Frequency and spectrum of MT-TT variants associated with Leber’s hereditary optic neuropathy in a Chinese cohort of subjects

Abstract Leber’s hereditary optic neuropathy (LHON) is a maternally inherited eye disease. In our previous investigations, we have reported the spectrum and frequency of mitochondrial MT-ND1, MT-ND4 and MT-ND6 gene in Chinese LHON population. This study aimed to assess the molecular epidemiology of MT-TT mutations in Chinese families with LHON. A cohort of 352 Chinese Han probands lacking the known LHON-associated mtDNA mutations and 376 control subjects underwent molecular analysis of mtDNA. All variants were evaluated for evolutionary conservation, structural and functional consequences. Fifteen variants were identified in the MT-TT gene by mitochondrial genome analysis of LHON pedigrees, which was substantially higher than that of individuals from general Chinese populations. The incidences of the two known LHON-associated mutations, m.15927G > A and m.15951A > G, were 2.27% and 1.14%, respectively. Nine putative LHON-associated variants were identified in 20 probands, translated into 2.1% cases of this cohort. Moreover, mtDNAs in 41 probands carrying the MT-TT mutation(s) were widely dispersed among nine Eastern Asian haplogroups. Our results suggest that the MT-TT gene is a mutational hotspot for these 352 Chinese families lacking the known LHON-associated mutations. These data further showed the molecular epidemiology of MT-TT mutations in Chinese Han LHON pedigrees.


Introduction
Leber's hereditary optic neuropathy (LHON, #535000) is a maternally inherited mitochondrial disorder leading to vision failure by the preferential loss of retinal ganglion cells (RGCs), and marked adult male bias (Yu-Wai-Man et al. 2011;Jurkute and Yu-Wai-Man 2017). The minimum prevalence of visual impairment due to LHON was 3.22 per 100,000 in northeast of England (Yu-Wai-Man et al. 2003). Maternal inheritance of LHON indicated the involvement of mutations in mitochondrial DNA (mtDNA) (Wallace et al. 1988;Howell 2003). The human mitochondrial genome is a 16,569 bp, doublestranded, circular molecule that codes for 13 subunits of respiratory chain complexes, two rRNAs (12S and 16S rRNA) and 22 mitochondrial tRNAs (Andrews et al. 1999). The majority of subjects with LHON (90%-95%) harbor one of three primary LHON-associated mtDNA mutations (m.3460G > A, m.11778G > A, and m.14484T > C) in some western countries (Brown et al. 1995;Mackey et al. 1996;Mashima et al. 1998), while these mutations are only responsible for 38.3% and 41.1% cases in two large cohorts of Chinese LHON subjects, respectively (Jia et al. 2006;Liang et al. 2014;Jiang et al. 2015;). Thus, other mtDNA genes including mitochondrial tRNA genes are the hotspots associated with LHON (Ruiz-Pesini et al. 2007;Zheng et al. 2012;Xue et al. 2016). In our previous four investigations, four tRNA mutations (MT-TM 4435A > G, MT-TE 14693A > G, MT-TT 15927G > A and 15951A > G) have been identified as LHONassociated mutations Qu et al. 2006;Tong et al. 2007;Zhang et al. 2018). These studies tested only in relatively small sized samples of pedigrees, while the association of MT-TT mutations with LHON in a large population remains to be explored. The purpose of this present study was to perform a comprehensive test of the hypothesis that MT-TT variants play an important role in the pathogenesis of LHON. For this objective, we recruited a cohort of 352 genetically unrelated Chinese Han patients with LHON (269 males and 82 females) and 376 Chinese control subjects performed the Sanger sequence analysis of the DNA fragments spanning MT-TT gene, and then, investigated mutational spectrum and incidences of MT-TT gene. This analysis showed the identification of 15 nucleotide changes among MT-TT gene. To identify deleterious mutations from polymorphisms, these variants were further evaluated using the criteria shown in the previous studies (Bandelt et al. 2009;Zheng et al. 2012;Kirchner and Ignatova 2015;Xue et al. 2016). These analyses showed that 9 tRNA variants might have higher evolutionary conservation index, structural and functional alterations. Moreover, these mtDNAs of 20 subjects carrying the putative variants were assigned to the Asian mtDNA haplogroups using the nomenclature of mtDNA haplogroups (Tanaka et al. 2004;Kong et al. 2006).

Subjects
A totally of 352 unrelated Han Chinese LHON subjects lacking the known LHON-associated mtDNA mutations were recruited for this investigation. This study was in compliance with the Declaration of Helsinki (Liang et al. 2014;Jiang et al. 2015;. The institutional review boards of Wenzhou Medical University and Zhejiang University approved this study. A cohort of 376 Chinese control subjects obtained from the same areas were screened for the presence of mtDNA variants.

Ophthalmologic examinations
These probands and other members of these families received ophthalmological examinations at School of Ophthalmology and Optometry, Eye Hospital, Wenzhou Medical University and were diagnosed as LHON. The degree of visual impairment was defined according to the visual acuity as follows: normal > 0.3, mild ¼ 0.3-0.1, moderate <0.1-0.05, severe <0.05-0.02, and profound <0.02 (Qu et al. 2009;Liu et al. 2011).

Mutational analysis of mitochondrial genomes
Genomic DNA was isolated from whole blood of participants (352 probands lacking the known LHON-associated mutations and 376 Chinese control subjects) using QIAamp DNA Mini Kit (Qiagen). Subjects' DNA fragments spanning the MT-TT gene were amplified, purified and subsequently analyzed by direct sequencing in an ABI 3700 automated DNA sequencer using the BigDye Terminator Cycle sequencing reaction kit (Applied Biosystems) (Jia et al. 2018;Zhang et al. 2018). These sequence results were compared with the revised Cambridge Reference Sequence (rCRS, NC_012920.1) (Bandelt et al. 2014). For defining the mitochondrial haplogroups, the entire mitochondrial genomes of 41 subjects with MT-TT mutations were PCR amplified in 24 overlapping fragments using sets of the light (L) strand and the heavy (H) strand oligonucleotide primers, as described previously (Rieder et al. 1998). The analysis of variants was evaluated according to the previous description (Zou et al. 2010;Zhang et al. 2011Zhang et al. , 2012.

Evolutionary conservation and structural analysis
Evolutionary conservation analysis for certain mtDNA variant was performed by comparing human mtDNA to 43 different vertebrate species, as shown in our previous studies (Ruiz-Pesini and Wallace 2006; Carelli et al. 2017). The conservation index (CI) of certain variant was defined by the percentage of species for a list of 44 different vertebrate species (including Homo species). The secondary cloverleaf and tertiary structure of human MT-TT were analyzed by the online software (Sprinzl and Vassilenko 2005;J€ uhling et al. 2009;Lott et al. 2013).

Haplogroup classification
The mtDNA sequences of eight probands carrying m.15927G > A, four subjects carrying the m.15951A > G mutation, as well as 20 subjects carrying the putative MT-TT variants are assigned to the Asian mtDNA haplogroups by using the nomenclature of mitochondrial haplogroups (Kong et al. 2006;Zou et al. 2010).

Statistical analysis
Statistical analysis was performed by the v 2 test contained in Microsoft Office Excel (Version 2017). p value indicates the significance, according to the v 2 test, of the difference between mutant and control mean. Differences were considered significant at a p < .05.

Study samples
The study samples lacking the known LHON-associated mtDNA mutations consisted of 269 males and 83 females. All participants were Han Chinese subjects recruited from eye clinics of 25 provinces in China, as shown in Figure 1. Ophthalmologic evaluation showed that all affected subjects exhibited the variable severity and age at onset of optic neuropathy. Of these, 38 subjects exhibited profound visual impairment, 50 subjects had severe visual impairment, 53 individuals suffered from moderate visual impairment, and 212 subjects had mild visual impairment. The age at onset of optic neuropathy ranged from 1 to 52 years, with an average of 17.5 years. Comprehensive family medical histories of those probands showed no other clinical abnormalities, including diabetes, muscular diseases, hearing loss, and other neurological disorders.

Mutational analysis of MT-TT gene
Deoxyribonucleic acid fragments spanning MT-TT gene were PCR-amplified from genomic DNA of 352 Chinese subjects with LHON and 376 Han Chinese control individuals. Each fragment was purified and subsequently analyzed by DNA sequencing. Comparison of the resultant sequences in 352 affected subjects with the Cambridge consensus sequence identified 15 known nucleotide changes in the MT-TT gene (Lott et al. 2013), but only seven variants in this gene were identify in 376 control individuals, as shown in the Table 1. The m.15927G > A and m.15951A > G mutations were the two known LHON-associated MT-TT gene mutations Zhang et al. 2018). All the nucleotide changes were identified by sequence analysis of both strands and appeared to be homoplasmy. In the mutational screening, no MT-TT nucleotide changes were detected in the 311 patients and 353 controls, while at least one MT-TT variant was identified in 41 affected subjects and 23 control individuals (p ¼ .0085). That indicates that MT-TT gene is a mutational hotspot for Chinese LHON pedigrees. Among these, 12 individuals carried one of the known LHON-associated MT-TT mutations, including eight subjects carrying the m.15927G > A mutation and four individuals with the m.15951A > G mutation.
Furthermore, the nine variants carrying in twenty probands were considered as putative LHON-associated variants and other four variants belonged to the polymorphisms.

Evaluation of the MT-TT variants
These variants in MT-TT were first evaluated by the phylogenetic analysis of these variants and amino acid sequences from other 43 vertebrates. The conservation index among   these residues ranged from 22.7% to 100%, as shown in Table 1. Of these, conservation indexes of 9 variants were greater than 70%, with potential functional significance ( Figure 2) (Ruiz-Pesini and Wallace 2006). As shown in Table 1, eight variants were absent in 376 Chinese controls, while the frequencies of 7 variants ranged from 0.27% to 3.46% in this control population. Furthermore, we analyzed the structural alteration of tRNA Thr by these variants based on the predicated secondary and tertiary structure. As shown in Figure , which were present in the controls or lower conservation indexes, appeared to be the polymorphisms.

Characterization of 20 Chinese probands
Comprehensive medical histories of 20 probands carrying one of nine putative LHON-associated MT-TT variants and other members in these families showed no other clinical abnormalities, including diabetes, muscular diseases, hearing loss, and neurological disorders. As shown in Table 2 and Figure 3, these families exhibited a wide range of severity, age at onset, and penetrance of optic neuropathy. Of these, only one matrilineal relative per family in fifteen pedigrees suffered from optic neuropathy, while five pedigrees (WZ1008, WZ1011, WZ1012, WZ1014, and WZ1020) had a history of optic neuropathy. The putative variants were first examined in all available members of these pedigrees. The mtDNA mutations were presented in matrilineal relatives in each family in the homoplasmic form, but not in other members of every family. To assess the contribution that mtDNA variants make toward the variable penetrance and expressivity of optic neuropathy in these Chinese pedigrees, we analyzed entire mtDNA sequences in 20 Chinese probands (Genbank accession numbers: MK795825-MK795844). These affected individuals exhibited distinct sets of mtDNA polymorphisms including 217 known and 4 novel variants (Table  3)

Analysis of entire mtDNA sequences in probands
The past study examined the entire mtDNA sequences of 8 subjects consisted of five females and three males carrying the m.15927G > A mutation . MT-TT m.15951A > G mutation may have a potential modifier role in increasing the penetrance and expressivity of the primary LHON-associated m.11778G > A mutation in a Chinese family . In this study, we determined the complete mtDNA sequence analysis of additional four probands carrying m.15951A > G mutation. Furthermore, these probands exhibited distinct sets of mtDNA polymorphisms including 70 known variants. We further performed the haplogroup analysis of mtDNAs carrying the m.15951A > G mutation. As shown in Table 4, the mtDNAs from four Chinese families carrying the m.15951A > G mutation belong to Eastern Asian mtDNA haplogroup D. The frequency of mtDNA haplogroups D in 41 LHON families carrying the MT-TT mutation were 29.3%; while that of 376 Chinese controls was 21.5%. And then, we determined the complete mtDNA sequence analysis of additional 20 probands carrying putative LHONassociated variants, these probands exhibited distinct sets of mtDNA polymorphisms including 217 known variants and 4 unknown variants (Table 3). Thus, the frequency of haplogroup B in the Chinese pedigrees carrying the MT-TT mutations were significantly higher than that in 376 Chinese controls and other Asian populations. Meanwhile, that of haplogroup M8 was much lower than control individuals. This discrepancy between the different ethnic origins may be attributed to evolution.

Discussion
The majority of patients with LHON (90%-95%) harbors one of three primary mtDNA point mutations, including m.3460G > A, m.11778G > A, and m.14484T > C, while these mutations are only responsible for 38.3% and 41.1% cases in two large cohorts of Chinese LHON subjects, respectively (Jia et al. 2006;Liang et al. 2014;Jiang et al. 2015;. A number of LHON-associated mtDNA mutationts have been reported (Table 5), with some still awaiting full confirmation for pathogenicity, having been identified in only single families (Lott et al. 2013). MT-TT gene region is thought to be "mutational hotspot", harboring other LHON-causing mutations, in addition to m.15927G > A and m.15951A > G Zhang et al. 2018). The coexistent of the m.15924A > G and m.3635G > A in some Chinese families indicate that m.15924A > G mutation may play a synergistic role in the phenotypic manifestation of LHON associated MT-ND1 3635G > A mutation ). The marked male bias and variability in the clinical phenotypes suggest nuclear modifier gene(s) or environmental factor(s) appear to play a role in the phenotypic expression in these 20 Chinese pedigrees (Yu-Wai-Man et al. 2011). Nuclear modifier genes were proposed to increase the susceptibility to LHON-associated mtDNA mutations (Chen et al. 2015). Three studies using microsatellite markers have confirmed significant linkage on the X-chromosome, with some of these candidate regions showing areas of overlap (Shankar et al. 2008;Ji et al. 2010). A genome-wide study of nine large m.11778G > A Thai pedigrees found two SNPs (rs3749446 and rs1402000), located within PARL (Presenilin-associated rhomboid-like) were associated with a statistically increased risk of phenotypic expression among LHON carriers (Phasukkijwatana et al. 2010). However, the association between these two PARL SNPs and visual loss was not replicated in an independent cohort of Chinese m.11778G > A LHON pedigrees ). In our previous study, we identified a mutation in YARS2 as a nuclear modifier for the phenotypic manifestation of LHON-associated m.11778G > A mutation .
In the present study, using the Sanger sequence of MT-TT gene, we identified 15 known variants in MT-TT gene were identified in a cohort of 352 Han Chinese subjects with LHON. These variants could have potential structural alterations and functional significance of MT-TT. In particular, these variants could affect the processing of the tRNAs 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. Seven variants at tRNA stems, abolishing the Watson-Crick (WC) base pairs of mitochondrial tRNAs, likely lead to the tRNA aminoacylation, editing, and modification, which might result in low efficiency and accuracy of mitochondrial protein synthesis . The m.15908T > C mutation affected a highly conserved thymine at position 23 at the DHU-stem of MT-TT, destabilizing the conservative base pairing (12A-23T). That may alter the secondary structure and function of MT-TT, Two mutations at the acceptor stem 15951A > G and 15949G > A may alter the secondary structure and function  of tRNAs, as in the case of the MT-TS 7511T > C (A4) ) and MT-TH 12201T > C (U68) (Gong et al. 2014) mutations. Moreover, two variants (m.15924A > G and m.15928G > A) at the anticodon stem may affect the function of tRNAs, as in the case of the MT-TT 15927G > A mutation (Jia et al. 2018;Zhang et al. 2018). Finally, m.15943T > C at the T-stem may also affect the structure and function of tRNAs. However, the functional significances of these putative LHON-associated tRNA variants should be further investigated. . Classification tree of 20 complete mtDNA sequences, plus the revised Cambridge reference sequence (rCRS). The synonymous and non-synonymous coding-region variants in the mtDNA sequences are denoted by "/s" and "/ns," respectively. Variants in the ribosomal RNA genes and tRNA genes are denoted by "/r" and "/t." Recurrent mutations are underlined. The failures in tRNA metabolisms caused by these putative LHON-associated variants would lead to the impairment of mitochondrial protein synthesis and deficient respirations, as in the case of other mitochondrial tRNA mutations Jiang et al. 2016;Xue et al. 2016;Jia et al. 2018;Zhang et al. 2018). The MT-TE m.14693A > G variant may act as modifiers influencing the phenotypic manifestation of LHONassociated m.3460G > A mutation (Tong et al. 2007). Furthermore, our previous investigation showed that MT-TM m.4435A > G and MT-TT m.15951A > G mutations modulate the phenotypic expression of the LHON-associated m.11778G > A mutation in Chinese families Qu et al. 2006). However, the tissue specificity of these tRNA variants is likely attributed to tissue-specific tRNA metabolism or the involvement of nuclear modifier genes (Dittmar et al. 2006;Chen et al. 2016). The homoplasmic nature of these mitochondrial tRNA variants hints to mild nature of mutations. These suggest that the tRNA variants may be insufficient to produce a clinical phenotype by itself but the inherited risk factor(s) is necessary for the development of LHON. Nuclear modifier genes, environmental and epigenetic factors, as well as personal lifestyles such as smoking and drinking may also contribute to the development of LHON in these subjects carrying the mtDNA variants (Carelli et al. 2016;Jiang et al. 2016).
Here, mtDNAs in 41 LHON families carrying the MT-TT variants were widely dispersed among 9 Eastern Asian subhaplogroups. Indeed, the occurrences of mtDNA haplogroups D in families carrying the m.15951A > G mutation were higher than those in controls. Moreover, the frequencies of mtDNAs in haplogroups G and B in eight Chinese families carrying the m.15927G > A mutation, while 20 pedigrees carrying one of nine putative variants were similar to those in controls. Thus, the frequencies of haplogroups G, B, and F in the LHON probands carrying the MT-TT mutations were significantly higher than those in 478 Chinese controls and other Asian populations (Tanaka et al. 2004;Kong et al. 2006). mtDNA haplogroups M7b1'2 and M8a affect clinical expression of LHON in Chinese families with the m.11778G > A Mutation (Ji et al. 2008). This discrepancy implicates a role of mtDNA haplotypes in the phenotypic manifestation of LHON-associated mtDNA mutations (Liang et al. 2014;Jiang et al. 2015;.
In summary, this is the first study to investigate the frequency and spectrum of mutations in MT-TT gene in Chinese subjects with LHON. The two known LHON-associated MT-TT mutations, m.15927G > A and m.15951A > G, in Chinese cohort accounted for 3.41% cases of 352 Chinese subjects with LHON. Furthermore, the nine putative LHON-associated mtDNA variants were the rare mutations, accounting for 5.66% cases in this Chinese cohort. A total of 41 subjects carrying one of the MT-TT mutations accounted for 9.07% cases of 352 Chinese subjects with LHON. These data further support that the MT-TT gene is a hotspot for mutations associated with LHON. Thus, our findings may provide valuable information for the further understanding of pathophysiology and management of LHON.

Availability of data and materials
The analyzed data and materials generated during the study are available from the corresponding author on reasonable request.

Ethical approval
The present study was approved by the Ethics Committee of Wenzhou Medical University. Written informed consent was obtained from each patient involved in the study.