Overexpression of mitochondrial histidyl-tRNA synthetase restores mitochondrial dysfunction caused by a deafness-associated tRNAHis mutation

The deafness-associated m.12201T>C mutation affects the A5-U68 base-pairing within the acceptor stem of mitochondrial tRNAHis. The primary defect in this mutation is an alteration in tRNAHis aminoacylation. Here, we further investigate the molecular mechanism of the deafness-associated tRNAHis 12201T>C mutation and test whether the overexpression of the human mitochondrial histidyl-tRNA synthetase gene (HARS2) in cytoplasmic hybrid (cybrid) cells carrying the m.12201T>C mutation reverses mitochondrial dysfunctions. Using molecular dynamics simulations, we demonstrate that the m.12201T>C mutation perturbs the tRNAHis structure and function, supported by decreased melting temperature, conformational changes, and instability of mutated tRNA. We show that the m.12201T>C mutation-induced alteration of aminoacylation tRNAHis causes mitochondrial translational defects and respiratory deficiency. We found that the transfer of HARS2 into the cybrids carrying the m.12201T>C mutation raises the levels of aminoacylated tRNAHis from 56.3 to 75.0% but does not change the aminoacylation of other tRNAs. Strikingly, HARS2 overexpression increased the steady-state levels of tRNAHis and of noncognate tRNAs, including tRNAAla, tRNAGln, tRNAGlu, tRNALeu(UUR), tRNALys, and tRNAMet, in cells bearing the m.12201T>C mutation. This improved tRNA metabolism elevated the efficiency of mitochondrial translation, activities of oxidative phosphorylation complexes, and respiration capacity. Furthermore, HARS2 overexpression markedly increased mitochondrial ATP levels and membrane potential and reduced production of reactive oxygen species in cells carrying the m.12201T>C mutation. These results indicate that HARS2 overexpression corrects the mitochondrial dysfunction caused by the tRNAHis mutation. These findings provide critical insights into the pathophysiology of mitochondrial disease and represent a step toward improved therapeutic interventions for mitochondrial disorders.

Furthermore, HARS2 overexpression markedly increased mitochondrial ATP levels and membrane potential and reduced production of reactive oxygen species in cells carrying the m.12201T>C mutation. These results indicate that HARS2 overexpression corrects the mitochondrial dysfunction caused by the tRNA His mutation. These findings provide critical insights into the pathophysiology of mitochondrial disease and represent a step toward improved therapeutic interventions for mitochondrial disorders.
Mitochondrial DNA (mtDNA) 2 -dependent defects have been associated with hearing deficit, either the nonsyndromic form (where hearing loss is the only obvious medical problem) or the syndromic form (hearing loss with other medical problems, such as diabetes) (1)(2)(3)(4)(5). The majority of these deafnesslinked mtDNA mutations are located in the mitochondrial 12S rRNA and tRNA genes (1)(2)(3)(4)(5)(6). The m.1555AϾG and m.1494CϾT mutations in the 12S rRNA gene have been associated with both aminoglycoside-induced and nonsyndromic deafness in many families worldwide (1,2,(7)(8)(9). Mitochondrial tRNAs are the hot spots for mutations associated with hearing loss (3,10,11). The syndromic deafness-associated mtDNA mutations are the MELAS-associated tRNA Leu(UUR) 3243AϾG mutation (12) and MERRF-associated tRNA Lys 8344AϾG mutation (13) and MDD-associated tRNA Glu 14692AϾG (14), whereas the nonsyndromic deafness-associated mtDNA mutations included the tRNA Ser(UCN) 7445AϾG, 7505TϾC, and 7511TϾC; tRNA His 12201TϾC; tRNA Asp 7551AϾG; and tRNA Ile 4295AϾG mutations (15)(16)(17)(18)(19)(20). These tRNA mutations have structural and functional consequences, including the processing of the tRNA from the primary transcripts, stability of the folded secondary structure, the charging of the tRNA, or the codon-anticodon interaction in the process of translation (14 -19). Of these, the m.12201TϾC mutation resided at the uridine at position 68 (U68) and abolished a base-pairing (5A-68U) on the acceptor stem of the tRNA His that may play an important role in the stability and identity of tRNA (18,21,22) (Fig. 1A). The primary defect in this mutation was the aberrant aminoacylation of the tRNA His (18). The deficient aminoacylation of tRNA His mainly contributed to a shortage of tRNA His , thereby causing the reduced rate of mitochondrial protein synthesis and respiration defects (18).
A specific cognate amino acid is charged or aminoacylated to each tRNA catalyzed by aminoacyl tRNA synthetase (10,11). Therefore, the deficient aminoacylation of tRNAs may be restored by the transfer of aminoacyl tRNA synthetase in cells carrying the tRNA mutation. In the previous studies, the overexpression of human mitochondrial leucyl-tRNA synthetase in the cybrid cells carrying the tRNA Leu(UUR) 3243AϾG mutation improved the efficiency of aminoacylation and stability of mitochondrial tRNA Leu(UUR) and oxidative phosphorylation (23)(24)(25)(26). The aberrant tRNA Ala and tRNA Val metabolisms were restored by overexpression of AARS2 and VARS2 in the cells carrying the tRNA Ala 5565AϾG or tRNA Val 1624CϾT mutation, respectively (27,28). In fact, human mitochondrial histidyl-tRNA synthetase is a highly conserved enzyme composed of 506 amino acids with a mitochondrial signal sequence (29). Thus, it is anticipated that the overexpression of human HARS2 in the cybrid cells carrying the m.12201TϾC mutation would improve the aminoacylation capacity of tRNA His , enhance the stability of tRNA, and then increase the rates of mitochondrial translation and respiration, consequently correcting the mitochondrial dysfunction. To test this hypothesis, stable transfectants were constructed by transferring a human HARS2 cDNA into a cybrid cell line carrying the m.12201TϾC mutation and a control cybrid cell line harboring the WT version of tRNA His . Human HARS2 was further characterized by examining subcellular locations. These stable transfectants were analyzed for the aminoacylation capacity of tRNAs, the stability of the tRNA His , the rates of mitochondrial translation and respiration, the levels of mitochondrial ATP, and the mitochondrial membrane potential (⌬⌿ m ) as well as the production of reactive oxygen species (ROS).

MD simulation analyses
To assess the impact of m.12201TϾC mutation on the tertiary structure of tRNA His , we carried out the molecular dynamics simulation using the acceptor stems (15 nt) of both WT and mutant tRNA His by the 100-ns all-atom method. As shown in Fig. 1B, the root mean square deviation (RMSD) curve of the mutated acceptor stem fluctuated more significantly than those of the WT counterpart, suggesting that the mutated acceptor stem exhibited more instability than its WT counterpart. Using the cpptraj program Amber14, the U68 formed a canonical base pair with A5 in the WT tRNA His through hydrogen bonds with an occupancy of 47 and 64%, respectively (Fig. 1, C and D). In contrast, the newly formed interaction between A5 and C68 in the mutant tRNA His was decreased to one hydrogen bond with a lower occupancy of 38%. These data implied that the mutant tRNA His molecule with the mismatch of 5A and C68 may be less stable than those in the WT counterpart.

Aberrant stability and conformation of tRNA His
To experimentally test whether there was an effect of the m.12201TϾC mutation on the stability of tRNA His , we measured the melting temperatures (T m ) of WT and mutant tRNA His transcripts by calculating the derivatives of absorbance against a temperature curve. As shown in Fig. 2A, the T m values of WT (U68) and mutant (C68) transcripts were 47.0 Ϯ 1.0 and 43.0 Ϯ 1.7°C, respectively. These data were in a good agreement with data of molecular stimulation, indicating that the m.12201TϾC mutation led to the instability of tRNA His .
These transcripts were then assessed for conformational change by PAGE analysis under denaturing and native conditions. As shown in Fig. 2B, electrophoretic patterns showed that the mutant (C68) tRNA His transcript migrated faster than the WT (U68) tRNA His transcript under native conditions. However, there were no differences in the migration pattern between WT (U68) and mutant (C68) tRNA His transcripts under denaturing conditions. To further test whether the m.12201TϾC mutation affected the conformation of tRNA His in vivo, total RNAs isolated from mutant and control cybrids were electrophoresed through 10% native polyacrylamide gel in Tris-glycine buffer and then electroblotted onto a positively charged nylon membrane for hybridization analysis with digoxigenin (DIG)-labeled oligodeoxynucleotide probes for tRNA His and tRNA Leu(UUR) , respectively. As shown in Fig. 2C, electrophoretic patterns showed that the tRNA His in mutant cybrids carrying the m.12201TϾC mutation migrated much faster than control cybrids lacking this mutation.
Furthermore, the probes specific for tRNA His , tRNA Leu(UUR) , and cytosol-tRNA His were validated using total RNAs isolated from mutant (E1) and control (C9) cybrids as well as 143B cells and derived mtDNA-less o 206 cell lines. As shown in Fig. 2D, mitochondrial tRNA His and tRNA Leu(UUR) were only absent in 0 206 cells, whereas the ct-tRNA His were present in the 0 206 as well as 143B, C9, and E1 cells. These data suggested that the structural alterations by the m.12201TϾC mutation caused the conformational change of tRNA His .

Subcellular location of human HARS2
To determine the subcellular localization of human HARS2, pEGFP-N1-HARS2 expressing the HARS2-GFP fusion protein was transfected into the 143B cell line. Fig. 3A shows that the immunofluorescence pattern of transfected 143B cells was double-labeled with an mAb specific for the GFP and Mitotrack probes, which contain a mildly thiol-reactive chloromethyl moiety for labeling mitochondria. A typical mitochondrial staining pattern was observed, and superimposition of two panels showed the complete overlap of two patterns, demonstrating that human HARS2 localizes exclusively at mitochondria.

Construction of stable transfectants expressing the human HARS2
A 1.6-kb human HARS2 cDNA expressed in a pCDH-puro vector or the vector only was transfected into the mutant cybrid tRNA synthetase transfer rescues mitochondrial dysfunctions cell line (E1) carrying the m.12201TϾC mutation and control cell line (C9) lacking this mutation (18). These stable transfectants were isolated by culturing cells in DMEM supplemented with 1 g/ml puromycin and 10% FBS for 2 weeks.
The expression levels of the HARS2 cDNA in resultant stable transfectants were examined by Western blot analysis, as shown in Fig. 3B. The levels of exogenous HARS2 in transfectants C9 and E1 were more than 8-fold higher than those of our own HARS2. These four transfectants [C9V (vector only), C9H (exogenous HARS2), E1V (vector only), and E1H (exogenous HARS2)] were then used for further characterization.
To test whether HARS2 overexpression affected the expression of other mitochondrial aminoacyl-tRNA synthetases, the levels of LARS2 (leucyl-tRNA synthetase 2) and SARS2 (seryl-tRNA synthetase 2) were determined by Western blot analysis. As illustrated in Fig. 3B, the levels of LARS2 and SARS2 in C9H and E1H cell lines were comparable with those in four other cell lines lacking the exogenous HARS2 expression. This suggested that the overexpression of HARS2 did not affect the expression levels of these mitochondrial aminoacyl-tRNA synthetases.

Enhancing aminoacylation capacity of mitochondrial tRNA His
We investigated the effects of the transfer of human HARS2 on the aminoacylation of tRNA His in various cell lines ex vivo. The aminoacylation levels of tRNA His in these cell lines were determined by using electrophoresis in an acid polyacrylamide/ urea gel system to separate uncharged and charged tRNA species, electroblotting and hybridizing with specific probes for tRNA His as well as tRNA Leu(UUR) , tRNA Met , tRNA Ile , and tRNA Ser(AGY) , respectively. To further distinguish nonaminoacylated tRNA from aminoacylated tRNA, samples of mitochondrial tRNAs were deacylated by being heated for 10 min at 60°C at pH 8.3 and then run in parallel (30,31). As shown in Fig.  4, the overexpression of HARS2 increased the aminoacylated  (18,22). Arrow, location of the m.12201TϾC mutation. Nucleotides in the dashed box in the acceptor stem of tRNA Ser(UCN) were used for MD simulation analysis. B, time evolution of the root mean square deviation (RMSD) values of all backbone atoms on the acceptor stem for the WT (black lines) and mutant (MT) (red lines) of tRNA His . C, superimposed models of tertiary structures of acceptor stems of WT (gray) and mutant (steel blue) tRNA His . The sequences of mutant and WT models were shown in regular mode, respectively. D, hydrogen bonds (black dashes) between A5 and U68 in WT (gray) or A5 and C68 in mutant tRNA His (steel blue). The occupancies of hydrogen bonds are shown in red. tRNA synthetase transfer rescues mitochondrial dysfunctions levels of tRNA His mutant cell line E1 but not those of control cell line C9. In particular, the proportions of aminoacylated tRNAs in the E1H cell lines were 75.0, 67.5, 61.6, 43.6, and 47.9% in the tRNA His , tRNA Leu(UUR) , tRNA Met , tRNA Ile , and tRNA Ser(AGY) , respectively, whereas 56.3% of tRNA His , 67.6% of tRNA Leu(UUR) , 59.6% of tRNA Met , 44.3% of tRNA Ile , and 47.8% of tRNA Ser(AGY) were aminoacylated in the parental E1 cell lines. In contrast, there were no significant differences in the aminoacylated levels of these five tRNAs between the C9H cell lines and parental C9 cell lines.

The increasing steady-state levels of tRNAs
To evaluate whether the overexpression of HARS2 enhanced the levels of tRNA His in transfectants carrying the m.12201TϾC mutation, we subjected mitochondrial RNAs from various cell lines to Northern blot analysis and hybridized . The calculations were based on three independent determinations. The graph shows the results of a representative experiment. B, assessment of conformational changes by PAGE analysis under native and denaturing conditions. The transcripts of WT and MT tRNA His were electrophoresed through native or denaturing polyacrylamide gel stained with ethidium bromide. C, Northern blot analysis of tRNAs under native conditions. D, Northern blot analysis of tRNAs from various cell lines under denatured conditions. Five micrograms of total cellular RNAs from mutant and control cell lines were electrophoresed through native polyacrylamide gel, electroblotted, and hybridized with DIG-labeled oligonucleotide probes specific for the tRNA His , tRNA Leu(UUR) , and ct-RNA His , respectively.

Figure 3. Subcellular localization of human HARS2 in 143B cells.
A, cells were transiently transfected with a HARS2 cDNA fused with GFP. The fusion protein was visualized by indirect immunofluorescence using antibodies to GFP. Mitotracker Red-stained mitochondria and 4Ј,6-diamidino-2-phenylindole-stained nuclei were identified by red and blue fluorescence respectively. Scale bars, 50 m. B, Western blot analysis of mitochondrial aminoacyl-tRNA synthetases in these four transfectants (C9V (vector only), C9H (exogenous HARS2), E1V (vector only), and E1H (exogenous HARS2)) and their parental cell lines E1 and C9. Twenty micrograms of total cellular proteins from various cell lines were electrophoresed through a denaturing polyacrylamide gel, electroblotted, and hybridized with HARS2, LARS2, and SARS2, respectively, and with TOM20 as a loading control.

tRNA synthetase transfer rescues mitochondrial dysfunctions
them with a DIG-labeled oligodeoxynucleotide probe specific for tRNA His , five other tRNAs and nucleus-encoded 5S rRNA for normalization (32)(33)(34). For comparison, the average level of each tRNA in control or mutant cell lines was normalized to the average levels in the same cell line for reference 5S rRNA. As shown in Fig To investigate whether there was an effect of m.12201TϾC mutation on the half-life of tRNA His , we treated the various cell lines with EtBr (250 ng/ml) to block mitochondrial RNA synthesis (35) and extracted total cellular RNA at various time points after the addition of the drug. We then subjected mitochondrial RNAs from various cell lines to Northern blot analysis and hybridized them with a DIG-labeled oligodeoxynucleotide probe specific for tRNA His and 5S rRNA as a loading control. As shown in Fig. 5C, there were no significant differences in the half-lives of tRNA His between transfectants and parental mutant and cell lines. This suggested that m.12201TϾC mutation and overexpression of HARS2 did not have a significant effect on the half-life of tRNA His .

Northern blot analysis of mRNA and rRNAs
We then examined whether the m.12201TϾC mutation affected the expression/processivity of polycistronic transcripts. RNA transfer hybridization experiments were performed with total cellular RNAs from various mutant and control cell lines, using a set of DIG-labeled RNA probes: ND6 from L-strand transcripts; COX1, COX2, CYTB, 12S rRNA, and 16S rRNA from H-strand transcripts (32,33); and ␤-actin as a control, respectively. As shown in Fig. 6, the levels of ND6, COX1, COX2, CYTB, 12S rRNA, and 16S rRNA in mutant cell lines, normalized with respect to those of ␤-actin mRNA, were comparable with those in the control cell lines. These results indicated that the m.12201TϾC mutation did not affect the expression of these mRNAs and rRNAs.

Elevated levels of mtDNA-encoding proteins
To assess whether the overexpression of HARS2 affected the mitochondrial protein synthesis, a Western blot analysis was performed to examine the levels of eight mtDNA-encoding proteins in transfectants as well as their parental cell lines with Tom20 as a loading control. As shown in Fig. 7 (A and B), the overexpression of HARS2 significantly increased the levels of these mitochondrial proteins in the mutant cell line E1. Of these, the levels of ND1, ND4, ND5, and ND6 (subunits 1, 4, 5, and 6 of NADH dehydrogenase (complex I)); CYTB (cytochrome b of ubiquinone cytochrome c oxidoreductase (complex III)); CO2 (subunit 2 of cytochrome c oxidase (complex IV)); and ATP6 and ATP8 (subunits 6 and 8 of the H ϩ -ATPase (complex V)) in E1H cell lines were 66  To test whether the m.12201TϾC-induced deficiency affected the mitochondrial proteostasis, we measured the levels of Clpp involved in mitochondrial ribosome assembly (36) and ATP family gene 3-like 2 (Afg3l2) proteases involved in the turnover of misfolded proteins markers for proteostasis stress (37), in the various cell lines. As shown in Fig. 7C and Fig. S1, there was no significant difference in the levels of Afg3l2 and Clpp among mutant and control cell lines. These data indicated that m.12201TϾC mutation may not affect the proteostasis stress.

The restoration of respiration deficiency
To evaluate whether the overexpression of HARS2 rescued the respiratory deficiency caused by the m.12201TϾC mutation, we measured the activities of respiratory complexes by the use of isolating mitochondria of various cell lines. Complex I activity was measured by following the oxidation of NADH with ubiquinone as the electron acceptor (38,39). The activity of complex II (succinate ubiquinone oxidoreductase) exclusively encoded by the nuclear DNA was examined by the artificial electron acceptor DCPIP (40,41). Complex III (ubiquinone cytochrome c oxidoreductase) activity was measured as the reduction of cytochrome c (III) using D-ubiquinol-2 as the electron donor. The activity of complex IV was monitored by following the oxidation of cytochrome c (II). As shown in Fig. 8, the overexpression of HARS2 enhanced the activities of complex I, III, and IV but not complex II in cells carrying the m.12201TϾC mutation. As shown in Table S1, the activities of complexes I, III, and IV in the E1H cell line were 150.2, 145.3, and 209.3% of those in the parental cell line E1, respectively. By contrast, the activities of complexes I, III, and IV in the C9H cell line were 89.4, 107.6, and 139.5% of those in the parental C9 cell line, respectively.
Oxygen consumption rate (OCR) is an indicator of mitochondrial respiration. Using a Seahorse Bioscience XF-96 extracellular flux analyzer, we can measure mitochondrial respiratory control, including basal respiration, O 2 consumption

tRNA synthetase transfer rescues mitochondrial dysfunctions
attributed to ATP production, proton leak, maximum respiratory rate, reserve capacity, and nonmitochondrial respiration (42,43). As shown in Fig. 9, the basal OCR in the E1H cell line increased 47.9% compared with those in the parental cell line E1, whereas the basal OCR of C9H was comparable with those in the C9 cell line. The drug-insensitive OCRs were then measured after the sequential addition of oligomycin (to inhibit the ATP synthase), carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) (to uncouple the mitochondrial inner membrane and allow for maximum electron flux through the ETC), rotenone (to inhibit complex I), and antimycin A (to inhibit complex III). As shown in Fig. 9, the ATP-linked OCR, maximal OCR and reserve capacity OCR in the E1H cell line significantly increased 67.4, 89.5, and 144.7%, as compared with those in the parental cell line E1, respectively. However, basal OCR, ATPlinked OCR, maximal OCR, and reserve capacity OCR in the E1H cell line were significantly lower than those in the C9H cell line. By contrast, the OCRs for C9H cell line were comparable with those of the parental C9 cell line. These data indicated that the respiration of the E1H cell line was significantly improved but not restored to the levels of WT cell line.

Increasing levels of mitochondrial ATP
To further examine the effect of overexpression of HARS2 on the oxidative phosphorylation, we used the luciferin/luciferase assay to measure the levels of cellular and mitochondrial ATP. Populations of cells were incubated in the media in the presence of glucose and of 2-deoxy-D-glucose with pyruvate (18). As shown in Fig. 10, the levels of mitochondrial ATP (the presence of pyruvate and 2-deoxy-D-glucose to inhibit the glycolysis) in the E1H cell line were ϳ133% of those in the parental cell line E1, whereas the levels of mitochondrial ATP in the C9H cell line were 103% of those in the parental C9 cell line, respectively. On the contrary, the overexpression of HARS2 did not significantly change the levels of total cellular ATP (the presence of glucose) in the mutant and control cell lines.

Enhancement of mitochondrial membrane potential
The mitochondrial membrane potential (⌬⌿ m ) generated by complexes I, III, and IV is an essential component in the process of energy storage during oxidative phosphorylation (44). The ⌬⌿ m levels in various cell lines were measured using a fluorescence probe JC-10 assay via flow cytometry. As shown in Fig. 11, the ⌬⌿ m of the C9V, C9H, E1, E1V, and E1H cell lines was 101.0, 93.5, 53.6, 60.4, and 86.5% relative to the mean values in the control cell line C9, respectively. In contrast, the populations with a normal level of ⌬⌿ m in these six cell lines were comparable with those in the presence of FCCP.

Overexpression of HARS2 reduced the production of mitochondrial ROS
The levels of mitochondrial ROS among these cells were determined using a MitoSOX assay via flow cytometry (45). Geometric mean intensity was recorded to measure and delineate the rate of ROS of each sample. As shown in Fig. 12 (A and  B), the levels of mitochondrial ROS production in the cell lines C9V, C9H, E1, E1V, and E1H were 90.5, 87.0, 163.3, 147.7, and 109.2% relative to the mean values in the control cell line C9, respectively.
To test whether the m.12201TϾC mutation-induced mitochondrial ROS production affected the antioxidant systems, we examined the levels of three antioxidant enzymes, SOD2 in the mitochondrion and SOD1 and catalase in the cytosol (46), in the various cell lines. As shown in Fig. 12C, the mutant cell lines  E1 and E1V revealed marked increases in the levels of SOD2 and mild increases in the levels of SOD1 and catalase, as compared with those in the WT cell line C9. Notably, overexpression of HARS2 in the mutant cell line led to pronounced reductions in the level of SOD2 but relatively mild decreases in the levels of SOD1 and catalase.

Discussion
The objective of this study was to further elucidate the molecular pathogenesis of the deafness-associated tRNA His 12201TϾC mutation and to test whether human HARS2 overexpression in the cybrid cells carrying the m.12201TϾC mutation reverses the mitochondrial dysfunctions. The m.12201TϾC mutation destabilized the canonical A5-U68 base-pairing within the aminoacyl acceptor stem of this mitochondrial tRNA His (18,21,47,48). Thus, we hypothesized that the m.12201TϾC mutation led to the structural and functional effects on this tRNA, including the alteration of conformation, thermal stability, and aminoacylation. MD studies indicated that the U68 formed a canonical base pair with A5 of WT tRNA His through hydrogen bonds with an occupancy of 47 and 64%, respectively, whereas the newly formed interaction between A5 and C68 in the mutant tRNA His was decreased to one hydrogen bond with a lower occupancy of 38% (49). These findings indicated that the mutant tRNA His molecule with the mismatch of 5A and C68 may be less stable than those in the WT counterpart with the canonical A5-U68 base-pairing. In fact, the T m in mutant tRNA His molecule was 4°C lower than those in the WT counterpart. The instability of the mutant tRNA molecule was further evidenced by the drastically reduced level of tRNA His in the mutant cell lines carrying the  tRNA synthetase transfer rescues mitochondrial dysfunctions m.12201TϾC mutation (18). Furthermore, the m.12201TϾC mutation caused the conformational change of tRNA His , as suggested by faster electrophoretic mobility of mutated tRNA with respect to the WT molecule in vitro or ex vivo, consistent with the conformational changes of tRNAs carrying the m.4435AϾG and m.3253TϾC mutations (50,51). However, the m.12201TϾC mutation did not affect the expression/processivity of polycistronic transcripts, in contrast with the aberrant processing of the polycistronic transcripts observed in the cell lines carrying the tRNA Leu(UUR) 3243AϾG and tRNA Ser(UCN) 7445AϾG mutations and the m.4401AϾG mutation in the precursor of tRNA Met /tRNA Gln mutations (15,24,52). Moreover, the abolishment of A5:U68 base-pairing substitution may affect the tRNA His interaction with mitochondrial histidyl-tRNA synthetase, thereby altering the aminoacylation properties of tRNA His by either charging inefficiently or mischarging with mitochondrial histidyl-tRNA synthetase (53,54). In this study, the cell line bearing the m.12201TϾC mutation displayed increasing levels of aminoacylated tRNA His and faster electrophoretic mobility of mutated tRNA with respect  ) were added at the indicated times to determine different parameters of mitochondrial functions. B, graphs present the ATP-linked OCR, proton leak OCR, maximal OCR, reserve capacity, and nonmitochondrial OCR in six cell lines. Nonmitochondrial OCR was determined as the OCR after rotenone/antimycin A treatment. Basal OCR was determined as OCR before oligomycin minus OCR after rotenone/antimycin A. ATP-lined OCR was determined as OCR before oligomycin minus OCR after oligomycin. Proton leak was determined as basal OCR minus ATP-linked OCR. Maximal was determined as the OCR after FCCP minus nonmitochondrial OCR. Reserve capacity was defined as the difference of maximal OCR after FCCP minus basal OCR. The data were based on three determinations for each cell line. Graph details and symbols are explained in the legend to Fig. 4. Error bars, S.D.

tRNA synthetase transfer rescues mitochondrial dysfunctions
to the WT cell line (18). The mutant tRNA His may be metabolically less stable and subject to turnover, thereby lowering the steady-state level of tRNA His . As a result, a failure in tRNA metabolism is responsible for defective mitochondrial protein synthesis and an impaired oxidative phosphorylation and increasing production of oxidative reactive species (18,48). The resultant mitochondrial dysfunction would lead to the dysfunction or death of cochlear cells, thereby producing a phenotype of hearing loss.
In the present investigation, we have shown that the overexpression of human HARS2 in the cybrid cells carrying the m.12201TϾC mutation corrected the mitochondrial dysfunctions. In particular, the overexpression of HARS2 in the mutant cybrids bearing the m.12201TϾC mutation raised the level of aminoacylated tRNA His from 56.3 to 75.0% but did not change the aminoacylation levels of other tRNAs. Furthermore, overexpression of HARS2 did not increase significantly the level of aminoacylated tRNA His in the WT cybrids, suggesting that 50% of aminoacylated tRNA His in cells appeared to be the maximum threshold level to maintain the normal function, as in the case of overexpression of LARS2 in the WT cybrid cell line HSI (24). The increasing proportions of aminoacylated tRNA led to the elevating steady-state level of tRNA His in the cybrids harboring the m.12201TϾC mutation from 41.8 to 55.9% of those in the control cybrids lacking the mtDNA mutation. The increasing levels of mutant tRNA His were likely due to the improvement of stability and function of mutant tRNA His . It is worthwhile to note that overexpression of HARS2 also enhanced the steadystate levels of noncognate tRNAs, such as tRNA Ala , tRNA Gln , tRNA Glu , tRNA Leu(UUR) , tRNA Lys , and tRNA Met , but did not affect the levels of mRNAs and rRNAs in cells bearing the m.12201TϾC mutation. In fact, overexpression of LARS2 increased the levels of mutated tRNA Val in the cells in the cybrid cell line carrying the m.1624CϾT mutation (26). These findings suggested that the overexpression of HARS2 may mediate the noncognate tRNA metabolisms in cells carrying the pathogenic tRNA mutation. The improved efficiency of aminoacylation and stability of mitochondrial tRNAs by overexpression of HARS2 resulted in ϳ80% increasing levels of mitochondrial translation in the cybrid cell line harboring the m.12201TϾC mutation. The restoration of deficient mitochondrial translation by overexpression of HARS2 appeared to be more efficient than those in myoblasts carrying the m.3243AϾG mutation by overexpression of EFTu and EFG2 (55). The facilitated synthesis of these mtDNA-encoding polypeptides improved the respiration capacity. In particular, the enhancing levels of OCR and mitochondrial ATP in the cybrid bearing the m.12201TϾC mutation by overexpression of HARS2 were correlated well with the increasing levels of mitochondrial protein, suggesting that the improvement of mitochondrial translation was responsible for the ameliorated oxidative phosphorylation deficiency. Furthermore, the elevating activities of respiratory chain complexes caused by overexpression of HARS2 yielded ϳ61.4% increase of ⌬⌿ m in the cell line bearing m.12201TϾC mutation. Indeed, ⌬⌿ m reflects the pumping of hydrogen ions across the inner membrane during the process of electron transport and oxidative phosphorylation (44). The improvement of both OXPHOS and ⌬⌿ m would reduce the production of ROS and expression of SOD2 in mutant cells carrying the m.12201TϾC mutation. The lower production of ROS can reduce a vicious cycle of oxidative stress in the mitochondria, thereby decreasing the damage of mitochondrial and cellular proteins, lipids, and nuclear acids (56).
In summary, our results demonstrated that overexpression of HARS2 corrected the mitochondrial dysfunction caused by the deafness-associated tRNA His 12201TϾC mutation. The biochemical phenotypes manifested by interplay between tRNA His 12201TϾC mutation and HARS2 gene may provide new insights into the pathophysiology of maternally inherited deafness. The restoration of m.12201TϾC mutation-induced mitochondrial dysfunctions by overexpression of HARS2 may be a step toward therapeutic interventions for these disorders.

Cell lines and culture conditions
The 143B.TK Ϫ cell line, the mutant cybrid cell lines (E1) carrying the m.12201TϾC mutation, and control cybrid cell lines (C9) belonging to the same mtDNA haplogroup Z3 but lacking the mutation (H7) were grown in DMEM (containing 4.5 mg of glucose and 0.11 mg of pyruvate per ml), supple-

tRNA synthetase transfer rescues mitochondrial dysfunctions
mented with 5% FBS (18). The mtDNA-less °206 cell line, derived from 143B.TK Ϫ (57) was grown under the same conditions as the parental line, except for the addition of 50 g of uridine/ml. The stable transfectants were grown in DMEM supplemented with 10% FBS and 1 g/ml puromycin.

tRNA synthetase transfer rescues mitochondrial dysfunctions Molecular dynamics simulation procedure
The acceptor stem of tRNA His containing 15 nucleotide bases spanning the A5-68U/C base-pairing (Fig. 1A) was constructed as the model for simulations. The initial coordinates for WT were extracted from the crystal structure of human mitochondrial tRNA Phe -PheRS complex (Protein Data Bank entry 3TUP). The coordinates of the backbones were maintained, and nucleotide bases were substituted with the mitochondrial tRNA His sequence by Chimera (58). In the mutated structure, the coordinates of U to C substitution at position 68 were generated by Chimera using the established WT model.
The acceptor stems of WT and mutant tRNA His were simulated using the same methods. The ff14SB force field parameters were employed in MD simulations with Amber14 (59). The initial models were surrounded by TIP3P water molecules and ions. The simulation systems were maintained in a solution containing 50 mM NaCl, made by the addition of 3Na ϩ and 3Cl Ϫ , and then 13 Na ϩ for neuralization. The SHAKE algorithm was used to constrain hydrogen atoms (60). Alternately, we minimized and equilibrated the whole system to relieve all unfavorable interactions of the initial model. The procedure of equilibration in NVT ensemble was heated to 300 K and was not subjected to pressure. NPT ensemble was applied to equilibrate the solvent at the final equilibration step, in which the temperature was kept at 300 K and pressure at 1 bar in periodic boundary conditions. Subsequently, 100-ns production simulations were performed for both systems with a time step of 2 fs. The MD trajectory of each system was observed with VMD (61). Extraction and analysis of trajectories were performed using the cpptraj program in Amber14. Distance cutoff for hydrogen bond and angle cutoff were set to 3.0 Å and 135°, respectively.

Measurement of melting temperature
UV melting assays were carried out as described previously (50,51). The WT and mutant tRNA His transcripts were produced using in vitro transcription by T7 RNA polymerase according to previous protocols (62). The tRNA His transcripts were dissolved in 50 mM sodium phosphate buffer (pH 7.0), containing 50 mM NaCl, 5 mM MgCl 2 , and 0.1 mM EDTA. Absorbance against melting temperature curves was measured at 260 nm with a heating rate of 1°C/min from 25 to 95°C via an Agilent Cary 100 UV spectrophotometer.

Isolation of human HARS2 cDNA
To construct the plasmid pHARS2 containing the entire coding region of HARS2 cDNA, RT-PCR was performed by using TaqDNA polymerase (Promega) and total RNA isolated from 143B cells as template, with the primers 5Ј-ATAATGTCTGA-CCCGCCTCCTT-3Ј (nt 159 -176) and 5Ј-CCTTGGTGGCT-AGTTGTTGTGA (nt 1705-1720) (GenBank TM accession no. NM_001278732.1). The predominant PCR product was purified by agarose gel electrophoresis and subsequently cloned into a pGEM-T vector (Promega). Nucleotide sequence was determined by Sanger sequencing.

Subcellular localization of human HARS2
The coding region of HARS2 cDNA lacking its natural stop codon was obtained by PCR using pHARS2 cDNA as the template. Primers 5Ј-CCGCTCGAGCCGGCGTCCTGCCGC (nt 204 -220) and 5Ј-CCGACCGGTGATCCAGACTCAGACAG (nt 2870 -2891) were used for the PCR amplification. PCR products were digested with AgeI and XhoI and then cloned into pEGFP-N1. After sequence determination, Resultant constructs were transfected into 143B cells using the jetPRIME TM transfection reagent (Polyplus Transfection) according to the manufacturer's protocol. Immunofluorescence analysis was performed as detailed elsewhere (63,64).

Construction of stable transfectants
The insert of pHARS2 was subcloned into pCDH-puro (Invitrogen). The resultant constructs or vector only were transfected into E1 and C9 cell lines using the jetPRIME TM transfection reagent (Polyplus Transfection) according to the manufacturer's protocol. The stable transfectants were isolated by culturing cells in DMEM supplemented with 1 g/ml puromycin and 10% FBS for 2 weeks. The resultant clones were examined for the expression of HARS2 by Western blot analysis.
For the tRNA mobility shift assay, 2 or 5 g of RNAs were electrophoresed through a 10% polyacrylamide native gel at 4°C with 50 mM Tris-glycine buffer. After electrophoresis, the gels were treated according to the Northern blot analysis as described above (50,51).
The aminoacylation assays including the isolation of total cellular and gel electrophoresis were as detailed elsewhere (30,45). The gels were then electroblotted onto a positively charged nylon membrane (Roche Applied Science) for the hybridization analysis with oligodeoxynucleotide probes as described above. Quantification of the density in each band was performed as detailed previously (30,45).
tRNA half-life measurements were performed as detailed elsewhere (35). Briefly, various cell lines were incubated in fresh medium containing 250 ng/ml EtBr for the times indicated in Fig. 5C. Ten micrograms of total cellular RNAs, extracted as above, were subjected to Northern blot analysis as detailed above.

Enzymatic assays
The enzymatic activities of complexes I-IV were measured as detailed elsewhere (38 -40, 67).

Measurements of oxygen consumption
The rates of oxygen consumption in lymphoblastoid cell lines were assayed with a Seahorse Bioscience XF-96 extracellular flux analyzer (Seahorse Bioscience), as detailed previously (42,43,67). The protein content of each well was then measured to normalize OCR values.

Measurements of ATP levels
The Cell Titer-Glo luminescent cell viability assay kit (Promega) was used for the measurement of cellular and mitochondrial ATP levels, following the modified manufacturer's instructions (18,51).

Assessment of mitochondrial membrane potential
Mitochondrial membrane potential was assessed with the JC-10 assay kit-microplate (Abcam) according to the manufacturer's general recommendations with some modifications, as detailed elsewhere (44,51).

Statistical analysis
Statistical analysis was performed by the analysis of variance test contained in the StatView program SAS (version 9.4) (SAS Institute) and entering individual replicate values. Unless indicated otherwise, a p value Ͻ 0.05 was considered statistically significant.