Glucocerebrosidase Is Shaking Up the Synucleinopathies

Marina Siebert; Ellen Sidransky; Wendy Westbroek


Brain. 2014;137(5):1304-1322. 

In This Article

Regulators of GBA1 and/or Glucocerebrosidase Expression and Their Implications in the Synucleinopathies

It is now clear that Gaucher disease encompasses a broad spectrum of clinical phenotypes, with limited correlation between genotype and phenotype (Sidransky, 2004, 2012). This suggests the involvement of modifier genes that can interact with the disease-causing allele and influence its phenotypic expression (Goker-Alpan et al., 2005). Several potential modifiers for Gaucher disease have been proposed (Latham et al., 1990; Winfield et al., 1997; Montfort et al., 2004). Although mutations in GBA1 are a common risk factor for the development of Parkinson's disease, only a fraction of patients with Gaucher disease and carriers for GBA1 mutations develop Parkinson's disease (Sidransky et al., 2009). This leads us to speculate that potential disease modifiers in this process might serve as additional risk factors that, in combination with GBA1 mutations, might favour the development and progression of synucleinopathies.

SCARB2/LIMP2: A Genetic Modifier of GBA1

Although the majority of lysosomal enzymes reach their destination via the mannose-6-phosphate receptor pathway, a subset gets sorted through mannose-6-phosphate receptor-independent pathways (Coutinho et al., 2012). One of these enzymes is glucocerebrosidase, which reaches the lysosome via lysosomal integral membrane protein type 2 (LIMP-2) mediated trafficking (Reczek et al., 2007) (Fig. 2). LIMP-2, encoded by the scavenger receptor class B member 2 (SCARB2) gene located on chromosome 4q13-21, belongs to the CD36 family of scavenger receptor proteins (Fujita et al., 1991; Febbraio et al., 2001; Berkovic et al., 2008). LIMP-2, which is ubiquitously expressed, is one of the most abundant transmembrane proteins in the lysosomal membrane (Eskelinen et al., 2003); mutation and over-expression studies suggest that it plays a role in the biogenesis and maintenance of late endosomes and lysosomes (Kuronita et al., 2002; Eskelinen et al., 2003), as well as in the fusion between lysosomes and autophagosomes (Gleich et al., 2013).

Figure 2.

Regulators of GBA1 expression and glucocerebrosidase activity and trafficking. (1) Phosphorylated TFEB is located in the cytoplasm. Under starvation conditions, dephosphorylated TFEB translocates to the nucleus where it regulates the transcription of genes involved in the CLEAR network, including GBA1. (2) GBA1 messenger RNA is translated into glucocerebrosidase. The interaction with its receptor LIMP-2 facilitates translocation of glucocerebrosidase to the lysosomes via the endoplasmic reticulum (ER), Golgi and late endosomes (LE). (3) In the lysosomes, glucocerebrosidase dissociates from LIMP-2. The glucocerebrosidase activator, saposin C (SAPC), lifts glucocerebroside from the lysosomal membrane and/or membranes within the lysosome and makes it available for glucocerebrosidase-mediated breakdown. (4) When LIMP-2 is deficient or absent, the glucocerebrosidase enzyme cannot be correctly sorted to the lysosomes. As a consequence, glucocerebrosidase is secreted into the extracellular environment (ECE). (5) Saposin C deficiency leads to the impairment of glucocerebroside degradation since the substrate is not available to glucocerebrosidase and subsequently accumulates inside the lysosomes. CM = Cell Membrane.

It was not until 2007 that LIMP-2 was identified as the receptor required for the trafficking of glucocerebrosidase to the lysosomes (Reczek et al., 2007). Protein interaction studies showed a pH-dependent interaction between LIMP-2 and glucocerebrosidase, which was regulated by the pH-sensor amino acid histidine 171 (Zachos et al., 2012). This direct interaction between glucocerebrosidase and LIMP-2 is initiated at neutral pH within the endoplasmic reticulum, and is disrupted upon reaching the acidic lysosome (Reczek et al., 2007; Blanz et al., 2010; Zachos et al., 2012). Studies of SCARB2 knock-out mice showed GBA1 messenger RNA levels were not affected, but there was a decrease in glucocerebrosidase activity and protein levels along with lysosomal accumulation of glucocerebroside (Reczek et al., 2007). Conditioned media taken from SCARB2 knock-out cells in culture demonstrate that glucocerebrosidase is secreted into the extracellular environment as a result of impaired trafficking of glucocerebrosidase (Reczek et al., 2007; Velayati et al., 2011). Mutations in SCARB2 are associated with action myoclonus-renal failure syndrome (OMIM #254900), an autosomal recessive disorder characterized by renal pathology, progressive myoclonus epilepsy and ataxia (Berkovic et al., 2008; Blanz et al., 2010). Studies on four different mutant SCARB2 lines showed that the mutant proteins are retained in the endoplasmic reticulum and affect glucocerebrosidase activity in the lysosomes by binding to glucocerebrosidase in the endoplasmic reticulum and preventing its translocation to the lysosomes (Balreira et al., 2008; Blanz et al., 2010). SCARB2 was recently identified as a genetic modifier for GBA1 in a study of a unique pair of siblings who had discordant Gaucher disease phenotypes but identical genotypes. One sib suffered myoclonic seizures and sequencing of his SCARB2 gene revealed a novel heterozygous c.1412A>G (p.Glu471Gly) mutation in one allele, which was absent from the brother. Expression studies in fibroblasts from this patient revealed significant downregulation of LIMP-2 and glucocerebrosidase protein levels, as well as glucocerebrosidase enzyme activity. Secretion of mature glucocerebrosidase into the extracellular environment was observed (Velayati et al., 2011). As LIMP-2 is crucial for the correct trafficking of glucocerebrosidase, and LIMP-2 malfunction can lead to a reduction in glucocerebrosidase levels and activity, it is tempting to speculate a role for LIMP-2 in the development of Parkinson's disease. SCARB2 mutations and Parkinson's disease could be related through the modulation of glucocerebrosidase protein levels and activity in the cell. LIMP-2 deficiency could lead to glucocerebrosidase secretion instead of proper delivery to the lysosome, which could result in accumulation of glucocerebroside substrate, alterations in lysosomal function, and aggregation of proteins such as α-synuclein inside the lysosomes. It was demonstrated in a cell model over-expressing α-synuclein that less glucocerebrosidase was bound to LIMP-2, which indicates less translocation of glucocerebrosidase to the lysosome (Gegg et al., 2012).

It still remains unclear how a variation near or inside SCARB2 could be associated with Parkinson's disease (Hopfner et al., 2013). Recent genetic-based evidence has suggested an association between SCARB2 and Parkinson's disease. Genome-wide association studies identified an association between rs6812193, a single nucleotide polymorphism located upstream of the SCARB2 gene, and Parkinson's disease (OR = 0.84) in a population of European ancestry (Do et al., 2011). The single nucleotide polymorphism is located in an intron of FAM47E, a gene encoding a protein of unknown function (Do et al., 2011). This association was confirmed by the International Parkinson's Disease Genomics Consortium (2011) in a two-stage meta-analysis, and further supported by an independent genotyping study of 984 patients with Parkinson's disease and 1014 controls of German/Austrian descent (Hopfner et al., 2013). However, the association was not seen in a Chinese study of 449 patients with Parkinson's disease and 452 controls (Chen et al., 2012; Hopfner et al., 2013). A candidate gene screen, performed on 347 subjects with sporadic Parkinson's disease and 329 controls from Greece, revealed an additional single nucleotide polymorphism, rs6825004, located within intron 2 of SCARB2, that appeared to be associated with Parkinson's disease (OR = 0.68). However, the authors recognized the limitations in their study because of the small sample size (Michelakakis et al., 2012). In another small study, the presence of rs6812193 and/or rs6825004 single nucleotide polymorphisms and corresponding SCARB2 and LIMP-2 expression levels were assessed in 15 lymphocyte and leucocyte samples derived from individuals without Parkinson's disease. There was no indication that the SCARB2 single nucleotide polymorphism genotypes described were associated with the modulation of SCARB2 messenger RNA and LIMP-2 protein expression levels (Maniwang et al., 2013).

Saposin C: An Activator of Glucocerebrosidase

Mature saposin C (SAPC) is a glucocerebrosidase enzyme activator in lysosomes and is essential in the hydrolysis of glucocerebroside (Beutler and Grabowski, 2001; Sidransky, 2004; Vaccaro et al., 2010), but the mechanism for this activation is not fully understood. Saposin A, B, C, and D are small homologous glycoproteins with six cysteine residues forming disulphide bridges. The bridges are crucial for saposin C function (Tamargo et al., 2012). Saposin proteins are generated through proteolytic cathepsin D-mediated cleavage of its precursor prosaposin (Hiraiwa et al., 1997; Yuan and Morales, 2011). Biophysical experimental evidence indicates that saposin C-mediated extraction and solubilization of glucocerebroside exposes the lipid substrate to glucocerebrosidase for subsequent hydrolysis (Alattia et al., 2006, 2007) (Fig. 2). An additional role for saposin C is the protection of glucocerebrosidase against proteolytic breakdown, which is demonstrated by a significant reduction in levels of glucocerebrosidase protein and enzyme activity in saposin C-deficient cells (Sun et al., 2003, 2010). Because of its essential role as a glucocerebrosidase activator, saposin C could be a modifier gene for Gaucher disease and potentially Parkinson's disease. Indeed, crossbreeding studies with a mouse model of saposin C, created by a knock-in mutation in exon 11 of the prosaposin gene, and the V394L homozygous Gaucher mouse (Xu et al., 2003; Hruska et al., 2008; Sun et al., 2010), revealed that the combined deficiencies exacerbate the Gaucher disease phenotype, with progressive neurological complications resulting in early death, greater glucocerebrosidase activity reduction, significant defects in glucocerebroside 18:0 species breakdown in the brain, and increased storage of the substrates glucocerebroside and glucosylsphingosine (Sun et al., 2013b). This model confirmed that saposin C could act as a disease modifier for Gaucher disease. Only six patients with saposin C deficiency have been described in the literature and a correlation was observed between the type of mutation and the nature of their Gaucher-like phenotype. Patients with mutations in the crucial cysteine residues in the saposin C domain of prosaposin had a clinical phenotype similar to Gaucher disease type 3, whereas those with other mutations resembled non-neuronopathic type 1 Gaucher disease (Christomanou et al., 1986, 1989; Schnabel et al., 1991; Rafi et al., 1993; Diaz-Font et al., 2005; Tylki-Szymanska et al., 2007, 2011; Vaccaro et al., 2010). Complete deficiency of prosaposin and consequently all saposins, resulted in a severe fatal neurological infantile sphingolipidosis (Hulkova et al., 2001). As patients with both saposin C and glucocerebrosidase deficiencies have never been identified, it is difficult to assess the role of saposin C as a modifier gene in human samples. Interestingly, patient fibroblasts with cysteine saposin C mutations showed an accumulation of autophagosomes, which was believed to be caused by reduced protein levels and enzymatic activity of both cathepsin B and D (Tatti et al., 2011, 2012, 2013). Exogenous over-expression of both cathepsins restored autolysosomal degradation (Tatti et al., 2013). This secondary effect of saposin C deficiency is of great interest as malfunctions in the autophagy clearance pathway and their role in the development of Parkinson's disease are well documented (Lim and Zhang, 2013), as is the role of cathepsin D in proteolytic breakdown of α-synuclein (Cullen et al., 2009). Although saposin C can act as a modifier gene for Gaucher disease in a mouse model (Sun et al., 2010), appropriate saposin C expression studies on samples from patients with Gaucher disease with discordant phenotypes are currently lacking. Also, considering the recent observations of reduced wild-type glucocerebrosidase activity in Parkinson's disease models and patient samples, it is possible that altered saposin C levels in patients with Parkinson's disease (with or without GBA1 mutations) could be a crucial determinant in the development of synucleinopathies.

TFEB: A Regulator of GBA1/Glucocerebrosidase Expression

A majority of the genes involved in both lysosomal function and biogenesis are part of the coordinated lysosomal expression and regulation network. Expression of gene members of this network is positively regulated by the basic helix-loop-helix leucine zipper transcription factor EB (TFEB), which binds to the GTCACGTGAC motif element within their promoter region (Sardiello et al., 2009). Work by Ballabio and colleagues established that TFEB is part of a signalling pathway by which lysosomes self-regulate. Indeed, experimental data in Drosophila S2, human HEK293T cells, and a cell-free system support an 'inside-out' model in which accumulated amino acids inside the lysosome initiate signalling through the v-ATPase-Ragulator protein complex to Rag-GTPases, which, in turn, recruit mammalian target of rapamycin (mTOR) to the surface of the lysosomes (Sancak et al., 2008, 2010; Zoncu et al., 2011). TFEB interacts with mTOR on the lysosomal surface, where phosphorylation of multiple serine residues by mTOR prevents TFEB translocation to the nucleus (Settembre and Ballabio, 2011; Settembre et al., 2012; Martina and Puertollano, 2013). Cell starvation, which includes amino acid depletion within the lysosome, results in inhibition of the 'inside-out' signalling pathway, and eventual mTOR release from the lysosome surface. TFEB is no longer phosphorylated, and translocates to the nucleus, where it activates transcription of the coordinated lysosomal expression and regulation network genes and autoregulates its own expression through a feedback loop (Settembre et al., 2012, 2013). In addition to its role in lysosomal function and biogenesis, TFEB is also a key player in lipid metabolism (Settembre et al., 2013), autophagosome formation and autophagosome-lysosome fusion (Settembre and Ballabio, 2011), and Ca2+-mediated lysosomal exocytosis, which can positively affect cellular substrate clearance in select lysosomal storage disorders, including Batten disease, Pompe disease, neuronal ceroid lipofuscinoses, multiple sulphatase deficiency, and mucopolysaccharidosis type IIIA (Medina et al., 2011).

TFEB over-expression and silencing studies in HeLa cells showed that TFEB positively regulated GBA1 messenger RNA expression (Fig. 2). Additionally, chromatin immunoprecipitation analysis confirmed that GBA1 is a direct target of TFEB (Sardiello et al., 2009). The TFEB field is still in its infancy and very few studies on its role in neurodegeneration are available. One study showed that adenovirus-mediated over-expression of human α-synuclein in the midbrain of rats induced TFEB retention in the cytoplasm, blockage of lysosomal function, accumulation of α-synuclein in autophagosomes, and progressive build-up of α-synuclein oligomers. Co-immunoprecipitation experiments showed an interaction between α-synuclein and TFEB, suggestive of a role for α-synuclein in cytoplasmic sequestration of TFEB. These observations were confirmed in nigral dopaminergic neurons of post-mortem Parkinson's disease midbrains. In the α-synuclein rat model, both over-expression of TFEB or activation through pharmacological inhibition of mTOR resulted in a block in the progression of α-synuclein-mediated neurodegeneration. This study puts TFEB on the map as a key player in Parkinson's disease (Decressac et al., 2013). Recently, reduced wild-type glucocerebrosidase protein levels were observed in samples from patients with synucleinopathies (Balducci et al., 2007; Parnetti et al., 2009; Gegg et al., 2012). It is possible that this could be because of α-synuclein-induced TFEB retention in the cytoplasm with consequently lower transcription of GBA1 messenger RNA. Currently, TFEB is the only known transcription factor for GBA1, but a study of the promoter and regulatory regions of GBA1 revealed several conserved transcription factor-binding sites resulting in altered GBA1 expression levels when mutated. This suggests that these regions might be involved in transcriptional regulation of GBA1 and potentially contribute to the complex phenotypic diversity observed in Gaucher disease including the development of Parkinson's disease (Blech-Hermoni et al., 2010).