SYNAPTIC TAGGING AND THE E-LTP/L-LTP TRANSITION
In 1997, Uwe Frey and Richard Morris published an interesting series of studies where they formulated the "synaptic tag" hypothesis of L-LTP induction (see reference 94 and figure). Without repeating the entirety of the details of their seminal paper and a variety of subsequent work in this area, the basic idea is as follows. L-LTP is dependent on protein synthesis and presumably on altered gene expression as well. How is it that one can have synapse specificity, which is known to occur with L-LTP, in the face of a certain central (nuclear) source of mRNA and potentially a central (rough ER) source of newly synthesized proteins? Frey and Morris proposed that local activity-regulated generation of persisting signals establishes a "synaptic tag" that marks synapses for potentiation when the synapse experiences an L-LTP-inducing stimulation.
The synaptic tag allows the capture of new gene products (mRNAs or proteins) sent out from the nucleus and cell body (also triggered by LTP-inducing stimulation), localizing these potentiating products at the appropriate synapses. Frey and Morris also showed in their original paper that the generation of the synaptic tag could be pharmacologically isolated from the more generalized induction of L-LTP. In other words, they could generate the tag in the absence of protein synthesis by giving E-LTP-inducing stimulation. Then, subsequent L-LTP-inducing stimulation at another set of synapses allowed the original group of synapses to capture the potentiating gene products. While the biochemical identity of the synaptic tag is unknown at present, any one of the variety of persisting post-translational modifications that we are discussing in this chapter are viable can didate mechanisms for contributing to the synaptic tag.
The synaptic tagging work of Frey and Morris also has another very important implication as a mechanism for generating long-lasting synaptic plasticity. Their findings imply that, after a cell has received an L-LTP-inducing strong stimulation, subsequent weaker stimuli can capture the L-LTP-inducing products at their synapses. Thus, a weaker stimulation, when following a strong stimulation, could produce L-LTP. This is a powerful mechanism for temporal integration of signals across time (95). Depending on the timing of the signals, at any particular time, a signal of a given strength may or may not trigger LTP depending on the prior recent "experience" of the cell.
I find this a fascinating finding, in part because this and similar mechanisms of temporal integration have the potential to explain one of the long-standing mysteries in learning and memory. As described in Chapter 2, a highly reproducible feature of learning across species and learning paradigms is the improved efficacy of "spaced" versus "massed" training. Ten training sessions separated by 15 minutes is much more effective in producing long-lasting and robust memory than ten training sessions back-to-back, for example. Synaptic tagging and other temporal integration mechanisms of this sort have the capacity to explain this phenomenon. Subsequent stimuli, when timed appropriately, can have stronger effects than they would otherwise. A weak signal, that normally might not cross the threshold for triggering change, can be converted to a long-lasting signal if it follows a previous training session. If the subsequent training
SYNAPTIC TAGGING AND THE E-LTP/L-LTP TRANSITION
trials come too soon after the initial stimulus, the synaptic tag generated by the weak stimulus may have nothing to capture because the nuclear products have not diffused far enough to be captured. If the weak stimulus comes too late (i.e., the training is too spaced out), the genomic products will have been generated but not captured before they decayed. Thus, temporal integration mechanisms of this sort involving synaptic tagging and the generation of other persisting but transient signals allow for the cell to build an optimal time window for the efficacy of repeated stimuli.
regulation are microtubule-associated protein 1B (MAP1B) and the PSD-95 associated protein SAPAP4. These cytoskeletal/ scaffolding protein targets are a potential means by which FMRP might regulate spine structure.
Other well-known and extensively characterized products of local protein synthesis are CaMKII (85) and Arc (86).
While the synthesis of these proteins is probably not directly regulated by FMRP, they certainly may play an important role as targets of regulated protein synthesis in dendrites. As we have already discussed in various parts of this book, Arc mRNA and protein are selectively localized to active synapses, and Arc as a cytoskeleton-associated protein may serve a morphological/structural role. CaMKII, of course, is a dominant player in post-synaptic function and structure. Thus, while it is quite early to try to synthesize a complete model for the regulation and targets of local dendritic protein synthesis, it is not unreasonable to think that this process plays a key role in various processes including the regulation of spine structure, morphology, and PSD stabilization, as well as the regulation of signal transduction locally.
Finally, I should note that while for the present I am defining L-LTP as that phase of LTP dependent upon changes in both gene expression and protein synthesis, there are recent publications suggesting that there could be an intermediate phase of LTP dependent on altered protein synthesis but not altered gene expression (e.g., 93). This is certainly a reasonable idea, and, in the future, we may need to break down LTP into additional substages beyond E-LTP and L-LTP based on these criteria.
Was this article helpful?
A course in techniques and skills for mentalists, magicians and students. For students, improve your grades with less effort! But this book is also.... The ideal for any stage mentalist or magician by establishing credibility of amazing skills with an easy to follow instructional book on using the amazing power of your memory.