Using Extracellular miRNA Signatures to Identify Patients with LRRK2-Related Parkinson’s Disease

Abstract

AbstractBackgroundMutations and polymorphisms in the Leucine Rich Repeat Kinase 2 gene were found to be highly relevant in both sporadic and familial cases of Parkinson’s disease. Specific therapies are entering clinical trials and patient stratification is crucial but remains challenging. Dysregulated microRNA expression levels have been observed in cases of sporadic Parkinson’s disease and were proposed as biomarker candidates. In this proof-of concept study we evaluated the potential of using the extracellular miRNA signature of patients to identify LRRK2-driven molecular patterns in Parkinson’s disease.MethodsWe quantified the expression levels of 91 miRNAs via RT-qPCR in ten individuals with sporadic Parkinson’s disease, ten patients carrying aLRRK2mutation and eleven healthy controls using both plasma and cerebrospinal fluid. We first compared the miRNA signatures of all individuals using heatmaps and t-tests. Next, we applied group sorting algorithms (Least Absoulte Shrinkage and Selection Operator, Principal Component Analysis, Random Forest) and tested the sensitivity and specificity of their group predictions. Finally, we extracted a collection of discriminatory miRNAs from each analysis.ResultsmiR-29c-3p was differentially expressed betweenLRRK2mutation carriers and sporadic cases, with miR-425-5p trending towards significance. Individuals clustered in principal component analysis along mutation status. Group affiliation was predicted with high accuracy in the least absolute shrinkage and selection operator regression model (sensitivity of 89%, specificity of 70%) and the random forest model (sensitivity of 89%, specificity of 70%). miRs-128-3p, 29c-3p, 223-3p and 424-5p were identified as promising discriminators among all analyses.ConclusionsIdentifying LRRK2-driven molecular patterns in Parkinson’s disease by characterizing the extracellular miRNA signatures and using group sorting algorithms potentially provides an easily scalable method for molecular subtyping of patients. Knowing that LRRK2 also plays a role in patients with sporadic forms of the disease, this approach could enable identification of patients that are likely to benefit from drugs targeting LRRK2. In light of small sample size, this approach needs to be validated in larger cohorts.