B

BOX 3, cont'd (B) Potential signal transduction routes leading to PI-3-K activation in the hippocampus Mechanisms coupling cAMP to PI-3-K are speculative but based on the available literature. Red circles represent the specific phosphorylation events that can be measured using available phospho-specific antibodies. Abbreviations: PI-3-K, phosphatidylinositol-3-kinase; PIPx, (poly) phosphorylated derivatives of phosphatidylinositol; pH domain, pleckstrin homology domain; RAS, the low-molecular weight G protein ras; GEF, guanine nucleotide exchange factor; PKA, cAMP-dependent protein kinase; GFR, growth factor receptor; RSK, ribosomal S6 kinase; CREB, cAMP response element binding protein; AKT/PKB, the kinase AKT, also known as protein kinase B; Rap-1, RAF, MEK, and ERK are all components of the extracellular signal-regulated kinase (ERK) cascade, a subfamily of the mitogen-activated protein kinases. Reproduced from Sweatt (97).

BOX 3, cont'd (B) Potential signal transduction routes leading to PI-3-K activation in the hippocampus Mechanisms coupling cAMP to PI-3-K are speculative but based on the available literature. Red circles represent the specific phosphorylation events that can be measured using available phospho-specific antibodies. Abbreviations: PI-3-K, phosphatidylinositol-3-kinase; PIPx, (poly) phosphorylated derivatives of phosphatidylinositol; pH domain, pleckstrin homology domain; RAS, the low-molecular weight G protein ras; GEF, guanine nucleotide exchange factor; PKA, cAMP-dependent protein kinase; GFR, growth factor receptor; RSK, ribosomal S6 kinase; CREB, cAMP response element binding protein; AKT/PKB, the kinase AKT, also known as protein kinase B; Rap-1, RAF, MEK, and ERK are all components of the extracellular signal-regulated kinase (ERK) cascade, a subfamily of the mitogen-activated protein kinases. Reproduced from Sweatt (97).

(60-62). A second proposed mechanism is that the steady-state level of membrane AMPA receptor protein is increased in a dynamic fashion by CaMKII through regulation of AMPA receptor trafficking and stabilization (9, 63). Finally, AMPA receptors can be inserted into previously "silent" synapses, increasing the strength of connections between two neurons in an essentially all-or-none fashion (reviewed in reference 64), and this mechanism has been proposed as contributing to E-LTP. In the next section, I will briefly review some of the mechanisms underlying these three processes, based on the current literature. The following discussion draws extensively from groundbreaking work in this area by the laboratories of Roberto Malinow, Rob Malenka, Tom Soderling, and Rick Huganir.

B. Direct Phosphorylation of the AMPA Receptor

AMPA receptors mediate the majority of fast synaptic transmission throughout the nervous system, including at Schaffer-collateral synapses in area CA1. Four homologous alpha subunits (GluR1-GluR4) combine in a mix-and-match fashion into a multiunit complex (likely tetrameric), which forms a functional AMPA receptor. The GluR1 alpha subunit can be phos-phorylated at Ser831 by CaMKII or PKC in vitro, in cultured hippocampal neurons and in the hippocampal slice preparation. Phosphorylation of GluR1 at this site increases the receptor's ionic conductance, providing a direct route for CaMKII or PKC to enhance synaptic efficacy during LTP (65). In fact, in a key paper by Andres Barria and co-workers, phosphorylation of GluR1 at this site was shown to be increased in E-LTP (62). In addition, LTP induction is associated with increased conductance of AMPA receptors as well (61), which is consistent with increased phosphorylation at the Ser831 site. Thus, phosphorylation of GluR1 at Ser831 has been shown to occur in LTP, and this mechanism is sufficient to enhance synaptic transmission. Although most models for E-LTP posit the phos-phorylation of Ser831 to be mediated by CaMKII, persistently active PKC could perform this role as well. An interesting variation on this idea is that CaMKII and PKC might be functionally redundant in E-LTP, each phosphorylating AMPA receptors and serving as a fail-safe mechanism for maintaining synaptic potentiation.

The AMPA receptor is also a substrate for PKA, and PKA phosphorylation likewise increases AMPA channel activity. In a fascinating series of studies by Hei-sung Lee in Rick Huganir's lab, regulation of AMPA receptors by PKA was found to predominantly be involved in de-depression of synaptic strength in area CA1 (66). These studies showed that potentiation and depotentiation revolved around the CaMKII/PKC phosphorylation site, while LTD and dedepression revolved around the PKA site. This important study thereby separated mechanisms for synaptic potentiation from mechanisms for synaptic depression by demonstrating that one process is not simply the reversal of the other. Similar recent work, using a different approach, from Dan Madison's lab has also supported this idea (67).

C. Regulation of Steady-State Levels of AMPA Receptors

Active CaMKII can also lead to an increase in the level of postsynaptic AMPA receptor density in hippocampal pyramidal neurons in culture. A wide variety of sophisticated studies by Robert Malinow's lab have shown that transfection of active CaMKII into pyramidal neurons leads to increased trafficking of AMPA receptors into the dendritic spine and into active synapses (64). Similar processes also occur with LTP induction in cultured neurons in vitro (68). Thus, one target for CaMKII in E-LTP is regulation of steady-state levels of AMPA receptors postsynaptically.

The mechanism for this increased trafficking and membrane insertion is under active investigation. It is known to be independent of Ser831 phosphorylation, clearly rendering this mechanism as distinct and separable from the mechanism described earlier for increasing current flow through the AMPA channel (68). Current models posit that one component of E-LTP is activity-dependent delivery of GluRl/ 2-containing AMPA receptors into the spine and postsynaptic density, a process distinct from a second constitutive pathway that delivers GluR2/3-containing receptors and maintains baseline synaptic transmission.

It is important to note that insertion of AMPA receptors into the postsynaptic membrane is not a mechanism for the maintenance of E-LTP. Postsynaptic membrane AMPA receptors turn over with a lifetime of about 15 minutes (69, 70). Thus, regulation of AMPA receptor insertion is an active process maintained by some other persisting signal. As mentioned previously, one relevant persisting signal is autophos-phorylated, autonomously active CaMKII.

How is it that CaMKII increases steady-state levels of AMPA receptors? The mechanisms are currently under investigation and are complex. There also is not unanimity of opinion in this rapidly evolving area of research. What follows is a hybrid model drawn from the recent work on AMPA receptor trafficking referred to earlier, theoretical work by John Lisman and his colleagues, and findings from Michael Browning's laboratory. Although many investigators will likely disagree with some of the particulars, it will give a flavor of the current thinking about mechanisms for kinase regulation of AMPA receptor expression in E-LTP.

Figure 12 summarizes the model. Calcium influx through the NMDA receptor leads to activation of CaMKII and triggers membrane insertion of AMPA receptors. The calcium signal also causes CaMKII autophos-phorylation, which as we discussed earlier leads to a self-perpetuating increase in autophosphorylated CaMKII. Autophos-phorylated CaMKII binds with high affinity to the cytoplasmic domain of the NMDA receptor and to the actin-binding protein alpha-actinin. The actinin linkage couples the CaMKII/NMDA receptor complex to actin, and as we discussed in the last chapter actin filaments cross-link to AMPA receptors via a number of mechanism including by binding through the 4.1 protein and SAP97. Thus, by this mechanism, autophosphory-lated CaMKII stabilizes AMPA receptors in the PSD by linking them to the more stable NMDA receptor complex.

An additional component of the model is that the calcium signal acting through persistently activated PKC and src leads to increased insertion of NMDA receptors in the postsynaptic membrane. This has been shown to occur with LTP-inducing stimulation by Grosshans, Clayton, Coultrap, and Browning (71). This second signal increases the number of NMDA receptor "anchors" in the PSD and, by this mechanism, contributes to elevating the steady-state level of AMPA receptors postsynaptically.

Notable components of the model distinguish it from Ser831 phosphorylation of AMPA receptors as a mechanism for increasing synaptic strength. First, it does not require CaMKII phosphorylation of AMPA receptors or indeed any other CaMKII substrate beside CaMKII itself— autophosphorylated CaMKII serves a structural not a catalytic role. Second, it involves a number of receptor-interacting proteins in the PSD that stabilize the presence of AMPA receptors. Third, it involves the cytoskeleton and thus could serve as a signaling system beyond simply regulating AMPA receptor function. Fourth, it involves

FIGURE 12 A model for glutamate receptor insertion and stabilization in E-LTP. See text for details. Adapted from Lisman and Zhabotinsky (9).

the NMDA receptor as well as the AMPA receptor, with the NMDA receptor serving as a PSD-organizing molecule. Finally, it involves parallel actions of PKC and CaMKII, as opposed to the two kinases converging on the same target phospho-rylation site.

D. Silent Synapses

As we discussed in Chapter 4 on LTP physiology, de novo insertion of AMPA receptors is an additional potential mechanisms for enhanced synaptic strength in LTP. "Silent" synapses containing NMDA receptors but not functional AMPA receptors occur with reasonable frequency in prenatal and neonatal brain. NMDA receptor-dependent triggering of AMPA receptor insertion into silent synapses occurs in an activity-dependent fashion in neurons, likely by mechanisms quite similar to those described earlier for elevating AMPA receptor levels in the PSD. Thus, activation of silent synapses through AMPA receptor insertion is clearly a potential mechanism for E-LTP, and "AMPA-fication" of synapses occurs under a number of experimental conditions (reviewed in reference 64).

However, the quantitative contribution of silent synapse activation in LTP in the adult hippocampus is unclear. Activation of silent synapses has been difficult to observe in acute slices from adult animals, although the process is quite robust in cultured neurons in vitro and in immature animals. Thus, at the present level of understanding, it appears that increasing AMPA receptor ionic conductance and regulating the steady-state levels of AMPA receptors at the synapse may be the predominant mechanisms for E-LTP at adult synapses, while activation of silent synapses may be more important in the context of developmental synaptic plasticity.

This also raises an important general point. Overall, the magnitude of the contribution of each of the three mechanism for augmenting AMPA receptor function is unclear as well. The extent of receptor insertion, stabilization, and phosphory-lation are subject to many variables including the recent history of the synapse (see reference 67) and developmental age. Thus, there still is an open question concerning the precise mechanisms for enhancing AMPA receptor function, and different mechanisms may operate under a wide number of different experimental conditions and in different synaptic states in vivo.

E. Proteolysis

A final note is that proteolysis of cytoskeletal and AMPA receptor-associated proteins has also been proposed to play a role in the generation of enhanced synaptic strength. Michel Baudry and Gary Lynch have published a variety of evidence supporting their model that the calcium-activated protease calpain cleaves postsynaptic scaffolding and cytoskeletal proteins, and via this mechanism synaptic strength is enhanced. One specific mechanism that has been proposed is that the AMPA receptor interacting protein GRIP (72) is proteolyzed by calpain and that by this mechanism AMPA receptor function is enhanced. This idea is supported by the finding that inhibition of calpain can block LTP (73). An additional relevant mechanism is that the actin-binding protein spectrin undergoes proteolysis with LTP-inducing stimulation (74), suggesting that calpain is indeed activated with LTP induction. Further work will be required in order to define the role of this potential mechanism in elevating AMPA receptor function in E-LTP.

F. Presynaptic Changes— Increased Release

LTP is typically measured as an increase in the initial slope of the EPSP (or EPSP magnitude), and this effect can be subserved by an increase in postsynaptic receptor number or efficacy, as described previously, or by an increase in presynaptic neurotransmitter release, or both of these mechanisms together. The locus of LTP expression (pre- versus postsynaptic) has been widely debated and is a source of continuing controversy, as was described in Chapter 5. The evidence for postsynaptic changes in LTP seems quite convincing at this point, and not much argument about this aspect of LTP exists. The physiologic evidence for changes in neurotransmitter release is less convincing than that for changes in postsynaptic responsiveness in my opinion, but many experts in this area disagree on the interpretation of the relevant data. However, there is much evidence indicative of a change in release presynap-tically in E-LTP, and there is convincing biochemical evidence presynaptic changes do indeed occur in LTP. Thus, we will briefly discuss mechanism to account for lasting changes in the presynaptic compartment.

One of the most interesting and convincing physiologic studies on this topic was carried out by Dan Madison's laboratory (see Figure 13). Dan's lab has

FIGURE 13 Selective reversal of E-LTP by a presynaptic protein kinase inhibitor. Presynaptic injection of the protein kinase inhibitor H-7 (100 pM) inhibits LTP but does not affect basal transmission. Controls, open circles; H-7, filled circles. (A) Monitoring basal transmission over a period of 80 minutes. Injection of H-7 into the presynaptic neuron did not suppress basal EPSC amplitudes compared with control (n = 10). (B) Responses to 1-Hz stimulation for 1 minute, such as those used to induced LTP, were unaffected by H-7 injection into the presynaptic cell. (C) Inhibitory effect of H-7 on LTP. The graph shows data from pairs in which the postsynaptic cell was obtained first with the amphotericin perforated patch technique, enabling basal transmission to be monitored for 40 minutes before pairing. After pairing there was some initial potentiation, but this decayed rapidly leaving no significant potentiation after 20 minutes. For H-7 experiments, n = 7; for control experiments n = 17. Inset, Sample sweeps from before and after pairing. (D) Summary of all H-7 experiments (both perforated patch and whole-cell mode) showing that LTP was reduced on average by H-7 (filled circles) when compared with controls (open circles and filled triangles). The baseline data in this panel have been truncated to the length of the experiments with the shortest baselines. Data and legend courtesy of Pavlidis, Montgomery, and Madison (75).

FIGURE 13 Selective reversal of E-LTP by a presynaptic protein kinase inhibitor. Presynaptic injection of the protein kinase inhibitor H-7 (100 pM) inhibits LTP but does not affect basal transmission. Controls, open circles; H-7, filled circles. (A) Monitoring basal transmission over a period of 80 minutes. Injection of H-7 into the presynaptic neuron did not suppress basal EPSC amplitudes compared with control (n = 10). (B) Responses to 1-Hz stimulation for 1 minute, such as those used to induced LTP, were unaffected by H-7 injection into the presynaptic cell. (C) Inhibitory effect of H-7 on LTP. The graph shows data from pairs in which the postsynaptic cell was obtained first with the amphotericin perforated patch technique, enabling basal transmission to be monitored for 40 minutes before pairing. After pairing there was some initial potentiation, but this decayed rapidly leaving no significant potentiation after 20 minutes. For H-7 experiments, n = 7; for control experiments n = 17. Inset, Sample sweeps from before and after pairing. (D) Summary of all H-7 experiments (both perforated patch and whole-cell mode) showing that LTP was reduced on average by H-7 (filled circles) when compared with controls (open circles and filled triangles). The baseline data in this panel have been truncated to the length of the experiments with the shortest baselines. Data and legend courtesy of Pavlidis, Montgomery, and Madison (75).

developed a very nice cell culture model system wherein they study LTP at synapses between CA3 pyramidal neurons in culture—synapses which appear to produce LTP quite similar to that at Schaffer-collateral synapses. The beauty of the system is that they can simultaneously have electrodes in both the presynaptic and postsynaptic neurons, giving unprecedented control over the presynaptic neuron, at least for mammalian CNS neurons. Using this preparation, Pavlides, Montgomery, and Madison (75) found that injection of the nonspecific protein kinase inhibitor H7 into the presynaptic neuron caused a selective reversal of E-LTP without perturbing baseline synaptic transmission.

What is the identity of this presynaptic kinase? Various evidence is consistent with the hypothesis that it is PKC. As described in the first section of this chapter, a variety of evidence indicates that persistently activated PKC is generated in LTP. More direct evidence that presynaptic PKC activity is increased is also available. As has been emphasized by one of the pioneers in this area, Aryeh Routtenberg, this is nicely illustrated by biochemical studies of LTP-associated increases in phosphory-lation of the presynaptic protein GAP-43

(also known as B50, F1, and neuromodulin (76). There is very good biochemical evidence that PKC-mediated GAP-43 phosphorylation increases in LTP (32, 33, 77). GAP-43 is a calmodulin-binding protein similar to the protein neurogranin, which we discussed in the last chapter, and PKC phosphorylation of GAP-43 may increase presynaptic calmodulin levels and help facilitate neurotransmission by this mechanism (see Figure 14). In addition, a number of studies have shown that presy-naptic PKC can increase neurotransmitter release by more direct effects on the release mechanism itself (37, 38, 78).

Taken together with data from Eric Klann's group showing persistent oxidative activation of PKC in E-LTP, these data allow a parsimonious model for retrograde signaling and facilitation of neurotrans-mitter release (see Figure 14). In this model, calcium-induced generation of reactive oxygen species postsynaptically allows retrograde signaling to PKC presynap-tically. Persistently activated PKC presynap-tically phosphorylates GAP-43 and other targets, leading to increased neurotrans-mitter release and synaptic potentiation.

This model is speculative, but the reader should not miss a very important point that

Nmda Receptor Ltp Model
FIGURE 14 Retrograde signaling in E-LTP. Two potential presynaptic sites of PKC action are illustrated—direct effects on the release process and the calmodulin binding protein GAP43.

arises from the underlying data, which is independent of the particulars of the model. Consider the GAP-43 phosphory-lation data together with other biochemical evidence from studies of other presynaptic proteins (79-83), plus the observations from Dan Madison's lab of presynaptic physiologic changes in their cultured neuron system. With these data in aggregate, a strong case can be made that NMDA receptor-dependent, postsynaptically induced presynaptic changes occur in LTP. The conclusion that can be drawn from these observations is profound; neurons in the CNS are capable of retrograde signaling. This conclusion stands independently of whether or not the particulars of the model are correct, or indeed independent of whether increased neurotransmitter release contributes to the expression of LTP. However it occurs and whatever its effects, retrograde signaling is an important component of the cell biological armamentarium available to the central neuron.

G. Postsynaptic Changes in Excitability?

As we have already discussed, LTP, as originally defined by Bliss and Lomo, is manifest as two physiologic components. The first component is an increase in synaptic strength, and the second component of LTP is referred to as EPSP-slope (E-S) potentiation. E-S potentiation is a general term used to refer to the post-synaptic cell having an increased probability of firing an action potential at a constant strength of synaptic input. E-S potentiation can be explained based on alterations in recurrent inhibitory connections in area CA1 (see Chapter 4). The other possibility is that E-S potentiation is a manifestation of increased excitability in the postsynaptic neuron, but the molecular mechanisms that could account for this aspect of LTP are completely mysterious at present. In an extension of the variety of mechanisms that we have been talking about in the context of AMPA receptor regulation, one can hypothesize that persistently activated CaMKII or PKC could regulate potassium or sodium channels in order to increase excitability. Investigations to test whether such phenomena occur are currently under way, although these experiments are quite difficult due to the necessity of recording directly from CA1 pyramidal neuron dendrites after LTP induction. It also is important to remember that post-synaptic changes in voltage-dependent potassium and sodium channels might also alter the postsynaptic depolarization produced by AMPA receptor activation and by that mechanism contribute directly to EPSP potentiation as well as postsynaptic excitability.

Was this article helpful?

0 0
How To Win Your War Against Anxiety Disorders

How To Win Your War Against Anxiety Disorders

Tips And Tricks For Relieving Anxiety... Fast Everyone feels anxious sometimes. Whether work is getting to us or we're simply having hard time managing all that we have to do, we can feel overwhelmed and worried that we might not be able to manage it all. When these feelings hit, we don't have to suffer. By taking some simple steps, you can begin to create a calmer attitude, one that not only helps you feel better, but one that allows you the chance to make better decisions about what you need to do next.

Get My Free Ebook


Post a comment