Rapid and complete shift of LpCSfoHR cells (mtDNA heteroplasmic cells) to a rps4 -null state by the expression of mitochondrion-targeted Sfo I. (a) Diagram of wild-type-mtDNA (Wt-mtDNA; outer green circle) and mutant-mtDNA (Mut-mtDNA; inner blue circle) indicating the positions of the restriction sites (Sfo I, 52,908; Nde I, 980 and 51,540) and the oligonucleotide probes (gray arrows "1", "2", and "3") used. The dia3 gene cluster (yellow line) containing mutated rps4 gene (red line) in Mut-mtDNA is slightly shorter than that in Wt-mtDNA, because the Sfo I site and a 5'-half of rps4 coding region were deleted. (b) MB35 cells and LpCSfoHR cells grown with (+Tet) or without (-Tet) tetracycline were washed twice in BSS and shaken in BSS for the indicated times (hr) at 22°C. This was followed by extraction of the total RNA and Northern hybridization using the probe "1" that covers the whole rps4 gene. The expression of dia3 gene in MB35 cells is usually transient and reaches to the maximum level after about 2 hours of starvation in BSS. The large polycistronic dia3 transcript (6.5 kb) was detected in LpCSfoHR (+Tet) cells, because they have heteroplasmically Wt-mtDNA and Mut-DNA under the +Tet condition. In LpCSfoHR cells grown under the -Tet condition, however, only Wt-mtDNA but not Mut-mDNA was selectively cleaved by Sfo I, thus resulting in formation of rps4-null cells. Actually, Wt-mtDNA (wild-type dia3 transcript) as detected by the probe "1" was never observed in LpCSfoHR (-Tet) cells. V, vegetative growth phase. (c) When the dia3 transcripts from Wt-mtDNA and/or Mut-mtDNA were hybridized by the probe "2", they gave a positive band at the position of about 6.5 kb, because this probe must detect the mutated dia3 transcript with a slightly shorter length as well as the wild-type dia3 transcript. (d) Southern blot analysis of total DNAs extracted from MB35, LpCSfo and LpCSfoHR cells. DNAs were digested with the indicated restriction enzymes (N, Nde I; NS, Nde I and Sfo I), followed by electrophoresis and Southern blot analysis, as described in the legend of Figure 1d. The numbers shown below the cells used indicate incubation times (hr) in PS medium under the Tet-minus conditions. As probes, the probe "2" (for mitochondrial rps4), "3" (mitochondrial nad7 as a region other than rps4) and a specific probe for nuclear dd-trap1 were used. Since LpCSfo cells grown without tetracycline lose their mtDNA and become ρ0 cells, the whole mtDNA (about 50 kb) detected by the probe "3" disappeared. Accordingly, bands of 5 kb and 3.6 kb as detected by the probe "1" were never observed in the ρ0 cells. As was expected, almost the same amount of the nuclear gene dd-trap1 was detected in all of the samples examined. (e) Staining of MB35, LpCSfo, LpCSG and LpCSfoHR cells with DAPI. Cells were grown in PS medium in the absence of Tet for 48 hours at 22°C and stained with DAPI. LpCSfo cells are completely devoid of mitochondrial DAPI-staining, though dot-like stains of nuclei are retained under the -Tet condition. In LpCSG cells, the hsEGFP signal with Sfo I was specifically observed in mitochondria. Mitochondria in LpCSfoHR cells were stained with DAPI, because of the presence of Mut-mDNA. Bars, 10 μm. (f) PCR analysis of Wt-mtDNA and Mut-mtDNA in MB35, LpCSfo and LpCSfoHR cells. Quantitative mitochondrial genomic-PCR analysis was carried out using DA3RP1 and DA3RP2 primers (Additional data file 3). The numbers shown below the cells used indicate incubation times (hr) in PS medium under the Tet-minus conditions. It is evident that the bands of 2 kb (Wt-mtDNA) are shifted to 1.5 kb (Mut-mtDNA) under the Tet-minus conditions in LpCSfoHR cells. To control for equal loading, a 0.8 kb fragment of the dd-trap1 gene was amplified at the same time.