51 AGGREGATION REDUCES THE VARIATION OF GENE EXPRESSION LEVELS IN BOVINE CLONES

Abstract

Mammalian somatic cell clones frequently exhibit abnormal gene expression that presumably results from errors in reprogramming of the transplanted genome. In the mouse, aggregation of 4-cell stage clones with each other improves reprogramming with respect to Oct-4 expression in blastocysts and an increase in term development (Boiani et al. 2003 EMBO J. 22, 5304-5312). To determine if clone-clone aggregation has a similar beneficial effect in the bovine, we aggregated 8-16 cell bovine clones with each other and profiled gene expression levels in bovine clones and clone-clone aggregates at the blastocyst stage. Clone embryos were produced from fibroblasts and cultured in vitro in SOF supplemented with fetal bovine serum at 39°C in an atmosphere of 5% CO₂, 5% O₂, and 90% N₂. For aggregation of embryos, we first removed the zonae pepellucidae by treatment with 0.5% pronase at the 8-16 cell stage and then placed two zona-free embryos per well into deep microwells produced on the bottom of a culture dish by pressing a heated darning needle onto the surface. Seven to 10 microwells in close proximity were covered by a culture 50-µL drop of culture medium, and embryos were cultured until Day 7. Real-time RT-PCR analysis for Oct-4, DNA methyltransferase 1 (Dnmt1), Dnmt3, glucose transporter 1 (Glut1), Glut3, and Poly(A) polymerase (PolyA) was performed on reusable Dynabead Oligo (dT)₂₅-cDNA libraries synthesized from individual blastocysts at Day 7. In vitro-fertilized embryos were used as controls. To compare the variation of gene expression in each embryo within the group, the coefficient of variation (COV; standard deviation/mean) was calculated. Although spatial distribution of Oct-4 transcript is normal in bovine blastocyst stage clones (Kurosaka et al. 2004 Reprod. Fertil. Dev. 16, 147), we detected disturbances in the level of Oct-4 expression in clones: 44.4% (8 of 18) of clones expressed Oct-4 within a range of 0.5- and 1.5-fold of the average level of expression in IVF embryos, compared to 81.8% (9 of 11) of IVF embryos. Only 22.2% (4 of 18) of clones expressed all genes examined within a range of 0.5- and 2.0-fold of the average level of IVF embryos, versus 45.5% (5 of 11) of IVF embryos. Clone-clone aggregation did not increase the proportion of clones with normal expression levels but did reduce the coefficient of variation of gene expression levels between individual clones for the genes Oct-4, Dnmt1, Dnmt3a and PolyA, but not for Glut1 and Glut3. Interestingly, bovine clone-clone aggregates (n = 25) had less variation between individual embryos compared to IVF aggregates (n = 11) for all genes except Glut1 and Glut3, although variation of single clones was larger than that of single IVF embryos. Analysis of Oct-4 and ²-Actin transcripts in mouse clone blastocysts indicated a similar decrease in gene expression variation subsequent to aggregation of mouse clones. These results demonstrate that bovine pre-implantation stage clones exhibit a high degree of variation in gene expression levels and suggest that aggregation of clones is beneficial in reducing the variation in expression of some genes.