In this study, we sequenced the PPP2R5D gene in 203 EOPD and 465 LOPD Han Chinese patients, and identified three de novo exonic variants, one in a LOPD patient and two in EOPD patients. All three variants are novel and potentially pathogenic with respect to PD. The exonic mutation rate of PPP2R5D in our Han Chinese patients was 0.9% among those with EOPD and 0.2% among those with LOPD. As far as we know, this is the first study to assess genetic sequences of exonic variants in the PPP2R5D gene based on data from a large cohort of EOPD and LOPD patients.
The PPP2R5D gene encodes the B subunit (B56) of protein phosphatase 2A (PP2A). PP2A plays a role in key neuronal and developmental regulation processes by regulating PI3K/AKT and GSK3β-mediated cell growth, chromatin remodeling, and gene transcription[8,9] (Fig. 1C). The E420K mutation of PPP2R5D has been shown in cell culture studies to constitutively activate AKT/mTOR signaling, leading to larger cells and uncoordinated cell growth[10] . Following phosphorylation of S129, PP2A can also regulate α-synuclein, which is an integral part of the Lewy bodies and a pathophysiological hallmark of PD[11]. Increased activity of PP2A activates tyrosine hydroxylase and, consequently, dopamine synthesis[12]. Knocking out the PPP2R5D gene in mice leads the microtubule-associated protein tau to become progressively phosphorylated[13]; such tau can aggregate in PD, either contributing to the disease or simply reflecting its progression[14]. Our study justifies further research into the mechanisms behind the apparent association between PPP2R5D and PD.
A recent study involving three men carrying the p.E200K mutation in the PPP2R5D gene and one patient reported severe atrophy of the substantia nigra but the absence of Lewy body pathology (Patients 4 in Table 1)[5]; similar pathologic characteristics were observed in PD patients with mutations in the parkin gene. All three patients with the p.E200K mutation showed levodopa response, sporadic early-onset parkinsonism, and mild intellectual disability. In another study, a patient with a PPP2R5D p.E250K mutation showed levodopa-responsive, early-onset parkinsonism[6]: FP-CIT (DaTscan) SPECT imaging performed in his early 20s showed lower dopamine uptake in the right basal ganglia than in the left, and the patient rapidly developed motor fluctuations and dyskinesias. Furthermore, a 29-year-old woman with a p.E198K mutation in the PPP2R5D gene showed PPP2R5D-related neurodevelopmental disorder as well as levodopa-responsive, early-onset parkinsonism, indicating that PPP2R5D can play a role in both PPP2R5D-related neurodevelopmental disorder[15,16] and parkinsonism[7]. Compared to patients who carry the E200K mutation, those with the E198K mutation tend to experience more severe intellectual disability and developmental delay[16]. This reflects the critical role of E198K in subunit interaction and binding.
Our study identified three patients with three unique exonic mutations in the PPP2R5D gene. Patient 1 was a 40-year-old man carrying the R91S mutation in exon 3. He developed the first symptoms of PD when he was 39 years old; he had no familial history of neurodegenerative disease.Although the patient experienced mild anxiety, he showed no obvious signs of depression, and his Moca and MMSE scores were normal. Furthermore, he experienced no motor complications, had a H-Y stage of 1.0, and responded well to a rotigotine transdermal patch (6 mg). Patient 2 was a 47-year-old man carrying the R523L mutation in exon 13. He reported no familial history of neurodegenerative disease. He began showing bradykinesia of his left arm when he was 43 years old, followed by the development of serious gait disorder, Freezing of gate(FOG), and wearing-off phenomenon over the next three years. These symptoms responded to treatment using levodopa.Patient 3 was a 59-year-old man carrying the E8A mutation in exon 1. The patient developed PD when he was 56 years old. After treatment with levodopa and piribedil sustained-release tablets, he was able to achieve 70% remission, resulting in the wearing-off phenomenon at approximately three years after his initial diagnosis.
Consistent with previous studies, our study identified two EOPD patients carrying PPP2R5D mutations [6-8]. The present study appears to be the first to detect a PPP2R5D mutation in an LOPD patient. In silico analyses predict that the p.E8A and the p.R91S variants are “disease-causing” mutations. Since these mutations occur at the 5’ promoter region of PPP2R5D, they could potentially affect transcription. In contrast, the R523L mutation is located at the armadillo-type fold of PP2R5D; this structure has been reported to be functionally important for other genes(eg.beta-catenin)[17]. Further studies should be performed to understand the functions of these mutations in vivo.
We must consider the results of our study in the light of certain limitations. The related small number of patients in the EOPD cohorts, as well as our exclusion of patients with a familial history of PD may have caused a bias in our understanding of the pathophysiology of PD. Furthermore, we could not examine the effects of other genes on PD because we restricted our sequencing analysis to the PPP2R5D gene and conducted whole-exome sequencing only for the three patients who carried the variants. Since we only predicted the function of the three mutations using in silico tools, functional experiments are needed to understand the role of these variants in PD.
Despite these limitations, our findings suggest that mutations in the PPP2R5D gene may associated with PD in Han Chinese. Further studies must be conducted to understand the effects of these three exonic variants of PPP2R5D in PD patients, as well as identify other mutations that might be associated with the disease. Based on previous studies and our findings, we recommend adding the PPP2R5D gene to the gene panel sequencing analysis conducted to detect early-onset PD.