The Problem Of Detecting Brain Plasticity

Important modifications in brain circuits can be quite difficult to detect. Small changes in synaptic strength, when distributed across many synapses and neurons, can be difficult to measure and quantify. Thus, plasticity is often demonstrated under rather unnatural conditions with the assumption that similar but less obvious changes occur under more natural conditions. One form of synaptic plasticity, long-term potentiation or LTP, for example, is typically studied in living slices of brain tissue rather than in the intact brain, so that electrodes can be properly placed and variables better manipulated and controlled. Electrical stimulation of axons replaces activity induced by natural stimuli in these experiments. The tissue itself, the hippocampus, is picked for study, not only because of its critical role in the early stages of learning, but because it has a highly laminar type of tissue organization that permits such electrical stimulation. Such studies have been more difficult in the neocortex, where neurons also demonstrate LTP.

In a similar manner, we have chosen the large, orderly representation of the hand in primary somatosensory cortex (S1) of monkeys for many of our studies of cortical plasticity. The hand is represented in S1 or area 3b (using the numerical scheme of Brodmann, 1909), so that separate territories exist for each digit and pad of the palm. The digit territories proceed from digits 1-5 in a lateromedial sequence in area 3b, and each digit is represented from tip to base in a rostrocaudal sequence. The pads of the palm are arranged in order in cortex caudal to the digits. In properly prepared sections of the cortex through this representation cut parallel to the surface of the brain, the orderly arrangement of the representations of the digits and pads can even be seen (Jain et al., 1998). The neural fibers associated with each digit in cortex form an elongated oval, separated from each other by a narrow fiber-poor septum. This morphological map is laid down during early brain development, and it does not change in adulthood.

Given this orderly map of the hand (and other body parts) in the cortex, we can ask the following question: "Does it change as a result of sensory experience or deprivation?" The organization of the normal map can be determined in great detail by recording receptive fields for neurons at many places in the map (hundreds of places) with penetrating microelectrodes, and using this information to reconstruct the map. All of the neurons recorded in the territory of digit 3, for example, will have receptive fields centered on digit 3, and even have most or all of each receptive field on digit 3. After any manipulation that might alter the organization of the map, microlectrode recordings can be used to characterize the potentially altered map, and the results can be compared with previously obtained maps from normal or inexperienced monkeys, a map from the same monkey before the manipulation, and even the unchanged morphological map from the manipulated monkey. Prolonged, intense stimulation of digit 3, for example, might enlarge the territory of digit 3 at the expense of other digits (see Jenkins et al., 1990), so that neurons over a larger than normal extent of cortex would have receptive fields centered on digit 3, and this might be detected by comparing the sizes of territories for digit 3 in normal and stimulated monkeys. Other changes such as reductions or increases in receptive field sizes might also be considered, but most studies have been concerned with detecting changes in territory. Given the ability that we now have to non-invasively image evoked activity patterns in the human brain, large changes in cortical territories can even be demonstrated in humans. For example, there is evidence from functional brain imaging studies that the cortical representations of the digits of the hand used in playing string instruments are larger in skilled musicians than in non-players (Elbert et al., 1995).

Of course, there are other representations in the brain besides S1 that can be studied. Somatosensory afferents enter the spinal cord and lower brainstem to terminate on an elongated sheet of neurons in the dorsal horn of the spinal cord and its extension into the brainstem. These neurons form an elongated map of tactile and other inputs. Branches of afferents also ascend in the dorsal fiber columns of the spinal cord or travel in the trigeminal nerve of the face to terminate in the dorsal-column-trigeminal nucleus complex in the brainstem, where a second map of skin receptors occurs. Neurons in the complex then project to the opposite side of the brainstem where they ascend to the ventroposterior nucleus on the opposite side of the upper brainstem forming a third representation of skin receptors. These subcortical representations of the body were studied long ago for alterations due to nerve damage, with plastic changes and somatotopic reorganizations often being reported (see Snow and Wilson, 1991 for review). However, the results were not always very convincing, since the small brainstem structures were difficult to map accurately, and reported changes were often of proportions that were close to the error of measurement. Because more obvious results can be obtained in the larger, more accessible cortical maps, recent investigators have concentrated on cortex. In addition, cortex reflects changes relayed from brainstem structures, and cortex may be more plastic than brainstem structures.

Another possibility in studies of plasticity is to evaluate other cortical maps for changes. Area 3b is the homolog in monkeys and humans of S1 in rats and cats (Kaas, 1983), but there are other cortical representations. Just rostral to the area 3b representation of tactile receptors, area 3a represents muscle-spindle receptors. In addition, strip-like areas 1 and 2, just caudal to area 3b, represent tactile receptors (area 1) or a mixture of tactile and muscle-spindle receptors (area 2). These fields project to other representations, including the second somatosensory area, S2, and the parietal ventral somatosensory area, PV (see Kaas, 1993 for review). Some of these representations, such as areas 3a and 1, and S2, have been included in studies of plasticity, but they have neurons with larger, and often less easily defined receptive fields (e.g., 3a, 2). These representations may also be smaller in size (e.g., S2, PV). Thus, convincing data on plasticity and reorganization are more difficult to obtain, and so the number of studies on these areas has been limited. Investigators have also considered auditory, visual, and motor systems, with the most easily explored primary representation more commonly considered. Some of these studies, especially those on motor cortex, are reviewed here.

0 0

Post a comment