Hcr1

Figure 3. Hcrt-evoked depolarization demonstrates voltage dependence. (a1,2) Two depolarizing responses to a puff application of Hcrt-2 (as in Fig. 1). Both responses were recorded in the same cell, either at resting membrane potential (a1), or 10 min later when the cell was hyperpolarized to -85mV (a2). Note a smaller depolarization in the latter. (b) Three puff applications of Hcrt-2 (30 s application) to the same cell, at 10 min intervals and different membrane potentials as indicated. The first application evoked a clear depolarization at resting membrane potential (b1). The second application, after the membrane potential was shifted to approx. -90 mV, produced no depolarization (b2). The third application again depolarized the cell when the membrane potential was returned to the previous resting level (b3). TTX (1 uM) and Co2+ (1 mM) were added to the bath 3 min before trace b1 was recorded, and were continuously present during the following records. Note that Ca2+ spikes, and associated membrane oscillations, but not the Hcrt-induced depolarization, were blocked by Co2+. The dotted lines with the numbers below indicate the level of membrane potential in all records. a and b represent LC neurons from different slices. Taken from ref 7.

Figure 3. Hcrt-evoked depolarization demonstrates voltage dependence. (a1,2) Two depolarizing responses to a puff application of Hcrt-2 (as in Fig. 1). Both responses were recorded in the same cell, either at resting membrane potential (a1), or 10 min later when the cell was hyperpolarized to -85mV (a2). Note a smaller depolarization in the latter. (b) Three puff applications of Hcrt-2 (30 s application) to the same cell, at 10 min intervals and different membrane potentials as indicated. The first application evoked a clear depolarization at resting membrane potential (b1). The second application, after the membrane potential was shifted to approx. -90 mV, produced no depolarization (b2). The third application again depolarized the cell when the membrane potential was returned to the previous resting level (b3). TTX (1 uM) and Co2+ (1 mM) were added to the bath 3 min before trace b1 was recorded, and were continuously present during the following records. Note that Ca2+ spikes, and associated membrane oscillations, but not the Hcrt-induced depolarization, were blocked by Co2+. The dotted lines with the numbers below indicate the level of membrane potential in all records. a and b represent LC neurons from different slices. Taken from ref 7.

terminal innervation than other nearby regions, indicating that it is a highly favored target.

Although Hcrt neurons are located within a circumscribed area of the hypothalamus, it was not known if the dense LC innervation originated in a specific region or cluster of

Hcrt neurons. Thus, we sought to determine if the Hcrt neurons were composed of projection specific subpopulations. As a first step in that effort we analyzed the distribution and density of Hcrt neurons in the rat and monkey hypothalamus to determine if there was evidence of topographical subdivisions. As shown in Fig. 4, in rat we delineated 3 subregions of the hypothalamus area that contains Hcrt neurons: the dorsomedial, perifornical and lateral hypothalamus (mediorostral to laterocaudal, respectively). In monkey, four clusters of Hcrt neurons were evident along the rostro-caudal axis of the hypothalamus. These rough delineations were based upon apparent groupings of Hcrt neurons in the rostrocaudal and mediolateral axes, and proximity of Hcrt neurons to hypothalamic landmarks.

To determine if the topographical delineations that we observed represented segregation of projection specific populations of neurons, we conducted tract-tracing studies with the beta subunit of cholera toxin (CTb), a retrograde tracer. Injections of CTb that encompassed the core LC nucleus and peri-LC dendritic zone yielded numerous retrogradely labeled neurons in the hypothalamus of both rats and monkeys. Many of these retrogradely labeled LC afferent neurons also stained for Hcrt in both species, confirming the hypothalamic input that is presumed to underlie the Hcrt innervation of the LC (Fig. 4). Notably, there were differences among the subgroups of Hcrt neurons in the numbers and percentages of cells that were retrogradely labeled from the LC. In rat, a significantly higher percentage of LC afferents in the DMH and perifornical region stained for Hcrt than in the lateral hypothalamus (16%, 15% and 10%, respectively; p<0.05). Analyzed differently, we found that a significantly higher percentage of Hcrt neurons in the DMH were retrogradely labeled from the LC than in the perifornical or lateral hypothalamus (18%, 6% and 5%, respectively; p<0.05). These results indicate that the bulk of Hcrt input to the rat LC originates in the DMH and perifornical hypothalamus, and that Hcrt neurons in the DMH are more likely to project to the LC than Hcrt neurons in other hypothalamic subfields. The functional significance of this topography of Hcrt projections to the LC warrants further investigation, but may correspond to suprachiasmatic nucleus (SCN)-mediated circadian regulation of LC activity and arousal, as described later in this chapter.

4. GABA INTERNEURONS IN THE PERI-LC: TARGET FOR HCRT INPUTS

We recently examined inputs to the LC nucleus from the neurons in the dendritic field of the LC using microinjection of retrograde tract-tracers into the LC nucleus proper 18. Focal microinjections of the sensitive tracer wheat germ agglutinin conjugated to apo (inactivated) horseradish peroxidase and coupled to colloidal gold (WGA-Au) labeled a discrete population of small neurons in the peri-LC surround. Double labeling studies revealed that these neurons often stained for the GABA synthetic enzyme glutamate decarboxylase (GAD; Fig. 5). Counts of neurons in 12 sections from 3 rats demonstrated that approximately 45% of WGA-Au-labeled afferents to the LC in the peri-LC zone also stained for GAD. Thus, these local neurons contribute to the dense GABA fiber input known to innervate the LC nucleus.18

In this same study, we also used the retrograde transynaptic tracer Pseudorabies virus (PRV) to study the possible inputs to distal LC dendrites in the peri-LC area.1718 This tracer is particularly well suited for such a study because following focal injection into a cell body area such as the LC it is transported out distal LC dendrites where it

Rat Hypothalamus

Figure 4. Upper: Frontal sections at low-power through the rat hypothalamus double stained for Hcrt (brown) and CTb (retrogradely transported from the LC, black). These images depict the different areas that were defined for counting Hcrt+ neurons: DMH - dorsomedial hypothalamus; PeF - perifornical hypothalamus; LH -lateral hypothalamus. Other abbreviations: 3V - third ventricle; mt - mammilothalmic tract; f - fornix. Lower: High-power photo taken from box in upper panel, showing doubly labeled neurons in the DMH and PeF regions.

Figure 4. Upper: Frontal sections at low-power through the rat hypothalamus double stained for Hcrt (brown) and CTb (retrogradely transported from the LC, black). These images depict the different areas that were defined for counting Hcrt+ neurons: DMH - dorsomedial hypothalamus; PeF - perifornical hypothalamus; LH -lateral hypothalamus. Other abbreviations: 3V - third ventricle; mt - mammilothalmic tract; f - fornix. Lower: High-power photo taken from box in upper panel, showing doubly labeled neurons in the DMH and PeF regions.

transsynaptically labels afferents to those dendrites.19 Following focal PRV injections into the LC, we found that neurons in areas known to be direct afferents to the LC (e.g., the nucleus paragigantocellularis)20 became labeled with PRV at about 23 hr of survival. Many neurons in the peri-LC also exhibited PRV labeling at this time, consistent with the observations above made with WGA-Au indicating direct inputs to the LC proper. However, a time-course analysis revealed that PRV+ neurons became much more numerous in the peri-LC at 35 hr of survival (188.2 + 20.5 vs. 74.8 + 9.7 cells per section, respectively). These results indicate that peri-LC labeling is mediated by PRV transport and replication rather than spread of injection outside of the LC. We interpret the differing magnitude of infection of peri-LC neurons at 23 hours and 35 hours as evidence for two populations of neurons with differing synaptic relations with the LC. Specifically, we postulated that the differing temporal course of infection of the two populations of neurons reflected either first-order infection through projections into the LC nuclear core (23 hour group) or second-order infection resulting from transneuronal

Brain Gad Staining

Figure 5. LC afferents in peri-LC zone stain for glutamic acid decarboxylase (GAD). Photomicrographs of frontal sections through the ventral LC and ventromedial peri-LC showing neurons labeled for GAD and the retrograde tracer WGA-Au after injection into the LC nuclear core. Arrows indicate neurons stained for both WGA-Au and GAD. A, B, Bright-field photomicrographs of mid- ( A) and rostral ( B) peri-LC areas stained for both GAD (brown) and the retrograde tracer WGA-Au (black particulate) in a case with a focal WGA-Au injection in the LC nuclear core. C, D, High-power photos taken from areas in A and B. Note that WGA-Au-labeled LC afferents (filled with black particles) often also stained for GAD (diffuse brown stain). Arrows indicate neurons stained for both WGA-Au and GAD. For all panels, ventral is down, medial is left. Scale bars: (in A) A, B, 250 um; C, D, 50 um. Taken from.18

Figure 5. LC afferents in peri-LC zone stain for glutamic acid decarboxylase (GAD). Photomicrographs of frontal sections through the ventral LC and ventromedial peri-LC showing neurons labeled for GAD and the retrograde tracer WGA-Au after injection into the LC nuclear core. Arrows indicate neurons stained for both WGA-Au and GAD. A, B, Bright-field photomicrographs of mid- ( A) and rostral ( B) peri-LC areas stained for both GAD (brown) and the retrograde tracer WGA-Au (black particulate) in a case with a focal WGA-Au injection in the LC nuclear core. C, D, High-power photos taken from areas in A and B. Note that WGA-Au-labeled LC afferents (filled with black particles) often also stained for GAD (diffuse brown stain). Arrows indicate neurons stained for both WGA-Au and GAD. For all panels, ventral is down, medial is left. Scale bars: (in A) A, B, 250 um; C, D, 50 um. Taken from.18

passage of virus from dendrites of infected LC neurons (35 hours group) 18. This time course is consistent with the replication cycle of PRV.21

We extended these studies using electron microscopy (EM) to ultrastructurally examine the innervation of LC neurons by these peri-LC GABA neurons. In counts of EM material, we found that 101/131 (77%) PRV-labeled neurons in the peri-LC stained for GABA. Moreover, GABA terminals frequently made symmetrical synaptic contacts onto PRV+ or TH+ dendrites in the peri-LC region. Some of these GABA terminals contained PRV particles, as did some of the TH+ dendrites that received input from GABA terminals (Fig. 6). These results are consistent with the interpretation that PRV virions were transported into peri-LC GABA neurons from LC dendrites, indicating a synaptic connection from local peri-LC GABA neurons to LC distal dendrites. We also found that GABA+ somata and dendrites in the peri-LC received numerous symmetrical and asymmetrical synaptic inputs, indicating that these GABA neurons integrate numerous inputs to this region.

As noted above, the peri-LC dendritic zone that contains these GABA interneurons also contains dense Hcrt innervation. In recent studies, we carried out ultrastructural

Figure 6. Ultrastructure of GABAergic LC afferents. EM photomicrograph of a neuron in the rostroventral peri-LC immunolabeled for PRV (stained with DAB) and GABA (stained with silver-intensified gold). Highpower images in B and C are taken from areas indicated in the low-power image in A. Note the virions in the nucleus (N) at arrows in A and the nuclear inclusions in A and B, indicative of PRV infection (at arrowheads). Note also the cytoplasmic virions (indicated by arrows in B and C), and abundant GABA staining in the cytoplasm. Scale bars: A, 2 um; B, C, 0.5 um. Taken from ref 18.

Figure 6. Ultrastructure of GABAergic LC afferents. EM photomicrograph of a neuron in the rostroventral peri-LC immunolabeled for PRV (stained with DAB) and GABA (stained with silver-intensified gold). Highpower images in B and C are taken from areas indicated in the low-power image in A. Note the virions in the nucleus (N) at arrows in A and the nuclear inclusions in A and B, indicative of PRV infection (at arrowheads). Note also the cytoplasmic virions (indicated by arrows in B and C), and abundant GABA staining in the cytoplasm. Scale bars: A, 2 um; B, C, 0.5 um. Taken from ref 18.

studies using immunocytochemistry to examine whether these GABA interneurons receive direct Hcrt innervation.22 For this, we examined Hcrt and GABA staining in the peri-LC of rats that had received focal PRV injections into the LC. Results revealed that a high percentage of Hcrt terminals contacted GABA+ profiles, and virtually all of the Hcrt-GABA synaptic contacts were asymmetrical. This high prevalence of Hcrt-GABA contacts indicates that GABA elements are a prominent Hcrt target in the peri-LC region. The majority of such contacts were onto dendrites. In cases with PRV injections into the LC nuclear core, PRV profiles were frequently seen in GABA+ dendrites or somata that were post-synaptic to Hcrt+ terminals (Fig. 7). Again, the prevalent synaptic arrangement was asymmetrical. These results indicate that Hcrt fibers not only make contact onto noradrenergic LC neurons as previously reported, but also make frequent contact onto neighboring GABAergic neurons that in turn contact LC neurons. The finding that the bulk of these Hcrt-GABA contacts are asymmetrical indicates that Hcrt may be excitatory on these GABA interneurons.

As GABA is strongly inhibitory on LC neurons,23'24 these results in turn indicate that Hcrt has both direct excitatory effects on LC neurons (as reviewed above) and inhibitory effects on LC neurons via a feedforward circuit involving local GABA interneurons. Thus, the effect of activity in Hcrt neurons on LC function is more complex than originally envisioned. As these GABA interneurons that synapse onto LC cells integrate a number of different inputs, the effect of Hcrt inputs on LC activity may be gated or modulated by other afferents onto these GABA neurons. Additional studies are called for to understand the functional consequences of this additional level of integration for Hcrt influences on LC activity.

Figure 7. Electron micrographs from the peri-LC area in a case in which PRV was injected into the LC nuclear core. Tissue was immunolabeled for Hcrt (DAB reaction product) and GABA (silver-intensified gold). PRV profiles (indicated at arrowheads) were present within GABA+ profiles (cells and dendrites) either as individual virions (panels A and B) or sequestered within multivesicular bodies (panels C and D). The morphology of virions and their presence within multivesicular bodies is consistent with the life cycle previously established for PRV in the rodent CNS (Card et al., 1993). Arrows indicate synaptic contacts between Hcrt+ and GABA+ profiles infected with PRV.

Figure 7. Electron micrographs from the peri-LC area in a case in which PRV was injected into the LC nuclear core. Tissue was immunolabeled for Hcrt (DAB reaction product) and GABA (silver-intensified gold). PRV profiles (indicated at arrowheads) were present within GABA+ profiles (cells and dendrites) either as individual virions (panels A and B) or sequestered within multivesicular bodies (panels C and D). The morphology of virions and their presence within multivesicular bodies is consistent with the life cycle previously established for PRV in the rodent CNS (Card et al., 1993). Arrows indicate synaptic contacts between Hcrt+ and GABA+ profiles infected with PRV.

Feedforward Inhibition
Figure 8. Illustration of direct excitatory Hcrt innervation of LC neurons, and of feedforward inhibitory Hcrt influence on LC via GABA interneurons that project to LC.

5. A CIRCUIT FROM THE SCN TO THE LC AND CIRCADIAN REGULATION OF AROUSAL: A POSSIBLE ROLE FOR HCRT?

Neurons in the hypothalamic SCN exhibit an endogenous circadian rhythm of activity that plays a prominent role in the temporal organization of behaviors. However, the projections of the SCN are largely confined to the hypothalamus and there is no direct projection to the LC. Given that both the SCN and the LC play prominent roles in the regulation of behavioral state we sought to determine if the SCN influences LC activity through polysynaptic connections. Toward that end we demonstrated that focal microinjection of PRV into the rat LC infected a large number of neurons in the SCN.25 This labeling occurred at a survival time of greater than 44 hr, indicating an indirect circuit connection, consistent with prior work indicated that there is no direct projection from the SCN to the LC. To identify the relay through which the SCN became infected we conducted a series of double labeling anatomical and lesion studies. Our results showed that the dorsomedial nucleus of the hypothalamus (DMH) was a major relay in this circuit. For example, as shown in Fig. 9 cell body-specific lesions of the DMH substantially decreased the number of PRV-labeled neurons in the SCN ipsilateral to the injected LC.25

These results, in view of the role of the SCN as the brain's primary regulator of circadian rhythmicity, indicated that the LC may have a circadian rhythm. This

Was this article helpful?

0 0
Natural Cures For Insomnia

Natural Cures For Insomnia

How Would You Love To Finally Get A Decent Night's Sleep? Do you toss and turn all night long without getting much of a wink of sleep? Are you tired of counting sheep for hours without falling asleep? Wouldn't you love to be able to fall asleep simply, easily and naturally, without pills, potions or harmful medicine?

Get My Free Ebook


Post a comment