Download Structural Changes in AMPA-Receptive Neurons in the Nucleus of

Structural Changes in AMPA-Receptive Neurons in the
Nucleus of the Solitary Tract of Spontaneously
Hypertensive Rats
Sue A. Aicher, Sarita Sharma, Jennifer L. Mitchell
Downloaded from http://hyper.ahajournals.org/ by guest on May 7, 2017
Abstract—The baroreceptor reflex is critical for homeostatic regulation of blood pressure and is initiated centrally by
glutamate release from baroreceptive afferents onto neurons in the nucleus of the solitary tract that activates AMPA-type
glutamate receptors. The GluR1 subunit of the AMPA receptor is located at postsynaptic sites within the nucleus of the
solitary tract, particularly in dendritic spines, which are important sites for synaptic plasticity. We tested whether the
distribution of GluR1 changes after sustained hypertension, which alters baroreceptor afferent activity. We examined
the distribution of GluR1 in the nucleus of the solitary tract of both hypertensive (spontaneously hypertensive) and
normotensive (Wistar-Kyoto) rats at the light microscopic and electron microscopic levels. There were more
GluR1-containing dendritic spines in the nucleus of the solitary tract of hypertensive rats compared with normotensive
rats, which was attributable to an increase in the proportion of dendritic spines containing GluR1 as well as an increase
in the total number of dendritic spines. The differences were only seen after the development of hypertension and were
not seen in rostral regions of the nucleus of the solitary tract. In the spontaneously hypertensive rat, many synapses on
GluR1-containing dendrites had the morphological features of synapses undergoing dynamic changes, including the
presence of perforated synapses. These results suggest that changes in afferent activity to the nucleus of the solitary tract
during sustained hypertension alter both the dendritic structure and AMPA receptor content of some neurons. These
structural changes may be a substrate for central resetting of the baroreceptor reflex. (Hypertension. 2003;41:12461252.)
Key Words: rats, spontaneously hypertensive 䡲 rats, inbred WKY 䡲 central nervous system 䡲 blood pressure
䡲 baroreflex 䡲 glutamate receptor
B
have found similar results,16 and we also noted that the GluR1
subunit was most heavily located in dendritic spines, suggesting a potential role for AMPA receptors containing this
subunit in synaptic plasticity within the NTS. We set out to
determine if structural changes occur in neurons that contain
GluR1 and may be an underlying substrate for the neural
changes that occur during hypertension, including central
resetting of the baroreceptor reflex.
A variety of animal models of hypertension have been
developed to facilitate the study of the underlying mechanisms of this disorder. This study used the spontaneously
hypertensive rat (SHR), a genetic/developmental model that
is probably the most widely used model of hypertension.17
We asked whether a change in the distribution of the AMPA
receptor subunit, GluR1, is seen after sustained hypertension.
We also tested whether the change was specifically related to
hypertension by examining GluR1 in the NTS of young,
normotensive SHRs and by examining regions of the NTS
that do not receive cardiovascular afferents. We found
changes in both the morphology and AMPA receptor content
aroreceptors are stretch receptors found in large blood
vessels around the heart. Activation of these receptors is
relayed to the brain by afferents to the nucleus of the solitary
tract (NTS),1 which release glutamate.2,3 A central reflex arc
through the brain stem and spinal cord serves to return blood
pressure to homeostatic values. Although the underlying of
hypertension is likely to be heterogeneous, at least some
forms of hypertension may be caused by, or lead to, dysfunction within the central baroreceptor reflex arc.4,5 Functional
changes in properties of NTS neurons as well as altered blood
pressure responses to glutamate microinjections into this
region have been reported during hypertension.6 Although it
has been postulated that changes in central autonomic neurons may occur during hypertension,7,8 these changes have
been difficult to confirm.4,9
␣ -Amino-3-hydroxy-5-methyl-4-isoxazole-propionate
(AMPA)-type glutamate receptors10 are involved in mediating the effects of baroreceptor stimulation within the
NTS.11–13 Recent studies have shown that 3 of the 4 known
AMPA receptor subunits are present within the NTS.14,15 We
Received December 27, 2002; first decision January 24, 2003; revision accepted March 18, 2003.
From the Neurological Sciences Institute, Oregon Health and Science University, Beaverton, Ore.
Correspondence to Sue A. Aicher, Neurological Sciences Institute, Oregon Health and Science University, 505 NW 185th Ave, Beaverton, OR 97006.
E-mail [email protected]
© 2003 American Heart Association, Inc.
Hypertension is available at http://www.hypertensionaha.org
DOI: 10.1161/01.HYP.0000069007.98987.E0
1246
Aicher et al
GluR1 in NTS of SHR
1247
Figure 1. Light microscopic distribution of AMPA
receptor subunit GluR1 in NTS. A, GluR1 was
densely localized within area postrema (AP) and
was seen primarily in punctate profiles. Boxed areas
indicate regions of NTS sampled in quantitative
studies (see Methods). B, GluR1 labeling was dense
in punctate profiles (arrowhead) throughout medial
NTS. GluR1 was also in other profiles, but only
puncta were quantified at light microscopic level
(see Methods). ts indicates solitary tract. Scale bars:
A⫽100 ␮m, B⫽20 ␮m.
of NTS neurons after sustained hypertension. Because the
NTS receives primary baroreceptive afferents that release
glutamate, our results support the notion that hypertension
can alter the structure, and possibly function, of NTS neurons.
Systolic blood pressure was assessed in restrained rats by use of an
indirect tail-cuff method (Kent Scientific Corp). The cuff pressure
and pulsatile blood pressure signal were detected by a piezoelectric
pulse sensor and were monitored on a computer interfaced to an A/D
board with Spike 2 software (Cambridge Electronic Design). After
the rat was acclimated and a strong pulsatile blood pressure signal
was recorded, the tail occlusion cuff was inflated until the pulsatile
blood pressure signal was lost and then was released for a gradual
deflation (2 to 4 seconds). Systolic pressure was the pressure of the
occlusion cuff at which the pulsatile blood pressure signal was again
detected. The average systolic pressure for 3 to 6 trials was the
systolic pressure for that day. The systolic blood pressure on the last
day before perfusion was used for statistical analyses. Systolic
pressures between groups were compared by means of a t test,
(␣⫽0.05).
mounted on slides for light microscopic analysis; the other half were
prepared for electron microscopic analysis. An observer blinded to
the experimental conditions determined the area of the NTS to be
examined in a selected tissue section from each animal. To ensure
that the volume of NTS did not change, the entire subpostremal
region was outlined and the total pixel area of the outlined region
was determined with the use of NIH Image software. The outline of
NTS excluded the area postrema, the central canal, the dorsal motor
nucleus of the vagus, and the dorsal column nuclei. An observer
blinded to the experimental condition measured the area of NTS for
each case. To determine the number of GluR1-labeled puncta within
the NTS, puncta were counted in 2 rectangular regions that were
selected from a single vibratome section. Two regions within the
medial NTS (0.05 mm⫻0.1 mm⫽0.005 mm2), at the level of the area
postrema, were examined (Figure 1). One region was located just
medial to the solitary tract, whereas the other was located ventrolateral to the area postrema (boxes in Figure 1A). Because the results of
GluR1 counts were similar for these 2 subregions, counts were
pooled for the final statistical analyses and are shown in the analyses
as the combined area (0.01 mm2) of both regions.
Puncta were counted from a digital image captured with a Spot 2
camera and Twain software (with background subtraction), and
counts were verified by 2 independent blinded observers. Punctate
profiles were considered small, round regions of intense labeling that
were not continuous with other labeled profiles (Figure 1B). Paired
t tests were used to compare the number of GluR1-labeled puncta for
similar groups, such as 16-week-old SHR versus 16-week-old WKY
or 5-week-old SHR versus 5-week-old WKY (␣⫽0.05). Because of
potential variations in immunocytochemical conditions, comparisons
between different runs may lead to invalid comparisons at the light
microscopic level. In the current study, all comparisons were made
between cases with identical tissue processing conditions (eg, 2
strains at 16 weeks of age or 2 strains at 5 weeks of age).
Perfusion and Immunocytochemistry
Electron Microscopy
Rats were overdosed with sodium pentobarbital (150 mg/kg) and
perfused transcardially with the following sequence of solutions: (1)
10 mL heparinized saline; (2) 50 mL 3.8% acrolein in 2% paraformaldehyde; and (3) 200 mL 2% paraformaldehyde (in 0.1 mol/L
phosphate buffer). The brain stem was sectioned (40 ␮m) on a
vibrating microtome and collected into 0.1 mol/L phosphate buffer.
Sections through the NTS were processed for immunoperoxidase
localization18,19 of GluR1. A rabbit polyclonal antibody directed
against the R1 (GluR1) subunit of the AMPA receptor20 was
obtained from Chemicon and was used at a 1:25 dilution.16 Tissue
sections were incubated in the primary antibody for 40 hours at 4°C,
and the bound antibody was detected with a biotinylated goat
anti-rabbit IgG (1:400; Vector Laboratories).
Tissue sections were osmicated and embedded in plastic after
immunocytochemical procedures. Ultrathin sections (75 nm) through
the surface of each section were collected onto copper grids,
counterstained, and examined with the use of a Phillips CM10 or a
JEOL1200EX electron microscope. Only cases with successful
labeling and optimal morphological preservation were included in
the electron microscopic analysis. For the ultrastructural analysis of
the GluR1 labeling in SHR and WKY, a 3025-␮m2 area of the medial
NTS (3 SHR, 3 WKY at each age) was examined (Figure 1). For
each case, a region was selected that contained labeled tissue over
one entire grid square (55 ␮m2) that was at a consistent tissue depth
(eg, no plastic regions were seen). For each case, all GluR1-labeled
profiles and unlabeled profiles of similar type (eg, dendrites or
spines) were counted and measured. Profiles were examined and
categorized by an observer who was blinded to the experimental
condition. The cross-sectional diameter of dendrites determined their
classification as either large (⬎1.5 ␮m), small (0.5 to 1.5 ␮m), or
spine (⬍0.5 ␮m and receiving an asymmetric synapse21). A ␹2
analysis was used to determine if there were different proportions for
GluR1-labeled versus unlabeled profiles in SHR versus WKY in the
fixed area of tissue. Separate ␹2 analyses were conducted for each
profile type (eg, spines, small dendrites, or large dendrites) and at
each age (5 weeks versus 16 weeks). A level of ␣⫽0.05 was used for
all statistical comparisons.
Methods
Subjects
Downloaded from http://hyper.ahajournals.org/ by guest on May 7, 2017
All methods were approved by the IACUC at Weill Medical College
of Cornell University, where this work was initiated. Male rats used
in these experiments were purchased from Taconic Laboratories
(Germantown, NY). Age-matched pairs of SHR and Wistar-Kyoto
rats (WKY) were received at either 4 weeks (n⫽4 pairs) or at 15
weeks (n⫽4 pairs) of age. Rats were tested and perfused 1 week
later, at 5 and 16 weeks of age, respectively.
Assessment of Blood Pressure
Light Microscopic Analyses
For GluR1 immunoperoxidase labeling in the NTS, pairs of animals
(1 SHR, 1 WKY; same age) were processed through identical
immunocytochemical conditions to ensure comparable labeling conditions. A puncture through the brain stem was performed on one
brain in each pair, and tissue sections from the 2 animals were
combined in the same vials during all antibody incubations. Tissue
sections were separated again after the completion of the immunocytochemical procedures by means of the tissue puncture to identify
sections from one animal in the pair. Half of the tissue sections were
1248
Hypertension
June 2003
Downloaded from http://hyper.ahajournals.org/ by guest on May 7, 2017
Figure 2. Both systolic blood pressure and GluR1 puncta
counts in NTS are increased in 16-week-old SHR compared
with WKY control rats but are not different at 5 weeks of age.
Systolic blood pressure is similar between the 2 strains at 5
weeks of age (A) but is significantly increased at 16 weeks of
age in SHR (B). Number of GluR1 puncta in a fixed region of
medial NTS is similar between the 2 strains at 5 weeks of age
(C) but is greater in SHR at 16 weeks of age compared with
age-matched WKY (D).
Results
In a previous study,16 we examined the ultrastructural distribution of AMPA receptor subunits that are present in the NTS
and are likely candidates for involvement in mediating the
baroreflex. The distribution of GluR1 was most intriguing. At
the light microscopic level, GluR1 was found primarily in
small punctate structures and in cell bodies (Figures 1A and
1B). At the electron microscopic level, GluR1 was found
primarily at postsynaptic sites, particularly in small dendrites
and dendritic spines (see below). Based on the comparisons
between the light microscopic and electron microscopic
distributions of GluR1, most of the small punctate structures
seen at the light microscopic level probably correspond to
small dendrites and dendritic spines. For quantitative analyses, these punctate structures were compared between groups
at the light microscopic level, whereas both dendrites and
spines were examined at the electron microscopic level.
We first measured systolic arterial pressure in each animal
to verify that our SHR were hypertensive at 16 weeks of age
but not at 5 weeks of age (Figures 2A and 2B). We then found
that there were more GluR1-labeled punctate profiles in the
NTS of the hypertensive rats (Figures 2C and 2D) compared
with normotensive control animals at the light microscopic
level. This increased density of GluR1-labeled puncta per unit
area was not due to shrinkage of the nucleus, since there was
no difference in the size of NTS between these groups of
animals (P⫽0.40).
We tested whether these differences in GluR1 puncta
density are due to inherent genetic differences between SHR
and WKY or may actually be related to blood pressure. If the
differences in GluR1 labeling are not related to blood
pressure, then in younger animals with similar blood pres-
sure, we should still see differences in GluR1 labeling
between the 2 strains. We measured blood pressure in
5-week-old SHR and WKY and then examined their GluR1
puncta density in the NTS. The 5-week-old SHR were not
hypertensive (Figure 2A), and they did not have a greater
number of GluR1-labeled puncta compared with age-matched
WKY (Figure 2C). These results suggest that the differences
in GluR1 density are not due to inherent strain differences but
are related to the onset of hypertension. This hypothesis is
supported by linear regression analyses of blood pressure and
GluR1 puncta counts in NTS. At 5 weeks of age, there was no
relationship between systolic blood pressure and the number
of GluR1 puncta in the NTS (R⫽0.21), but at 16 weeks of
age, there was a strong correlation between blood pressure
and GluR1 puncta count (R⫽0.83). It is apparent from this
analysis that across a range in systolic pressures between
SHR and WKY at 16 weeks of age, rats with the highest
systolic pressures also had the highest GluR1 puncta counts.
These results support the hypothesis that GluR1 labeling in
the NTS is related to the level of arterial blood pressure.
Comparison of the values for puncta counts in the NTS
across ages and rat strains (Figures 2C and 2D) indicates that
there are actually more GluR1 puncta in the NTS of both
strains at 5 weeks of age (Figure 2C) and there is a decrease
in the number of puncta at 16 weeks in the WKY only (Figure
2D). However, this suggestion is not born out at the electron
microscopic level (see below), where the number of spines
and small dendrites are similar between WKY at both ages.
These observations indicate that the detection of GluR1labeled puncta is somewhat more variable at the light microscopic level and may be due to variations in the immunocytochemical conditions between experiments. This is why only
comparisons between cases run under identical immunocytochemical conditions were compared.
If the changes in GluR1-labeled puncta are related to
changes in cardiopulmonary or baroreceptor afferent activity,
we would expect the increase in GluR1 density to be confined
to regions of the NTS that receive baroreceptor afferents. As
a control for the specificity of the changes in GluR1 density,
we examined GluR1 labeling in the rostral NTS, which
receives gustatory rather than baroreceptive afferents. The
density of GluR1 puncta in the rostral NTS was similar in
16-week-old SHR (14⫾2.2 per 0.01 mm2) compared with
16-week-old WKY (12⫾4.1 per 0.01 mm2). Thus, GluR1
puncta labeling is selectively increased in the region of NTS
that receives baroreceptive afferents and is increased only in
hypertensive rats.
The above results indicate that there are potentially more
GluR1-containing spines in the NTS of SHR after the
development of hypertension. We can imagine at least 2
possible scenarios for an increased number of GluR1containing spines in the NTS after sustained hypertension
(although these scenarios are not mutually exclusive). These
2 models are illustrated schematically in Figure 3. The bottom
panel of Figure 3 shows the result after hypertension in which
there is an increase in the number of GluR1-containing
dendritic spines on each neuron. The 2 possible models
indicate 2 possible mechanisms: In the first model (Figure
3A), the elevation in blood pressure causes an increase in the
Aicher et al
GluR1 in NTS of SHR
1249
Figure 3. Schematic shows 2 potential mechanisms for increasing the number of GluR1-labeled spines after hypertension. A,
More GluR1 is trafficked to preexisting spines that would not be
visible by light microscopy if they were not labeled. B, There is
an increase in the number of dendritic spines whereas the proportion of spines containing GluR1 is constant. Distinction
between models requires electron microscopy in which both
GluR1-labeled and unlabeled dendritic spines can be observed.
Downloaded from http://hyper.ahajournals.org/ by guest on May 7, 2017
synthesis of GluR1 and an enhanced trafficking of the
receptor to preexisting spines. In this model (Figure 3A),
there is no change in the total number of spines per neuron,
but after hypertension, a greater proportion of the spines
contain GluR1. In an alternative model (Figure 3B), the
elevation in blood pressure leads to an increase in the total
number of spines in NTS through the addition of new
dendritic spines on each NTS neuron. In this second model,
the proportion of spines that contain GluR1 is constant even
after the addition of new spines to each neuron (Figure 3B).
To determine if there is an increase in spine number or an
increased distribution of GluR1 to preexisting spines, we
examined the distribution of GluR1 at the electron microscopic level. Both labeled and unlabeled spines are visible
with electron microscopy but not with light microscopy.
Consistent with our light microscopic observations, there
was a significant increase in the number of GluR1-labeled
spines in SHR compared with WKY in 16-week-old animals
but not in 5-week-old animals (Figures 4A and 4B). There
were no differences in the number of unlabeled spines
between the groups at either age (Figures 4A and 4B). This
finding suggests that hypertension leads to both a greater
number of spines in the NTS and an increase in the proportion
of those spines that contain GluR1 (so both mechanisms
illustrated in Figure 3 appear to occur in the NTS after
hypertension). This morphological change was specific to
spines, since there was not a significant difference between
SHR and WKY with regard to the proportion of small
dendrites (Figures 4C and 4D) or large dendrites (not shown)
containing GluR1. To confirm that these morphological
changes were indeed related to blood pressure and not
inherent genetic differences, we did similar comparisons
between SHR and WKY at 5 weeks of age, before the onset
of hypertension in the SHR. There were no differences in
GluR1 labeling of spines (Figure 4A) or small dendrites
(Figure 4C) at the electron microscopic level between the 2
strains of rats at 5 weeks of age.
Although not quantitatively assessed, we did note some
interesting morphological features in GluR1-labeled spines of
hypertensive rats. In normotensive WKY, small dendrites
(Figures 5A and 5B) and dendritic spines (Figure 5C)
contained GluR1. GluR1-containing dendrites received asymmetric synapses from unlabeled axon terminals, and many of
Figure 4. By electron microscopy, the number of GluR1-labeled
spines was similar for SHR (filled bars) and WKY (open bars) at
5 weeks of age (A). Number of GluR1-labeled spines was
increased in NTS of SHR compared with WKY at 16 weeks of
age (B). There were no differences in the number of unlabeled
spines between strains at either age (A, B). There were no differences in number of GluR1 small dendrites between strains of
rats at 5 weeks of age (C) or at 16 weeks of age (D).
these afferent terminals contained dense core vesicles (Figure
5C). Most of these afferents contacted a single dendritic
target (Figures 5B and 5C), but other terminals formed
glomeruli with multiple dendritic targets, one or more of
which contained GluR1 (Figure 5A). In the SHR, GluR1containing dendrites also received asymmetric synapses from
axon terminals (Figure 6). However, in the SHR, many of
these axon terminals formed perforated synapses (Figures 6A
and 6B) or exhibited extensive synaptic densities (Figure 6C).
Instead of forming typical glomeruli, these afferents often
contacted GluR1-containing dendrites that were in direct
apposition to each other (Figure 6A). In both WKY and SHR,
we often noted that astrocytic glial processes surround both
the axon terminal and its dendritic target, particularly at
asymmetric synaptic junctions (Figures 5B and 5C). However, in the SHR, these glial processes appeared to be more
abundant and possibly enlarged (Figure 6). The morphology
of the contacts between unlabeled axon terminals and their
GluR1-containing dendritic targets in the SHR is suggestive
of snapshots of synapses undergoing plasticity,22,23 potentially in the process of dividing dendritic targets into smaller
dendrites or spines.24
Discussion
AMPA receptors in the NTS are critical for baroreceptor
transmission11 and are present in baroreceptive neurons15 in
the medial NTS. These observations, together with the
prevalence of cardiopulmonary and barosensitive neurons
within the medial NTS,25 support the notion that many of the
AMPA-receptive neurons in the medial NTS are related to
cardiovascular function. We found that in the SHR model of
hypertension, the number of GluR1-containing spines was
increased within the medial NTS. These results indicate that
1250
Hypertension
June 2003
Downloaded from http://hyper.ahajournals.org/ by guest on May 7, 2017
Figure 5. GluR1 in dendrites of NTS
neurons in 16-week-old WKY. A, Large
unlabeled axon terminal (ut) forms asymmetric synapses (curved arrows) with 2
dendrites (GluR1-d1 and GluR1-d2) containing immunoperoxidase labeling for
GluR1. B, Unlabeled terminal (ut) forms
asymmetric synapse (curved arrow) with
dendrite containing immunoperoxidase
labeling for GluR1. Astrocytic glial processes (*) surround both axon terminal
and its dendritic target. C, Unlabeled
axon terminal (ut) containing both small,
clear vesicles and dense core vesicles
(dcv) forms asymmetric synapse (curved
arrow) with dendritic spine (GluR1-s) that
contains immunoperoxidase for GluR1.
Terminal and portions of the spine are
surrounded by astrocytic glial processes
(*). Scale bars⫽500 nm.
some glutamate-receptive NTS neurons in the hypertensive
brain change morphologically, and more spines develop. The
majority of these additional spines contained the GluR1
subunit of the AMPA receptor, supporting a role for AMPA
receptors in synaptic plasticity22 in the NTS during
hypertension.
Our finding of an increase in GluR1 within NTS neurons is
in contrast to a recent study26 that found no significant
increase in AMPA receptor binding in the NTS of hypertensive animals with radioactive AMPA receptor ligands used to
assess binding. However, our results indicate that the increase
in receptor density may not be sufficient to be detected with
the use of some sampling methods. For example, although we
clearly were able to detect an increase in the density of
GluR1-labeled spines in the NTS of hypertensive rats, this
increase was not sufficient to increase the optical density of
GluR1 staining in the whole nucleus. This suggests that
changes may be subtle, may occur in only a subset of NTS
neurons, and may require different techniques to be detected.
An increase in the number of AMPA receptors might
appear to be counterintuitive, since there is a blunting of the
baroreflex during hypertension. If AMPA receptors mediate
the baroreflex, their increase would be expected to enhance
reflex function if they were located on excitatory projection
neurons in the NTS. Indeed, a large upregulation of GluR1
and other glutamate receptors in the NTS by heat shock has
been reported to augment baroreflex responses.27 However, in
the current study, the GluR1 increase was selectively localized to dendritic spines in a subset of neurons. These AMPA
receptors on dendritic spines may be calcium permeable,
since they appear to lack GluR2 AMPA subunits.16 The
calcium signal produced by activation of these receptors may
be isolated in the spine, but it is also possible that afferent
activation of AMPA receptors on individual spines may have
less impact on the postsynaptic cell than activation of
synapses on larger dendrites.28 –30 We would speculate that
the segregation of these GluR1-containing AMPA receptors
to dendritic spines could create a situation in which activation
of their afferents would have less impact on NTS neurons,
except perhaps under conditions in which there is synchronous activation of many of these afferents. This is a feasible
mechanism contributing to the decreased sensitivity of the
baroreflex during hypertension.
It is possible that the changes in cellular morphology and
the changes in receptor distribution are 2 independent events.
Factors influencing the formation of dendritic spines may be
distinct from mechanisms regulating glutamate receptor trafficking and/or turnover. An alternative mechanism that may
lead to an increase in the density of GluR1 would be a
reduced turnover of this receptor population. If the mechanisms regulating receptor production are independent, a
reduced turnover could lead to an increase in receptor density.
However, this would not explain the increase in the formation
of dendritic spines.
In other brain regions, increases in dendritic spine density
have been correlated with enhanced afferent activation and
glutamatergic stimulation.31–33 Recent studies have shown
that long-term potentiation leads to an increase in the number
of perforated synapses in hippocampal neurons,34 whereas
others have shown an increased expression of AMPA receptors in the dendrites of perforated synapses.22 Only recently
have attempts been made to determine if changes in spine
density lead to changes in the activity of the postsynaptic
neurons, and it is clear that we still do not fully understand the
Aicher et al
GluR1 in NTS of SHR
1251
Downloaded from http://hyper.ahajournals.org/ by guest on May 7, 2017
ever, the long-term consequences of hypertension on baroreceptor afferents are not fully understood, nor are the central
changes in response to alterations in afferent activity. There
also may be increased activity in other glutamatergic afferents to the NTS, such as cardiopulmonary afferents.
The morphology of the GluR1-labeled synapses in SHR
was intriguing, and these synapses resemble synapses exposed to increased afferent activity,24 including the abundance of perforated synapses,37 which are thought to be
related to plasticity.23 We also noted that the ensheathing of
these synapses by glial processes was extensive. Astrocytic
processes have long been noted surrounding asymmetric
synaptic complexes,38 but in light of recent evidence showing
the presence of glutamate transporters39 and receptors on glial
processes,20,40,41 it is possible that these astrocytes may play
a greater role in shaping these synapses and their function
than previously thought possible.42,43
Our results indicate that even at the first relay site of the
baroreceptor reflex pathway, there are structural and molecular changes in response to sustained increases in blood
pressure. These observations are consistent with previous
studies showing altered responses to glutamate injections in
the NTS of hypertensive rats8 as well as altered membrane
properties of NTS neurons in the hypertensive rat.7 A more
recent study has shown altered responses of second-order
baroreceptive neurons in another model of hypertension.6
Together these results support the idea that since baroreceptor
activity is mediated through glutamate and baroreflex function is altered during hypertension, at least some of these
functional changes are manifested in glutamate-receptive
neurons within the NTS. Our findings do not preclude the
possibility that other receptors may also change within these
neurons or that other changes may occur at other sites along
the baroreflex arc.
Figure 6. GluR1 in dendrites of NTS neurons in 16-week-old
SHR. A, Large unlabeled axon terminal (ut) forms asymmetric
synapses (curved arrows) with 2 dendrites (GluR1-d1 and
GluR1-d2) containing immunoperoxidase labeling for GluR1.
Synaptic density with GluR1-d1 appears to be perforated. Dendrites and portions of axon terminal are surrounded by glial processes (*). B, Unlabeled axon terminal (ut) forms perforated
asymmetric synapse (curved arrows) with dendrite (GluR1-d)
containing immunoperoxidase labeling for GluR1. Both terminal
and dendrite are surrounded by thick astrocytic glial processes
(*). C, Unlabeled axon terminal (ut) forms a large synaptic contact with dendrite (GluR1-d) containing immunoperoxidase for
GluR1. Axon terminal contains filaments at sites distal to the synapse (small, straight arrows), suggesting that this is an en passant
synapse. *Much of the axon terminal and its dendritic target are
surrounded by astrocytic processes. Scale bars⫽500 nm.
function of dendritic spines or the consequences of such
structural changes for the neuron.35 However, the evidence
strongly suggests that alterations in spine density, together
with changes in AMPA receptor distribution, probably are
related to changes in the activity of glutamatergic afferents to
the neuron. One puzzling aspect of these studies is that the
morphological changes seen in the SHR are consistent with
an increase in glutamatergic synaptic activity, not a decrease.
Aortic baroreceptors are known to reset36 and thus are
thought to decrease their activity during hypertension. How-
Perspectives
The functional consequences of an increase in spine density
and how this alters other components of the reflex pathway,
remain to be determined. Even for an individual neuron, we
do not understand the functional significance of an increase in
spine density. Two recent studies examined spine formation
and stability in the living cerebral cortex44,45 and came to
dramatically different conclusions.46 The first study44 showed
that cortical neurons have a relatively unstable population of
dendritic spines, whereas the second study45 concluded that
neurons in a different region of the cortex had extremely
stable dendritic spines. The meaning of these cellular changes
with regard to information storage or relative weighting of
synaptic inputs is premature. It will be interesting to see if the
structural changes seen in NTS neurons are preventable
and/or reversible. Insights will require more detailed information about the morphology of baroreceptive neurons
within the NTS, the organization of afferents to these neurons, and the functional consequences of morphological
changes in these neurons.
Acknowledgments
Grant support was from the American Heart Association and
National Institutes of Health (HL56301). The authors thank Alla
1252
Hypertension
June 2003
Goldberg, James Kraus, Aarti Patel, and Kristin Swanson for
technical assistance. Some of the experiments were conducted at the
Weill Medical College of Cornell University in the Department of
Neurology and Neuroscience, Division of Neurobiology.
References
Downloaded from http://hyper.ahajournals.org/ by guest on May 7, 2017
1. Dampney RAL. Functional organization of central pathways regulating
the cardiovascular system. Physiol Rev. 1994;74:323–364.
2. Talman WT, Perrone MH, Reis DJ. Evidence for L-glutamate as the
neurotransmitter of primary baroreceptor afferent nerve fibers. Science.
1980;209:813– 815.
3. Sykes RM, Spyer KM, Izzo PN. Demonstration of glutamate immunoreactivity in vagal sensory afferents in the nucleus tractus solitarius of the
rat. Brain Res. 1997;762:1–11.
4. Ferrario CM, Averill DB. Do primary dysfunctions in neural control of
arterial pressure contribute to hypertension? Hypertension. 1991;18(suppl
I):I-38 –I-51.
5. Head GA. Baroreflexes and cardiovascular regulation in hypertension.
J Cardiovasc Pharmacol. 1995;26(suppl 2): S7–S16.
6. Zhang J, Mifflin SW. Integration of aortic nerve inputs in hypertensive
rats. Hypertension. 2000;35:430 – 436.
7. Sundaram K, Johnson SM, Felder RB. Altered expression of delayed
excitation in medial NTS neurons of spontaneously hypertensive rats.
Neurosci Lett. 1997;225:205–209.
8. Talman WT, Lewis SJ. Altered cardiovascular responses to glutamate and
acetylcholine microinjected into the nucleus tractus solitarii of the SHR.
Clin Exp Hypertens [A]. 1991;13A:661– 668.
9. Lohmeier TE. The sympathetic nervous system and long-term blood
pressure regulation. Am J Hypertens. 2001;14:147S-154S.
10. Hollmann M, Heinemann S. Cloned glutamate receptors. Annu Rev
Neurosci. 1994;17:31–108.
11. Andresen MC, Yang MY. Non-NMDA receptors mediate sensory
afferent synaptic transmission in medial nucleus tractus solitarius. Am J
Physiol Heart Circ Physiol. 1990;90:H1307–H1311.
12. Gordon FJ, Leone C. Non-NMDA receptors in the nucleus of the tractus
solitarius play the predominant role in mediating aortic baroreceptor
reflexes. Brain Res. 1991;568:319 –322.
13. Yen JC, Chan JYH, Chan SHH. Differential roles of NMDA and
non-NMDA receptors in synaptic responses of neurons in nucleus tractus
solitarii of the rat. J Neurophysiol. 1999;81:3034 –3043.
14. Kessler JP, Baude A. Distribution of AMPA receptor subunits GluR1– 4
in the dorsal vagal complex of the rat: a light and electron microscope
immunocytochemical study. Synapse. 1999;34:55– 67.
15. Botsford SA, Dean C, Hopp FA, Seagard JL. Presence of glutamate
receptor subtypes on barosensitive neurons in the nucleus tractus solitarius of the dog. Neurosci Lett. 1999;261:113–117.
16. Aicher SA, Sharma S, Mitchell JL. Co-localization of AMPA receptor
subunits in the nucleus of the solitary tract in the rat. Brain Res. 2002;
958:454 – 458.
17. Okamoto K, Aoki K. Development of a strain of spontaneously hypertensive rat. Jpn Circ J. 1963;27:282–293.
18. Hsu SM, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex
(ABC) in immunoperoxidase techniques: a comparison between ABC and
unlabeled antibody (PAP) procedures. J Histochem Cytochem. 1981;29:
557–580.
19. Pickel VM. Immunocytochemical methods. In: Heimer L, Robards MJ,
eds. Neuroanatomical Tract Tracing Methods. New York, NY: Plenum
Publishing; 1981:483–509.
20. Ambalavanar R, Ludlow CL, Wenthold RJ, Tanaka Y, Damirjian M,
Petralia RS. Glutamate receptor subunits in the nucleus of the tractus
solitarius and other regions of the medulla oblongata in the cat. J Comp
Neurol. 1998;402:75–92.
21. Gray EG. Axo-somatic and axo-dendritic synapses of the cerebral cortex:
an electron microscope study. J Anat. 1959;93:420 – 433.
22. Desmond NL, Weinberg RJ. Enhanced expression of AMPA receptor
protein at perforated axospinous synapses. Neuroreport. 1998;9:
857– 860.
23. Geinisman Y, Morrell F, deToledo-Morrell L. Perforated synapses on
double-headed dendritic spines: a possible structural substrate of synaptic
plasticity. Brain Res. 1989;480:326 –329.
24. Geinisman Y, Berry RW, Disterhoft JF, Power JM, Van der Zee EA.
Associative learning elicits the formation of multiple-synapse boutons.
J Neurosci. 2001;21:5568 –5573.
25. Deuchars J, Li YW, Kasparov S, Paton JFR. Morphological and electrophysiological properties of neurones in the dorsal vagal complex of the rat
activated by arterial baroreceptors. J Comp Neurol. 2000;417:233–249.
26. Ashworth-Preece MA, Chen F, Jarrott B, Lawrence AJ. Visualisation of
AMPA binding sites in the brain stem of normotensive and hypertensive
rats. Brain Res. 1999;834:186 –189.
27. Chan SH, Chang KF, Ou CC, Chan JY. Up-regulation of glutamate
receptors in nucleus tractus solitarii underlies potentiation of baroreceptor
reflex by heat shock protein 70. Mol Pharmacol. 2002;61:1097–1104.
28. Vu ET, Krasne FB. Evidence for a computational distinction between
proximal and distal neuronal inhibition. Science. 1992;255:1710 –1712.
29. Korkotian E, Segal M. Bidirectional regulation of dendritic spine
dimensions by glutamate receptors. Neuroreport. 1999;10:2875–2877.
30. Sabatini BL, Maravall M, Svoboda K. Ca(2⫹) signaling in dendritic
spines. Curr Opin Neurobiol. 2001;11:349 –356.
31. Lissin DV, Carroll RC, Nicoll RA, Malenka RC, Von Zastrow M. Rapid,
activation-induced redistribution of ionotropic glutamate receptors in
cultured hippocampal neurons. J Neurosci. 1999;19:1263–1272.
32. Kossel AH, Williams CV, Schweizer M, Kater SB. Afferent innervation
influences the development of dendritic branches and spines via both
activity-dependent and non-activity-dependent mechanisms. J Neurosci.
1997;17:6314 – 6324.
33. Yuste R, Bonhoeffer T. Morphological changes in dendritic spines associated with long-term synaptic plasticity. Annu Rev Neurosci. 2001;24:
1071–1089.
34. Toni N, Buchs PA, Nikonenko I, Povilaitite P, Parisi L, Muller D.
Remodeling of synaptic membranes after induction of long-term potentiation. J Neurosci. 2001;21:6245– 6251.
35. Harris KM, Kater SB. Dendritic spines: cellular specializations imparting
both stability and flexibility to synaptic function. Ann Rev Neurosci.
1994;17:341–371.
36. Xie P, McDowell TS, Chapleau MW, Hajduczok G, Abboud FM. Rapid
baroreceptor resetting in chronic hypertension: implications for normalization of arterial pressure. Hypertension. 1991;17:72–79.
37. Peters A, Kaiserman-Abramof IR. The small pyramidal neuron of the rat
cerebral cortex: the synapses upon dendritic spines. Z Zellforsch. 1969;
100:487–506.
38. Peters A, Palay SL, Webster HD. The Fine Structure of the Nervous
System: Neurons and Their Supporting Cells. 3rd ed. Oxford, UK: Oxford
University Press; 1991.
39. Seal RP, Amara SG. Excitatory amino acid transporters: a family in flux.
Annu Rev Pharmacol Toxicol. 1999;39:431– 456.
40. Aicher SA, Sharma S, Pickel VM. N-methyl-D-aspartate receptors are
present in vagal afferents and their dendritic targets in the nucleus tractus
solitarius. Neuroscience. 1999;91:119 –132.
41. Teichberg VI. Glial glutamate receptors: likely actors in brain signaling.
FASEB J. 1991;5:3086 –3091.
42. Araque A, Parpura V, Sanzgiri RP, Haydon PG. Tripartite synapses: glia,
the unacknowledged partner. Trends Neurosci. 1999;22:208 –215.
43. Newman EA, Zahs KR. Modulation of neuronal activity by glial cells in
the retina. J Neurosci. 1998;18:4022– 4028.
44. Trachtenberg JT, Chen BE, Knott GW, Feng G, Sanes JR, Welker E,
Svoboda K. Long-term in vivo imaging of experience-dependent synaptic
plasticity in adult cortex. Nature. 2002;420:788 –794.
45. Grutzendler J, Kasthuri N, Gan WB. Long-term dendritic spine stability
in the adult cortex. Nature. 2002;420:812– 816.
46. Ottersen OP, Helm PJ. How hardwired is the brain? Nature. 2002;420:
751–752.
Structural Changes in AMPA-Receptive Neurons in the Nucleus of the Solitary Tract of
Spontaneously Hypertensive Rats
Sue A. Aicher, Sarita Sharma and Jennifer L. Mitchell
Downloaded from http://hyper.ahajournals.org/ by guest on May 7, 2017
Hypertension. 2003;41:1246-1252; originally published online April 14, 2003;
doi: 10.1161/01.HYP.0000069007.98987.E0
Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2003 American Heart Association, Inc. All rights reserved.
Print ISSN: 0194-911X. Online ISSN: 1524-4563
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://hyper.ahajournals.org/content/41/6/1246
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Hypertension can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Hypertension is online at:
http://hyper.ahajournals.org//subscriptions/