P2Y Receptors in the Mammalian Nervous System: Pharmacology, Ligands and Therapeutic Potential

Abstract

P2Y receptors for extracellular nucleotides are coupled to activation of a variety of G proteins and stimulate diverse intracellular signaling pathways that regulate functions of cell types that comprise the central nervous system (CNS). There are 8 different subtypes of P2Y receptor expressed in cells of the CNS that are activated by a select group of nucleotide agonists. Here, the agonist selectivity of these 8 P2Y receptor subtypes is reviewed with an emphasis on synthetic agonists with high potency and resistance to degradation by extracellular nucleotidases that have potential applications as therapeutic agents. In addition, the recent identification of a wide variety of subtype-selective antagonists is discussed, since these compounds are critical for discerning cellular responses mediated by activation of individual P2Y receptor subtypes. The functional expression of P2Y receptor subtypes in cells that comprise the CNS is also reviewed and the role of each subtype in the regulation of physiological and pathophysiological responses is considered. Other topics include the role of P2Y receptors in the regulation of blood-brain barrier integrity and potential interactions between different P2Y receptor subtypes that likely impact tissue responses to extracellular nucleotides in the CNS. Overall, current research suggests that P2Y receptors in the CNS regulate repair mechanisms that are triggered by tissue damage, inflammation and disease and thus P2Y receptors represent promising targets for the treatment of neurodegenerative diseases.

INTRODUCTION

P2 receptors are a diverse group of cell surface nucleotide receptors that can be separated into two structurally distinct subtypes: the P2X receptors that are ligand-gated ion channels and the G protein-coupled P2Y receptors. Genes for seven P2X receptor subtypes (i.e., P2X1-7) have been characterized and their physiological function and therapeutic relevance are discussed in other chapters in this issue. P2Y receptors are seven transmembrane spanning proteins that bind extracellular nucleotides to cause the activation of heterotrimeric G proteins and the transduction of intracellular signaling pathways that regulate a wide variety of cellular responses. To date, eight different subtypes of P2Y receptors have been cloned and characterized in mammalian cells of multiple species and have been shown to be activated by structurally distinct nucleotide agonists, including adenosine 5’-triphosphate (ATP), adenosine 5’-diphosphate (ADP), uridine 5’-triphosphate (UTP), uridine 5’-diphosphate (UDP) and UDP-glucose [1-3], as described in Table 1. Endogenous P2Y receptor (P2YR) agonists are released from cells in the central nervous system (CNS) under various conditions, including exocytosis at nerve terminals [4, 5], the opening of pannexin 1 hemi-channels [6-9], oxygen deprivation and apoptosis [4, 6], whereupon they activate all 8 subtypes of P2Y receptors (i.e., P2Y₁, P2Y₂, P2Y₄, P2Y₆, P2Y_11-14) that are expressed in cells that comprise the CNS (e.g., neurons, astrocytes, microglia and endothelial cells) [3, 10-19]. This review will describe the signaling pathways coupled to activation of each P2YR subtype and the physiological and pathophysiological responses regulated by P2YRs in the CNS. For example, P2YRs have been shown to regulate neurotransmission, cell growth, inflammatory responses and apoptosis [10-13, 20-23]. Recent studies have provided evidence that activation of P2YRs has both neuroprotective and neurodegenerative consequences in the CNS [24, 25] and, therefore, P2YRs represent novel therapeutic targets for the treatment of neurodegenerative diseases, as will be described for each P2YR subtype. In addition, we will summarize the current state of P2YR subtype-selective agonists and antagonists that could potentially represent promising drugs for modulation of cellular functions in the CNS.

Table 1.

P2Y Receptor Agonists and Antagonists

Receptor	Agonists	Antagonists	Potential Therapeutic Pathways
P2Y1	Endogenous: ADP Synthetic: 2-MeS-ADP, MRS 2365, ATP-α-B derivatives, (Ap5(γ-B)A), β,γ-Me-ATP, Di-(2-MeS)-adenosine 5',5"- P1,P4,α,β-methylene-tetraphosphate, 2-MeS-β,γ-CCl2-ATP	PPADS, Suramin, Reactive blue 2, ATPαS, β,γ-methylene-ATP, A3P5PS, A3P5P, A2P5P, MRS2179, MRS2500, MRS2279	Antagonists can prevent cytokine/chemokine-induced damage following ischemia in mice, agonists can induce axonal elongation in neurons and modulation of pain sensation, antagonists can reduce anxiolytic behavior in rats
P2Y2	Endogenous: ATP or UTP Synthetic: UTPγS, Ap4A, INS365 (Diquafosol), INS37217 (Denusofol), INS45973, 2'-amino-2'-deoxy-UTP, 6-nitro-UTP, UTPαS, 2'-deoxy-UTPaS, 2-thio-UTP, MRS2768, PSB1114	PPADS, Suramin, Reactive blue 2, PSB716	Agonists can increase the migration of glial cells through P2Y2R/integrin interactions, proliferation of glial cells, non- amyloidogenic APP processing in neurons, and the uptake and degradation of neurotoxic forms of Aβ1-42 by microglial cells
P2Y4	Endogenous: UTP (humans), UTP or ATP (mice, rats) Synthetic: UTPγS, 5-bromo-UTP, INS365 (Diquafosol), INS37217 (Denusofol), 2’- azido-2’-deoxy-UTP	PPADS, Reactive blue 2	P2Y4R activation can inhibit presynaptic glutamate release, modulate blood-brain barrier function, and inhibit K+ currents in rat myocytes
P2Y6	Endogenous: UDP Synthetic: UDPpS, MRS2693, PSB0474, INS48823, α,β-methylene-UDP, 5-bromo-UTP	PPADS, Suramin, Reactive blue 2, MRS2567, MRS2578, MRS2575	P2Y6R activation can increase phagocytic activity of microglia and regulate repair mechanisms in response to CNS injury
P2Y11	Endogenous: ATP Synthetic: ATPγS, BzATP, UTP, AR-C67085, NF546, NAD+, NADP+	Suramin, Reactive blue 2, AMPαS, NF157, NF340	P2Y11R activation delays pathogen- or inflammation-induced apoptosis in neutrophils, inhibits TLR signaling and modulates cytokine release
P2Y12	Endogenous: ADP Synthetic: 2-MeS-ATP, 2-MeS-ADP	Clopidogrel (Plavix), Ticlopidine (Ticlid), Prasugrel (Effient), Ticagrelor (Brillinta), Elinogrel, AR- C67085, AR-C66096, AR-C69931, CT50547, BX-667, MRS2395, PSB0739	P2Y12R antagonists are in widespread clinical use as inhibitors of platelet aggregation, and P2Y12R activation can regulate glial cell migration and increase cell proliferation
P2Y13	Endogenous: ADP Synthetic: 2-MeS-ADP, 2-MeS-ATP, ADPβS, BzATP	PPADS, Suramin, Reactive blue 2, Ap4A, AR-C69931, AR-C67085, MRS2211, MRS2603	P2Y13R activation enhances glycine transport in the synaptic cleft, and promotes cell survival through a PI3K/Akt-dependent mechanism
P2Y14	Endogenous: UDP-glucose Other: UDP-galactose, UDP-glucuronic acid, UDP-N-acetylglucosamine Synthetic: MRS2690		P2Y14R activation can modulate inflammatory responses through chemokine and cytokine production, and may play a role in muscle contraction

The EC₅₀ values for endogenous agonists, the chemical names of agonists and antagonists and the associated references are given in the main text.

COMMON STRUCTURAL AND SIGNALING FEATURES OF P2Y RECEPTORS

The primary and secondary structures of P2YRs indicate the presence of an extracellular N-terminus followed by seven transmembrane spanning domains that delineate 3 extracellular loops and 3 intracellular loops and an intracellular C-terminal tail that varies significantly among the P2YR subtypes [3, 26-29]. The number of amino acids in human P2YR subtypes ranges from 328 (for P2Y₆R) to 377 (for P2Y₂R) and these subtypes have ~20 to 50% homology [30]. Binding of the negatively charged P2YR agonists within the transmembrane domain is likely facilitated by several positively charged amino acids that are conserved among the P2YR subtypes [27, 31, 32] and information from site-directed mutagenesis experiments has suggested a model of the ligand-binding site for some subtypes [33, 34]. Once activated by nucleotide agonists, P2YRs stimulate GDP-GTP exchange by heterotrimeric G proteins whereupon dissociation of the GTP-Gα subunit modulates the activities of intracellular enzymes, including phospholipase C and adenylyl cyclase, to activate intracellular signaling cascades that regulate multiple cell type-specific responses [3, 35]. Among the P2YR subtypes, the P2Y₁, P2Y₂, P2Y₄, P2Y₆ and P2Y₁₁ receptors are coupled to G_q protein and the activation of phospholipase C which promotes the hydrolysis of the plasma membrane phospholipid phosphatidylinositol 4,5-bisphosphate (PIP₂) to generate two second messengers, inositol 1,4,5-trisphosphate (IP₃) and diacylglycerol that induce release of Ca²⁺ from intracellular stores and activate protein kinase C (PKC), respectively [3, 36]. Alternatively, P2Y₁₂, P2Y₁₃ and P2Y₁₄ receptors are coupled to G_i/o proteins which regulate the activity of adenylyl cyclase and production of the downstream effector cyclic adenosine 5’-monophosphate (cAMP) [3, 36]. P2YRs have also been shown to activate additional subfamilies of G proteins either directly or indirectly through interactions with other plasma membrane receptors, as described below.

P2YRs, like most G protein-coupled receptors, are desensitized by prolonged agonist treatment due to phosphorylation of intracellular sites in the receptor that promotes the binding of β-arrestins and subsequent receptor internalization and/or degradation [37]. β-Arrestins can regulate receptor desensitization by interfering with G protein coupling and interacting with clathrin and AP2 to induce endocytosis [37]. Most P2YRs interact with β-arrestin-2, although the P2Y₂ and P2Y₄ receptors also interact with β-arrestin 1 [38]. Receptor endocytosis regulated by β-arrestins has been shown to activate intracellular mitogen-activated protein kinases [37], such as ERK1/2, a target of many P2YR signaling cascades, and differential coupling of P2YRs to β-arrestins may explain variations in ERK1/2 activation induced by different P2YR agonists [38]. Agonist-induced desensitization of most P2YRs, except for the P2Y₂R, also is dependent on the protein dynamin [38].

P2Y RECEPTOR SUBTYPES

The P2Y₁ Receptor

The P2Y₁R is activated by the endogenous agonist ADP (EC₅₀ = 10 nM) and is directly coupled to activation of G_q proteins [39]. When the human P2Y₁R was overexpressed in Sf9 insect cells, then purified and reconstituted in proteoliposomes along with heterotrimeric G proteins, it was demonstrated that receptor activation was also coupled to activation of Gα₁₁ proteins (determined by increases in 2-methylthio-ADP-induced GTP hydrolysis) whose activity was enhanced by the presence of the GTPase-activating proteins RGS4 and PLCβ₁ [40]. The synthetic agonist 2-methylthio-ADP (2-MeS-ADP; EC₅₀ = 2 nM) has been widely used to induce P2Y₁R activation, although it can also activate the P2Y₁₂ and P2Y₁₃ receptors [41]. The 2-MeS-ADP analog (N)-methanocarba-2-methylthio-ADP (MRS2365) was found to be a highly potent and stable agonist at the human P2Y₁R with an EC₅₀ = 0.4 nM [42]. Other studies have synthesized derivatives of adenosine 5'-O-(1-boranotriphosphate) (ATP-α-B) and β,γ-methyl-ATP that are P2Y₁R agonists and are relatively insensitive to hydrolysis by nucleoside 5’-triphosphate diphosphohydrolase (NTPDase), as compared to endogenous agonists, suggesting their potential as therapeutic agonists with long-term stability [43, 44]. Similarly, dinucleoside polyphosphate derivatives including diadenosine (γ-borano)pentaphosphate (Ap₅(γ-B)A) have proven to be potent and selective P2Y₁R agonists [45, 46]. Di-(2-MeS)-adenosine 5',5"-P¹,P⁴,α,β-methylene-tetraphosphate and 2-MeS-β,γ-CCl₂-ATP also have been shown to be stable P2Y₁R agonists with potential for the treatment of human disease due to their ability to stimulate insulin secretion in rats and relieve intraocular pressure in rabbits, respectively [47, 48].

P2Y₁R activation has been shown to be antagonized by the ATP analogs, ATPaS and β,γ-methylene-ATP, which act as competitive antagonists, whereas pyridoxal-phosphate-6-azophenyl-2',4'-disulphonic acid (PPADS) and suramin were also shown to inhibit P2Y₁R activity [49], although these latter compounds are not selective for the P2Y₁R [50]. Adenosine-3'-phosphate-5'-phosphosulfate (A3P5PS), adenosine-3'-phosphate-5'-phosphate (A3P5P), adenosine-2'-phosphate-5'-phosphate (A2P5P) and adenosine-2',5'-diphosphate are competitive antagonists at the turkey and human P2Y₁Rs and A3P5PS and A3P5P were devoid of agonist or antagonist activities at the P2Y_2, P2Y₄ and P2Y₆ receptors [51, 52]. These initial observations describing P2Y₁R antagonism by bisphosphates (i.e., A3P5P) led to the development of synthetic analogs such as N6-methyl 2'-deoxyadenosine 3',5'-bisphosphate (N6MABP; or MRS2179) which also antagonizes the P2Y₁R without effects on the human P2Y₂, P2Y₄ and P2Y₆ receptors [53, 54]. Additional synthetic bisphosphate antagonists of the P2Y₁R include 2-iodo-N(6)-methyl-(N)-methanocarba-2'-deoxyadenosine-3',5'-bisphosphate (MRS2500), which was shown to be a potent and selective P2Y₁R antagonist [55, 56], and 2-chloro N(6)-methyl-(N)-methanocarba-2'-deoxyadenosine-3',5'-bisphosphate (MRS2279), which was shown to be a selective competitive P2Y₁R antagonist without an effect on P2Y₂, P2Y₄, P2Y₆, P2Y₁₁ and P2Y₁₂ receptors [57]. Other studies have indicated that, whereas N6-methyl modification enhances the antagonist potency of 2'-deoxyadenosine 3',5'-bisphosphate at the P2Y₁R, this effect is inhibited by benzoylation or dimethylation of the N6-amino group of adenosine or by its replacement with methylthio, chloro, or hydroxy groups [58].

The P2Y₁R is widely expressed in the mammalian brain, including in the cerebral cortex, hippocampus, caudate nucleus, putamen, globus pallidus, habenula, subthalamic nucleus, midbrain and cerebellum [59-61]. The P2Y₁R is expressed in Purkinje cells in the cerebellum, in regions of the cerebral cortex, in ischemia-sensitive areas of the hippocampus [61] and in oligodendrocytes and astrocytes in brain and optic nerves [14, 59]. P2Y₁Rs also have been shown to regulate glial cell functions [61]. In hippocampal astrocytes, P2Y₁R activation has been suggested to provide neuroprotection from oxidative stress via increased interleukin-6 (IL-6) release [62] and to play roles in brain development and repair [63] and sensory reception [64, 65]. Other studies indicate that P2Y₁Rs also are expressed in microglial cells [59, 66], rat neuroprogenitor cells [63], and dorsal root ganglia and horn neurons [65, 67, 68]. Furthermore, recent studies with P2Y₁R knockout mice indicate that P2Y₁Rs mediate neurotransmission in the gastrointestinal tract [69].

Consistent with studies indicating the role of the P2Y₁ receptor in increasing cytokine (i.e., IL-6) release [62], intracerebroventricular administration of the P2Y₁R antagonist, MRS2179, significantly decreased the expression of IL-6, phospho-RelA (p-RelA), tumor necrosis factor-α, monocyte chemotactic protein-1/chemokine (C-C motif) ligand 2 (CCL2), and interferon-inducible protein-10/chemokine (C-X-C motif) ligand 10 (CXCL10) mRNA in a rat model of cerebral ischemia/reperfusion [25]. While previous studies indicated that P2Y₁R-induced IL-6 release provided neuroprotection [62], intracerebroventricular administration of the P2Y₁R agonist MRS2365 was shown to increase cerebral infarct volume due to cerebral ischemia/reperfusion, whereas administration of the P2Y₁R antagonist MRS2179 decreased infarct volume [25]. Additionally, P2Y₁R and p-RelA colocalized with glial fibrillary acidic protein-positive astrocytes, suggesting that P2Y₁R expression in cortical astrocytes mediates cytokine/chemokine-induced damage that occurs during cerebral ischemia/reperfusion, which can be prevented by antagonists of the P2Y₁R [25]. The effects of P2Y₁R agonism/antagonism appear to be experimental model-dependent making it difficult to determine the potential therapeutic role in treatment of ischemia-related damage. In brain sections from Alzheimer’s disease (AD) patients, P2Y₁Rs in neurons have been shown to colocalize with neurofibrillary tangles and neuritic plaques, as compared to samples from control patients [70]. While the significance of these observations is unclear, P2Y₁R-mediated stimulation of G_q protein has been shown to activate the small GTPase cytoskeletal modulator Rac [71], which recent studies have shown enhances axonal elongation [72], although this effect was shown to occur through a mechanism involving adenylate cyclase 5 and the phosphatidylinositol 3-kinase (PI3K)/Akt pathway. Nonetheless, these data suggest that the P2Y₁R may play a neuroprotective role in AD by promoting axonal elongation to counteract the neurotoxic effects of neurofibrillary tangles. While further research is needed to validate this speculative therapeutic pathway, it stands to reason that selective P2Y₁R agonists and antagonists should be investigated for therapeutic utility in the modulation of neurodegenerative disease phenotypes, such as AD, amyotrophic lateral sclerosis (ALS) and Parkinson’s disease.

P2Y₁R activation also has been shown to cause PKC-dependent phosphorylation of the capsaicin receptor (a VR1 cation channel) that can alter the perception of pain [73]. The potential therapeutic use of P2Y₁R antagonists in pain perception is highlighted by a recent study showing significantly decreased tactile allodynia in a rat model of bone pain following treatment with the P2Y₁R antagonist MRS2179 [74]. Additionally, intracerebroventricular injection of the P2Y₁R agonist ADPβS was shown to increase anxiolytic-like behavior in a rat model of anxiety [75]. Furthermore, this behavior was attenuated by injection of the P2Y₁R antagonist MRS2179, suggesting another possible therapeutic pathway that may be exploited using specific P2Y₁R antagonists.

The P2Y₂ Receptor

The P2Y₂R is activated equipotently by the fully ionized forms of ATP or UTP (i.e., by NTP⁴⁻) with an EC₅₀ ~ 0.5-3 µM [76-79], suggesting that this P2YR subtype is primarily activated in close proximity to released nucleotides or under conditions associated with high levels of extracellular ATP and UTP (e.g., inflammation and apoptosis). It seems likely that the negatively-charged triphosphate moieties of these structurally distinct purine (i.e., ATP) and pyrimidine (i.e., UTP) nucleotides promote agonist binding to the P2Y₂R, since selective deletion of positively charged amino acids in the 6^th and 7^th transmembrane domains inhibits P2Y₂R activity [27]. Uridine-5'-O-(3-thio)triphosphate (UTPyS) has been shown to be equally potent to UTP as a P2Y₂R agonist and relatively resistant to hydrolysis by alkaline or acid phosphatase and apyrase [78]. Diadenosine tetraphosphate (Ap₄A) displays relatively potent (EC₅₀ = 720 nM) agonist activity at the P2Y₂R and has been suggested to be an endogenous agonist [80]. The observation that dinucleoside polyphosphates have agonist activity at P2Y receptors led to pharmacological investigations of similar compounds, such as P¹,P⁴-di(uridine-5’-) tetraphosphate (Up₄U; INS365; Diquafosol), which was shown to be a potent agonist of the P2Y₂R (EC₅₀ = 100 nM) and also displayed agonist activity at the P2Y₄R and the P2Y₆R [81]. Similarly, P1-(inosine 5’-)P⁴-(uridine 5’-) tetraphosphate (Ip₄U; INS45973), which has EC₅₀ values for the P2Y₂R and the P2Y₄R of ~ 280 nM and an EC₅₀ for the P2Y₆R of >10 µM, has been used intravenously in mice to demonstrate that activation of the P2Y₂R decreases blood pressure and increases renal Na⁺ excretion [82]. The relative potency and selectivity of these compounds along with the development of the more hydrolytically stable P2Y₂R agonist P(1)-(uridine 5')-P(4)-(2'-deoxycytidine 5')-tetraphosphate (dCp₄U; INS37217; Denusofol) [83, 84] led to several clinical investigations into the therapeutic potential of these P2Y₂R agonists to increase fluid flow across epithelial cell membranes that has been shown to be regulated by the P2Y₂R [85-88]. However, recent Phase III clinical trials for INS37217 in the treatment of cystic fibrosis and INS365 in the treatment of dry eye, two epithelial-based diseases with underlying defects in fluid flow, failed to achieve their primary endpoints. Other documented P2Y₂R agonists include 2'-amino-2'-deoxy-UTP, 6-nitro-UTP, UTPaS and 2'-deoxy-UTPaS [89], 2-thio-UTP [90], uridine-5'-tetraphosphate 8-phenyl ester (MRS2768), dihalomethylene phosphonate analogues and the 2-thio analogue of INS37217 [91]. A recently developed P2Y₂R agonist, 4-thiouridine-5'-O-(β,γ-difluoromethylene) triphosphate (PSB1114), shows 60-fold increased selectivity at the P2Y₂R, as compared to the P2Y₄R or the P2Y₆R [92], which potentially provides an agent for selective in vivo activation of the P2Y₂R among the known uridine nucleotide receptor subtypes.

P2Y₂R activation mediates G_q-dependent stimulation of phospholipase C (PLC), although the P2Y₂R has been shown to induce GTPy[³⁵S] incorporation into G_q, G_o and G₁₂ proteins [93-94]. Only G_q appears to be directly coupled to P2Y₂R activation, since activation of G_o and G₁₂ proteins was shown to require the presence of the consensus integrin-binding arginine-glycine-aspartic acid (RGD) motif in the first extracellular loop of the P2Y₂R whose deletion prevents receptor interaction with α_v integrins and the activation G_o and G₁₂, but not G_q proteins [93-95]. Although activation of PLC by the P2Y₂R has been shown to require GTP binding to Gα_q11 [96], the P2Y₂R also has been found to modulate phospholipase C activities via Gα_q16 [97] and Gβγ_i3 [96].

There have been few P2Y₂R-selective antagonists described, although PPADS and suramin are non-selective P2Y₂R antagonists [98, 99]. The non-nucleotide compound 1-amino-4-(2-methoxyphenyl)-2-sulfoanthraquinone (PSB716) has been shown to be a potent P2Y₂R antagonist with an IC₅₀ value in the low micromolar range [100]. To date, the most reliable means of inhibiting P2Y₂R function has been the use of P2Y₂R-selective antisense oligonucleo-tides or siRNA or deletion of the P2Y₂R in transgenic mice [24, 101, 102]. Accordingly, there is a critical need for the development of P2Y₂R-selective antagonists and agonists, particularly those that can cross the blood-brain barrier, since this P2YR subtype represents a promising target in the treatment of neuroinflammatory and neurodegenerative diseases, such as Alzheimer’s disease, as described below.

The P2Y₂R subtype likely plays a role in CNS functions predominantly under pathophysiological conditions, including inflammation and bacterial infection [103-106]. The P2Y₂R has been found to be upregulated in cell and animal models under a variety of conditions associated with inflammation or injury [16, 107-111], including spinal cord injury [112] and brain trauma [113], which suggests the possibility that the P2Y₂R plays a protective role in the CNS. P2Y₂R expression is upregulated by treatment of rat primary cortical neurons with the proinflammatory cytokine interleukin-1β (IL-1β) [16], and levels of IL-1β have been shown to be elevated in the brains of AD patients, as compared to normal controls [114, 115]. Consistent with a role in inflammation, P2Y₂R expression under proinflammatory conditions has been shown to require binding to the P2Y₂R promoter of NF-κB [116], a transcription factor known to regulate gene expression in inflammation [117].

P2Y₂R activation has been shown to increase the migration and proliferation of astrocytic and microglial cells [24, 93, 118-120]. The mechanism of P2Y₂R-mediated cell migration appears to require more than G_q-coupled activation of PLC, since cytoskeletal rearrangements and cell migration induced by the P2Y₂R agonist UTP were abolished in astrocytoma cells that express a mutant P2Y₂R in which aspartic acid in a consensus integrin-binding RGD motif was modified to glutamic acid to generate RGE, which prevents the P2Y₂R from binding to α_v integrins [93-94]. This result suggests a requirement for P2Y₂R interaction with α_v integrins to mediate UTP-induced cell migration. Consistent with this conclusion, UTP-induced G_o-dependent Rac1 and G₁₂-dependent RhoA activation, pathways well known to regulate actin polymerization and depolymerization required for cell migration [121, 122], were absent in cells expressing the RGE-mutant of the P2Y₂R [93-95]. Other data suggest that increases in cell proliferation caused by P2Y₂R activation are due to the ability of the P2Y₂R to induce phosphorylation and activation of growth factor receptors that leads to increases in the activities of the mitogen-activated protein kinases ERK1/2 and the related adhesion focal tyrosine kinase (RAFTK; Pyk 2) via a pathway dependent upon Src and Shc/Grb2 [123-125]. The P2Y₂R contains Src-homology-3 (SH3) binding motifs in the intracellular C-terminal domain that when deleted can prevent growth factor receptor activation [125, 126], suggesting that agonist-induced Src binding to these SH3-binding sites can regulate P2Y₂R-mediated cell proliferation. Other studies indicate that the C-terminal domain of the P2Y₂R can interact with the actin-binding protein filamin A, a regulator of cytoskeletal rearrangements [101], and the C-terminal domain also has been shown to play a role in P2Y₂R desensitization and internalization [127]. The P2Y₂R also can regulate cell migration by activation of growth factor receptors [128], suggesting that a wide variety of cellular functions are likely mediated by P2Y₂R interactions with integrins, growth factor receptors and cytoskeletal proteins well beyond the ability of the P2Y₂R to mediate the G_q-dependent activation of PLC.

Current data indicate that P2Y₂R expression is relatively low in neurons, but can be upregulated by the proinflammatory cytokine IL-1β [16]. Increased P2Y₂R expression in neurons is likely to have neuroprotective effects, since P2Y₂R activation promotes neurite outgrowth [129] and the α-secretase-dependent degradation of amyloid precursor protein (APP) to generate the non-amyloidogenic soluble APPα peptide, rather than the neurotoxic Aβ_1-42 peptide associated with the pathophysiology of Alzheimer’s disease [16, 130]. In mouse primary microglial cells, the P2Y₂R is upregulated by Aβ_1-42, whereupon P2Y₂R activation increases the phagocytosis and degradation of neurotoxic forms of Aβ [24, 105, 131]. Interestingly, studies have shown that P2Y₂R expression is decreased in the parietal cortex of AD patients and this loss of P2Y₂R expression was correlated with neuropathology and synapse loss [132]. These observations suggest that the loss of neuroprotective functions provided by the P2Y₂R may contribute to disease pathogenesis in AD. Other potentially neuroprotective effects mediated by the P2Y₂R include the regulation of synaptic transmission through the induction of intracellular calcium waves in astrocytic cells [133] and the upregulation of anti-apoptotic protein expression which promotes cell survival [134]. Activation of P2Y₂Rs also has been shown to resensitize ionotropic P2X₂ receptors in bladder sensory neurons [135] and vanilloid type 1 channels (TRPV1) in kidney sensory neurons [136]. The ability of the P2Y₂R to activate metalloproteases, including TNF-α converting enzyme (TACE; or ADAM17) [16], may contribute to the AD phenotype through the production of TNFα, which has been shown to be upregulated in the brain and cerebrospinal fluid of AD patients [137, 138]. Taken together, the P2Y₂R represents a promising target to promote neuroprotection in neurodegenerative diseases through the combined activation of P2Y₂Rs in astrocytes, microglial cells and neurons.

The development of neurodegenerative diseases has been shown to be preceded by malfunctions in the cerebrovascular system, including decreased blood flow in the brain and breakdown of the blood-brain barrier (BBB) [139-141]. It is known that microvascular cells (i.e., endothelium and pericytes) affect neuronal functions by altering blood flow, BBB function and glucose/nutrient supply [140, 141]. These cells are also responsible for the clearance of toxic molecules and the secretion of trophic factors and extracellular matrix molecules required for proper neuronal functioning [140, 142]. It is, therefore, important to consider cerebrovascular function in relation to brain health.

Studies in various vasculature systems, including the cerebral microvasculature, have shown that intravenous or intraluminal application of ATP and/or UTP induces endothelium-dependent relaxation of blood vessels and a decrease in blood pressure, which has been attributed to activation of the endothelial P2Y₂R [82, 143-148]. Fewer in vivo studies have explored the effects of nucleotides on vascular permeability properties, although studies in frogs have shown that ATP increases the ion permeability of cerebral venules [149] and mesenteric microvascular permeability as measured by changes in hydraulic conductivity [150]. P2Y₂Rs, which are expressed along with P2Y_1,4,6 receptors in mouse aortic endothelium [151], increase blood flow through the release of NO [145]. In human mammary arteries, several P2Y receptor subtypes, including P2Y_1,2,4,6 receptors, are thought to mediate vasodilatation by releasing NO, prostanoids or endothelium-derived hyperpolarizing factor [152]. In contrast, vascular smooth muscle cells express many P2X and P2Y receptors [153] that cause blood vessel constriction, which together help regulate vascular tone.

Cerebral microvessels are closely associated with several types of brain cells, including microglia and neuronal processes, and are encircled by the end feet of astrocytes, which are also thought to play a role in the regulation of local blood flow and barrier function [154]. Immunohistochemistry studies in rat brain slices indicate that the P2Y₂R and the P2Y₄R are strongly expressed in astrocytic end feet at the gliovascular interface and application of ATP to freshly prepared brain slices triggers calcium waves in the astrocytic end feet that are propagated along the vessel wall, which the authors speculate is important for regulation of vascular permeability and tone [155]. Furthermore, Lewis et al., demonstrated in rat cerebral tissue that exogenous or extraluminal application of ATP evokes transient vasoconstriction of arterioles followed by sustained vasodilation, whereas UTP and UDP evoke sustained vasoconstriction [156]. Thus, the type of nucleotide and the route of administration (intravenous vs extraluminal) cause profound differences in vascular responses that are important to consider for drug development.

The P2Y₄ Receptor

The human P2Y₄R is activated by UTP (EC₅₀ = 73 nM), but not by ATP, whereas the rat and mouse P2Y₄Rs are activated equipotently by ATP and UTP, similar to the P2Y₂R [26, 89, 157-159]. Molecular docking studies indicate that the agonist binding site for the P2Y₄R is similar to the P2Y₂R with some variation in the second transmembrane domain and the second extracellular loop [89]. Activation of the P2Y₄R has been shown to couple to G_q- and G_o-dependent stimulation of PLCβ [26, 160] and the activation of Rho [161]. ATP has been shown to act as an antagonist at the human P2Y₄R [160] and amino acid variations in the second extracellular loop and the N-terminal domain were found to play a role in determining whether ATP functions as an agonist or antagonist at mammalian P2Y₄Rs [162]. In addition to UTP and ATP, P2Y₄R agonists include UTPγS [163], 5-bromo-UTP [26], INS365, INS37217, and INS45973 [81-83]. However, all five of these compounds also display agonist activity at other P2Y receptors, especially the P2Y₂R. To date, selective P2Y₄R agonists have not been generated, although the compound 2’-azido-2’-deoxy-UTP has been shown to display slight selectivity for the P2Y₄R over the P2Y₂R (human P2Y₄R: EC₅₀ = 1 μM; human P2Y₂R: EC₅₀ = 5 μM) [89]. Similarly, definitive selective P2Y₄R antagonists have not been generated, although the non-specific P2Y receptor inhibitor PPADS was shown to potently inhibit the UTP response at the human P2Y₄R [160].

While little is known about the physiological or pathological effects of P2Y₄R activation in the CNS, P2Y₄R mRNA has been shown to be expressed in human brain [164] and in rat hippocampal pyramidal neurons where it has been suggested to have a presynaptic inhibitory role on the release of glutamate [21]. P2Y₄R expression in astrocytes and microglial cells also has been extensively documented [14, 66, 157, 165], however, the physiological role of the receptor is unclear and may be complementary to the P2Y₂R. P2Y₄Rs are expressed in glial end feet in proximity to blood vessel walls, similar to P2Y₂Rs [155], where their activation has been suggested to regulate BBB function, blood flow, metabolic trafficking and water homeostasis [155, 166]. The P2Y₄R also has been shown to inhibit K⁺ currents in myocytes of rat cerebral arteries in a Rho-dependent manner [161].

The P2Y₆ Receptor

The P2Y₆R is activated preferentially by UDP (EC₅₀ = 15 nM) and to a lesser extent by UTP [167]. UDPβS has been shown to be a more stable P2Y₆R agonist, although this compound also displays agonist activity at the P2Y₁₄R [163]. The synthetic compound 5-iodo-UDP (MRS2693) has been shown to be a selective agonist for the human P2Y₆R [168]. Interestingly, intraperitoneal injection of MRS2693 was shown to have a protective effect on skeletal muscle in a mouse model of ischemia [169]. The compound 3-phenacyl-UDP (PSB0474) is a relatively selective agonist of the P2Y₆R, as compared to the P2Y₂R and the P2Y₄R [90]. The diuridine triphosphate analog INS48823 also has been shown to be a potent and selective P2Y₆R agonist (EC₅₀ = 200 nM) [81] that, similar to INS365 and INS37217, can increase fluid flow across epithelial cell membranes [170]. Additionally, INS48823 has been shown to activate the NF-κB pathway in osteoclasts to increase cell survival [171]. Other P2Y₆R agonists include α,β-methylene-UDP [91] and 5-bromo-UTP [167]. The P2Y₆R activates PLCβ in a G_q-dependent manner [172], but has been shown to regulate pertussis toxin (PTX)-sensitive, voltage-gated Ca²⁺ currents in rat sympathetic neurons, suggesting the involvement of G_i/o [173]. The P2Y₆R also can mediate G_12/13-dependent Rho activation, similar to the P2Y₂R, although the involvement of integrins in this pathway has not been determined [174].

Diisothiocyanate derivatives of 1,2-diphenylethane (MRS2567) and 1,4-di-(phenylthioureido)butane (MRS2578) have been demonstrated to be potent and selective antagonists at the rat and human P2Y₆R (IC₅₀ values ~ 40-130 nM), with no effect on responses induced by agonists of P2Y₁, P2Y₂, P2Y₄ and P2Y₁₁ receptors [175]. A derivative of 1,4-phenylendiisothiocyanate (MRS2575) also displays P2Y₆R antagonist activity, but only at the human P2Y₆R [175]. Among the non-selective P2YR antagonists, reactive blue 2 was shown to be more potent than PPADS and suramin at the P2Y₆R [176].

P2Y₆R mRNA is expressed throughout the human brain, but at highest levels in amygdala, cingulate gyrus, nucleus accumbens and putamen [164]. P2Y₆R mRNA also has been identified in mouse superior cervical ganglion [177], rat dorsal-root ganglion neurons [65, 67] and rat pyramidal hippocampal neurons [21]. Astrocytes in the cerebellum and cortex of rats also have P2Y₆R activity [14, 178]. Similar to the P2Y₂R, P2Y₆R activation has been shown to increase the phagocytotic activity of microglia [179, 180], a response that was inhibited by the P2Y₆R antagonist MRS2578. These observations suggest that a range of endogenous nucleotides and synthetic agonists/antagonists that act at different P2YR subtypes can be exploited therapeutically to induce neuroprotective responses, such as the microglia-dependent phagocytosis of neurotoxic forms of Aβ [24, 131]. Injury can enhance P2Y₆R expression in rat astroglial cells [113] and P2Y₆Rs in microglial cells are activated in response to bacterial lipopolysaccharide, implicating a role in neuroinflammation [66]. Thus, these data suggest that the P2Y₆R, along with the other uridine nucleotide receptors, (i.e., P2Y₂R and P2Y₄R) can regulate protective mechanisms in the CNS under a variety of pathophysiological conditions. Other studies indicate that P2Y₄ and P2Y₆ receptors form homo- and hetero-oligomeric complexes in neuronal cells and dimeric P2Y₄Rs and monomeric P2Y₆Rs can partition into lipid rafts in synaptosomes [181], suggesting the possibility of complex P2YR interactions in the CNS.

The P2Y₁₁ Receptor

The human P2Y₁₁R is activated by ATP and leads to the activation of both PLC (EC₅₀ for IP₃ production = 65 µM) and adenylyl cyclase (EC₅₀ for cAMP production = 17 µM), although ATPyS and 3’-O-(4-benzoyl)benzoyl adenosine 5’-triphosphate (BzATP) are more potent agonists than ATP [182]. UTP, in addition to ATP, has been shown to activate the P2Y₁₁R expressed in astrocytoma cells [183]. Interestingly, the ATP analogue 2-propylthio-β,γ-dichloromethylene-D-ATP (AR-C67085), which had previously been shown to inhibit platelet aggregation through P2Y₁₂R antagonism [184], was shown to be a potent P2Y₁₁R agonist (EC₅₀ for IP₃ production = 9 µM; EC₅₀ for cAMP production = 1.5 µM) [182]. The non-nucleotide phosphoric acid derivative 4,4'-(carbonylbis(imino-3,1-phenylene-carbonylimino-3,1-(4-methyl-phenylene)carbony-limino))-bis1,3-xylene-α,α'-diphosphonic acid (NF546) is a selective P2Y₁₁R agonist that has been shown to stimulate IL-8 release from human dendritic cells with similar potency as ATPγS [185]. Nicotinic adenine dinucleotide (NAD+) and nicotinic adenine dinucleotide phosphate (NADP+) have been shown to be agonists of P2Y₁₁Rs expressed in human granulocytes and astrocytoma cells, respectively [186, 187]. The P2Y₁₁R can regulate both G_s-dependent activation of adenylyl cyclase and G_q-dependent activation of PLC [183, 188-190]. P2Y₁₁Rs expressed in astrocytoma cells mediate UTP-induced PLC activation by a pertussis toxin (PTX)-sensitive mechanism and ATP-induced PLC activation by a PTX-insensitive mechanism, indicating coupling to PLC via both G_o and G_q proteins [183]. The coupling of the P2Y₁₁R to multiple G proteins may be due to the ability of the P2Y₁₁R to form heterodimers with other P2YRs, such as the P2Y₁R [191].

The nucleotide analogue AMPaS (5’-AMPS) has been shown to be an effective P2Y₁₁R antagonist [182, 192]. Suramin was a more potent antagonist of the P2Y₁₁R than reactive blue 2, whereas PPADS was completely inactive [182], and the antagonist potency and selectivity of suramin at the P2Y₁₁R over P2Y₁, P2Y₂ and several ionotropic P2X receptors could be further increased by substitution of methyl groups in suramin with fluorine [193]. The resulting suramin derivative, 8,8'-(carbonylbis(imino-3,1-phenylene-carbonylimino(4-fluoro-3,1-phenylene)carbonylimino))bis-1,3,5-naphthalenetrisulfonic acid (NF157), was shown to prevent NAD+-induced human granulocyte migration through P2Y₁₁R antagonism [187]. Another non-nucleotide antagonist of the P2Y₁₁R, 4,4’-(carbonylbis(imino-3,1-(4-methyl-phenylene)carbonylimino))bis-naphthalene-2,6-disulfonic acid (NF340), showed relative selectivity at the P2Y₁₁R over the P2Y₁, P2Y₂, P2Y₄, P2Y₆, P2Y₁₂, and P2X1-3 receptors [185].

Although the rodent P2Y₁₁R gene has not been cloned, studies have suggested that P2Y₁₁R mRNA is expressed in the nucleus accumbens, parahippocampal gyrus, putamen and striatum of rats [1]. Additional studies using human P2Y₁₁R primers for PCR and anti-human P2Y₁₁R antibodies have localized the P2Y₁₁R to rat hippocampal pyramidal neurons and Purkinje cells in the adult rat cerebellum [21, 194]. Since a rodent P2Y₁₁R gene has not been cloned, it is difficult to conclude that rodents express a P2Y₁₁R. In human neutrophils, P2Y₁₁R activation has been shown to delay pathogen- or inflammatory mediator-induced apoptosis by a cAMP-dependent mechanism, suggesting a protective role for the P2Y₁₁R under neuroinflammatory conditions [190]. Similarly, in human monocytes, P2Y₁₁R activation inhibits toll-like receptor signaling by increasing cAMP production [189]. P2Y₁₁Rs in monocyte-derived dendritic cells also have been shown to regulate thrombospondin-1 secretion and inhibition of lipopolysaccharide-stimulated interleukin-12 release [185]. Thus, P2Y₁₁Rs represent a promising therapeutic target for the treatment of neuroinflammatory diseases.

The P2Y₁₂ Receptor

The P2Y₁₂ receptor is activated by the endogenous agonist ADP (EC₅₀ = 60 nM), whereas 2-MeS-ATP and 2-MeS-ADP are more potent agonists [195]. Activation of the P2Y₁₂R couples to both G_i protein [196, 197] and G_o protein [196, 198]. Vesicle reconstitution of the P2Y₁₂R with different G proteins demonstrated that the P2Y₁₂R couples more effectively to Gα_i2 than to Gα_i1 and Gα_i3, but does not couple to Gα_o or Gα_q proteins [199], suggesting that G_o coupling may be due to formation of heteromeric receptors with other P2YRs.

A wide variety of P2Y₁₂R antagonists have been described, due to the central role of the P2Y₁₂R in the initiation of platelet aggregation [200-203]. The thienopyridine-ADP family of P2Y₁₂R antagonists include clopidogrel (methyl (+)-(S)-α-(2-chlorophenyl)-6,7-dihydro-thieno[3,2-c]pyridine-5(4H)acetate sulfate; Plavix), ticlopidine (5-[(2-chlorophenyl)methyl]-4,5,6,7-tetrahydrothieno[3, 2-c] pyridine hydrochloride; Ticlid) and prasugrel (5-[(1RS)-2-cyclopropyl-1-(2-fluorophenyl)-2-oxoethyl]-4,5,6,7-tetrahydrothieno[3,2-c]pyridin-2-yl acetate hydrochloride; LY640 315; CS-747; Effient) [204, 205]. Ticlopidine was discovered over 30 years ago [204, 206] and the anti-platelet aggregation properties of clopidogrel were found decades before the cloning and identification of its endogenous target, the P2Y₁₂R [41, 207]. Following conversion to their respective active thiol derivatives in the liver, these orally active antagonists irreversibly bind to cysteine residues in the P2Y₁₂R and prevent ADP-induced platelet aggregation. In the case of clopidogrel, binding to the receptor disrupts the formation of lipid raft-associated P2Y₁₂R oligomers, the speculated functional form of the receptor, suggesting a possible mechanism of antagonism [208]. Therapeutic use of Plavix, Ticlid and Effient has been FDA-approved for prevention of a number of cardiovascular pathologies, including myocardial infarction, stroke, and peripheral artery disease [204].

Observations that ATP can act as a P2Y₁₂R antagonist [209] led to investigations of the antagonist activity of stable ATP analogs [41]. The ATP analog family of competitive P2Y₁₂R antagonists includes the AR compounds AR-C67085, AR-C66096 (2-(propylthio)adenosine-5'-O-(β,γ-di-fluoromethylene)triphosphate), AR-C69931 (Cangrelor; N6-(2-methylthioethyl)-2-(3,3,3-trifluoropropylthio)-5'-adenylic acid) [184, 210, 211] and the orally active non-phosphorylated AstraZeneca compound AZD6140 (Ticagrelor; Brillinta; (1S,2S,3R,5S)-3-[7-[[(1R,2S)-2-(3,4-difluorophenyl) cyclopropyl]amino]-5-(propylthio)-3H-1,2,3-triazolo[4,5-d]pyrimidin-3-yl]-5-(2-hydroxyethoxy)-1,2-cyclopentanediol) [212]. Similar to AZD6140, the orally active drug PRT-060128 (Elinogrel; N-[(5-chlorothiophen-2-yl)sulfonyl]-N’-(4-[6-fluoro-7-(methylamino)-2, 4-dioxo-1, 4-dihydroquinazolin-3(2H)-yl]phenyl)urea) was shown to be an effective reversible antagonist of P2Y₁₂R [213]. While Elinogrel failed to make it out of Phase II clinical trials, Brillinta (AZD6140) has been approved by the FDA for prevention of adverse cardiovascular events. Additional P2Y₁₂R antagonists include the tricyclic benzothiazolo[2,3-c]thiadiazine derivative CT50547 [214], BX-667 ((S)-4-((4-[1-(ethoxycarbonyl)-1-methylethoxy]-7-methyl-2-quinolyl)carbamoyl)-5-[4-(ethoxycarbonyl) piperazin-1-yl]-5-oxopentanoic acid) [215], MRS2395 (2,2-dimethyl-propionic acid 3-(2-chloro-6-methylaminopurin-9-yl)-2-(2,2-dimethyl-propionyloxymethyl)-propylester) [216], and the reactive blue 2 analog PSB0739 (1-amino-4-[4-phenylamino-3-sulfophenylamino]-9,10-dioxo-9,10-dihydroanthracene-2-sulfonate) [217].

Besides in platelets, the P2Y₁₂R is highly expressed in astrocytes [207, 218], including in the rat nucleus accumbens of the cortex and cerebellum [14, 113, 219] and in rat hippocampal pyramidal neurons [21]. P2Y₁₂Rs have also been shown to be expressed in rat brain microglial cells where they are suggested to regulate microglial cell migration to neurons in response to injury [220] and directional migration of glial cell processes to axons during premyelination [221]. Accordingly, transgenic mice with selective deletion of the P2Y₁₂R show reduced directional microglial branch extension in vivo, as compared to control mice who exhibited significantly less P2Y₁₂R expression in activated microglia than in resting microglia, suggesting a role for the P2Y₁₂R in early stages of CNS injury and a potential therapeutic pathway that may be exploited for treatment of neurodegenerative diseases [222]. However, another study indicated that expression of the P2Y₁₂R in the CNS is limited to oligodendrocytes [221].

The P2Y₁₂R has been shown to increase the proliferation of glioma cells by the Gα_i-dependent activation of RhoA, ROCK and PKCζ, independent of ERK1/2 activation [197], although P2Y₁₂R-mediated proliferation of Chinese hamster ovary cells was linked to activation of PI3K/Akt and ERK1/2, whereas a separate PTX-insensitive pathway leading to RhoA and ROCK activation was found to regulate actin cytoskeletal reorganization [196]. Interestingly, the P2Y₁₂R has been shown to play a role in glioma progression and malignancy where antagonism of the receptor with clopidogrel retarded the growth of NTPDase2-overexpressing gliomas in rats [223]. The P2Y₁₂R is also expressed in mouse dendritic cells, where it mediates the G_o-dependent macropinocytosis of antigens [198]. Reconstitution of the P2Y₁₂R with different G proteins indicates that the P2Y₁₂R preferentially couples to Gα_i2-dependent inhibition of adenylyl cyclase and the activation of PI3K, Akt, Rap1b and potassium channels, in comparison to Gα_i1 and Gα_i3, and does not couple to Gα_o or Gα_q proteins [199].

The P2Y₁₃ Receptor

The P2Y₁₃R is activated by ADP with an EC₅₀ = 60 nM [224], and 2-MeS-ADP has been shown to be a potent agonist [225]. The P2Y₁₃R is activated more potently by 2-MeS-ADP, adenosine 5'-O-2-(thio)diphosphate (ADPβS) and 2-MeS-ATP than by the endogenous agonist ADP [224]. BzATP also has been shown to be a P2Y₁₃R agonist in rat cerebellar astrocytes [226]. Similar to the ADP-activated P2Y₁₂R, the P2Y₁₃R couples to G_i-mediated inhibition of adenylyl cyclase [224], but can also couple to G_s-dependent activation of adenylyl cyclase at high agonist concentrations [227]. The P2Y₁₃R, like other P2YRs, can activate RhoA and ROCK [228, 229], suggesting that it will contribute to cytoskeletal rearrangements induced by adenine nucleotides in cells of the CNS.

The P2Y₁₂R antagonists AR-C69931 and AR-C67085 are potent non-competitive antagonists of the P2Y₁₃R with IC₅₀‘s = 4 nM and 1 nM, respectively, whereas diadenosine tetraphosphate, reactive blue 2, suramin and PPADS also antagonized the effects of ADP at the P2Y₁₃R [227]. Derivatives of PPADS, including the 2-chloro-5-nitro (MRS2211) and 4-chloro-3-nitro (MRS2603) analogues, were significantly more potent P2Y₁₃R antagonists than PPADS, whereas MRS2211 showed significant specificity for antagonism of the P2Y₁₃R, as compared to the P2Y₁ and P2Y₁₂ receptors [230]. MRS2211 was also shown to inhibit BzATP-induced increases in the intracellular calcium concentration in cerebellar astrocytes through antagonism of the P2Y₁₃R [226].

P2Y₁₃Rs are expressed in astrocytes and glutamatergic neurons and have been shown, along with P2Y₁ and P2Y₁₂ receptors, to enhance Na⁺ and Cl⁻-dependent glycine transport in the synaptic cleft [231]. The P2Y₁₃R also has been suggested to enhance cell survival by increasing the PI3K/Akt-dependent translocation of the glycogen synthase kinase-3 substrate β-catenin to the nucleus to promote the expression of cell survival genes [232]. Thus, the P2Y₁₃R fits the general profile of the P2YR family with respect to their likely roles in the regulation of protective or reparative processes in the CNS.

The P2Y₁₄ Receptor

The P2Y₁₄R is unique among the P2YR family in its ability to be activated by UDP-glucose (EC₅₀ = 80 nM), UDP-galactose (EC₅₀ = 125 nM), UDP-glucuronic acid (EC₅₀ = 370 nM) and UDP-N-acetylglucosamine (EC₅₀ = 710 nM), but not by adenine or uridine nucleotides (e.g., ATP, ADP, UTP and UDP) [225, 233, 234]. The 2-thio-modified UDP-glucose analog MRS2690 (diphosphoric acid 1-alpha-d-glucopyranosyl ester 2-[(2-thio)uridin-5"-yl] ester) is a potent P2Y₁₄R agonist [235]. Modifications in the uracil moieties of P2Y₁₄R agonists abolish activity [236]. The P2Y₁₄R has been shown to couple to G_i/o protein activation [235].

The effects of P2Y₁₄R activation and its therapeutic relevance in the treatment of CNS disorders are largely unknown. The P2Y₁₄R is expressed in human astrocytes [237] and in rat cortical and cerebellar astrocytes [14, 219]. P2Y₁₄R expression was increased in rat primary microglial cells in response to bacterial lipopolysaccharide [66], suggesting a role during neuroinflammation. P2Y₁₄Rs expressed in immature dendritic cells have been suggested to have an immunoregulatory function [23, 238]. P2Y₁₄R activation also can increase the production of chemokines and cytokines involved in the recruitment of neutrophils, including IL-8, macrophage inflammatory protein-2 and keratinocyte-derived cytokine [239], and the P2Y₁₄R has been shown to play a role in muscle contractility in rats and gastric function in mice [240].

CONSIDERATIONS FOR THE THERAPEUTIC USE OF P2Y RECEPTOR AGONISTS AND ANTAGONISTS

Blood-Brain Barrier

P2Y receptors in the CNS, particularly in the brain, represent promising targets in the treatment of neuroinflammatory and neurodegenerative diseases and there is a critical need for the development of more stable and selective P2Y receptor agonists/antagonists. Of greatest need are drugs that can cross the blood-brain barrier (BBB), which represents a major obstacle to the efficient delivery into the brain of drugs designed to mimic the negatively-charged structure of nucleotides that is known to be critical for agonist binding to P2YRs. The BBB is composed of cerebral endothelial cells, pericytes and astrocytes that regulate the flow of proteins and macromolecules into the CNS [241]. Potential therapies involving P2YR agonists and antagonists that target the CNS must consider modes of transport across the BBB, which for small polar molecules such as nucleotides likely can occur via plasma membrane transport proteins, receptor- or adsorptive-mediated transcytosis, or passage through endothelial cell tight junctions [241]. Studies have shown that P2Y receptors and P1 adenosine receptors are able to modulate the permeability of the blood brain barrier [149, 242]. Remarkably, treatment of murine models with the FDA approved A_2A adenosine receptor agonist regadenoson (Lexiscan) can increase BBB permeability to intravenously injected macromolecules, including anti-β-amyloid antibody, through a mechanism involving alterations in endothelial cell tight junctions [243]. Recent studies, including unpublished studies in our lab [Erb et al., submitted manuscript], indicate that activation of P2Y₂Rs in vascular endothelium can increase the transendothelial permeability of leukocytes [109, 244, 245], suggesting that P2Y₂R activation in brain microvessels that comprise the BBB may be a novel target for increasing drug delivery to the brain. It seems likely that any P2Y receptor agonists/antagonists targeted to the CNS must possess properties that are conducive to passage across the BBB or these compounds will have to be administered along with modulators of barrier function.

P2Y Receptor Interactions

Another important consideration in the targeting of P2Y receptors for therapeutic treatments is whether the functions being affected are a product of homomeric P2YR activation or are regulated by heteromeric interactions between different P2YR subtypes. For example, in vitro experiments have suggested that the P2Y₁₁R can form a hetero-oligomeric complex with the P2Y₁R which is internalized upon activation with ATP or 2-MeS-ADP and has a distinct agonist/antagonist profile that differs from both P2Y₁R and P2Y₁₁R [246]. Heteromeric interactions between P2Y receptors and P1 adenosine receptors also have been reported. The P2Y₁R has been shown to form functional hetero-oligomers with the A1 receptor (A1R) both in vitro [247] and in the rat cortex, hippocampus, and cerebellum [248, 249]. While the physiological contributions of this heteromeric P2Y₁:A1 receptor are unknown, it has been demonstrated that addition of the P2Y₁R agonist 2-MeS-ADP decreases A1R-mediated functional responses [250], whereas addition of the A1R agonist N⁶-cyclohexyladenosine enhances the 2-MeS-ADP-induced G protein coupling to P2Y₁Rs [248]. In addition, multiple homomeric P2Y and P2X receptor subtypes in a single cell type are likely to be activated simultaneously by the same nucleoside triphosphate (i.e., ATP or UTP) or breakdown products (i.e., ADP or UDP) and, therefore, may interact downstream of receptor activation via overlapping signal transduction pathways. Considering that multiple cell types that comprise the brain are likely to have a cell-specific complement of P2 receptor subtypes that are differentially expressed under a variety of physiological and pathophysiological conditions, it is very difficult to assess the contributions in vivo of individual P2Y receptor subtypes in specific cell types to neurological responses to released nucleotides in the brain. Although the deletion of specific P2YR subtypes in mice has helped reveal the physiological role of some P2YR subtypes, the use of these mouse models to evaluate P2YR interactions is in its infancy and more information is available from in vitro studies using receptor antagonists. For example, a recent study has shown that ADP increases axonal elongation in cultured rat hippocampal neurons [72]. However, ADP can activate multiple P2Y receptors in neurons, including the P2Y₁R and the P2Y₁₃R. Through the use of selective P2YR antagonists (MRS2179 for the P2Y₁R and MRS2211 for the P2Y₁₃R) and the suppression of P2Y₁R or P2Y₁₃R expression in neurons with siRNA, it was shown that axonal growth in response to ADP was coordinately regulated by both P2Y₁R and P2Y₁₃R activation and their opposing effects on the activation of adenylate cyclase 5. Whereas activation of adenylyl cyclase 5 by the P2Y₁R stimulated axonal elongation likely via G_q-dependent PI3K/Akt activation, P2Y₁₃R-mediated G_i activation inhibited adenylate cyclase 5 activity and axonal elongation. Furthermore, antagonism of the P2X7R also increased axonal elongation by modulating adenylate cyclase 5 activity [72]. The complexity of axonal elongation in response to ATP or ADP, as demonstrated by this study, accurately depicts the challenges of targeting P2Y receptors for therapy. Alternatively, this complexity could be advantageous in that a desired physiological response (e.g., axonal growth) can be targeted through activation or inhibition of individual P2YR subtypes.

Indirect interactions between the G_q protein-coupled P2Y₂ receptor and the ionotropic P2X7 receptor have been suggested to mediate neuroprotective responses in neuroinflammatory diseases, such as AD [35, 105, 251]. Under inflammatory conditions, such as those found in AD [114-115], ATP is released from apoptotic cells into the extracellular milieu where it activates the P2X7R to induce the processing and release of the proinflammatory cytokine IL-1β [4, 252, 253]. In response to IL-1β, P2Y₂R expression has been shown to be upregulated in rat [16] and mouse [Weisman et al., submitted manuscript] primary cortical neurons through a NF-κB-dependent mechanism [16, 116]. As described above, activation of the P2Y₂R has been shown to induce neuroprotective responses, such as stimulation of the non-amyloidogenic processing of amyloid precursor protein to generate soluble APPa, rather than the neurotoxic beta-amyloid peptide [130], and the stimulation of neurite outgrowth [254; Weisman et al., submitted manuscript]. Additionally, P2Y₂R activation in microglial cells has been shown to increase uptake and degradation of neurotoxic Aβ_1-42 [24]. IL-1β also has been shown to be elevated in the brains of AD patients [114, 115] and the P2X7R is upregulated around Aβ plaques in an AD mouse model [255]. These findings suggest that a cascade of events starting with tissue damage can induce the activation of P2X7Rs and IL-1β receptors to increase the expression of P2Y₂Rs, thereby counteracting a neurodegenerative response (i.e., IL-1β and ATP release) by enhancing a neuroprotective pathway (i.e., P2Y₂R-mediated neurite extension and neurotoxic Aβ elimination).

CONCLUSION

Recent research has demonstrated the existence of 8 P2Y receptor subtypes (i.e., P2Y_{1,2,4,6,11-14}) that are expressed in cells of the CNS (e.g., neurons, astrocytes, microglia and endothelial cells). When ATP, UTP and other nucleotides are released from cells in the CNS under a variety of conditions, including neurotransmission, apoptosis, inflammation and cell damage, the activation of these P2Y receptor subtypes stimulates intracellular signal transduction pathways that regulate physiological and pathological responses, including neurite outgrowth, glial cell migration and proliferation, glial cell-mediated phagocytosis, amyloid precursor protein processing, gene expression, cytokine release and diapedesis. These P2Y receptor-mediated responses to extracellular nucleotides are induced by the direct or indirect activation of G proteins (e.g., G_s, G_i, G_q, and G_12/13) and the modulation of intracellular protein activities, including adenylyl cyclase, phospholipase C, small GTPases, matrix metalloproteases, integrins, growth factor receptors and protein kinases. Current data suggest that P2Y receptor activation in the CNS, particularly in the brain, can serve a neuroprotective role. Therefore, novel agonists and antagonists are being actively pursued that are relatively selective for single P2Y receptor subtypes and may potentially have a wide variety of therapeutic applications in the treatment of pathologies of the CNS.

ABBREVIATIONS

2-MeS-ADP

2-Methylthio-ADP

2-MeS-ATP

2-Methylthio-ATP

AD

Alzheimer’s disease

ADP

Adenosine 5’-diphosphate

ATP

Adenosine 5’-triphosphate

APP

Amyloid precursor protein

BBB

Blood-brain barrier

cAMP

Cyclic adenosine 3-,5’-monophosphate

CNS

Central nervous system

GTP

Guanosine 5’-triphosphate

IL-1β

Interleukin-1β

IP₃

Inositol 1,4,5-trisphosphate

P2YR

P2Y receptor

PCR

Polymerase chain reaction

PKC

Protein kinase C

PLC

Phospholipase C

PPADS

Pyridoxal-phosphate-6-azophenyl-2',4'-disulphonic acid

UDP

Uridine 5’-diphosphate

UTP

Uridine 5’-triphosphate