, whereas 10% serum was beneficial to transfected fibroblasts, the opposite was true for nontransfected fibroblasts. The calving rate of nontransgenic clones was twice the rate in the serum-starved group compared with the 10% serum group. We have confirmed that finding with nontransfected BFF10 fibroblasts (data not shown).
Growth of mammalian cells in culture was enhanced and population doublings extended by approximately 25% when oxygen tension was reduced from 20% to 10% [52]. The length of telomere shortening, as a result of cell division, is influenced by oxidative stress. Atmospheric (20%) oxygen tension in fibroblast cultures can cause telomere shortening of more than four times that observed in low O2 cultures [53]. Atmospheric oxygen concentration (in which most tissue cultures are conducted) is considerably higher than the oxygen tension found in most tissues, which is closer to 5% [43]. Cultured cells from monkey, mouse, and rat exhibited equal or improved colony formation under low (13%) oxygen tension [52, 54, 55]. However, the beneficial effects of low oxygen tension for cultured cells has not always been demonstrated [43, 56]. In a parallel study, plating efficiency (i.e., colony formation from single cells) of hip muscle fibroblasts from the same fetuses described in this report confirmed that low oxygen tension did not enhance colony formation [57]. However, as with the serum concentrations, the influence of oxygen tension during fibroblast culture on somatic cell nuclear transfer efficiency was fetus dependent. The three fetuses tested were sired by three different bulls. Therefore, it is not possible to distinguish between a fetus or genetic affect.
Over the three experiments, blastocyst development rate ranged from 17% to 30% (mean ± SEM, 22% ± 3%) for the six fetuses tested. Contemporaneous in vitro-produced and parthenogenote blastocyst development rates ranged from 2% to 63% (33% ± 2%) and 13% to 59% (36% ± 9%), respectively. There was a clear clustering in blastocyst production based on sire. Fetuses of bull C (BFF 11 and 12) produced nuclear transfer blastocysts most efficiently and those of bull A (BFF 7, 8 and 9) least efficiently. However, favorable in vitro development did not predict in vivo success. Fetuses of bull C produced no live offspring. It should be noted that only 104 embryos representing bull C were transferred (about half as many as for the other two bulls). Even so, based on the data presented here, we would have expected at least one calf from bull C. These observations suggest there may be a genetic component to the success of fetal cells as nuclear donors, but genetics that favor in vitro development may not be the most desirable.
These experiments were conducted over a 3-yr period. The overall strategy was to superimpose a new treatment on the best conditions defined in the previous experiment (Expt. 1, Hip, MEM; Expt. 2, 0.5% serum; Expt. 3, 20% O2). So it may not be too surprising that our calving rate improved over time (Expt. 1, 1% of embryo transfers [ETs]); Expt. 2, 3% of ETs; Expt. 3, 10% of ETs, P = 0.015). However, the improvement with time may simply be attributable to our increasing competence with the process or random chance.
The 8-wk pregnancy rate may serve as a predictor of the success of bovine nuclear transfer experiments. The average pregnancy rate for recipients receiving embryos generated from BFF7 and BFF10, the two fetal fibroblast sources that produced live calves, was 23%. The aggregate pregnancy rate for the other four fetuses was 7%. From the work reported here, one would expect three or four pregnant recipients at 56 days from 15 to 20 embryo transfers from good fibroblast sources, and no pregnancies or possibly one pregnancy for embryos derived from fibroblast sources such as fetuses 8, 9, 11, and 12. Interestingly, the 8-wk pregnancy rate may not be universally predictive. In a similarly sized study, differences did not appear apparent until approximately 160 days of gestation [5].
There are few published studies that provide enough detail to determine if somatic cell nuclear transfer calving efficiencies observed are attributable to fibroblast cell lines or to a fetal effect or possibly a genetic effect. To try to ensure that we were comparing the influence of fetuses as the source of variation, we pooled all the cells harvested from a given fetus before aliquoting and freezing. The differences observed were therefore due to experimental variation, fibroblast treatments, or fetal fibroblast source. In our hands, the most profound differences in nuclear transfer efficiency were attributable to fetuses from which fibroblast were harvested. Though the data set is too small to be compelling, there is a suggestion that the genetic background of the fetus may contribute to the overall efficiency of producing somatic cell clones. The use of cells from fetuses produced by sires A and B resulted in pregnancy rates of 21% and 22%, respectively, at 56 days postestrus, but the pregnancy rate was only 2% for recipients receiving embryos generated from sire C fetuses.
These results support the notion that the source of the donor cells may be one of the most important factors in determining the success of a somatic cell nuclear transfer project [8] and suggest that it may be possible to make a reasonable assessment of the clonability of bovine nuclear transfer donor cells by 60 days posttransfer.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Received: 2 January 2004.
First decision: 4 February 2004.
Accepted: 3 March 2004.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. K Berg, C. Li, G. Asher, D. N Wells, and B. Oback Red Deer Cloned from Antler Stem Cells and Their Differentiated Progeny Biol Reprod, September 1, 2007; 77(3): 384 - 394. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Shi, F. Lu, Y. Wei, K. Cui, S. Yang, J. Wei, and Q. Liu Buffalos (Bubalus bubalis) Cloned by Nuclear Transfer of Somatic Cells Biol Reprod, August 1, 2007; 77(2): 285 - 291. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Yu, C. Ding, E. Wang, X. Chen, X. Li, C. Zhao, Y. Fan, L. Wang, N. Beaujean, Q. Zhou, et al. Piezo-assisted nuclear transfer affects cloning efficiency and may cause apoptosis Reproduction, May 1, 2007; 133(5): 947 - 954. [Abstract] [Full Text] [PDF] |
||||
![]() |
Q. Zhou, S.H. Yang, C.H. Ding, X.C. He, Y.H. Xie, T.B. Hildebrandt, S.M. Mitalipov, X.H. Tang, D.P. Wolf, and W.Z. Ji A comparative approach to somatic cell nuclear transfer in the rhesus monkey Hum. Reprod., October 1, 2006; 21(10): 2564 - 2571. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. D. Michael, I. M. Alvarez, O. M. Ocon, A. M. Powell, N. C. Talbot, S. E. Johnson, and A. D. Ealy Fibroblast Growth Factor-2 Is Expressed by the Bovine Uterus and Stimulates Interferon-{tau} Production in Bovine Trophectoderm Endocrinology, July 1, 2006; 147(7): 3571 - 3579. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |