Michael T. Shipley

the connections of the mouse olfactory bulb: A study using orthograde and retrograde transport of wheat germ agglutinin conjugated to horseradish peroxidase

The efferent and centrifugal afferent connections of the main olfactory bulb (MOB) of the mouse were studied by orthograde and retrograde transport of wheat germ agglutinin conjugated to horseradish peroxidase (WGA-HRP). MOB projects ipsilaterally to the anterior olfactory nucleus, taenia tecta, anterior hippocampal continuation, indusium grisium, olfactory tubercle, and the lateral and medial divisions of the entorhinal area. In the region of the anterior one-half to two-thirds of the posterior division of the insular cortex the projection from MOB extends into the insular cortex. The only efferent projection of MOB to the contralateral half of the brain was to the anterior olfactory nucleus. All efferent projections of MOB, thus, are to telencephalic structures. By contrast the centrifugal afferents to MOB originate from every major division of the neuraxis. Neurons projecting to the bulb were found ipsilaterally in all divisions of the anterior olfactory nucleus (AON). In some cases, labeling in the external division of AON was weak or absent. In the contralateral AON, pars externa was the most intensively labeled sub-division. Retrogradely labeled neurons were also present in all other subdivisions of the contralateral AON but were fewer in number and less heavily labeled than in the ipsilateral AON. Ipsilaterally, positive neurons were also present in taenia tecta, and the anterior hippocampal continuation. There was profuse retrograde labeling of neurons in the entire extent of the ipsilateral piriform cortex (PC). There was a rostral to caudal gradient of labeling in PC with more positive neurons in rostral than caudal parts. Labeled neurons were present in the lateral entorhinal cortex LEC and in the transitional cortex between LEC and PC. Very heavy retrograde labeling was present in the nuclei of the horizontal and vertical limbs of the diagonal band (HDB and VDB). More cells were labeled in HDB than in VDB. Neurons were labeled in the ipsilateral nucleus of the lateral olfactory tract (NLOT) and, when the injection spread into the accessory olfactory bulb, labeled neurons were present ventral to NLOT in accessory NLOT. A few lightly labeled neurons were always present in the posterolateral and medial cortical amygdaloid areas. Neurons were labeled in the zona inserta and scattered throughout several hypothalamic nuclei. There was massive retrograde labeling of neurons in the locus coeruleus and neurons were abundantly labeled in the dorsal and medial raphe nuclei and nucleus raphe pontis. In general, the labeling of MOB connections was more extensive than that which has been reported in closely related species.(ABSTRACT TRUNCATED AT 400 WORDS)

Transport of molecules from nose to brain: Transneuronal anterograde and retrograde labeling in the rat olfactory system by wheat germ agglutinin-horseradish peroxidase applied to the nasal epithelium ☆

Transneuronal anterograde labeling with the conjugate wheat germ agglutinin-horseradish peroxidase (WGA-HRP) has been documented in the mammalian and immature avian visual system [6,14]. Transneuronal retrograde labeling was significant only in the chick [6]. The present study was performed to determine whether transneuronal labeling could be shown in the mammalian olfactory system, whether the phenomenon was robust in adults, and whether transneuronal retrograde transport could label several transmitter-specific centrifugal afferent projections to the olfactory bulb. In addition we wished to learn whether molecules that enter the nasal cavity can undergo transport to brain neurons. Gelfoam implants soaked with 1% WGA-HRP, surgically implanted into the nasal cavity, produced transneuronal labeling patterns that affirmed all of these questions. Transneuronal anterograde transport labeled the appropriate zones in the olfactory bulb and in all second order olfactory targets. In addition, there was transneuronal retrograde labeling of neurons in the olfactory bulb, anterior olfactory nucleus and in transmitter-specific projection neurons from the diagonal band (cholinergic), raphe (serotonergic) and locus coeruleus (noradrenergic). Transneuronal labeling was robust and consistent. The patterns of labeling indicated that transneuronal anterograde and retrograde transport occurred along known, specific circuits in the olfactory system. The present results suggest that nasal epithelial application of WGA-HRP may be a useful tool for assessing regeneration of primary olfactory neurons and the status of central circuitry following regeneration. The method should also facilitate the study of central olfactory connections after surgical or genetic lesions of the olfactory bulb. Finally, these experiments suggest the possibility that inhaled molecules including, possibly substances of abuse, may be transported to, and, possibly, influence the function of neurons in the brain, including some (diagonal band, raphe, locus coeruleus) which have extensive projections to wide areas of the CNS.

Anatomical evidence for convergence of olfactory, gustatory, and visceral afferent pathways in mouse cerebral cortex

Flavor perception requires the neural integration of olfactory, gustatory and, possibly, visceral afferent information. Presently, it is not known where, or how this integration takes place in the brain. Neuroanatomical data presented here suggest that pathways subserving these sensory modalities converge in mouse insular cortex after surprisingly few synaptic relays. Orthograde transport of wheat germ agglutinin conjugated to horseradish peroxidase (WGA-HRP) was used to label main olfactory bulb (MOB) efferents. A projection into layer I of insular cortex was present in every case. Bulb transections were made to provoke anterograde degeneration and EM analysis confirmed that the olfactory projection to insular cortex was a terminal pathway. WGA-HRP injections in the MOB-recipient zone of insular cortex resulted in ortho and retrograde labeling of ascending and descending gustatory-visceral afferent pathways. It is concluded that in the mouse, there is a remarkably direct convergence of olfactory and gustatory-visceral sensory pathways in insular cortex. Together with the descending connections from insular cortex to the amygdala and to brainstem autonomic structures, it is possible that the cortical integration of olfactory and gustatory-visceral information could modulate mechanisms involved in food selection and autonomic reactions relating to the chemical senses. Basic mechanisms subserving flavor perception might be usefully modelled in mouse insular cortex.

Topographical Specificity of Forebrain Inputs to the Midbrain Periaqueductal Gray: Evidence for Discrete Longitudinally Organized Input Columns

Over two decades ago it was discovered that electrical stimulation of the periaqueductal gray (PAG) caused profound analgesia (Reynolds, 1969). It was subsequently found that “PAG-analgesia” is, at least in part, mediated by opiate and neurotensin systems acting via PAG projections to the rostral medulla (Basbaum and Fields, 1984; Behbehani, 1981; Behbehani and Fields, 1979; Behbehani and Pert, 1984; Behbehani et al., 1987; Lakos and Basbaum, 1988; Reichling et al., 1988; Shipley et al., 1987). As a result of the observation that a discrete CNS structure exerted such a potent regulation of pain, much subsequent research has focused on the role of PAG in antinociception. At the same time there has been growing evidence that PAG plays a key role in the “defense reaction” (Bandler and Carrive, 1988; Bandler and Depaulis, this volume; Bandler et al., 1985a, 1991; Depaulis and Vergnes, 1986; Depaulis et al., 1989; Zhang et al., 1990), vocalization (Jurgens, 1976; Jurgens and Richter, 1986; Larson, 1985; Larson and Kistler, 1984; 1986), and in certain sexual behaviors (Ogawa et al., this volume; Sakuma and Pfaff; 1979a,b).

Golgi-like, transneuronal retrograde labelling with CNS injections of herpes simplex virus type 1.

The use of HSV1 as a retrograde transneuronal marker for the CNS was assessed in several neuroanatomical systems of the rat brain including the olfactory, visual and somatosensory systems. In all systems, retrograde transneuronal transport was observed; with appropriate survival times transport was evident in third and fourth order neurons in established neuronal circuits. A striking observation was the high frequency of neurons labelled in a Golgi-like manner. The visualization of even the finest dendritic processes provides information about the architecture of neurons several synapses removed from the site of injection. The Golgi-like labelling is so complete that it is possible to identify and process distal parts of dendrites for EM analysis. Thus, it should be feasible to identify synaptic inputs to the dendrites of neurons two or more synapses removed from the site of injection. There was spotty evidence for anterograde transport but the vast majority of the labelling could be accounted for by retrograde transport. With increased survival time, some regions, especially those located one synapse removed from the injection site, became necrotic and the virus spread to glia cells in addition to neurons in those regions. However, in regions more than one synapse removed from the injection there was negligible labelling of glial cells. Taken together, these results suggest that transneuronal retrograde labelling with HSV1 is a useful tool in the analysis of neural circuits.

Activation of serotonin1A receptors inhibits midbrain periaqueductal gray neurons of the rat.

The midbrain periaqueductal gray (PAG) is involved in a variety of functions including pain modulation, vocalization, autonomic control, fear and anxiety. This area contains serotonin receptors, particularly 5-HT1A that are known to play a role in the above functions. The goals of this study were to characterize the effects of 8-OH-DPAT, a selective 5-HT1A agonist, on the firing characteristics and membrane properties of PAG neurons. Both in vivo and in vitro preparations were used. The effects of 8-OH-DPAT on baseline activity of 91 neurons were tested in the in vivo preparation. In 50/91 cells, 8-OH-DPAT produced a decrease in the firing rate that ranged between 21 and 98% (mean +/- S.E.M. decrease of 49 +/- 1.9%). This inhibitory effect was dose dependent and could be blocked by spiperone. In 10/91 cells, 8-OH-DPAT produced an increase in the firing rate that ranged between 13 and 290%, with mean increase of 83 +/- 7.4%. The baseline firing rate of the remaining 31 cells was not affected by 8-OH-DPAT. In the PAG slice preparation, the effects of 8-OH-DPAT on synaptic and membrane properties of 17 PAG neurons were tested using whole-cell voltage clamp-recording procedures. In 14 cells, application of 8-OH-DPAT produced hyperpolarization that ranged between 6 and 21 mV, with mean of 8.4 +/- 2.0 mV. This hyperpolarization was associated with a decrease in membrane impedance that ranged between 8 and 45%, with mean decrease of 21.6 +/- 4.5%. The remaining three neurons did not respond to 8-OH-DPAT.(ABSTRACT TRUNCATED AT 250 WORDS)

The indusium griseum in the mouse: Architecture, Timm's histochemistry and some afferent connections

The Indusium griseum (IG) is an enigmatic cortical field classically felt to be a part of the hippocampus (HC). In the mouse, IG lies just dorsal to the corpus callosum at the base of the anterior half of cingulate cortex. In coronal sections the field is small but constitutes a fairly long rostro-caudal strip. The connections of the IG are poorly understood. The Timms staining pattern of the IG is reminiscent of a mini-dentate gyrus (DG) comprising a layer of granule cells with two bands of staining in the molecular layer. In the DG there are three bands which correspond to inputs from the lateral and medial entorhinal area (LEA and MEA) and the ipsi- and contralateral association systems. Using anterograde transport of HRP we have found that the LEA and MEA also terminate in the molecular layer of the IG. This suggests that the IG is a displaced portion of the DG. The olfactory system is known to have a strong indirect influence on the HC via primary and secondary bulbar projections to the LEA. Wheat germ agglutinin-HRP injections confined to the main olfactory bulb (MOB) show a direct projection from the MOB to IG. Both the olfactory bulb itself and retrobulbar structures such as the piriform cortex (PC) convey olfactory information to the LEA; the LEA supplies a major input to the DG. Our results suggest that there is a more direct pathway whereby olfactory information may influence a cortical region, IG, whose histochemistry and direct afferents from the entorhinal cortex suggest that it is part of or closely related to the DG. Thus, IG may represent a phylogenetically old olfacto-recipient outpost of the hippocampus.

Processing and Analysis of Neuroanatomical Images

Today, neuroanatomists have an almost bewildering array of techniques that are being used to study the organization of the nervous system along dimensions that were not even conceived 10 to 15 years ago. Brain circuits are no longer viewed as wiring diagrams interlinking classical brain structures but rather as subpopulations of neurons differing along morphological and neurochemical lines having interconnections with other equally specific subpopulations of neurons in multiple other brain structures. No longer are neural circuits “excitatory” or “inhibitory.” Neurons may contain several transmitters/modulators, and the actions of modulators may depend on the moment-to-moment status of their target cells. The levels of these neuroac-tive molecules may vary with the animal’s functional, humoral, or developmental state. On the postsynaptic side, multiple receptor subtypes are well established for many transmitter systems; these link presynaptic inputs to a growing multitude of channel types and second messenger systems, which may or may not have direct actions on the genome or gene products of the target cell. Methods for studying these phenomena in brain sections are now available, and there are exciting possibilities to apply similar techniques to in vitro brain-slice preparations to study functioning neuroanatomy.

Characterization of the effect of cholecystokinin (CCK) on neurons in the periaqueductal gray of the rat: immunocytochemical and in vivo and in vitro electrophysiological studies

The periaqueductal gray (PAG) is an important integration site for pain, autonomic functions, vocalization, fear and anxiety. Cholecystokinin (CCK) is a major neurotransmitter in the PAG and CCK receptors are heterogeneously distributed within the PAG. Since CCK antagonists are anxiolytic and potentiate morphine analgesia, it is possible that these effects of CCK are mediated through alteration of neuronal activities in the PAG. The goals of this study were to examine the anatomical and physiological properties of the PAG CCK containing systems. The distribution of CCK-containing axons and boutons in PAG was examined using immunohistochemical procedures. These studies show that CCK-like immunoreactive (CCK-LIR) fibers and terminals are present throughout PAG, but are particularly heavily concentrated in a focal column that runs longitudinally throughout the rostrocaudal axis of dorsolateral PAG and in nucleus cuneiformis which represents a caudolateral extension of PAG. The physiological effects of CCK on PAG neurons were examined in both in vivo and in vitro preparations. In the in vivo experiments multibarreled electrodes were used to record from PAG neurons and to apply CCK and the CCK antagonists, CR1409 and proglumide. Of 37 neurons recorded in vivo, CCK caused excitation in 25 cells, inhibited 7 cells and had no effect on 5 cells. The excitatory effect was blocked by CR1409 in 11/11 cells tested. Proglumide blocked the excitatory response of CCK in 12/14 cells. Proglumide blocked the inhibitory effect in 2 of 7 cells, but CR1409 had no effect on CCK-evoked inhibition in 7 cells tested. Extracellular, conventional intracellular and whole cell patch clamping procedures were used to study CCK actions in the in vitro slice preparation. In the extracellular recording experiments, responses of PAG cells to CCK were measured in slices that were maintained at 22 degrees C (room temperature) and at 32 degrees C. CCK excited 40/56, inhibited 7/56 and had no effect on 9/56 cells; excitatory responses were blocked by CR1409 in 32/36 cells and by proglumide in 25/27 cells tested. Inhibitory responses to CCK were unaffected by CR1409, but were blocked in 3/7 cells by proglumide. Conventional intracellular recordings were made from 13 cells.(ABSTRACT TRUNCATED AT 400 WORDS)

Pilocarpine-induced convulsions in rats : evidence for muscarinic receptor-mediated activation of locus coeruleus and norepinephrine release in cholinolytic seizure development

We recently reported that systemic administration of the anticholinesterase, soman, caused rapid depletion of forebrain norepinephrine (NE) in convulsive but not in nonconvulsive rats. As neurons in nucleus locus coeruleus (LC) provide the bulk of NE innervation to most of the forebrain and the sole source of NE input to the cortex and the olfactory bulb, soman-induced NE depletion was hypothesized to result from activation of LC neurons. This activation was thought to be due to inhibition of acetylcholinesterase by soman, leading to rapid, sustained accumulation of acetylcholine in LC, causing these cells to fire at a high sustained rate. Support for this hypothesis was provided by neurophysiological findings showing that: (i) Systemic administration of soman in anesthetized rats caused a sustained, fivefold increase in the mean firing rate of LC neurons and (ii) microinjections of soman directly into LC caused a similar increase in the firing rate of LC neurons. Soman-induced activation of LC occurred prior to and even in the absence of seizures. As systemic administration of the muscarinic receptor antagonist, scopolamine, rapidly and completely reversed soman-induced activation of LC, it was further hypothesized that activation of LC neurons following soman administration is due to muscarinic receptor stimulation. The rapid release of NE by cholinolytic agents, thus, may play an important role in the initiation and/or maintenance of convulsions. To further test the hypothesis that NE release in soman-intoxicated rats is due to muscarinic activation of LC, we have investigated the effects of the muscarinic receptor agonist, pilocarpine, on NE release and LC discharge. In one set of experiments, rats were injected with a periconvulsive dose of pilocarpine (300 mg/kg, ip); both convulsive and nonconvulsive rats were sacrificed between 1 and 96 h and monoamine levels in the rostral forebrain and olfactory bulb were determined by HPLC with electrochemical detection. NE levels declined substantially only in convulsive rats; forebrain NE levels in convulsive rats rapidly decreased to 50% of control levels at 1 h and to 37% of controls level between 2 and 4 h. The time course and magnitude of these changes were similar to those observed following soman administration in our previous study. Recovery of forebrain NE began at 8 h and was complete by 96 h following pilocarpine administration. Neither dopamine (DA) nor serotonin (5-HT) levels were changed in the forebrain and olfactory bulb of either convulsive or nonconvulsive rats.(ABSTRACT TRUNCATED AT 400 WORDS)