One of the most intriguing questions in behavioral neuroscience concerns the manner in which the brain can modify its structure and ultimately its function throughout one's lifetime. Although the concept of changing brain structure has a long history, it has only been in recent years that the extensiveness of experience-dependent alterations in the brain has been realized. In addition, the definition of experience has expanded considerably to include drug and hormone experience, brain injury, and aging.
Neuronal plasticity during development is no surprise, although the role of neuronal activity and experience in directing neuronal organization and synaptic connectivity may seem surprising at first. Our brains are best categorized as changing, rather than static. This is certainly evident when one considers the malleability of the cerebral cortex based upon experience and damage-induced reorganization.
In our sensory and motor cortical areas, there are maps or representations of the sensory or motor surface. For example, we have the retinotopic map in the visual cortex, the somatotopic homunculus (little man) in the somatosensory cortex where each part of the body is represented, and we have the tonotopic map in the auditory cortex with each area of the primary cortex responding to a particular frequency of tone. A similar organization occurs in the motor cortex.
These maps change, depending upon experience (see [1]). For example, Merzenich and colleagues demonstrated that if an owl monkey were trained on a tactile discrimination task or a task that required them to keep one finger on a tactile stimulator until a tone was presented, that the size of the cortical representation of that finger would increase [2]. Similarly, the cortical representation of the ventral body surface is larger in nursing rat mothers than in non-nursing female rats [3]. There is evidence of similar remapping due to experience in humans, as well. In Germany, Elbert and colleagues [4] reported that the somatosensory cortical area devoted to the fingers in the left hand of musicians who played stringed instruments (e.g violins, guitars) were larger compared to controls. This was not observed in the area devoted to the right hand, which is involved in bowing, rather than fine motor and sensory articulation. Similarly, there is increased cortical representation for the index finger used in reading by blind Braille readers [5]. Such plasticity is not restricted to the somatosensory/motor cortical areas. Auditory discrimination training can lead to reorganization of the auditory cortex [6]. Interestingly, in these experiments, it was found that attention was required for the cortical changes to occur. It is believed that activation of neuromodulatory circuits such as Ach contribute to this.
But just as sensory experience can modify the cortical representation, so can damage-related alterations in afferentation.
For example, Merzenich demonstrated that amputation of a finger would lead to reorganization of the somatosensory map, with sensory input into adjacent fingers now activating the cortical area previously innervated by the amputated finger [7].
Some have suggested that this deafferentation-induced reorganization might explain the clinical phenomena of phantom limb.
Phantom limb was first described by Silas Mitchell in 1866, not in a scientific journal but in the Atlantic Monthly, as an anonymously written short story.
We discussed the phenomena during class.
The first question is, what gives rise to this phantom limb?
· Activation and irritation of axonal terminals in the stump due to scarring and neuromas. This seems the most intuitive explanation and consistent with this, often the phantom may be more vivid or resurrected if the stump is rubbed or hit. For years, this was the classic textbook explanation, but if it were true, then anesthetizing the stump or surgically removing neuromas would consistently eliminate the phantom.
· Melzack [8] suggested that the image of our body is encoded in an extensive neuromatrix, or network, that spans many brain regions that, when activated, serve as a neurosignature or memory of the missing limb. Melzack would argue that this is "pre-wired" or that we are born with this matrix, although it may be modified by experience. He supports this contention with cases of individuals with congenital limb loss who report phantom limb experiences.
· Ramachandran suggests that the remapping of the somatosensory cortex leads to phantom limb, which he discusses in the paper on your reading list.
What evidence is there of remapping?
Early evidence:
· Patrick Wall discovered that deafferentation could lead to changes in the spinal cord. Recordings from neurons in the dorsal column indicated receptive field changes shortly after partial denervation [9]. He postulated that these alterations in receptive fields were related to the uncovering of 'silent' synapses. Similar effects are believed to occur following damage to the retina-that change in receptive fields of neurons in the visual pathway "fill in" visual scotomas.
· Merzenich demonstrated that amputation of a finger in an adult owl monkey led to an invasion of 1-2 mm of adjacent areas in the cortical representation of the deafferented finger in primary somatosensory cortex [7]. It was long assumed that a few millimeters was the maximum reorganization that could take place. And this seemed to suggest that the reorganization may be due to sprouting, since axonal sprouting could account for expansions over several millimeters. But then along came…
· Pons [10] found more extensive reorganization of about 1-2 cm in adult macaque who had undergone deafferentation of an upper extremity 12 years earlier. This suggested that the reorganization was much more massive. The magnitude also suggested that the mechanism involved might be more than just sprouting of dendritic trees. It was noted that the cortical area previously activated by the arm was now activated by stimulation to the face.
· This led Ramachandran to use brain imaging (MEG) to examine the extent of remapping and begin a systematic investigation of the perceptual phenomenon of phantom limb referral to other body areas. These referrals were modality specific.
Others have found similar effects [11].
Ramachandran also argues that since individuals who have stroke and damage to the cortex may experience phantom experiences, that the phenomenon is due largely to cortical reorganization. He postulated that in cases of damage to the primary somatosensory cortex, reorganization occurs in the secondary somatosensory cortex.
How would remapping occur?
· Uncovering of subthreshold "silent" synapses
· Sprouting and formation of new synaptic connections
Although both may occur, there have been claims that phantom referral of a hand to the face can occur within 24 hours [12]. If this is true, it would indicate that silent synapses are uncovered. Also, given that one can have referral of a phantom on the contralateral limb suggests that existing synaptic connections are disinhibited.
Where does it occur?
Many argue that given the extent of reorganization, it is likely due to synaptic reorganization in the cortex [13]. Nevertheless, others argue that it may occur in the spinal column, or thalamus as well. It is simply not known.
But the remapping theory does not explain everything. For example, how does it explain the telescoping? Thus, it is likely that the phenomena is probably due to a combination of factors and integration of several experiences:
1-Stump neuromas
2-Remapping
3-Monitoring of motor commands to the limb
4-Internal "image" of one's body
5-Vivid somatic memories
Why is it often painful?
Again, it is not well understood and it could be related to events at many different levels.
Initially, the focus was on peripheral factors. It could be due to the neuromas and activation of cut nerves in the stump. There is some evidence that blockade or interruption of sympathetic supply to the stump temporarily alleviates phantom pain. And there appears to be a relationship between the severity of phantom limb pain and the pathology in the stump, as well as a co-occurrence in terms of frequency and intensity with pain in the residual limb. Local anesthetic can eliminate phantom limb pain in some cases (although it does not necessarily eliminate the phantom limb). However, efforts to eliminate the pain by cutting the sensory nerves of the stump often fail, and the pain would eventually return.
It is also possible that it is related to pain memories in the spinal cord (see work of Basbaum [14]). Wall has suggested that disinhibition of pain pathways in the spinal cord lead to phantom pain.
Others suggest that the pain is related to the cortical reorganization. For example, Flor [15] found that the extent of reorganization is correlated with the amount of pain experienced in the phantom limb (see also [16] and [17].
Consistent with this hypothesis, Birbaumer [18] injected local anesthetic into the region near the amputation, which resulted in relief of phantom limb pain in half of the subjects tested. In those subjects that had pain relief, there was a reduction in the amount of apparent cortical reorganization following the injection of local anesthetic.
Treatments:
We certainly did not go into an exhaustive discussion of treatments, but two rather novel treatments have been devised based on the theories behind phantom limbs and phantom limb pain.
· Mirror virtual reality box
· Since it has been shown that the phantom limb pain is more likely to occur if the limb was painful prior to amputation, many are examining whether injection of anesthetic prior to amputation might successfully reduce the severity of phantom pain. That is, if the phantom limb pain is related to "sensory memories", then altering those memories prior to limb removal might reduce the sensation of pain in the phantom related to the idea that some aspects of the phantom limb phenomena are related to "sensory memories." (see [19, 20]
So the phenomenon of phantom limb provides an interesting clinical scenario that may be the result of neuronal plasticity. How this reorganization relates to perception of body image is not well understood. Indeed, the mechanisms, sites of action, and controlling factors are also not well understood. But it does posit some interesting thoughts. What does it tell us of conscious experience? What is the functional significance of remapping? Do we not need a body to feel a body? It is a challenge to both medical management and psychology.
Is this only found in somatosensation?
Some have suggested that tinnitus, a condition characterized by a ringing in the ears following damage to the cochlea may be an example of an auditory phantom. Similarly, cortical reorganization in the primary auditory cortex has been found in cases of tinnitus [21].
Here's an interesting thought. If experience can alter the topographic maps in our cortical areas, could discrimination training alter phantom experiences? Several groups are currently investigating this possibility.
But what are the limits to this cortical plasticity?
What happens when an entire sensory modality is lost? What happens to the cortical area normally devoted to that sensory modality?
There is evidence of cross-modal reorganization in the developing brain. Sur and colleagues [22, 23] demonstrated that visually responsive neurons may "rewire" the primary auditory cortex. In their experiment, they re-routed the visual afferents from the retina to the auditory thalamus (which is connected to the primary auditory cortex) so that visual information was now directed to the auditory cortex. How did this influence the development of the primary auditory cortex?
The visually activated neurons induced the auditory cortex to take on a topographic organization that parallels the normal visual cortex. This new cortical area has visual orientation columns, and cells sensitive to stimulus direction, just like we saw in the primary visual cortex. That is, neurons that have a preference for a particular line orientation are found arranged in columns. They were later able to demonstrate that these cortical neurons were functional and processed visual information [23].
Okay, sure, but that is with quite a bit of experimenter manipulation. What if neurons are not redirected experimentally? Do other sensory modalities change to compensate for the loss? And what happens to the cortical areas normally devoted to processing this now missing modality?
Historically, there has been a lot of debate whether blind or deaf individuals have a perceptual advantage or disadvantage in processing information of their intact modalities. Some argue that other sensory modalities are enhanced, other argue that they may be hindered. And the data produced over the years have been mixed. But recent data suggest that there may indeed be changes in other sensory modalities.
For example, if the eyelids of a cat are sutured at a young age so that they are blind, when tested on a sound localization task, they perform with better precision. What's more, if you look at the neurons that are activated by auditory stimuli, they have expanded and expanded into the cortical areas normally used for visual processing!
In addition to auditory changes, there may also be somatosensory changes. Cats that are blinded early in life have hypertrophied whiskers and an increase in the cortical area devoted to the whiskers. Somatosensation expands into cortical area normally used for visual processing! See [24]
But what about humans?
There is some evidence that deaf individuals respond more quickly and more accurately than hearing subjects to visual stimuli. Similarly, there is some evidence that blind individuals are better at auditory discriminations and localization compared to controls [25] see also [26]. (Keep in mind that this does not mean that they have different thresholds of activation nor that non-blind individuals would not achieve these levels with additional training).
What about their brains? PET and ERP studies indicate that cortical areas normally activated during vision may be activated during auditory or tactile processing in blind individuals.
For example, in early blind individuals, PET studies illustrate that during an auditory localization task, the occipital areas are activated [27]. PET measurements also indicate that activation of the occipital cortex occurs in early-blind individuals during a tactile discrimination task and during Braille reading [28]. Interestingly, no activity was observed if the subjects were asked to sweep their fingers over a surface covered with Braille dots, suggesting that the activation requires attention and not just simple sensory activation. The idea that the visual cortex plays a role in Braille reading is further supported by Pascual-Leone's studies that illustrate that if the occipital cortex is stimulated with transcranial magnetic stimulation (to disrupt neuronal activity) then Braille reading is disrupted.
But most of these studies examined in early-blind individuals. It was commonly believed that the large-scale cross-modal reorganization of brain functions in sensory deprivation could only occur in an immature nervous system. Early animal studies suggested that this was true. There is now evidence to suggest that such changes may occur even in late-blind individuals [29]. Kujala reports changes in electrophysiological responses (event-related potentials, ERPs) occur in visual areas during an auditory discrimination task. Similarly, Buchel [30] reported activation of the striate cortex in the occipital lobe in individuals who became blind after puberty. Many are debating whether this is possible and some studies find differences only in early-blind individuals, so there is hot debate.
Probably the most extreme claim comes from Pascual-Leone, who claims that the visual cortex is activated and processes tactile information in individuals who have been blindfolded for one week, suggesting that this cross-modal plasticity may occur relatively rapidly. These data have not been published in a peer-reviewed journal, so we will just have to see……
There are still many questions that need to be answered. What exactly are the visual areas processing in blind individuals? Are they involved in higher sensory processing? In attention?
And what are the mechanisms for this kind of plasticity? It could be that multimodal thalamic nuclei or neurons from other cortical areas send projections to the striate cortex and that these are masked or even degenerate given normal vision. Or it could be that cortico-cortical connections re-route information. There is much that is currently not well understood.
Finally, this dynamicism of the cortex and our perceptions provides further evidence of how experience can change the brain. Identification of the controlling factors of this plasticity could lead to development of novel treatments for phantom limb, or tinnitus, or other perceptual disorders.
Despite Joseph Altman's suggestion that neurogenesis occurs in adult rats back in the 1960s [31], the idea that adult brains generate new neurons was met with skepticism. The focus on neuroplasticity was on the changes in dendrites and synapses related to experience and it was believed that the only way for the adult brain to compensate for damage was via new connections among surviving nerve cells.
Then, in the 1980s, Fernando Nottebohm of the Rockefeller University demonstrated neurogenesis in song birds. He found that neurogenesis occurred in brain areas involved with song, like the higher vocal center (an area homologous to the human cortex), during seasons when the bird engaged in singing. During seasons when the birds were not singing, the number of cells were reduced and neurogenesis did not occur. So, in the bird higher vocal center, there is an increase during the spring, when the adult birds acquired their songs and this fluctuation occurs annually. This increase is mediated by increased levels of testosterone, which increase based on length of daylight.
Song areas were not the only areas to involve neurogenesis in the bird. Chickadees store food in cache sites prior to winter so that they have sources during the wintery months. It was discovered that neurogenesis occurs in the hippocampus, which is important for spatial memory, during seasons that place high demands on the birds' memory systems.
There was also evidence of neurogenesis in rats, with neurogenesis occurring in the hippocampus and olfactory systems.
BUT, it was assumed that this did not occur in primates, because Pasko Rakic at Yale failed to find new brain neurons in adult rhesus monkeys. This was in keeping with the idea that neurogenesis is increasingly restricted throughout evolution, as the brain becomes more complex. And in many ways, this would make sense, because how would new neurons get integrated into the orderly flow of neuronal signals?
Peter Eriksson for Goteborg, Sweden had completed a sabbatical at the Salk here in San Diego, when he discovered that bromodeoxyuridine (BrdU) was given to some terminally ill patients with cancer of the tongue or larynx as part of a study to monitor tumor growth. After getting consent from the participants, he and Fred Gage investigated their brains and found labeled cells in the hippocampus of these patients, indicating that there were new neurons, specifically granule cells, in the dentate gyrus [32]. At the same time, Gould's and Rakic's groups both reported that nerve cell production takes place in the hippocampus of adult rhesus monkeys.
The next question is, do these new neurons work? Do they form functional connections? And what is the functional significance of neurogenesis?
Many new neurons die soon after being produced. But some will migrate and attain the morphological characteristics of neurons and glial cells, complete with neurite outgrowth. It has been demonstrated that these new cells express receptors, establish connections, and become functional.
What is the functional significance of this neurogenesis? There is a strong correlation between the size of the higher vocal center in songbirds and singing behavior. And a strong correlation between spatial memory demands and hippocampal neurogenesis in birds. Also, in rats, there is a relationship between neurogenesis and performance on learning tasks. Studies are now being conducted to manipulate neurogenesis and behavior to better understand the functional significance.
For example, Kempermann and colleagues [33] demonstrated that adult mice placed in an enriched environment grew 60 percent more new granule cells in the dentate gyrus than did controls. They also performed better on a learning task. It also had an effect in very old mice, which have a lower basal rates of neuronal production than younger adults. This suggests that experience and activity may enhance brain plasticity during aging, providing evidence for the old adage, "use it or lose it."
Factors that control neurogenesis:
As mentioned by Caroline, many factors may affect each of the steps involved in the generation, incorporation, and survival of new neurons. Consider that an increase in cell proliferation will yield a net rise in new neurons if the rates of survival and differentiation remain constant, but the neuronal number may not rise if the survival and differentiation rates are altered. So, if an investigator finds a greater number of labeled neurons, it could be due to alterations at each of those steps, indicating that the processes that regulate neurogenesis are complex and multi-level.
These factors include:
Experiences:
· Enriched environment increases
o Promotion of cell survival
· Running increases
o Higher rate of cell division
· Learning tasks increase
· Stress reduces
Neurochemicals
· Growth factors increase
o Epidermal growth factor favors differentiation into glial cells, whereas fibroblasts growth factors promote neuronal production
o Neurotrophic factors probably mediate the effects of some of these other factors, like hormones
· Serotonin increases
· Estrogen increases
o We will be returning to the effects of estrogen on plasticity and on cognitive functioning later in the course
· Glucocorticoids reduce
· NMDA receptor antagonists increase (this will make more sense later in the course)
· NMDA receptor agonists reduce
Damage:
· Epileptic seizures increase
· Stroke increase
Genetics
Concerns with the research
Although it has become increasingly accepted that the adult brain generates new neurons, it is important that researchers are rigorous in their methods. Bromodeoxyuridine (BrdU) will label not only DNA replication, but also DNA repair. So are the labeled cells really new? This can be very problematic if the investigator is studying the effects of brain damage on neurogenesis. Other methodologies that are important to note are the methods by which they estimate the number of cells and the amount of BrdU labeling.
What are the implications of neurogenesis?
If we trace the cues and internal events that control neurogenesis, perhaps new therapies could be devised. If one could stimulate stem cells to migrate into areas where they usually do not go and mature into specific kinds of nerve cells, it would preclude the difficulty of rejection of transplanted material. Could they be used to replace cells lost to stroke or neurodegenerative diseases?
The hypothesis that mood disorders are related to neurochemical alterations has recently been expanded. Evidence has accumulated that structural changes in the limbic system may be related to major depression. Duman [34] suggests that major depression and other affective disorders could result in loss of neuroplasticity. The loss of neural plasticity and synaptic interactions could contribute to a loss of neurotrophic action and thus cell survival and synaptic efficacy, a vicious cycle.
Keep in mind that stress reduces neurogenesis and 5-HT increases neurogenesis. Incidentally, antidepressants like fluoxetine, d-fenfluramine, and electroconvulsive shock therapy all lead to an increase in neurogenesis. We will be talking in greater detail of the role of neuroplasticity in mood disorders in a few weeks.
1. Kaas, J.H., Plasticity of sensory and motor maps in adult mammals. Annual Review of Neuroscience, 1991. 14: p. 137-167.
2. Recanzone, G.H., et al., Topographic reorganization of the hand representation in cortical area 3b of owl monkeys trained in a frequency discrimination task. Journal of Neurophysiology, 1992. 67: p. 1031-1056.
3. Xerri, C., J.M. Stern, and M.M. Merzenich, Alterations of the cortical representation of the rat ventrum induced by nursing behavior. Journal of Neuroscience, 1994. 14: p. 1710-1721.
4. Elbert, T., et al., Increased cortical representation of the fingers of the left hand in string players. Science, 1995. 270: p. 305-307.
5. Pascual-Leon, A. and F. Torres, Plasticity of th sensorimotor cortex representation of the reading finger of braille readers. Brain, 1993. 116: p. 39-52.
6. Recanzone, G.H., C.E. Schreiner, and M.M. Merzenich, Plasticity in the frequency representation of primary auditory cortex following discrimination training in adult owl monkeys. Journal of Neuroscience, 1993. 13: p. 87-103.
7. Merzenich, M.M., et al., Somatosensory cortical map changes following digital amputation in adult monkey. Journal of Comparative Neurology, 1984. 224: p. 591-605.
8. Melzack, R., Phantom limbs and the concept of a neuromatrix. Trends in Neuroscience, 1990. 13: p. 88-92.
9. Wall, P.D., The presence of ineffective synapses and the circumstances which unmask them. Philosophical Transactions of the Royal Society of London B, 1977. 278: p. 361-372.
10. Pons, T.P., et al., Massive cortical reorganiation after sensory deafferentation in adult macaques. Science, 1991. 252: p. 1857-1860.
11. Halligan, P.W., et al., Thumb in cheek? Sensory reorganization and perceptual plasticity after limb amputation. Neuroreport, 1993. 4: p. 233-236.
12. Borsook, B., et al., Acute plasticity in the human somatosensory cortex follwoing amputation. Society for Neuroscience Abstracts, 1997. 23: p. 438.
13. Kew, J.J., et al., Abnormal access of axial vibrotactile input to deafferentaed somatosensory cortex in human upper limb amputees. Journal of Neurophysiology, 1997. 77: p. 2753-2764.
14. Basbaum, A.I., Memories of pain. Scientific American Medicine, 1996: p. 22-31.
15. Flor, H., et al., Phantom limb pain as a perceptual correlate of cortical reorganization following arm amputation. Nature, 1995. 375: p. 482-484.
16. Grusser, S.M., et al., The relationship of perceptual phenomena and cortical reorganization in upper extremity amputees. Neuroscience, 2001. 102: p. 263-272.
17. Karl, A., et al., Reorganization of motor and somatosensory cortex in upper extremity amputees with phantom limb pain. Journal of Neuroscience, 2001. 21: p. 3609-3618.
18. Birbaumer, N., et al., Effects of regional anesthesia on phantom limb pain are mirrored in changes in cortical reorganization. Journal of Neuroscience, 1997. 17: p. 5503-5508.
19. Schug, S.A., et al., Pre-emptive epidural analgeisa may prevent phatom limb pain. Regional Anesthesia, 1995. 20: p. 256.
20. Danshaw, C.B., An anesthetic approach to amputtion and pain syndromes. Physical Medicine and Rehabilitation Clinics of North America, 2000. 11: p. 553-557.
21. Mulnickel, W., et al., Reorganization of auditory cortex in tinnitus. Proceedings of the National Academy of the Sciences, 1998. 95: p. 10340-10343.
22. Sharma, J., A. Angelucci, and M. Sur, Induction of visual orientation modules in auditory cortex. Nature, 2000. 404: p. 841-847.
23. von Melchner, L., S.L. Pallas, and M. Sur, Visual behvaioru mediated by retinal projections directed to the auditory pathway. Nature, 2000. 404: p. 871-875.
24. Rauschecker, J.P., Compensatory plasticity and sensory substitution in the cerebral cortex. Trends in Neuroscience, 1995. 18: p. 36-43.
25. Weeks, R., et al., A positron emission tomographic study of auditory localization in the congenitally blind. Journal of Neuroscience, 2000. 20: p. 2664-2672.
26. Kujala, T., K. Alho, and R. Naatanen, Cross-modal reorganization of human cortical functions. Trends in Neuroscience, 2000. 23: p. 115-120.
27. De Volder, A.G., et al., Changes in occipital cortex activity in early blind humans using sensory substitution device. Brain Research, 1999. 826: p. 128-134.
28. Sadato, N., et al., Activation of the primary visual cortex by Braille reading in blind subjects. Nature, 1996. 380: p. 526-530.
29. Kujala, T., et al., Electrophysiological evidence for cross-modal plasticity in humans with early- and late-onset blindness. Psychophysiology, 1997. 34: p. 213-216.
30. Buchel, C., et al., Different activation patterns in the visual cortex of late and congenitally blind subjects. Brain, 1998. 121: p. 409-419.
31. Altman, J. and G.D. Das, Autoradiographicand histological evidence o postnatal hippocampal neurogenesis in rats. Journal of Comparative Neurology, 1965. 124: p. 319-335.
32. Eriksson, P.S., et al., Neurogenesis in the adult human hippocampus. Nature Medicine, 1998. 4(11): p. 1313-1317.
33. Kempermann, G., H.G. Kuhn, and F.H. Gage, More hippocampal neruons in adult mice living in an enriched environment. Nature, 1997. 386: p. 493-495.
34. Duman, R.S., et al., Neuronal plasticity and survival in mood disorder. Biological Psychiatry, 2000. 48: p. 732-739.