GPR50, a potential factor involved in psychiatric disorders interacts with Alzheimer’s disease-related protein β- secretase (BACE1)

Abstract

GPR50, an X-linked orphan G protein-coupled receptor (GPCR), is a risk factor for bipolar disorder (BD) in female subjects. It has been shown that GPR50 plays a part in neurite outgrowth, glucocorticoid receptor signalling and leptin signalling by interacting with major factors involved in these events. Yeast two-hybrid screens have identified multiple putative GPR50 interactors involved in neurodevelopment, stress response and apoptosis, lipid and glucose metabolism, as well as regulation of NMDA receptors and GABA transmission. Among these interactors, RTN3, RTN4, SREBP2 and SNX6 are known regulators of β-secretase (BACE1), a key enzyme in Aβ generation, myelination of the central/peripheral nerve, and neurite outgrowth/synapse formation. Preliminary data indicated that GPR50 expression significantly increased endogenous BACE1 activity in HEK293 cells, so I hypothesised that there is a functional interaction between the two. In this thesis, I investigated the relationship between GPR50 and BACE1 by identifying the effects of GPR50 on BACE1 expression and function, which may provide an explanation of GPR50’s potential association with psychiatric disorders and Alzheimer’s disease. Firstly, studies on expression levels revealed that when GPR50 was over-expressed, BACE1 protein expression was up-regulated in SH-SY5Y cells, but down-regulated in HEK293 cells, suggesting a differentiated regulative system between cell lines. Then I confirmed the physical association between endogenous GPR50 and BACE1 in HEK293 cells by co-localisation and co-immunoprecipitation studies. Their putative interaction sites were located at the plasma membrane and the filopodia/lamellipodia-like structures in HEK293 cells, and at the neurites in mouse primary neuronal cells. Subcellular fractionation of adult mouse brain revealed that endogenous Gpr50 and Bace1 were co-fractionated in the presynaptic vesicles. Secondly, I showed that, in contrast to HEK293 cells, GPR50 overexpression had no effects on β-secretase activity in mouse primary cortical neurons. However, the BD-associated variant GPR50del significantly decreased β-secretase activity compared to the more common variant GPR50, and showed a trend of diminishing β-secretase activity compared to the control condition. Subcellular fractionation experiments showed that in HEK293 cells, there was an increased ratio of mature BACE1 against immature BACE1 localised in the plasma membrane fractions, indicating a role in regulation of BACE1 trafficking to one of its putative activity sites; whereas in mouse primary cortical neurons, GPR50del increased co-fractionation of immature Bace1 with endoplasmic reticulum (ER) marker calreticulin, thus potentially retarding the maturation of Bace1. Importantly, the regulative trend of GPR50/GPR50del on β-secretase activity is cell line-specific and is highly correlated to their effects on β-secretase intracellular distribution. Thirdly, I found that the mRNA levels of human GPR50 and BACE1 were negatively correlated in the dorsolateral prefrontal cortex of female subjects sampled after birth. Mouse Gpr50 and Bace1 mRNA levels were negatively correlated across the telencephalon regions, and had a trend of negative correlation across the hypothalamic regions. Co-localisation of the two proteins was detected in multiple mouse brain regions, with the strongest co-localised signals occurring in CA2 pyramidal neurons, arcuate hypothalamic nucleus and dorsomedial nucleus of the hypothalamus. Finally, preliminary experiments in Alzheimer’s disease model TgSwDI mice, suggested that the expression level of Gpr50 in layer V of the entorhinal cortex was positively correlated with Aβ deposition. Decreased Gpr50 expression was identified in the hippocampus of 9 months transgenic animals compared with age-matched controls. This indicates that Gpr50 expression might be altered in this mouse model co-ordinately with Aβ deposition. The findings in this thesis provide further evidence of GPR50’s correlation to psychiatric illnesses and its interaction with enzyme BACE1 highlights a potential link to neurodegenerative disease.