GLUCOCEREBROSIDASE MUTATIONS IN 108 NEUROPATHOLOGICALLY CONFIRMED CASES OF MULTIPLE SYSTEM ATROPHY

Abstract

Parkinson disease (PD), Lewy body dementia (LBD), and multiple system atrophy (MSA) are synucleinopathies whose primary pathogenic event is the deposition of inclusions composed of aberrantly fibrillized α-synuclein.1 In PD and LBD, Lewy bodies are the key aggregate, whereas in MSA, α-synuclein accumulates in the form of oligodendroglial and neuronal cytoplasmic inclusions (GCIs and NCIs).2,3 Parkinsonian manifestations have been noted in a subset of patients with Gaucher disease and there is evidence that parkinsonism is more frequent among carrier relatives of patients with Gaucher disease.4 In a remarkable study, the glucocerebrosidase (GBA) gene was sequenced in an American PD brain bank series where GBA mutations were detected at a much higher frequently than in controls (PD 21% vs control 4.5%).5 These findings have since been replicated, mainly in Ashkenazi patient groups who have a higher mutation frequency but also in patients with clinically and pathologically diagnosed PD and LBD in a number of studies in different populations.4 In a study of 75 neuropathologically confirmed synucleinopathies, GBA mutations were found in 23% of the cases with Lewy bodies.6 The frequency of GBA mutations around the world between 2.3 and 31% (depending on population) indicates that GBA mutations are one of the commonest genetic risk factors for PD. GBA mutation carriers have a wide spectrum of phenotypes, ranging from classic l-dopa-responsive PD to LBD. In neuropathologic studies of PD/LBD cases, GBA mutations, α-synuclein inclusions, and Lewy bodies have been seen. This spectrum of clinical and pathologic features would suggest that MSA should also be a candidate to have GBA mutation.3 Only 12 cases of MSA have been analyzed for GBA mutations and defects were seen in this handful of cases.6 We extracted DNA from the brain tissue of 108 neuropathologically confirmed British MSA cases that had been diagnosed according to brain bank criteria and 257 normal British controls. Mean age at onset was 58.2 ± 10.7 years (range 34–83), mean age at death 64.5 ± 10.2 years (39–87), mean disease duration 6.8 ± 2.9 years (2–16), and 48% were men. All exons and flanking intronic regions of the GBA gene were sequenced in MSA and control cases. To avoid amplifying and sequencing the GBA pseudogene we employed long range GBA PCR and then BigDye sequencing as previously described.7 In our MSA study group of 108 cases, we identified one heterozygous GBA mutation (c.904C>T; R262H), giving a mutation frequency of 0.92%. In the British controls, three heterozygous mutations (V497L, N409S, and R269Q) out of 257 cases were identified (1.17%). There was no significant difference between the two groups (p = 0.66). The single MSA case with the heterozygous R262H mutation was a woman with an age at onset of 44 years. She had parkinsonian, cerebellar, and autonomic features (MSA–mixed type) with no family history. She died at age 51 years and the neuropathology revealed widespread GCIs and NCIs with a predominance in striatonigral structures. There were no Lewy bodies. One limitation of our study is the small sample size. Our study has a power of 80% to detect variants with an OR >1.61 or <0.63 at a significance level of 0.05. The results of this study indicate that GBA mutations are not common etiologic players in Caucasian patients with MSA. We cannot exclude that GBA mutations confer modest or low risk to disease. Furthermore, we did not sequence risk variants in regulatory regions (such as the promotor region or untranslated regions). Mutations in these regions would therefore have been missed. The unexpected role of GBA mutations has been demonstrated in several populations and is undoubtedly a highly significant risk factor for PD and LBD. More importantly, GBA mutations reveal a direct link between the lysosomal protein pathway and the clearance or the development of α-synuclein aggregates into Lewy bodies. Our study indicates that GBA mutations are not associated with MSA in the population that we analyzed, and that this branch of the ceramide pathway is unlikely to be associated with all types of primary α-synuclein deposition.