Alternatively Spliced Genes
137
of the 3
0
ss. A typical U12-class intron lacks
a polypyrimidine tract between the branch
site and the 3
0
ss.
In the yeast,
S. cerevisiae
,on
lyasma
l
l
fraction of genes contain introns, and these
are usually short introns (approximately
240 introns, averaging 270 nucleotides
in
length).
Splicing
signals
in
yeast
introns are highly conserved and may
contain sufFcient information for deFning
splice
junctions,
especially
considering
the
number
of
genes
devoted
to
or
associated with pre-mRNA splicing in the
yeast genome.
In mammals, pre-mRNA transcripts are
usually much longer and contain multiple
introns of variable sizes. In humans, the
average size of exons is 150 nucleotides,
and that of introns is approximately 3500
nucleotides. Mammalian introns can be
a
sl
a
rg
ea
s500kbp
.Th
eb
a
s
i
csp
l
i
c
ing
signals in mammalian pre-mRNAs are de-
generate, especially in the case of U2-type
of introns. The branch sites for U2-type
introns are highly divergent. At both the
5
0
ss and 3
0
ss, only two nucleotides (/GT
at the 5
0
ss and AG/ at 3
0
ss) are highly
conserved. As a
result,
the
nucleotide
sequences surrounding the splice junc-
tions, the 5
0
ss and 3
0
ss, usually contain
only a limited amount of information. This
is not sufFcient for conferring the speci-
Fcity required to achieve accurate splice
site selection. In mammals, the recog-
nition of not only exon–intron junction
sequences but also the regulatory elements
in intronic and exonic regions is important
for deFning splice junctions and maintain-
ing splicing Fdelity. In addition, multiple
networks of interactions among the ma-
chineries for transcription, cap formation,
splicing, and polyadenylation may also in-
fluence splice site selection. This high
degree of degeneracy in the splicing sig-
nals in mammalian pre-mRNA transcripts
provides the flexibility for alternative selec-
tion and pairing of different splice sites,
a fundamental mechanism for regulating
alternative splicing.
1.3
Spliceosomal UsnRNP Biogenesis
As essential subunits of the splicing ma-
chinery, spliceosomal UsnRNPs contain
not only uridine-rich snRNAs but also
a number of polypeptides. U1, U2, U4,
and U5snRNAs are transcribed by RNA
polymerase II as precursors containing
additional 3
0
nucleotides. After acquir-
ing a monomethylated guanosine (m7G)
cap
structure,
these
pre-UsnRNAs
are
exported to the cytoplasm in a pathway de-
pendent on the m7G cap, the cap-binding
complex (CBC), RanGTP, and phospho-
rylated adaptor for RNA export. In the
cytoplasm, pre-UsnRNAs interact with Sm
proteins including B/B
0
, D3, D2, D1, E,
±, and G (Table 1A) to form the snRNP
core structure.
The Sm protein-binding sites in U1, U2,
U4, and U5snRNAs are highly conserved,
containing two stem-loop structures flank-
ing PuAU4-6GPu sequence (±ig. 3a). A
number of proteins interacting with Sm
proteins have been identiFed, including
the
protein
product
of
SMN
(survival
of motor neuron) gene, SMN-interacting
protein/Gemin2, Gemin3, and Gemin4.
Genetic defects in the SMN gene cause
spinal muscular atrophy (SMA), possi-
bly by interfering with UsnRNP Sm core
assembly and therefore deFciency in Us-
nRNP biogenesis.
±ollowing UsnRNP Sm core assembly,
the m7G cap of their snRNA is con-
verted to the 2,2,7-tri-methylated guano-
sine (m3G), and the 3
0
extra nucleotides
of pre-UsnRNAs are removed. These core
UsnRNPs are imported into the nucleus to
previous page 137 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online next page 139 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online Home Toggle text on/off