606
chapter 26
Regulation of Gene Expression
Genomic Imprinting
Approximately two dozen autosomal genes in humans are
inherited in a silent state from one parent and in a fully
active state from the other parent. Such genes are said to
be “imprinted” by the parents during gamete formation
and the phenomenon is referred to as
genomic imprint-
ing,
which does not change the nucleotide sequence of
DNA. Rather, it is an epigenetic phenomenon in which
the DNA at a particular locus is marked according to
gender, and this determines whether or not the locus is
expressed during embryonic development. Although re-
garded as a normal epigenetic phenomenon, two clin-
ically different disorders are characterized by genomic
imprinting.
Prader-Willi
and
Angelman syndromes
are distinct dis-
orders associated with multiple abnormalities and mental
retardation. Both disorders are caused by mutations at the
proximal end of chromosome 15 that silence one or more
genes. In Prader-Willi syndrome, the maternally inherited
chromosome is silent; in Angelman syndrome, the pater-
nally inherited chromosome is silent. For both syndromes,
a small deletion is usually responsible for the genomic
imprinting that silences the relevant gene(s). Both syn-
dromes can also result from having both copies of chromo-
some 15 derive from only one parent, a condition called
uniparental disomy. Angelman syndrome
is caused by
mutation in a single gene whereas
Prader-Willi syndrome
is caused by mutations in more than one gene. Thus, An-
gelman syndrome can also be caused by mutations in the
responsible gene itself, which is not the case for Prader-
Willi syndrome. Diagnosis of both can be confirmed by
analysis of DNA methylation in the respective genes since
methylation is the mechanism used for imprinting.
Regulation of RNA Processing
Initiation of transcription ultimately leads to production
of a primary transcript, which in higher eukaryotes is pro-
cessed to form an mRNA. Alternative processing patterns
can yield different mRNAs. One example comes from
chicken skeletal muscle in which two forms of the muscle
protein myosin, LC1 and LC3, are produced. The myosin
gene has two different TATA sequences that yield two
different primary transcripts. These two transcripts are
processed differently to form mRNA molecules encoding
distinct forms of the protein (Figure 26-10).
Another example of regulation of RNA processing is
found in adenovirus, which has only a few promoters but
which makes numerous primary transcripts and a large
number of mRNAs. A fairly small amount of DNA is used
efficiently because proteins are translated in all reading
frames and different regions of the DNA are, by virtue
of distinct RNA splicing patterns, used to form different
proteins.
The amount of each mRNA is regulated with respect
to the time after infection. RNAs 1 and 2 are both formed
shortly after the primary transcript is made, although more
of mRNA-2 is made. Later in the viral life cycle, mRNA-1
is not made and mRNA-2 is abundant. If cycloheximide,
an inhibitor of protein synthesis, is added to the in-
fected cells before the shift in splicing pattern takes place,
the shift does not occur. This inhibition implies that a
newly synthesized protein is a positive effector of the
shift. Cycloheximide has a similar effect on other mRNA
species derived from other primary transcripts at late
times.
Adenovirus has a single promoter for all RNA made late
in the cycle of infection. The primary transcripts terminate
at five major polyadenylation sites. Each termination site
influences the splicing pattern by allowing particular in-
trons or intron termini to be present or not. The five sets
are not used with equal frequency, with the result that
most mRNAs encoding various genes are not the same.
This is the primary mechanism for determining the rel-
ative amounts of the different structural proteins synthe-
sized late in the adenovirus life cycle.
Another type of regulation of processing involves
choice of different sites of polyadenylation. One exam-
ple is the differential synthesis of the hormone
calcitonin
in different tissues; another is the synthesis of two forms of
the heavy chain of immunoglobulins (Chapter 35). In both
cases, the differential processing includes distinct patterns
of intron excision (i.e., splicing), but they are necessitated
by an earlier event in which differential poly(A) sites are
selected from the primary transcript. That is, when the
poly(A) site nearer the promoter is selected, a splice site
used in the larger primary transcript is not present, so a
different splice pattern results. Thus, slightly different pro-
teins are synthesized.
The calcitonin gene consists of five exons and uses two
alternative polyadenylation sites that respond to differ-
ent signals in different tissues. In the thyroid calcitonin
is produced by a signal that produces a pre-mRNA con-
sisting of exons 1^1. The introns are then spliced out
to give the mature mRNA. However, in neural tissue a
different polyadenylation site is activated that produces
a pre-mRNA consisting of exons 1-5 and the interven-
ing introns. When this larger pre-mRNA is processed,
the introns are spliced out but so is exon 4, produc-
ing a mature mRNA consisting of exons 1-3 and 5.
This is translated into a growth factor called
calcitonin-
related gene peptide (CRGP).
Mutations in the calci-
tonin gene can result in both adrenal and thyroid tumors,
