70 P
the brainstem was frozen and cut into 14 mm-thick sections out in seven sections for each animal one section every
using a Reichert–Jung cryostat. The sections were stored 500 mm. The mean number of immunostained neurons
in a solution of 0.1 sodium azide, 0.25 in phosphate Fos1, GABA1 Fos1 and GABA2 Fos1 on each side
buffered saline PBS, 0.1 M. of the brain and for each animal were then calculated.
Brainstem sections were first immunostained for Fos. These means were used to compare stained neurons during
For this purpose, free-floating sections from selected brain AS-carbachol and wakefulness n53 and n54, respective-
stem sections were incubated overnight in a rabbit poly- ly.
clonal Fos antiserum Fos Ab5; Oncogene Research Prod- The serotonergic area extends beyond the cytoarchitec-
ucts Calbiochem, La Jolla, CA, USA at a dilution of tonic limits of the DRN [42,60]. In order to standardize the
1:40 000 in PBS. The sections were rinsed four times in area of analysis, the lateral borders were defined by a
PBS for a total duration of 30 min and then incubated for parasagittal plane traced from the lateral border of the
90 min in a biotinylated donkey anti-rabbit immuno- main fascicle of the medial longitudinal fascicle mlf, see
globulin G 1:700. Subsequently, the sections were incu- Fig. 3B. The dorsal and ventral borders of the DRN were
bated with the ABC complex Vector ABC Elite kit, determined according to standard criteria [52]. Serotoner-
1:500 for 60 min. After rinsing again, the tissue was gic neurons were observed throughout the defined area of
reacted for 10–20 min with 0.6 nickel ammonium analysis. In caudal sections, special care was taken to
sulfate, 0.02
diaminobenzidine tetrahydrochloride
avoid including neurons of the dorsal tegmental nucleus of Sigma and 0.015 hydrogen peroxide in 50 ml of 50
Gudden in this analysis. mM tris buffer, pH 7.5.
Photomicrographs were obtained using a Spot digital In order to identify GABAergic neurons, polyclonal
camera attached to an Olympus B360 microscope. Images
guinea pig antibodies raised against GABA-keyhole limpet were analyzed using Adobe
PHOTOSHOP
software with a hemocyanin conjugated with glutaraldehyde NT-108,
Power Macintosh
G3 computer.
Conventional and
Protos were employed. After the sections were treated for Nomarsky optics were employed. A 1003 oil immersion
Fos immunostaining, they were incubated overnight with objective lens was used to measure the diameter of soma
GABA antibody 1:3500 and normal donkey serum profiles. The distribution of labeled neurons was deter-
NDS; 3. Then, after the sections were rinsed, they were mined from drawings using a camera lucida attachment.
incubated for 60 min with biotinylated donkey anti-guinea The level of statistical significance of the difference
pig antibody 1:300 plus NDS. After another rinse, the between the mean number of immunoreactive neurons in
tissue was incubated in the ABC complex 1:200 for 60 control and AS-carbachol cats was evaluated using the
min. Finally, the tissue was exposed to diaminobenzidine Student’s t-test. The criterion chosen to discard the null
without nickel enhancement. hypothesis was P,0.05.
Serotonin immunocytochemistry was performed on brainstem sections of both an AS-carbachol cat and an
awake cat which were also both employed for GABA and
3. Results
Fos analysis, in order to determine the distribution of serotonergic
neurons. Following
Fos immunocytoch-
Following the administration of carbachol, all the ex- emistry, the sections were incubated overnight with poly-
perimental animals exhibited the AS-carbachol state. The clonal goat serotonin antiserum Incstar, 1:10 000 and
latency to the onset of AS-carbachol was 2.061.3 min NDS. Subsequently, the tissue was incubated with
mean6S.E.M. and the duration of this induced state was biotinylated donkey anti-goat immunoglobulin 1:1000
107.864.8 min. Table 1 presents these data for each followed by incubation with ABC complex 1:400. Final-
animal. ly, the tissue was reacted with diaminobenzidine without
nickel enhancement. 3.1. Characteristic of immunostained neurons
In all procedures, four final rinses with 0.01 M PBS preceded the mounting of sections on slides. Control
As illustrated in Fig. 1, Fos immunoreactivity was omission experiments were performed without the expo-
restricted to the neuronal nuclei. Fos1 neurons were sure of the tissue to the primary antibody. In representative
identified by their dark-stained nuclei. sections, Fos immunocytochemistry was followed by a
Serotonergic immunoreactive neurons exhibited brown- Pyronin-Y counterstain.
Table 1
2.5. Data analysis
Active sleep induced by carbachol Cat
Duration min Latency min
Brainstem coronal sections from approximately A 0.5
AS-C 1 108.0
4.5
which corresponds to the rostral DRN to approximately P
AS-C 2 99.3
1.0
2 2.5 which corresponds to the caudal DRN were uti-
AS-C 3 116.0
0.5
lized. Counts of the different types of cells were carried
P . Torterolo et al. Brain Research 884 2000 68 –76
71
Fig. 1. Photomicrographs of the DRN illustrating serotonin, GABA and Fos immunoreactive neurons in AS-carbachol A–H and awake cats I. A Serotonin and Fos immunoreactive neurons in the DRN during AS-carbachol. Serotonergic neurons are stained in brown. The nuclei of the serotonergic
cells do not exhibit Fos immunostaining. Non-serotonergic Fos1 neurons, which are identified by their black-stained nucleus and empty cytoplasm arrowhead are located near serotonergic neurons. B GABAergic neurons that express c-fos GABA1 Fos1 intermingled with GABA1 Fos2 cells
arrow. Oval-shaped GABA1 Fos1 neurons are indicated by arrowheads Fos is immunostained black. GABA1 Fos1 neurons of this type were the most commonly observed within the DRN. C Oval-shaped GABA1 Fos1 neurons filled arrowheads near a GABA2 Fos1 neuron open arrowhead.
D Multipolar-shaped GABA1 Fos1 neuron arrowhead. This type of GABA1 Fos1 neuron was scarce in the DRN. E Fusiform GABA1 Fos1 neuron arrowhead. These neurons were mainly located in the ventromedial region of the DRN. The arrow indicates an adjacent GABA1 Fos2 neuron.
F Small, oval-shaped GABA1 Fos1 neuron arrowhead in close proximity to a larger GABA1 Fos2 neuron arrow. G and H Examples of non-GABAergic Fos-immunoreactive neurons GABA2 Fos1, open arrowheads during AS-carbachol. In G the neuronal body is surrounded by
GABAergic immunoreactive profiles. I GABA2 Fos1 neuron open arrowhead in the vicinity of a GABAergic Fos2 neuron arrow during wakefulness. All photomicrographs were taken from 14 mm-thick sections and processed with the diaminobenzidine method enhanced with nickel.
Calibration bars: A 50 mm; B and C 20 mm; D–I 10 mm.
stained soma and dendrites Fig. 1A. The distribution of non-serotonergic neurons. We have previously analyzed
serotonergic neurons observed in the present study was c-fos expression in serotonergic neurons [63], therefore
similar to that previously reported in the cat [20]. As this analysis was not carried out in the present study.
illustrated in the photomicrograph of Fig. 1A, during GABA immunoreactive neurons, which were observed
AS-carbachol, c-fos expression was observed mainly in in the DRN, exhibited a diffuse brown-stained soma Fig.
72 P
Table 2
1. Other structures that are known to contain GABAergic
a
Mean cell counts from individual cats
neurons, e.g. cerebellum, globus pallidus, dorsal nucleus of
Cat GABA1 Fos1
GABA2 Fos1 Total Fos
the lateral lemniscus, etc, also exhibited strong immuno- staining [35]. No GABA immunoreactivity was observed
Controls
in control omission experiments.
C1 6.2
9.5 16.7
C2 3.4
1.6 5.0
In brainstem sections immunostained for GABA and
C3 3.3
15.3 18.9
Fos, double-labeled neurons GABA1 Fos1 neurons
C4 3.1
11.0 15.1
displayed a brown cytoplasm and a black nucleus. Non- GABAergic neurons that expressed c-fos GABA2 Fos1
AS-carbachol
neurons were also observed. In about 4 of the Fos1
AS-C1 30.7
14.0 46.6
AS-C2 24.3
7.1 33.1
neurons it was not possible to determine whether or not
AS-C3 19.4
7.3 27.9
they were GABAergic because of their small size and the
a
Each value represents an average of neurons in seven sections see
intensity of background staining. These neurons were
Material and methods. The ‘Total Fos’ group included Fos1 neurons
counted as Fos1, but were not classified as either GABA1
that were impossible to identify either as GABA1 or as GABA2 see
Fos1 or GABA2 Fos1. Representative examples of
text.
GABA1 Fos1, GABA2 Fos1 and GABA1 Fos2 neurons are presented in Fig. 1.
3.3. GABAergic neurons GABA1 3.2. Fos immunoreactivity during AS-carbachol
The camera lucida drawing in Fig. 3B depicts the The mean number of Fos1 neurons GABA1 Fos1
distribution of GABAergic neurons in the DRN. These and GABA2 Fos1 was larger during AS-carbachol
neurons were mainly located in the lateral regions of the compared to wakefulness 35.965.6 vs. 13.964.4, P,
DRN; very few were found in the central region. As shown 0.05, see bar chart in Fig. 2A. Table 2 presents the mean
in Fig. 3A and B, the distribution of Fos1 neurons was cell counts from individual cats. The photomicrograph in
similar to the distribution of GABAergic neurons. Fig. 3A, illustrates the distribution of Fos1 neurons in the
There was a 620 increase in the number of GABA1 dorsal raphe in a section counterstained with Pyronin-Y.
Fos1 neurons during AS-carbachol compared to control Fos1 neurons were located adjacent to the mlf and in the
24.863.3 vs. 4.061.0, P,0.001, as illustrated in Fig. 2B lateral portions of the raphe nucleus. They were scarce in
see also Table 2. Of the total number of Fos1 neurons, the central raphe, where the largest concentration of
69 were GABAergic in the AS-carbachol animals and serotonergic neurons is located.
only 29 in awake animals 24.8 GABA1 Fos1 out of
Fig. 2. Mean numbers of immunoreactive neurons in the DRN during wakefulness filled bars and AS-carbachol empty bars. Compared to the waking state, there was a statistically significant increase in the mean number of Fos1 and GABA1 Fos1 neurons A and B during AS-carbachol. There was no
statistical difference in the mean number of GABA2 Fos1 neurons C in these two conditions. The error bars represent the SEM. , P,0.05; , P,0.001, unpaired two-tailed Student’s t-test. n54 for wakefulness and n53 for AS-carbachol.
P . Torterolo et al. Brain Research 884 2000 68 –76
73
of 8.760.3 mm. GABA1 Fos1 neurons were also heterogeneous in shape. The majority 51 of these
neurons were oval-shaped Fig. 1B,C and F, 39 were fusiform Fig. 3E and 10 were multipolar Fig. 3D.
Fig. 4A consists of camera lucida drawings from waking and AS-carbachol cats. Note the larger number of GABA1
Fos1 neurons that are present in the section obtained from the AS-carbachol cat.
The number of GABA1 Fos1 neurons in the AS- carbachol cats was larger ipsilateral to the injection site
than on the contralateral side 17.362.2 vs. 7.561.2, P,0.05; Fig. 4B. No significant differences were ob-
served between the right and left side of the sections obtained from cats that were awake. In the saline-injected
animal, the number of GABA1 Fos1 neurons was similar between the ipsilateral 1.460.7, average of seven sec-
Fig. 3. Distribution of Fos1 and GABA1 neurons in the DRN. A This photomicrograph illustrates Fos immunostaining with Pyronin-Y counters-
taining in the ventromedial region of the DRN during AS-carbachol. c-fos-expressing neurons were mainly distributed in the periphery close to
the medial longitudinal fasciculus examples are indicated by ar- rowheads. Fos immunoreactivity was not observed in the large, centrally
located neurons. The photomicrograph was obtained from a 14-mm thick section and processed with the diaminobenzidine method enhanced with
nickel. Calibration bar5100 mm. B Camera lucida drawing that depicts
Fig. 4. A Camera lucida drawings illustrating the distribution of the distribution of neurons immunostained for GABA in a representative
GABA1 Fos1 in representative brainstem sections of awake and AS- section P 21.5 of the DRN. Each dot represents one immunolabeled
carbachol cats P 21.5. Compared to wakefulness, the increase in the neuron. Note that GABAergic neurons were concentrated mainly in the
number of GABA1 Fos1 neurons in the section obtained from AS- lateral regions of the DRN. Dashed lines indicate the borders of the area
carbachol cat is readily apparent. On the side ipsilateral to the carbachol that was included in the present analysis. A and B were obtained from
microinjection, GABA1 Fos1 neurons were more abundant than in the adjacent sections of the same cat. 4V, fourth ventricle; mlf, medial
contralateral side. Each mark indicates one labeled neuron. 4V, fourth longitudinal fasciculus.
ventricle; mlf, medial longitudinal fasciculus. The dashed line designates the midline. B Regional differences in GABA1 Fos1 neurons during
AS-carbachol. This plot presents the number of c-fos-expressing neurons in brainstem sections from AS-carbachol cats both ipsilateral and
contralateral to the injection site. The mean number of GABA1 Fos1
35.9 Fos1 neurons, and 4.0 out of 13.9, respectively.
neurons was larger ipsilaterally filled bars than contralaterally empty
GABA1 Fos1 neurons were small, with an average major
bars , P,0.05, paired two-tailed Student’s t-test, n53. The error bars
diameter of 15.760.6 mm and an average minor diameter
represent the S.E.M.
74 P
tions and the contralateral sides 1.760.8, average of AS also support the previously reported increase in meta-
seven sections. bolic activity that has been described in the DRN during
During AS-carbachol, the number of GABA1 Fos1 this state [27]. In addition, Maloney et al. [28] have shown
neurons was significantly larger on both in the ipsilateral that the number of GABAergic neurons that express c-fos
17.362.2, P,0.001 and contralateral sides 7.561.2, in the DRN of the rat was positively correlated with the
P,0.005 compared to sections from the animals that were percent time spent in AS during recovery from AS
awake 1.960.6, right side; 2.160.5, left side. deprivation.
In contrast to the GABAergic neurons, there was no 3.4. Non-GABAergic neurons GABA2
significant difference in the number of non-GABAergic Fos1 GABA2 Fos1 neurons in AS-carbachol and
Compared to control conditions, no statistically signifi- awake cats.
cant changes in the number of GABA2Fos1 neurons An important question involves the source of the
were observed in AS-carbachol cats 9.562.3 vs. 9.364.1; activation of GABAergic neurons of the DRN during AS.
Fig. 2C, see also Table 2. The photomicrographs of Fig. 1 We suggest that neurons in the NPO may be involved in
are examples of GABA2 Fos1 neurons in brainstem this process for the following reasons. Electrical stimula-
sections from AS-carbachol cats Fig. 1C, G and H and tion of the NPO produces an inhibitory response in DRN
from a cat during wakefulness Fig. 1I. These neurons neurons that is blocked by GABA antagonists, but not by
A
were intermingled with GABA1 Fos1 Fig. 1C as well glycine antagonists [59]. Neurons in the NPO are also
as GABA1 Fos2 Fig. 1I neurons. activated during natural AS and directly by carbachol
[13,14,18,29]. There are also fiber projections to the DRN from the NPO and surrounding areas [22,42,44]. There-
4. Discussion fore, it is also possible that GABAergic DRN neurons are