Mitochondrial tRNAGln 4394C>T Mutation May Contribute to the Clinical Expression of 1555A>G-Induced Deafness

The mitochondrial 1555A>G mutation plays a critical role in aminoglycoside-induced and non-syndromic hearing loss (AINSHL). Previous studies have suggested that mitochondrial secondary variants may modulate the clinical expression of m.1555A>G-induced deafness, but the molecular mechanism has remained largely undetermined. In this study, we investigated the contribution of a deafness-associated tRNAGln 4394C>T mutation to the clinical expression of the m.1555A>G mutation. Interestingly, a three-generation family with both the m.1555A>G and m.4394C>T mutations exhibited a higher penetrance of hearing loss than another family harboring only the m.1555A>G mutation. At the molecular level, the m.4394C>T mutation resides within a very conserved nucleotide of tRNAGln, which forms a new base-pairing (7T-66A) and may affect tRNA structure and function. Using trans-mitochondrial cybrid cells derived from three subjects with both the m.1555A>G and m.4394C>T mutations, three patients with only the m.1555A>G mutation and three control subjects without these primary mutations, we observed that cells with both the m.1555A>G and m.4394C>T mutations exhibited more severely impaired mitochondrial functions than those with only the m.1555A>G mutation. Furthermore, a marked decrease in mitochondrial RNA transcripts and respiratory chain enzymes was observed in cells harboring both the m.1555A>G and m.4394C>T mutations. Thus, our data suggest that the m.4394C>T mutation may play a synergistic role in the m.1555A>G mutation, enhancing mitochondrial dysfunctions and contributing to a high penetrance of hearing loss in families with both mtDNA pathogenic mutations.


Introduction
Deafness is a common sensory defect, with an incidence of 2.8 in 1000 children [1]. This disease is related to single-gene mutations or environmental risk factors such as aminoglycoside antibiotics (AmAn). In fact,~50% of AINSHL cases have a genetic etiology, and >150 nuclear genes, together with >1200 mutations, have been confirmed to be associated with hearing loss (http://deafnessvariationdatabase.org/ (accessed on 11 September 2022) [2,3]. Mitochondrial DNA (mtDNA) mutations have been thought to be associated with both syndromic and nonsyndromic hearing loss [4,5]. Among these, the m.1555A>G and m.1494C>T mutations in the extremely conserved A-site of 12S rRNA have been found to be associated with AINSHL in many families worldwide [6][7][8]. Without the use of AmAn, individuals with the m.1555A>G mutation exhibit a variable degree of hearing impairment, from profound to normal [9,10]. Furthermore, functional characterization of cell lines derived from subjects with the m.1555A>G mutation shows that it confers a mild impairment of mitochondrial function and is sensitive to AmAn [11,12], suggesting that this mutation is insufficient by itself to form enough phenotypes, and that therefore a combination of environmental factors, nuclear genes and mitochondrial genetic variants In addition, pure-tone audiometry (PTA) was carried out in a sound-controlled room at frequencies ranging from 250 to 8000 Hz, as suggested in our recent study [24]. The levels of hearing loss were divided into five grades: normal: <20 decibels (dB); mild: 20-40 dB; moderate: 41-70 dB; severe: 71-95 dB; and profound >95 dB, according to the study as previously described [25].

mtDNA Analysis
The genomic DNA of the affected matrilineal relatives of these pedigrees was isolated  In addition, pure-tone audiometry (PTA) was carried out in a sound-controlled room at frequencies ranging from 250 to 8000 Hz, as suggested in our recent study [24]. The levels of hearing loss were divided into five grades: normal: <20 decibels (dB); mild: 20-40 dB; moderate: 41-70 dB; severe: 71-95 dB; and profound >95 dB, according to the study as previously described [25].

mtDNA Analysis
The genomic DNA of the affected matrilineal relatives of these pedigrees was isolated using Puregene DNA isolation kits (Biomega). 24 PCR primers spanning complete mtDNA genes were used to amplify the mitochondrial genomes, as suggested in a previous study [26]. The PCR products were purified and analyzed, and subsequently compared with an updated version of the human mitochondrial genome sequence to detect the mutations (GenBank Accessible No: NC_012920.1) [27,28].

Data Analyses
The ClustalW program (http://www.ebi.ac.uk/Tools/msa/clustalw2/ (accessed on 11 September 2022) was used to align the mtDNA sequences. A total of 14 species were selected for this analysis. The conservation index (CI) was measured by comparing the human mtDNA with that of other vertebrates. CI ≥ 75% was considered to have functional significance [29]. Moreover, the Phylotree (http://www.phylotree.org/ (accessed on 11 September 2022)) and East Asia phylogeny were used to determine the mtDNA haplogroups of these two pedigrees [21,30].

Screening for Nuclear Gene Mutations
To see the contributions of common nuclear modified genes (GJB2, GJB3, GJB6, SLC26A4 and TRMU) to m.1555A>G-induced deafness, we conducted mutational screening for these genes in two pedigrees (HZD055 and HZD510). PCR-Sanger sequencing was performed using the method described previously [31,32].

Generation of Cell Lines and Culture Conditions
Nine trans-mitochondrial cells which derived from three patients with both the m.1555A>G and m.4394C>T mutations (HZD510: II-6, II-8 and III-7), three individuals with only the m.1555A>G mutation (HZD055: II-8, III-4 and III-7) and three controls without these mtDNA mutations (C1, C2 and C3). These were constructed by introducing mitochondria from immortalized lymphoblastoid cell lines into human ρ • cells, which were generated from bromodeoxyuridine (BrdU)-resistant 143B cells, according to the protocols described elsewhere [33]. All cells were cultured in RPMI1640 medium (Invitrogen), supplemented with 10% cosmic calf serum.

Analysis of Mitochondrial Membrane Potential (MMP)
The levels of MMP in nine cybrid cells were evaluated using JC-1 dye (Life Technology, Carlsbad, CA, USA). At a low concentration, JC-1 existed as a monomer and was detected as green fluorescence, while at a high concentration, JC-1 existed as a multimer and was detected as red fluorescence [35]. To measure the MMP, JC-1 staining buffer was first added to the cells for 20 min. Subsequently, the solution was removed, and the cells were washed with JC-1 twice. After adding PBS per well, the fluorescence signal of the cells was observed under flow cytometry (BD Biosciences, Franklin Lakes, NJ, USA) [36].

Mitochondrial Reactive Oxygen Species (ROS) Analysis
A total of 2 × 10 6 cells were maintained in fluorescent probe 2 -7 dichlorofluorescin diacetate (DCFH-DA) with a final concentration of 10 µM at 37 • C for 0.5 h. Subsequently, the cells were washed twice with PBS and analyzed using a fluorescence plate reader (Millipore, Burlington, MA, USA) at 485 nm excitation and 535 nm emission [37].

mtDNA Quantification
Quantitative PCR (qPCR) was applied to quantify the mtDNA content. The mtDNA ND1 was used to determine the mtDNA copy numbers, and the nuclear β-globin gene was used as a control. The primer sequence for the amplification of the mt-ND1 was: forward: 5 -AACATACCCATGGCCAACCT-3 ; reverse: 5 -AGCGAAGGGTTGTAGTAGCCC-3 . The primer for the genetic amplification of the β-globin gene was: forward: 5 -GAAGAGCCAAGGACAGGTAC-3 ; reverse: 5 -CAACTTCATCCACGTTCACC-3 . The qPCR data were normalized to β-globin, then compared with the wild-type control values and analyzed using a relative quantification method (2 −∆∆CT ) [38].

Measurement of mt-RNA Transcription
The total RNA was isolated from nine cybrids using the TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, USA). In brief, 500 ng of RNA was used with a reverse transcription kit (Takara, Kusatsu, Shiga, Japan). Then, fluorogenic SYBR Green (Bio-Rad, Hercules, CA, USA) was used for qPCR, following the protocol suggested previously [39]. The primers for the qPCR amplification are listed in Table 1.

Assays of Activities of Respiratory Chain Complexes
The enzymatic activities of complexes I, II, III and IV were assayed as detailed elsewhere [16], and their activities were normalized by citrate synthase activity.

Qualification of 8-Hydroxy-2 -Deoxyguanosine (8-OHdG)
The presence of 8-OHdG is commonly regarded as an indicator of oxidative stress and mitochondrial dysfunction [40]. The plasma concentrations of 8-OHdG in three subjects with both the m.1555A>G and m.4394C>T mutations, three subjects with only the m.1555A>G mutation and three controls were analyzed using enzyme-linked immunosorbent assays (ELISA), according to the protocol provided by the manufacturer (Nikken Foods, St. Louis, MO, USA).

Assigning Pathogenicity to the mt-tRNA Gln Mutation
The updated pathogenicity scoring system was used to assess the m.4394C>T mutation [41]. The mutation was classified as "probably pathogenic" with a score of ≥11 points, "possibly pathogenic" with a score of 7-10 points and a "neutral polymorphism" with a score of ≤6 points.

Computer Analysis
All data were expressed as the mean ± standard deviation (SD). The statistical analyses were carried out using the unpaired, two-tailed Student's t test contained in the GraphPad Prism 5 program (GraphPad Software). p < 0.05 was believed to be statistically significant.

Clinical Features of HZD055 and HZD510 Pedigrees
We examined two Chinese families with maternally transmitted AINSHL in the Otology Clinic of Hangzhou First People's Hospital (Figure 1). In the HZD055 pedigree, the proband (III-7) lived in Hangzhou city of Zhejiang Province. She had a history of using gentamycin when she was 19 years old, and she began to suffer from bilateral hearing loss two months after she started using gentamycin. Clinical examinations revealed that she exhibited bilateral hearing loss (106 dB for the right ear, 86 dB for left ear). Interestingly, her matrilineal relatives (II-8 and III-4) also had a history of using AmAn. As shown in Figure 2, they had profound hearing impairment. In the HZD510 pedigree, five of the ten matrilineal relatives exhibited deafness as a sole clinical phenotype. Genetic counseling suggested that the matrilineal relatives (II-1, II-4, II-6, II-8 and III-7) were deaf patients. Audiological examinations revealed that subject II-6 had mild hearing loss, subject II-8 had moderate hearing loss and proband (III-7) developed severe hearing impairment ( Figure 2 and Table 2). A comprehensive genetic counseling revealed that these subjects did not have any other clinical abnormalities, such as cardiovascular disease, visual loss, trauma, infectious diseases or neurological disorders.  In the HZD510 pedigree, five of the ten matrilineal relatives exhibited deafness as a sole clinical phenotype. Genetic counseling suggested that the matrilineal relatives (II-1, II-4, II-6, II-8 and III-7) were deaf patients. Audiological examinations revealed that subject II-6 had mild hearing loss, subject II-8 had moderate hearing loss and proband (III-7) developed severe hearing impairment ( Figure 2 and Table 2). A comprehensive genetic counseling revealed that these subjects did not have any other clinical abnormalities, such as cardiovascular disease, visual loss, trauma, infectious diseases or neurological disorders. Abbreviations: AmAn: aminoglycoside antibiotics; PTA: pure tone audiometry; dB: decibels.
As shown in Figure 1, the family histories of HZD055 and HZD510 were consistent with maternal inheritance. Notably, the penetrance of hearing loss in HZD055 and HZD510 was 30% and 50% (AmAn included), respectively; when the AmAn was excluded, the penetrance of these pedigrees was 0% and 10%, respectively. The age at onset of deafness in HZD055 and HZD510 varied from 19 to 45, with an average of 27 years.

Construction of Cybrid Cells from Two Chinese Families with AINSHL
Immortalized lymphoblastoid cell lines were generated from three matrilineal relatives (HZD510: II-6, II-8 and III-7) harboring both the m.1555A>G and m.4394C>T mutations, three subjects (HZD055: II-8, III-4 and III-7) belonging to the mtDNA haplogroup D4g2a, and three controls without these pathogenic mutations (C1, C2 and C3) ( Figure 6). The cybrid cells were used to further the biochemical analysis.

Decreased in ATP Production
ATP is the key energy-carrying molecule in all cells. As shown in Figure 7A, the ATP levels in cell lines with both the m.1555A>G and m.4394C>T mutations and the cells with only the m.1555A>G mutation were 48% (p < 0.001) and 74% (p = 0.0012) of the average value measured in the control cell lines, respectively.

Construction of Cybrid Cells from Two Chinese Families with AINSHL
Immortalized lymphoblastoid cell lines were generated from three matrilineal relatives (HZD510: II-6, II-8 and III-7) harboring both the m.1555A>G and m.4394C>T mutations, three subjects (HZD055: II-8, III-4 and III-7) belonging to the mtDNA haplogroup D4g2a, and three controls without these pathogenic mutations (C1, C2 and C3) ( Figure 6). The cybrid cells were used to further the biochemical analysis.

Construction of Cybrid Cells from Two Chinese Families with AINSHL
Immortalized lymphoblastoid cell lines were generated from three matrilineal relatives (HZD510: II-6, II-8 and III-7) harboring both the m.1555A>G and m.4394C>T mutations, three subjects (HZD055: II-8, III-4 and III-7) belonging to the mtDNA haplogroup D4g2a, and three controls without these pathogenic mutations (C1, C2 and C3) ( Figure 6). The cybrid cells were used to further the biochemical analysis.

Decreased in ATP Production
ATP is the key energy-carrying molecule in all cells. As shown in Figure 7A, the ATP levels in cell lines with both the m.1555A>G and m.4394C>T mutations and the cells with only the m.1555A>G mutation were 48% (p < 0.001) and 74% (p = 0.0012) of the average value measured in the control cell lines, respectively.

Decreased in ATP Production
ATP is the key energy-carrying molecule in all cells. As shown in Figure 7A, the ATP levels in cell lines with both the m.1555A>G and m.4394C>T mutations and the cells with only the m.1555A>G mutation were 48% (p < 0.001) and 74% (p = 0.0012) of the average value measured in the control cell lines, respectively.

Reductions in MMP
Depolarization of MMP is often associated with apoptosis by the cytochrome c released from mitochondria [45]. To see whether the m.4394C>T mutation affected MMP, the JC-1 assay was used to evaluate the MMP in both mutant and control cells. As shown in Figure 7B, the levels of MMP in the mutant cell lines carrying both the m.1555A>G and m.4394C>T mutations and the cells carrying only the m.1555A>G mutation were 53% (p < 0.0001) and 74% (p = 0.0004) of the mean values measured in the control cell lines, respectively.

The Increase in ROS Production
ROS plays an active role in many cellular process [46]. As illustrated in Figure 7C, the levels of ROS generation in mutant cell lines with both the m.1555A>G and m.4394C>T mutations and the cells carrying only the m.1555A>G mutation were 192% (p < 0.0001) and 137% (p = 0.0015) of the mean values measured in the control cell lines, respectively.

mtDNA Copy Number Decrease
The mtDNA copy number is a promising biomarker of mitochondrial dysfunction [47]. To see whether the m.4394C>T mutation affected mtDNA content, Qpcr was used to

Reductions in MMP
Depolarization of MMP is often associated with apoptosis by the cytochrome c released from mitochondria [45]. To see whether the m.4394C>T mutation affected MMP, the JC-1 assay was used to evaluate the MMP in both mutant and control cells. As shown in Figure 7B, the levels of MMP in the mutant cell lines carrying both the m.1555A>G and m.4394C>T mutations and the cells carrying only the m.1555A>G mutation were 53% (p < 0.0001) and 74% (p = 0.0004) of the mean values measured in the control cell lines, respectively.

The Increase in ROS Production
ROS plays an active role in many cellular process [46]. As illustrated in Figure 7C, the levels of ROS generation in mutant cell lines with both the m.1555A>G and m.4394C>T mutations and the cells carrying only the m.1555A>G mutation were 192% (p < 0.0001) and 137% (p = 0.0015) of the mean values measured in the control cell lines, respectively.

mtDNA Copy Number Decrease
The mtDNA copy number is a promising biomarker of mitochondrial dysfunction [47]. To see whether the m.4394C>T mutation affected mtDNA content, Qpcr was used to assess the mtDNA copy number variations. As shown in Figure 7D, the levels of the mtDNA copy number in the mutant cell lines carrying both the m.1555A>G and m.4394C>T mutations and the cells carrying only the m.1555A>G mutation were 52% (p < 0.0001) and 78% (p = 0.0064) of the mean values measured in control group, respectively.

Effect of the m.4394C>T Mutation on mt-RNA Transcription
To explore the synergistic roles of the m.4394C>T mutation on m.1555A>G-induced mitochondrial dysfunction, we evaluated the mt-RNA transcription in patient-derived cell lines and found that significantly lower levels of ND1, ND2, ND3, ND4, A6 and A8 mRNA were present in the cell lines carrying both the m.1555A>G and m.4394C>T mutations when compared with the cells carrying only the m.1555A>G mutation and controls without these primary mutations (Figure 8). This result indicated that the m.4394C>T mutation might, at least partially, block mt-RNA transcription. m.4394C>T mutations and the cells carrying only the m.1555A>G mutation were 52% (p < 0.0001) and 78% (p = 0.0064) of the mean values measured in control group, respectively.

Effect of the m.4394C>T Mutation on mt-RNA Transcription
To explore the synergistic roles of the m.4394C>T mutation on m.1555A>G-induced mitochondrial dysfunction, we evaluated the mt-RNA transcription in patient-derived cell lines and found that significantly lower levels of ND1, ND2, ND3, ND4, A6 and A8 mRNA were present in the cell lines carrying both the m.1555A>G and m.4394C>T mutations when compared with the cells carrying only the m.1555A>G mutation and controls without these primary mutations (Figure 8). This result indicated that the m.4394C>T mutation might, at least partially, block mt-RNA transcription.

Reduced Activities of Complexes I~IV
To see the effect of the m.4394C>T mutation on OXPHOS functions, we analyzed the activities of respiratory complexes in three cell lines with both the m.1555A>G and m.4394C>T mutations, three cell lines derived from patients with only the m.1555A>G mutation, and three cells without these mutations. As shown in Figure 9, we noticed that Complexes I~IV were markedly decreased in cells with both mtDNA mutations, as compared with controls (p < 0.05 for all).

Reduced Activities of Complexes I~IV
To see the effect of the m.4394C>T mutation on OXPHOS functions, we analyzed the activities of respiratory complexes in three cell lines with both the m.1555A>G and m.4394C>T mutations, three cell lines derived from patients with only the m.1555A>G mutation, and three cells without these mutations. As shown in Figure 9, we noticed that Complexes I~IV were markedly decreased in cells with both mtDNA mutations, as compared with controls (p < 0.05 for all).

The m.4394C>T Mutation Increases DNA Damage
To see whether the m.4394C>T mutation enhanced the DNA damage, the serum 8-OHdG levels were measured using ELISA. As shown in Figure 10, we found that subjects with both the m.1555A>G and m.4394C>T mutations exhibited higher levels of 8-OHdG than patients with only the m.1555A>G mutation (p = 0.0002) and controls without those mutations (p < 0.0001).

The m.4394C>T Mutation Increases DNA Damage
To see whether the m.4394C>T mutation enhanced the DNA damage, the serum 8-OHdG levels were measured using ELISA. As shown in Figure 10, we found that subjects with both the m.1555A>G and m.4394C>T mutations exhibited higher levels of 8-OHdG than patients with only the m.1555A>G mutation (p = 0.0002) and controls without those mutations (p < 0.0001).

Mutational Screening for Common Deafness-Associated Nuclear Genes
To uncover the roles of nuclear gene mutations in deafness, we screened for the mutations in common deafness-associated genes (GJB2, GJB3, GJB6 SLC26A4 and TRMU) in the matrilineal relatives of the HZD055 and HZD510 pedigrees. However, PCR and Sanger sequencing revealed that no functional variants could be identified in these genes, suggesting that nuclear modified genes may not play an active role in m.1555A>G-induced deafness.

Possible Pathogenic Effect of m.4394C>T in AINSHL
As shown in Table 4, according to the pathogenicity scoring system [41], the total score of the m.4394C>T mutation was 8 points, meaning it is "possibly pathogenic" and may cause AINSHL.

Mutational Screening for Common Deafness-Associated Nuclear Genes
To uncover the roles of nuclear gene mutations in deafness, we screened for the mutations in common deafness-associated genes (GJB2, GJB3, GJB6 SLC26A4 and TRMU) in the matrilineal relatives of the HZD055 and HZD510 pedigrees. However, PCR and Sanger sequencing revealed that no functional variants could be identified in these genes, suggesting that nuclear modified genes may not play an active role in m.1555A>G-induced deafness.

Possible Pathogenic Effect of m.4394C>T in AINSHL
As shown in Table 4, according to the pathogenicity scoring system [41], the total score of the m.4394C>T mutation was 8 points, meaning it is "possibly pathogenic" and may cause AINSHL.

Discussion
The present study has explored the synergistic role of the m.4394C>T mutation in clinical manifestations of deafness-associated m.1555A>G mutation. Hearing loss as a sole clinical phenotype only presented in matrilineal relatives, but not in other members in these pedigrees, suggesting that mitochondrial dysfunctions caused by mtDNA mutations are the molecular basis for this disease. In fact, the m.1555A>G mutation was originally identified in Arab-Israeli families with AmAn-associated hearing loss [48]. Without the usage of AmAn, subjects with the m.1555A>G mutation exhibited a wide range of clinical phenotypes that ranged from mild to severe hearing loss, and even normal hearing, indicating that while the m.1555A>G mutation is a major risk factor for hearing loss, by itself it is not sufficient to produce the phenotype [49], for which it requires other factors such as nuclear genes, AmAn or mtDNA secondary variants [50,51]. In particular, mtDNA haplogroup B4C1C exhibited a higher penetrance and expressivity of m.1555A>G-induced deafness [52]. Recent experimental studies indicate that mitochondrial haplogroup B enhances the risk for hearing impairment in Chinese patients carrying the m.1555A>G mutation [53]. Moreover, mitochondrial haplogroup D4 specific m.5802T>C, m.10454T>C, m.12224C>T and m.11696G>A mutations may enhance m.1555A>G-induced deafness [15]. Sequence characterization of the complete mitochondrial genome of matrilineal relatives from these pedigrees suggested the presence of sets of polymorphisms belonging to mtDNA haplogroup D4g2a [21]. We noticed that the penetrance of hearing loss in HZD055 was 50% including AmAn and 10% excluding AmAn, while in HZD510 it was 30% including AmAn and 0% excluding AmAn. Compared with previous studies, the penetrances of m.1555A>G-induced hearing loss ranged from 22.2% to 73.9% (average: 50.9%, including AmAn), whereas excluding the effects of AmAn, the penetrances varied from 11.1% to 60.8% (average: 30.8%, Table 5) [15,16,20,54]. The m.4394C>T mutation resided at position 7 in the 5 end of tRNA Gln , which was well conserved from different vertebrates. Notably, the m.4394C>T mutation introduced a novel 7T-66A base-pairing and was critical for the steady-state level of the tRNA Gln . Interestingly, nucleotides at the same position in tRNA Gly (9997T>C) and in tRNA Ile (4269A>G) were regarded as pathogenic mutations for cardiomyopathy [55,56]. Hence, the m.4394C>T mutation may be similar to the m.9997T>C or m.4269A>G mutation, and cause mitochondrial dysfunction that is involved in hearing impairment.
Trans-mitochondrial technology was frequently used to assess the contribution of mtDNA genetic polymorphisms to OXPHOS function since noise from the nuclear genetic background was adjusted. We demonstrated that cybrid cells bearing both the m.1555A>G and m.4394C>T mutations exhibited much more severe effects on mitochondrial dysfunctions than those with only the m.1555A>G mutation and controls. Specifically, we observed drops of approximately 52% and 26% in ATP production in mutant cells carrying both the m.1555A>G and m.4394C>T mutations or mutant cell lines with only the m.1555A>G mutation, respectively, which are below the threshold ATP level for developing a clinical phenotype [57]. Notably, the deficiency of mitochondrial respiratory chain functions will frequently affect MMP, and in this study, approximately 47% and 26% reductions in MMP were found in cell lines carrying both the mtDNA mutations, and in cell lines with only the m.1555A>G mutation, respectively. It had been recognized that loss of MMP may cause cytochrome c to release into the cytoplasm and then induce programmed cell death [58]. Moreover, MMP constitutes an important component of mitochondrial respiration complexes and is associated with many mitochondrial functions, such as ATP formation and mitochondrial protein import [59,60]. In addition, a much lower level in mtDNA content was found in patients with both mutations than in patients with only the m.1555A>G mutation. Indeed, the mtDNA copy number is an important element for mitochondrial replication and transcription, which is critical for cell functions [61]. Alternations in mtDNA copy number are associated with diverse human pathologies [62]. Thus, the abnormal mtDNA copy number and MMP may lead to an increased in ROS production. ROS are the by-products of oxidative stress, and include peroxides, superoxides, hydroxyl radicals and singlet oxygen [63]. It has been documented that ROS plays a crucial role in human physiological and pathophysiological processes [64]. High ROS concentration will react with proteins, lipids, RNA or DNA, and frequently lead to irreversible functional alterations, including regulating mitochondrial functions, apoptosis and signaling transduction [65]. These results indicated that the ATP, MMP, ROS and mtDNA copy number were impaired in cells with both mtDNA mutations.
Next, examination of the RNA level of mtDNA-encoded OXPHOS subunits revealed that the mRNA levels of ND1, ND2, ND3, ND4, A6 and A8 were lower in cybrids carrying both mtDNA mutations than in cybrids with only the m.1555A>G mutation (p < 0.05 for all). This result indicates that defective mt-RNA translation caused by both the m.1555A>G and m.4394C>T mutations may be responsible for multiple OXPHOS complex deficiencies in cybrid cells. The altered tRNA Gln metabolism caused by the m.4394C>T mutation led to a defect in mitochondrial translation. In fact, cell lines with both the m.1555A>G and m.4394C>T mutations would perturb the activities of Complexes I-IV and then worsen the defects in mitochondrial respiratory phenotypes associated with the m.1555A>G mutation, as in the case of the m.14692A>G mutation [66].
In addition, the level of serum 8-OHdG was much higher in patients with both the m.1555A>G and m.4394C>T mutations, suggesting that the mitochondrial dysfunctions caused by the m.1555A>G mutation may be worsened by the m.4394C>T mutation. Furthermore, the absent of any functional variants in deafness-related common nuclear genes (GJB2, GJB3, GJB6, SLC26A4 and TRMU) in the matrilineal relatives in these pedigrees suggests that nuclear modified genes may not play an active role in deafness expression, hence, the mt-tRNA Gln 4394C>T mutation may enhance the penetrance of m.1555A>Ginduced deafness.

Conclusions
In conclusion, the mt-tRNA Gln 4394C>T mutation may aggravate mitochondrial dysfunction and increase the penetrance of hearing impairment caused by the m.1555A>G mutation.

Informed Consent Statement:
Written informed consent to publish this paper has been obtained from the patients.

Data Availability Statement:
All the data supporting the results of this study are included in the article.