Effects of aerobic exercise training on cognitive function and cortical vascularity in monkeys.

This study examined whether regular exercise training, at a level that would be recommended for middle-aged people interested in improving fitness could lead to improved cognitive performance and increased blood flow to the brain in another primate species. Adult female cynomolgus monkeys were trained to run on treadmills for 1 h a day, 5 days a week, for a 5 month period (n=16; 1.9+/-0.4 miles/day). A sedentary control group sat daily on immobile treadmills (n=8). Half of the runners had an additional sedentary period for 3 months at the end of the exercise period (n=8). In all groups, half of the monkeys were middle-aged (10-12 years old) and half were more mature (15-17 years old). Starting the fifth week of exercise training, monkeys underwent cognitive testing using the Wisconsin General Testing Apparatus (WGTA). Regardless of age, the exercising group learned to use the WGTA significantly faster (4.6+/-3.4 days) compared to controls (8.3+/-4.8 days; P=0.05). At the end of 5 months of running monkeys showed increased fitness, and the vascular volume fraction in the motor cortex in mature adult running monkeys was increased significantly compared to controls (P=0.029). However, increased vascular volume did not remain apparent after a 3-month sedentary period. These findings indicate that the level of exercise associated with improved fitness in middle-aged humans is sufficient to increase both the rate of learning and blood flow to the cerebral cortex, at least during the period of regular exercise.

Organization of cerebellar and area "y" projections to the nucleus reticularis tegmenti pontis in macaque monkeys.

Axonal projections to the nucleus reticularis tegmenti pontis (RTP) were studied in 11 macaque monkeys by mapping axonal degeneration from lesions centered in the dentate and interpositus anterior (IA) nuclei and by mapping anterograde transport of tritiated amino acid precursors injected into the dentate nucleus. Projections from the dentate and IA nuclei overlap in central parts of the body of RTP, but the terminal field of dentate axons extends dorsomedial and rostral to the terminal field of IA axons, and IA terminal fields extend more ventrolaterally. A caudal to rostral topography of projections from each nucleus onto dorsal to ventral parts of RTP was seen. Projections from rostral parts of both nuclei terminate in a sublemniscal part of the nucleus. The topography of dentate and IA projections onto central to ventrolateral RTP appears to match somatotopic maps of these cerebellar nuclei with the somatotopic map of projections to RTP from primary motor cortex. Projections from caudal and ventral parts of the dentate nucleus appear to overlap oculomotor inputs to rostral, dorsal, and medial RTP from the frontal and supplementary eye fields, the superior colliculus, and the oculomotor region of the caudal fastigial nucleus. Projections to the paramedian part of RTP from vestibular area "y" were also found in two cases that correlated with projections to vertical oculomotor motoneurons. The maps of dentate and IA projections onto RTP correlate predictably with maps of dentate and IA projections to the ventrolateral thalamus and subnuclei of the red nucleus that were made from these same cases (Stanton [1980b] J. Comp. Neurol. 192:377-385).

Topography of projections to posterior cortical areas from the macaque frontal eye fields.

Frontal eye field (FEF) projections to posterior cortical areas were mapped by autoradiography of tritiated amino acids (Leu, Pro) in six macaque monkeys. In three monkeys, the large saccade part of the FEF (IFEF) was identified by microstimulation and injected with tracers. In a fourth monkey, the small saccade part of the FEF (sFEF) was identified by microstimulation and injected with tracer. Tracer injections were placed into the sFEF region of two other monkeys using anatomical landmarks. The IFEF and sFEF generally had distinct and largely segregated projections to posterior cortical areas, and the overall pattern of labeling in visual areas with established topology indicates that IFEF neurons preferentially project to areas having large and eccentric receptive fields, whereas sFEF neurons project to areas having smaller, more centrally located fields. The terminal fields from the sFEF were more widespread than those from IFEF. Projections from sFEF terminated in the lateral intraparietal area (LIP), the ventral intraparietal area (VIP), and the parietal part of visual area V3A, in the fundus of the superior temporal visual area (FST), the middle temporal area (MT), the medial superior temporal area (MST), the temporal part of visual area V4, the inferior temporal area (IT), and the temporal-occipital area (TEO) and in occipital visual areas V2, V3, and V4. Projections from IFEF terminated in parietal areas 7a, LIP, and VIP and the medial part of parietal area PE; in temporal areas MST and the superior temporal polysensory area (STP); and in occipital area V2 and posterior cingulate area 23b. Projections from IFEF and sFEF appeared to terminate in different parts of common target areas in MST, LIP, and V2. The topography of IFEF and sFEF projections to LIP suggests that this posterior eye field may also be organized by saccade amplitude. Most terminal labeling from FEF injections was bilaminar to layers I and V/VI, but labeling in area LIP, area MT, the medial part of area PE, and area 23b was columnar-form to all layers.

A three-dimensional reconstruction of the human hypothalamus.

A previous three-dimensional reconstruction of the rat hypothalamus revealed an organization of nuclei into three major clusters. This clustering may relate to the presence of three hypothalamic anlagen in the embryo and to a restriction of marker proteins and transcription factors to specific regions of the hypothalamus during development. To see if a similar clustering was apparent in the human hypothalamus, a reconstruction of the hypothalamus from a male cadaver was prepared. The reconstruction showed a clustering of nuclei into three anterior-posterior groups. With reference to the well-defined human supraoptic nucleus, the suprachiasmatic and lateral mammillary nuclei were proportionately smaller in this single human specimen than would be expected from data in the rat. A high degree of homology between hypothalamic structures in the rat and human were generally observed.

Thalamic afferents of area 4 and 6 in the dog: a multiple retrograde fluorescent dye study.

In the present study, we compared the distribution of thalamocortical afferents of cortical area 4 to that of cortical area 6 in the dog, using fluorescent tracers. Multiple injections of combinations of two dyes (diamidino yellow dihydrochloride, Evans blue, fast blue, granular blue) were made into either the anterior and posterior sigmoid gyri or into the medial and lateral regions of the anterior sigmoid gyrus in the anesthetized dog. We found that the thalamic afferents of areas 4 and 6 arise from topographically organized bands of cells that traverse several thalamic nuclei and extend throughout the rostrocaudal extent of the thalamus. The most medial band included area 6-projecting neurons in the anterior nuclei, the rhomboid nucleus, the ventral anterior nucleus (VA), ventromedial nucleus (VM) and mediodorsal nucleus (MD). Within this band, cells projecting to medial area 6 a alpha tended to be more numerous in the anterior nuclei, anterior parts of VA and VM and anterior and caudal parts of MD. Fewer cells in MD but more cells in caudal parts of VA and VM projected to lateral area 6 a beta. Lateral bands of cells in central through lateral parts of VA and VL projected topographically to lateral area 4 on the anterior sigmoid gyrus and lateral through medial parts of postcruciate area 4. The most lateral band of cells in VL continued ventrally into the zona incerta. Area 4 also received input from VM and the central lateral (CL) and centrum medianum (CM) nuclei.(ABSTRACT TRUNCATED AT 250 WORDS)

Cell-cell communication correlates with pattern formation in molting Manduca midgut epithelium.

The midgut epithelium of larval Manduca sexta is constructed of single goblet cells surrounded by a one-cell-thick reticulum of columnar cells. This pattern is expanded at each molt by the addition of new cells. Between molts, these epithelial cells are not dye coupled, even though gap junctions are present. Proliferating stem cells are dye coupled in small groups early in the molt. Then, at mid-molt, the whole epithelium temporarily becomes dye coupled. This is when the new (expanded) pattern is being established. Later, at the end of the molt, the epithelium returns to the non-coupled state. These results suggest that cell communication via gap junctions may play a role in cell patterning.

Topography of projections to the frontal lobe from the macaque frontal eye fields.

Efferents from the frontal eye fields (FEF) to the ipsilateral frontal lobe were studied by autoradiography of tritiated tracers (leucine, proline, and fucose) in seven macaque monkeys that were used previously to describe subcortical connections. In four of the cases, tracer injection sites were confirmed by low thresholds for the electrical elicitation of saccadic eye movements. Cases were grouped as lFEF of sFEF cases according to large or small saccades that were characteristic of the injection site. Projections from the FEF terminated in five frontal regions: 1) area FD on the dorsomedial convexity; 2) area FC (containing SEF) medial to the upper limb of the arcuate sulcus; 3) areas FD and FD delta along the walls of the principal sulcus; 4) area FCBm on the deep, posterior wall of the arcuate sulcus inferior to the sulcal spur; and 5) the inferolateral cortex (area FDi) on the convexity and lateral two thirds of the anterior wall of the arcuate sulcus. Projections in sFEF cases tended to be confined to medial parts of dorsomedial FD and FC and the lateral wall of the principal sulcus and inferolateral convexity. Neither lFEF nor sFEF appeared to project to the SMA or pericingulate cortex. Label in these areas was found only in the cases in which tracer spread into non-FEF areas. FEF projections terminated in column-like patches of about 500-600 microns in diameter. Labeled axons and terminals were seen in all cortical layers regardless of location in the frontal lobe.

A three-dimensional reconstruction of the rat hypothalamus.

Tracings of structures present in serial coronal frozen sections of the rat hypothalamus were entered into an IBM-PC and sections were aligned in space using a program for 3-dimensional reconstruction. The positions and relative volumes of 16 major hypothalamic nuclei were accurately displayed in lateral, medial, and superior views of the hypothalamus. Three major clusters of hypothalamic nuclei were apparent, reinforcing embryological concepts of "neuromeres" from which adult structures develop. A better knowledge of the spatial locations of hypothalamic nuclei, which determine the pathways of intrahypothalamic connections, should be of aid in interpreting studies which disrupt such connections.

Cytoarchitectural characteristic of the frontal eye fields in macaque monkeys.

The cytoarchitecture of the prearcuate gyrus, including the region of the physiologically defined frontal eye fields (FEF), was studied in four macaque monkeys (Macaca fascicularis, M. mulatta) to determine if the FEF could be anatomically identified. Brain sections were stained with standard Nissl and, in some cases, myelin stains. Two nonstandard planes of section were used: one tangential to the prearcuate gyrus and the second normal to the most posterior bend of the prearcuate gyrus. The first plane of section was advantageous for studying the location of the FEF with reference to the entire medial-lateral extent of the gyrus and the second allowed good comparisons of the FEF to adjacent anterior and posterior cortical areas. Frontal plane sections through the prearcuate gyrus were also examined in 15 macaque monkeys for comparison with sections cut normal to the posterior bend of the gyrus and tangential to the gyrus. Intracortical microstimulation was performed in three monkeys. The FEF was defined as the area from which low-threshold (less than or equal to 50 microA) saccades could be evoked. The area extended about 10 mm along the anterior bank of the arcuate sulcus. Within the area, saccade amplitudes were represented in a mediolateral, large-to-small topography. No topography of saccade direction was noted within FEF but reversals of saccade direction for any given electrode pass were found. These results confirm the results from our earlier mapping study of FEF (Bruce et al.: J. Neurophysiol. 54:714-734, '85). Cell bodies of large pyramidal cells in layers III and V of the prearcuate gyri from three hemispheres were measured with the aid of an image-combining computer microscope. The distribution of cells of greater than 22 microns diameter or cross-sectional areas of greater than 500 microns 2 were plotted. In one monkey, marker lesions made at microstimulation sites within the FEF or in adjacent non-FEF areas were also plotted. The location of the FEF appeared to coincide with the concentration of large layer V pyramidal cells in the prearcuate gyrus rather than with any previously mapped cytoarchitectonic area. The numbers of large pyramids in layer V were noticeably reduced along the lip of the prearcuate gyrus and at dorsomedial and ventrolateral locations which were outside the physiologically defined FEF.

Frontal eye field efferents in the macaque monkey: I. Subcortical pathways and topography of striatal and thalamic terminal fields.

Anterograde tracers (tritiated leucine, proline, fucose; WGA-HRP) were injected into sites within the frontal eye fields (FEF) of nine macaque monkeys. Low thresholds (less than or equal to 50 microA) for electrically evoking saccadic eye movements were used to locate injection sites in four monkeys. Cases were grouped according to the amplitude of saccades evoked or predicted at the injection site. Dorsomedial prearcuate injection sites where large saccades were elicited were classified as lFEF cases, whereas ventrolateral prearcuate sites where small saccades were evoked were designated sFEF cases. One control case was injected in the medial postarcuate area 6. We found five descending fiber bundles from FEF; fibers to the striatum, which enter the caudate nucleus at or just rostral to the genu of the internal capsule; fibers to the claustrum, which travel in the external capsule; and transthalamic, subthalamic, and pedunculopontine fibers. Our results indicate that transthalamic and subthalamic pathways supply all terminal sites in the thalamus, subthalamus, and tegmentum of the midbrain and pons, whereas pedunculopontine fibers appear to terminate in the pontine and reticularis tegmenti pontis nucleus exclusively. Frontal eye field terminal fields in the striatum were topographically organized: lFEF projections terminated dorsal and rostral to sFEF projections. Thus, lFEF terminal fields were located centrally in the head and body of the caudate nucleus and a small dorsomedial portion of the putamen, whereas sFEF terminal fields were located in ventrolateral parts of the caudate body and ventromedial parts of the putamen. In the claustrum, lFEF projections terminated dorsal and rostral to sFEF projections. Projections from FEF terminated in ventral and caudal parts of the subthalamic nucleus without a clear topography. By comparison, terminal fields from medial postarcuate area 6 were located more caudally and laterally in the striatum and claustrum than projections from FEF, and more centrally in the subthalamic nucleus. In the thalamus, FEF terminal patches in some thalamic nuclei were also topographically organized. Projections from lFEF terminated in dorsal area X, dorsolateral medial dorsal nucleus, pars parvicellularis (MDpc), and the caudal pole of MDpc, whereas projections from sFEF terminated in ventral area X, medial dorsal nucleus, pars multiformis, and caudal medial dorsal nucleus pars densocellularis.(ABSTRACT TRUNCATED AT 400 WORDS)

Frontal eye field efferents in the macaque monkey: II. Topography of terminal fields in midbrain and pons.

Frontal eye field (FEF) projections to the midbrain and pons were studied in nine macaque monkeys that were used to study FEF projections to the striatum and thalamus (Stanton et al.: J. Comp. Neurol. 271:473-492, '88). Injections of tritiated amino acids or WGA-HRP were made into FEF cortical locations where low-level microstimulation (less than or equal to 50 microA) elicited saccadic eye movements, and anterograde axonal labeling was mapped. The injections were made into the anterior bank of the arcuate sulcus from dorsomedial sites where large saccades were evoked (lFEF) to ventrolateral sites where small saccades were evoked (sFEF). The largest terminal fields of FEF fibers were located in the ipsilateral superior colliculus (SC). Projections to SC were topographically organized: lFEF sites projected to intermediate and deep layers of caudal SC, sFEF sites projected to intermediate and superficial layers of rostral SC, and FEF sites between these extremes projected to intermediate locations in SC. Patches of terminal labeling were located ipsilaterally in the lateral mesencephalic reticular formation near the parabigeminal nucleus and the ventrolateral pontine reticular formation. These patches were larger from lFEF injections. Small, dense terminal patches were seen in the ipsilateral pontine gray, mostly along the medial and dorsal borders of these nuclei but occasionally in central and dorsolateral regions. Patches of label like those in the pontine nuclei were located ipsilaterally in the reticularis tegmenti pontis nucleus in lFEF cases and bilaterally in sFEF cases. Small terminal patches were found in the nucleus of Darkschewitsch and dorsal and medial parts of the parvicellular red nucleus in most FEF cases. In the pretectal region, labeled terminal patches were consistently found in the nucleus limitans of the posterior thalamus, but we could not determine if label in the nucleus of the pretectal area and dorsal parts of the nucleus of the posterior commissure marked axon terminals or fibers of passage. We found small, lightly labeled terminal patches in the pontine raphe between the rootlets of the abducens nerve (three cases) or in the adjacent paramedian pontine reticular formation (one case). Omnipauser cells in this region are important in initiating saccades. In one sFEF case, very small patches of label were located in the supragenual nuclei anterior to the abducens nuclei and in the ipsilateral nucleus prepositus hypoglossi posterior to the abducens nucleus. Presaccadic burster neurons in the periabducens region are known to fire immediately before horizontal saccades.(ABSTRACT TRUNCATED AT 400 WORDS)

Thalamic afferents to cytoarchitectonic subdivisions of area 6 on the anterior sigmoid gyrus of the dog: a retrograde and anterograde tracing study.

The cytoarchitecture and thalamic afferents of cortical area 6 located on the anterior sigmoid gyrus were mapped and analyzed in the dog by means of cytoarchitectonic, horseradish peroxidase (HRP), and autoradiographic methods. Cytoarchitectonically, area 6 consists of medial and lateral subdivisions that correspond, respectively, to areas 6a alpha and 6a beta in the cat. In the dog, area 6a alpha is characterized by a wide layer III, the merging of borders between layers III and V, the presence of small-to-medium-size pyramidal cells in layer V, and a pallisade arrangement of cells in layer VI. Area 6a beta appears more stratified, with a relatively acellular layer present between layers V and VI and the presence of large pyramidal cells in layer V. Neither area 6a alpha nor 6a beta contains a layer IV. Data obtained from injections of HRP into areas 6a alpha or 6a beta revealed that labeled thalamic neurons were distributed in a longitudinal band extending from the rostral part of the ventral anterior nucelus (VA) through the caudal part of the mediodorsal nucleus (MD). Labeled cells were observed in the ventral lateral and ventral medial thalamic nuclei as well as in several of the intralaminar nuclei including the central lateral, central medial, parafascicular, and centrum medianum nuclei. A few labeled cells were also located in the suprageniculate nucleus. The densest thalamic labeling was present in VA and MD following injections into area 6a alpha. Equivalent or even larger injections into area 6a beta resulted in much less thalamic labeling. The band of labeled cells also extended into the hypothalamus, zona incerta, amygdala, claustrum, periaqueductal gray of the midbrain, and the nucleus of Darkschewitsch. Results from autoradiographic experiments showed that area 6 subdivisions receive a loosely organized topographic input from VA. Injections of tritiated amino acids were made into selected regions of VA and into the caudal part of MD, areas in which the largest numbers of HRP-labeled cells were located. Area 6a alpha receives afferents primarily from the rostromedial part of VA and the caudal part of MD while area 6a beta receives its principal input from the caudal and lateral parts of VA with minimal input from MD. Axons originating from VA terminate in both layers I and III of area 6 while those originating from the caudal part of MD terminate only in layer III.(ABSTRACT TRUNCATED AT 400 WORDS)

Primate frontal eye fields. II. Physiological and anatomical correlates of electrically evoked eye movements.

We studied single neurons in the frontal eye fields of awake macaque monkeys and compared their activity with the saccadic eye movements elicited by microstimulation at the sites of these neurons. Saccades could be elicited from electrical stimulation in the cortical gray matter of the frontal eye fields with currents as small as 10 microA. Low thresholds for eliciting saccades were found at the sites of cells with presaccadic activity. Presaccadic neurons classified as visuomovement or movement were most associated with low (less than 50 microA) thresholds. High thresholds (greater than 100 microA) or no elicited saccades were associated with other classes of frontal eye field neurons, including neurons responding only after saccades and presaccadic neurons, classified as purely visual. Throughout the frontal eye fields, the optimal saccade for eliciting presaccadic neural activity at a given recording site predicted both the direction and amplitude of the saccades that were evoked by microstimulation at that site. In contrast, the movement fields of postsaccadic cells were usually different from the saccades evoked by stimulation at the sites of such cells. We defined the low-threshold frontal eye fields as cortex yielding saccades with stimulation currents less than or equal to 50 microA. It lies along the posterior portion of the arcuate sulcus and is largely contained in the anterior bank of that sulcus. It is smaller than Brodmann's area 8 but corresponds with the union of Walker's cytoarchitectonic areas 8A and 45. Saccade amplitude was topographically organized across the frontal eye fields. Amplitudes of elicited saccades ranged from less than 1 degree to greater than 30 degrees. Smaller saccades were evoked from the ventrolateral portion, and larger saccades were evoked from the dorsomedial portion. Within the arcuate sulcus, evoked saccades were usually larger near the lip and smaller near the fundus. Saccade direction had no global organization across the frontal eye fields; however, saccade direction changed in systematic progressions with small advances of the microelectrode, and all contralateral saccadic directions were often represented in a single electrode penetration down the bank of the arcuate sulcus. Furthermore, the direction of change in these progressions periodically reversed, allowing particular saccade directions to be multiply represented in nearby regions of cortex.(ABSTRACT TRUNCATED AT 400 WORDS)

[3H]Choline labeling of cerebellothalamic neurons with observations on the cerebello-thalamo-parietal pathway in cats.

[3H]Choline injected into the ventral lateral thalamic nucleus (VL) labeled cell bodies of the deep cerebellar nuclei and adjacent vestibular nuclei by retrograde axoplasmic transport. Injections in caudal and dorsal parts of VL labeled cells in ventral parts of the dentate nucleus and interpositus posterior nucleus. Injections in rostral and ventral parts of VL labeled cells in the interpositus anterior nucleus and dorsal parts of the dentate nucleus. A few labeled cell bodies were found throughout the rostrocaudal extent of the fastigial nucleus and in adjacent parts of the vestibular nuclei. A combined injection of [3H]choline and [3H]amino acids labeled cells in the deep cerebellar nuclei and axon terminals in layer I of the middle suprasylvian gyrus (areas 5, 7).

The ventral lateral thalamic nucleus in the dog: cytoarchitecture, acetylthiocholinesterase histochemistry, and cerebellar afferents.

The canine ventral lateral nucleus was defined on the basis of Nissl cytoarchitecture, acetylthiocholinesterase (AChE) chemoarchitecture, and the distribution of cerebellar afferents. It consists of at least two subdivisions: a dorsal and rostral division (VLd) that stains intensely for AChE and contains densely packed cells, and a larger principal division (VL), located ventrocaudally, that stains moderately for AChE and contains a mixture of cells with moderate packing density. Results from autoradiographic, silver degeneration, and AChE material indicate that both divisions receive input from the deep cerebellar nuclei.

Brain stem afferents to the periabducens reticular formations (PARF) in the cat. An HRP study.

Six injections of HRP were placed in the periabducens reticular formation (PARF). Two were placed ventromedial to the caudal half of the abducens nucleus (VIn), two were placed further laterally and ventral to the rostral half of the nucleus, and two were placed rostral to the nucleus. Most injections in PARF produced cell labeling in the vestibular and perihypoglossal nuclei bilaterally and labeled cells in the reticularis gigantocellularis (Rgc) and reticularis pontis caudalis (Rpc) nuclei contralateral to the injection site. Few labeled neurons were found in the caudal part of the paramedian pontine reticular formation (PPRF). In the mesencephalon, bilateral but more numerous ipsilateral labeled cells were found in the medial mesodiencephalic region including the nuclei of Cajal, Darkschewitsch and the posterior commissure. Injections placed caudomedial to VIn resulted in a characteristic concentration of labeled cells in the ipsilateral nucleus cuneiformis and rostral half of the contralateral superior colliculus (SC). Injections placed rostral to VIn in PARF produced cell labeling in the nucleus campi Foreli. The results are related to physiological evidence which suggests that PARF is an important premotor center for coordination of oculomotor, head and body movements.

Topographical organization of ascending cerebellar projections from the dentate and interposed nuclei in Macaca mulatta: an anterograde degeneration study.

Discrete electrolytic lesions were placed in the dentate and interpositus cerebellar nuclei in th monkey Macaca mulatta and anterograde degeneration was traced in the ascending limb of the brachium conjunctivum using the Wiitanen technique. The dentate and interpositus anterior (IA) nuclei project topographically onto the two somatotopically organized divisions of the red nucleus and the ventrolateral-ventral intermediate nuclei of the thalamus (VL-Vim). Caudal parts of the dentate project to face areas in medial parts of the parvocellular red nucleus (RNpc) and VL-Vim, rostrolateral parts to central arm areas in these nuclei. There is a small projection from the rostroventral dentate to lateral hindlimb areas in RNpc and VL. Dorsal parts of the dentate project to ventral parts of RNpc and VLc-Vim (distal limb musculature area) exclusively, whereas ventral parts of the dentate project to dorsal parts of RNpc with some overlap ventrally dorsomedial VLc (proximal limb musculature and premotor areas). Caudal parts of IA project to medial forelimb and rostral parts to lateral hindlimb areas in the magnocellular red nucleus (RNmc) and VL-Vim. Projections from IA are greatest to distal musculature areas in VL-Vim. No projections from the interpositus posterior nucleus (IP) to the red nucleus could be identified positively, but projections to VL-Vim are to proximal hindlimb musculature and premotor areas.