In somata of sensory neurons in the pleural ganglia of Aplysia californica, serotonin (5-HT) modulates at least three K+ currents: the S K+ current (I(K,S)), a slow component of the Ca2+-activated K+ current (I(K,Ca)), and the delayed or voltage-dependent K+ current (I(K,V)). The modulation of I(K,S) and the slow component of I(K,Ca) by 5-HT has been shown previously to be mediated via adenosine 3',5'-cyclic monophosphate (cAMP). To determine whether the modulation of I(K,V) by 5-HT also is mediated via cAMP, we used two-electrode voltage-clamp techniques to compare the modulation of membrane current by cAMP and 5-HT. Current responses were elicited by brief (200 ms) voltage-clamp pulses before and after the bath application of analogues of cAMP. At all voltage-clamp potentials examined (-40-30 mV), analogues of cAMP reduced the amplitude of the current response. The properties of the cAMP-sensitive component of membrane current were revealed by computer subtraction of current responses elicited in the presence of the analogue of cAMP from current responses elicited before application of the analogue. The characteristics of the resulting cAMP difference current (I(cAMP)) suggested that cAMP modulated a component of membrane current that was relatively voltage independent, did not inactivate, and was active over a wide range of membrane potentials. In addition, the current-voltage (I-V) relationship of the cAMP difference current had a positive slope. These properties of the cAMP difference current were consistent with those of I(K,S) but did not indicate that I(K,V) was modulated by cAMP. The cAMP-independent modulation of membrane current by 5-HT was examined by eliciting current responses first in the presence of an analogue of cAMP and again after the addition of 5-HT to the bath, which still contained the analogue. The presence of the analogue of cAMP occluded further modulation of I(K,S) by 5-HT. However, the analogue of cAMP did not occlude the modulation of I(K,V) by 5-HT. This cAMP-independent effect of 5-HT on membrane current was revealed by computer subtraction of current responses elicited in the presence of 5-HT from current responses elicited before the application of 5-HT (the analogue of cAMP was present throughout). The resulting cAMP-independent 5-HT difference current (I(5-HT)) was highly voltage dependent, had complex kinetics, and its I-V relationship had a negative slope at membrane potentials above 0 mV. This voltage-dependent and cAMP-independent component of the 5-HT difference current has been shown previously to represent modulation of I(K,V) by 5-HT. To examine the relative contributions that the cAMP-mediated and cAMP-independent processes make to the overall modulatory effects of 5-HT, we used two-electrode current-clamp techniques to compare the effects of cAMP and 5-HT on the duration of individual action potentials and excitability in the somata of sensory neurons. Bath application of 5-HT alone produced a threefold increase in the duration of action potentials and doubled the number of action potentials elicited during measurements of neuronal excitability. Bath application of analogues of cAMP produced a modest (12%) increase in the duration of action potentials and doubled neuronal excitability. The subsequent addition of 5-HT to the bath, which still contained the analogue of cAMP, produced a significant (3-fold) increase in the duration of action potentials but did not produce any further increase in excitability. These results indicate that the modulation of I(K,V) by 5-HT is not mediated by cAMP and may require an as yet unidentified second messenger system. In addition, the results indicate that the cAMP-mediated modulation of membrane current by 5-HT plays a key role in regulating the excitability of the somata of sensory neurons, whereas the cAMP-independent modulation of membrane current by 5-HT has an important role in regulating the duration of the somatic action potential.
