The vertebrate limb has been used and continues to be used as a model system for the
study of many developmental and evolutionary processes. More often than not,
studies have involved intensive laboratory work. However, as early as 30 years ago
researchers such as Wilby and Ede (1975) recognised the potential of computational
tools in providing a deeper understanding of the development of the vertebrate limb.
Towards this end, the goal of this thesis was to integrate practical and computational
approaches for the investigation and analysis of vertebrate limb morphogenesis.
Before a complete picture of vertebrate limb development can be formed the relevant
components of the system require thorough analysis. One important component is the
changing spatial distribution of cellular proliferation throughout the limb bud tissue
during morphogenesis. To date, all of the proliferation studies completed on the
vertebrate limb are not truly quantitative or comprehensive. It was with this
limitation in mind that a new approach was sought to capture this information. This
new approach involved both optimisation of the experimental technique (BrdU-IddU
double-staining) and development of new computational tools to estimate cell cycle
times in the early vertebrate limb. These developments have allowed, for the first
time, a comprehensive spatio-temporal map of quantitative cell cycle times in the
early vertebrate limb.
A second key question of limb morphogenesis is how genes create the digit pattern.
An example of such a gene is Sox9, which is an early marker of chondrogenesis and
is, therefore, assumed to follow a pattern similar to early stages of digit patterning.
Classical chondrogenic experiments, suggest digital regions are patterned by the
intermediate formation of a
"digital arch" from which the digits arise in a posterior to
anterior order. In contrast, a thorough analysis of a large number of Sox9 in situs
revealed digital regions 1, 2 and 3 branch from a region reminiscent of the tibia
(anterior zeugopod) and digital regions 4 and 5 branch from a fibula-like region
(posterior zeugopod). Moreover, the Sox9 pattern first arises in digital regions 2, 3
and 4, followed by digital regions 5 and 1. The Sox9 in situ analysis was achieved
using newly developed software for the 3D analysis of optical projection
tomographic (OPT) images at a very high spatial resolution.
These studies have highlighted the importance of integrating practical and
computational tools in order to close the gaps in our knowledge and understanding of
limb development, and developmental processes as a whole. The computational tools
generated for the proliferation studies are valuable in offering a thorough means of
analysis of cell cycle times and the new OPT software will be invaluable for the
study of both weak and strong gene expression patterns in whole embryos. In the
future, the proliferation data and 3D Sox9 in situ data can be incorporated into
simulation software, the results of which should shed light upon the interactive
effects of different factors upon the process of limb morphogenesis