«Maria Svensson Faculty of Natural Resources and Agricultural Sciences Department of Plant Biology and Forest Genetics Uppsala Doctoral thesis Swedish ...»
Studies of Genes Involved in
Regulating Flowering Time in
Faculty of Natural Resources and Agricultural Sciences
Department of Plant Biology and Forest Genetics
Swedish University of Agricultural Sciences
Acta Universitatis Agriculturae Sueciae
© 2006 Maria Svensson, Uppsala
Tryck: SLU Service/Repro, Uppsala 2006
Svensson, M. 2006. Studies of genes involved in regulating flowering time in Arabidopsis thaliana.
ISSN 1652-6880, ISBN 91-576-7051-X.
Transition from a vegetative growth phase to flowering in plants occurs in response to both environmental conditions and endogenous signals. Identification of genes that are involved in regulating the time of flowering is of great importance in agri- and horticulture.
Flowering-time genes can be used for crop improvement by, for instance, engineering plants to flower earlier. This shortening of the time to flowering could result in an extended growing season that could enable farmers to grow more than one crop each year.
In this work, a gene knockout approach using T-DNA tagging and in vivo gene fusion has been employed to identify and characterise genes that are involved in regulating flowering time in the model plant Arabidopsis thaliana. This approach resulted in the identification of two genes, At4g20010 and its homologue At1g31010. Expression studies and GUS histochemical analysis of a reporter gene revealed that At4g20010 is mainly expressed in rapid growing tissues such as root tips, shoot apex, flowers and stem nodes. T-DNA insertional mutants of At4g20010 and At1g31010 exhibit a late-flowering phenotype that can largely be repressed by application of gibberellin. Plants with an insertional mutation in At4g20010 contain a reduced amount of the bioactive gibberellin GA4 compared to wildtype plants. The decreased level of GA4 is not due to a transcriptional repression of the GAbiosynthetic genes AtGA3ox1 or AtGA20ox1, since their expressions were increased in the mutant plants. In silico analyses revealed that the C-terminal protein sequences encoded by At4g20010 and At1g31010 contain RNA-binding motifs, whereas the N-terminal sequences have three-dimensional structures similar to single stranded nucleic acid-binding proteins.
To conclude, At4g20010 and At1g31010 may encode two RNA-binding proteins that are involved in regulating flowering time in A. thaliana by affecting the metabolism of GA.
This can be possible either by a positive regulation of GA3ox at the post-transcriptional level or by a negative regulation of GA2ox.
Keywords: Arabidopsis thaliana, flowering time, fold recognition, GA biosynthesis, promoter trapping, RNA-binding, T-DNA tagging Author’s address: Maria Svensson, Department of Life Sciences, University of Skövde, Box 408, SE-541 28 Skövde, Sweden. E-mail: Maria.Svensson@his.se Appendix This thesis is based on the following papers, which will be referred to by their
I. Svensson, M., Lundh, D., Ejdebäck, M. & Mandal, A. 2004. Functional prediction of a T-DNA tagged gene of Arabidopsis thaliana by in silico analysis. Journal of Molecular Modeling 10, 130-138.
II. Svensson, M., Lundh, D., Bergman, P. & Mandal, A. 2005. Characterisation of a T-DNA-tagged gene of Arabidopsis thaliana that regulates gibberellin metabolism and flowering time. Functional Plant Biology 32, 923-932.
III. Svensson, M., Lundh, D., Bergman, P. & Mandal, A. 2005. At4g20010 and its homologue At1g31010 encode two putative nucleic acid-binding proteins involved in regulating flowering time in Arabidopsis thaliana. (Manuscript).
Papers I and II are reproduced with permission from the publishers.
Contents Introduction, 7 Arabidopsis thaliana, 7 T-DNA tagging, 8 Protein structure prediction, 10 Flowering time, 13 Gibberellins, 15 GA mutants, 15 GA biosynthesis, 15 Feedback and feed-forward regulation of GA biosynthesis, 19 Aims of this study, 21 Results and discussion, 21 Functional prediction of a T-DNA tagged gene of Arabidopsis thaliana by in silico analysis (I), 21 Characterisation of a T-DNA-tagged gene of Arabidopsis thaliana that regulates gibberellin metabolism and flowering time (II), 23 At4g20010 and its homologue At1g31010 encode two putative nucleic acid-binding proteins involved in regulating flowering time in Arabidopsis thaliana (III), 25 Conclusions, 27 Future perspectives, 27 References, 29 Acknowledgments, 36 Abbreviations CDP ent-copalyl diphosphate CPS ent-copalyl diphosphate synthase FMI Floral meristem identity GA Gibberellin GA13ox GA 13-hydroxylase GA20ox GA 20-oxidase GA2ox GA 2-oxidase GA3ox GA 3β-hydroxylase GA-3P Glyceraldehyde-3-phosphate GGDP Geranylgeranyl diphosphate I-PCR Inverse-PCR IPP Isopentenyl diphosphate KAO ent-kaurenoic acid oxidase KO ent-kaurene oxidase KS ent-kaurene synthase MVA Mevalonic acid (Mevalonate) PDB Protein Data Bank RMSD Root mean square deviation RT-PCR Reverse transcriptase PCR UTR Untranslated region Introduction Arabidopsis thaliana During the last decades Arabidopsis thaliana has become one of the most widely used model plant in biological research. Its relatively small genome (~125Mbp) with low amounts of repetitive sequences, short generation time, and its close relationship to the organisms it is meant to model, are some of the reasons why Arabidopsis has been adopted as a model system for molecular and genetic studies (Meyerowitz, 1987; Meinke et al., 1998). Arabidopsis’s value as a model plant increased even more at the end of year 2000 when it was the first plant to have its genome sequenced (The Arabidopsis Genome Initiative, 2000). At that time it was the third genome of a higher eukaryote, after Caenorhabditis elegans and Drosophila melanogaster, that was completely sequenced (The C. elegans sequencing consortium, 1998; Adams et al., 2000). The Arabidopsis ecotype Columbia was sequenced by a public consortium (The Arabidopsis Genome Initiative, 2000), whereas the private company Cereon sequenced the ecotype Landsberg erecta (Jander et al., 2002). After completion of the sequencing process, the genes and other features of the entire Arabidopsis genome were annotated (The Arabidopsis Genome Initiative, 2000). The annotation process is iterative and on-going, and since the end of year 2000 the genome has been reannotated by TIGR, The Institute for Genomic Research (Wortman et al., 2003).
The total number of Arabidopsis genes was initially estimated to 25,498; however, the latest released version of the TIGR ATH1 genome (Version 5) includes an estimated number of 30,700 genes.
Approximately one third of the initially predicted genes could not be assigned any biological function (based on homology searches), and only about 10% of the annotated genes have yet been thoroughly established to have a definitive function (The Arabidopsis Genome Initiative, 2000; Ostergaard & Yanofsky, 2004). These numbers will probably increase drastically in the next few years as the genetic resources in Arabidopsis research have recently been boosted (Ostergaard & Yanofsky, 2004). For example, the development of different public collections of T-DNA tagged lines enables researchers to search a sequence database and find a mutant line with an insertion in their gene of interest. The largest insertion collection for Arabidopsis thaliana was created by Ecker and co-workers at the Salk Institute and it contains more than 225,000 T-DNA tagged lines. For approximately 90,000 of these lines the location of the T-DNA has been determined by sequencing, and it revealed that about 22,000 of the Arabidopsis genes contain T-DNA insertions (Alonso et al., 2003). The SALK T-DNA collection together with several other insertion collections that have been developed for Arabidopsis thaliana, including the SAIL collection, very much contributes to the process of determining gene function through reverse genetics (Balzergue et al., 2001; Sessions et al., 2002; Till et al., 2003).
T-DNA tagging The insertion of foreign DNA into a plant genome is a powerful approach for identifying new genes and determining gene function. A knockout mutation can be generated by inserting a DNA segment with a known sequence into a plant gene, for instance by disrupting the expression of the gene. The knockout of the plant gene may consequently result in plants with a recognisable mutant phenotype. The insertion of the T-DNA does not need to be occurred in the exon of a gene to result a mutant phenotype. Several researchers have shown that insertions can also occur in introns and in 5´ or 3´ non-coding regions as well as resulting plants with mutant phenotypes (reviewed by Azpiroz-Leehan & Feldmann, 1997). The insertional mutagen does not create only a mutation, it also ´tags´ the affected gene. This enables researchers to identify the gene in question. The tagged gene can be identified and isolated by amplifying and sequencing the plant DNA flanking the known insert.
One of the most commonly used methods for transferring foreign DNA into plants genome is Agrobacterium-mediated transformation (Topping et al., 1995;
Tinland, 1996; Zupan et al., 2000). In this method researchers take advantage of the soil bacterium Agrobacterium tumefaciens’s natural ability to transfer a fragment of its own DNA into plant genomes. The DNA that is being transferred (T-DNA) is flanked by 25 bp imperfect direct repeat border sequences, named right and left border. In theory, only the sequence within these borders (the T-DNA sequence) is transferred to the plant genome by Agrobacterium in a random manner. However, in practice sequences outside the T-DNA borders that belong to the transformation vector can also be transferred to the plants (Ramanathan & Veluthambi, 1995; Kononov, Bassuner & Gelvin, 1997; De Buck et al., 2000).
Other rearrangements of the T-DNA and the plant DNA sequence at the site of insertion have also been observed by several researchers (Mayerhofer et al., 1991;
Ohba et al., 1995; Forsbach et al., 2003). Despite these rearrangements that can occur, the Agrobacterium mediated T-DNA transfer system is usually the method of choice since, in comparison with other transformation methods, it usually results in stable transgenes that are intact, non-rearranged and that exist in a low copy number (Gelvin, 1998). Feldmann (1991) showed that the average number of independent inserts was 1.5 per diploid genome, where 57% of the transformed plants contained a single insert and 25% of the plants contained two inserts.
Similar results have also been obtained for other T-DNA insertion collections (McElver et al., 2001; Alonso et al., 2003).
An additional feature of the T-DNA tagging approach is the use of in vivo gene fusion technology. In this technique, a T-DNA vector containing a promoterless or enhancerless reporter gene placed at the right or left end of the T-DNA is employed for gene tagging. Following random insertion of the T-DNA into plant genomes a transcriptional or translational gene fusion between the plant gene and the promoterless reporter gene can be achieved and identified by screening the plants for the activity of the reporter gene (Topping & Lindsey, 1995). In a promoter trap approach the promoterless reporter gene will be activated when inserted downstream of a native plant gene promoter (Fig. 1). An advantage of the promoter trap approach compared to regular T-DNA insertion mutagenesis is that it relies not only on the ability of generating a mutant phenotype but also reveals information about the expression pattern of the tagged gene. This is because of the fact that the expression pattern of the reporter gene usually reflects the expression of the tagged gene (Topping et al., 1995). The first vector designed for promoter trapping was developed by Koncz and co-workers (1989) and contained aminoglycoside (kanamycin) phosphotransferase as a reporter gene. Other reporter genes used frequently in promoter traps are the uidA (β-glucuronidase; GUS), green fluorescent protein (GFP) and the luciferase genes (Riggs & Chrispeels, 1987; Kertbundit et al., 1991; Topping & Lindsey, 1995; Stewart, 2001; Ryu et al., 2004).
Although several different genes have been identified by the promoter trap approach it is not always easy to demonstrate an exact correlation between the expression pattern of the reporter gene and the tagged gene (Pereira, 2000). For example, Stangeland and co-workers (2005) observed GUS activity even when the promoterless gus reporter gene was inserted in intergenic regions and in inverted orientation in respect of the direction of the promoter of the tagged gene. This activation can be explained by the presence of cryptic promoters or by promoters of still unannotated genes (Stangeland et al., 2005). However, when the promoterless reporter gene is inserted in the same orientation as the tagged gene in the 5´UTR or in the intron region, the expression pattern of the reporter gene greatly reflects the pattern of the tagged gene (Stangeland et al., 2005).
Fig. 1. Schematic presentation of T-DNA mediated gene tagging using a promoter trap and in vivo gene fusion. LB, left border; RB, right border; p, promoter; -p reporter, promoterless reporter gene; KmR, kanamycin resistant selectable marker gene.