Biology Reference
In-Depth Information
A number of metazoans, including the nematode
C. elegans, the fruit fly D. melanogaster, the sea urchin
Strongylocentrotus purpuratus, and mouse, as well as
human cell lines, have been utilized to investigate various
aspects of gene regulation in animals. In C. elegans, gene-
centered Y1H approaches have been used to delineate
several medium-scale GRNs
[28,105
e
107]
. Although
C. elegans has more protein-coding genes and TFs than
Drosophila, intergenic regions are shorter, and it is not clear
whether complex CRMs such as enhancers play a major
role in gene regulation as they do in Drosophila, or if most
regulation is conferred through the proximal gene
promoter. Since C. elegans are transparent throughout their
lifetime, green fluorescent protein (GFP) expression can be
simply examined in living animals by light microscopy (see
Chapter 19). For instance, by using transgenic animals that
express GFP under the control of a variety of gene
promoters it has been found that promoters often contain all
elements required to correct tissue-specific expression
[156
e
158]
. This suggests that long-range gene control
occurs much less frequently in C. elegans than in other
complex metazoans.
Transcriptional enhancers have been studied extensively
in Drosophila.InDrosophila, as in mammals, there are often
multiple tissue-specific enhancers per gene, and enhancers
can be found far from the gene(s) they regulate
[159]
.
Detailed studies of cis-regulatory logic in transcriptional
enhancers have been carried out in early blastoderm devel-
opment
[85]
and embryonic mesoderm development
[89,160]
. Although dense clustering of TF-binding sites is
often thought of as a central feature of transcriptional
enhancers in Drosophila, recent analysis of 280 experi-
mentally verified CRMs has revealed that binding-site
clustering is typical of only a particular subclass of
Drosophila enhancers
[161]
.BothD. melanogaster and
C. elegans served as model organisms in efforts by the
modENCODE Consortium to catalog the functional regu-
latory elements in those genomes
[80,81]
, resulting in
genomic data sets, in particular by ChIP, on in vivo TF-
binding sites for a small number of TFs and hence the
positions of putative DNA regulatory elements. Further,
because only few TFs have been characterized by ChIP in
these studies (22 of 937 for C. elegans and 41 of 755 for
Drosophila), this has not yet led to extensive GRN models.
Finally, testing the activity of putative CREs, such as tissue-
specific transcriptional enhancers, is still only a moderate-
throughput endeavor even in these model organisms.
The sea urchin S. purpuratus (see Chapter 11) offers
several advantages. First, it has fewer gene duplications
than vertebrates, with the practical consequence that there
are fewer functionally redundant genes that need to be
perturbed experimentally in genetic experiments of gene
function. Second, it has a transparent embryo, allowing
direct visualization of development and reporter gene
GRNs: Model Organisms
A variety of model organisms, ranging from relatively
simple unicellular organisms to complex metazoans, have
been employed in the study of GRNs. Each of the different
major model organisms provides particular advantages and
limitations, as discussed briefly here and in more detail in
Box 4.3
. Consequently, these organisms have been used to
investigate different aspects of gene regulation in general
and GRNs in particular. In the sections that follow, the
major model organisms are discussed, starting with simpler
organisms with compact genomes (i.e., microbes), and
moving on to more complex organisms with larger
genomes (i.e., animals and plants).
E. coli has been a long-standing bacterial model system
for the investigation of basic phenomena in TF
e
DNA inter-
actions, such as cooperativity in TF binding to multiple
adjacent DNA recognition sites
[140]
, and the roles of such
interactions in the regulation of gene expression. However,
a variety of features limit the general utility of E. coli as
a model organism for understanding eukaryotic gene regu-
lation. These include a compact genome, expression of
polycistronic transcripts from operons, the lack of a nucleus,
the lack of nucleosomes, and different inherent promoter
activity in the absence of activators or repressors compared to
eukaryotes
[141]
. Despite these limitations, studies in E. coli
have provided important insights into basic biological
mechanisms andGRNs, and E. coli is regaining attentionwith
the growing number of microbiome studies and their impor-
tance for human health
[142]
.
S. cerevisiae is one of the most useful eukaryotic model
systems to study the systems biology of gene expression in
eukaryotes because many basic biological processes,
including transcriptional components such as histones and
chromatin remodeling complexes, are conserved in yeast.
In addition, it has a compact genome, a short doubling time,
and can be easily genetically manipulated, and various
genomic resources are available (see Chapters 6 and 8).
S. cerevisiae served as a primary model system in the devel-
opment of numerous genomic and proteomic technologies,
including gene expression microarrays
[143,144]
,large
collections of bar-coded gene knockouts
[145]
,ChIP-chip
[146
e
148]
(
Box 4.1
), genome-wide screens of protein
e
protein interactions by yeast two-hybrid assays
[149,150]
or by affinity purification coupled with mass spectrom-
etry
[151,152]
, large-scale genetic interaction screens
[153,154]
, as well as computational approaches for pre-
dicting DNA regulatory elements (
Box 4.2
). The wealth
of available genome sequence data and genomic data sets
has permitted a number of pioneering, integrated
approaches in GRN analysis in yeast
[68,155]
. Despite
all the advantages of yeast, it has much simpler CREs
and less combinatorial regulatory input by TFs than is
typically found in higher eukaryotes.
