The laminar organization of cell types in macaque cortex and its relationship to neuronal oscillations.

The canonical microcircuit (CMC) has been hypothesized to be the fundamental unit of information processing in cortex. Each CMC unit is thought to be an interconnected column of neurons with specific connections between excitatory and inhibitory neurons across layers. Recently, we identified a conserved spectrolaminar motif of oscillatory activity across the primate cortex that may be the physiological consequence of the CMC. The spectrolaminar motif consists of local field potential (LFP) gamma-band power (40-150 Hz) peaking in superficial layers 2 and 3 and alpha/beta-band power (8-30 Hz) peaking in deep layers 5 and 6. Here, we investigate whether specific conserved cell types may produce the spectrolaminar motif. We collected laminar histological and electrophysiological data in 11 distinct cortical areas spanning the visual hierarchy: V1, V2, V3, V4, TEO, MT, MST, LIP, 8A/FEF, PMD, and LPFC (area 46), and anatomical data in DP and 7A. We stained representative slices for the three main inhibitory subtypes, Parvalbumin (PV), Calbindin (CB), and Calretinin (CR) positive neurons, as well as pyramidal cells marked with Neurogranin (NRGN). We found a conserved laminar structure of PV, CB, CR, and pyramidal cells. We also found a consistent relationship between the laminar distribution of inhibitory subtypes with power in the local field potential. PV interneuron density positively correlated with gamma (40-150 Hz) power. CR and CB density negatively correlated with alpha (8-12 Hz) and beta (13-30 Hz) oscillations. The conserved, layer-specific pattern of inhibition and excitation across layers is therefore likely the anatomical substrate of the spectrolaminar motif. Significance Statement: Neuronal oscillations emerge as an interplay between excitatory and inhibitory neurons and underlie cognitive functions and conscious states. These oscillations have distinct expression patterns across cortical layers. Does cellular anatomy enable these oscillations to emerge in specific cortical layers? We present a comprehensive analysis of the laminar distribution of the three main inhibitory cell types in primate cortex (Parvalbumin, Calbindin, and Calretinin positive) and excitatory pyramidal cells. We found a canonical relationship between the laminar anatomy and electrophysiology in 11 distinct primate areas spanning from primary visual to prefrontal cortex. The laminar anatomy explained the expression patterns of neuronal oscillations in different frequencies. Our work provides insight into the cortex-wide cellular mechanisms that generate neuronal oscillations in primates.

Functionally matched domains in parietal-frontal cortex of monkeys project to overlapping regions of the striatum.

Projections of small regions (domains) of primary motor cortex (M1), premotor cortex (PMC) and posterior parietal cortex (PPC) to the striatum of squirrel monkeys were revealed by restricted injections of anterograde tracers. As many as 8 classes of action-specific domains can be identified in PPC, as well as in PMC and M1, and some have been identified for injections by the action evoked by 0.5 s trains of electrical microstimulation. Injections of domains in all three cortical regions labeled dense patches of terminations in the matrix of the ipsilateral putamen, while providing sparse or no projections to corresponding regions of the contralateral putamen. When two or three of these domains were injected with different tracers, projection fields in the putamen were highly overlapped for injections in functionally matched domains across cortical areas, but were highly segregated for injections placed in functionally mismatched domains. While not all classes of domains were studied, the results suggest that the striatum potentially has separate representations of eight or more classes of actions that receive inputs from domains in three or more cortical regions in sensorimotor cortex. The overlap/segregation of cortico-striatal projections correlates with the strength of cortico-cortical connections between injected motor areas.

Modular segregation of functional cell classes within the postcentral somatosensory cortex of monkeys.

The distribution of two functionally distinct cell types, presumably related to slowly and rapidly adapting mechanoreceptors in the skin, was explored within the representation of the glabrous hand in area 3b of the somatosensory cortex of monkeys. The two cell classes lie in relatively segregated alternating anteroposterior bands within the middle layers of the cortex.

Representation of the visual field on the medial wall of occipital-parietal cortex in the owl monkey.

The medial visual area is located on the medical wall of occipital-parietal cortex. A much larger proportion of this area is devoted to the representation of the more peripheral parts of the visual field than in any other cortical area or subcortical visual structure than has been mapped previously in any species of primate.

Genetic abnormality of the visual pathways in a "white" tiger.

"White"tigers show an inherited reduction of pigment, produced by an autosomal recessive gene. The brain of one of these tigers shows an abnormality of the visual pathways similar to abnormalities that are associated with albinism in many other mammals. There is a close relationship between the reduced pigment formation, the pathway abnormality, and strabismus.

Recovery of normal topography in the somatosensory cortex of monkeys after nerve crush and regeneration.

After median nerve fibers to glabrous skin on the hands of monkeys were crushed and allowed to regenerate, normal topographical organization was recovered in the representation of the hand in primary somatosensory cortex. Similar recovery of normal cortical organization may underlie the sensory restoration that usually follows nerve crush injury in humans.

Large-scale sprouting of cortical connections after peripheral injury in adult macaque monkeys.

Distributions of thalamic and cortical connections were investigated in four macaque monkeys with long-standing, accidental trauma to a forelimb, to determine whether the growth of new connections plays a role in the reorganization of somatosensory cortex that occurs after major alterations in peripheral somatosensory inputs. In each monkey, microelectrode recordings of cortical areas 3b and 1 demonstrated massive reorganizations of the cortex related to the affected limb. Injections of tracers in area 1 of these monkeys revealed normal patterns of thalamocortical connections, but markedly expanded lateral connections in areas 3b and 1. Thus, the growth of intracortical but not thalamocortical connections could account for much of the reorganization of the sensory maps in cortex.

Multiple representations of the body within the primary somatosensory cortex of primates.

Microelectrode mapping experiments indicate that the classical primary somatosensory cortex of monkeys consists of as many as four separate body representations rather than just one. Two complete body surface representations occupy cortical fields 3b and 1. In addition, area 2 contains an orderly representation of predominantly "deep" body tissues. Area 3a may constitute a fourth representation.

Reorganization of retinotopic cortical maps in adult mammals after lesions of the retina.

The organization of the visual cortex has been considered to be highly stable in adult mammals. However, 5 degrees to 10 degrees lesions of the retina in the contralateral eye markedly altered the systematic representations of the retina in primary and secondary visual cortex when matched inputs from the ipsilateral eye were also removed. Cortical neurons that normally have receptive fields in the lesioned region of the retina acquired new receptive fields in portions of the retina surrounding the lesions. The capacity for such changes may be important for normal adjustments of sensory systems to environmental contingencies and for recoveries from brain damage.

Massive cortical reorganization after sensory deafferentation in adult macaques.

After limited sensory deafferentations in adult primates, somatosensory cortical maps reorganize over a distance of 1 to 2 millimeters mediolaterally, that is, in the dimension along which different body parts are represented. This amount of reorganization was considered to be an upper limit imposed by the size of the projection zones of individual thalamocortical axons, which typically also extend a mediolateral distance of 1 to 2 millimeters. However, after extensive long-term deafferentations in adult primates, changes in cortical maps were found to be an order of magnitude greater than those previously described. These results show the need for a reevaluation of both the upper limit of cortical reorganization in adult primates and the mechanisms responsible for it.

Spinal cord atrophy and reorganization of motoneuron connections following long-standing limb loss in primates.

Primates with long-standing therapeutic amputations of a limb at a young age were used to investigate the possibility that deefferented motor nerves sprout to new muscle targets. Injections of anatomical tracers into the muscles proximal to the amputated stump labeled a larger extent of motoneurons than matched injections on the intact side or in normal animals, including motoneurons that would normally supply only the missing limb muscles. Although the total numbers of distal limb motoneurons remained normal, some distal limb motoneurons on the amputated side were smaller in size and simpler in form. These results suggest that deprived motoneurons survive and retain function by reinnervating new muscle targets. The sprouted motor efferents may account for some of the reorganization of primary motor cortex that follows long-standing amputation.

'What' and 'where' processing in auditory cortex.

Tracing of auditory cortical connections suggests that the primate auditory system, like the visual and somatosensory systems, may be organized into 'what' and 'where' pathways.

Simultaneous encoding of tactile information by three primate cortical areas.

We used simultaneous multi-site neural ensemble recordings to investigate the representation of tactile information in three areas of the primate somatosensory cortex (areas 3b, SII and 2). Small neural ensembles (30-40 neurons) of broadly tuned somatosensory neurons were able to identify correctly the location of a single tactile stimulus on a single trial, almost simultaneously. Furthermore, each of these cortical areas could use different combinations of encoding strategies, such as mean firing rate (areas 3b and 2) or temporal patterns of ensemble firing (area SII), to represent the location of a tactile stimulus. Based on these results, we propose that ensembles of broadly tuned neurons, located in three distinct areas of the primate somatosensory cortex, obtain information about the location of a tactile stimulus almost concurrently.

Human visual cortex. Progress and puzzles.

A wealth of data is now available on the functional organization of the human visual cortex. Caution is necessary in basing interpretations of such data on information gained from studies of the monkey visual cortex.

Thalamic connections of the dorsal and ventral premotor areas in New World owl monkeys.

Thalamic connections of two premotor cortex areas, dorsal (PMD) and ventral (PMV), were revealed in New World owl monkeys by injections of fluorescent dyes or wheat-germ agglutinin conjugated to horseradish peroxidase (WGA-HRP). The injections were placed in the forelimb and eye-movement representations of PMD and in the forelimb representation of PMV as determined by microstimulation mapping. For comparison, injections were also placed in the forelimb representation of primary motor cortex (M1) of two owl monkeys. The results indicate that both PMD and PMV receive dense projections from the ventral lateral (VL) and ventral anterior (VA) thalamus, and sparser projections from the ventromedial (VM), mediodorsal (MD) and intralaminar (IL) nuclei. Labeled neurons in VL were concentrated in the anterior (VLa) and the medial (VLx) nuclei, with only a few labeled cells in the dorsal (VLd) and posterior (VLp) nuclei. In VA, labeled neurons were concentrated in the parvocellular division (VApc) dorsomedial to VLa. Labeled neurons in MD were concentrated in the most lateral and posterior parts of the nucleus. VApc projected more densely to PMD than PMV, especially to rostral PMD, whereas caudal PMD received stronger projections from neurons in VLx and VLa. VLd projected exclusively to PMD, and not to PMV. In addition, neurons labeled by PMD injections tended to be more dorsal in VL, IL, and MD than those labeled by PMV injections. The results indicate that both premotor areas receive indirect inputs from the cerebellum (via VLx, VLd and IL) and globus pallidus (via VLa, VApc, and MD). Comparisons of thalamic projections to premotor and M1 indicate that both regions receive strong projections from VLx and VLa, with the populations of cells projecting to M1 located more laterally in these nuclei. VApc, VLd, and MD project mainly to premotor areas, while VLp projects mainly to M1. Overall, the thalamic connectivity patterns of premotor cortex in New World owl monkeys are similar to those reported for Old World monkeys.

The emergence and evolution of mammalian neocortex.

Cortical variation in mammals and other terrestrial vertebrates, re-examined by current comparative methodology (out-group analysis), indicates that separate lateral (olfactory), dorsal and medial (hippocampal) pallial or cortical formations arose with the origin of vertebrates. Although the exact origin of mammalian isocortex (so-called neocortex) is still disputed, it appears that the earliest mammals already had a six-layered isocortex with ten to 20 functional subdivisions. Among placental mammals, at least, isocortex has expanded numerous times, producing additional cortical subdivisions. Because these expansions were independent transformations of a simpler cortex, they produced subdivisions that are not homologous.

Dynamic features of sensory and motor maps.

Recent data support the idea that the functional organizations of sensory and motor maps in the mature brain are dynamically maintained. Experiments employing peripheral injuries or other manipulations indicate that these maps are capable of extensive reorganization. A number of candidate mechanisms for these changes have been suggested, providing avenues for further research.

Hierarchical, parallel, and serial arrangements of sensory cortical areas: connection patterns and functional aspects.

Recent studies have led to a better understanding of the organization and connections of somatosensory and visual cortex in a number of mammalian species. Lesion studies have provided information on the significance of particular connections. The variable effectiveness of cortical lesions in deactivating target areas suggests that serial processing may be emphasized in higher primates.

The organization of sensory cortex.

Recent studies of primary visual cortex (V1) redefine layers 3 and 4 of V1 in monkeys and show that monkeys, apes and humans have different laminar specializations. Projections from V1 define a smaller, but complete, third visual area, and a dorsomedial area. The middle temporal visual area has two types of motion-sensitive modules with inputs from cytochrome oxidase columns in V1. Second-level somatosensory areas have been described in humans, and a second-level auditory area is shown to respond to somatosensory stimuli.

Auditory processing in primate cerebral cortex.

Auditory information is relayed from the ventral nucleus of the medial geniculate complex to a core of three primary or primary-like areas of auditory cortex that are cochleotopically organized and highly responsive to pure tones. Auditory information is then distributed from the core areas to a surrounding belt of about seven areas that are less precisely cochleotopic and generally more responsive to complex stimuli than tones. Recent studies indicate that the belt areas relay to the rostral and caudal divisions of a parabelt region at a third level of processing in the cortex lateral to the belt. The parabelt and belt regions have additional inputs from dorsal and magnocellular divisions of the medial geniculate complex and other parts of the thalamus. The belt and parabelt regions appear to be concerned with integrative and associative functions involved in pattern perception and object recognition. The parabelt fields connect with regions of temporal, parietal, and frontal cortex that mediate additional auditory functions, including space perception and auditory memory.