Huntington’s Disease

Huntington’s disease (HD) is a devastating autosomal dominant inherited neurodegenerative disorder characterized by progressive motor, cognitive, and psychiatric symptoms, leading to death, inevitably [1, 2]. Cognitive and personality alterations are early symptoms, followed by chorea and loss of balance. Movement difficulties, which progressively worsen over time, are associated with both involuntary and voluntary movement. Among the psychiatric disturbances that take place in HD, affective disorders are the most commonly prevalent, with documented rates of major depression as high as 50% [3] and mania or hypomania as high as 12% [3, 4].

A polyglutamine expansion in the amino-terminal region of the huntingtin (htt) protein is the cause of HD [5]. The length of the polyglutamine repeat is inversely correlated with the age of disease onset and directly correlated with the severity of symptoms [6]. However, patient sex, environmental factors and genetic modifiers can alter the variability of clinical expression. Although the htt mutation is pointed to as the cause of HD, the mechanisms responsible for mutant htt pathogenicity are still largely unknown. There is no data explaining why the mutant protein, which is expressed throughout the body results in the selective death of medium sized spiny neurons (MSNs). In addition, it is still unclear whether HD pathology progresses due to either a lack of function of the htt protein or to a gain of toxic function of the mutant htt. Normal htt protein has been shown to be anti-apoptotic [7, 8] and essential for normal embryonic development [7, 9, 10]. On the other hand, mutant htt and htt aggregation triggers a cascade that leads to neuronal dysfunction through oxidative stress, transcriptional dysregulation, glutamate excitotoxicity, activation of apoptotic cascade, mitochondrial dysfunction and energy depletion [11–14].

Cleavage of polyglutamine expanded htt leads to the release of amino-terminal fragments containing the polyglutamine repeats, which can aggregate in neurites, cytoplasm, and nuclei. Importantly, htt aggregate formation and loss of striatal neurons strongly correlate with HD symptom severity [15]. MSNs in the striatum, containing GABA and enkephalin, are affected early in the disease and are the primary neurons targeted in HD. Over time, htt aggregates and inclusions spread to the remainder of the basal ganglia with subsequent dissemination through the cortex and substantia nigra. However, it is not known if the accumulation of htt aggregates results in cell death or if the soluble form of the protein is the toxic one [16–18].

Glutamate Receptors and Cell Signaling

Glutamate, the major excitatory neurotransmitter in the brain, is essential for a wide variety of physiological processes, such as integrative brain function and neuronal cell development. However, glutamate is also implicated in neuronal cell death and has been postulated to play an important role in the pathogenesis and excitotoxic neuronal cell loss that takes place in HD [19–22].

Glutamate exerts its actions by interacting with ionotropic glutamate receptors, which are ligand-gated ion channels that mediate fast excitatory neurotransmission, and metabotropic glutamate receptors (mGluRs), which are members of the family C of G protein-coupled receptor (GPCR) [23–27]. There are at least three different types of ionotropic glutamate receptor, N-methyl-d-aspartate receptor (NMDA), alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), and kainate receptors, and eight distinct mGluRs, which are divided into three subgroups based on sequence homology and G protein coupling specificity [22, 23, 28, 29]. Group I mGluRs (mGluR1 and mGluR5) are predominantly coupled to the activation of phospholipase C (PLC) via Gαq/11, whereas Group II (mGluR2 and mGluR3) and Group III (mGluR4, mGluR6, mGluR7 and mGluR8) mGluRs negatively regulate adenylyl cyclase via Gαi.

mGluRs are differentially localized in presynaptic and postsynaptic neuronal regions [30, 31]. Group II and Group III mGluRs are mainly localized presynaptically and act as autoreceptors to inhibit glutamate release [32]. Group I mGluRs can be localized at both presynaptic and postsynaptic sites, although at synapses they are mainly located peri-synaptically at the postsynaptic neuronal membrane, where they function to regulate neuronal excitability by modulating currents mediated by ionotropic glutamate receptors [33–36].

Group I mGluR stimulation can lead to activation of a wide variety of cell signaling pathways, generating very complex responses [27] (Fig. 1). Activation of PLC by mGluR1/5 leads to diacylglycerol and inositol-1,4,5-triphosphate (InsP3) formation [37]. InsP3 binding to its receptor leads to release of calcium from intracellular stores. Both calcium and diacylglycerol lead to the activation of protein kinase C (PKC), which has been proposed to activate phospholipase D (PLD), phospholipase A2 (PLA2) and mitogen-activated protein kinase (MAPK), as well as to modulate a variety of ion channels [38]. mGluR1/5 can also modulate calcium channels in a PLC/PKC-independent and G protein-dependent manner [39, 40]. Moreover, Group I mGluRs can also modulate potassium channels employing different cell signaling strategies [41, 42]. Ionotropic glutamate receptors can also be regulated by mGluRs, as stimulation of mGluR5 leads to PKC activation, which can activate NMDAR by increasing its open probability, leading to calcium influx [43]. Interestingly, it has been shown that mGluR5 and NMDAR can be cross-linked through the interaction with Homer and Shank proteins [44, 45]. Group I mGluR stimulation also leads to activation of other signaling pathways important for cell survival and proliferation, such as extracellular signal-regulated kinase (ERK) and AKT [46–48]. mGluR1/5 signaling complexity may underlie different responses, depending on activation context (agonist type and concentration, disease state, neuronal cell type, etc.). Group I mGluRs signaling can be modified in various disease states [27, 49] and understanding mGluRs regulation will be essential for developing efficient pharmacological strategies to treat such disorders.

Fig. 1
figure 1

Cell signalling pathways activation by Group I mGluRs. Group I mGluRs activate the hydrolysis of PIP2 by phospholipase C (PLC) following the activation of the heterotrimeric G protein Gαq resulting in increases in intracellular DAG and InsP3 (IP3) levels. InsP3 activates the InsP3 receptor (IP3R) resulting in increased intracellular Ca2+ concentrations which in conjunction with DAG activate protein kinase C (PKC) PKC activation can lead to the activation of ERK1/2 phosphorylation and the phosphorylation of the NMDA receptor. Homer interacts with the C-tails of Group I mGluRs and can either contribute to the activation of Akt via PI3 kinase (PI3K) or can directly regulate NMDA receptor activity via its association with Shank

Full size image

Cell Signaling Alterations Caused by Mutant htt

Mutant htt protein has been implicated in a wide variety of cell signaling alterations that are appearing as important features underlying the progression of HD pathology. Mutant htt protein promotes cellular alterations mainly by modifying htt-protein interaction patterns [50] and/or by altering transcription of key cellular components [51]. Htt interaction with other proteins can be modified by htt polyglutamine expansion resulting in altered htt-mediated cellular processes. In particular, alteration in mutant htt protein interaction can modify htt functions involved in clathrin-mediated endocytosis, neuronal transport processes, postsynaptic signaling, apoptosis and cell survival [52–56].

Numerous studies in both HD patients and mouse models have demonstrated that mutant htt protein can alter neuronal function and cell signaling by disrupting transcriptional pathways and altering striatal gene expression profiles [57]. The down-regulation of several specific transcripts, such as preproenkephalin (PPE) [58, 59], dopamine and cyclic AMP-regulated phosphoprotein, 32 kDa (DARPP-32) [60], D1 and D2 dopamine receptors [61], adenosine A2A receptor [62], and the CB1 cannabinoid receptor [63, 64], has been particularly well studied and quantified. Thus, the htt polyglutamine expansion can affect its interactions with other proteins and transcriptional regulation of multiple genes, causing a wide variety of cell signaling alterations that are crucial for disease progression.

Htt-Mediated Cell Signaling Alterations Involving the Glutamatergic System

An important cell signaling alteration promoted by the htt polyglutamine expansion is an increase in intracellular Ca2+ levels, which may contribute to the neuronal cell death that takes place in HD. Mutant htt leads to NMDAR sensitization, increasing Ca2+ influx into the cell [65, 66], as discussed further in “Role of Glutamate Receptors in Neuronal Survival in HD”. Moreover, mutant htt protein can also cause an increase in intracellular Ca2+ levels by destabilizing mitochondrial Ca2+ regulation [67, 68] and sensitizing InsP3 receptor-mediated release of Ca2+ from intracellular stores [69, 70]. As a consequence of InsP3 receptor sensitization, the stimulation of Gαq/11 coupled receptors (e.g., mGluR1/5) leads to an increase in Ca2+ release from intracellular stores [49, 70].

Mutant htt protein can alter Group I mGluR cell signaling by a mechanism involving its interaction with optineurin, a protein that has been demonstrated to contributed to the attenuation of mGluR1/5 signaling [56]. Although wild-type htt protein does not enhance optineurin mediated mGluRs desensitization, mutant htt functions to synergistically increase optineurin-mediated mGluR desensitization in HEK 293 cells [56]. New data from our group confirmed that InsP3 formation stimulated by endogenous mGluR5 is selectively attenuated in the striatum of a knock-in mouse model of HD (Hdh Q111/Q111 mice), as compared to control mice (Hdh Q20/Q20 mice) [49]. However, InsP formation in response to activation of muscarinic acetylcholine receptors remains unaltered in the Hdh Q111/Q111 mice. Interestingly, the attenuation of mGluR5 signaling observed in Hdh Q111/Q111 mice is PKC dependent and is only present in young asymptomatic mice, and is lost in mice that are older than 11 months [49]. PKC-dependent mGluR1/5 attenuation might be protective, as activation of mGluRs in striatal neurons and inhibition of PKC lead to increased death of neurons derived from Hdh Q111/Q111 mice without affecting survival of Hdh Q20/Q20 neurons [49]. Despite decreased InsP formation, DHPG-mediated Ca2+ release is higher in Hdh Q111/Q111 than in Hdh Q20/Q20 neurons. Thus, it is possible that the PKC-mediated mGluR1/5 desensitization is protective as it avoids further increases in calcium release that could result in increased cell death.

In addition to the attenuation of mGluR1/5-mediated InsP3 formation, other Group I mGluRs signaling pathways that may be protective against cell death are increased in HD [49]. mGluR5 activation leads to higher levels of ERK1/2 phosphorylation in Hdh Q111/Q111 striatal neurons than in Hdh Q20/Q20 neurons. Moreover, basal levels of Akt are increased in Hdh Q111/Q111 neurons, and inhibition of either NMDA receptors or mGluR5 decreases phospho-Akt levels in Hdh Q111/Q111 neurons as compared to Hdh Q20/Q20 neurons [49, 71]. Phosphorylation of Akt by glutamate receptors is particularly important because it has been shown that Akt activation can protect against neuronal death [72, 73]. The activation of Akt by mGluR5 involves phosphoinositide 3-kinase (PI3K) and phosphoinositide-dependent kinase (PDK1) [47, 48]. A PI3K enhancer (PIKE) couples group I mGluRs to PI3K via Homer proteins [47]. mGluR activation leads to formation of the functional complex mGluR-Homer-PIKE, allowing PI3 kinase activation by PIKE, which results in reduced apoptosis [47]. Thus, mGluR-mediated Akt activation is independent of InsP3/PLC pathway.

Akt is emerging as a key component to protect against neuronal cell death in a growing number of neurodegenerative disorders. In HD, Akt can promote phosphorylation of mutant Htt protein, which leads to reduced Htt aggregate formation and neuronal cell death, providing a protective pathway in HD [74, 75]. It has also been demonstrated that a myristoylated form of Akt, which is constitutively active, has a potent in vivo anti-apoptotic effect on dopaminergic neurons of the substantia nigra in a mouse model of Parkinson’s disease [76]. In addition, Akt also has an important protective role in Alzheimer’s disease (AD), as Akt activation leads to inhibition of glycogen synthase kinase-3 (GSK-3) and GSK-3 promotes phosphorylation of tau leading to neurofibrilary tangle formation and neuronal cell death [77]. Interestingly, amyloid-beta can activate GSK-3 by inhibiting PI3K/Akt pathway and Akt activation can reverse amyloid-beta toxic effects [78]. Thus, Akt occupies a pivotal role in Alzheimer’s disease pathology, modulating both amyloid-beta and tau pathologies, constituting a potential therapeutic target to treat AD patients. It is possible that the observed increase in Akt activation in Hdh Q111/Q111 striatal neurons plays a similar role in HD by protecting neurons against cell death during the asymptomatic stages of the disease.

Htt-Mediated Cell Signaling Alterations Involving the Dopaminergic System

The dopaminergic system appears to play a role in HD, as a dopamine transporter knock-out mice display both spontaneous striatal death and behavioural alterations that resemble HD [79]. Moreover, exacerbation of HD symptoms and augmentation of aggregate formation occur when these mice are mated to a knock-in HD mice [80]. In vitro data demonstrate that dopamine stimulation can lead to the formation of reactive oxygen species and the activation of pro-apoptotic pathways, as well as aggregate formation and mitochondria dysfunction via D2 dopamine receptor activation [81, 82]. Furthermore, D2 antagonist treatment protects against aggregate formation and striatal dysfunctions induced by mutant htt [81, 83]. D2-mediated activation of aggregate formation involves a Rho/ROCK signaling pathway, as inhibition of ROCK activity reverses D2-promoted aggregate formation, neuritic retraction and neuronal death induced by mutant htt [84]. Glutamate and dopamine signaling pathways can also act synergistically to induce apoptosis of MSNs via elevated Ca2+ signaling [85, 86]. Moreover, it has been shown that, in cells expressing mutant htt, NMDAR activation potentiates D1-induced Cdk5 phosphorylation, which can lead to neurotoxicity and apoptosis [86].

Htt-Mediated Cell Signaling Alterations Involving the Cannabinoid System

Cannabis sativa, popularly known as marijuana, has been used as recreational drug for the past 4,000 years by numerous cultures. In 1964, Gaoni and Mechoulam [87] identified Δ9-tetrahidrocanabinol (Δ9-THC) as the major psychoactive constituent of the plant. The effects of Δ9-THC are mediated mainly by two cannabinoid receptors named CB1 and CB2 which were cloned and characterized in the early 1990s [88, 89] and one of the most surprising findings was that these receptors could bind endogenous ligands recently known as endocannabinoids (ECs). Five ECs have been identified thus far, including anandamide (AEA) and 2-arachidonoyl glycerol (2-AG), which are the two most studied ECs [90, 91]. The ECs are released by post-synaptic neurons and act predominantly at pre-synaptic neurons [92]. This retrograde signaling pathway has emerged as being important in synaptic plasticity and, recently, neurobiologists have increasingly turned their attention to EC system as it has been implicated in numerous neurophysiologic functions such as pain, appetite, learning and memory, and motor functions [93–96]. It has been shown that AEA can also act as a full agonist of transient receptor potential vanilloid 1 (TRPV1) and, since these receptors are expressed both in the periphery and in the CNS, this AEA endovanilloid activity may influence many physiological brain functions [97]. A role for ECs in a variety of CNS disorders, especially neurodegenerative diseases, is suggested by the high levels of expression of CB1 receptors in brain regions involved in cognition and motor activity [98].

The activation of EC signaling by direct receptor agonists and/or inhibition of EC metabolism has powerful effects on the control of movement, mostly inhibitory [99, 100]. The motor effects of the EC system are related to the capacity of this system to modulate the activity of glutamate, GABA and dopamine which participate in the control of basal ganglia function [101, 102]. The presence of CB1 receptors at GABAergic and glutamatergic synapses within the basal ganglia, as well as the presence of TRPV1 receptors in nigrostriatal dopaminergic neurons, enables ECs to directly control the function of these key neurotransmitters [102]. Thus, the main function of cannabinoid system within the basal ganglia is to modulate GABAergic and glutamatergic synapses through a retrograde signaling mechanism [101].

The first evidence linking EC signaling with HD was provided by Glass et al. [103]. They demonstrated a loss of approximately 97% of CB1 receptors in the subtantia nigra of human HD brains [103]. This loss of CB1 receptors preceded the loss of D1 and D2 dopamine receptors and occurred even before the onset of major HD symptoms [104]. Thus, CB1 receptors may play a central role in either the pathogenesis and/or progression of the neurodegeneration in HD patients. These findings are consistent with the observation that medium spiny-GABAergic neurons, which specifically express CB1 receptors, are the predominant neuronal population lost in the basal ganglia in HD [98].

The EC system has also been studied in a variety of animal models of HD [105–108]. These mouse models develop many HD features such as striatal atrophy, intraneuronal aggregates and progressive dystonia and also exhibit decreased CB1 mRNA levels and activity in basal ganglia. This reduction CB1 mRNA expression occurs prior to the development of motor symptoms and neuronal degeneration. TRPV1 also seems to be involved in HD as the stimulation of this receptor subtype, located at nigrostriatal dopaminergic neurons of basal ganglia, reduces hyperkinesia in HD animal models [109, 110].

Htt-Mediated Cell Signaling Alterations Involving Trophic Factors

Trophic factors, such as brain-derived neurotrophic factor (BDNF), largely influence neuronal survival and function [111]. Wild-type, but not mutant, huntingtin promotes the transcription and vesicular transport of BDNF [112–114]. Polyglutamine expansion of the htt protein results in the reduction of BDNF transcription and axonal transport, which may affect the survival of both striatal and cortical neurons [112, 113]. It has also been shown that BDNF is able to prevent the death of striatal projection neurons in a quinolinic acid model of HD and in a 3-NP-induced toxicity mouse model, which causes abnormal movement, cognitive deficits and neuronal degeneration similar to that seen in HD patients [115–117]. Most of BDNF neuroprotective effects are mediated by TrkB receptor-induced activation of pro-survival signaling pathways, including: PLC-γ, Ras/MEK/MAPK and PI3K/Akt pathways [118]. BDNF also protects cortical neurons from 3-NP toxicity through the activation of PI3K and ERK1/2 intracellular signaling pathways resulting in decreased mitochondrial abnormalities and apoptosis [119]. These data highlight the importance of altered BDNF signaling in HD pathology.

Research on BDNF and HD has focused on drugs that could boost BDNF production, as this trophic factor does not cross the blood–brain barrier. Recent data have demonstrated that ampakine, a positive modulator of AMPA glutamate receptors, can up-regulate endogenous BDNF levels, rescuing plasticity and reducing learning problems in a HD mouse model [120]. However, ampakine treatment has no measurable effect on decreased locomotor activity. Nevertheless, as ampakines are well tolerated by patients, they may represent a novel strategy for treatment of the cognitive difficulties that occur in HD, as well as for preventing neuronal cell death [120].

Role of Glutamate Receptors in Neuronal Survival in HD

Glutamate receptors are appearing as important pharmacological targets in HD. It is well known that glutamate plays an important role in neuronal excitotoxicity through the activation of ionotropic receptors [11, 19, 121]. Excitotoxicity is one of the most extensively studied processes of neuronal cell death, and plays an important role in many CNS disorders, including ischemia, trauma, and neurodegenerative disorders, such as AD, HD, Parkinson’s disease, and amyotrophic lateral sclerosis [122–128]. Excitotoxicity is characterized by an excessive synaptic release of glutamate, leading to glutamate receptor over-stimulation and Ca2+ overload, compromising mitochondria function and leading to cell death [129, 130]. Elevation of intracellular Ca2+ by glutamate is mainly achieved by activation of the ionotropic NMDAR via calcium influx and, to a lesser degree, Group I mGluRs (mGluR1 and mGluR5), which are coupled to Ca2+ release from intracellular stores [25, 29].

Several studies have implicated NMDAR signaling in excitotoxic neuronal loss in HD [11, 19, 121]. Radioligand binding studies, using post-mortem brain tissue from HD patients in the early symptomatic phase, showed a loss of striatal NMDARs suggesting that striatal neurons with high NMDAR expression are more vulnerable and are lost early during disease progression [131, 132]. Moreover, a mouse model of HD was created by injecting the NMDAR agonist quinolinic acid into the striatum [133, 134]. This HD mouse model exhibits many of the HD-like lesions and symptoms [133, 134]. Furthermore, NMDAR-mediated excitotoxicity may explain why MSNs are more vulnerable in HD. NMDARs that are comprised of the NR1A/NR2B, but not NR1A/NR2A subunit combination, can be sensitized by mutant htt [65, 135]. Interestingly, MSNs mainly express the NR1A and NR2B subunits [136], whereas other brain regions express combinations of both NR2A and NR2B with a variety of NR1 splice variants [137, 138]. Thus, NMDAR-specific subunit expression might underlie the preferential death of MSNs in the striatum. Group I mGluRs may also play a role in the selective loss of MSNs in the striatum as DHPG stimulation strongly enhances membrane depolarization and intracellular calcium accumulation induced by NMDAR in MSNs, but not in cholinergic striatal interneurons, which are spared in HD [20].

The prominent role of NMDARs in neuronal excitotoxicity and cell death has led to a concerted effort to design and assess NMDAR antagonists such as ketamine, phencyclidine, and MK-801 for the treatment of neurological disorders involving cell death. However, although ionotropic glutamate receptors are likely to be essential for the neuronal cell loss that occurs in HD, pharmacological approaches targeting these receptors is not an ideal strategy. This is because the blockage of ionotropic receptors leads to many toxic effects, including psychosis, nausea, memory impairment, and neurotoxicity, which have led to their failure in clinical trials and a search for alternative therapeutic targets [139, 140]. Group I mGluRs have a modulatory rather than excitatory role in neurotransmission, making these receptors exciting targets for new therapeutic strategies for a number of neurological disorders, including HD.

It is not clear yet whether Group I mGluRs have a role in HD. However, a direct link between htt and Group I mGluRs has been established by our group, as we have shown that mGluR1/5 interact with both Htt and optineurin, which is also a Htt-interacting protein [50, 56]. Nevertheless, the role of mGluR5 in HD and neuronal cell death is very controversial. Treatment of an HD transgenic mouse model with a mGluR5 antagonist increases survival, indicating that mGluR5 activation can accelerate HD progression [141]. In addition, disturbed calcium signaling and apoptosis observed in primary cultured MSNs of an HD mouse model has been attributed to activation of mGluR1/5 and NMDA receptors containing the NR2B subunit [69, 70]. Treatment with mGluR1 antagonists results in a decrease in the tissue infarct size and cell death in an in vivo animal models of ischemic stroke [142–146]. On the other hand, it has been reported that mGluR1 knock-out mice do not show any difference in infarct size when compared to control mice [147]. Moreover, other studies have provided evidence that Group I mGluRs activation may be protective. For example, when cortical neuronal cultures are consecutively incubated two times with DHPG (3,5-dihydroxyphenylglycine), an agonist for group I mGluRs, NMDA excitotoxicity is attenuated [148]. In rat hippocampal organotypic slices, DHPG stimulation protects CA1 hippocampal cells and this effect is lost when mGluR1 antagonists are applied [149]. As outlined in “Glutamate Receptors and Cell Signaling”, data from our group suggest that mGluR1/5 signaling is modified in a mouse model of HD during the presymptomatic phase of the disease and we suggest that these alterations have a protective role [49, 56]. Thus, depending on the context of activation, group I mGluR stimulation is found to be either neurotoxic or neuroprotective and this may be related to the precise molecular mechanism by which mGluR signaling is achieved.

Concluding Remarks

Polyglutamine expansion of the htt protein results in altered htt protein interactions and gene transcript that results in perturbations in cell signaling that affect neuronal homeostasis. It is possible that several of these alterations in htt function occur in the pre-symptomatic phase of the disease and might be crucial to determine the rate of HD progression. Understanding these early changes in cell signaling caused by mutant htt will be essential for the development of pharmacological therapies to prevent neuronal cell death in HD and slow disease progression. Mutant htt can induce alterations of the dopaminergic and cannabinoid system and these alterations can also influence the glutamatergic system. Moreover, mutant htt-induced decrease in BDNF function may also contribute to neuronal cell loss and motor alterations. In addition, Htt mutation promotes several alterations in the glutamatergic system, involving both ionotropic and metabotropic glutamate receptors. NMDARs are sensitized by mutant htt and Group I mGluRs signaling is significantly altered in mouse models of HD. The glutamatergic system is closely linked to the regulation of neuronal cell death. However, glutamate receptor activation may be either excitotoxic or protective depending upon the context of activation. In particular, it is possible that pharmaceutical agents may be developed that selectively activate mGluR-stimulated pro-survival pathways as opposed to activating mGluR signaling that leads to neuronal cell death.