Abstract
Introduction:
Mutations in the gene encoding DNA polymerase gamma (POLG) impair its ability to proofread mitochondrial DNA (mtDNA) during replication [1]. This results in a high frequency of randomly distributed mtDNA mutations and thus a wide range of phenotypes, including seizures, neuropathy, and cerebellar ataxia [1, 2]. We document a phenotype associated with the rare POLG variant c.1370G>A (p.R457Q).
Methods:
Over 10 years, we performed electrodiagnostic and neuropsychologic on a patient who presented with a variety of neurologic symptoms.
Results:
Testing revealed an axonal sensorimotor polyneuropathy, depression and executive function difficulties, and asymmetric ataxia. Genetic testing revealed a POLG variant of uncertain significance (c.1370G>A, p.R457Q) in a heterozygous state.
Conclusions:
We have identified a mutation in POLG that could result in a diverse array of symptoms and signs of our patient. However, interpreting pathogenicity of rare variants such as R457Q is challenging and will likely require identification of patients with similar phenotypes caused by the variant of uncertain significance.
INTRODUCTION
DNA polymerase gamma (POLG) plays a critical role in DNA repair during the replication of mitochondrial DNA (mtDNA). DNA repair is important for maintaining DNA integrity during replication as the repair process protects against cancer and age-related diseases. Mutations in the POLG gene that disrupt the proofreading activity of the POLG enzyme cause a high frequency of randomly distributed mtDNA mutations. Thus, such mutations can affect a number of genes which, in addition to mitochondrial heteroplasmy, are likely responsible for the wide-range of phenotypes observed. These can include seizures, neuropathy, cerebellar ataxia, myopathy, and progressive external ophthalmoplegia, among others [3]. Currently, almost 300 POLG variants are listed in ClinVar, an NIH-supported online variant database. Among these, 65 are classified as pathogenic, and 25 as likely to be pathogenic. Additionally, 128 benign or likely benign variants of POLG are known to exist [4]. Combined with the large number of pathogenic variants and the potential for many genes to be affected by one variant, the effects of POLG mutations can be highly varied, diverse and difficult to determine. Additionally, POLG-related diseases change over time and have overlapping symptoms with varying timings of onset and severity of disease [5].
Symptoms caused by mutations in POLG may become evident during infancy and continue throughout late adulthood. These are often widespread and can include: a severe childhood-onset encephalopathy called Alpers-Huttenlocher syndrome, childhood myocerebrohepatopathy spectrum, myoclonic epilepsy myopathy sensory ataxia (MEMSA), autosomal dominant and recessive forms of progressive external ophthalmoplegia (PEO), the ataxia neuropathy spectrum, which includes the phenotypes previously referred to as mitochondrial recessive ataxia syndrome (MIRAS) and sensory ataxia neuropathy dysarthria and ophthalmoplegia (SANDO). The wide-range of symptoms associated with mutations in POLG make it challenging to discern the cause of the phenotype and highlight the importance of describing rare mutations.
Important clues to determining whether a condition’s pathology is mitochondrial dysfunction are involvement of multiple organ systems requiring a lot of ATP to function and a progressive phenotype. Once a mitochondrial dysfunction is suspected the search for a molecular cause should take into account family history, age of onset, and the specific clinical findings. Potential molecular causes of mitochondrial dysfunction include mutations in the mitochondrial genome but also nuclear genes. We report on a patient whose diagnosis was uncertain for many years prior to the identification of a POLG variant of uncertain significance, c.1370G>A, p.R457Q.
CASE REPORT
Informed consent was obtained from the patient. The patient is a 68 year-old right-handed man who was the product of a normal pregnancy and delivery. He reportedly reached developmental milestones on time. As a child he was an average runner and had no known difficulties with balance, falling, or weakness. He had a mild tremor as a child, but this was a minor issue. He played intramural football and wrestled, but never excelled in sports. He biked but never skated; he did not have trouble finding shoes that fit. His neurological symptoms and findings are outlined as follows:
At this time the patient presented to our neurogenetics clinic for the above-described array of symptoms. The patient described no bowel or bladder complaints and no difficulty swallowing; however, he noted some mild fasciculations in the quadriceps and sometimes in the hand muscles. Of note, the abnormal speech and asymmetric cerebellar ataxia progressed well after phenytoin was discontinued. He had a history of social cigarette smoking (1/4th pack per day for 40 years) and occasional alcohol use.
The patient’s examination revealed significant problems with speech that was characterized as variable and ataxic dysarthria. Ocular movements were full and conjugate; saccades were slightly slowed in the horizontal plane but nystagmus was not detected. Atrophy of the first dorsal interosseous and anterior tibialis muscles was observed. Motor examination demonstrated weakness in the first dorsal interosseous, abductor pollicis brevis, abductor digiti minimi, anterior tibialis, foot eversion, and great toe dorsi and plantar flexion. The patient’s response to pinprick of the toes and above the ankles was decreased bilaterally. Also, vibration at toes and ankles, as well as sensing of joint position at the toes, were decreased.
The patient showed left worse than right finger-to-nose as well as difficulty with finger tapping (Video 1) and heel-to-shin dyssynergy. He had an ataxic tremor that was markedly more pronounced on the left side. Cerebellar testing was abnormal, as elicited by ataxic finger-to-nose, alternating hand movements, and slowed, arrhythmic finger-to-thumb tapping (Video 1). He was diffusely areflexic. His gait was characterized as a widened-base, steppage gait with mild to moderate unsteadiness; the Romberg sign was positive. He had no rigidity or abnormal facial expression.
Laboratory testing
A laboratory evaluation with serum immunoelectrophoresis in 2005 was unremarkable. Tests for vitamin E, vitamin B12, lactic acid, methylmalonic acid, copper, antigliadin level, anti-glutamic acid decarboxylase (GAD) antibody, tissue transglutaminase, fragile×syndrome, thyroid stimulating hormone, fasting glucose level, oral glucose tolerance test, peripheral smear for acanthocytes, serum paraneoplastic panel, and hemoglobin A1C were all normal. The level of frataxin was normal. A muscle and nerve biopsy were offered, but the patient declined having this procedure.
Imaging
Magnetic resonance imaging (MRI) and magnetic resonance angiogram (MRA) of the head and neck were unrevealing. A CT scan of the chest, abdomen, and pelvis, as well as a mammogram, were all negative for malignancy.
Electrodiagnostic testing
Electromyography and nerve conduction studies revealed an axonal sensorimotor polyneuropathy (Table 1).
Electrophysiological Data (normal values in parenthesis)
EMG of the first dorsal interosseous and tibialis anterior showed 2 + fibs, positive sharp waves, and reduced recruitment with high amplitude, long duration motor unit action potentials. The left vastus lateralis, left thoracic paraspinal, and left trapezius musculature was normal. Repetitive stimulation of the left facial nerve was normal.
Pedigree
A pedigree was unrevealing except that the patient’s father was reported to possibly have had ataxia, tremor, and seizures. Also, the father was confined to a wheelchair during the last 4 years of his life, and had a shunt for obstructive hydrocephalus. He could not be genetically tested because he was deceased.
Genetic testing
Genetic testing for Charcot Marie Tooth disease type 1A was performed prior to our evaluation and was negative. In the neurogenetics clinic, we investigated a panel of genes including POLG, SLC25A4, and C10ORF2. Results revealed a POLG missense variant of uncertain significance, c.1370G>A, p.R457Q, NM_002693.2 in a heterozygous state. Using the ACMG standards and guidelines to report the pathogenicity of variants, our mutation is classified a variant of uncertain significance [6]. We hypothesize the variant is causing the abnormalities present in our patient. We have not been able to identify other causes and his exam was consistent with that reported with other POLG mutations. Accordingly, we then attempted to follow the standards and guidelines for interpreting sequence variants recently proposed by the American College of Medical Genetics and Genomics and the Association for Molecular Pathology [6]. We were unable to do segregation studies, as both parents of the proband are deceased. There is limited data regarding the proband’s family and the disorder came later on in life. Arginine at amino acid position 457 of the POLG protein is weakly conserved during evolution. In addition, the mutation has been reported at a low allele frequency in dbSNP [7] and ExAc [8] (rs372911506). The mutation was not identified in the “POLG Pathogenicity Prediction Server” [1]. SIFT and PolyPhen-2 yield discordant predictions as to the pathogenicity of this variant. In gnomAD [9], the overall frequency is 1/12,998; in European ancestry it is 0/8,598 and in African ancestry, it is 1/4,400. Therefore, although the variant fits clinically we can really say no more than it is a true variant of uncertain significance (VUS). Given our limited data, it is possible that this mutation could be in a trans configuration to a non-coding POLG variant, thus resulting in a recessive POLG syndrome. Additionally whole genome sequencing and multiplex ligation-dependent probe amplification could be used to look for intronic variants and large deletions. Although autosomal recessive inheritance is a possibility, most dominant POLG mutations exhibit late onset disease manifestations [10].
DISCUSSION
POLG and mitochondrial related mutations are extremely variable and result in an extensive array of symptoms of the central and peripheral nervous systems. Patients present with disease of irregular severity, onset, and progression, making early diagnosis difficult; additionally, the POLG clinical manifestations vary so widely even among patients with the same POLG mutations [1, 11]. As was evident in our patient, neurological signs and symptoms evolve over time. POLG mutations are relatively rare with a carrier frequency of 1/100 in the Western hemisphere, thus making the diagnosis even more challenging [1, 12].
Analysis of our patient suggests that in cases of asymmetric ataxia with axonal sensorimotor polyneuropathy, depression, cognitive deficits, and no significant family history it is important to consider mitochondrial mutations as a differential. Given that a clinical diagnosis of POLG-related disorders can be challenging, a genetic test for the POLG gene can potentially result in a reliable diagnosis [13]. Examples of limitations of other forms of analysis include the following: MRI of the brain may not or may not show cortical, subcortical, or cerebellar atrophy; cerebrospinal fluid (CSF) may or may not show increased serum CSF lactate and pyruvate as well as positive oligoclonal bands with elevated CSF protein and pleocytosis. Analysis of muscle pathology may reveal phenotypes ranging from no pathology (normal appearance) to severe abnormalities indicating a mitochondrial myopathy (e.g., the presence of cytochrome oxidase (COX)-deficient, ragged-blue, and ragged-red fibers), and electrodiagnostic testing may be normal or could show reduced sensory or motor nerve action potentials with preserved conduction velocity.
Because no blood relatives of the patient are living, because the variant of uncertain significance has been observed in presumably asymptomatic individuals and because of the in silico predictive studies it is difficult to confirm that the identified variant of uncertain significance is responsible for the pathogenicity in this patient; however, clinical aspects of the case support that the POLG mutations could be disease causing (Table 2). Other patients with POLG mutations that present in the ataxia neuropathy spectrum have had symptoms similar to those observed in our patient, including peripheral neuropathy, mild cognitive impairment, horizontal slowing of saccades, epilepsy and hypoacusis (Table 2). As is the case with other POLG-related disorders, patients on the ataxia neuropathy spectrum typically exhibit depletion of the mitochondrial DNA (mtDNA) in the tissues affected by the condition, such as the brain. The resulting reduction in oxidative phosphorylation results in a decrease in the amount of energy available to the cell, which may account for the wide variety of signs and symptoms in the patients.
Diagnostic clues to distinguishing between Freidreich’s Ataxia and the POLG mutation resulting in an Ataxia Neuropathy Spectrum
In murine models, mitochondria can be relatively tolerant of point mutations and a 500-fold increase in these mutations did not affect the lifespan of the animals [1, 14]. These mutations, however, can be highly pathogenic when in a compound heterozygous form [1]. The location of the mutation is also very important— those mutations located near a polymerase active site can result in symptoms like PEO, muscle weakness, and other various neurological symptoms [1]. At this point, we have limited data regarding the mutation we describe in our subject; the low number of reported patient cases for the dominant mutations and lack of consistent genetic testing make the dominant mutations more difficult to evaluate [1].
Effective treatment of mitochondrial and POLG-related disorders requires a large and coordinated medical team that includes a neurologist, physical therapist, occupational therapist, nutritionist, genetic counselor, gastroenterologist, and psychiatrist, and possibly also an ophthalmologist. The prevention of seizures is a priority, and requires careful selection of antiepileptic medications as some can cause serious adverse side effects. For example, valproate is associated with an increase in the risk of liver failure [15]. Speech therapy and a percutaneous endoscopic gastrostomy (PEG) tube may be necessary, and possibly surgery to correct ptosis. Recent reports suggest that deficiency of folate in the CSF may be common in POLG disorders [15]. Although the data in support of such a deficiency are limited, a previous study recommends that folate be supplemented if the level is below the normal range as well as treatment with levocarnitine, coenzyme q10, and a B-complex vitamin in spite of the limited evidence regarding the efficacy of this treatment [15]. Occupational therapy, physical therapy, bilateral ankle/foot orthotics and a balance vest (thought to help with balance, stability, and postural control) were used on our patient. We note that the balance vest resulted in only limited improvement for our patient, but our patient has done well with the use of ankle foot orthotics. The folic acid has not seemed to stop the progression of the disease.
In summary, our study reports a POLG variant of uncertain significance, c.1370G>A, p.R457Q, in a patient exhibiting a variety of symptoms along the ataxia neuropathy spectrum, and shows that accurate diagnosis requires the use of genetic testing. Although there is a difficulty in classifying this variant of uncertain significance as benign or pathogenic, we believe it is important to report such variants as it is only by reporting such cases in the literature that other individuals may be identified with similar phenotypes associated with the same variant.
CONFLICTS OF INTEREST
Dr. Jerath and Dr. Shy have no conflicts of interest to disclose.
Footnotes
ACKNOWLEDGMENTS
The work was supported by a grant from the National Institute of Neurological Disorders and Stroke (MES) and Office of Rare Diseases (MES, U54NS065712), Muscular Dystrophy Association (MES), Charcot-Marie-Tooth Association (MES), and MDA Clinical Research Training grant (NUJ). We would like to acknowledge the hard work and effort of genetic counselors, Tiffany Grider and Sandra Peacock.
