|
| |||||||||||||||||||||||||||||||||||||||||||||||||
Background HallervordenSpatz syndrome is an autosomal recessive disorder characterized by dystonia, parkinsonism, and iron accumulation in the brain. Many patients with this disease have mutations in the gene encoding pantothenate kinase 2 (PANK2); these patients are said to have pantothenate kinaseassociated neurodegeneration. In this study, we compared the clinical and radiographic features of patients with HallervordenSpatz syndrome with and without mutations in PANK2.
Methods One hundred twenty-three patients from 98 families with a diagnosis of HallervordenSpatz syndrome were classified on the basis of clinical assessment as having classic disease (characterized by early onset with rapid progression) or atypical disease (later onset with slow progression). Their genomic DNA was sequenced for PANK2 mutations.
Results All patients with classic HallervordenSpatz syndrome and one third of those with atypical disease had PANK2 mutations. Whereas almost all mutations in patients with atypical disease led to amino acid changes, those in patients with classic disease more often resulted in predicted protein truncation. Patients with atypical disease who had PANK2 mutations were more likely to have prominent speech-related and psychiatric symptoms than patients with classic disease or mutation-negative patients with atypical disease. In all patients with pantothenate kinaseassociated neurodegeneration, whether classic or atypical, T2-weighted magnetic resonance imaging (MRI) of the brain showed a specific pattern of hyperintensity within the hypointense medial globus pallidus. This pattern was not seen in any patients without mutations.
Conclusions PANK2 mutations are associated with all cases of classic HallervordenSpatz syndrome and one third of cases of atypical disease. A specific MRI pattern distinguishes patients with PANK2 mutations. Predicted levels of pantothenate kinase 2 protein correlate with the severity of disease.
Recently we discovered that many cases of HallervordenSpatz syndrome result from mutations in a gene located on chromosome 20p13.8 The culprit gene (PANK2) encodes a novel pantothenate kinase, a key regulatory enzyme in the biosynthesis of coenzyme A.8 We speculated that a deficiency of this enzyme could lead to neurodegeneration and iron accumulation, and we proposed a new descriptive term for the resulting autosomal recessive error of metabolism: pantothenate kinaseassociated neurodegeneration (Online Mendelian Inheritance in Man number 234200).
In the current study, we determined the genotype of patients with classic and atypical HallervordenSpatz syndrome and reviewed their medical records to assess whether there was a correlation between clinical manifestations or radiographic findings and the presence of PANK2 mutations.
Methods
Subjects
Subjects were recruited internationally by means of announcements posted in neurology and genetics journals, on a pediatric-neurology list server, and on the Web sites of the HallervordenSpatz Syndrome Association (http://www.hssa.org) and the National Organization for Rare Disorders (http://www.rarediseases.org). We identified 186 patients from 145 families who had extrapyramidal defects and radiographic evidence of iron deposition in the basal ganglia. We examined 28 of the subjects. For the remainder, samples were sent by referring physicians and clinical and laboratory data were obtained from records and MRI scans. Detailed clinical information and DNA samples were collected from the patients after they had given written informed consent. The protocol was approved by the institutional review boards of the Oregon Health and Science University or the University of California, San Francisco. We gathered all possible clinical information, including the age at onset of disease and presenting symptoms; results of neurologic and ophthalmologic examinations; results of electrical studies, including electroencephalography, electromyography, electroretinography, and visual evoked potentials; and brain MRI scans. Clinical information was insufficient for 63 of the 186 patients, and our analysis is therefore based on the remaining 123 patients from 98 families.
Patients were classified as having either classic disease (66 patients) or atypical disease (57 patients). Early-onset, rapidly progressive (classic) disease was usually evident by the age of 10 years and was manifested by dystonia, progression to severe disability by 20 years of age, and radiographic changes indicating a high iron content in the basal ganglia. Those with atypical disease included all other patients with extrapyramidal dysfunction and radiographic evidence of iron accumulation in the basal ganglia. The atypical disease usually also had a later onset and a more slowly progressive course.
Analysis of Mutations
Primers were designed to amplify each of the seven exons of PANK2. Polymerase-chain-reactionamplified DNA was sequenced in both the forward and the reverse directions and compared with control DNA. Mutations were classified as null if they were predicted to result in premature protein termination (including frame shifts, nonsense mutations, and mutations altering splice-donor and acceptor sequences); missense mutations were defined as those causing the substitution of one amino acid for another.
Statistical Analysis
Chi-square analyses were conducted with the use of an online tool (http://www.georgetown.edu/cball/webtools/web_chi.html).
Results
Genetic Findings
PANK2 mutations were found in 66 of the 98 families of patients with HallervordenSpatz syndrome (Table 1). Of 49 families whose members had classic disease, all had mutations in PANK2. Of 49 families whose members had atypical disease, mutations were found in 17 (35 percent). Null mutations were found in 36 of 92 alleles in patients with classic disease, but in only 2 of 31 alleles in patients with atypical disease. All patients with two null alleles had the classic form of the disease.
|
An intriguing feature of the G411R mutation is that in six families harboring this mutation (four with classic disease and two with atypical disease), no mutation was detected on the other chromosome. Families with only one identified mutation were not distinguishable from those with two. With our current strategy, some mutations would be undetectable (e.g., promoter mutations). However, of nine families with single mutant alleles, six had an allele with a G411R mutation. This observation is striking because mutations in both alleles were detected in nearly all families, and it suggests that G411R might be semidominant, with one allele sufficient to cause disease given certain genetic backgrounds. Contrary to this hypothesis, no disease phenotype was observed in G411R-heterozygous carrier parents of affected persons. Environmental exposure or modifier effects of other genes, including those for enzymes downstream in the coenzyme A synthetic pathway, might also have a role in the pathogenesis of the disease, in concert with the G411R allele.
Clinical Findings
On the basis of the existing clinical information, the 123 patients who were studied were not different from the 63 who were excluded because of insufficient clinical information. Information about each clinical characteristic was not available for every patient included in our study cohort.
The clinical features of our cohort of 66 PANK2-mutationpositive patients with classic disease were remarkably homogeneous. Pantothenate kinaseassociated neurodegeneration usually presented before the age of 6 years (in 88 percent of cases), with a mean (±SD) age at onset of 3.4±3.0 years (range, 0.5 to 12). The most common presenting symptoms were gait or postural difficulties, which occurred in 40 of 51 patients for whom information was available (78 percent). These symptoms were much less common among patients who had the later-onset form of the disease or who did not have PANK2 mutations (P<0.001).
The predominant neurologic features were extrapyramidal and included dystonia, dysarthria, rigidity, and choreoathetosis (51 of 52 patients [98 percent]). Dystonia was a nearly constant early manifestation (45 of 52 patients [87 percent]). Early dystonia often involved the cranial and limb musculature, with axial dystonia predominating later. Involvement of the corticospinal tract, with spasticity, hyperreflexia, and extensor toe signs, was common (13 of 52 patients [25 percent]), as was cognitive decline (15 of 52 patients [29 percent]). Seizures were not reported in any patient with classic disease. Forty-five of 66 patients with classic disease (68 percent) had clinical or electroretinographic evidence of retinopathy. Optic atrophy was infrequent, occurring in only 2 of 66 patients (3 percent). Acanthocytosis was reported in 8 percent of patients with classic disease. Since acanthocytosis is not sought routinely, its true prevalence among patients with pantothenate kinaseassociated neurodegeneration remains uncertain. We observed that classic pantothenate kinaseassociated neurodegeneration progressed at a nonuniform rate, with periods of marked deterioration, often lasting one to two months, interspersed with longer periods of clinical stability. The majority of patients with classic pantothenate kinaseassociated neurodegeneration (85 percent) became nonambulatory within 15 years after the onset of the disease.
The clinical features of the 23 patients with atypical HallervordenSpatz syndrome and PANK2 mutations were heterogeneous. These patients were significantly older at the onset of the disease than patients with classic disease (13.7±5.9 years [range, 1 to 28] vs. 3.4±3.0 years [range, 0.5 to 12], P<0.001). In rare cases, these patients had very early nonspecific problems (3 of 20 patients for whom information was available [15 percent]), including developmental delay (2 of 20 patients [10 percent]). Extrapyramidal defects developed in 16 of 22 patients with atypical disease (73 percent), but dystonia and rigidity were generally less severe and more slowly progressive than in patients with classic disease. Most of these patients (14 of 22 [64 percent]) continued to be able to walk into adulthood, but in many the disease eventually progressed to loss of independent ambulation. Spasticity, hyperreflexia, and other signs of corticospinal tract involvement were common (3 of 17 patients [18 percent]) and progressive, eventually limiting ambulation. Freezing was reported in 3 of 20 patients with atypical disease (15 percent). Clinical evidence of retinopathy or optic atrophy was much less common than in patients with classic disease (3 of 15 patients [20 percent], P<0.001).
An unexpected finding was that in 9 of the 23 patients with atypical pantothenate kinaseassociated neurodegeneration (39 percent), difficulty with speech, including palilalia (repetition of words or phrases) and dysarthria, was either the sole presenting feature or part of the early disease. In contrast, no patient with classic pantothenate kinaseassociated neurodegeneration presented with a speech defect (although dysarthria developed later in 16 of these patients). Psychiatric symptoms with cognitive decline, reminiscent of frontotemporal dementia, were prominent in patients with atypical pantothenate kinaseassociated neurodegeneration (6 of 18 patients for whom information was available [33 percent]) and rare in patients with classic pantothenate kinaseassociated neurodegeneration; these symptoms included personality changes with impulsivity and violent outbursts, depression, and emotional lability.
Additionally compelling is the clinical comparison between patients with a diagnosis of atypical HallervordenSpatz syndrome who had a PANK2 mutation and those who did not. Among the patients for whom the presenting symptoms were noted in the medical records, 6 of 18 with atypical disease and PANK2 mutations presented with speech difficulties, whereas none of the 17 with atypical disease who had no PANK2 mutations had this presentation (P<0.05). Psychiatric symptoms developed in 6 of 18 patients with atypical disease and PANK2 mutations but in no patients with atypical disease without PANK2 mutations (P<0.05). Otherwise, patients without PANK2 mutations were similar to those with mutations: they generally had extrapyramidal and corticospinal tract dysfunction; their mean age at onset of the disease was 7.0±9.9 years (range, 0.5 to 38); and their family histories indicated that they had affected siblings or that their cases were sporadic, both findings that are consistent with autosomal recessive inheritance.
Radiographic Findings
A striking correlation was found between the MRI findings and the presence or absence of PANK2 mutations in patients with HallervordenSpatz syndrome. All MRI scans reviewed from 28 patients with PANK2 mutations (24 with classic disease and 4 with atypical disease) showed bilateral areas of hyperintensity within a region of hypointensity in the medial globus pallidus on T2-weighted images, a pattern known as "eye of the tiger"9 (Figure 1). Moreover, reports of MRI scans from 41 additional mutation-positive patients (36 with classic and 5 with atypical disease) described them in detail as showing these specific changes. Indeed, no PANK2-mutationpositive patients lacking the eye-of-the-tiger sign have been found.
|
On the basis of this correlation, we assessed the value of the brain MRI alone in predicting mutation status. In a small subgroup of symptomatic patients with HallervordenSpatz syndrome who were not included in our study because of insufficient clinical information, we identified six patients solely by the presence of the eye-of-the-tiger sign and analyzed their DNA for PANK2 mutations. All six patients were found to have PANK2 mutations on both chromosomes, a result further supporting the correlation between the presence of these mutations and the eye-of-the-tiger sign.
Discussion
We identified mutations in the PANK2 gene in a large subgroup of patients with HallervordenSpatz syndrome. All patients with classic disease had PANK2 mutations, suggesting that patients with early-onset, rapidly progressive disease will consistently prove to have inherited defects in pantothenate kinase 2. Among patients with atypical disease, those with PANK2 mutations were much more likely to have speech and psychiatric problems than were mutation-negative patients. Most compelling is our finding of a one-to-one correlation between the MRI eye-of-the-tiger pattern and the presence of a PANK2 mutation, regardless of the severity of the disease. In Table 2, we propose clinical descriptions of the major forms of pantothenate kinaseassociated neurodegeneration.
|
Two clinical observations warrant further discussion. First, although the profile of classic pantothenate kinaseassociated neurodegeneration is consistent with that presented in the many reports of patients with early-onset HallervordenSpatz syndrome,1,3,10,11,12 pigmentary degeneration of the retina was much more common in our patients than has been reported in others.1,13 Retinopathy develops early, though it may be recognized only when a full diagnostic evaluation is performed. As a corollary, we found it uncommon for retinopathy to develop later in patients with normal results on funduscopic examination at the time of diagnosis.
Second, we have delineated the dominant neurologic features of patients with atypical pantothenate kinaseassociated neurodegeneration, which differ from those seen in the classic form of the disease. These features, which include severe palilalia, dysarthria, dystonia (which may be intermittent and fluctuating), perseverative behavior and movements, and freezing, have not been widely recognized in this disease.14,15,16 Freezing during ambulation, especially when turning corners or encountering surface variations, is strikingly similar to changes seen in Parkinson's disease.15 Psychiatric symptoms, including depression and psychosis, were common in the patients with atypical disease in our study. In several patients with atypical disease whose first major symptom was palilalia, psychiatric abnormalities were misinterpreted as the cause of the speech defect. Progressive dementia occurs in atypical pantothenate kinaseassociated neurodegeneration.
The one-to-one correlation between a specific radiographic finding, the eye-of-the-tiger sign, and mutations in a single gene, PANK2, is extraordinary. Clearly, the medial globus pallidus is especially sensitive to deficiency of pantothenate kinase 2, one of four pantothenate kinases encoded by the human genome. The eye-of-the-tiger pattern may reflect tissue necrosis and edema (seen on T2-weighted MRI as hyperintensity) within a region of iron deposition (seen as hypointensity), but how a deficiency in pantothenate kinase 2 leads to these changes remains speculative. Previously, we suggested that accumulated cysteine, which would normally condense with phosphopantothenate, and cysteine-containing compounds may form complexes with iron and exacerbate oxidative damage in this brain structure.8
Pantothenate kinaseassociated neurodegeneration is one of three extrapyramidal disorders associated with increased amounts of brain iron for which the molecular basis has been defined. The other two disorders, neuroferritinopathy, which results from mutations in the ferritin light-chain gene,17 and aceruloplasminemia, which results from mutations in the gene encoding ceruloplasmin,18 are of adult onset and can be distinguished from pantothenate kinaseassociated neurodegeneration on the basis of the clinical presentation, MRI findings, and results of genetic testing.
Additional biochemical markers of pantothenate kinaseassociated neurodegeneration may help us understand its pathologic effects and develop therapeutic targets. Acanthocytosis and a defect in plasma lipoproteins have been associated with mutations in PANK2.19 These systemic features will help focus future biochemical investigations. With better understanding of the metabolic perturbations in pantothenate kinaseassociated neurodegeneration, we may be able to devise neuroprotective interventions to prevent exacerbations and thereby slow the progression of the disease.
Currently, however, we are left with no imaging or DNA-based tool for the definitive diagnosis of the disease in two thirds of patients with atypical HallervordenSpatz syndrome. Conceivably, pantothenate and coenzyme A metabolism may be perturbed in these patients or in patients with other neurodegenerative disorders. For example, patients with normal amounts of brain iron who have early-onset parkinsonism, severe speech impairment, perseverative movements, or pigmentary retinopathy may have alterations in these metabolic pathways. Indeed, the spectrum of diseases that are associated with inborn errors of pantothenate or coenzyme A metabolism is likely to expand.
The eponymous term "HallervordenSpatz syndrome" has fallen into disfavor in view of the unethical activities of the German neuropathologists Hallervorden and Spatz during World War II.20 We encourage the use of the term "pantothenate kinaseassociated neurodegeneration" for the majority of patients with HallervordenSpatz syndrome who have proved or suspected mutations in PANK2. For the remainder, we propose the term "neurodegeneration with brain iron accumulation."
Supported by grants from the National Eye Institute and the Sandler Neurogenetics Center at the University of California, San Francisco. Dr. Gitschier is an investigator of the Howard Hughes Medical Institute.
We are indebted to the patients and their families; to our clinical collaborators Drs. J. Jankovic, M. Thomas, J. Guimarães, J. Molinuevo, F. Martí, M.A.M. Salih, J. Coppeto, J.P. Harpey, M. Shevell, M. Piñeda, G. Kurlemann, S. Davis, H. Hattori, K. Sethi, M. Pandolfo, L. Angelini, N. Nardocci, A. Malandrini, J. Penzien, G. Mortier, M. Hoeltzenbein, J.A. Urtizberea, D. Buckley, C. Haenggeli, A. Bottani, B. Beinlich, J. Østergaard, S. Bundey, F. Stögbauer, K. Nørgaard Hansen, C. Yalcinkaya, A. Feigenbaum, Z. Liptai, J. Carlo, P. Blasco, A. Zimmerman, R. Cilio, E. Bertini, G. Worley, U. Thyen, M. Melis, G. Cossu, J. Menkes, K. Hollódy, A. Barrett, S. Simpson, C. Schrander-Stumpel, H. Chaabouni, R. Gatti, H. Topaloglu, M. Nigro, F. Hisama, N.R.M. Buist, B. Ben-Zeév, A. Macaya, B. Korf, P. Heydemann, S. Abbs, R. Robinson, L. Shinobu, E. Dooling, P. Wheeler, P. Rosman, W. Wasiewski, P. Castelnau, P. Evrard, R. Haslam, M. Filocamo, M. Karwacki, T. Kmieæ, S. Frucht, T. Konishi, M. Regina Reyes, M. Al-Mateen, K. Weidenheim, M. Delgado, S. Johnsen, S. Golembowski, W. Ondo, S. Bohlega, R. Bustamante, O. Fernandez, M. Wiznitzer, H. Morgan, I. Butler, T. Babb, D. Sanderson, M. Williams, C. Harding, R. Steiner, S. Toor, E. Thompson, J. MacKenzie, J. Clyman, J.J. Higgins, and D. Fornoff; to P. Hogarth and J. Nutt for critical review of the manuscript; to the Oregon Health and Science University Molecular Microbiology and Immunology Sequencing Core; and to the HallervordenSpatz Syndrome Association.
Source Information
From the Departments of Molecular and Medical Genetics, Pediatrics, and Neurology, School of Medicine, Oregon Health and Science University, Portland (S.J.H., S.K.W., M.A.J., K.H.L.C.); and the Howard Hughes Medical Institute and the Departments of Medicine and Pediatrics, University of California, San Francisco (B.L., B.Z., J.G.).
Address reprint requests to Dr. Hayflick at the Department of Molecular and Medical Genetics, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd., Mailcode L-103a, Portland, OR 97201-3098, or at hayflick{at}ohsu.edu.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2010 Massachusetts Medical Society. All rights reserved. |