Reshaping the cortical motor map by unmasking latent intracortical connections.

The primary motor cortex (MI) contains a map organized so that contralateral limb or facial movements are elicited by electrical stimulation within separate medial to lateral MI regions. Within hours of a peripheral nerve transection in adult rats, movements represented in neighboring MI areas are evoked from the cortical territory of the affected body part. One potential mechanism for reorganization is that adjacent cortical regions expand when preexisting lateral excitatory connections are unmasked by decreased intracortical inhibition. During pharmacological blockade of cortical inhibition in one part of the MI representation, movements of neighboring representations were evoked by stimulation in adjacent MI areas. These results suggest that intracortical connections form a substrate for reorganization of cortical maps and that inhibitory circuits are critically placed to maintain or readjust the form of cortical motor representations.

Learning-induced LTP in neocortex.

The hypothesis that learning occurs through long-term potentiation (LTP)- and long-term depression (LTD)-like mechanisms is widely held but unproven. This hypothesis makes three assumptions: Synapses are modifiable, they modify with learning, and they strengthen through an LTP-like mechanism. We previously established the ability for synaptic modification and a synaptic strengthening with motor skill learning in horizontal connections of the rat motor cortex (MI). Here we investigated whether learning strengthened these connections through LTP. We demonstrated that synapses in the trained MI were near the ceiling of their modification range, compared with the untrained MI, but the range of synaptic modification was not affected by learning. In the trained MI, LTP was markedly reduced and LTD was enhanced. These results are consistent with the use of LTP to strengthen synapses during learning.

Shared neural substrates controlling hand movements in human motor cortex.

Voluntary hand movements in humans involve the primary motor cortex (M1). A functional magnetic resonance imaging method that measures relative cerebral blood flow was used to identify a distributed, overlapping pattern of hand movement representation within the posterior precentral gyrus, which contains M1. The observed pattern resembles those reported in nonhuman primates and differs from a somatotopically organized plan typically used to portray human motor cortex organization. Finger and wrist movements activated a wide expanse of the posterior precentral gyrus, and representations for different finger movements overlapped each other and the wrist representation. Multiple sites of activation occurred in the precentral gyrus for all movements. The overlapping representations may mediate motor and cognitive functions requiring coordinated neural processing for finger and wrist actions rather than discrete control implied by somatotopic maps.

Strengthening of horizontal cortical connections following skill learning.

Learning a new motor skill requires an alteration in the spatiotemporal pattern of muscle activation. Motor areas of cerebral neocortex are thought to be involved in this type of learning, possibly by functional reorganization of cortical connections. Here we show that skill learning is accompanied by changes in the strength of connections within adult rat primary motor cortex (M1). Rats were trained for three or five days in a skilled reaching task with one forelimb, after which slices of motor cortex were examined to determine the effect of training on the strength of horizontal intracortical connections in layer II/III. The amplitude of field potentials in the forelimb region contralateral to the trained limb was significantly increased relative to the opposite 'untrained' hemisphere. No differences were seen in the hindlimb region. Moreover, the amount of long-term potentiation (LTP) that could be induced in trained M1 was less than in controls, suggesting that the effect of training was at least partly due to LTP-like mechanisms. These data represent the first direct evidence that plasticity of intracortical connections is associated with learning a new motor skill.

Plasticity of adult sensorimotor representations.

Plasticity of sensory and motor cortical and subcortical representations in the adult brain appears to be a general phenomenon in animals that has now been extended to humans. There is a growing understanding of the mechanisms and rules that regulate the form and extent of reorganization; these appear to include activity-dependent control of synaptic efficacy, details of circuit arrangements, and growth of new axonal arbors. Of particular relevance to plasticity of cerebral cortical sensorimotor representations is recent evidence for the participation of intracortical horizontal pathways. These fibers provide a substrate for reorganization and contain mechanisms for increases or decreases in synaptic efficacy that depend on particular spatiotemporal activation patterns.

BCI Meeting 2005--workshop on clinical issues and applications.

This paper describes the outcome of discussions held during the Third International BCI Meeting at a workshop charged with reviewing and evaluating the current state of and issues relevant to brain-computer interface (BCI) clinical applications. These include potential BCI users, applications, validation, getting BCIs to users, role of government and industry, plasticity, and ethics.

Gaze direction modulates finger movement activation patterns in human cerebral cortex.

We investigated whether gaze direction modified the pattern of finger movement activation in human cerebral cortex using functional magnetic resonance imaging (MRI). Participants performed a sequential finger-tapping task or made no finger movements while maintaining gaze in the direction of the moving hand (aligned conditions) or away from the location of the moving hand. Functional MR signals, measured in the hemisphere contralateral to the moving hand, revealed finger movement-related activation in primary motor cortex, lateral and medial premotor cortex, and a wide extent of the lateral superior and inferior parietal lobules. In each area, the extent of the finger movement activation increased when static gaze was more aligned with the moving hand compared to when gaze was directed away from the moving hand. These data suggest the existence of large-scale cortical networks related to finger actions and indicate that skeletomotor processing in the cerebral cortex is consistently modified by gaze direction signals.

Neuronal interactions improve cortical population coding of movement direction.

Interactions among groups of neurons in primary motor cortex (MI) may convey information about motor behavior. We investigated the information carried by interactions in MI of macaque monkeys using a novel multielectrode array to record simultaneously from 12-16 neurons during an arm-reaching task. Pairs of simultaneously recorded cells revealed significant correlations in their trial-to-trial firing rate variation when estimated over broad (600 msec) time intervals. This covariation was only weakly related to the preferred directions of the individual MI neurons estimated from the firing rate and did not vary significantly with interelectrode distance. Most significantly, in a portion of cell pairs, correlation strength varied with the direction of the arm movement. We evaluated to what extent correlated activity provided additional information about movement direction beyond that available in single neuron firing rate. A multivariate statistical model successfully classified direction from single trials of neural data. However, classification was consistently better when correlations were incorporated into the model as compared to one in which neurons were treated as independent encoders. Information-theoretic analysis demonstrated that interactions caused by correlated activity carry additional information about movement direction beyond that based on the firing rates of independently acting neurons. These results also show that cortical representations incorporating higher order features of population activity would be richer than codes based solely on firing rate, if such information can exploited by the nervous system.

A collateral pathway to the neostriatum from corticofugal neurons of the rat sensory-motor cortex: an intracellular HRP study.

A projection from large pyramidal cells in layer V of the rat somatic sensory-motor (SSM) cortex both to the neostriatum and the brainstem was demonstrated by intracellular recording and injection of horseradish peroxidase (HRP). Layer V neurons that project to the brainstem were identified either by antidromic activation from the cerebral peduncle or by tracing the HRP-labeled axon into the internal capsule in histochemically processed sections. Intracellular responses to stimulation of the hindlimb, forelimb or mystacial pad were also examined. Five of 20 HRP-injected neurons that project to the brainstem had a fine collateral branch within the striatum. These branched corticostriatal cells respond at short latency (7--12 msec) to somatic sensory stimulation. All of the injected corticofugal neurons that had a striatal collateral were large pyramidal neurons located in layer Vb of the forelimb and head areas of SSM cortex. Branched corticofugal neurons have a rich basal dendritic field and a prominent apical dendrite that arborizes in the superficial cortical layers. Intracortical axon collaterals from the branched cells ramify in layers V and VI, and also project to the upper layers of cortex near the apical dentrite. Beyond the cortex, the main axon has no collateral branches, except for a single laterally directed branch in the neostriatum. The diameter of the striatal collateral is small (about 0.5 micrometer) compared to that of the main axon (2.0--2.5 micrometers). It is concluded that these branched cells provided a parallel input to the neostriatum and to brainstem or spinal motor centers.

The organization of thalamic projections to the parietal cortex of the Virginia opossum.

The thalamic projections to somatic sensory-motor (SSM) cortex and adjacent cortical areas of the Virginia opossum were studied using anterograde and retrograde axoplasmic transport techniques. Large injections of horseradish peroxidase and/or tritiated amino acids were made in the parietal cortex to identify all of the thalamic nuclei that are interconnected with this large cortical area. Very restricted injections were then made in physiologically identified subdivisions of SSM cortex, in the remaining posterior portion of parietal cortex, and in the anteriorly adjacent postorbital cortex. The results show that the parietal cortex is reciprocally connected with a number of thalamic nuclei. Different combinations of these thalamic areas project to specific subregions within the parietal field. All parts of the SSM cortex, which occupies the anterior four-fifths of parietal cortex, receive input from the ventrobasal complex (VB), the ventrolateral complex (VL), the central intralaminar nucleus (CIN), the central lateral nucleus (CL), and the ventromedial nucleus (VM). We could detect no segregation of VL and VB inputs in any part of SSM cortex. Projections from all of these thalamic nuclei, except VM, show at least some degree of topographic organization. Anterior-posterior strips of SSM cortex receive input from clusters of thalamic neurons that extend dorsoventrally and rostrocaudally through VB and VL. The posterior one-fifth of the parietal cortex (the posterior parietal area) receives input from VL, the posterior nuclear complex, and the lateral complex, as well as input from CL, CIN, and VM. Postorbital cortex receives input mainly from intralaminar, midline, and medial thalamic nuclei. We conclude that the projection field of VB in the parietal cortex coincides precisely with the first somatic sensory area (SI) as defined by single unit studies (Pubols et al., '76). The VB projection field also delineates the area of the first motor (MI) representation. Thus, there is no separation of SI and MI cortex in the opossum. The posterior parietal area lies outside of SSM cortex and has thalamic connections similar to the posterior parts of parietal cortex in other mammals.

The laminar distribution and ultrastructure of fibers projecting from three thalamic nuclei to the somatic sensory-motor cortex of the opossum.

The projections of the ventrobasal complex (VB), the ventrolateral complex (VL), and the central intralaminar nucleus (CIN) to the somatic sensory-motor (SSM) cortex of the Virginia opossum were studied with light and electron microscopic autoradiographic methods. VB, VL, and CIN have overlapping projections to SSM cortex and each one also projects to an additional cortical area. Unit responses to somatic sensory stimulation and the areal and laminar distribution of axons in cortex is different for VB, VL, and CIN, but the axons from each form similar round asymmetrical synapses, predominantly with dendritic spines. As in other mammals, VB units in the opossum have discrete, contralateral cutaneous receptive fields. VB projects somatotopically to SSM cortex and also projects to the second somatic sensory representation. Within the cortex, VB axons terminate densely in layer IV and the adjacent part of layer III. A few axons also terminate in the outermost part of layer I and the upper part of layer VI. Most VB axons terminate upon dendritic spines (86.6%), but they also contact dendritic shafts (10%) and neuronal cell bodies (3%). Neurons in VL have no reliable response to somatic stimulation under our recording conditions. VL projects to the SSM cortex and to the posterior parietal area. Throughout this entire projection field VL fibers terminate in layers I, III, and IV most densely, and sparsely in the other cortical layers. The density of termination in the mid-cortical laminae is quite sparse compared to VB, but the projection to layer I is considerably greater. Nearly all (93%) of VL axons contact dendritic spines, the remainder (7%) end on dendritic shafts. CIN is a thalamic target of ascending medial lemniscal, cerebellar, spinal, and reticular formation axons. Neurons in CIN respond to stimulation restricted to a particular body part, but typically responses may be evoked from larger areas and at longer latencies than neurons in VB that are related to the same body part. CIN neurons require a firm tap or electrical stimulation within their receptive field to elicit a response in the anesthetized preparation. CIN axons terminate throughout the entire parietal cortex, but unlike VB and VL, CIN fibers end almost exclusively in the outer part of layer I. Approximately 21% of CIN fibers contact dendritic shafts in layer I, which is twice the percentage of shafts contacted by VL or VB axons. All of the other CIN synapses are formed with dendritic spines. These experiments demonstrate three different pathways to SSM cortex. The results suggest that each projection has a unique role in controlling the patterns of activity of neurons within the SSM cortex.

The motor cortex of the rat: cytoarchitecture and microstimulation mapping.

The first motor (MI) cortex of the rat was identified as the region from which movements could be evoked by the lowest intensity of electrical stimulation. The location of this region was correlated with cytoarchitecture in the frontal and parietal cortex. Two frontal areas can be discerned in Nissl-stained sections: (1) the medial agranular field, marked by a pale-staining layer III and a compact layer II, and (2) the lateral agranular field, which has more homogeneous superficial layers and a broad layer V containing large, densely staining cells. Both of these regions project to the spinal cord and can therefore be included in the somatic sensorimotor cortex. MI in the rat coincides with the lateral agranular field but also overlaps with part of the adjacent granular cortex of the first somatic sensory (SI) representation. We conclude that the rat MI cortex can be identified by microstimulation techniques and by cytoarchitecture in the rat.

Afferent connections of the lateral agranular field of the rat motor cortex.

Retrograde axonal transport techniques were used to identify the afferent connections of the lateral agranular field (AG1) of the rat frontal cortex. This cytoarchitectonically distinct cortical field forms the bulk of the primary motor cortex (MI) as defined by intracortical microstimulation studies (Donoghue and Wise, '82). Following injections of horseradish peroxidase (HRP) or wheat germ agglutinin conjugated to HRP into AG1, retrogradely labeled cells are found in the forebrain, thalamus, and brainstem. Within the cerebral cortex labeled neurons are mainly present in the first somatic sensory area (SI), the second somatic sensory area (SII), and the medial agranular field, which lies medial and rostral to AG1. In SI, labeled cells are found primarily in a cytoarchitecturally distinct region called the dysgranular field of SI. Labeled neurons are present in layers II and III, Va, and the deepest part of layer VI in this field and in SII. Labeled cells are also present in layers Va and VI of the densely granular field of SI, which is the part of SI that is strongly activated by cutaneous inputs. Commissural inputs to AG1 arise from layers II-VI of the contralateral AG1 and thalamic inputs arise from the ventrolateral, ventromedial, posterior medial, and central lateral nuclei. Additional afferent fibers originate from neurons in the basal forebrain, the ventral thalamus, the midbrain raphé nuclei, and the locus coeruleus. This combination of inputs to AG1 from somatic sensory and frontal cortical fields, thalamic motor centers, and several other subcortical areas implies that AG1 forms a subdivision of sensorimotor cortex that is important in centrally directed movements as well as those that are guided by somatic sensory feedback.

Neocortex and hippocampus contain distinct distributions of calcium-calmodulin protein kinase II and GAP43 mRNA.

Calcium-calmodulin protein kinase II and GAP43 are two molecules which have been linked to synaptic plasticity. Localization of mRNA for these molecules identifies the neuronal populations which have the potential to utilize these mechanisms. General descriptions for calcium-calmodulin protein kinase II or GAP43 mRNA have been previously reported. In light of recent evidence that suggests that at some sites these two molecules may interact, we sought to determine the cortical distribution in detail, and to examine the extent of overlap between neuronal populations containing each mRNA. To this end we have used in situ hybridization techniques to study the distribution of calcium-calmodulin protein kinase II and GAP43 mRNA in adjacent sections of adult rat forebrain. Overall, the distribution patterns were distinct but partially overlapping. For both calcium-calmodulin protein kinase II and GAP43, mRNA levels were highest in hippocampus, allo- and neocortex, compared to moderate to low levels in striatum and thalamic nuclei. Within the heavily labeled regions certain populations expressed both calcium-calmodulin protein kinase II and GAP43 mRNA at high levels, while other populations were selective for calcium-calmodulin protein kinase II. In the hippocampus, the stratum pyramidale of CA1-3 expressed high levels of both calcium-calmodulin protein kinase II and GAP43 mRNA. Granule cells of the fascia dentata and the stratum radiatum of CA3 both contained moderate to high levels of calcium-calmodulin protein kinase II mRNA, but near background levels of GAP43 mRNA label. Within the neocortex, deep layers were distinguished from superficial layers by their lack of calcium-calmodulin protein kinase II mRNA expression within the neuropil, and the presence of GAP43 mRNA in neurons located in layer V and the deepest part of layer VI. Thus, layer V and deep layer VI neurons showed high levels of label for both GAP43 and calcium-calmodulin protein kinase II mRNA, while neurons of superficial layers contained only calcium-calmodulin protein kinase II mRNA. These markers differentiate neuronal populations which can also be distinguished on the basis of their ability to undergo specific forms of synaptic plasticity. These different forms of plasticity may be due in part to the laminar-specific patterns of GAP43 and calcium-calmodulin protein kinase II mRNA that we have described.

N-methyl-D-aspartate receptor mediated component of field potentials evoked in horizontal pathways of rat motor cortex.

To identify potential sites of synaptic modification of intrinsic cortical circuits, the contribution of the N-methyl-D-aspartate type of glutamate receptors to field potentials evoked in horizontal and oblique intracortical pathways was examined in rat motor cortex slice preparations. Presumably monosynaptic, short latency responses with a prominent negativity (-0.4 to -2.0 mV) were recorded in both superficial (across layer III) and deep (across layer V) horizontal pathways at a distance of approximately equal to 500 microns lateral to electrical stimulation sites and in oblique V-III pathway (-0.3 to -1.6 mV). Bath application of the N-methyl-D-aspartate receptor antagonist D,L-2-amino-5-phosphonovaleric acid (100 microM) reversibly decreased field potentials. Although decreases were observed in all components of the waveform, the most pronounced effect was on the late phase of the response. D,L-2-Amino-5-phosphonovaleric acid produced on average a 22% decrease in area, 12% in initial slope and 11% in peak amplitude of responses. Combined application of 100 microM D,L-2-amino-5-phosphonovaleric acid and a non-N-methyl-D-aspartate glutamate receptor antagonist, 6-cyano-7-nitro- or 6,7-dinitro-quinoxaline-2,3- dione (10-20 microM), eliminated all but a small, early and presumably non-synaptic response. In 18 of 23 cases, the relative contribution of the D,L-2-amino-5-phosphonovaleric acid-sensitive component was unrelated to field potential magnitude, suggesting that this component is present in all fiber classes. It is concluded that glutamate is the major transmitter of horizontal connections of layers II/III and layer V, as well as in the oblique V-III pathway. While most glutamatergic transmission is relayed by other glutamate receptor subtypes, N-methyl-D-aspartate receptor activation contributes a small but consistent part of ordinary transmission in each of these pathways in vitro. The results further suggest that a potential for N-methyl-D-aspartate receptor-mediated synaptic modification exists in intrinsic horizontal pathways of both superficial and deep layers of rat motor cortex.

The VEX-test for myocardial scintigraphy with thallium-201 and sestamibi: effect on abdominal background activity.

UNLABELLED: High abdominal background activity of 99mTc-sestamibi may interfere with the diagnosis in studies in which a coronary vasodilator is used; supplemental dynamic exercise might reduce this problem. METHODS: Clinical and angiographic determinants of subdiaphragmatic-to-myocardial activity ratios were measured on immediate poststress left anterior oblique images and on corresponding tomographic studies 1 hr after injection in 600 sestamibi studies. Similar measurements were made in 550 historic controls with planar 201Tl imaging. Patients performed symptom-limited ergometry when there were no limiting factors, dipyridamole-handgrip in which ergometry was not possible and VEX (vasodilator followed by symptom-limited ergometry) in which exercise capacity was reduced. RESULTS: Abdominal activity was higher with sestamibi than with 201Tl, in women versus men, and with dipyridamole-based tests compared to exercise alone. Compared to the dipyridamole-handgrip, 3 min of ergometry as part of VEX decreased abdominal background (p < or = 0.02) by 18% on immediate 201Tl images, by 13% on immediate sestamibi images and by 12% on 1-hr delayed sestamibi tomoacquisitions. CONCLUSIONS: Poststress abdominal background activity is influenced by similar factors with both agents. Supplemental exercise following dipyridamole reduces potentially interfering abdominal activity but perhaps not as efficiently with sestamibi as with 201Tl.

Pulmonary thallium-201 uptake following dipyridamole-exercise combination compared with single modality stress testing.

Angiographic and clinical determinants of pulmonary uptake of thallium-201 were assessed in a laboratory setting where supine bicycle exercise is used for stress testing in the absence of limiting pharmacologic or physical factors, and where symptom-limited exercise is added to intravenous dipyridamole infusion in other cases. Angiographic correlation was available in 400 patients, including 130 tested with exercise, 94 in whom only handgrip or abbreviated bicycle exercise could be used after dipyridamole, and 176 in whom intravenous dipyridamole was combined with a significant level of exercise. For each test mode, lung/myocardial ratios on the immediate image were highly correlated (p less than or equal to 0.001) with a score based on the number of critical coronary artery stenoses, with grading by contrast ventriculography, and with the number of stenosed (greater than or equal to 50%) arteries; relationships (p less than 0.05) to history of myocardial infarction and to gender were also present. Multiple regression analysis showed the critical stenosis score and ventricular dysfunction to be the only significant determinants. When dipyridamole based tests were compared with exercise, curves of receiver-operating characteristics showed a tendency to better diagnostic performance. When dipyridamole is incorporated in stress testing, the value of increased lung uptake as an ancillary diagnostic sign is similar to that established for exercise.

Contrasting properties of neurons in two parts of the primary motor cortex of the awake rat.

Features of neuronal activity in two subdivisions of primary motor cortex (MI) were recorded in awake rats. Neurons in the caudal part of MI, which overlaps part of the somatic sensory cortex, discharge with brief bursts in conjunction with isometric bar pressing with the forelimb. Cells in this caudal region are activated by cutaneous stimuli. In the rostral part of MI, neurons discharge prior to and during forelimb force changes, begin to discharge earlier than in the caudal zone, and have non-cutaneous or unidentifiable receptive fields. These results suggest separate motor control functions for rostral and caudal parts of rat MI.

Neostriatal projections from individual cortical fields conform to histochemically distinct striatal compartments in the rat.

A number of recent studies have demonstrated two chemically distinct compartments in the neostriatum: opiate receptor-rich patches and a surrounding matrix. Using axonal transport and receptor localization techniques in the rat brain, we have found that striatal projections from architectonically distinct cortical fields conform to this compartmentalized plan. The prelimbic cortex has bilateral projections that concentrate within the striatal patches. In contrast, the agranular motor cortex and cingulate cortex have bilateral projections to the matrix, while the somatic sensory cortex and visual cortex have ipsilateral matrix projections. Each matrix input occupies a characteristic striatal district. The projection to the patches distributes prelimbic input throughout the striatum, which may allow for prelimbic interactions with input from all other cortical areas.

Cholinergic modulation of sensory responses in rat primary somatic sensory cortex.

Responses evoked in single neurons of the primary somatic sensory cortex following tactile stimulation were examined before, during and after local iontophoretic application of acetylcholine (ACh) in urethane-anesthetized rats. The most common effect of ACh was an enhancement of the discharge evoked by sensory stimuli. Some cells responded to sensory input only in the presence of ACh. Response enhancement was observed in both supra- and infragranular layers, whereas response suppression was the most common effect in layer IV. These studies suggest that cholinergic systems modify sensory processing in cerebral neocortex by modulating the effectiveness of afferent inputs to cortical neurons in all layers.