Info

Heparin-binding growth-associated molecule (knockout) (74) LIM Kinase (Williams syndrome) (91) Fragile X2 protein (92) PSD-95 (93)

5HT1A receptor (94) Cav2.3 channel (95) PKCP (9) Ataxin-1 (96) L1 adhesion molecule (97) Truncated TrkB receptors (98) Kv1.1 (99)

CaMKII (100-102)

BDNF (106-109)

mGluR1 (111)

NMDAR tail mutants (32, 112)

AC1/8 double knockout (115)

Integrin-associated protein (118)

PACAP receptor 1(mossy fiber LTP) (125) Acid-sensing ion channel (126) Mitochondrial VDAC (128) Ras GRF (130)

Neurofibromatosis Type 1 (103) Angelman syndrome gene (ubiquitin ligase) (110)

Extracellular superoxide dismutase

(transgenic) (113) SOD1 (transgenic) (116) Zif268 (114)

Constitutively active CaMKII (29, 117) CREB/ATF family transcription factors (120) CaMKIV (122) PKA (123, 124)

Inbred mouse lines—CBA and DBA (127) Calbindin/ D28 (129)

FIGURE 1 Loss of LTP does not correlate with a loss of spatial learning in the Morris water maze. (A) Summary graphs of extracellular fEPSP slopes evoked in the tetanized (filled circles) and untetanized (open circles) pathways from slices of wild-type (WT; n = 5-17; five mice) and GluR-A deficient mice (n = 20; six mice). Data were obtained in bicuculline methochloride (10 mM) and 4 mM Ca2+ and Mg2+. Arrow, time of tetanic stimulation. (B) Spatial learning. (a) In a Morris water maze the mean latency (± SEM) to escape from the pool to the submerged platform (eight trials per day in blocks of four) is presented as a function of trial block for male wild-type (n = 19) (filled squares) and GluR-A-/- (n = 21) (open squares) mice. (b) Swim test gives the mean distance covered by the two genotypes in 50 s in the pool without a platform before training. (c) In a transfer test after trial block 10, the platform was removed and wild-type and GluR-A-/- mice were allowed to search for 60 seconds. Both groups searched selectively in the target quadrant (SE). Ordinate, percent time spent in each quadrant. (d) Quadrant preference was not observed when distal visual cues were invisible in the transfer test. Figure and Legend reproduced from Zamanillo et al. (7).

FIGURE 1 Loss of LTP does not correlate with a loss of spatial learning in the Morris water maze. (A) Summary graphs of extracellular fEPSP slopes evoked in the tetanized (filled circles) and untetanized (open circles) pathways from slices of wild-type (WT; n = 5-17; five mice) and GluR-A deficient mice (n = 20; six mice). Data were obtained in bicuculline methochloride (10 mM) and 4 mM Ca2+ and Mg2+. Arrow, time of tetanic stimulation. (B) Spatial learning. (a) In a Morris water maze the mean latency (± SEM) to escape from the pool to the submerged platform (eight trials per day in blocks of four) is presented as a function of trial block for male wild-type (n = 19) (filled squares) and GluR-A-/- (n = 21) (open squares) mice. (b) Swim test gives the mean distance covered by the two genotypes in 50 s in the pool without a platform before training. (c) In a transfer test after trial block 10, the platform was removed and wild-type and GluR-A-/- mice were allowed to search for 60 seconds. Both groups searched selectively in the target quadrant (SE). Ordinate, percent time spent in each quadrant. (d) Quadrant preference was not observed when distal visual cues were invisible in the transfer test. Figure and Legend reproduced from Zamanillo et al. (7).

entertain alternative ways of thinking about how LTP fits into the larger issue of how memory happens.

Some of the other dissociations of LTP and memory that are illustrated in Table 1 are fairly easily explained. No change in LTP with an attendant memory deficit can be explained simply by the presence of a deficit in another relevant brain region. A likely example of this is the PKC beta knockout mouse characterized by Ed Weeber in my lab (9). These animals have a pronounced contextual fear-conditioning deficit but no hippocampal LTP deficit. However, PKC beta is prominently expressed in the basolateral nucleus of the amygdala, and the knockout animals have a deficit in cued fear conditioning. Thus, their deficit in the hippocampus-dependent memory paradigm is easily explained as being the result of a defect in amygdalar plasticity. A similar rationale can be applied to other findings where memory is affected without an attendant change in hip-pocampal LTP.

Similarly, one can explain animals that exhibit an enhancement of LTP with an attendant memory deficit in a fairly straightforward fashion. Any manipulation that perturbs LTP may disrupt the plastic capacity of the entire hippocampal system. An aberrantly high net baseline synaptic transmission caused by a lifelong overexpression of LTP may be functionally equivalent to a hippocampal lesion. In a more specific scenario, if hippocampal synapses are driven to saturation as a result of excessively robust LTP capacity, this may occlude the plasticity necessary for memory formation. We will return shortly to this as a specific type of experimental manipulation. The observation that saturating LTP artificially causes memory deficits is, in fact, interpreted as being consistent with a role for LTP in memory formation.

Thus, we have seen from our analysis of Table 1 that LTP clearly does not equal memory. One should not be surprised when experimental analysis reveals that hippocampal LTP and hippocampus-dependent memory are not co-varying. Nevertheless, further consideration of the experimental observations reveals that the data can be interpreted in ways that are consistent with a role for LTP in memory formation of various sorts. LTP does not equal memory, but it is still quite the viable candidate as a mechanism contributing to memory. Moreover, while hypothesis testing is not a democratic process, it is somewhat reassuring that in the vast majority of the cases so far there has been good agreement between effects of genetic manipulation on hippocampal LTP and their attendant effects on hippocampus-dependent memory.

Finally, I should note that we have so far dealt with only one specific variation of the block experiment, wherein genetic engineering is used as the experimental tool. The armamentarium is not limited to this single bullet—many drug infusion studies have addressed the question of whether blocking LTP blocks memory formation as well. Many of these have been reviewed in the excellent paper by Martin et al. (2), and in general the findings are quite supportive of a role for LTP in hippocampus-dependent memory. Later in the chapter, I will return to a couple of specific, and seminal, findings from Richard Morris's lab using this approach.

B. The Mimic Experiment

The mimic experiment is a fascinating thought experiment. In one specific variation, we would take an animal, place electrodes in its hippocampus, and use defined LTP-inducing stimulation patterns to place a memory in its CNS for the location of a food reward. Placing the stimulated but otherwise naive animal into a maze with a hidden food pellet would reveal that the animal could proceed directly to the reward - this despite never having received any direct experience concerning the location of the reward. A positive outcome in an experiment of this sort would be strongly supportive of the hypothesis of a role for LTP in memory formation.

The principal thing that we can say about the mimic experiment is that we are nowhere near being able to execute it at our present level of understanding and technology, at least for behaviors relevant to the hippocampus. Consider for a moment what the proper design and implementation of the experiment would entail. First, we would have to know the precise neuronal circuit underlying the behavior and how those circuits tied into the hippocampus. The details of the circuit would have to be known down to the level of the specific synapses involved in the hippocampus, if such is the case (perhaps the hippocampus never specifies synapses on a predetermined basis but rather allocates synaptic resources for computation on the fly). We would have to know that potentiating those synapses would translate into the formation of the relevant behavioral pattern—that is, that synaptic potentiation there would translate into the appropriate change in neuronal circuit properties somewhere down the line (it seems unlikely that the hippocampus is directly "in-line" as part of a circuit directly mediating a behavior). We would have to be able to induce LTP selectively at the relevant synapses with pinpoint accuracy. Overall, this is an experiment unlikely to be achievable anytime soon and perhaps never depending on how exactly the hippocampus operates.

Nevertheless, spending some time considering the mimic experiment is quite useful in my opinion for several reasons. It helps define what it is that we don't understand about the hippocampus—highlighting important areas of future pursuit. It also emphasizes the importance of being able to do this type of experiment in those memory paradigms where it is beginning to be practicable, such as in amygdala- and cerebellum-dependent memory paradigms (see references 10 and 11).

The mimic experiment also has import beyond the memory field, testing an important prediction of the general model driving all contemporary neuroscience. I am referring to the general theory that changes in the CNS mediate behavioral change, a point for which there substantial correlative evidence but scant direct evidence. Considering the mimic experiment also impinges upon the critical issues of the neural basis of consciousness—after all, the animal presumably is unable to distinguish real experience from artificially implanted experience. Thus, considering the mimic experiment brings us face to face with some of the key issues confronting modern neurobiology and philosophy. All in all, it is an obdurate topic of great significance.

Given the technical incapacity to execute the mimic experiment, what approaches of this sort are we left with? A very important variation of this idea that has been previously attempted by several pioneering labs and executed lately with great sophistication is the "occlusion" variant. The rationale here is that if we can go in and saturate LTP in the hippocampus, further naturally occurring LTP is not possible; thus, learning deficits should arise because of the lost capacity of the hippocampus to trigger its necessary synaptic plasticity.

Pioneering studies using this approach were executed by Bruce McNaughton and Carol Barnes's groups, which indicated that saturating hippocampal LTP produced learning deficits (12, 13). A more recent collaborative study by the Moser/Morris consortium has confirmed and solidified the original conclusions (14). The studies from both groups used a similar rationale and approach, although there were appreciable and significant differences in the technical execution of the studies and the types of data analysis brought to bear. Regardless, the overall conclusions were the same—saturating LTP at the perforant path inputs to the dentate gyrus leads to a loss of the capacity for hippocampus-dependent memory formation (Figure 2). These data strongly support the hypothesis that synaptic changes of a sort similar or identical to LTP are necessary for memory formation in vivo.

What distinguishes this experiment from the block experiment? At one level, they use the same approach, certainly. However, the appeal of the occlusion variation is that all changes are brought about by synaptic activity occurring endogenously. The only molecules brought into play are those normally and already in existence in the animal. The blocking manipulation utilizes the animal's own axons, neurotransmitters, and synapses. Overall the experiment has a much more physiologic "feel" to it.

Thus, while the mimic experiment is not now and may never be practical for hippocampus-dependent memory formation, its cousin the occlusion experiment has provided some satisfaction. The findings also are complementary to and consistent with the variety of inhibitor and gene-engineering experiments that we have already discussed. The conclusion from these disparate studies is that LTP or a very similar phenomenon is necessary for hippocampus-dependent memory formation. In the next section, we turn our attention to the issue of whether LTP or something similar in fact happens when the animal learns.

FIGURE 2 Saturating hippocampal LTP occludes Morris water maze learning. Four groups of animals were tested. Two groups were controls that received no LTP-inducing stimulation (nonstimulated and low frequency). Another group received LTP-inducing stimulation, but that did not saturate LTP (high-frequency test LTP > 10%). The final group, which exhibited saturated LTP (test LTP < 10%) had learning deficits. (A) Records of the search pattern of a representative animal from each group during the final spatial probe test (60 seconds). (B) Time spent inside a circle (radius of 35 cm) around the platform position (black bar) and in corresponding, equally large zones in the three other pool quadrants (diagonally striped, horizontally striped, and white bars) during the final spatial probe test (60 seconds). The dotted line indicates the chance level. Error bars indicate SEM. Reproduced from Moser et al. (14).

FIGURE 2 Saturating hippocampal LTP occludes Morris water maze learning. Four groups of animals were tested. Two groups were controls that received no LTP-inducing stimulation (nonstimulated and low frequency). Another group received LTP-inducing stimulation, but that did not saturate LTP (high-frequency test LTP > 10%). The final group, which exhibited saturated LTP (test LTP < 10%) had learning deficits. (A) Records of the search pattern of a representative animal from each group during the final spatial probe test (60 seconds). (B) Time spent inside a circle (radius of 35 cm) around the platform position (black bar) and in corresponding, equally large zones in the three other pool quadrants (diagonally striped, horizontally striped, and white bars) during the final spatial probe test (60 seconds). The dotted line indicates the chance level. Error bars indicate SEM. Reproduced from Moser et al. (14).

C. The Measure Experiment

While not as prohibitively difficult as the mimic experiment, the measure experiment is nevertheless a tough nut to crack. In the ideal experiment, we would be able to put stimulating and recording electrodes into the hippocampus, have the animal learn, and then directly measure an increased strength of synaptic connections in situ attendant with the memory formation. This would demonstrate that indeed LTP is occurring with memory formation in the behaving animal.

However, the practical constraints on executing this experiment are legion. What behavioral paradigm do you use? At what time point during learning do you expect LTP to happen? What hippocampal subregion do you record from? Are a significant fraction of the synapses in that region potentiated with a single learning experience? Might LTP formation be counterbalanced by simultaneous LTD elsewhere in the same population of cells, yielding no net change?

For these many reasons, in all probability, no one has yet successfully recorded endogenously triggered hip-pocampal LTP in the behaving animal (15). This does not mean that there have not been important related findings, however. As we discussed in Chapter 3, in vivo recordings from the hippocampus have revealed a number of important and relevant occurrences. For example, the formation of place field firing patterns among hippocampal pyramidal neurons implies altered hippocampal connectivity. Physiologic changes such as altered excitability occur with spatial learning, that are consistent with the induction and maintenance of LTP-like phenomena. Endogenous learning-related firing patterns are consistent with the occurrence of LTP-triggering high-frequency synaptic activity. Moser et al. also have directly observed synaptic potentiation in the dentate gyrus with learning (16, 17); however, the observable potentiation is fairly short-lived. Thus, the lack of capacity to observe LTP in the hippocampus with spatial learning has not been demonstrated directly, but a circumstantial case can be made that it happens. To the chagrin of us hippocampal types, the occurrence of LTP in amygdala-dependent learning has been nailed down (see Box 2).

If we cannot observe LTP directly using physiologic approaches, are there alternate ways to get at the "measure" question? A similar, molecular approach has been utilized in a variety of instances. The general rationale for these studies is that if one can identify specific molecular events associated with LTP induction and show that those same events occur in the behaving animal with hippocampus-dependent learning, then the inference can be made that LTP has indeed happened in vivo. It is worth noting that this is not the only motivation for the experiment, of course. Identifying molecular events associated with learning is also key to identifying the molecular basis of learning independent of the role of LTP in memory formation. For our purposes here, however, a large number of experiments in this line have worked out as quite supportive of the idea that hippocampal LTP does occur with spatial learning in the behaving animal.

For illustrative purposes I will review one example of this type of approach. I will focus on data from studies of ERK activation with learning because we have worked in this area in my laboratory (see reference 18). Moreover, as described extensively in prior chapters, a number of studies have shown that ERK is necessary for LTP induction. It also is clearly activated with LTP-inducing stimuli, allowing one to treat it as a biochemical marker of LTP induction.

Text continued on pg. 279

Was this article helpful?

0 0
Eliminating Stress and Anxiety From Your Life

Eliminating Stress and Anxiety From Your Life

It seems like you hear it all the time from nearly every one you know I'm SO stressed out!? Pressures abound in this world today. Those pressures cause stress and anxiety, and often we are ill-equipped to deal with those stressors that trigger anxiety and other feelings that can make us sick. Literally, sick.

Get My Free Ebook


Post a comment