| |
|
| ||
       |
 |
 |
|
|
 |
| This Document | ||
 |
SummaryPlus | |
 |
Full Text + Links | |
|
PDF (277 K) | |
|
| ||
| Actions | ||
|
Cited By | |
|
Save as Citation Alert | |
|
E-mail Article | |
|
Export Citation | |
Cite or link using doi
Physiological consequences of morphologically detectable synaptic plasticity:
potential uses for examining recovery following damage
Tammy L. Ivancoa
and William T. Greenough
, 
, a,
b
a Beckman Institute, University of
Illinois, 405 N. Mathews Avenue, Urbana, IL 61801, USA
b Departments of Psychology, Psychiatry, and Cell and
Structural Biology, University of Illinois, Urbana/Champaign, IL, USA
Accepted 4 January 2000. Available online 25 February 2000.
A growing literature indicates that brain structure is modified in various ways with experience. In this paper we briefly survey evidence that the brain retains the capacity to modify its organization in response to demands, including demands resulting from learning, throughout the lifetime. We attempt to address whether these experience-induced changes are accompanied by physiological changes that indicate a functional reorganization of the brain. The kinds of morphological changes that have been observed following brain injury appear to be very similar to those seen after learning. The similarity suggests that many of the basic mechanisms of synaptic change in the brain may be utilized for both functions. This suggests that we can take advantage of some of the methods used to test the changes in physiology with behavioral manipulations to examine the damaged brain. We advocate utilizing electrophysiological techniques to measure functional recovery from brain injury as these may be useful in evaluating both spontaneous recovery from damage and the therapeutic benefits of training, or other therapies.
Author Keywords: Electrophysiology; Complex environment
(EC); Learning; Lesions; Long-term potentiation (LTP); Memory; Neurogenesis;
Synaptogenesis; Fragile Mental Retardation Protein (FMRP)
An exciting question in neurobiology revolves around what is needed within the mammalian brain to retain new information that is learned. Although the brain is able to undergo many forms of plastic change, considerable evidence indicate that growth or change in existing structure, particularly synapse number and structure, predominantly underlies learning. Recovery from damage may depend on similar changes as well. If morphological changes are involved in memory, this should be reflected in the physiology of the pathways that they mediate. This paper describes some of the morphological changes brought about by experience and discusses whether these experience-induced changes are accompanied by physiological changes that indicate that there has been a functional reorganization of the brain. The ability to measure functional reorganization utilizing simple physiological techniques may aid in evaluating recovery from damage in animal models, and potentially, in clinical situations as well.
Until recently it was thought that the adult human brain did not undergo neurogenesis. It is still presumed that mature neurons in the adult brain have lost most of the ability to proliferate and migrate (Shihabuddin et al., 1999). Proliferating cells can be identified using bromodeoxyuridine (BrdU), which labels cellular DNA during the synthesis, or S-phase. The human hippocampus has been shown to generate new cells that survived and differentiated into phenotypic neurons throughout life ( Eriksson et al., 1998). The rate of proliferation was not very high, nor have the new cells been shown to be functional.
The neurogenesis may be a result of locally proliferating neurons, but new cells also arise from the subependymal zone. The lateral ventricle subependyma in the adult mammalian forebrain contains multipotent neural stem cells and progenitor cells. The post-mitotic fate of a population of constitutively proliferating cells from the subependyma has been shown to be cell death (Morshead and van der Kooy, 1992) or neuronal differentiation after migration to the olfactory bulb ( Lois and Alvarez-Buylla, 1994). A recent study indicated that cell proliferation occurred along the migratory pathway, as well ( Craig et al., 1999). In vivo and in vitro, growth factors promote the migration of subependyma cells away from the lateral ventricles into adjacent tissues where they differentiate into neurons and glia (e.g. Craig et al., 1996). The biological significance and biological cost of cell genesis in the adult brain remain to be determined.
Recent evidence suggests that neurogenesis may play a role in learning. To examine the role of behavioral experience on proliferation in the hippocampus, Kempermann et al. (1997) placed mice in special environmentally complex (EC) cages or in standard laboratory cages. They found that mice reared in the complex housing condition for 40 days did not show an increase in BrdU-labeled neurons above that seen in animals from the standard housing condition one day after the final BrdU injections, indicating that there was no effect on neuronal proliferation. The environmental complexity did promote the survival of the proliferating cells, as the EC mice had significantly more BrdU-labeled cells remaining one month later. In the senescent dentate gyrus (20 months), EC housing had an even greater influence on the survival of BrdU-labeled neurons, than in younger (8 months) mice (Kempermann et al., 1998). Nilsson et al. (1999) also found that rats reared in EC housing did not show neurogenesis at a level above that seen in controls. There was a marked increase in the number of surviving cells one month later in the animals that had been placed in EC housing. Voluntary running in mice produced an enhanced proliferation of new cells in the dentate gyrus, such that the net number of cells added was similar to that seen in mice in an EC environment, whereas maze training (learning) did not enhance the survival of new neurons ( van Praag et al., 1998). The learning, however, may have been poorly timed to affect cell survival measures ( Greenough et al., 1999). In contrast to the maze training result, trace eyeblink conditioning, which requires hippocampal function, enhanced dentate cell survival, whereas delay eyeblink conditioning, which is unaffected by hippocampal ablation, did not ( Gould et al., 1998). The new neurons that are being generated within the hippocampus may have an increased sensitivity to a survival-promoting effect of learning ( Greenough et al., 1999). Evidence of neuronal proliferation in human hippocampus ( van Praag et al., 1998) and in cerebral cortex of monkeys ( Gould et al., 1998) suggests continued attention to possible roles of neurogenesis in learning is warranted.
Evidence from the insect literature implies a smaller role, if any, of neurogenesis in invertebrate learning. The mushroom bodies are the insect brain structures most often associated with learning. For example, these structures exhibit structural plasticity in drosophila (Heisenberg et al., 1995) and honey bees ( Fahrbach et al., 1995) in response to changes in the environment and demands of the environment, respectively. Fahrbach et al. (1995), however, have shown in a comprehensive study that neurogenesis in the adult honey bee is a rare event. Bees were fed or injected with BrdU or treated in vitro with BrdU, and proliferating cells were never seen. Thus, neurogenesis cannot explain the complex behavioral adaptations that the bee makes.
The subependymal zone increases metabolic activity following cortical injury, suggesting that the subependymal cells are proliferating at a higher rate in response to the injury (Valla et al., 1999). Lesions in the rostral migratory stream path of subependymal cells (within the prefrontal cortex) impede the migration to the olfactory bulb, inducing accumulation of progenitors within the lesion cavity that subsequently differentiate into phenotypic neurons ( Alonso et al., 1999). Kolb et al. (1998) have shown that the filling in of such lesion cavities with neurons in the medial prefrontal cortex has the potential to produce functional recovery if the lesion occurs in young animals. Similar lesions in the posterior parietal cortex, which is outside of the migratory pathway, do not fill in and there is less functional recovery ( Kolb and Cioe, 1998).
Although cells proliferate in response to brain injury, the identity of differentiated cells appears to be dependent, at least in part, on the type and loci of the damage. Damage that targets the hippocampus directly, such as seizure activity (Holmes et al., 1998) and ischemia ( Liu et al., 1998), results in BrdU incorporation into newborn granule cells identified with neuron-specific markers, whereas indirect damage, such as fimbria–fornix lesions ( Weinstein et al., 1996), increases the proliferative rate of cells that are identified with astrocyte-specific markers.
The evidence indicates that proliferation of new functional neurons may not always be the response of the mammalian brain to damage. Furthermore, the generation of functional neurons may not always be a good thing for functional recovery following damage. Parent et al. (1997) investigated the effects of seizure activity on the network reorganization within the dentate gyrus. Following induction of status epilepticus there was an increase in proliferation and neuronal markers indicated that the mitotically active cells differentiated into neurons in ectopic dentate locations. These new cells were shown to project aberrant axons into CA3 and the dentate inner molecular layer. The authors suggest that these results are indicative of a newborn population of neurons remodeling connections within the hippocampus, and that this may also be the basis of aberrant reorganization seen in human patients with temporal lobe epilepsy (e.g. Sutula et al., 1989). Two interesting speculations may be warranted here. First, one wonders what the effect of anti-mitotic drugs currently used in (e.g. methotrexate, 5-fluorouracil), and being developed for (e.g. vinblastine, benzimidazoles), central nervous system malignancy chemotherapy are on proliferation of new neurons and whether cognitive deficiencies might be observed during such treatment regimens. Second, one wonders whether anti-mitotic drugs might be valuable in the management of certain types of temporal lobe epilepsy.
Although neurogenesis may be involved in learning in some cases, it is almost certainly not the only mechanism involved in either learning or recovery from damage. The idea that existing neurons that are activated would grow has been around for some time. Cajal (1894) and Tanzi (1893) first suggested and later, in 1949, Hebb proposed in a detailed hypothesis that the synapse that was the critical site for plastic change. Hebb's postulate, that increments in synaptic efficacy occur during learning when firing of one neuron repeatedly produces firing in another neuron to which it is connected, has been used as a construct for many models of synaptic plasticity. Neurons show various kinds of plasticity that appear to follow Hebbian rules. Only a brief sample of such plasticity is reviewed below.
Although the mushroom bodies in arthropods are known to be involved in learning, the effects of EC on these organisms is not well understood. Heisenberg et al. (1995) investigated the effects of housing on the Kenyon cell fibers within the mushroom bodies of Drosophila melanogaster. Embryos were placed in vials containing 60 or 200 embryos and mushroom bodies were investigated one hour after hatching. The flies grown in high-density conditions had larger mushroom bodies than siblings grown in low-density conditions. Female flies living alone in food vials had smaller lobulae than females that were housed for an equal amount of time in larger, populated flight cages. The authors suggest that changes in the size and number of terminal branches, swelling, or spine formation and retraction are likely to account for the differences observed.
Turner and Greenough (1985) compared the synapse per neuron ratio in rats reared in a complex environment (EC) with animals reared in social conditions (SC), which was defined as animals housed in pairs in standard laboratory cages, and individually housed controls (IC), which were animals housed individually in standard laboratory cages. The EC condition is one in which animals are housed in a group within a large cage. The cage is filled with toys, which are changed daily, for a visually and motorically stimulating environment. The highest synapse per neuron ratio was found in animals that were housed in the EC environment. This finding was consistent with predictions from previous Golgi studies (e.g.