Recently, two metabolites of paracetamol were shown to activate TRPA1, a receptor involved in neurogenic airway inflammation1 and thermal, mechanical and chemical nociception.2 Thus p-benzoquinone (p-BQ) and n-acetyl-p-benzoquinoneimine (NAPQI) increased [Ca2+]i in CHO cells expressing TRPA1;and a subset of cultured DRG neurones displaying such Ca2+ responses was strongly depolarised, suggesting a spinal and presynaptic role in APAP analgesia2; our in-vivo studies address this possibility. In anaesthetised rats (Urethane, i.p.) the spinal cord was exposed, and L1-S1 dorsal roots prepared for recording either orthodromic sensory nerve action potentials or ones arising centrally and conducted antidromically ( Fig1, lower trace). These arise in small diameter myelinated and C-fibres in contrast to the larger spikes recorded from the peripheral end of the same cut root in response to tactile or proprioceptive stimulation (not shown). The upper trace shows the mean spike rate counted in 5.0 sec bins. Within seconds of application to the spinal cord root exit zone 10micro litres of 1.0 microM p-BQ evoked a five-fold increase in net impulse rate persisting for 2 minutes before return to baseline level. The next dose (10µl p-BQ at 10 microM) evoked a prompt sixteen-fold rate increase at its peak with recruitment of new spikes; this activity was sustained for a further 3 minutes after the illustrated trace. This response was elicited on each occasion (35 in 18 rats) with BQ concentrations ranging from 300nM to 600microM ; we attribute this to variations both in the integrity of the pia at the site and dispersion of the applied drug. Although retrograde impulse activity can be directly attributed to the depolarisation of TRPA1 afferent terminals some glutamatergic collaterals may excite internuncial pathways causing presynaptic depolarisation via GABA or glycine receptors. However, this does not seem relevant for APAP analgesia. The P-wave of the cord dorsum potential evoked by sural nerve stimulation, an electrical sign of presynaptic inhibition, was unaffected in 5 rats followed for up to 2 hours after s.c. injection of the same doses (300mg/kg) used in the behavioural studies. These results thus support the hypothesis that paracetamol analgesia is, in part, mediated by its metabolite p-BQ activating TRPA1 receptors leading to depolarisation and consequent presynaptic inhibition of pain pathway(s) mediating antinociception2.