Cerebellar ataxia and coenzyme Q10 deficiency

Cerebellar ataxia and coenzyme Q10 deficiency C. Lamperti, MD; A. Naini, PhD; M. Hirano, MD; D.C. De Vivo, MD; E. Bertini, MD; S. Servidei, MD; M. Valeriani, MD; D. Lynch, MD; B. Banwell, MD; M. Berg, MD; T. Dubrovsky, MD; C. Chiriboga, MD; C. Angelini, MD; E. Pegoraro, MD; and S. DiMauro, MD

Abstract—The authors measured coenzyme Q10 (CoQ10) concentration in muscle biopsies from 135 patients with genetically undefined cerebellar ataxia. Thirteen patients with childhood-onset ataxia and cerebellar atrophy had markedly decreased levels of CoQ10. Associated symptoms included seizures, developmental delay, mental retardation, and pyramidal signs. These findings confirm the existence of an ataxic presentation of CoQ10 deficiency, which may be responsive to CoQ10 supplementation. NEUROLOGY 2003;60:1206 –1208

In 2001, we described six patients with cerebellar ataxia and severe deficiency of coenzyme Q10 (CoQ10, ubiquinone) in skeletal muscle.1 Because this syndrome appeared to be inherited as an autosomal-recessive trait, but Friedreich ataxia and all known genetic causes of spinocerebellar ataxia had been excluded, we suggested that this was a new form of primary CoQ10 deficiency. Recognition of this entity is important because all patients have benefited from oral CoQ10 administration.1 During the past year, we have received more than 100 muscle biopsies from patients with undiagnosed cerebellar ataxia for determination of CoQ10. Here, we describe 13 new patients with childhood-onset cerebellar ataxia and marked CoQ10 deficiency in skeletal muscle. These findings confirm the existence of an ataxic syndrome associated with CoQ10 deficiency and apparently responsive to CoQ10 supplementation. Methods. Patients. We received 135 muscle biopsy specimens from patients with genetically undefined ataxic syndromes. We chose a CoQ10 concentration of 15 ␮g/g fresh tissue as the cutoff value for the definition of CoQ10 deficiency. We considered this value indicative of primary CoQ10 deficiency because it falls well below 2 SD of the mean CoQ10 concentration in 121 normal human muscle samples (27.64 ⫾ 4.43). Eighteen patients (13%) had abnormally low CoQ10 concentrations. Four of these had adultonset ataxia and will be described separately, and one child with mevalonic aciduria was excluded because this condition can cause cerebellar ataxia.2 Biochemical analyses. Spectrophotometric assays for respiratory chain complexes and for citrate synthase were performed as described.3 CoQ10 concentration in muscle extracts and in blood was measured by reverse-phase high-performance liquid chroma-

tography as described.1 We have good evidence that CoQ10 is stable in frozen muscle even after prolonged storage at temperatures of ⫺70 °C or lower. The concentration of CoQ10 was determined in at least two separate assays for each sample. Results. The clinical features of the 13 patients with infantile or childhood onset are summarized in the table. There were five male and eight female patients, ranging in age from 6 to 35 years. The first symptoms were hypotonia and motor delay in five individuals, apparent before 2 years of age; ataxia in six, manifesting between 2 and 9 years of age; and seizures in two, at ages 2 and 6 years. Ataxia became apparent in all patients by 10 years of age and affected trunk, limbs, and speech. In fact, Patient 2 never learned to walk independently, Patient 10 requires a walker at age 12, and Patients 1 and 5 never developed intelligible speech, although they interact with signs. Accurate clinical assessment of intellect is difficult owing to the significant impact of the disorder on speech and motor skills. However, most of the patients are interactive, with good verbal comprehension. Patients 2 and 3 attend regular school, and Patient 4 has a normal IQ. On the other hand, Patients 5, 6, and 9 are developmentally delayed, while Patient 10 attends a special school and is 2 years behind her grade level. Four patients had seizures, which were generalized tonic-clonic in two, psychomotor in one, and partial motor in another. Pyramidal signs were described in five patients, ranging from hyperreflexia with bilateral Babinski signs in Patient 8 to spastic paraparesis (Patient 6) or tetraparesis (Patient 9). Proximal more than distal limb weakness was described in Patients 6 and 10, but distal weakness predominated in two sisters (Patients 12 and 13). Other neurologic signs included both spontaneous and action myoclonus (Patient 4), recurrent right hemiplegia with contralateral hemispheric slowing (Patient 5), and bilateral ophthalmoparesis (Patient 10). The course is progressive. Laboratory abnormalities were mostly confined to cerebellar atrophy, documented by MRI in all patients and affecting both cerebellar hemispheres and the vermis. In Patient 5, MRI of the brain was normal at age 10 months but showed severe cerebellar atrophy at 34 months. Patient 6 had increased lactate in blood and CSF. Amino acid and organic acid profiles and liver function tests were normal in all patients. Electromyograms were also

From the Departments of Neurology (Drs. Lamperti, Naini, Hirano, and DiMauro) and Pediatrics (Drs. De Vivo and Chiriboga), Columbia University College of Physicians and Surgeons, New York, NY; the Pediatric Hospital “Bambino Gesù” (Drs. Bertini and Valeriani), Rome; the Institute of Neurology (Dr. Servidei), Catholic University, Rome, Italy; the Division of Neurology (Dr. Lynch), Children’s Hospital of Philadelphia, PA; the Division of Neurology, Department of Pediatrics (Dr. Banwell), Hospital for Sick Children, Toronto, Ontario, Canada; the Department of Neurology (Dr. Berg), Strong Memorial Hospital, Rochester, NY; the Office of Pediatric Neurology (Dr. Dubrovsky), Hollywood, FL, and the Department of Neurology (Drs. Angelini and Pegoraro), University of Padova, Italy. Supported by NIH grant NS11766, and by a grant from the National Ataxia Foundation. Also supported by a fellowship from the Associazione Amici del Centro Dino Ferrari (C.L.), Ospedale Maggiore, Istituto di Ricovero e cura a Carattere Scientifico (IRCCS), Milan, Italy. Tishcon Corp. donated CoQ10 to the patients. Received September 5, 2002. Accepted in final form December 16, 2002. Address correspondence and reprint requests to Dr. Salvatore DiMauro, 4-420 College of Physicians & Surgeons, 630 West 168th Street, New York, NY; e-mail, [email protected] 1206

Copyright © 2003 by AAN Enterprises, Inc.

Table Clinical features and muscle CoQ10 concentrations in 13 patients

Patient no.

Age, y/sex

Age at onset

Ataxia/ cerebellar atrophy

Seizure

Developmental motor delay

Mental retardation

⫹

⫺

Other symptoms

Muscle CoQ10, ␮g/g

1

24/F

2 mo

⫹

2

8/F

Birth

⫹

⫺

⫹

⫺

3

9/F

6y

⫹

Psychomotor/ pseudoabsences

⫺

⫺

4

11/M

9y

⫹

⫺

⫺

⫺

Myoclonus

10 mo

⫹

⫺

⫹

⫹

Hemiplegia

12.1

2y

⫹

Generalized tonic-clonic

⫹

⫹

Spastic paraparesis, muscle weakness

11.0

2 mo

⫹

Partial motor

⫹

⫺

Generalized tonic-clonic

12.8 Muscle weakness

9.9 2.9 5.9

5

9/M

6

12/F

7

6/M

8

10/M

5y

⫹

⫺

⫺

⫺

Pyramidal signs

9

35/F

3y

⫹

⫺

⫹

⫹

Spastic tetraparesis

9.2

10

12/F

Birth

⫹

⫺

⫹

⫹

Muscle weakness, ophthalmoparesis

12.2

8.2 12.6

11

11/M

9y

⫹

⫺

⫺

⫺

12*

27/F

2y

⫹

⫺

⫺

⫺

Pyramidal signs, distal weakness

8.7

13*

30/F

8y

⫹

⫺

⫺

⫺

Pyramidal signs, distal weakness, pes cavus

14.8

2.9

* Siblings.

normal in all patients but showed neurogenic features in Patients 9, 12, and 13. Correspondingly, muscle biopsy was morphologically normal in most patients but showed neurogenic features in Patients 9, 12, and 13. Axonal neuropathy was documented by nerve biopsy in Patient 9, who had distal sensory loss and paresthesia. There was no known consanguinity, but Patients 12 and 13 were sisters, and the brother of Patient 7 had severe behavioral problems. The patients belonged to diverse ethnic groups, including Anglo-Saxon American, African American, Ashkenazi Jewish, Italian, Albanian, and East Indian. Genetic tests for all known causes of spinocerebellar ataxia and for Friedreich ataxia were negative. We measured the activities of the mitochondrial respiratory chain enzymes in muscle extracts from nine patients: in only two (Patients 1 and 3) the activities of complex I⫹III and II⫹III were markedly decreased, suggesting a block at the CoQ10 level (data not shown). Similar data were found in another laboratory (Dr. Brian Robinson, Toronto, Canada, 2002) for Patient 10. CoQ10 concentrations were low by definition, below 15 ␮mol/g tissue in all patients, and below 10 ␮mol/g in seven (see the table). The mean value (⫾ SD) of muscle CoQ10 in the 117 patients with cerebellar ataxia not included in this series was 23 ⫾ 6.9 ␮g/g tissue. The lower mean value and greater SD in these patients than in nonataxic control subjects reflects our stringent definition of CoQ10 deficiency. Some ataxic patient with milder CoQ10 deficiency may also have primary CoQ10 deficiency. The pathologic range of CoQ10 in muscle remains to be defined, and we cannot formally exclude that low CoQ10 in muscle might be a nonspecific feature in ataxic patients. All patients have been started on oral CoQ10 supplementation, but the daily doses varied considerably (from 200 mg to 900 mg) due to local availability of the drug and financial limitations. The three adult patients (Patients 9, 12, and 13) did not respond to therapy. In all other patients, anecdotal feedback from physicians, physical therapists, and family members are positive, including improved posture and gait (Patients 1– 4, 9, 11), enhanced postural stability (Patient 10), shorter episodes of marching clonus (Patient 7), better speech articulation (Patients 7, 8, and 10). It is noteworthy that seizures, which were poorly controlled by valproic

acid, disappeared in Patient 3 and became more responsive to lower doses of diazepam in Patient 7.

Discussion. Of 135 muscle biopsies sent to us from patients with genetically undiagnosed cerebellar ataxia and atrophy, 18 (13%), had CoQ10 concentrations below 15 ␮mol/g fresh tissue. After excluding four patients with adult-onset ataxia and a child with mevalonic aciduria, we were left with 13 patients, who are remarkably similar to those we described in our initial report.1 The identification of 13 additional cases confirms the existence of an ataxic presentation of primary CoQ10 deficiency. Cerebellar ataxia and cerebellar atrophy have been present in all 19 patients described thus far. Seizures are the most common associated feature, afflicting seven (37%) of all patients. Pyramidal signs (5/13 as compared to 1/6) and mental retardation (5/13 as compared to 2/6) were more common in the present series. Weakness was reported in all six initial patients but in only five patients in the current series. Nevertheless, delayed motor milestones were noted in half of the present cohort. Routine laboratory tests are not very useful: lactic acidosis is inconsistent, and both amino acid and organic acid profiles are normal. We have found low serum CoQ10 in a few patients, but we could not obtain pretherapy levels in all and we are still establishing the range of normal blood values. If a correlation could be documented between blood and muscle CoQ10 levels, blood tests would offer a valuable indication for muscle biopsy. April (1 of 2) 2003

NEUROLOGY 60

1207

In contrast to the myopathic form of CoQ10 deficiency (see below), muscle morphology is not helpful in ataxic patients, and muscle biopsies are often read as normal. Likewise, muscle enzymology only inconsistently shows the expected block in the transfer of electrons to complex III, probably due to the partial deficiency of CoQ10. Thus, presently the diagnosis of this form of ataxia requires measurement of CoQ10 in skeletal muscle. Primary CoQ10 deficiency is a mitochondrial encephalomyopathy with heterogeneous clinical presentations. An apparently rare myopathic form is dominated by exercise intolerance and recurrent myoglobinuria but is associated with CNS dysfunction, including ataxia, seizures, and developmental delay.4-7 Muscle biopsy in these patients characteristically shows both ragged-red fibers and lipid storage myopathy. A third clinical variant, described in one family, included infantile-onset encephalomyopathy (with ataxia) and renal disease in three siblings, one of whom died of renal failure at age 8 years.8 Yet another presentation of CoQ10 deficiency was adult Leigh syndrome in two sisters with encephalopathy, growth retardation, ataxia, deafness, and lactic acidosis.9 This clinical heterogeneity suggests that different biochemical and molecular defects may be involved in causing individual syndromes and raises the question whether the ataxic variant is, in fact, a single entity. This question can only be answered by studies aimed at identifying the specific faulty step

1208 NEUROLOGY 60

April (1 of 2) 2003

in the complex biosynthetic pathway of CoQ1010 and the underlying genetic defects. Within 1 year, we have identified 13 new cases of cerebellar ataxia associated with CoQ10 deficiency, indicating that this syndrome may be a relatively common cause of cerebellar atrophy in children. Diagnosis is important because there is some evidence that patients may benefit from early CoQ10 supplementation. References 1. Musumeci O, Naini A, Slonim AE, et al. Familial cerebellar ataxia with muscle coenzyme Q10 deficiency. Neurology 2001;56:849 – 855. 2. Hoffmann GF, Charpentier C, Mayatepek E, et al. Clinical and biochemical phenotype in 11 patients with mevalonic aciduria. Pediatrics 1993;91:915–921. 3. DiMauro S, Servidei S, Zeviani M, et al. Cytochrome c oxidase deficiency in Leigh syndrome. Ann Neurol 1987;22:498 –506. 4. Ogasahara S, Engel AG, Frens D, Mack D. Muscle coenzyme Q deficiency in familial mitochondrial encephalomyopathy. Proc Natl Acad Sci USA 1989;86:2379 –2382. 5. Sobreira C, Hirano M, Shanske S, et al. Mitochondrial encephalomyopathy with coenzyme Q10 deficiency. Neurology 1997;48:1238 –1243. 6. Di Giovanni S, Mirabella M, Spinazzola A, et al. Coenzyme Q10 reverses pathological phenotype and reduces apoptosis in familial CoQ10 deficiency. Neurology 2001;57:515–518. 7. Boitier E, Degoul F, Desguerre I, et al. A case of mitochondrial encephalomyopathy associated with a muscle coenzyme Q10 deficiency. J Neurol Sci 1998;156:41– 46. 8. Rötig A, Appelkvist E-L, Geromel V, et al. Quinone-responsive multiple respiratory-chain dysfunction due to widespread coenzyme Q10 deficiency. Lancet 2000;356:391–395. 9. Van Maldergem L, Trijbels F, DiMauro S, et al. Coenzyme Q-responsive Leigh encephalopathy in two sisters. Ann Neurol 2002;52:750 –754. 10. Murthy V. Coenzyme-Q and related isoprenoid compounds: biosynthesis, regulation, functions and biomedical implications. In: Ebadi M, Marwah J, Chopra R, eds. Mitochondrial ubiquinone (coenzyme Q10). Scottsdale: Prominent Press; 2001,231–345.