Biology Reference
In-Depth Information
concept is essentially that of a hardwired genomic program
for development. There is no concept more intrinsic to
systems developmental biology than that of process control
by a preformed genomic program, in exactly the same
sense as the term 'program' is used for the code directing
a complex stepwise computational operation.
We have two reasons to begin our chapter in this manner.
First, we wish to make clear that our definition of systems
developmental biology is primarily conceptual rather than
one dependent on technology, throughput, or data mass. For
us, systems developmental biology is the means of achieving
a causal explanation of development, which begins with
genomic regulatory information and extends vertically to all
the different levels of biological organization directly
affected by the developmental genomic control system. The
explanation must encompass both the spatial allocation of
gene expression and its temporal sequence. This is funda-
mentally a conceptual as well as an experimental discovery
problem in information processing and regulatory logic, on
a system-wide scale. Second, by alluding briefly to the deep
roots of this conceptual problem, we can see that it was to
a large extent formulated independently of the gigantic mass
of current knowledge of molecular, biochemical, and cell
biological detail. Indeed, the subject of this chapter is just
how the 'material configuration' of the control system that
Wilson deduced actually causes the process which, from then
until now, has been termed 'development'.
'material configuration'
and the outcome of this mate-
rial configuration is the spatial allocation of cell types,
tissues, organs and appendages that define the adult
organism.
The organization of the body plan is determined by the
assignment of cell fates in specific spatial arrangements
oriented with respect to the body axes. The earliest
assignment of spatial cell fate domains occurs at the tran-
scriptional level, long before any morphogenetic functions
are deployed. That is, embryonic organization depends
directly on the transcription factors that come to be
expressed in each cell and in each spatial domain. When
and where transcription factors and signaling molecules are
expressed during embryonic development are in turn
determined by their transcriptional control apparatus. We
can extend this principle from the earliest expression of the
zygotic genome to the differentiated cell types of the adult
organism. The formation of spatial regulatory domains
depends on the genomically encoded structure of GRNs.
These networks include both the regulatory genes and the
regulatory interactions between them, such that each node
of the network represents a regulatory gene and its cis-
regulatory control system. Multiple regulatory interactions
are required to control the expression of each regulatory
gene, and each regulatory factor in turn provides inputs into
multiple target genes; hence the network character of
GRNs. GRNs control the spatial organization of the
embryo, by determining the regulatory character of cell fate
domains and by setting their boundaries.
How are GRNs encoded in the genome and how could
they account for developmental complexity? Separate
genomic entities encode the GRN nodes and linkages.
Whereas the identity of the nodes is given by the protein-
coding sequences of the regulatory genes, the GRN link-
ages are encoded in the cis-regulatory sequences associated
with those genes. The transcriptional control system relies
on multiple combinatorial information-processing func-
tions:
[1]
each gene is regulated by several cis-regulatory
modules, each with a defined spatial and temporal activity
profile;
[2]
the activity of each cis-regulatory module is
controlled by a specific combination of transcription
factors, determined by the specificities of transcription
factor-binding sites encoded in its regulatory sequences;
[3]
the cis-regulatory logic processing functions (AND, OR,
NOT) define the relation between these inputs and the
regulatory output of the module;
[4]
the transcriptional
activity of any particular gene depends on the state of each
cis-regulatory module at each time and place, and on the
relative contributions of these regulatory modules.
In summary, developmental GRNs include a number of
combinatorial logic processing functions which are essen-
tial for the increase in complexity during development.
They explain why such diversity in developmental outcome
can be achieved with a limited set of transcription factors.
e
DEVELOPMENT IS A SYSTEM-WIDE DIRECT
OUTPUT OF THE GENOME
Biological organization underlies development at every
level from genome to organism. If the question is how
many genes are needed to form an adult organism, then the
answer will be close to the total number of genes encoded
in the genome. Numerous genetic screens have attempted
to catalogue genes based on the developmental processes
they are involved in. However, even though these studies
have demonstrated a great many molecules involved in
specific developmental processes, they are not designed to
reveal how these individual components actually work
together to form any given body part in a normally
developing organism. It is the goal of systems biology to
ask not only which molecular components are required for
any given process, but also how these components operate
on one another so as to promote function. Here we outline
a system-level approach to development, addressing
specifically how body plan organization is controlled,
what types of molecule are required for this control and
how they interact, and how both molecules and interac-
tions are encoded in the genome so as to generate the
developmental process. The developmental process is thus
determined by the genomic control system
Wilson's
e
