Spastic Paraplegia 4, Autosomal Dominant

A number sign (#) is used with this entry because autosomal dominant spastic paraplegia-4 (SPG4) is caused by heterozygous mutation in the SPAST gene (604277) on chromosome 2p22.

Description

The hereditary spastic paraplegias (SPG, HSP) are a group of clinically and genetically diverse inherited disorders characterized predominantly by progressive lower extremity spasticity and weakness. SPG is classified by mode of inheritance (autosomal dominant, autosomal recessive, and X-linked) and whether the primary symptoms occur in isolation ('uncomplicated') or with other neurologic abnormalities ('complicated').

Pure SPG4 is the most common form of autosomal dominant hereditary SPG, comprising up to 45% of cases (Svenson et al., 2001; Crippa et al., 2006).

For a general phenotypic description and a discussion of genetic heterogeneity of autosomal dominant spastic paraplegia, see SPG3A (182600).

Clinical Features

In 5 of 7 French families and in 1 large Dutch pedigree with a form of autosomal dominant familial spastic paraplegia, Hazan et al. (1994) found linkage to a locus, which they termed FSP2 (also known as SPG4), on chromosome 2p. This finding distinguished the disease from autosomal dominant spastic paraplegia-3 (182600), which had been mapped to chromosome 14. Age of onset in the 6 families showing linkage to 2p varied widely within families and the mean age at onset ranged from 20 to 39 years. Thus, age of onset may be a poor criterion for classifying autosomal dominant spastic paraplegia. Anticipation in the age of onset was observed in 2 of the kindreds.

Durr et al. (1996) reported 12 families with autosomal dominant spastic paraplegia linked to the SPG4 locus on chromosome 2. Age of onset ranged from infancy to 63 years. The clinical expression of the disorder within a family included asymptomatic patients who were unaware of their condition, mildly affected individuals who had spastic gait but were able to walk independently, and severely affected patients who were wheelchair bound. Durr et al. (1996) commented on the extensive intra- and interfamilial clinical variation.

Nielsen et al. (1998) evaluated 5 families with 2p-linked pure spastic paraplegia. In 2 families, nonprogressive 'congenital' spastic paraplegia was seen in some affected members, whereas adult-onset progressive spastic paraplegia was present in others. Low backache was reported as a late symptom by 47% of the 63 at-risk members in the 5 families. Brain and total spinal cord MRI disclosed no significant abnormalities. Nielsen et al. (1998) concluded that SPG4 is a phenotypically heterogeneous disorder, characterized by both interfamilial and intrafamilial variation.

Nance et al. (1998) found striking variation in clinical features in 4 families with spastic paraplegia with linkage to chromosome 2 markers. Only mild neurologic signs were observed in some subjects. The clinical features of 1 family had previously been described by Boustany et al. (1987). Onset was generally in the third to fifth decades with an average onset age of 35 years (range, 5 to 61 years). All clearly affected patients had scissoring gait, and in all who were examined at least 2 of the following were found: extensor plantar responses, increased knee and ankle reflexes, increased tone, muscle spasms, or leg cramps. Urinary urgency or other symptoms compatible with a neurogenic bladder, leg weakness, and decreased vibration sense were present in some but not all patients.

Byrne et al. (1997, 1998) presented a family with autosomal dominant hereditary spastic paraplegia and a specific form of cognitive impairment who showed linkage to the SPG4 locus on chromosome 2. The pattern of cognitive impairment in this family was characterized primarily by deficits in visual-spatial functions. Dysfunction manifested itself by difficulty in carrying out new tasks, forgetfulness, poor spatial perception, and impaired visual-motor coordination. By haplotype analysis the presence of the mutant gene was identified in an individual who, at the age of 57, had the same pattern of cognitive impairment but no spastic paraplegia. Furthermore, 6 individuals who presented with the disease haplotype had normal neurologic and neuropsychologic examinations. All 6 were below the maximal age of onset in the family, namely, 60 years. In this Irish family the cognitive impairment was considered to be a manifestation of the SPG4 gene mutation.

Reid et al. (1999) investigated 35 individuals from 4 families of Welsh origin, 22 of whom had 'pure' hereditary spastic paraplegia, for the presence of subclinical cognitive impairment. They found significant reductions in scores on the Mini-Mental State Examination (MMSE) among affected individuals compared to controls. To assess whether the lower MMSE scores were restricted to subjects older than 50 years, scores for affected subjects 50 years of age or younger were compared to those of controls. A significant difference in score remained. One of the families was linked to the chromosome 2 locus, while 2 others showed linkage to none of the loci known at that time. There was no significant difference between the results of these 2 groups.

McMonagle et al. (2000) compared the phenotypic expressions of autosomal dominant hereditary spastic paresis in several families with a mutation in the SPG4 gene and several families without a mutation in SPG4. In the mutation-positive group, age of onset was later, disability score was greater, progression of disease was faster, wheelchair use was greater (40.9% vs 4.8% in the mutation-excluded group), there was greater abnormal vibration sensation in the lower limbs (68.2% vs 19%), and fewer individuals were asymptomatic (18.2% vs 42.9%). Dementia was more prevalent in the mutation-positive group. McMonagle et al. (2000) emphasized the finding of cognitive impairment as a feature of SPG4 mutations.

White et al. (2000) reported a patient with familial SPG4 who had clinical dementia. Postmortem neuropathologic examination showed neuronal loss and tau- (MAPT; 157140) immunoreactive neurofibrillary tangles in the hippocampus and tau-immunoreactive balloon cells in the limbic area and neocortex. Lewy bodies were present in the substantia nigra. White et al. (2000) suggested that these findings confirmed an association of dementia with SPG4.

McMonagle et al. (2004) used several measures of cognitive function to assess 11 patients from 3 families in whom SPG4 was confirmed by genetic analysis or linkage. SPG4 patients scored significantly lower on the Cambridge cognitive examination (CAMCOG) (mean score of 73.5 compared to 91.7 in controls). After approximately 3 years, the patients' mean score fell to 64.4, whereas the mean control score declined slightly to 90.8. Deficits in the SPG4 patients were noted in attention, language expression, memory, and abstraction. Behavior assessment found that SPG4 patients exhibited agitation, aggression, apathy, irritability, depression, and disinhibition. Accounting for age, McMonagle et al. (2004) concluded that subtle changes in cognitive function in patients with SPG4 may begin after age 40 years, with more severe decline after age 60.

Orlacchio et al. (2004) reported 32 patients from 9 families from southern Scotland with SPG4. Age at onset varied from 11 to 53 years. In addition to classic features of hereditary spastic paraplegia, 2 of the 32 patients had mental retardation and 2 other patients had a thin corpus callosum and cerebellar atrophy. All affected members had the same mutation in the SPG4 gene (604277.0014), and haplotype analysis suggested a founder effect.

Orlacchio et al. (2004) reported a large Italian family in which all 16 members who had SPG4 also had congenital arachnoid cysts at the cerebellopontine angle ranging in size from 21 to 31 mm. Six patients also had mental retardation. Genetic analysis confirmed a mutation in the SPG4 gene.

McDermott et al. (2006) reported a patient with SPG4 who developed walking difficulties in his late teens with deteriorating gait in his 20s; he was wheelchair-dependent at age 35. He later developed stiffness in the upper limbs, bladder dysfunction, dysarthria, and swallowing difficulties. In his 40s, he developed respiratory insufficiency and distal muscle wasting in the lower limbs. Molecular analysis identified a mutation in the SPG4 gene (S445R; 604277.0021). The findings of bulbar and respiratory involvement, as well as lower motor neuron degeneration, broadened the phenotype associated with mutations in the SPG4 gene.

Orlacchio et al. (2008) reported a large 4-generation Italian family with SPG4 confirmed by genetic analysis. The mean ages at onset were 17.5 and 18.8 years for symptoms of the lower and upper limbs, respectively. There was a general impression of genetic anticipation spanning the 4 generations. All affected individuals had spasticity of the lower limbs and pyramidal tract signs such as hyperreflexia, extensor plantar responses, or both, and pes cavus. All patients also had weak intrinsic hand muscles, with severe amyotrophy most relevant in the thenar eminence. Peroneal muscle wasting was reported in five patients, and many used a cane. Other associated features included impaired vibration sensation and cognitive dysfunctions. All patients except 1 had temporal lobe epilepsy with partial complex seizures associated with hippocampal sclerosis.

Murphy et al. (2009) reported a family in which 12 members had SPG4 due to a deletion of exon 17 in the SPG4 gene (Beetz et al., 2007). Cognitive assessment performed over a 7-year period found that all 4 patients who were older than 60 years developed mild to moderate cognitive decline. Two younger patients aged 48 and 40, respectively, had mild cognitive impairment. Genetic analysis of this family was unusual because 4 patients with the SPG4 deletion also carried a microdeletion in the NIPA1 gene (608145), which causes SPG6 (600363); only 2 of these 4 had cognitive impairment. Five patients with only the SPG4 deletion had cognitive impairment, including 2 who did not have clinical signs of SPG. Another family member with only the NIPA1 microdeletion lacked clinical signs of SPG or cognitive impairment at age 57. Murphy et al. (2009) concluded that SPG4 is associated with cognitive decline, and that the SPG6 microdeletion does not have a clinical phenotype in this family. Postmortem examination of the proband, who had both deletions as well as SPG and cognitive impairment, showed a markedly atrophic spinal cord with degeneration of the corticospinal tracts, and superficial spongiosis and widespread ubiquitin-positive inclusions in the neocortex and white matter.

Mapping

In 5 French families and 1 large Dutch pedigree with autosomal dominant spastic paraplegia, Hazan et al. (1994) found linkage markers in the 2p24-p21 region. An analysis of recombination events and multipoint linkage placed this form of the disease within a 4-cM interval flanked by loci D2S400 and D2S367. In 4 Caucasian North American families and in 1 family from Tunisia, Hentati et al. (1994) found linkage of late-onset SPG to DNA markers on chromosome 2p in 4 of the families. Pathologic findings in a member of one of the chromosome 2-linked families had previously been reported by Sack et al. (1978).

Scott et al. (1997) examined 11 Caucasian pedigrees with autosomal dominant 'uncomplicated' familial spastic paraplegia for linkage to the previously identified loci on 2p, 14q (SPG3A), and 15q (SPG6; 600363). Chromosome 15q was excluded for all families. Five families showed evidence for linkage to 2p, 1 family to 14q, and 5 families remained indeterminate. Recombination events reduced the 2p minimum candidate region to a 3-cM interval between D2S352 and D2S367, and supported the previously reported 7-cM minimum candidate region for 14q. Age of onset was highly variable, indicating that subtypes of SPG are more appropriately defined on a genetic basis than by age of onset. Comparison of age of onset in parent-child pairs was suggestive of anticipation, with a median difference of 9.0 years (p less than 0.0001).

Molecular Genetics

Using the repeat expansion detection (RED) method, Nielsen et al. (1997) analyzed 21 affected individuals from 6 SPG4 Danish families with SPG linked to 2p24-p21. They found that 20 of 21 affected individuals showed CAG repeat expansions of the SPG4 gene (604277.0006) versus 2 of 21 healthy spouses, demonstrating a strongly statistically significant association between the occurrence of the repeat expansion and the disease. Presumably, CAG repeat expansion is involved as a dynamic mutation in SPG4. They estimated the expansion to be equal to or greater than 60 CAG repeat copies in the affected individuals.

Benson et al. (1998) analyzed 20 familial spastic paraplegia families, including 4 for which there was evidence for linkage to the SPG4 region on 2p24-p21, and found that in most cases the repeat expansion detected by the RED method was due to nonpathogenic expansions of the chromosome 18q21.1 SEF2-1 locus (TCF4; 602272) or the 17q21.3 ERDA1 locus (603279). Polymorphic expansions at SEF2-1 and ERDA1 appeared frequent and can confound RED studies in the search for genes causing disorders demonstrating anticipation. In 6 SPG families, however, the CAG repeat expansion was detected in a subset of affected and at-risk individuals that did not result from expansion of either of these loci. Overall, 11 of 37 (30%) of the SPG patients with a CAG/CTG repeat expansion were unaccounted for by the SEF2-1 and ERDA1 loci, compared with 2 of 23 (9%) of the unaffected at-risk individuals and none of 19 controls. In the majority of cases the novel expansions were shorter than those previously reported.

Fonknechten et al. (2000) analyzed DNA from 87 unrelated patients with autosomal dominant hereditary spastic paraplegia and detected 34 novel mutations scattered along the coding region of the SPG4 gene. They found missense (28%), nonsense (15%), and splice site point (26.5%) mutations as well as deletions (23%) and insertions (7.5%). Mean age at onset was 29 +/- 17 years, with a range of 0 to 74 years. Disease severity was highly variable among patients, and disease progression was actually faster in the late-onset group. Penetrance was age-dependent and incomplete even in older mutation carriers (85% after 40 years). Six percent of 238 mutation carriers were asymptomatic, while 20% of carriers were unaware of their symptoms. There was no difference in either age of onset or clinical severity among groups of patients with missense mutations versus truncation mutations.

Svenson et al. (2001) stated that pure hereditary spastic paraplegia type 4 is the most common form of autosomal dominant hereditary SPG. They screened the spastin gene (604277) for mutations in 15 families showing linkage to the SPG4 locus and identified 11 mutations, 10 of which were novel. In 15 of 76 unrelated individuals with hereditary spastic paraplegia, Meijer et al. (2002) identified 5 previously reported mutations and 8 novel mutations in the SPG4 gene.

Svenson et al. (2004) identified 2 rare polymorphisms in the SPG4 gene: ser44 to leu (S44L; 604277.0015) and pro45 to gln (P45Q; 604277.0017). In affected members of 4 SPG4 families, the presence of either the S44L or P45Q polymorphism in addition to a disease-causing SPG4 mutation (see, e.g., 604277.0016; 604277.0018) resulted in an earlier age at disease onset. Svenson et al. (2004) concluded that the S44L and P45Q polymorphisms, though benign alone, modified the SPG4 phenotype when present with another SPG4 mutation.

Depienne et al. (2006) identified 19 different mutations in the SPG4 gene in 18 (12%) of 146 unrelated mostly European patients with progressive spastic paraplegia. Most of the patients had no family history of the disorder.

In 13 (26%) of 50 unrelated Italian patients with pure hereditary spastic paraplegia (HSP), Crippa et al. (2006) identified 12 different mutations in the SPG4 gene, including 8 novel mutations. All 5 of the familial cases analyzed carried an SPG4 mutation, confirming that the most common form of autosomal dominant HSP is caused by mutations in this gene. Eight (18%) of 45 sporadic patients had a SPG4 mutation. No mutations were identified in 10 additional patients with complicated HSP. Genotype-phenotype correlations were not observed.

In 24 (20%) of 121 probands with autosomal dominant SPG in whom mutations in the SPG4 gene were not detected by DHPLC, Depienne et al. (2007) identified 16 different heterozygous exonic deletions in the SPG4 gene using multiplex ligation-dependent probe amplification (MLPA). The deletions ranged in size from 1 exon to the whole coding sequence. The patients with deletions showed a similar clinical phenotype as those with point mutations but an earlier age at onset. The findings confirmed that haploinsufficiency of SPG4 is a major cause of autosomal dominant SPG and that exonic deletions account for a large proportion of mutation-negative SPG4 patients, justifying the inclusion of gene dosage studies in appropriate clinical scenarios. Depienne et al. (2007) stated that over 150 different pathogenic mutations in the SPG4 gene had been identified to date.

Using MLPA analysis, Beetz et al. (2007) identified partial deletions of the SPG4 gene in 7 of 8 families who had been linked to the region, but in whom mutation screening had not identified mutations. The families had been previously reported by Lindsey et al. (2000), McMonagle et al. (2000), Meijer et al. (2002), and Svenson et al. (2001). The findings indicated that large genomic deletions in SPG4 are not uncommon and should be part of a workup for autosomal dominant SPG.

Mitne-Neto et al. (2007) identified a heterozygous tandem duplication of exons 10 through 12 of the SPG4 gene (604277.0022) in affected individuals of a large Brazilian kindred with spastic paraplegia, originally reported by Starling et al. (2002). In this family, Starling et al. (2002) noted that there were 24 affected men and only 1 affected woman, but X-linked inheritance was ruled out. The authors found strong linkage to the SPG4 locus, but no mutations were identified in the coding region of the SPG4 gene. The results of Mitne-Neto et al. (2007) thus confirmed the diagnosis of SPG4. At the time of the latter report, 12 of 30 mutation carriers had no clinical complaints. Among these patients, 9 of 14 female carriers had no complaints, indicating sex-dependent penetrance in this family, with women being partially protected.

Shoukier et al. (2009) identified SPG4 mutations in 57 (28.5%) of 200 unrelated, mostly German patients with SPG. There were 47 distinct mutations identified, including 29 novel mutations. In a review of other reported mutations, the authors found that most (72.7%) of the mutations were clustered in the C-terminal AAA domain of the SPG4 gene. However, clustering was also observed in the MIT (microtubule interacting and trafficking), MTBD (microtubule-binding domain), and an N-terminal region (residues 228 to 269). In the original cohort of 57 patients, there was a tentative genotype-phenotype correlation indicating that missense mutations were associated with an earlier onset of the disease.

Population Genetics

Among 49 Japanese probands with autosomal dominant SPG who underwent analysis of candidate SPG genes, Ishiura et al. (2014) found that SPG4 was the most common type, accounting for 55.1% of patients.

Animal Model

Du et al. (2010) showed that exogenous expression of wildtype Drosophila or human spastin rescued behavioral and cellular defects in spastin-null flies equivalently. Flies coexpressing 1 copy of wildtype human spastin and 1 copy with the K388R catalytic domain mutation in the fly spastin-null background exhibited aberrant distal synapse morphology and microtubule distribution, similar to but less severe than spastin nulls. R388 or a separate nonsense mutation acted dominantly and were sufficient to confer partial rescue. As in humans, both L44 (604277.0015) and Q45 (604277.0017) were largely silent when heterozygous, but exacerbated mutant phenotypes when expressed in trans with R388.

Cytogenetics

Miura et al. (2011) reported a 4-generation Japanese family from the Miyazaki prefecture in southern Japan with autosomal dominant SPG. RT-PCR and sequencing of affected individuals identified a heterozygous 70-kb deletion of 2p23 encompassing exons 1 to 4 of the SPAST gene as well as exons 1 to 3 of the neighboring DPY30 (612032) gene, located approximately 24 kb upstream of SPAST in a head-to-head orientation. The clinical features included early childhood onset of slowly progressive spastic paraplegia, decreased vibration sense at the ankles, urinary disturbances, and mild cognitive impairment. All 4 affected females had miscarriages, which Miura et al. (2011) speculated may have resulted from loss of DPY30, which plays a role in the regulation of X chromosome dosage compensation and possibly affects sex determination in C. elegans (Hsu and Meyer, 1994).

Nomenclature

Hazan et al. (1993) referred to the form of autosomal dominant spastic paraplegia encoded by a gene on chromosome 14q as FSP1, and Hazan et al. (1994) referred to the form encoded by a gene on chromosome 2p as FSP2. The genes for these 2 disorders are also symbolized SPG3 and SPG4, respectively.