Introduction
Complex I is the first and largest component of the mitochondrial respiratory chain. Impairment of complex I accounts for most cases of respiratory chain deficiency, and results in a wide range of clinical manifestations (1,2), including Leigh syndrome (LS) (3). Complex I deficiency is also the most common cause of LS (4). In LS patients with complex I deficiency, complex I activity may be reduced up to 66% (4), whereas the activity of other complexes is normal.
LS is a subacute, necrotizing encephalopathy, characterized by bilateral symmetrical necrotic lesions of gray matter nuclei in the basal ganglia, diencephalon, cerebellum, or brainstem (5). Onset of LS typically occurs at early infancy. Patients present psychomotor retardation, seizures, nystagmus, ophthalmoparesis, optic atrophy, ataxia, dystonia, or respiratory failure, due to the progressive decline of the central nervous system functions. In most cases, LS patients die before age 5 (6).
The mutation m.10197 G>A in the mitochondrial NADH dehydrogenase, subunit 3 (ND3) gene was found to cause LS. This mutation, causing the typical manifestations of LS, is considered to lead to a severe phenotype. In this study, we describe the clinical and molecular features of the first Chinese LS pedigree with the mutation m.10197 G>A, and compare the phenotype of these LS patients to those previously described for LS patients with the same mutation. We also describe follow-up results of treatment with the coenzyme Q10.
Materials and methods
Pedigree and ethics
The study was approved by the Ethics Committee of Fujian Medical University Union Hospital. Informed consent was obtained from all patients included in the study.
A two-generation Han Chinese family living in southeast China near the Min River with 4 family members was studied (Fig. 1A). The proband, a 16-year-old girl (40 kg weight, 157 cm height), was admitted in the Union Hospital in 2012 with slight facial paralysis. She had been delivered with mild asphyxia, due to a 6-h-long second stage of labor with an 8-h duration from full dilatation to delivery. She presented symptoms of anorexia and fatigue following physical activities. Therefore, she did not pass any physical agility test at school. With regards to academic performance, she ranked middle in her class. There was no evidence of mental or psychomotor retardation. She scored 28 in both the Mini-Mental State Examination and Montreal Cognitive Assessment. Neurological examination revealed horizontal nystagmus, slightly shallow left nasolabial fold, slight hypalgesia in the left low face, hypotonia, and indications of bilateral Babinski sign. The levels of serum ceruloplasmin, serum creatine kinase, serum creatine kinase-MB, and antinuclear antibody were normal. An elevated plasma lactate level was noted [4.0 mmol/l; normal level (N) <2.7]. Electroencephalogram and electromyography examinations revealed no abnormality. Brain magnetic resonance imaging (MRI) showed prolonged T1- and T2-weighted signals in the basal ganglia, thalamus, midbrain, pons, and medulla oblongata (Fig. 1B–E). Magnetic resonance spectroscopy showed a markedly increased lactate doublet in the lesions (Fig. 1F and G). The pressure, routine, biochemical, cytological, and IgG index examinations of the cerebrospinal fluid (CSF) were normal, and oligoclonal bands were absent. Histochemical examination of a muscle biopsy specimen stained with hematoxylin and eosin (Fig. 2A), modified Gomori trichrome (Fig. 2B), nicotinamide adenine dinucleotide (Fig. 2C), as well as electron microscopy examination (Fig. 2D) showed no characteristic abnormality. The mother of the proband, and her 8-year-old brother were normal. There were no abnormalities in their physical examination and MRI brain imaging. The proband received treatment with 90 mg/day of the coenzyme Q10 for a year. Physical examination, MRI brain imaging, and plasma lactate level assessment were performed at the end of the treatment.
Mutational analysis
Genomic DNA was extracted from blood lymphocytes and skeletal muscle of the pedigree using QIAamp DNA Mini kit (Qiagen, Dusseldorf, Germany). Full-length mitochondrial DNA (mtDNA) was amplified by polymerase chain reaction using 24 pads of overlapping primers (Sangon Biotech, Shanghai, China). PCR was performed in a total volume of 25 μl, containing 100 ng of genomic DNA, 0.25 M of each primer, 10 M deoxynucleotide triphosphates, 1.25 U of Taq enzyme and 2.5 μl of 10X buffer. PCR amplification was performed using an Applied Biosystems 9700 thermocycler (Life Technologies, Carlsbad, CA, USA) with an initial denaturation step at 94°C for 5 min followed by 35 cycles of denaturation at 94°C for 30 sec, annealing at 59°C for 45 sec, and extension at 72°C for 1 min, with a final extension step at 71°C for 6 min. Sequencing was performed on an Applied Biosystems 3730 DNA automatic sequencer (Life Technologies) and resulting sequences were compared to the revised Cambridge reference sequence (rCRS) of the human mtDNA (7,8).
Quantification of heteroplasmy
The mutant load of m.10197 G>A was determined by restriction fragment length polymorphism analysis. A 476-bp fragment was amplified using the fluorescent-labeled FAM forward primer 5′FAM-TTTGTAGCCACAGGCTTCC-3′ and the reverse primer 5′-AAGGAGGGCAATTTCTAGAT-3′. The fragment was then digested with the restriction enzyme Cac8I (restriction site: GCN*NGC). This restriction site is present only in the wild-type mtDNA, and is thus expected to give 402- and 74-bp products upon wild-type mtDNA digestion. The digested fragments were subjected to electrophoresis and DNA sequencing, and the fluorescent fragments were detected using the Applied Biosystems GeneMapper® software V3.3 (Life Technologies). Heteroplasmy or mutant load was calculated by dividing the 476-bp peak area by the sum of the 476- and 402-bp peak areas.
Haplogroup identification in the pedigree
MtDNA haplogroup analysis was performed by full-length mitochondrial DNA sequencing, using the same method described above for mutational analysis. The results were confirmed by referring to Kivisild’s study (9).
Discussion
LS is a severe, progressive, metabolic neurodegenerative, mitochondrial disorder. The molecular defects and pathogenetic background of LS would provide considerable information to the patients in terms of management, genetic counseling, and prognosis. In this study, we described the clinical and molecular features of the first Chinese LS pedigree with the m.10197 G>A mutation, which leads to a severe phenotype, and report promising effects of treatment with the coenzyme Q10, as assessed by both clinical feature assessment and brain imaging.
The mutation m.10197 G>A was first reported in 2007 in 3 French pedigrees, as present in totally 8 LS patients (11). As shown in Table I, patients generally showed early-onset LS with a severe clinical outcome. Seventy-five percent of the patients died 1 year after the onset. A number of studies have provided evidence for the pathogenic role of the mutation m.10197 G>A. The mutation involves the codon 47 of the ND3 gene. This codon encodes alanine, a highly conserved residue in higher eukaryotes such as Homo sapiens, Bos Taurus, Equus caballus, Mus musculus, and Gallus gallus (Fig. 4). The conservation index of the mutation m.10197 G>A based on 39 species (conservation in 36 species except Xenopus laevis, Drosophila melanogaster and Sciurus vulgaris) is 92.3% (11). This value reached to the average level (93±13%) of conservation index of 22 well-characterized human pathogenic mtDNA mutation studied in the same 39 species (12). The mutation identified in the proband is a substitution from an alanine to a threonine. These two residues are quite different in structure and chemical property. Alanine is a hydrophobic amino acid without any side-chain. By contrast, threonine is a hydrophilic amino acid with a hydroxyl side-chain. These differences in amino acid properties may underlie the pathogenic role of the m.10197 G>A mutation in causing complex I deficiency and a severe LS phenotype.
It is notable that the individuals carrying the mutation in our study did not show the typical clinical manifestations, as previously reported (9,14). Certain common hypotheses postulated to explain the phenotypic variability of mtDNA diseases can be excluded based on the results of the Chinese pedigree reported in our study. First, the proband and her brother both presented a rather high mutant load, which has been commonly considered the most important factor affecting the severity of the disease (14). Thus, it is unlikely that the phenotypic variability of the patient may be due to mutant load. Second, the mutation was invariably present in both the blood and muscle tissues of the proband. Thus, it is unlikely that the mild clinical manifestation of the proband might be due to a different tissue distribution of the mutation. Third, the severe pathogenicity associated with certain mutations causing LS is expected to affect the early development of the affected individual. In Leshinsky-Silver’s study, brain lesions could be generated even at the fetal stage, before clinical manifestations of the syndrome become apparent (14). However, the normal brain image of the proband’s brother suggests that this mutation may not generate visible anatomical changes at the early stage.
Additional factors potentially contributing to the variability of phenotypes of mtDNA diseases include ethnicity, haplogroup and the environment. For instance, in the Leber hereditary optic neuropathy also caused by a mutation, the risk of visual loss is significantly increased in the haplogroups J2 in Europeans and M7b1’2 in Han Chinese, in contrast to the respective H and M8a haplogroups (16,17). As Table I shows, the severity of the disease was different between European patients and Asian ones, although they carried the same mutation m.10197G>A. European patients had an earlier onset than Asian ones (3.3±1.9 month v.s. 9±6.2 year) and the former progressed so rapidly that 75% of them succumbed to the disease within 1 year. The haplogroup of previously reported LS patients presented in Table I is regrettably not available. The Chinese pedigree presented herein was characterized as haplogroup N9a, which is not a common haplogroup in European populations. Individuals of the N9a haplogroup were reported to be resistant or protected against certain metabolic diseases such as type 2 diabetes (18,19) and metabolic syndrome (20). Thus, it is likely that N9a has an additive effect to the insusceptibility to LS. It is notable that another Han Chinese pedigree with the mutation m.10197 G>A, but in a different genetic background (haplogroup D4b) presented a different phenotype, of Leber hereditary optic neuropathy and dystonia (LDYT) (21).
The prognosis of LS is generally poor, and no effective treatment is available for this syndrome. The coenzyme Q10 has a crucial role as the electron acceptor in complexes I and II of the mitochondrial electron transport chain. Coenzyme Q10 affects the apoptosis of mitochondria though a number of mechanisms, including interference with mitochondrial depolarization (22), protection against reactive oxygen species (23), intervention in the production of ceramide (24), and mitochondrial protein uncoupling (25). Q10 is the most common drug used for treatment of mitochondrial diseases. In the present study, the condition of the patient was not serious. However, the typical brain lesions, particularly lower brainstem lesions, indicated that the LS patient would suffer from severe brainstem dysfunction in the near future (26). Low-dose coenzyme Q10 (2 mg/kg/day) treatment was applied on the proband as an experimental treatment, although patients with LS have rarely shown an improvement following administration of Q10 (27). Surprisingly, the patient showed a clear improvement in both clinical manifestations and the brain image, while its plasma lactate level also decreased down to a normal level. This is thus the first report on successful treatment of an LS patient with mutation m.10197 G>A. It is likely that the coenzyme Q10 may attenuate the mitochondrial dysfunctions caused by the m.10197 G>A mutation.
In conclusion, the present study reported the first Chinese LS pedigree with mutation m.10197 G>A, and discussed a number of factors potentially affecting the pathogenic potential of the mutation m.10197 G>A and contributing to the phenotypic variability of LS, such as molecular defects, ethnicity, haplogroup and environment. Our study further indicated that treatment with the coenzyme Q10 might be useful for LS patients with the mutation m.10197 G>A.