|Title||Optimizing the design of oligonucleotides for homology directed gene targeting|
|Publication Type||Journal Article|
|Year of Publication||2011|
|Authors||Miné-Hattab J, Fleury G, Prévost C, Dutreix M, Viovy J-L|
BACKGROUND: Gene targeting depends on the ability of cells to use homologous recombination to integrate exogenous DNA into their own genome. A robust mechanistic model of homologous recombination is necessary to fully exploit gene targeting for therapeutic benefit. METHODOLOGY/PRINCIPAL FINDINGS: In this work, our recently developed numerical simulation model for homology search is employed to develop rules for the design of oligonucleotides used in gene targeting. A Metropolis Monte-Carlo algorithm is used to predict the pairing dynamics of an oligonucleotide with the target double-stranded DNA. The model calculates the base-alignment between a long, target double-stranded DNA and a probe nucleoprotein filament comprised of homologous recombination proteins (Rad51 or RecA) polymerized on a single strand DNA. In this study, we considered different sizes of oligonucleotides containing 1 or 3 base heterologies with the target; different positions on the probe were tested to investigate the effect of the mismatch position on the pairing dynamics and stability. We show that the optimal design is a compromise between the mean time to reach a perfect alignment between the two molecules and the stability of the complex. CONCLUSION AND SIGNIFICANCE: A single heterology can be placed anywhere without significantly affecting the stability of the triplex. In the case of three consecutive heterologies, our modeling recommends using long oligonucleotides (at least 35 bases) in which the heterologous sequences are positioned at an intermediate position. Oligonucleotides should not contain more than 10% consecutive heterologies to guarantee a stable pairing with the target dsDNA. Theoretical modeling cannot replace experiments, but we believe that our model can considerably accelerate optimization of oligonucleotides for gene therapy by predicting their pairing dynamics with the target dsDNA.