Because most plants are sessile organisms they have had to evolve different mechanisms to defend
against herbivores. Cyanogenic glucosides (CNglcs) are ancient defence compounds found in
certain plant and a few insect species, which releases cyanide upon attack. The biosynthesis of
CNglcs in plants and insects is performed by three enzymes converting an amino acid into its
corresponding CNglc. The pathways have been shown to have evolved convergently with the
enzymes from plants and insects having less than 20 percent sequence identity at the amino acid
level. The first enzyme in the pathway is from the cytochrome P450 (CYP) 79 and CYP405 family
in plants and insects respectively, which convert the amino acid into its corresponding oxime
through two sequential (N)-hydroxylation’s followed by a dehydration, decarboxylation and
isomerization step.
Enzymes from both families display a high degree of substrate specificity, unlike many
mammalian cytochromes P450. CYP405A2 from Zygaena filipendulae and CYP79D3 from Lotus
corniculatus both convert isoleucine and valine into their corresponding oximes, but neither will
convert leucine neatly illustrating the high degree of specificity the enzymes possess. Previous
work on CYP79A1 has shown that tyrosine is the only substrate. However, in-depth studies
measuring substrate affinities and investigating the structural features responsible for achieving
the specificity have not been performed for any of the families. To investigate the structural
features determining substrate specificity in the CYP79 and CYP405 families we chose to initiate
a large-scale project, attempting to express and purify multiple P450s with the aim of obtaining
crystal structures.
The research presented in this thesis demonstrates how to engineer plant and insect microsomal
cytochromes P450 to express them in a soluble state in E. coli, using various modifications of the
primary sequence as well as optimizing the growth conditions. The modified enzyme CYP79A1
was purified and attempted structurally elucidated using crystallography. As this was not
successful, within the time frame of this PhD study, a combined approach of visible-light
fluorescence spectroscopy and in silico methods was used to investigate structural features
important for determining the specificity in these proteins. We identified residue in the F and I
helix as well as the β1-4 and β4-2 strands as likely determinants of substrate specificity, although
possibly not the only determinants.
The results obtained in this PhD, represent an advance in our understanding of how these enzymes
function and have achieved their high degree of specificity. Furthermore, the accumulated
knowledge this thesis represents regarding expression and purification of soluble CYP79s and
CYP405s can be used in the future as a stepping stone for more detailed structural investigation.