Dr. Gregory D. May:
Double strand breaks occur routinely during DNA replication and during meiosis. They can also be induced by DNA damaging agents and transposable elements. Double strand breaks are repaired via two basic classes of pathways: homologous recombination (HR) and non-homologous end joining. All organisms possess both pathways, but organisms with less compact genomes appear to repair most breaks generated in mitotic cells via non-homologous end joining. The knowledge of base sequences in the model plant Arabidopsis has opened up possibilities of gaining precise information on the genetic basis of non-homologous end joining in plants. The best way to accomplish that objective is to apply both a reverse genetic and classical genetic approach. The use of these two well known methods has provided investigators a clue (a) whether the genes controlling non-homologous end joining in yeast and mammals have any homolog in Arabidopsis, and (b) whether recently identified novel mutants unique to plants, are involved in the processing of double strand breaks.
To date, targeted gene replacements occur in higher plants at such a low frequency as to render it an unfeasible approach for everyday use by biologists. The ability to generate targeted gene changes in plants would have as great an impact on fundamental research and crop improvement as did the original plant transformation experiments that took place more than two decades ago.
In a review article published in the February, 2003 issue of Trends in Plant Science (vol. 8: 90-95), Anne B. Britt at UCD and Gregory D. May at Noble Foundation discuss the present advancements in methods to disrupt plant gene function and the mechanisms by which they operate, in order to integrate genes of interest at targeted sites in the host plant.
Reverse Genetic approaches to plant functional genomics: The authors begin the article by mentioning a variety of sources that provide a total of 200,000 random T-DNA or transposable element insertion mutants in Arabidopsis. These mutants are useful but they have limitations in that they are unable to produce null mutants and they fail to target genes of smaller base pairs. The authors describe some of the mechanisms used for gene silencing. In one, termed homology-directed gene silencing, interactions between structurally similar host- and transgenes lead to reduced levels of mRNAs for both resident and non-resident (transgene) loci. Such interactions occur in most plant species at both the transcriptional or posttranscriptional levels. Transcription of the targeted gene is substantially reduced by hypermethylation of regulatory regions. In posttranscription, on the other hand, small interfering RNAs (siRNA) of 21 to 25 nucleotides are formed. These small siRNAs participate in the cleavage of a double-stranded RNA into siRNAs. Directed by a nuclease complex, these small RNAs degrade the target mRNA. Since somatic tissues are more responsive to posttranscription via the expression of dsRNAs, it is a practical way for inhibiting dominant gene function. However, the method has its limited applicability in that its expression, homology-directed gene silencing lacks tissue specificity.
Virus-induced gene silencing: The authors describe another recent method in which viral vectors are used to suppress gene expression. This method, called virus-induced gene silencing, only applies in situations where virus vectors share exon-containing regions of the host gene. Among the vectors used successfully for this purpose are TMV (tobacco mosaic virus) and potato virus X. Although a number of genes in a wide variety of plants were silenced by this method, it is not universally applicable. Whether a particular plant species will respond to this treatment depends upon identifying a suitable viral vector.
Transgene insertion via double-strand- break repair two competing pathways:
The authors point out that transgene replacements occur in yeast (= Saccharomyces cerevisiae), through homologous recombination. On the other hand, integration of plasmid-mediated transgenes in all higher plant- and most of vertebrate tissues takes place in a random fashion via nonhomologous end joining process. The notion that homologous recombination occurs in taxa of smaller genomic size does no longer hold good. For instance, in the moss species, Physcomitrella, the genome size is three times larger than that of Arabidopsis, and yet gene replacements in that species occur through homologous recombination. The authors interpret the above finding as indicating that it is the pattern of expression of proteins and not the genomic size that determines gene integration through homologous recombination. This event encourages the authors to speculate that in the future it might be possible to integrate transgene at predetermined sites by altering the pattern of expression of these genes aided by proteins required for homologous recombination.
The ratio between homologous- and non- homologous recombination in Arabidopsis is approximately 1:3000. According to the authors, one of the ways to increase the chances of homologous recombination is to eliminate the competing non-homologous recombination pathway and to identify the novel genes controlling double strand breaks repair. Some of these proteins have been identified in mammals and yeast; in plants the homologs have been identified, but their roles in gene replacement have not yet been elucidated. The authors speculate that elimination of the non- homologous recombination pathway, reported in Ku mutant plants of Arabidopsis is likely to increase the homologous to nonhomologous transformation ratio, leading to a higher frequency of gene replacement events per introduced transgene.
Transgene integration via homologous recombination; possible targets for upregulation: Results of a recent study in this area indicate that the repair of chromosomal double strand breaks takes place by a process called synthesis-dependent strand-annealing (SDSA). This process involves synthesis of a new single-stranded DNA fragment, matching the base sequences of the break. This mechanism protects the involved chromosome from undergoing deletion, duplications and translocations during mitotic divisions.
Applying double strand breaks repair pathways for the integration of transgenes via homologous recombination: For integration of a transgene at the recipient locus of the host cell, the free ends on the transgene must be available, so that it can be integrated or somehow copied onto the target. The authors explain how the integration of transgenes may take place either through a break-induced replication model or a process which requires induction of a nick on flaps on the displaced strand.
The authors discuss the role of specific mice enzymes (encoded by mErcc1 and mXPF) required for generating target-site nicks during gene replacements. Recently, homologs of these genes and their mutants have been identified in the Arabidopsis database. Attempts to stimulate the HR pathways in plants, by overexpressing E. coli RecA or RuvC Holliday junction resolvase proteins have been found to increase the repair rate of breaks without corresponding improvement in the gene replacement frequency.
Oligonucleotide-directed targeting: Some recent experiments demonstrated that a defect in each of g-globin and alkaline phosphatase genes were corrected using specifically designed chimeric RNA/DNA oligonucleotides. It is presumed that the oligos induced single base changes via base pairing with a homologous sequences, followed by correction of mismatched bases. Gene correction efficiency rates in these reported cases were in the range from 1 to 20%.
In plants, gene conversions are non-specific except when plastid DNA is the substrate.