Pkc

-7 ± 5 (8)

*25 ±15 (11)

■5 ±11 (7)

11 ± 23 (7)

25 + 11 (12)

10 ±11 (10)

a-CaMKII

■2 ±7 (8)

7 ±11 (11)

8 ± 10 (7)

-3 ± 16 (7)

13 ±23 (14)

47 ± 20 (10)

FIGURE 15 Hippocampal protein kinase activation in fear-conditioning. Two different fear-conditioning paradigms were used, cued and cued plus contextual (see text). Protein kinase activation was assessed using phospho-selective antisera to measure kinase phosphorylation. Percent change in phosphorylation from control for each protein kinase is shown. Number of animals used are in parentheses. All protein phosphorylation measurements were normalized to corresponding protein kinase amounts. Shaded boxes are statistically significant. The asterisk (*) denotes a nonsignificant (p = .1) increase in autophosphorylated Protein Kinase C (PKC) 1 hour after contextual conditioning. Hippocampal CaMKII was selectively activated with cued-plus-contextual fear conditioning. Data courtesy of Coleen Atkins (10).

FIGURE 15 Hippocampal protein kinase activation in fear-conditioning. Two different fear-conditioning paradigms were used, cued and cued plus contextual (see text). Protein kinase activation was assessed using phospho-selective antisera to measure kinase phosphorylation. Percent change in phosphorylation from control for each protein kinase is shown. Number of animals used are in parentheses. All protein phosphorylation measurements were normalized to corresponding protein kinase amounts. Shaded boxes are statistically significant. The asterisk (*) denotes a nonsignificant (p = .1) increase in autophosphorylated Protein Kinase C (PKC) 1 hour after contextual conditioning. Hippocampal CaMKII was selectively activated with cued-plus-contextual fear conditioning. Data courtesy of Coleen Atkins (10).

have surprises in store for us as we move from behavioral studies, where we select a particular behavioral output to measure and in essence use the animal as a filter, to molecular studies where molecular readouts of a wide variety of different events are likely to occur.

Don't Forget Synapses are Made of Molecules

Finally, another variation of the measure experiment in learning is that there have been anatomical changes in various CNS regions, especially in the cerebral cortex, identified after raising animals in enriched environments. This approach was pioneered by Bill Greenough and has been extensively studied in his laboratory. A significant part of these anatomical changes in the fine structure of cortical neurons is presumed to be the result of (or at least influenced by) learning and memory during the growth of the animal. Among the changes identified are changes in synaptic density and dendritic morphology. Even though I am not going to elaborate further on these anatomical studies for now, I do wish to make one final point. It is important to bear in mind that these anatomical changes are of necessity subserved by molecular changes. The molecular events involved are not limited to the induction of these changes—their maintenance must also be a manifestation of molecular alterations. I will illustrate my point by reducing it to the simplest example. Suppose that there are 20% more synapses in an enriched environment brain due to a variety of learning events. If nothing else, this will require a steady-state increase in the rate of synthesis of all the proteins comprising those synapses because proteins are constantly turning over in the cell. This may seem like a simple feat to accomplish, but somewhere in those cells there must be some molecular mechanism(s) maintaining that increased net rate of synthesis. Once again, this adds a layer of molecular complexity to what otherwise appears to be a straightforward application of behavioral approaches in a "measure" experiment.

Using Behavioral Paradigms as an Assay of Learning in Block Experiments

Block experiments are by far the most common use of rodent behavioral assessment paradigms. In these experiments, the behavioral paradigms we have reviewed are used to assess whether an animal has a learning or memory deficit when a particular process is blocked. The behavioral read-out of the learned behavior is used to assess whether the animal has learned. A "memory" variation commonly used is to assess animals behaviorally to determine if an experimental treatment causes a faster decrement of learned behavior over time. The three most common examples of experimental manipulations in the application of the block approach to behavior, roughly in historical order, are anatomical lesions, drug infusion studies, and, more recently, gene manipulation experiments. What is of interest in these types of experiments is determining what structures or molecules are necessary for an animal to learn, remember, and recall a learned event.

We will be returning to a great many specific examples of these types of experiments later, experiments that have implicated specific molecules and categories of molecules in learning and memory and synaptic plasticity. Thus, for the present, I will discuss these types of experiments only in general terms. Of course, there are a great many caveats in interpreting these types of experiments, and we will focus our discussion here on five general considerations that must be kept in mind. I review them here in general terms so that I can avoid repeating them throughout the book when we discuss specific experiments. They also lead us to the final section of the chapter where we will talk about a variety of behavioral assessments in rodents that are used as an adjunct to experiments designed specifically to look for learning and memory deficits (see references 11 and 12).

The first consideration in behavioral "block" experiments is that there may have been nonspecific effects of the manipulation, as is always the case with inhibitors or lesions of any sort. The drug may not be specific for the molecule of interest, the knockout animal may have not developed a normal CNS, or the anatomical lesion may have destroyed fibers of passage connected to distal brain regions. This limitation is practical in nature and, in general, has received a great degree of attention in the literature. The specifics also vary greatly depending on the specific experiment under consideration, so we will not address this point further for the present.

The second limitation is largely conceptual. In these behavioral learning experiments, you are training an animal using environmental signals and measuring at some later time point a complex behavioral read-out. Many things are occurring during the training, learning, memorizing, recalling, and execution of the read-out. It is fundamentally difficult with the basic experimental design employed to distinguish among effects on learning, memory, or recall. Imagine the simplest case where an animal has a molecular deficit throughout the experiment—it is clear that no conclusion can be drawn concerning whether the animal has a deficit in learning, memory, or recall.

Two basic variations of the experiment are used to try to begin to distinguish among these possibilities. A transient inactivation experiment, where a structure or molecule is inhibited for a limited period of time, allows one to begin to parse effects on learning/memory versus recall, for example. However, it is still difficult with the transient inactivation design to distinguish between effects on learning versus effects on early memory consolidation, for example. This brings us to the second variation, where memory is assessed at short time points versus long time points. If an animal with a molecular or anatomical deficit is able to perform normally at short time periods after training and has a selective deficit at longer time periods, this implies that learning has occurred but that there is a loss of longer-term memory. The principal limitation to this approach is that it assumes that the learning mechanisms for short-term memory are identical to the mechanisms used for long-term memory.

A third consideration that must be kept in mind in interpreting block experiments is that compensation may have occurred, acutely or chronically. For example, the loss of a brain structure or molecule may force the CNS to utilize an ancillary mechanism that is capable of doing the job, but that normally is never brought into play. From a hypothesis-testing perspective, this leads to a false negative result—we conclude that molecule or structure X is not necessary but in fact it is necessary under normal circumstances. The enormous plasticity of the CNS in general makes this a particularly bothersome concern. I will use Lashley's classic lesioning experiments to illustrate this point, precisely because they have been so important in shaping modern thinking about memory. Lashley trained rats in mazes and made post-training cortical lesions in order to try to localize the maze memory trace anatomically. Lashley observed that, by and large, no single lesion could erase a memory, but rather that maze performance declined in relation to the overall extent of cortical lesioning. Thus, Lashley concluded that memories likely are "distributed" throughout the cortex, and that there was no discrete memory trace. The caveat is that there may have been multiple, redundant, memory traces, and that only when the last one was destroyed was there an appreciable decline in maze perfomance. Unfortunately, very little can be done in the way of control experiments to address this general limitation clearly— for the most part, it must simply be left as a caveat to the interpretation.

The fourth and fifth overall considerations in interpreting block behavioral experiments are at least well-defined enough that control experiments can be brought to bear—these are performance deficits and sensory processing deficits. In hypothesis testing terms, both of these limitations lead to potential false positive results. In many of the types of memory paradigms we have been discussing, fairly sophisticated control experiments can be executed to rule out these limitations. However, the biology has to be working to the advantage of the experiments, and, of course, we have no control over that. One of my favorite examples is finding a selective deficit in contextual versus cued fear conditioning. If a lesioned animal performs normally in cued fear conditioning but has a deficit in contextual fear conditioning, one can make a reasonable interpretation that it can feel the foot shock (thus no sensory deficit for foot shock) as well as exhibit freezing behavior (no performance deficit in freezing). Another favorite is the selective effects on long-term versus short-term memory—if short-term memory is intact then it is reasonable to conclude that the lesioned animal both perceived the environmental stimuli and is capable of performing the necessary behavioral read-out of memory.

In many cases, however, the biology does not work to your advantage and these two types of sophisticated control experiments cannot be used. In that case then, more indirect measures must be employed to bolster the case that the experimental manipulation has not led to general deficits in overall health, motivation, perception, or motor performance. In the following section, I will briefly describe a number of behavioral assays that typically are used as controls in this situation. It's important to keep in mind that many of these behaviors are important behavioral tests in their own right, but that, for our purposes, I am simply presenting them in a context of their use as general control experiments for learning and memory assessments.

Advanced Memory Techniques

Advanced Memory Techniques

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.

Get My Free Ebook


Post a comment