have a viable embryonic disc (Fischer-Brown et al., 2005). Such an effect could help to explain

increases in fetal loss associated with in vitro produced embryos.

 While it is clear that the production of embryos in vitro can have long term affects on fetal,

placental and neonatal development, the precise reasons for these alterations are not fully

understood. One hypothesis is that manipulation of the early embryo during in vitro embryo

production alters the expression of imprinted genes. Consistent with this idea, IGF-2 mRNA

abundance was altered in day 70 bovine fetuses derived from in vitro produced embryos

compared to fetuses that developed from superovulated embryos (Blondin et al., 2000).

Moreover, expression of IGF-2 and IGF-2 receptor were also altered in bovine embryos

produced in vitro (Bertolini et al., 2002a). Thus, it is possible that bovine embryo production in

vitro can affect the methylation patterns that regulate monoallelic expression of imprinted genes.

While this has not been reported for in vitro produced bovine embryos, such an effect has been

reported in mice (Khosla et al., 2001) and sheep (Young et al., 2001).

 The expression of non-imprinted genes can also be altered by in vitro embryo production

(Bertolini, 2002a; Crosier et al., 2002; Miles et al., 2004; Miles et al., 2005). Following

fertilization, the paternal DNA undergoes an active, rapid process of demethylation while the

maternal DNA undergoes a passive demethylation. During this period, epigenetic marks on non-

imprinted genes are erased (Morgan et al., 2005). Embryonic methylation patterns are re-

established during development to the blastocyst stage in cattle (Reik et al., 2001; Li, 2002) by

the actions of two enzymes, DNA methyltransferase 3a and 3b (Reik et al., 2001; Reik et al.,

2003). It is also possible that embryonic manipulation as part of in vitro embryo production

could affect the re-methylation of non-imprinted genes and thereby alter post-transfer survival