No retraction, correction or concern found when last checked on Oct 17, 2026.
Mitochondrial optic neuropathies – Disease mechanisms and therapeutic strategies
Patrick Yu-Wai-Man, Philip G. Griffiths, Patrick F. Chinnery
Progress in Retinal and Eye Research · 2011 · Review · CC BY
doi:10.1016/j.preteyeres.2010.11.002
Abstract
Leber hereditary optic neuropathy (LHON) and autosomal-dominant optic atrophy (DOA) are the two most common inherited optic neuropathies in the general population. Both disorders share striking pathological similarities, marked by the selective loss of retinal ganglion cells (RGCs) and the early involvement of the papillomacular bundle. Three mitochondrial DNA (mtDNA) point mutations; m.3460G>A, m.11778G>A, and m.14484T>C account for over 90% of LHON cases, and in DOA, the majority of affected families harbour mutations in the OPA1 gene, which codes for a mitochondrial inner membrane protein. Optic nerve degeneration in LHON and DOA is therefore due to disturbed mitochondrial function and a predominantly complex I respiratory chain defect has been identified using both in vitro and in vivo biochemical assays. However, the trigger for RGC loss is much more complex than a simple bioenergetic crisis and other important disease mechanisms have emerged relating to mitochondrial network dynamics, mtDNA maintenance, axonal transport, and the involvement of the cytoskeleton in maintaining a differential mitochondrial gradient at sites such as the lamina cribosa. The downstream consequences of these mitochondrial disturbances are likely to be influenced by the local cellular milieu. The vulnerability of RGCs in LHON and DOA could derive not only from tissue-specific, genetically-determined biological factors, but also from an increased susceptibility to exogenous influences such as light exposure, smoking, and pharmacological agents with putative mitochondrial toxic effects. Our concept of inherited mitochondrial optic neuropathies has evolved over the past decade, with the observation that patients with LHON and DOA can manifest a much broader phenotypic spectrum than pure optic nerve involvement. Interestingly, these phenotypes are sometimes clinically indistinguishable from other neurodegenerative disorders such as Charcot-Marie-Tooth disease, hereditary spastic paraplegia, and multiple sclerosis, where mitochondrial dysfunction is also thought to be an important pathophysiological player. A number of vertebrate and invertebrate disease models has recently been established to circumvent the lack of human tissues, and these have already provided considerable insight by allowing direct RGC experimentation. The ultimate goal is to translate these research advances into clinical practice and new treatment strategies are currently being investigated to improve the visual prognosis for patients with mitochondrial optic neuropathies.
Full text
From PubMed Central, U.S. National Library of Medicine, as current on Oct 17, 2026. Paragraph numbers are ours, so an answer can point at one.
Introduction
Inherited optic neuropathies affect at least 1 in 10,000 individuals and as a group, they represent an important cause of chronic visual impairment. Historically, these inherited optic nerve disorders were classified according to their mode of inheritance, and whether they were isolated or part of a more complicated syndromal variant. The identification of the underlying genetic defects in a large number of these inherited optic neuropathies now allows for a more accurate molecular classification, which has greatly benefited diagnostic accuracy and genetic counselling. The two classical prototypes are Leber hereditary optic neuropathy (LHON) and autosomal-dominant optic atrophy (DOA), which are both characterised by the preferential loss of retinal ganglion cells (RGCs). LHON is due to primary mitochondrial DNA (mtDNA) mutations, whereas the majority of patients with DOA harbour pathogenic mutations within the OPA1 gene, which codes for a mitochondrial inner membrane protein. As the genetic basis for other inherited optic neuropathies were uncovered, it became apparent that mitochondrial dysfunction is a recurrent molecular theme underlying the loss of RGCs in these disorders. In this review, we will cover basic aspects of mitochondrial biology and genetics, and how disruption of these critical biological systems contributes to optic nerve degeneration in different mitochondrial disease models.
Evolutionary origin
Mitochondria are ubiquitous intracellular organelles and they fulfil a fundamental role by providing most of the adenosine triphosphate (ATP) requirements of eukaryotic cells. The prevailing endosymbiotic hypothesis suggests that mitochondria evolved from aerobic α-proteobacteria, which were then gradually assimilated by primitive glycolytic eubacteria in a symbiotic relationship. During evolution, the α-proteobacteria gradually transferred the majority of their genetic material to the eubacteria’s nuclear chromosomes, creating the prototypal eukaryotic cell. Phylogenetic comparison of mtDNA between modern humans and other organisms, including Rickettsia prowazekii, an α-proteobacterium, supports this common evolutionary origin for the mitochondrial genome.
Structure
Mitochondria are tubular-shaped organelles bounded by an outer and an inner membrane, and these delineate two distinct compartments: an intermembrane space and an internal matrix space. The outer membrane allows passive diffusion of low molecular weight molecules up to 10 kDa, and this permeability is conferred by a family of channel-forming proteins known as porins, or voltage dependent anion channels (VDAC). The inner membrane is highly convoluted and these multiple infoldings, known as cristae, greatly increase its effective surface area. Compared to the outer membrane, the inner membrane is relatively impermeable except for specific active transport channels, allowing an electrochemical gradient to be established across this barrier. The inner membrane also contains a number of highly specialised proteins, including the respiratory chain complexes and members of the mitochondrial membrane protease family. The mitochondrial matrix compartment contains mtDNA molecules packaged within nucleoid structures, and it is also the site of multiple metabolic pathways essential for normal cellular function: the citric acid cycle, β-oxidation of fatty acids, steroid, amino acid, and pyrimidine biosynthesis.
Oxidative phosphorylation
The mitochondrial respiratory chain comprises four multi-subunit polypeptide complexes (I–IV) which are embedded within the inner mitochondrial membrane (Fig. 1). The production of ATP is tightly regulated and it is the end-product of a process known as oxidative phosphorylation (OXPHOS). Acetyl-CoA, an intermediate product of glycolysis and β-oxidation, is metabolised further by the citric acid cycle to the reducing equivalents nicotinamide adenine dinucleotide hydrogen (NADH) and flavin adenine dinucleotide hydrogen (FADH2). NADH and FADH2 are then re-oxidised by donating electrons to complexes I and II respectively. The energy released by the shuttling of these high energy electrons along the respiratory chain complexes allows protons to be pumped from the matrix compartment into the intermembrane space. Two additional carriers, ubiquinone (Co-enzyme Q10) and cytochrome c, also play critical roles in the efficient transfer of electrons through these successive oxidation–reduction reactions. The electrochemical gradient generated across the inner mitochondrial membrane is then used by complex V (ATP synthase) to catalyze the conversion of adenosine diphosphate (ADP) and inorganic phosphate (Pi) to ATP.
Mitochondrial genetics
Mitochondria are unique in having their own genome in the form of a circular, double-stranded molecule 16,569 bp long (Fig. 2). It is a high-copy number genome with hundreds to thousands of mtDNA molecules per cell, depending on their specific energy requirements. MtDNA replicates continuously and this process is independent of nuclear genome replication, occurring in both mitotic and post-mitotic cells. The mitochondrial genome codes for 2 ribosomal RNAs (12S and 16S rRNA), 22 transfer RNAs (tRNAs), and 13 polypeptide subunits of the respiratory chain complexes. The majority of these genes is located on the H-strand, with only MTND6 and eight tRNA genes found on the L-strand. MtDNA is a very compact molecule with overlapping gene regions devoid of introns, and a small 1.1 kb non-coding region, known as the D-loop, which is involved in mtDNA transcription and replication.
Mitochondrial haplogroups
The mitochondrial genome accumulates mutations at a significantly faster rate compared to the nuclear genome, and several factors contribute to this higher mutational rate: the absence of protective histones, the lack of effective repair mechanisms, the high mtDNA replication rate increasing the likelihood of errors, and the close proximity of mtDNA molecules to the respiratory chain complexes where they are exposed to high levels of reactive oxygen species (ROS). The mutational rate varies between different mtDNA regions, and it is much faster within the two hypervariable regions (HVR I and II) of the D-loop where a mutation is estimated to occur every 30 maternal generations. MtDNA is therefore highly polymorphic, and during human evolution, a number of relatively benign mitochondrial sequence variants have become fixed in different populations. As mtDNA is maternally inherited, these polymorphisms have accumulated sequentially along radiating female lineages, following the pattern of human migration from Africa into the various continents some 150,000 years ago. The human phylogenetic tree contains 18 major mtDNA haplogroups, and these comprise a total of 497 haplogroup-defining polymorphic variants. Individuals of European ancestry belong to one of nine haplogroups: H, I, J, K, T, U, V, W and X, with haplogroup H accounting for nearly half of all cases.
Heteroplasmy and mutational threshold
There are about 10,000 mtDNA molecules per cell, with each mitochondrion containing multiple copies. Two possible situations can therefore arise, known as homoplasmy and heteroplasmy. In the heteroplasmic state, two or more mtDNA variants are present at a specific nucleotide position, and the same phenomenon can also occur for mtDNA re-arrangements such as deletions. Most mtDNA mutations are heteroplasmic, a feature which supports the concept of a mutational threshold for pathogenicity. The relationship between mutational load and respiratory chain activity has been extensively investigated in different tissues, and the deleterious consequences of most mtDNA mutations on OXPHOS usually become apparent when the proportion of the mutant species exceeds 60–80%. There are mutation- and tissue-specific variations in this biochemical threshold, and although these could account for the pattern of organ involvement and clinical severity associated with a particular mtDNA defect, the molecular mechanisms are likely to be much more complex.
Nuclear-mitochondrial interactions
Mitochondria only have limited autonomy and they rely heavily on the nuclear genome for the majority of their structural and functional subunits (Fig. 3). Mitochondrial disorders can therefore arise secondary to both primary mtDNA mutations and nuclear genetic defects which disrupt mitochondrial-related proteins. In 2001, the first nuclear genes, POLG1 and PEO1, were identified among families with autosomal-dominant chronic progressive external ophthalmoplegia (CPEO) associated with multiple mtDNA deletions. Since then, the number of genes causing nuclear mitochondrial disorders has expanded continuously (Table 1), allowing significant progress to be made in elucidating the fundamental mechanisms that underpin mitochondrial physiology in both normal and disease states. As a result, we have gained a better understanding of the complex interactions between subunits of the respiratory chain complexes, and the crucial role played by accessory proteins in ensuring their proper assembly and stability along the inner mitochondrial membrane. These sometimes rare neurodegenerative and metabolic disorders have also provided important insights into the molecular components required for mtDNA maintenance, and the translational machinery that regulates intra-mitochondrial protein synthesis.
Nuclear mitochondrial disorders represent an important group of human diseases. They often pose significant diagnostic challenges related to their genetic and phenotypic heterogeneity, but they are increasingly being recognised, helped by greater clinical awareness and easier access to molecular genetic testing. A common feature shared by all these disorders is impaired mtDNA maintenance, which can lead to a reduction in mtDNA copy number, the accumulation of high levels of somatic mtDNA mutations, or both. The identification of these quantitative and qualitative mtDNA abnormalities in diagnostic specimens is therefore a key finding, suggesting an underlying nuclear defect, and helping to direct appropriate molecular investigations. MtDNA depletion is the pathological hallmark of several early-onset mitochondrial syndromes, and the clinical prognosis is often poor, due to the marked bioenergetic crisis caused by such a gross reduction in mtDNA copy number. Interestingly, the observed mtDNA depletion can be highly tissue-specific, which partly explains the variability in disease presentation and severity.
A mosaic pattern of cytochrome c oxidase (COX) deficient fibres is frequently observed in muscle biopsies of patients with both primary mtDNA and nuclear mitochondrial disorders, with some of these fibres exhibiting abnormal accumulation of mitochondria in the subsarcolemmal space, giving the classical appearance of “ragged-red fibres” (RRFs) (Fig. 4). For nuclear genetic defects involving POLG1, POLG2, PEO1, SLC25A4, TYMP, and more recently OPA1, the COX-defect is secondary to the accumulation of multiple mtDNA deletions, which have clonally expanded within individual cells to reach suprathreshold levels >70%. These deleted mtDNA species can be detected in homogenate DNA samples with Southern blot and long-range polymerase chain reaction (PCR), or more accurately quantified at the single-fibre level using real-time PCR assays. As most mtDNA deletions involve critical tRNA and protein-encoding genes, OXPHOS is adversely affected, and this eventually leads to apoptotic cell loss and tissue dysfunction. Mutations in POLG1 and TYMP have also been linked with the accumulation of somatic mtDNA point mutations. Although still speculative and controversial, these point mutations could compromise the replication machinery located within the D-Loop, thereby contributing to the formation of mtDNA deletions. It is fascinating that different mutations within the same gene can result in such a varied spectrum of secondary mtDNA abnormalities. The clarification of the secondary factors which dictate whether depletion, deletions, or point mutations predominate will provide crucial insights into the underlying disease mechanisms in nuclear mitochondrial disorders.
Epidemiology
The North of England has been relatively stable in terms of migratory flux, with a population of about three million, predominantly white, inhabitants (Fig. 5). As a result of the nature of healthcare provision in this region, over the past 20 years, patients with unexplained visual failure and suspected inherited optic neuropathies have been referred to the neuro-ophthalmology (PYWM and PGG) and neurogenetics (PFC) services in Newcastle upon Tyne for further assessment. This centralised referral pattern, in addition to active contact tracing, allowed us to determine for the first time the prevalence of LHON in a defined geographical region. We found a minimum estimate of 1 in 31,000, and fairly comparable prevalence figures of 1 in 39,000 and 1 in 50,000 have since been reported in Dutch and Finnish population studies, respectively. In Australia, the cause of visual impairment for about 2% of individuals on the national blind registry was optic atrophy secondary to LHON.
Primary mitochondrial DNA mutations
The majority of patients with LHON (90–95%) harbour one of three primary mtDNA point mutations: m.3460G>A, m.11778G>A, and m.14484T>C. The m.11778G>A mutation was identified in 1988 by, and it is of special historical significance, being the first mtDNA substitution confirmed to cause human disease. The most prevalent LHON mutation in Northern Europe, Australia, and the Far East is m.11778G>A, but as a result of a founder event, m.14484T>C is the most common mutation (87%) among French Canadians. A number of other pathogenic mtDNA LHON variants have since been reported (Table 2), with some still awaiting full confirmation for pathogenicity, having been identified in only single families. The distribution of these rarer mtDNA defects is not uniform, and the MTND1 and MTND6 gene regions are thought to be “mutational hotspots”, harbouring other LHON-causing mutations, in addition to m.3460G>A and m.14484T>C. On close questioning, up to 60% of affected individuals report other family members with a pattern of early-onset visual failure, and a detailed family history should always be sought in suspected inherited optic neuropathy cases. De novo m.3460G>A and m.14484T>C mutations have been reported in LHON, but these are rare. Most presumed singleton cases are therefore probably due to difficulties in tracing back a more extensive family history.
As a follow-on to our original epidemiological study, about 3000 umbilical cord blood samples from a local birth cohort in the North of England were screened for specific mtDNA point mutations. Nine healthy neonates were found to harbour one of the three primary LHON mutations, about 1 in every 350 births. In six of these neonates, the mutation was heteroplasmic, and of these cases, four were present at mutational levels less than 70%. There is clearly a large pool of these primary LHON mutations in the general population, and for heteroplasmic variants, the “mitochondrial bottleneck” will introduce shift in mitochondrial allele frequencies among future generations, directly influencing the risk of disease expression. The “mitochondrial bottleneck” is thought to be a protective mechanism which allows the rapid removal of deleterious mtDNA mutations from the genetic pool. The fertilised oocyte contains over 100,000 mtDNA molecules and during the early stages of development, there is a dramatic reduction in mtDNA copy number, down to 200–2000 copies before mtDNA replication is re-initiated. This decrease in the number of mitochondrial genomes repopulating the offspring of the next generation causes a sampling effect and accounts for the rapid changes in heteroplasmy levels. A pathogenic mtDNA variant would either be lost during transmission to the next generation, or it would quickly reach suprathreshold levels within an oocyte, increasing the likelihood of developmental arrest and its elimination. Even if a mature oocyte carrying a high proportion of the mutant species is successfully fertilised and a live birth results, there is a high probability that the affected individual’s fertility will be subnormal, which again serves to limit the transmission of mtDNA mutations. This situation clearly does not apply to LHON, as there is no evidence that mutational carriers have reduced fertility, and the majority of individuals, both affected and unaffected, are homoplasmic for the mtDNA mutation (Section 3.4.1).
Pre-symptomatic phase
In some asymptomatic LHON carriers, fundal abnormalities such as telangiectatic vessels around the optic discs, and fluctuating levels of retinal nerve fibre layer oedema have previously been observed. More detailed psychophysical testing can also uncover more subtle features of optic nerve dysfunction in some individuals, with loss of colour discrimination along the red–green axis, minimal central visual field changes on automated static perimetry, reduced contrast sensitivity, and subnormal visual electrophysiology.
Acute phase
Disease onset among LHON carriers is characterised by acute, painless loss of central vision, which is bilateral in about 25% of cases. If unilateral, the fellow eye is usually affected within six to eight weeks. There are rare cases of unilateral optic neuropathy, but these are the exceptions, second-eye involvement in LHON occurring invariably within 1 year of disease onset. The majority of carriers become symptomatic in the second and third decades of life, and over 90% of carriers who will experience visual failure will do so before the age of 50 years. However, visual deterioration can occur anytime during the first to the seventh decade of life and LHON should be part of the differential diagnosis for all cases of bilateral, simultaneous or sequential optic neuropathy, irrespective of age, and especially in male patients. Although one report identified females harbouring the m.11778G>A mutation as having a slightly increased age of onset compared to other groups, gender and mutational status are not thought to significantly influence the timing or initial severity of visual loss.
Visual loss worsens over a period of four to six week, and it is severe, dropping to levels of 6/60 or worse, with a dense central or centrocaecal scotoma, and marked impairment in colour vision. Importantly, the pupillary light reflexes are thought to be relatively preserved in affected LHON patients compared with the extent of visual loss, and this can be a useful clinical sign. In the acute stage, dilated fundal examination can be particularly informative, classical LHON cases exhibiting several distinct abnormalities such as vascular tortuosity of the central retinal vessels, swelling of the retinal nerve fibre layer, and a circumpapillary telangiectatic microangiopathy. However, it must be stressed that in about 20% of LHON cases, the optic disc looks entirely normal, and these patients are sometimes labelled as having functional visual loss.
Chronic phase
Within six weeks, optic nerve pallor becomes apparent, initially more marked temporally due to early axonal loss within the papillomacular bundle. Pathological cupping of the optic disc can occur with more extensive loss of RGC axons, and it is not an uncommon finding in longstanding LHON cases. If a patient is first assessed at this late stage, it can be difficult to exclude compressive, infiltrative or inflammatory causes of a bilateral optic neuropathy, especially when there is no convincing maternal history of early-onset visual failure. The results of molecular genetic testing in some diagnostic laboratories can take up to 2 months, and in these cases, the appropriate investigations, including neuroimaging of the anterior visual pathways, should not be delayed in order to exclude the possibility of reversible causes.
Visual prognosis
LHON causes significant visual impairment and in the majority of cases, visual recovery is minimal, the patient remaining within the legal requirement for blind registration. In the first year following disease onset, visual fields can improve with the appearance of small islands of vision. These fenestrations can help with scanning vision, especially if the central scotoma becomes concurrently less dense. The likelihood of visual recovery is greatest with the m.14484T>C mutation, and least with the m.11778G>A mutation, the m.3460G>A mutation having an intermediate visual prognosis. To objectively document the level of visual handicap experienced by LHON patients, we used the well-validated Visual Function Index (VF-14) questionnaire in a large cohort of 125 LHON pedigrees. LHON had a negative detrimental impact on most activities of daily living, and quality of life, as assessed by the overall VF-14 score, was the worst compared with other acquired and inherited ophthalmological disorders.
Extra-ocular LHON features
Although visual failure is the cardinal clinical feature, cardiac arrhythmias and neurological abnormalities such as peripheral neuropathy, myopathy, dystonia, and myoclonus have been reported to be more common among LHON carriers compared to controls. In a small number of families from Holland, Australia and North America, the reported extra-ocular features were particularly severe, with variable combinations of psychiatric disturbances, spastic dystonia, ataxia, and juvenile onset encephalopathy complicating the optic neuropathy. The phenotypic severity of these so-called “LHON plus” syndromes has been linked with specific mtDNA variants at m.4160T>C, m.11696A>G and/or m.14596T>A, and m.14459G>A. Two mtDNA point mutations affecting complex I activity, m.3376G>A and m.3697G>A, have also been identified in individuals with overlap clinical features of LHON and mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS).
originally described an intriguing association between the m.11778G>A primary mutation among female LHON carriers and demyelination. Following the onset of visual loss, these patients developed clinical and neuroimaging features indistinguishable from multiple sclerosis (MS), with characteristic periventricular white matter lesions on magnetic resonance imaging (MRI), and unmatched oligoclonal bands in their cerebrospinal fluid. Since this first description of Harding’s disease, further evidence has emerged in LHON and other mitochondrial disorders, which suggest that this association is unlikely to be a chance occurrence. LHON female carriers are twice more likely to develop an MS-like illness compared with male carriers, and although there is a preponderance for the m.11778G>A mutation, this phenotype has also been observed with the m.3460G>A and m.14484T>C primary mutations. More robust confirmatory epidemiological studies are required, but demyelination has been estimated to affect up to 1 in 20 LHON carriers, which is fifty times higher than the prevalence of MS in the general population. It has not yet been determined whether subclinical white matter MRI changes are present in asymptomatic LHON carriers or those who only manifest pure optic nerve involvement. If present, these again would suggest more widespread central nervous involvement in LHON, which becomes clinically manifest in only a subgroup of at-risk individuals. Interestingly, the proton magnetic resonance spectroscopic (1H-MRS) profile of both affected and unaffected LHON carriers were found to be abnormal compared to healthy controls, with reduced creatine (Cr) and N-acetylaspartate (NAA) levels in normal-appearing white matter regions, suggesting an underlying mitochondrial metabolic deficit. A number of pathophysiological mechanisms have been put forward linking RGC loss, oligodendrocyte survival, and mitochondrial dysfunction in patients with LHON and MS-like features (Section 8).
Incomplete penetrance and gender bias
Two key features of LHON still remain unexplained; the marked incomplete penetrance and the significant gender bias in disease predisposition, with only 50% of male and 10% of female carriers eventually losing vision in their lifetime. The primary LHON mutation is a prerequisite, but secondary factors are clearly modulating the risk of visual loss. Their identification has proven challenging, and the accumulated evidence favours a complex disease model, with both genetic and environmental factors interacting to precipitate optic nerve dysfunction.
Mitochondrial genetic factors
Among heteroplasmic LHON carriers, visual loss only occurs if the mutational load exceeds 60%, the threshold required for triggering a bioenergetic defect. However, incomplete penetrance is still observed among heteroplasmic carriers harbouring suprathreshold mutational levels, and over 80% of all LHON pedigrees are homoplasmic for the primary mtDNA mutation. Another possible mitochondrial modulating factor is the haplogroup background on which the LHON mutation is segregating. In a meta-analysis of 159 Caucasian LHON pedigrees, there was a significantly increased risk of visual failure when the m.11778G>A and m.14484T>C mutations occurred on a haplogroup J background, whereas m.3460G>A carriers were more likely to experience visual loss if they belonged to haplogroup K. A protective effect was conferred by haplogroup H, but only among m.11778G>A mutational carriers. The mitochondrial background also influenced the clinical expression of the m.11778G>A mutation among mainland Chinese LHON carriers, with haplogroup M7b1’2 increasing the risk of disease conversion, and haplogroup M8a having a protective effect. MtDNA haplogroups are defined by combinations of various polymorphic substitutions within the mitochondrial genome (Section 2.5). Some of these are non-synonymous, and they result in amino acid changes within mitochondrially-encoded subunits of the respiratory chain. Although it is convenient to view them as separate entities (Fig. 1), mitochondrial respiratory chain complexes do not exist in isolation, but they interact closely with one another forming so-called supercomplexes. Although speculative, these amino acid changes could induce subtle conformational changes, which affect the assembly and stability of these putative supercomplexes. In support of this hypothesis, cybrid cell lines harbouring the m.11778G>A mutation had a lower oxygen consumption and a longer doubling time on a haplogroup J background, compared with other mtDNA haplogroups. However, the link between specific mtDNA haplogroups and the risk of visual failure in LHON is not entirely clear-cut. A study of South-East Asian m.11778G>A LHON pedigrees found no significant association between specific mtDNA polymorphisms and the risk of developing overt optic nerve dysfunction. Similarly, using in vivo 31P-MRS measurements, haplogroup J did not induce a more pronounced mitochondrial biochemical defect in the brain and skeletal muscle of affected m.11778G>A mutational carriers.
Nuclear genetic factors
The marked male bias seen in LHON cannot be explained by mitochondrial genetic factors. Based on an extensive analysis of 31 large pedigrees totalling more than 1200 individuals, have proposed a two-locus model for visual failure in LHON. The segregation pattern was consistent with a visual-loss susceptibility gene on the X-chromosome, acting in synergy with the primary mtDNA mutation to precipitate visual loss among at-risk carriers. Male carriers have only one X-chromosome, and unlike female carriers, they cannot compensate for the inheritance of a putative X-linked visual-loss susceptibility allele. Three studies using microsatellite markers have now confirmed significant linkage on the X-chromosome, with some of these candidate regions showing areas of overlap (Figs. 6 and 7). The actual gene or genes involved have yet to be identified, and more sophisticated bioinformatic tools are currently being applied for candidate gene analysis and to narrow down specific areas of interest. LHON could be an even more complex disorder than originally considered and the existence of autosomal nuclear modifiers remains a distinct possibility. A recent genome-wide study of nine large m.11778G>A Thai pedigrees found evidence of significant linkage on areas of chromosomes 3, 12, 13, and 18. Candidate gene regions were analysed with a tagging single nucleotide polymorphism (SNP) methodology, and 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. 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.
Hormonal factors
Although much attention has been focused on possible secondary genetic modifiers in LHON, hormonal factors could also influence phenotypic expression. This hypothesis has recently been investigated by using osteosarcoma-derived cybrid cell lines harbouring one of the three primary LHON mutations: m.3460G>A, m.11778G>A, and m.14484T>C. These mutant cybrids exhibited elevated ROS levels, decreased mitochondrial membrane potential, increased rates of apoptosis, and hyper-fragmented mitochondrial networks compared with controls. Interestingly, treatment with 17β-oestradiol had a mitigating effect on these pathological features. In addition, supplementation of these LHON cybrids with 17β-oestradiol led to increased cellular levels of the anti-oxidant enzyme superoxide dismutase (SOD) and to more efficient mitochondrial biogenesis. These results are very interesting, providing another explanation for the protective effect of female gender on LHON penetrance, and supporting the possible therapeutic use of oestrogen-like compounds in this disorder.
Environmental factors
Two pairs of discordant monozygotic twins have been described, where one sibling has remained visually unaffected on long-term follow-up. These rare observations support an environmental component to the pathophysiology of optic nerve dysfunction in LHON, and there is increasing evidence in the literature supporting this hypothesis. The role of smoking and alcohol in LHON has been studied in a number of relatively small case-control studies, with contradictory findings. In one study, which included affected and unaffected siblings from 80 LHON sibships, high alcohol and tobacco consumption were not linked with an increased likelihood of visual failure. To further clarify this important issue, we conducted a multi-centre study of potential triggers in LHON, comparing the environmental exposure between 196 affected and 206 unaffected carriers. Smoking was strongly associated with an increased risk of visual loss, and interestingly, there was a dose-response relationship, with the risk of visual loss being much greater in heavy smokers compared to light smokers. There was also a trend towards an increased risk of visual failure among heavy drinkers, but this effect was not as strong as smoking. Based on these results, LHON carriers should be strongly advised not to smoke and to moderate their alcohol intake, especially avoiding binge drinking episodes. Although no functional studies were performed, smoking could further impair mitochondrial OXPHOS, either through a direct effect on complex I activity, or by reducing arterial oxygen concentration. Several other environmental triggers have been reported in LHON, including head trauma, acute physical illness, psychological stress, occupational exposure to chemical toxins such as 2,5-hexanedione, antiretroviral drugs, and anti-tuberculous agents. Most of these reports are clinical descriptions and they do not provide conclusive evidence for a causal relationship. However, have shown that 2,5-hexanedione had a mitochondrial toxic effect on LHON cybrids harbouring the m.11778G>A and m.14484T>C primary mutations. Of particular interest, an increased sensitivity to undergo apoptosis was noted on the haplogroup J mtDNA background, further highlighting the possible synergistic interactions between environmental and genetic risk factors.
Biochemical defect in LHON
All the three primary LHON mutations involve complex I subunits and their impact on mitochondrial OXPHOS has been extensively investigated using a wide range of cell types and biochemical assays. These in vitro studies have not always been consistent regarding the extent of the respiratory chain defect in LHON, but the overall evidence supports a predominant impairment in complex I-driven ATP synthesis (Table 3). In vivo 31P-MRS studies have also confirmed an underlying biochemical defect in LHON, with the m.11778G>A mutation resulting in the most pronounced reduction in mitochondrial ATP synthesis, followed by m.14484T>C, and m.3460G>A. With the caveat that RGCs were not directly assessed, an interesting observation from all these studies is that no significant difference in biochemical profile has ever been clearly demonstrated between affected and unaffected LHON carriers.
Oxidative stress leads to free radical production and ROS levels were found to be significantly increased in transmitochondrial cybrids carrying one of the three primary LHON mutations. LHON patients also have reduced α-tocopherol/lipid ratios and elevated 8-hydroxy-2-deoxygaunosine levels in their blood leukocytes, both of these biomarkers being indicative of increased ROS production. LHON cybrids have impaired EAAT1 (Excitatory amino acid transporter 1) activity, which is highly relevant to RGC survival, as these transporters are actively involved in the uptake of glutamate into Muller cells of the inner retina. The glutamate transport defect could be partially reversed by supplementing these LHON cybrids with antioxidants, supporting the use of these compounds in RGC neuroprotection. Increased ROS and glutamate excitotoxicity are potent inducers of apoptotic cell death, and in LHON, this process is likely to be both Fas-dependent and caspase-independent.
Epidemiology
DOA affects at least 1 in 35,000 individuals in the North of England, a figure which is comparable to LHON. Based on the Danish Registry for Hereditary Eye Diseases, the prevalence of this optic nerve disorder in Denmark has been estimated at 1 in 12,000, a higher figure which could reflect the methodological criteria used. This study was also performed in the pre-molecular era and patients were included based on a clinical diagnosis of DOA.
Clinical features
Visual loss in DOA is insidious, invariably starting in the first two decades of life, and with a mean age of onset of 6–10 years. Compared with LHON, it has a milder phenotype and up to 1 in 4 patients are visually asymptomatic, optic atrophy only being detected through contact tracing of an affected proband. Mean visual acuities of 6/18–6/60 have been reported in various case series, with visual acuities ranging from 6/6 to the detection of hand movement. There is a marked inter- and intra-familial variability in the rate of disease progression, but a significant proportion of patients (50–75%) will experience further worsening of their visual function in later life. Although the overall visual prognosis is better compared with LHON, DOA still causes significant visual impairment with half of all patients failing the driving standards and being registered legally blind.
Most patients with DOA have a generalised dyschromatopsia and only a minority (10%) has pure tritanopia, which used to be considered a pathognomonic feature of this optic nerve disorder. Due to the primary involvement of the papillomacular bundle, the most common visual defects are central, centrocaecal, and paracentral scotomas, with sparing of the peripheral field. Both magnocellular and parvocellular pathways seem to be similarly affected in DOA. Similar to LHON, the retino-tectal fibres mediating the pupillary light reflex are relatively preserved, and patients do not normally exhibit an afferent pupillary defect.
Optic disc pallor in DOA can be either diffuse, involving the entire neuro-retinal rim, or it can show a characteristic temporal wedge (Fig. 8). Pallor of the neuro-retinal rim can be subtle and in one report, nearly a third of all affected patients had normal looking optic discs on slit lamp biomicroscopy. Based on our own experience, measurement of retinal nerve fibre layer (RNFL) thickness with optical coherence tomography (OCT) can be particularly helpful in equivocal cases, by confirming pathological RNFL thinning, especially in the temporal quadrant involving the papillomacular bundle. Although not characteristic, other morphological disc features have been reported in DOA; saucerisation of the neuro-retinal rim, peripapillary atrophy, and enlarged cup-to-disc ratios greater than 0.5, the latter often leading to an erroneous diagnosis of normal tension glaucoma.
Mutational spectrum
The majority (50–60%) of patients with DOA harbour mutations in the OPA1 gene (OMIM 165500), and over 200 pathogenic mutations have been identified (eOPA1 database at http://lbbma.univ-angers.fr/lbbma.php?id=9, Accessed 31st of August 2010) (Table 4). OPA1 consists of 30 exons spread over 100 kb of genomic DNA and alternative splicing of exons 4, 4b and 5b results in eight different mRNA isoforms. OPA1 mutations cluster in the GTPase domain (Exons 8–15) and dynamin central region (Exons 16–23), with single base-pair substitutions (69%) representing the most common mutational subtype, followed by deletions (26%), and insertions (5%). OPA1 testing is usually performed with PCR-based sequencing protocols, and although cost and access are important practical issues, about 10–20% of patients who are found to be negative using these screening methods will harbour large-scale OPA1 re-arrangements. The majority of OPA1 mutations result in premature termination codons, with truncated mRNAs which are unstable and mostly degraded by protective cellular mechanisms. The reduction in OPA1 protein level is a major disease mechanism in DOA, and the importance of haploinsufficiency is further emphasised by rare families with microdeletions which span the entire OPA1 coding region. However, about 30% of patients with DOA harbour missense OPA1 mutations, and those located within the catalytic GTPase domain are more likely to exert a dominant-negative effect (Section 4.4).
Other genetic loci have been identified in families with DOA phenotypes (Table 4), but of these, only the causative gene for OPA3 has been characterised. OPA3 mutations were first identified in Iraqi Jewish families with Type III 3-methylglutaconic aciduria (Costeff syndrome); an autosomal recessive neurodegenerative disorder characterised by optic atrophy, progressive neurodegeneration, increased urinary levels of 3-methylglutaconic acid, and elevated plasma 3-methylglutaric acid levels. Pathogenic mutations were subsequently found in two independent French families segregating both optic atrophy and premature cataract in an autosomal-dominant mode of inheritance (ADOAC). However, OPA3 mutations are likely to be extremely rare, especially in isolated optic atrophy cases. We did not identify any pathogenic OPA3 variants in two large case series of OPA1-negative individuals, even with the use of a comparative genomic hybridization (CGH) assay to screen for large-scale OPA3 re-arrangements.
Expanding clinical phenotypes
Even in the pre-molecular era, a number of pedigrees with DOA were described where optic neuropathy occurred in parallel with other clinical features such as sensorineural deafness and CPEO. OPA1 mutations have been confirmed as the causative genetic defects in these syndromal forms of DOA, but until recently, these were thought to be rare manifestations among isolated families. In our North of England inherited optic neuropathy cohort, DOA+ phenotypes were observed in 1 in 6 OPA1 carriers, which clearly indicate that these syndromal variants affect a significant patient subgroup. We then conducted a multi-centre study of 104 patients manifesting DOA+ to define the phenotypic spectrum and natural history of these additional neurological complications. Bilateral sensorineural deafness beginning in late childhood and early adulthood was the most frequently observed extra-ocular manifestation, affecting nearly two-thirds of all cases. Other prominent clinical features then developed from the third decade of life onwards; ataxia, myopathy, peripheral neuropathy and CPEO. Unexpectedly, OPA1 mutations were also identified in families previously labelled as having autosomal-dominant hereditary spastic paraplegia (HSP), and in individuals with visual failure complicated by a multiple sclerosis-like illness. Of note, there was a two- to three-fold increased risk of developing multi-system neurological disease with missense OPA1 mutations located within the GTPase domain, suggesting deleterious gain-of-function mechanisms. Although these syndromal DOA+ variants show significant phenotypic variability even within the same family, a consistent finding is a worse visual prognosis among this patient subgroup. These observations are of major pathophysiological importance, highlighting the widespread deleterious consequences of OPA1 mutations, not only for RGCs, but also for other neuronal populations, skeletal muscle, and extra-ocular muscle. A key feature of these OPA1 mutations is the induction of secondary mtDNA deletions, and their possible roles in triggering multi-system cellular dysfunction are currently being investigated.
OPA1 and OPA3 protein functions
OPA1 is highly expressed within the RGC layer, but it is a ubiquitous protein, and abundant levels have also been identified in photoreceptors, and other non-ocular tissues such as the inner ear and the brain. OPA1 belongs to a large family of mechanoenzymes, which is characterised by a highly conserved, dynamin GTPase domain. OPA1 is a transmembrane protein embedded within the mitochondrial inner membrane, and it mediates several interrelated cellular functions. One important aspect relates to its pro-fusion properties and unsurprisingly, the pathological hallmark of OPA1 mutations is mitochondrial network fragmentation, the isolated mitochondria becoming dysmorphic, with aberrant balloon-like enlargements, disorganised cristae, and paracrystalline inclusion bodies. GTP hydrolysis activates membrane tubulation and this GTPase activity is thought to be enhanced following interaction of OPA1 with inner membrane phospholipids such as cardiolipin. Cytochrome c molecules are normally sequestered within the tight cristae junctions, and as they leach out into the cytosol, the apoptotic cascade is potentiated contributing to cell death and tissue dysfunction. OPA1 also regulates OXPHOS by interacting directly with the respiratory chain complexes, controlling their delicate assembly, and facilitating the efficient shuttling of electrons between complexes. Fibroblasts harbouring OPA1 mutations have reduced mitochondrial membrane potentials and ATP synthesis was found to be significantly impaired, secondary to a predominantly complex I defect. With the use of in vivo phosphorus magnetic resonance spectroscopy (31P-MRS), these biochemical defects have also been demonstrated in the calf muscles of patients harbouring OPA1 mutations. Interestingly, it is now apparent that OPA1 plays a critical role in maintaining the integrity of the mitochondrial genome. High levels of COX-negative fibres have been identified in skeletal muscle biopsies from patients with DOA+ phenotypes, and this has been conclusively linked to the accumulation of multiple mtDNA deletions.
OPA3 has a mitochondrial targeting domain, and it was thought to localise to the mitochondrial inner membrane. However, a recent study challenged this view, suggesting instead that OPA3 is an integral component of the mitochondrial outer membrane, with its C-terminus exposed to the cytosol. Mitochondrial fragmentation was induced both with mutant forms of the protein, and following overexpression of the wild-type protein, indicating an important pro-fission role for OPA3. The cells studied had an increased sensitivity to various pro-apoptotic stimuli, a phenomenon which was also observed in fibroblasts collected from an affected patient belonging to one of the French ADOAC families.
Charcot-Marie-Tooth disease
Charcot-Marie-Tooth (CMT) disease is a heterogeneous group of inherited peripheral neuropathies, and as a group they are one of the most common inherited human disorders, affecting at least 1 in 2500 individuals. Both motor and sensory nerves are affected resulting in distal limb weakness, sensory loss, decreased deep tendon reflexes, and foot deformities. A specific CMT subtype, hereditary motor and sensory neuropathy type VI (HMSN-VI, OMIM 601152), is caused by MFN2 mutations (1p36.2). In addition to early onset, severe peripheral neuropathy, affected individuals develop progressive optic nerve dysfunction starting in later childhood. Visual acuity usually deteriorates to levels of 6/60 or worse, but a subset of patients with HMSN-VI can experience sometimes dramatic visual recovery in later life. MFN1 and MFN2 are mitochondrial outer membrane proteins with dynamin GTPase domains, and they share a remarkable degree of structural and functional complementarity with OPA1. These three proteins interact closely with each other to coordinate the various steps involved in mitochondrial membrane fusion. MFN2 is also thought to exert a direct influence on mitochondrial biogenesis by regulating the expression of nuclear-encoded respiratory chain subunits. Consistent with this hypothesis, fibroblasts from patients harbouring MFN2 mutations exhibit a mitochondrial coupling defect, with impaired membrane potential and reduced OXPHOS capacity.
Hereditary spastic paraplegia
The clinical hallmark of HSP is slowly progressive lower limb spasticity and weakness. Its prevalence has been estimated at 3–10 per 100,000 in Europe, and the age of onset varies according to the causative genetic defect. HSP is classified into pure and complicated forms, depending on whether additional clinical features are present besides spastic paraplegia, such as optic atrophy, ataxia, peripheral neuropathy, extrapyramidal deficits, and cognitive decline. HSP is genetically heterogeneous with 41 mapped loci. So far, 17 genes have been identified, providing valuable insights into the various pathogenetic mechanisms which trigger axonal degeneration along the corticospinal tracts.
OPA1 is cleaved by mitochondrial proteases following import into the mitochondrial intermembrane space, and these post-translational maturational steps are critical, generating long (L) and short (S) forms of the protein. On their own, the L and S forms of OPA1 have little functional activity, but when co-expressed, they complement each other, triggering mitochondrial network fusion. One of these key mitochondrial proteases is paraplegin, which is encoded by SPG7 (16q24.3). Mutations in SPG7 have been identified in an autosomal recessive form of HSP, and in some patients, bilateral optic neuropathy is a prominent clinical feature further complicating the neurological phenotype (HSP-7, OMIM 607259). Muscle biopsies from two severely affected individuals with HSP-7 showed classical histochemical changes of mitochondrial dysfunction with ragged-red fibres and COX-negative fibres. Transmission electron microscopy confirmed an accumulation of abnormal mitochondria containing paracrystalline inclusion bodies. Biochemical studies performed on cultured myoblasts harbouring SPG7 mutations have revealed a reduction in citrate synthase-corrected complex I activity, again suggesting that impaired OXPHOS plays an important role in the pathogenesis of HSP, and by extension the optic neuropathy observed in a subgroup of patients.
Friedreich ataxia
Friedreich ataxia (FRDA) is an autosomal recessive disorder caused by pathological GAA trinucleotide repeat expansions in the FXN gene (9q13-q21.1, OMIM 229300). The encoded protein frataxin is directed to the mitochondrial inner membrane and it is involved in the assembly of iron–sulphur clusters, which are critical components of the mitochondrial respiratory chain complexes. FXN mutations therefore impair OXPHOS and they also result in abnormal accumulation of intra-mitochondrial iron, which eventually reaches toxic levels. Because frataxin has anti-oxidant properties, cellular defences against ROS are impaired in FRDA, and this likely further exacerbates neuronal loss.
Patients with FRDA usually become symptomatic in the second decade of life, with progressive gait ataxia, loss of the deep tendon reflexes, dysarthria, distal limb weakness, pes cavus, scoliosis, and arrhythmias secondary to hypertrophic cardiomyopathy. In a recent study of 26 patients with genetically-confirmed FRDA, all patients had evidence of optic nerve dysfunction, although only five were visually symptomatic. The optic neuropathy differed from that observed in LHON and DOA, with a diffuse and progressive pattern of RNFL loss, and no preferential involvement of the papillomacular bundle. Interestingly, the pathological process in FRDA extended to the post-geniculate optic radiations, and the involvement of the anterior and posterior visual pathways seemed to proceed independently of each other. Altogether, these findings suggest that FXN mutations probably cause RGC loss via other disease mechanisms compared with OPA1 and the primary mtDNA LHON mutations.
Autosomal recessive non-syndromal optic atrophy
Autosomal recessive optic neuropathies are rare and the visual phenotype is usually overshadowed by other more prominent neurodegenerative features. Using homozygosity mapping, identified mutations in TMEM126A (11q14.1-q21, OMIM 612989) in a large, inbred, Algerian family segregating pure optic atrophy. TMEM126A is conserved in higher eukaryotes and it encodes a transmembrane mitochondrial protein, which is found at high levels within the RGC layer and optic nerve head. The exact localisation and function of this protein remain to be clarified, but TMEM126A mutations do not cause mitochondrial network fragmentation or mtDNA depletion. A novel nonsense TMEM126A mutation has since been reported where affected family members had subclinical auditory neuropathy in addition to progressive visual failure. The clinical phenotype associated with TMEM126A mutations could therefore expand further as more families with syndromal optic atrophy are screened for this particular gene.
Mitochondrial protein-import disorders
Mohr–Tranebjaerg syndrome or deafness-dystonia-optic neuronopathy (DDON) syndrome is caused by loss-of-function mutations in the TIMM8A gene (Xq22, OMIM 300356). The clinical phenotype includes prelingual or postlingual sensorineural deafness, dystonia and ataxia in late childhood, visual failure with optic atrophy from the age of 20 years, and cognitive decline with psychiatric disturbances before the age of 50 years. Electrophysiological studies indicate that the visual loss in DDON is secondary to RGC loss, and the visual prognosis is poor, with most patients registered legally blind by the age of 40 years. TIMM8A assembles as a 70 kDa hetero-oligomeric complex in the mitochondrial intermembrane space, and in conjunction with TIMM13, another mitochondrial membrane translocase, this complex facilitates the import and insertion of inner membrane proteins. COX-negative fibres were not present in muscle biopsy specimens from affected patients and there was no alteration in mitochondrial network morphology. However, there was evidence of an underlying mitochondrial biochemical defect, with reduced complex I–IV enzymatic activities.
Dilated cardiomyopathy with ataxia (DCMA) or 3-methylglutaconic aciduria, type V (OMIM 610198) is an autosomal recessive disorder found in the Dariusleut Hutterite population of Canada and the Northern United States. The clinical phenotype is characterised by growth failure, severe, early-onset cardiomyopathy, and a cerebellar syndrome with ataxia. Optic atrophy has been reported in some patients with DCMA, although it is not a cardinal feature of this syndrome. Homozygous mutations in the DNAJC19 gene (3q26.33) have recently been identified in these inbred families, a G–C transversion within the splice acceptor site of intron 3. The DNAJC19 protein localises to the mitochondrial compartment and although its subcellular localisation requires further investigation, it is thought to be the human orthologue of the yeast Tim14 protein. Tim14 is an integral mitochondrial inner membrane protein and together with other mitochondrial translocases, it actively participates in protein import into the mitochondrial matrix compartment. Both DDON and DCMA are therefore due to defective mitochondrial protein-import systems, and it is fascinating that in both disorders optic nerve involvement is observed, albeit as part of a more widespread multi-systemic degeneration.
Mitochondrial encephalomyopathies
Optic neuropathy can also develop in the following classical mitochondrial syndromes, although it is usually a secondary feature, overshadowed by other more prominent neurological and ocular manifestations: MELAS, myoclonic epilepsy and ragged-red fibres (MERRF), CPEO, KSS, maternally inherited Leigh syndrome (MILS), and mitochondrial neurogastrointestinal encephalomyopathy (MNGIE).
Overlapping phenotypes
The concept of mitochondrial optic neuropathies has greatly expanded over the years and this trend is set to continue as new genes with mitochondrial-related functions are identified in other inherited optic nerve disorders. What is truly remarkable is the overlapping phenotypes seen with this group of disorders, for example among OPA1 carriers, some patients develop neurological features indistinguishable from HSP, others develop a pattern of peripheral neuropathy with a similar disease course to CMT, and others still will develop a prominent cerebellar syndrome consistent with FRDA. These shared clinical features also apply to the optic nerve phenotype, as seen with LHON and DOA, the two most common inherited optic neuropathies diagnosed in the general population. As described in earlier sections, LHON and DOA have distinct clinical presentations which usually allow for their easy differentiation, but with the greater availability of genetic testing, it is now clear that a degree of phenotypic overlap exists. Although unusual, some patients with genetically-confirmed OPA1 mutations have been described with acute and even reversible visual loss, and a subgroup of LHON patients can present with a slowly progressive optic neuropathy, more suggestive of DOA.
Smoking, alcohol, and nutritional deprivation
Individuals with excessive alcohol and tobacco intake can develop a bilateral optic neuropathy, with slowly progressive visual loss, dyschromatopsia, and centrocaecal scotomas. This entity is sometimes referred to as “tobacco-alcohol amblyopia”, but there is persisting debate regarding its actual aetiology. It is possible that alcohol and tobacco per se are not the main or only causative factors, with other confounding variables such as nutritional deprivation contributing to the observed RGC-toxic effect. An epidemic of optic neuropathy was observed in Cuba in the early 1990s at a time of deteriorating socioeconomic conditions within this country. This outbreak of optic neuropathy has been linked with chronic malnutrition combined with high levels of tobacco and cassava consumption. Cassava contains naturally-occurring cyanogenic glucosides and the occurrence of cyanide poisoning has been well described, especially during periods of famine. In terms of disease mechanisms, it is interesting that primary mtDNA LHON mutations have been detected in patients labelled as having “tobacco-alcohol amblyopia”. Moreover, smoking has recently been identified as a strong risk factor for visual loss among LHON mutational carriers, with heavy smokers having a higher risk than light smokers. However, it should be noted that during the Cuban optic neuropathy outbreak, there was no evidence of increased disease penetrance in a large Cuban m.11778G>A family.
Chloramphenicol
With the advent of safe, broad spectrum antibiotics, systemic chloramphenicol is now rarely used for treating severe, disseminated infections. However, until the late 1970s, it was frequently used in the management of patients with cystic fibrosis, and a rare but serious complication noted in this group of patients was bilateral optic neuropathy. Post-mortem studies performed on patients who died shortly afterwards showed complete loss of RGCs within the distribution of the papillomacular bundle, with relative sparing of the nasal and peripheral retina. There was no evidence of inflammation within the optic nerve but there was mild gliosis and demyelination in both the pre- and post-laminar segments. Transmission electron microscopy of haemopoetic cells collected from patients with chloramphenicol-induced bone marrow suppression revealed swollen mitochondria with disrupted cristae architecture, in addition to abnormal intra-mitochondrial iron deposits. These findings strongly suggest that the pathological changes in chloramphenicol optic neuropathy are due to a similar mitochondrial toxic effect. Chloramphenicol is a powerful inhibitor of both bacterial and mitochondrial ribosomal protein synthesis, resulting in mitochondrial stress and a marked reduction in cellular ATP biosynthesis.
Linezolid
Linezolid is an oxazolidinone antibiotic that is increasingly being used to treat drug-resistant, gram positive organisms such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus (VRE). Overall, it is a safe antibiotic agent when used for up to 28 days, with only minor side effects reported in large Phase III clinical trials. However, about 20 cases of linezolid-associated optic neuropathy (LION) have been reported in the literature, and a common feature of this rare association seems to be an extended duration of treatment (5–50 months). Patients present with typical features of a bilateral optic neuropathy and at first presentation, visual acuities ranging from 6/9 to counting fingers have been reported. Early in the course of the disease, optic disc swelling and marked thickening of the RNFL have also been documented, the latter indicating axonal RGC swelling and strikingly reminiscent of the pathological process in acute LHON. Unlike linezolid-induced peripheral neuropathy which is irreversible, full visual recovery is possible following discontinutation of linezolid therapy. It is therefore crucial for clinicians to be aware of this possibly devastating complication among patients receiving linezolid treatment beyond the 28-day safe limit. Intriguingly, mitochondrial dysfunction seems to be central to the pathophysiology of LION, and a number of hypotheses have been proposed based upon the mechanism of action of linezolid. The antibacterial effect of linezolid is due to inhibition of bacterial protein synthesis, with the active moiety binding to the 23S rRNA of the 50S ribosomal subunit, thereby preventing the formation of the 70S RNA initiation complex. Mammalian ribosomes lack the 50S component and they are therefore not affected by linezolid. Although speculative, it is possible that mitochondrial ribosomes are susceptible to the sustained effect of long-term linezolid because of their closer evolutionary similarities to bacterial ribosomes. Consistent with this mechanism, mitochondrial respiratory chain activity was significantly depressed in a range of tissues; muscle, liver, and kidney, biopsied from a patient who developed optic neuropathy, encephalopathy, skeletal myopathy, lactic acidosis, and renal failure after a 4-month course of linezolid. This finding was further substantiated in an experimental rat model, with linezolid inducing a dose- and time-dependent decrease in the activity of complex I, and to a lesser extent that of complex IV.
Erythromycin
reported on a 23-year-old male who developed bilateral optic neuropathy following treatment with erythromycin. His visual acuity was severely reduced to counting fingers in both eyes and bilateral diffuse optic nerve pallor was observed, more marked temporally. A homoplasmic m.11778G>A mutation was identified, and to investigate a possible link between erythromycin and the onset of visual failure in this LHON carrier, cybrid osteosarcoma cells were established harbouring the patient’s mtDNA. When cultured in media containing erythromycin, a marked impairment in mitochondrial protein synthesis was noted and this was a dose-dependent effect. Similar to linezolid, erythromycin binds to the 23S rRNA molecule of the 50S ribosomal subunit blocking the exit of the growing polypeptide chain. Erythromycin and, by extension, other macrolide antibiotics are therefore potential mitochondrial toxins and their use should be avoided in patients with mitochondrial genetic disorders, especially if other alternatives are available.
Ethambutol
Ethambutol is an important first-line drug used for treating tuberculosis and 2–6% of patients receiving this anti-mycobacterial drug will develop optic neuropathy, a serious side effect which is duration- and dose-dependent. Optic nerve involvement is rare with treatment duration of less than 2 months, and although visual function can improve following discontinuation of the drug, in some cases, it is irreversible. Several studies have confirmed a direct, toxic effect of ethambutol on RGC survival, and various pathophysiological mechanisms have been proposed including glutamate excitotoxicity, zinc-mediated lysosomal membrane permeabilisation, and disruption of mitochondrial complex IV activity through a copper-chelating action. were the first to describe the onset of bilateral, subacute visual loss in a LHON m.11778G>A carrier, 8 months following the initiation of treatment for pulmonary tuberculosis. Since then, other case reports of ethambutol-induced optic neuropathy among LHON carriers have been reported, suggesting a synergistic deleterious effect of this anti-mycobacterial drug on the background of a pre-existing pathogenic mtDNA mutation. Against this argument, the m.11778G>A mutation did not further enhance ethambutol cytotoxicity in NT2/D1 teratoma-derived cybrids, but this could simply reflect a tissue-specific effect, the latter being more pronounced in RGCs.
The link between ethambutol therapy and inherited mitochondrial optic neuropathies was further highlighted by. They unexpectedly identified a pathogenic OPA1 mutation in a male patient who developed bilateral, sudden, central visual loss 3 months after starting treatment for primary tuberculosis on a standard regimen of ehthambutol, isoniazid, and rifampicin. His visual acuity was severely affected, with caecocentral scotomas, and no visual recovery was noted following withdrawal of ethambutol. The acute presentation observed in this case is somewhat atypical for DOA and functional studies performed on cultured fibroblasts supported a role for ethambutol in precipitating an accelerated loss of RGCs. Compared with control fibroblasts similarly treated with ethambutol, a more marked biochemical uncoupling defect was observed, with loss of mitochondrial membrane potential, mitochondrial fragmentation, disturbed calcium homeostasis, and the accumulation of abnormal intracellular vacuoles.
Antiretroviral drugs
There are suggestive reports of visual loss being triggered among LHON carriers following the initiation of highly active antiretroviral therapy (HAART) for newly diagnosed HIV infections. These HIV patients were all male and harboured either the m.11778G>A or the m.14484T>C primary mtDNA mutations. The possible mechanisms have focused mainly on a particular class of antitretrovirals, the nucleoside analogue reverse transcriptase inhibitors (NRTIs), which can induce mitochondrial toxicity. NRTIs can inhibit mitochondrial polymerase gamma (POLG), a key component of the mtDNA replication machinery. A well reported side effect of NRTI therapy is mtDNA depletion, the resulting biochemical defect leading to lactic acidosis, hepatotoxicity, cardiomyopathy, generalised myopathy, and peripheral neuropathy. Fascinatingly, HAART has also been linked with the development of a CPEO phenotype in a limited number of HIV patients, with a gradual onset of ophthalmoplegia and ptosis. One attractive hypothesis is that patients harbouring specific SNPs within POLG are at a higher risk of developing mitochondrial toxicity to NRTIs, with the HIV virus itself having an additive detrimental effect, further exacerbating mitochondrial dysfunction. The mitochondrial protein profile of treatment-native HIV patients and those receiving HAART therapy have recently been investigated in blood leukocytes using a proteomic-based approach. Both groups of patients showed downregulated levels of prohibitins, important mitochondrial enzymes closely involved in OXPHOS and the maintenance of the mitochondrial network. Additional studies are needed to determine whether fibroblasts from HIV patients, on or off HAART therapy, show any evidence of mitochondrial network fragmentation. Given the emerging links between HIV, HAART therapy, and mitochondrial dysfunction, it is not unreasonable to speculate that LHON carriers would be at an increased risk of visual loss in this particular context, and should therefore be adequately counselled should they require treatment with antiretrovirals.
Normal tension glaucoma
Primary open angle glaucoma (POAG) affects over 60 million people worldwide and it is the second most common cause of legal blindness in developed countries. RGCs are lost at an accelerated rate compared with normal ageing, resulting in pathological disc cupping and visual field defects. POAG is a complex disease and several genetic susceptibility loci have been identified in populations from different ethnic backgrounds. The major risk factor for POAG is elevated intraocular pressure (IOP), but in about a third of patients, IOPs never exceed the statistical normal threshold of 21 mmHg. Although lowering IOP can reduce visual field progression in normal tension glaucoma (NTG), additional IOP-independent risk factors are considered to be more important in the pathophysiology of NTG than in high tension glaucoma (HTG). The pattern of RGC loss in these two disease subgroups has been extensively investigated, and in most studies, NTG patients tended to have steeper, more focal field defects, which were located closer to fixation compared with HTG patients. It is therefore not surprising that patients with DOA are often misdiagnosed as having NTG, especially if there is no family history of early-onset visual loss. Furthermore, disc pallor in DOA can be quite subtle, involving only the temporal segment, and about half of all patients have excavated discs, with cup-to-disc ratios greater than 0.5. Some investigators have even argued, rather provocatively, that NTG is in actual fact an inherited optic neuropathy representing a forme fruste of DOA.
OPA1 polymorphisms
OPA1 was an obvious candidate disease-susceptibility gene in glaucoma and we recently carried out a systematic screen of the entire coding region using a well-characterised POAG cohort from the North of England. No pathogenic sequence variants were identified ruling out OPA1 as a monogenic risk determinant for this disorder. Instead, we found a statistically significant association between two specific intronic OPA1 SNPs, IVS8+4c>t and IVS8+32t>c, and the risk of developing NTG but not HTG. Three other studies have found a similar association between these two OPA1 SNPs and NTG, supporting a possible role of the OPA1 protein in modulating disease susceptibility to glaucoma. However, these findings have not been replicated in other populations, and functional studies are needed to clarify the mechanisms by which OPA1 SNPs could exacerbate RGC loss and increase the risk of developing NTG.
OPTN mutations
Mutations in four genes have been identified in families with POAG: optineurin (OPTN, OMIM 602432), myocilin (MYOC, OMIM 610652), CYP1B1 (OMIM 601771), and WDR36 (OMIM 609669). OPTN mutations are present in about 17% of families with autosomal-dominant NTG, but overall they are relatively rare, accounting for less than 2% of sporadic cases. These studies, including the original report of OPTN mutations in NTG, also suggest that the p.M98K OPTN variant could be an important disease risk factor, being present in 10–15% of glaucoma subjects compared with only 2–4% of normal controls. In one study, the p.M98K variant was not found to be associated with DOA and LHON, and it did not influence the severity or risk of visual loss.
Intriguingly, OPTN mutations have also recently been identified in Japanese individuals from consanguineous marriages and with a familial form of amyotrophic lateral sclerosis (ALS). OPTN is a multifaceted protein and one of its important functions is to protect cells against high levels of oxidative stress by regulating the activation of NF-κB, an important component of the extrinsic apoptotic pathway. NF-κB is also found in both the cytosol and mitochondrial compartments and its upregulation inhibits cytochrome c oxidase and cytochrome b mRNA, disrupting the structural integrity of the mitochondrial respiratory chain and adversely affecting ATP synthesis. Furthermore, OPTN co-localises to the Golgi apparatus and it probably plays an important role in membrane trafficking and vesicular transport. The neuropathology in ALS is characterised by the degeneration of upper and lower motor neurones, and it is intriguing that OPTN mutations seem to have a predilection for long neuronal populations, including RGCs, with mitochondrial dysfunction being one of the likely pathogenetic mechanisms.
Demyelination, optic neuritis, and multiple sclerosis
MS is an inflammatory demyelinating disease of the central nervous system and as the disease progresses, extensive axonal loss is observed in the brain and spinal cord. The prevalence of MS varies worldwide and in populations of Northern European extraction, the average quoted figure is 1 in 1000. Up to half of all patients with MS will experience an episode of optic neuritis, and in 15–20% of cases, it is the first manifestation of an underlying demyelinative disorder. Optic neuritis classically presents acutely, with loss of central vision, marked dyschromatopsia, and pain on eye movement. Spontaneous visual recovery usually occurs two to three weeks after the first onset of symptoms and it is nearly complete within the next four weeks. The overall visual prognosis is very good and 1 year after an attack of optic neuritis, more than 90% of patients have recovered their premorbid level of visual acuity. However, in longitudinal studies of MS patients, including those without a previous history of optic neuritis, progressive RNFL thinning was noted on OCT, clearly indicating a chronic process of subclinical RGC loss.
Female LHON carriers are twice more likely to develop an MS-like illness than male carriers, despite the fact that the risk of visual loss is five times higher among males (Section 3.3.5). We propose two main hypotheses to account for this association. It is possible that subclinical disease in at-risk LHON carriers results in immunogenic mitochondrial peptides being exposed, which trigger or potentiate an autoimmune cascade against oligodendrocytes. Conversely, the autoimmune process could be the primary pathological factor and the deleterious consequences for oligodendrocytes are magnified by the presence of an underlying mtDNA defect, the latter putting additional stress on mitochondrial oxidative metabolism. The optic neuropathy in LHON patients who develop an MS-like illness is invariably severe with poor visual recovery and as such behaves differently from the natural history of acute demyelinative optic neuritis. Presumably, RGC axonal loss in these LHON-MS patients is amplified by dysfunctional oligodendrocytes leading to demyelination or failure of remyelination.
The link between mitochondrial dysfunction and MS is not limited to the primary mtDNA LHON mutations. We recently reported the occurrence of a more disseminated demyelinative process in two independent DOA families with genetically-confirmed OPA1 mutations. Compound heterozygous POLG1 mutations have also been recently identified in a 4-year-old child with presumed acute disseminated encephalomyelitis (ADEM), and in two patients in their thirties presenting initially with unilateral optic neuritis, which subsequently evolved into clinically-definite MS with classical white matter MRI abnormalities and unmatched CSF oligoclonal bands. In both adult MS cases, muscle biopsy revealed numerous COX-negative fibres, with prominent RRFs and multiple mtDNA deletions on long-range PCR analysis.
Optic nerve head morphology and disease susceptibility
Unaffected LHON carriers have larger optic disc areas compared with affected carriers and normal controls. Among affected LHON carriers, visual recovery was associated with larger vertical but not horizontal disc diameters, and there was a trend towards a better visual outcome with larger optic disc areas. About 10% of LHON carriers experience visual loss before the age of 10 years, and in these childhood cases, two patterns of visual loss have been described; an acute presentation similar to the one seen in adults, and a milder variant with a slowly progressive course. Compared to the slowly progressive group, children with acute LHON had worse visual acuities, thinner peripapillary RNFL thickness, and significantly smaller optic disc areas. The results from these two studies therefore suggest a possible link between optic disc anatomy and the phenotypic expression of the primary mtDNA LHON mutations. Similar to anterior ischaemic optic neuropathy (AION), a small crowded optic disc could predispose LHON carriers to a higher risk of visual loss. Transient and segmental RNFL swelling have been observed in asymptomatic LHON carriers, and with the advent of OCT imaging, thickening of the temporal RNFL has been documented more objectively, confirming these earlier clinical observations. The swelling of these RGC axons is secondary to axoplasmic stasis and it reflects the compromised bioenergetic state within these neurones, which are struggling to maintain energy-dependent cytoskeletal transport mechanisms (Section 10.3). If the optic nerve axons are congenitally packed within a relatively confined space, under the right circumstances, a compartment syndrome could result, further impeding axonal transport. As the fine capillary network around the lamina cribosa becomes compressed by the surrounding tissue oedema, the resulting ischaemia could set up a vicious cycle, which eventually precipitates the onset of visual loss. The concept of a “disc-at-risk” is also attractive in that it could explain the role of environmental factors such as heavy smoking in increasing the risk of disease conversion among LHON carriers. Long-term heavy smokers are more likely to have atherosclerotic vascular changes and these could further exacerbate a local hypoxic insult at the optic nerve head.
In one study, OPA1 carriers had significantly smaller optic disc areas compared with normal controls, and although speculative, these findings were ascribed to the role played by OPA1 in early embryonic development (Section 11.3.2). By accentuating developmental apoptosis, OPA1 mutations could result in a greater loss of RGCs and therefore a correspondingly smaller optic nerve head. In support of this hypothesis, OPA1 carriers have thinner peripapillary RNFL measurements than controls at all age groups, and extrapolating the data, it is likely that these patients had lower RGC counts at birth (Fig. 9). If OPA1 does exert a developmental influence on optic nerve head morphology, this could have important implications for other ocular disorders such as NTG and optic nerve hypoplasia. One could speculate that specific combinations of OPA1 variants could have a subtle effect on the biomechanical characteristics of the lamina cribosa, which is the principal site of IOP-related damage to RGC axons. The mechanical effect of IOP on load-bearing connective tissues is accentuated at the lamina cribosa, and this will vary depending on the unique susceptibility conferred by an individual’s optic nerve head configuration. In individuals with high-risk OPA1 genotypes and relatively weak optic nerve head biomechanics, axonal damage could occur at relatively low IOPs, increasing the risk of developing NTG. The aetiology of optic nerve hypoplasia is poorly defined and in 20% of cases, this abnormality occurs in isolation, without more widespread cerebral malformations or hypothalamic dysfunction. It would therefore be of great interest to screen for OPA1 in a cohort of patients with isolated optic nerve hypoplasia to determine whether some patients harbour pathogenic mutations, or whether there is any association with specific polymorphic sequence variants.
Anatomical considerations
Both OPA1 and the primary LHON mutations result in a mitochondrial biochemical defect, but the same situation applies to other cell types such as photoreceptors and substantia nigra neurones, which have similar, if not higher, oxidative requirements than RGCs. What other factors therefore contribute to this greater vulnerability of RGCs in mitochondrial optic neuropathies? The optic nerve has some rather peculiar anatomical constraints which could influence the pathological consequences of these genetic defects. One such feature is the fact that RGC axons only acquire a myelin sheath beyond the lamina cribosa. In the pre-laminar region, the generation and propagation of action potentials are therefore less efficient and require greater energetic input. To investigate this further, we performed a series of histological studies of human optic nerves, first staining them for mitochondrial COX and succinate dehydrogenase (SDH) enzyme activities. As we expected, the pre-laminar, unmyelinated optic nerve demonstrated much more intense COX staining compared with the post-laminar, myelinated segment, clearly indicating a differential concentration of mitochondria across the lamina cribosa (Fig. 10). The increased mitochondrial content within the unmyelinated RGC axons was associated with a high density of voltage gated sodium channels, reflecting the greater energy requirements for nerve conduction in the absence of myelination (Fig. 11). The pre-laminar segment of the optic nerve could therefore be viewed as a vulnerable metabolic “chokepoint”, magnifying the consequences of any impairment in mitochondrial OXPHOS, however subtle. Another important point to consider is that parvocellular RGCs within the papillomacular bundle have relatively small cross-sectional areas, which could put them at an even greater physiological disadvantage in terms of their mitochondrial reserve, compared with the larger magnocellular RGCs.
Another cranial nerve marked by such a dramatic transition between unmeylinated and myelinated segments is the cochlear nerve. Nearly two-thirds of all patients with DOA+ phenotypes develop sensorineural deafness. It is therefore very revealing that in two family members with optic atrophy and hearing loss due to a pathogenic OPA1 mutation, auditory investigations localised the defect to the terminal unmyelinated portions of cochlear nerve. Transtympanic electrocochleography revealed sub-threshold, abnormally-prolonged negative potentials within the unmyelinated auditory nerve terminals suggesting a failure to generate action potentials at the first node of Ranvier.
Mitochondrial network dynamics
The mitochondrial network is in a constant state of flux and it relies on a delicate balance between fusional and fissional forces. Disruption of mitochondrial fusion can have dramatic consequences on cell survival as clearly illustrated by OPA1 and MFN2 mutations. The role of mitochondrial network dynamics in maintaining cellular homeostatis was confirmed further when a heterozygous, dominant-negative, DLP1 mutation was identified in a female infant with a complex neurodegenerative phenotype characterised by microcephaly, abnormal brain development, optic atrophy, and disturbed metabolic function. The mitochondrial network in the patient’s cultured fibroblasts was elongated, forming tangled tubular structures, which were clumped around the nucleus. DLP1 codes for DRP1, a pro-fission, dynamin-related GTPase protein, and the mutation leads to a dominant-negative loss of function, resulting in unopposed mitochondrial fusion. A dysfunctional mitochondrial network can obviously impact on other cellular processes such as axonal transport, but as such, it still does not explain the selective vulnerability of RGCs. Some important clues were uncovered by, who transfected cultured rat RGCs and cerebellar granule cells (CGCs) with short interfering RNAs (siRNAs) against Opa1 transcripts. RGCs were much more susceptible to these Opa1-blocking siRNAs and they exhibited a greater degree of mitochondrial fragmentation compared with CGCs, reflecting the predominant optic nerve phenotype seen in patients with DOA.
Mitochondrial–cytoskeletal interactions
Mitochondria do not exist in isolation and an intricate cytoskeletal system is essential for their proper localisation to areas of increased energetic demands (Fig. 12). These mitochondrial–cytoskeletal interactions apply not only to the lamina cribosa, but also to other sites requiring the maintenance of a differential mitochondrial gradient such as the nodes of Ranvier. Axonal transport along the relatively long RGC axons is mainly dependent on the microtubule tracks, which are powered by the energy-dependent dynein and kinesin family of motor proteins. Conceptually, abnormal axonal transport could therefore be disrupted by two processes, either a mitochondrial biochemical defect or a primary problem with the assembly and maintenance of the microtubule network, as seen in the HSP group of disorders. The distinction is somewhat artificial in that one will precipitate the other, setting up a vicious circle leading to axonal swelling due to axoplasmic stasis, and ultimately axonal degeneration marked by the onset and progression of clinical symptoms.
Light and mitochondrial toxicity
RGCs are constantly being exposed to light in the 400–760 nm wavelength spectra, the cornea and the lens effectively blocking ultraviolet light below 400 nm. The blue component of visible light has the shortest wavelength and in studies of the retinal pigment epithelium and photoreceptors, it had the greatest potential for inducing cellular ROS production. Photoreceptors are partially shielded by macular lutein and zeoxanthin pigments, whereas mitochondria within the pre-laminar, unmyelinated RGC axons are directly exposed to light and do not have the benefit of these protective mechanisms. It is therefore possible that chronic light exposure could tip the balance in RGCs already compromised by a genetically-determined mitochondrial respiratory chain defect. This hypothesis has been partially tested in RGC-5 cell lines exposed to incremental light intensities, and cultured in both normal media and conditions of serum deprivation. Light exposure resulted in the generation of a significant amount of ROS in these RGC-5 cells, an effect which was intensity-dependent and exacerbated in conditions of nutritional deprivation. This light-induced elevation in ROS levels affected cellular viability by suppressing the expression of RGC-specific mRNAs and triggering the apoptotic cascade. Additional studies are therefore warranted to investigate the consequences of light exposure on RGC survival in mitochondrial optic nerve disorders, and for that purpose, both patient cell lines and existing animal models could be used.
Melanopsin retinal ganglion cells
Much attention has focused on the selective vulnerability of RGCs in mitochondrial optic neuropathies. However, immunohistochemical analysis of post-mortem eyes from patients with LHON and DOA has revealed relative sparing of a specific subpopulation of melanopsin-containing RGCs. These melanopsin RGCs constitute only about 1% of the total RGC population, but they subserve an important evolutionary role, regulating the body’s circadian rhythm as part of the retino-hypothalamic tract. The relative sparing of these melanopsin RGCs in LHON and DOA also explain to a certain extent the normally preserved pupillary light reflexes in these two disorders. The inherent resistance of melanopsin RGCs to mitochondrial dysfunction is tantalising and it could reveal key protective mechanisms that could be applied to neuroprotection of the larger pool of susceptible RGCs.
Drosophila dOpa1 mutants
have recently established a Drosophila model harbouring a specific dOpa1 mutation (CG8479), and the ocular phenotypes associated with heterozygous and homozygous carriers were determined. Homozygous mutant flies developed a rough and glossy eye phenotype due to the loss of hexagonal lattice cells, and decreased lens and pigment deposition. The dOpa1 mutation caused an increase in ROS levels and mitochondrial fragmentation, which damaged both cone and pigment cells. In a series of elegant experiments, these investigators were then able to demonstrate that the rough and glossy eye phenotype could be partially reversed by dietary supplementation with SOD-1 and vitamin E, and by genetic overexpression of human SOD1. Heterozygous adult flies did not exhibit any ocular abnormalities, but similar to the homozygous mutants, they also demonstrated elevated ROS levels and a greater susceptibility to oxidative stress. The heteterozygous drosophila carriers showed irregular and dysmorphic mitochondria in their muscle, and they had a significantly shortened lifespan, which was only partially restored by anti-oxidant treatment.
Zebrafish opa3 mutants
High levels of opa3 mRNA transcripts are present in the central nervous system of zebrafish (Danio rerio), including the optic nerve and retinal layers. Zebrafish embryos were therefore engineered harbouring a 5.2 kb retroviral DNA insertion immediately downstream of the mitochondrial leader signal, which disrupts the opa3 reading frame and results in a premature termination codon. Five-days-old homozygous mutant embryos (opa3ZM/ZM) had increased MGA levels, recapitulating the biochemical signature of patients with Costeff syndrome (Section 4.3). A significant loss of RGCs together with a reduction in optic nerve diameter was also observed on histological analysis of 1-year-old opa3ZM/ZM zebrafish mutants, confirming the role of opa3 in preserving RGC function. In addition, these homozygous mutants demonstrated extra-ocular deficits with dramatic alterations in swimming behaviour secondary to ataxia, loss of buoyancy control, and hypokinesia.
LHON
There is still no animal model where the primary LHON mutations have been successfully introduced into the mitochondrial genome. In spite of these technical difficulties, four experimental techniques have been developed which can replicate the optic nerve degeneration seen in LHON: (i) intravitreal injection of a respiratory chain poison such as rotenone, (ii) stereotactic injection of biodegradable rotenone-loaded microspheres into the optical layer of the superior colliculus, (iii) downregulation of nuclear-encoded complex I subunits (e.g. NFUFA1) with specific mRNA-degrading ribozymes, and (iv) allotopic expression of mutant subunits (e.g. MTND4) which are then imported into the mitochondria. These disease models will be indispensable when testing the feasibility of gene therapy in LHON, and different strategies are currently being pursued (Section 12.4).
Opa1
Two DOA mouse models have been developed which harbour pathogenic mutations in exon 8 (c.1051C>T) and intron 10 (c.1065+5g>a) of the Opa1 gene. These mutations are truncative in nature and they result in a 50% reduction in the overall expression of the Opa1 protein. In both models, homozygous mutant mice (Opa1−/−) died in utero during embryogenesis, highlighting the central role played by the Opa1 protein in early development. Heterozygous Opa1+/− mice faithfully replicated the human phenotype exhibiting a slowly progressive optic neuropathy and demonstrating objective reduction in visual function on psychophysical testing. Visual evoked potential (VEP) measurements showed significantly reduced amplitudes, but no change in latencies, supporting an ascending progress of degeneration from the soma towards the axon. Histological and retrograde labelling experiments confirmed a gradual loss of RGCs and an associated thinning of the retinal nerve fibre layer. The surviving optic nerve axons had an abnormal morphology with swelling, distorted shapes, irregular areas of demyelination and myelin aggregates. These features of optic nerve degeneration were seen as early as 9 months, but they were much more visible by 24 months. Loss of dendritic arborisation was observed for on- but not off-centre RGCs, these early features of neuronal dysfunction preceding the onset of axonal loss. An increased number of autophagosomes was also noted in the RGC layer of heterozygous Opa1+/− mice at later time points, probably due to the accumulation of dysfunctional mitochondria. Mitochondria within these axons showed disorganised cristae structures on transmission electron microscopy and cultured fibroblasts showed increased mitochondrial network fragmentation. To further investigate the extent of mitochondrial dysfunction, we carried out additional histochemical and molecular genetic studies on both these DOA mouse models. COX deficiency and multiple mtDNA deletions were not detected in sketelal muscle and the RGC layer of heterozygous mutant mice, implicating other Opa1-related disease mechanisms in precipitating optic nerve degeneration.
A third Opa1 mouse model is currently being characterised harbouring a heterozygous c.2708-2711delTTAG mutation in exon27. Heterozygous mutant mice are showing signs of optic nerve degeneration from the age of 6 months, with subnormal VEPs, and abnormal mitochondrial distribution, especially in the lamina cribosa region (Dr Guy Lenaers, Montpellier, personal communication). All these Opa1 mouse models only manifest pure optic atrophy, and it will be important to create a DOA+ mouse model in order to dissect the mechanisms which contribute to multi-system organ involvement.
The role played by Opa1 in early mouse embryogenesis was recently confirmed in the lilR3 mutant mouse where a proportion of embryos was noted to die at midgestation (E11.5) displaying growth retardation, exencephaly, and abnormal patterning along the anterior–posterior axis. Using a meiotic mapping strategy, uncovered the genetic basis for this embryonic lethality by identifying a homozygous splice site mutation within intron 19 of the Opa1 gene, which results in a 6-bp intronic sequence insertion. As a result of this Opa1 mutation, the mutant Opa1 protein is mislocalised, remaining within the cytosol instead of being properly translocated to the mitochondrial inner membrane.
Opa3
An Opa3 mouse model carrying a c.365T>C (p.L122P) mutation has been reported. Heterozygous Opa3+/− mice were not compromised, whereas homozygous Opa3−/− mice developed multi-system organ failure with cachexia, dilated cardiomyopathy, extrapyramidal features, and a reduced lifespan of less than 4 months. These mice had severely impaired visual function, and although all the retinal layers were affected, cell loss was much more prominent within the RGC layer, further reinforcing the selective vulnerability of this specific cell type.
Glaucoma
A series of papers has been reported by the Weinreb group exploring the relationships between Opa1 and IOP-mediated RGC loss in murine glaucoma models. Several interesting observations were made: (i) increased IOPs led to apoptotic RGC loss, which was preceded by mitochondrial network fragmentation, and marked Opa1 and cytochrome c release into the cytosol, (ii) the IOP effect was mediated by glutamate receptor activation, which could be blocked by the N-methyl-d-aspartate (NMDA) glutamate receptor antagonist memantine, and (iii) overexpression of Opa1 with an adenovirus-associated virus (AAV) vector was protective against RGC death in glaucomatous optic neuropathy.
In vivo imaging of retinal ganglion cells
Analysis of fixed retinal whole mount preparations is time-consuming and rather limited given that the animal in question needs to be sacrificed. A better understanding of the complex pathological mechanisms underlying inherited optic neuropathies will therefore require more sophisticated in vivo techniques for visualising RGCs, with the need for both time-lapse imaging and long-term serial monitoring of various cellular processes. Apoptosis has been implicated in all mitochondrial optic neuropathies and a non-invasive system which allows RGC death to be visualised directly would be a major research tool. Glaucoma investigators are working on different experimental paradigms, which include (i) retrograde labelling of RGCs, with tracer injection at the superior colliculus, (ii) transduction of RGCs following intravitreal injection of a recombinant AAV vector carrying GFP, and (iii) DARC (Detection of apoptosing retinal cells) with intravitreal injection of fluorescently-labelled annexin 5, which binds to exposed phosphatidylserine in early stages of apoptosis.
All these apoptosis-imaging techniques are invasive and they require repeated injections, with the possibility of introducing non-disease-related artefacts. To overcome these technical issues, various groups have generated transgenic mice which express a fluorescent marker such as GFP, under a specific RCG promoter such as Thy-1. Although further improvement in transfection efficiency is needed, RGCs which are successfully transfected can be visualised using a standard confocal scanning laser ophthalmoscope (CSLO), with the mouse anaesthesised on an appropriately adapted microscope stage. Using this protocol, extremely detailed RGC images can be obtained including cell body morphology and the extent of dendritic arborisation. Furthermore, coupled with the intravitreal injections of tracer agents, time-lapsed sequences would allow axonal transport to be quantified both in the pre- and post-laminar segments of the optic nerve. The application of these new imaging modalities to both existing and future animal models of mitochondrial optic neuropathies will be a major breakthrough, allowing for example early subclinical RGC abnormalities to be detected, and the therapeutic potential of putative neuroprotective agents to be assessed at a more functional level.
Genetic counselling
Male carriers harbouring mtDNA point mutations can be reassured that their children are not at risk of inheriting their genetic defect. Female carriers will transmit the mutation to all their offspring, but if the mutation is heteroplasmic, it is not possible to reliably predict the mutational level that will be transmitted. As discussed earlier, significant variations in heteroplasmy levels can occur as a result of the mitochondrial genetic bottleneck. The use of amniocentesis or chorionic villus sampling for prenatal testing is therefore limited in this situation, as the mutational load detected in amniocytes and chorionic villus cells could differ from other foetal tissues, especially those at greater risk from a specific mtDNA point mutation such as RGCs. Although it is not possible to accurately predict whether a LHON carrier will eventually lose vision, individuals can be counselled based on the two major identifiable risk factors in this disorder, age and gender. Male carriers have about a 50% lifetime risk of visual failure compared with only 10% for female carriers, and most patients will experience visual loss in their late teens and twenties. The probability of becoming affected decreases subsequently and disease conversion in LHON is rare over the age of 50 years. The risks of transmission for nuclear mitochondrial disorders follow the laws of Mendelian inheritance, but if a specific nuclear defect has not been identified, only an approximate risk can be provided based on the family history. As a result of the marked inter- and intra-familial phenotypic variability seen with both pure and syndromal inherited optic neuropathies, genetic counselling for patients and their families remains a challenging area of practice.
Supportive measures
The treatment options for patients with inherited optic neuropathies are currently limited. However, there are several practical steps that can be taken as part of a multi-disciplinary team to improve a patient’s quality of life, and minimise long-term morbidity. Depending on their needs, these patients should be provided with access to facilities such as low visual aids and occupational therapy, and clinicians can help with financial assistance through their local social services. It is also important to aggressively manage related medical problems such as diabetes and epilepsy, and clinicians need to be vigilant to the development of new complications such as cardiomyopathy and sensorineural deafness, which could be amenable to therapeutic intervention. Patients with mitochondrial disorders should be strongly advised not to smoke and to minimise their alcohol intake, not only as a general health measure, but smoking, and to a lesser extent excessive alcohol intake, have been linked with an increased risk of visual loss among LHON carriers. Patients with CPEO often get significant benefit from simple conservative measures such as ptosis props or Fresnel prisms for symptomatic ocular misalignment. In some cases, strabismus and ptosis surgery are indicated but these should be performed by experienced surgeons, because of the increased risk of complications such as corneal exposure secondary to poor orbicularis oculi function and impaired Bell’s phenomenon.
Neuroprotection
On the basis of limited, mostly anecdotal evidence, various combinations of pharmacological agents have been used to treat patients with mitochondrial disorders including multi-vitamin supplements, co-enzyme Q10 (CoQ10) and its derivatives, and putative free radical scavengers. There is relatively more clinical data on the use of CoQ10, which has shown a clear benefit for patients with primary CoQ10 deficiency. In one randomised, double-blind, crossover study of 16 patients with mitochondrial cytopathies, a combination of CoQ10 with creatine monohydrate and α-lipoic acid, reduced the levels of resting plasma lactate and oxidative stress markers.
Idebenone is a synthetic analogue of CoQ10, and it is currently being investigated as a treatment option for LHON and other neurodegenerative disorders such as FRDA. Idebenone is able to operate under low oxygen tension and it is thought to have anti-oxidant properties, in addition to optimising ATP production by the respiratory chain complexes. Initial reports of idebenone therapy in FRDA showed promising results with an improvement in both cardiac and neurological status. However, preliminary data released from two recently completed randomised controlled trials have proven rather disappointing, with no significant difference in primary cardiac and neurological endpoints between the idebenone-treated and placebo groups. (http://www.santhera.com/index.php?docid=212&vid=&lang=en&newsdate=201005&newsid=1417424&newslang=en, Accessed 1st of December 2010). In a retrospective study of 28 affected LHON patients, half of whom received a combination of idebenone, vitamin B2, and vitamin C for at least 1 year, the treated group showed a faster rate of visual recovery. However, in a more recent report, megadoses of idebenone, vitamin C, and riboflavin did not prevent second-eye involvement in two m.11778G>A LHON carriers treated after the onset of unilateral visual loss, and both patients showed no improvement in visual function. The benefits of idebenone in LHON therefore remain unclear, but being a safe drug, it is often recommended by clinicians or self-prescribed by patients. In collaboration with clinical partners in Germany and Canada, we recently completed a double-blind randomised placebo-controlled trial of idebenone in LHON. The preliminary results suggest a beneficial effect, especially among patients with recent disease onset, and relatively good remaining visual acuity in the second eye to become affected. (http://www.santhera.com/index.php?docid=212&vid=&lang=enamp;newsdate=201006&newsid=1424223&newslang=en, Accessed 1st of December 2010). If these initial findings are confirmed, idebenone could also be considered as a neuroprotective agent for other mitochondrial optic neuropathies such as DOA, although long-term follow-up will be required given the relatively slower rate of RGC loss in this disorder.
Brimonidine is a topical α-2 agonist used in the management of glaucoma for its IOP-lowering properties. In addition, studies based on models of optic nerve ischaemia have suggested that brimonidine could have an anti-apoptotic effect, mitigating RGC loss. On this basis, topical brimonidine has been tested as prophylactic agent for second-eye involvement in an open-labelled study of 9 patients with unilateral acute vision from LHON. Brimonidine failed to prevent fellow eye involvement and there was no evidence of improved visual benefit following the onset of visual loss. In the long term, neuroprotection for LHON remains an attractive treatment option given the easy accessibility of ocular tissues to various forms of manipulation. Various agents have been shown to have RGC-neuroprotective properties in other models of optic nerve degeneration such as memantine, valproic acid, prostaglandin J2 (PGJ2), and SIRT-1 activators. It is not inconceivable an effective agent will soon become available in clinical practice, which could be safely delivered intravitreally, providing an instantaneous local protective effect.
Gene therapy
Classical gene therapy is proving very challenging for primary mitochondrial disorders, because the tools required for successful gene transfer into the mitochondrial genome is still not available. The mitochondrial inner membrane represents a significant physical barrier that needs to be overcomed, and given the large number of mitochondria per cell, a highly efficient vector will be required to achieve an adequate level of transfection. To bypass these difficulties, allotopic strategies have been developed where the gene of interest is transfected into the nuclear genome, usually with an AAV vector. The protein product has a mitochondrial targeting sequence and as a result, it gets imported into the mitochondrial compartment, either replacing the missing protein or complementing the dysfunctional mutant protein.
In both in vitro and in vivo experimental LHON models, RGC loss was dramatically reduced by transfecting them with an AAV vector containing the human SOD2 gene. The increased expression of the superoxide dismutase enzyme is thought to improve RGC survival by minimising free radical damage, and decreasing the cell’s susceptiblity to apoptosis. Allotopic rescue, with the replacement of the defective mitochondrial complex subunit, is another attractive option for gene therapy in LHON. Proof-of-principle for this approach has been demonstrated in a rat model expressing a defective ND4 gene containing the m.11778A>G primary LHON mutation. The loss of visual function in these rats was reversed by transfecting RGCs with the wild-type ND4 gene, using an in vivo electroporation technique instead of an AAV vector. The transgene became stably integrated within the nuclear genome and the level of expression achieved was sufficient for successful RGC neuroprotection. These early studies of allotopic rescue in LHON are promising, but the results need to be replicated in larger animals, and long-term safety data is essential before human clinical trials can be contemplated. The same caveats will apply to proposed gene therapy strategies for nuclear mitochondrial disorders. Issues of safety and efficacy are especially relevant given the significant concerns that have been raised recently about the limitations of allotopic rescue in the treatment of primary mtDNA disorders. As yet, there is no conclusive experimental evidence that allotopically-expressed mitochondrial subunits can be properly integrated into fully-assembled OXPHOS complexes. Mitochondrially-encoded complex I subunits are also highly hydrophobic and if a proportion of these polypeptides is not imported, they could have deleterious consequences by physically aggregating into the cytosol or triggering an inappropriate immune response.
Preventing transmission of pathogenic mutations
Following successful fertilisation, two distinct structures can be observed within the fertilised oocyte, known as the male and female pronuclei. Given the lack of treatment for mitochondrial disorders, pronuclear transfer is currently being investigated as a method to prevent the transmission of mtDNA mutations in human embryos. This strategy involves the replacement of the entire mitochondrial population, by transferring karyoplast containing the pronuclei from a donor zygote to an enucleated recipient zygote. In a landmark study using abnormally-fertilised human zygotes, have shown that pronuclear transfer resulted in minimal carry-over of donor zygote mtDNA, and it was compatible with successful progression to the blastocyst stage. Preimplantation genetic diagnosis for nuclear defects is well established and it does not involve the same degree of complexity faced with mtDNA mutations.
Future directions
The past decade has seen significant advances in the way both clinicians and basic scientists approach inherited optic neuropathies. With greater access to molecular genetic testing, the phenotypic spectrum of classical optic nerve disorders such as LHON and DOA has expanded to encompass a much wider range of clinical features, some overlapping with other neurodegenerative disorders such as CMT, HSP, and even MS. Mitochondrial optic neuropathies could therefore be used as models to understand key pathophysiological aspects of these complex disorders, the optic nerve being relatively accessible and amenable to direct measurements with non-invasive imaging modalities. A fundamental change in our understanding is the realisation that in a significant number of genetically-determined optic nerve disorders, including glaucoma, mitochondrial dysfunction is likely to be part of a final common pathway mediating RGC loss. However, there are still more questions than answers – What combination of secondary factors will eventually prove to be the basis for the marked incomplete penetrance and male bias in LHON? What disease mechanisms explain multi-system tissue involvement in OPA1 carriers with DOA+ phenotypes, when other family members harbouring the same mutation only have pure optic nerve involvement? Conversely, why is it that only some patients with MFN2 and SPG7 mutations develop visual failure and optic atrophy? With rapid advances in genetic technology, bioinformatics, and molecular biology, the next decade will be an exciting time in this area of research. We are hopeful that these future breakthroughs will translate into much-needed, effective treatment strategies for LHON, DOA, and other mitochondrial optic neuropathies.
Conflicts of interest
None of the authors have any financial interests to disclose.
References 508
- 1.
Ahn J., Kim N.J., Choung H.K., Hwang S.W., Sung M., Lee M.J. Frontalis sling operation using silicone rod for the correction of ptosis in chronic progressive external ophthalmoplegia. British Journal of Ophthalmology. 2008;92:1685-1688.
Not in our library10.1136/bjo.2008.144816
- 2.
Ahrlich K.G., De Moraes C.G.V., Teng C.C., Prata T.S., Tello C., Ritch R. Visual field progression differences between normal-tension and exfoliative high-tension glaucoma. Investigative Ophthalmology & Visual Science. 2010;51:1458-1463.
Not in our library10.1167/iovs.09-3806
- 3.
Aijaz S., Erskine L., Jeffery G., Bhattacharya S.S., Votruba M. Developmental expression profile of the optic atrophy gene product: OPA1 is not localized exclusively in the mammalian retinal ganglion cell layer. Investigative Ophthalmology & Visual Science. 2004;45:1667-1673.
Not in our library10.1167/iovs.03-1093
- 4.
Akepati V.R. Characterization of OPA1 isoforms isolated from mouse tissues. Journal of Neurochemistry. 2008;106:372-383.
Not in our library10.1111/j.1471-4159.2008.05401.x
- 5.
Alavi M.V., Bette S., Schimpf S., Schuettauf F., Schraermeyer U., Wehrl H.F. A splice site mutation in the murine Opa1 gene features pathology of autosomal dominant optic atrophy. Brain. 2007;130:1029-1042.
Not in our library10.1093/brain/awm005
- 6.
Alberio S., Mineri R., Tiranti V., Zeviani M. Depletion of mtDNA: syndromes and genes. Mitochondrion. 2007;7:6-12.
Not in our library10.1016/j.mito.2006.11.010
- 7.
Alexander C., Votruba M., Pesch U.E.A., Thiselton D.L., Mayer S., Moore A. OPA1, encoding a dynamin-related GTPase, is mutated in autosomal dominant optic atrophy linked to chromosome 3q28. Nature Genetics. 2000;26:211-215.
Not in our library10.1038/79944
- 8.
Alward W.L., Feldman F., Cashwell L.F., Wilensky J., Geijssen H.C., Greeve E. The effectiveness of intraocular pressure reduction in the treatment of normal-tension glaucoma. American Journal of Ophthalmology. 1998;126:498-505.
Not in our library10.1016/s0002-9394(98)00272-4
- 9.
Amati-Bonneau P., Guichet A., Olichon A., Chevrollier A., Viala F., Miot S. OPA1 R445H mutation in optic atrophy associated with sensorineural deafness. Annals of Neurology. 2005;58:958-963.
Not in our library10.1002/ana.20681
- 10.
Amati-Bonneau P., Odent S., Derrien C., Pasquier L., Malthiery Y., Reynier P. The association of autosomal dominant optic atrophy and moderate deafness may be due to the R445H mutation in the OPA1 gene. American Journal of Ophthalmology. 2003;136:1170-1171.
Not in our library10.1016/s0002-9394(03)00665-2
- 11.
Amati-Bonneau P., Valentino M.L., Reynier P., Gallardo M.E., Bornstein B., Boissiere A. OPA1 mutations induce mitochondrial DNA instability and optic atrophy plus phenotypes. Brain. 2008;131:338-351.
Not in our library10.1093/brain/awm298
- 12.
Anderson S., Bankier A.T., Barrell B.G., de Bruijn M.H., Coulson A.R., Drouin J. Sequence and organization of the human mitochondrial genome. Nature. 1981;290:457-465.
Not in our library10.1038/290457a0
- 13.
Andrews R.M., Griffiths P.G., Johnson M.A., Turnbull D.M. Histochemical localisation of mitochondrial enzyme activity in human optic nerve and retina. British Journal of Ophthalmology. 1999;83:231-235.
Not in our library10.1136/bjo.83.2.231
- 14.
Andrews R.M., Kubacka I., Chinnery P.F., Lightowlers R.N., Turnbull D.M., Howell N. Reanalysis and revision of the Cambridge reference sequence for human mitochondrial DNA. Nature Genetics. 1999;23:147.
Not in our library10.1038/13779
- 15.
Anikster Y., Kleta R., Shaag A., Gahl W.A., Elpeleg O. Type III 3-methylglutaconic aciduria (optic atrophy plus syndrome, or Costeff optic atrophy syndrome): identification of the OPA3 gene and its founder mutation in Iraqi Jews. American Journal of Human Genetics. 2001;69:1218-1224.
Not in our library10.1086/324651
- 16.
Araie M., Kitazawa M., Koseki N. Intraocular pressure and central visual field of normal tension glaucoma. British Journal of Ophthalmology. 1997;81:852-856.
Not in our library10.1136/bjo.81.10.852
- 17.
Arnold A.C. Evolving management of optic neuritis and multiple sclerosis. American Journal of Ophthalmology. 2005;139:1101-1108.
Not in our library10.1016/j.ajo.2005.01.031
- 18.
Artal-Sanz M., Tavernarakis N. Prohibitin and mitochondrial biology. Trends in Endocrinology and Metabolism. 2009;20:394-401.
Not in our library10.1016/j.tem.2009.04.004
- 19.
Assink J.J.M., Tijmes N.T., ten Brink J.B., Oostra R.J., Riemslag F.C., deJong P. A gene for X-linked optic atrophy is closely linked to the Xp11.4-Xp11.2 region of the X chromosome. American Journal of Human Genetics. 1997;61:934-939.
Not in our library10.1086/514884
- 20.
Aung T., Ocaka L., Ebenezer N.D., Morris A.G., Krawczak M., Thiselton D.L. A major marker for normal tension glaucoma: association with polymorphisms in the OPA1 gene. Human Genetics. 2002;110:52-56.
Not in our library10.1007/s00439-001-0645-7
- 21.
Balcer L.J. Optic neuritis. New England Journal of Medicine. 2006;354:1273-1280.
Not in our library10.1056/nejmcp053247
- 22.
Ban T., Heymann J.A.W., Song Z.Y., Hinshaw J.E., Chan D.C. OPA1 disease alleles causing dominant optic atrophy have defects in cardiolipin-stimulated GTP hydrolysis and membrane tubulation. Human Molecular Genetics. 2010;19:2113-2122.
Not in our library10.1093/hmg/ddq088
- 23.
Baracca A., Solaini G., Sgarbi G., Lenaz G., Baruzzi A., Schapira A.H. Severe impairment of complex I-driven adenosine triphosphate synthesis in leber hereditary optic neuropathy cybrids. Archives of Neurology. 2005;62:730-736.
Not in our library10.1001/archneur.62.5.730
- 24.
Barbet F., Gerber S., Hakiki S., Perrault I., Hanein S., Ducroq D. A first locus for isolated autosomal recessive optic atrophy (ROA1) maps to chromosome 8q. European Journal of Human Genetics. 2003;11:966-971.
Not in our library10.1038/sj.ejhg.5201070
- 25.
Barbet F., Hakiki S., Orssaud C., Gerber S., Perrault I., Hanein S. A third locus for dominant optic atrophy on chromosome 22q. Journal of Medical Genetics. 2005;42.
Not in our library10.1136/jmg.2004.025502
- 26.
Barbiroli B., Montagna P., Cortelli P., Iotti S., Lodi R., Barboni P. Defective brain and muscle energy-metabolism shown by in-vivo P-31 magnetic-resonance spectroscopy in nonaffected carriers of 11778-mtDNA mutation. Neurology. 1995;45:1364-1369.
Not in our library10.1212/wnl.45.7.1364
- 27.
Barboni P., Carbonelli M., Savini G., Foscarini B., Parisi V., Valentino M.L. OPA1 mutations associated with dominant optic atrophy influence optic nerve head size. Ophthalmology. 2010;117:1547-1553.
Not in our library10.1016/j.ophtha.2009.12.042
- 28.
Barboni P., Carbonelli M., Savini G., Ramos C.D., Carta A., Berezovsky A. Natural history of Leber’s hereditary optic neuropathy: longitudinal analysis of the retinal nerve fiber layer by optical coherence tomography. Ophthalmology. 2009.
Not in our library10.1016/j.ophtha.2009.07.026
- 29.
Barboni P., Savini G., Valentino M.L., La Morgia C., Bellusci C., De Negri A.M. Leber’s hereditary optic neuropathy with childhood onset. Investigative Ophthalmology & Visual Science. 2006;47:5303-5309.
Not in our library10.1167/iovs.06-0520
- 30.
Barboni P., Savini G., Valentino M.L., Montagna P., Cortelli P., De Negri A.M. Retinal nerve fiber layer evaluation by optical coherence tomography in Leber’s hereditary optic neuropathy. Ophthalmology. 2005;112:120-126.
Not in our library10.1016/j.ophtha.2004.06.034
- 31.
Barnils N., Mesa E., Munoz S., Ferrer-Artola A., Arruga J. Response to idebenone and multivitamin therapy in Leber’s hereditary optic neuropathy. Archivos de la Sociedad Española de Oftalmología. 2007;82:377-380.
Not in our library10.4321/s0365-66912007000600012
- 32.
Barron M.J., Griffiths P., Turnbull D.M., Bates D., Nichols P. The distributions of mitochondria and sodium channels reflect the specific energy requirements and conduction properties of the human optic nerve head. British Journal of Ophthalmology. 2004;88:286-290.
Not in our library10.1136/bjo.2003.027664
- 33.
Beck R.W., Cleary P.A., Trobe J.D., Kaufman D.I., Kupersmith M.J., Paty D.W. The effect of corticosteroids for acute optic neuritis on the subsequent development of multiple-sclerosis. New England Journal of Medicine. 1993;329:1764-1769.
Not in our library10.1056/nejm199312093292403
- 34.
Beretta S., Mattavelli L., Sala G., Tremolizzo L., Schapira A.H.V., Martinuzzi A. Leber hereditary optic neuropathy mtDNA mutations disrupt glutamate transport in cybrid cell lines. Brain. 2004;127:2183-2192.
Not in our library10.1093/brain/awh258
- 35.
Bern C., Philen R.M., Fredman D., Bowman B.A., Newman N.J., Gerr F. Epidemic optic neuropathy in cuba – clinical characterization and risk-factors. New England Journal of Medicine. 1995;333:1176-1182.
Not in our library10.1056/nejm199511023331803
- 36.
Berninger T.A., Jaeger W., Krastel H. Electrophysiology and color perimetry in dominant infantile optic atrophy. British Journal of Ophthalmology. 1991;75:49-52.
Not in our library10.1136/bjo.75.1.49
- 37.
Besch D., Leo-Kottler B., Zrenner E., Wissinger B. Leber’s hereditary optic neuropathy: clinical and molecular genetic findings in a patient with a new mutation in the ND6 gene. Graefes Archive for Clinical & Experimental Ophthalmology. 1999;237:745-752.
Not in our library10.1007/s004170050307
- 38.
Bette S., Schlaszus H., Wissinger B., Meyermann R., Mittelbronn M. OPA1, associated with autosomal dominant optic atrophy, is widely expressed in the human brain. Acta Neuropathologica. 2005;109:393-399.
Not in our library10.1007/s00401-004-0970-8
- 39.
Bhat N.K., Niranjan B.G., Avadhani N.G. The complexity of mitochondrial translation products in mammalian-cells. Biochemical and Biophysical Research Communications. 1981;103:621-628.
Not in our library10.1016/0006-291x(81)90496-4
- 40.
Biermann J., Grieshaber P., Goebel U., Martin G., Thanos S., Di Giovanni S. Valproic acid-mediated neuroprotection and regeneration in injured retinal ganglion cells. Investigative Ophthalmology & Visual Science. 2010;51:526-534.
Not in our library10.1167/iovs.09-3903
Cited by papers we hold 7
- Oocytes maintain ROS-free mitochondrial metabolism by suppressing complex I · 2022No notice
- Impaired complex I repair causes recessive Leber’s hereditary optic neuropathy · 2021No notice
- Mitochondrial Dysfunction and Multiple Sclerosis · 2019No notice
- Efficacy and Safety of rAAV2-ND4 Treatment for Leber’s Hereditary Optic Neuropathy · 2016No notice
- A neurodegenerative perspective on mitochondrial optic neuropathies · 2016No notice
- The primary vascular dysregulation syndrome: implications for eye diseases · 2013No notice
- Dominant optic atrophy · 2012No notice
Reads
Read by people on this page
None yet.
Checked or read by AI through naulon
None yet.
A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.
Are you an author?
Is this your paper?
Claim it to add your note and be paid when agents cite it. Free.