Projections from the parabrachial nucleus to the vestibular nuclei: potential substrates for autonomic and limbic influences on vestibular responses

Brain Research 996 (2004) 126 – 137 www.elsevier.com/locate/brainres

Research report

Projections from the parabrachial nucleus to the vestibular nuclei: potential substrates for autonomic and limbic influences on vestibular responses Carey D. Balaban * Departments of Otolaryngology and Neurobiology, Eye and Ear Institute, University of Pittsburgh, 203 Lothrop Street, Pittsburgh, PA 15213, USA Accepted 20 October 2003

Abstract Previous anatomical studies in rabbits and rats have shown that the superior vestibular nucleus (SVN), medial vestibular nucleus (MVN) and inferior vestibular nucleus (IVN) project to the parabrachial nucleus (PBN) and Ko¨lliker – Fuse (KF) nucleus. Adult male albino rabbits and Long – Evans rats received iontophoretic injections of biotinylated dextran amine, Phaseolus vulgaris leucoagglutinin, Fluoro-Gold or tetramethylrhodamine dextran amine into either the vestibular nuclei or the PBN and KF nuclei. The results were similar in both rats and rabbits. Injections of retrograde tracers into the vestibular nuclei produced retrogradely labeled neurons bilaterally in caudal third of the medial, external medial, and external lateral PBN in both species, with more variable labeling in KF. Rats also had consistent bilateral (predominantly contralateral) labeling in the ventrolateral PBN. The most prominent labeling was produced from injections that included the SVN, with fewer labeled neurons observed from injections in the caudal MVN and the IVN. Anterograde transport of BDA from injections into the PBN and KF nuclei of rabbits revealed prominent projections to the SVN, dorsal aspect of the rostral MVN, caudal MVN, pars beta of the LVN and IVN. These connections appear to contain a component that is reciprocal to the vestibulo-parabrachial pathway and a nonreciprocal component to regions connected with the vestibulocerebellum and vestibulo-motor reflex pathways. These connections support the concept that a synthesis of autonomic, vestibular and limbic information is an integral property of pathways related to balance control in both the brain stem and forebrain. It is suggested that these projections may contribute broadly to both performance tradeoffs in vestibularrelated pathways during variations in the behavioral context and affective state and the close association between anxiety and balance function. D 2003 Published by Elsevier B.V. Theme: Motor systems and sensorimotor integration Topic: Vestibular system Keywords: Vestibular nucleus; Parabrachial nucleus; Anxiety; Balance

1. Introduction Recent anatomic and physiologic studies have demonstrated direct connections between the vestibular nuclei and brain stem regions that influence sympathetic and parasympathetic outflow (review: [8]). These pathways originate from a region within the vestibular nuclei, which includes the dorsal aspect of the superior vestibular nucleus (SVN), pars alpha (or caudoventral aspect) of the lateral vestibular

* Tel.: +1-412-647-2298; fax: +1-412-647-0108. E-mail address: [email protected] (C.D. Balaban). 0006-8993/$ - see front matter D 2003 Published by Elsevier B.V. doi:10.1016/j.brainres.2003.10.026

nucleus (LVN), and the caudal half of the medial vestibular nucleus (MVN) and the inferior vestibular nucleus (IVN) [3,8,40,41,46,47]. The caudal MVN and the IVN can influence parasympathetic and sympathetic outflow, either directly via projections to the brain stem or indirectly via relays in the parabrachial nucleus (PBN). Projections from the caudal MVN and IVN to the nucleus of the solitary tract and the rostral ventrolateral medullary reticular formation are likely to contribute to sympathetic components of responses to body movements with respect to gravity, such as blood pressure changes, heart rate changes and alterations in muscle sympathetic nerve activation [10,32]. Projections from the same vestibular nuclear regions to

C.D. Balaban / Brain Research 996 (2004) 126–137 Table 1 Rabbits: summary of locations of injection sites and retrograde labeling loci Case

Site in VN

Retrograde labeling in PBN and KF

95014 95016 98003 98006 98007 99013 99015 99019 20002 20004

S, Lb, Lg S, Lg, La M, Lb S, Lb, M Lb S S S S S, Lb

mpb(bi), em(bi), el(bi) mpb(bi), em(bi), el(bi), kf(bi) mpb(bi), em(bi) mpb(bi), em(bi), el(bi), kf (cont) mpb(bi), em(bi), el(i), lat(i) mpb(bi), em(bi) mpb(bi), em(bi), el(i) mpb(bi), em(bi), el(bi), kf(i) mpb(bi), em(bi), el(i) mpb(i), em(i), el(i)

Abbreviations: La, pars alpha of LVN; Lb, pars beta of LVN; Lg, pars gamma of LVN; M, medial vestibular nucleus; S, superior vestibular nucleus; mpb, medial parabrachial nucleus; em, external medial PBN; el, external lateral PBN; lat, lateral PBN; KF, Ko¨lliker – Fuse nucleus. The laterality of retrograde labeling is indicated parenthetically for each nuclear region: I, ipsilateral; cont, contralateral; bi, bilateral.

preganglionic parasympathetic neurons in the dorsal motor vagal nucleus and nucleus ambiguous have been suggested to contribute to alterations in gastrointestinal function and

127

‘vasovagal’ features of vestibular disorders. A recently described projection from the SVN, MVN and LVN (pars alpha) to preganglionic parasympathetic neurons innervating the eye [6] may contribute to accommodative vergence and papillary constriction during linear vestibulo-ocular reflexes and to intraocular blood flow control during postural shifts. An ascending pathway also originates from the dorsal aspect of the SVN, pars alpha of the LVN, and the caudal half of the MVN and the IVN. This ascending projection terminates densely in a caudal, vestibulo-recipient region of the PBN [3,11,40], in a region that includes the medial, external medial and external lateral parabrachial nuclei and the Ko¨lliker – Fuse (KF) nucleus. Neurons in this region respond to whole body angular velocity and position (relative to gravity) in alert monkeys, indicating that these neurons receive both semicircular canal- and otolith organderived signals [11]. The presence of vestibular responses is significant because the PBN forms a bi-directional link between brain stem autonomic and telencephalic structures: it has reciprocal connections with the amygdala, hypothal-

Fig. 1. Charting of four BDA injection sites in the vestibular nuclei that produced retrograde labeling in the PBN. The sites, defined as both the dense core and halo region, are indicated by case numbers and are charted on a standard series of camera lucida drawings of transverse sections from a paraffin embedded reference brain (sectioned at 10 Am). The sections are arranged in series from a middle level of the vestibular nuclei (a, lower right) to the rostral pole of the SVN (g, upper left). The nomenclature for the vestibular nuclei includes the SVN (S), MVN (M), LVN pars alpha (La), LVN pars beta (Lb), LVN pars gamma (or Deiters nucleus, Lg), and group y (y). The inferior cerebellar peduncle (ICP) is also indicated in this figure.

128

C.D. Balaban / Brain Research 996 (2004) 126–137

amus and prefrontal cortex and descending projections to autonomic output and spinal pathways. In particular, interconnections between the PBN, central amygdaloid and infralimbic and prefrontal cortex are believed to be important for the development and expression of conditioned aversion and fear responses, and panic disorder [17,24, 34]. Hence, it has been proposed that these structures may also be a substrate for the close clinical linkage between

balance disorders and panic with agoraphobia [4,9,10]. This study demonstrates the existence of a descending projection from the PBN to the vestibular nuclei in the two species with most extensively studied PBN connections, rats and rabbits. This parabrachio-vestibular pathway may provide integrated sensory and limbic contextual information to vestibular nucleus neurons that influence autonomic and affective responses.

Fig. 2. Camera lucida drawings of transverse sections through the ipsilateral caudal third of the parabrachial region after an injection of BDA into the SVN. Sections are arranged from rostral (A) to caudal (B). Anterogradely labeled axons and retrogradely labeled cell bodies are illustrated in the medial parabrachial (m) and external medial parabrachial (em) nuclei and intercalated among fibers of the SCP. The borders of the ventrolateral PBN (vl) are also shown.

C.D. Balaban / Brain Research 996 (2004) 126–137

2. Materials and methods 2.1. Surgical procedures The experimental protocols were reviewed and approved by the University of Pittsburgh Institutional Animal Care and Use Committee. 2.1.1. Rabbits In addition to a series of animals produced for this study, this report includes data from animals utilized in previous studies of the organization of vestibular nuclear output pathways [3,8]. New Zealand white rabbits (2.7 – 4.4 kg body weight) were premedicated with atropine methylnitrate (0.06 – 0.2 mg, s.c.) and anesthetized with sodium pentobarbital (40 mg/kg, 4 ml/kg total volume, i.v.). The rabbits were given 13% mannitol (2.3 – 3.0 ml/kg, i.v.) to increase the exposure of the floor of the fourth ventricle by osmotically shrinking the cerebellum and brain stem. Sur-

129

gical procedures were performed under aseptic conditions. The head was fixed with zygoma clamps in a stereotaxic apparatus (Narishige Instruments, Tokyo, Japan) with the head tilted 45j nose-down. Lidocaine (1 –2%, s.c.) was injected along the incision line. The cervicoauricularis muscle aponeurosis was divided and the underlying muscles were retracted to expose the occipital bone, atlas and atlanto-occipital membrane. The atlanto-occipital membrane was removed and the foramen magnum was enlarged with rongeurs medulla to expose the medulla and posterior aspect of the cerebellum. Using the obex and the floor of the fourth ventricle as landmarks, rabbits were given an iontophoretic injection of Phaseolus vulgaris leucoagglutinin (PHAL, Vector Laboratories, 2.5% solution in sodium phosphate buffered saline (PBS), pH 8.0) and/or biotinylated dextran amine (BDA, 10,000 MW, Molecular Probes, 7 –10% solution in PBS, pH 7.0) into the vestibular nuclei and/or nucleus prepositus hypoglossi (10 – 15 Am tip diameter, 4 AA positive current,

Fig. 3. Photomicrographs of retrogradely labeled cells in the PBN. (A) Large multipolar neuron in the rabbit medial PBN, labeled retrogradely after a BDA injection into the ipsilateral SVN. (B) Medium-sized multipolar neuron in the rabbit medial PBN, labeled retrogradely after a BDA injection into the ipsilateral SVN. (C) Medium-sized neuron in the rabbit medial PBN, labeled retrogradely after a BDA injection into the contralateral SVN. (D) Medium-sized neuron in rabbit medial PBN labeled retrogradely from a BDA injection into the ipsilateral SVN. (E) Fluoro-Gold labeled neurons in the rat external medial PBN labeled retrogradely from the ipsilateral vestibular nuclei. (F) A cluster of retrogradely labeled neurons in the rat external PBN contralateral to a vestibular nuclear injection of Fluoro-Gold. Calibration bar: 100 Am in A – D, 200 Am in E – F.

130

C.D. Balaban / Brain Research 996 (2004) 126–137

10 min). This communication is restricted to data from sites confined to the vestibular nuclei and nucleus prepositus hypoglossi, with no evidence of spread to the cerebellum, nucleus tractus solitarius or the dorsal medullary reticular formation. After completion of the injections, the craniotomy was packed with Gelfoam or Surgicel and the soft tissues were sutured in layers. Postsurgical analgesia was provided with a single injection of ketaprofen (2 mg/kg, s.c.). Penicillin

(80,000 – 100,000 U/day, i.m.) was administered during the survival period if a break in sterility was suspected during the surgery. 2.1.2. Rats Adult male Long –Evans rats were anesthetized with sodium pentobarbital (25 mg/kg, i.p.) combined with either Innovar Vet (0.02 mg/kg, i.m.) or ketamine (75 mg/kg, i.m.). Two surgical approaches were used: (1) a

Fig. 4. Chartings of retrogradely labeled neurons in the rabbit parabrachial nuclear complex after injections of BDA in the vestibular nuclei. The respective injection sites are shown in Fig. 1. The labeling is charted on a series of transverse sections from caudal (lower section) to rostral (upper section) through the caudal third of the parabrachial nuclear region. Note bilateral labeling in the medial parabrachial (m), external medial parabrachial (em) and external lateral parabrachial (el) nuclei. The locations of the lateral (l) and ventrolateral (vl) parabrachial and KF nuclei, the mesencephalic trigeminal nucleus (5m), LC and the SCP are also shown.

C.D. Balaban / Brain Research 996 (2004) 126–137

stereotaxically guided injection through a burr hole in the dorsal surface of the cranium or (2) a direct approach through an occipital craniotomy. Micropipettes (20 – 40 Am O.D.) filled with either 10% Fluoro-Gold or 10% tetramethylrhodamine dextran amine in PBS were introduced to make iontophoretic injections bilaterally (4 –5 AA DC, tip positive, 7 –10 min). The craniotomy was then packed lightly with Gelfoam and the skin was sutured.

131

2.2. Histological, immunohistochemical and histochemical procedures 2.2.1. Rabbits After survival times ranging from 4 to 10 days, the rabbits were euthanized with a pentobarbital overdose and perfused transcardially with PBS followed by the paraformaldehyde – lysine – sodium metaperiodate (PLP) fixative of McLean and Nakane [37]. The brains were post-fixed for

Fig. 5. Camera lucida drawing of anterograde transport of BDA from the caudal aspect of the external medial and external lateral parabrachial nuclei to the vestibular nuclei. The sections are arrayed from caudal (lower section) to rostral (upper section). The nomenclature for the vestibular nuclei includes the SVN (S), MVN (M), LVN pars alpha (La), LVN pars beta (Lb), and IVN (I). Nucleus prepositus hypoglossi (PH) and nucleus tractus solitarius (NTS) are also indicated in this figure.

132

C.D. Balaban / Brain Research 996 (2004) 126–137

18 –24 h at 4 jC in a solution of 4% paraformaldehyde– 30% sucrose in 50 mM phosphate buffer and cryoprotected in a 30% sucrose – 50 mM phosphate buffer solution for 2 – 3 days. Frozen sections (40 Am, transverse plane) were cut on a sliding microtome and sets of every fourth to sixth section were placed in 50 mM phosphate buffer (pH 7.2– 7.4). For longer term storage, sections were maintained at 20 jC in a solution of 30% sucrose –30% ethylene glycol solution in 50 mM phosphate buffer. Axonally transported PHA-L was visualized immunohistochemically by standard published methods [3]. For visualizing BDA transport, free-floating 40 Am frozen sections were rinsed successively in distilled water (3  10 min), 0.9% H2O2 and distilled water to suppress endogenous peroxidase activity, followed by a preincubation for 2 h at room temperature in 0.5% Triton X-100 in PBS. After a rinse in PBS, the sections were incubated for 1 h in VectastainR ABC peroxidase (avidin – biotin conjugate horseradish peroxidase) reagent (Vector Laboratories), rinsed in buffer and reacted for visualizing sites of peroxidase activity with either a nickel-enhanced DAB or a standard DAB (2 mg DAB, 8.3 Al H2O2 (ACS reagent grade, 30.8%, Sigma Chemical Co.) in 10 ml 500 mM sodium acetate buffer, pH 6.0) chromagen. Sections were mounted on subbed slides, dehydrated through a graded alcohol series, cleared in xylene and coverslipped with either permount or non-fluorescent DPX (Fluka). 2.2.2. Rats After survival times ranging from 2 to 3 days, rats were euthanized by sodium pentobarbital overdose) and perfused transcardially with PBS followed by a paraformaldehyde – lysine– periodate (PLP) fixative solution [37]. The brains

were removed from the cranium and cryoprotected for at least 2 days at 4 jC in a 50 mM phosphate-buffered, 4% paraformaldehyde solution containing 30% sucrose. The brains were then sectioned in the transverse plane at a thickness of 40 Am on a sliding microtome equipped with a dry ice freezing stage. Sections were mounted on gelatinchromalum subbed slides, dried, cleared in xylene and coverslipped with DPX. Consistent with NIH requirements, the procedures in these rat and rabbit studies have been reviewed and approved by the University of Pittsburgh Institutional Animal Care and Utilization Committee.

3. Results Ten rabbits had BDA injections confined to the rostral half of the vestibular nuclei, with no evidence of spread into the cerebellar white matter, superior cerebellar peduncle (SCP), reticular formation, locus coeruleus (LC) or the caudal pole of the parabrachial nuclear complex. The injection sites involved the SVN, MVN, and both pars beta and pars gamma of the LVN. The sites and findings are summarized in Table 1; four representative sites are shown in Fig. 1. Since all ten cases displayed the same basic pattern of anterograde and retrograde labeling in the PBN complex, four representative cases have been illustrated to show the range of variability in retrograde labeling between animals (Figs. 1, 2 and 4). Fig. 2 shows a camera lucida drawing of the ipsilateral PBN after an injection of BDA in the SVN. Anterogradely labeled varicose axons and terminal arborizations were located caudally within the medial parabrachial, external

Fig. 6. Photomicrographs of rabbit parabrachiovestibular axons labeled anterogradely with BDA. Panels A and B show larger caliber fibers that contributed terminal varicosities and en passage varicosities in the SVN. Panel C shows an example of a smaller caliber parabrachiovestibular fiber in the SVN. Panel D shows a larger caliber fiber parabrachiovestibular fiber and terminals in the rostral MVN (rMVN). Calibration bar: 50 Am.

C.D. Balaban / Brain Research 996 (2004) 126–137 Table 2 Rats: summary of locations of injection sites and retrograde labeling loci Case

Site in VN

Retrograde labeling in PBN and KF

329A

S, La, Lg, I, y (grazed Lb, M) La, I Mc, I Mc, I M, Lb (grazed La) M, La, Lb, Lg, I S, M, La, Lg Mc, La, I

m (bi), em (I), vl (cont)

329C 607B 711B 712B 712C 712D 396113

m m m m m m m

(bi), em (bi), el (bi), vl (bi) (i), em (i), el (i), vl (cont), kf(i) (i), vl (i) (bi), em (bi), el (bi), vl (bi) (i), em (i), el (i), vl (i), kf(i) (i), em (i), el (i), vl (cont), kf(i) (bi), em (bi), el (bi), vl (bi), kf(bi)

Abbreviations: I, inferior vestibular nucleus; La, pars alpha of LVN; Lb, pars beta of LVN; Lg, pars gamma of LVN; M, medial vestibular nucleus; S, superior vestibular nucleus; y, group y; m, medial PBN; em, external medial PBN; el, external lateral PBN; vl, ventrolateral PBN; KF, Ko¨lliker – Fuse nucleus. The laterality of retrograde labeling is indicated parenthetically for each nuclear region: I, ipsilateral; cont, contralateral; bi, bilateral.

medial parabrachial, external lateral parabrachial and KF nuclei. Retrogradely labeled neurons were also interspersed within this terminal region, ranging from heavily labeled multipolar neurons (Fig. 3a and b) to more lightly labeled cuboid (Fig. 3c) and fusiform (Fig. 3d) cells. The labeled somata confined almost exclusively within the medial parabrachial, external medial parabrachial and external lateral parabrachial nuclei (Fig. 4). These labeled neurons were present bilaterally, with more labeled cells ipsilateral than contralateral to the iontophoretic injection site. Significantly, the retrogradely labeled neurons were confined to the region of the parabrachial nuclear complex that contained anterogradely labeled fibers in each animal. 3.1. Anterograde and retrograde transport from PBN Two rabbits had iontophoretic injections of BDA confined within the parabrachial nuclear complex. These ani-

133

mals displayed both anterogradely labeled axons and retrogradely labeled somata in the vestibular nuclei. The distribution of retrogradely labeled neurons reproduced the pattern reported previously [3]. The pattern of anterograde transport to the vestibular nuclei is summarized in a series of camera lucida drawings from an injection centered in the external medial and external lateral subnuclei of the PBN in Fig. 5. Anterogradely labeled axons were traced caudally from the injection site to the dorsal border and the dorsolateral margin of the rostral pole of the SVN. The fibers along the dorsal border continued caudally, forming both en passage varicosities and terminal varicosities in the medial aspect of the SVN and in the rostral MVN. Coarse axons formed both terminal varicosities and varicosities en passage in the neuropil and near somata in superior (Fig. 6a and b) and rostral medial (Fig. 6d) vestibular nuclei. Finer caliber fibers formed a more extensive plexus in both regions (Fig. 6c). Some fibers continued caudally from the dorsolateral margin of the SVN to form en passage and terminal varicosities in the lateral vestibular (pars alpha and beta), caudal medial vestibular and inferior vestibular nuclei. The distal branches of these caudally projecting fibers were almost exclusively of fine caliber. A few anterogradely labeled axons were also observed in the contralateral vestibular nuclei. These fibers originated from large caliber axons that entered the ipsilateral SCP and traveled caudally, dorsally and medially to enter the cerebellar white matter. These fibers followed a trans-cerebellar course similar to crossed parabrachio-PBN projections in the subfastigial bundle in rats [38]. The axons traveled medially, decussated in the white matter rostral to the fastigial nucleus and entered the medial aspect of the contralateral SCP. The fibers then turned caudally and contributed a sparse projection to the contralateral vestibular

Fig. 7. Charting of injection sites in the rat vestibular nuclei that produced retrograde labeling in the caudal aspect of the parabrachial nuclear region. The sections are arranged from caudal (lower drawing) to rostral (upper drawing). Abbreviations are identical to Fig. 1.

134

C.D. Balaban / Brain Research 996 (2004) 126–137

nuclei, following the same pattern as the ipsilateral projections (Fig. 5). 3.1.1. Rats This analysis is based upon the results from eight rats with Fluoro-Gold injections confined to the vestibular nuclei (Table 2). Retrogradely labeled neurons were observed in the medial (m), external medial (em), ventral lateral (vl), and external lateral (el) parabrachial subnuclei and the KF nucleus. Examples of labeled neurons in the external medial and external lateral parabrachial nuclei are shown in Fig. 3e and f. There were no discernable differences in the distribution of retrogradely labeled neurons associated with

differences in the locations of the injection sites. All cases displayed retrogradely labeled neurons ipsilaterally in the medial parabrachial nuclei; half of the cases also showed contralateral labeling. Four representative injection sites are shown in Fig. 7 and the distributions of retrogradely labeled neurons in the PBN are charted in Fig. 8. The retrogradely labeled neurons were all distributed in the caudal third of the parabrachial nuclear complex, corresponding to the region that receives vestibular nucleus input [40]. Seven of the eight cases had retrogradely labeled neurons in the ipsilateral external medial and external lateral parabrachial nuclei, with three cases also showing bilateral labeling. The laterality of retrograde labeling in the ventral lateral PBN

Fig. 8. Retrogradely-labeled parabrachiovestibular somata are charted on a series of transverse sections through the rat PBN. Each case is charted on a series of spaced sections through the caudal half of the PBN. The lower section in each series is at approximately ear bar 0.80, the middle section is at level ear bar 0.50 and the upper section is at approximately level ear bar 0.20. The abbreviations are identical to Fig. 4.

C.D. Balaban / Brain Research 996 (2004) 126–137

showed more inter-animal variability: it was bilateral in three cases, strictly ipsilateral in two cases and strictly contralateral in three cases. A few retrogradely labeled cells were present in the caudal aspect of KF in only four rats (three strictly ipsilateral to the injection and one bilaterally).

4. Discussion This study has demonstrated the existence of descending projections from the caudal aspect of the PBN (and, to a lesser extent, the KF nucleus) to the vestibular nuclei. Retrograde tracing data indicated that these projections originate bilaterally from the medial, external medial and external lateral parabrachial subnuclei in both rats and rabbits, with a less prominent contribution from the caudal aspect of the KF nucleus. In rats, there was also a contribution from the caudal aspect of the ventrolateral parabrachial subnucleus. The significance of this apparent species difference is unclear. All of these regions of the parabrachial nuclear complex receive afferents from the vestibular nuclei in the respective species [3]; corresponding PBN regions of alert primates respond to whole body rotation, displaying sensitivities to rotational velocity and static position that are consistent with dynamic response properties of vestibular nucleus neurons [5,11]. Further, these PBN regions all contain neurons that project to the central amygdaloid nucleus in the respective species; the origin of projections to the central amygdaloid nucleus includes the ventrolateral parabrachial subnucleus in rats [23,30] but not rabbits [31]. Hence, despite the species differences in projections from the ventrolateral subnucleus, parabrachiovestibular projections appear to originate from groups of neurons with similar connectivity in both rats and rabbits. The projections from the caudal parabrachial and KF nuclei to the SVN, rostral MVN, IVN and caudal MVN are consistent with reciprocal connections between a component of the parabrachiovestibular pathway and vestibular nucleus regions that project to the PBN. This type of PBN-brainstem reciprocal connectivity was reported previously by Herbert et al. [29]. They found reciprocal connections between the ‘respiratory part’ of nucleus of the solitary tract (dorsal respiratory group) and the KF nucleus and between the rostral ventrolateral reticular nucleus, periambiguus region and parvicellular reticular area and the parabrachial and KF nuclei. These brainstem connections share the property of involvement in sensorimotor integration, either for automatic movements (e.g. respiration) or autonomic control. The ascending vestibulo-autonomic path to mediate the automatic panic-like aspects associated with falling (or a perception of falling) and the clinical linkage between balance disorders and panic disorder with agoraphobia [4,9]. Hence, a reciprocal pattern of organization of vestibuloparabrachial and parabrachiovestibular connections is consistent with parabrachial connections with other brainstem sensorimotor integration pathways for relatively automatic responses.

135

The projections to the SVN, rostral MVN, IVN and caudal MVN are consistent with reciprocity between a component of the parabrachiovestibular pathway and vestibuloparabrachial connections [3,40]. These connections are likely to influence information processing in the ascending vestibulo-autonomic pathway. However, it is likely that these connections function as more than a reciprocal processing loop in the ascending vestibulo-autonomic path. Parabrachiovestibular projections to the IVN and caudal MVN also have the potential to influence (1) descending vestibulo-autonomic projections to the solitary nucleus, nucleus ambiguus/parambiguus, rostral ventrolateral medulla and lateral medullar tegmentum [8,40] and (2) vestibulospinal motor projections to abdominal musculature [14,48]. These connections of the vestibular-related regions of the PBN are consistent with the view that they are involved in coordinating somatic, autonomic and affective responses to linear and angular acceleration challenges to factors as diverse as blood distribution, respiratory movements and control of body segments (review: [10]). The primary sites of origin of parabrachiovestibular connections (the caudal aspect of the medial, external medial, external lateral and ventral lateral parabrachial subnuclei) have three common features: they receive afferents from the vestibular nuclei [3,40], paratrigeminal nucleus [22,42] and the infralimbic and insular cortex [39]. Because the paratrigeminal nucleus receives chemoceptive, mechanoreceptive and nociceptive afferents from the oral cavity, nasal cavity and pharynx [42], it seems reasonable to suggest that a global characteristic of parabrachiovestibular projection regions may be integration of information about head motion, oropharyngeal (i.e. head-referenced) visceral and somatic sensation, and descending signals from ‘limbic’ cortex. Nested within the termination region of these common input sources, though, are smaller regions that were classified by Herbert et al. [29] as sites showing predominant gustatory or respiratory patterns of connectivity. The parabrachiovestibular projections from ‘gustatory’ PBN regions may constitute a head-centered representation for information processing within vestibulo-autonomic pathways. The overlap between the vestibulo-recipient and gustatory PBN regions encompasses the caudal aspects of the medial parabrachial and external medial parabrachial subnuclei. Anatomical data indicate that this region receives convergent vestibular information and oropharyngeal chemoceptive, mechanoreceptive and nociceptive inputs from both the rostral pole of the nucleus of the solitary tract [29] and the paratrigeminal nucleus [42]. Infralimbic and insular cortex also contribute projections to these parabrachial subnuclei [39]. Since the gustatory region of the PBN mediates the development of conditioned taste aversions [25 –27,44], it is possible that this vestibulo-oropharyngeal reciprocal circuit may contribute to both the development of conditioned taste aversions with environments evoking motion sickness and, subsequently, the detection of adequate stimuli to trigger the conditioned response.

136

C.D. Balaban / Brain Research 996 (2004) 126–137

By contrast, the participation of respiratory regions of the parabrachial complex in reciprocal connections with the vestibular nuclei may constitute a torso and airway-related representation within vestibulo-autonomic pathways. These regions of the parabrachial complex include the external lateral parabrachial, ventral lateral parabrachial and caudal KF nuclei. They receive projections from the ventrolateral nucleus of the solitary tract (or ‘‘dorsal respiratory group’’) [29], the medial nucleus of the solitary tract (a visceral sensory relay region) [29] and the amygdala [29,39]. The ventral lateral PBN also receives projections from the infralimbic and insular cortex [39]. A convergence of respiratory-related, visceral sensory and vestibular information in this reciprocal circuit would be a logical contributor to phenomena such as the entrainment of the respiratory cycle with otolith organ stimulation during off-vertical axis rotation [32] or the entrainment of respiratory movements with locomotor activity [15]. In a more general sense, though, muscles involved in ventilation are active during both postural adjustments and activities as diverse as defecation, deglutition, vocalization, and emesis [10]. Therefore, their patterns of activation reflect dynamic trade-offs between voluntary and automatic task demands; for example, the trading off a less urgent need to take a breath for a more urgent desire to speak or to generate a postural response to a slip on an icy street. The multimodal information processing in reciprocal vestibuloparabrachial connections is a candidate mechanism for matching the appropriate responses to a complex behavior context. The parabrachiovestibular projection also extends to a region of the vestibular nuclei beyond the location of vestibuloparabrachial cells and their dendritic fields. This is particularly evident for projections to pars alpha and pars beta of the LVN, which are not involved in vestibuloparabrachial pathways. Further, much of the projection field in the inferior and caudal medial vestibular nuclei lies outside the region of dendrites and somata of cells that project to either the PBN or the solitary nucleus [3,8]. These terminal regions, though, are likely to be associated with flocculonodular lobe terminal regions (e.g. [2,28]), related vestibulo-ocular reflex pathways (e.g. [1,2,7]), the origins of vestibular projections the anteromedian nucleus [6] and sites of origin of secondary vestibulo-flocculonodular lobe pathways [12]. The mechanisms are currently unknown for well-documented phenomena such as modulation of vestibulo-ocular reflex performance as a function of ‘arousal’ [19 –21], ‘mental set’ [18] and a subject’s frame of reference [13], context-dependence of vestibulo-ocular reflex adaptation [16] and [33,43,45] context-related alterations of velocity storage characteristics of vestibulo-ocular and optokinetic responses [35,36]. As in the case the reciprocal component of vestibulo-parabrachial connections, it is suggested that these projections may contribute broadly to performance tradeoffs in vestibular-related pathways during variations in the behavioral context and affective state.

Acknowledgements The author wishes to thank Maria Freilino, Gloria Limetti and Jean Betsch for expert surgical and histological assistance. These studies were supported by R01 DC00739 and P01 DC03417. A Core Grant for Vision Research (EY08098) provided technical support for maintenance of critical laboratory equipment.

References [1] C.D. Balaban, Olivovestibular and cerebellovestibular connections in albino rabbits, Neuroscience 12 (1984) 129 – 149. [2] C.D. Balaban, Distribution of inferior olive projections to the vestibular nuclei in albino rabbits, Neuroscience 24 (1988) 119 – 134. [3] C.D. Balaban, Vestibular nucleus projections to the parabrachial nucleus in rabbits: implications for vestibular influences on autonomic function, Exp. Brain Res. 108 (1996) 367 – 381. [4] C.D. Balaban, Vestibular autonomic regulation, Curr. Opin. Neurol. 12 (1999) 29 – 33. [5] C.D. Balaban, Neural substrates linking balance control and anxiety, Physiol. Behav. 77 (2002) 469 – 475. [6] C.D. Balaban, Vestibular projections to the Edinger – Westphal and anteromedian nuclei of rabbits, Brain Res. 963 (2003) 121 – 131. [7] C.D. Balaban, G. Beryozkin, Organization of vestibular nucleus projections to the dorsal cap of Kooy in rabbits, Neuroscience 62 (1994) 1217 – 1236. [8] C.D. Balaban, G. Beryozkin, Vestibular nucleus projections to nucleus tractus solitarius and the dorsal motor nucleus of the vagus nerve: potential substrates for vestibulo-autonomic interactions, Exp. Brain Res. 98 (1994) 200 – 212. [9] C.D. Balaban, J.F. Thayer, Neurological bases for balance – anxiety links, J. Anxiety Disord. 15 (2001) 53 – 79. [10] C.D. Balaban, B.J. Yates, Vestibulo-autonomic interactions: a teleologic perspective, in: S. Highstein, R. Fay, A. Popper (Eds.), Springer Handbook of Auditory Research: Vestibular Anatomy and Physiology, vol. 19, Springer-Verlag, New York, pp. 286 – 342. [11] C.D. Balaban, D.M. McGee, J. Zhou, C.A. Scudder, Responses of primate caudal parabrachial nucleus and Ko¨lliker – Fuse nucleus neurons to whole body rotation, J. Neurophysiol. 88 (2002) 3175 – 3193. [12] N.H. Barmack, R.W. Baughman, F.P. Eckenstein, H. Shojaku, Secondary vestibular cholinergic projection to the cerebellum of the rabbit and rat as revealed by choline acetyltransferase immunohistochemistry, retrograde and orthograde tracers, J. Comp. Neurol. 317 (1992) 250 – 270. [13] C.C. Barr, L.W. Schulteis, D.A. Robinson, Voluntary, non-visual control of the human vestibulo-ocular reflex, Acta Otolaryngol. 81 (1976) 365 – 375. [14] I.A. Billig, J.M. Foris, J.P. Card, B.J. Yates, Transneuronal tracing of neural pathways controlling an abdominal muscle, rectus abdominis, in the ferret, Brain Res. 820 (1999) 31 – 44. [15] D.M. Bramble, F.A. Jenkins Jr., Mammalian locomotor-respiratory integration: implications for diaphragmatic and pulmonary design, Science 262 (1993) 235 – 240. [16] S.C. Cannon, R.J. Leigh, D.S. Zee, L.A. Abel, The effect of the rotational magnification of corrective lenses on the quantitative evaluation of the VOR, Acta Otolaryngol. 100 (1985) 81 – 88. [17] D.S. Charney, A. Deutsch, A functional neuroanatomy of anxiety and fear: implications for the pathophysiology and treatment of anxiety disorders, Crit. Rev. Neurobiol. 10 (1996) 419 – 446. [18] W.E. Collins, Effects of mental set upon vestibular nystagmus, J. Exp. Psychol. 63 (1962) 191 – 197.

C.D. Balaban / Brain Research 996 (2004) 126–137 [19] W.E. Collins, in: H.H. Kornhuber (Ed.), Vestibular System, vol. 6/2, Springer, Berlin, 1974, pp. 361 – 368. [20] W.E. Collins, F.E. Guedry Jr., Arousal effects and nystagmus during prolonged constant acceleration, Acta Otolaryngol. (Stockholm) 54 (1961) 349 – 362. [21] W.E. Collins, R.H. Poe, Amphetamine, arousal and human vestibular nystagmus, J. Pharmacol. Exp. Ther. 138 (1962) 120 – 125. [22] K. Feil, H. Herbert, Topographic organization of spinal and trigeminal somatosensory pathways to the rat parabrachial and Ko¨lliker – Fuse nuclei, J. Comp. Neurol. 353 (1995) 506 – 528. [23] C.E. Fulweiler, C. Saper, Subnuclear organization of the efferent connections of the parabrachial nucleus in the rat, Brain Res. Rev. 7 (1984) 229 – 259. [24] J.M. Gorman, J.M. Kent, G.M. Sullivan, J.D. Coplan, Neuroanatomical hypothesis of panic disorder, revised, Am. J. Psychiatry 157 (2000) 493 – 505. [25] P.S. Grigson, T. Shimura, R. Norgren, Brain stem lesions and gustatory function: II. The role of the nucleus of the solitary tract in Na+ appetite, conditioned taste aversion, and conditioned odor aversion in rats, Behav. Neurosci. 111 (1997) 180 – 187. [26] P.S. Grigson, S. Reilly, G. Scalera, R. Norgren, The parabrachial nucleus is essential for acquisition of a conditioned odor aversion in rats, Behav. Neurosci. 112 (1998) 1104 – 1113. [27] P.S. Grigson, S. Reilly, T. Shimura, R. Norgren, Ibotenic acid lesions of the parabrachial nucleus and conditioned taste aversion: further evidence for an associative deficit in rats, Behav. Neurosci. 112 (1998) 160 – 171. [28] R.T. Henry, J.D. Connor, C.D. Balaban, Nodulus-uvula depressor response: central GABA-mediated inhibition of a-adrenergic outflow, Am. J. Physiol. 256 (1989) H1601 – H1608. [29] H. Herbert, M.M. Moga, C.B. Saper, Connections of the parabrachial nucleus with the nucleus of the solitary tract and medullary reticular formation in the rat, J. Comp. Neurol. 293 (1990) 540 – 580. [30] L. Jasmin, A.R. Burkey, J.P. Card, A.I. Basbaum, Transneuronal labeling of a nociceptive pathway, the spino-(trigemino-)parabrachioamygdaloid, in the rat, J. Neurosci. 17 (1997) 3751 – 3765. [31] B.S. Kapp, C.G. Markgraf, J.S. Schwaber, T. Bilyk-Spafford, The organization of dorsal medullary projections to the central amygdaloid nucleus and parabrachial nucleus in the rabbit, Neuroscience 30 (1989) 717 – 732. [32] H. Kaufman, I. Biaggioni, A. Voustianiouk, A. Diedrich, F. Costa, M. Gizzi, T. Raphan, B. Cohen, Vestibular control of sympathetic activity: an otolith-sympathetic reflex in humans, Exp. Brain Res. 143 (2002) 463 – 469. [33] P.D. Kramer, M. Shelhamer, G.C.Y. Peng, D.S. Zee, Context-specific short-term adaptation of the phase of the vestibulo-ocular reflex, Exp. Brain Res. 120 (1998) 184 – 192.

137

[34] J.E. LeDoux, J. Iwata, P. Cicchetti, D.J. Reis, Different projections of the central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear, J. Neurosci. 8 (1988) 2517 – 2529. [35] M. Magnusson, Effect of alertness on the vestibulo-ocular reflex and the slow rise in optokinetic nystagmus in rabbits, Am. J. Otolaryngol. 7 (1986) 353 – 359. [36] M. Magnusson, I. Pyykko¨, L.O. Schale´n, H. Enbom, The effect of alertness on the velocity storage mechanism, Adv. Oto-Rhino-Laryngol. 41 (1988) 53 – 57. [37] I.W. McLean, P.K. Nakane, Periodate – lysine – paraformaldehyde fixative. A new fixative for immunoelectron microscopy, J. Histochem. Cytochem. 22 (1974) 1077 – 1083. [38] M. Miura, K. Takayama, The site of origin of the so-called fastigial pressor response, Brain Res. 473 (1988) 352 – 358. [39] M.M. Moga, H. Herbert, K.M. Hurley, Y. Yasui, T.S. Gray, C.B. Saper, Organization of cortical, basal forebrain, and hypothalamic afferents to the parabrachial nucleus in the rat, J. Comp. Neurol. 295 (1990) 624 – 661. [40] J.D. Porter, C.D. Balaban, Connections between the vestibular nuclei and regions that mediate autonomic function in the rat, J. Vestibular Res. 7 (1997) 63 – 76. [41] D.A. Ruggiero, E.P. Mtui, K. Otake, M. Anwar, Vestibular afferents to the dorsal vagal complex: substrate for vestibulo-autonomic interactions in the rat, Brain Res. 743 (1996) 294 – 302. [42] D. Saxon, D.A. Hopkins, Efferent and collateral organization of paratrigeminal nucleus projections: an anterograde and retrograde fluorescent tracer study in the rat, J. Comp. Neurol. 402 (1998) 93 – 110. [43] M. Shelhamer, D.A. Robinson, H.S. Tan, Context-specific adaptation of the gain of the vestibulo-ocular reflex in humans, J. Vestibular Res. 2 (1992) 89 – 96. [44] A.C. Spector, R. Norgren, H. Grill, Parabrachial gustatory lesions impair taste aversion learning in rats, Behav. Neurosci. 106 (1992) 147 – 161. [45] S. Yakushin, T. Raphan, B. Cohen, Context-specific adaptation of the vertical vestibuloocular reflex with regard to gravity, J. Neurophysiol. 84 (2000) 3067 – 3071. [46] B.J. Yates, L. Grelot, I.A. Kerman, C.D. Balaban, J. Jakus, A.D. Miller, The organization of vestibular inputs to nucleus tractus solitarius (NTS) and adjacent structures in the cat brainstem, Am. J. Physiol. 267 (1994) R974 – R983. [47] B.J. Yates, C.D. Balaban, A.D. Miller, K. Endo, Y. Yamaguchi, Vestibular inputs to the lateral tegmental field of the cat: potential role in autonomic control, Brain Res. 689 (1995) 197 – 206. [48] B.J. Yates, J.A. Smail, S.D. Stocker, J.P. Card, Transneuronal tracing of neural pathways controlling activity of diaphragm motoneurons in the ferret, Neuroscience 90 (1999) 1501 – 1513.