MAPT association with REM sleep behavior disorder

Idiopathic REM sleep behavior disorder (IRBD) is a REM sleep parasomnia comprising unpleasant dreams, dream-enacting behaviors, and loss of muscle atonia during REM sleep. Longitudinal studies have demonstrated that because most patients with IRBD develop with time Parkinson disease (PD) and other synucleinopathies including dementia with Lewy bodies (DLB) or multiple system atrophy (MSA), IRBD represents a prodromal stage of these diseases.¹ In PD, 5%–10% of cases are caused by nonfrequent mendelian mutations segregating with disease in pedigrees, but the vast majority of cases are sporadic. Unbiased genome-wide association studies in sporadic PD (sPD) have shown that single nucleotide polymorphisms (SNPs) in the α-synuclein (SNCA) and microtubule-associated protein tau (MAPT) genes modulate disease susceptibility.² Given that IRBD often antedates sPD and that, akin to PD, familial clustering of IRBD is rare, we hypothesized that genetically shared variation at SNCA and MAPT may influence disease susceptibility to both conditions. Accordingly, we genotyped PD-associated genetic variants in SNCA and MAPT in a cohort of Spanish patients with IRBD.

Methods.

The SNPs rs356219 (A/G) in SNCA and rs1800547 (H1/H2) in MAPT have been consistently associated with sPD in the Spanish population.^3,4 Here, we genotyped these markers in a Spanish Caucasian sample of 121 patients with IRBD and 175 healthy controls. This population was studied in a previous report.⁵ The IRBD group was recruited at the Multidisciplinary Sleep Unit of the Hospital Clínic de Barcelona, had a polysomnography-confirmed diagnosis, and consisted of 21/100 women/men (17.4/82.6%), with age at IRBD onset of 68.3 ± 6.2 years and age at sample collection of 71.2 ± 6.4 years. Patients fulfilled the diagnosis of IRBD and were free of parkinsonism, mild cognitive impairment, and dementia at the time of sample collection. After 6.8 ± 4.1 years of follow-up from IRBD diagnosis, 38 patients (31.4%) were clinically diagnosed with a synucleinopathy (DLB in 19 patients, PD in 18, and MSA in 1). Sex-, age-, and demographic-matched controls were recruited at the Multidisciplinary Sleep Unit of the Hospital Clínic de Barcelona (n = 45) and the Clínica Universitaria de Navarra (n = 129), and consisted of 21/154 women/men (13.6/86.4%) with age at sample collection of 71.7 ± 6.4 years. DNA was extracted from the peripheral blood using standard procedures. Genotyping was performed using the predesigned TaqMan assays C-1020193-10 (SNCA rs356219) and C-7563692-10 (MAPT rs1800547) in a StepOnePlus Real-time PCR System (Applied Biosystems, Foster City, CA). Allelic and genotypic frequencies were compared using SNPstats software (bioinfo.iconcologia.net/SNPstats).

Standard protocol approvals, registrations, and patient consents.

Samples were collected after patient's signed consent, and the local ethics committee approved the study.

Results.

We found no allelic or genotypic distribution differences of the SNCA marker rs356219 in patients with IRBD and in controls. However, we found an association of the MAPT polymorphism rs1800547 with IRBD. More specifically, we found that the frequency of the MAPT H2/H2 genotype which is protective for PD^3,4 was underrepresented in our patients with IRBD. This association held statistically significant after adjusting for sex and age as covariates (odds ratio = 0.16; 95% confidence interval = 0.04–0.69; p = 0.002; table). Moreover, we observed that MAPT genotypes in patients with IRBD, but not in controls, were not in the Hardy-Weinberg equilibrium (p = 0.049), a finding which is consistent with previous observations for disease-linked loci in disease-affected individuals.⁶

Table.

Distribution of genotypic frequencies of SNCA and MAPT polymorphisms in patients with IRBD (n = 121) and sex- and age-matched healthy controls (n = 175)

Discussion.

We found that the MAPT marker rs1800547 is associated with IRBD. Consistently, we detected that the H2 variant which is protective for PD was underrepresented in our patients with IRBD. On the contrary, we did not find an association of the SNCA polymorphism rs356219 with IRBD. Altogether, these results are in agreement with another IRBD study reporting association for MAPT but not for SNCA polymorphisms.⁷ Yet, markers in both studies were different. These data raise the question whether IRBD shares with PD the same genetic risk factors or whether, alternatively, IRBD represents a specific endophenotype of PD with only certain genetic risk factors shared in common by both conditions. The later might be a plausible explanation because, although IRBD is considered a prodrome of sPD, not all patients with sPD develop RBD prior to parkinsonism. Supporting this view, we have recently reported that pathogenic mutations in the leucine-rich repeat kinase 2 gene (LRRK2), which are the most frequent genetic cause of PD, are absent in our IRBD cohort.⁵ One limitation of our study is the reduced number of patients with IRBD. Thus, future studies in larger cohorts are warranted to validate the association of MAPT with IRBD and also to further elucidate whether or not SNCA polymorphisms play a role in disease. We found that one polymorphism in MAPT which is associated with PD also modulates the propensity to IRBD. These results point toward an at least in part overlapping genetic susceptibility to both conditions in our Spanish population.