GPR4 Knockout Improves the Neurotoxin-Induced, Caspase-Dependent Mitochondrial Apoptosis of the Dopaminergic Neuronal Cell

In Parkinson’s disease, mitochondrial oxidative stress-mediated apoptosis is a major cause of dopaminergic neuronal loss in the substantia nigra (SN). G protein-coupled receptor 4 (GPR4), previously recognised as an orphan G protein coupled-receptor (GPCR), has recently been claimed as a member of the group of proton-activated GPCRs. Its activity in neuronal apoptosis, however, remains undefined. In this study, we investigated the role of GPR4 in the 1-methyl-4-phenylpyridinium ion (MPP+) and hydrogen peroxide (H2O2)-treated apoptotic cell death of stably GPR4-overexpressing and stably GPR4-knockout human neuroblastoma SH-SY5Y cells. In GPR4-OE cells, MPP+ and H2O2 were found to significantly increase the expression levels of both mRNA and proteins of the pro-apoptotic Bcl-2-associated X protein (Bax) genes, while they decreased the anti-apoptotic B-cell lymphoma 2 (Bcl-2) genes. In addition, MPP+ treatment activated Caspase-3, leading to the cleavage of poly (ADP-ribose) polymerase (PARP) and decreasing the mitochondrial membrane potential (ΔΨm) in GPR4-OE cells. In contrast, H2O2 treatment significantly increased the intracellular calcium ions (Ca2+) and reactive oxygen species (ROS) in GPR4-OE cells. Further, chemical inhibition by NE52-QQ57, a selective antagonist of GPR4, and knockout of GPR4 by clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 decreased the Bax/Bcl-2 ratio and ROS generation, and stabilised the ΔΨm, thus protecting the SH-SY5Y cells from MPP+- or H2O2-induced apoptotic cell death. Moreover, the knockout of GPR4 decreased the proteolytic degradation of phosphatidylinositol biphosphate (PIP2) and subsequent release of the endoplasmic reticulum (ER)-stored Ca2+ in the cytosol. Our results suggest that the pharmacological inhibition or genetic deletion of GPR4 improves the neurotoxin-induced caspase-dependent mitochondrial apoptotic pathway, possibly through the modulation of PIP2 degradation-mediated calcium signalling. Therefore, GPR4 presents a potential therapeutic target for neurodegenerative disorders such as Parkinson’s disease.


Introduction
Parkinson's disease (PD) is a neurodegenerative disorder characterised by dopamine deficiency. An important pathological basis of PD is the loss of dopaminergic neurons due to apoptotic cell death in the substantia nigra (SN) of the brain [1]. An array of evidence suggests that reactive oxygen species (ROS)-induced oxidative stress is a major cause of the dopaminergic neuronal loss in the SN [2]. Mitochondria are key players in apoptosis during this neurodegeneration. In a cell undergoing apoptosis in a PD model. In particular, through our study of the overexpression and genetic deletion of GPR4, we investigate the role of GPR4 in the mitochondrial apoptosis pathway.

Expression of GPR4 Is Upregulated in Neurotoxin-Stimulated Apoptosis in SH-SY5Y Cells
To investigate the concentrations of MPP + and H 2 O 2 that precipitated a cell death of nearly 50% in the SH-SY5Y cells, 24 h serum-starved SH-SY5Y cells were treated with MPP + (0.25, 0.5, and 1 mM) or H 2 O 2 (50, 75, and 125 µM) for 24 h. As is shown in Figure 1A, when treated with the various concentrations of MPP + (1 mM; 56.511 ± 1.55%) and H 2 O 2 (125 µM; 53.12 ± 2.34%), half of the cell population in the MTT assay died. Furthermore, the mRNA and protein expressions of GPR4 in SH-SY5Y cells in both MPP + -(1 mM) and H 2 O 2 -(125 µM) treated serum-free media gradually increased in a time-dependent manner (3-24 h; Figure 1B).  (3,6,12,18, and 24 h) after stimulation with MPP + (1 mM) and H 2 O 2 (125 µM) in serum-free media. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and β-actin were utilised as the internal controls. Mean ± standard error of the mean (SEM; n = 3) was employed to express the data. Tukey's multiple comparison test was performed using a one-way analysis of variance (ANOVA). Each * p < 0.05 refers to the other sample concentrations compared with the control cells.

Knockout of GPR4 Protects SH-SY5Y Cells from Neurotoxin-Stimulated Apoptosis in SH-SY5Y Cells
To assess the effect of GPR4 overexpression and knockout on MPP + -induced apoptotic cell death, 24 h serum-starved SH-SY5Y cells were treated with MPP + (1 mM) for 24 h in serum-free media ( Figure 2). Following the MPP + (1 mM) treatment for 24 h in serum-free media, the number of SH-SY5Y viable cells decreased. Furthermore, the cells became rounded, displayed an increased neurite retraction, and were found to be loosely attached to the plate. Under bright-field optics, the GPR4-OE cells treated with MPP + (1 mM) exhibited less cell viability, with increased rounded cells, increased neurite retraction, and loose attachment to the surface. In contrast, the GPR4-KO cells treated with MPP + (1 mM) were more viable, strongly attached, neuronal shaped, and demonstrated less neuronal retraction than both the control and the GPR4-OE cells (Figure 2A). The morphology of SH-SY5Y GPR4-OE and GPR4-KO cells was observed through bright-field microscopy. (B) Cell viability was evaluated using an MTT assay. Mean ± SEM (n = 3) was employed to express the data. Tukey's multiple comparison test was performed using a one-way ANOVA. Each * p < 0.05 refers to the other sample concentrations compared with the control cells.
Cell viability was assessed with an MTT assay. The control SH-SY5Y cells presented a 55.67 ± 5.22% cell survival rate, whereas only 42.00 ± 2.01% of the GPR4-OE cells treated with MPP + (1 mM) survived. In contrast, the MPP + -treated GPR4-KO cells had a significantly higher cell survival rate (71.63 ± 3.54%), at 15% higher than for the MPP + -treated control SH-SY5Y cells and almost 30% higher than for the MPP + -treated GPR4-OE cells ( Figure 2B).

Knockout of GPR4 Decreases the Bax/Bcl-2 mRNA Ratio during Neurotoxin-Induced Apoptosis in SH-SY5Y Cells
To determine the role of GPR4 in both MPP + -(1 mM) and H 2 O 2 -(125 µM) stimulated apoptotic cell death, we investigated the expression levels of the Bcl-2 family proteins (Bax and Bcl-2). Many studies suggest that the Bcl-2 family plays a critical role in the mitochondrial apoptotic pathway. Bax enhances the release of cytochrome C from the space of the mitochondrial intermembrane to the cytosol, resulting in apoptosis. In contrast, Bcl-2 prevents apoptosis through its prevention of cytochrome C release, thereby maintaining mitochondrial cellular integrity [29,30]. In this study, an RT-PCR was employed to assess the mRNA expression levels of GPR4, Bax, and Bcl-2 in 24 h serum-starved SH-SY5Y cells treated with either MPP + (1 mM) or H 2 O 2 (125 µM; Figure 3A). A semi-quantification of the GPR4 mRNA and Bax/Bcl-2 mRNA expressions relative to GAPDH. This semi-quantification of the respective mRNA expression levels was performed on ImageJ software; GAPDH was utilised as an internal control. Mean ± SEM (n = 3) was employed to express the data. Tukey's multiple comparison test was performed using a one-way ANOVA. Each * p < 0.05 refers to the sample concentration compared with the same group of non-treated cells.
A semiquantitative analysis ( Figure 3B) of the RT-PCR bands highlighted a more than 4-fold increase in the expression of GPR4 in the GPR4-OE cells without any treatment, compared with the non-treated SH-SY5Y cells. In comparison with the non-treated SH-SY5Y cells, neurotoxins increased the expression of GPR4 in the GPR4-OE cells by 3-4-fold (MPP + , 3.65 ± 0.03; H 2 O 2 , 3.31 ± 0.17), whereas no significant difference in the GPR4 expression of the GPR4-OE cells (MPP + , 0.53 ± 0.003; H 2 O 2 , 0.04 ± 0.003) was observed.

Effect of a GPR4 Antagonist on the Cellular Morphology and GPR4 mRNA Expression of SH-SY5Y Cells
To investigate the effect of the pharmacological inhibition of GPR4, we adopted a GPR4 antagonist, NE52-QQ57. At the physiological pH, NE 52-QQ57 has been reported to effectively block the cAMP that is released by GPR4 activation (IC 50 26.8 nM) in HEK293 cells [3]. In this study, the impact of GPR4 antagonist, NE52-QQ57, on the cellular morphology and mRNA expression of GPR4 in MPP + -(1 mM) treated, 24 h serum-starved SH-SY5Y cells was assessed ( Figure 4). The SH-SY5Y cells were treated with NE52-QQ57 (1 µM) at pH 7.4 and incubated for 24 h in serum-free culture media, to evaluate the effect of NE52-QQ57 on cell morphology and viability. However, no morphological alteration or cellular toxicity was observed in the control SH-SY5Y, GPR4-OE, or GPR4-KO cells ( Figure 4A). This semi-quantification of the respective mRNA expression levels was performed on ImageJ software; GAPDH was utilised as an internal control. Mean ± SEM (n = 3) was employed to express the data. Tukey's multiple comparison test was performed using a one-way ANOVA. Each * p < 0.05 refers to the sample concentration compared with the same group of non-treated cells.
For an RT-PCR, RNA was isolated from the cells pre-treated with NE52-QQ57 (1 µM) at pH 7.4; this was incubated for an additional 24 h in serum-free media, with or without MPP + (1 mM) stimulation. The resulting levels of GPR4 mRNA expression illustrated that the NE52-QQ5 (100 nM) had effectively blocked the expression of GPR4 on the control SH-SY5Y and the GPR4-OE cells, similar to genetically GPR4-KO cells ( Figure 4B).

Knockout of GPR4 Decreases the Bax/Bcl-2 Protein Ratio and the Cleavage of PARP Expression in Neurotoxin-Stimulated SH-SY5Y Cells
MPP + -treated SH-SY5Y cells were assessed with immunoblotting to evaluate the effect of GPR4 antagonist, NE52-QQ57, on GPR4; the pro-apoptotic proteins, Bax and Bcl-2; and cleaved PARP expression ( Figure 5). The SH-SY5Y cells were pre-treated with NE52-QQ57 (100 nM) at pH 7.4. This was followed by 24 h incubation, or MPP + (1 mM) stimulation for 24 h in serum-free media. β-Actin was utilised as an internal control. Mean ± SEM (n = 3) was employed to express the data. Tukey's multiple comparison test was performed using a one-way ANOVA. Each * p < 0.05 refers to the sample concentration compared with the same group of non-treated cells.

Knockout of GPR4 Decreases the Caspase-3 Activity and Lowers the ROS Generation in Neurotoxin-Stimulated SH-SY5Y Cells
Caspases are important factors that trigger apoptosis. Caspase-3, in particular, is a crucial biomarker and executor of neuronal apoptosis [31]. To evaluate Caspase-3 activity, immunoblotting and a caspase activity assay were performed ( Figure 6). SH-SY5Y cells were treated with MPP + (1 mM) for 24 h in serum-free media, while cell lysates were analysed through a western blot and a caspase activity assay. MPP + -treated GPR4-OE cells demonstrated a significant increase in their cleaved Caspase-3 protein levels, whereas knockout of GPR4 prevented an MPP + stimulated increase in the level of cleaved Caspase-3 ( Figure 6A). β-Actin was utilised as an internal control. Mean ± SEM (n = 3) was employed to express the data. Tukey's multiple comparison test was performed using a one-way ANOVA. Each * p < 0.05 refers to the sample concentration compared with the same group of non-treated cells.
We evaluated the effects of GPR4 overexpression and knockout on H 2 O 2 -induced intracellular ROS generation in SH-SY5Y cells [19]. In our study, the protective effect of the knockout of GPR4 against H 2 O 2 resulted in lower intracellular ROS levels measured in the SH-SY5Y cells. DCFDA, a fluorescent dye that in the presence of ROS is oxidised to fluorescent DCF, was utilised for the detection of intracellular ROS levels. Treatment of the SH-SY5Y cells with H 2 O 2 (300 µM) for 1 h led to a marked increase in their intracellular ROS levels. In H 2 O 2 -treated GPR4-OE cells, the level of intracellular ROS generation was 1.56 ± 0.01 folds higher than that of the non-treated GPR4-OE cells, whereas the H 2 O 2 -treated GPR4-KO cells demonstrated a lower level of ROS generation (1.20 ± 0.04 folds) in comparison with both the H 2 O 2 -treated control SH-SY5Y (1.42 ± 0.001 folds) and the H 2 O 2 -treated GPR4-OE cells (1.56 ± 0.01 folds; Figure 6B).

Knockout of GPR4 Increases the Mitochondrial Membrane Potential (MMP) in Neurotoxin-Stimulated SH-SY5Y Cells
Excess intracellular ROS leads to swelling of the mitochondrial matrix and rupture of the outer membrane, which opens up the mitochondrial permeability transition pores (mPTPs). As a result, the mitochondrial membrane potential (MMP) is disrupted and mitochondrial oxidative stress-mediated apoptosis is initiated [4].
In this study, to measure the MMP, 24 h serum-starved SH-SY5Y cells were treated with MPP + (1 mM) for 24 h in serum-free culture media. MMP was then determined through a JC-10 fluorescence quantitative assay. Similarly, in a separate experiment in a 6-well plate, cells were utilised for JC-10 fluorescence microscopy to visualise the red and green fluorescence.
Aggregated JC-10 is an indicator of MMP; the greater the ratio of red/green fluorescence, the higher the level of MMP. In a quantitative JC-10 fluorescence microplate assay, the MPP + -treated SHSY-5Y cells presented a lower red/green fluorescence ratio (44.75 ± 0.82%) than the untreated SH-SY5Y cells ( Figure 7A). The MPP + -treated GPR4-OE cells, meanwhile, displayed the lowest red/green fluorescence ratio of all the samples (39.44 ± 0.39%), indicating a loss of MMP. In contrast, GPR4 knockout prevented the loss of MMP for the MPP + -treated GPR4-KO cells, as indicated by a higher red/green fluorescence ratio (65.44 ± 0.99%). Therefore, the MPP + -stimulated GPR4-KO cells demonstrated a higher level of MMP than either the MPP + -stimulated SH-SY5Y (44.75 ± 0.82%) or the GPR-OE (39.44 ± 0.39%) cells.
In this study, the aggregated JC-10 created red fluorescence in the polarised mitochondrial membrane. When the MMP collapsed in apoptotic cells, the JC-10 retained its monomeric form, which is characterised by green fluorescence. An increase in the red/green fluorescence intensity ratio indicated intact mitochondria. In a separate experiment, to visualise the JC-10 fluorescence dye in the MPP + -treated cells, JC-10 fluorescence microscopy was employed. In the control SHSY-5Y cells, both red and green fluorescence was observed, with a high level of red fluorescence and low level of green fluorescence. In contrast, the level of green fluorescence was higher and the red fluorescence remarkably lower in both the MPP + -treated SHSY-5Y and GPR4-OE cells, when compared with the control SHSY-5Y cells. In the MPP + -treated GPR4-KO cells, the red fluorescence was restored close to that of the control SHSY-5Y cells, while the level of green fluorescence was decreased ( Figure 7B). These results suggest that GPR4 knockout restores the MPP + -induced a loss of MMP in dopaminergic neurons.

Knockout of GPR4 Decreases the Intracellular Calcium in Neurotoxin-Stimulated SH-SY5Y Cells
Increases in intracellular Ca 2+ in association with MPP + -or H 2 O 2 -mediated apoptotic cell death have been previously reported [32]. Several studies have suggested that an increase in the intracellular Ca 2+ released from the ER store by the inositol trisphosphate receptor (IP 3 R) is directly responsible for mitochondrial Ca 2+ overload [33,34]. However, the exact mechanism by which MPP + or H 2 O 2 stimulation increases the intracellular calcium is not clearly understood.
Interestingly, several studies have demonstrated that H 2 O 2 -/MPP + -mediated mitochondrial oxidative stress is associated with an intracellular Ca 2+ spike, which increases the Bax/Bcl-2 ratio, the release of cytochrome C, mitochondrial depolarisation, and the Caspase-3 activity in neuronal cells [19,32].
Previous reports have suggested that many G protein coupled-receptors (GPCRs), such as GPR4, which releases G βγ and activates G i , are capable of Ca 2+ signalling. Few GPCRs, however, harness G βγ -dependent activation of PLC β to release ER-stored Ca 2+ into the cytoplasm through PIP 2 degradation [35,36]. In this study, the MPP + -treated GPR4-OE cells demonstrated an increased proteolytic degradation of PIP 2 , in comparison with the SH-SY5Y cells treated with MPP + . Contrastingly, the MPP + -stimulated GPR4-KO cells presented a particularly low degradation of PIP 2 compared with both the MPP + -stimulated SH-SY5Y and GPR4-OE cells ( Figure 8A). To evaluate whether GPR4 overexpression increased intracellular calcium through G βγ modulation of the PLC β -PIP 2 pathway, SH-SY5Y cells were treated with MPP + (1 mM) for 24 h in serum-free media. Cell lysates were analysed through western blotting to determine the degradation of PIP 2 . The SH-SY5Y cells were treated with H 2 O 2 (200 µM) for 2 h 30 min to determine their relative intracellular Ca 2+ , utilising a Fluo-4 AM calcium indicator in a fluorescence microplate assay. Similarly, in a separate 6-well plate, cells stained with a Fluo-4 AM calcium indicator were observed under a fluorescence microscope.
In a separate experiment to visualise intracellular Ca 2+ levels in the H 2 O 2 -treated cells, Fluo-4 AM fluorescence microscopy was employed. This round of microscopy demonstrated similar results to those obtained from the quantitative microplate assay. H 2 O 2 -treated GPR4-OE cells displayed the highest levels of green Fluo-4 AM fluorescence, while H 2 O 2 -treated GPR4-KO cells produced lower levels of green Fluo-4 AM fluorescence than both the H 2 O 2 -treated SH-SY5Y and GPR4-KO cells ( Figure 8C). Overall, these data suggest that the increase in intracellular calcium associated with H 2 O 2 -or MPP + -mediated mitochondrial oxidative stress is exaggerated by GPR4 overexpression, whereas GPR4 knockout prevents an increase in intracellular Ca 2+ through the decrease of PIP 2 degradation, and thus restricts the release of Ca 2+ from the ER by preventing the degradation of PIP 2 . Therefore, GPR4-PLC β -PIP 2 signalling may act as a key factor through which GPR4 increases intracellular calcium and potentiates mitochondrial oxidative stress-mediated apoptosis.

Discussion
In this study, we investigated the roles of GPR4 overexpression, pharmacological inhibition, and genetic knockout in the mitochondrial oxidative stress-induced apoptotic cell death that is associated with PD. Although many studies have reported the activation of GPR4 at the physiological pH range (7.0-7.4), overexpression of GPR4 showed relatively high GPR4 activity at neutral pH 7.4 [37]. In transiently GPR4-overexpressing HEK293 cells, GPR4 is inactive at pHs higher than 8.0, whereas it is highly active at the physiological pH, 7.4, and substantially less active at pHs down to 6.8 (plausible in the range of physiological acidification) [38]. The pH sensitivity of GPR4 has been reported to vary for different cells, though potentially due to the methods employed in different laboratories [25]. In the natively GPR4-expressing cell, HUVEC, pHs from 7.4 to 7.0 have been shown to result in a 1.5-fold activation of GPR4 [25]. In this study, we found an increase in GPR4 mRNA expression at pH 7.4 in both SH-SY5Y and stably GPR4-OE cells in serum-starved media (data not added). A very slight increase in the expression of GPR4 was observed at pH 6.4. Therefore, to maintain consistency, we conducted all the experiments at a pH~7.4. This was also the pH of the culture media that we employed.
Human-derived neuroblastoma SH-SY5Y cells are widely used in neuroscientific research as an in vitro model for the investigation of neuronal differentiation and neuroprotective events. Stimulation with several neurotoxins, such as MPP + , MPTP, rotenone, 6-OHDA, and H 2 O 2 , has been utilised to induce oxidative stress-mediated apoptotic death, thereby mimicking neurodegenerative diseases, including PD and aging [39][40][41]. To determine the final concentration of H 2 O 2 and MPP + , SH-SY5Y cells were treated with H 2 O 2 at different concentrations, ranging from 75 µM to 125 µM, for 24 h, as well as with MPP + , ranging from 250 µM to 1 mM. H 2 O 2 and MPP + both decreased the cell viability in a concentration-dependent manner, with optimum cytotoxicity being observed at concentrations of 125 µM for H 2 O 2 and 1 mM for MPP + ; these concentrations were selected for further experiments to determine their cytotoxicity in the serum-free SH-SY5Y cell line. In our study, GPR4 mRNA and protein expressions were increased in a time-dependent manner for 24 h in both MPP + -and H 2 O 2 -treated SH-SY5Y cells. Hence, GPR4 is directly linked with MPP + -and H 2 O 2 -induced apoptotic cell death.
Both the pro-apoptotic protein, Bax, and the anti-apoptotic protein, Bcl-2, are members of the Bcl-2 family and are directly involved in apoptotic cell death. The balance between these two proteins of the Bcl-2 family, or an increase in the Bax/Bcl-2 ratio, indicates the early phases of an apoptotic cascade [29,30]. Significant increases in ROS, or the Bax/Bcl-2 ratio, result in the collapse of the mitochondrial membrane potential, the release of cytochrome C, the activation of Caspase-3, the cleavage of PARP, and, subsequently, apoptotic cell death [6,7]. Both MPP + -and H 2 O 2 -induced apoptotic cell deaths bear the characteristic hallmarks of an increase in the Bax/Bcl-2 ratio, the release of cytochrome-C, and the activation of the proteolytic enzyme, Caspase-3, which cleaves PARP and induces apoptotic cell death [7,8]. In our study, the overexpression of GPR4 in SH-SY5Y cells significantly increased the effect of either MPP + or H 2 O 2 and increased the Bax/Bcl-2 ratio, as was seen in both the immunoblot and RT-PCR. As a result, this significantly increased the protein level of the cleaved Caspase-3, the Caspase-3 mediated cleavage of PARP, and the Caspase-3 activity. On the contrary, the CRISPR/Cas9 knockout of GPR4 was found to result in a lesser increase in the Bax/Bcl-2 mRNA and protein ratio in both MPP + -and H 2 O 2 -treated cells. Knockout of GPR4 was also shown to reduce the cleavage of PARP after MPP + treatment. NE52-QQ57, a selective antagonist of GPR4, demonstrated a similar level of the inhibition of GPR4 expression, as was determined through both our immunoblots and RT-PCR.
We further investigated the effect of GPR4 on mitochondrial oxidative stress-induced increases in intracellular ROS generation and MMP. Surprisingly, GPR4-OE was found to significantly increase tricellular ROS generation in SH-SY5Y cells, whereas GPR4-KO generated a lower level of intracellular ROS accumulation, after a high concentration of H 2 O 2 treatment. Similarly, through both a JC-10 assay and fluorescence microscopy, the knockout of GPR4 was found to decrease mitochondrial membrane depolarisation. In JC-10-tagged fluorescence microscopy, knockout of GPR4 was seen to prevent MPP + stimulated decrease red fluorescence and increase green fluorescence. The latter was highly increased in the case of GPR4-OE as membrane depolarisation occurs 24 h after MPP + treatment.
Besides mitochondrial dysfunction, abnormal protein aggregation and dysregulated Ca 2+ homeostasis are other factors that may be involved in the neurodegeneration observed in individuals with PD [42]. Recent findings suggest that increases in cytosolic Ca 2+ occur at both early and late stages of the apoptotic pathway. In both cases, ER Ca 2+ channels are linked with the release of Ca 2+ to the cytoplasm [33,34]. However, the exact mechanism by which intracellular Ca 2+ modulates mitochondrial oxidative stress-mediated apoptosis remains elusive. Many studies have suggested that MPP + -and H 2 O 2 -induced apoptosis are associated with an increase in intracellular calcium levels [43]. For example, Sing et al. (2016) demonstrated that the administration of Nimodipine, an L-type calcium channel blocker, protected from MPTP-induced dopaminergic neuronal death in an animal model of PD. More importantly, providing evidence for Nimodipine as a means to improve mitochondrial integrity and function. In the study, Nimodipine attenuated the MPTP-induced loss of tyrosine hydroxylase-positive dopaminergic neurons in the SN. It also improved mitochondrial oxygen consumption and inhibited ROS production, as well as improving mitochondrial integrity and function in striatal mitochondria [43]. These findings provide evidence in support of the notion that calcium signalling is linked with neurotoxin-induced mitochondrial dysfunction and neurodegeneration. GPR4 is a G s -coupled receptor that signals through adenylate cyclase and also via G proteins G 13 and G q/11 . GPR4 is well known for its ability to recognise phospholipase C β (PLC β ) as its canonical target [44]. G q class α subunits, or G βγ released by GPCR, activate Ca 2+ signalling through G βγ -dependent activation of PLC β . Upon activation, PLC β hydrolyses PIP 2 to generate IP 3 . IP 3 binds to the ER-resident IP 3 receptors, which act as Ca 2+ release channels to release ER-stored Ca 2+ into the cytoplasm [35]. In this study, overexpression of GPR4 significantly increased the intracellular calcium level in both MPP + -and H 2 O 2 -treated cells (MPP + data not given), whereas knockout generated very little change in the intracellular calcium. These findings were also observed in our study when employing the Fluo-4 AM indicator. To determine how GPR4 can modulate the intracellular calcium level, we investigated GPCR-mediated calcium signalling. We found that GPR4 knockout decreases the breakdown of PIP 2 , which is a critical step in Ca 2+ release from the ER to the cytoplasm. Therefore, decreased intracellular Ca 2+ may be responsible for GPR4-mediated neuroprotection against MPP +or H 2 O 2 -induced apoptotic cell death. Our study, for the first time, demonstrated that the knockout of GPR4 protects SH-SY5Y cells from both MPP + -and H 2 O 2 -stimulated mitochondrial apoptotic cell death, in association with a decrease in intracellular Ca 2+ .
In summary, our study suggests that overexpression of GPR4 potentiates neurotoxin-induced mitochondrial oxidative stress, whereas a knockout or pharmacological inhibition of GPR4 improves the neurotoxin-induced, caspase-dependent mitochondrial apoptosis of dopaminergic neuronal cells. This study has also found that GPR4 can increase intracellular Ca 2+ through the degradation of PIP 2 . Further investigation is required to determine how GPR4-mediated calcium signalling can mitigate the neuronal cell death seen in neurodegenerative disorders, including PD.

Cell Culture and Transfection
The human dopaminergic neuroblastoma SH-SY5Y cell line was acquired from the American Type Culture Collection (ATCC; Manassas, VA, USA). SH-SY5Y cells were cultured in DMEM/F12 with or without phenol red and HEPES, supplemented with 100 U/mL penicillin/streptomycin and 10% (v/v) inactivated foetal bovine serum. The SH-SY5Y cells were maintained in a 5% CO 2 and 95% humidified air incubator at 37 • C for the time indicated in the experiments. MPP + and H 2 O 2 were dissolved in three-times distilled water (3DW).
Both the GPR4-overexpression and knockout-silencing genes were designed to carry puromycin-resistance genes and were produced using the Lentivector Expression System (Applied Biological Materials Inc. (ABM), Canada). Stable single clones were selected following 3-5 weeks of puromycin treatment (1 µg/µL). GPR4 overexpression and knockout in the stably infected clones were assessed through RT-PCR and western blotting. A sequence analysis of the GPR4 insert was also employed.

Measurement of Cell Viability
The cytotoxicity of the MPP + -and H 2 O 2 -treated SH-SY5Y cells was measured with an MTT assay, involving the reduction of formazan crystals [41]. SH-SY5Y cells (2.2 × 10 4 cells/mL) were pre-treated in 24-well plates with NE 52-QQ57 (100 nM) and left in serum-free cell culture media for 1 h; this was followed by stimulation with or without MPP + (1 mM) or H 2 O 2 for 24 h. After MPP + or H 2 O 2 stimulation, the medium was replaced with 0.5 mg/mL MTT solution, before the plates were incubated for 3 h at 37 • C. The supernatant was carefully removed and the formazan crystals were dissolved in dimethyl sulfoxide (DMSO) by gentle shaking for 10 min. A microplate reader (Molecular device, Sunnyvale, CA, USA) was utilised to measure the absorbance at 550 nm.

Total RNA Isolation for RT-PCR
SH-SY5Y cells (2.2 × 10 4 cells/mL) were pre-treated in 60 mm cell culture dishes with NE 52-QQ57 (100 nM), then left in serum-free cell culture media for 1 h, followed by stimulation with or without MPP + (1 mM) or H 2 O 2 for 18 h, once again in serum-free media. TRIzol (Invitrogen; Burlington, ON, Canada) was employed to extract the total RNA from the cells. 2.5 µg total RNA from each group was reverse-transcribed using a first-strand cDNA synthesis kit (Invitrogen).

Immunoblot Analysis
SH-SY5Y cells (2.2 × 10 4 cells/mL) were pre-treated in 60 mm cell culture dishes with NE 52-QQ57 (100 nM) and left in serum-free media for 1 h. They were then stimulated with or without MPP + (1 mM) or H 2 O 2 for 24 h, again in a serum-free media. Next, the cells were washed two times with PBS and lysed for 10 min at 4 • C using an RIPA lysis buffer (with protease and phosphatase inhibitors). Supernatants were collected for further investigation after the cell lysates were centrifuged at 14,000 rpm, at 4 • C. The protein concentration of each sample was measured and normalised using a DC Protein Assay kit (Bio-Rad). Equal amounts of proteins (20-30 µg) were loaded and separated electrophoretically in 8, 10, and 12% sodium dodecyl sulphate-polyacrylamide gels; these were then transferred to polyvinylidene difluoride membranes (Millipore; Bedford, MA, USA). The membranes were incubated overnight at 4 • C, with corresponding primary antibodies, GPR4 (1:500) from Novus Biologicals Using ImageJ (NIH) software, the pixel intensity for each band in the photographs was measured and normalised to the band intensity of β-Actin, to quantify its relative expression.

Detection of Intracellular ROS
The ROS-sensitive fluorescent dye, 2 ,7 -dichlorofluorescein diacetate (DCFDA; Sigma-Aldrich), was utilised to measure the intracellular ROS levels. SH-SY5Y cells (2.2 × 10 4 cells/mL) were cultured in black 96-well plates in DMEM/F12 without phenol red. Then, 60-70% confluence cells were stimulated with H 2 O 2 (300 µM) for 1 h in serum-free media and then washed twice with PBS, followed by a 30 min incubation with DCFDA (10 µM) in PBS. The cells were then rinsed with PBS twice. Finally, 200 µL PBS was added, and fluorescence was measured using 485 nm excitation and 535 nm in a fluorescence microplate reader (Molecular Device; Sunnyvale, CA, USA).

Assessment of Caspase-3 Activity
Caspase-3 activity was measured using a Colorimetric Caspase-3 Assay Kit (Sigma-Aldrich; St. Louis, MO, USA), as described previously [41]. The reaction mixture (total volume, 200 µL) was distributed in 96-well plates and incubated at 37 • C for 90 min. Absorbance values were measured at wavelengths of 405 nm in a Tecan Microplate Reader (Meilen; Zurich, Switzerland).

Assessment of Mitochondrial Membrane Potential (MMP)
A JC-10-based Mitochondrial Membrane Potential Assay Kit (Abcam; Cambridge, United Kingdom) was employed to assess MMP, while a fluorescence microscope was utilised to visualise the JC-10 staining, according to the manufacturer's instructions. In brief, SH-SY5Y cells (2.2 × 10 4 cells/mL) were cultured in black, 96-well plates for quantification and in 6-well plates for imaging in DMEM/F12 without phenol red. Then, 60-70% confluence cells were stimulated with MPP + (1 mM) for 24 h in serum-free media. The cells were incubated with a JC-10 dye loading solution at 37 • C for 1 h and protected from the light. For the 96-well plates, their fluorescence intensities (λ ex = 490/λ em = 525 nm) and (λ ex = 540/λ em = 590 nm) the red-green fluorescence ratios were measured using a fluorescence microplate reader (Molecular Device; Sunnyvale, CA, USA), while confocal images were acquired with a Nikon Eclipse Ts2-FL diascopic and epi-fluorescence illumination microscope.

Detection of Intracellular Calcium
Intracellular calcium was assessed with a Fluo-4 AM dye (Abcam; Cambridge, United Kingdom) and confocal microscopy, following the manufacturer's instructions. Fluo-4 AM was diluted in DMSO containing 2 mM probenecid and 0.02% pluronic F-127. In brief, the SH-SY5Y cells (2.2 × 10 4 cells/mL) were cultured in black, 96-well plates for quantification and in 6-well plates for imaging in DMEM/F12 without phenol red. Then, 60-70% confluence cells were stimulated with H 2 O 2 (200 µM) for 2 h 30 min in serum-free media. The cells were washed with PBS containing probenecid (2 mM) at room temperature. The cells were incubated with the Fluo-4 AM (2 µM) dye loading solution at 37 • C for 30 min and protected from light, then washed with PBS containing probenecid (2 mM) at room temperature for 30 min. For the 96-well plates, the fluorescence intensities (λ ex = 488/λ em = 515 nm) were measured using a fluorescence microplate reader (Molecular Device; Sunnyvale, CA, USA), and confocal images were acquired with a Nikon Eclipse Ts2-FL diascopic and epi-fluorescence illumination microscope.

Statistical Analyses
Statistical analyses were performed using GraphPad Prism software, version 5 (GraphPad, La Jolla, CA, USA). Data are expressed as means ± standard error (SEM) of at least three independent experiments. One-way analysis of variance (ANOVA) followed by Tukey's post hoc analysis were performed to determine the significant differences between the groups. p-values < 0.05 were considered statistically significant.  T-Cell Death-Associated Gene 8 OGR1 The Ovarian Cancer G Protein-Coupled Receptor 1