Anthology of Human Repetitive DNA
285
Interspersed repeats can affect alter-
native splicing and cause inclusion of
intronic repeats into cellular mRNAs. This
can alter or destroy the mRNA or pro-
tein functionality and be manifested as a
disease phenotype. Notably, Alu repeats
contain cryptic splice sites in the an-
tisense orientation that, after acquiring
speciFc mutations, can give rise to new
Alu-derived exons (±ig. 13a). The major-
ity of such Alu cassettes are alternatively
spliced and produce premature termina-
tion (stop) codons. The affected mRNAs
are probably destroyed by cellular mRNA
quality control machinery, particularly by
the nonsense-mediated RNA decay (NMD)
pathway, which degrades mRNAs with pre-
mature stop codons at the time of protein
translation. However, if the new exon be-
comes constitutively spliced in, the NMD
control prevents efFcient translation of the
affected mRNA, and thus causes a dele-
terious phenotype at the protein level. So
far, there are two known cases of disease-
related Alu splicing into cellular mRNAs.
One, reported in a case of ornithine-delta
aminotransferase deFciency, was caused
by alternative Alu splicing, creating a pre-
mature stop codon. In a second instance,
a splice-mediated insertion of an Alu se-
quence in the precursor of alpha 3 type
IV collagen mRNA caused autosomal re-
cessive Alport syndrome in the affected
patient. The affected proteins were not ex-
pressed in either case, despite the fact that
the insertion preserved the open reading
frame in the second case.
Insertion into genomic DNA is an
integral
part
of
the
ampliFcation
of
interspersed repeats. An integration into
coding sequences is likely to cause an
insertional inactivation of the targeted
gene, which subsequently manifests itself
as a genetic abnormality. To date, there are
41 published repeat insertions associated
with human diseases (Table 9). All the
insertions are either L1 elements (14
cases), L1-dependent Alu (23 cases), or
SVA elements (4 cases). The Alu and L1
insertions come nearly exclusively from
the two active subfamilies AluY and
L1-Ta respectively (Table 9). No recent
insertions of endogenous retroviruses or
DNA transposons have been reported.
±rom Table 9 it is clear that some genes
are overrepresented among insertion tar-
gets, particularly coagulation factors VIII
and XI and the dystrophin gene. ±urther-
more, 42% (19 out of 41) of all documented
insertions are on chromosome X, which
represents only 5% of the human genome.
The bias is particularly strong for L1
insertions 11/14 (79%), and less so for L1-
dependent nonautonomous Alu and SVA
elements, which represent 26% (6/23) and
25% (1/4) of all documented insertions re-
spectively. Chromosome X is present in
only one functional copy in both males
and females, due to female imprinting
(inactivation) of one X copy. Therefore,
theX-linkedgenesaremorevu
lnerab
leto
mutations. The excess of detected inser-
tions on the X chromosome may reflect
the detection bias for dominant pheno-
types rather than any preferential insertion
patterns on this chromosome.
The majority of insertions listed in
Table 9 occurred in exons and the remain-
ing intronic insertions appear to have dis-
rupted splicing or other essential signals
located within introns. The deleterious ef-
fect of intraexonic insertions is mainly due
to the incorporation of premature stop
codons into the reading frame and mRNA
degradation by NMD (±ig. 13b). Other de-
fects can be caused by intron inclusion
or exon skipping (±ig. 13a, c), leading to
either truncation of the protein or induc-
tion of premature stop codons and NMD
sensitivity.
previous page 285 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online next page 287 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online Home Toggle text on/off