Molecular Basis of Long-Term Sensitization

Sensitization is a simple form of nonassociative learning that involves the enhancement of the response to a weak stimulus that occurs after the presentation of a strong or noxious stimulus. Sensitization usually occurs in two forms that differ in their duration and underlying mechanisms. Short-term sensitization lasts seconds to minutes and involves the modification of neuronal membrane properties and synaptic efficacy, often through the alteration of the phosphorylation state of existing proteins. Long-term sensitization lasts from days to weeks, depending on the training protocol. Unlike the short-term version, long-term sensitization requires synthesis of new macromolecules—the inhibition of either gene transcription into mRNA or translation of mRNA into protein blocks long-term sensitization. In its most persistent form, long-term sensitization involves morphological changes and neuronal growth.

The marine mollusk Aplysia has proved a useful model for gaining insights into the underlying neural and molecular mechanisms of long-term sensitization. Aplysia has a simple nervous system with large, individually identifiable neurons that are accessible for detailed anatomical, biophysical, biochemical, and molecular studies. Researchers have identified many of the neural circuits involved in sensitization of several reflexive withdrawal behaviors and have characterized individual neurons within these circuits. They have also identified specific locations of learning-related modifications in these circuits and have analyzed the underlying cellular mechanisms.

The Siphon Withdrawal Reflexes of Aplysia

Sensitization studies in Aplysia have focused chiefly on the withdrawal reflexes of the siphon gill and the tail siphon. The gill, the animal's respiratory organ, is in the mantle cavity. The siphon is a fleshy, tubelike extension of the body wall protruding from the mantle cavity. A stimulus delivered to the siphon elicits withdrawal of both the siphon and the gill: This is the siphon-gill withdrawal reflex. Eliciting the tailsiphon withdrawal reflex requires the delivery of a stimulus to the tail, causing withdrawal of both the tail and the siphon. In the wild, sensitization might be induced by an unsuccessful attack by a predator (e. g., pinching or scratching by a crab). In the laboratory, electrical shocks to the body wall are the most common means of sensitizing the animal because it is easier to deliver the same stimulus repeatedly. Long-term sensitization requires prolonged training (multiple shocks delivered over an extended interval—e.g., one to two hours). One training session can sensitize the animal for several days, whereas multiple training sessions repeated over several days can sensitize the animal for weeks.

The afferent signal is carried by mechanosensory neurons. About twenty-four sensory neurons innervate the siphon, and another 200 sensory neurons innervate the body wall (a subset of only about twenty of these innervate the tail). These sensory neurons make excitatory synapses onto motor neurons that cause contractions of muscles in the gill, siphon, and tail. Sensory neuron membrane properties and pre-synaptic release machinery are the principal sites of plasticity during sensitization. Sensory neurons also synapse with a variety of interneurons. Some interneurons convey sensory information from one body region to the circuits that control other regions. Other interneurons secrete modulatory transmitters such as serotonin that modify the biophysical properties of sensory and motor neurons and the characteristics of neurotransmitter release from the sensory neuron (i.e., heterosynaptic facilitation, discussed in more detail below).

Cellular Correlates of Long-Term Sensitization

One advantage of Aplysia for studies on long-term learning is that the nervous system, or portions of it, can be removed for studies in vitro. Studies in reduced preparations derived from sensitized and control animals have shown that changes in the animal's behavior correlate with changes in the properties of both the presynaptic sensory neuron and the postsynaptic motor neuron. Although these are unlikely to be the only sites of plasticity associated with sensitization in the nervous system, changes in the sensory and motor neurons are likely to be important because they occur in the basic circuit of the reflex arc.

Modulation of the Presynaptic Sensory Neuron

The first correlate of long-term sensitization to be described was a decrease in sensory-neuron-membrane potassium current (Scholz and Byrne, 1987). Researchers had found that modulation of the same current for short-term sensitization was sensitive to serotonin (5-hydroxytryptamine, 5-HT) and cyclic adenosine monophosphate (cAMP) treatments. The decrease in membrane current after long-term sensitization training leads to an increase in the number of action potentials elicited for a fixed amount of injected current. Thus, sensory neurons fire more action potentials in response to a given stimulus after sensitization (Cleary, Lee, and Byrne 1998).

Another correlate of long-term sensitization is an increase in synaptic efficacy at the synapses between sensory neurons and motor neurons (Cleary, Lee, and Byrne 1998; Frost, Castellucci, Hawkins, and Kandel, 1985). A major component of this synaptic modulation appears to be due to an increase transmitter release from the presynaptic sensory neurons (Dale, Schacher, and Kandel, 1988), although the nature of this increase is still poorly understood. The prime targets of of modulation are as follows: broadening action potential, increased numbers of release-ready synaptic vesicles, and increased efficiency of the synaptic-release machinery. It is not yet known whether other synapses made by sensory neurons (e.g., onto interneurons) are also enhanced. In addition, postsynaptic mechanisms may also contribute to increased synaptic efficacy (Cleary, Lee, and Byrne 1998; see below).

More extensive training (e.g., four days) produces more persistent sensitization. After four days of training, anatomical studies have revealed remodeling of sensory neuron release sites and the growth of additional axonal and dendritic branches as well as new synaptic contacts that may contribute to increased synaptic efficacy (Bailey and Chen, 1983; Wainwright, Zhang, Byrne, and Cleary, 2002). The recruitment of this growth requires several days of training rather than a single day (Wainwright, Zhang, Byrne, and Cleary, 2002). The down regulation of theAplysia neuronal cell adhesion molecule, ApCAM, plays an important role in this growth-associated plasticity (Abel and Kandel, 1998). Sensory neurons reduce their synthesis of new ApCAM protein and internalize ApCAM already present at their plasma membranes. The signal for internalization of ApCAM is phosphorylation by extracellular signal-regulated kinase (ERK, discussed below). The decrease in ApCAM in sensory-neuron axonal membranes may allow the formation of new branches and additional neurotransmitter release sites.

Modulation of the Postsynaptic Neuron

Changes in the properties of the postsynaptic neuron may also contribute to the increase in synaptic efficacy underlying long-term sensitization. After training, there are changes in two motor-neuron membrane properties: an increase in the resting-membrane potential and a decrease in the spike threshold (Cleary et al, 1998). In addition, motor neurons show a protein-synthesis-dependent increased responsiveness to glutamate twenty-four hours after 5-HT treatment (Trudeau and Castellucci, 1995). This result suggests an increase in the number of postsynaptic receptors that accompanies the increase in presynaptic transmitter release. However, blocking this increase in postsynaptic responsiveness does not block long-term facilitation.

Heterosynaptic Modulation

What is the modulatory signal that produces these presynaptic and postsynaptic changes? It seems that the sensory neurons that convey information about the sensitizing stimulus into the central nervous system synapse onto one or more interneurons. These interneurons, in turn, release a modulatory transmitter to produce the changes described above. A growing body of evidence indicates that the transmitter used by these interneurons is serotonin (5-HT) (Byrne and Kandel, 1996). Indeed, 5-HT mimics the many effects of sensitization training: Prolonged or multiple applications of 5-HT are necessary to produce long-term changes in vitro. As an analog of sensitization, 5-HT use has proved invaluable in studies seeking to delineate the cellular mechanisms that underlie long-term sensitization.

In addition to 5-HT, other transmitter molecules may induce long-term sensitization. In the late 1990s, researchers found evidence that a member of the transforming growth factor β family was involved in long-term sensitization (Zhang et al., 1997). Indeed, treatment of Aplysia neurons with human transforming growth factor β results in long-term facilitation and increased sensory-neuron excitability (Chin et al., 1999; Zhang et al., 1997). Furthermore, a scavenger molecule specific for transforming growth factor β blocks long-term facilitation induced by 5-HT treatment. These data suggest that the Aplysia homologue of the growth factor may be necessary for the induction or expression of long-term plasticity by acting downstream of 5-HT. Transforming growth factor β activates two protein kinases (PKC and ERK, see below) in Aplysia sensory neurons (Chin et al., 2002; Farr, Mathews, Zhu, and Ambron 1999)

Second Messengers and Protein Kinase Cascades

In the early 1980s, researchers found that sensitization training and 5-HT each increased cytoplasmic levels of the intracellular second messenger, cAMP (Bernier, Castellucci, Kandel, and Schwartz, 1982; Ocorr, Tabata, and Byrne, 1986). Regulation of this second messenger is critical for long-term memory in fruit flies, bees, and rodents. In Aplysia, injection of cAMP into sensory neurons produces many of the long-term changes previously correlated with sensitization training, including decreased potassium conductance, increased membrane excitability, and neurite growth (O'Leary, Byrne, and Cleary, 1995; Schacher et al., 1988, 1993; Scholz and Byrne, 1988). Indeed, cAMP appears to be sufficient for the induction of several of the major forms of long-term neuronal plasticity associated with sensitization.

The key effector of cAMP for long-term neuronal plasticity is the cAMP-dependent protein kinase (PKA). Inhibition of PKA during the induction and early stages of consolidation of long-term memory blocks synaptic facilitation one day after 5-HT treatment (Abel and Kandel, 1998), and injection of the active catalytic subunit of PKA can induce long-term facilitation in the absence of upstream signaling (Chain et al., 1999). Among the substrates of PKA that are critical for the induction of long-term plasticity are several transcription factors that regulate gene expression. The principal target of PKA is the cAMP/Ca²⁺-responsive element binding protein (CREB, discussed below), a transcription factor that plays important roles in many forms of long-term memory. In addition, PKA activity persists at least twenty-four hours after sensitization training or treatment with 5-HT (Chain et al., 1999; Müller and Carew, 1998). It is possible, then, that persistent kinase activity is also important for long-term changes.

Although the activation of the cAMP/PKA cascade appears to be sufficient to produce long-term plasticity associated with sensitization, other kinases play important roles in the induction of sensitization. One of the kinases is protein kinase C (PKC). Sensitization training or extended 5-HT treatment results in a prolonged activation of PKC that lasts about two hours after training or treatment. During this period PKC contributes to the regulation of translation and the activation of a transcription factor (C/EBP, see below) that regulates the expression of late genes necessary for long-term sensitization.

Another kinase cascade that plays an essential role in the induction of long-term facilitation is the extracellular signal-regulated kinase (ERK), a member of the mitogen-activated protein kinase family. ERK can be activated by several intracellular cascades. Researchers have yet to pinpoint the one that leads to its activation during sensitization training or by 5-HT treatment. Yet in Aplysia sensory neurons cAMP can mediate both the activation of ERK and its translocation to the nucleus. The cAMP cascade appears to be important for activation of ERK during the early stages of long-term sensitization induction. ERK can also be activated by PKC, a cascade that may be important during later stages of induction.

The protein synthesis required for long-term sensitization occurs during a critical period that begins at the onset of training and continues until training ends (Castellucci, Blumenfeld, Goelet, and Kandel, 1989; Levenson et al., 2000). This period corresponds to the time when CREB-dependent transcription occurs.

There is evidence that a second round of protein synthesis occurs during the formation of long-term sensitization. Growth of sensory neurons in response to cAMP injections is blocked by protein-synthesis inhibitor up to seven hours after injection (O'Leary, Byrne, and Cleary, 1995). This second round of protein synthesis may correspond to the gene expression that stabilizes long-term sensitization.

CREB and Memory

As noted above, long-term memory differs from short-term memory chiefly in requiring protein synthesis. During the induction of long-term memory, activated PKA translocates from the sensory neuron cytoplasm into the nucleus, where it phosphorylates the transcriptional activator CREB1 (Abel and Kandel, 1998). CREB1 activates transcription by binding to the DNA of certain genes that contain cAMP/Ca²⁺ responsive element (CRE) sequences in their regulatory domains. Another form of CREB, CREB2, represses transcription of these genes. CREB2 is a substrate of ERK, which also translocates into the nucleus during the induction phase. Thus, expression of CRE-containing genes requires both activation by CREB1 and derepression by CREB2.

Immediate Early Genes

CREB-dependent transcription leads to the expression of immediate early genes. Researchers have identified two of these genes. One is a key component of the ubiquitin-proteosome pathway for the degradation of proteins, Aplysia ubiquitin C-terminal hydrolase (ApUCH), and the other is another transcription factor, CCAAT/enhancer binding protein (C/EBP). ApUCH is an enzyme that associates with the proteosome and increases its proteolytic activity. ApUCH is an immediate early gene involved in the induction of long-term sensitization (Abel and Kandel, 1998; Alberini, 1999). CREB activation increases the expression of ApUCH. An important substrate of the proteosome in sensory neurons is the regulatory subunit of PKA. Increased expression of ApUCH results in the proteosome-mediated degradation of a subset of the PKA regulatory subunits (those binding cAMP). As the ratio of the regulatory to catalytic subunits decreases, PKA activity becomes independent of cAMP and is prolonged. Thus, CREB activation leads to degradation of the PKA regulatory subunit and thereby prolongs the activity of the PKA catalytic subunit, which is now independent of cAMP.

C/EBP, the CCAAT/enhancer binding protein, is another transcription factor. In Aplysia sensory neurons, C/EBP is an immediate early gene expressed in response to CREB activation (Alberini, 1999). C/EBP can form both homodimers with other C/EBP proteins or heterodimers with other transcription factors to activate transcription of late genes. In order to activate transcription of its target genes, C/EBP must be phosphorylated by ERK. Although the synthesis of a number of proteins are known to be regulated by sensitization, researchers have yet to identify C/EBP target genes in Aplysia.

Although there are other sites of neural plasticity, the detailed study of the modifications that occur in sensory neurons and their synapses with motor neurons (i. e., the fundamental reflex arc) during long-term sensitization have led to an understanding of the basic mechanisms that underlie this simple form of learning. Research using the relatively simple nervous system of Aplysia has provided insights into memory formation at the cellular, molecular, and biophysical levels. These basic mechanisms appear to form the platform upon which more complex forms of learning (i. e., classical and operant conditioning) develop.

See also:MORPHOLOGICAL BASIS OF LEARNING AND MEMORY: INVERTEBRATES; PROTEIN SYNTHESIS IN LONG-TERM MEMORY IN VERTEBRATES

Bibliography

Abel, T., and Kandel, E. R. (1998). Positive and negative regulatory mechanisms that mediate long-term memory storage. Brain Research Reviews 26, 360-378.

Alberini, C. (1999). Genes to remember. Journal of Experimental Biology 202, 2,887-2,891.

Bailey, C. H., and Chen, M. (1983). Morphological basis of long-term habituation and sensitization in Aplysia. Science 220, 91-93.

Bernier, L., Castellucci, V. F., Kandel, E. R,. and Schwartz, J. H. (1982). Facilitatory transmitter causes a selective and prolonged increase in adenosine 3',5'-monophosphate in sensory neurons mediating the gill and siphon withdrawal reflex in Aplysia. Journal of Neuroscience 2, 1,682-1,691.

Byrne, J. H., and Kandel, E. R. (1996). Presynaptic facilitation revisited: State and time dependence. Journal of Neuroscience 16, 425-435.

Castellucci, V. F., Blumenfeld, H., Goelet, P., and Kandel, E. R. (1989). Inhibitor of protein synthesis blocks long-term behavioral sensitization in the isolated gill-withdrawal reflex of Aplysia. Journal of Neurobiology 20, 1-9.

Chain, D. G., Casadio, A., Schacher, S., Hegde, A. N., Valbrun, M., Yamamoto, N., Goldberg, A. L., Bartsch, D., Kandel, E. R., and Schwartz, J. H. (1999). Mechanisms for generating the autonomous cAMP-dependent protein kinase required for long-term facilitation in Aplysia. Neuron 22, 147-156.

Chin, J., Angers, A., Cleary, L. J., Eskin, A., and Byrne, J. H. (1999). TGF- β 1 in Aplysia: Role of long-term changes in the excitability of sensory neurons and distribution of T β R-II-like immunoreactivity. Learning and Memory 6, 317-330.

—— (2002). TGF- β 1 alters synapsin distribution and modulates synaptic depression in Aplysia. Journal of Neuroscience 22, RC220.

Cleary, L. J., Lee, W. L., and Byrne, J. H. (1998). Cellular correlates of long-term sensitization in Aplysia. Journal of Neuroscience 18, 5,988-5,998.

Dale, N., Schacher, S., and Kandel, E. R. (1988). Long-term facilitation in Aplysia involves increase in transmitter release. Science 239, 282-285.

Farr, M., Mathews, J., Zhu, D. F., and Ambron, R. T. (1999). Inflammation causes a long-term hyperexcitability in the nociceptive sensory neurons of Aplysia. Learning & Memory 6, 331-340.

Frost, W. N., Castellucci, V. F., Hawkins, R. D., and Kandel, E. R. (1985). Monosynaptic connections made by the sensory neurons of the gill-and siphon-withdrawal reflex in Aplysia participate in the storage of long-term memory for sensitization. Proceedings of the National Academy of Sciences of the United States of America 82, 8,266-8,269.

Levenson, J., Endo, S., Kategaya, L. S., Fernandez, R. I., Brabham, D. G., Chin, J., Byrne, J. H., and Eskin, A. (2000). Long-term regulation of neuronal high-affinity glutamate and glutamine uptake in Aplysia. Proceedings of the National Academy of Sciences of the United States of America 97, 12,858-12,863.

Ocorr, K. A., Tabata, A. M., and Byrne, J. H. (1986). Stimuli that produce sensitization lead to an elevation of cyclic AMP levels in tail sensory neurons of Aplysia. Brain Research 371, 190-192.

O'Leary, F. A., Byrne, J. H., and Cleary, L. J. (1995). Long-term structural remodeling in Aplysia sensory neurons requires de novo protein synthesis during a critical time period. Journal of Neuroscience 15, 3,519-3,525.

Schacher, S., Castellucci, V. F., and Kandel, E. R. (1988). cAMP evokes long-term facilitation in Aplysia sensory neurons that requires new protein synthesis. Science 240, 1,667-1,669.

Schacher, S., Kandel, E. R., and Montarolo, P. (1993). cAMP and arachidonic acid simulate long-term structural and functional changes produced by neurotransmitters in Aplysia sensory neurons. Neuron 10, 1,079-1,088.

Scholz, K. P., and Byrne, J. H. (1987). Long-term sensitization in Aplysia: biophysical correlates in tail sensory neurons. Science 235, 685-687.

—— (1988). Intracellular injection of cAMP induces a long-term reduction of neuronal K⁺ currents. Science 240, 1,664-1,666.

Trudeau, L.-E., and Castellucci, V. F. (1995). Postsynaptic modifications in long-term facilitation in Aplysia: Upregulation of excitatory amino acid receptors. Journal of Neuroscience 15, 1,275-1,284.

Wainwright, M. L., Zhang, H., Byrne, J. H., and Cleary, L. J. (2002). Localized neuronal outgrowth induced by long-term sensitization training in Aplysia. Journal of Neuroscience 22, 4,132-4,141.

Zhang, F., Endo, S., Cleary, L. J., Eskin, A., and Byrne, J. H. (1997). Role of transforming growth factor- β in long-term synaptic facilitation in Aplysia. Science 275, 1,318-1,320.

VincentCastellucci

Revised byGregg A.Phares

andJohn H.Byrne