De Novo Occurrence of Novel SPG3A/Atlastin Mutation Presenting As Cerebral Palsy

OBSERVATION De Novo Occurrence of Novel SPG3A/Atlastin Mutation Presenting as Cerebral Palsy

Shirley Rainier, PhD; Carron Sher, MD; Orit Reish, MD; Donald Thomas, BS; John K. Fink, MD

Background: Mutations in the SPG3A gene (atlastin pro- Results: We identified a novel SPG3A mutation (L157W) tein) cause approximately 10% of autosomal-dominant he- in the proband and her affected child. This mutation was reditary spastic paraplegia. For many subjects with an absent in the proband’s unaffected parents. Results of mi- SPG3A mutation, spastic gait begins in early childhood and crosatellite polymorphism analysis were consistent with does not significantly worsen even over many years. Such paternity as reported. These results indicate that this novel subjects resemble those with spastic diplegic cerebral palsy. SPG3A mutation arose de novo in the proband. To date, only 9 SPG3A mutations have been reported. Conclusions: We report the de novo occurrence of a novel SPG3A mutation in a subject with childhood- Objective: To analyze the SPG3A coding sequence in onset, nonprogressive, spastic diplegia who had no pre- an individual with childhood-onset spastic gait, who, prior vious family history of hereditary spastic paraplegia un- to the birth of her similarly affected child, had no pre- til the birth of her similarly affected son. Although rare, vious family history of hereditary spastic paraplegia. the occurrence of a de novo hereditary spastic paraplegia gene mutation must be considered in subjects with Methods: The SPG3A coding sequence was analyzed in spastic diplegic cerebral palsy for whom no other cause DNA samples from the proband, her affected child, her un- is identified. This is extremely important for correct ge- affected parents, and control subjects by polymerase-chain- netic counseling because recurrence risk may be as high reaction amplification of each exon followed by direct DNA as 50% when a mutation is detected. sequencing. Seventeen microsatellite polymorphisms were amplified and analyzed to confirm reported paternity. Arch Neurol. 2006;63:445-447

PG3A GENE (ATLASTIN PRO- infancy and her 8-year-old son (subject III:1, tein) mutations cause approxi- Figure 1A), for whom spastic paraplegia was ob- mately 10% of autosomal- served before age 1 year, presented for evalua- dominant hereditary spastic tion and genetic counseling. Each individual was paraplegia (HSP). To date, only the product of full-term, uncomplicated gesta- 9 SPG3A mutations have been reported. We tion, labor, and delivery. There was no parental S consanguinity. Neurologic examination of the report about a 34-year-old woman in whom woman and her son revealed brisk lower limb childhood-onset, nonprogressive, spastic tendon reflexes, clonus, and waddling gait with gait was considered to represent spastic normal bulbar and upper limb function, nor- diplegic cerebral palsy. The diagnosis was mal bowel and urinary control, and normal in- changed to autosomal-dominant HSP when telligence. Diagnostic evaluation results, includ- her child developed similar symptoms at 10 ing those of a computed tomographic scan of the months of age. Each subject was heterozy- brain and routine laboratory studies, were nor- gous for a novel SPG3A mutation (SPG3A mal for both the mother and child. A diagnosis cDNA nucleotide 638 T→C) that resulted of autosomal-dominant, uncomplicated, early- in the substitution of tryptophan for leu- onset HSP was made on the basis of neurologic cine at amino acid 157 (L157W). Genetic findings and family history. Author Affiliations: analysis of the proband’s parents (unaf- Control subjects were older than age 60 years and had been interviewed and exam- Department of Neurology, fected) indicated that the mutation arose de ined by neurology and psychiatry house offic- University of Michigan, novo in the proband. Although rare, the oc- Ann Arbor, (Drs Rainier and ers. All subjects were found to be free of per- Fink and Mr Thomas); currence of de novo HSP gene mutation sonal or family history of neurologic disease. Assaf Harofeh Medical Center, must be considered in subjects with spas- The DNA samples were prepared from pe- Tel Aviv, Israel (Drs Sher and tic diplegic cerebral palsy for whom no other ripheral blood samples from the proband (af- Reish); Sackler School of cause is identified. fected), her affected child, and each of her par- Medicine, Tel Aviv University, ents (unaffected subjects I:1 and I:2, Figure 1A). Tel Aviv (Dr Reish); and The SPG3A exon was polymerase-chain- Geriatric Research, Education, METHODS reaction amplified and sequenced as previ- and Clinical Center, Veterans ously described.1 Microsatellite polymor- Affairs Medical Center, A 34-year-old woman (subject II:3, Figure 1A) phisms (D1S199, D1S207, D1S230, D1S238, Ann Arbor (Dr Fink). with nonprogressive spastic paraplegia dating to D1S2697, D1S2868, D1S413, D1S425,

(REPRINTED) ARCH NEUROL / VOL 63, MAR 2006 WWW.ARCHNEUROL.COM 445

A A C T T T G

III

A A C T T T G A A C T T N G

Normal Control Subject HSP Subject

Figure 1. A, Hereditary spastic paraplegia (HSP) kindred with novel SPG3A mutation (L157W). Arrow indicates proband. B, SPG3A sequence results. cDNA indicates complementary DNA; nt, nucleotide.

A161P F151S R239C H247P I315S cDNA 1688insA L157W A217Q H258R S259Y

Exon 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Figure 2. SPG3A mutations in hereditary spastic paraplegia. The SPG3A coding sequence is divided into 14 exons (black boxes). The location of hereditary spastic paraplegia–specific SPG3A mutations is shown.

D19S865, D19S1034, D19S216, D19S894, D19S120, D19S247, sis of emerging genotype-phenotype correlations may im- D19S922, D19S406, and D14S976) were amplified and ana- prove the ability to provide counseling to patients with 2 lyzed as previously described. SPG3A mutations. The very early onset of apparently nonprogressive spastic paraplegia in the present kin- RESULTS dred is very similar to that observed in affected subjects with 247His→Pro,4 239Cys→Arg,1 and 258His→Arg,1 Sequence analysis revealed that the proband (affected) and somewhat younger than in subjects with either her affected child were each heterozygous for a missense 259Ser→Tyr (onset usually less than 5 years) (written mutation at SPG3A complementary DNA (cDNA) posi- communication, T. Heiman Patterson, November 2001) tion 638 (Figure 1B). This occurs in SPG3A exon 4 and sub- or 217Arg→Gln (average, 8.3 years),5 and significantly stitutes tryptophan for leucine at amino acid 157 (L157W). younger than in subjects with the frameshift mutation The remainder of the SPG3A coding sequence, including (1688insA) that causes premature translation termina- intron-exon boundaries, was analyzed and found to be nor- tion at residue 522.5 Subjects with a 1688insA mutation mal. Sequencing SPG3A exon 4 in samples from each of reported symptoms beginning on average at age 18 years the proband’s parents and 100 control subjects did not re- (range, 5-39 years).6 It is possible that relatively delayed veal any mutation. Analysis of genotypes from 17 micro- onset of symptoms in subjects with a 1688insA muta- satellite polymorphisms (data not shown) were com- tion is due to some degree of residual atlastin function pletely consistent with paternity as reported. present in the protein (which is terminated at residue 522 instead of 559). Both childhood-onset HSP and late- COMMENT onset HSP (after age 40 years) occurred with the re- cently reported 161Ala→Pro mutation.4 This wide range SP3A mutations cause approximately 10% of domi- of ages at symptom onset for unrelated subjects with the nantly inherited, uncomplicated HSP. Among domi- same SPG3A mutation indicates the influence of other nantly inherited HSP kindreds with early childhood on- modifying factors (presumably modifying genes) in de- set, however, approximately 25% had SPG3A mutations.3 termining the phenotype. Clinical information was not Ten SPG3A mutations have been reported (including reported for the other known SPG3A mutations the novel mutation described herein) (Figure 2). Analy- (151Phe→Ser and 315Ile→Ser).7

(REPRINTED) ARCH NEUROL / VOL 63, MAR 2006 WWW.ARCHNEUROL.COM 446

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 The presence of the mutation in the proband but Author Contributions: Study concept and design: Rainier, not her unaffected parents raised the possibility of Reish, Sher, and Fink. Acquisition of data: Rainier, Reish, either nonpaternity or de novo mutation. Analysis of Sher, and Thomas. Analysis and interpretation of data: genotypes from 17 informative microsatellite polymor- Rainier, Reish, and Fink. Drafting of the manuscript: Rainier phisms (data not shown) were completely consistent and Fink. Critical revision of the manuscript for impor- with paternity as reported. This supports the occur- tant intellectual content: Rainier, Sher, Reish, Thomas, and rence of the proband’s novel SPG3A mutation as a de Fink. Obtained funding: Fink. Administrative, technical, and novo or spontaneous event. material support: Rainier, Reish, and Fink. Study super- The function of SPG3A’s encoded protein atlastin is vision: Rainier, Sher, Reish, and Fink. unknown. As noted previously, atlastin contains con- Funding/Support: This research was supported by grants served GTPase motifs and shows homology to human gua- from the Veterans Affairs Merit Review (J.K.F.) and grants nyl binding protein 1 (hGBP1).1 The novel SPG3A mu- NINDS R01NS33645, R01NS36177, and R01NS38713 from tation we identified (T638C) disrupts a putative the National Institutes of Health, Bethesda, Md (J.K.F.). phosphorylation site, indicating the functional impor- Acknowledgement: We gratefully acknowledge the ex- tance of this motif. The emerging “morbid map” (Figure 2) pert secretarial assistance of Lynette Girbach and the par- of SPG3A mutations may provide insight into important ticipation of subjects with hereditary spastic paraplegia and functional domains, which in turn gives clues to atlas- their family members, without whom our investigations tin’s function and the molecular mechanisms that un- of hereditary spastic paraplegia would not be possible. derlie HSP. The early-onset and relatively nonprogressive nature REFERENCES of lower extremity spasticity in this family resembled that of spastic diplegic cerebral palsy. The proportion of 1. Zhao X, Alvarado D, Rainier S, et al. Mutations in a novel GTPase cause autoso- subjects diagnosed with spastic diplegic cerebral palsy mal dominant hereditary spastic paraplegia. Nat Genet. 2001;29:326-331. whose disorder is due to de novo SPG3A mutation sub- 2. Hedera P, Rainier S, Alvarado D, et al. Novel locus for autosomal dominant he- jects is unknown. Although rare, the occurrence of de reditary spastic paraplegia on chromosome 8q. Am J Hum Genet. 1999;64:563- 569. novo HSP gene mutation must be considered in sub- 3. Alvarado DM, Ming L, Hedera P, et al. Atlastin gene analysis in early onset he- jects with spastic diplegic cerebral palsy for whom no reditary spastic paraplegia. Am J Hum Genet. 2001;69:597. other cause is identified. This is extremely important 4. Sauter SM, Engel W, Neumann LM, Kunze J, Neesen J. Novel mutations in the for correct genetic counseling because recurrence risk Atlastin gene (SPG3A) in families with autosomal dominant hereditary spastic may be as high as 50% when a mutation is detected, and paraplegia and evidence for late onset forms of HSP linked to the SPG3A locus. Hum Mutat. 2004;23:98. families may be interested in using this information for 5. Muglia M, Magariello A, Nicoletti G, et al. Further evidence that SPG3A gene mu- prenatal genetic testing. tations cause autosomal dominant hereditary spastic paraplegia. Ann Neurol. 2002; 51:794-795. Accepted for Publication: May 5, 2005. 6. Tessa A, Casali C, Damiano M, et al. SPG3A: an additional family carrying a new atlastin mutation. Neurology. 2002;59:2002-2005. Correspondence: John K. Fink, MD, 5214 CCGCB, 1500 7. Abel A, Hofer A, Del Turco D, Klimpe S, Deller T, Auburger G. The SPG3 disease E Medical Center Dr, Ann Arbor, MI 48109-0940 (jkfink gene atlastin: new mutations and the characterization of its murine orthologue. @umich.edu). Am J Hum Genet. 2003;73:547.

(REPRINTED) ARCH NEUROL / VOL 63, MAR 2006 WWW.ARCHNEUROL.COM 447