Neurochemical organization of paratrigeminal nucleus projections to the dorsal vagal complex in the rat

Somatotopy in the Medullary Dorsal Horn As a Basis for Orofacial Reflex Behavior

The somatotopy of the trigeminocervical complex of the rat was defined as a basis for describing circuitry for reflex behaviors directed through the facial motor nucleus. Thus, transganglionic transport of horseradish peroxidase conjugates applied to individual nerves/peripheral receptive fields showed that nerves innervating oropharyngeal structures projected most rostrally, followed by nerves innervating snout, periocular, and then periauricular receptive fields most caudally. Nerves innervating mucosae or glabrous receptive fields terminated densely in laminae I, II, and V of the trigeminocervical complex, while those innervating hairy skin terminated in laminae I-V. Projections to lamina II exhibited the most focused somatotopy when individual cases were compared. Retrograde transport of FluoroGold (FG) deposited into the facial motor nucleus resulted in labeled neurons almost solely in lamina V of the trigeminocervical complex. The distribution of these labeled neurons paralleled the somatotopy of primary afferent fibers, e.g., those labeled after FG injections into a functional group of motoneurons innervating lip musculature were found most rostrally while those labeled after injections into motoneurons innervating snout, periocular and preauricular muscles, respectively, were found at progressively more caudal levels. Anterograde transport of injections of biotinylated dextran amine into lamina V at different rostrocaudal levels of the trigeminocervical complex confirmed the notion that the somatotopy of orofacial sensory fields parallels the musculotopy of facial motor neurons. These data suggest that neurons in lamina V are important interneurons in a simple orofacial reflex circuit consisting of a sensory neuron, interneuron and motor neuron. Moreover, the somatotopy of primary afferent fibers from the head and neck confirms the "onion skin hypothesis" and suggests rostral cervical dermatomes blend seamlessly with "cranial dermatomes." The transition area between subnucleus interpolaris and subnucleus caudalis is addressed while the paratrigeminal nucleus is discussed as an interface between the somatic and visceral nervous systems.

Somatostatin in the rat rostral ventrolateral medulla: Origins and mechanism of action

Somatostatin (SST) or agonists of the SST-2 receptor (sst2 ) in the rostral ventrolateral medulla (RVLM) lower sympathetic nerve activity, arterial pressure, and heart rate, or when administered within the Bötzinger region, evoke apneusis. Our aims were to describe the mechanisms responsible for the sympathoinhibitory effects of SST on bulbospinal neurons and to identify likely sources of RVLM SST release. Patch clamp recordings were made from bulbospinal RVLM neurons (n = 31) in brainstem slices prepared from juvenile rat pups. Overall, 58% of neurons responded to SST, displaying an increase in conductance that reversed at -93 mV, indicative of an inwardly rectifying potassium channel (GIRK) mechanism. Blockade of sst2 abolished this effect, but application of tetrodotoxin did not, indicating that the SST effect is independent of presynaptic activity. Fourteen bulbospinal RVLM neurons were recovered for immunohistochemistry; nine were SST-insensitive and did not express sst2a . Three out of five responsive neurons were sst2a -immunoreactive. Neurons that contained preprosomatostatin mRNA and cholera-toxin-B retrogradely transported from the RVLM were detected in: paratrigeminal nucleus, lateral parabrachial nucleus, Kölliker-Fuse nucleus, ventrolateral periaqueductal gray area, central nucleus of the amygdala, sublenticular extended amygdala, interstitial nucleus of the posterior limb of the anterior commissure nucleus, and bed nucleus of the stria terminalis. Thus, those brain regions are putative sources of endogenous SST release that, when activated, may evoke sympathoinhibitory effects via interactions with subsets of sympathetic premotor neurons that express sst2 .

Localization and chemical coding of the dorsal motor vagal nucleus (DMX) neurons projecting to the porcine stomach prepyloric area in the physiological state and after stomach partial resection

The aim of our study was to localize and define immunocytochemical characteristic of the dorsal motor nucleus of the vagus (DMX) neurons projecting to the porcine stomach prepyloric region in the physiological state and after gastric partial resection. To identify the stomach-projecting perikarya, the neuronal retrograde tracer--Fast Blue (FB) was injected into the studied region of control and resection group (RES). In the RES group, on 22nd day after FB injection, the partial resection of the stomach region previously injected with FB was performed. Sections were immunostained with ChAT, pituitary adenylate cyclase-activating peptide (PACAP), vasoactive intestinal polypeptide (VIP), nitric oxide synthase (NOS), galanin (GAL), substance P (SP), leu-enkephalin (LENK), and cocaine- and amphetamine-regulated transcript (CART). In the DMX of control and RES group, the stomach-projecting perikarya were found in the entire extent of the nucleus bilaterally. Within control animals, 30.08 ± 1.97 % of the gastric DMX perikarya expressed PACAP, while other substances were found only in the neuronal fibers. In the RES group DMX, PACAP was found in 45.58 ± 2.2 %, VIP in 28.83 ± 3.63 %, NOS in 21.22 ± 3.32 %, and GAL in 5.67 ± 1.49 % of the FB-labeled gastric perikarya. Our data implicate PACAP, VIP, NOS, and GAL as neuronal survival promoting substances and the CART-, LENK-, SP- NOS-, and GAL-immunoreactive processes in control of the gastric vagal neurons in the pig.

Selective distribution and function of primary afferent nociceptive inputs from deep muscle tissue to the brainstem trigeminal transition zone

Orofacial injury activates two distinct regions in the spinal trigeminal complex, the subnuclei interpolaris/caudalis (Vi/Vc) transition zone and the laminated Vc, or medullary dorsal horn (MDH). Studies suggest that the Vi/Vc transition zone plays an important role in processing orofacial deep input. To test this hypothesis, we employed a double-tracing strategy to compare central projections of primary afferent neurons that innervate the masseter muscle and the overlying skin. Different tracers were injected either centrally (Fluoro-Gold: ventral Vi/Vc, or MDH) or peripherally (wheat germ agglutinin-conjugated horseradish peroxidase or cholera toxin B: masseter or overlying skin) in the same rat. Trigeminal ganglion tissue sections were processed for single or double immunohistochemistry. The double labeling of ganglion neurons indicates their site of peripheral and central innervations. A population of small to medium-sized neurons was doubly labeled after injections of the tracers into the masseter-Vi/Vc, masseter-MDH, or the skin-MDH. However, only a few double-labeled neurons were occasionally observed after injections of the tracers into the skin-Vi/Vc. Injection of an N-methyl-D-aspartate receptor antagonist, AP-5, into the Vi/Vc and MDH attenuated masseter inflammatory hyperalgesia. In contrast, hyperalgesia after inflammation of the skin overlying the masseter was attenuated by injection of AP-5 into the MDH but not Vi/Vc. These results indicate that while both masseter and cutaneous inputs project to the MDH, masseter afferents provide an additional input to the Vi/Vc. These findings provide further evidence to support a role of the trigeminal transition zone in response to orofacial deep injury.

Role of the paratrigeminal nucleus in nocifensive responses of rats to chemical, thermal and mechanical stimuli applied to the hind paw

Anatomical and immunohystochemical data suggest the paratrigeminal nucleus (Pa5) may play a role in nociceptive processing. The current study examines the influence of unilateral Pa5 lesion on nocifensive responses of conscious rats to noxious thermal (Hargreaves test), mechanical (electronic von Frey and Randall-Selitto tests), and chemical (formalin 2.5%; 50 microl) stimuli applied to the hind paw. Lesion of the Pa5 induced by ibotenic acid did not affect the latency for radiant heat-induced withdrawal of either paw. In contrast, the mean mechanical threshold for withdrawal of the contralateral (but not ipsilateral) paw in Pa5-lesioned rats was reduced by approximately 45% and 20%, in electronic von Frey and Randall-Selitto tests, respectively, when compared to sham-operated animals. Conversely, animals with Pa5 lesions injected with formalin in the contralateral paw spent less time engaged in focused (licking, biting or scratching the injected paw) and total nocifensive behavior (i.e., focused nocifensive behavior plus protection of the injected paw during movements) in both the first and second phases of the test [ approximately 50% inhibition of each parameter during first phase (0-5 min) and at 20, 25, and 30 min of second phase, relative to the sham-operated group], but the number of paw-jerks was unaffected. Pa5 lesion also delayed the onset of second phase focused pain induced by formalin in the ipsilateral paw. The results suggest that the Pa5 integrates the supraspinal pain control system and plays a differential modulatory role in the central processing of mechanical and chemical nociceptive information.

Ultrastructure and synaptology of the paratrigeminal nucleus in the rat: primary pharyngeal and laryngeal afferent projections

The paratrigeminal nucleus (PTN) receives primary afferent projections from the aerodigestive tract and orofacial regions and plays a role in the integration of visceral and somatic information. This study describes the fine structure of the rat PTN and the synaptology of primary afferent projections from the pharynx and larynx. Injections of wheat germ agglutinin-horseradish peroxidase (WGA-HRP) or cholera toxin-HRP (CT-HRP) were made into the wall of the pharynx or larynx to label primary afferent axon terminals. Light microscopic observations demonstrated that afferent axons terminated bilaterally in overlapping fields in the PTN. Electron microscopic observations of the PTN revealed that there were three distinct classes of neurons, based on morphology and axosomatic contacts. The most abundant neurons, Type 1, were fusiform in shape and received very few or no axosomatic contacts. Type 2 neurons contained prominent Nissl substance (rough endoplasmic reticulum) and few axosomatic contacts, while Type 3 neurons had many axosomatic synapses. Terminals containing round, clear vesicles and forming asymmetric contacts (round asymmetric, RA) with dendrites were the predominant synaptic type in the PTN. Primary afferent terminals from the pharynx and larynx were of the RA type and formed synaptic contacts with small-diameter (<1 microm) dendrites. Visceral primary afferent inputs from the pharynx and larynx overlap with trigeminal somatic afferents in the PTN and have similar synaptic morphology. The results support the concept that the PTN provides an anatomical substrate for mediating viscerovisceral and somatovisceral reflexes via efferent connections with autonomic centers in the brainstem.

Cyto- and chemoarchitecture of the sensory trigeminal nuclei of the echidna, platypus and rat

We have examined the cyto- and chemoarchitecture of the trigeminal nuclei of two monotremes using Nissl staining, enzyme reactivity for cytochrome oxidase, immunoreactivity for calcium binding proteins and non-phosphorylated neurofilament (SMI-32 antibody) and lectin histochemistry (Griffonia simplicifolia isolectin B4). The principal trigeminal nucleus and the oralis and interpolaris spinal trigeminal nuclei were substantially larger in the platypus than in either the echidna or rat, but the caudalis subnucleus was similar in size in both monotremes and the rat. The numerical density of Nissl stained neurons was higher in the principal, oralis and interpolaris nuclei of the platypus relative to the echidna, but similar to that in the rat. Neuropil immunoreactivity for parvalbumin was particularly intense in the principal trigeminal, oralis and interpolaris subnuclei of the platypus, but the numerical density of parvalbumin immunoreactive neurons was not particularly high in these nuclei of the platypus. Neuropil immunoreactivity for calbindin and calretinin was relatively weak in both monotremes, although calretinin immunoreactive somata made up a large proportion of neurons in the principal, oralis and interpolaris subnuclei of the echidna. Distribution of calretinin immunoreactivity and Griffonia simplicifolia B4 isolectin reactivity suggested that the caudalis subnucleus of the echidna does not have a clearly defined gelatinosus region. Our findings indicate that the trigeminal nuclei of the echidna do not appear to be highly specialized, but that the principal, oralis and interpolaris subnuclei of the platypus trigeminal complex are highly differentiated, presumably for processing of tactile and electrosensory information from the bill.

Calbindin D28k-containing neurons in the paratrigeminal nucleus receive convergent nociceptive information and project to nucleus of the solitary tract in rat

The paratrigeminal nucleus (PTN) receives orofacial somatic and visceral afferent fibers and contains many calbindin-D28k neurons (CB-containing neurons) that project to nucleus of the solitary tract (NTS). In the present study, retrograde and transganglionic tracing methods combined with immunofluorescence histochemistry and confocal laser scanning microscopy were used. After Fluoro-gold (FG) injection into the unilateral NTS, 74.4% FG-labeled neurons of ipsilateral PTN were double-labeled with CB. Furthermore, 41.0% and 32.5% FG/CB double-labeled neurons co-existed with Fos induced by nociceptive stimulation of the lips and the upper alimentary tract, respectively. In the PTN unilateral to FG injection site, 26.6% CB-LI neurons were double-labeled with PAG, 61.5% and 79.0% CB/PAG double-labeled neurons were triple-labeled with FG and Fos, and 22.9% FG/CB double-labeled neurons were triple-labeled with PAG, 84.3% FG/PAG double-labeled neurons expressed Fos induced by the upper alimentary tract stimulation. In the intact animals, 62.8% CB-LI neurons and 88.3% PAG-LI neurons co-existed with GABA(B)R, respectively. In addition, some terminals from the inferior alveolar nerve (IAN) were closely apposed to CB/Fos double-labeled or CB single-labeled neurons. These results suggested that CB-containing neurons in the PTN receive the nociceptive information converge from the orofacial area and visceral organs, and comprising the glutamatergic excitatory transmission pathway from the PTN to the NTS. This pathway might be modulated by GABA via the GABA(B) receptor.

Immunohistochemical profiles of spinal lamina I neurones retrogradely labelled from the nucleus tractus solitarii in rat suggest excitatory projections

Three morphologically distinct types of lamina I neurones, fusiform, flattened and pyramidal, project from the spinal cord to the caudal part of the nucleus tractus solitarii in the rat, and may represent a pathway whereby peripheral stimuli can modify autonomic functions. The neurochemistry of these three types of projection neurones was investigated using retrograde neuronal tracing with cholera toxin B-subunit combined with dual and triple immunofluorescence labelling for different neuroactive substances. None of the lamina I neurones with immunoreactivity for GABA or glycine were found to project to the nucleus tractus solitarii, whereas high levels of glutamate immunoreactivity, which may indicate a glutamatergic phenotype, were found in 18.4% of fusiform, 9.6% of pyramidal and 2.1% of flattened projection neurones. Immunoreactivity for calbindin-D28K was present in 34.9% of fusiform cells, 18.3% of pyramidal cells and 10.5% of flattened cells, and nitric oxide synthase immunoreactivity was detected in 13.8% of fusiform cells, 1.1% of pyramidal cells and 4.2% of flattened cells that had projections to the nucleus tractus solitarii. Calbindin immunoreactivity was co-localised in major subpopulations of projection neurones of each morphological type that contained glutamate immunoreactivity, whereas co-localisation of nitric oxide synthase immunoreactivity in these neurones was relatively uncommon. The pyramidal cell was the only retrogradely labelled cell type found to be immunoreactive for substance P, but few (<5%) of these neurones were immunolabelled. These data are consistent with the hypothesis that lamina I neurones projecting to the dorsal vagal complex are not inhibitory, and that some of them, belonging mostly to the fusiform and pyramidal types, may exert excitatory, glutamate- or substance P-mediated effects upon inhibitory interneurones in the nucleus tractus solitarii. Such excitatory pathways could be involved in the attenuation of the reflex control of blood pressure by both painful and innocuous peripheral stimuli, such as those arising in injury and exercise.

Projections of the paratrigeminal nucleus to the ambiguus, rostroventrolateral and lateral reticular nuclei, and the solitary tract

The paratrigerminal nucleus (Pa5), a constituent of the spinal interstitial system, was linked to the pressor effect caused by bradykinin injected in the dorsal lateral medulla of the rat. The nucleus receives primary afferent sensory fibers contained in branches of the trigeminal, glossopharyngeal and vagus nerves. In this investigation connections of the paratrigeminal nucleus to other medullary structures were studied with the use of retrograde and anterograde neuronal tracers. Fluorescent light microscopy analyses of medullary sections of rats injected with the retrograde transport tracer Fluoro-gold in the nucleus of the solitary tract (NTS) or in the pressor area of the rostral ventrolateral medulla (RVLM) revealed labeled neuronal cell bodies in the ipsi- and contralateral Pa5. FluoroGold microinjections in the caudal ventrolateral medulla (CVLM) did not produce fluorescent labeling of Pa5 neurons. Microinjection of the anterograde transport neuronal tracer biocytin in the Pa5 produced bilateral labeling of the solitary tract (sol). rostroventrolateral reticular nucleus (RVL), ambiguus nucleus (Amb), lateral reticular nucleus (LRt) and ipsilateral parabrachial nuclei, but not the contralateral Pa5. Confocal laser microscopy showed fluorescence labeling of fibers and presumptive terminal varicosities in the NTS, RVL, Amb and LRt. The present findings showing the paratrigeminal nucleus interposed between sensory afferent and stuctures associated to cardiovascular and respiratory functions, suggest that the structure may act as a medullary relay nucleus for sensory stimuli directly connecting primary afferents to structures mediating cardiovascular and respiratory reflexes.

Nitric oxide producing neurones in the rat medulla oblongata that project to nucleus tractus solitarii

The production of nitric oxide in neurones of the rat medulla oblongata that project to the nucleus tractus solitarii (NTS) was examined by simultaneous immunohistochemical detection of nitric oxide synthase (NOS) and of cholera toxin B-subunit (CTb), which was injected into the caudal zone of the NTS. Neurones immunoreactive for CTb and neurones immunoreactive for NOS were widely co-distributed and found in almost all the anatomical divisions of the medulla. Dual-labelled cells, containing both CTb and NOS immunoreactivities were more numerous ipsilaterally to the injection sites. They were concentrated principally in the more rostral zone of the NTS, raphé nuclei, dorsal, intermediate and lateral reticular areas, spinal trigeminal and paratrigeminal nuclei and the external cuneate and medial vestibular nuclei. Isolated dual-labelled neurones were also scattered throughout most of the divisions of the reticular formation. These observations indicate that many areas of the medulla that are known to relay somatosensory and viscerosensory inputs contain NOS immunoreactive neurones that project to the NTS, and may, therefore, contribute to the dense NOS-immunoreactive innervation of the NTS. The release of nitric oxide from the axon terminals of these neurones may modulate autonomic responses generated by NTS neurones in relation to peripheral sensory stimuli, and thus ultimately regulate sympathetic and/or parasympathetic outflow.

Persistent Fos protein expression after orofacial deep or cutaneous tissue inflammation in rats: implications for persistent orofacial pain

This study was designed to systematically examine the effects of persistent orofacial tissue injury on prolonged neuronal activation in the trigeminal nociceptive pathways by directly comparing the effects of orofacial deep vs. cutaneous tissue inflammation on brainstem Fos protein expression, a marker of neuronal activation. Complete Freund's adjuvant (CFA) was injected unilaterally into the rat temporomandibular joint (TMJ) or perioral (PO) skin to produce inflammation in deep or cutaneous tissues, respectively. Rats were perfused 2 hours, 24 hours, 3 days, or 10 days following CFA injection. The TMJ and PO inflammation-induced Fos expression paralleled the intensity and course of inflammation over the 10-day observation period, suggesting that the increase in intensities and persistence of Fos protein expression may be associated with a maintained increase in peripheral input. Compared to PO CFA injection, the injection of CFA into the TMJ produced a significantly stronger inflammation associated with a greater Fos expression. In TMJ- but not in PO-inflamed rats, Fos-like immunoreactivity (LI) spread from superficial to deep upper cervical dorsal horn as the inflammation persisted and there was a dominant ipsilateral Fos-labeling in the paratrigeminal nucleus. Common to TMJ and PO inflammation, Fos-LI was induced in the trigeminal subnuclei interpolaris and caudalis, C1-2 dorsal horn, and other medullary nuclei. Substantial bilateral Fos-LI was found in the interpolaris-caudalis trigeminal transition zone. Further analysis revealed that Fos-LI in the ventral transition zone was equivalent bilaterally, whereas Fos-LI in the dorsal transition zone was predominantly ipsilateral to the inflammation. The differential induction of Fos expression suggests that an increase in TMJ C-fiber input after inflammation and robust central neuronal hyperexcitability contribute to persistent pain associated with temporomandibular disorders.

Orofacial deep and cutaneous tissue inflammation differentially upregulates preprodynorphin mRNA in the trigeminal and paratrigeminal nuclei of the rat

Preprodynorphin (PPD) and preproenkephalin (PPE) gene expression in a rat model of orofacial inflammation were examined in order to further characterize the neurochemical mechanisms underlying orofacial inflammation and hyperalgesia. Deep and cutaneous orofacial inflammation was produced by a unilateral injection of complete Freund's adjuvant (CFA) into the rat temporomandibular joint (TMJ) or perioral skin (PO), respectively. RNA blot analysis of the tissues including the spinal trigeminal complex revealed that the PPD mRNA level ipsilateral to TMJ inflammation was increased by 56.5+/-14.7% (n=4) when compared to the Naive group, and was significantly greater than the contralateral PPD mRNA level (p<0.05). The distribution of neurons that exhibited PPD mRNA after inflammation was localized by in situ hybridization (naive approximately 0). In TMJ-inflamed rats (n=6) PPD mRNA-positive neurons were found ipsilaterally in the medial portion of laminae I-II of the upper cervical dorsal horn (4.5+/-0.3), the dorsal portion of the subnucleus caudalis and caudal subnucleus interpolaris (5.2+/-0.3), and the paratrigeminal nucleus (6.4+/-1.2). A very localized induction of PPD mRNA was also identified in a group of neurons in the intermediate portion of the subnucleus caudalis (2.4+/-0.4) in PO-inflamed rats (n=6). The distribution of these PPD mRNA-positive neurons was somatotopically relevant to the site of injury. There were no significant changes in PPE mRNA expression in both TMJ- and PO-inflamed rats. These results indicate that TMJ inflammation resulted in a more intense and widespread increase in PPD mRNA expression when compared to PO inflammation. These changes may contribute to persistent central hyperexcitability and pain associated with temporomandibular disorders.