Normal Dopaminergic Nigrostriatal Innervation in SPG3A Hereditary Spastic Paraplegia

Abstract

SPG3A/atlastin-1 gene mutations cause an autosomal dominant form of hereditary spastic paraplegia (SPG3A-HSP). We used positron emission tomography with [¹¹C]DTBZ to assess nigrostriatal dopaminergic integrity in two unrelated adults with SPG3A-HSP due to the common SPG3A/atlastin-1 mutation, R239C. Nigrostriatal dopaminergic terminal density was normal. A difference from the human pattern of neurodegeneration is a critical limitation of this Drosophila model of SPG3A-HSP. This major difference between human SPG3A/atlastin-1 mutations and the Drosophila atl^l phenotype has several possible explanations.

INTRODUCTION

The hereditary spastic paraplegias (HSPs) are a genetically heterogeneous group of disorders characterized by lower extremity spasticity and weakness (Fink, 2007). Postmortem studies of uncomplicated HSP reveal axonal degeneration in distal ends of corticospinal tracts and fasciculus gracilis fibers.

SPG3A/atlastin-1 gene mutations cause an autosomal dominant form of hereditary spastic paraplegia (SPG3A-HSP) that usually manifests with early-childhood-onset spastic gait ( Zhao et al., 2001). SPG3A/atlastin-1 encodes a 63.5-kD protein (atlastin-1) that contains conserved GTPase motifs and bears a sequence similarity to guanylate-binding protein 1, a member of the dynamin family of large GTPases (Zhao et al., 2001). Through database analysis, Zhu et al. (2003) identified two highly similar proteins, designated atlastin-2 and -3.

Lee et al. (2008) reported that Drosophila bearing a null mutation (designated atl^l) in an SPG3A/atlastin-1 homolog exhibited paralysis following mechanical shock, premature mortality, and age-related degeneration of dopaminergic neurons. The motor impairment could be rescued by feeding atl^l mutant Drosophila the dopamine precursor L-dopa or the dopaminergic D1 receptor agonist, SK&F 38393. The entire phenotype could be rescued by transgenic expression of wild-type atl. Observations of dopaminergic neuron vulnerability in atl-mutant Drosophila raise the possibility that degeneration of central dopaminergic neurons could contribute to the phenotype of SPG3A-HSP. Lee et al. point out correctly that this would have treatment implications for individuals with SPG3A-HSP. We evaluated this possibility by assessing nigrostriatal terminal integrity in two unrelated subjects with HSP due to a SPG3A/atlastin-1 mutation.

METHODS

This investigation was approved by our Institutional Review Board.

Clinical

We studied two unrelated female subjects, ages 38 and 37, with autosomal-dominant SP3A-HSP due to SPG3A/atlastin-1 mutation R239C, one of the common SP3A/atlastin-1 mutations. Each subject had a childhood onset of spastic gait that did not significantly worsen in subsequent decades. There was no evidence of cogwheel rigidity, tremor, bradykinesia, hypomimia, hypokinetic dysarthria, slow tongue movements, slowed finger tapping, or other clinical signs of nigrostriatal dopamine deficiency.

Neuroimaging

To quantify nigrostriatal terminal integrity, both subjects underwent [¹¹C]dihydrotetrabenazine positron emission tomography (DTBZ-PET), as described previously (Bohnen et al., 2005). Nigrostriatal terminal density was quantified in dorsal caudate, ventral striatum, anterior putamen, and posterior putamen. DTBZ binding was measured as peak distribution volume (DV) in each subregion. DTBZ binds to the type 2 vesicular monoamine transporter (VMAT2), the protein responsible for pumping monoamines from the cytosol into synaptic vesicles. VMAT2 is expressed at uniquely high levels by nigrostriatal terminals. DV is a measure of binding-site density. Results were compared with a dataset of 22 age-matched controls.

RESULTS

Both HSP subjects had normal striatal DTBZ binding in all striatal subregions (Table 1 and Figure 1).

Table 1.

Striatal Control and Individual HSP Subject Data

Region	Dorsal caudate	Ventral striatum	Anterior putamen	Posterior putamen
Controls: mean BP	2.422	1.668	2.516	2.398
(COV%)	(17.9%)	(19.7%)	(15.7%)	(17.4%)
HSP subject 1	2.484	1.505	2.613	2.596
% of control mean BP	102.6%	90.3%	103.9%	108.3%
HSP subject 2	2.104	1.742	2.337	2.368
% of control mean BP	86.9%	104.5%	92.9%	98.7%

Values are the binding potential (BP; DVR – 1).

Figure 1.

Transaxial parametric DTBZ-PET images of nigrostriatal terminals in SPG3A-HSP subjects and representative control. SPG3A-HSP and control are identical.

DISCUSSION

DTBZ-PET demonstration of normal nigrostriatal innervation correlates with the absence of Parkinsonism in our patients and previously described SPG3A subjects. Although Drosophila appears to be a useful species in which to study the molecular pathogenesis of other forms of HSP, Drosophila bearing the null atl^l mutation do not reproduce faithfully the distribution of neurodegeneration seen in SPG3A-HSP (Orso et al., 2005). DTBZ-PET and other PET methods assessing dopaminergic nigrostriatal terminal density are specific and sensitive measures of nigrostriatal projection integrity. DTBZ-PET was used to detect modest reductions of nigrostriatal terminal density in rapid eye movement sleep behavior disorder, which is often a precursor of Parkinsonism (Albin et al., 2000). Similar PET methods have been used to detect the loss of nigrostriatal terminals in presymptomatic carriers of mutant alleles, causing Mendelian forms of Parkinson's disease (Adams et al., 2005). The natural history of SPG3A-HSP is discrepant with the course of the atl^l mutation phenotype. Lee et al. described age-dependent motor impairment, premature mortality, and age-dependent neurodegeneration. The SPG3A-HSP phenotype, in contrast, develops in childhood and then remains static.

There at least three explanations for the observed species-specific differences in the patterns and time course of neurodegeneration due to SPG3A/atlastin-1 mutation and its Drosophila orthologue. First, these differences could reflect the nature of the discrete mutations in SPG3A/atlastin-1 and atl mutations investigated. The dominant character of the SPG3A-HSP raises the possibility that the human disease results from a gain of function effect. Some data, however, suggest that SPG3A-HSP results from haploinsufficiency. Muglia et al. (2002) described a pedigree with an Arg217Gln missense mutation in a highly conserved part of the GTPase domain of atlastin, probably altering the GTPase active site. Tessa et al. (2002) describe another pedigree with a frameshift mutation leading to a premature translation termination. Second, it is possible that the discrepancy between human SPG3A/atlastin-1 mutations and the atl^l phenotype is related to species-specific differences in selective neuronal vulnerability. Third, it is possible that there are significant differences in the molecular functions of SPG3A/atlastin-1 mutation and its Drosophila ortholog.

CONCLUSIONS

It is notable that while there is one Drosophila atlastin-like locus, human and murine atlastins are a small family of related proteins (atlastin-1, -2, and -3). The possibility exists that dopaminergic degeneration in the Drosophila atl^l null mutation is analogous to the loss of function of either atlastin-2 or -3. It might be worthwhile to evaluate these loci (atlastin-2 or -3) as potential causes of Mendelian forms of Parkinson's disease.