Partial deletion of AFG3L2 causing spinocerebellar ataxia type 28

Katrien Smets, MD Tine Deconinck, MSc Jonathan Baets, MD, PhD Anne Sieben, MD Jean-Jacques Martin, MD, PhD Iris Smouts, BSc Shuaiyu Wang, MSc Franco Taroni, MD Daniela Di Bella, MD, PhD Wim Van Hecke, PhD Paul M. Parizel, MD, PhD Christina Jadoul, MD Robert De Potter, MD Francine Couvreur, MD Elena Rugarli, MD Peter De Jonghe, MD, PhD

Correspondence to Dr. De Jonghe: [email protected]

ABSTRACT

Objective: To identify the genetic cause of autosomal dominant spinocerebellar ataxia type 28 (SCA28) with ptosis in 2 Belgian families without AFG3L2 point mutations and further extend the clinical spectrum of SCA28 through the study of a brain autopsy, advanced MRI, and cellbased functional assays exploring the underlying disease mechanism.

Methods: Two large families were clinically examined in detail. Linkage analysis and multiplex amplicon quantification were performed. A brain autopsy was obtained. Brain MRI with voxel-based morphometry and diffusion tensor imaging was performed. RNA and Western blot analysis and blue native–polyacrylamide gel electrophoresis experiments were performed. Results: MRI analysis demonstrated a significant cerebellar atrophy, as well as white matter degeneration in the cerebellar peduncles, corticospinal tracts, corpus callosum, and cingulum. A brain autopsy showed severe atrophy of the upper part of the cerebellar hemisphere. Ubiquitin and p62 immunoreactive intranuclear inclusions were found in cerebral and cerebellar cortical neurons, in neurons of the hippocampus, and in pontine and medullary nuclei. An identical heterozygous partial deletion of exons 14 to 16 of the AFG3L2 gene was found in both families. Additional functional assays in patient-derived cell lines revealed haploinsufficiency as the underlying disease mechanism.

Conclusions: Our study expands the phenotypic characterization of SCA28 by means of brain pathology and diffusion tensor imaging/voxel-based morphometry MRIs. The identification of a partial AFG3L2 deletion and the subsequent functional studies reveal loss of function as the most likely disease mechanism. Specific testing for deletions in AFG3L2 is warranted because these escape standard sequencing. Neurology® 2014;82:2092–2100 GLOSSARY AFG3L2 5 ATPase Family Gene 3-Like 2; BN-PAGE 5 blue native–polyacrylamide gel electrophoresis; DRPLA 5 dentatorubralpallidoluysian atrophy; DTI 5 diffusion tensor imaging; M 5 musculus; MAQ 5 multiplex amplicon quantification; mRNA 5 messenger RNA; SCA 5 spinocerebellar ataxia; SCA28 5 spinocerebellar ataxia type 28; SPAX5 5 spastic ataxia 5; SPG7 5 spastic paraplegia 7; VBM 5 voxel-based morphometry.

Supplemental data at Neurology.org

Spinocerebellar ataxias (SCAs) are a clinically and genetically heterogeneous group of neurodegenerative cerebellar disorders.1,2 Autosomal dominant SCA type 28 (SCA28) represents one of 11 autosomal dominant SCA subtypes not caused by repeat expansions.3–9 SCA28 is characterized by a juvenile-onset slowly progressive cerebellar ataxia, with ophthalmoparesis and ptosis as prominent features. Heterozygous mutations in AFG3L2 cause SCA28.6 A total of 17 families have been published to date.3–9 AFG3L2 encodes a subunit of the m-AAA protease, a component of the ATP-dependent metalloprotease, located on the inner mitochondrial membrane. AFG3L2 is highly homologous to paraplegin, the causal gene of autosomal recessive hereditary spastic paraplegia type 7 (SPG7).10 A homozygous AFG3L2 missense mutation has been reported in a single autosomal recessive spastic ataxia 5 (SPAX5) family.11 We expanded the genetic From the Neurogenetics Group, VIB–Department of Molecular Genetics (K.S., T.D., J.B., I.S., P.D.J.), and Laboratories of Neurogenetics and Neuropathology, Institute Born-Bunge (K.S., T.D., J.B., A.S., J.-J.M., P.D.J.), University of Antwerp; Departments of Neurology (K.S., J.B., J.-J.M., I.S., P.D.J.) and Radiology (W.V.H., P.M.P.), Antwerp University Hospital; Department of Neurology (A.S.), Ghent University Hospital, Belgium; Biocenter (S.W., E.R.), University of Cologne; Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (S.W., E.R.), Cologne, Germany; Unit of Genetics of Neurodegenerative and Metabolic Diseases (F.T., D.D.B.), Fondazione IRCCS Istituto Neurologico Carlo Besta, Milan, Italy; Icometrix (W.V.H., P.M.P.), Leuven; Department of Neurology (C.J.), AZ Nicolaas, Sint-Niklaas; Department of Neurology (R.D.P.), AZ Sint-Lucas, Gent; and Department of Neurology (F.C.), AZ Klina, Brasschaat, Belgium. Go to Neurology.org for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article.

2092

© 2014 American Academy of Neurology

spectrum of SCA28 through the identification of an identical partial deletion of exons 14 to 16 of AFG3L2 in 2 autosomal dominant SCA families. Furthermore, we provide detailed brain autopsy results and MRIs, confirming superior vermis atrophy in this disease. With additional functional work, we identified haploinsufficiency as the underlying disease mechanism in our families. METHODS Patients. Genealogic studies showed a clear autosomal dominant inheritance in both family 1 (F1) (figure 1A) and family 2 (F2) (figure 1C). Both families are of Belgian origin and live in the same region in Flanders. We collected clinical information and blood samples from all affected (table) and unaffected individuals. For this study, all living patients were clinically reexamined (K.S. and P.D.J.). MRI studies, muscle and brain pathology, and electrophysiologic studies were performed in the University Hospital of Antwerp.

Standard protocol approvals, registrations, and patient consents. All patients or legal representatives signed informed consent before enrollment. The local institutional review board approved this study.

Molecular genetic analyses. Testing for repeat expansions in the SCA1–3, 6–8, 10, 12, and 17 and DRPLA (dentatorubralpallidoluysian atrophy) genes was performed using classic PCRbased methods. A genome scan was performed in F2 (e-Methods on the Neurology® Web site at Neurology.org). Sanger sequencing of the 17 coding exons and exon-intron boundaries of AFG3L2 was performed. To determine whether a copy number variation within AFG3L2 could be the cause of SCA28, we also screened

Figure 1

F2 with the multiplex amplicon quantification (MAQ) technique (http://www.multiplicom.com/). This technique consists of a multiplex PCR amplification of fluorescently labeled target and control amplicons followed by fragment analysis on an ABI 3730 DNA analyzer (Applied Biosystems, Foster City, CA). The assay consisted of 8 target amplicons located in AFG3L2, and 5 control amplicons located at randomly selected genomic positions outside the AFG3L2 region and other known copy number variations. These 13 amplicons were PCR-amplified in a single reaction containing 20 ng of genomic DNA. Peak areas of the target amplicons were normalized to the control amplicons. Comparison of normalized peak areas between patients and controls resulted in a dosage quotient for each target amplicon, calculated by the MAQ software package. Dosage quotient values below 0.75 are indicative of a deletion. All deletion variants were confirmed by an independent MAQ experiment of the original or newly obtained DNA sample. A quantitative PCR with SYBR green incorporation was performed on genomic DNA with 2 amplicons residing in exons 14 to 16 of AFG3L2. The average cycle threshold value obtained was normalized against RNPP30 (ribonuclease P/MRP 30-kDa subunit).

MRI analysis. MRI data were acquired from 6 patients and 6 control subjects, matched for age and sex. All MRI examinations were performed on a 3-tesla platform scanner. For volumetric measurements, a high-resolution anatomical T1-weighted magnetic resonance dataset was obtained, using a magnetization-prepared rapid-acquisition gradient echo sequence with 176 slices and a field of view of 256 3 192 mm, and a resolution of 1 3 1 3 1 mm3. Repetition time and echo time were 1,910 and 3.37 milliseconds, respectively; the flip angle was 15°. Diffusion tensor imaging (DTI) datasets were acquired by obtaining diffusion-weighted images in 64 directions with a b value of 1,000 s/mm2. The sequence contained 60 slices with 2-mm slice thickness, field of view of 256 3 256 mm, and voxel size of 2 3 2 3 2 mm3. Repetition time and echo time were

Pedigrees of families 1 and 2, and photograph of patient with SCA28

(A) Pedigree of family 1 and (C) pedigree of family 2. Filled symbols indicate affected subjects. Open symbols indicate unaffected spouses and presently asymptomatic family members. Gray symbols indicate possibly affected individuals. A diamond is shown to anonymize sex. Deceased subjects are marked by a diagonal line. Asterisk indicates that DNA was available. (B) Photograph of SCA28 patient with severe ptosis. SCA28 5 spinocerebellar ataxia type 28. Neurology 82

June 10, 2014

2093

2094

Table

Clinical features and investigations of Belgian individuals with AFG3L2 partial deletion Family, generation, and patient no. F1 I.1

F2 II.2

II.3

III.1

I.2

II.1

II.3

II.6

III.1

III.2

III.4

III.5

III.7

III.9

III.11

III.13

IV.1

IV.3

IV.5

Neurology 82 June 10, 2014

Onset age, y, sex

30, F

54, M

35, M

36, M

50, F

70, M

48, F

48, F

43, M

49, M

20, F

35, M

33, F

30, F

28, F

30, M

33, F

26, F

28, F

Examination age, y

74

55

47

49

/

88

60

60

62

68

28

54

59

58

57

52

33

26

28

X: current age, y

X: 75

X: 75

X: 52

49

X

89

X

X

63

69

X

54

59

58

57

52

33

26

28

Disease duration, y

45

21

17

13

?

19

?

?

20

20

?

19

26

28

29

22

,1

,1

,1

Gait ataxia

11

1

1

1

1

1

1

1

1

1

1

1

1

11

1

11

1

2

2

Dysarthria

11

11

1

1

/

1

1

1

1

11

1

1

1

11

11

11

2

2

2

Ophthalmoparesis

1

1

1

1

/

2

1

1

2

1

1

1

2

1

2

11

2

2

2

Hypometric saccades

1

1

1

1

/

1

1

1

1

1

1

1

1

1

1

11

2

1

1

Ptosis

11

11

11

1

1

2

1

11

11

11

1

2

2

11

2

11

2

2

2

Nystagmus

11

1

1

1

/

1

1

1

1

1

1

2

2

11

1

11

2

1

2

Increased reflexes in LL

1

2

2

1

/

2

1

11

2

1

11

1

2

2

2

11

2

2

2

Babinski sign

2

2

2

2

/

2

1

2

2

1

1

2

2

2

2

2

2

2

2

Spasticity

2

2

2

2

/

2

2

1

2

1

1

2

2

2

2

2

2

2

2

Dementia

2

2

2

2

/

2

2

2

2

2

2

2

2

2

2

2

2

2

2

Epilepsy

2

2

2

1

/

2

2

2

2

2

1

2

2

2

2

2

2

2

2

Muscle weakness

2

LL

LL

2

/

2

2

2

2

1

1

2

2

2

2

2

2

2

2

Muscle atrophy

2

pLL

pLL

2

/

2

2

2

2

2

2

2

2

2

2

2

2

2

2

Wheelchair

No

Yes

Yes

No

No

No

No

No

Yes

No

Yes

No

No

No

No

Yes

No

No

No

Initial MRI, age, y

VA, ?

VA, 55

/

CA, 46

/

/

/

VA, 68

CA, 47

CA, 53

CA, 22

CA, 35

CA, 34

CA, 36

VA, 36

COA, 44

/

/

/

MRI DTI 1 VBM, age, y

/

/

/

VA, 48

/

/

/

/

/

/

/

VA, 53

VA, 58

VA, 57

VA, 55

VA, 51

/

/

/

Muscle biopsy

/

/

N

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

EMG

/

N

N

/

/

/

N

/

N

N

N

N

N

N

/

N

/

/

/

Brain autopsy

COA

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

/

Abbreviations: CA 5 cerebellar atrophy; COA 5 cerebello-olivary atrophy; DTI 5 diffusion tensor imaging; F 5 family; LL 5 lower limbs; N 5 normal; pLL 5 proximal lower limbs; VA 5 vermis atrophy; VBM 5 voxelbased morphometry; X 5 died. Symbols: 1 5 present and moderate; 11 5 present and severe; 2 5 absent; / 5 not done; ? 5 unknown.

9,300 and 88 milliseconds, respectively. Voxel-based morphometry (VBM) and DTI were performed (e-Methods).

Brain autopsy, muscle biopsy, and EMG. Postmortem brain examination was performed in patient F1-I.1 (figure 1, A and B; table). Frozen sections were examined using the method of Spielmeyer for myelin, Holzer for fibrillary glia, and cresyl violet for cytology. Paraffin sections were stained with cresyl violet and hematoxylin and eosin for cytology and with the method of Bodian for axons. In addition, immunohistochemistry on 5-mm sections of the original paraffin-embedded tissue was performed including a staining with specific AFG3L2 antibodies (e-Methods).6 A biopsy of the left deltoid muscle had previously been examined in patient F1-II.3 using routine light microscopy and electron microscopy methods. An EMG examination was performed in patients F1-II.2 and F1-II.3 at the onset of their disease; the muscle atrophy was mild in both patients at that time. The musculus (M) deltoideus, M biceps, M triceps, M thenar, M first interosseus, M gluteus maximus, M quadriceps, M tibialis anterior, and M peroneus longus were examined. AFG3L2 expression studies. To test whether the putative shorter AFG3L2 messenger RNA (mRNA) escapes nonsensemediated mRNA decay resulting in the expression of a truncated protein, RNA and proteins were isolated from patients’ lymphoblasts (F1-III.1; F2-II.1, F2-II.6, and F2-III.1) and from an unaffected control (F2-II.7). For Western blot analysis, lymphoblasts were cultured, and subsequently mitochondria from those lymphoblasts were isolated for blue native–polyacrylamide gel electrophoresis (BN-PAGE) experiments (e-Methods). The AFG3L2-specific antibodies used were previously described.6

Detailed clinical findings in 19 patients are shown in the table. The core phenotype consisted of a slowly progressive cerebellar gait ataxia and dysarthria. The mean onset age was 38.7 years in F1 (range 30–54 years) and 38.1 years in F2 (range 20–70 years). Intrafamilial disease severity varied markedly. Ptosis was markedly severe in all patients in F1 while it was less common and usually less prominent in F2 in which several individuals are still young and paucisymptomatic. No optic atrophy was seen, but hypometric saccades were seen by detailed ophthalmologic examination in 17 of 18. Pyramidal tract signs were sometimes present; brisk reflexes were observed in 7 of 18, with Babinski signs in 3 of 18 and overt spasticity in 3 of 18. Patients F2-III.4 and F2-III.13 were wheelchair-bound because of the severe spasticity in addition to cerebellar ataxia. In 2 brothers (F1-II.2 and F1-II.3), atrophy of the gluteus and quadriceps muscles was present with a residual strength of 4/5 on the Medical Research Council Scale. Both brothers displayed Trendelenburg gait and were wheelchair-bound in the end stage of their disease. Two patients developed epilepsy: individual F1-III.1 experienced tonic-clonic seizures and had generalized sharp waves on EEG, and individual F2-III.4 had partial epilepsy with right-sided motor seizures; the

RESULTS Clinical features.

EEG showed sharp waves in the left hemisphere. Both were treated with valproate. All of our patients had normal development in childhood and adulthood, and no signs of cognitive decline were noticed. Molecular genetic analyses. Testing for repeat expansions causing SCA1–3, 6–8, 10, 12, 17, and DRPLA was negative in individual F2-II.6. Linkage analysis showed suggestive linkage to the centrosomal region of chromosome 18 harboring the SCA28 locus (highest LOD score 5 2.8 for marker D18S453). There were no alternative regions of linkage; all LOD scores were ,1.5. Sanger sequencing of the 17 coding exons and exon-intron boundaries of AFG3L2 revealed no mutation. Subsequently, MAQ detected a deletion of a region of 8 to 18 kb, corresponding to a heterozygous deletion of exons 14 to 16 of AFG3L2 (figure 2). This deletion was absent from 91 healthy Belgian controls. A quantitative PCR revealed a decrease of approximately 50% for both amplicons residing in exons 14 to 16 of AFG3L2 in 3 patients compared with 2 unrelated controls. Haplotype reconstruction showed that both families share a common haplotype of 9.9 to 10.6 Mb, suggesting that the 2 SCA28 families have a common ancestor. All deletion carriers were also screened for SPG7, and no mutations were found. MRI. In 6 patients, MRI DTI and VBM were per-

formed (table). VBM analysis showed significant cerebellar gray matter atrophy in the patients with SCA compared with healthy control subjects (figure 3). In addition, DTI analysis revealed fractional anisotropy decrease and mean diffusivity increase in the patient population in the cerebellar peduncles, corticospinal tracts, corpus callosum, and cingulum (figure e-1). Brain pathology. Histologic analysis of the brain

showed cerebellar atrophy with diffuse loss of Purkinje cells in both cerebellar hemispheres predominating in their dorsal part and in the culmen of the vermis, associated with an increase of Bergmann glial cells and of glial cells in the molecular layer (figure 4, e-Results, figure e-2). A decrease of the amount of neurons was noted in the dorsal part of the nucleus olivaris inferior principalis (figure 4, B and C). Immunohistochemistry with antibodies against phosphorylated tau (AT8) and Ab1 amyloid (4G8) showed 5 to 10 neurofibrillary tangles/mm2 in the gyrus parahippocampalis and sparse to moderate amounts of senile plaques in the iso- and allocortex. Calbindin showed a marked decreased number of Purkinje cells. The dendritic tree of remaining Purkinje cells was hypotrophic (figure 4, D and E). Staining with an antibody against ubiquitin showed a few immunoreactive neuronal intranuclear inclusions in the cerebral and cerebellar cortex. Inclusions in glial nuclei were Neurology 82

June 10, 2014

2095

Figure 2

MAQ analysis

(A) Electropherogram of the MAQ analysis for a patient carrying the partial AFG3L2 deletion. x-Axis: length in basepairs (bp); y-axis: signal strength in fluorescence units (FU). Blue peaks indicate areas of amplification of the patient sample, and gray peaks are from the reference sample. Yellow bins correspond to the target amplicons in the AFG3L2 region; gray bins refer to control amplicons located at randomly selected genomic positions outside the AFG3L2 region and other known copy number variations. Arrows indicate peak areas where patient samples show half the amplification compared with the reference samples. (B) Dosage plot of the MAQ analysis for a patient carrying the partial AFG3L2 deletion. Dots in the gray region (dosage quotient between 0.8 and 1.2) correspond to copy number 2. The 2 dots at dosage quotient 5 0.5 correspond to copy number 1 or the heterozygous partial deletion of AFG3L2. MAQ 5 multiplex amplicon quantification.

much more visible with the p62 antibody and were also found in the hippocampus, the pontine, and medullary nuclei (figure 4, F–I). Antibody 1C2 against expanded polyglutamine tracts was negative. Staining with TDP-43 (TAR DNA-binding protein 43) and FUS (fused in sarcoma) antibodies showed normal immunoreactivity. The AFG3L2 antibody was raised against residues 67 to 305 of the protein so it did not target the exons 14 to 16 deletion.6 Because this antibody was previously only used in confocal immunofluorescence and in situ hybridization analysis, we first performed several stainings using different 2096

Neurology 82

June 10, 2014

dilutions on cerebellum of a control patient. No immunoreactivity could be elicited in patient and control. Muscle pathology. A muscle biopsy of the deltoid muscle showed no ragged-red fibers on routine light microscopy, and no other specific abnormalities were found. Electron microscopy revealed no mitochondrial abnormalities (data not shown). EMG, performed at the onset of the disease, was normal. AFG3L2 expression at the RNA and protein level. RNA and Western blot analysis showed the presence of a

Figure 3

MRI, voxel-based morphometry

Axial, coronal, and sagittal slices are shown of the increased gray matter densities in the control group compared with the patient group. Results are shown for an uncorrected statistical threshold of p , 0.001, for a minimum cluster size of 5 voxels. Mean age of patients 5 53.6 years and mean age of controls 5 56.0 years; mean disease duration of patients 5 22.8 years.

shorter AFG3L2 mRNA and the corresponding truncated AFG3L2 protein, suggesting that this partial deletion escapes nonsense-mediated decay. BNPAGE demonstrated reduced levels of AFG3L2 complexes in patients, indicating that this truncated protein does not assemble in functional m-AAA proteolytic complexes. This confirmed that the disease in our patients is due to haploinsufficiency (e-Results, figure e-3). DISCUSSION In this report, we describe 2 SCA28 families with a novel mutation, namely, a partial AFG3L2 deletion. The core phenotype is similar to that in previous reports and consists of gait ataxia,

dysarthria, ophthalmoparesis, and/or ptosis. SCA28 was initially described as a slowly progressive disorder and with onset age as early as 12 years with a mean of 19.5 years.7 In 9 other European SCA28 families,4 mean onset age was 30.7 years (range 6–60 years). Our families confirm the later onset with a mean of 38.2 years (range 20–70 years). Brisk reflexes and spasticity can be part of the phenotype, justifying its classification among the spastic ataxias. In a 4-generation Italian family,7 brisk reflexes were found in 90% of patients, but in our families, pyramidal features are less prominent (38%). Febrile seizures have been previously described in 2 patients with SCA28,5 while a severe generalized myoclonic epilepsy syndrome was observed Neurology 82

June 10, 2014

2097

Figure 4

Microscopy and immunohistochemistry of brain autopsy

(A) SCA28 patient; cerebellar cortex, cresyl violet. (B) SCA28 patient; nucleus olivaris inferior principalis (dorsal part), cresyl violet. (C) SCA28 patient; nucleus olivaris inferior principalis (ventral part), cresyl violet. (D) Control patient; cerebellar cortex, calbindin. (E) SCA28 patient; cerebellar cortex, calbindin. (F) SCA28 patient; neocortex, p62. (G) SCA28 patient; granular layer, molecular layer, and Bergmann glia in the cerebellar cortex. (H) SCA28 patient; nucleus vestibularis medianus, p62. (I) SCA28 patient; nucleus olivaris inferior principalis (dorsal part), p62. There is a severe loss of Purkinje cells in the cerebellar cortex and an astrocytic gliosis of the molecular layer (A). Also note the marked neuronal loss in the dorsal part of the nucleus olivaris inferior principalis (B) compared with the ventral part (C). Calbindin antibody elicits the dendritic tree in the control patient (D), whereas the perikaryon and dendrites of the Purkinje cells are hypotrophic and nearly nonexistent in the SCA28 patient (E). Neurons of the neocortex show NII immunoreactive for p62 antibody (arrow) (F). NII are also present in Bergmann glia (arrow). There is again a severe loss of Purkinje cells (G). In the nucleus vestibularis medianus, the same NII are found (arrow) (H). Many NII are also present in the dorsal part of the nucleus olivaris inferior principalis (arrow) (I). NII 5 neuronal intranuclear inclusions; SCA28 5 spinocerebellar ataxia type 28.

2098

Neurology 82

June 10, 2014

in 2 patients with SPAX5.11 In our study, patient F1III.1 had generalized epilepsy and patient F2-III.4 had partial seizures. Patients F1-II.2 and F1-II.3 had a progressive proximal muscle atrophy of the legs. Proximal muscle weakness in the lower limbs has not been described before in SCA28, only in the 2 SPAX5 patients with lower extremity weakness and distal amyotrophy with electrophysiologic evidence of axonal sensorimotor neuropathy.11 Muscle pathology in one patient did not show specific myopathic or neuropathic changes; however, this biopsy was taken in a clinically unaffected deltoid muscle. MRI reports in SCA28 described superior vermis atrophy in the cerebellum without brainstem involvement.3 DTI is a new imaging technique and has been used to study SCA1 and SCA2.12 In our 6 patients with SCA28 who underwent MRI DTI and VBM, marked vermis atrophy was observed, confirming previous MRI observations in patients with SCA28.13 Limited degeneration of the central pyramidal motor system was observed consistent with the mild pyramidal tract signs and spasticity observed in our patients, unlike previously described patients with SCA28. To further explore the involvement of the pyramidal motor system, the corticospinal tracts were reconstructed with DTI. However, no significant differences of fractional anisotropy or mean diffusivity of the corticospinal tract were found between patients and controls. Given the small sample size, no significant VBM or DTI differences were observed after correction for multiple comparisons. Brain pathology in patient F1-I.1 revealed a cerebello-olivary atrophy predominating in the dorsal part of the vermis and cerebellar hemispheres and the dorsal part of the inferior olivary nuclei. Neuronal and glial intranuclear inclusions were found in the cerebellar cortex, dentate gyrus of the hippocampus, pontine and medullary nuclei, and in the cerebral cortex. The relevance of these lesions is unknown and needs confirmation in additional SCA28 autopsies. No immunoreactivity with the AFG3L2 antibody could be obtained in brain tissue of both patient and control. Because the AFG3L2 antibody did show the truncated AFG3L2 protein in Western blot, we assume that the lack of immunoreactivity is due to technical problems, so no additional conclusion can be drawn regarding the pathomechanism. A partial deletion of exons 14 to 16 of AFG3L2 causing SCA28 has not been previously described. Before this study, 14 missense mutations in AFG3L2 and one frameshift mutation were reported.4–6,8,9,11 AFG3L2 has an AAA domain and a proteolytic domain. Our partial deletion targets the proteolytic domain of the protein, more precisely the C-terminal peptidase M41 domain, containing the central pore,

surrounding the exit of the proteolytic chamber. The proteolytic domain is a crucial and highly conserved region in m-AAA and m-AAA–related proteins. Almost all pathogenic missense mutations are located in this domain, except for the N432T mutation in the AAA domain.6 Previous reports stated that deletions are probably not disease-causing in SCA28 and that rearrangements should not be considered in SCA28.4 In the SCA28 families described here, mRNA and Western blot analysis detected a shorter mRNA fragment and a corresponding truncated AFG3L2 protein. BNPAGE experiments in patient lymphoblasts showed reduced levels of assembled AFG3L2 in mitochondria. Both AFG3L2 and paraplegin (encoded by SPG7) form the hetero-oligomeric m-AAA protease in the inner mitochondrial membrane. AFG3L2 can form homooligomeric AFG3L2/AFG3L2 complexes as well, whereas paraplegin is not able to form homooligomeric complexes.14 The relative abundance of mAAA protease subunits can vary strongly among tissues, potentially explaining the specific phenotypic consequences of mutations in m-AAA protease subunits.14 It is remarkable that patients carrying SPG7 mutations share similar clinical features with SCA28 patients, showing a spastic paraplegia often complicated by ataxia,15,16 although no optic atrophy was seen in our SCA28 families, but it is common in SPG7 families. It is known that the cerebellum is the neuronal tissue most vulnerable to reduced levels of AFG3L2.17 The fact that truncated AFG3L2 does not assemble in functional m-AAA proteolytic complexes argues against a dominant-negative effect of the mutant protein, and strongly suggests haploinsufficiency as the disease-causing mechanism in the 2 Belgian SCA28 families. Our result is consistent with previous studies on the yeast m-AAA protease, which have identified the proteolytic domain as a key determinant for assembly of the m-AAA protease subunits.18 Of note, m-AAA protease complexes of different sizes were detected in BN-PAGE experiments. This confirms previous results in yeast, showing that upon digitonin solubilization, the m-AAA protease interacts with the prohibitin complex.14 Whether other proteins are part of this complex still awaits investigation. Previous data postulated haploinsufficiency as an unlikely cause of SCA28 because of observations in 18p deletion syndromes. A mother and son with a deletion of the complete short arm of chr18 supposedly encompassing AFG3L2 had dystonia.19 The age of those patients was not mentioned, so they may still be at risk of developing ataxia. Likewise, a Korean male with a partial deletion of 2.7 kb of exons 10 and 11 in AFG3L2 did not have ataxia at the age of 30 years, but again, may still develop SCA28. This partial deletion is also not located in the proteolytic domain of AFG3L2.20

In contrast, studies of mutant mice have provided evidence for haploinsufficiency as the pathomechanism of SCA28. Two mutant mice, the AFG3L2 null mice (AFG3L2emv66/emv66) and the AFG3L2par/par, presented with the same severe neuromuscular phenotype.21 The heterozygous AFG3L2par/1 mouse had normal appearance and fertility,21 but late-onset degeneration was not studied in detail, whereas the AFG3L2emv66/1 mouse, carrying the heterozygous lossof-function mutation, was further investigated in detail and developed a phenotype with similarities to SCA28, namely, severe motor incoordination, mitochondrial dysfunction, and Purkinje cell degeneration.22 In this last model, haploinsufficiency was the likely mechanism, because the mutant allele did not give rise to protein translation. Furthermore, yeast coexpression studies revealed that 5 human AFG3L2 missense mutations fall in 2 classes, a group of dominant-negative mutants and another with haploinsufficiency or a weak dominant-negative effect.6 A recently identified frameshift mutation in AFG3L2 further strengthens haploinsufficiency as a disease mechanism.9 Our study of 2 SCA28 deletion families broadens the genetic and clinical spectrum of m-AAA–related diseases. Brain autopsy and MRI DTI-VBM imaging confirm vermis atrophy in patients with SCA28. Rearrangements in the SCA28 locus should be considered in patients and families presenting with dominantly inherited ataxia and ptosis. Appropriate techniques should be used because partial deletions are overlooked by Sanger sequencing. Finally, at least for this particular partial deletion, haploinsufficiency is the underlying disease mechanism. AUTHOR CONTRIBUTIONS K. Smets: acquisition of data, contributing material of participants, study supervision, study concept and design, critical revision of the manuscript for important intellectual content. T. Deconinck: technical assistance, analysis and interpretation of the data. J. Baets: study supervision, study concept and design, critical revision of the manuscript for important intellectual content. J.-J. Martin and A. Sieben: acquisition of data, contributing material of participants, critical revision of the manuscript for important intellectual content. I. Smouts: acquisition of data, contributing material of participants, technical assistance. S. Wang: technical assistance, analysis and interpretation of the data. F. Taroni and D. Di Bella: acquisition of data, technical assistance. W. Van Hecke: acquisition of data, analysis and interpretation of the data, critical revision of the manuscript for important intellectual content. P.M. Parizel: acquisition of data, analysis and interpretation of the data. C. Jadoul, R. De Potter, and F. Couvreur: contributing material of patients, acquisition of data. E. Rugarli: acquisition of data, technical assistance, critical revision of the manuscript for important intellectual content. P. De Jonghe: acquisition of data, contributing material of participants, study supervision, study concept and design, critical revision of the manuscript for important intellectual content.

ACKNOWLEDGMENT The authors are grateful to the family members for their participation. The authors thank Inge Bats, Karen Sterckx, and Lieve Vits for technical assistance, the VIB Genetic Core Facility for sequencing and cell-maintenance support, and Multiplicom for providing the MAQ technology. Neurology 82

June 10, 2014

2099

10.

STUDY FUNDING Professor Franco Taroni expresses thanks for the Telethon-Italian grant GGP09301.

DISCLOSURE K. Smets, T. Deconinck, J. Baets, A. Sieben, J. Martin, I. Smouts, S. Wang, F. Taroni, and D. Di Bella report no disclosures relevant to the manuscript. W. Van Hecke reports to be the cofounder of and working with icoMetrix, a spinoff company of the Universities of Leuven and Antwerp (Belgium), and the Antwerp University Hospital (Belgium). P. Parizel reports, C. Jadoul, R. De Potter, F. Couvreur, E. Rugarli, and P. De Jonghe report no disclosures relevant to the manuscript. Go to Neurology.org for full disclosures.

Received September 3, 2013. Accepted in final form March 5, 2014. REFERENCES 1. Durr A. Autosomal dominant cerebellar ataxias: polyglutamine expansions and beyond. Lancet Neurol 2010;9:885–894. 2. Klockgether T. Sporadic ataxia with adult onset: classification and diagnostic criteria. Lancet Neurol 2010;9:94–104. 3. Mariotti C, Brusco A, Di Bella D, et al. Spinocerebellar ataxia type 28: a novel autosomal dominant cerebellar ataxia characterized by slow progression and ophthalmoparesis. Cerebellum 2008;7:184–188. 4. Cagnoli C, Stevanin G, Brussino A, et al. Missense mutations in the AFG3L2 proteolytic domain account for approximately 1.5% of European autosomal dominant cerebellar ataxias. Hum Mutat 2010;31:1117–1124. 5. Edener U, Wollner J, Hehr U, et al. Early onset and slow progression of SCA28, a rare dominant ataxia in a large four-generation family with a novel AFG3L2 mutation. Eur J Hum Genet 2010;18:965–968. 6. Di Bella D, Lazzaro F, Brusco A, et al. Mutations in the mitochondrial protease gene AFG3L2 cause dominant hereditary ataxia SCA28. Nat Genet 2010;42:313–321. 7. Cagnoli C, Mariotti C, Taroni F, et al. SCA28, a novel form of autosomal dominant cerebellar ataxia on chromosome 18p11.22-q11.2. Brain 2006;129:235–242. 8. Lobbe AM, Kang JS, Hilker R, Hackstein H, Muller U, Nolte D. A novel missense mutation in AFG3L2 associated with late onset and slow progression of spinocerebellar ataxia type 28. J Mol Neurosci 2014;52:493–496. 9. Musova Z, Kaiserova M, Kriegova E, et al. A novel frameshift mutation in the AFG3L2 gene in a patient with spinocerebellar ataxia. Cerebellum Epub 2013 Nov 23.

11.

12.

13. 14.

15.

16.

17.

18.

19.

20.

21.

22.

Casari G, De Fusco M, Ciarmatori S, et al. Spastic paraplegia and OXPHOS impairment caused by mutations in paraplegin, a nuclear-encoded mitochondrial metalloprotease. Cell 1998;93:973–983. Pierson TM, Adams D, Bonn F, et al. Whole-exome sequencing identifies homozygous AFG3L2 mutations in a spastic ataxia-neuropathy syndrome linked to mitochondrial m-AAA proteases. PLoS Genet 2011;7:e1002325. Mandelli ML, De Simone T, Minati L, et al. Diffusion tensor imaging of spinocerebellar ataxias types 1 and 2. AJNR Am J Neuroradiol 2007;28:1996–2000. Mariotti C, Bella DD, Di Donato S, Taroni F. Spinocerebellar ataxia type 28. Handb Clin Neurol 2012;103:575–579. Koppen M, Metodiev MD, Casari G, Rugarli EI, Langer T. Variable and tissue-specific subunit composition of mitochondrial m-AAA protease complexes linked to hereditary spastic paraplegia. Mol Cell Biol 2007;27:758–767. van Gassen KL, van der Heijden CD, de Bot ST, et al. Genotype-phenotype correlations in spastic paraplegia type 7: a study in a large Dutch cohort. Brain 2012;135:2994–3004. Klebe S, Depienne C, Gerber S, et al. Spastic paraplegia gene 7 in patients with spasticity and/or optic neuropathy. Brain 2012;135:2980–2993. Martinelli P, La Mattina V, Bernacchia A, et al. Genetic interaction between the m-AAA protease isoenzymes reveals novel roles in cerebellar degeneration. Hum Mol Genet 2009;18:2001–2013. Lee S, Augustin S, Tatsuta T, Gerdes F, Langer T, Tsai FT. Electron cryomicroscopy structure of a membrane-anchored mitochondrial AAA protease. J Biol Chem 2011;286:4404– 4411. Nasir J, Frima N, Pickard B, Malloy MP, Zhan L, Grunewald R. Unbalanced whole arm translocation resulting in loss of 18p in dystonia. Mov Disord 2006;21:859–863. Kim JI, Ju YS, Park H, et al. A highly annotated wholegenome sequence of a Korean individual. Nature 2009; 460:1011–1015. Maltecca F, Aghaie A, Schroeder DG, et al. The mitochondrial protease AFG3L2 is essential for axonal development. J Neurosci 2008;28:2827–2836. Maltecca F, Magnoni R, Cerri F, Cox GA, Quattrini A, Casari G. Haploinsufficiency of AFG3L2, the gene responsible for spinocerebellar ataxia type 28, causes mitochondria-mediated Purkinje cell dark degeneration. J Neurosci 2009;29:9244–9254.

Subspecialty Alerts by E-mail! Customize your online journal experience by signing up for e-mail alerts related to your subspecialty or area of interest. Access this free service by visiting http://www.neurology.org/site/subscriptions/etoc.xhtml or click on the “E-mail Alerts” link on the home page. An extensive list of subspecialties, methods, and study design choices will be available for you to choose from—allowing you priority alerts to cutting-edge research in your field!

2100

Neurology 82

June 10, 2014

Partial deletion of AFG3L2 causing spinocerebellar ataxia type 28 Katrien Smets, Tine Deconinck, Jonathan Baets, et al. Neurology 2014;82;2092-2100 Published Online before print May 9, 2014 DOI 10.1212/WNL.0000000000000491 This information is current as of May 9, 2014 Updated Information & Services

including high resolution figures, can be found at: http://www.neurology.org/content/82/23/2092.full.html

Supplementary Material

Supplementary material can be found at: http://www.neurology.org/content/suppl/2014/05/09/WNL.000000000 0000491.DC1.html

References

This article cites 21 articles, 9 of which you can access for free at: http://www.neurology.org/content/82/23/2092.full.html##ref-list-1

Subspecialty Collections

This article, along with others on similar topics, appears in the following collection(s): All Clinical Neurology http://www.neurology.org//cgi/collection/all_clinical_neurology All Genetics http://www.neurology.org//cgi/collection/all_genetics Gait disorders/ataxia http://www.neurology.org//cgi/collection/gait_disorders_ataxia

Permissions & Licensing

Information about reproducing this article in parts (figures,tables) or in its entirety can be found online at: http://www.neurology.org/misc/about.xhtml#permissions

Reprints

Information about ordering reprints can be found online: http://www.neurology.org/misc/addir.xhtml#reprintsus

Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it is now a weekly with 48 issues per year. Copyright © 2014 American Academy of Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.

Partial deletion of AFG3L2 causing spinocerebellar ataxia type 28.

To identify the genetic cause of autosomal dominant spinocerebellar ataxia type 28 (SCA28) with ptosis in 2 Belgian families without AFG3L2 point muta...
1MB Sizes 0 Downloads 6 Views