An araC-controlled bacterial cre expression system to produce DNA minicircle vectors for nuclear and mitochondrial gene therapy.

The presence of CpG motifs and their associated sequences in bacterial DNA causes an immunotoxic response following the delivery of these plasmid vectors into mammalian hosts. We describe a biotechnological approach to the elimination of this problem by the creation of a bacterial cre recombinase expression system, tightly controlled by the arabinose regulon. This permits the Cre-mediated and -directed excision of the entire bacterial vector sequences from plasmid constructs to create supercoiled gene expression minicircles for gene therapy. Minicircle yields using standard culture volumes are sufficient for most in vitro and in vivo applications whereas minicircle expression in vitro is significantly increased over standard plasmid transfection. By the simple expedient of removing the bacterial DNA complement, we significantly reduce the size and CpG content of these expression vectors, which should also reduce DNA-induced inflammatory responses in a dose-dependent manner. We further describe the generation of minicircle expression vectors for mammalian mitochondrial gene therapy, for which no other vector systems currently exist. The removal of bacterial vector sequences should permit appropriate transcription and correct transcriptional cleavage from the mitochondrial minicircle constructs in a mitochondrial environment and brings the realization of mitochondrial gene therapy a step closer.

Introduction of chloramphenicol resistance into the modified mouse mitochondrial genome: cloning of unstable sequences by passage through yeast.

Despite increasing awareness of the importance of the mitochondrial genome in human pathology, very few attempts have been made so far toward genetic engineering of mitochondrial DNA (mtDNA). One of the reasons for this slow progress is the difficulty of cloning mtDNA in Escherichia coli, a trait in common with repetitive or palindromic sequences, and some viral sequences. We have previously made a construct containing the entire mouse mitochondrial genome and a cDNA sequence coding for human ornithine transcarbamylase in a yeast/bacterial shuttle vector, which can be stably maintained in E. coli. We wished to modify this vector for mitochondrial gene therapy by the addition of mitochondrial chloramphenicol resistance, conferred by a point mutation in the 16S rRNA gene. Attempts to modify this construct by a straightforward cloning approach in E. coli proved unsuccessful. Two successful strategies for modification of large unstable constructs in both E. coli and the yeast Saccharomyces cerevisiae are compared here.