Search

Bambara groundnut (Vigna subterranea (L.) Verdc.) is an important underutilised legume crop, grown mainly by female subsistence farmers in Africa under traditional low input agricultural systems. Bambara groundnut is known as being of high nutritional value, as an atmospheric nitrogen fixer and to possess high levels of drought, pest and disease tolerance. Bambara groundnut is a predominantly self-pollinated crop and is grown as locally adapted landraces. These are expected to exist as non-identical inbred lines and are generally low yielding. Strategies involving genetic analysis of this species could provide important data for breeding programmes that could enhance food security in Africa. A set of 124 SSR primers designed from different library sources were tested to screen a ‘narrow’ genetic cross (F3) and a ‘wide’ genetic cross (F2) . The former is a cross between domesticated landraces (DipC and Tiga necaru) while the latter is a cross between a domesticated landrace and a wild ancestor (DipC and VSSP11). Residual heterozygosity in the F3 ‘narrow’ cross was confirmed to be around 25% based on 33 polymorphic SSR primers, consistent with an F3 population. A ‘narrow’ cross linkage map was constructed for the first time in bambara groundnut using 269 polymorphic markers (236 DArT and 33 SSR). The map consisted of 238 markers in 21 linkage groups of two or more linked markers, totalling 608.1cM and covering a predicted 54% of the bambara groundnut genome, although the high marker-marker linkage (at 89%) suggests a more comprehensive coverage. QTL analysis was carried out for 73 bulked lines of an F3 population and plants were evaluated for traits in a controlled glasshouse suite and a field trial in Indonesia. Data from single plant analysis of the F2 generation of this cross grown in a controlled environment glasshouse was also used. Most of the QTLs detected were clustered on linkage groups 1, 4 and 12. Major QTLs for internode length and biomass dry weight were detected on LG4 and LG1, respectively, for the FutureCrop glasshouse and field datasets. The highest LOD score of 9.7 was detected for peduncle length and was located within the confidence interval for a QTL for internode length locus. Marker locus bgPabg-596774 was detected to be associated with QTL for six traits; node no./plant, pod no/plant, pod weight, seed no./plant, seed yield and biomass dry weight, on LG1 within one LOD score of confidential interval, potentially suggesting pleiotropic effects of a more limited number (or even one) gene(s). One hundred and fifty-nine additional markers (136 DArT and 23 SSR) were used to improve the existing partial ‘wide’ map (141 AFLP, 1 SSR) constructed in an F2 population of 98 plants. A total of 194 markers were assigned to 20 linkage groups spanning a total of 901 cM. The linkage map derived from the ‘wide’ cross (DipC x VSSP11) had an expected genome coverage of 79.6%. An attempt to combine both maps through 32 common markers allowed a common QTL for days to emergence to be detected in both populations in close association with the common DArT markers 601384 and 601748. The main segregating traits were found to be plant spread, internode length, growth habit, peduncle length, pod weight, seed yield and biomass dry weight. Detecting the same QTL positions for a number of traits, suggested that common underlying genes might be responsible. The QTL-DNA marker associations developed in this study could be used practically for MAS in a future breeding program of this crop.

Mobile DNA sequences - transposable elements (TEs) that amplify and move within genomes represent a high proportion of the DNA in most eukaryotes. The present study aimed to define TE nature, structure and abundance in two contrasting groups of diploid and polyploids crop genera, Brassica (dicotyledon) and Musa (monocotyledon). Rather than starting with known TE sequences, a sequence-data driven approach was used, comparing homologous and homoeologous BAC pairs. Over ~100 kb regions, any stretch of sequence was characterized that was inserted or deleted in the evolutionary time since divergence of the two BAC genomic sequences. Almost all the sequences were indeed TEs, representatives of existing and a few novel superfamilies. Polymorphisms due to activity were measured by PCR with flanking primers in 40 (Brassica) or 96 (Musa) accessions, and some families were localized on chromosomes by fluorescent in situ hybridization. Autonomous and non-autonomous TEs were found; class I retrotransposons like Copia and Gypsy (LTR) predominated in both genera, while SINEs and LINEs (Non-LTR) were abundant in Brassica genomes. Large retrotransposon derivatives (LARDs) were in both genera, with a very few terminal-repeat in miniature (TRIM) elements. Class II DNA transposons included CACTA, hAT, Harbinger, Mariner and Mutator like MITEs in Brassica, while CACTA and Mutator were uncommon in Musa. Among miniature inverted-repeat transposable elements (MITEs), Stowaway, Tourist, and Mutator-like MITEs were abundant with several novel families identified and characterized. In diploid and allopolyploid Brassica species, A- or C-genome specific elements were found while others were more active. PCR enabled accession identification and phylogenetic reconstruction in Brassica and Musa. As well as known element families, few novel types of TEs were identified, including several variable, short elements with characteristic structural features. The analysis provides insight into the nature and diversity of TEs as an important genomic component; results are useful for genome annotation and understanding evolution and variation within these crops and the associated pool of wild germplasm.

It has been shown that under high light (HL) conditions, Arabidopsis bundle sheath cell (BSC) , chloroplasts accumulate hydrogen peroxide (H202) as part of a retrograde signalling network associated with the induction of HL-responsive ASCORBATE PEROXIDASE2 (APX2). It was postulated that this network is mediated by both the "core" abscisic acid (ABA) signalling pathway and G protein signalling and involves the heterotrimeric Ga subunit (GP AI), type 2C protein phosphatases (PP2Cs), SNFI-related kinases 2 (SnRK2s) and respiratory burst oxidase homo logs (RBOHs). The main objective of this study was to determine to what extent these two pathways were implicated in the regulation of APX2 and establish whether the same pathway regulates the expression of other HL-responsive genes in Arabidopsis. Finally, the project looked at identifying if GPAI and SnRK2.6 were integrated into the same HL, ABA-directed signalling pathway. Screening of null mutants under HL showed that GP Al was responsible for the negative regulation of APX2 and extracellular H202 together with ABI2. It was predicted that RBOH F acts upstream of RBOH D where it may interact with SnRK2.6 and in turn regulates ROS production from RBOH D. Screening of the F3 population from the cross between ost1-1 and gpa1-4 under HL indicated that both GPAI and OSTI do not feature on the same signalling pathway. A microarray analysis of HL- responsive gene overlaps between PP2C and SnRK2 mutants revealed that only a proportion of genes regulated by ABA and HL were regulated by this pathway. The promoters of these genes contained both heat shock elements (HSEs) and ABA-responsive elements (ABREs). This highlighted the fact that only a few genes conformed to the same signalling that occurs for APX2. In summary, ABA-directed signalling may control three separate pathways that are regulated by HSFs, ABFs/ AREBS and genes that respond indirectly to HL via other regulatory pathways.

The jasmonate (JA)-dependent signalling gene network in Arabidopsis thaliana differentially regulates responses to biotic (e.g. pathogens, wounding) and abiotic (e.g. drought) stress. In this study, three transcriptional networks using gene expression data (a large data set of microarray experiments) have been constructed in order to model the interactions between individual genes. The modelling procedures make use of gene co-expression signatures measured by partial correlation. This study presents the novel idea of integrating these signatures by generating edges between two genes in the network on the basis of profiles of partial correlations to a curated set of genes known to be involved in the response. The networks assess the interactions between regulators in the JA pathway and their cross-talk with other hormone-dependent pathways. The results confirm previously known relationships between genes based on experimental evidence and highlight additional connections to test experimentally. This method gives new insights for the discovery of novel genes in the transcriptional coordination of the JA signalling pathway.

Reactive oxygen species (ROS) such as superoxide, peroxide, hydroxyl ion and singlet oxygen radical are by-products of specific metabolic reactions in plants. ROS damage cellular biomolecules and are thus actively scavenged. However, ROS also act as signal transducers during development and stress responses. Redox enzymes such as superoxide dismutase (SOD) contribute to achieving a balance between ROS being scavenged and being available as signal transducers. Spatio-temporally differential expression of SODs in various tissues plays a critical role in scavenging the superoxide radical. Using a nitroblue-tetrazolium-riboflavin-based in-gel SOD assay, up-regulation of two low mobility (~96 kDa and ~92 kDa) putative SODs was observed in rice spikelets. The two proteins were purified by gel electrophoresis and both were identified through mass spectrometry as homotetrameric versions of a single germin-like protein (GLP), encoded by the gene LaC _Os08g08960 (Rice Genome Annotation Project). Staining with a glycoprotein gel stain and enzymatic deglycosylation showed that the protein is post-translationally modified by N-glycosylation; differential modification probably accounts for_the difference between the two forms. Biochemical characterization- of this developmentally regulated spikelet-specific GLP (ssGLP) indicated that its SOD activity is Mn-SOD type and is highly stable in denaturing agents and at elevated temperatures. Tetrameric ssGLP lacks oxalate oxidase activity. The biological functions of GLPs have not been fully defined; among the different functions attributed to them their role in biotic stress tolerance is relatively well studied. Rice germins, especially some of those with genes on chromosome 8 in the same cluster as the LaC _Os08g08960 gene, although not this specific gene, have previously been found to contribute to the basal mechanism of resistance to pathogens causing rice blast and sheath blight disease. ssGLP was estimated to contribute nearly 11% of the total SOD activity in rice spikelets, indicating a crucial developmental role. Expression analysis with the in-gel SOD assay at the spatio-temporal level, as well as treatment with various hormones, chemicals and abiotic variables such as wounding and light indicated that the expression of ssGLP is affected by developmental rather than stress stimuli. Further, expression analysis at the spatio-temporallevel through RT-PCR confirmed the developmental patterns observed previously at the protein level. In situ mRNA hybridization studies provided evidence for ssGLP gene expression in pollen and endosperm and in localized regions of leaves, and for production of antisense RNA. Consistent with the up-regulation in spikelets, in silico comparison of the rice GLP promoters revealed more seed and pollen specific motifs in LOC_Os08g08960 than in other GLP genes. Additionally, presence of a GAF protein domain typical of cyclic nucleotide phosphodiesterases is unique to ssGLP. Preliminary enzyme assays and in silico results suggested that ssGLP may act as a nucleotide-sugar pyrophosphatase/phosphodiesterase (NSPPase) similarly to a previously-described barley GLP. As a first step to a fuller definition of the enzyme activity of ssGLP and in particular to testing the hypothesis that a different oligomeric configuration acts as NSPPase, a gene for recombinant ssGLP was constructed, cloned and expressed in Pichia pastoris.

To investigate the role of extracytoplasmic function sigma factors in the Rhizobium-legume symbiosis, the genome sequence of Rhizobium leguminosarum bivoar viciae 3841 was analysed using bioinformatics approaches. In total, twenty sigma factors were identified, including sixteen extracytoplasmic function sigma factors. Twelve of the sixteen ECF sigma factors appeared to be co-transcribed with a regulatory gene. The role of the ECF sigma factors in responding to legume produced compounds was investigated by growing strain 3841 containing various ecf reporter plasmids with pea root exudate. Three ECF sigma factors showed different expression patterns when grown with root exudate compared to minimal medium. One of the three sigma factor aenes. ecfR, which was induced by pea root exudates, was investigated further and was shown to be expressed in pea infection threads. Over-expression of ecfR was lethal to R. leguminosarum 3841, but co-expression with the adjacent gene, asfR, prevented the lethal effect.

Expression vectors based on the bipartite Cowpea Mosaic Virus (CPMV) have been investigated for use in molecular farming in plants. Agroinfiltration into Nicotiana benthamiana of two full-length RNA-2 constructs containing different marker genes in the presence of RNA-1 showed that two foreign proteins can be efficiently expressed within the same cell in inoculated tissue. However, segregation of the RNA-2 molecules was observed upon systemic spread of the recombinant viruses. Thus efficient assembly of heteromeric proteins is likely to occur only in inoculated tissue. Therefore vectors based on full-length RNA-2 molecules do not have an advantage over vectors based on deleted versions of RNA-2 that are restricted to infiltrated tissue.

Transformation of the chloroplast genome is a technique that allows for the production of recombinant proteins in plants. Chloroplast transformation allows the very high expression of a transgene while limiting the gene flow to other species. In addition, multiple genes can be transferred in one transformation event and no gene silencing as been shown so far. Despite these advantages, little work has been directed at assessing the commercial feasibility of using the chloroplast as a means of expressing high-value proteins. In this thesis, a new expression system was developed to allow the very high expression of transgenes in the chloroplast under contained conditions. Tobacco plants were transformed with a tobacco chloroplast vector expressing green fluorescent \ protein. Cell suspensions were induced from these leaves. Using a bioreactor, I was able to demonstrate that coupling temporary immersion with a hormonal shift triggered the rapid production of plant tissue whilst retaining a high level of expression of green fluorescent protein. Another aspect of this thesis was to assess the potential of several fusion tags to improve the solubility and purification of various target proteins in transformed tobacco chloroplasts. The main proteins studied were a mannanase from coffee, which is involved in the detergent, pulp and, more recently, the bioethanol industry, as well as alpha defensin 1 peptide, which has potential therapeutic value in the treatment of several diseases such as HIV and Herpes. N- and C-terminal constructions were created with oleosin, dehydrin, fibrillin, maltose-binding protein and glutathione-S-transferase as tags. Constructs were first evaluated in Escherichia coli before being integrated into the tobacco plastome. Apart from oleo sin, all fusion proteins were successfully expressed in transplastomic tobacco. My work has identified dehydrin, GST and MBP as promising affinity-tags to be used in chloroplast transformation experiments. Finally, I describe the development of experimental tools and procedures for the transformation of the chloroplast genome of coffee, which is one of the world's major cash crops. For this, coffee-specific vectors were created and direct somatic embryogenesis employed to propagate transformed tissue.

Gene flow amongst AA genome rice species Oryza sativa (cultivated rice) and 0. rufipogon and 0. nivara (wild relatives) is known to occur in rice agro-ecosystems in many Asian countries. Gene flow leads to the formation of hybrid plants that may coexist with parental species. These hybrids (especially 'weedy rice'- plants that are very similar to cultivars but possessing undesirable traits) are a threat to rice production as they may reduce cultivated rice yields, both qualitatively and quantitatively. The precise origins of 'weedy' rice (WR) genotypes are unknown but two main hypotheses have been proposed: endoferality (by segregation and selection of weedy traits from existing heterozygous cultivars in areas where no wild rice exists) and exoferality via gene flow between cultivated and wild rice species. With a continuous growing world population especially in Asia, food demand has to be satisfied with an increase in rice production, playing a key role.

The uncertainty of hazards resulting from consuming GM foods over the course of the last decade has been unique. During this time, the risk scene has observed continuous debate, change and public health concerns, in particular those that concern food 'scares' or 'panics'. In response, there is now a growing trend for policymakers to adopt deliberative democratic methods in order to gather public views in an attempt to pacify their concerns. This thesis explores two main schools of thought. First, it delves into the 'world' of food policy. Second, it explores precautionary approaches to GM food consumption 'hazards' or 'risks'.

Any novel transgenic crop requires rigorous testing prior to commercialisation using an ecological risk assessment framework. The aim of this study is to help inform this process using the parasitoid Cotesia marginiventris, the host Spodoptera littoralis, purified Cry1Ab insecticidal protein and Bt176 transgenic maize. Under direct exposure in injecting trials, parasitoid tibia length and sex ratio showed no significant difference between C. marginiventris injected with CrylAb or distilled water. Mean pupation was significantly longer in Cry1Ab infected treatment (9.5 days) when compared to the injected control group (9.1 days). C. marginiventris raised on S. littoralis hosts ingesting the toxin at 100μg g⁻¹ Cry1Ab showed a significantly extended pupation time (11.1 days) when compared to controls (9.5 days), mean tibia length was also significantly shorter at this concentration. The negative effects on pupation time and tibia length were therefore only measurable at 5-10 times plant expression levels. ELISA analysis confirmed that S. littoralis hosts ingested high amounts of Cry1Ab whilst feeding on both plant tissue and artificial diet treatments. However, whilst concentrations of Cry1Ab were found in S. littoralis larvae, frass and parasitoid cocoons none was detectable in parasitoid adults. Therefore the toxin is not accumulating up the food chain and appears to be secreted by C. marginiventris juveniles prior to adult emergence. Further laboratory feeding experiments were carried out to establish the host mediated effects of Cry1Ab on C. marginiventris attack rate, host preference, parasitism success, emergence success and tibia length when reared on S. littoralis hosts ingesting 200, 20, 0μg g⁻¹ Cry1Ab, Bt176 transgenic maize or its near isoline. Host preference of adult parasitoids was also assessed to establish if adults showed any avoidance behaviour towards hosts reared on Bt176 maize, no avoidance behaviour was found. There was no significant difference between attack rate, proportion of attacks resulting in cocoon formation or proportion of adults emerging between the treatments. Tibia length was significantly shorter in the Bt176, 200 and 20μg g⁻¹ CrylAb treatments when compared to the near isoline and control artificial diet treatments. However, there was a trend towards reduced parasitism and emergence success following exposure to the Cry1Ab toxin. C. marginiventris larvae were exposed to the toxin via S. littoralis hosts at 100μg g⁻¹ over two generations. Following two generations of exposure to the toxin pupation time was significantly longer (10.4 days) when compared to controls (10.0 days). Neither tibia length, parasitism or emergence success were significantly different following exposure over two generations.

Extensins are hydroxyproline-rich glycoproteins, which are abundant in the cell wall of most plant species. It has been found that expression of extensin genes is induced in areas of the plant that are subjected to mechanical stresses. The promoter of the extA extensin gene from oilseed rape (Brassica napus) has been previously characterised and a negative regulatory region (NRR), which repressed expression in non-phloem cell types, was located between -664 and -433bp relative to the transcription start site. This repression was shown to be overcome in regions of the plant subjected to mechanical stresses, resulting in high expression levels. In the current work, the NRR from the extA gene promoter was dissected further using a series of PCR-generated 5' promoter truncations fused to the β-glucuronidase (gus) reporter gene. These truncations terminated at -636bp, -619bp, -586bp, -569bp, -547bp, -524bp, -505bp, -481bp and -452bp, relative to the transcription start site. Transgenic tobacco plants containing these truncations were regenerated and extA-driven gus expression was analysed. Histochemical analysis of mature stem cross sections from transgenics containing the -636bp, -619bp, -586bp, -569bp truncations revealed that GUS activity was located in the phloem. Expression became non-specific in transgenics containing the -547bp, -524bp, -505bp, -481 bp, and -452bp truncations, suggesting that a negative regulatory element, repressing expression in non-phloem cell types, exists between -569 and -54 7bp. The ability of the -569 to -547bp region to bind a putative repressor protein was investigated in electrophoretic mobility shift assays (EMS As) using specific oligonucleotides. A MYB core sequence (CTGTTA), which is disrupted in the - 54 7bp truncation construct, was located and EMSAs revealed that an oligonucleotide containing the MYB core and 44bp of flanking sequence bound rape leaf nuclear proteins and formed two specific complexes of reduced mobility. These specific complexes were designated B 1 and B2. Northern analysis identified developmental stages when extA was expressed; the extA transcript is absent in mature laminae and present to high levels in mature petiole. B1 and B2 complexes were only formed with nuclear protein extracts from mature B. napus lamina and not with extracts from mature petiole, implying that the complexes consist of repressor proteins, preventing extA expression in the lamina. Different regions of the NRR formed additional complexes with nuclear proteins extracted from rape leaves, indicating that the NRR is functionally composite.

Particle bombardment of immature scutella was used to generate multiple lines of transgenic wheat with either PHY A, PHYB or PHYC from Arabidopsis or PHY A from oat.;Responses of segregating transgenic seedlings to cFR light were used as a screen for biologically active lines. The 'short coleoptile' phenotype under cFR was shown to segregate with the oat PHYA transgene and rtPCR detection of transcript. All of the transgenic lines containing the Arabidopsis PHYA and several oat PHYA lines demonstrated phenotypes indistinct to wild-type, possibly demonstrating transcriptional silencing.;Biologically active lines with Mendelian segregation ratios of 3:1 were selected for further analysis and homozygous individuals for each identified by PCR. Southern analysis of selected lines demonstrated that all originated from a distinct transformation event and harboured at least a single intact copy of the transgene. The complexity of gene integration was related to observed gene dosage and silencing effects.;Adult transgenic plants under white light with high R:FR ratio, demonstrated an increased chlorophyll content, a reduced production of superfluous tillers, a reduction in sensitivity to shortened day-length and an increased developmental rate leading to a reduction in days to heading.;The transgenic lines demonstrated an altered response to shade. In addition to the phenotypes observed under white light, plant height was significantly reduced and harvest index increased compared to the nulls, these responses were distinct to the those observed under white light and suggest an altered response to reduced R:FR.;The results suggest that oat PHYA is biologically active in transgenic wheat and that over-expressing PHYA alters the responses of etiolated seedlings to cFR, the light grown morphology and development of adult plants, and the responses of the plants to shade.