RNA Processing in C. elegans - Caenorhabditis Elegans: Molecular Genetics and Development - page 190

most common intron size in flies is 59 nt, and in humans most introns are several
kilobases long. It is clear that the C. elegans splicing machinery has evolved to
process small introns. Studies have shown that short C. elegans introns cannot be
processed by mammalian spliceosomes ( Ogg et al., 1990 ), in which a length of
approximately 80 nt is necessary for efficient splicing to occur ( Wieringa et al.,
1984 ). The unusually short length of the introns in C. elegans may be, in part,
explained by how they are recognized and processed.
4. Splice Site Recognition
The C. elegans 5 0 splice site is defined by the canonical metazoan consensus
sequence AG/GURAGU, indicating that it is recognized by the U1 snRNP, as occurs
in other eukaryotes ( Blumenthal and Steward, 1997 )( Fig. 1c ). The 3 0 splice site
consensus sequence, however, is much more extensive than those found in vertebrate
or yeast introns. In C. elegans, there is a very highly conserved sequence UUUUCAG/
R at the boundary between the intron and the next exon ( Csank et al., 1990; Sheth
et al., 2006; Spieth and Lawson, 2006 ). This 3 0 splice-site sequence is not efficiently
recognized by the mammalian spliceosome ( Kay et al.,1987 ). All C. elegans introns
also lack ANY branch-point consensus sequence (YURAY in mammals), even though
worms do encode a SF1/BBP protein containing the conserved domain that recog-
nizes this sequence ( Blumenthal and Thomas, 1988; Mazroui et al.,1999 ).
Additionally, the C. elegans U2 RNA has an antisense branch-point sequence iden-
tical to that found in mammalian U2 ( Thomas et al.,1990 ). The only other apparent
information content of C. elegans intron 3 0 splice sites is a peak of adenosines at -16 to
-18 nt from the 3 0 splice site, which presumably serves as the site of branching. Such a
variably positioned branch-point adenosine is not without precedent. It has been found
that when possible branch-point adenosines are removed from mammalian introns,
splicing proceeds through nearby alternative cryptic branch points ( Ruskin et al.,
1985 ). A study in plants has also shown that several different adenosines in the last
third of an intron can act as branch points during splicing ( Goodall and Filipowicz,
1989 ). This may also be the case in nematodes, whose introns bear many similarities
to those of plants. In fact, it has been observed in C. elegans that mutation of putative
branch-point adenosines does not affect 3 0 splice site choice ( Conrad et al., 1993b ).
Finally, as has been described in plants, C. elegans introns often lack the polypyr-
imidine tract found immediately upstream of the 3 0 splice-site ( Csank et al., 1990;
Goodall and Filipowicz, 1989; Spieth and Lawson, 2006 ). Generally, the short uridine
stretch characterized as part of the UUUUCAG/R 3 0 splice site serves as the only
polypyrimidine tract ( Blumenthal and Thomas, 1988 ).
These modified intron features are indicators of some important differences in the
C. elegans splicing mechanism. Both U2AF 65 and U2AF 35 , shown to recognize the
polypyrimidine tract and 3 0 splice site in most metazoans, have been identified in
C. elegans ( Zorio et al., 1997; Zorio and Blumenthal, 1999b ). An early examination
of the UUUUCAG/R 3 0 splice site showed that mutation of the uridine residues
immediately upstream of the 3 0 splice site reduced proper splice site recognition