Yeast cells deleted for MSK1 do not grow on non-fermentable carbon sources as they are unable to aminoacylate mitochondrial tRNA^Lys (tRK3). This results in the inhibition of mitochondrial protein synthesis and loss of mtDNA. Before investigating the ability of human Kars to import tRK1 into yeast mitochondria, we determined whether Kars could substitute for the yeast Msk1 in the aminoacylation of tRK3. To test this, we introduced the human KARS cDNA or yeast MSK1 on a high/low copy plasmids into a yeast strain heterozygously deleted for MSK1 by insertion of a KANMX4 cassette (NS104). The resulting yeast strains were sporulated on fermentable and nonfermentable carbon containing media plates. In initial tests, all four meiotic progeny from diploid strains that contained high copy yeast MSK1 formed colonies on both fermentable and non-fermentable carbon sources. Progeny from the diploid strain that contained high copy human KARS plasmid yielded four spores on fermentable and nonfermentable carbon sources. Out of the four spores, two were larger and wild type for MSK1 gene while other two were smaller and positive for MSK1 deletion (data not shown). Haploid progeny with KARS on a low copy plasmid failed to grow on non-fermentable carbon sources (data not shown). These results suggest that human KARS can substitute yeast MSK1 deletion partially.

We investigated further the growth properties of the haploid progeny of strain NS104 containing high copy plasmids expressing human mitochondrial KARS or yeast MSK1 under respiring and fermenting conditions. Cells were grown at 30°C in selective medium, serially diluted and spotted on rich carbon sources (YEPD) and on non-fermentable carbon sources (YEP plates supplemented with 3% glycerol/ethanol). Cells expressing high levels of human KARS partially suppressed the growth defect of msk1Δ cells on both fermentable and on non-fermentable carbon sources (Figure 1). The extent of suppression was comparable to that achieved by high levels of ectopically expressed MSK1. Cells deleted for MSK1 exhibited poor growth on rich carbon sources compared to cells wild type for MSK1 (NS101; Figure 1). The reason for the slow growth of the msk1Δ strain on rich carbon sources is not known, but it could be strain specific. However, loss of MSK1 displayed a similar phenotype on non-fermentable carbon sources in other strain backgrounds ([3], present study). Further, human KARS poorly complements the growth defect at 37°C caused by MSK1 deletion (unpublished results). This may be due to requirement of Msk1 at elevated temperature for the import of tRK1 [9].

Nevertheless, the results show that human KARS partially complements yeast MSK1 deletion. We hypothesize that the human Kars can be imported into yeast mitochondria and substitute the function of yeast Msk1.

We used an in vitro import system with bacterially expressed human Kars to investigate whether human Kars can import tRK1 into yeast mitochondria. Yeast mitochondria isolated from strain D273-10B [13] and were incubated with in vitro transcribed [³²P]-labeled tRK1 under standard import conditions and the import efficiency was determined by RNase protection assay. tRK1 was imported into yeast mitochondria in the presence of wild-type yeast cytosol (Figure 2A, lane 1). Yeast cytosol is expected to contain both the cytosolic lysyl-tRNA synthetase and trace amounts of pre-Msk that are required for tRK1 import [3]. tRK1 was not imported into yeast mitochondria in the presence of a yeast cytosolic extract derived from a msk1Δ strain (Figure 2A, lane 2). However, the addition of purified human Kars (Figure 2A, lanes 6 and 7) or yeast Msk1 (Figure 2A, lanes 3–5) resulted in import of tRK1 into yeast mitochondria. A dose dependent increase in the import of tRK1 into yeast mitochondria was observed with increasing concentrations of added human Kars (Figure 2A, lanes 6 and 7). tRK1 was not imported into yeast mitochondria in the presence of yeast Msk1 or human Kars alone (data not shown). Similar results were obtained when yeast mitochondria substituted with mitochondria isolated from rat liver or human 143b cells (Figures 2B and 2C). These results show that human Kars can substitute for yeast Msk1 in targeting tRK1 into mitochondria in vitro in the presence of cytosol.

Cytosolic extract from human cells was used to investigate whether human cytosol can complement the yeast cytosol in directing the import of tRK1 into mammalian and yeast mitochondria. Isolated mitochondria from human 143b cells, rat liver or yeast were incubated with ³²P labeled ATP and processed as described in Methods. tRK1 was not imported into mitochondria in the absence of yeast or human cytosol (Figure 2D). Addition of mammalian cytosol to the reaction stimulated the import of tRK1 into human, rat liver or yeast mitochondria (Figure 2D). Cytosolic extract from a MSK1 deletion strain failed to import the tRK1 either into yeast or mammalian mitochondria but the addition of purified human Kars or yeast Msk1 stimulated the import of tRK1 into mitochondria. These results show that human Kars or yeast Msk1 are equally efficient in stimulating the import of tRK1 and that human cytosol can substitute the role of yeast cytosol in directing the import of tRK1 into mitochondria in vitro.

Next, we analyzed the distribution of tRK1 in wild type and MSK1 deletion strains. High resolution northern blot analysis suggested that more than 95% of the tRK1 is associated with cytoplasm and less than 5% is associated with mitochondria in wild-type cells (data not shown), consistent with previous studies [14]. Earlier studies showed the absence of tRK1 in the mitochondria of msk1Δ strain [3]. However, in our preliminary studies, a small but significant amount of tRK1 was associated with mitochondria in the msk1Δ strain (data not shown).

The association of tRK1 with mitochondria isolated from msk1Δ cells could be due to contamination of mitochondria preparation with cytosol. Loss of mitochondrial DNA as a result of MSK1 deletion could change the mitochondrial morphology and result in enhanced association with the cytosolic fraction. However, we have observed decreased levels of mitochondrial-associated tRK1 in a rho⁰ strain that was generated by ethidium bromide treatment of parental strain NS101 (data not shown). To determine whether tRK1 was localized inside mitochondria or is a cytosolic contaminant, we used several methods that could eliminate cytosolic contamination with little reduction in the endogenous mitochondrial tRNA levels. First, we selectively permeabilized the mitochondrial membranes by digitonin detergent. Low concentrations of digitonin disrupt and partially solubilize the outer mitochondrial membrane as well as removing other membranous structures contaminating the mitochondrial fraction. To determine the concentration of digitonin that is required to reduce the cytosolic contamination, aliquots of mitochondria were treated with increasing concentrations of digitonin prior to centrifugation. Initially, the pellet and supernatant fractions were analyzed for the selective solubilization of specific proteins that serve as markers for the different mitochondrial subfractions. In subsequent experiments, nucleic acids were isolated from the digitonin soluble supernatant and insoluble pellet fractions and specific tRNA species were detected and quantitated by northern analysis.

Treatment of mitochondria with 0.05% digitonin solubilized specifically the mitochondrial outer membrane, as shown by the release of the intermembrane space marker protein, CCPO into the supernatant (Figure 3A). Higher concentrations of digitonin solubilized the mitochondrial inner membrane, shown by the release into the supernatant of the matrix marker protein Put2 and the inner membrane protein Tim23 (Figure 3A). Endogenous tRK3 and imported tRK1 were released into the supernatant at the same concentrations of digitonin, which released the matrix protein Put2 (compare Figures 3A and 3B). These results indicate that mitochondrial membranes are intact and that the imported tRK1 fragments are found in the same intramitochondrial fraction as the endogenous mtDNA-encoded tRNA, tRK3. The levels of tRK2, a non-imported cytosolic tRNA, are reduced by 60–70% compared to the untreated mitochondria at 0.075% concentration (Figure 3B). It appears that digitonin resistant structures are still associated with nonspecific cytosolic tRNAs.

We determined if treatment of mitochondria with MNase (Micrococcal Nuclease) would reduce the contamination of cytosolic tRNAs. However, mitochondria treated with MNase alone showed contamination of cytosolic RNAs as evidenced by the presence of tRK2 (Figure 3D, lanes 2–4). We therefore developed alternative approach in which we treated mitochondria with increased concentrations of digitonin and fixed concentration of MNase or fixed concentration of digitonin and increasing concentrations of MNase to reduce non-specific association of cytosolic RNAs. It was previously shown that nuclear encoded small RNAs non-specifically associated with highly purified mitochondrial preparations but were sensitive to digitonin and MNase treatment [15].

In the first approach, mitochondria were treated with increasing concentrations of digitonin (Figure 3C, lanes 2–4) and then with a fixed concentration of MNase (Figure 3C, lanes 6–8). Aliquots of mitochondria were treated with increasing concentrations of digitonin (0–0.2%) for 15 minutes on ice. Mitochondria were re-isolated and treated with MNase at 500 U/ml for 25 minutes on ice and processed as described in the Methods. As shown in the figure 3C, low concentrations of digitonin (lanes 1–3) failed to reduce the cytosolic RNA contamination whereas the higher concentrations completely eliminates the cytosolic RNA levels but also reduces the endogenous tRNA levels significantly (Figure 3C, lane 4).

In another approach, aliquots of mitochondria were incubated with 0.05% digitonin to solubilize the outer membrane and then treated with different concentrations of MNase for 25 minutes on ice. Total nucleic acids were isolated and the presence of various tRNAs was detected by northern blot using specific probes. The contamination of tRK2 in samples treated with MNase alone was almost completely eliminated by treatment of the mitochondrial fraction with 0.05% digitonin and MNase (Figure 3D, lanes 5–8). This treatment had little effect on the amount of mitochondrial tRNA (represented by tRK3, Figure 3), even at the highest concentration of MNase.

The above experiments showed that neither digitonin nor MNase treatment alone was sufficient to remove the contaminating cytosolic tRNAs. However, we could reduce the amount of contaminating cytosolic RNAs by 97–99% without greatly reducing the endogenous RNA levels by a combined treatment of the mitochondrial fractions with digitonin and MNase. We used this combined treatment to further investigate import of tRK1 into yeast mitochondria.

To further investigate whether isolated mitochondria from msk1Δ strain contained any significant amounts of tRK1, we used the combined digitonin and MNase treatment described above to analyze the tRNAs associated with mitochondria. Mitochondria isolated from wild type and msk1Δ cells were treated with 0.05% digitonin for 30 minutes on ice, re-isolated, and treated with MNase at 500 U/ml for 25 minutes on ice. Total mitochondrial nucleic acids were then extracted and separated by urea-acrylamide gel electrophoresis and examined for the presence of representative cytosolic tRNAs (tRK1, tRK2, tRNA^Cys, tRNA^Phe) by northern analysis (Figure 4A). In the absence of digitonin and MNase treatment, as expected, a small portion of the cytosolic tRNAs are associated with mitochondria isolated from wild type and msk1Δ strains (Figure 4A, lanes 3 and 5). Treatment of mitochondria with digitonin and MNase reduced the level of most cytosolic tRNAs (tRK2, tRNA^Cys, tRNA^Phe). However, the amount of tRK1 is at least 10-fold greater than other cytosolic tRNAs, in both wild type and msk1Δ mitochondria (Figure 4A, lanes 4 and 6; Figure 4B). In mitochondria isolated from msk1Δ, the amount of tRK1 is decreased by 50% but its inaccessibility to digitonin and MN treatment (Figure 4A, lane 6) indicates that tRK1 is imported in these cells, despite the lack of Msk1. Results from three independent experiments suggest that tRK1 can be imported into mitochondria in the absence of Msk1 (Figure 4B).

We excluded the possibility that the MSK1 gene was not deleted or partially deleted in our commercially obtained msk1Δ strain, by PCR analysis. We isolated total genomic DNA from the parent and from the msk1Δ strain and performed PCR with primers flanking the open reading frame of MSK1. The putative msk1Δ strain contained the larger KanMX4 marker gene as the entire coding region of MSK1 is replaced (Figure 5, lanes 1 and 2). We also used primers annealing to regions within the MSK1 gene. As expected, MSK1 gene product was obtained from the parent strain; no product was amplified from the msk1Δ strain (Figure 5, lanes 3 and 4). This analysis confirmed that msk1Δ cells indeed lacked the MSK1 gene. We conclude that yeast mitochondrial lysyl-tRNA synthetase does not play a significant role in the import of native tRK1 into the mitochondrial matrix.

BY4741 (MATa/MATα his3Δ1 leu2Δ0 met15Δ0 ura3Δ0), BY4742 (MATα his3Δ1 leu2Δ0 lys2Δ0 ura3Δ) and msk1Δ (MATa msk::KAN his3Δ1 leu2Δ0 met15Δ0 ura3Δ0) strains were obtained from Research Genetics Inc. Heterozygous diploid strain NS104 (MATa/MATα his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 met15Δ/MET ura3Δ0/ura3Δ0 LYS/lys2Δ0) was constructed by crossing strain BY4742 with msk1Δ. Strain NS108 is isogenic to NS104 but contains MSK1 on a high copy plasmid (pTEF MSK1 URA3-2μ). Strain NS112 (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0) contains human KARS on URA3-2μ plasmid. Standard yeast genetics and techniques were used [27].

The complete coding sequence of human mitochondrial lysyl-tRNA synthetase (KARS) cDNA was amplified by PCR using Thermo polymerase (Ambion) with KARS-A primer (ACTAGTGAATTCATGTTGACGCAAGCTGCT) containing an EcoRI site and KARS-B primer (GGATCGATCTCGAGGACAGAAGTGCCAACTGT) containing a XhoI site. Human KARS and was cloned into the pGEM-T Easy vector to generate plasmid pNS33. pNS33 was digested with EcoRI and XhoI and the KARS ORF was isolated and cloned into a 2μ vector (pTEF-URA3) at EcoRI and XhoI sites [28] (a kind gift from Erica Johnson, TJU) to generate pNS37. MSK1 was amplified using yeast genomic DNA as a template and primers MSK1-A (GGCACGTGACTAGTATGAATGTGCTGTTAAAA) and MSK1-B (GGATCGATAAGCTTTTACTGCCTGTTTACATC) and the PCR product was cloned into pGEM-T Easy vector to create pNS32. A SpeI/HindIII digestion product of pNS32 containing MSK1 was inserted into pTEF-URA to generate pNS34. PCR was performed to check the insertion of KAN marker at MSK1 locus with sense primer MSK1-C (TAGTCTTTTATTCGTGATAAAAGCGAAAAT) and antisense primer MSK1-D (CGTAGCGTAGTTTATTGGTGTAGAGAAAAA). To exclude the possibility of partial deletion of MSK1 gene, PCR was performed with sense primer MSK1-C (TAGTCTTTTATTCGTGAAAAAAGCGAAAAT) and antisense primer MSK1-E (AATACGCAACTCCAAATGGTCGAAGTC CTT).

The human mitochondrial lysyl-tRNA synthetase cDNA cloned in pET 24d vector (Novagen) was expressed in E. coli BL21 (DE3) Codon Plus (RIL) cells and purified to 80% homogeneity as described [12]. The complete coding sequence of yeast mitochondrial lysyl-tRNA synthetase (MSK1) was PCR amplified and cloned into E. coli expression vector pET24d to generate a hexahistidine tag at the carboxyl terminal of synthetase. This construct was expressed in E. coli BL21 (DE3) Codon Plus (RIL) (Stratagene) at 15°C for 12 hours in the presence of 0.5 mM isopropyl-1-thio-β-D-galactopyranoside. The histidine tagged soluble pre-protein was partially purified using Talon metal affinity resin (Clontech) and concentrated. The purified Msk constituted approximately 70% of the total purified protein.

A synthetic gene encoding tRK1 downstream from a T7 polymerase promoter was constructed by annealing ten overlapping oligonucleotides and cloning into EcoRI/BamHI digested pUC19 vector. This plasmid was linearized with Bst NI and used as a template for in vitro transcription using T7 polymerase. The tRK1 in vitro transcript was gel purified and stored in TE pH 8.0. tRK1 was labeled with α-[³²P]-ATP at the 3′ end using E. coli terminal nucleotidyl transferase as described [29]. Prior to import assays, the tRNA was denatured at 80°C followed by slow cooling to 37°C in the presence of 15 mM MgCl₂.

Mitochondria were isolated from yeast [30], rat liver [31] or from 143b osteosarcoma cell lines [29] as described. Yeast cytosolic factors were prepared as described [3]. 0.2 pmol of in vitro synthesized [³²P]-labeled tRK1 was preincubated with 2 µg of yeast cytosolic factors with or without purified human Kars or yeast Msk1 on ice for 10 minutes prior to the import assay. Import of tRNA into mitochondria was performed in a volume of 100 µl containing 100 µg of mitochondria, 20 mM HEPES-KOH pH 7.2, 0.6 M sorbitol, 4 mM ATP, 1 mM GTP, 5 mM MgCl₂, 25 mM KCl and 0.2 pmol of [³²P]-labeled tRK1. The import reaction was carried out at 30°C for 25 minutes. Following import, mitochondria were re-isolated, suspended in import buffer and treated with 2.5 µg/ml RNases (Roche Biochemicals) for 30 minutes on ice to remove non-imported tRK1. Then mitochondria were reisolated and washed three times with SEM buffer (250 mM Sucrose, 10 mM MOPS pH 7.2, 1 mM EDTA). The washed mitochondrial pellet was resuspended in 0.1 M potassium acetate (pH-5.2), 0.5% (w/v) SDS and mitochondrial RNA was extracted using phenol-chloroform as described [3]. The total RNA was resolved on 12% urea-polyacrylamide gel electrophoresis and tRK1 was detected and quantitated by autoradiography or phosporimaging using a Molecular Dynamics PhosphorImager. Import of tRNA into isolated rat liver mitochondria or human cell lines were essentially performed as described [31]. In brief, isolated mitochondria were incubated with tRNA in the presence of energy mix at 30°C for 30 minutes and treated with RNases and processed as above. The Animal Ethics Committee of the University of Hyderabad approved the experiment protocol.

For generation of mitoplast, digitonin was used to permeabilize the outer mitochondrial membrane of mitochondria. Mitochondria were suspended at a concentration of 1 mg/ml in SEM buffer. 250 µg of mitochondria were incubated for 25 minutes on ice in 250 µl of SEM buffer containing 0 to 0.2% digitonin. The resulting mitoplasts were harvested by spinning at 14,000 rpm for 10 minutes and the supernatant was saved for further analysis. The mitochondrial pellet was washed once again with SEM buffer and was used either for micrococcal nuclease treatment or for the isolation of mitochondrial nucleic acids. To determine the efficiency of digitonin treatment, 100 µg of mitochondria were treated as above with digitonin, centrifuged, and the pellet and supernatant fractions separated by SDS-PAGE and transferred to a polyvinylidene difluoride membrane (Immobilon P, Millipore). Immunoblots were performed using antibodies specific for cytochrome c peroxidase (CCPO) (intermembrane space), TIM23 (inner membrane), delta-1-pyrroline-5-carboxylate dehydrogenase (Put2p) (matrix) and porin (outer membrane).

For the Micrococcal nuclease (MNase) treatment, aliquots (250 µg) of mitochondria were resuspended in 2 mM CaCl₂, 250 mM sucrose and 10 mM MOPS pH 7.1. MNase was added to 500 U/ml unless otherwise specified. MNase treatment was performed for 25 minutes on ice with occasional mixing. MNase was then inactivated by the addition of 10 mM EGTA on ice for 5 minutes. The mitochondrial suspension was diluted with SEM buffer containing 2 mM EGTA and centrifuged at 14000 g for 10 minutes. The reisolated mitochondrial pellet was washed one more time in SEM buffer. The resulting pellet was used to isolate total mitochondrial RNA.

Mitochondrial nucleic acids were isolated from the various samples and were separated on 12% urea-acrylamide gels. The transfer of nucleic acids to the nylon membranes was performed as described [29]. Membranes were probed using T4 polynucleotide kinase-labeled oligonucleotide probes specific for tRK1, tRK2, tRK3, tRNA^cys and tRNA^Phe. To detect tRK1, we used the oligonucleotide probe anti tRK1 (GTAGGGGGCTCGAACCCCTAACC) for tRK2, the probe anti tRK2 (GCCGAACGCTCTACCAACTCAGC) for tRK3, the probe anti tRK3 (TAGCTGGAGTTGAACCAAGCATG) for cytosolic tRNA^cys, the probe anti tRNA^cys (TCAGGATCGAACTAAGGACCAAG) and for cytosolic tRNA^Phe, the probe anti tRNA^Phe (TGAGAATCGAACTACATGTAAAT).