Maternal-fetal transplacental leakage of mitochondrial DNA in bovine nuclear transfer pregnancies: potential implications for offspring and recipients

The synepitheliochorial placenta of ruminants is constructed of multiple tissue layers that separate maternal and fetal blood. In nuclear transfer cloned ruminants, however, placental anomalies such as abnormal vascular development and hemorrhagic cotyledons have been reported. We have investigated the possible exchange of genetic material between somatic cell nuclear transfer cloned (SCNT) bovine fetuses and recipients at day 80 of gestation using mitochondrial DNA (mtDNA) as a marker. Twenty-three recovered SCNT-fetuses and their recipients were screened for divergent and thus informative mtDNA combinations. Twenty-one fetuses generated by in vitro fertilization (IVF) or multiple ovulation embryo transfer (MOET) and the corresponding recipients served as controls. A search for recipient mtDNA haplotype in DNA extracts from fetal blood by PCR-RFLP analysis revealed three cases of chimerism (two SCNT, one IVF) among a total of 19 informative fetus-recipient pairs (eight SCNT, seven IVF, four MOET). Placental anomalies have also been observed in some IVF fetuses and the present data therefore suggests transplacental leakage of cell components or cells from the recipient into some fetuses generated by in vitro techniques. Further studies are necessary to determine (i) the nature of leaked material, (ii) whether there is bi-directional leakage, and (iii) whether leaked material is present in recipients and calves after parturition, i.e. whether leakage takes place in vivo. If recipients were chimeric for DNA or cells derived from genetically modified SCNT (or IVF) embryos, their subsequent utilization might be affected.     

Induction of a senescent-like phenotype does not confer the ability of bovine immortal cells to support the development of nuclear transfer embryos

Previously, we reported that cloned embryos derived from an immortalized bovine mammary epithelial cell line (MECL) failed to develop beyond 12- to 16-cell stage. To analyze whether induction of a senescent-like phenotype in MECL can improve their ability to support the development after transfer into enucleated oocytes, we treated MECL with DNA methylation inhibitor 5-aza-2-deoxycytidine (Aza-C), histone deacetylase inhibitors trichostatin A (TSA), sodium butyrate (NaBu), or 5-bromodeoxyuridine and used those cells for nuclear transfer. Primary bovine fetal fibroblasts (BFF) were used as control. All agents were capable to induce features of senescence including reduced cell proliferation, enlarged cell size with a considerable proportion of cells stained positive for acidic senescence-associated beta-galactosidase and G1/S cell cycle boundary arrest in MECL. Aza-C treatment induced genome demethylation. Acetylation of H3 and H4 was increased after TSA treatment in both MECL and BFF, whereas no obvious changes in global H3 or H4 acetylation were detected after NaBu treatment. Nuclear transfer experiments following diverse treatments demonstrated that the induced senescent-like phenotype of MECL did not confer their ability to support embryonic development, although 7.3% of reconstructed embryos derived from NaBu-treated cells developed to morula stage. Intriguingly, a much higher proportion of cloned embryos developed to blastocysts when using NaBu-treated BFF, compared with using untreated BFF (59% versus 26%). Our results suggest that the developmental failure of donor nuclei from bovine immortal cells could not be reversed by induction of senescent-like phenotype. The beneficial effect of NaBu on the developmental potential of cloned embryos reconstructed from BFF merits further studies.     

Nuclear transfer in cattle with non-transfected and transfected fetal or cloned transgenic fetal and postnatal fibroblasts

The efficiency of nuclear transfer (NT) using two primary cultures of fetal fibroblasts (FF1 and FF2) was compared vs. the same cultures transfected with an expression vector in which the bovine prochymosin coding sequence is placed under the control of the bovine alpha(S1)-casein promoter (TFF1 and TFF2). In addition, fibroblasts of a cloned transgenic fetus (TRFF1) derived from TFF1 and ear skin fibroblasts of a 1-month-old cloned transgenic calf (TRCF1) derived from TRFF1 were used as nuclear donors. Embryos reconstructed from FF1 (44%) and FF2 (52%) developed to the blastocyst stage at a significantly (P < 0.05) higher rate than those derived from TFF1 (24%) and TFF2 (27%). The proportions of cleaved embryos and blastocysts were significantly (P < 0.05) higher with TRFF1 than with TRCF1 used as nuclear donors (75 vs. 66% and 33 vs. 16%, respectively). Transfer of NT embryos derived from FF2 and TFF2 to recipients resulted in similar pregnancy rates on day 30 (52 and 48%, respectively). However, with TFF2 embryos, the majority of pregnancies (8/11; 73%) was lost in the first and second trimesters of gestation, whereas 4/11 (36%) pregnancies with FF2 embryos were lost during the full period of in vivo development. Of 11 FF2 and 6 TFF2 born calves (25 and 13% of transferred embryos, respectively), 6 and 3 survived including one oversized FF2 calf. After transfer of TRFF1 and TRCF1 NT embryos to recipients, initial pregnancy rate was as a tendency higher in the TRFF1 (49%) than in the TRCF1 group (30%). The majority (14/17) of TRFF1 pregnancies and all TRCF1 pregnancies were lost in the first and second trimester. A high proportion of TRFF1 calves (5/8) showed increased body weights, and only two calves which were also large survived. These findings demonstrate that (i) extended culture associated with transfection and selection procedures may induce changes of donor cells which markedly decrease the efficiency of nuclear transfer and (ii) these changes are not reversed by recloning.     

Behavior of M-phase synchronized blastomeres after nuclear transfer in cattle

M-phase synchronized bovine blastomeres were used to study the effect of nuclear-cytoplasmic synchronization on the developmental potential after nuclear transfer (NT). The capacity of nocodazole and benomyl to reversibly synchronize blastomeres from embryos in M-phase was evaluated. Nocodazole reversibly arrested bovine embryos at the studied stages and induced high rates of M-phases in morulae and compact morulae. In contrast, benomyl was less efficient than nocodazole to synchronize in M-phase. After transfer of an M-phase blastomere, premature chromatin condensation was the prevalent finding 1 hr post-fusion (hpf). Condensed chromosomes non-arranged in the equatorial plate (1-3 hpf) that acquired an organized structure over time (3-7 hpf) were subsequently observed. Anaphase-telophase structures were predominantly recorded at 4-9 hpf. About 50% of the embryos activated at both 3-4 and 6-7 hpf extruded a polar body-like structure 5 hr after activation, but this was not observed in embryos activated immediately after fusion. A significantly lower activation rate was observed for oocytes activated 3-4 hpf compared to those activated 6-7 hpf. However, the ability to undergo first cleavage was significantly lower in the latter group. Reconstructed embryos activated immediately after fusion showed no difference in the rate of activation compared to those activated 6-7 hpf, although the cleavage rate was higher. DNA synthesis was observed at a significantly higher rate in embryos activated both immediately and 3-4 hpf that did not extrude a PB-like structure than in those activated 3-4 hpf that extruded a polar body-like structure. Under the conditions tested M-phase donor cells cannot be properly remodeled after NT in cattle to trigger normal embryonic development. Our observations of chromatin structures together with DNA synthesis suggest that the failure in the development may be due to improper chromatin remodeling of mitotic nuclei after NT, which may result in chromosomal abnormalities incompatible with normal embryo development.     

Nuclear-cytoplasmic interactions affect in utero developmental capacity, phenotype, and cellular metabolism of bovine nuclear transfer fetuses

We generated a clone of bovine somatic cell nuclear transfer embryos using oocyte pools from defined maternal sources to study nuclear-cytoplasmic interactions. Nucleocytoplasmic hybrids were reconstructed with Bos taurus (Brown Swiss) granulosa cells and oocytes that contained B. taurus A (Simmental), B. taurus B (Simmental), or Bos indicus (Dwarf Zebu) cytoplasm. Another set of embryos was reconstructed with randomly selected Brown Swiss (B. taurus R) oocytes. Embryo transfer resulted in nine (12.5%), nine (13.8%), three (50%), and 11 (16.7%) Day 80 fetuses, of which eight (11.1%), three (4.6%), three (50%), and 10 (15.2%) were viable, respectively. The proportion of viable fetuses was affected by cytoplasm (likelihood ratio test, P < 0.02) and was higher for embryos with B. indicus cytoplasm than for the B. taurus A (P < 0.05) and B (P < 0.01) groups. Furthermore, the proportion of surviving Day 80 fetuses was reduced for B. taurus B as compared with B. taurus A and B. taurus R cytoplasm (P < 0.05 and P < 0.02). Body weight of nucleocytoplasmic hybrid fetuses was not significantly different from Brown Swiss control fetuses produced by artificial insemination (AI), but fetuses reconstructed with random cytoplasts of the same breed as the nuclear donor exhibited overgrowth (P < 0.01) and a higher coefficient of variation in weight. Furthermore, body weight, crown rump length, thorax circumference (P < 0.05), and femur length (P < 0.01) of fetuses with B. taurus A cytoplasm differed from fetuses with B. taurus R cytoplasms. Fetal skin, heart, and liver cells with B. indicus cytoplasm showed a greater increase in number per time period (P < 0.001) and oxygen consumption rate per cell (skin and liver, P < 0.001; heart, P < 0.08) in comparison with their counterparts with B. taurus A cytoplasm. These data point to complex oocyte cytoplasm-dependent epigenetic modifications and/or nuclear DNA-mitochondrial DNA interactions with relevance to nuclear transfer and other reproductive technologies such as ooplasmic transfer in human assisted reproduction.     

Effects of serum starvation and re-cloning on the efficiency of nuclear transfer using bovine fetal fibroblasts.

The developmental potential of bovine fetal fibroblasts was evaluated using nuclear transfer. Fibroblasts from a 37-day-old fetus were fused to enucleated oocytes before activation. Nuclei of starved (cultured for 8 days in medium containing 0.5% serum) fibroblasts supported the development of reconstructed embryos to the blastocyst stage significantly better than those of non-starved fibroblasts (39% versus 20%; P < 0.05). When nuclear transfer morulae derived from starved or non-starved fibroblasts were used for re-cloning, the proportion of blastocysts (52 and 55%, respectively) obtained with these embryonic nuclei was significantly higher than it was with fibroblast nuclei used in the first round of nuclear transfer (P < 0.05 and P < 0.001, respectively). After transfer of blastocysts derived from non-starved and starved fibroblasts, respectively, 33% (1/3) and 78% (7/9) of recipients were pregnant on day 30 as assessed by ultrasonography. On day 90, the corresponding pregnancy rates were 33% (1/3) and 63% (5/8). Two live male twin calves, derived from non-starved fibroblasts, were delivered by Caesarean section at day 281 of gestation. This study demonstrates a positive effect of serum starvation on the efficiency of nuclear transfer using bovine fetal fibroblasts. The efficiency of nuclear transfer could be further increased by recloning.     

Remodeling of the Nuclear Envelope and Lamina during Bovine Preimplantation Development and Its Functional Implications

The present study demonstrates a major remodeling of the nuclear envelope and its underlying lamina during bovine preimplantation development. Up to the onset of major embryonic genome activation (MGA) at the 8-cell stage nuclei showed a non-uniform distribution of nuclear pore complexes (NPCs). NPCs were exclusively present at sites where DNA contacted the nuclear lamina. Extended regions of the lamina, which were not contacted by DNA, lacked NPCs. In post-MGA nuclei the whole lamina was contacted rather uniformly by DNA. Accordingly, NPCs became uniformly distributed throughout the entire nuclear envelope. These findings shed new light on the conditions which control the integration of NPCs into the nuclear envelope. The switch from maternal to embryonic production of mRNAs was accompanied by multiple invaginations covered with NPCs, which may serve the increased demands of mRNA export and protein import. Other invaginations, as well as interior nuclear segments and vesicles without contact to the nuclear envelope, were exclusively positive for lamin B. Since the abundance of these invaginations and vesicles increased in concert with a massive nuclear volume reduction, we suggest that they reflect a mechanism for fitting the nuclear envelope and its lamina to a shrinking nuclear size during bovine preimplantation development. In addition, a deposit of extranuclear clusters of NUP153 (a marker for NPCs) without associated lamin B was frequently observed from the zygote stage up to MGA. Corresponding RNA-Seq data revealed deposits of spliced, maternally provided NUP153 mRNA and little unspliced, newly synthesized RNA prior to MGA, which increased strongly at the initiation of embryonic expression of NUP153 at MGA.     

Generation of transgenic cattle by lentiviral gene transfer into oocytes

The potential benefits of transgenic cattle range from the production of large quantities of pharmaceutically relevant proteins to agricultural improvement. However, the production of transgenic cattle is presently time-consuming and expensive because of the inefficiency of the classical DNA microinjection technique. Here, we report the use of lentiviruses for the efficient generation of transgenic cattle. Initial attempts to produce transgenic cattle by lentiviral infection of preimplantation embryos were not successful. In contrast, infection of bovine oocytes with lentiviral vectors carrying an enhanced green fluorescent protein (eGFP) expression cassette followed by in vitro fertilization resulted in the birth of transgenic calves. Furthermore, all of the calves generated by infection of oocytes were transgenic, and 100% of these animals expressed eGFP as detected by in vivo imaging and Western blotting. In addition, a transgenic calf was produced by infection of fetal fibroblasts followed by nuclear transfer into enucleated oocytes. Taken together, after adjusting lentiviral transgenesis to cattle, unprecedented high transgenesis and expression rates were achieved.