Differential Effects of Serotonin, FMRFamide, and Small Cardioactive Peptide on Multiple, Distributed Processes Modulating Sensorimotor Synaptic Transmission in Ap/ysia

form (Pieroni and Byrne, 1989, 1990). Materials and Methods Aplysiu culifornicu (125-300 gm) were obtained from Alacrity Marine Biological Specimens (Redondo Beach, CA), Marine Specimens Unlimited (Pacific Palisades. CA). and Marinus. Inc. (Lone. Beach. CA). AnI I I ~imals were maintained in individual perforated cages at 15”C’in aquaria filled with aerated artificial seawater (ASW; Instant Ocean, Aquarium Systems, Mentor, OH). Animals were exposed to a 12 hr lighti hr dark cycle and were fed an amount of dried seaweed sufficient to maintain a constant body weight. Prior to dissection, animals were anesthetized by injection ofa volume of isotonic MgCl, equal to approximately one-half of their body volume. Abdominal ganglia were removed and rinsed in a cold mixture of artificial seawater (ASW, buffered to pH 7.6 with 10 mM Trizma, Sigma) and isotonic MgCl, (50:50, v/v) containing 0.5% glutaraldehyde (Sigma, St. Louis, MO) for 35 set in order to reduce TEAand 5-HT-induced contractions in the connective tissue sheath (Mirolli and Gorman, 1968; Byrne et al., 1979). Ganglia were then pinned to the floor of a Sylgard (Dow Coming, Midland, MI)-lined chamber (300 ~1 volume) and maintained in ASW/isotonic MgCl, (50:50, v/v) during dissection. Cells in the ventral left hemiganglion were exposed by surgical removal of the sheath, and subsequently the ASWfMgCl, solution was replaced with ASW. Sensory neurons in the LE cluster (Byrne et al., 1974) and follower neurons (usually LFS cells; Frost et al., 1988) were identified by their size, location, and electrophysiological properties. Sensory neurons with resting membrane potentials greater than -35 mV were considered acceptable, as were follower neurons with resting membrane potentials greater than -20 mV (in ASW containing TEA). The static ASW bathing solution contained 100 mM TEA chloride (Eastman Kodak, Rochester, NY) and 240 PM TTX (Calbiochem, La Jolla, CA). Pretreatment of the ganglion with TE.4 blocks ZK,” and the calcium-dependent K+ current (ZK,-; Baxter and Byrne, 1989; Walsh and Byrne, 1989) and broadens the sensory neuron spikes into a range of spike durations in which additional broadening alone has little or no effect on release of transmitter (Gingrich and Byrne, 1985; Hochner et al., 1986; see also Goldsmith and Abrams, 199 1). Under this condition, facilitatory changes in release occur relatively independently of spike broadening and therefore are due primarily to activation of a process(es) of a spike duration-independent nature (Gingrich and Byrne, 1985; Hochner et al., 1986). TTX was added to the bath in order to avoid multiple action potentials being triggered in the sensory neuron. TTX also reduced both spontaneous PSPs in the follower neuron and polysynaptic input (elicited by neurons interposed between the sensory and follower neurons) to the follower neuron. Serotonin (creatinine sulfate complex) was obtained from Sigma, and FMRFamide and SCP, were obtained from Peninsula Laboratories (Belmont, CA). These agents were delivered to the static bath in 5 ~1 volumes, using the same solution as contained in the bath solution (100 mM TEA in ASW) as vehicle. The concentrations are similar to those used in other studies on sensory neurons (e.g., Abrams et al., 1984; Ocorr and Byrne, 1985, 1986; Brezina et al., 1987; Baxter and Byrne, 1989, 1990; Blumenfeld et al., 1990; Edmonds et al., 1990; Schacher et al., 1990). Two concentrations were used in experiments with SCP, (4 and 25 FM), but the effects of these concentrations were indistinguishable and apparently maximal (a concentration of 50 hi, used in separate unpublished experiments, did not produce greater effects). Appropriate concentrations of stock solutions were made daily and kept on ice. Serotonin was protected from exposure to light. Experiments were conducted at room temperature (approximately 2 1°C). Presynaptic, calcium-mediated (in TTX) spikes were initiated in sensory neurons by suprathreshold depolarizing constant current pulses (10-17 nA, 1.5 msec) at 60 set intervals. Follower neurons were hyperpolarized (via a second microelectrode) by approximately 30 mV from the resting membrane potential to prevent spiking during evoked EPSPs. In some experiments, input resistance of follower neurons was assessed by delivering brief (1.5 set) hyperpolarizing current pulses at 60 set intervals (30 set before each stimulation). In the figures, spikes and EPSPs labeled initial were generated approximately 10 min (rest period) after a synaptic connection was established. Data were digitized and stored on a computer, and also recorded with a Gould pen recorder. The peak amplitude of the monosynaptic EPSPs and the half-amplitude spike duration (duration from peak of the spike to 50% of its repolarization to the resting membrane potential) were calculated by a computer program. Input resistances were calculated from records made on the Gould recorder. Data (normalized to initial values) are presented as mean percentage The Journal of Neuroscience, July 1992, 72(7) 2635