Keeping the global population fed in times of climate change and population growth is considered to be one of the greatest challenges of the 21st Century. Plant stress is defined as any external factor that negatively impacts on growth, productivity, reproductive capacity or survival. The use of salinized water in agriculture is likely to become a more regular occurrence, as diminishing freshwater supplies are available for crop irrigation. High salt drastically affects growth and it is therefore necessary that crops be bred to be able to withstand such adversity. Recent advancements in technology allow us to measure gene expression on a genome wide scale, techniques resulting in the development of theoretical models of regulation and the identification of key regulatory genes have been used in Arabidopsis. There is need to transfer this knowledge from model plant to crop, ensuring the application of such technologies to the issue of food security. A large microarray experiment was performed during this project in which the expression of over 60,000 genes were measured in Brassica oleracea GD33DH over a period of 36 hours following salt shock. The use of bioinformatics tools allowed the identification of 7,141 significantly differentially expressed genes in the early response to salt shock in GD33DH. Additional information on the time of differential expression revealed potential genes and mechanisms indicating that metabolism was highly affected by salt shock. Germplasm from crop wild relatives in breeding programmes is a crucial source of genetic material to replace variation lost through years of selective breeding allowing the development of crops with higher stress tolerance. By screening a collection of wild C-genome Brassica species for salt shock tolerance, tolerant germplasm was identified and sequenced alongside susceptible germplasm. Comparative analyses revealed the genes and mechanisms used by wild Brassica species protect themselves from the adverse effects of salt shock. Whole genome duplication events occurring in the recent evolutionary history of C-genome Brassica was examined whereupon it was found that stress specific duplicate genes are on average expressed more highly than single copy suggesting that WGD has implications on the response to stress. These results provide a wealth of potential gene targets for future study and germplasm that can be used in the development of stress tolerant B. oleracea varieties.

Auxin (Indole-3-acetic acid; IAA) can be considered one of the most important hormones in plant development as it coordinates plant development through transcriptional regulation. For years the principal auxin receptor was sought and only relatively recently was it identified as an F-box protein known as TIR1, with five homologues; the auxin F-Box proteins (AFB1–5). The elucidation of a crystallographic structure elegantly demonstrated IAA’s integral role in forming a ternary complex between auxin F-box proteins and Aux/IAA proteins, revealing the mechanism of auxin perception. An exact chemical description of an auxin has been pursued for decades and researchers have always faced the challenge of factoring in the complex nature of multiple auxin-receptor classes, transport and metabolism. My project aimed to develop a better understanding of auxin chemical specificity at the receptor level, focusing on TIR1 and one of its most distantly related homologues, AFB5. We employed a structure activity relationship study with a rational selection of compounds and have defined a pharmacophore for auxin activity for the TIR1 receptor. The thesis also describes a receptor-led, in silico rational drug design approach in which we replaced the carboxylic acid moiety with a tetrazole, giving a novel compound that demonstrated auxin-like activity both in vitro and in vivo. Furthermore this bioisosteric replacement serendipitously demonstrated a novel selectivity for TIR1, with no activity against AFB5. Preliminary in silico docking studies of the TIR1 binding site could not discern between active and in active compounds generating many false positive results, leading us to develop TomoDock; a novel in silico docking approach to study the entire TIR1 receptor binding pocket geometries. Results from TomoDock suggest that binding is a two phase process with active ligands firstly engaging with a niche, which orients them, allowing passage past and molecular filter region before interacting with the binding site. This process contributes to the mechanism of compound selection by the TIR1 pocket receptor. A range of multidisciplinary approaches utilised in this project have allowed us to investigate and report many new insights on the mechanism of auxin perception. Such rational approaches may also be used in other drug discovery programmes alike to help researchers discern compound perception and selectivity.

Genetic studies aimed at onion improvement have been limited because of outcrossing, high heterozygosity and a very large genome size with a high level of repetitive DNA. Onion bulb initiation is photoperiod-dependent, which places a significant barrier to adapting new varieties for growth at different latitudes. In comparison to photoperiodic regulation of flowering, relatively little is known about genetic regulation of the bulbing process. This project aims to test the hypothesis that the genetic regulation of bulb formation in response to daylength is analogous to the daylength regulation of flowering and to identify genes involved in daylength adaptation in onion. A comprehensive set of developmental, diurnal and spatial mRNA and quantitative expression experiments were carried out to investigate the bulbing response. Bulbing ratios were used to measure the bulbing response of onion plants and the reversibility of the bulbing process under long day (LD) and short day (SD) conditions. RNA-Seq analysis provided a large number of differentially expressed transcripts in onion in response to daylength. Five FT and three COL genes were identified in onion including two novel COL sequences. AcFT1 was expressed in LD, which might induce bulb formation, while AcFT4 was expressed in SD, which might inhibit bulb formation. AcFT5 and AcFT6 were expressed in LD and might also be involved in bulb formation itself. FKF1, GI and COL2 showed good diurnal expression patterns consistent with photoperiod sensing and regulation of FT1. All FT genes exhibited different diurnal expression patterns peaking at different times of the day. FKF1, COL2, COL3, FT1, FT4, LFY and GA3ox1 genes showed distinctive patterns of tissue specific expression in onion. FT genes did not show any variation in expression that would account for the difference in critical daylength between the LD and SD varieties of onion.

The aim of this project was to down-regulate ACC oxidase (ACO) 1 and 2 and ACC synthase (ACS) in broccoli (Brassica oleracea var. italica) to lengthen post-harvest shelf-life. The ACO 1 and 2 and ACS cDNAs of broccoli were ligated into pSCV1.0 in sense and antisense orientations in relation to a CaMV 35S promoter and nos terminator. They were electroporated into the Agrobacterium rhizogenes co-integrate strain LBA 9402 pRi1855::GFP, and used to co-transform GDDH33, a doubled haploid line derived from the calabrese cultivar Green Duke. 150 transgenic hairy roots were identified by GFP fluorescence, and 18 were regenerated into whole plants. Four of these lines showed severe rol phenotype, which did not appear to be related to TL₋ DNA insert copy number. The floral buds from T₀ broccoli heads were assayed for post-harvest production of ethylene and chlorophyll levels. T₀ lines with ACC oxidase 1 and 2 constructs produced significantly less ethylene than the control plants, and chlorophyll loss was significantly reduced. A positive correlation between post-harvest bud ethylene production and chlorophyll loss was described by the equation y= 0.2386x-23.041. There were two copies of aco1 and aco2 in the Brassica oleracea genome of which one was mapped for each to linkage group 1 and 8, respectively.

Self-incompatibility is adopted by many flowering plants to prevent inbreeding, and is controlled by a multi-allelic \(S\)-locus. In \(Papaver\) \(rhoeas\), the pistil \(S\)-determinant is PrsS (a small secreted protein); the pollen \(S\)-determinant is PrpS (a novel transmembrane protein). Cognate PrpS-PrsS interaction induces DEVDase-mediated programmed cell death of incompatible pollen. Here, we examined the role of proteasome during the \(Papaver\) SI response and showed that the proteasome is a target of the \(Papaver\) SI response, and is distinct from the SI-induced DEVDase activity. Our main focus here is translational work, attempting to move the Papaver SI system into \(A\). \(thaliana\). We previously demonstrated that PrpS:GFP expressed in \(A\). \(thaliana\) pollen was functional \(in\) \(vitro\). Here, we expressed the female \(S\)-determinant, PrsS, in \(A\). \(thaliana\) and investigated function \(in\) \(vivo\). We present data demonstrating that transgenic A. thaliana stigmas expressing PrsS pollinated with \(A\). \(thaliana\) pollen expressing PrpS:GFP inhibited pollen tube growth in an S-specific manner, and virtually no seed was set. Transformation of both \(PrpS\)\(:\)\(GFP\) and \(PrsS\) into \(A\). \(thaliana\) generated self-incompatible plants that set no self-seed. This demonstrates that transfer of the \(Papaver\) SI system into a highly diverged self-compatible species can result in a fully functional SI system.

Climate change is predicted to have far-reaching deleterious impacts worldwide; agriculture in particular is expected to be effected by significant loss of suitable land and crop yields in the world’s most populous and poorest regions. Crop wild relatives (CWR) are a rich source of underutilised genetic diversity which could help to mitigate climate change for agriculture through breeding new resilient varieties. However, CWR are under-conserved and threatened in the wild. This thesis researches and develops systematic methodologies to advance knowledge and support action on in situ CWR conservation at the global level. Methods included developing a global inventory of CWR associated with crops important for food security worldwide, species distribution modelling, climate change analysis, in situ gap analysis, reserve planning and prioritisation, and, examining the congruence of CWR distributions with regions of high biodiversity and crop diversity. The methods described here can be applied to CWR at both the national and regional level to ensure robust in situ CWR conservation. A principal success of this research is the global CWR inventory, which has been used in several national strategies and as the basis of a major ex situ germplasm collection mission worldwide.

Self-incompatibility (SI) is a genetic mechanism which prevents self-fertilisation via the recognition and rejection of ‘self’ pollen. In the self-incompatible species Papaver rhoeas L., rejection of incompatible pollen is achieved through interaction of the female and male S-determinants, PrsS and PrpS, respectively. This interaction results in a Ca\(^{2+}\)-dependent signalling cascade in the ‘self’ pollen, which mediates programmed cell death (PCD). To date, many downstream effects of SI signalling cascade have been identified, including actin depolymerization, the formation of actin foci, and the activation of caspase-like activities. Work presented in this thesis identified the involvement of Reactive Oxygen Species (ROS) and Nitric Oxide (NO) in the SI response, and the temporal and spatial patterns were characterized. Other studies identified SI-induced cytosolic acidification as a key step in SI. Moreover, investigation of the role of ROS, NO and H\(^+\) revealed that they all play a role in triggering key features of SI: actin foci formation and caspase-3-like activity. Other studies also provided the first evidence for vacuolar breakdown in SI in this species. Data presented also show the first documentation of SI-induced alterations in phospholipids. Together these data further our understanding of mechanisms involved in the complex SI signalling network.

Amblyseius swirskii and Phytoseiulus longipes are targeted as biological control agents for the horticultural pest Tetranychus urticae. This study applies a standardised protocol to evaluate the risk of establishment of introduced species and investigates temperature related behavioural thresholds for all three species. Laboratory results demonstrate a low level of cold tolerance in A. swirskii and no diapause. Field studies recorded 100% mortality within two weeks of outdoor winter exposure. Amblyseius swirskii has a higher activity threshold temperature than it’s target prey T. urticae. Amblyseius swirskii lacks cold tolerance and is unlikely to establish outdoors and thus can be considered a ‘safe candidate’ for release. Laboratory results demonstrate that P. longipes can not diapause yet is more cold tolerant than A. swirskii. Field studies report 100% mortality after 73 days of winter exposure. Phytoseiulus longipes demonstrates mid-range cold tolerance yet is unlikely to survive an entire winter outdoors. Phytoseiulus longipes has lower activity threshold temperatures than T. urticae. Further studies are required on other factors attributable to establishment potential before it can be classified a ‘safe candidate’. As a consequence of the findings of the present study A. swirskii was granted a license for release into the UK in 2006.

Many higher plants use self-incompatibil ity (SI) mechanism to prevent inbreeding and thus encouraging outcrossing. Upon a self-challenge in Papaver rhoeas, a Ca2+-dependent-signal ling-cascade is initiated resulting in the destruction ofthe self-pollen by Programmed Cell Death. Upstream ofPCD, several Sl-specific events are triggered in incompatible pollen, including phosphorylation of soluble inorganic pyrophosphatases (sPPases); alterations to actin; increases in Reactive Oxygen Species (ROS) and Nitric Oxide (NO). In Papaver pollen, sPPases play an important role, as they provide the driving force for biosynthesis; data suggested that Ca2+ and phosphorylation inhibits the sPPases activities, contributing to pollen tube inhibition. Work presented in this thesis characterized Pr-p26.1 sPPases and analysis of phosphomimic mutants in the SI signalling. These studies provide good evidence that, together with Ca2+, phosphorylation, H20 2 and pH dramatically affect sPPase activity. As previous studies showed that increases in ROS and NO are triggered by Sl in incompatible pollen, to provide insights into SI-mediated events, this project investigated protein-targets in pollen modified by oxidation and Snitrosylations after Sl, including actin and actin-associated proteins. Using a mass spectrometry approach we identified several proteins that were modified by oxidation and S-nitrosylation. This has provided us with several potential new mechanisms involved in Sl.

This thesis comprises two Parts and concerns the cytology of Lycopersicon and Solanum. Part I was a study of pachytene in Lycopersicon and began as an attempt to assess the usefulness of this stage as a source of information to aid classification. The principal findings of this thesis were as follows: 1. Chromosomes at pachytene were examined by the 'isolated chromosome technique', the precision of which was augmented by reference to probability theory. 2. In no case was it possible to identify all 12 chromosomes in any species examined but in L. esculentum sub sp humboldtii 9 chromosome types were elucidated. 3. Detailed interspecific comparisons of two distinctive chromosomes revealed that differences were of two kinds (a) difference in chromomere size and number, (b) differences in chromomere sequence presumably due to inversion. 4. Since such a small proportion of the genome was available for reliable interspecies comparisons pachytene morphology probably has little taxonomic value. 5. The occurrence of distinctive types once rather than twice in each nucleus provided morphological evidence for a basic number of 12 rather than 6. 6. Examination of diakinesis confirmed that at this stage the bulk of the chiasmata were confined to the achromatic region. This observation was however reinterpreted as evidence of terminalisation. 7. Evidence was obtained that chiasma formation occurred in both chromatic and achromatic regions. 8. The observation that chiasmata occurred at random throughout the tomato chromosome considered jointly with the existence of very small chromosome changes provided instances of 'cryptic structural differentiation’ and may account for a proportion at least of the genetic breakdowns observed in F\(_2\) populations of interspecific Lycopersicon and Solanum hybrids. Part II was a study of meiosis in polyploid species of Solanum and the principal findings were as follows: 1. Solanum juzepczukii may well have originated from the cross S. acaule x S. stenotomum. 2. S. curtilobum probably originated from a cross between S. juzepczukii and S. tuberosum. 3. During this study an attempt was made to define precisely types of chromosome configuration encountered.