Chronic GM1 Gangliosidosis with Characteristic 'Wish Bone Sign' on Brain MRI: Another Type of Neurodegeneration with Brain Iron Accumulation?

View the entire article with references and supplemental information on the Wiley Online Library. 
Note:  Reference links embedded in the article below will also take you to the article on the Wiley Online Library.

Return to Table of Contents

Videos show her clinical course over the past 10 years with cardinal features of slowly progressive generalized dystonia with marked oromotor dystonia. Segment 1: Video at 18 years depicts significant generalized dystonia with oromotor dystonia, which is prominent in NBIAs. She is still able to walk with support. Strabismus, which is described among one of the early findings of late-onset or juvenile GM1 gangliosidosis is also observed in this patient. Segment 2: Video at 22 years shows that she gradually lost her ability to walk and now is wheelchair bound. Segment 3: Video at 27 years shows her slow gradual progression of symptoms. However, all three videos show that, compared to her motor symptoms, her cognitive and emotional capacity is well preserved and can respond well to commands with good understanding.

Authors: Omkar Hajirnis and Anaita Udwadia-Hegde

Article first published online:2 JUN 2015 | DOI: 10.1002/mdc3.12197

Case Report

This 27-year-old female presented with delay in motor milestones noted in infancy followed by slowly progressive motor disability starting at around 3 to 4 years of age and by speech difficulties appearing shortly thereafter.

She was born at term from nonconsanguineous healthy parents. Family history was negative. She started to walk late, at 2 years of age, with an unsteady gait. She developed normal speech until around 3 years of age, when she could speak approximately three- to four-word sentences, after which she was noted to have started regressing in her motor followed by language milestones, which started with a stammer. She progressed over the next few years from being ataxic for a while followed by development of marked generalized dystonia, which has slowly progressed since then, along with significant oromotor dystonia. Speech impairment worsened to the point of unintelligibility in the subsequent years. Cognitively, she remained well with no evidence of deterioration. No seizures were ever noted. Furthermore, no visual impairment was observed. Her subsequent clinical course over the years is characterized by progressive spastic-dystonic tetraparesis with anarthria with marked oromotor dystonia.


Findings on general physical examination when first observed by us at 18 years of age were normal with no obvious dysmorphic features. Neurological examination revealed extrapyramidal signs with severe generalized dystonia and rigidity. Speech was severely impaired because of oromandibular dystonia. The remainder of her other systemic examination was unremarkable with no evidence of organomegaly. Fundoscopy revealed no abnormality.

Workup for her over several years included routine blood tests, alpha-fetoprotein, vitamin E, copper, and ceruloplasmin titers, which were negative. Metabolic workup, including blood and urine chromatography of amino acids and urinary organic acids, as well as peroxisomal function tests, were unremarkable. Neurophysiology included visual and brainstem auditory evoked potentials, electromyography, and nerve conduction velocities, which were all normal.

MRI done at 3.5 years of age was reported to show patchy hyperintense lesions in the para ventricular deep white matter (WM) along the posterior part of bodies of the lateral ventricles and occipital horns on both sides on T2-weighted images. Repeat MRI brain at 13 and 25 years, apart from showing the previous WM changes, also showed bilateral hypointensities of the globus pallidus (GP) on T1-weighted images with corresponding low signal intensity on gradient-echo T2-weighted images, which was consistent with susceptibility effects from paramagnetic iron deposition. The susceptibility-weighted images (SWIs) done on the last scan revealed progression of the iron deposition involving the GP and spreading to the anterior SN. However, specific signal-intensity abnormalities, such as the “eye-of-the-tiger” sign, was not observed in any of the scans.

In view of the above-described findings, DNA sequence analysis of the PANK2 gene was sent in suspicion of atypical pantothenase kinase-associated neurodegeneration, which did not reveal any abnormalities. She was evaluated for other causes of neurodegeneration with brain iron accumulation (NBIA), such as neuroferrinopathy, infantile neuroaxonal dystrophy, and aceruloplasminemia, with iron studies, including serum ferritin levels, PLA2G6 sequence analysis, and serum ceruloplasmin levels, respectively, which all were negative. Peripheral smears for acanthocytes were also negative. Serum magnesium levels were normal. Considering a treatable differential diagnosis of Wilson's disease, she was evaluated and ruled out for the same.

In view of inconclusive investigations and slowly progressive course over several years, a genetic panel for dystonia was sent followed by advanced metabolic testing, confirming a diagnosis of chronic GM1 gangliosidosis.

Table 1 depicts the index child's and her parent's genetic variations along with their enzyme levels.

Table 1. β-galactosidase enzyme levels with correlated genetics of the family
  β-Galactosidase Levels Normal Range Genetics and Co- relationa Mutation Status Clinical Status
  1. a  DNA isolated from blood was used to perform targeted region capture using a custom capture kit. Libraries were sequenced to 80 to 100× coverage on an Illumina sequencing platform. Sequences obtained were aligned to human reference genome (GRCh37/hg19) using the BWA program and analyzed using the Picard and GATK-Lite toolkit to identify variants in whole exome relevant to clinical indication. Only nonsynonymous and splice site variants found in the targeted genes were used for clinical interpretation. Bold indicates abnormal findings.
Index patient 4.8 70–324 nmol/h/mg (1) Unreported heterozygous missense variation in exon 3 of the GLB1 gene (chr3: 33110442; T> T/C), which resulted in an amino acid substitution of arginine for histidine at codon 89 (p.H89R; ENST00000307363), was detected Compound heterozygote Symptomatic
(2) A previously reported heterozygous missense variation in exon 13 of the GLB1 gene (chr3: 33059962; C>C/T), which resulted in the amino acid substitution of glutamine for arginine at codon 442 (p.R442Q; ENST00000307363), was detected.
Father 55   Chr3: 33110442; T> T/C (Het); c.266A>A/G; p.H89R in exon 3 of GLB1 gene Present (heterozygous state) Asymptomatic
Chr3: 33059962; C>C/T (Het);c.1325G >G/A; p.R442Q in exon 13 of GLB1 gene Absent
Mother 55   Chr3: 33059962; C>C/T (Het);c.1325G >G/A; p.R442Q in exon 13 of GLB1 gene Present (heterozygous state) Asymptomatic
Chr3: 33110442; T> T/C (Het); c.266A>A/G; p.H89R in exon 3 of GLB1 gene Absent

Reviewing literature on the MRI imaging findings in GM1 gangliosidosis, we found that her initial scans, which were suggestive of T2-weighted signal intensity of the supratentorial WM, could be consistent with abnormal myelination observed in late infantile or juvenile-onset GM1 gangliosidosis.[1] Repeat scans over the years showed susceptibility effects from progressive paramagnetic iron deposition. This has been uncommonly reported in the literature.[2-6] Distribution of the deposition is typically in a “wish bone” pattern with the medial and lateral parts of globus pallidi forming the forked ends and the extension to the anterior SN and red nucleus forming the stem of the wish bone. Hence, we would like to coin the wish bone sign to this characteristic image finding (Fig. 1). In GM1 gangliosidosis, this deposition is owing to iron overload resulting from a defect in intralysosomal recycling.[6]

Serial MRIs in a patient with GM1 gangliosidosis

Figure 1. Serial MRIs in a patient with GM1 gangliosidosis at 13  (A and B) and 25 years of age (C and D). (A) Axial T2-weighted images show bilateral symmetric hypointense lesions in the globus pallidi (white arrows) with hyperintensities observed in the putamen (black arrow), which is described in type III GM1 gangliosidosis. Furthermore, asymmetric lesions are observed in the subcortical WM with cortical atrophy. (B) Corresponding axial gradient echo images (G.R.E) reveal bilaterally symmetrical iron deposition in the globus pallidi (black arrows). (C) Axial T2-weighted images again show hypointense pallida (white arrows) with mild progression, compared to previous scan, putaminal hyperintensities, atrophy of the caudate nuclei, and diffuse cortical atrophy with ventriculomegaly. (D) Evidence of exaggerated SWI hypointensities point to progression of iron deposition in bilateral globus pallidi with similar deposits in the SN, red nucleus, and subthalamic nuclei. The arrow shows the proposed “wish bone” sign, which takes its name from the pattern of iron accumulation seen here, wherein the medial (GPi) and lateral parts (GPe) of globus pallidi make the forked ends and the extension downward the pallidi to the anterior SN and red nucleus form the stem of the wish bone.

In conclusion, we have highlighted that chronic GM1 gangliosidosis, though rare, should be kept in mind in young patients presenting with slowly progressive generalized dystonia, even without the presence of visceromegaly or dysmorphic features. It is an important differential diagnosis of NBIA not only clinically, but also on imaging findings and hence we propose should be included in the list as a new type of NBIA.[7

PWT is, per definition, task specific and is elicited either by writing (type A: task-induced tremor) or by adopting a writing hand position (type B: positional sensitive tremor).[2] Although present in the very first patient described with this condition,[1] it is atypical for PWT that other tasks elicit a tremor.

Apart from overt dystonic posturing upon writing, several clinical characteristics in the presented case are reminiscent of features frequently encountered in dystonic conditions: As shown in Segment 3 of Video 1, the occurrence of the tremor depends on the task attempted. Furthermore, only writing executed by distal, but not by proximal, muscles provokes the tremor (see Video 1, Segment 4), which had actually been described in the very first PWT case as well.[1] Ultimately, the occurrence of a mirror tremor, elicited by writing with the contralateral hand, could be viewed in analogy to mirror dystonia, a particularly common finding in focal hand dystonia.[8] To our knowledge, this is the first video documentation of a mirror tremor, although the phenomenon has been described previously.[2]

Taken together, dystonic posturing, task specificity, triggering by distal, but not proximal, muscle activity and the mirror tremor phenomenon argue for a dystonic origin of the writing tremor observed in this case.

Concerning the debated issue of whether PWT is a variant of dystonic tremor or ET, there have been cases of PWT in both hands with a positive family history of ET and positive response to propranolol,[9] as well as cases of PWT and writers cramp in the same patient described in the literature.[7] The clinical features noted in our patient at least make a strong argument for dystonia being the pathophysiological basis for the writing tremor observed in this case.


Leave a Comment

The following required items were not provided or are in the wrong format. Please provide the required responses and submit again:

Comment Title
Comment: 1000 characters
  [[put error message here]]