Comparative Study of Visuo-Vestibular Conditioning in Lymnaea stagnalis
http://www.100md.com
《生物学报告》
Department of Biological Science and Technology, School of High-Technology for Human Welfare, Tokai University, Nishino 317, Numazu 410-0321, Shizuoka, Japan
Abstract
In this review, we compare the current understanding of visuo-vestibular conditioning in Hermissenda crassicornis and Lymnaea stagnalis on the basis of behavioral, electrophysiologic, and morphologic studies. Paired presentation of a photic conditioned stimulus (CS) and an orbital rotation unconditioned stimulus (US) results in conditioned escape behavior in both species. In Hermissenda, changes in excitability of type B photoreceptors and morphologic modifications at the axon terminals follow conditioning. Caudal hair cells, which detect mechanical turbulence, have reciprocal inhibition with type B photoreceptors. In Lymnaea, the interaction between photoreceptors and hair cells is dependent on statocyst location. Furthermore, the organization of the Lymnaea eye is complex, with more than 100 photoreceptors distributed in a uniquely folded retina. Although the optimal conditions to produce long-term memory (memory persistent for >1 week) are almost identical in Hermissenda and Lymnaea, physiologic and morphologic differences suggest that the neuronal mechanisms underlying learning and memory are distinct.
Introduction
Gastropod molluscs are established animal models in the study of the neuronal mechanisms of learning and memory. Examples of their use include Aplysia californica as a model of sensitization of the gill-withdrawal reflex (Kandel, 1976), Hermissenda crassicornis as a model of classical conditioning (Alkon, 1987), and Lymnaea stagnalis as a model of classical conditioning and operant conditioning (Lukowiak et al., 1996; Benjamin et al., 2000; Kawai et al., 2004b). The present review focuses on visuo-vestibular conditioning in Lymnaea and Hermissenda, with an emphasis on behavior, neuronal architecture, and electrophysiologic responses.
Conditioning With Paired Light and Rotation
During conditioning, specimens of Hermissenda are presented with a temporal sequence of a light flash as the conditioned stimulus (CS) and mechanical turbulence as the unconditioned stimulus (US). The animal initially responds to light with phototactic behavior (conditioned response: CR) and moves toward the light, but in response to mechanical turbulence it clings to surfaces (unconditioned response; UR). After several paired presentations of the CS and US over a period of days, the phototactic response decreases significantly (Alkon, 1983). Hereafter, I refer to this conditioning paradigm as visuo-vestibular conditioning (VVC). Phototactic behavior in Hermissenda is estimated from measurement of foot length; the animal responds to light with foot elongation in the na?ve state and with foot contraction after acquisition of learning (Lederhendler et al., 1986). Recent findings indicate that even a single presentation of paired light and orbital rotation (produced by a laboratory shaker with a 4-mm shaking motion) effectively induces phototactic suppression lasting for several minutes (Epstein et al., 2003, 2004). Short-term memory (STM), lasting a few minutes in Hermissenda, can be distinguished from consolidated or long-term memory (LTM), which is induced by 3 days of 100 to 150 paired presentations per training day and lasts for at least 1 week (Matzel et al., 1990). Protein synthesis is associated with the formation of LTM, but not STM (Alkon et al., 1998). The mechanisms of classical conditioning and the characteristics of memory in Hermissenda are discussed in more detail in other papers in this issue (see Crow and Tian, Kuzirian et al.).
The onset of light produces a CR in Hermissenda but is a neutral stimulus in Lymnaea. Lymnaea does, however, display a whole-body withdrawal (pulling its body into the shell) in response to a vestibular turbulence, such as mechanical shock or vibration, or to certain visual stimuli, such as light offset or a moving shadow. The withdrawal response is the sole escape behavior available to Lymnaea (Cook, 1970). At the cellular level, the left and right PeD11 neurons, located on both sides of the pedal ganglia, mediate the withdrawal behavior (Ferguson and Benjamin, 1991a,b; Syed and Winlow, 1991). After VVC, Lymnaea responds to the onset of light stimulation with a withdrawal response. This classical conditioning in Lymnaea is one of the few examples of the truly classical conditioning as defined by Pavlov, as there is no observable response to the CS in na?ve animals. Because this behavior is quite robust, the response latency of the withdrawal response can be determined after a flash of light. The reciprocal of the latency is used as the behavioral index and is proportional to the degree of learning (Sakakibara et al., 1998b).
For Hermissenda, VCC paradigms consist of pairing a 6-s flash of light (CS) and a 4-s orbital rotation (US) with a 2-s delay, and terminating the stimuli simultaneously. For Lymnaea, the CS is presented for 3 s and the US is presented for 2 s with a 1-s delay, and the stimuli simultaneously are terminated. In control experiments, light and rotation are presented randomly without overlap, or animals are maintained in the training apparatus without presentation of light or rotation. The interstimulus interval and CS-US timing have been studied in depth for Hermissenda and Lymnaea (Grover and Farley, 1987; Lederhendler and Alkon, 1989; Matzel et al., 1990; Ono et al., 2002). The optimal VVC conditions are identical for the two species: the CS precedes the US by 1 or 2 s, and the CS and US terminate simultaneously (Lederhendler and Alkon, 1989; Matzel et al., 1990; Ono et al., 2002). The minimum number of paired presentations necessary for LTM formation is greater than 100 trials per day for 3 days in Hermissenda (Matzel et al., 1990), and more than 10 times per day for 3 days in Lymnaea (Ono et al., 2002). The differences in number of paired presentations necessary to form LTM might not be significant because recent studies in Hermissenda revealed that 9 is the minimum number of paired presentations needed to induce LTM (Epstein et al., 2004; Alkon et al., 2005).
The total number of paired presentations is not as important in LTM formation as is the distribution of the number of trials across days. Only animals that receive 30 trials per day for 3 days learn, while neither 90 pairings within a single day nor 45 pairings per day for 2 days results in acquisition of the CR in Lymnaea (Sakakibara et al., 1998b). This, together with the marked stabilization of the learned response observed on day 3 of training, suggests that long-term changes to the associative neural circuitry occur between days 2 and 3 of training. The induction of specific mRNA observed after 3 days of associative conditioning in Hermissenda (Nelson and Alkon, 1990) raises the possibility that modification of biochemical, morphologic, and genetic expression is involved in the formation and stabilization of LTM.
In the field of psychology, it is well known that spaced training—interposing a rest interval between multiple training sessions—produces stronger and longer-lasting memory than the same number of training sessions with no rest interval (Ebbinghaus, 1885; Hintzman, 1974). This phenomenon has also been demonstrated in Drosophila (Tully et al., 1994), Aplysia (Pinsker et al., 1973), honeybee (Bitterman et al., 1983), turtle (Ishida and Papini, 1997), and Lymnaea (Sakakibara et al., 1998b). It has been suggested that spaced training results in the formation of LTM. The cyclic AMP response element binding protein (CREB) is thought to be involved in the transformation from STM to LTM in Drosophila (Yin et al., 1994), Aplysia (Dash et al., 1990), Lymnaea (Hatakeyama et al., 2004a, 2006), honeybee (Eisenhardt et al., 2003), and mice (Bourtchuladze et al., 1994). In Hermissenda, the 22-kDa protein calexcitin is synthesized following VVC (Nelson et al., 1996). Phosphorylation of calexcitin by protein kinase C is critical for the transformation of STM to LTM, a process that is mimicked by the selective protein kinase C activator bryostatin (Alkon et al., 2005). Calexcitin is also found in Lymnaea neural circuits of the withdrawal response (Hatakeyama et al., 2004b), suggesting that it is involved in classical conditioning of withdrawal behavior. Thus, Lymnaea might have two mechanisms for conversion of STM to LTM, one involving CREB and one utilizing calexcitin.
The ability to convert STM to LTM changes during development. We previously demonstrated that animals with a shell length longer than 18 mm (sexually mature) acquire and retain associative memory, while immature animals (shell length < 15 mm) acquire but do not retain the memory to the following day (Ono et al., 2002). I speculate that this same developmental relationship holds also for Hermissenda because, though we do not know the difference in learning ability between mature and immature Hermissenda, learning ability depends on the maturity of the neuronal circuit involved in memory formation. During memory formation, the neural circuitry controlling the withdrawal response is thought to be modified at synapses between sensory neurons and motoneurons. The withdrawal response is mediated by the columellar and dorsal longitudinal muscles, which are innervated by the motoneuron network in the cerebral and pedal ganglia (Ferguson and Benjamin, 1991a,b). The motoneuron network involved in the CR appears to be developed in immature animals, suggesting that it is the neural circuitry necessary for memory retention that is not fully developed. All of the neurons forming the motoneuron network are electrically coupled, and the coupling ratio changes with developmental age, decreasing from 60% at day 90 of postnatal development (immature) to 30% at day 400 (adult) (Wildering et al., 1991). This decrease in electrical coupling might be necessary for the stabilization of associative memory and might reflect the corresponding morphologic changes observed during in vivo and in vitro conditioning in Hermissenda (Alkon et al., 1990; Kawai et al., 2002), though we do not know whether this decrease in electrical coupling induces the CR. The underlying neural circuitry of the clinging behavior in Hermissenda is distinct from that of Lymnaea, and the ontogenic modifications in the motoneuron network during development of Lymnaea are not known.
Neuronal Loci of Visual and Vestibular Information
In molluscs, visual and vestibular information is conveyed by ocular photoreceptors located in the eye, and vestibular information is transmitted by hair cells of the statocyst. In Hermissenda, these receptors are found within 50 μm of each other and intertwine with the optic ganglion neurons. In Lymnaea, a pair of ocular eyes is located at the head surface beneath the tentacles, and the statocysts are located on the right and left pedal ganglia. Though invertebrates have dermal photoreceptors located on the skin around their body surface (Zylstra, 1972; Stoll, 1973; Chono et al., 2002), dermal photoreceptors are not involved in VVC in Lymnaea (Ono et al., 2002; Tsubata et al., 2003). Electrophysiologic studies demonstrate major differences in the neuronal processing of visuo-vestibular information in Hermissenda (Tabata and Alkon, 1982) and Lymnaea (Sakakibara et al., 2005a).
Ocular photoreceptors
Five photoreceptors, subdivided into types A and B, are located around the eye in the circumesophageal nervous system in the body cavity just beneath a tentacle in Hermissenda (Stensaas et al., 1969). Two type A and three type B photoreceptors are distinguished by their location and light response (Fig. 1). Type A photoreceptors are less sensitive to light than type B photoreceptors and are characterized by a large spike amplitude (Fig. 1) (Alkon and Fuortes, 1972).
The light-induced response in Hermissenda involves coordinated control of a number of foot muscle groups. In na?ve animals, the foot undergoes a lengthening in response to light onset and an accompanying orienting movement toward the light source. After VVC, light onset elicits foot contraction. This CR might involve synaptic connections between type B photoreceptors and another group of interneurons in the cerebropleural ganglia. Some neurons receive direct excitatory input from both type B photoreceptors and statocyst hair cells. These cerebropleural interneurons might be an important convergent site where learning-induced changes in visual input result in the transfer of the unconditioned vestibular-elicited foot contraction to the light-elicited CR (Goh and Alkon, 1984).
Type B photoreceptors are believed to be the location of the trace of the associative memory. The conclusion that Type B photoreceptors have an important role in forming an association between visual and vestibular information is based on physiologic measures after acquisition of learning, such as cumulative depolarization following the offset of a light flash (long-lasting depolarization) (West et al., 1982), and persistent inactivation of a K+ current (Alkon et al., 1985). These indices are proportional to the degree of phototactic behavior displayed by the animal. After acquisition of learning, type B photoreceptors also undergo morphologic modification, such as a reduction in the volume of terminal arborization (Alkon et al., 1990). The same morphologic alterations are observed following conditioned learning in vitro. This modification parallels the physiologic indices of type B photoreceptors such as increased input resistance and longer long-lasting depolarization (Kawai et al., 2002, 2003, 2004a). In Hermissenda, the arrangement of ocular photoreceptors and statocyst hair cells provides a useful in vitro conditioning model. Because the Hermissenda eye is relatively simple, containing only five photoreceptors, the phototransduction mechanism in type B photoreceptors is well characterized. Inositol trisphosphate is an important second messenger for transduction of visual information (Sakakibara et al., 1994, 1998a).
Ocular photoreceptors are classified into two types, type I and type II, on the basis of ultrastructural observations (Bobkova, 1998). The physiologic characterization, arrangement, and terminal projections of the photoreceptors in Lymnaea, however, are not fully understood. Recent intracellular recordings using electrodes filled with a tracer dye reveal that there are at least two types of photoreceptors, type A and type T, which have distinct responses to light and terminal arborizations (Sakakibara et al., 2005a).
The response of type A photoreceptors is characterized by a smooth waveform without spiking activity, reduced light sensitivity, and a maximum action spectrum at 480 to 500 nm (the same wavelength as rhodopsin). Type A photoreceptors have very thin axons terminating at the cerebral ganglion, and their cell bodies are located all around the eye (Fig. 4). Type T photoreceptors are characterized by a transient photoresponse overlapping with small-amplitude spiking activity (10 mV) and an intensity-dependent response latency. The action spectrum is broader than for type A cells, peaking between 460 and 540 nm. The photoresponse from type T cells is characterized by a transient "shoulder" followed by a longer latency "hump." This transient shoulder is reversibly reduced in Ca2+-free saline. This two-component photoresponse is consistent with extracellular recordings of optic nerve (Sharko and Osipov, 1981). Type T photoreceptors are characterized by a flat perikaryon (Fig. 4). A single microinjection of tracer dye (Lucifer yellow) into a type T photoreceptor sometimes stains adjacent cell bodies, with two or more axons observed running in parallel to the cerebral ganglion (Fig. 4). The existence of dye coupling with a small-molecular-weight tracer dye is consistent with the existence of gap junctions (Mills and Massey, 1995; Xin and Bloomfield, 1997; Ladewig et al., 1998). The physiologic function of gap junctions between type T photoreceptors is unknown. Figure 5 summarizes the distribution of the two types of photoreceptors. Type A photoreceptors are distributed around the retina. There is no relationship between the distribution of the type A or type T photoreceptors and retinal folding.
Statocyst hair cells
Although 13 statocyst hair cells are located on both sides of the pedal ganglion in Hermissenda and Lymnaea, their interactions with the photoreceptors are quite different in the two species. In Hermissenda, impulses of the caudal hair cells (located on the tail half of the cyst on the dorsoventral equator) cause a cessation of excitatory postsynaptic potentials (EPSPs) received by ipsilateral type B cells. This effect might result from the inhibition of a single optic ganglion cell (the presynaptic source of EPSPs) and inhibitory postsynaptic potentials (IPSPs) produced by the caudal hair cells. Another type of hair cell, the In hair cell adjacent to the caudal hair cells, causes synaptic inhibition of ipsilateral type B photoreceptors as well as cessation of the EPSP. The In hair cell also causes synaptic inhibition of at least one of the two ipsilateral type A photoreceptors. Excitation of the type B photoreceptors causes synaptic inhibition of the cephalic hair cells and cessation of large spontaneous IPSPs received by the caudal hair cells (Alkon et al., 1978; Tabata and Alkon, 1982).
Synaptic interactions in Lymnaea hair cells, unlike those in Hermissenda, are not well understood. Lymnaea changes its geotactic behavior under specific conditions. In O2-rich water, the animal is positively geotactic and moves downward. In O2-deficient water, the animal displays negative geotaxis and moves to the water surface to breathe air. For both positive and negative geotactic behaviors, Lymnaea uses statocyst hair cells. The firing rate of the hair cells depends on the relative geotactic body position and the O2 concentration in the water stream above the mantle area (Janse et al., 1988). Tactile stimulation of the lips, tentacles, mantle edge, and pneumostome effectively produces geotactic responses. The responses are thought to be mediated synaptically because they diminish after perfusion with saline that is high in Mg2+ and low in Ca2+ (Janse et al., 1988). The chemosensory stimulus triggers pneumostome opening by activating sensory cells in the pneumostome-osphradial area, which in turn provide excitatory afferent input to RPeD1 neurons. Activation of RPeD1 initiates central pattern generator activity, which underlies respiratory rhythm. Tactile stimulation of the pneumostome area evokes pneumostome closure and stops aerial respiratory behavior. This response, mediated by the RPeD11 neurons and characterized by a whole-body withdrawal (Inoue et al., 1996a,b), is an important component of the withdrawal response, and therefore tactile stimulation of the pneumostome can be used as an aversive stimulus in operant conditioning experiments (Lukowiak et al., 2003a,b). Furthermore, Janse et al. (1988) demonstrated that at light onset, hair cells depolarize as a result of electrical stimulation of the optic nerve exiting from an eye. Our recent observations of the photoresponse of statocyst hair cells indicate that rostral and caudal hair cells respond to the onset of the photic stimulus, while lateral, medial, and central hair cells do not (Sakakibara et al., 2005a). Moreover, there are regional differences in the photoresponse: caudal hair cells depolarize; rostral cells hyperpolarize. The caudal hair cell response is dependent on light intensity. Increasing light intensity results in a shorter response latency (>100 ms) to the brightest light tested and a larger depolarizing response with action potentials superimposed on the depolarizing potential. In previous work on Helix, Ovchinnikov (1986) demonstrated that hair cells respond to light onset with membrane depolarization and an increase in the superimposed firing frequency of impulse activity, with a latency ranging from 0.3 to 2 s depending on the light intensity (Ovchinnikov, 1986). The photo-induced response recorded in the statocyst hair cells is likely a result of synaptic input from ocular photoreceptors via monosynaptic pathways, because the photoresponse is abolished after eye enucleation or perfusion of Ca2+-free saline but remains following perfusion of high-Ca2+–high-Mg2+ saline. On the basis of our morphologic observations, we suggest that the monosynaptic interaction between statocyst hair cells and photoreceptors occurs at the terminal branches in the central neuropil of the cerebral ganglion (Sakakibara et al., 2005a).
Our recent study reveals the kinetics and pharmacology of the voltage-, time-, and Ca2+-dependent outward K+ currents in both statocyst hair cells and RPeD1 in Lymnaea (Sakakibara et al., 2005b). We modeled type B photoreceptors with the relevant ionic currents using equations of the Hodgkin-Huxley type (Sakakibara et al., 1993), which have been used in neuro-informatic studies of learning mechanisms in Hermissenda (Blackwell, 2000, 2002, 2004). A similar description of the ionic currents in statocyst hair cells will enhance our understanding of the cellular basis of associative learning in Lymnaea. The family of currents in Lymnaea hair cells differs qualitatively and quantitatively from the family of currents in Hermissenda hair cells (Yamoah, 1997). Voltage-dependent Na+ currents are apparent in Lymnaea hair cells, but they are absent in Hermissenda. In both Lymnaea and Hermissenda hair cells, there are three voltage-dependent K+ currents, IA, IKV, and ICa-K, but there are quantitative differences between the values of each of these currents for each species.
The intrinsic membrane properties of hair cells are hypothesized to change as a result of classical conditioning. Because hair cells receive input from type T photoreceptors of the eye (Sakakibara et al., 2005a), Lymnaea learns to withdraw in response to photic stimulation, resulting in VVC. Thus, we expect that following the formation of memory, hair cells will be more excitable to photic stimulation. A possible means of increasing the excitability of the hair cells is to decrease the contribution made by the K current, IA.
In Hermissenda, type A and B photoreceptors differ not only in their photosensitivity but also in their roles mediating phototactic behavior. Type B photoreceptors are inhibitory, whereas type A cells are excitatory (Akaike and Alkon, 1980). In Hermissenda, type B cells are more excitable after conditioning, thus decreasing animal movement toward the light (West et al., 1982). The increased excitability is thought to result from synaptic interactions between visual and vestibular sensory neurons.
In Lymnaea, the behavioral consequence of pairing a photic stimulus with a rotational stimulus is different from that in Hermissenda. Visuo-vestibular conditioning in Lymnaea increases the probability that a photic stimulus will cause withdrawal. In Hermissenda, VCC results in clinging behavior. Thus, although the sensory modalities used in the Hermissenda and Lymnaea studies are the same, the conditioned behaviors are different. Moreover, because the synaptic interactions between these two sensory modalities are not identical, we expect that the neuronal mechanisms underlying learning and memory are also distinct.
Acknowledgments
Some parts of the morphologic study on photoreceptors were performed in collaboration with Dr. N. Katagiri of the Tokyo Women’s Medical University. I thank Ms. C. Toyoshima for assistance with the experiment.
Footnotes
Received 28 October 2005; accepted 6 March 2006.
Literature Cited
Akaike, T., and D. L. Alkon. 1980. Sensory convergence on central visual neurons in Hermissenda. J. Neurophysiol. 44:501–513.
Alkon, D. L. 1983. Learning in a marine snail. Sci. Am.249:70–84.
Alkon, D. L. 1987. Memory Traces in the Brain. Cambridge University Press, New York.
Alkon, D. L., and M. G. Fuortes. 1972. Responses of photoreceptors in Hermissenda. J. Gen. Physiol. 60:631–639.
Alkon, D. L., T. Akaike, and J. Harrigan. 1978. Interaction of chemosensory, visual, and statocyst pathways in Hermissenda crassicornis. J. Gen. Physiol. 71:177–194.
Alkon, D. L., M. Sakakibara, R. Forman, J. Harrigan, I. Lederhendler, and J. Farley. 1985. Reduction of two voltage-dependent K+ currents mediates retention of a learned association. Behav. Neural. Biol.44:278–300.
Alkon, D. L., H. Ikeno, J. Dworkin, D. L. McPhie, J. L. Olds, I. Lederhendler, L. Matzel, B. G. Schreurs, A. Kuzirian, and C. Collin. 1990. Contraction of neuronal branching volume: an anatomic correlate of Pavlovian conditioning. Proc. Natl. Acad. Sci. USA87:1611–1614.
Alkon, D. L., H. Epstein, A. Kuzirian, M. C. Bennett, and T. J. Nelson. 2005. Protein synthesis required for long-term memory is induced by PKC activation on days before associative learning. Proc. Natl. Acad. Sci. USA102:16432–16437.
Andrew, R. J., and H. Savage. 2000. Appetitive learning using visual conditioned stimuli in the pond snail, Lymnaea. Neurobiol. Learn. Mem. 73:258–273.
Benjamin, P. R., K. Staras, and G. Kemenes. 2000. A systems approach to the cellular analysis of associative learning in the pond snail Lymnaea. Learn. Mem. 7:124–131.
Bitterman, M. E., R. Menzel, A. Fietz, and S. Schafer. 1983. Classical conditioning of proboscis extension in honeybees (Apis mellifera). J. Comp. Psychol.97:107–119.
Blackwell, K. T. 2000. Evidence for a distinct light-induced calcium-dependent potassium current in Hermissenda crassicornis. J. Comput. Neurosci. 9:149–170.
Blackwell, K. T. 2002. Calcium waves and closure of potassium channels in response to GABA stimulation in Hermissenda type B photoreceptors. J. Neurophysiol.87:776–792.
Blackwell, K. T. 2004. Paired turbulence and light do not produce a supralinear calcium increase in Hermissenda. J. Comput. Neurosci. 17:81–99.
Bobkova, M. V. 1998. Structural and functional organization of the optic system’s peripheral area in the pond snail Lymnaea stagnalis. Zh. Evol. Biokhim. Fiziol. 34:716–731.
Bourtchuladze, R., B. Frenguelli, J. Blendy, D. Cioffi, G. Schutz, and A. J. Silva. 1994. Deficient long-term memory in mice with a targeted mutation of the cAMP-responsive element-binding protein. Cell79:59–68.
Chono, K., Y. Fujito, and E. Ito. 2002. Non-ocular dermal photoreception in the pond snail Lymnaea stagnalis. Brain Res. 951:107–112.
Cook, A. 1970. Habituation in a freshwater snail (Lymnaea stagnalis). Anim. Behav.178:463–474.
Dash, P. K., B. Hochner, and E. R. Kandel. 1990. Injection of the cAMP-responsive element into the nucleus of Aplysia sensory neurons blocks long-term facilitation. Nature345:718–721.
Ebbinghaus, H. 1885. über das Ged?chtnis. Dover, New York.
Eisenhardt, D., A. Friedrich, N. Stollhoff, U. Muller, H. Kress, and R. Menzel. 2003. The AmCREB gene is an ortholog of the mammalian CREB/CREM family of transcription factors and encodes several splice variants in the honeybee brain. Insect Mol. Biol.12:373–383.
Epstein, H. T., F. M. Child, A. M. Kuzirian, and D. L. Alkon. 2003. Time windows for effects of protein synthesis inhibitors on Pavlovian conditioning in Hermissenda: behavioral aspects. Neurobiol. Learn. Mem.79:127–131.
Epstein, H. T., A. M. Kuzirian, F. M. Child, and D. L. Alkon. 2004. Two different biological configurations for long-term memory. Neurobiol. Learn. Mem.81:12–18.
Ferguson, G. P., and P. R. Benjamin. 1991a. The whole-body withdrawal response of Lymnaea stagnalis. I. Identification of central motoneurones and muscles. J. Exp. Biol.158:63–95.
Ferguson, G. P., and P. R. Benjamin. 1991b. The whole-body withdrawal response of Lymnaea stagnalis. II. Activation of central motoneurones and muscles by sensory input. J. Exp. Biol.158:97–116.
Goh, Y., and D. L. Alkon. 1984. Sensory, interneuronal, and motor interactions within Hermissenda visual pathway. J. Neurophysiol.52:156–169.
Grover, L. M., and J. Farley. 1987. Temporal order sensitivity of associative neural and behavioral changes in Hermissenda. Behav. Neurosci. 101:658–675.
Hatakeyama, D., Y. Fujito, M. Sakakibara, and E. Ito. 2004a. Expression and distribution of transcription factor CCAAT/enhancer-binding protein in the central nervous system of Lymnaea stagnalis. Cell Tissue Res. 318:631–641.
Hatakeyama, D., S. Inamura, E. Ito, M. Sakakibara, T. Nelson, and D. L. Alkon. 2004b. Calexcitin-like immunoreactivity in the pond snail Lymnaea stagnalis. Neurosci. Res. Commun. 35:32–40.
Hatakeyama, D., H. Sadamoto, T. Watanabe, A. Wagatsuma, S. Kobayashi, Y. Fujito, M. Yamashita, M. Sakakibara, G. Kemenes, and E. Ito. 2006. Requirement of new protein synthesis of a transcription factor for memory consolidation: paradoxical changes in mRNA and protein levels of C/EBP. J. Mol. Biol.356:569–577.
Hintzman, D. L. 1974. Theoretical implications of the spacing effect. Pp.77–99 in Theories of Cognitive Psychology, R. L. Solso, ed. Erlbaum, Potomac, MD.
Inoue, T., M. Takasaki, K. Lukowiak, and N. Syed. 1996a. Inhibition of the respiratory pattern-generating neurons by an identified whole-body withdrawal interneuron of Lymnaea stagnalis. J. Exp. Biol. 199:1887–1898.
Inoue, T., M. Takasaki, K. Lukowiak, and N. I. Syed. 1996b. Identification of a putative mechanosensory neuron in Lymnaea: characterization of its synaptic and functional connections with the whole-body withdrawal interneuron. J. Neurophys.76:3230–3238.
Ishida, M., and M. R. Papini. 1997. Massed-trial overtraining effects on extinction and reversal performance in turtle (Geoclemys reevesii). Q. J. Exp. Psychol.50B:1–16.
Janse, C., G. J. van der Wilt, M. van der Roest, and A. W. Pieneman. 1988. Modulation of primary sensory neurons and its relevance to behaviour in the pond snail Lymnaea stagnalis. Symp. Biol Hung. 36:559–568.
Kandel, E. R. 1976. Cellular Basis of Behavior. Freeman, San Francisco.
Kawai, R., T. Horikoshi, T. Yasuoka, and M. Sakakibara. 2002. In vitro conditioning induces morphological changes in Hermissenda type B photoreceptor. Neurosci. Res.43:363–372.
Kawai, R., T. Yasuoka, and M. Sakakibara. 2003. Dynamical aspects of in vitro conditioning in Hermissenda type B photoreceptor. Zool. Sci.20:1–6.
Kawai, R., T. Horikoshi, and M. Sakakibara. 2004a. Involvement of the ryanodine receptor in morphologic modification of Hermissenda type B photoreceptors after in vitro conditioning. J. Neurophysiol.91:728–735.
Kawai, R., H. Sunada, T. Horikoshi, and M. Sakakibara. 2004b. Conditioned taste aversion with sucrose and tactile stimuli in the pond snail Lymnaea stagnalis. Neurobiol. Learn. Mem. 82:164–168.
Ladewig, T., W. Hanke, M. Goldermann, A. C. Guimaraes de Almeida,V. M. Fernandes de Lima. 1998. Distribution of gap junctions in the chicken retina. J. Hirnforsch.39:77–86.
Lederhendler, I. I., and D. L. Alkon. 1989. The interstimulus interval and classical conditioning in the marine snail Hermissenda crassicornis. Behav. Brain Res. 35:75–80.
Lukowiak, K., E. Ringseis, G. Spencer, W. Wildering, and N. Syed. 1996. Operant conditioning of aerial respiratory behaviour in Lymnaea stagnalis. J. Exp. Biol. 199:683–691.
Lukowiak, K., S. Sangha, C. McComb, N. Varshney, D. Rosenegger, H. Sadamoto, and A. Scheibenstock. 2003a. Associative learning and memory in Lymnaea stagnalis: How well do they remember? J. Exp. Biol.206:2097–2103.
Lukowiak, K., S. Sangha, A. Scheibenstock, K. Parvez, C. McComb, D. Rosenegger, N. Varshney, and H. Sadamoto. 2003b. A molluscan model system in the search for the engram. J. Physiol. Paris97:69–76.
Matzel, L. D., B. G. Schreurs, I. Lederhandler, and D. L. Alkon. 1990. Acquisition of conditioned associations in Hermissenda: additive effects of contiguity and the forward interstimulus interval. Behav. Neurosci.104:597–606.
Mills, S. L., and S. C. Massey. 1995. Differential properties of two gap junctional pathways made by AII amacrine cells. Nature377:734–737.
Nelson, T. J., and D. L. Alkon. 1990. Specific high molecular weight mRNAs induced by associative learning in Hermissenda. Proc. Natl. Acad. Sci. USA 87:269–273.
Nelson, T. J., S. Cavallaro, C. L. Yi, D. McPhie, B. G. Schreurs, P. A. Gusev, A. Favit, O. Zohar, J. Kim, S. Beushausen, G. Ascoli, J. Olds, R. Neve, and D. L. Alkon. 1996. Calexcitin: a signaling protein that binds calcium and GTP, inhibits potassium channels, and enhances membrane excitability. Proc. Natl. Acad. Sci. USA93:13808–13813.
Ono, M., R. Kawai, T. Horikoshi, T. Yasuoka, and M. Sakakibara. 2002. Associative learning acquisition and retention depends on developmental stage in Lymnaea stagnalis. Neurobiol. Learn. Mem. 78:53–64.
Ovchinnikov, A. V. 1986. Intersensory interactions in the CNS of Helix lucorum. Neirofiziologiya 18:17–26.
Pinsker, H. M., W. A. Hening, T. J. Carew, and E. R. Kandel. 1973. Long-term sensitization of a defensive withdrawal reflex in Aplysia. Science 182:1039–1042.
Sakakibara, M., H. Ikeno, S. Usui, C. Collin, and D. L. Alkon. 1993. Reconstruction of ionic currents in a molluscan photoreceptor. Biophys. J.65:519–527.
Sakakibara, M., D. L. Alkon, T. Kouchi, H. Inoue, and T. Yoshioka. 1994. Induction of photoresponse by the hydrolysis of polyphosphoinositides in the Hermissenda type B photoreceptor. Biochem. Biophys. Res. Commun.202:299–306.
Sakakibara, M., H. Inoue, and T. Yoshioka. 1998a. Evidence for the involvement of inositol trisphosphate but not cyclic nucleotides in visual transduction in Hermissenda eye. J. Biol. Chem.273:20795–20801.
Sakakibara, M., R. Kawai, S. Kobayashi, and T. Horikoshi. 1998b. Associative learning of visual and vestibular stimuli in Lymnaea. Neurobiol. Learn. Mem. 69:1–12.
Sakakibara, M., T. Aritaka, A. Iizuka, H. Suzuki, T. Horikoshi, and K. Lukowiak. 2005a. Electrophysiological responses to light of neurons in the eye and statocyst of Lymnaea stagnalis. J. Neurophysiol. 93:493–507.
Sakakibara, M., F. Okuda, K. Nomura, K. Watanabe, H. Meng, T. Horikoshi, and K. Lukowiak. 2005b. Potassium currents in isolated statocyst neurons and RPeD1 in the pond snail, Lymnaea stagnalis. J. Neurophysiol. 94:3884–3892.
Sharko, N. V., and B. S. Osipov. 1981. Electrical responses of the eye and optic nerve of the pond snail Lymnaea stagnalis. Neirofiziologiya 13:652–654.
Stensaas, L. J., S. S. Stensaas, and O. Trujillo-Cenoz. 1969. Some morphological aspects of the visual system of Hermissenda crassicornis (Mollusca: Nudibranchia). J. Ultrastruct. Res.27:510–532.
Stoll, C. J. 1973. On the role of eyes and non-ocular light receptors in orientational behaviour of Lymnaea stagnalis (L.). Proc. K. Ned. Akad. Wet. C76:203–214.
Stoll, C. J., and A. Bijlsma. 1973. Optic nerve responses in Lymnaea stagnalis to photic stimulation of the eye. Proc. K. Ned. Akad. Wet. C76:406–413.
Syed, N. I., and W. Winlow. 1991. Coordination of locomotor and cardiorespiratory networks of Lymnaea stagnalis by a pair of identified interneurones. J. Exp. Biol.158:37–62.
Tabata, M., and D. L. Alkon. 1982. Positive synaptic feedback in visual system of nudibranch mollusk Hermissenda crassicornis. J. Neurophysiol. 48:174–191.
Tsubata, N., A. Iizuka, T. Horikoshi, and M. Sakakibara. 2003. Photoresponse from the statocyst hair cell in Lymnaea stagnalis. Neurosci. Lett. 337:46–50.
Tully, T., T. Preat, S. C. Boynton,M. Del Vecchio. 1994. Genetic dissection of consolidated memory in Drosophila. Cell 79:35–47.
West, A., E. Barnes, and D. L. Alkon. 1982. Primary changes of voltage responses during retention of associative learning. J. Neurophysiol.48:1243–1255.
Wildering, W. C., M. van der Roest, T. A. de Vlieger, and C. Janse. 1991. Age-related changes in junctional and non-junctional conductances in two electrically coupled peptidergic neurons of the mollusc Lymnaea stagnalis. Brain Res. 547:89–98.
Xin, D., and S. A. Bloomfield. 1997. Tracer coupling pattern of amacrine and ganglion cells in the rabbit retina. J. Comp. Neurol.383:512–528.
Yamoah, E. N. 1997. Potassium currents in presynaptic hair cells of Hermissenda. Biophys. J. 72:193–203.
Yin, J. C., J. S. Wallach, M. Del Vecchio, E. L. Wilder, H. Zhou, W. G. Quinn, and T. Tully. 1994. Induction of a dominant negative CREB transgene specifically blocks long-term memory in Drosophila. Cell 79:49–58.
Zylstra, U. 1972. Histochemistry and ultrastructure of the epidermis and the subepidermal gland cells of the freshwater snails Lymnaea stagnalis and Biomphalaria pfeifferi. Z. Zellforsch. Mikrosk. Anat. 130:93–134.(Manabu Sakakibara)
Abstract
In this review, we compare the current understanding of visuo-vestibular conditioning in Hermissenda crassicornis and Lymnaea stagnalis on the basis of behavioral, electrophysiologic, and morphologic studies. Paired presentation of a photic conditioned stimulus (CS) and an orbital rotation unconditioned stimulus (US) results in conditioned escape behavior in both species. In Hermissenda, changes in excitability of type B photoreceptors and morphologic modifications at the axon terminals follow conditioning. Caudal hair cells, which detect mechanical turbulence, have reciprocal inhibition with type B photoreceptors. In Lymnaea, the interaction between photoreceptors and hair cells is dependent on statocyst location. Furthermore, the organization of the Lymnaea eye is complex, with more than 100 photoreceptors distributed in a uniquely folded retina. Although the optimal conditions to produce long-term memory (memory persistent for >1 week) are almost identical in Hermissenda and Lymnaea, physiologic and morphologic differences suggest that the neuronal mechanisms underlying learning and memory are distinct.
Introduction
Gastropod molluscs are established animal models in the study of the neuronal mechanisms of learning and memory. Examples of their use include Aplysia californica as a model of sensitization of the gill-withdrawal reflex (Kandel, 1976), Hermissenda crassicornis as a model of classical conditioning (Alkon, 1987), and Lymnaea stagnalis as a model of classical conditioning and operant conditioning (Lukowiak et al., 1996; Benjamin et al., 2000; Kawai et al., 2004b). The present review focuses on visuo-vestibular conditioning in Lymnaea and Hermissenda, with an emphasis on behavior, neuronal architecture, and electrophysiologic responses.
Conditioning With Paired Light and Rotation
During conditioning, specimens of Hermissenda are presented with a temporal sequence of a light flash as the conditioned stimulus (CS) and mechanical turbulence as the unconditioned stimulus (US). The animal initially responds to light with phototactic behavior (conditioned response: CR) and moves toward the light, but in response to mechanical turbulence it clings to surfaces (unconditioned response; UR). After several paired presentations of the CS and US over a period of days, the phototactic response decreases significantly (Alkon, 1983). Hereafter, I refer to this conditioning paradigm as visuo-vestibular conditioning (VVC). Phototactic behavior in Hermissenda is estimated from measurement of foot length; the animal responds to light with foot elongation in the na?ve state and with foot contraction after acquisition of learning (Lederhendler et al., 1986). Recent findings indicate that even a single presentation of paired light and orbital rotation (produced by a laboratory shaker with a 4-mm shaking motion) effectively induces phototactic suppression lasting for several minutes (Epstein et al., 2003, 2004). Short-term memory (STM), lasting a few minutes in Hermissenda, can be distinguished from consolidated or long-term memory (LTM), which is induced by 3 days of 100 to 150 paired presentations per training day and lasts for at least 1 week (Matzel et al., 1990). Protein synthesis is associated with the formation of LTM, but not STM (Alkon et al., 1998). The mechanisms of classical conditioning and the characteristics of memory in Hermissenda are discussed in more detail in other papers in this issue (see Crow and Tian, Kuzirian et al.).
The onset of light produces a CR in Hermissenda but is a neutral stimulus in Lymnaea. Lymnaea does, however, display a whole-body withdrawal (pulling its body into the shell) in response to a vestibular turbulence, such as mechanical shock or vibration, or to certain visual stimuli, such as light offset or a moving shadow. The withdrawal response is the sole escape behavior available to Lymnaea (Cook, 1970). At the cellular level, the left and right PeD11 neurons, located on both sides of the pedal ganglia, mediate the withdrawal behavior (Ferguson and Benjamin, 1991a,b; Syed and Winlow, 1991). After VVC, Lymnaea responds to the onset of light stimulation with a withdrawal response. This classical conditioning in Lymnaea is one of the few examples of the truly classical conditioning as defined by Pavlov, as there is no observable response to the CS in na?ve animals. Because this behavior is quite robust, the response latency of the withdrawal response can be determined after a flash of light. The reciprocal of the latency is used as the behavioral index and is proportional to the degree of learning (Sakakibara et al., 1998b).
For Hermissenda, VCC paradigms consist of pairing a 6-s flash of light (CS) and a 4-s orbital rotation (US) with a 2-s delay, and terminating the stimuli simultaneously. For Lymnaea, the CS is presented for 3 s and the US is presented for 2 s with a 1-s delay, and the stimuli simultaneously are terminated. In control experiments, light and rotation are presented randomly without overlap, or animals are maintained in the training apparatus without presentation of light or rotation. The interstimulus interval and CS-US timing have been studied in depth for Hermissenda and Lymnaea (Grover and Farley, 1987; Lederhendler and Alkon, 1989; Matzel et al., 1990; Ono et al., 2002). The optimal VVC conditions are identical for the two species: the CS precedes the US by 1 or 2 s, and the CS and US terminate simultaneously (Lederhendler and Alkon, 1989; Matzel et al., 1990; Ono et al., 2002). The minimum number of paired presentations necessary for LTM formation is greater than 100 trials per day for 3 days in Hermissenda (Matzel et al., 1990), and more than 10 times per day for 3 days in Lymnaea (Ono et al., 2002). The differences in number of paired presentations necessary to form LTM might not be significant because recent studies in Hermissenda revealed that 9 is the minimum number of paired presentations needed to induce LTM (Epstein et al., 2004; Alkon et al., 2005).
The total number of paired presentations is not as important in LTM formation as is the distribution of the number of trials across days. Only animals that receive 30 trials per day for 3 days learn, while neither 90 pairings within a single day nor 45 pairings per day for 2 days results in acquisition of the CR in Lymnaea (Sakakibara et al., 1998b). This, together with the marked stabilization of the learned response observed on day 3 of training, suggests that long-term changes to the associative neural circuitry occur between days 2 and 3 of training. The induction of specific mRNA observed after 3 days of associative conditioning in Hermissenda (Nelson and Alkon, 1990) raises the possibility that modification of biochemical, morphologic, and genetic expression is involved in the formation and stabilization of LTM.
In the field of psychology, it is well known that spaced training—interposing a rest interval between multiple training sessions—produces stronger and longer-lasting memory than the same number of training sessions with no rest interval (Ebbinghaus, 1885; Hintzman, 1974). This phenomenon has also been demonstrated in Drosophila (Tully et al., 1994), Aplysia (Pinsker et al., 1973), honeybee (Bitterman et al., 1983), turtle (Ishida and Papini, 1997), and Lymnaea (Sakakibara et al., 1998b). It has been suggested that spaced training results in the formation of LTM. The cyclic AMP response element binding protein (CREB) is thought to be involved in the transformation from STM to LTM in Drosophila (Yin et al., 1994), Aplysia (Dash et al., 1990), Lymnaea (Hatakeyama et al., 2004a, 2006), honeybee (Eisenhardt et al., 2003), and mice (Bourtchuladze et al., 1994). In Hermissenda, the 22-kDa protein calexcitin is synthesized following VVC (Nelson et al., 1996). Phosphorylation of calexcitin by protein kinase C is critical for the transformation of STM to LTM, a process that is mimicked by the selective protein kinase C activator bryostatin (Alkon et al., 2005). Calexcitin is also found in Lymnaea neural circuits of the withdrawal response (Hatakeyama et al., 2004b), suggesting that it is involved in classical conditioning of withdrawal behavior. Thus, Lymnaea might have two mechanisms for conversion of STM to LTM, one involving CREB and one utilizing calexcitin.
The ability to convert STM to LTM changes during development. We previously demonstrated that animals with a shell length longer than 18 mm (sexually mature) acquire and retain associative memory, while immature animals (shell length < 15 mm) acquire but do not retain the memory to the following day (Ono et al., 2002). I speculate that this same developmental relationship holds also for Hermissenda because, though we do not know the difference in learning ability between mature and immature Hermissenda, learning ability depends on the maturity of the neuronal circuit involved in memory formation. During memory formation, the neural circuitry controlling the withdrawal response is thought to be modified at synapses between sensory neurons and motoneurons. The withdrawal response is mediated by the columellar and dorsal longitudinal muscles, which are innervated by the motoneuron network in the cerebral and pedal ganglia (Ferguson and Benjamin, 1991a,b). The motoneuron network involved in the CR appears to be developed in immature animals, suggesting that it is the neural circuitry necessary for memory retention that is not fully developed. All of the neurons forming the motoneuron network are electrically coupled, and the coupling ratio changes with developmental age, decreasing from 60% at day 90 of postnatal development (immature) to 30% at day 400 (adult) (Wildering et al., 1991). This decrease in electrical coupling might be necessary for the stabilization of associative memory and might reflect the corresponding morphologic changes observed during in vivo and in vitro conditioning in Hermissenda (Alkon et al., 1990; Kawai et al., 2002), though we do not know whether this decrease in electrical coupling induces the CR. The underlying neural circuitry of the clinging behavior in Hermissenda is distinct from that of Lymnaea, and the ontogenic modifications in the motoneuron network during development of Lymnaea are not known.
Neuronal Loci of Visual and Vestibular Information
In molluscs, visual and vestibular information is conveyed by ocular photoreceptors located in the eye, and vestibular information is transmitted by hair cells of the statocyst. In Hermissenda, these receptors are found within 50 μm of each other and intertwine with the optic ganglion neurons. In Lymnaea, a pair of ocular eyes is located at the head surface beneath the tentacles, and the statocysts are located on the right and left pedal ganglia. Though invertebrates have dermal photoreceptors located on the skin around their body surface (Zylstra, 1972; Stoll, 1973; Chono et al., 2002), dermal photoreceptors are not involved in VVC in Lymnaea (Ono et al., 2002; Tsubata et al., 2003). Electrophysiologic studies demonstrate major differences in the neuronal processing of visuo-vestibular information in Hermissenda (Tabata and Alkon, 1982) and Lymnaea (Sakakibara et al., 2005a).
Ocular photoreceptors
Five photoreceptors, subdivided into types A and B, are located around the eye in the circumesophageal nervous system in the body cavity just beneath a tentacle in Hermissenda (Stensaas et al., 1969). Two type A and three type B photoreceptors are distinguished by their location and light response (Fig. 1). Type A photoreceptors are less sensitive to light than type B photoreceptors and are characterized by a large spike amplitude (Fig. 1) (Alkon and Fuortes, 1972).
The light-induced response in Hermissenda involves coordinated control of a number of foot muscle groups. In na?ve animals, the foot undergoes a lengthening in response to light onset and an accompanying orienting movement toward the light source. After VVC, light onset elicits foot contraction. This CR might involve synaptic connections between type B photoreceptors and another group of interneurons in the cerebropleural ganglia. Some neurons receive direct excitatory input from both type B photoreceptors and statocyst hair cells. These cerebropleural interneurons might be an important convergent site where learning-induced changes in visual input result in the transfer of the unconditioned vestibular-elicited foot contraction to the light-elicited CR (Goh and Alkon, 1984).
Type B photoreceptors are believed to be the location of the trace of the associative memory. The conclusion that Type B photoreceptors have an important role in forming an association between visual and vestibular information is based on physiologic measures after acquisition of learning, such as cumulative depolarization following the offset of a light flash (long-lasting depolarization) (West et al., 1982), and persistent inactivation of a K+ current (Alkon et al., 1985). These indices are proportional to the degree of phototactic behavior displayed by the animal. After acquisition of learning, type B photoreceptors also undergo morphologic modification, such as a reduction in the volume of terminal arborization (Alkon et al., 1990). The same morphologic alterations are observed following conditioned learning in vitro. This modification parallels the physiologic indices of type B photoreceptors such as increased input resistance and longer long-lasting depolarization (Kawai et al., 2002, 2003, 2004a). In Hermissenda, the arrangement of ocular photoreceptors and statocyst hair cells provides a useful in vitro conditioning model. Because the Hermissenda eye is relatively simple, containing only five photoreceptors, the phototransduction mechanism in type B photoreceptors is well characterized. Inositol trisphosphate is an important second messenger for transduction of visual information (Sakakibara et al., 1994, 1998a).
Ocular photoreceptors are classified into two types, type I and type II, on the basis of ultrastructural observations (Bobkova, 1998). The physiologic characterization, arrangement, and terminal projections of the photoreceptors in Lymnaea, however, are not fully understood. Recent intracellular recordings using electrodes filled with a tracer dye reveal that there are at least two types of photoreceptors, type A and type T, which have distinct responses to light and terminal arborizations (Sakakibara et al., 2005a).
The response of type A photoreceptors is characterized by a smooth waveform without spiking activity, reduced light sensitivity, and a maximum action spectrum at 480 to 500 nm (the same wavelength as rhodopsin). Type A photoreceptors have very thin axons terminating at the cerebral ganglion, and their cell bodies are located all around the eye (Fig. 4). Type T photoreceptors are characterized by a transient photoresponse overlapping with small-amplitude spiking activity (10 mV) and an intensity-dependent response latency. The action spectrum is broader than for type A cells, peaking between 460 and 540 nm. The photoresponse from type T cells is characterized by a transient "shoulder" followed by a longer latency "hump." This transient shoulder is reversibly reduced in Ca2+-free saline. This two-component photoresponse is consistent with extracellular recordings of optic nerve (Sharko and Osipov, 1981). Type T photoreceptors are characterized by a flat perikaryon (Fig. 4). A single microinjection of tracer dye (Lucifer yellow) into a type T photoreceptor sometimes stains adjacent cell bodies, with two or more axons observed running in parallel to the cerebral ganglion (Fig. 4). The existence of dye coupling with a small-molecular-weight tracer dye is consistent with the existence of gap junctions (Mills and Massey, 1995; Xin and Bloomfield, 1997; Ladewig et al., 1998). The physiologic function of gap junctions between type T photoreceptors is unknown. Figure 5 summarizes the distribution of the two types of photoreceptors. Type A photoreceptors are distributed around the retina. There is no relationship between the distribution of the type A or type T photoreceptors and retinal folding.
Statocyst hair cells
Although 13 statocyst hair cells are located on both sides of the pedal ganglion in Hermissenda and Lymnaea, their interactions with the photoreceptors are quite different in the two species. In Hermissenda, impulses of the caudal hair cells (located on the tail half of the cyst on the dorsoventral equator) cause a cessation of excitatory postsynaptic potentials (EPSPs) received by ipsilateral type B cells. This effect might result from the inhibition of a single optic ganglion cell (the presynaptic source of EPSPs) and inhibitory postsynaptic potentials (IPSPs) produced by the caudal hair cells. Another type of hair cell, the In hair cell adjacent to the caudal hair cells, causes synaptic inhibition of ipsilateral type B photoreceptors as well as cessation of the EPSP. The In hair cell also causes synaptic inhibition of at least one of the two ipsilateral type A photoreceptors. Excitation of the type B photoreceptors causes synaptic inhibition of the cephalic hair cells and cessation of large spontaneous IPSPs received by the caudal hair cells (Alkon et al., 1978; Tabata and Alkon, 1982).
Synaptic interactions in Lymnaea hair cells, unlike those in Hermissenda, are not well understood. Lymnaea changes its geotactic behavior under specific conditions. In O2-rich water, the animal is positively geotactic and moves downward. In O2-deficient water, the animal displays negative geotaxis and moves to the water surface to breathe air. For both positive and negative geotactic behaviors, Lymnaea uses statocyst hair cells. The firing rate of the hair cells depends on the relative geotactic body position and the O2 concentration in the water stream above the mantle area (Janse et al., 1988). Tactile stimulation of the lips, tentacles, mantle edge, and pneumostome effectively produces geotactic responses. The responses are thought to be mediated synaptically because they diminish after perfusion with saline that is high in Mg2+ and low in Ca2+ (Janse et al., 1988). The chemosensory stimulus triggers pneumostome opening by activating sensory cells in the pneumostome-osphradial area, which in turn provide excitatory afferent input to RPeD1 neurons. Activation of RPeD1 initiates central pattern generator activity, which underlies respiratory rhythm. Tactile stimulation of the pneumostome area evokes pneumostome closure and stops aerial respiratory behavior. This response, mediated by the RPeD11 neurons and characterized by a whole-body withdrawal (Inoue et al., 1996a,b), is an important component of the withdrawal response, and therefore tactile stimulation of the pneumostome can be used as an aversive stimulus in operant conditioning experiments (Lukowiak et al., 2003a,b). Furthermore, Janse et al. (1988) demonstrated that at light onset, hair cells depolarize as a result of electrical stimulation of the optic nerve exiting from an eye. Our recent observations of the photoresponse of statocyst hair cells indicate that rostral and caudal hair cells respond to the onset of the photic stimulus, while lateral, medial, and central hair cells do not (Sakakibara et al., 2005a). Moreover, there are regional differences in the photoresponse: caudal hair cells depolarize; rostral cells hyperpolarize. The caudal hair cell response is dependent on light intensity. Increasing light intensity results in a shorter response latency (>100 ms) to the brightest light tested and a larger depolarizing response with action potentials superimposed on the depolarizing potential. In previous work on Helix, Ovchinnikov (1986) demonstrated that hair cells respond to light onset with membrane depolarization and an increase in the superimposed firing frequency of impulse activity, with a latency ranging from 0.3 to 2 s depending on the light intensity (Ovchinnikov, 1986). The photo-induced response recorded in the statocyst hair cells is likely a result of synaptic input from ocular photoreceptors via monosynaptic pathways, because the photoresponse is abolished after eye enucleation or perfusion of Ca2+-free saline but remains following perfusion of high-Ca2+–high-Mg2+ saline. On the basis of our morphologic observations, we suggest that the monosynaptic interaction between statocyst hair cells and photoreceptors occurs at the terminal branches in the central neuropil of the cerebral ganglion (Sakakibara et al., 2005a).
Our recent study reveals the kinetics and pharmacology of the voltage-, time-, and Ca2+-dependent outward K+ currents in both statocyst hair cells and RPeD1 in Lymnaea (Sakakibara et al., 2005b). We modeled type B photoreceptors with the relevant ionic currents using equations of the Hodgkin-Huxley type (Sakakibara et al., 1993), which have been used in neuro-informatic studies of learning mechanisms in Hermissenda (Blackwell, 2000, 2002, 2004). A similar description of the ionic currents in statocyst hair cells will enhance our understanding of the cellular basis of associative learning in Lymnaea. The family of currents in Lymnaea hair cells differs qualitatively and quantitatively from the family of currents in Hermissenda hair cells (Yamoah, 1997). Voltage-dependent Na+ currents are apparent in Lymnaea hair cells, but they are absent in Hermissenda. In both Lymnaea and Hermissenda hair cells, there are three voltage-dependent K+ currents, IA, IKV, and ICa-K, but there are quantitative differences between the values of each of these currents for each species.
The intrinsic membrane properties of hair cells are hypothesized to change as a result of classical conditioning. Because hair cells receive input from type T photoreceptors of the eye (Sakakibara et al., 2005a), Lymnaea learns to withdraw in response to photic stimulation, resulting in VVC. Thus, we expect that following the formation of memory, hair cells will be more excitable to photic stimulation. A possible means of increasing the excitability of the hair cells is to decrease the contribution made by the K current, IA.
In Hermissenda, type A and B photoreceptors differ not only in their photosensitivity but also in their roles mediating phototactic behavior. Type B photoreceptors are inhibitory, whereas type A cells are excitatory (Akaike and Alkon, 1980). In Hermissenda, type B cells are more excitable after conditioning, thus decreasing animal movement toward the light (West et al., 1982). The increased excitability is thought to result from synaptic interactions between visual and vestibular sensory neurons.
In Lymnaea, the behavioral consequence of pairing a photic stimulus with a rotational stimulus is different from that in Hermissenda. Visuo-vestibular conditioning in Lymnaea increases the probability that a photic stimulus will cause withdrawal. In Hermissenda, VCC results in clinging behavior. Thus, although the sensory modalities used in the Hermissenda and Lymnaea studies are the same, the conditioned behaviors are different. Moreover, because the synaptic interactions between these two sensory modalities are not identical, we expect that the neuronal mechanisms underlying learning and memory are also distinct.
Acknowledgments
Some parts of the morphologic study on photoreceptors were performed in collaboration with Dr. N. Katagiri of the Tokyo Women’s Medical University. I thank Ms. C. Toyoshima for assistance with the experiment.
Footnotes
Received 28 October 2005; accepted 6 March 2006.
Literature Cited
Akaike, T., and D. L. Alkon. 1980. Sensory convergence on central visual neurons in Hermissenda. J. Neurophysiol. 44:501–513.
Alkon, D. L. 1983. Learning in a marine snail. Sci. Am.249:70–84.
Alkon, D. L. 1987. Memory Traces in the Brain. Cambridge University Press, New York.
Alkon, D. L., and M. G. Fuortes. 1972. Responses of photoreceptors in Hermissenda. J. Gen. Physiol. 60:631–639.
Alkon, D. L., T. Akaike, and J. Harrigan. 1978. Interaction of chemosensory, visual, and statocyst pathways in Hermissenda crassicornis. J. Gen. Physiol. 71:177–194.
Alkon, D. L., M. Sakakibara, R. Forman, J. Harrigan, I. Lederhendler, and J. Farley. 1985. Reduction of two voltage-dependent K+ currents mediates retention of a learned association. Behav. Neural. Biol.44:278–300.
Alkon, D. L., H. Ikeno, J. Dworkin, D. L. McPhie, J. L. Olds, I. Lederhendler, L. Matzel, B. G. Schreurs, A. Kuzirian, and C. Collin. 1990. Contraction of neuronal branching volume: an anatomic correlate of Pavlovian conditioning. Proc. Natl. Acad. Sci. USA87:1611–1614.
Alkon, D. L., H. Epstein, A. Kuzirian, M. C. Bennett, and T. J. Nelson. 2005. Protein synthesis required for long-term memory is induced by PKC activation on days before associative learning. Proc. Natl. Acad. Sci. USA102:16432–16437.
Andrew, R. J., and H. Savage. 2000. Appetitive learning using visual conditioned stimuli in the pond snail, Lymnaea. Neurobiol. Learn. Mem. 73:258–273.
Benjamin, P. R., K. Staras, and G. Kemenes. 2000. A systems approach to the cellular analysis of associative learning in the pond snail Lymnaea. Learn. Mem. 7:124–131.
Bitterman, M. E., R. Menzel, A. Fietz, and S. Schafer. 1983. Classical conditioning of proboscis extension in honeybees (Apis mellifera). J. Comp. Psychol.97:107–119.
Blackwell, K. T. 2000. Evidence for a distinct light-induced calcium-dependent potassium current in Hermissenda crassicornis. J. Comput. Neurosci. 9:149–170.
Blackwell, K. T. 2002. Calcium waves and closure of potassium channels in response to GABA stimulation in Hermissenda type B photoreceptors. J. Neurophysiol.87:776–792.
Blackwell, K. T. 2004. Paired turbulence and light do not produce a supralinear calcium increase in Hermissenda. J. Comput. Neurosci. 17:81–99.
Bobkova, M. V. 1998. Structural and functional organization of the optic system’s peripheral area in the pond snail Lymnaea stagnalis. Zh. Evol. Biokhim. Fiziol. 34:716–731.
Bourtchuladze, R., B. Frenguelli, J. Blendy, D. Cioffi, G. Schutz, and A. J. Silva. 1994. Deficient long-term memory in mice with a targeted mutation of the cAMP-responsive element-binding protein. Cell79:59–68.
Chono, K., Y. Fujito, and E. Ito. 2002. Non-ocular dermal photoreception in the pond snail Lymnaea stagnalis. Brain Res. 951:107–112.
Cook, A. 1970. Habituation in a freshwater snail (Lymnaea stagnalis). Anim. Behav.178:463–474.
Dash, P. K., B. Hochner, and E. R. Kandel. 1990. Injection of the cAMP-responsive element into the nucleus of Aplysia sensory neurons blocks long-term facilitation. Nature345:718–721.
Ebbinghaus, H. 1885. über das Ged?chtnis. Dover, New York.
Eisenhardt, D., A. Friedrich, N. Stollhoff, U. Muller, H. Kress, and R. Menzel. 2003. The AmCREB gene is an ortholog of the mammalian CREB/CREM family of transcription factors and encodes several splice variants in the honeybee brain. Insect Mol. Biol.12:373–383.
Epstein, H. T., F. M. Child, A. M. Kuzirian, and D. L. Alkon. 2003. Time windows for effects of protein synthesis inhibitors on Pavlovian conditioning in Hermissenda: behavioral aspects. Neurobiol. Learn. Mem.79:127–131.
Epstein, H. T., A. M. Kuzirian, F. M. Child, and D. L. Alkon. 2004. Two different biological configurations for long-term memory. Neurobiol. Learn. Mem.81:12–18.
Ferguson, G. P., and P. R. Benjamin. 1991a. The whole-body withdrawal response of Lymnaea stagnalis. I. Identification of central motoneurones and muscles. J. Exp. Biol.158:63–95.
Ferguson, G. P., and P. R. Benjamin. 1991b. The whole-body withdrawal response of Lymnaea stagnalis. II. Activation of central motoneurones and muscles by sensory input. J. Exp. Biol.158:97–116.
Goh, Y., and D. L. Alkon. 1984. Sensory, interneuronal, and motor interactions within Hermissenda visual pathway. J. Neurophysiol.52:156–169.
Grover, L. M., and J. Farley. 1987. Temporal order sensitivity of associative neural and behavioral changes in Hermissenda. Behav. Neurosci. 101:658–675.
Hatakeyama, D., Y. Fujito, M. Sakakibara, and E. Ito. 2004a. Expression and distribution of transcription factor CCAAT/enhancer-binding protein in the central nervous system of Lymnaea stagnalis. Cell Tissue Res. 318:631–641.
Hatakeyama, D., S. Inamura, E. Ito, M. Sakakibara, T. Nelson, and D. L. Alkon. 2004b. Calexcitin-like immunoreactivity in the pond snail Lymnaea stagnalis. Neurosci. Res. Commun. 35:32–40.
Hatakeyama, D., H. Sadamoto, T. Watanabe, A. Wagatsuma, S. Kobayashi, Y. Fujito, M. Yamashita, M. Sakakibara, G. Kemenes, and E. Ito. 2006. Requirement of new protein synthesis of a transcription factor for memory consolidation: paradoxical changes in mRNA and protein levels of C/EBP. J. Mol. Biol.356:569–577.
Hintzman, D. L. 1974. Theoretical implications of the spacing effect. Pp.77–99 in Theories of Cognitive Psychology, R. L. Solso, ed. Erlbaum, Potomac, MD.
Inoue, T., M. Takasaki, K. Lukowiak, and N. Syed. 1996a. Inhibition of the respiratory pattern-generating neurons by an identified whole-body withdrawal interneuron of Lymnaea stagnalis. J. Exp. Biol. 199:1887–1898.
Inoue, T., M. Takasaki, K. Lukowiak, and N. I. Syed. 1996b. Identification of a putative mechanosensory neuron in Lymnaea: characterization of its synaptic and functional connections with the whole-body withdrawal interneuron. J. Neurophys.76:3230–3238.
Ishida, M., and M. R. Papini. 1997. Massed-trial overtraining effects on extinction and reversal performance in turtle (Geoclemys reevesii). Q. J. Exp. Psychol.50B:1–16.
Janse, C., G. J. van der Wilt, M. van der Roest, and A. W. Pieneman. 1988. Modulation of primary sensory neurons and its relevance to behaviour in the pond snail Lymnaea stagnalis. Symp. Biol Hung. 36:559–568.
Kandel, E. R. 1976. Cellular Basis of Behavior. Freeman, San Francisco.
Kawai, R., T. Horikoshi, T. Yasuoka, and M. Sakakibara. 2002. In vitro conditioning induces morphological changes in Hermissenda type B photoreceptor. Neurosci. Res.43:363–372.
Kawai, R., T. Yasuoka, and M. Sakakibara. 2003. Dynamical aspects of in vitro conditioning in Hermissenda type B photoreceptor. Zool. Sci.20:1–6.
Kawai, R., T. Horikoshi, and M. Sakakibara. 2004a. Involvement of the ryanodine receptor in morphologic modification of Hermissenda type B photoreceptors after in vitro conditioning. J. Neurophysiol.91:728–735.
Kawai, R., H. Sunada, T. Horikoshi, and M. Sakakibara. 2004b. Conditioned taste aversion with sucrose and tactile stimuli in the pond snail Lymnaea stagnalis. Neurobiol. Learn. Mem. 82:164–168.
Ladewig, T., W. Hanke, M. Goldermann, A. C. Guimaraes de Almeida,V. M. Fernandes de Lima. 1998. Distribution of gap junctions in the chicken retina. J. Hirnforsch.39:77–86.
Lederhendler, I. I., and D. L. Alkon. 1989. The interstimulus interval and classical conditioning in the marine snail Hermissenda crassicornis. Behav. Brain Res. 35:75–80.
Lukowiak, K., E. Ringseis, G. Spencer, W. Wildering, and N. Syed. 1996. Operant conditioning of aerial respiratory behaviour in Lymnaea stagnalis. J. Exp. Biol. 199:683–691.
Lukowiak, K., S. Sangha, C. McComb, N. Varshney, D. Rosenegger, H. Sadamoto, and A. Scheibenstock. 2003a. Associative learning and memory in Lymnaea stagnalis: How well do they remember? J. Exp. Biol.206:2097–2103.
Lukowiak, K., S. Sangha, A. Scheibenstock, K. Parvez, C. McComb, D. Rosenegger, N. Varshney, and H. Sadamoto. 2003b. A molluscan model system in the search for the engram. J. Physiol. Paris97:69–76.
Matzel, L. D., B. G. Schreurs, I. Lederhandler, and D. L. Alkon. 1990. Acquisition of conditioned associations in Hermissenda: additive effects of contiguity and the forward interstimulus interval. Behav. Neurosci.104:597–606.
Mills, S. L., and S. C. Massey. 1995. Differential properties of two gap junctional pathways made by AII amacrine cells. Nature377:734–737.
Nelson, T. J., and D. L. Alkon. 1990. Specific high molecular weight mRNAs induced by associative learning in Hermissenda. Proc. Natl. Acad. Sci. USA 87:269–273.
Nelson, T. J., S. Cavallaro, C. L. Yi, D. McPhie, B. G. Schreurs, P. A. Gusev, A. Favit, O. Zohar, J. Kim, S. Beushausen, G. Ascoli, J. Olds, R. Neve, and D. L. Alkon. 1996. Calexcitin: a signaling protein that binds calcium and GTP, inhibits potassium channels, and enhances membrane excitability. Proc. Natl. Acad. Sci. USA93:13808–13813.
Ono, M., R. Kawai, T. Horikoshi, T. Yasuoka, and M. Sakakibara. 2002. Associative learning acquisition and retention depends on developmental stage in Lymnaea stagnalis. Neurobiol. Learn. Mem. 78:53–64.
Ovchinnikov, A. V. 1986. Intersensory interactions in the CNS of Helix lucorum. Neirofiziologiya 18:17–26.
Pinsker, H. M., W. A. Hening, T. J. Carew, and E. R. Kandel. 1973. Long-term sensitization of a defensive withdrawal reflex in Aplysia. Science 182:1039–1042.
Sakakibara, M., H. Ikeno, S. Usui, C. Collin, and D. L. Alkon. 1993. Reconstruction of ionic currents in a molluscan photoreceptor. Biophys. J.65:519–527.
Sakakibara, M., D. L. Alkon, T. Kouchi, H. Inoue, and T. Yoshioka. 1994. Induction of photoresponse by the hydrolysis of polyphosphoinositides in the Hermissenda type B photoreceptor. Biochem. Biophys. Res. Commun.202:299–306.
Sakakibara, M., H. Inoue, and T. Yoshioka. 1998a. Evidence for the involvement of inositol trisphosphate but not cyclic nucleotides in visual transduction in Hermissenda eye. J. Biol. Chem.273:20795–20801.
Sakakibara, M., R. Kawai, S. Kobayashi, and T. Horikoshi. 1998b. Associative learning of visual and vestibular stimuli in Lymnaea. Neurobiol. Learn. Mem. 69:1–12.
Sakakibara, M., T. Aritaka, A. Iizuka, H. Suzuki, T. Horikoshi, and K. Lukowiak. 2005a. Electrophysiological responses to light of neurons in the eye and statocyst of Lymnaea stagnalis. J. Neurophysiol. 93:493–507.
Sakakibara, M., F. Okuda, K. Nomura, K. Watanabe, H. Meng, T. Horikoshi, and K. Lukowiak. 2005b. Potassium currents in isolated statocyst neurons and RPeD1 in the pond snail, Lymnaea stagnalis. J. Neurophysiol. 94:3884–3892.
Sharko, N. V., and B. S. Osipov. 1981. Electrical responses of the eye and optic nerve of the pond snail Lymnaea stagnalis. Neirofiziologiya 13:652–654.
Stensaas, L. J., S. S. Stensaas, and O. Trujillo-Cenoz. 1969. Some morphological aspects of the visual system of Hermissenda crassicornis (Mollusca: Nudibranchia). J. Ultrastruct. Res.27:510–532.
Stoll, C. J. 1973. On the role of eyes and non-ocular light receptors in orientational behaviour of Lymnaea stagnalis (L.). Proc. K. Ned. Akad. Wet. C76:203–214.
Stoll, C. J., and A. Bijlsma. 1973. Optic nerve responses in Lymnaea stagnalis to photic stimulation of the eye. Proc. K. Ned. Akad. Wet. C76:406–413.
Syed, N. I., and W. Winlow. 1991. Coordination of locomotor and cardiorespiratory networks of Lymnaea stagnalis by a pair of identified interneurones. J. Exp. Biol.158:37–62.
Tabata, M., and D. L. Alkon. 1982. Positive synaptic feedback in visual system of nudibranch mollusk Hermissenda crassicornis. J. Neurophysiol. 48:174–191.
Tsubata, N., A. Iizuka, T. Horikoshi, and M. Sakakibara. 2003. Photoresponse from the statocyst hair cell in Lymnaea stagnalis. Neurosci. Lett. 337:46–50.
Tully, T., T. Preat, S. C. Boynton,M. Del Vecchio. 1994. Genetic dissection of consolidated memory in Drosophila. Cell 79:35–47.
West, A., E. Barnes, and D. L. Alkon. 1982. Primary changes of voltage responses during retention of associative learning. J. Neurophysiol.48:1243–1255.
Wildering, W. C., M. van der Roest, T. A. de Vlieger, and C. Janse. 1991. Age-related changes in junctional and non-junctional conductances in two electrically coupled peptidergic neurons of the mollusc Lymnaea stagnalis. Brain Res. 547:89–98.
Xin, D., and S. A. Bloomfield. 1997. Tracer coupling pattern of amacrine and ganglion cells in the rabbit retina. J. Comp. Neurol.383:512–528.
Yamoah, E. N. 1997. Potassium currents in presynaptic hair cells of Hermissenda. Biophys. J. 72:193–203.
Yin, J. C., J. S. Wallach, M. Del Vecchio, E. L. Wilder, H. Zhou, W. G. Quinn, and T. Tully. 1994. Induction of a dominant negative CREB transgene specifically blocks long-term memory in Drosophila. Cell 79:49–58.
Zylstra, U. 1972. Histochemistry and ultrastructure of the epidermis and the subepidermal gland cells of the freshwater snails Lymnaea stagnalis and Biomphalaria pfeifferi. Z. Zellforsch. Mikrosk. Anat. 130:93–134.(Manabu Sakakibara)