Plastid transformation experiments have demonstrated an efficient homologous recombination pathway in plastids mediated by a RecA-homologue that appears to be active throughout shoot development. The presence of this pathway is consistent with a new emerging view of plastid DNA maintenance in which recombination plays a predominant role. WT plastid DNA is comprised of a mixture of circular and linear DNA molecules, which form a multimeric series from monomer to at least the octomer, and high MW DNA complexes (Section 3). Deleted plastid genomes in grasses contain sub-genomic circles and linear hairpin DNA molecules (Section 3.1, 6). The relationship between the mechanisms responsible for the maintenance of WT plastid DNA and the formation and replication of small linear DNA molecules in plastids is not understood. Replication models to account for the different topological forms of WT plastid DNA (circular DNA, linear DNA, branched complex DNA) have been proposed (Fig. 5). To identify which of these models are correct requires experimental confirmation beyond further descriptions of topological forms in WT chloroplasts. Progress in this research area requires the identification of proteins involved and mutants to determine the impact of loss or downregulation of these DNA-RRR proteins on plastid DNA levels and topological forms.
Whilst several approaches have localised putative replication origins in plastid DNA from flowering plants they have been mapped to different positions (Fig. 6) hindering the application of a universal model. Multiple locations for replication origins might reflect multiple origins in plastid DNA and differential usage of replication origins in different cells or differences in the accuracies of the methods used. The possibility of alternative modes of replication in different plastid types (Wang et al. 2003) increases the complexity of studying plastid genome maintenance. Distinguishing between these replication pathways might require the isolation and analysis of mutants affecting specific pathways. Recombination-dependent DNA replication plays an important role in genome maintenance in bacteria and has been suggested to be active in plastids to account for the complex branched DNA structures found in Z. mays plastids (Bendich 2004; Oldenburg and Bendich 2004b). Linear DNA molecules with heterodisperse or defined ends could invade template DNA molecules to prime DNA replication by recombination (Fig. 5d). The identification of linear DNA molecules with fixed ends that map to potential origins in the large inverted repeat (Oldenburg and Bendich 2004b; Scharff and Koop 2006) is interesting and characterisation of the structures of these ends might reveal the mechanisms involved in their formation. Whether recombination-dependent DNA replication is limited to a relatively small number of specific sites in plastids, possibly corresponding to the natural ends mapped in Z. mays (Oldenburg and Bendich 2004b) and N. tabacum (Scharff and Koop 2006), can be addressed by mutating these sites in recombinant plastid genomes (Scharff and Koop 2007). The finding that a high proportion of plastid DNA (50%) is comprised of complex branched DNA molecules in Z. mays seedlings (Oldenburg and Bendich 2004b) warrants further investigation in other species us ing additional techniques such as DNA fibre-based FISH with plastid DNA probes (Lilly et al. 2001) to study the organisation of these complexes.
A highly active homologous recombination pathway in plastids is consistent with recombination-dependent DNA replication. Widespread inter-molecular and intra-molecular recombination between large inverted repeat sequences or between repeated copies of the unit genome would be expected to produce a large number of isomers. If the molecules are linked by strand-invasion and recombination-dependent DNA replication this will give rise to a complex mixture of interconnected high-molecular-weight complexes (Oldenburg and Bendich 2004). The organisation of plastid DNA as high MW multi-genome complexes (containing linear and branched forms) has been suggested to underlie the packaging of plastid DNA into nucleoids (Bendich 2004). Random replication and recombination events are thought to contribute to the random segregation patterns observed for plastid genomes (Birky 1994, 2001). This raises the question of whether all genomes and topological forms have an equal chance of being replicated? Our current knowledge is too limited to address such a question. Other interesting areas worth exploring in future work include the relationship between topological forms and transcription, and the maintenance of heteroplasmic states. Distinct plastid genomes in heteroplasmic plants, where both genomes are required for survival, might be expected to segregate to different high MW DNA complexes within a plastid. However, the maintenance of different plastid genomes in the same high MW complexes might be possible and shed light on the dynamics of plastid genome maintenance. In normal WT plants copy-correction involving DNA repair pathways would be expected to ensure the maintenance of a uniform population of plastid DNA molecules.
Plastids have been evolving in the cytoplasm of their eukaryotic hosts for several billion years and have acquired proteins of nuclear or mitochondrial origin that were not present in the original cyanobacterial symbiont (see for example Wagner and Pfannschmidt 2006). Elucidation of DNA-RRR pathways in plastids should confirm roles for eukaryotic proteins (Mukherjee et al. 1994) in addition to roles for homologues of well known prokaryotic DNA-RRR proteins. It seems likely that the proteins, mechanisms, and regulation of plastid DNA-RRR pathways will have diverged substantially from the eubacterial DNA-RRR model. The availability of whole plant genome sequences allow genomic approaches to identify genes encoding proteins of prokaryotic (Table 2) and eukaryotic origin involved in plastid genome maintenance. A major problem is the prediction of plastid location due to the difficulty in identifying the N-termini of proteins from gene and cDNA sequences, and because computer programs (Emanuelsson et al. 2000) are only partially successful in predicting plastid-targeted proteins. Approximately 30% of chloroplast proteins do not contain recognisable plastid targeting signals (Kleffmann et al. 2004). Proteomics provides an alternative method to identify plastid DNA-RRR proteins. However, proteomic studies on chloroplasts (Kleffmann et al. 2004) have not uncovered the suite of DNA-RRR present in plastids possibly because of their limited abundance. More success has been achieved by proteomics of purified subfractions enriched in nucleoids (Sakai et al. 1999; Phin-ney and Thelen 2005). Alternatively, shoot tissues with actively dividing cells ex press a number of DNA-RRR proteins (Sakai et al. 1999; Saotome et al. 2006) and might provide better material for proteomic studies on plastid DNA-RRR proteins.
Reverse genetics provides a powerful tool to elucidate the roles of candidate DNA-RRR proteins in plastid genome maintenance. Knockout, using T-DNA insertions or transposons, and knockdown approaches, using RNAi, can be used to identify genes with important roles in plastid DNA maintenance. Knockdowns are particularly suitable for studying essential genes by allowing the isolation of viable plants. This provides the plant material in which to study the impact of DNA-RRR protein deficiencies on plastid genome maintenance. Plastid transformation allows the DNA substrates of plastid DNA-RRR pathways to be manipulated. Combining plastid transformation technologies with knockouts and knockdowns in nuclear genes is a particularly attractive method for studying plastid DNA-RRR pathways. Functional assays (see for example Fig. 11c) based on recombinant plastid genomes (Mühlbauer et al. 2002; Khakhlova and Bock 2006; Kode et al. 2006) will enable the impact of plastid DNA-RRR deficiencies on plastid DNA maintenance to be monitored. These new experimental approaches involving ge-nomics, reverse genetics and transplastomic technologies, where both the transacting proteins and cis-acting DNA sequences can be manipulated, are likely to provide the functional studies needed to improve our understanding of the DNA-RRR pathways responsible for the maintenance of plastid genomes.
Was this article helpful?