| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Journal of Physiology (2001), 537.1, pp. 161-177
© Copyright 2001 The Physiological Society
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The paraventricular nucleus (PVN) of the hypothalamus is an important integrating site for autonomic and endocrine function (Swanson & Sawchenko, 1980). This nucleus is a complex heterogeneous region consisting of magnocellular and parvocellular neurones that are largely segregated into specific anatomical compartments (Swanson & Sawchenko, 1983). The parvocellular PVN comprises different neuronal types, including neuroendocrine neurones that project to the median eminence and regulate the release of hormones from the anterior pituitary gland, and pre-autonomic neurones that send long descending projections to brainstem and spinal cord regions that are important with respect to autonomic control. These include areas in the ventral and dorsal medulla, such as the nucleus of the tractus solitarius (NTS), the dorsal motor nucleus of the vagus (DMX) and the rostral ventrolateral medulla (RVLM) (Armstrong et al. 1980; Swanson & Kuypers, 1980). Both anatomical and electrophysiological studies also demonstrate that sympathetic preganglionic motoneurones located in the intermediolateral cell column of the thoraco-lumbar spinal cord are directly innervated by neurones in the PVN (Saper et al. 1976; Coote et al. 1998; Ranson et al. 1998), whose firing activity was shown to follow changes in blood pressure (Lovick & Coote, 1988).
A large bulk of physiological data supports a role for the PVN in cardiovascular function. For example, the PVN has been shown to play a role in the control of the baroreceptor reflex (Zhang & Ciriello, 1985a,b). Electrical and chemical activation of the PVN evoked vasodilatation and vasoconstrictor responses, with concomitant changes in blood pressure (Porter & Brody, 1985, 1986; Kannan et al. 1988; Malpas & Coote, 1994). Parvocellular neurones located in the medial and ventral subdivisions of the PVN are activated during haemorrhage (Badoer et al. 1993), and changes in renal sympathetic nerve discharge have also been attributed to PVN activation (Kannan et al. 1988). Moreover, it has been suggested that the PVN is involved in the development and/or maintenance of cardiovascular-related diseases such as hypertension (Goto et al. 1981; Ciriello et al. 1984) and ischaemic heart failure (Patel & Zhang, 1996; Patel et al. 2000).
Despite the strong evidence for the PVN as an important site for cardiovascular control, less is known about the cytoarchitectural and physiological properties of identified PVN pre-autonomic parvocellular neurones. Recently, Barrett-Jolley et al. (2000) demonstrated the expression of voltage-gated K+ (Kv) currents in dissociated PVN neurones, retrogradely identified as projecting to the intermediolateral column in the spinal cord. Although previous in vitro studies have looked in detail into the electrophysiological properties of PVN neurones, none of these studies were carried out on identified PVN neurones projecting to autonomic-related areas (Hoffman et al. 1991; Tasker & Dudek, 1991; Luther & Tasker, 2000). In the present study, identification of PVN pre-autonomic neurones was achieved by combining in vivo anatomical retrograde tracing techniques with in vitro patch-clamp recordings of labelled neurones visualized in adult hypothalamic brain slices. Using these techniques, a detailed characterization of the cellular properties of identified PVN pre-autonomic neurones was obtained. These results indicate that even though PVN pre-autonomic neurones display distinctive morphological and electrical properties that distinguish them from other PVN neuronal types, they constitute a heterogeneous neuronal population.
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Male Sprague-Dawley rats (250-350 g) were purchased from Harlan Laboratories (Indianapolis, IN, USA). Rats were maintained under a 12 h:12 h light-dark cycle and given free access to food and water.
All procedures were carried out in agreement with the Wright State University Institutional Animal Care and Use Committee guidelines and were approved by the committee. All surgical procedures were performed in the morning and animals were closely monitored until the dark cycle. After this period, animals were inspected daily for signs of infection at the surgical incision point. Animal weight and daily indices of food and water intake were monitored to assess the health status of the animals. Post-surgical pain associated with surgical incisions was minimized by topical application of the local analgesic bupivicaine (5 %, 0.1 ml, S.C., single dose). Typically, animals were conscious and moving around within 1-2 h, and showed no obvious sign of pain or discomfort. A total of 45 rats were used for these studies.
Retrograde labelling of PVN pre-autonomic neurones
Rats were anaesthetized by I.P. injection of a ketamine-xylazine mixture (90 and 5 mg kg-1, respectively). The adequacy of anaesthesia was determined by the absence of a response to noxious stimulation (tail pinch method). A single dose of anaesthetic was sufficient to keep the animal adequately anaesthetized for the whole surgical procedure. The head of the rat was placed in a stereotaxic frame and the dorsal medulla was exposed after retraction of the overlying muscles and occipital membranes. A small part of the occipital bone was removed to increase exposure of the medulla. The tracer 1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine (DiI; Molecular Probes, Eugene, OR, USA) was dissolved in DMSO (1 %) and was pressure-injected unilaterally (100-200 nl) into the dorsal vagal complex (DVC) at the level of the obex, which included the NTS and the DMX (Fig. 1A). The injection point was 1.0 mm lateral to the midline, and 0.8 mm below the dorsal surface. After the injection, the muscles were sutured together and the wound was closed. Animals usually recovered from the anaesthesia within about 20-30 min. Five to seven days after surgery the rats were anaesthetized with sodium pentobarbitone (50 mg kg-1 I.P.) and rapidly decapitated, and the brain dissected for electrophysiological experiments. As verified histologically (see below), this tracer injection protocol reliably labelled PVN neurones in the ventral parvocellular (PaV) and posterior parvocellular (PaPo) subnuclei of the PVN (see Fig. 1B).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 1. Identification of PVN pre-autonomic neurones in brain slices using retrograde labelling techniques A, low-power photomicrograph depicting the retrograde tracer injection site at the level of the DVC. Two images (bright light and fluorescence light) were taken from the same section and superimposed, to show clearly the location and extension of the DiI injection (red area, arrow). B, confocal digital image of DiI-labelled neurones in a coronal slice (300 µm) containing the PVN (at the level of the PaV subnucleus). Slices were obtained 7 days after an injection of DiI into the DVC. Note that most of the labelling was unilateral. C, example of a PVN DiI-labelled neurone at higher magnification. D, photomicrograph of the same neurone as in C after ABC-DAB staining. E, for morphometric analysis, the neuronal dendritic tree was traced and reconstructed in three dimensions using a computer-assisted program (see Methods). 3V, third ventricle; CC, central canal. | ||
Hypothalamic slices
Coronal hypothalamic slices (300 µm thick) containing the PVN were obtained as described previously (Stern et al. 1999). Rats were anaesthetized with sodium pentobarbitone (50 mg kg-1, I.P.) and perfused through the heart with a cold medium in which NaCl was replaced by an equiosmolar amount of sucrose. Slices were cut using a vibroslicer (S.S.K., Microslicer, Ted Pella, Redding, CA, USA). An ice-cold standard solution was used during slicing, which contained (mM): 126 NaCl, 2.5 KCl, 1.25 NaH2PO4, 1 MgSO4, 2 CaCl2, 26 NaCO3, 10 glucose and 0.4 ascorbic acid; pH 7.4 (315-320 mosmol l-1). After sectioning, slices were placed in a holding chamber containing standard solution at 32-34 °C for 60 min and then stored at room temperature until they were used. After an incubation period of at least 1 h, slices were transferred to a submersion-type recording chamber. The solutions bathing the slices (~2 ml min-1) were kept at room temperature (22-24 °C) and bubbled continuously with a gas mixture of 95 % O2-5 % CO2.
Recording and data analysis
Patch pipettes (3-5 M
) were pulled from thin-walled (1.5 mm o.d., 1.17 mm i.d.) borosilicate glass (GC150T-7.5, Clark, Reading, UK) on a horizontal electrode puller (P-97, Sutter Instruments, Novato, CA, USA). The pipette internal solution contained (mM): 135 potassium gluconate, 20 KCl, 10 Hepes, 4 MgATP, 20 phosphocreatine (Na+), 0.3 NaGTP and 0.2 EGTA; pH 7.3 (295 mosmol l-1). For labelling neurones, biocytin (0.2 %) was added to the internal solution (Horikawa & Armstrong, 1988). Whole-cell recordings from PVN pre-autonomic neurones were made under visual control using a combination of epifluoresence illumination and infrared differential interference contrast (IR-DIC) videomicroscopy. Briefly, DiI-labelled neurones were first identified with the aid of epifluorescence illumination (rhodamine filter, Fig. 1C). A tight giga-ohm seal was subsequently obtained in the selected neurone under IR-DIC illumination. Recordings were obtained with an Axopatch 200B amplifier (Axon Instruments, Foster City, CA, USA). The data shown was not corrected for the pipette liquid junction potential (measured to be ~10 mV). The voltage output was digitized at 16-bit resolution in conjunction with pCLAMP 8 software (Digidata 1320, Axon Instruments). Data were digitized at 10 or 20 kHz and transferred to a disk. The series resistance was monitored frequently and the experiment was terminated if it was not stable throughout the recording. Cell input resistance was calculated in voltage-clamp mode using a 5 mV pulse while holding the cells at -70 mV. To measure action potential properties, neurones were current clamped close to threshold and a 5 ms, 10 pA depolarizing pulse was applied. Action potential height and width (measured at 50 % of the peak, from threshold) were calculated using algorithms provided by computer software (Axograph and Clampfit, Axon Instrutments). The action potential threshold was defined as the point at which the first derivative of voltage with respect to time exceeded 4 V s-1, which corresponded to a sharp inflection in the voltage trace. Current-evoked firing behaviour was evaluated by measuring successive interspike intervals (ISIs) during a 200-280 ms pulse using different current injection amplitudes, while current clamping the neurones at a membrane potential just below the action potential threshold. Time-dependent inward rectification was quantified as the percentage change in membrane potential from the initial peak to the end of a 400 ms hyperpolarization to -90 mV. The firing rate of spontaneously active neurones was calculated as 1/mean ISI. Following electrophysiological recordings, the pipette was withdrawn from the neurone and the slice was processed for histology and/or immunohistochemistry (see below). Recordings were also obtained from magnocellular neurones located in the lateral magnocellular subnucleus (n = 4). These neurones were subsequently identified immunologically as vasopressin-neurophysin-containing cells, as described below.
Histology, immunohistochemistry and neuronal reconstruction
After the recordings, slices were fixed in 4 % paraformaldehyde and 0.2 % picric acid, dissolved in 0.15 M phosphate buffer (pH ~7.3). Magnocellular neurones recorded in this study were identified immunologically, as described previously (Stern & Armstrong, 1997). Briefly, slices were rinsed in phosphate-buffered saline (PBS) and incubated overnight in avidin-AMCA (Vector Labs) diluted 1:2000 in PBS containing 0.5 % Triton X-100. Incubation in avidin-AMCA allowed the visualization under fluorescent light of the recorded (biocytin-filled) neurone (Kita & Armstrong, 1991). Slices were then incubated in PBS containing 0.5 % Triton X-100 in the presence of a rabbit antiserum specific for vasopressin-neurophysin (provided by Alan Robinson, UCLA School of Medicine) at a dilution of 1:30 000, followed by the appropriate fluorescein-labelled, donkey anti-rabbit IgG secondary antibody, at a dilution of 1:250 (Jackson ImmunoResearch).
For neuronal reconstruction of PVN pre-autonomic neurones, slices containing the recorded cells were stained with the avidin-biotin complex (ABC)-diaminobenzidine tetrahydrochloride (DAB) method (Fig. 1D). Briefly, slices were incubated overnight in ABC (Vector Laboratories) diluted 1:100 in PBS containing 0.5 % Triton X-100. After thorough rinsing in PBS, sections were reacted with DAB (60 mg (100 ml)-1) in the presence of H2O2 (0.006 %) and nickel ammonium sulphate (0.05 %) for 10-20 min. Sections were then rinsed, mounted on gelatin-coated glass slides and dried for 24 h. In order to verify the location of the recorded neurone within the various subnuclei of the PVN (Armstrong et al. 1980; Swanson & Kuypers, 1980), sections were counterstained with cresyl violet (0.3 %), dehydrated and differentiated in an ascending series of ethyl alcohol concentrations, immersed in two changes of xylene, and coverslips were applied with Permount (Fischer Scientific).
For a morphometric analysis of the dendritic trees, neurones were reconstructed in three dimensions, using a computer-aided tracing system (Neurolucida, Microbrightfield, Colchester, VT, USA; see for example Fig. 1E). The detailed reconstruction of the dendrites was made using a 
60 objective. The course of each dendrite was traced by digitizing the X, Y and Z coordinates and the width of the dendrite along its entire extent. The length and surface area of the dendritic branches were calculated by algorithms provided by Microbrightfield software. Total dendritic length (TDL) and the total number of branches were obtained by summing the data for each dendritic tree. The mean dendritic length (MDL) is the average length of individual branches. Path length is the distance from the origin of a first order branch to the end point of a terminal branch. The concentric sphere method of Sholl (1953) was also used to analyse the branching patterns of dendritic trees. Briefly, concentric spheres of a constant interval of 20 µm, with the centre of the soma as the origin, were drawn for each filled neurone. The number of dendritic intersections encountered within each circle was counted, and the means and standard errors were calculated and plotted as a function of the distance to the soma.
Statistical analysis
A one-way ANOVA, followed by Scheffé's t test was used for comparing electrophysiological and morphological properties among neuronal types. To compare the incidence of different firing patterns, as well as the incidence of a dendritic vs. somatic axonal origin, frequencies of observations were arranged in a contingency table, and a chi-square (
2) statistic was used for analysis (Zar, 1984). Correlation of morphological and electrophysiological data was obtained using a correlation Z test. All results shown represent the mean ± S.E.M.
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Whole-cell patch-clamp recordings were obtained from 79 identified PVN pre-autonomic neurones, located in the PaV (n = 44) and the PaPo (n = 35) subnuclei of the PVN. Of these, 65 were reconstructed for morphometric analysis. All recorded neurones had resting membrane potentials of at least -50 mV (range, -70 to -50 mV), an apparent input resistance larger than 250 M (range, 250-2400 M) and action potentials larger than 60 mV (range, 60-96 mV).
As shown in Fig. 2, PVN pre-autonomic neurones constitute a heterogeneous population in which at least three distinct morphological cell types could be distinguished. The first category of neurones displayed an oblique orientation with respect to the third ventricle (3V), with at least one long dendritic tree running ventromedially towards the ventricle. These neurones were located exclusively in the PaV subnucleus (Fig. 2A, 41/79 (52 %) of recorded neurones), and will henceforth be referred to as PVN pre-autonomic type A neurones. A second category of neurones had a perpendicular orientation with respect to the 3V, and showed elongated bipolar or tripolar dendritic trees that ran and branched mediolaterally, in a perpendicular plane with respect to the 3V. These neurones were located exclusively in the PaPo subnucleus (Fig. 2B, 20/79 (25 %) of recorded neurones), and will henceforth be referred to as PVN pre-autonomic type B neurones. The third category of neurones (18/79 (23 %) of recorded neurones) showed a more compact and concentric dendritic configuration, with dendrites running both in the mediolateral and dorsomedial direction, but mostly towards the ventricle. The great majority of these neurones were located in the PaPo subnucleus (15/18), while the rest were located in the PaV subnucleus (Fig. 2C). These neurones will be referred to as PVN pre-autonomic type C neurones. Thus, neuronal heterogeneity in PVN pre-autonomic neurones can be in part related to the anatomical distribution of neurones within the PVN, with type A neurones located in the PaV subnucleus, and type B and C neurones concentrated in the PaPo subnucleus (P < 0.0001, 2 test).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 2. Morphological subtypes of PVN pre-autonomic neurones A-C, typical examples of type A, B and C PVN pre-autonomic neurones, respectively. Arrows point to the axons. In the insets, neurones were scaled to the same proportion. Their location within the PVN and their respective dendrograms are displayed. | ||
With respect to their electrophysiological characteristics, most PVN pre-autonomic neurones (88 %), regardless of their morphological type, were characterized by the presence of a low-threshold spike (LTS), which evoked a burst of action potentials (see Fig. 3 and Fig. 7). This property distinguished PVN pre-autonomic neurones from neighbouring magnocellular vasopressin neurones, which were characterized by a strong transient outward rectification, as shown previously (Hoffman et al. 1991; Renaud & Bourque, 1991; Armstrong et al. 1994). These properties correspond very well with those described previously for type I (putative magnocellular) and type II/III (putative parvocellular) PVN neurones (Hoffman et al. 1991; Tasker & Dudek, 1991; Luther & Tasker, 2000).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 3. Typical features of PVN pre-autonomic type A, B and C neurones Examples of PVN pre-autonomic neurone types A (A1-A3), B (B 1-B 3) and C (C 1-C 3) located in the PaV (A1) and in the PaPo (B 1 and C 1) subnuclei, respectively. A2-C 2, photomicrographs corresponding to the boxed areas shown in A1-C 1, respectively. Arrows point to the axons. Note that the axon in the type A (A2) neurone shown originated from a proximal dendrite, whereas the axon in the type B (B 2) and type C (C 2) neurones shown originated from the soma. A3-C 3, electrophysiological recordings obtained from the same neurones. Low-threshold spikes (LTSs; arrows), were evoked by depolarizing pulses (0.05-0.07 nA) while clamping the neurones at approximately -80 mV (upper traces), or at the offset of hyperpolarizing pulses (-0.05 to -0.07 nA), while clamping the neurones at approximately -50 mV (lower traces). Note also the presence of an inward sag during the hyperpolarizing pulses (arrowheads in lower trace). | ||
Several electrophysiological characteristics, including apparent input resistance, action potential waveform and repetitive firing properties, differed between PVN pre-autonomic neuronal types, and will be described below for each category of neurone.
Morphological properties of PVN pre-autonomic neurones
PVN pre-autonomic type A neurones. PVN pre-autonomic type A neurones were located exclusively in the PaV subnucleus and were the most common cell type labelled, accounting for 52 % of all labelled neurones in the PVN (Fig. 3A). They had a mean cross-sectional soma area of 134.8 ± 15.4 µm2. Dendrites were often varicose, and short spinous processes were occasionally observed. A common observation (also found for the other neuronal types) was that distal dendritic branches extended beyond the boundaries of the subnucleus and tended to approach the walls of the 3V. A morphometric analysis of reconstructed neurones showed that type A neurones (n = 33) had 2.7 ± 0.1 primary dendrites, which gave rise to 6.4 ± 0.7 branches. The TDL, MDL and mean path length (MPL) were 2093.7 ± 215.8 µm, 766.6 ± 91.4 µm and 399.7 ± 33.2 µm, respectively (see Fig. 4, for comparison with the other neuronal types). Axons were identified by their thinner diameter and beaded appearance. In most cases, axons were artificially cut at the slice surface, but could be traced for several hundred micrometres, running laterally or ventrolaterally towards the lateral hypothalamic area. In 25/33 (78 %) type A neurones, axons arose from a primary dendrite at a mean distance from the soma of 41 ± 5 µm (Fig. 3A1 and A2). In the remaining type A neurones, axons arose directly from the soma.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 4. Morphometric analysis of the dendritic organization of PVN pre-autonomic neurones A, the number of primary dendrites and dendritic branches was significantly larger in type B neurones as compared to the other groups (* P < 0.001 and # P < 0.02). B, the total dendritic length (TDL) was also significantly larger in type B neurones (* P < 0.02). No significant differences in mean dendritic length (MDL) and mean path length (MPL) were observed among neuronal types. C, frequency distribution of the number of branches into branch orders in PVN pre-autonomic neurones. In all neuronal types, most branches corresponded to the second-order type. D, plot of the number of dendritic intersections as a function of the distance from the soma in PVN pre-autonomic neurones (Sholl's analysis; Sholl, 1953). Note the more distal extension of dendrites in type B and C neurones. The inset shows a diagram depicting the use of the Sholl method. Concentric spheres of a constant interval of 20 µm, with the centre of the soma as the origin, were drawn for each tracer-filled neurone, and the number of dendritic intersections encountered within each sphere was counted. | ||
PVN pre-autonomic type B neurones. PVN pre-autonomic type B neurones were located exclusively in the PaPo subnucleus, and accounted for 25 % of the recorded neurones (Fig. 3B). They had a mean cross-sectional soma area of 188.7 ± 4.2 µm2, which was not significantly different from the other PVN pre-autonomic neurones (P > 0.5, one-way ANOVA). A morphometric analysis of reconstructed neurones revealed that type B pre-autonomic neurones (n = 15) had the most complex dendritic arborization, as compared to the other neuronal types. Type B neurones had 3.8 ± 0.1 primary dendrites, which gave rise to 10.7 ± 1.9 branches (both parameters were significantly larger than the other neuronal types, P < 0.001 and P < 0.02, respectively, one-way ANOVA, Fig. 4). The TDL was also significantly larger than in the other neuronal types (5329.3 ± 1500 µm, P < 0.02, one-way ANOVA, Fig. 4). The MDL and MPL were 778.6 ± 310.7 µm and 390.5 ± 104.4 µm, respectively (not different from the other neuronal types). Similar to type A neurones, in the majority of cases (12/15, 80 %) axons arose from a primary dendrite, at a mean distance from the soma of 38 ± 6 µm.
PVN pre-autonomic type C neurones. PVN pre-autonomic type C neurones accounted for 23 % of recorded neurones, with the great majority of them (83 %) located in the PaPo subnucleus (Fig. 3C). They had a mean cross-sectional soma area of 165.4 ± 40 µm2, with 3.1 ± 0.2 primary dendrites, which gave rise to 7.9 ± 0.9 branches (n = 17). Similar to the other neuronal types, dendrites were often varicose and tended to approach the walls of the 3V. In two cases, dendrites of type C neurones were observed to cross to the contralateral PVN (see Fig. 1D). The TDL, MDL and MPL of type C neurones were 2168.2 ± 750 µm, 692.9 ± 220 µm and 348.5 ± 86 µm, respectively. In contrast to type A and B neurones, in the majority of type C neurones (10/17, 60 %), axons arose from the soma (Fig. 3C 1 and C 2). However, the incidence of axon origin was not significantly different between cell types (P = 0.1, 2 test).
In order to obtain further information on the organization of dendritic trees, a frequency distribution analysis of dendrites into branch orders was performed (Fig. 4C). In all neuronal types, the number of branches varied as a function of the branch order (P < 0.001, two-way ANOVA), with most of the branches corresponding to second-order type (~35 % of the total number of branches). A larger number of branches was observed in type B neurones, when compared to the other groups (P < 0.02, Scheffé's post hoc test). However, no significant interactions in the analysis were observed (P = 0.3), indicating that the distribution pattern into branch orders was similar among groups.
The dendritic branching pattern of PVN pre-autonomic neurones was also analysed using Sholl's method (Sholl, 1953; see also Methods). Figure 4D shows a plot of the number of dendritic intersections encountered on concentric spheres of increasing radius (20 µm steps) from the centre of the soma. A significantly different pattern was observed between the different types of pre-autonomic neurone (P < 0.001, two-way ANOVA). Whereas most of the branches in type A neurones were concentrated within a radius of 400 µm from the soma, branches in types B and C neurones were encountered in more distal regions, up to a radius of 600 µm (type C) and 800 µm (type B) from the soma. Furthermore, the peak for encountered intersections in type A neurones was located more proximal to the soma than for type B and C neurones (type A neurones, 62.5 ± 10.3 µm; type B neurones, 120.0 ± 25.3 µm; and type C neurones, 144 ± 27.1 µm; P < 0.02, Scheffé's post hoc test).
Electrophysiological properties of PVN pre-autonomic neurones
General electrophysiological properties. Data on basic membrane properties and Na+ action potential waveforms of the three types of PVN pre-autonomic neurone are summarized in Table 1. While the resting membrane potential was similar among neurones, significant differences in apparent input resistance were observed (P < 0.001, one-way ANOVA), with higher values corresponding to type C neurones (see Table 1). Cell input resistance is a reflection of the total neuronal surface area and specific membrane resistivity. In order to determine whether cell input resistance in PVN pre-autonomic neurones is influenced by their cytoarchitectural organization, a correlation between apparent input resistance and several dendritic parameters was obtained. Interestingly, a statistically significant negative correlation between input resistance and TDL, and between input resistance and the total number of branches was observed only for type A neurones (Z values: -3.2 and -2.2 for TDL and number of branches, respectively, n = 33, P < 0.01).
The amplitude of Na+ action potentials differed among groups (P < 0.05, one-way ANOVA), with larger values observed for type A neurones (See Table 1). On the other hand, action potential width and threshold were not significantly different among groups.
Inwardly rectifying current-voltage relationships in response to hyperpolarizing current pulses were observed in the great majority of PVN pre-autonomic neurones studied (75 % (12/16) of type A, 80 % (8/10) of type B and 90 % (9/10) of type C neurones). Typical examples for the different neuronal types are shown in Fig. 5. In most cases, a time-dependent inward rectification, in which a sag developed during the response, was observed (Fig. 5B and C). However, in a few cases (three type A neurones and two type C neurones), a fast inward rectification, which persisted throughout the response, was observed (Fig. 5A). Even though a tendency for a smaller sag was observed in type C neurones, differences were not statistically significant (type A neurones, 8.8 ± 1.8 %; type B neurones, 8.9 ± 0.5 %; type C neurones, 5.2 ± 1.6 %, P > 0.5, one-way ANOVA). No significant correlations between inward rectification and other morphometric or electrophysiological parameters were observed in any type of PVN pre-autonomic neurone (results not shown).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 5. PVN pre-autonomic neurones displayed inwardly rectifying current-voltage relationships in response to hyperpolarizing pulses A-C, examples of fast (A), and time-dependent inward rectification (B and C) in PVN pre-autonomic neurones. Note the absence of a depolarizing sag during the hyperpolarizing pulses in A. The lower panels show the current-voltage relationships obtained at the peak ( | ||
Properties of LTSs in PVN pre-autonomic neurones. LTSs with variable shapes and amplitudes, ranging from small-amplitude 'humps' to long-lasting plateaus, were observed in 35/41 (85 %) type A, 20/20 (100 %) type B and 15/18 (85 %) type C neurones. The incidence of LTSs was not significantly different among neuronal types (P > 0.2, 2 test). Examples of LTSs recorded from the different neuronal types are shown in Fig. 3 and Fig. 6-8. In order to quantify and compare the LTS amplitude and waveform between neuronal types, LTSs were evoked at a holding potential of -80 mV, in the presence of TTX (0.5 µM; Fig. 6). Whereas LTS amplitude was not different among PVN pre-autonomic neuronal types (P > 0.5, one-way ANOVA), LTS threshold was significantly more hyperpolarized in type B neurones, as compared to the other neuronal types (P < 0.05, one-way ANOVA; Fig. 6B).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 6. General properties of LTSs in PVN pre-autonomic neurones A, LTSs with varying shapes (arrows), including fast spikes (A1), slow humps (A2) and long-lasting plateaus (A3), were observed in PVN pre-autonomic neurones. In order to isolate LTSs, recordings were done in the presence of TTX (0.5 µM). Examples in A1-A3 were obtained from type A, B and C PVN pre-autonomic neurones, respectively. B, the LTS threshold was significantly more hyperpolarized in type B neurones (*P < 0.05). On the other hand, the LTS amplitude was similar among PVN pre-autonomic neuronal types (P > 0.05; C). | ||
As observed in other neuronal types (Llinas & Yarom, 1981; Zhan et al. 1999), the latency for the LTS onset and evoked fast action potentials riding on the LTS decreased with increasing current injection steps. This is shown in Fig. 7A, where a steeper depolarizing slope leading to the LTS and action potential burst can be observed during the larger current step. The LTS observed in PVN pre-autonomic neurones resembles that first described in inferior olivary neurones (Llinas & Yarom, 1981), which results from the activation of a low-threshold Ca2+ conductance (T-type Ca2+ current, IT). The data presented here suggest that IT also underlies LTSs in PVN pre-autonomic neurones. Firstly, the LTS and burst amplitude varied as a function of the holding potential (Fig. 7B). When neurones were hyperpolarized enough to ensure removal of IT inactivation, depolarizing current steps evoked a fully activated LTS with a shorter burst of action potentials. However, when neurones were held at progressively more depolarized membrane potentials, the amplitude of the LTS and the induced burst decreased, and eventually disappeared, resulting in tonic firing activity. Secondly, LTSs in PVN pre-autonomic neurones were insensitive to TTX (0.5 µM; see Fig. 6), but were blocked by low concentrations of Ni2+ (50-100 µM), a relatively specific blocker of IT (Fox et al. 1987; Fig. 7C). Under these conditions, LTSs were evoked either by depolarizing pulses from a hyperpolarized membrane potential, or by repolarization following hyperpolarizing pulses from resting membrane potentials, an almost complete block by Ni2+ (50 µM) being observed in both cases. Similar effects were observed in four other neurones (2 type A, 1 type B and 1 type C).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 7. Voltage dependency and Ni2+ sensitivity of LTSs in PVN pre-autonomic neurones A, example of LTSs and bursts of action potentials in a type A PVN pre-autonomic neurone evoked with two incremental depolarizing current steps (A1 and A2). A portion of both traces superimposed and expanded is shown in the inset (the arrow corresponds to A2). Note the decreased latency for evoked action potentials during the larger current injection. B, voltage dependency of LTS amplitude. At a membrane potential of approximately -80 mV, a fully activated LTS is evoked, inducing a short burst of action potentials (B 1). As the holding potential is depolarized to approximately -60 mV, the LTS and burst amplitude are diminished (B 2), and eventually disappear (at a membrane potential of approximately -50 mV), resulting in a tonic firing activity (B 3). Traces were obtained from a type B neurone. Action potentials in A and B are truncated. C, example of the LTS sensitivity to low concentrations of Ni2+ (50 µM). LTSs were evoked either by a depolarizing step from hyperpolarized membrane potentials (C 1) or at the offset of hyperpolarizing pulses from depolarized membrane potentials (C 2). Note the complete blockade after 5 min exposure to 50 µM Ni2+ (arrows in C 1 and C 2). The membrane potentials shown indicate the initial holding potential. | ||
The present data suggest that the LTS plays a key role in controlling the excitability of PVN pre-autonomic neurones. In other neuronal types, T-type Ca2+ channels underlying the LTS have been shown to be concentrated in dendritic trees (Kavalali et al. 1997; Mouginot et al. 1997; Magee et al. 1998). The morphological and electrophysiological data obtained in the present study from individual PVN pre-autonomic neurones showing different dendritic configurations allowed us to test the hypothesis that the properties of the LTS are influenced by these different dendritic structures (Fig. 8). Interestingly, it was found that LTS amplitude in PVN pre-autonomic type B neurones was negatively correlated with TDL (Fig. 8; Z value: -2.2, P < 0.05). On the other hand, the correlation was not statistically significant in the other neuronal types (Z values: -1.1 and -1.3, P > 0.05, for type A and C neurones, respectively). No other significant correlations were found between the LTS and other electrophysiological or morphological properties.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 8. Correlation between LTS and dendritic structure in PVN pre-autonomic neurones A-C, examples of reconstructed type A, B and C PVN pre-autonomic neurones, respectively, with their respective evoked LTSs (arrows). Arrowheads point to the axons. D, plot of LTS amplitude vs. TDL. Note the better linear regression fit obtained for type B neurones (type A neurones, r2 = 0.25, P = 0.3; type B neurones, r2 = 0.65, P < 0.05; and type C neurones, r2 = 0.50, P = 0.2). | ||
Spontaneous and repetitive firing properties of PVN pre-autonomic neurones. The spontaneous firing properties of PVN pre-autonomic neurones were analysed in 30 neurones. Spontaneous activity was observed in 13/17 (76 %) type A neurones, 5/8 (63 %) type B neurones and 6/8 (75 %) type C neurones, in which a wide spectrum of firing activities was observed, including tonic regular, tonic irregular and burst firing (Fig. 9A-C). Even though bursting and tonic activity were predominant in type B and C neurones, respectively (Fig. 9D), the incidence of firing patterns was not statistically different among neuronal types (P > 0.5, 2 test). Importantly, 93 % (28/30) of the studied neurones expressed LTSs. One of the neurones that lacked LTSs was silent, whereas the other one was tonically active. Thus, these data suggest that the presence or absence of the LTS is not the main factor determining the expression of different firing patterns in PVN pre-autonomic neurones.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 9. Spontaneous firing activity in PVN pre-autonomic neurones Different spontaneous firing patterns were exhibited by PVN pre-autonomic neurones. Shown are typical examples of tonic regular (A), tonic irregular (B) and bursting spiking activity (C). Traces shown in A, B and C were obtained from a type A, B and C neurone, respectively, although these patterns were observed in all three neuronal types. D, pie graphs depicting the incidence of spontaneous activity and firing patterns in the different subtypes of PVN pre-autonomic neurones. No significant differences in the incidence of firing activity or patterns were observed between neuronal types (P > 0.5). | ||
The repetitive firing properties of PVN pre-autonomic neurones (n = 33) were studied using depolarizing current pulses while holding the neurones at depolarized membrane potentials (approximately -50 mV). The data are summarized in Table 2 and Fig. 10. A varying degree of spike frequency adaptation (SFA) was observed in PVN pre-autonomic neurones, with some neurones showing a prominent SFA, and others showing little or no SFA. An example of SFA recorded in each neuronal type is shown in Fig. 10A. The time constant of SFA was similar among PVN pre-autonomic neuronal types (P > 0.5, one-way ANOVA; Table 2). In magnocellular neuroendocrine neurones, as well as in other neuronal types (Bourque & Brown, 1987; Lorenzon & Foehring, 1992; Armstrong et al. 1994; Sah, 1996), SFA time course was shown to be dependent on the expression of a slow after-hyperpolarizing potential (AHP). To determine whether this was also the case in PVN pre-autonomic neurones, AHP properties and their correlation with SFA were studied in identified PVN pre-autonomic neurones. Since the amplitude of the AHP is linearly related to the number of action potentials evoked during a pulse (Andew & Dudek, 1984), AHP amplitudes were calculated per action potential. Examples of evoked AHPs in each neuronal type are shown in Fig. 10C. Whereas the peak amplitude of the AHP was not different among neuronal types (P > 0.5 one-way ANOVA), the AHP decay time constant was significantly different among neurones, with slower kinetics observed in type B pre-autonomic neurones (P < 0.05, one-way ANOVA; Table 2). Although not statistically significant, the slower decay kinetics observed in type B neurones resulted in a larger AHP area in these neurones (Table 2). To determine whether there is a relationship between AHP and SFA properties in PVN pre-autonomic neurones, a multiple correlation analysis was performed. A significant correlation between SFA decay time constant and AHP amplitude, area and decay time course was observed only in type A neurones (Z values: -1.95, -2.1 and -1.92, respectively, P < 0.05).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 10. Repetitive firing properties of PVN pre-autonomic neurones A, PVN pre-autonomic neurones showed varying degrees of spike frequency adaptation (SFA) in response to a 280 ms depolarizing pulse. B, plots of instantaneous firing rate vs. time obtained from traces in A. The time constant of SFA for the type A and C neurones are also shown. Note the little SFA in the type B neurone shown. In this case, the data could not be fitted by an exponential function. No differences in the time course of SFA were observed between neuronal types (see Results). C, examples of after-hyperpolarizations (AHPs; arrows) obtained from the same PVN pre-autonomic neurones as shown in A. The inset in C 3 shows averaged AHP traces (n = 5) obtained from the same neurones, which were scaled to the same peak amplitude to facilitate the comparison of their time course. D, examples of the progressive increase in action potential duration during repetitive firing obtained from the same PVN pre-autonomic neurones as shown in A. The first five action potentials of an evoked train were aligned at action potential onset to show clearly the progressive spike broadening. | ||
As shown previously in unidentified parvocellular PVN neurones (Bains & Ferguson, 1999), a progressive increment in action potential duration (spike broadening) was observed during repetitive firing in identified pre-autonomic neurones. Examples of spike broadening for each neuronal type are shown in Fig. 10D, where the first five action potentials evoked during a depolarizing pulse were aligned at threshold. Spike broadening was quantified as the ratio of the duration of the fifth action potential to that of the first. No differences in the degree of spike broadening were observed among neuronal types (Table 2).
A multiple correlation analysis of repetitive firing and morphometric neuronal properties revealed that the decay time course and area of AHPs in type B neurones were positively correlated with the number of branches (Z values: 3.2 and 4.2 for AHP decay time course and area, respectively, P < 0.001). No other significant correlations were observed (results not shown).
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In order to gain information on the cellular properties underlying the intrinsic excitability of PVN pre-autonomic neurones, which are known to play an important role in cardiovascular function (Coote, 1995; Blair et al. 1996), a detailed characterization of their electrophysiological and morphological properties was obtained in the present study. This was achieved by combining fluorescent retrograde labelling techniques with patch-clamp recordings in hypothalamic slices. When possible, recorded neurones were reconstructed in three-dimensions for morphometric analysis. This is the first study on identified PVN pre-autonomic neurones where electrophysiological and morphometric data were obtained from individual neurones, allowing for a direct correlation between the two parameters.
Methodological considerations
Retrogradely labelled PVN pre-autonomic neurones were identified and recorded using fluorescence and IR-DIC videomicroscopy. An intrinsic limitation of this technique is that only neurones relatively close to the surface, an area that is more prone to damage during slice preparation, can be visualized. Thus, structural preservation is not always optimal. Damaged (e.g. somatic swelling, missing or cut dendritic trees), or weakly labelled neurones were discarded from the analysis. Another technical limitation of this study was the restriction of the morphometric analysis to a single plane of slice sectioning. For instance, if dendritic trees from the different subnuclei were differentially oriented in the rostro-caudal plane, the degree of dendritic pruning could have been differentially affected by the slicing procedure. Such differential damage, however, was not evident, since the majority of the dendritic endings of reconstructed neurones were contained within the slice preparation (e.g. they were not cut at the slice surface), and the reconstructed dendritic trees were oriented mostly in parallel with the slice plane. However, the possibility that some dendritic trees were completely cut off, and not detected by this method, cannot be completely ruled out. This could occur, for example, if a dendrite was running at right angles to the surface of the slice closest to the recorded neurone, which might have been hidden by the presence of the soma. Future studies on reconstructed neurones recorded from slices cut at different planes will help to clarify this issue. On the other hand, axonal projections were always cut at the slice surface, limiting a detailed study of their topographical orientation, as well as the presence of ramifications at locations beyond the limits of the slice.
A further technical limitation for this study is related to the procedure used for labelling PVN pre-autonomic neurones. In this study, injection of the retrograde tracer DiI was limited to the dorsal vagal complex (DVC), centred at the level of the obex, an area that comprises both the NTS and the DMX. The lipophilic nature of DiI (Honig & Hume, 1989) raises the concern that both axonal terminals and fibres of passage might have been labelled by these injections, resulting in non-specific labelling in the PVN. However, based on the restricted area of the injections, as well as the trajectory and topography of descending axonal projections from the PVN, it is very unlikely that PVN neurones projecting to targets other than the DVC were labelled during this procedure. In this sense, previous studies have shown that PVN descending pathways at the level of the medulla run ventrally (Swanson, 1977; Luiten et al. 1985). Whereas fibres that innervate the DVC sweep dorsomedially from ventrolateral parts of the reticular formation, those innervating the RVLM and spinal cord continue to run ventrally (Swanson 1977; Luiten et al. 1985). Thus, even though axons arising from areas other than the PVN that project to, or pass through the injected area might be labelled by the DiI injections, the clear anatomical segregation of PVN axonal projections at this level of the medulla precludes the simultaneous labelling of different PVN descending pathways. A similar labelling pattern in the PVN was obtained when Fluorogold, a tracer that is not taken up by fibres of passage (Schmued & Fallon, 1986; Pieribone & Aston-Jones, 1988) was used instead of DiI (author's unpublished observations). This also argues against non-specific or heterogeneous labelling in the PVN due to the properties of DiI. Further evidence in this sense is the fact that essentially separate populations of neurones in the PVN project to sympathetic and parasympathetic preganglionic neurones (Portillo et al. 1996, 1998), and that virtually no neurones project both to the DVC and the pituitary or the median eminence (Swanson & Kuypers, 1980; Swanson et al. 1980). Thus, based on all of the above considerations, PVN pre-autonomic neurones recorded in this study are probably circumscribed only to those innervating the DVC.
Morphological heterogeneity among PVN pre-autonomic neurones
The present study demonstrates that although PVN pre-autonomic neurones display common morphological and electrophysiological properties that distinguish them from other PVN neuronal types, such as magnocellular neuroendocrine neurones, they constitute a heterogeneous neuronal population. Based on their morphological appearance, neurones were categorized into three types, named A, B and C. Neuronal heterogeneity was related in part to their subnuclear location within the PVN. Whereas type A and B neurones were restricted to the PaV and PaPo subnuclei, respectively, type C neurones were present in both areas, although more predominantly in the PaPo subnucleus. Besides their topographical location within the PVN, neurones also differed in their dendritic configuration, which could differentially influence their integrative properties. For example, the electrotonic properties, and consequently the propagation of distal synaptic inputs into the soma, are known to be dependent on the cytoarchitecture of dendritic trees (Rall, 1959, 1977). Moreover, since most of the synaptic contacts in the PVN neurones are axodendritic (Kiss et al. 1983), the distinct dendritic configurations may also determine a difference in the number and/or type of synaptic inputs impinging on these neurones. Future studies on the electrotonic properties and synaptic activity of identified PVN pre-autonomic neurones will further aid in understanding their integrative properties.
In general agreement with previous studies using different morphological and neuronal tracing techniques (Armstrong et al. 1980; van den Pol, 1982; Rho & Swanson, 1989), some interesting common features to the different types of PVN pre-autonomic neurone were observed in this study. For instance, dendritic trees of PVN pre-autonomic neurones were not confined to their respective subnuclei, but rather extended into the lateral posterior magnocellular, the periventricular or the medial parvocellular parts of the nucleus. Occasionally, dendrites crossed to the respective contralateral subnucleus. This finding is in agreement with previous work, where dendrites from other neuronal populations within the PVN were also found to extend beyond their subnucleus boundaries (Armstrong et al. 1980; Rho & Swanson, 1989). Thus, although the PVN is in general anatomically compartmentalized, caution should be taken when considering the selectivity of a particular synaptic input that preferentially innervates a particular region of the nucleus.
Electrophysiological heterogeneity in PVN pre-autonomic neurones
Neurones categorized as types A, B and C shared a similar electrophysiological profile, including the expression of LTSs and inwardly rectifying current-voltage relationships. These properties distinguish them from vasopressin magnocellular neuroendocrine neurones (see also Hoffman et al. 1991; Tasker & Dudek, 1991; Luther & Tasker, 2000). However, differences in their basic membrane properties were observed. For example, cell input resistance was significantly lower in type A neurones, and was negatively correlated with TDL and the number of branches. This was not the case for type B and C neurones. Thus, differences in input resistance among neuronal types cannot be attributed exclusively to differences in dendritic configuration, and may be related to differences in ionic conductances operating at resting membrane potentials.
One of the most prominent intrinsic properties observed in the great majority of PVN pre-autonomic neurones was a Ca2+-dependent LTS. The LTS could be evoked either by depolarizing steps from hyperpolarizing membrane potentials, or at the offset of hyperpolarizing pulses (see Fig. 3). Even though the LTS amplitude and waveform varied among neurones, LTSs in all cases were able to reach the threshold for activating Na+-dependent action potentials, evoking a brief burst of firing activity. That the expression of the LTS in PVN pre-autonomic neurones is mediated by the activation of IT, a transient, low-threshold Ca2+ current, is supported by the voltage-dependent properties and sensitivity to low, micromolar concentrations of Ni2+. This is in agreement with recent data by Luther & Tasker (2000), who demonstrated the expression of IT in non-identified parvocellular PVN neurones. Similarly, the involvement of IT underlying the LTS has been shown previously in other neuronal types (Llinas & Yarom, 1981; Fox et al. 1987; Zhan et al. 1999). It is also possible that the current underlying the inward rectification observed in PVN pre-autonomic neurones, most probably IH (Mayer & Westbrook, 1983), also contributes to shaping the LTS, by slowly de-activating upon membrane depolarization from hyperpolarized membrane potentials. However, the involvement of IH does not seem to play a critical role in mediating the LTS, due to the fact that LTSs were almost completely blocked by Ni2+. Moreover, no significant correlations between inward rectification and LTS properties were observed for any of the PVN pre-autonomic neuronal types (P > 0.05, Z test).
One of the main properties of IT is that it contributes to the expression of different firing patterns, depending on its degree of activation/inactivation (determined for example by the resting membrane potential of the neurone). Thus, if neurones expressing IT are hyperpolarized enough to remove IT inactivation, incoming excitatory inputs will fully activate IT and induce a burst of action potentials (bursting mode). On the other hand, if neurones are initially depolarized so IT is inactivated, the neurone will respond to the same depolarizing inputs with a tonic discharge of action potentials (tonic mode). Thus, factors controlling the resting membrane potential and/or the properties of IT will have a profound impact on the firing activity of PVN pre-autonomic neurones. In other neuronal types, T-type Ca2+ channels have been shown to be concentrated mainly in dendritic compartments (Kavalali et al. 1997; Mouginot et al. 1997; Magee et al. 1998). Interestingly, LTS amplitude in PVN pre-autonomic type B neurones in the present study was negatively correlated with TDL. These results suggest that the topographical distribution of ion channels in these neurones is different from other neuronal types. Alternatively, the negative correlation between LTS amplitude and dendritic length could be related to a differential space clamp between neurones with elongated and compact dendritic trees. If T-type Ca2+ channels were preferentially located in dendrites distal enough to be partially unclamped (due to space-clamp problems), they would be less de-inactivated upon membrane hyperpolarization in neurones with elongated trees (in which space-clamp problems would be accentuated), giving rise to LTSs of smaller amplitude.
Interestingly, the majority of recorded PVN pre-autonomic neurones in this study (~70 %) were spontaneously active. Previous in vivo recordings of spinally projecting PVN pre-autonomic neurones showed that these neurones were quiescent at rest (Lovick & Coote, 1988; Bains & Ferguson, 1995). One possibility for this difference is that in the intact animal, PVN pre-autonomic neurones are tonically inhibited by an extrinsic input that is lost in the slice preparation. In this sense, unilateral vagotomy in the intact animal increased the number and excitability of spontaneously active cells, suggesting that baroreceptor inputs exert a potent tonic inhibition on these neurones (Lovick & Coote, 1988). Alternatively, differences in the degree of spontaneous activity among PVN pre-autonomic neurones could be related to their innervation target, with spontaneous activity being more predominant in DVC-projecting PVN pre-autonomic neurones (as shown in this study), as compared to spinally projecting neurones (Lovick & Coote, 1988; Bains & Ferguson, 1995).
Spontaneously active neurones displayed either bursting or tonic firing patterns, the expression of which was not circumscribed to a specific neuronal type. As shown in Fig. 7, individual PVN pre-autonomic neurones could be induced to fire in either of these modes, depending on the initial holding potential, probably reflecting the expression and inactivation, respectively of IT. LTSs were observed in the great majority of spontaneously active neurones, suggesting that the expression of these patterns was not determined by the presence or absence of LTSs. Alternatively, the different spontaneous firing patterns observed in PVN pre-autonomic neurones could be dependent upon the resting membrane potential, and thus, the degree of activation/inactivation of IT. Future studies will aim to address whether individual neurones could switch from one pattern to another, as well as the precise ionic mechanisms underlying the expression of these different firing patterns.
When tonic firing was induced by depolarizing current steps from depolarized membrane potentials, a varying degree of SFA and spike broadening was observed. In magnocellular neuroendocrine neurones, as well as in other neuronal types (Bourque & Brown, 1987; Lorenzon & Foehring, 1992; Armstrong et al. 1994; Sah, 1996), SFA time course was shown to be dependent upon the expression of an AHP. Although a pronounced slow AHP was observed in PVN pre-autonomic neurones, a direct relationship between the slow AHP and SFA was not always observed. For example, although longer-lasting and larger AHPs were observed in type B neurones, the SFA time course was similar among neurones. Moreover, a significant correlation between SFA decay time constant and AHP properties was observed only in type A neurones. These results suggests that SFA in PVN pre-autonomic neurones might also be dependent upon other K+ conductances, such as IA (see for example Gean & Shinnick-Gallagher, 1989).
Functional considerations
The coordinated activity of distinct magnocellular and parvocellular neuronal populations within the PVN results in integrated neuroendocrine and autonomic responses that are fundamental for bodily homeostasis. Within these populations, pre-autonomic parvocellular neurones play an important role in the control of the cardiovascular system (Coote, 1995; Blair et al. 1996). Furthermore, there is a growing consensus for the involvement of PVN pre-autonomic neurones in cardiovascular disorders, including hypertension (Goto et al. 1981; Ciriello et al. 1984) and congestive heart failure (Patel & Zhang, 1996; Patel et al. 2000). In order to understand the cellular mechanisms involved in their contribution to these pathological processes, it is necessary first to characterize the basic mechanisms underlying their neuronal excitability under physiological conditions. An important finding from this study is that PVN pre-autonomic neurones constitute a heterogeneous neuronal population, which is only partially related to their topographical distribution within the PVN. One important factor that could be associated with the heterogeneity described herein is their innervation target. In this sense, PVN pre-autonomic neurones are known to send projections to either the autonomic-related neurones in brainstem areas (including the NTS, DMX and RVLM), or preganglionic sympathetic neurones located in the intermediolateral column of the spinal cord. In a small proportion of cases, simultaneous projections to the RVLM and the spinal cord have also been observed (Shafton et al. 1998; Pyner & Coote, 2000). These data suggest that different subsets of PVN pre-autonomic neurones could differentially affect parasympathetic or sympathetic neuronal outflows through projections to distinct target areas. Thus, the different morphological and electrophysiological profiles observed in PVN pre-autonomic neurones could be related to their innervation target. Based on the anatomical distribution of axons projecting to the different areas innervated by PVN pre-autonomic neurones (see 'Methodological considerations' above), it is unlikely that the neuronal heterogeneity described in this study results from the non-specific labelling of PVN pre-autonomic neurones that project to targets other than the DVC. However, since the DVC comprises both the NTS and DMX, heterogeneity could be in part related to neurones innervating these different nuclei. Experiments using more anatomically precise tracer injections will be needed to test this hypothesis.
Alternatively, heterogeneity could be related to different neurochemical identities among PVN pre-autonomic neurones. In this sense, a variety of neuroactive substances, including oxytocin, vasopressin, somatostatin, enkephalin and dynorphin, have been shown to be expressed by PVN pre-autonomic neurones (Sawchenko & Swanson, 1982; Hallbeck et al. 2001; Puder & Papka, 2001). Moreover, these neuropeptides are known to differentially affect the cardiovascular system (Sun & Guyenet, 1989; Ishizuka et al. 1993; Okada et al. 1994; Michelini & Morris, 1999; Braga et al. 2000). Thus, whether heterogeneity in PVN pre-autonomic neurones is related to their chemical phenotype remains to be determined. Future experiments will aim to shed light on the nature of cellular heterogeneity of PVN pre-autonomic neurones, and to determine whether this heterogeneity is related to a differential role in the control of the autonomic system.
Summary
Based on the combination of an in vivo retrograde tracing technique with in vitro patch-clamp recordings, a detailed electrophysiological and morphological characterization of identified PVN pre-autonomic neurones of adult rats was obtained. The results of the present study show that PVN pre-autonomic neurones constitute a heterogeneous neuronal population. However, some consistent properties such as the expression of LTS, strong inward rectification and highly branched dendrites, distinguish them from other neurones within the PVN, such as magnocellular neuroendocrine neurones. The role of PVN pre-autonomic neuronal heterogeneity in the control of a balanced autonomic outflow remains to be established.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| ANDREW, R. D. & DUDEK, F. E. (1984). Intrinsic inhibition in magnocellular neuroendocrine cells of rat hypothalamus. Journal of Physiology 353, 171-185 | [Abstract] |
| ARMSTRONG, W. E., SMITH, B. N. & TIAN, M. (1994). Electrophysiological characteristics of immunochemically identified rat oxytocin and vasopressin neurones in vitro. Journal of Physiology 475, 115-128 | [Abstract] |
| ARMSTRONG, W. E., WARACH, S., HATTON, G. & MCNEILL, T. (1980). Subnuclei in the rat hypothalamic paraventricular nucleus: a cytoarchitectural horseradish peroxidase and immunocytochemical analysis. Neuroscience 5, 1931-1958 | [Medline] |
| BADOER, E., OLDFIELD, B. J. & MCKINLEY, M. J. (1993). Haemorrhage-induced production of Fos in neurons of the lamina terminalis: role of endogenous angiotensin II. Neuroscience Letters 159, 151-154 | [Medline] |
| BAINS, J. S. & FERGUSON, A. V. (1995). Paraventricular nucleus neurons projecting to the spinal cord receive excitatory input from the subfornical organ. American Journal of Physiology 268, R625-633 | [Medline] |
| BAINS, J. S. & FERGUSON, A. V. (1999). Activation of N-methyl-D-aspartate receptors evokes calcium spikes in the dendrites of rat hypothalamic paraventricular nucleus neurons. Neuroscience 90, 885-891 | [Medline] |
| BARRETT-JOLLEY, R., PYNER, S. & COOTE, J. H. (2000). Measurement of voltage-gated potassium currents in identified spinally-projecting sympathetic neurones of the paraventricular nucleus. Journal of Neuroscience Methods 102, 25-33 | [Medline] |
| BLAIR, M. L., PIEKUT, D., WANT, A. & OLSCHOWKA, J. A. (1996). Role of the hypothalamic paraventricular nucleus in cardiovascular regulation. Clinical and Experimental Pharmacology and Physiology 23, 161-165 | [Medline] |
| BOURQUE, C. W. & BROWN, D. A. (1987). Apamin and d-tubocurarine block the afterhyperpolarization of rat supraoptic neurosecretory neurons. Neuroscience Letters 82, 185-190 | [Medline] |
| BRAGA, D. C., MORI, E., HIGA, K. T., MORRIS, M. & MICHELINI, L. C. (2000). Central oxytocin modulates exercise-induced tachycardia. American Journal of Physiology 278, R1474-1482 | |
| CIRIELLO, J., KLINE, R. L., ZHANG, T. X. & CAVERSON, M. M. (1984). Lesions of the paraventricular nucleus alter the development of spontaneous hypertension in the rat. Brain Research 310, 355-359 | [Medline] |
| COOTE, J. H. (1995). Cardiovascular function of the paraventricular nucleus of the hypothalamus. Biological Signals 4, 142-149 | [Medline] |
| COOTE, J. H., YANG, Z., PYNER, S. & DEERING, J. (1998). Control of sympathetic outflows by the hypothalamic paraventricular nucleus. Clinical and Experimental Pharmacology and Physiology 25, 461-463 | [Medline] |
| FOX, A., NOWYCKY, M. & TSIEN, R. (1987). Kinetic and pharmacological properties distinguishing three types of calcium currents in chick sensory neurones. Journal of Physiology 394, 149-172 | [Abstract] |
| GEAN, P. W & SHINNICK-GALLAGHER, P. (1989). The transient potassium current, the A-current, is involved in spike frequency adaptation in rat amygdala neurons. Brain Research 480, 160-169 | [Medline] |
| GOTO, A., IKEDA, T., TOBIAN, L., IWAI, J. & JOHNSON, M. A. (1981). Brain lesions in the paraventricular nuclei and catecholaminergic neurons minimize salt hypertension in Dahl salt-sensitive rats. Clinical Science 61, 53S-55S | [Medline] |
| HALLBECK, M., LARHMMAR, D. & BLOMQVIST, A. (2001). Neuropeptide expression in rat paraventricular hypothalamic neurons that project to the spinal cord. Journal of Comparative Neurology 433, 222-238 | [Medline] |
| HOFFMAN, N., TASKER, J. & DUDEK, E. (1991). Immunohistochemical differentiation of electrophysiologically defined neuronal populations in the region of the rat hypothalamic paraventricular nucleus. Journal of Comparative Neurology 307, 405-416 | [Medline] |
| HONIG, M. G. & HUME, R. I. (1989). DiI and diO: versatile fluorescent dyes for neuronal labelling and pathway tracing. Trends in Neurosciences 12, 333-335 | [Medline] |
| HORIKAWA, K. & ARMSTRONG, W. E. (1988). A versatile means of intracellular labeling: injection of biocytin and its detection with avidin conjugates. Journal of Neuroscience Methods 25, 1-11 | [Medline] |
| ISHIZUKA, T., WEI, X. & KUBO, T. (1993). Cardiovascular effects of microinjections of thyrotropin-releasing hormone, oxytocin and other neuropeptides into the rostral ventrolateral medulla of the rat. Archives Internationales de Pharmacodynamie et de Therapie 322, 35-44 | [Medline] |
| KANNAN, H., NIIJIMA, A. & YAMASHITA, H. (1988). Effects of stimulation of the hypothalamic paraventricular nucleus on blood pressure and renal sympathetic nerve activity. Brain Research Bulletin 20, 779-783 | [Medline] |
| KAVALALI, E. T., ZHUO, M., BITO, H. & TSIEN, R. W. (1997). Dendritic Ca2+ channels characterized by recordings from isolated hippocampal dendritic segments. Neuron 18, 651-663 | [Medline] |
| KISS, J. Z., PALKOVITS, M., ZABORSZKY, L., TRIBOLLET, E., SZABO, D. & MAKARA, G. B. (1983). Quantitative histological studies on the hypothalamic paraventricular nucleus in rats: I. Number of cells and synaptic boutons. Brain Research 262, 217-224 | [Medline] |
| KITA, H. & ARMSTRONG, W. (1991). A biotin-containing compound N-(2-aminoethyl)biotinamide for intracellular labeling and neuronal tracing studies: comparison with biocytin. Journal of Neuroscience Methods 37, 141-150 | [Medline] |
| LLINAS, R. & YAROM, Y. (1981). Properties and distribution of ionic conductances generating electroresponsiveness of mammalian inferior olivary neurons in vitro. Journal of Physiology 315, 569-584 | [Abstract] |
| LORENZON, N. M. & FOEHRING, R. C. (1992). Relationship between repetitive firing and afterhyperpolarizations in human neocortical neurons. Journal of Neurophysiology 67, 350-363 | [Medline] |
| LOVICK, T. A. & COOTE, J. H. (1988). Electrophysiological properties of paraventriculo-spinal neurones in the rat. Brain Research 454, 123-130 | [Medline] |
| LUITEN, P. G., TER HORST, G. J., KARST, H. & STEFFENS, A. B. (1985). The course of paraventricular hypothalamic efferents to autonomic structures in medulla and spinal cord. Brain Research 329, 374-378 | [Medline] |
| LUTHER, J. A. & TASKER, J. G. (2000). Voltage-gated currents distinguish parvocellular from magnocellular neurones in the rat hypothalamic paraventricular nucleus. Journal of Physiology 523, 193-209 | [Abstract/Full Text] |
| MAGEE, J., HOFFMAN, D., COLBERT, C. & JOHNSTON, D. (1998). Electrical and calcium signaling in dendrites of hippocampal pyramidal neurons. Annual Review of Physiology 60, 327-346 | [Abstract/Full Text] |
| MALPAS, S. C. & COOTE, J. H. (1994). Role of vasopressin in sympathetic response to paraventricular nucleus stimulation in anesthetized rats. American Journal of Physiology 266, R228-236 | [Medline] |
| MAYER, M. L. & WESTBROOK, G. L. (1983). A voltage-clamp analysis of inward (anomalous) rectification in mouse spinal sensory ganglion neurones. Journal of Physiology 340, 19-45 | [Abstract] |
| MICHELINI, L. C. & MORRIS, M. (1999). Endogenous vasopressin modulates the cardiovascular responses to exercise. Annals of the New York Academy of Sciences 897, 198-211 | [Medline] |
| MOUGINOT, D., BOSSU, J. L. & GAHWILER, B. H. (1997). Low-threshold Ca2+ currents in dendritic recordings from Purkinje cells in rat cerebellar slice cultures. Journal of Neuroscience 17, 160-170 | [Abstract/Full Text] |
| OKADA, J., TAKAYAMA, K., XIONG, Y. & MIURA, M. (1994). Influence of humoral control peptides on medullary vasomotor control neurons: microstimulation and double-labeling studies using SHR and WKY rats. Journal of the Autonomic Nervous System 49, 171-182 | [Medline] |
| PATEL, K. P. & ZHANG, K. (1996). Neurohumoral activation in heart failure: role of paraventricular nucleus. Clinical and Experimental Pharmacology and Physiology 23, 722-726 | [Medline] |
| PATEL, K. P., ZHANG, K., KENNEY, M. J., WEISS, M. & MAYHAN, W. G. (2000). Neuronal expression of Fos protein in the hypothalamus of rats with heart failure. Brain Research 865, 27-34 | [Medline] |
| PIERIBONE, V. A. & ASTON-JONES, G. (1988). The iontophoretic application of Fluoro-Gold for the study of afferents to deep brain nuclei. Brain Research 475, 259-271 | [Medline] |
| PORTER, J. P. & BRODY, M. J. (1985). Neural projections from paraventricular nucleus that subserve vasomotor functions. American Journal of Physiology 248, R271-281 | [Medline] |
| PORTER, J. P. & BRODY, M. J. (1986). A comparison of the hemodynamic effects produced by electrical stimulation of subnuclei of the paraventricular nucleus. Brain Research 375, 20-29 | [Medline] |
| PORTILLO, F., CARRASCO, M. & VALLO, J. J. (1996). Hypothalamic neuron projection to autonomic preganglionic levels related with glucose metabolism: a fluorescent labelling study in the rat. Neuroscience Letters 210, 197-200 | [Medline] |
| PORTILLO, F., CARRASCO, M. & VALLO, J. J. (1998). Separate populations of neurons within the paraventricular hypothalamic nucleus of the rat project to vagal and thoracic autonomic preganglionic levels and express c-Fos protein induced by lithium chloride. Journal of Chemical Neuroanatomy 14, 95-102 | [Medline] |
| PUDER, V. A. & PAPKA, R. E. (2001). Hypothalamic paraventricular axons projecting to the female rat lumbosacral spinal cord contain oxytocin immunoreactivity. Journal of Neuroscience Research 64, 53-60 | [Medline] |
| PYNER, S. & COOTE, J. H. (2000). Identification of branching paraventricular neurons of the hypothalamus that project to the rostroventrolateral medulla and spinal cord. Neuroscience 100, 549-556 | [Medline] |
| RALL, W. (1959). Branching dendritic trees and motoneuron membrane resistivity. Experimental Neurology 1, 491-527 | |
| RALL, W. (1977). Core conductor theory and cable properties of neurons. In Handbook of Physiology, The Nervous System, vol. I, Cellular Biology of Neurons, ed. KANDEL, E. R., BROOKHART, J. M. & MOUNTCASTLE, V. B., pp. 39-97. American Physiological Society, Bethesda, MD, USA | |
| RANSON, R. N., MOTAWEI, K., PYNER, S. & COOTE, J. H. (1998). The paraventricular nucleus of the hypothalamus sends efferents to the spinal cord of the rat that closely appose sympathetic preganglionic neurones projecting to the stellate ganglion. Experimental Brain Research 120, 164-172 | [Medline] |
| RENAUD, L. P. & BOURQUE, C. W. (1991). Neurophysiology and neuropharmacology of hypothalamic magnocellular neurons secreting vasopressin and oxytocin. Progress in Neurobiology 36, 131-169 | [Medline] |
| RHO, J. H. & SWANSON, L. W. (1989). A morphometric analysis of functionally defined subpopulations of neurons in the paraventricular nucleus of the rat with observations on the effects of colchicine. Journal of Neuroscience 9, 1375-1388 | [Abstract] |
| SAH, P. (1996). Ca2+-activated K+ currents in neurones: types, physiological roles and modulation. Trends in Neurosciences 19, 150-154 | [Medline] |
| SAPER, C. B., SWANSON, L. W. & COWAN, W. M. (1976). The efferent connections of the ventromedial nucleus of the hypothalamus of the rat. Journal of Comparative Neurology 169, 409-442 | [Medline] |
| SAWCHENKO, P. E. & SWANSON, L. W. (1982). Immunohistochemical identification of neurons in the paraventricular nucleus of the hypothalamus that project to the medulla or to the spinal cord in the rat. Journal of Comparative Neurology 205, 260-272 | [Medline] |
| SCHMUED, L. C. & FALLON, J. H. (1986). Fluoro-Gold: a new fluorescent retrograde axonal tracer with numerous unique properties. Brain Research 377, 147-154 | [Medline] |
| SHAFTON, A. D., RYAN, A. & BADOER, E. (1998). Neurons in the hypothalamic paraventricular nucleus send collaterals to the spinal cord and to the rostral ventrolateral medulla in the rat. Brain Research 801, 239-243 | [Medline] |
| SHOLL, D. (1953). Dendritic organization in the neurons of the visual and motor cortices of the cat. Journal of Anatomy 87, 387-406 | |
| STERN, J. E. & ARMSTRONG, W. E. (1997). Sustained outward rectification of oxytocinergic neurones in the rat supraoptic nucleus: ionic dependence and pharmacology. Journal of Physiology 500, 497-508 | [Abstract] |
| STERN, J. E., GALARRETA, M., FOEHRING, R. C., HESTRIN, S. & ARMSTRONG, W. E. (1999). Differences in the properties of ionotropic glutamate synaptic currents in oxytocin and vasopressin neuroendocrine neurons. Journal of Neuroscience 19, 3367-3375 | [Abstract/Full Text] |
| SUN, M. K. & GUYENET, P. G. (1989). Effects of vasopressin and other neuropeptides on rostral medullary sympathoexcitatory neurons 'in vitro'. Brain Research 492, 261-270 | [Medline] |
| SWANSON, L. & KUYPERS, H. (1980). The paraventricular nucleus of the hypothalamus: Cytoarchitectonic subdivisions and organization of projections to the pituitary, dorsal vagal complex, and spinal cord as demonstrated by retrograde fluorescence double-labeling methods. Journal of Comparative Neurology 194, 555-570 | [Medline] |
| SWANSON, L. W. (1977). Immunohistochemical evidence for a neurophysin-containing autonomic pathway arising in the |
) and at steady state (
) during the hyperpolarizing pulses.