Re-engineering the mitochondrial genomes in mammalian cells

Abstract

Mitochondria are subcellular organelles composed of two discrete membranes in the cytoplasm of eukaryotic cells. They have long been recognized as the generators of energy for the cell and also have been known to associate with several metabolic pathways that are crucial for cellular function. Mitochondria have their own genome, mitochondrial DNA (mtDNA), that is completely separated and independent from the much larger nuclear genome, and even have their own system for making proteins from the genes in this mtDNA genome. The human mtDNA is a small (~16.5 kb) circular DNA and defects in this genome can cause a wide range of inherited human diseases. Despite of the significant advances in discovering the mtDNA defects, however, there are currently no effective therapies for these clinically devastating diseases due to the lack of technology for introducing specific modifications into the mitochondrial genomes and for generating accurate mtDNA disease models. The ability to engineer the mitochondrial genomes would provide a powerful tool to create mutants with which many crucial experiments can be performed in the basic mammalian mitochondrial genetic studies as well as in the treatment of human mtDNA diseases. In this review we summarize the current approaches associated with the correction of mtDNA mutations in cells and describe our own efforts for introducing engineered mtDNA constructs into the mitochondria of living cells through bacterial conjugation.

Keywords: bacterial conjugation; lipophilic cations; mitochondrial genome engineering; mitochondrial targeting; mtDNA delivery.

Fig. 1

Schematic representation of the human mitochondrial genome. The genome encodes two ribosomal RNAs (12S and 16S), 22 transfer RNAs (indicated by single letter abbreviation) between the coding genes, and 13 essential genes that encode subunits of the oxidative phosphorylation enzyme complexes. The major noncoding regions in the genome are the D-loop region which includes heavy and light-strand promoters (HSP and LSP), and the origin of L-strand replication (O_L).

Fig. 2

Experimental scheme for transferring engineered mtDNA constructs into the mitochondrial networks of mammalian cells by bacterial conjugation. The circular mouse mtDNA genome is isolated from mouse liver and cloned in E. coli. The E. coli is modified to contain invasive and conjugative ability. The E. coli then invades the cytoplasm and conjugatively transfers the mtDNA construct to the mitochondrial networks of the cell. The cells that the engineered mtDNA construct is incorporated in their mitochondria are then used to generate mouse models of human mitochondrial diseases.

Fig. 3

Invasive E. coli. The arrows indicate invaded E. coli into the cytoplasm of a cell. This invasive E. coli occupies the entire cytoplasm of the cell and eventually kills the cell. The lower panel shows the gentamicine protection assay. Typically gentamicin that is applied to the culture medium kills any bacteria outside the tissue culture cells and only bacteria that have invaded into the cytoplasm are protected from the antibiotic. The EIEC strain that has been invaded into the cells was able to form many colonies after gentamicin treatment whereas in the control plate only a background level of colonies formed when the laboratory strain of E. coli DH5α was applied to tissue culture cells.

Fig. 4

Characterization of modified E. coli. (A) Temperature sensitive EIEC strain. This E. coli can grow at a lower temperature (30℃) whereas is not able to replicate at a higher temperature (42℃). (B) Generation of a population of a non-replicating but metabolically active (conjugative) EIEC daughter cells by temperature and antibiotic (ampicillin) treatment. Non-replicating EIEC donor cells (see plate 2) can actively conjugate with the recipient cells and transfer DNA with which the recipient DH5α can grow on the selection plate (see plate 4). (C) Non-replicating EIEC are still able to actively invade the cytoplasm of tissue culture cells. Note that the non-replicating E. coli are more elongated than the replicating forms. (D) Invasiveness of the non-replicating EIEC was improved by adding the invasin gene from Yersinia pseudotuberculosis. Multiple non-replicating bacteria that invaded in the cell are shown by GFP labeling.