Lasting changes in a network of interneurons after synapse regeneration and delayed recovery of sensitization.

Regeneration of neuronal circuits cannot be successful without restoration of full function, including recovery of behavioral plasticity, which we have found is delayed after regeneration of specific synapses. Experiments were designed to measure neuronal changes that may underlie recovery of function. Sensitization of the leech withdrawal reflex is a non-associative form of learning that depends on the S-interneuron. Cutting an S-cell axon in Faivre's nerve disrupted the capacity for sensitization. The S-cell axon regenerated its electrical synapse with its homologous cell after 3-4 weeks, but the capacity for sensitization was delayed for an additional 2-3 weeks. In the present experiments another form of non-associative conditioning, dishabituation, was also eliminated by S-cell axotomy; it returned following regeneration. Semi-intact preparations were made for behavioral studies, and chains of ganglia with some skin were used for intracellular recording and skin stimulation. In both preparations there was a similar time-course, during 6 weeks, of a lesion-induced decrease and delayed restoration of both S-cell action potential threshold to depolarizing pulses and S-cell firing in response to test stimuli. However, the ability of sensitizing stimuli to decrease S-cell threshold and enhance S-cell activity in response to test stimuli did not fully return after regeneration, indicating that there were lasting changes in the circuit extending beyond the period necessary for full recovery of behavior. Intracellular recordings from the axotomized S-cell revealed a shift in the usual balance of excitatory and inhibitory input, with inhibition enhanced. These results indicate that loss of behavioral plasticity of reflexive shortening following axotomy in the S-cell chain may be related to reduced S-cell activity, and that additional processes underlie full recovery of sensitization of the whole body shortening reflex.

Neuronal competition for action potential initiation sites in a circuit controlling simple learning.

The spatial and temporal patterns of action potential initiations were studied in a behaving leech preparation to determine the basis of increased firing that accompanies sensitization, a form of non-associative learning requiring the S-interneurons. Little is known at the network level about mechanisms of behavioral sensitization. The S-interneurons, one in each ganglion and linked by electrical synapses with both neighbors to form a chain, are interposed between sensory and motor neurons. In sensitized preparations the strength of shortening is related to S-cell firing, which itself is the result of impulses initiating in several S-cells. Because the S-cells, as independent initiation sites, all contribute to activity in the chain, it was hypothesized that during sensitization, increased multi-site activity increased the chain's firing rate. However, it was found that during sensitization, the single site with the largest initiation rate, the S-cell in the stimulated segment, suppressed initiations in adjacent ganglia. Experiments showed this was both because (1) it received the earliest, greatest input and (2) the delayed synaptic input to the adjacent S-cells coincided with the action potential refractory period. A compartmental model of the S-cell and its inputs showed that a simple, intrinsic mechanism of inexcitability after each action potential may account for suppression of impulse initiations. Thus, a non-synaptic competition between neurons alters synaptic integration in the chain. In one mode, inputs to different sites sum independently, whereas in another, synaptic input to a single site precisely specifies the overall pattern of activity.

Learning in simple systems.

Cellular processes that mediate learning and memory show a remarkable level of conservation between vertebrates and invertebrates. Recent studies have shown that learning and memory formation in invertebrates, so-called 'simple systems', involves a highly complex arrangement of cellular pathways. Some pathways contribute to a single stage of memory formation, whereas others impact on multiple stages of memory development. Distinct cellular pathways may also act in series or in parallel during various stages of memory formation.

Multiple sites of action potential initiation increase neuronal firing rate.

Sensory input to an individual interneuron or motoneuron typically evokes activity at a single site, the initial segment, so that firing rate reflects the balance of excitation and inhibition there. In a network of cells that are electrically coupled, a sensory input produced by appropriate, localized stimulation can cause impulses to be initiated in several places. An example in the leech is the chain of S cells, which are critical for sensitization of reflex responses to mechanosensory stimulation. S cells, one per segment, form an electrically coupled chain extending the entire length of the CNS. Each S cell receives input from mechanosensory neurons in that segment. Because impulses can arise in any S cell and can reliably propagate throughout the chain, all the S cells behave like a single neuron with multiple initiation sites. In the present experiments, well-defined stimuli applied to a small area of skin evoked mechanosensory action potentials that propagated centrally to several segments, producing S cell impulses in those segments. Following pressure to the skin, impulses arose first in the S cell of the same segment as the stimulus, followed by impulses in S cells in other segments. Often four or five separate initiation sites were observed. This timing of impulse initiation played an important role in increasing the frequency of firing. Impulses arising at different sites did not usually collide but added to the total firing rate of the chain. A computational model is presented to illustrate how mechanosensory neurons distribute the effects of a single sensory stimulus into spatially and temporally separated synaptic input. The model predicts that changes in impulse propagation in mechanosensory neurons can alter S cell frequency of firing by changing the number of initiation sites.

Non-associative learning and serotonin induce similar bi-directional changes in excitability of a neuron critical for learning in the medicinal leech.

In studies of the cellular basis of learning, much attention has focused on plasticity in synaptic transmission in terms of transmitter release and the number or responsiveness of neurotransmitter receptors. However, changes in postsynaptic excitability independent of receptors may also play an important role. Changes in excitability of a single interneuron in the leech, the S-cell, were measured during non-associative learning of the whole-body shortening reflex. This interneuron was chosen because it is known to be necessary for sensitization and full dishabituation of the shortening response. During sensitization, S-cell excitability increased, and this enhancement corresponded to facilitation of the shortening reflex and increased S-cell activity during the elicited response. During habituation training, there was a decrement in both the shortening reflex and the elicited S-cell activity, along with decreased S-cell excitability. Conversely, dishabituation facilitated both the shortening response and S-cell activity during shortening, with an accompanying increase in S-cell excitability. Bath application of 1-10 micrometer serotonin (5HT), a modulatory neurotransmitter that is critical for sensitization, for full dishabituation, and for associative learning, increased S-cell excitability. S-cell excitability also increased after stimulation of the serotonergic Retzius cells. However, focal application of serotonin onto the S-cell soma hyperpolarized the interneuron, and bath application of a lower dose of serotonin (0.1 micrometer) decreased excitability. The observed changes in postsynaptic excitability appear to contribute to non-associative learning, and modulatory neurotransmitters, such as serotonin, evidently help regulate excitability. Such changes in S-cell excitability may also be relevant for more complex, associative forms of learning.

Nerve injury induces a rapid efflux of nitric oxide (NO) detected with a novel NO microsensor.

An early step in repair of the leech CNS is the appearance of endothelial nitric oxide synthase (eNOS) immunoreactivity and NOS activity, but coincident generation of NO at the lesion after injury has not been shown. This is important because NO can regulate microglial cell motility and axon growth. Indirect measurement of NO with the standard citrulline assay demonstrated that NO was generated within 30 min after nerve cord injury. A polarographic NO-selective self-referencing microelectrode that measures NO flux noninvasively was developed to obtain higher spatial and temporal resolution. With this probe, it was possible to demonstrate that immediately after the leech CNS was injured, NO left the lesion with a mean peak efflux of 803 +/- 99 fmol NO cm(-2) sec(-1). NO efflux exponentially declined to a constant value, as described through the equation f(t) = y(o) + ae(-t/tau), with tau = 117 +/- 30 sec. The constant y(o) = 15.8 +/- 4.5 fmol cm(-2) represents a sustained efflux of NO. Approximately 200 pmol NO cm(-2) is produced at the lesion (n = 8). Thus, injury activates eNOS already present in the CNS and precedes the accumulation of microglia at the lesion, consistent with the hypothesis that NO acts to stop the migrating microglia at the lesion site.

Action potential reflection and failure at axon branch points cause stepwise changes in EPSPs in a neuron essential for learning.

In leech mechanosensory neurons, action potentials reverse direction, or reflect, at central branch points. This process enhances synaptic transmission from individual axon branches by rapidly activating synapses twice, thereby producing facilitation. At the same branch points action potentials may fail to propagate, which can reduce transmission. It is now shown that presynaptic action potential reflection and failure under physiological conditions influence transmission to the same postsynaptic neuron, the S cell. The S cell is an interneuron essential for a form of nonassociative learning, sensitization of the whole body shortening reflex. The P to S synapse has components that appear monosynaptic (termed "direct") and polysynaptic, both with glutamatergic pharmacology. Reflection at P cell branch points on average doubled transmission to the S cell, whereas action potential failure, or conduction block, at the same branch points decreased it by one-half. Each of two different branch points affected transmission, indicating that the P to S connection is spatially distributed around these branch points. This was confirmed by examining the locations of individual contacts made by the P cell with the S cell and its electrically coupled partner C cells. These results show that presynaptic neuronal morphology produces a range of transmission states at a set of synapses onto a neuron necessary for a form of learning. Reflection and conduction block are activity-dependent and are basic properties of action potential propagation that have been seen in other systems, including axons and dendrites in the mammalian brain. Individual branch points and the distribution of synapses around those branch points can substantially influence neuronal transmission and plasticity.

Nitric oxide influences injury-induced microglial migration and accumulation in the leech CNS.

Damage to the leech or mammalian CNS increases nitric oxide (NO) production and causes accumulation of phagocytic microglial cells at the injury site. The aim of this study was to determine whether NO plays a role in microglial migration and accumulation at lesions in which NO is generated by a rapidly appearing endothelial nitric oxide synthase (eNOS) in leeches. Immunohistochemistry and cytochemistry demonstrated active eNOS before and throughout the period of microglial accumulation at the lesion. Decreasing NO synthesis by application of the NOS inhibitor N(w)-nitro-L-arginine methyl ester (1 mM) significantly reduced microglial accumulation, whereas its inactive enantiomer N(w)-nitro-D-arginine methyl ester (1 mM) resulted in microglial accumulation similar to that in crushed controls. Increasing NO with the donor spermine NONOate (SPNO) (1 mM) also inhibited accumulation, but not in the presence of the NO scavenger 2-(4-carboxyphenyl)-4,4,5, 5-teramethylimidazoline-oxyl-3-oxide (50 microM). The effect of SPNO was reversed by washout. SPNO application reduced average microglial migratory speeds and even reversibly arrested cell movement, as measured in living nerve cords. These results suggest that NO produced at a lesion may be a stop signal for microglia to accumulate there and that it can act on microglia early in their migration. Thus, NO may assume a larger role in nerve repair and recovery from injury by modulating accumulation of microglia, which appear to be important for axonal regeneration.

Serotonin depletion does not prevent intrinsic sensitization in the leech.

Intrinsic sensitization is a form of behavioral facilitation that is distinct from the extrinsic sensitization normally studied. To examine whether intrinsic and extrinsic sensitization are mediated by different physiological processes, the effects of 5,7-dihydroxytryptamine-induced serotonin (5-HT) depletion on intrinsic sensitization of the leech whole-body shortening response were observed. Previous experiments have shown that 5-HT depletion disrupts dishabituation and extrinsic sensitization of this behavior in the leech. Intrinsic sensitization was observed in preparations from both control and 5-HT-depleted animals, indicating that this form of behavioral facilitation was not affected by 5-HT depletion. The differences in the effects of 5-HT depletion on intrinsic versus extrinsic sensitization suggest that there are distinct neurophysiological processes mediating these two forms of behavioral facilitation. In addition, 5-HT depletion appeared to disrupt a putative extrinsic form of habituation of the shortening reflex. These data support the hypothesis that both intrinsic and extrinsic processes of neuromodulation mediate habituation and sensitization.

Modulatory effects of myomodulin on the excitability and membrane currents in Retzius cells of the leech.

Ion channel modulation by the peptide myomodulin (MM) has been demonstrated in a wide variety of organisms including Aplysia, Lymnaea, and Pleurobranchaea. This neural and muscular modulation has been shown to be important for shaping and modifying behavior. In this paper, we report that MM modulates several distinct ionic channels in another species, the medicinal leech Hirudo medicinalis. Experiments have focused on the Retzius cell (R) because the R cell is a multifunction neuron that has been implicated in a number of behaviors including feeding, swimming, secretion, thermal sensing, and the touch elicited shortening reflex and its plasticity. Previous work had identified a MM-like peptide in the leech and demonstrated that this peptide modulated the excitability of the R cell. Using combined current- and voltage-clamp techniques to examine the effects of MM on the R cell, we found that in response to a step pulse, MM increased the excitability of the R cell such that the cell fires more action potentials with a shorter latency to the first action potential. We found that this effect was mediated by the activation of a Na+-mediated inward current near the cell resting membrane potential. Second, we found that MM differentially modulated the potassium currents IA and IK. No effect of MM was found on IA, whereas MM significantly reduced both the peak and steady-state amplitudes of IK by 49 +/- 2.9% and 43 +/- 7.2%, respectively (means +/- SE). Finally we found that MM reduced the amplitude of the Ca2+ current by approximately 20%. The ionic currents modulated by MM are consistent with the overall effect of MM on the cellular activity of the R cell. An understanding of the cellular mechanisms by which MM modulates the activity of the R cell should help us to better understand the roles of both MM and the R cell in a variety of behaviors in the leech.

Identification and characterization of a myomodulin-like peptide in the leech.

A novel myomodulin-like peptide, GMGALRLamide, has been purified and sequenced from extracts of 1000 medicinal leech nerve cords. Synthetic leech myomodulin-like peptide blocked the specific staining pattern of leech ganglia by the antiserum against Aplysia myomodulin A PMGMLRLamide. Moreover, the synthetic leech myomodulin-like peptide GMGALRLamide showed identical neuronal modulation effect on the giant leech Retzius cell compare to that by the synthetic Aplysia myomodulin A PMGMLRLamide.

Injury-induced expression of endothelial nitric oxide synthase by glial and microglial cells in the leech central nervous system within minutes after injury.

It is known that nitric oxide (NO) is produced by injured tissues of the mammalian central nervous system (CNS) within days of injury. The aim of the present experiments was to determine the cellular synthesis of NO in the CNS immediately after injury, using the CNS of the leech which is capable of synapse regeneration, as a step towards understanding the role of NO in nerve repair. We report that within minutes after crushing the nerve cord of the leech, the region of damage stained histochemically for NADPH diaphorase, which is indicative of nitric oxide synthase (NOS) activity, and was immunoreactive for endothelial NOS (eNOS). On immunoblots of leech CNS extract, the same antibody detected a band with a relative molecular mass of 140,000, which is approximately the size of vertebrate eNOS. Cells expressing eNOS immunoreactivity as a result of injury were identified after freezing nerve cords, a procedure that produced less tissue distortion than mechanical crushing. Immunoreactive cells included connective glia and some microglia. Calmodulin was necessary for the eNOS immunoreactivity: it was blocked by calmodulin antagonist W7 (25 microM), but not by similar concentrations of the less potent calmodulin antagonist W12. Thus in the leech CNS, in which axon and synapse regeneration is successful, an increase in NOS activity at lesions appears to be among the earliest responses to injury and may be important for repair of axons.

Generalization of habituation and intrinsic sensitization in the leech.

Using the shortening reflex of the medicinal leech Hirudo medicinalis we examined stimulus generalization of habituation learning. Preparations received mechanosensory stimulus at two positions on the leech body wall, one site used to carry out habituation training and a second novel site to test for generalization of habituation. After training, the specific mechanosensory neurons activated by each stimulus were assessed using intracellular recordings. As expected, the closer the two sites were to each other, the greater the degree of generalization of habituation at the novel site and the more sensory cells were shared. However, a form of behavioral facilitation was observed at the trained site that resembled behavioral sensitization, but differed from the standard sensitization process in several respects. (1) Facilitation was induced by stimulation of the novel site before habituation training at the trained site, although the stimulus intensity at the novel site was equivalent to the training stimuli and was not the strong, noxious stimuli that normally induce sensitization. (2) The magnitude of the facilitating effect was proportional to the proximity of the novel and trained stimulation sites. (3) Although behavior at the trained site was facilitated, behavior at the novel site was habituated, indicating that the induced behavioral facilitation did not generalize throughout the animal, as normally occurs during sensitization, but was limited to a single stimulus-response pathway.

Regeneration of a central synapse restores nonassociative learning.

Sensitization is a form of nonassociative learning in which a strong or noxious stimulus persistently enhances the response produced by a weaker stimulus. In the leech Hirudo medicinalis, the S-interneuron is required for sensitization of the shortening response. A single S-cell axon was surgically separated from its sole synaptic partner, the neighboring S-cell. This consistently eliminated sensitization without impairing reflexive shortening itself, as measured in semi-intact specimens. Sensitization of the shortening reflex returned after 3 weeks when the severed axon grew and regenerated its specific electrical synapse within the nerve cord, as shown by restored conduction of impulses between S-cells. This confirms the essential role of one neuron, the S-cell, in sensitization, and it demonstrates that regeneration of the synapse between S-cells restores this example of nonassociative learning.

Localization of the myomodulin-like immunoreactivity in the leech CNS.

The distribution of myomodulinlike immunoreactivity in the leech CNS was determined using an antiserum raised against Aplysia myomodulin. Segmental ganglia contained approximately 60 immunoreactive neurons. In addition, numerous fibers containing immunoreactive varicosities were found throughout the neuropil. Using a combination of Lucifer Yellow injections and immunocytochemistry, we identified neurons including the anterior Pagodas (AP), annulus erector (AE), motor neurons, Leydig, longitudinal muscle motoneurons (L), S cells, and coupling interneurons, all of which are active during the touch-elicited shortening reflex. FMRF-amide-like immunoreactivity in three of these cells (L, AP, and AE) was previously demonstrated. Specific staining for myomodulin was abolished by preadsorption of the antiserum with synthetic myomodulin, but not with FMRF-amide. These results suggest a potential role for myomodulin in both intrinsic and extrinsic modulation of the leech touch-elicited shortening reflex. Further, it is possible that several neurons mediating this reflex contain multiple neuromodulatory peptides.

What we have learned from the study of learning in the leech.

The use of invertebrate preparations has contributed greatly to our understanding of the neural basis of learning. The leech is especially useful for studying behavioral changes and their underlying neuronal mechanisms. Learning in the leech is essentially identical to that found in other animals, both vertebrate and invertebrate. Using anatomical and physiological techniques on leeches as they learn, we have begun to characterize the properties of individual neurons and neuronal networks that play a role in learning. We have been able to show two neuronal mechanisms that have not been previously associated with associative conditioning. The first has to do with the importance of contingency: one stimulus [the conditional stimulus (CS)] becomes associated with a second stimulus [the unconditional stimulus, (US)] in proportion to the ability of the CS to predict the US. We have found that important properties for encoding predictability, such as circuit reconfiguration, may lie in the US pathway. The firing of the serotonergic Retzius cells is taken as the US; consistent CS prediction of a US prevents "dropout" of a critical component of one US pathway. Throughout training, predicted USs continue to elicit a barrage of action potentials in these cells. Recurring unpredicted USs degrade both the learning and the response of the Retzius cell to the US. A second insight is that at least two US pathways contribute to learning, the Retzius cell pathway and the nociceptive (N) cell pathway. This second pathway persists after the elimination of the Retzius cell pathway. The observation of multiple US pathways raises a host of issues concerning CS-US convergence and the functional significance of distinct US pathways, and our results are discussed in terms of implications to current models of learning.

Serotonin depletion impairs but does not eliminate classical conditioning in the leech Hirudo medicinalis.

The goal of these experiments was to test the role of serotonin (5HT) in classical conditioning of the touch-elicited shortening reflex in the leech (Hirudo medicinalis). The toxin 5,7-dihydroxytryptamine (5,7-DHT) was used to deplete serotonin. The results indicated that 5HT depletion significantly impairs the expression of conditioned responding; however, depleted leeches experiencing conditioned stimulus-unconditioned stimulus (CS-US) pairings still performed significantly better than depleted leeches experiencing unpaired CS-US presentations, suggesting that a 5HT-dependent mechanism does not account fully for learning in this preparation. Moreover, the residual pairing dependent effect is observed, although the depletions eliminate sensitization, suggesting that the amplification of sensitization may not be sufficient to account for classical conditioning of this reflex. Histological analyses of the ganglia revealed an absence of staining in 100% of the Retzius cells in the toxin group.

The S cell: an interneuron essential for sensitization and full dishabituation of leech shortening.

Sensory neurons in the leech excite the S interneuron, which in turn excites motoneurons that shorten the leech, although activity in the S cell reportedly cannot by itself shorten the animal. Experiments were performed in semi-intact leeches using established dishabituation and sensitization protocols. S-cell activity increased during reflexive shortening once the animal was sensitized or dishabituated with a strong shock. S-cell activity otherwise was not associated with shortening. To test the role of the S-cell in dishabituation and sensitization of the shortening reflex, single S cells were ablated in vivo by intracellular injections of pronase. S-cell lesions reduced but did not eliminate dishabituation; however, sensitization was completely disrupted. This was consistent with recent evidence that separate processes contribute to dishabituation and sensitization. Since the S cell in each ganglion is a link in a rapidly conducting chain along the length of the animal, it may be sufficient to break the chain at a single point to eliminate sensitization.

Associative learning modifies the shortening reflex in the semi-intact leech Hirudo medicinalis: effects of pairing, predictability, and CS preexposure.

Three experiments addressed the importance of the inter-event relationships of contiguity and contingency for associative learning in the semi-intact leech. It was found that both of these relationships are important for the leech to acquire a learned association between a touch (conditional stimulus, CS) and shock (unconditional stimulus, US). The learning can be extinguished if training is followed by explicitly unpaired presentations of the CS and US, which removes the contiguity between the stimuli. Learning is degraded by the introduction of unpredicted USs, as well as by unreinforced presentations of the CS (CS preexposure), both manipulations reduce the contingency between the CS and US. These results suggest that the associative process in both vertebrates and invertebrates share considerable functional similarity in the inter-event relationships important to learning.

FMRF-amide modulates the electrical activity of the leech Retzius cell.

The effect of the peptide FMRF-amide on the electrical activity of the leech Retzius (R) cell was investigated using electrophysiological techniques. FMRF-amide and six structurally related analogs increased the excitability of the R cell in several distinct ways that could act in concert to modulate transmitter release. 'Puffs' of FMRF-amide transiently depolarized the cell leading to a barrage of action potentials. This depolarization was followed by a phase of rhythmical bursting that appeared intrinsic to the neuron. FMRF-amide also broadened the plateau of the Ca(2+)-dependent action potential. The results suggest that the terminal Phe and Arg as well as the C-terminal amide are critical for the activity of these peptides.