45 121 Categories of Activators

wea25324_ch12_314-354.indd Page 315
11/25/10
8:08 PM user-f469
/Volume/204/MHDQ268/wea25324_disk1of1/0073525324/wea25324_pagefile
12.1 Categories of Activators
In addition, eukaryotic DNA is complexed with protein
in a structure called chromatin. Some chromatin, called
heterochromatin, is highly condensed and inaccessible to
RNA polymerases, so it cannot be transcribed. Other
chromatin (euchromatin) still contains protein, but it is relatively extended. Much of this euchromatin, even though it
is relatively open, contains genes that are not transcribed
in a given cell because the appropriate activators are not
available to turn them on. Instead, other proteins may hide
the promoters from RNA polymerase and general transcription factors to ensure that they remain turned off. In
this chapter, we will examine the activators that control
eukaryotic genes. Then, in Chapter 13, we will look at the
crucial relationship among activators, chromatin structure,
and gene activity.
12.1 Categories of Activators
Activators can either stimulate or inhibit transcription by
RNA polymerase II, and they have structures composed of
at least two functional domains: a DNA-binding domain
and a transcription-activating domain. Many also have a
dimerization domain that allows the activators to bind to
each other, forming homodimers (two identical monomers
bound together), heterodimers (two different monomers
bound together), or even higher multimers such as tetramers. Some even have binding sites for effector molecules
like steroid hormones. Let us consider some examples of
these three kinds of structural–functional domains, bearing
in mind an important principle we discussed in Chapters 6
and 9: A protein does not have just one shape. Rather, it is
a dynamic molecule that assumes many possible conformations. Some of these may be especially advantageous for
binding to other molecules, such as a specific DNA sequence, and these conformations would be stabilized by
binding to such DNA sequences. Thus, when we refer to the
shape of a DNA-binding protein, or a domain within such a
protein, we mean one of many possible shapes, which happens to fit particularly well with the DNA in question.
DNA-Binding Domains
A protein domain is an independently folded region of a
protein. Each DNA-binding domain has a DNA-binding
motif, which is the part of the domain that has a characteristic shape specialized for specific DNA binding. Most
DNA-binding motifs fall into the following classes:
1. Zinc-containing modules. At least three kinds of zinccontaining modules act as DNA-binding motifs. These
315
all use one or more zinc ions to create the proper
shape so an a-helix within the motif can fit into the
DNA major groove and make specific contacts there.
These zinc-containing modules include:
a. Zinc fingers, such as those found in TFIIIA and
Sp1, two transcription factors we have already
encountered.
b. Zinc modules found in the glucocorticoid receptor
and other members of this group of nuclear
receptors.
c. Modules containing two zinc ions and six cysteines,
found in the yeast activator GAL4 and its relatives.
2. Homeodomains (HDs). These contain about 60 amino
acids and resemble in structure and function the helixturn-helix DNA-binding domains of prokaryotic
proteins such as the l phage repressor. HDs, found in
a variety of activators, were originally identified in
activators called homeobox proteins that regulate
development in the fruit fly Drosophila.
3. bZIP and bHLH motifs. The CCAAT/enhancer-binding
protein (C/EBP), the MyoD protein, and many other
eukaryotic transcription factors have a highly basic
DNA-binding motif linked to one or both of the protein
dimerization motifs known as leucine zippers and
helix-loop-helix (HLH) motifs. (By the way C/EBP is
different from the CCAAT-binding transcription factor
[CTF, Chapter 10]).
This list is certainly not exhaustive. In fact, several
transcription factors have now been identified that do not
fall into any of these categories.
Transcription-Activating Domains
Most activators have one of these domains, but some have
more than one. So far, most of these domains fall into three
classes, as follows:
1. Acidic domains. The yeast activator GAL4 typifies this
group. It has a 49-amino-acid domain with 11 acidic
amino acids.
2. Glutamine-rich domains. The activator Sp1 has two
such domains, which are about 25% glutamine. One
of these has 39 glutamines in a span of 143 amino acids.
In addition, Sp1 has two other activating domains that
do not fit into any of these three main categories.
3. Proline-rich domains. The activator CTF, for instance,
has a domain of 84 amino acids, 19 of which are
prolines.
Our descriptions of the transcription-activating domains are necessarily nebulous, because the domains themselves are rather ill-defined. The acidic domain, for example,
has seemed to require nothing more than a preponderance
of acidic residues to make it function, which led to the
name “acid blob” to describe this presumably unstructured