Spastic Paraplegia 3, Autosomal Dominant

A number sign (#) is used with this entry because autosomal dominant spastic paraplegia-3A (SPG3A) is caused by heterozygous mutation in the ATL1 gene (606439) on chromosome 14q22.

See also autosomal dominant hereditary sensory neuropathy type 1D (HSN1D; 613708), an allelic disorder with a different phenotype.

Description

The hereditary spastic paraplegias are a group of clinically and genetically diverse disorders characterized by progressive, usually severe, lower extremity spasticity; see reviews of Fink et al. (1996) and Fink (1997).

SPG is classified according to both the mode of inheritance (autosomal dominant, autosomal recessive (see 270800), and X-linked (see 303350)) and whether progressive spasticity occurs in isolation ('uncomplicated SPG') or with other neurologic abnormalities ('complicated SPG'), including optic neuropathy, retinopathy, extrapyramidal disturbance, dementia, ataxia, ichthyosis, mental retardation, and deafness. The major neuropathologic feature of autosomal dominant, uncomplicated SPG is axonal degeneration that is maximal in the terminal portions of the longest descending and ascending tracts (crossed and uncrossed corticospinal tracts to the legs and fasciculus gracilis, respectively). Spinocerebellar fibers are involved to a lesser extent. Since the description of 'pure' hereditary spastic paraparesis of late onset by Strumpell (1904), many 'complicated' forms of the disorder have been reported and the question as to whether a 'pure' form exists has been raised off and on. Probably in large part because of their exceptional length, the pyramidal tracts are unusually vulnerable to both acquired and genetic derangement. Although a majority of reported families have displayed recessive inheritance, 10 to 30% of families have a dominant pattern and in fact recessive inheritance of a 'pure' spastic paraplegia may be rare.

Genetic Heterogeneity of Autosomal Dominant Spastic Paraplegia

In addition to SPG3A, other forms of autosomal dominant spastic paraplegia for which the molecular basis is known include SPG4 (182601), caused by mutation in the SPAST gene (604277) on 2p22; SPG6 (600363), caused by mutation in the NIPA1 gene (608145) on 15q11; SPG8 (603563), caused by mutation in the KIAA0196 gene (610657) on 8q24; SPG9A (601162), caused by mutation in the ALDH18A1 gene (138250) on 10q24; SPG10 (604187), caused by mutation in the KIF5A gene (602821) on 12q13; SPG12 (604805), caused by mutation in the RTN2 gene (603183) on 19q13; SPG13 (605280), caused by mutation in the SSPD1 gene (118190) on 2q33.1; SPG31 (610250), caused by mutation in the REEP1 gene (609139) on 2p11; SPG33 (610244), caused by mutation in the ZFYVE27 gene (610243) on 10q24; SPG72 (615625), caused by mutation in the REEP2 gene (609347) on 5q31; SPG73 (616282), caused by mutation in the CPT1C gene (608846) on 19q13; and SPG80 (618418), caused by mutation in the UBAP1 gene (609787) on chromosome 9p13.

Autosomal dominant spastic paraplegia has been mapped to chromosomes 9q (SPG19; 607152), 1p31-p21 (SPG29; 609727), 12q23-q24 (SPG36; 613096), 8p21.1-q13.3 (SPG37; 611945), 4p16-p15 (SPG38; 612335), and 11p14.1-p11.2 (SPG41; 613364).

Clinical Features

In the Amish of Lancaster County, Pa., a kindred with spastic paraplegia in 3 generations was observed (McKusick, 1965). In this closed community the origin of the de novo mutation could be identified with considerable certainty. The disease was early in onset but very slowly progressive or even static. This same type of congenital stationary familial paraplegia was described in 7 members of 2 generations by Hohmann (1957). In contrast to the early-onset, static form of disease in the Amish family, Thurmon and Walker (1971) reported a family from Deer Isle, Maine, in which many affected members had onset in the second or third decade with steady progression of neurologic defect.

Schwarz and Liu (1956) reported several families including one originally reported by Bayley (1897) which in 1956 contained 22 affected persons in 6 generations. Aagenaes (1959) described a family with 31 cases in 4 generations. Prognosis for life was good. Histopathologic changes were found bilaterally in the lateral corticospinal tracts in the thoracic cord and in the fasciculus gracilis. The confusion of the spinocerebellar degenerations is illustrated by the fact that some members of Aagenaes' family had ataxia in addition to spastic paraplegia. Behan and Maia (1974) studied 6 families. In 2 cases autopsy studies were performed. The authors concluded that distal axonal degeneration of the long ascending and descending tracts in the spinal cord is characteristic. McLeod et al. (1977) found no abnormality of motor and sensory nerve conduction in 10 persons in 3 families. In one family 4 generations were affected, in a second, 3 generations, and in a third 2 brothers were affected, possibly with the X-linked form. Sack et al. (1978) described affected members of 6 generations of a kindred. Onset was in the fourth decade or later, with symptoms of progressive gait difficulties, lower limb spasticity, and weakness. No sensory cerebellar and cranial nerve changes were associated. Anatomic changes in 1 affected person studied at autopsy were confined to the lateral corticospinal tracts and the fasciculus gracilis. Opjordsmoen and Nyberg-Hansen (1980) described a family from northern Norway with spastic paraplegia and type III syndactyly (fusion of fingers 4 and 5). The two traits were transmitted together through 3 generations and 9 affected persons. The spastic paraplegia was of unusual type: neurogenic bladder was the earliest manifestation. Indeed, the spastic paraplegia easily escaped attention. Are these two genes linked? Harding (1981) reviewed 22 families with 'pure' spastic paraplegia and found autosomal dominant inheritance in 19 and autosomal recessive in 3. She identified 2 forms on the basis of age of onset: type I with onset mainly before age 35 years; type II with onset usually after age 35 years.

Cooley et al. (1990) identified 71 affected individuals in 7 generations of a New England family; of these they examined 17 cases. Onset occurred at or before 3 years of age with involvement limited to the lower limbs. They suggested that although in the first year of life the physical examination is normal, in the second year long tract signs are evident on examination and there is a rapid increase in spasticity followed by a delay in walking. Crutches are occasionally necessary in the teens and often necessary after the age of 18. No progression of spasticity was observed after age 7. Cooley et al. (1990) reviewed the medical records of 25 family members and examined 16 of them. They felt that children free of signs by age 3 years could be assumed to be unaffected. They further suggested that early, aggressive, habilitative intervention may result in more functional ambulation for the youngest family members. No significant progression was observed after 3 years of age.

Scheltens et al. (1990) described a Dutch family with 15 affected members in 3 generations. Onset of clinical signs was in the fourth or fifth decade. The disease was mild; only a few of the affected persons became chairbound late in life. Mild sphincter disturbances were noted in 6 patients. There were no sensory changes.

Polo et al. (1993) described the genetic and clinical features of 46 patients in 9 families. Inheritance was autosomal dominant in 7, but was thought to be autosomal recessive in 2. The evidence for recessive inheritance was the occurrence in males and females in one generation with consanguineous parents (see 270800). Among the dominant kindreds, 5 corresponded to type I with onset before 35 years and 2 to type II with onset over 35 years. Irrespective of genetic type, serial evaluation demonstrated that the main symptom was slowly progressive spastic gait, extremely variable in severity, associated in some patients with decreased vibratory sense and micturition disorders (generally as late features). Among dominant families, the disease tended to be more severe in late-onset cases. No patient had symptoms in the upper limbs and plantar responses were flexor in 6 symptomatic patients.

Durr et al. (1994) studied 23 families with pure autosomal dominant spastic paraplegia and found a unimodal distribution of age of onset. The clinical manifestations of early-onset and late-onset patients were not significantly different. There was no evidence of anticipation or imprinting. Spasticity, sphincter disturbance, decreased vibratory sense, and muscle weakness increased with disease duration. Except for 1 family with electrophysiologic evidence of an axonal neuropathy, there were no clinical features by which the families could be distinguished.

Schady and Smith (1994) reported a large kindred transmitting typical 'pure' hereditary spastic paraplegia and found electrophysiologic evidence of a sensory polyneuropathy, with normal motor nerve conduction velocities. Sural nerve biopsies demonstrated severe loss of large diameter fibers and relative preservation of small myelinated and nonmyelinated fibers. Members of this family had previously been shown to have delayed central motor conduction (Schady et al., 1991). The mild sensory changes, the absence of mutilating ulcers, and the dominant mode of inheritance clearly distinguished the disorder in this family from autosomal recessive hereditary sensory neuropathy with spastic paraplegia (256840). Although the sensory changes were subclinical and therefore may have been missed in other cases of 'pure' hereditary spastic paraplegia, the authors speculated that HSP with abnormal sensory action potentials may be a distinct entity. In their own studies of 17 kinships with HSP (Schady et al., 1991; Schady and Sheard, 1990), they had found only 2 other patients with abnormal sensory action potentials.

Thurmon et al. (1999) restudied the large Deer Isle, Maine, family reported by Thurmon and Walker (1971). Analysis of age of onset was found to be consistent with anticipation in this family. The findings were considered consistent with an unstable trinucleotide repeat occurring primarily in the female germline. On reexamination they were impressed with variable spasticity and Babinski responses. Indeed, spasticity was said not to be a prominent aspect of the disorder; most affected relatives exhibited leg paralysis, with little or no spasticity. Only individuals with long duration of the disorder (more than 22 years) typically manifested a combination of paraplegic gait, hyperreflexia, and Babinski sign. One patient was thought to have been homozygous for the mutation. He was affected with spastic quadriplegia and mental retardation and died at the age of 11.5 years of pneumonia. The parents were consanguineous. The father was known to be affected at the time of report in 1971; since that time the boy's mother had become affected.

In a family with 6 members affected with an early-onset severe form of spastic paraplegia, Dalpozzo et al. (2003) identified a heterozygous mutation in the SPG3A gene (606439.0006). All affected members had onset in infancy with delayed motor milestones, gait impairment, spastic paraparesis, distal atrophy, and lower limb weakness. Because of the early onset, the first patients were misdiagnosed with cerebral palsy, and the index patient (mother of 5 affected members) was unaware that she had a genetically transmissible disease. Two patients had the unusual sign of mild hand atrophy.

Durr et al. (2004) identified mutations in the atlastin gene in 12 of 31 (39%) families in France with early-onset autosomal dominant SPG. Mean age at onset was 4.6 years (range, birth to 14 years). The overall clinical phenotype was of a pure spastic gait disorder. Scoliosis was present in 22% of patients, mild pes cavus in 15%, and brisk upper limb reflexes in 10%. Sensation was not impaired, and only 13% of patients reported decreased vibration sense in the ankles. Two patients had postural tremor in the upper limbs. One family showed incomplete penetrance.

Rainier et al. (2006) reported a mother and son with SPG3A confirmed by the finding of a mutation in the SPG3A gene (L157W; 606439.0008). Genetic analysis of family members indicated that the mutation occurred de novo in the mother. The mother was a 34-year-old woman with uncomplicated nonprogressive spastic paraplegia since infancy who was originally diagnosed with spastic diplegic cerebral palsy. She was correctly diagnosed with SPG after her son developed similar clinical symptoms at age 10 months. Both patients showed brisk lower limb reflexes, clonus, and spastic gait with normal bulbar and upper limb function, normal bowel and urinary control, and normal intelligence. Rainier et al. (2006) emphasized the importance of the correct diagnosis of SPG for genetic counseling because the recurrence risk may be as high as 50%.

Ivanova et al. (2007) identified SPG3A mutations in 12 (6.6%) of 182 European or Australian probands with spastic paraplegia. Mean age at onset in SPG3A probands was 3 years. In the 12 probands and 24 affected family members, age of onset was before 10 years of age, except in 1 family with mean onset of 14 years and notable variability (range, 8 to 28 years). In addition to typical features of SPG, 6 (17%) of 36 affected individuals had an axonal, predominantly motor peripheral polyneuropathy, confirmed by pathologic and electrophysiologic studies. The 6 patients with neuropathy originated from 5 unrelated families, and 4 of these patients had pes cavus.

Clinical Variability

Orlacchio et al. (2011) reported a 3-generation Zulu family from South Africa with an unusual form of late-onset SPG3A. The 68-year-old proband presented with progressive walking difficulties at age 56, and required a walking aid since age 66. He had mild mental retardation (IQ of 62), urinary incontinence, and thin corpus callosum without cerebellar involvement or white matter abnormalities. Inheritance was clearly autosomal dominant. Other affected family members had a similar disease course, with late onset (range, 38-51 years), spasticity restricted to the lower limbs, mental impairment, and thin corpus callosum on brain imaging. Genomewide linkage analysis followed by direct sequencing identified a heterozygous mutation in exon 12 of the ATL1 gene in affected individuals (R416C; 606439.0013).

Inheritance

SPG3A is transmitted in an autosomal dominant pattern and may show incomplete penetrance (Durr et al., 2004).

Varga et al. (2013) reported 2 unrelated families with hereditary spastic paraplegia in which family pedigree analysis suggested different inheritance patterns, but whole-exome sequencing identified pathogenic mutations in the ATL1 gene, consistent with SPG3A. Four Moroccan sibs in the first family, in which there were several consanguineous marriages, showed onset of lower limb spasticity in the first decade. Two patients also had upper limb involvement. The pedigree pattern suggested autosomal recessive inheritance of the disorder, but whole-exome sequencing identified an ATL1 mutation (R415Q; 606439.0014): 3 sibs were homozygous and 1 was heterozygous for the mutation. The heterozygous R415Q mutation was then found in 3 unaffected family members and in 2 family members who had very subtle signs of the disorder (hyperreflexia). These findings suggested complete penetrance for the mutation in homozygous state and incomplete penetrance for the mutation in heterozygous state. The second family had previously been reported by Raggio et al. (1973) as showing X-linked transmission of a pure spastic paraplegia. Whole-exome sequencing of 1 of the affected males identified a heterozygous ATL1 mutation (R415W; 606439.0007). The mutation was then identified in 3 affected and 3 unaffected members of the family, consistent with incomplete penetrance. Two of the unaffected carriers were women, and family history indicated that most unaffected women were obligate carriers. These findings were consistent with sex-associated reduced penetrance of this mutation. Varga et al. (2013) also identified heterozygosity for the R415W mutation in 1 of 83 Spanish patients with apparent sporadic HSP and in 2 of 28 Russian patients with dominant HSP. Evidence again suggested incomplete penetrance in these families. Both mutations, c.1244A-G (R415Q) and c.1243C-T (R415W), occur at a CpG nucleotide (on the plus and minus strands, respectively) and thus may represent a mutation hotspot due to spontaneous deamination of methylated cytosines. R415 is a highly conserved residue that does not localize to a known protein domain. Varga et al. (2013) suggested that female ATL1 mutation carriers may be more protected from developing disease compared to male carriers.

Possible Autosomal Recessive Inheritance

Khan et al. (2014) reported a consanguineous Pakistani family in which 6 males presented with pure SPG before 2 years of age. Clinical features included spastic gait, toe walking, hyperreflexia of the lower limbs, pes cavus, reduced vibration sense, peripheral numbness and tingling, urinary bladder hyperactivity, and scoliosis. Cognitive function was normal. Whole-exome sequencing identified a homozygous missense variant in the ATL1 gene (R118Q; 606439.0015) that was present in all affected family members. Seven family members were heterozygous for the variant, all of whom were asymptomatic except for 1 woman who had subclinically reduced vibration sensation. Functional studies of the variant were not performed. Khan et al. (2014) concluded that SPG3A in this family was transmitted in an autosomal recessive pattern, adding to the clinical complexity of the disorder.

Diagnosis

Schule et al. (2006) presented a 13-item scale designed to rate functional impairment in pure forms of spastic paraplegia. The scale measures items including walking distance, gait quality, maximum gait speed, spasticity, weakness, and pain. The scale can be performed in an outpatient setting, requires no special equipment, and was found to be a reliable and valid measure of disease severity.

Population Genetics

In a nationwide survey of Japanese patients, Hirayama et al. (1994) estimated the prevalence of all forms of spinocerebellar degeneration to be 4.53 per 100,000; of these, 3.9% were thought to have hereditary spastic paraplegia.

Mapping

Boustany et al. (1987) studied a family with 33 members affected with autosomal dominant 'pure' familial spastic paralysis. Linkage studies excluded close linkage to HLA, C8, PGM1, and P blood group. Positive lod scores were obtained with GC (139200) and Rh (111700). In 1 of 3 families, Hazan et al. (1993) found close linkage to a group of markers on 14q; maximum multipoint lod score = 10. On the other hand, the chromosome 14q candidate region was entirely excluded in the 2 other families, providing evidence of genetic heterogeneity within a clinically homogeneous form of familial spastic paraplegia.

Gispert et al. (1995) reported 3 large German pedigrees with autosomal dominant pure familial spastic paraplegia. One of the pedigrees demonstrated linkage to a 7-cM region on 14q, whereas the other 2 pedigrees were excluded from this region, confirming genetic heterogeneity of the disease.

Dube et al. (1997) undertook linkage analysis with 21 families with uncomplicated autosomal dominant hereditary spastic paraplegia, testing for linkage to the 3 known loci: SPG3 on 14q, SPG4 on 2p, and SPG4a on 15q. (The chromosome 15-linked form of hereditary spastic paraplegia, referred to as SPG4a by Dube et al. (1997), is referred to here as SPG6.) Linkage to SPG4 was found in 3 of the families and was excluded in several other families by multipoint linkage analysis. They developed a combined analytical approach which permitted conclusive linkage analysis on small-to-medium sized families, under the restrictions of genetic heterogeneity.

Among the 33 SPG kindreds reported by the hereditary spastic paraplegia working group (Fink et al., 1996), linkage to 2p was the most common, being observed in 15 kindreds (45%). Two kindreds (6%) were linked to 14q, and linkage to 15q was observed in 1 kindred (3%). Known SPG loci on these 3 chromosomes were excluded in 15 (45%) of 33 autosomal dominant SPG kindreds.

Huang et al. (1997) mapped autosomal dominant spastic paresis in a family in Northern Tibet to 14q11.2-q24.3 using microsatellite markers. Linkage to 2p24-p21 (SPG4) and 15q11.1 (SPG6), which have been found for other pedigrees with dominant spastic paresis, was excluded. In all affected individuals in the Tibetan family, signs of spastic paraplegia were confined to the legs. The younger individuals showed a positive Babinski. All managed to survive and provide adequately for their living, despite the extreme conditions in Northern Tibet, at an altitude of about 4,300 m.

Heterogeneity

Subramony et al. (2009) reported a large 6-generation Caucasian American family in which 8 living individuals had adult-onset spastic paraplegia inherited in an autosomal dominant pattern. Linkage to 14 known autosomal dominant SPG loci was excluded. The phenotype was relatively homogeneous, with onset after age 35 years (except in 1 individual) of a pure spastic paraplegia mainly affecting the lower limbs and hyperreflexia of the upper limbs. Most individuals had loss of vibratory sense in the lower limbs, and 3 had urinary urgency. Two complained of mild recall deficits at age 77 and 80 years, respectively. There was also evidence of anticipation in later generations. Subramony et al. (2009) referred to the candidate locus as SPG40.

Molecular Genetics

Zhao et al. (2001) analyzed 5 autosomal dominant hereditary spastic paraplegia kindreds showing linkage to the SPG3A locus on 14q. They identified an obligate recombinant individual, permitting a reduction of the interval containing the SPG3A locus to 2.7 cM, and screened candidate genes in this interval for disease-causing mutations (Rainier et al., 2001). Zhao et al. (2001) reported the identification of disease-specific missense mutations in a novel gene, SPG3A (ATL1; 606439), in affected individuals from these 5 SPG3A-linked kindreds. SPG3A is expressed predominantly in the central nervous system. It does not have homology to genes that cause other forms of HSP. By contrast, the peptide encoded by SPG3A, termed atlastin, shows significant homology with several GTPases, particularly guanylate-binding protein-1 (GBP1; 600411), which maps to chromosome 1 and is a member of the dynamin family of large GTPases. The hereditary spastic paraplegia in the families of the SPG3A variety is characterized by early onset (before age 10 and usually before age 5 years).

In an Italian family with HSP characterized by a mean age of onset of 8.3 years and progressive lower extremity weakness and spasticity, Muglia et al. (2002) identified a mutation in the ATL1 gene (606439.0004), which resulted in an arg217-to-gln substitution in a conserved area of GTPases.

Abel et al. (2004) identified mutations in the ATL1 gene (see, e.g., 606439.0001) in affected members of the families reported by Hazan et al. (1993) and Gispert et al. (1995).

Namekawa et al. (2006) stated that 19 mutations in the ATL1 gene had been identified in 40 different families. More than 90% of the mutations were located in exons 4 (12.5%), 7 (27.5%), 8 (17.5%), and 12 (35%). They identified mutations in the ATL1 gene in 7 (13.5%) of 52 families with autosomal dominant SPG with onset before age 20 years and 7 (31.8%) of 22 families with onset before age 10 years. Among a total of 106 mostly European families, no ATL1 mutations were identified in patients with onset after age 10 years. ATL1 mutations were twice as frequent as SPAST (604277) mutations in patients with onset before age 10 years.

Rainier et al. (2006) stated that SPG3A accounts for approximately 10% of dominantly inherited, uncomplicated SPG.

Ivanova et al. (2007) identified 12 different heterozygous ATL1 mutations in 12 (6.6%) of 182 European or Australian probands with spastic paraplegia. Seven mutations were novel, and 3 were de novo.