Search

OBJECTIVEInitial studies applying deep brain stimulation (DBS) of the posteromedial hypothalamus (PMH) to patients with pathological aggressiveness have yielded encouraging results. However, the anatomical structures involved in its therapeutic effect have not been precisely identified. The authors’ objective was to describe the long-term outcome in their 7-patient series, and the tractography analysis of the volumes of tissue activated in 2 of the responders.METHODSThis was a retrospective study of 7 subjects with pathological aggressiveness. The findings on MRI with diffusion tensor imaging (DTI) in 2 of the responders were analyzed. The authors generated volumes of tissue activated according to the parameters used, and selected those volumes as regions of interest to delineate the tracts affected by stimulation.RESULTSThe series consisted of 5 men and 2 women. Of the 7 patients, 5 significantly improved with stimulation. The PMH, ventral tegmental area, dorsal longitudinal fasciculus, and medial forebrain bundle seem to be involved in the stimulation field.CONCLUSIONSIn this series, 5 of 7 medication-resistant patients with severe aggressiveness who were treated with bilateral PMH DBS showed a significant long-lasting improvement. The PMH, ventral tegmental area, dorsal longitudinal fasciculus, and medial forebrain bundle seem to be in the stimulation field and might be responsible for the therapeutic effect of DBS.

Hypothalamic deep brain stimulation (DBS) has been used for more than a decade to treat cluster headache (CH) but its mechanisms remain poorly understood. The authors have successfully treated a patient with CH using hypothalamic DBS and found that the contact used for chronic stimulation was located in a white matter region posterior to the mammillary bodies. Fiber tracts crossing that region were the medial forebrain bundle and those interconnecting the hypothalamus and brainstem, including the dorsal longitudinal fasciculus. Because the stimulation of axons is an important mechanism of DBS, some of its clinical effects in CH may be related to the stimulation of fibers interconnecting the hypothalamus and brainstem.

The suprachiasmatic nucleus of the hypothalamus is the chief circadian pacemaker in the brain, and is entrained to day-night cycles by visual afferents from melanopsin containing retinal ganglion cells via the inferior accessory optic tract. Tracer studies have demonstrated efferents from the suprachiasmatic nucleus projecting to the paraventricular nucleus of the hypothalamus, which in turn project to first-order sympathetic neurons in the intermedio-lateral grey of the spinal cord. Sympathetic projections to the pineal gland trigger the secretion of the sleep inducing hormone melatonin. The current study reports the first demonstration of potential sympathopetal hypothalamic projections involved in circadian regulation in humans with in vivo virtual white matter dissections using probabilistic diffusion tensor imaging (DTI) tractography. Additionally, our data shows a correlation between individual differences in white matter microstructure (measured with fractional anisotropy) and increased daytime sleepiness [measured with the Epworth Sleepiness Scale (ESS, Johns, 1991)]. Sympathopetal connections with the hypothalamus were virtually dissected using designated masks on the optic chiasm, which served as an anatomical landmark for retinal fibres projecting to the suprachiasmatic nucleus, and a waypoint mask on the lateral medulla, where hypothalamic projections to the sympathetic nervous system traverse in humans. Sympathopetal projections were demonstrated in each hemisphere in twenty-six subjects. The tract passed through the suprachiasmatic nucleus of the hypothalamus and its trajectory corresponds to the dorsal longitudinal fasciculus traversing the periaqueductal region and the lateral medulla. White matter microstructure (FA) in the left hemisphere correlated with high scores on the ESS, suggesting an association between circadian pathway white matter microstructure, and increased daytime sleepiness.

Aims Gulf War Illness (GWI) is manifested as multiple chronic symptoms, including chronic pain, chronic fatigue, sleep problems, neuropsychiatric disorders, respiratory, gastrointestinal, and skin problems. No single target tissue or unifying pathogenic process has been identified that accounts for this variety of symptoms. The brainstem has been suspected to contribute to this multiple symptomatology. The aim of this study was to assess the role of the brainstem in chronic sleep problems and pain in GWI veterans. Materials and methods We enrolled 90 veterans (Age = 50 ± 5, 87% Male) who were deployed to the 1990–91 Gulf War and presented with GWI symptoms. Sleep quality was evaluated using the global Pittsburgh Sleep Quality Index. Pain intensities were obtained with the Brief Pain Inventory sum score. Volumes in cortical, subcortical, brainstem, and brainstem subregions and diffusion tensor metrics in 10 bilateral brainstem tracts were tested for correlations with symptom measures. Key findings Poorer sleep quality was significantly correlated with atrophy of the whole brainstem and brainstem subregions (including midbrain, pons, medulla). Poorer sleep quality also significantly correlated with lower fractional anisotropy in the nigrostriatal tract, medial forebrain tract, and the dorsal longitudinal fasciculus. There was a significant correlation between increased pain intensity and decreased fractional anisotropy in the dorsal longitudinal fasciculus. These correlations were not altered after controlling for age, sex, total intracranial volumes, or additional factors, e.g., depression and neurological conditions. Significance These findings suggest that the brainstem plays an important role in the aberrant neuromodulation of sleep and pain symptoms in GWI.

BACKGROUND The brainstem is one of the most challenging areas for the neurosurgeon because of the limited space between gray matter nuclei and white matter pathways. Diffusion tensor imaging-based tractography has been used to study the brainstem structure, but the angular and spatial resolution could be improved further with advanced diffusion magnetic resonance imaging (MRI). OBJECTIVE To construct a high-angular/spatial resolution, wide-population-based, comprehensive tractography atlas that presents an anatomical review of the surgical approaches to the brainstem. METHODS We applied advanced diffusion MRI fiber tractography to a population-based atlas constructed with data from a total of 488 subjects from the Human Connectome Project-488. Five formalin-fixed brains were studied for surgical landmarks. Luxol Fast Blue-stained histological sections were used to validate the results of tractography. RESULTS We acquired the tractography of the major brainstem pathways and validated them with histological analysis. The pathways included the cerebellar peduncles, corticospinal tract, corticopontine tracts, medial lemniscus, lateral lemniscus, spinothalamic tract, rubrospinal tract, central tegmental tract, medial longitudinal fasciculus, and dorsal longitudinal fasciculus. Then, the reconstructed 3-dimensional brainstem structure was sectioned at the level of classic surgical approaches, namely supracollicular, infracollicular, lateral mesencephalic, perioculomotor, peritrigeminal, anterolateral (to the medulla), and retro-olivary approaches. CONCLUSION The advanced diffusion MRI fiber tracking is a powerful tool to explore the brainstem neuroanatomy and to achieve a better understanding of surgical approaches. ABBREVIATIONS CN, cranial nerveCPT, corticopontine tractCST, corticospinal tractCTT, central tegmental tractDLF, dorsal longitudinal fasciculusHCP, Human Connectome ProjectML, medial lemniscusMLF, medial longitudinal fasciculusRST, rubrospinal tractSTT, spinothalamic tract.

The molecules and mechanisms involved in patterning the dorsoventral axis of the developing vertebrate spinal cord have been investigated extensively and many are well known. Conversely, knowledge of mechanisms patterning cellular distributions along the rostrocaudal axis is relatively more restricted. Much is known about the rostrocaudal distribution of motoneurons and spinal cord cells derived from neural crest but there is little known about the rostrocaudal patterning of most of the other spinal cord neurons. Here we report data from our analyses of the distribution of dorsal longitudinal ascending (DoLA) interneurons in the developing zebrafish spinal cord. We show that, although apparently distributed irregularly, these cells have cryptic organisation. We present a novel cell-labelling technique that reveals that DoLA interneurons migrate rostrally along the dorsal longitudinal fasciculus of the spinal cord during development. This cell-labelling strategy may be useful for in vivo analysis of factors controlling neuron migration in the central nervous system. Additionally, we show that DoLA interneurons persist in the developing spinal cord for longer than previously reported. These findings illustrate the need to investigate factors and mechanisms that determine "irregular" patterns of cell distribution, particularly in the central nervous system but also in other tissues of developing embryos.

We investigated the direct projections from the paraventricular hypothalamic nucleus (PVH) to the spinal cord. When Phaseolus vulgaris leucoagglutinin (PHA-L) was injected into the PVH, labeled descending fibers were observed running bilaterally through three pathways. The first pathway ran into the dorsal longitudinal fasciculus and projected to the central gray matter, Edinger-Westphal nucleus, pedunculopontine tegmental nucleus, nucleus of the locus ceruleus and parabrachial nucleus. The second and third pathways coursed through the medial forebrain bundle, ventral tegmental area, and ventral part of the medulla oblongata. At the medulla oblongata, the second pathway curved dorsolaterally and joined Rexed's lamina V of C1 after giving many projections to the nucleus ambiguus, nucleus of the solitary tract, dorsal motor nucleus of the vagus, and a few to the area postrema. The fibers descended through lamina V until C5, and coursed through lamina I from C6 to the upper coccygeal segments. They gave collateral projections to lamina I from the cervical through the upper coccygeal segments. The third pathway coursed laterally and descended through the lateral funiculus after giving projections to the lateral reticular nucleus and the marginal layer of the spinal trigeminal nucleus. These fibers gave off many projections to the intermediolateral cell column of the thoracic cord and the sacral parasympathetic nucleus. Lamina X received many projections from the fibers of the lateral funiculus at C5 through the b-segment of sacral spinal cord. These results indicate that the PVH may integrate directly with the medullary and spinal autonomic regulatory nuclei, including the vagus complex, sympathetic intermediolateral cell column, laminae I and X, and sacral parasympathetic nucleus.

NE is produced primarily by neurons in the locus coeruleus (LC), which is located at the pontomesencephalic junction at the floor of the fourth ventricle in the pontine region of the brain. Other nuclei with noradrenergic neurons include the lateral tegmentum and the nucleus of the solitary tract. Retrograde and ultrastructural examinations reveal that the core of the LC and its surrounding regions receive afferent projections from the insular cortex, the central amygdala, preoptic area, bed nucleus of the stria terminalis, hypothalamus, nucleus paragigantocellularis, prepositus hypoglossi, paraventricular nucleus, nucleus of solitary tract, lateral habenula, and cerebral cortex.– These afferent projections provide multiple neurochemical inputs to the LC with changes in LC neuronal firing being a highly coordinated event. Neurochemical interactions within the LC have been documented for numerous neurotransmitters including glutamate, gamma-aminobutyric acid (GABA), glycine, serotonin, dopamine, orexin, corticotropin-releasing factor (CRF), enkephalin, acetylcholine, and substance P.– Noradrenergic neurons of the LC express multiple receptor types for these various neurochemical inputs to the LC., Some of the neurochemical inputs to the LC have received a great deal of attention in the recent literature with regard to depression and these are considered in greater detail later.NE containing fibers from the LC innervate nearly the entire brain. These fibers leave the LC largely by five distinct efferent tracts. The central tegmental tract, central gray dorsal longitudinal fasciculus, and the ventral tegmental medial forebrain tract leave the LC to innervate the hypothalamus, limbic regions, thalamus, cortex, and olfactory bulb. A fourth pathway travels through the cerebellar peduncle to terminate in the cerebellum. A fifth tract descends into the spinal cord and mesencephalon. NE release can occur at individual synaptic boutons in terminal fields or through en passant connections along the noradrenergic axons, thus allowing NE to be released throughout innervated regions of the brain.

An 18-year-old boy with refractory epilepsy and aggressiveness associated to a hypothalamic hamartoma was submitted to a stereotactically guided lesion by thermocoagulation. The target was based on magnetic resonance (MR) images merged with computed tomography scan images taken on the day of surgery while patient was on a stereotactic frame. In order to reveal structures not discernible in MR images, the Schaltenbrand digital brain atlas was merged onto the patient's images. Target and trajectory of the depth electrode were chosen based on three-dimensional imaging reconstructions. A surgical plan was devised to disconnect the hypothalamic hamartoma from the hypothalamus, medial forebrain bundle, fasciculus princeps, and dorsal longitudinal fasciculus. Our target was placed at the inferior portion of the posterolateral component of the hamartoma, bordering the normal hypothalamus. The patient evolved with marked lessening of aggressiveness. Seizure frequency was reduced from several seizures per day to less than one tonic-clonic seizure during sleep per month and only two episodes suggestive of partial complex seizures during daytime. These results have remained consistent over a 24-month postoperative follow-up. Functional neuroanatomy of hypothalamic connections involved in seizure propagation and aggressive behavior was reviewed.

Hypothalamus The hypothalamus is the region of the mammalian brain that is most important in the coordination of behaviors essential for the maintenance and continuation of the species. Although the hypothalamus occupies only about 0.15% of the volume of the human brain, it plays a major role in the regulation and release of hormones from the pituitary gland, maintenance of body temperature, and organization of goal-seeking behaviors such as feeding, drinking, mating, and aggression. It is the primary center for the control of autonomic function. It is also the region of the brain that is essential for behavioral adjustments to changes in the internal or external environment (Figure 8.1). The hypothalamus is a very old structure with striking similarity between humans and lower animals. It is made up of a number of nuclei and scattered cell groups. Some hypothalamic cell groups control specific functions (e.g., blood pressure, heart rate, etc.) through the coordinated action of short intrahypothalamic connections. Other nuclei operate by projections to structures outside the confines of the hypothalamus. Anatomical and behavioral considerations The hypothalamus lies on either side of the walls of the third ventricle below the level of the hypothalamic sulcus (Figure 8.2, see Figures 9.2, 9.3, and 14.6). It is bounded in front (rostrally) by the lamina terminalis and optic chiasm, laterally by the optic tracts, and behind by the mamillary bodies. Some hypothalamic nuclei are continuous across the floor of the third ventricle.

Based upon our previous findings that a reciprocal relationship exists between intracranial self-stimulation (ICSS) and rapid eye movement (REM) sleep, we hypothesized that the ICSS neural network is a part of the REM sleep system and fires during REM sleep. A series of experiments was designed to test a corollary to this hypothesis, that some sites involved in triggering REM sleep are also ICSS sites. We explored five locations: the locus coeruleus (LC), the medial longitudinal fasciculus (MLF) (both posterior midbrain and pontine level), the dorsal longitudinal fasciculus (DLF) (midbrain), and the ventral raphe (pontine level). We obtained rate-intensity functions from each animal that displayed ICSS. High to very high (15,000 per hour) rates of ICSS were obtained at several dorsal brain sites. These sites included the LC, pontine aspects of the MLF, and posterior midbrain aspects of the MLF and DLF. ICSS from the LC is interpreted as evidence that noradrenergic ICSS sites are part of the REM sleep neural network.

The projection of enkephalin-like immunoreactive (EnkI) neurons in the ventromedial hypothalamic nucleus (VMH) in the rat was investigated using immunocytochemistry. These neurons project to the dorsal portion of the midbrain central gray matter (DCG), because the destruction of the VMH resulted in a marked ipsilateral reduction of EnkI fibers there. Destruction of the dorsomedial portion of the VMH caused the appearance of fibers accumulating EnkI ventrolateral to the lesion and these fibers could be traced to the ventrolateral portion of the VMH where many EnkI neurons were concentrated. Transection of the brain at the level of the premammillary nucleus, mammillary body and superior colliculus caused the appearance of fibers accumulating EnkI in the posterior hypothalamic nucleus (PH), central gray matter medial to the fasciculus retroflexus Mynert and in the DCG, respectively. We concluded that the axons from these neurons first project dorsomedially to situate in the PH. These fibers ran caudodorsally and entered into the dorsal longitudinal fasciculus in the DCG at the midbrain.

The adrenergic innervation of the thalamus, epithalamus, metathalamus, and subthalamus in the rat has been investigated by means of the recently introduced glyoxylic acid fluorescence method. Many areas of the thalamus and adjoining regions, that appear sparsely innervated by catecholamine (CA) fibers in specimens processed according to the standard Falck-Hillarp formaldehyde method, were found to be richly supplied with such fibres in the glyoxylic acid-treated specimens. Moreover, the glyoxylic acid method allows the tracing of the CA axons from the cell bodies up to the terminals, and in combination with stereotaxic lesions the following CA systems to the thalamus could be established: 1 The locus coeruleus system. Most of these axons ascend in the so-called dorsal tegmental bundle through the mesencephalon and the zona incerta into the medial forebrain bundle. From this bundle branches were traced along several routes, giving rise to extensive terminal systems in many thalamic, metathalamic and pretectal areas, most notably the anterior, ventral and lateral nuclear complexes, and the medial and lateral geniculate bodies. 2 The dorsal periventricular bundle, which constitutes a previously not described adrenergic component of the dorsal longitudinal fasciculus. This system originates in cell bodies (defined as the A11 cell group) in the dorsal raphe region, the central gray of the mesencephalon, and in the periventricular gray of the caudal thalamus. The axons ascend within the dorsal longitudinal fasciculus and give rise to a thalamic and hypothalamic periventricular system, projecting to medial and midline thalamic, epithalamic and pretectal regions. 3 Part of the terminals in the paraventricular thalamic nucleus was identified with a non-locus projection from cell bodies in the pontine or medullary reticular formation. 4 A system of delicate, probably dopamine-containing axons was revealed in the caudal thalamus, the zona incerta and the dorsal and anterior hypothalamus. This system probably originates in the dopamine cell bodies of the diencephalic A11 and A13 cell groups, forming a hitherto unknown intradiencephalic dopaminergic system. The adrenergic afferent systems to the thalamus can, to a large extent, be regarded as adrenergic components of known ascending reticular projections. The information on the adrenergic systems obtained with the glyoxylic acid method revealed new features of the organization of the thalamic projections from the brain stem reticular formation.

Summary Circulating prolactin levels were measured by radioimmunoassay after electrical stimulation of discrete sites in the forebrain of 420 lactating rats. The animals were prepared under initial anaesthesia, were decorticated except for frontal cortex and their brain stems were transected at the mid-collicular level. An ascending pathway for prolactin release, probably involving the dorsal longitudinal fasciculus, enters the caudal hypothalamus from the midbrain on a broad front. It passes rostrally in the dorsal periventricular area, moves laterally to the mediodorsal and then to the lateral hypothalamus and continues rostrally to the lateral preoptic area (POA) within the medial forebrain bundle. A descending pathway from prefrontal sulcal cortex passes through the striatum and claustrum to reach the diagonal band of Broca, septum and POA, where it merges with the ascending path. This common path then moves medially into the medial POA, and then caudally, and appears to terminate in the periventricular region of the anterior hypothalamic area (AHA). It is suggested that the ascending path may be operative in suckling-induced release of prolactin, that the descending path may be concerned with reinforcing this release and may also play a role in olfactory-mediated release of prolactin. The nature and possible mechanism of action of the pathway between AHA and median eminence is discussed.

To identify descending pathways with cardiovascular function originating in the hypothalamus, unilateral electrolytic lesions were made in adult cats at hypothalamic sites from which arterial hypertension and cardioacceleration were elicited by electrical stimulation. After a survival period of 5 to 11 days, histological sections of the hypothalamus, midbrain, pons, medulla, and spinal cord were stained using modifications of the Nauta techniques for degenerating axons and terminals. The main caudal projection from lesions in the dorsal hypothalamic region could be traced ipsilaterally into the periaqueductal gray and was found to continue in the region of the dorsal longitudinal fasciculus throughout the midbrain, pons, and medulla. Fibers terminated primarily in the Edinger-Westphal nucleus, in the inferior colliculus, in a region dorsolateral to the superior olivary nucleus, and in the parahypoglossal region. The degeneration patterns resulting from lesions in the posterior hypothalamic region and in the ventrolateral hypothalamic-medial forebrain bundle regions were essentially similar. Lesions in these regions demonstrated the existence of dorsomedial and ventrolateral descending pathways (i) through the medial hypothalamus and periaqueductal gray and (ii) through the medial forebrain bundle and ventrolateral mesencephalic reticular formation. The projection through the periaqueductal gray was similar to that traced in animals with lesions in the dorsal hypothalamic region, but dense terminal degeneration was seen bilaterally in the caudal aspect of the accessory cuneate nucleus and the medial aspect of the nucleus tractus solitarius. The medial forebrain bundle partially decussated just rostral to the red nucleus and descended bilaterally into the region of the ventral tegmental area of Tsai, where some fibers terminated. The pathway continued through and around the lateral reticular nucleus in the medulla and into the spinal cord. Dense terminal degeneration was noted in the intermediomedial nucleus primarily in segments from C8 to T3, but extended to the lower thoracic segments as low as T12. No terminal degeneration was observed in the upper cervical, lumbar, or sacral cord, or in the intermediolateral nucleus of the thoracolumbar cord. These results suggest the existence of two distinct descending pathways that may be involved in the mediation of cardiovascular responses elicited by stimulation of the hypothalamus.

The distribution of norepinephrine (NE) was chemically mapped in discrete regions of the rat habenula (Hb). In addition, the contribution of both peripheral sympathetic and central noradrenergic sources to habenula NE was quantified following selective deafferentations. It was found that NE is distributed hetereogeneously in discrete Hb regions. Furthermore, it was demonstrated that after superior cervical ganglionectomy a small (26%) but significant (P less than 0.05) decrease in NE occurred only in the medial Hb (MHb), while NE in the lateral Hb (LHb) remained unaltered. In contrast, deafferentation of central noradrenergic innervation, such as lesions in the ascending dorsal noradrenergic bundle (DNB), caused a marked decrease in NE in both MHb (53%) and LHb (50%). Lesioning the dorsal longitudinal fasciculus resulted in NE loss in both MHb and LHb by 31%. These data corroborate the histofluorescent demonstration of the noradrenergic innervation of the Hb.

1. The fibres 'of the intrapontine facial nerve do not only take the schematic course-the so-called first part, genu and second part, as hither-to said, but passing through or passing over all the sides of the abducens nucleus in various directions, surround the abducens nucleus and then assemble into the second part. 2. A part of the fibres of the facial nerve decussates at the genu. 3. Some of the fibres derived from the trapezoid body end in the motor nucleus of the facial nerve. 4. Some of the fibres derived from the lateral portion of the abducens nucleus seem to join the second part of the facial nerve. 5. Some of the fibres derived from the spinal tract of the trigeminal nerve end in the motor nucleus of the facial nerve. 6. A part of the fibres of the dorsal longitudinal fasciculus (Schutz's bundle) seems to join the second part of the facial nerve. 7. The Mendel's so-called upper facial nerve, originating in the nucleus of the oculomotor nerve, can be seen in my observation, too.

The effect of brain-stem lesions on lactation in the cat was studied. Lesions in the rostral mesencephalon or caudal hypothalamus, which involved the dorsal longitudinal fasciculus, caused persistent supression of milk secretion and involution of the mammary glands. This effect seems to be due to interruption of the afferent pathway for prolactin secretion. Control lesions in other parts of the brain caused only a transient impairment of lactation.

Summary The ascending pathways which convey mesencephalic stimuli capable of inhibiting gonadotropin secretion have been investigated. Stimulation by electrolytic deposition of iron (1 mA, 10 sec) from stainless steel unipolar electrodes was used. The inhibitory impulses originated by stimulating the midbrain are conveyed to the hypothalamus by the dorsal longitudinal fasciculus (DLF) and medial forebrain bundle (MFB). This was indicated by the facts that: (1) ovulation and LH secretion were blocked after stimulation of the DLF or the MFB at hypothalamic levels, and (2) transection of either MFB or DLF prevented the blockade of ovulation which follows stimulation of the ventral tegmental area (VTA). The blocking effect of VTA stimulation was also suppressed by transection of the medial corticohypothalamic tract (CHT), indicating that the hippocampus is also involved. On the other hand stimulation of the MFB resulted in blocked ovulation when the DLF or the CHT were interrupted independently, but not when both tracts were cut simultaneously, showing that the impulses can be conveyed to the hypothalamus by either one of these two pathways. Stimulation of the DLF exerts a blocking effect on ovulation independently of other hypothalamic afferents and depresses serum LH concentration in ovariectomized-estrogen primed rats in which the preoptic-hypothalamic connections have been interrupted, indicating that the inhibitory impulses have direct access to the hypophysiotropic area. It can be concluded that the mesencephalic inhibitory signals are transmitted to the hypothalamus directly via the DLF or through the MFB-hippocampus-CHT system.

The ascending projections of the central gray substance and adjacent tegmentum in the rat and cat were studied with the Fink-Heimer (1967) modification of the Nauta method. Following lesions of the circumaqueductal gray, degeneration was traced along the dorsal longitudinal fasciculus of Schutz, a reciprocal association pathway between the central gray and the hypothalamus, into the dorsal and posterior hypothalamus, midline and intralaminar regions of the thalamus, and the zona incerta of the subthalamus. Another group of fibers travelled ventrally from the central grey to reach the ventral tegmental area of Tsai; some of these fibers continued rostrally to terminate in the lateral hypothalamus. Other fibers distributed to the midbrain tegmentum as components of Weisschedel's radiatio grisea tegmenti. Lesions which extended into the adjacent tegmentum produced a larger quantity of degeneration than did lesions confined to the central gray, but the patterns of degeneration were similar in both cases. The ascending projections of the central gray in the rat and cat were found to have similar distributions but were slightly more massive in the cat. No evidence of fiber degeneration was detected in the ventromedial hypothalamus. Negative results, however, must be interpreted with caution, for they may be due to an excessively small caliber of axons and/or peculiar modes of axon termination in hypothalamic tissue.

An electrophysiological study has been made of pathways from mesencephalic central gray to forebrain in sixteen rabbits. Attention was focussed on those potentials evoked by electrical stimulation of central gray which, on the basis of latency (less than 7 msec), stability to high-frequency stimulation, and resistance to barbiturate anesthesia or anoxemia, were considered to be mediated by direct fibers. Such responses had a relatively limited distribution, being mainly confined to the hypothalamus, although they were also obtained from caudate nucleus and thalamus. Responses of a different type were recorded in the tuberal region of the hypothalamus, consisting of an initial small component of short latency (sometimes biphasic), followed by a broad wave. Whereas the initial components were depressed by barbiturate, the broad component increased in amplitude and duration. The present study suggests that the dorsal longitudinal fasciculus and periventricular-diencephalic system of fibers may be an important afferent pathway to the hypothalamus, including the rostral portion of the latter and the hypothalamohypophyseal system.

The midbrain course of descending pathways mediating sham rage behavior has been investigated in thalamic cats. Electrical stimulation of the lateral hypothalamus has been tested for rage outbursts before and following various lesions in the midbrain. The following lesions were found not to impair transmission of the descending hypothalamic influences: transection of the rostral third of the central periaqueductal gray and adjacent tegmentum, with interruption of the hypothalamic component of Schutz's dorsal longitudinal fasciculus; midline lesions, including the medial component of the medial forebrain bundle to the ventral tegmental area of Tsai and to the caudal periaqueductal gray; and lateral tegmental lesions involving the lateral component of the medial forebrain bundle to the lateral midbrain tegmentum. Large lesions involving most, though not all, of the tegmentum at the midcollicular level impaired, but did not abolish, rage responses to lateral hypothalamic stimulation. It is concluded that descending connections from the lateral hypothalamus, responsible for the peripheral manifestations of rage, are diffuse through the midbrain, including both the lateral and the medial components of the medial forebrain bundle, and possibly the dorsal longitudinal fasciculus as well.

Projections from the area postrema and adjacent parts of the medial solitary nucleus are demonstrated with the Nauta method following lesions limited exclusively to these structures. Experiments are controlled with lesions involving adjacent bulbar regions, cerebellum, and spinal cord. Ascending pathways in the dorsal and lateral columns of the spinal cord project ipsilaterally to the area postrema and bilaterally to a para-alar nucleus in the ventral periphery of the nucleus gracilis. Neurons in the area postrema project mainly inspilaterally to the dorsal and medial regions of the medial solitary nucleus. Neurons in the posterior half of the medical solitary nucleus project ipsilaterally to the lateral solitary nucleus, dorsal vagal nucleus, ambigus, retrofacial nucleus, and dorsal and lateral bulbar reticular formation. Projections to nuclei intercalatus and prepositus hypoglossi, bilaterally, and to the ipsilateral dorsal tegmental nucleus by way of the dorsal longitudinal fasciculus are also shown. No direct projections to the diencephalon are demonstrated. Control lesions in the dorsal column nuclei reveal projections to the contralateral inferior olive and thalamic reticular and ventrobasal nuclei, but not to the projection sites of the solitary nucleus. Evidence is given to support the hypothesis that ascening visceral pathways are interruped in the bulbar reticular formation and dorsal tegmental nucleus before reaching the diencephalon. Correlations are suggested with functional aspects of the central autonomic and reticular activating systems.

Twenty-two albino rats were divided into three groups for chronic brain electrode implantation. The effects of subsequent stimulation and ablation of the subcommissural organ (SCO), the bilateral lateral hypothalamic nuclei (LH), and the dorsal longitudinal fasciculus (FLD) on volume of water intake, and on concentration of urinary excretion were observed over twenty-two hour periods of free access to water both before and after treatments. It was observed that neither stimulation nor lesion of the SCO resulted in disruption of water intake or of urinary volume and concentration. Similar stimulation of LH potentiated drinking, while similar lesions produced a hypodipsia that necessitated tube feeding. Lesions of LH produced, in addition, marked elevation of urine volume, but no change in urine concentration. Sectioning of the FLD was followed by effects similar to those following lesions of the LH nuclei.

Electrolytic lesions centered in the reticular formation or partial hemisections of the brain stem were induced stereotaxically at various levels from the medulla to the diencephalon. Nauta and Fink-Heimer methods were used to reveal degenerated fibers ascending through the upper brain stem. Degenerated fibers coursing rostrally through the subthalamus and ventrally to enter the lateral hypothalamus at tuberal levels were observed after lesions at all levels rostral to the medulla. The trajectory of this ascending pathway through the midbrain and diencephalon was distinct from that of the dorsal longitudinal fasciculus which is medial and that of the medial forebrain bundle which is ventral. Several lines of evidence suggest the hypothesis that this pathway is a component of the ascending reticular system which carries sensory information into the hypothalamus for correlation with hypothalamic visceral and behavioral functions.

SUMMARY:Median eminence and ventromedial hypothalamus have in the past been the principal foci of research in neuroendocrine and neurovisceral control mechanisms. The present report provides an overview of work involving the dorsomedial hypothalamic nucleus (DMN). This structure is located dorsal to the ventromedial hypothalamic nucleus (VMN) and extends anteroposteriorly from the plane of the largest cross section of the VMN to the plane of the dorsal premammillary nucleus. Fibers from the DMN pass with the periventricular system and the dorsal longitudinal fasciculus of Schütz and have been traced to the midbrain tegmentum and reticular formation. Intrahypothalamic connections involve intensive networks between DMN, lateral hypothalamic nucleus (LHN) and VMN. Regarding neurotransmitters, recent studies indicate that the DMN receives noradrenergic innervation along two pathways, a dorsal and a ventral one. Monoamine-containing systems approach the DMN from the lateral hypothalamus and the bulk of these fibers are carried in the medium forebrain bundle from their cells of origin in the brain stem. Studies of the vascular supply indicate that both VMN and DMN receive their blood supply from the internal carotid artery. It has been recently demonstrated that the DMN is involved in the control of food intake and possibly water intake as well. Discrete lesions in the DMN have caused hypophagia and hypodipsia, and implantation of epinephrine and norepinephrine in this area has initiated eating. Many years ago, electrical stimulation of this area was reported to cause eating. Although DMN lesions cause hypodipsia, they do not result in the reduced water/food intake ratios that are so characteristic of the VMN syndrome. DMN lesions are also followed by reduced spontaneous activity (running wheel), but this reduced activity is not accompanied by increased weight gain and accretion of adipose tissue, the latter being consistently observed in the VMN rat. Rather, carcass fat remains normal in the DMN rat and carcass protein is either normal or slightly increased. Many of the aforementioned changes in weanling rats with DMN lesions, however, are not matched by similar alterations in the intermediary metabolism of carbohydrate and lipid. Possibly this is due to a “resetting” of a central autonomic control system that makes it possible for the DMN rat to adapt more efficiently to a reduced influx of substrate, i.e. the consistent hypophagia. From a review of the literature it appears that the DMN and their circuitry are involved in only a few neuroendocrine, i.e. hypothalamohypophyseal control mechanisms. Both lesion and cervical stimulation experiments suggest an involvement of the DMN in the control of LTH. Circumstantial evidence points to the DMN as a possible formation and/or storage site of growth hormone inhibiting factor (GIF). Although DMN rats show reduced ponderal and linear growth, they have been found to have normal or elevated plasma growth hormone (GH) levels. Both lesion and stimulation studies have yielded the impression that the DMN is not involved in thyroid, i.e., thyrotropin stimulating hormone releasing factor (TSHRF) control. Electrical stimulation of the DMN has been reported to result in a positive correlation between adrenal blood flow and adrenal corticoid release in hypophysectomized dogs. This has been interpreted as a coordinated response at the level of a “dorsomedial sympathetic vasodilator relay” rather than a “true” neuroendocrine effect via corticotropin releasing factor (CRF). Experiments that failed to demonstrate a relationship between the DMN and the tonic and cyclic control of luteinizing hormone releasing factor (LHRF) are discussed. The data reviewed indicate the existence in the dorsomedial hypothalamus of an area that exerts a profound influence on many aspects of neurovisceral and some neuroendocrine control systems.

The hypothalamus, a continuation of the central periaqueductal gray matter, comprises the center of the neuroendocrine system; it surrounds the third ventricle with groups of neurons anteriorly and central gray matter posteriorly. The periventricular fibers, the medial forebrain bundle, and the dorsal longitudinal fasciculus are important fiber tracts that pass through the region. Relevant to the third ventricle because of its location, the posterior hypothalamus lies between the midcommissural point and the posterior commissure, below the anterior commissure-posterior commissure (AC — PC) line and includes the area 5–6 mm from the midline.

In female rats anesthetized with urethane, 151 neurons in and around the ventromedial nucleus of the hypothalamus were identified by antidromic activation as having axonal projection to the mesencephalic central gray at the midcollicular level. Identified neurons were most numerous in the rostral part and at the borders of the nucleus. Antidromic spike latencies, constant for a given cell to stimulation with fixed intensity at a low repetition rate, had a wide range across cells (1.4–41.5 ms). In 37 cells, gradual increases in stimulus intensity allowed sudden discrete latency decreases as large as 9.8 ms. These may reflect activation of separate axonal branches of terminal arborizations. Eleven among 43 tested cells were antidromically driven from the dorsal longitudinal fasciculus (DLF) at the diencephalic-mesencephalic junction as well as from the central gray. Latencies of DLF responses were always shorter than those from central gray. From this and collision experiments between central gray-evoked and DLF-evoked antidromic spikes, it was concluded that at least one quarter of mesencephalic projections from the ventromedial nucleus descend through DLF. The mean conduction velocity of these axons was 0.8 m/s, indicating that they belong to thin unmyelinated C-group fibers. Thirty percent of the cell population studied received excitatory input from the cortical or medial nucleus of the amygdala. Four cells were identified as having projections both to the central gray and the amygdala. Estrogen treatment of ovariectomized female rats caused no major changes in antidromic latency, absolute refractory period or resting activity of these identified hypothalamic neurons. However, the stimulation threshold for antidromic activation was significantly lower in the estrogen-treated animals. Axons to the central gray from ventromedial hypothalamic neurons provide for hypothalamic bias on brain stem reflex paths, for reproductive and other behaviors.

Light microscope radioautography was used to visualize ascending serotonin (5-HT) systems in adult rat brain, following prolonged latero-ventricular instillation of [ 3 H]5-HT. In paraventricular and paracisternal regions, the cell bodies, axonal projections and terminal fields of 5-HT neurons were distinctly labelled, allowing comprehensive analysis of their organizational features. Two major systems of ascending 5-HT fibers could be defined: a transtegmental system originating mainly from nucleus raphe dorsalis (B-7 and B-6) and less prominently from nucleus raphe medianus (B-8), and a periventricular system predominantly issuing from the rostral pole of nucleus raphe dorsalis (B-7). The transtegmental system converges ventral-ward across the decussation of the superior cerebellar peduncles and sweeps rostral-ward, through the ventral tegmental decussation, to enter the ventral tegmental area. The periventricular system closely follows the dorsal longitudinal fasciculus of Schu¨tz. It branches off dorsally toward the superior and inferior colliculi and the subcommissural organ, but the bulk of its fibres arch abruptly ventralward, beneath the posterior commissure, to reach the dorsal hypothalamus. Along their way, these axons most likely contribute to the dense 5-HT innervation of the periventricular gray matter of the midbrain and caudal diencephalon. Both ascending 5-HT systems merge in the medial forebrain bundle area of caudal hypothalamus, wherefrom their fibres borrow several limbic pathways to reach distant territories of innervation: fasciculus retroflexus to medial habenula; striae medullaris and terminalis toward the habenula and the amygdala, respectively; fornix to hippocampus; diagonal band of Broca to septum; induseum griseum and cingulate bundle to hippocampus and neocortex. Other fibers spread more laterally, across the internal capsule, to innervate the globus pallidus and neostriatum. Finally, at the base of anterior forebrain, some 5-HT axons traverse the anterior olfactory nucleus and end in the glomerular layer of olfactory bulb. These and additional results concerning the heterogeneous distribution of 5-HT axonal varicosities in various parts of forebrain (e.g. thalamus, hypothalamus, septal area and amygdaloid complex) complement earlier data gathered by means of other neuroanatomical techniques and make it possible to present a useful model of the general organization of ascending 5-HT systems in the mammalian brain.

In 9 chloralosed spinal cats (C7) electrical stimulation of the posteromedial hypothalamus (PMH) on either side was found to inhibit the vagal bradycardia elicited by stimulation of the right nucleus of tractus solitarius (NTS) and right dorsal nucleus medullae oblongatae centralis (d-MOC) but not the bradycardia elicited from the right n. ambiguus (NA). In 21 chloralosed spinal cats, field potentials were recorded at 405 sites in the right and left medulla during stimulation of the ipsilateral and contralateral PMH. The field potentials were located in the inferior olivary n., lateral reticular n., NTS, ventral and d-MOC, NA, parahypoglossal area, and dorsal longitudinal fasciculus. The field potentials had a peak latency of 19–53 msec and did not follow frequencies greater than 26 Hz, were not affected by muscle paralysis, but were abolished by barbiturate and by asphyxia. Ipsilateral medullary field potentials were abolished by ipsilateral hemitransection at the pontomedullary junction whereas contralateral field potentials were reduced by ipsilateral transection and abolished by contralateral transection. It is concluded that the inhibitory effect of PMH stimulation on reflex vagal bradycardia is mediated by crossed and uncrossed pathways which alter the electrical activity of neurons in discrete medullary structures.

To examine the role of the ventral midbrain in the control of release of ACTH, we stimulated electrically 92 sites in the mesencephalon of 15 cats anesthetized with chloralose/urethane. Responses of arterial pressure could not account for change of release of ACTH. Three active areas were identified. First, in a dorsal facilitatory area that includes the dorsal longitudinal fasciculus, electrical stimulation led to changes in ACTH of +106, +117, and +90 pg/ml at 1.5, 3.5, and 6.5 min, respectively (P less than 0.05). Second, in a more ventral inhibitory area that includes the mammillary peduncle, electrical stimulation led to changes in ACTH of -63, -72, and -47 pg/ml, respectively (P less than 0.05). Third, in a ventral facilitatory area that includes the ventral tegmental area of Tsai, electrical stimulation led to changes in ACTH of +57, +56, and +59 pg/ml, respectively (P less than 0.01). The inhibitory and facilitatory areas of the ventral midbrain appeared to be continuous, respectively, with the inhibitory and facilitatory areas mediating control of ACTH in the dorsal rostral pons and in the hypothalamus. Anatomical evidence indicates projections from these ACTH-active areas of the midbrain and of the pons to ACTH-active areas of the hypothalamus. Thus, the present results suggest that the midbrain areas identified may represent pathways from ACTH-active areas of the pons to the hypothalamus.

Summary Connections to the paraventricular and supraoptic nuclei from cerebral structures rostral to the pons related to neurohypophysial hormones release were determined by milk-ejection, pressor, evoked potential and degeneration techniques. Milk-injection (oxytocic) responses were readily obtained from stimulating the hippocampal rudiment and nucleus accumbens. Oxytocic and vasopressor responses resulted from stimulation of the midbrain central gray, subthalamus and posterior hypothalamus. Evoked potentials having a latency of 7 msec or less were recorded from the paraventricular and supraoptic nuclei following stimulation of the former structures. Responses and latencies greater than 7 msec were registered in the supraoptic nucleus when the cingulate cortex, claustrum, putamen, medial lemniscus, superior colliculus and prectum were stimulated. Degeneration patterns from lesions in the hippocampal rudiment reveal direct projections to the paraventricular nucleus via the superior fornix anf fornix column but not to the supraoptic nucleus. Following lesions in the midbrain central gray, connections were traced to the hypothalamus and subthalamus via the dorsal longitudinal fasciculus, mammillary peduncle and medial forebrain bundle. From lesions in the subthalamus and posterior hypothalamus, degenerated fibers entered the medial forebrain bundle and projected to the supraoptic and paraventricular nuclei, respectively. These results show that the hippocampal rudiment, central gray, subthalamus and posterior hypothalamus are intimately involved in mechanisms of

Electrolytic lesions were stereotaxically placed in the dorsal and ventral areas of the mesencephalic periaqueductal gray in the cat. A dorsal angular approach was made so that mechanical electrode damage was contralateral to the lesion, or a posterior approach through the fourth ventricle was employed to avoid electrode damage. The brains were sectioned coronally and sagittally and stained with a modifield Nauta-Gygax or Fink-Heimer stain. Degeneration from the dorsal lesions was chiefly in a radial pattern to the superior colliculus, inferior colliculus and mesencephalic reticular area. Also, fibers angled out of the dorsal gray, crossed the midline and joined the commissure of the superior colliculus, some taking a ventrolateral course through the colliculi and others running ventrally along the borders of the gray. Degeneration was traced caudally to the cuneiform nucleus and adjacent reticular area. Rostrally, fibers traveled in the dorsal longitudinal fasciculus to the pretectal area, lateral habenular nucleus and finally, the posterior hypothalamic area. Ventral lesions showed the same radial pattern of degeneration and fibers in the superior colliculus commissure. Caudally, fibers could be traced to the cuneiform nucleus, reticular area and inferior olive. The rostral course of fibers in the dorsal longitudinal fasciculus was similar, though there were additional connections with the ventral tegmental area of Tsai, the fields of Forel, and the parafascicular nucleus of the thalamus.

Summary 1. Following unilateral removal of the cervical sympathetic ganglion of new-bom and young animals, we found that the periventricular fiber system at the hypothalamic level as well as the dorsal longitudinal fasciculus at the level of the midbrain showed marked degeneration homolaterally. As the dorsal longitudinal fasciculus (Schtutz) continues cranial-wards in the periventricular fiber system, it is elucidated that these fiber systems relate themselves with the peripheral sympathetic system. 2. After severing, of the vagus nerve the nerve cells of c zone Kurotsu's and the fiber system belonging to the medial forebrain bundle at the hypothalamic level got degenerated on the operated