T-DNA promoter tagging in Nicotiana tabacum : a thesis presented in fulfilment of the requirements for the degree of Master of Philosophy in Genetics at Massey University, Palmerston North, New Zealand

Abstract

Plant development is primarily controlled at the level of gene expression. In order to analyse this regulation it is necessary to isolate genes which are involved in organ development through cellular and tissue determination or which respond to environmental signals. Promoter tagging was chosen in order to identify genes potentially associated with plant development by their spatial and temporal pattern of expression. The introduction of a promoterless reporter gene tag allows the expression patterns of plant genes to be readily characterised. A new series of promoter tagging vectors were constructed from the plasmid pPCV604 (Koncz, 1989). The selectable kanamycin resistance marker gene from pBin6 (Bevan, 1984) was cloned into pPCV604 to create pGT. The hygromycin phosphotransferase gene in pGT was then replaced with a promoterless (β-glucuronidase (gus) gene coupled with octopine synthase termination sequence subcloned from pKiwi101a (Janssen and Gardner, 1989) creating pGTG. This binary transformation vector required the helper pRK replication functions of Agrobacterium tumefaciens strain GV3101. In order to bypass this restriction, the vector sequence of pBin19 was combined with the T-DNA of pGTG to create pBin19-GTG. The latter plasmid was found to have a higher Agrobacterium tumefaciens-mediated Nicotiana tabacum transformation efficiency in strain LBA4404 than pGTG in strain GV3101. In both the pGTG and pBin19-GTG promoter tagging vectors the promoterless gus gene has an initiation codon 62 base pairs inside the T-DNA. This sequence includes translation termination codons in all three reading frames. Therefore, insertion of the T-DNA into a plant gene could lead to activation of the gus gene, under the control of the plant gene promoter, via transcriptional fusion. Nicotiana tabacum leaf segments were transformed with pGTG or pBin19-GTG and transgenic plants selected on kanamycin. A population of 87 transgenic tobacco plants were fluorometrically screened for GUS activity in leaf and root material; 37% were found to contain GUS activity, indicating a high frequency of promoter tagging. Two transgenic plants with root specific gus expression were analysed histochemically. Progeny after self-fertilisation lacked GUS activity, though this was restored in progeny of one plant with 5-azacytidine treatment, suggesting involvement of methylation in the gene silencing. Southern hybridisation, inverse PCR cloning of T-DNA flanking sequences and segregation on kanamycin indicated the presence of multiple T-DNA copies within the primary transformants. Furthermore, inverse PCR sequence from one plant indicated multiple and truncated T-DNA insertions at one or more loci. A further population of transformed plants was generated with pBin19-GTG and histochemically screened for GUS activity in roots (14 positive from 147 tested), shoots (27 positive from 147) and floral organs (14 positive from 56). Overall, combining results from all plant organs tested, an average of 33% of plants were found with GUS activity in one or more organs. A diverse range of patterns of gus expression were observed and described including patterns involving root branching. Forty four plants from this population were analysed for T-DNA copy number via Southern hybridisation with a gus probe (right border junction T-DNA) and nptII probe (central T-DNA). Multiple copies were frequently found with an average of 3.3 T-DNA copies per transgenic plant. Overall, an average of 11% of T-DNA insertions were found to be involved in gus activation. Comparison of the fluorometric (37% positive, 87 plants tested) and histochemical (22% positive, 147 plants tested) screens for GUS activity in root and shoot material was discussed and it is suggested that further care is needed in assigning promoter tagging hits from fluorometric screening. Variable expression was observed with promoter tagged genes. It is suggested that further research is required to determine whether this variation was due to silencing, perhaps by methylation, or was a result of the tagged promoters' normal expression patterns.