Novel 8-amino-1,2,4-triazolo[4,3-a]pyrazin-3-one derivatives as potent human adenosine A₁ and A_2A receptor antagonists. Evaluation of their protective effect against β-amyloid-induced neurotoxicity in SH-SY5Y cells

Highlights

• Several derivatives possess nanomolar hA₁ and hA_2A receptor affinities.

• Two compounds showed high affinity and a complete selectivity for the hA_2A AR.

• Two compounds were able to prevent β-amyloid peptide (25–35)-induced neurotoxicity.

• Docking studies were employed to rationalize the affinity data.

Abstract

In this work, an enlarged series of 1,2,4-triazolo[4,3-a]pyrazin-3-ones was designed to target the human (h) A_2A adenosine receptor (AR) or both hA₁ and hA_2A ARs. The novel 8-amino-1,2,4-triazolopyrazin-3-one derivatives 1–25 featured a phenyl or a benzyl pendant at position 2 while different aryl/heteroaryl substituents were placed at position 6. Two compounds (8 and 10) endowed with high affinity (K_i = 7.2 and 10.6 nM) and a complete selectivity for the hA_2A AR were identified. Moreover, several derivatives possessed nanomolar affinity for both hA₁ and hA_2A ARs (both K_i < 20 nM) and different degrees of selectivity versus the hA₃ AR. Two selected compounds (10 and 25) demonstrated ability in preventing β-amyloid peptide (25–35)-induced neurotoxicity in SH-SY5Y cells. Results of docking studies at the hA_2A and hA₁ AR crystal structures helped us to rationalize the observed affinity data and to highlight that the steric hindrance of the substituents at the 2- and 6-position of the bicyclic core affects the binding mode in the receptor cavity.

Introduction

Adenosine is a ubiquitous neuromodulator which controls many physiological and pathological processes, both in the central and peripheral nervous system. Adenosine exerts its effects through activation of G protein-coupled receptors, subdivided into the four subtypes A₁, A_2A, A_2B and A₃ [1], [2]. A₁ and A₃ adenosine receptors (ARs) are negatively coupled to adenylate cyclase while A_2A and A_2B subtypes activate the enzyme. The hA₁ AR subtype is the most abundant in the brain and is traditionally considered a neuroprotective receptor due to its inhibitory effects [3]. For instance, it curtails excitatory neurotransmission thus exerting a protective role in diverse pathological conditions linked to glutamate excitotoxicity such as cerebral ischemia or epilepsy. Nevertheless, there is recent evidence that A₁ AR-sustained activation after stroke or ischemia induces AMPA receptor endocytosis and the consequent, persistent synaptic depression may contribute to enhanced neuronal death [4], [5], [6].

The hA_2A AR subtype has less widespread localization at central level where it shows higher density in the basal ganglia and lower in the cortex and hippocampus. Blockade of this AR subtype exerts a protective effect in different models of cerebral ischemia and neurodegenerative disorders, such as Parkinson’s or Alzheimer’s diseases [4], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16]. In the last decade, several human studies have highlighted that consumption of caffeine, a non-selective A₁ and A_2A AR antagonist, negatively correlated with the risk of developing AD and PD [9], [11], [12]. The protective effect of caffeine, investigated in animal models of AD and PD was ascribed to antagonism of the A_2A AR subtype, among other pathways [10], [11], [12], [13], [14]. Related to AD models, both caffeine and the potent A_2A AR antagonist ZM 241,385 (4-(2-[7-amino-2-(2-furyl)-1,2,4-triazolo[2,3-a]triazin-5-ylamino]ethyl]phenol) prevented cell death after exposure of rat cultured cerebellar granule neurons to the β-amyloid peptide (25–35) [13]. A_2A AR antagonists demonstrated an ability to alleviate cognitive deficits caused by administration of the β-amyloid peptides in different in vivo rodent models [14], [15]. However, more recently, also A₁ AR antagonism was recognized as affording neuroprotection in a model of combined neurotoxicity. In fact, the protective effect of dual A₁ and A_2A AR blockade in preventing β-amyloid toxicity in neuroblastoma cells exposed to aluminium chloride was demonstrated [16].

In recent years, as part of a research program aimed at finding new adenosine receptor antagonists [17], [18], [19], [20], [21], [22], [23], [24], [25], we disclosed the 8-amino-1,2,4-triazolo[4,3-a]pyrazin-3-one as a new decorable scaffold to obtain potent antagonists able to bind selectively the hA_2A AR or both the hA₁ and hA_2A subtypes [21] (Fig. 1). The SAR study showed that the unsubstituted phenyl ring at position 2 was the best group for obtaining an efficient hA_2A receptor interaction, and that the presence of small para alkoxy groups (OMe, OEt, O-isopropyl, O-propargyl) on the 6-phenyl pendant afforded nanomolar affinity and a complete selectivity for this AR subtype.

To continue investigations of the 8-amino-1,2,4-triazolopyrazin-3-one series, we designed and synthesized an enlarged series of derivatives to obtain antagonists for the hA_2A AR or both hA₁ and hA_2A ARs. These types of ligands were of our interest for their ability to induce neuroprotection in different neurodegenerative diseases. Thus, a new set of triazolopyrazines, featuring the unsubstituted phenyl ring at position 2 and heteroaryl or aryl groups at position 6 (Fig. 1 compounds 1–22), were synthesized. Simple substituents, endowed with different electronic, steric and lipophilic properties, were probed on the 6-phenyl ring (X = OMe, NO₂, NH₂, Br, Cl). The NH₂ group was also inserted to construct the piperazine moiety appended on the 6-phenyl moiety (compounds 17–22). Finally, derivatives featuring a benzyl chain at position 2 and a phenyl ring or a 2-furyl/2-(5-methylfuryl) group at position 6 were synthesized (derivatives 23–25) since the benzyl pendant and the furyl moiety were thought to enhance compound solubility.