Abstract
Parkinson’s disease (PD) is chronic neurodegenerative movement disorder and affects approximately 1% of people older than 55 years. It is due to a progression of the death of dopaminergic neurons in the brain. The most effective current treatment is dopamine replacement therapy. It involves drugs such as L-3,4-dihydroxyphenylalanine (L-DOPA) and dopamine receptor agonists to restore dopamine levels in the brain and reduce the motor symptoms of PD. However, prolonged use of L-DOPA leads to unwanted side effects, such as abnormal involuntary movement. Hence the need for a new and more effective approach to treat PD. Recent research has revealed the therapeutic potential of targeting G protein-coupled receptors (GPCRs). These receptors have two binding sites, an orthosteric binding site and an allosteric binding site. Metabotropic glutamate receptor 4 (mGluR4) is a type of GPCR and is currently under research as a potential target for treating the motor symptoms of PD. These studies looked at the therapeutic effects of agonists and allosteric modulators of mGluR4 on rat models which mimicked the symptoms of PD. Currently, there has been no preclinical and clinical studies for allosteric modulators of mGluR4 for the treatment of PD. However, positive allosteric modulators of mGluR4 have great potential in the treatment of PD. This dissertation will evaluate the previous and current research for potentially targeting mGluR4 in treating the motor symptoms of PD and conclude the future prospects of this type of treatment in PD. In conclusion, allosteric modulators of mGluR4 may be useful in treating the motor symptoms of PD as they have shown to have great therapeutic potential.
Introduction
G protein-coupled receptors (GPCRs) have 7-transmembrane spanning domains, an intracellular C-terminal and an extracellular N-terminal (Figure 1) (Challiss, 2016). These receptors are activated by an endogenous ligand. This causes a conformational change, where the G protein subunits split, which leads to intracellular signalling within the cell, resulting in important physiological responses in the immune system, neurotransmission within the brain and homeostasis (Gentry et al., 2015). Around half of current pharmacological therapies are using GPCRs as targets. Most drugs in pharmaceutics target the orthosteric binding site (Wang et al., 2009). Traditionally, drugs were made to bind at the orthosteric binding site of the GPCR which would either copy (agonist) or stop (antagonist) the activity of the endogenous ligand. This sometimes led to unwanted side effects and lack of selectivity for the receptor subtype due to the orthosteric binding site being highly conserved between GPCRs (Christopoulos et al., 2004).
Allosteric modulators target at a site away from the orthosteric site known as the allosteric binding site, which is less conserved. These allosteric ligands can either enhance (positive allosteric modulation) or block (negative allosteric modulation) the effects of the endogenous ligand on its receptor (Nickols and Conn, 2014). The orthosteric and allosteric binding sites are conformationally linked which means when the allosteric modulator binds to the GPCR, the receptor changes shape and modifies the signalling or binding actions of the orthosteric ligand. The use of allosteric modulators over orthosteric ligands has become more popular as they have more advantages. One advantage is that allosteric modulators display higher receptor subtype selectivity than orthosteric ligands as the allosteric binding site is less conserved between GPCRs. Another advantage by targeting the allosteric binding site would be that there is less likely to be unwanted side effects as allosteric modulators will have negative cooperativity, which means their effect will be limited as the receptor’s affinity decreases for the allosteric modulator (Christopoulos et al., 2004).
Two allosteric modulators have been approved for clinical use so far. They show different effects on GPCR function by allosteric modulation. The first approved drug is Cinacalcet. It’s a positive allosteric modulator of the calcium sensing receptor (CaSR). Cinacalcet is used to treat secondary hyperparathyroidism and parathyroid carcinoma. It increases the affinity for calcium of CaSR, and blocks the parathyroid hormone from secreting. The second approved drug is Maraviroc. It’s a non-competitive allosteric antagonist of the chemokine receptor, CCR5. It has been approved as salvage therapy, which means it is a last resort for advanced cases of HIV disease. CCR5 acts as a cell surface receptor for the HIV virus and allows binding of Maraviroc to cause the receptor to change shape. This results in blocking the HIV virus from binding, which leads to a reduction in HIV infection and producing a significant decrease in systemic viral load. However, Cinacalcet and Maraviroc are unusual allosteric modulators as they target atypical GPCRs, a calcium sensing receptor and a chemokine receptor (Wang et al., 2009).
The recent advancements in the development of allosteric modulators has shown evidence for their potential use in many central nervous system (CNS) disorders. These include neurodegenerative diseases for example Alzheimer’s disease, Parkinson’s disease, and Huntington’s disease in addition to psychiatric and neurobehavioral diseases for example anxiety, schizophrenia, and addiction (Nickols and Conn, 2014).
One focus of treating CNS disorders has been to target metabotropic glutamate receptors (mGluRs), which are a type of GPCR, as they bind to glutamate which is the major excitatory neurotransmitter in the CNS. These receptors are able to alter cell excitability and transmission of neurotransmitters across synapses. They are found in neurons and glial cells, including astrocytes, oligodendrocytes, and microglia. Group I mGluRs, mGluR1 and mGluR5, are found postsynaptically in neurons and couple to Gq/11 proteins, which regulate the IP3 signalling pathway for Ca2+ release. mGluR2,3,4,7,8 are mostly found in presynaptic terminals of synapses and modulate the release of neurotransmitters (Figure 2). Group III mGluRs (mGluR4,6,7,8) and group II mGluRs (mGluR2,3) couple to adenylyl cyclase through Gi/o proteins (Nickols and Conn, 2014).
An example of a CNS disorder that is currently under research to use allosteric modulators as treatment is Parkinson’s disease (PD). Parkinson’s disease is a chronic neurodegenerative movement disorder. It is due to the death of dopamine neurons in the substantia nigra pars compacta. Some of the motor symptoms include bradykinesia, postural instability, and muscular rigidity. The non-motor symptoms include cognitive impairment, sleep disturbance, and olfactory dysfunction (Kalia and Lang, 2015). PD affects approximately 1% of people older than 55 years (Nickols and Conn, 2014). Current treatments include dopamine-replacement therapy and deep-brain stimulation. The drugs used in the dopamine-replacement therapies such as L-3,4-dihydroxyphenylalanine (L-DOPA) and dopamine receptor agonists are initially effective in treating the motor symptoms of PD (Kalia and Lang, 2015). However, prolonged use of these drugs lead to reduced efficacy and undesirable side effects such as dyskinesia, cognitive impairment, and on/off oscillations (Stacy and Galbreath, 2008). Hence, the need for an alternative therapeutic method such as allosteric modulators in order to treat PD.
Chapter 1: The Basal Ganglia Motor Circuit
In order to understand the pathology of PD and how potential treatments might work, it is crucial to understand the basal ganglia circuit which is responsible for muscle control.
The basal ganglia (BG) circuit is an assembly of subcortical nuclei which are responsible for the control of motor behaviour. The primary output nuclei of the BG are the substantia nigra pars reticulata (SNr) and the internal globus pallidus (GPi) and the primary input nucleus of the BG is the striatum. Many studies suggest that the globus pallidus (GP) has an important part in the pathophysiology of PD. The synapse between the medium spiny neurons of the striatum and the external globus pallidus (GPe) neurons is known as the striatopallidal synapse. It is the first synapse in the indirect pathway of the BG (Figure 3) (Valenti et al., 2003). The function of the GP in normal muscle control has been determined by primate studies which show that the speed of firing by the GPe neurons is linked to movement of the muscle (Mink and Thach, 1991). A better understanding of the BG motor circuit allows for research into mechanisms that can change the BG function without using dopamine replacement.
Due to the death of the dopaminergic neurons in the substantia nigra pars compacta (SNc), the striatum does not have enough dopamine (Figure 3). In the indirect pathway, an abnormally excessive inhibitory GABAergic signal is sent to the external globus pallidus (GPe), which in turn sends a smaller than normal inhibitory signal to the subthalamic nucleus (STN). This leads to a much bigger excitatory glutamatergic signal to the internal globus pallidus (GPi) and substantia nigra pars reticulata (SNr). A large inhibitory signal is sent to the thalamus (Thal) which gives a smaller excitatory signal to the motor cortex). This imbalance of inhibition and excitation in the basal ganglia motor circuit is why patients with PD have trouble with muscle movement. In the direct pathway, the reduced amount of inhibitory signal from the striatum to the GPi and SNr also plays a part in the motor symptoms of PD. The current treatment drug, L-DOPA, enters the striatum to restore dopamine levels however, there are side effects. An alternative therapeutic approach would be to reduce the excessive inhibitory GABAergic transmission at the striatopallidal synapse (Marino et al., 2003).
The striatopallidal synapse is between the striatum and GP. It is an inhibitory synapse which releases gamma-aminobutyric acid (GABA) as its neurotransmitter. The striatum is excited by the motor cortex (Figure 4). This allows glutamate to bind to the ionotropic glutamate receptors the postsynaptic terminal of the medium spiny neuron (MSN) and generate an action potential. Depolarisation of the membrane opens voltage-gated Ca2+ channels on the neuron and Ca2+ ions enter the cell. Ca2+ initiates a signalling cascade and causes the vesicles of GABA to fuse with the membrane, releasing GABA into the synaptic cleft. GABA binds to the GABAA receptors on the postsynaptic terminal of the GPe and activates them. This results in the GPe neuron becoming hyperpolarised and inhibiting an action potential so there is less neurotransmission. This leads to less GABA being released from the postsynaptic terminal of the GPe neurons. This causes an inhibitory effect on the STN, so less movement takes place. In PD, due to the degeneration of the SNc neurons, there is no dopamine in the striatum which would have had an inhibitory effect on neurotransmission. This means more GABA is released from the MSN, which results in reduced GABA transmission from the GPe to the STN. This leads to the movement disorder in PD patients. mGluR4 is presynaptically localised at the striatopallidal synapse. This was determined by Bradley et al. in 1999 by using immunohistochemistry. When mGluR4 receptors are activated by endogenous glutamate, the G subunit of the Gi/o protein inhibits the voltage-gated Ca2+ channels so there is no influx of Ca2+ ions. This means less GABA is released from the striatum so fewer GABAA receptors are activated. This leads to an increased amount of GABA being released from the GPe to the STN so movement is restored (Nickols and Conn, 2014).
Therefore, a potential therapeutic method to treat the motor deficits of PD would be to reduce the excessive inhibitory GABAergic transmission at the striatopallidal by activating mGluR4.
Previous studies in PD patients and rat models have proposed that reducing the overactive pathway between the substantia nigra and the striatum may have therapeutic potential in treating the motor deficits of PD. mGluR4 inhibits the neurotransmission of GABA across the synapse, which has increased activity in PD patients due to reduced levels of dopamine in the brain. Therefore, it has been predicted that mGluR4 positive allosteric modulators (PAMs) will be able to reduce this increased inhibition at the striatopallidal synapse in the indirect pathway of the BG motor circuit (Nickols and Conn, 2014). Activating mGluR4 with the group III selective agonist L-AP4 reduced the neurotransmission of GABA by inhibiting its release from the presynaptic terminals of the striatopallidal synapse (Valenti et al, 2003). In mGluR4 knockout mice, this effect didn’t take place which highlights the important function of mGluR4 in controlling neurotransmission in the brain. It has been shown that mGluR4 agonists are able to reduce motor deficits in preclinical rat models of PD. It has also been shown that mGluR4-selective PAMs have increased efficacy in treating the motor symptoms in many rat models of PD. VU0364770 and ADX88178 are mGluR4 PAMs which can enhance the antiparkinsonian effects of L-DOPA, thus showing potential as L-DOPA sparing agents. Some earlier research shows that mGluR4 PAMs may also have disease modifying effects by reducing the rate of death of the dopamine neurons. PHCCC, also a mGluR4 PAM, combined with the group III agonist ACPT-1 was able to reduce neuropathic pain in rat models, which may be useful in treating the non-motor symptoms of PD (Nickols and Conn, 2014).
Therefore, mGluR4 positive allosteric modulators have the potential to treat both motor and non-motor symptoms of PD. However, mGluR4 PAMS have not been clinically tested yet but there has been much advancement in recent studies by many researchers and pharmaceutical companies. These studies will lead to clinical testing of the potential of mGluR4 PAMs treating PD.
Chapter 2: L-AP4
This chapter will focus on a drug which is an agonist that binds at the orthosteric site of mGluR4 as a potential treatment for the motor symptoms of PD. This drug was studied by Valenti et al. in 2003. Their research involved many experimental methods such as electrophysiology and rodent models of PD.
2.1 L-AP4 reduces inhibition at striatopallidal synapse
In 2003, Valenti et al. studied an mGluR4-selective agonist called L-(+)-2-amino-4-phosphonobutyric acid (L-AP4) as a potential drug to improve the motor deficits of PD. An electrophysiology experiment was carried out to observe the effect of this agonist at the striatopallidal synapse. Whole-cell patch-clamp recordings were taken of the globus pallidus (GP) neurons in rat coronal brain slices. The striatum was electrically stimulated in order to produce GABAA-mediated inhibitory postsynaptic currents (IPSCs). IPSCs were evoked in the presence of blockers of AMPA (-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid), NMDA (N-Methyl-D-aspartic acid) and GABAB receptors. They were induced by single pulses in the range of 30 to 90 mV every 30-60 seconds with a holding potential of -50 mV. When 3 M of L-AP4 was added, the IPSC was inhibited until L-AP4 was washed out and its inhibitory effect reversed (Figure 5). L-AP4 gave a maximal inhibition of 85.2% (pre-drug 112.4 ± 29.9 pA; mean ± SEM; 3 M L-AP4 16 ± 4.6 pA; p = 0.004; paired t test). This means that L-AP4 is able to reduce inhibition at the striatopallidal synapse of the basal ganglia motor circuit (Valenti et al., 2003).
2.2 L-AP4 acts at mGluR4
The next thing that Valenti et al. wanted to find was at which metabotropic glutamate receptor (mGluR) L-AP4 is acting on at the striatopallidal synapse. Within the indirect pathway of the basal ganglia (BG) motor circuit, L-AP4 has low potency at other synapses. This means that mGluR7 is responsible for the actions at these other synapses as this receptor needs higher concentrations of L-AP4 to inhibit IPSC when compared to mGluR4, mGluR6 and mGluR8. Therefore, the effects on transmission produced by low doses of L-AP4 at the striatopallidal synapse must be regulated by mGluR4, mGluR6 or mGluR8. However, mGluR6 is expressed at very low non-significant levels within the central nervous system (Nakajima et al., 1993). Hence, mGluR4 or mGluR8 are responsible for the effects on transmission at the striatopallidal synapse. (S)-3,4-dicarboxyphenylglycine (DCPG) is a mGluR8-selective agonist and it was used to check if activating mGluR8 had any effect on transmission at the striatopallidal synapse. When 300 nM of DCPG was applied, there was no effect on striatopallidal transmission (pre-drug 63.1 ±7.3 pA; 300 nM DCPG 59.9 ± 9 pA; p = 0.3; paired t test). Therefore, this means that mGluR8 has no involvement in modifying transmission at the striatopallidal synapse. Previous studies have determined that mGluR4 is localised at the presynaptical terminals of the striatum by immunohistochemistry (Bradley et al., 1999). Thus, it was predicted that activating mGluR4 at the Schaffer collateral-CA1 synapse would have no effect on transmission. An electrophysiology experiment was used to test this prediction so whole-cell patch-clamp recordings of GP neurons in coronal rat brain slices were obtained. There was no change in transmission at the Schaffer collateral-CA1 synapse when L-AP4 was applied which means these results are consistent with where mGluR4 is localised in the brain (pre-drug 106.4 ± 27.5 pA; 3 M L-AP4 88.2 ± 32.4 pA; mean ± SEM; p>0.05; paired t test; n = 6). With the L-AP4 potency at different group III mGluRs and the lack of effect of DCPG, these findings show that mGluR4 is the receptor responsible for modifying transmission at the striatopallidal synapse when activated by a low concentration of L-AP4. This was further confirmed by studies performed in mGluR4 knock-out mice by Valenti et al. Electrophysiology studies in brain slices made from control 129X1/SvJ mice showed a significant inhibition of striatopallidal transmission when 3 M of L-AP4 was applied (Figure 6). However, this effect was not seen in the electrophysiology studies with brain slices prepared from mGluR4 knock-out mice. This further suggests that activation of mGluR4 is responsible for the modulation of transmission at the striatopallidal synapse (Valenti et al., 2003).
2.3 L-AP4 has a presynaptic mechanism of action
Valenti et al. then predicted that L-AP4 had a presynaptic mechanism of action as mGluR4 is located presynaptically at the striatopallidal synapse (Bradley et al., 1999). To find evidence of this hypothesis, the effect of L-AP4 was observed on paired-pulse plasticity and on tetrodotoxin (TTX)-resistant mIPSCs (miniature inhibitory postsynaptic currents). Pairs of IPSCs were evoked by two stimuli of equal strength and duration, separated by an interval of 50-100 msec. The second IPSC was enhanced compared to the first with these conditions. For recording mIPSCs, the potassium gluconate in the internal solution was replaced with potassium chloride to reverse the chloride gradient, which means more precise measurements of the mIPSCs can be taken. Recording of mIPSCs was done using the same mixture of receptor blockers used in the previous IPSC studies described above but with the addition of 1 M TTX (Valenti et al., 2003). TTX blocks voltage-gated Na+ channels to block the influx of Na+ ions, which stops the depolarisation of an action potential (Bane et al., 2014). The holding potential was -60mV for this study. Correlating with a presynaptic mechanism of action found by Zucker and Regehr in 2002, there was an increase in the paired pulse ratio (second IPSC/first IPSC) when 3-10 M of L-AP4 was applied (pre-drug 1.2 ± 0.2; L-AP4 1.9 ± 0.2; mean ± SEM; p<0.05; paired t test; n = 7). When 1 M of TTX was added, recordings of inward mIPSCs were taken from GP neurons. When 3 M of L-AP4 was applied, the frequency of mIPSCs reduced significantly with no effect on the amplitude. This demonstrates a presynaptic mechanism of action. With the findings combined, the data indicate that there is a presynaptic mechanism of action when L-AP4 activates mGluR4 and inhibits the transmission at the striatopallidal synapse (Valent et al., 2003).
2.4 L-AP4 pharmacology is not affected by reserpinisation
Valenti et al. then wanted to see if the pharmacology of the mGluR4 agonist, L-AP4, would be affected when dopamine levels were reduced by reserpine. Earlier studies showed that when reserpine depleted dopamine levels at the STN-SNr synapse, it reduced the ability of the group II mGluR agonists to reduce excitatory signals and the ability of the group III mGluR agonists to reduce inhibitory signals (Wittmann et al., 2002). Therefore, if depleted dopamine levels decrease the ability of L-AP4 acting on mGluR4 to reduce inhibitory striatopallidal transmission, this can affect any potential antiparkinsonian effects of this agonist. Using an overnight catecholamine depletion model, Valenti et al. investigated if L-AP4 was still able to reduce inhibitory striatopallidal transmission on animals with depleted dopamine levels. The rats were injected subcutaneously with reserpine (5 mg/kg) to deplete dopamine levels 18-24 hours before brain slices were prepared. This dose of reserpine induced catalepsy in the rats. When 3 M L-AP4 was applied, a significant reduction in striatopallidal transmission was seen in brain slices prepared from the reserpinised rats with an inhibition of 48.8% ± 5.8% (mean ± SEM). However, this effect was found to be smaller than in brain slices from normal rats which had an inhibition of 70.5% ± 9.8% (mean ± SEM; t test; p<0.01; n = 4-7). Overall the inhibition in striatopallidal transmission from reserpinised rat brain slices by L-AP4 was nevertheless a significant result (pre-drug 172.8 ± 53 pA; 3 M L-AP4 97.3 ± 36.9 pA; mean ± SEM; paired t test; p<0.01; n = 4) (Valenti et al., 2003).
2.5 L-AP4 produces antiparkinsonian effects in rodent models
Pharmacologically, L-AP4 is able to reduce inhibitory striatopallidal transmission by activating mGluR4. The next step was to look at if the agonist had any antiparkinsonian effect on behaviour in dopamine-depleted rats. Valenti et al. then carried out in vivo studies to see if L-AP4 is able to reverse motor deficits by using acute and chronic rodent models of PD. In the acute behavioural study, rats were subcutaneously injected with reserpine (5 mg/kg) to deplete dopamine levels so to induce akinesia in the rats. Motor activity was measured by the number by breaks in infrared beams every 30 mins when rats were placed in photocell cages. When an intracerebroventricular injection of L-AP4 (50 nmol) was given to the rats treated with reserpine, motor activity significantly increased when compared to the vehicle-treated rats (Figure 7A). These results show that activating mGluR4 with L-AP4 can produce a significant antiparkinsonian effect in this acute rodent model of PD. As PD is a chronic disease which is linked to major changes in plasticity, the next rodent model involved 6-hydroxydopamine (6-OHDA). 6-OHDA was injected unilaterally into the medial forebrain bundle of the rat brain. This depleted dopamine levels in one side of the rat brains, causing forelimb asymmetry in the rats. The aim was to see if activating mGluR4 could improve this movement disorder. The cylinder test was used to see whether the rat used the affected (contralateral) or non-affected (ipsilateral) forelimb as it lands on the base of the cylinder after rearing. Asymmetry scores were worked out by % ipsilateral paw – (% contralateral paw + % both paws). It was hypothesised that an anti-PD drug would reduce the forelimb asymmetry score by a significant amount. Both L-AP4 (100 nmol, i.c.v.) and L-DOPA (6 mg/kg, i.p.) reduced the forelimb asymmetry score by a significant amount when compared to the pre-treatment groups (Figure 7B). L-AP4 was found to have similar efficacy as the current anti-PD drug L-DOPA. With the results combined, the data show that activating mGluR4 can have a significant antiparkinsonian effect in both acute and chronic rodent models of PD (Valenti et al., 2003).
Activation of mGluR4 with L-AP4 reduces the inhibitory transmission at the striatopallidal synapse and shows improvement in motor deficits in rodent models of PD. Therefore, activation of group III mGluRs are a potential target for treating patients with PD. This treatment may be a more effective palliative therapy than dopamine replacement therapy.
Chapter 3: PHCCC
This chapter will focus on a drug which is an allosteric modulator of mGluR4 as a potential treatment for the motor symptoms of PD and why it is better drug than the mGluR4 agonist L-AP4.
3.1 PHCCC potentiates response of mGluR4 to L-AP4 in FLIPR assay
N-phenyl-7-(hydroxylimino)cyclopropa[b]-chromen-1a-carboxamide (PHCCC) is a positive allosteric modulator of mGluR4, but it is also a group I mGluR antagonist. In 2003, Marino et al. wanted to see if allosteric modulation of mGluR4 could be a potential approach to treating PD. First, they wanted to see what effect PHCCC has on the mGluR4 endogenous ligand, glutamate. They used Chinese hamster ovary cells which expressed recombinant human mGluR4 coupled to the chimeric G protein Gqi. This experimental set-up allows transfected receptors which normally inhibit the cAMP pathway, to couple to Gq signal transduction and mobilise Ca2+ (O’Brien et al., 2003). The cells were put in a fluorometric imaging plate reader (FLIPR) to measure their ability to mobilise Ca2+ in response to glutamate. Addition of glutamate (2 M) in the presence of PHCCC (10 M) produces a large potentiation when compared to glutamate on its own (Figures 8 and 9). In the absence of glutamate, PHCCC had no effect on the activity of mGluR4 (Marino et al., 2003).
Cell lines expressing different mGluR subtypes were put into FLIPR to measure their ability to mobilise Ca2+ in response to either glutamate or L-AP4. mGluR1b, mGluR4, mGluR5a and mGluR8 were expressed in Chinese hamster ovary cells. mGluR7 was expressed in human embryonic kidney cells. mGluR4 was co-expressed with chimeric G protein Gqi, and mGluR7 and mGluR8 were co-expressed with G to allow use in the Ca2+-sensitive fluorescence assay. The cells were preincubated with PHCCC or vehicle for 5 mins before adding the agonist. PHCCC (10 M) potentiated the response of mGluR4 to L-AP4 but did not activate or potentiate responses to any other mGluR subtype observed (Figure 9) (Marino et al., 2003).
These findings from the FLIPR assay show that PHCCC is able to potentiate the response of mGluR4 to its agonist L-AP4 in vitro. This experiment is the first piece of evidence that an allosteric modulator could potentially be a better drug than an agonist alone as its effects on the receptor are greater.
3.2 Electrophysiology shows PHCCC potentiates effects of L-AP4
Marino et al. next carried out electrophysiology experiments to see the potentiation effect of PHCCC on L-AP4 at the striatopallidal synapse of the basal ganglia (BG) motor circuit. First, they wanted to test for the ability of PHCCC to potentiate the effects of a low dose of L-AP4 on striatopallidal transmission. So, whole-cell patch-clamp recordings were taken of globus pallidus (GP) neurons in rat midbrain slices. The striatum was electrically stimulated to evoke GABAA-mediated inhibitory postsynaptic currents (IPSCs). IPSCs were evoked in the presence of blockers of AMPA, NMDA and GABAB receptors. They were induced by single pulses in the range of 30 to 90 mV every 30-60 seconds with a holding potential of -50 mV. Application of 1 M L-AP4 produced a small yet significant inhibition of striatopallidal transmission, as seen by the small reduction in the amplitude of IPSCs (Figures 10A and B). Application of vehicle (1% DMSO) or 30 M PHCCC alone had no effect on transmission at the striatopallidal synapse. However, when co-applying 30 M PHCCC and 1 M L-AP4, there was a more significant inhibition of the IPSCs produced, which means a more significant inhibition of striatopallidal transmission. The effect of L-AP4 in the presence of the PHCCC was significantly higher than the effect of L-AP4 alone (Figure 10C). Therefore, the use of allosteric modulators of group III mGluRs combined with agonists may be a better treatment for PD than using agonists alone (Marino et al., 2003).
The next electrophysiology experiment that Marino et al. carried out was to see if PHCCC was selective for mGluR4 in native brain slice studies as well as in recombinant studies.
Two synapses in the hippocampus that have been studied before were used and whole-cell patch-clamp recordings were obtained from them. These synapses are able to be modified by activating other group III mGluRs than mGluR4. The two synapses used in this experiment are the Schaffer collateral-CA1 synapse and the lateral perforant path-dentate gyrus synapse. Field excitatory postsynaptic potential recordings were obtained from these synapses. Activating a group III mGluR at the Schaffer collateral-CA1 synapse leads to inhibition of transmission acting from the presynaptic terminals as determined by earlier studies (Gereau and Conn, 1995). Due to the low potency of L-AP4 and the high expression levels of mGluR7 at this synapse, this reduction in transmission mediated by L-AP4 is most likely regulated by mGluR7. Earlier studies also show that activating mGluR8 at the lateral perforant path-dentate gyrus synapse reduces transmission (Macek et al., 1996). PHCCC did not enhance the inhibition of transmission mediated by L-AP4 at both synapses. There was also a small non-significant reduction in the effect of L-AP4 (Figure 11). These results are consistent with the previous recombinant studies. With the electrophysiology experimental results combined together, the data show that PHCCC is a selective allosteric modulator of mGluR4 in this in vitro study (Marino et al., 2003).
3.3 PHCCC produces antiparkinsonian effect in akinesia rodent model
Earlier studies showed that activating presynaptically localised mGluR4 reduced striatopallidal transmission, restoring balance within the indirect pathway of the BG motor circuit (Valenti et al., 2003). This underlies the antiparkinsonian effects of L-AP4 found in many rodent models of PD. Therefore, the ability of PHCCC to treat the motor disorder in rats with reserpine-induced akinesia was investigated. Rats were subcutaneously injected with reserpine (5 mg/kg; pre-drug) to deplete dopamine levels so rats were unable to initiate movement. 30 minutes later, rats received one intracerebroventricular injection (0.5 l/min) of PHCCC, CPCCOEt or vehicle control. CPCCOEt is a closely related analogue of PHCCC, but has no allosteric modulation activity on mGluR4 and has similar mGluR1 activity as PHCCC hence this compound was a suitable control for this in vivo study. Motor activity was recorded for both pre- and post-treatment groups. The activity of the rats was measured by the number by breaks in infrared beams durng 30 mins when rats were placed in photocell cages. Motor activity significantly increased in rats treated with PHCCC (Figure 12). However, CPCCOEt and vehicle had no effect significant effect on motor activity. These findings combined show that PHCCC has antiparkinsonian actions by reversing motor deficits in this reserpine-induced akinesia rodent model of PD (Marino et al., 2003).
To conclude, PHCCC potentiates the effects of L-AP4 on mGluR4 at the striatopallidal synapse and produces antiparkinsonian effects in rodent models of PD. Therefore, this shows that allosteric modulators have more therapeutic potential than agonists alone. These studies by Marino et al. further confirm that activating mGluR4 provides a potential therapeutic approach for treating the motor symptoms of PD.
Chapter 4: VU0364770
PHCCC was one of the first mGluR4 allosteric modulators to be discovered as a potential drug for the treatment of PD. However, PHCCC lacked many properties for further preclinical and clinical research. PHCCC had low potency so higher concentrations of this compound was needed for an effect. In addition, this compound did not have very good aqueous solubility. PHCCC also had antagonist activity at mGluR4 and mGluR1 with potencies of like values (Niswender et al., 2008). PHCCC cannot be used systemically as the drug requires to be injected into the brain directly to bypass the blood-brain barrier. Therefore, this makes PHCCC an impractical drug for the treatment of PD. Hence, there is a need for an allosteric modulator with better potency as a potential treatment for PD. In 2012, Jones et al. studied a more potent positive allosteric modulator of mGluR4, VU0364770, which has therapeutic potential for treating the motor symptoms of PD and has L-DOPA-sparing potential.
4.1 VU0364770 pharmacokinetic properties
To investigate the pharmacokinetic properties of N-(3-chlorophenyl)picolinamide (VU0364770), a Ca2+ mobilisation assay was carried out by Jones et al. Chinese hamster ovary cells which expressed human mGluR4 coupled to the chimeric G protein Gqi were used in this assay. Their ability to mobilise Ca2+ in response to glutamate was measured. There was a concentration-dependent improvement by VU0364770 with a potency of 1.1 ± 0.2M of the response to glutamate (10 M) and enhanced the maximal response to glutamate from 100 to 227 ± 17% (Figure 13A). To test the efficacy of VU0364770, Jones et al. wanted to see the effect of a 30 M concentration of this compound on the glutamate concentration-response curve. VU0364770 was able to shift this curve to the left by 31.4 ± 4.0-fold (Figure 13B). Overall, VU0364770 did not have any intrinsic allosteric agonist activity in this in vitro assay. This Ca2+ mobilisation assay shows that VU0364770 is an effective and potent positive allosteric modulator of mGluR4 (Jones et al., 2012).
As VU0364770 is an allosteric modulator of mGluR4, it can potentially bind to allosteric sites on other GPCRs. This would not be found in radioligand binding assays as they test orthosteric radioligands. To determine the selectivity of VU0364770 for mGluR4 compared to other GPCRs including the mGluRs, the effects of VU0364770 (10 M) on these receptors was tested using a panel of 176 GPCRs. The compound had very good pharmacology with very little agonist, allosteric, or antagonist activity at these receptors. VU0364770 had weak positive allosteric activity at mGluR6 and antagonist activity at mGluR5 when compared to mGluR4. When the compound was further assessed, VU0364770 had antagonist activity at mGluR5 with an EC50 value of 17.9 ± 5.5 M and positive allosteric activity at mGluR6 with an EC50 value of 6.8 ± 1.7 M (compared with VU0364770 EC50 value of 290 ± 80 nM on rat mGluR4). Overall, these results show that VU0364770 has good pharmacology for use in behavioural studies when activating mGluR4 in vivo. In addition, VU0364770 has shown an improvement in vivo pharmacokinetic properties compared to previously studied allosteric modulators of mGluR4 such as PHCCC. This compound has better central penetration and a total of brain-to-plasma ratio of more than 1 after administering a systemic dose of 10 mg/kg (Jones et al., 2012).
4.2 VU0364770 enhances efficacy of preladenant to reduce haloperidol-induced catalepsy
Early research showed that selective antagonists of adenosine A2A receptors have strong antiparkinsonian-like effects in dopamine-depleted animals. In 2011, phase 2 clinical trials with PD patients showed antiparkinsonian effects when preladenant, an adenosine A2A antagonist, was administered alone or with L-DOPA (Hauser et al., 2011). In 2008, it was found that co-administering a group III agonist and an adenosine A2A antagonist showed a bigger effect in reversing haloperidol-induced catalepsy than if the compounds were given alone (Lopez et al., 2008). Therefore, Jones et al. wanted to see if co-administration of preladenant and VU0364770 had potential of having a bigger reduction in haloperidol-induced catalepsy than when administered alone. Rats were randomised into different treatment groups and received an intraperitoneal injection of a 1.5 mg/kg dose of haloperidol to induce catalepsy (Jones et al., 2012). Haloperidol works by blocking the dopamine D2 receptors found on neurons in the striatum. This stops the transmission of striatal dopamine in the indirect pathway of the basal ganglia (BG) circuit (Duty and Jenner, 2011). The rats were then tested for catalepsy after administration of the compounds by how long the delay is before the rat removes one or both forepaws from a horizontal bar placed on the testing surface. VU0364770, at doses of 30 and 56.6 mg/kg when administered subcutaneously, was able to reduce haloperidol-induced catalepsy in the rats (Figure 14A). Preladenant also reduced haloperidol-induced catalepsy in the rats with an oral administration of 1-30 mg/kg (Figure 14B). When a dose of 10 or 30 mg/kg of VU0364770 was co-administered with a dose of 0.1-1 mg/kg of preladenant, there was a significant left shift of the preladenant dose-response curve (Figures 14C-E). It was found that there was also efficacy at previous ineffective doses of preladenant at 0.1 and 0.3 mg/kg. Both compounds had no significant change on the brain-to-plasma ratio. In conclusion, these results show that allosteric modulation of mGluR4 and antagonism of A2A receptors may provide a better antiparkinsonian actions in preclinical and clinical studies of PD. This could potentially provide a better therapeutic option for patients with PD than dopamine-replacement therapy for treatment of the motor symptoms associated with the disease (Jones et al., 2012).
4.3 VU0364770 has antiparkinsonian actions in 6-OHDA rodent model
To assess the antiparkinsonian actions of VU0364770 in behavioural studies, the ability of VU0364770 to reduce forelimb asymmetry in a 6-hydroxydopamine (6-OHDA) rodent model was measured. Rats were injected with 6-OHDA (4 l) unilaterally into the left medial forebrain bundle of the brain. This depleted dopamine levels in the left side of the brain, causing forelimb asymmetry in the rats. The cylinder test was used to see whether the rat used the affected (contralateral) or non-affected (ipsilateral) forelimb as it lands on the base of the cylinder after rearing. Rats orally administered with L-DOPA at doses of 2.5 and 4.5 mg/kg had a significant dose-dependent reduction in the forelimb asymmetry (FLA) index score (Figure 15A). Rats injected subcutaneously with VU0364770 with a dose of 100 mg/kg also showed some reduction in the FLA index score by ~32% when compared with the vehicle group (Figure 15B). Co-administration of L-DOPA, at a not as effective dose of 1.5 mg/kg, and VU0364770, at a dose of 100 mg/kg, caused a significant reversal of forelimb asymmetry of ~75% (Figure 15B). Brain-to-plasma ratios were unaffected when these compounds were administered together when compared to administered alone. Therefore, VU0364770 has antiparkinsonian actions by reversing forelimb asymmetry in rats alone or in combination with L-DOPA (Jones et al., 2012).
The efficacy of VU0364770 found in the rodent model of PD further supports the hypothesis that activating mGluR4 may provide a potential therapeutic approach for treating the motor symptoms associated with PD as a monotherapy and/or combined with dopamine replacement therapy. These studies also show that allosteric modulation of mGluR4 may potentially have a L-DOPA sparing effect. This would allow use of a lower dose of L-DOPA to avoid L-DOPA-induced dyskinesia which is a major side effect of this treatment (Jones et al., 2012).
4.4 VU0364770 is not effective with higher dose of L-DOPA
In 2015, a different set of authors conducted the same experiments as Jones et al. using VU0364770. They also compared this compound with an agonist of mGluR4, LSP1-2111. In the first experiment, an acute 6-OHDA rodent model was used. The rats received unilateral 6-OHDA lesions to the right medial forebrain bundle and received a higher dose of L-DOPA (6 mg/kg) than in the experiment by Jones et al, who used a dose of 1.5 mg/kg, to induce dyskinesia. Then the rats were treated with either VU0364770 (100 mg/kg), LSP1-2111 (15 mg/kg), or vehicles and abnormal involuntary movements (AIMs) were assessed. It was found that VU0364770 co-administered with L-DOPA had no effect in reducing how severe the dyskinesias were but the treatment did increase the time-action curve of the AIMs (Figure 16A). However, the effect of VU0364770 combined with L-DOPA on AIMs was not significantly different to the L-DOPA with vehicle treatment. The mGluR4 agonist, LSP1-2111 did not have an effect on how severe or how long the AIMSs were (Figure 16B) (Iderberg et al., 2015).
In the second experiment by Iderberg et al., rats were randomised into two groups for a chronic study using the 6-OHDA model. The first group of rats received daily injections of either L-DOPA (6 mg/kg) + VU0364770 (100 mg/kg) or L-DOPA (6 mg/kg) + LSP1-2111 (15 mg/kg) for 2-3 weeks. The second group of rats received daily injections of L-DOPA (6 mg/kg) + vehicle. For the duration of the treatment, VU0364770 and LSP1-2111 in addition to L-DOPA had no effect on the AIM scores (Figures 17A and D). Further evaluation of these findings by analysing AIM scores per monitoring period did not show any within-session effect of VU0364770 or LSP1-2111 on severity or duration of the AIM scores (Figures 17B and E). Due to these results being inconsistent with the prolonged AIMs by VU0364770 in the first experiment, automated rotation tests were used to further analyse. The rotation tests revealed that the total number and duration of L-DOPA-induced turns were significantly unaffected by VU0364770 or LSP1-2111 (Figures 17C and F) (Iderberg et al., 2015).
4.5 VU0364770 has L-DOPA sparing potential
In the third experiment by Iderberg et al., the L-DOPA sparing potential of VU0364770 was tested by using three different behavioural tests. These tests involved the automated rotation test, the cylinder test and the rotarod test. They were used to assess the actions of VU0364770 at a dose of 100 mg/kg when administered alone or in combination with L-DOPA. The L-DOPA dose of 1.5 mg/kg is a dose that has no effect on forelimb asymmetry when administered alone, which was determined by Jones et al. In the rotation test, 1.5 mg/kg of L-DOPA had a small but non-significant effect on rotational behaviour when compared to the saline-vehicle treatment (Figures 18A and B). When administered alone, VU0364770 had no effect on rotational behaviour but significantly increased the effects of L-DOPA when administered together (Figures 18A and B). The co-administration of VU0364770 and L-DOPA increased the total rotational behaviour five-times more than the saline-vehicle treatment (Figure 18B). When the cylinder test was conducted, rats treated with saline-vehicle used their forelimb contralateral to the lesion 17% of the time. Treatment with L-DOPA did not show any improvement in akinesia with only 15% of contralateral forelimb use (Figure 18C). Contralateral forelimb use was 20% when rats were treated with VU0364770 alone or 19% when administered with L-DOPA, but these results were not significant (Figure 18C). When conducting the rotarod test, baseline performance of the rats was enhanced by a significant amount by drug treatments. The L-DOPA treatment increased the baseline performance of the rats on the rotarod by 18% (Figure 18D). There was a greater improvement in baseline performance of 27% when L-DOPA was administered with VU0364770 (Figure 18D). However, the difference between the treatment with L-DOPA and the combined treatment with VU0364770 was not significant (Iderberg et al., 2015).
To conclude, these studies by Iderberg et al. confirm that VU0364770 has L-DOPA sparing potential. However, there was no significant difference in treatment of dyskinesias between a full dose of L-DOPA on its own and when VU0364770 is added which is in contrast with the research done by Jones et al. One positive outcome of these studies by both set of authors is that potentiation of mGluR4 may be combined with L-DOPA to reduce L-DOPA dose requirement. This may reduce the incidence of dyskinesias if the treatment is initiated early. These studies by Iderberg et al. shows that neither positive allosteric modulators nor agonists of mGluR4 possess any intrinsic antidyskinetic activity.
Chapter 5: Summary and Future Prospects
Early studies showed that activation of mGluR4 with an agonist, L-AP4, reduced inhibition of transmission at the striatopallidal synapse of the basal ganglia motor circuit. L-AP4 also produced antiparkinsonian effects in rodent models of PD by reducing reserpine-induced akinesia and reversing forelimb asymmetry. These studies suggest that activation of mGluR4 may be a potential target for treating the motor symptoms of PD. This treatment may be a more effective palliative treatment than dopamine replacement therapy (Valenti et al., 2003).
However, it was found that potentiating the effects of L-AP4 on mGluR4 using a positive allosteric modulator produced a more significant inhibition of transmission at the striatopallidal synapse than using the agonist alone. The positive allosteric modulator used in these studies was PHCCC. This compound also produced antiparkinsonian effects in a reserpine-induced akinesia rodent model of PD. These studies suggest that allosteric modulators have greater therapeutic potential in treating the motor symptoms of PD than using agonists alone. In addition, these studies further confirm that activation of mGluR4 provides potential therapeutic approach for the treatment of PD (Marino et al., 2003).
A more recent study developed a positive allosteric modulator of mGluR4 with better efficacy than PHCCC, which can also be used for systemic dosing. This compound, VU0364770, when combined with either L-DOPA or an adenosine A2A antagonist was able to potentiate the effects of both drugs in reversing the motor deficits in rodent models of PD. VU0364770 was found to have L-DOPA sparing potential which means a lower dose of L-DOPA can be used in combination with this compound. This may reduce the incidence of dyskinesias, which is a major side effect of L-DOPA, if the treatment is initiated early in PD patients (Jones et al., 2012; Iderberg et al., 2015). However, this L-DOPA sparing potential of VU0364770 needs to be studied further.
mGluR4 positive allosteric modulators have not been trialled in humans yet as further preclinical and clinical studies are needed. With the promising research so far, it can be said that positive allosteric modulators of mGluR4 have great therapeutic potential in treating the motor symptoms associated with PD.
Another area of therapeutic potential is targeting mGluR5 for the treatment of PD. mGluR5 negative allosteric modulators (NAM) decrease the effects of the agonist on its receptor. Fenobam is a mGluR5 NAM which decreases L-DOPA-induced dyskinesia in both rodent and primate models of PD (Rylander et al., 2010). Another mGluR5 NAM, AFQ056, was able to reach phase II clinical trials for the treatment of L-DOPA-induced dyskinesia however, this drug did not show antidyskinetic effects in two of the three phase II clinical trials (Litim, Morissette and Di Paolo, 2017). The most successful mGluR5 NAM in terms of clinical studies is ADX48621 which has completed phase II clinical trials and is ready for phase III clinical trials for the treatment of L-DOPA-induced dyskinesia. This drug has displayed good safety and tolerability. In addition, ADX48621 is able to reduce the severity of the L-DOPA-induced dyskinesia (Litim, Morissette and Di Paolo, 2017). Therefore, allosteric modulation of metabotropic glutamate receptors provides therapeutic potential in the treatment of PD.
Conclusion
To conclude this report, selective positive allosteric modulators provide a novel approach for treating motor symptoms of Parkinson’s disease. The selective potentiation of mGluR4 may provide a potential L-DOPA sparing mechanism which needs to be studied further. Positive allosteric modulators of mGluR4 have not been trialled in humans yet as further preclinical and clinical studies are needed. Therefore, allosteric modulators of mGluR4 could be useful for treating Parkinson’s disease.
