Heart interneurons (HN cells) in isolated ganglia of the medicinal leech were voltage-clamped with single microelectrodes. Hyperpolarizing voltage steps elicited a slow inward current (Ih), which underlies the characteristic depolarizing response of HN cells to injection of prolonged hyperpolarizing current pulses (Arbas and Calabrese, 1987a). The conductance underlying Ih begins to activate near -mV and is fully activated between -70 and -80 mV. The activation kinetics of Ih are slow and voltage dependent. The activation time constant (tau h) ranges from approximately 2 sec at -60 mV to near 700 msec at -100 mV. Ih persists in low Ca2+ (0.1 mM), 5 mM Mn2+ saline and exhibits a reversal potential of -21 +/- 5 mV. The reversal potential is shifted by altering [Na+]o or [K+]o but is unaffected by changes in [Cl-]o. Ih is blocked by extracellular Cs+ (1–5 mM) but not Ba2+ (5 mM) or TEA (25 mM). Low concentrations of Cs+ (100–200 microM) cause a partial block that exhibits strong voltage dependence. Temperature changes were also shown to affect Ih. Both the rate of activation and the steady-state amplitude of Ih are enhanced by temperature increases. HN cells are interconnected by inhibitory chemical synapses, and their normal electrical activity consists of bursts of action potentials separated by periods of inhibition. During the inhibitory phase of rhythmic bursting activity, HN cells hyperpolarize to a voltage range where Ih is activated. Block of Ih with extracellular Cs+ (4 mM) disrupted the normal bursting activity of HN cells. These results are consistent with the hypothesis that Ih contributes to escape from inhibitory inputs during normal bursting activity.

Synaptic transmission between reciprocally inhibitory heart interneurons (HN cells) of the medicinal leech was examined in the absence of Na-mediated action potentials. Under voltage clamp, depolarizing steps from a holding potential of -60 mV elicited 2 kinetically distinct components of inward current in the presynaptic HN cell: an early transient current that inactivates within 200 msec and a persistent current that only partially decays over several seconds. Both currents begin to activate near -60 mV. Steady-state inactivation occurs over the voltage range between -70 and -45 mV and is completely removed by 1–-2-sec hyperpolarizing voltage steps to -80 mV. The inward currents are carried by Ca2+, Ba2+, or Sr2+ ions, but not by Co2+, Mn2+, or Ni2+. These same inward currents underlie the burst-generating plateau potentials previously described in HN cells (Arbas and Calabrese, 1987a,b). With a presynaptic holding potential of -60 mV, the threshold for transmitter release is near -45 mV. Postsynaptic currents in the contralateral HN cell have a reversal potential near - 60 mV. The largest postsynaptic currents (300–400 pA) exhibit an initial peak response that is followed by a more slowly decaying component. The persistent component of Ca2+ current in the presynaptic neuron is strongly correlated with the prolonged component of the postsynaptic current, while the transient presynaptic Ca2+ current appears to correspond to the early peak of postsynaptic current. These data are consistent with the hypothesis that voltage-dependent calcium currents contribute to the oscillatory capability of reciprocally inhibitory HN cells by (1) generating the plateau potential that drives the burst of action potentials and (2) underlying the release of inhibitory transmitter onto the contralateral cell.

1. Leech ganglia were superfused with salines in which Ca2+ was replaced with equimolar concentrations of Co2+, Ni2+, or Mn2+. These salines elicited rhythmic membrane potential oscillations with cycle periods ranging from 8 to 25 s in all neurons examined within the ventral nerve cord. 2. Rhythmic activity consisted of a rapid depolarization to a prolonged (3-6 s) plateau level, followed by a rapid repolarization. Each depolarization elicited a burst of action potentials. Peak-to-trough amplitudes of the plateau depolarizations were up to 40 mV in some cells. The plateau depolarizations were separated by slowly depolarizing ramp potentials. 3. Oscillations in all neurons were synchronized (in phase) both within individual ganglia and between ganglia linked by connective nerves. Rhythmic activity in isolated ganglia persisted after the interposed connective nerves were cut. 4. The occurrence of oscillatory activity was strongly correlated with the block of chemical synaptic transmission. 5. Electrotonic interactions persisted during oscillatory activity and may be one mechanism by which oscillations are synchronized. 6. The phase of rhythmic impulse bursts monitored with extracellular electrodes could be reset by electrical stimulation of connective nerves but not by injection of current pulses into individual neurons. Phase reset appeared to occur within one cycle and to a fixed phase point (plateau termination). 7. Oscillatory activity was eliminated by 75-100% reductions of [Na+]o (Na+ replaced with N-methyl-D-glucamine). Smaller reductions of Na+ (by 25-50%) increased the cycle period of oscillations. 8. The Na(+)-K+ pump inhibitors ouabain and strophanthidin disrupted oscillations. Cells were depolarized by approximately 20 mV and fired tonically. After the initial washout of the inhibitors, cells repolarized and became quiescent. After several minutes of continued washing, oscillatory activity resumed. 9. A conceptual model is proposed to explain the mechanisms underlying oscillatory activity induced by Ca2+ channel blockers. According to this model, depolarizing plateaus are generated by a noninactivating Na+ conductance. Na+ influx during the plateau leads to an increase in [Na+]i, which activates an electrogenic Na(+)-K+ pump that contributes to plateau termination. 10. A quantitative computer simulation incorporating six types of currents (capacity, outward rectifying potassium, inward rectifying potassium, sodium, leakage, and an electrogenic sodium pump) demonstrates the plausibility of the conceptual model. 11. These data suggest that a novel Na(+)-based mechanism for membrane potential oscillation is revealed by blockade of Ca2+ channels in leech ganglia.

Postinhibitory rebound (PIR) is defined as membrane depolarization occurring at the offset of a hyperpolarizing stimulus and is one of several intrinsic properties that may promote rhythmic electrical activity. PIR can be produced by several mechanisms including hyperpolarization-activated cation current (I(h)) or de-inactivation of depolarization-activated inward currents. Excitatory swim motor neurons in the leech exhibit PIR in response to injected current pulses or inhibitory synaptic input. Serotonin, a potent modulator of leech swimming behavior, increases the peak amplitude of PIR and decreases its duration, effects consistent with supporting rhythmic activity. In this study, we performed current clamp experiments on dorsal excitatory cell 3 (DE-3) and ventral excitatory cell 4 (VE-4). We found a significant difference in the shape of PIR responses expressed by these two cell types in normal saline, with DE-3 exhibiting a larger prolonged component. Exposing motor neurons to serotonin eliminated this difference. Cs+ had no effect on PIR, suggesting that I(h) plays no role. PIR was suppressed completely when low Na+ solution was combined with Ca2+-channel blockers. Our data support the hypothesis that PIR in swim motor neurons is produced by a combination of low-threshold Na+ and Ca2+ currents that begin to activate near -60 mV.

The nematode nervous system is distinguished by the small number and morphological simplicity of its neurons. Recently, the shapes and synaptic interactions of each of the 302 neurons in the small free-living nematode, Caenorhabditis elegans, have been determined from reconstructions of serial sections by electron microscopy. Comparable anatomical studies of the large parasitic nematode Ascaris have concentrated on the dorsal and ventral nerve cords where reconstructions of motor neurons by light microscopy led to the identification of seven distinct types of motor neurons, each corresponding to a homologous cell type in C. elegans. In this study the shapes of the 13 neurons with cell bodies in the retrovesicular ganglion (RVG) of Ascaris suum were reconstructed from light micrographs of serial sections. In other preparations the morphology of RVG neurons was observed in whole mounts after the cells were impaled with microelectrodes and injected with the fluorescent dye Lucifer yellow. The intracellular electrodes also permitted electrical recordings and revealed that one type of cell, the AVF-like interneuron, expresses spontaneous repetitive plateau potentials. Comparisons of neuronal morphologies in the retrovesicular ganglia of Ascaris and C. elegans suggest that each neuron in Ascaris can be assigned a corresponding homolog in C. elegans. These data provide further evidence for a remarkable conservation of neuronal morphology in nematodes despite large differences in size and habitat.