498
showed similar downregulation patterns (Tang et al. 2007 ). This likely reflects the
fact that mature miRNAs do not carry any specific regulatory sequences, while their
elimination should not interfere with an upcoming accumulation of zygotic miR-
NAs. Maternal miRNAs are rapidly eliminated after fertilization also in the zebraf-
ish, as shown by small RNA cloning (Chen et al. 2005 ). However, some data suggest
that miRNA degradation after fertilization is the second wave of maternal miRNA
elimination. Northern blots of several maternal miRNAs in Xenopus show that
miRNA degradation is initiated already before fertilization as indicated by lower
miRNA levels in ovulated eggs (Watanabe et al. 2005 ). Furthermore, deep sequenc-
ing of fully grown and meiotically mature porcine oocytes also suggested reduction
of miRNA levels during meiosis (Yang et al. 2012 ).
Although the molecular mechanism of maternal miRNA degradation in verte-
brates has not been identified yet, it presumably involves RNA-specific ribonucleo-
tidyl transferases (Martin and Keller 2007 ). First, uridylation or adenylation of
RNAs by these enzymes is a well-established regulation of miRNA stability
(reviewed in Ha and Kim 2014 ). Second, a noncanonical poly(A) polymerase Wispy
was identified as the factor responsible for maternal miRNA adenylation and degra-
dation in Drosophila (Lee et al. 2014 ).
10.2.1.2 Short Interfering RNAs (siRNAs) from the RNAi Pathway
RNAi is sequence-specific mRNA degradation mediated by small RNAs generated
by Dicer from long double-stranded RNA (dsRNA). This pathway was originally
identified in C. elegans (Fire et al. 1998 ) and later in vertebrates (Svoboda et al.
2000 ; Wargelius et al. 1999 ; Wianny and Zernicka-Goetz 2000 ; Nakano et al. 2000 ).
Remarkably, oocytes were the first vertebrate cell type where RNAi was identified.
The vertebrate RNAi is somewhat enigmatic; the pathway employs the same protein
factors as the miRNA pathway, but endo-siRNAs (short interfering RNAs produced
by Dicer from dsRNA) are typically found in negligible numbers in vertebrate
somatic cells (e.g., Nejepinska et al. 2012b; Wei et al. 2012 ; Faunes et al. 2011 ).
This is in agreement with functional analyses of human Dicer, which showed that it
is efficiently processing miRNA precursors but not long perfect duplexes because of
an autoinhibitory function of its N-terminal helicase domain (Ma et al. 2008a;
Chakravarthy et al. 2010 ), which is conserved across vertebrates.
The RNAi pathway is apparently intact in vertebrate oocytes since mRNA
knockdown was efficiently induced with long dsRNA in zebrafish (Wargelius
et al. 1999 ) and Xenopus oocyte (Nakano et al. 2000 ). However, strong nonspe-
cific effects of dsRNA were also observed in the zebrafish model (Zhao et al.
2001a). Furthermore, RNAi efficiency in Xenopus oocytes and early embryos
might have reduced functionality due to limiting amounts of AGO2 protein (Lund
et al. 2011 ). In mammals, RNAi effects can be masked by the interferon (IFN)
pathway, which is a sequence-independent innate immunity response to dsRNA
(reviewed in Gantier and Williams 2007 ). However, mouse oocytes lack the IFN
response to dsRNA (Stein et al. 2005 ), and microinjected long dsRNA induces
P. Svoboda et al.