A question of balance e Positive versus negative allosteric modulation of GABA A receptor subtypes as a driver of analgesic efficacy in rat models of inflammatory and neuropathic pain Gordon Munro a, * , Helle K. Erichsen a , Mark G. Rae b , Naheed R. Mirza a a Department of Pharmacology, NeuroSearch A/S, 93 Pederstrupvej, DK-2750 Ballerup, Denmark b Department of Physiology, University College Cork, Cork, Ireland article info Article history: Received 2 February 2011 Received in revised form 14 March 2011 Accepted 15 March 2011 Keywords: Chronic constriction injury Formalin test FG-7142 NS11394 Spinal inhibition TPA023 abstract After injury GABA A receptor positive allosteric modulators (PAMs) mediate robust analgesia in animals via putative restoration of post-synaptic GABA A -a2 and -a3 receptor function within the spinal cord. GABA can also act at GABA A receptors localized on primary afferent neurones to inhibit presynaptic neurotransmitter release and produce analgesia via a process called primary afferent depolarization (PAD). Some forms of injury might sufficiently enhance PAD to shift it into a net excitatory process. Thus, negative allosteric modulators (NAMs) might also possess analgesic activity. We have compared compounds capable of either positively or negatively modulating GABA A receptors in rat models associated with injury-induced central sensitization. The subtype-selective PAMs NS11394 (1e10 mg/kg) and TPA023 (3e30 mg/kg) attenuated formalin-induced nocifensive behaviours. Similarly, both compounds reversed hindpaw mechanical hypersensitivity and weight bearing deficits in carrageenan- inflamed and nerve-injured rats. The non-selective PAM diazepam (1e5 mg/kg) was ineffective in all models. Surprisingly, both the non-selective NAM FG-7142 (3e30 mg/kg) and the a5-selective NAM a5IA-II (10e60 mg/kg) also attenuated formalin-induced nocifensive behaviours. In carrageenan- inflamed rats a5IA-II reversed mechanical hypersensitivity and weight bearing deficits whilst FG-7142 only attenuated weight bearing deficits. This picture was essentially reversed in nerve-injured rats for these two NAMs. With the exception of NS11394, all compounds attenuated exploratory motility behaviour in rats, either as a consequence of sedative or anxiogenic-like side-effects. These data indicate that the preferred selectivity and activity profiles for mediating analgesia upon activation of GABA A receptors might be more complex than previously anticipated, and is worthy of further exploration. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction GABA is the major inhibitory neurotransmitter in the mamma- lian CNS mediating its effects via activation of GABA A and GABA B receptors. GABA binding to GABA A receptors typically opens a Cl  permeant anion channel and the ensuing influx of Cl  ions hyper- polarizies the cell. The majority of CNS GABA A receptors consist of two a, two b and one g subunit (Sieghart and Sperk, 2002), that can interact with a range of compounds e via specific binding sites e that include agonists (e.g., GABA and muscimol) and antagonists (e.g., bicuculline), barbiturates, neurosteroids, anaesthetics and benzodiazepines (e.g., diazepam). The benzodiazepines and other related compounds do not directly activate the GABA A receptor, but rather function as positive allosteric modulators (PAMs) to enhance GABA-gated currents at GABA A receptors containing an a1, a2, a3, or a5 subunit. These four subunits are variously expressed at pre- and post- synaptic elements within the spinal dorsal horn, although a2- containing GABA A receptors predominate within superficial layers that receive nociceptive input from primary afferents (Knabl et al., 2008). Behavioural studies with genetically-modified mice and comparative pharmacology studies in rats have recently identified a2- and a3- (and possibly a5)-subunit-containing GABA A receptors as principal contributors to the spinal disinhibition that occurs after inflammatory or neuropathic injury (Knabl et al., 2008; Munro et al., 2008; Knabl et al., 2009). Together, these complimentary approaches have implied that it is the restoration of post-synaptic actions of GABA on intrinsic dorsal horn neurones which mediates this analgesia (Munro et al., 2009; Zeilhofer et al., 2009). * Corresponding author. Tel.: þ45 44608333; fax: þ45 44608080. E-mail address: gmu@neurosearch.dk (G. Munro). Contents lists available at ScienceDirect Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm 0028-3908/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropharm.2011.03.017 Neuropharmacology 61 (2011) 121e132