1. In current clamp recordings, 1 mu M prostaglandin E(2) (PGE(2)) increased the excitability of neonatal rat-dorsal root ganglion neurones. The current-threshold for firing was reduced, and the response to a constant suprathreshold stimulation was modified such that a single evoked action potential was converted to a train of action potentials. The excitatory action of PGE(2) was still apparent when action potentials were evoked in the presence of 500 nM tetrodotoxin. 2. In voltage-clamp experiments 1 mu M PGE(2) frequently increased the magnitude of the peak currents recorded, and caused a hyperpolarizing shift (of approximately 6 mV) in the activation curve for the tetrodotoxin-resistant sodium current (TTX-R I-Na). rn some cells, the hyperpolarizing shift in the activation curve was accompanied by a decrease in peak conductance. PGE(2) also caused a hyperpolarizing shift in the steady-state inactivation curve for the sodium current. 3. Extracellular application of the cAMP analogue dibutyryl cAMP (dbcAMP) at a concentration of 1 mM produced effects on both the current-voltage relationship and the steady-state inactivation curve for the TTX-R I-Na which were indistinguishable from those observed with PGE(2). Prior exposure of the neurones to dbcAMP occluded the effect of a subsequent treatment with PGE(2). 4. Forskolin (10 mu M), a direct activator of adenylate-cyclase, mimicked the effects of PGE(2) and dbcAMP on TTX-R I-Na. The inactive congener of forskolin, 1,9-dideoxyforskolin (10 mu M), reduced the amplitude of TTX-R I-Na, but did not evoke a hyperpolarizing shift in the activation curve. 5. Intracellular perfusion of the neurones with an inhibitor of protein kinase A inhibited the effect-of PGE(2) on TTX-R I-Na. 6. PGE(2) also reduced the amplitude of voltage-gated potassium currents (I-K), which will contribute to the excitatory action. The mechanisms underlying the changes in I-K have yet to be elucidated. 7. We propose that the PGE(2)-mediated increase in excitability in sensory neurones mag be due, at least in part, to the cAMP-protein kinase A-dependent modulation of the tetrodotoxin resistant sodium channel.