Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, NEUROSCIENCE ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 18 August 2017

Leech Mechanosensation

Summary and Keywords

The medicinal leech (Hirudo verbana) is an annelid (segmented worm) and one of the classic model systems in neuroscience. It has been used in research for over 50 years and was one of the first animals in which intracellular recordings of mechanosensory neurons were carried out. Remarkably, the leech has three main classes of mechanosensory neurons that exhibit many of the same properties found in vertebrates. The most sensitive of these neurons are the touch cells, which are rapidly adapting neurons that detect low-intensity mechanical stimuli. Next are the pressure cells, which are slow-adapting sensory neurons that respond to higher intensity, sustained mechanostimulation. Finally, there are nociceptive neurons, which have the highest threshold and respond to potentially damaging mechanostimuli, such as a pinch. As observed in mammals, the leech has separate mechanosensitive and polymodal nociceptors, the latter responding to mechanical, thermal, and chemical stimuli. The cell bodies for all three types of mechanosensitive neurons are found in the central nervous system where they are arranged as bilateral pairs. Each neuron extends processes to the skin where they form discrete receptive fields. In the touch and pressure cells, these receptive fields are arranged along the dorsal-ventral axis. For the mechano-only and polymodal nociceptive neurons, the peripheral receptive fields overlap with the mechano-only nociceptor, which also innervates the gut. The leech also has a type of mechanosensitive cell located in the periphery that responds to vibrations in the water and is used, in part, to detect potential prey nearby.

In the central nervous system, the touch, pressure, and nociceptive cells all form synaptic connections with a variety of motor neurons, interneurons, and even each other, using glutamate as the neurotransmitter. Synaptic transmission by these cells can be modulated by a variety of activity-dependent processes as well as the influence of neuromodulatory transmitters, such as serotonin. The output of these sensory neurons can also be modulated by conduction block, a process in which action potentials fail to propagate to all the synaptic release sites, decreasing synaptic output. Activity in these sensory neurons leads to the initiation of a number of different motor behaviors involved in locomotion, such as swimming and crawling, as well as behaviors designed to recoil from aversive/noxious stimuli, such as local bending and shortening. In the case of local bending, the leech is able to bend in the appropriate direction away from the offending stimuli. It does so through a combination of which mechanosensory cell receptive fields have been activated and the relative activation of multiple sensory cells decoded by a layer of downstream interneurons.

Keywords: Hirudo, mechanosensory, nociception, conduction block, synapse, mechanostimuli, T cells

Organization of Mechanosensory Afferents

Mechanosensation in the medicinal leech (Hirudo verbana) is mediated primarily by three different types of afferents (sensory neurons) whose cell bodies are located in the central nervous system (CNS) and organized into bilateral pairs, with each neuron innervating a similar region of the skin on the same side as (ipsilateral to) the cell body (Nicholls & Baylor, 1968). First are the three pairs of light touch-sensitive (T) cells. These are low-threshold, rapidly adapting sensory neurons that primarily encode the temporal qualities (i.e., the velocity) of applied mechanosensory stimuli during the onset and offset phases of the stimulation (Nicholls & Baylor, 1968; Carlton & McVean, 1995). [Adaptation refers to the degree to which neuron activity decreases or stops altogether, even though a stimulus is still being applied.] The receptive fields of each T cell innervates either the dorsal, lateral, or ventral receptive fields on one side (Figure 1). The fine processes from the T cells terminate just below the skin surface between the epethelial cells, and, unlike rapidly adapting afferents in mammals, these are bared nerve endings that lack accessory structures such as Pacinian corpuscles (Blackshaw, 1981). This lack of accessory structures indicates that what makes the T cells rapidly adapting is due to the properties of the mechanoactivated ion channels that transduce mechanostimuli, along with the complement of voltage-gated Na+, voltage-gated K+, and Ca2+-gated K+ channels that shape the action potential and regulate spike-firing frequency.

Second are the two pairs of neurons that respond to sustained pressure (P cells), which are higher-threshold, slow-adapting afferents that appear to encode the actual magnitude of the mechanostimulation. P cell receptive fields innervate the dorsal or ventral regions of the skin and overlap (Figure 1) (Kristan et al., 1982). The morphology of the fine processes in the skin has not been studied.

Finally, the two pairs of nociceptive neurons (N cells) respond to especially strong mechanostimulation and damage to the skin, for example, a pinch or a cut (Nicholls & Baylor, 1968; Carlton & McVean, 1995; Pastor et al., 1996). Similar to the P cells, these are slow-adapting afferents (Nicholls & Baylor, 1968; Pastor et al., 1996). The two N cells are further divided into the lateral N cells, which are polymodal nociceptors responding to noxious mechanical, thermal, and chemical stimuli and medial N cells that are strictly mechanosensitive nociceptors (Pastor et al., 1996; Summers et al., 2014, 2015). Both N cells innervate the same half of the leech body (Figure 1) with the mechanonociceptors also innervating the gut. Similar to nociceptors in other animal species (including mammals), leech N cells terminate with bare endings in the skin, albeit at a much deeper level compared to the T cells (Blackshaw et al., 1982).

Leech MechanosensationClick to view larger

Figure 1. Distribution of receptive fields for the three different types of mechanosensory neurons (not drawn to anatomical scale). Each T cell innervates either the dorsal (Td), lateral (Tl), or ventral (Tv) region of skin. Each P cell innervates either a dorsal (Pd) or ventral (Pv) region of skin. Both N cells innervate the same region of skin. Not shown is the potential overlap between adjacent receptive fields along the dorsal/ventral axis.

Comparable mechanosensory neurons have also been found in other species of leech, including Macrobdella, Haementaria, and Erpobdella (Kramer & Goldman, 1981; Johansen & Kleinhaus, 1986; Baltzley et al., 2010). In many cases, the sensory functions and the behaviors elicited by these homologous sensory neurons appear to be conserved (Kramer, 1981; Gibbons & Baltzley, 2014). However, this is not always the case. On the one hand, P cells in Hirudo elicit local bending characterized by asymmetrical contraction of muscles ipsilateral to the activated P cell (Kristan et al., 1982). Homologous P cells in Erpobdella, on the other hand, elicit symmetrical muscle contraction (referred to as “tensing”) (Baltzley et al., 2010). These species-specific differences in behavior appear to be due, in part, to differences in P-to-P synaptic connections.

All three types of afferents have both major and minor receptive fields (Figure 2). Major receptive fields are located in the same segment as the sensory cell, and the T, P, or N cell projects an axon that extends to the periphery via the segmental nerves (the segmental nerves carry both motor and sensory axons between the CNS and the periphery) (Nicholls & Baylor, 1968; Yau, 1976). Minor receptive fields are located in adjacent anterior and posterior segments. The axons carrying minor field signals take a different path, projecting via the connective nerves that link adjacent ganglia and then traveling out the segmental root in the adjacent segment (Yau, 1976). Any mechanosensory stimulus applied to the skin activates the major receptive field belonging to sensory neurons located in the same segment where the stimulus was applied, as well as the minor receptive fields belonging to neurons located in adjacent segment. Afferents stimulated via their major receptive field tend to be more strongly activated than those activated via their minor receptive fields (Kristan et al., 1982). This effect may contribute to stimulus localization.

Central Synapses and Their Modulation

In the CNS the T, P, and N cells form synapses onto targets in their “home” ganglion and onto targets in adjacent ganglia (Figure 2) (Baylor & Nicholls, 1969; Nicholls & Purves, 1970; Yau, 1976; Baccus et al., 2001). In all studies to date, glutamate appears to be the transmitter used by these afferents in their central synapses (Baccus, 1998; Wessel et al., 1999; Baccus et al., 2000; Li & Burrell, 2008; Yuan & Burrell, 2010; Higgins et al., 2013). An interesting distinction between the afferents is that N and P cell transmission is mediated by standard monosynaptic and polysynaptic chemical synapses, with perhaps small contributions made by electrical synapses as well (Nicholls & Baylor, 1968; Todd et al., 2010; Higgins et al., 2013). However, no monosynaptic chemical synapses have been reported in T cells, and these afferents appear to communicate via a combination of monosynaptic electrical synapses and polysynaptic chemical synapses (Nicholls & Purves, 1970; Muller & Scott, 1981; Li & Burrell, 2008). Even in the case of these polysynaptic chemical connections, the intervening neurons also appear to be driven by electrical synapses from the T cells, given that octanol treatment (which can inhibit electrical synaptic transmission) reversibly inhibits polysynaptic signaling (Johnston et al., 1980; Li, 2010). It is not clear why this distinction exists between synapses made by the T cells versus the other sensory neurons, although one possibility will be discussed in Behavioral Functions of Mechanosensory Neurons.

Leech MechanosensationClick to view larger

Figure 2. Organization of major and minor receptive fields, in this case in the P-cell. The sensory neurons project an axon to the periphery via the segmental nerve of their “home” ganglion that innervates the skin in that segment, forming the major receptive field. The same sensory neuron also has minor receptive fields in the adjacent anterior and posterior segments. The axon from these minor fields enters the CNS via the segmental nerve belonging to the adjacent ganglion and then travels to the home segment via the connective nerve. All three types of mechanosensory neurons form synapses onto target cells (labeled as “X”) in both the “home” ganglion and in the adjacent ganglia. Modified from Baccus et al. (2000).

In addition to synaptic transmission onto motor neurons and interneurons responsible for initiating behavior, T, P, and N cells also make synaptic contact with each other (Baylor & Nicholls, 1969; Burgin & Szczupak, 2003). These sensory-to-sensory neuron synaptic connections are observed both within the same ganglion and onto sensory cells in adjacent ganglia; they are a mixture of electrical synapses and monosynaptic and polysynaptic chemical connections, the latter often being inhibitory (Baylor & Nicholls, 1969). T cells, in particular, appear to be strongly inhibited by both P and N cells. However, this inhibition seems to be effective only when T cell activity is elicited centrally and does not affect activity elicited in the periphery (Burgin & Szczupak, 2003). The inhibitory signaling between afferents may represent a form of lateral inhibition, possibly as a mechanism to enhance the capacity to determine stimulus location by reducing the activity of sensory neurons with adjacent receptive fields.

Afferent synapses can be modulated through changes in behavioral state or through activity-dependent processes. One example of behavioral changes are the changes in P cell synaptic signaling that take place as a consequence of feeding. During feeding, leeches are unresponsive to mechanical stimuli that elicit swimming, bending, or shortening (Misell et al., 1998). These behaviors are all initiated by the P cells, and P cell synaptic transmission is depressed during feeding, with serotonin acting presynaptically to reduce neurotransmitter release (Gaudry & Kristan, 2009).

The inhibitory neurotransmitter GABA (gamma amino butyric acid) can also modulate mechanosensory synaptic transmission and potentially afferent activity as well, but the effects of GABA differ between the afferents (Sargent et al., 1977; Wang et al., 2015). GABA hyperpolarizes the P cells, and application of the GABA-A receptor antagonist bicuculline increases P cell synaptic transmission as a result of disinhibition at the presynaptic level. However, GABA depolarizes the N cells, and bicuculline treatment actually reduces N cell synaptic transmission, presumably owing to a loss of excitatory GABAergic input (disexcitation). The opposing effects of GABA are a consequence of differences in intracellular Cl levels (Wang et al., 2015). P cells have a relatively negative GABA equilibrium potential, indicating low levels of intracellular Cl, so that GABA receptor activation produces a Cl influx that hyperpolarizes the afferent. N cells have a relatively depolarized GABA equilibrium potential indicating elevated levels of intracellular Cl. Consequently, GABA receptor activation elicits a Cl efflux that depolarizes the N cell. T cell responses to GABA have not been tested, although they are likely similar to P cells since they clearly receive inhibitory synaptic input (Baylor & Nicholls, 1969; Burgin & Szczupak, 2003). Mammalian somatosensory afferents are also depolarized by GABA, although it is unclear if this is the case in all sensory neurons innervating the skin or only in the nociceptive afferents (Duchen, 1986; Gilbert et al., 2007).

Mechanosensory synapses in the leech are also modulated by endocannabinoids, which are often released as a result of stress or noxious stimuli (Butler & Finn, 2009; Katona & Freund, 2012). Endocannabinoids depress both polymodal and mechanical N cell synapses, acting at the presynaptic level; this depression leads to decreases in whole-body shortening, a defensive withdrawal reflex (Yuan & Burrell, 2010, 2012, 2013b; Wang & Burrell, 2016). These effects on N cells may contribute to endogenous antinociceptive modulatory processes that are also observed in mammals (Butler & Finn, 2009; Yang et al., 2016). Endocannabinoids also depress T cell synapses (Li & Burrell, 2009, 2010), and while the functional relevance of this form of modulation is unknown, it may contribute to associative learning (Li & Burrell, 2011). Surprisingly, endocannabinoids actually potentiate P cell synapses. This effect appears to be mediated not as a direct effect on the P synapses, but rather as a consequence of depression of inhibitory, GABAergic input to the P cells, resulting in disinhibition of the P synapses (Higgins et al., 2013; Wang & Burrell, 2016). It is possible that endocannabinoid-mediated disinhibition of P synapses may contribute to behavioral sensitization in the leech.

Leech mechanosensory neurons also undergo a variety of forms of activity-dependent synaptic plasticity, comparable to what is observed in mammalian synapses. In P cell synapses, long-term potentiation (LTP) can be elicited with either high-frequency stimulation of the P cell or paired activation of the presynaptic P cell and one of its postsynaptic targets (Burrell & Sahley, 2004; Grey & Burrell, 2010). In both cases, LTP is dependent on the N-methyl D-aspartate (NMDA) receptor (NMDAR). LTP can also be generated in T cell polysynaptic connections, but the cellular mechanisms differ based on how potentiation is elicited. When high-frequency stimuli are used, LTP in T synapses is not mediated NMDARs and is instead mediated by a combination of metabotropic glutamate receptors and voltage-gated Ca2+ channels (Burrell & Sahley, 2004; Burrell & Li, 2008). When LTP is induced by pairing presynaptic (T cell) and postsynaptic neuron activity (pre-before-post), an NMDAR-mediated form of potentiation is observed (Li & Burrell, 2011). Long-term depression of T cell signaling is also observed following either low-frequency stimulation or pairing protocols in which the activation of the postsynaptic target precedes T cell activity (post-before-pre). Interestingly, two forms of LTD are observed depending on the stimulation parameters (Li & Burrell, 2009, 2011). NMDAR-dependent LTD is observed when the duration of low-frequency stimulation is relatively short (7.5 mins) or when the post-before-pre interval is brief (< 10 secs). However, endocannabinoid-mediated LTD is observed following either long periods of stimulation (15 mins) or longer post-before-pre intervals (10–30 secs).

Mechanosensory afferents are also capable of modulating each other via heterosynaptic mechanisms. High-frequency stimulation of the T cell that elicits LTP in the activated pathway (a homosynaptic process) simultaneously elicits heterosynaptic LTD in the non-activated T cell synapses (Burrell & Sahley, 2004). Low-frequency stimulation of T cells can depress N cell synapses through an endocannabinoid-dependent mechanism, which has also been observed in the mammalian spinal cord (Yuan & Burrell, 2010, 2012, 2013a, 2013b; Yang et al., 2016). High-frequency stimulation of the polymodal N cells potentiates P cell synapses via endocannabinoid-mediated disinhibition of these synapses (Wang & Burrell, 2016). Interestingly, this same stimulation of the polymodal N cells elicits endocannabinoid-mediated depression of the mechano-only N cells.

Conduction Block and Action Potential Reflection

A different form of modulation in leech mechano-afferents involves changes in action potential propagation. As previously stated, these afferents have minor receptive fields in adjacent segments, and action potentials initiated in these minor fields have to travel via axons in the connective nerve until they enter the ganglion where the cell body and a majority of the central synapses are located. As they enter the “home” ganglion, the action potentials reach a branch point in which the relatively small-diameter fiber intersects with a large-diameter fiber, forming a “T” junction (Figure 2). Action potentials (or “spikes”) entering this branch point are susceptible to conduction block (Figure 3); this failure to propagate occurs because the current carrying the spike in the smaller fiber is insufficient to bring the larger fiber to threshold for the action potential to continue. (This phenomenon can also be observed in some branches carrying spikes from major fields; see Van Essen, 1973; Yau, 1976.) This conduction block decreases synaptic transmission because the action potentials do not stimulate neurotransmitter release at sites beyond the point of the block (see Figure 3) (Muller & Scott, 1981; Macagno et al., 1987; Gu et al., 1989; Gu, 1991). Conduction block can occur in an activity-dependent manner, with action potentials initially propagating successfully past branch points and then failing as activity increases in frequency. In this way, conduction block may act as a filter, decreasing the synaptic output of leech mechano-afferents as they become more persistently activated.

Leech MechanosensationClick to view larger

Figure 3. Examples of full conduction/propagation, reflection, and conduction block. Action potentials traveling down the relatively thin axon from the minor field are susceptible to propagation failures at the “T” junction when the thin axon meets the thick axon from the major field. During full conduction, action potentials are able to stimulate synapse on both sides of the branch point (normal synaptic transmission). During reflection, synapses prior to the branch point are actually stimulated twice (synaptic transmission is enhanced). During conduction block, only synapses prior to the branch point are stimulated (synaptic transmission is reduced). Modified from Baccus et al. (2000).

From a biophysical standpoint, conduction block is a consequence of the size disparity between the two axons forming the branch point and the membrane potential at the branch point. As the membrane potential becomes more negative (hyperpolarized), the propensity of conduction failure increases because it is more difficult to reach threshold (Van Essen, 1973). It is the changes in membrane potential that are activity-sensitive. As the frequency of spikes passing through the branch point increases, there is greater activity of both Na+/K+ pumps and Ca2+-gated K+ channels, which lead to increased hyperpolarization after activity (after hyperpolarization or AHP) and, consequently, increased conduction block (Jansen & Nicholls, 1973). Conduction block itself can be modulated by serotonin, which reduces the duration of block once it has been initiated (Mar & Drapeau, 1996). Other studies have focused on modulation of the AHP in T cells following repetitive stimulation, proposing that conduction block is occurring among the axon branches within the ganglion neuropil. These studies have found that serotonin can decrease the amplitude of the AHP by decreasing Na+/K+ pump and Ca2+-gated K+ channel activity. This decrease in AHP may prevent or reverse block and enhance synaptic transmission, although direct observations of action potential propagation were not made in these studies (Belardetti et al., 1984; Catarsi & Brunelli, 1991; Scuri et al., 2002; Cataldo et al., 2005; Scuri et al., 2007).

One issue that has not been studied is how conduction block is maintained as the leech grows (Baccus, 1998). The capacity for conduction block does not appear to differ between leeches of different sizes. This finding implies that the ratio of fiber diameters at the branch point is somehow maintained. It would be both interesting and innovative to study the interaction, and possibly homeostatic regulation, between morphology and electrical activity at the cellular level. A related issue is whether branch point morphology might change as an activity-dependent process, making the T-junction more or less likely to undergo conduction block. For example, one could envision spike-mediated Ca2+ entry acting on cytoskeletal elements to change the axon diameter (K. Muller, personal communication).

A second and remarkable form of modulation-based action potential propagation in leech mechanosensory neurons is referred to as “reflection” (Baccus, 1998). There is an intermediate state between full conduction and conduction block at the branch points in which an action potential traveling from the minor receptive fields is delayed in propagating from the small-diameter to large-diameter axon. This delay in propagation exceeds the duration of the action potential refractory period in the thin fiber. (Refractory period refers to the time, in milliseconds, in which a region of an axon cannot generate another spike and normally prevents action potentials from “backtracking” back the way they came.) This enables the currents from the action potential in the thick fiber to trigger a second action potential in the thin fiber that propagates back toward the minor field (in the direction from where the action potential originated; see Figure 3). Reflection results in a substantial increase in synaptic transmission, in part because a subset of presynaptic terminals are stimulated twice, once by the original action potential and again by the reflected spike (Baccus, 1998; Baccus et al., 2000). In addition, the interval between the original and reflected action potentials is short enough to result in short-term synaptic facilitation (Charlton & Bittner, 1978).

From a functional standpoint, conduction block may contribute to habituation, a nonassociative form of learning produced by repetitive stimulation (Groves & Thompson, 1970; Rankin et al., 2009). Reversal of conduction block or reflection may contribute to facilitatory forms of nonassociative learning, such as dishabituation or sensitization (Brunelli et al., 1997). As already stated, serotonin promotes full conduction from the blocked state, and serotonin has been shown to be critical for sensitization and dishabituation in the leech (Ehrlich et al., 1992; Zaccardi et al., 2004; Burrell & Sahley, 2005). Whether serotonin also promotes reflection is unknown, and no direct observations of conduction block or reflection have taken place during these forms of nonassociative learning.

Behavioral Functions of Mechanosensory Neurons

Mechanosensory stimuli initiate five main behaviors: the withdrawal reflexes local bending, localized shortening, and whole-body shortening, as well as the locomotory behaviors swimming and crawling (Kristan et al., 2005). The P and N cells are both capable of eliciting whole-body shortening, swimming, and crawling. At least in terms of whole-body shortening and swimming, there appears to be a bias along the anterior-to-posterior axis, in which anterior stimuli tend to elicit shortening and posterior stimuli tend to elicit swimming. This would appear to make sense if the goals of these behaviors are for the leech to move away from noxious stimuli. For the whole-body shortening reflex, N cell stimulation appears to be more effective than P cells at eliciting this reflex. Activation of multiple P cells is required to elicit whole-body shortening, whereas the same level of stimulation in a single N cell can elicit this reflex (Shaw & Kristan, 1995; Yuan & Burrell, 2013b).

The mechanisms by which mechanosensory stimuli elicit the local bending reflex have received considerable attention. Specifically, studies have focused on understanding how the leech determines the stimulus position on the body in order to produce the appropriate behavioral response, a bend away from the offending stimulus. This circuit provides an attractive model for examining how the nervous system encodes stimulus features that can vary along a continuum, such as touch position by the mammalian somatosensory cortex or object orientation by the primary visual cortex. Stimuli that produce local bending activate both T and P cells. However, the T cells do not appear to contribute to either the magnitude of the response (the T cell response rapidly saturates as stimulus intensity increases) or the stimulus position (Kristan et al., 1982; Lewis & Kristan, 1998a, 1998b; but also see the study by Pirschel & Kretzberg, 2016). This latter finding is remarkable given that the T cells have distinct receptive fields innervating the dorsal, lateral, and ventral skin on one side, while the P cell receptive fields innervate an entire dorsal or ventral quadrant of the body. Nevertheless, it appears that the P cells play a critical role in encoding the spatial position of the stimuli that ultimately determine the direction of the local bend. Because P cells have overlapping receptive fields, a bending-eliciting stimulus will likely activate two P cells. The degree to which each P cell is activated (number of action potentials) depends on the distance from the center of the receptive field; the closer to the center, the greater the activity (Kristan et al., 1982; Lewis & Kristan, 1998b). Therefore, the relative level of activity by adjacent P cells provides information about stimulus location. The relative response latency of the two P cells (the difference between the latency of the first spike by the first P cell to respond and the first spike of the second P cell to respond) may also provide positional information (Thomson & Kristan, 2006). The pattern of P cell activation stimulates a population of interneurons. The subset of these interneurons that are activated, in turn, generates the appropriate combination of excitation and inhibition in both excitatory and inhibitory motor neurons that will produce a bend in the appropriate direction (Lockery & Kristan, 1990; Lewis & Kristan, 1998a; Baca et al., 2005). At the motor neuron level, GABA-mediated lateral inhibition of the excitatory motor neurons, presumably initiated by P activity, also appears to play a role in producing the appropriate bend direction (Baca et al., 2008).

On their own T cells do not appear to be capable of eliciting any behaviors in the leech. What then is the functional role of these afferents? A study by Pirschel and Kretzberg (2016) suggests that T cells do contribute to encoding the position of tactile stimulation that ultimately produces the appropriate bend direction. Specifically, these authors propose that a population-coding mechanism derived from both T and P cell activity is being used in combination with the relative latency parameter to encode stimulus position and that the combined activity of T and P cells encodes the intensity of the applied stimulus. Resolving the ultimate role of T cells during local bending will require further investigation. One potential approach might be to lesion one or more T cells (Gu et al., 1989; Shaw & Kristan, 1999) and examine its effect on how the leech CNS translates the stimulus position to produce bending in the appropriate direction.

Other alternative functions of the T cells include the following. Since these afferents are rapidly adapting, they may be able to detect fine level details in the substrate with which the leech is in contact. This could be important for distinguishing preferred versus nonpreferred substrates or for detecting whether a leech is in contact with a potential prey.

Another potential role for T cells is that, instead of activating behaviors directly, perhaps these afferents activate modulatory networks that regulate the performance or gain of those behaviors. Kristan and colleagues have recently identified a “preparatory network” of neurons that are activated with a very short latency immediately prior to the production of multiple behaviors, including local bending and whole-body shortening (Frady et al., 2016). This preparatory network appears to activate motor neurons that increase muscle tension immediately prior to the actual motor response. One prominent element of the preparatory network is the S interneuron, which does not initiate any motor behaviors but is critical for the initiation of sensitization of the whole-body shortening reflex and may also contribute to the increase magnitude of shortening following sensitization (Sahley et al., 1994; Modney et al., 1997; Burrell et al., 2001; Burrell & Sahley, 2005). Although T cell input to the preparatory network has not been specifically examined, T cells have the strongest synaptic input to the S cell (Muller & Scott, 1981; Burrell & Sahley, 2004; Li & Burrell, 2011) and the temporal characteristics of T cell activity (i.e., that it is rapidly adapting) correspond to those of the preparatory network. This may also explain, in part, the functional relevance of T cell electrical synapses discussed earlier. Electrical synapses have a shorter latency compared to chemical synapses (Nicholls & Purves, 1970), and such fast-acting synaptic inputs would facilitate rapid communication of sensory information to the preparatory network. Activation of the preparatory network may not only prepare the leech to “do something,” but may also provide a degree of initial gain in control of the behavior so that it is performed with a shorter latency, at a higher/stronger magnitude, or for a longer duration.

Peripheral Mechanosensory Cells

At least two types of mechanosensory neurons are located in the periphery. First are the water motion detectors, which are located in peripheral sensory structures known as sensilla. Each segment in the leech has 14 sensilla arranged in a ring around the central annulus (the middle section of each segment). These sensilla contain both simple eyes and very small (1–9 µm long) hairs that extend from the sensilla and are thought to act as water motion detectors (Kretz et al., 1976; Derosa & Friesen, 1981; Friesen, 1981). Peripheral neurons in the sensilla carry information about water vibrations to a variety of targets in the CNS (Friesen, 1981; Phillips & Friesen, 1982). These cells have a lower threshold than the T cells and can detect the movement of water across the leech, which can be used to initiate orientation and locomotion toward the vibration source as part of the leech’s effort to find prey (Young et al., 1981; Dickinson & Lent, 1984; Harley et al., 2011).

A second type of peripheral mechanosensory neurons are stretch receptors that are located in the longitudinal muscle layer and that project a single axon to the CNS via the segmental nerve (Blackshaw & Thompson, 1988). Although the stretch receptors do generate overshooting action potentials in the cell body, these do not propagate to the CNS, implying that the stretch receptors communicate to postsynaptic targets by grade changes in neurotransmission. These stretch receptors are capable of encoding both the static and dynamic components of changes in muscle length. At least one function of this afferent input is to provide feedback to the swim central pattern generator circuit (Cang & Friesen, 2000; Cang et al., 2001). The stretch receptors respond to changes in muscle length by hyperpolarization (Blackshaw & Thompson, 1988), indicating that there is likely tonic signaling by these afferents and that increases in muscle stretch are communicated as a disinhibitory signal. This proposal is supported by at least one study in which stretch receptors were found to form nonrectifying electrical connections onto an oscillatory interneuron within the swim neural circuit that inhibits both motor neurons and other swim circuit interneurons (Cang et al., 2001; Cang & Friesen, 2002).


The study of mechanosensory neurons in the leech has a long, rich history of discovery. Detailed knowledge of how these sensory cells function, along with the ease of recording from these neurons and their postsynaptic targets, provides a basis for future experimentation. This system lends itself to a number of interesting questions. For example, how do the mechanosensory neurons encode positional information and transmit them to the rest of the CNS? The answer to this question will shed light on the fundamental mechanisms whereby sensory features are processed throughout the animal kingdom. Another question involves nociception: the leech provides an excellent system for understanding the neuromodulatory mechanisms that mediate sensitized responses to nociceptive and non-nociceptive forms of sensitization. As more molecular genetic tools become available for use in the leech, it will be possible to use these mechanosensory neurons as a system to study what gene/proteins are specific to the functioning of the various types of somatosensory afferents—for example, rapidly adapting versus slow-adapting; light touch versus pressure sensitive; or mechanical versus polymodal nociceptors. These sensory cells are also an excellent place to study how action potential propagation impacts synaptic transmission—potentially by using voltage-sensitive dyes to directly monitor propagation at the branch point itself, to examine interaction between activity and conduction block versus reflection, and to determine whether activity can impact the morphology of the branch points themselves. In closing, the leech mechanosensory system provides a unique opportunity within neuroscience to directly link cellular and physiological processes to system- and behavioral-level function.


Baca, S. M., Marin-Burgin, A., Wagenaar, D. A., & Kristan, W. B., Jr. (2008). Widespread inhibition proportional to excitation controls the gain of a leech behavioral circuit. Neuron, 57, 276–289.Find this resource:

Baca, S. M., Thomson, E. E., & Kristan, W. B., Jr. (2005). Location and intensity discrimination in the leech local bend response quantified using optic flow and principal components analysis. Journal of Neurophysiology, 93, 3560–3572.Find this resource:

Baccus, S. A. (1998). Synaptic facilitation by reflected action potentials: enhancement of transmission when nerve impulses reverse direction at axon branch points. Proceedings of the National Academy of Sciences USA, 95, 8345–8350.Find this resource:

Baccus, S. A., Burrell, B. D., Sahley, C. L., & Muller, K. J. (2000). Action potential reflection and failure at axon branch points cause stepwise changes in EPSPs in a neuron essential for learning. Journal of Neurophysiology, 83, 1693–1700.Find this resource:

Baccus, S. A., Sahley, C. L., & Muller K. J. (2001). Multiple sites of action potential initiation increase neuronal firing rate. Journal of Neurophysiology, 86, 1226–1236.Find this resource:

Baltzley, M. J., Gaudry, Q., & Kristan, W. B., Jr. (2010). Species-specific behavioral patterns correlate with differences in synaptic connections between homologous mechanosensory neurons. Journal of Comparative Physiology: A, Neuroethology, Sensory, Neural, and Behavioral Physiology, 196, 181–197.Find this resource:

Baylor, D. A., & Nicholls, J. G. (1969). Chemical and electrical synaptic connexions between cutaneous mechanoreceptor neurones in the central nervous system of the leech. Journal of Physiology, 203, 591–609.Find this resource:

Belardetti, F., Brunelli, M., Demontis, G., & Sonetti, D. (1984). Serotonin and Retzius cell depress the hyperpolarization following impulses of leech touch cell. Brain Research, 300, 91–102.Find this resource:

Blackshaw, S. E. (1981). Morphology and distribution of touch cell terminals in the skin of the leech. Journal of Physiology, 320, 219–228.Find this resource:

Blackshaw, S. E., Nicholls, J. G., & Parnas, I. (1982). Physiological responses, receptive fields and terminal arborizations of nociceptive cells in the leech. Journal of Physiology, 326, 251–260.Find this resource:

Blackshaw, S. E., & Thompson, S. W. (1988). Hyperpolarizing responses to stretch in sensory neurones innervating leech body wall muscle. Journal of Physiology, 396, 121–137.Find this resource:

Brunelli, M., Garcia-Gil, M., Mozzachiodi, R., Scuri, R., & Zaccardi, M. L. (1997). Neurobiological principles of learning and memory. Archives Italiennes Biologie, 135, 15–36.Find this resource:

Burgin, A. M., & Szczupak, L. (2003). Network interactions among sensory neurons in the leech. Journal of Comparative Physiology: A, Neuroethology, Sensory, Neural, and Behavioral Physiology, 189, 59–67.Find this resource:

Burrell, B. D., & Li, Q. (2008). Co-induction of long-term potentiation and long-term depression at a central synapse in the leech. Neurobiology of Learning and Memory, 90, 275–279.Find this resource:

Burrell, B. D., & Sahley, C. L. (2004). Multiple forms of long-term potentiation and long-term depression converge on a single interneuron in the leech CNS. Journal of Neuroscience, 24, 4011–4019.Find this resource:

Burrell, B. D., & Sahley, C. L. (2005). Serotonin mediates learning-induced potentiation of excitability. Journal of Neurophysiology, 94, 4002–4010.Find this resource:

Burrell, B. D., Sahley, C. L., & Muller, K. J. (2001). Non-associative learning and serotonin induce similar bi-directional changes in excitability of a neuron critical for learning in the medicinal leech. Journal of Neuroscience, 21, 1401–1412.Find this resource:

Butler, R. K., & Finn, D. P. (2009). Stress-induced analgesia. Progress in Neurobiology, 88, 184–202.Find this resource:

Cang, J., & Friesen, W. O. (2000). Sensory modification of leech swimming: rhythmic activity of ventral stretch receptors can change intersegmental phase relationships. Journal of Neuroscience, 20, 7822–7829.Find this resource:

Cang, J., & Friesen, W. O. (2002). Model for intersegmental coordination of leech swimming: central and sensory mechanisms. Journal of Neurophysiology, 87, 2760–2769.Find this resource:

Cang, J., Yu, X., & Friesen, W. O. (2001). Sensory modification of leech swimming: interactions between ventral stretch receptors and swim-related neurons. Journal of Comparative Physiology A, 187, 569–579.Find this resource:

Carlton, T., & McVean, A. (1995). The role of touch, pressure and nociceptive mechanoreceptors of the leech in unrestrained behaviour. Journal of Comparative Physiology A, 177, 781–791.Find this resource:

Cataldo, E., Brunelli, M., Byrne, J. H., Av-Ron, E., Cai, Y., & Baxter, D. A. (2005). Computational model of touch sensory cells (T Cells) of the leech: role of the afterhyperpolarization (AHP) in activity-dependent conduction failure. Journal of Computers in Neuroscience, 18, 5–24.Find this resource:

Catarsi, S., & Brunelli, M. (1991). Serotonin depresses the after-hyperpolarization through the inhibition of the Na+/K+ electrogenic pump in T sensory neurones of the leech. Journal of Experimental Biology, 155, 261–273.Find this resource:

Charlton, M. P., & Bittner, G. D. (1978). Presynaptic potentials and facilitation of transmitter release in the squid giant synapse. Journal of General Physiology, 72, 487–511.Find this resource:

Derosa, Y. S., & Friesen, W. O. (1981). Morphology of leech Sensilla: Observations with the scanning electron microscope. Biological Bulletin, 160, 383–393.Find this resource:

Dickinson, M. H., & Lent, C. M. (1984). Feeding behavior of the medicinal leech, Hirudo medicinalis L. Journal of Comparative Physiology A, 154, 449–455.Find this resource:

Duchen, M. R. (1986). Excitation of mouse motoneurones by GABA-mediated primary afferent depolarization. Brain Research, 379, 182–187.Find this resource:

Ehrlich, J. S, Boulis, N. M., Karrer, T., & Sahley, C. L. (1992). Differential effects of serotonin depletion on sensitization and dishabituation in the leech, Hirudo medicinalis. Journal of Neurobiology, 23, 270–279.Find this resource:

Frady, E. P., Kapoor, A., Horvitz, E., & Kristan, W. B., Jr. (2016). Scalable semisupervised functional neurocartography reveals canonical neurons in behavioral networks. Neural Computation, 28, 1453–1497.Find this resource:

Friesen, W. O. (1981). Physiology of water motion detection in the medicinal leech. Journal of Experimental Biology, 92, 255–275.Find this resource:

Gaudry, Q., & Kristan, W. B., Jr. (2009). Behavioral choice by presynaptic inhibition of tactile sensory terminals. Nature Neuroscience, 12, 1450–1457.Find this resource:

Gibbons, K. R., & Baltzley, M. J. (2014). Differing synaptic strengths between homologous mechanosensory neurons. Invertebrate Neuroscience, 14, 103–111.Find this resource:

Gilbert, D., Franjic-Wurtz, C., Funk, K., Gensch, T., Frings S., & Mohrlen, F. (2007). Differential maturation of chloride homeostasis in primary afferent neurons of the somatosensory system. International Journal of Developmental Neuroscience, 25, 479–489.Find this resource:

Grey, K. B., & Burrell, B. D. (2010). Co-induction of LTP and LTD and its regulation by protein kinases and phosphatases. Journal of Neurophysiology, 103, 2737–2746.Find this resource:

Groves, P. M., & Thompson, R. F. (1970). Habituation: a dual-process theory. Psychological Review, 77, 419–450.Find this resource:

Gu, X. N. (1991). Effect of conduction block at axon bifurcations on synaptic transmission to different postsynaptic neurones in the leech. Journal of Physiology, 441, 755–778.Find this resource:

Gu, X. N., Macagno, E. R., & Muller, K. J. (1989). Laser microbeam axotomy and conduction block show that electrical transmission at a central synapse is distributed at multiple contacts. Journal of Neurobiology, 20, 422–434.Find this resource:

Harley, C. M., Cienfuegos, J., & Wagenaar, D. A. (2011). Developmentally regulated multisensory integration for prey localization in the medicinal leech. Journal of Experimental Biology, 214, 3801–3807.Find this resource:

Higgins, A., Yuan, S., Wang Y., & Burrell, B. (2013). Differential modulation of nociceptive versus non-nociceptive synapses by endocannabinoids. Molecular Pain, 9, 26.Find this resource:

Jansen, J. K., & Nicholls, J. G. (1973). Conductance changes, an electrogenic pump and the hyperpolarization of leech neurones following impulses. Journal of Physiology, 229, 635–655.Find this resource:

Johansen, J., & Kleinhaus, A. (1986). Differential sensitivity of tetrodotoxin of nociceptive neurons in 4 species of leeches. Journal of Neuroscience, 6, 3499–3504.Find this resource:

Johnston, M. F., Simon, S. A., & Ramon, F. (1980). Interaction of anaesthetics with electrical synapses. Nature, 286, 498–500.Find this resource:

Katona, I., & Freund, T. F. (2012). Multiple functions of endocannabinoid signaling in the brain. Annual Review of Neuroscience, 35, 529–558.Find this resource:

Kramer, A. P., & Goldman, J. R. (1981). The nervous system of the Glossiphoniid leech Haementeria ghilianii. Journal of Comparative Physiology, 144, 435–448.Find this resource:

Kramer, A. P. (1981). The nervous system of the Glossiphoniid leech Haementeria ghilianii. II. Synaptic pathways controlling body wall shortening. Journal of Comparative Physiology, 144, 449–457.Find this resource:

Kretz, J. R., Stent, G. S., & Kristan, W. B. (1976). Photosensory input pathways in the medicinal leech. Journal of Comparative Physiology, 106, 1–37.Find this resource:

Kristan, W. B., Jr., Calabrese, R. L., & Friesen, W. O. (2005). Neuronal control of leech behavior. Progress in Neurobiology, 76, 279–327.Find this resource:

Kristan, w. B., McGirrr, S. J., & Simpson, G. V. (1982). Behavioural and mechanosensory neurone responses to skin stimulation in leeches. Journal of Experimental Biology, 96, 143–160.Find this resource:

Lewis, J. E., & Kristan, W. B., Jr. (1998a). A neuronal network for computing population vectors in the leech. Nature, 391, 76–79.Find this resource:

Lewis, J. E., & Kristan, W. B., Jr. (1998b). Representation of touch location by a population of leech sensory neurons. Journal of Neurophysiology, 80, 2584–2592.Find this resource:

Li, Q. (2010). Cellular mechanisms mediating synaptic plasticity in the leech T-to-S neuronal circuit. In Basic Biomedical Sciences, p. 174. Vermillion: University of South Dakota.Find this resource:

Li, Q., & Burrell, B. D. (2008). CNQX and AMPA inhibit electrical synaptic transmission: A potential interaction between electrical and glutamatergic synapses. Brain Research, 1228, 43–57.Find this resource:

Li, Q., & Burrell, B. D. (2009). Two forms of long-term depression in a polysynaptic pathway in the leech CNS: One NMDA receptor-dependent and the other cannabinoid dependent. Journal of Comparative Physiology A, 195, 831–841.Find this resource:

Li, Q., & Burrell, B. D. (2010). Properties of cannabinoid-dependent long-term depression in the leech. Journal of Comparative Physiology: A, Neuroethology, Sensory, Neural, and Behavioral Physiology, 196, 841–851.Find this resource:

Li, Q., & Burrell, B. D. (2011). Associative, bidirectional changes in neural signaling utilizing NMDA receptor- and endocannabinoid-dependent mechanisms. Learning and Memory, 18, 545–553.Find this resource:

Lockery, S. R., & Kristan, W. B., Jr. (1990). Distributed processing of sensory information in the leech. II. Identification of interneurons contributing to the local bending reflex. Journal of Neuroscience, 10, 1816–1829.Find this resource:

Macagno, E. R., Muller, K. J., & Pitman, R. M. (1987). Conduction block silences parts of a chemical synapse in the leech central nervous system. Journal of Physiology, 387, 649–664.Find this resource:

Mar, A., & Drapeau, P. (1996). Modulation of conduction block in leech mechanosensory neurons. Journal of Neuroscience, 16, 4335–4343.Find this resource:

Misell, L. M., Shaw, B. K., & Kristan, W. B., Jr. (1998). Behavioral hierarchy in the medicinal leech, Hirudo medicinalis: Feeding as a dominant behavior. Behavioural Brain Research, 90, 13–21.Find this resource:

Modney, B. K., Sahley, C. L., & Muller, K. J. (1997). Regeneration of a central synapse restores nonassociative learning. Journal of Neuroscience, 17, 6478–6482.Find this resource:

Muller, K. J., & Scott, S. A. (1981). Transmission at a “direct” electrical connexion mediated by an interneurone in the leech. Journal of Physiology, 311, 565–583.Find this resource:

Nicholls, J. G., & Baylor, D. A. (1968). Specific modalities and receptive fields of sensory neurons in CNS of the leech. Journal of Neurophysiology, 31, 740–756.Find this resource:

Nicholls, J. G., & Purves, D. (1970). Monosynaptic chemical and electrical connexions between sensory and motor cells in the central nervous system of the leech. Journal of Physiology, 209, 647–667.Find this resource:

Pastor, J., Soria, B., & Belmonte, C. (1996). Properties of the nociceptive neurons of the leech segmental ganglion. Journal of Neurophysiology, 75, 2268–2279.Find this resource:

Phillips, C. E., & Friesen, W. O. (1982). Ultrastructure of the water-movement-sensitive sensilla in the medicinal leech. Journal of Neurobiology, 13, 473–486.Find this resource:

Pirschel, F., & Kretzberg, J. (2016). Multiplexed population coding of stimulus properties by leech mechanosensory cells. Journal of Neuroscience, 36, 3636–3647.Find this resource:

Rankin, C. H., Abrams, T., Barry, R. J., Bhatnagar, S., Clayton, D. F., Colombo, J., … Thompson, R. F. (2009). Habituation revisited: an updated and revised description of the behavioral characteristics of habituation. Neurobiology of Learning and Memory, 92, 135–138.Find this resource:

Sahley, C. L., Modney, B. K., Boulis, N. M., & Muller, K. J. (1994). The S cell: An interneuron essential for sensitization and full dishabituation of leech shortening. Journal of Neuroscience, 14, 6715–6721.Find this resource:

Sargent, P. B., Yau, K. W., & Nicholls, J. G. (1977). Extrasynaptic receptors on cell bodies of neurons in central nervous system of the leech. Journal of Neurophysiology, 40, 446–452.Find this resource:

Scuri, R., Lombardo, P., Cataldo, E., Ristori, C., & Brunelli, M. (2007). Inhibition of Na+/K+ ATPase potentiates synaptic transmission in tactile sensory neurons of the leech. European Journal of Neuroscience, 25, 159–167.Find this resource:

Scuri, R., Mozzachiodi, R., & Brunelli, M. (2002). Activity-dependent increase of the AHP amplitude in T sensory neurons of the leech. Journal of Neurophysiology, 88, 2490–2500.Find this resource:

Shaw, B. K., & Kristan, W. B., Jr. (1995). The whole-body shortening reflex of the medicinal leech: motor pattern, sensory basis, and interneuronal pathways. Journal of Comparative Physiology A, 177, 667–681.Find this resource:

Shaw, B. K., & Kristan, W. B., Jr. (1999). Relative roles of the S cell network and parallel interneuronal pathways in the whole-body shortening reflex of the medicinal leech. Journal of Neurophysiology, 82, 1114–1123.Find this resource:

Summers, T., Holec, S., & Burrell, B. D. (2014). Physiological and behavioral evidence of a capsaicin-sensitive TRPV-like channel in the medicinal leech. Journal of Experimental Biology, 217, 4167–4173.Find this resource:

Summers, T., Wang, Y., Hanten, B., & Burrell, B. D. (2015). Physiological, pharmacological and behavioral evidence for a TRPA1 channel that can elicit defensive responses in the medicinal leech. Journal of Experimental Biology, 218, 3023–3031.Find this resource:

Thomson, E. E., & Kristan, W. B. (2006). Encoding and decoding touch location in the leech CNS. Journal of Neuroscience, 26, 8009–8016.Find this resource:

Todd, K. L., Kristan, W. B., Jr., & French, K. A. (2010). Gap junction expression is required for normal chemical synapse formation. Journal of Neuroscience, 30, 15277–15285.Find this resource:

Van Essen, D. C. (1973). The contribution of membrane hyperpolarization to adaptation and conduction block in sensory neurones of the leech. Journal of Physiology, 230, 509–534.Find this resource:

Wang, Y., & Burrell, B. D. (2016). Differences in chloride gradients allow for three distinct types of synaptic modulation by endocannabinoids. Journal of Neurophysiology, 116, 619–628.Find this resource:

Wang, Y., Summers, T., Peterson, W., Miiller, E., & Burrell, B. D. (2015). Differential effects of GABA in modulating nociceptive vs. non-nociceptive synapses. Neuroscience, 298, 397–409.Find this resource:

Wessel, R., Kristan, W. B., Jr., & Kleinfeld, D. (1999). Supralinear summation of synaptic inputs by an invertebrate neuron: dendritic gain is mediated by an “inward rectifier” K(+) current. Journal of Neuroscience, 19, 5875–5888.Find this resource:

Yang, F., Xu, Q., Shu, B., Tiwari, V., He, S. Q., Vera-Portocarrero, L. P., … Guan, Y. (2016). Activation of cannabinoid CB1 receptor contributes to suppression of spinal nociceptive transmission and inhibition of mechanical hypersensitivity by Abeta-fiber stimulation. Pain, 157, 2582–2593.Find this resource:

Yau, K. W. (1976). Receptive fields, geometry and conduction block of sensory neurones in the central nervous system of the leech. Journal of Physiology, 263, 513–538.Find this resource:

Young, S. R., Dedwylder, R. D., & Friesen, W. O. (1981). Responses of the medicinal leech to water waves. Journal of Comparative Physiology: A, Neuroethology, Sensory, Neural, and Behavioral Physiology, 144, 111–116.Find this resource:

Yuan, S., & Burrell B. D. (2010). Endocannabinoid-dependent LTD in a nociceptive synapse requires activation of a presynaptic TRPV-like receptor. Journal of Neurophysiology, 104, 2766–2777.Find this resource:

Yuan, S., & Burrell, B. D. (2012). Long-term depression of nociceptive synapses by non-nociceptive afferent activity: Role of endocannabinoids, Ca(2)+, and calcineurin. Brain Research, 1460, 1–11.Find this resource:

Yuan, S., & Burrell, B. D. (2013a). Endocannabinoid-dependent long-term depression in a nociceptive synapse requires coordinated presynaptic and postsynaptic transcription and translation. Journal of Neuroscience, 33, 4349–4358.Find this resource:

Yuan, S., & Burrell, B. D. (2013b). Nonnociceptive afferent activity depresses nocifensive behavior and nociceptive synapses via an endocannabinoid-dependent mechanism. Journal of Neurophysiology, 110, 2607–2616.Find this resource:

Zaccardi, M. L., Traina, G., Cataldo, E., & Brunelli, M. (2004). Sensitization and dishabituation of swim induction in the leech Hirudo medicinalis: Role of serotonin and cyclic AMP. Behavioural Brain Research, 153, 317–326.Find this resource: