RESULTS

Development of Embryos

Maturation rate of oocytes for nuclear transfer was 55%, and 203 of 215 matured and enucleated oocytes were successfully fused (94.4%) with nuclear donor cells. By Day 3, 139 KCCs (68.5%) had further developed; this resulted in 63 blastocysts (29.3%) by Day 7. In the IVF group, 120 oocytes subjected to maturation and fertilization procedures yielded 78 Day 3 embryos (65%), of which 32 (26.7%) developed into blastocysts by Day 7.

Phenotypic Characteristics of Fetuses and Fetal Membranes

For a detailed phenotypic analysis of fetuses derived from SCNT, IVF, and AI, fetuses and placentae were recovered 80 days after the initiation of embryonic development. Only SCNT fetuses with a body weight >109.9 g were selected for further analyses. This yielded a group of overgrown singleton SCNT (n = 6) fetuses. The investigated IVF (n = 5) and AI (n = 5) groups of fetuses were unselected for body weight but had been selected for females and singletons. SCNT and IVF fetuses revealed a distinct overgrowth phenotype (Fig. 1), which was further characterized by quantitative analysis of morphological parameters (see Fig. 3A).

In comparison with the AI controls, SCNT and IVF fetuses showed not only a significantly elevated body weight (61%, P < 0.001 and 28%, P < 0.003), liver weight (100%, P < 0.001 and 36%, P < 0.014) and thorax circumference (20%, P < 0.001 and 11%, P < 0.001) but also a higher relative liver weight (25%, P < 0.01 and 7%, not significant). The crown-rump length did not differ among the three groups of fetuses, but a reduced crown-rump length:thorax circumference ratio (1.175 ± 0.017 in SCNT and 1.292 ± 0.018 in IVF vs. 1.390 ± 0.018 in AI, P < 0.001 and P < 0.002) was the external hallmark of the disproportionate overgrowth phenotype observed in both groups of fetuses derived by in vitro techniques.

The absolute weight of fetal membranes was significantly higher in SCNT fetuses than in IVF or AI fetuses (49%, P < 0.012 and 36%, P < 0.035) but the ratio of fetus weight:placenta weight was not significantly different among the groups. There was no difference in the number of placentomes, but placentomes were significantly longer in SCNT placentae than in IVF or AI placentae (21%, P < 0.034 and 32%, P < 0.007). Placentomes in ScNt fetuses were also wider than in IVF or AI fetuses (36%, P < 0.011 and 21%, P < 0.077) (Figs. 2 and 3B).

Global 5-Methylcytosine Content in Liver and Cotyledon DNA and Relationship with Phenotypic Characteristics

In order to analyze the relationship between fetal overgrowth and DNA methylation, we determined the global cytosine methylation level of DNA samples from fetal liver and cotyledon tissue by capillary electrophoresis (Fig. 4A). The SCNT fetuses revealed significant hypermethylation of liver DNA in comparison with AI controls (3.457% ± 0.079% 5mC vs. 3.166% ± 0.087% 5mC, P < 0.03). The cytosine methylation level of liver DNA from IVF fetuses was intermediate (3.342% ± 0.087% 5mC) to but not significantly different (P > 0.18) from the other groups. The average 5mC content of placental cotyledon DNA was 46% lower than in liver DNA (P < 0.0001) but did not differ among fetus groups (Fig. 4B).

Regressions of fetal body and liver weight as well as thorax circumference on the cytosine methylation level of liver DNA were positive (P < 0.073-0.079) and showed very similar slopes (Fig. 5, A-C). Furthermore, a significant negative regression (P < 0.021) of the crown-rump length: thorax ratio on liver 5mC was observed (Fig. 5D). This demonstrates a statistically significant association between the level of DNA methylation and an important phenotypic characteristic of the observed fetal overgrowth.
Fig1Tissue-Specific Elevated Genomic-1
FIG. 1. Details of Day 80 bovine overgrowth phenotype. A) Control fetus with a body weight of 81 g. B) SCNT fetus with a body weight of 142 g. Note the heavier muscled neck and shoulder area in the SCNT fetus. Both fetuses are shown at identical magnification, bar = 5 cm.

Fig2Tissue-Specific Elevated Genomic-2
FIG. 2. Examples of uteri obtained from Simmental recipient heifers that contained Day 80 fetuses generated by artificial insemination (AI), in vitro fertilization (IVF), or somatic cell nuclear transfer (NT). Fetal membranes have been removed. The uteri are shown at identical magnification, bar = 10 cm.

Fig3Tissue-Specific Elevated Genomic-3
FIG. 3. Phenotypic characteristics of viable Day 80 fetuses and fetal membranes. Fetuses were generated by artificial insemination (AI, n = 5), in vitro fertilization (IVF, n = 5), or somatic cell nuclear transfer (NT, n = 6). A) Fetal weights and dimensions. Crown-rump:thorax = crown-rump length: thorax circumference ratio. B) Characteristics of the placenta. Fetus:placenta = fetus:placenta weight ratio. Standard error bars are indicated. Means with different superscripts differ significantly at P < 0.05.

Fig4Tissue-Specific Elevated Genomic-4
FIG. 4. Quantification of global cytosine methylation levels (% 5mC) in DNA samples from viable Day 80 fetuses generated by in vitro (NT, n = 6; IVF, n = 5) or in vivo (AI, n = 5) procedures. A) Examples of electro-pherograms obtained for DNA samples from individual fetuses of the different groups. The 5mC peaks are indicated by arrowheads. B) Mean cytosine methylation levels in fetal and placental tissue. Standard error bars are indicated. Means with different superscripts differ significantly at P < 0.05.

Fig5Tissue-Specific Elevated Genomic-5
FIG. 5. A-D) Regressions of phenotypic characteristics of all investigated Day 80 fetuses (n = 16) on the cytosine methylation level of liver DNA samples. Significance levels are indicated.