| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Reproductive Technology |
Department of Molecular Animal Breeding and Biotechnology,4 Gene Center of the Ludwig-Maximilian University, D-81377 Munich, Germany
Department of Animal Breeding and Genetics and Central Biotechnical Unit Strahlenzentrum,5 Justus-Liebig-University, D-35392 Giessen, Germany
Bavarian Research Station for Animal Breeding,6 D-85586 Grub, Germany
Bavarian Research Center for Biology of Reproduction (BFZF),7 D-85764 Oberschleissheim, Germany
Agrobiogen GmbH,8 D-86567 Hilgertshausen, Germany
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
assisted reproductive technology, conceptus, developmental biology, pregnancy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Accumulating evidence indicates that incomplete or inappropriate epigenetic reprogramming of donor nuclei is likely to be the primary cause of low embryo production efficiency, high embryonic and fetal losses, and abnormalities observed in individuals reconstructed by SCNT [9–11]. The effects of different donor cell types, donor cell cycle stages, and donor cell treatments on developmental potential of SCNT embryos have been [12] and still are [13] under extensive investigation, and recent data have shown remarkable differences in changes of epigenetic features such as histone methylation and acetylation of different donor cell types, which correlated with the developmental potential of cloned embryos [14].
The recipient oocyte cytoplasm, however, has received much less attention, although the fundamental importance of oocyte cytoplasmic factors for reprogramming of transferred nuclei and early embryonic development is well recognized [9, 15–17]. Previous studies in bovine SCNT have compared oocyte activation and maturation protocols [18] or follicular size of cytoplast donors [19], but investigations relating to the maternal source of oocytes are confined to a single report on the effect of maternal lineage of oocyte donor on SCNT efficiency [20].
Data obtained in mice show that oocyte maternal genetic and epigenetic factors may lead to complex nuclear-cytoplasmic interactions that affect blastomere fragmentation [21], blastocyst formation [22], embryonic gene expression [17], and adult phenotype [23]. Moreover, strain-specific differences in mouse oocytes with regard to their contributions to epigenetic inheritance have been observed [24], and it has been proposed that the genetic origin of the oocyte may, through epigenetic effects, also have a significant effect on the outcome of SCNT experiments [25]. The latter assumption is corroborated by results of a recent study in bovine, which reported significant effects of maternal lineage of oocyte donors on SCNT efficiency based on the number of transferable embryos obtained [20]. This suggests that identification of suitable combinations of nuclear donor cells and recipient oocyte cytoplasms for SCNT could improve efficiency in the production of viable and phenotypically normal offspring.
To test this hypothesis, we generated nucleocytoplasmic hybrids [23] by transferring nuclei from Brown Swiss granulosa cells into cytoplasts of distinct maternal origin (Simmental, Dwarf Zebu). Nuclear transfer embryos produced from the same donor cells and random Brown Swiss cytoplasts served as controls. Analysis of in utero developmental capacity, phenotype of Day 80 fetuses, and metabolism of fetal cells revealed complex nuclear-cytoplasmic interactions that could be further dissected by SCNT experiments.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals and Oocyte Cytoplasm
All oocyte donors used to collect oocytes with maternal lineage-specific cytoplasm were initially selected according to pedigree data. The type of cytoplasm was verified and distinguished by mitochondrial DNA (mtDNA) haplotype analysis. Oocytes with Bos taurus cytoplasm A or B were collected from two groups of Simmental cows and heifers produced by multiple ovulation embryo transfer from single founder dams and the same two Simmental bulls in each group [20]. Cytoplasms A and B were assigned arbitrarily based on mtDNA haplotype (GenBank accession AF386912 and AF386913). All Bos indicus oocytes were recovered from ovaries of maternally related Dwarf Zebu cattle (GenBank accession AF492437). Oocytes with random but breed-specific B. taurus (B. taurus R) cytoplasm were obtained from slaughtered Brown Swiss cattle identified by ID number. Nuclear donor granulosa cells were obtained from a B. taurus (Brown Swiss) cow with an mtDNA haplotype (GenBank accession AF492432) distinct from all predefined recipient oocyte mtDNA haplotypes. Simmental heifers were used as recipients for reconstructed embryos throughout. Artificial insemination (AI) control fetuses were produced from Brown Swiss heifers by using semen of a single Brown Swiss bull. All experiments were performed according to the relevant guidelines for the care and use of animals.
Oocyte Recovery and In Vitro Maturation
Oocytes with B. indicus (Dwarf Zebu) or random B. taurus (Brown Swiss) cytoplasm were obtained from ovaries of slaughtered animals. Ovaries were transported to the laboratory in PBS at 25°C–30°C within approximately 2 h. Cumulus-oocyte complexes (COCs) were obtained by aspiration and slicing of ovaries.
COCs with B. taurus (Simmental) cytoplasm A or B were collected by ultrasound-guided follicle aspiration, ovum pick-up, as described previously [20]. The collection medium consisted of 100 ml Tyrode lactate solution buffered with 10 mM Hepes supplemented with 2 ml fetal calf serum (FCS, heat-inactivated; Life Technologies, Karlsruhe, Germany) and 6 mg sodium heparin (175.5 IU; Kraeber, Hamburg, Germany). Collected COCs were washed with PBS enriched with 2 ml/L FCS and transferred into collection medium without sodium-heparin and supplemented with 10% FCS for transport. Oocytes were kept in this medium at room temperature for approximately 1 h.
All recovered oocytes were used for maturation without any preselection. COCs were washed three times in culture medium TCM 199 (Seromed, Berlin, Germany) supplemented with L-glutamine (100 mg/L), NaHCO3 (3 g/L), Hepes (1400 mg/L), pyruvate (250 mg/L), L-lactic-Ca-salt (600 mg/L), and gentamicin (55 mg/L; Seromed). Oocytes were transferred to four-well plates (Nunc, Roskilde, Denmark) with 400 µl TCM 199 containing 0.01 units bovine FSH and bovine LH (Sioux Biochem, Sioux Center, IA) supplemented with 10% estrous cow serum (ECS). After 18 h maturation at 39°C in an atmosphere of 5% CO2 and maximum humidity, oocytes were exposed for 5 min to modified PBS (mPBS; PBS plus 3 mg/ml bovine serum albumin) containing 3 mg/ml hyaluronidase, vortexed for 4 min, and stripped of cumulus cells by gentle pipetting.
Karyoplasts
Granulosa cells obtained after ovum pick-up were washed twice in saline solution and dispersed by exposure to 0.1% (w/v) trypsin (Gibco, Grand Island, NY). The cell suspension was then transferred into 5-cm culture dishes containing Dulbecco modified Eagle medium (DMEM; Gibco) supplemented with 10% (v/v) FCS (Biochrome, Berlin, Germany); 2 mM L-glutamine; 0.1 mM ß-mercaptoethanol; 2 mM nonessential amino acids; 100 IU/ml penicillin; and 100 µg/ml streptomycin. The cells were cultured until subconfluence at 37°C in a humidified atmosphere of 5% CO2 in air and then frozen in 10% (v/v) dimethyl-sulfoxide in FCS and stored in liquid nitrogen. For nuclear transfer experiments the cells were thawed and cultured for three to six passages until confluence just before nuclear transfer.
Nuclear Transfer
Oocytes were exposed for 15 min to 5 µg/ml cytochalasin B (Sigma) and 5 µg/ml Hoechst 33342. The metaphase chromosomes were removed within 1 h in mPBS containing 5 µg/ml cytochalasin B. Oocytes were visualized under epifluorescence microscope (Zeiss, Jena, Germany) to confirm the absence of chromatin. Single donor cells were transferred into the perivitelline space of enucleated oocytes, and the karyoplast-cytoplast complexes (KCCs) were exposed to a double electric pulse of 2.1 kV/cm for 10 µsec using the Zimmermann Cell Fusion Instrument (Bachofer, Reutlingen, Germany). KCCs were placed in the incubator in Ham F-12 medium supplemented with 0.3% bovine serum albumin.
Activation and Embryo Culture
Two hours postfusion, the KCCs were activated by a 5-min incubation in 7% ethanol followed by a 3-h culture in 10 µg/ml cycloheximide and 5 µg/ml cytochalasin B. Activated embryos were washed three times in culture medium before transfer into 100-µl drops of synthetic oviductal fluid medium supplemented with 2% basal medium Eagle amino acids (Gibco); 1% minimum essential medium (MEM) nonessential amino acids (Gibco); and 10% (v/v) ECS covered by paraffin oil (Merck, Darmstadt, Germany) and cultured at 39°C in a humidified atmosphere of 5% CO2, 5% O2, and 90% N2. At least five nuclear transfer sessions were performed with each type of cytoplasm over at least a 6-month period each, and sessions with different types of cytoplasm overlapped. Nuclear transfer involving the four different oocyte pools was performed by the same person (V.Z.).
Embryo Transfer and Artificial Insemination
On Day 7 after nuclear transfer, all viable embryos were transferred nonsurgically to synchronous recipients without further selection. On average, two to three embryos were transferred to each recipient (Table 1). Control fetuses were produced by inseminating heifers with frozen-thawed semen by standard procedures. Pregnancies were confirmed by ultrasonographic examination on Day 28 and by ultrasonographic examination or palpation before recipients were killed.
|
Recovery of Fetuses and Data Collection
Recipient heifers and inseminated controls diagnosed pregnant were killed in a local slaughterhouse at Day 80 after nuclear transfer and at Day 80.5 after AI. Logistic regression was used to determine the significance of type of cytoplasm on in utero developmental capacity of embryos (i.e., the percentage of viable fetuses recovered) and Fisher exact test was employed to test differences between groups in developmental capacity or survival rate of recovered fetuses. Fetal weight and dimensions (crown rump length, thorax circumference, femur length, umbilical cord length), organ weights (liver, kidney), and placenta weight were recorded, but not all fetuses had complete data records. Only female AI control fetuses were included in the control group (n = 8). The data were analyzed with the program SPSS for Windows version 11.0 (SPSS, Chicago, IL) using the general linear model procedure with the fixed effect of the fetus group to estimate group means. Differences between groups (t-test) were considered significant at P < 0.05.
Tissue samples were obtained from the biceps femoris muscle of all fetuses. Additional tissue samples (brain, heart, jejunum, kidney, liver, lung, skin, spleen, rumen, and cotyledons) were obtained from a subset of the 23 viable fetuses to search for nuclear donor cell mitochondria (heteroplasmy) in fetal tissues as previously described [26]. All samples were stored frozen at -20°C until further processing for mtDNA and microsatellite typing. Additional skin, liver, and heart tissue samples of four fetuses reconstructed with two different oocyte cytoplasms (two B. taurus A, two B. indicus) were collected in PBS for tissue culture experiments.
Tissue Culture and Oxygen Consumption Measurements
Fetal skin, heart, and liver tissue samples (100–200 mg) were minced and incubated for 10 min at 37°C in prewarmed trypsin/EDTA solution (Gibco). Disaggregated cells were transferred into DMEM supplemented with 10% (v/v) FCS (Gibco); centrifuged at 600 x g for 5 min; resuspended in fresh medium; and plated in culture dishes containing DMEM supplemented with 10% (v/v) FCS, 0.1 M MEM nonessential amino acids (Gibco), 0.1 M sodium pyruvate (Gibco), 1 mM L-glutamine (Gibco), 0.1 mM ß-mercaptoethanol, 100 IU/ml penicillin, and 100 IU/ml streptomycin. The cells were cultured until subconfluence at 37°C and 5% CO2, trypsinized, frozen in FCS with 10% (v/v) dimethylsulfoxide, and stored in liquid nitrogen.
For experiments evaluating the increase of cell numbers within defined time periods and cellular oxygen consumption, cells were passaged once (heart) or twice (skin, liver) in the culture medium described above and seeded at 7 x 105 cells in 28 cm2 culture dishes (four to six dishes per tissue and fetus, two independent experiments). Cultures of all four fetuses (two B. taurus A, two B. indicus) were set up and processed in parallel for each tissue and experiment. Cell cultures were trypsinized in late log phase after 48 h (skin) or 72 h (liver, heart), resuspended in 300 µl DMEM without glucose supplemented with 5% (v/v) FCS, and counted in Neubauer chambers. Five hemocytometer slides (improved Neubauer pattern) with two chambers each were used for each cell count. An aliquot of cells was frozen at -20°C for mtDNA analysis. Whole cell oxygen consumption measurements were performed with a Clark-type electrode [27]. A 200 µl aliquot of the counted cell suspension (2 to 3 x 106 cells) was transferred to the reaction chamber of a Hansatech DW1/AD oxygen electrode unit (Bachofer, Reutlingen, Germany) filled with 400 µl DMEM without glucose supplemented with 5% (v/v) FCS. Oxygen consumption rate was recorded at 37°C for 10–12 min. Heart cells tended to give nonlinear oxygen consumption rates in the first experiment; therefore, DMEM with glucose was used in the second experiment. In each experiment, cell cultures of all four fetuses were trypsinized, counted, and assayed for oxygen consumption in an alternating manner. It was thus possible to account statistically for even small amounts of drift of the oxygen electrode and for additional cell growth in the course of a 10-h experiment.
Statistical analysis of data obtained on the number of cells in culture and oxygen consumption rate per minute per cell from individual tissue culture experiments was performed with the program SPSS for Windows version 11.0 using the general linear model procedure. The mean cell number for each fetus was adjusted for the fixed effect of the nth culture dish analyzed to account for additional cell growth. Mean oxygen consumption per cell was adjusted for the nth culture dish measured to account for small amounts of electrode drift, and for the number of cells in each culture at the time of measurement, to account for the decrease in cellular metabolism with increasing confluence. A combined cytoplasm (B. taurus A or B. indicus)-specific analysis using the data of all four fetuses from two independent experiments for each cell type was performed (SAS release 8.02) by fitting a fetus effect within the fixed cytoplasm effect, in addition to fixed effects for the experiment, the nth culture dish, and the number of cells in each culture. Differences were considered significant at P < 0.05.
Microsatellite Typing, mtDNA Typing, and Sequencing of Mitochondrial Genomes
The nuclear genotype of all viable recovered fetuses was confirmed by typing genomic DNA samples with a panel of 13 standardized microsatellites recommended for parentage control in cattle by the International Society for Animal Genetics.
Mitochondrial DNA typing was performed with NlaIII as previously described [26]. Briefly, the complete mtDNA control region was amplified from total cellular DNA isolated from tissue samples with the E.Z.N.A. Tissue DNA Kit II (PEQLAB Biotechnologie GmbH, Erlangen, Germany) using Taq Polymerase (Qiagen, Hilden, Germany) and the primers mtDA (5'- CTCACCATCAACCCCCAAAGCT-3') and mtDB (5'-TCATCTAGGCATTTTCAGTG-3'). A 10-µl aliquot of the PCR product was then digested with NlaIII as recommended by the manufacturer (New England Biolabs GmbH, Frankfurt, Germany), and restriction fragment patterns were analyzed in 2% agarose (BIOzym Diagnostic GmbH, Hess. Oldendorf, Germany) gels using tris-borate-EDTA buffer according to standard procedures.
The mitochondrial genomes found in B. taurus A cytoplasm (GenBank accession AF386912) and in B. indicus cytoplasm (GenBank accession AF492437) were chosen for complete nucleotide sequence determination. Mitochondrial DNA was purified from liver tissue of slaughtered animals as described [28] and digested with EcoRI or EcoRI/HindIII as recommended by the manufacturer (New England Biolabs GmBH, Frankfurt, Germany). Restriction fragments of 1.6, 2.6, 4.8, and 7.3 kb that covered the complete mitochondrial genomes were isolated from agarose gels using the Qiaex II Kit (Qiagen) and cloned into the plasmid pZErO-2 using the Zero Background cloning kit (Invitrogen, Groningen, The Netherlands). The 2.6-kb B. taurus mtDNA fragment, but not the corresponding B. indicus fragment, was refractory to cloning. The sequence encoded by the 2.6-kb B. taurus fragment was, therefore, amplified by PCR from purified mtDNA in two overlapping segments of 1775 and 1412 bp using Pwo Polymerase (Hybaid, Heidelberg, Germany) with the primer combinations 5'-TGTCGTAGGTCCATATGG-3'/5'-CCCGATTCAGACAAGTAG-3' and 5'-GTCCCAGAAGTAACACAG-3'/5'-ATCTACTGAAGCTCCTGC-3'. Both strands of cloned and PCR-amplified fragments were sequenced by a primer walking approach using standard procedures on a LICOR 4200 (MWG Biotech, Ebersberg, Germany).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Embryo transfer resulted in nine (12.5%) B. taurus A, nine (13.8%) B. taurus B, three (50%) B. indicus, and 11 (16.7%) B. taurus R Day 80 fetuses, of which eight (11.1%), three (4.6%), three (50%), and 10 (15.2%) were viable, respectively (Table 1). Based on the number of transferred embryos, the proportion of viable fetuses was significantly affected by cytoplasm (likelihood ratio test, P < 0.02) and was higher for the limited number of embryos with B. indicus cytoplasm than for the B. taurus A (P < 0.05), B. taurus B (P < 0.01) and B. taurus R (P < 0.07) groups. Among the B. taurus groups, the proportion of viable fetuses was markedly lower for B. taurus B cytoplasm, but only the comparison with the B. taurus R group approached significance (P < 0.08). However, based on the total number of recovered fetuses, the proportion of surviving Day 80 fetuses was significantly lower for B. taurus B cytoplasm as compared with B. taurus A (P < 0.05) and B. taurus R (P < 0.02) cytoplasm.
Dead fetuses were in different physical condition with respect to maceration/resorption (Fig. 1A), but based on morphology and crown-rump lengths of fetuses that allowed measurements, we estimate that fetal death occurred around Day 60 of gestation.
|
Nuclear and Cytoplasmic Genes of Fetuses
All 23 viable recovered fetuses had the identical nuclear genotype expected from the donor cells (microsatellite data not shown). Mitochondrial DNA restriction fragment length polymorphism typing of fetal tissues and cultured cells was performed 1) to verify the cytoplasmic origin of individuals derived from defined oocyte pools (i.e., B. taurus A, B. taurus B, and B. indicus); 2) to estimate mtDNA variation among fetuses reconstructed from oocytes with random B. taurus (Brown Swiss) cytoplasm; and 3) rule out mtDNA heteroplasmy (detection limit 
2% [26]) in nuclear transfer fetuses or cultured cells.
The cytoplasm of fetuses reconstructed with oocytes from defined maternal lineages showed the NlaIII mtDNA control region profiles expected for B. taurus A, B. taurus B, and B. indicus mtDNA sequences, whereas fetuses derived from random B. taurus (Brown Swiss) oocytes revealed three different mtDNA haplotypes (Fig. 2). This indicates considerable cytoplasmic genetic diversity among fetuses with randomly collected B. taurus (Brown Swiss) cytoplasm, since NlaIII restriction sites surveyed only 20 of 998 PCR-amplified nucleotides. All intact fetuses, two dead fetuses that yielded DNA, and all tested cultured cell samples were homoplasmic in the restriction enzyme analysis.
|
The complete sequences of the B. taurus A and B. indicus mitochondrial genomes were 16 338 and 16 339 nucleotides long, and comparisons revealed extensive sequence polymorphism in coding and noncoding regions (Table 2). All respiratory chain enzyme complexes with mtDNA-encoded subunits, including complex V, were affected by nonsynonymous nucleotide substitutions.
|
Phenotype of Viable Day 80 Fetuses
Body and organ weights and dimensions of viable Day 80 SCNT and AI control group fetuses are presented in Figure 3. The SCNT fetuses are generally heavier, have a larger thorax circumference, and have a reduced crown rump length:thorax ratio. A more extreme example of this disproportionate phenotype is the overgrown fetus shown in Figure 1B. However, there are some remarkable differences among the SCNT fetuses depending on the type of cytoplasm used for embryo reconstruction. Body weight of nucleocytoplasmic hybrid fetuses with B. taurus A or B or B. indicus cytoplasm was not significantly different from control fetuses, but fetuses reconstructed with random B. taurus (Brown Swiss) cytoplasts clearly indicated fetal overgrowth when compared with AI controls (107.9 ± 5.8 g vs. 80.6 ± 6.5 g, P < 0.01). Furthermore, the coefficient of variation for body weight of eight fetuses with B. taurus A cytoplasm (14.4%) was similar to that for eight AI controls (10.9%), but much higher in 10 fetuses with random B. taurus (Brown Swiss) cytoplasm (25.7%). In addition, body weight, crown rump length, thorax circumference (P < 0.05), and Os femoris length (P < 0.01) of fetuses with B. taurus A cytoplasm differed significantly from fetuses with random B. taurus (Brown Swiss) cytoplasm.
|
Relative liver weight was higher than in AI controls in all but one group (B. taurus B) of SCNT fetuses. Fetuses reconstructed with B. indicus cytoplasm showed a markedly elevated absolute and relative kidney weight, umbilical cord length, and placenta weight. However, the mean placenta weight was still in the range encountered in individual AI controls, and placenta weight among all other SCNT groups and controls did not differ.
Growth and Oxygen Consumption of Cultured Fetal Cells
Two fetuses each from the B. taurus A and B. indicus nucleocytoplasmic hybrid groups were selected for tissue culture experiments. The cytoplasm of fetuses in both groups was the most diverged genetically (i.e., contained very different mitochondrial genomes; see Table 2). Primary cell cultures established from skin, liver, and heart tissue of the fetuses indicated significant cytoplasmic effects on the increase in cell number after defined time periods and on cellular oxygen consumption in all three investigated tissues in two independent experiments (Fig. 4, A and B). Cells derived from fetuses with B. indicus cytoplasm showed a marked increase in both cell number (skin fibroblasts, 22.3%, P < 0.001; liver cells, 17.5%, P < 0.001; cardiac cells, 39.7%, P < 0.001) and oxygen consumption (skin fibroblasts, 13.8%, P < 0.001; liver cells, 19.8%, P < 0.001; cardiac cells, 11.6%, P < 0.078) as compared with cells with B. taurus A cytoplasm (Fig. 4B).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The developmental capacity of embryos and fetuses was significantly affected by type of cytoplasm. Embryos with B. taurus B cytoplasm yielded the lowest proportion of viable fetuses and the highest number of fetuses in resorption among the three groups of embryos reconstructed with B. taurus cytoplasm. This is remarkable, because the B. taurus B cytoplasm was previously shown to yield a significantly lower proportion of transferable SCNT embryos in comparison with B. taurus A cytoplasm with the same nuclear donor cells as in the present study [20]. The poor capacity of B. taurus B cytoplasm to support development of SCNT embryos in vitro is thus extended to the in utero developmental capacity of transferred embryos. However, the cytoplasm-dependent developmental failure that leads to the marked differences in recovered fetuses seems to occur relatively late, during fetal development. This assumption is based on four lines of evidence: 1) the similar number of pregnancies per transferred embryo on Day 28 obtained with all three types of B. taurus cytoplasm, 2) the high number of dead fetuses with B. taurus B cytoplasm recovered on Day 80, 3) the similar number and proportion of fetuses with B. taurus cytoplasm recovered in total, and 4) the physical appearance of recovered dead fetuses.
The molecular nature of the cytoplasmic factors causing the observed differences is unknown and could involve epigenetic phenomena and/or cytoplasmic genes. Factors that influence epigenetic modifications after natural fertilization are apparently complex, and the precise interaction of the oocyte cytoplasm and the nuclear genome is crucial for normal development [9, 29]. Strain-specific maternal factors in mouse oocyte cytoplasm contribute to epigenetic inheritance and affect the developmental capacity of embryos, and thus provide clear evidence that epigenetic modifications that occur in the early embryo vary with genetic background [22, 24]. Several loci controlling epigenetic differences in oocyte cytoplasm, including postzygotic modifier loci, have been mapped by genetic linkage analysis in mouse models, but the underlying genes remain to be identified [25, 30]. Some of these loci could encode for ‘reprogramming factors' postulated to be present in oocyte cytoplasm and thought to be responsible for the ability of the oocyte to correctly reprogram somatic cell nuclei [9, 31]. Differences in regulatory epigenetic markers (e.g., DNA methylation, assumed to reflect reprogramming efficiency) have been described in bovine SCNT embryos [32–34]. Changes in epigenetic features can be brought about by maternal factors such as Dnmt1o, a variant form of Dnmt1 DNA (cytosine-5) methyltransferase synthesized and stored in the oocyte and responsible for maintaining DNA methylation patterns on alleles of imprinted genes [35]. Dnmt1o and Dnmt1 were recently shown to be abnormally regulated in SCNT mouse embryos [36]. Such defects could prevent clones from completing essential early developmental events and also contribute to developmental abnormalities and later failure. Genetically determined differences in how the oocyte cytoplasm modifies the incoming genome might thus explain the observed differences in reprogramming performance of bovine oocyte cytoplasms.
Maternal cytoplasmic genetic (i.e., mtDNA) effects could also directly influence or modulate developmental capacity. It has been speculated that induced artificial heteroplasmy could be responsible for low SCNT efficiency [37], but two fetuses in resorption that yielded DNA and could be assayed for heteroplasmy in our study turned out to be homoplasmic for oocyte mtDNA. Mitochondrial functions are nevertheless extremely diverse and complex [38], and natural variation in the mitochondrial genome affects a variety of parameters in natural and assisted reproduction [39]. Smith et al. [40] reported mtDNA effects on blastocyst rate in reconstructed mouse embryos; epigenetic effects seemed negligible in this model. This supports the previously suggested role of mtDNA haplotype in development of bovine SCNT embryos in vitro [20], which might also extend to the in utero developmental capacity of transferred embryos. The mtDNA sequence variation detected in the different bovine cytoplasms (see below) could therefore also play a role in SCNT success, and might even interact with epigenetic factors [41].
Some of the most intriguing findings of the present study are the differences in fetal phenotype depending on the type of oocyte cytoplasm used. Cytoplasm affects not only fetal weight and dimensions but also seems to influence organ weight. It is difficult to conceive how oocyte cytoplasmic factors could modulate specific body conformation traits such as Os femoris length or growth of specific organs in Day 80 fetuses. However, tissue-specific DNA methylation variation has been observed in SCNT mice [42], and adult phenotype (body weight, liver polypeptide expression) can be affected by epigenetic events caused by nuclear-cytoplasmic interactions in the early mouse embryo [23]. Such effects might be mediated by complex oocyte cytoplasm-dependent epigenetic marking processes that target DNA for later methylation in the fetus [30]. It is, therefore, entirely possible that oocyte cytoplasmic factors cause specific phenotypic effects in the different groups of bovine fetuses. Considering the problems with large offspring syndrome in SCNT animals, it is striking that this phenotype occurred only in fetuses with randomly collected B. taurus (Brown Swiss) cytoplasm and was clearly established in Day 80 fetuses. Comparisons between fetuses with defined B. taurus A (Simmental breed) cytoplasm and fetuses with randomly collected B. taurus (Brown Swiss breed) cytoplasm are of particular interest, because both cytoplasms yielded a similar number of fetuses with eight and 10 viable individuals, respectively. Body weight of fetuses with B. taurus A cytoplasm was not only normal but also much more homogeneous, with a similar coefficient of variation as in AI controls, which was 50% lower than in fetuses with random B. taurus (Brown Swiss) cytoplasm (Fig. 5). One explanation for the observed differences is obviously the variable cytoplasm with a diverse maternal genetic origin; this was also demonstrated by the mtDNA analysis. The high variation in phenotype could thus result from maternal nuclear genetic factors in the oocyte cytoplasm, maternal cytoplasmic genetic (i.e., mtDNA) effects, or both.
|
There is, however, another aspect that might account for phenotypic differences between the fetuses. The combination of Brown Swiss nuclei with Simmental breed cytoplasms (B. taurus A or B) and Dwarf Zebu (B. indicus) cytoplasms, but not with the randomly collected Brown Swiss cytoplasms, resulted in nucleocytoplasmic hybrids. None of the nucleocytoplasmic hybrids showed the marked overgrowth demonstrated by fetuses with Brown Swiss cytoplasm. The hybrid constellation could, therefore, also play a role in preventing the overgrowth phenotype. A hybrid vigor effect has been described in embryonic stem cell cloned mice, where pups reconstructed with donor cells from inbred lines invariably died shortly after birth, whereas most pups reconstructed with donor cells from F1 hybrids survived to adulthood [43]. This was, however, a classical nuclear genetic hybrid vigor effect, whereas the effect described here would be a novel nuclear-cytoplasmic interaction hybrid vigor effect. In this context, it is interesting that embryos with B. indicus cytoplasm showed a significantly higher in utero developmental capacity than any group of embryos with B. taurus cytoplasm. Although the number of transferred embryos reconstructed with B. indicus cytoplasm was limited, the data are unlikely to be confounded with experiment-specific effects, since the transferred embryos were reconstructed in three different nuclear transfer sessions that were each separated by several weeks over a period of several months. In addition, sessions that involved different sets of B. taurus oocytes were intermingled with B. indicus sessions. The greater vigor of embryos with B. taurus nucleus and B. indicus cytoplasm was accompanied by a marked increase in the number of cells recovered after defined time periods and oxygen consumption of cultured fetal cells as compared with cells with the same nucleus but B. taurus A cytoplasm. This suggests that the B. indicus cytoplasm in combination with the B. taurus nucleus yields a more vigorous phenotype, reminiscent of hybrid vigor effects described in numerous crossbreeding experiments with B. taurus and B. indicus cattle [44]. The extensive sequence variation in the mitochondrial genomes of B. taurus and B. indicus embryos and fetuses, in addition to epigenetic phenomena discussed above, could provide a molecular basis for such effects. Clearly, more data on B. taurus/B. indicus nucleocytoplasmic hybrids are needed, preferably in reciprocal combinations and with pregnancies carried to term, to assess final survival rate of fetuses.
We have shown that oocyte cytoplasm substantially affects embryonic and fetal development, fetal phenotype, and metabolisms of fetal cells. The observed differences are relevant for and could improve SCNT cloning efficiency and, if present in human, might have consequences for technologies such as ooplasmic transfer in assisted reproduction [39, 45]. Our data point to complex oocyte cytoplasm-dependent epigenetic modifications and/or nuclear DNA-mtDNA interactions that could be functionally dissected by SCNT after controlling the nuclear genome of oocyte donors by repeated backcrossings [38], or by using groups of transmitochondrial individuals [46] as oocyte donors for the production of embryos, fetuses, and offspring in further studies.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Stefan Hiendleder, Department of Molecular Animal Breeding and Biotechnology, Ludwig-Maximilian University Munich, Hackerstrasse 27, D-85764 Oberschleissheim, Germany. FAX: 49 89 315 2799; s.hiendleder{at}gen.vetmed.uni-muenchen.de
3 Permanent address: Department of Animal Biology, University of Sassari, 07100 Sassari, Italy
Received: 15 September 2003.
First decision: 20 October 2003.
Accepted: 12 December 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. F. Mastromonaco, L. A. Favetta, L. C. Smith, F. Filion, and W. A. King The Influence of Nuclear Content on Developmental Competence of Gaur x Cattle Hybrid In Vitro Fertilized and Somatic Cell Nuclear Transfer Embryos Biol Reprod, March 1, 2007; 76(3): 514 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X.-y. Yang, H. Li, Q.-w. Ma, J.-b. Yan, J.-g. Zhao, H.-w. Li, H.-q. Shen, H.-f. Liu, Y. Huang, S.-Z. Huang, et al. Improved efficiency of bovine cloning by autologous somatic cell nuclear transfer. Reproduction, November 1, 2006; 132(5): 733 - 739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Constant, M. Guillomot, Y. Heyman, X. Vignon, P. Laigre, J.L. Servely, J.P. Renard, and P. Chavatte-Palmer Large Offspring or Large Placenta Syndrome? Morphometric Analysis of Late Gestation Bovine Placentomes from Somatic Nuclear Transfer Pregnancies Complicated by Hydrallantois Biol Reprod, July 1, 2006; 75(1): 122 - 130. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
