Behavioural Brain Research 138 (2003) 17 /28
www.elsevier.com/locate/bbr
Research report
Predatory hunting and exposure to a live predator induce opposite
patterns of Fos immunoreactivity in the PAG
E. Comoli, E.R. Ribeiro-Barbosa, Newton Sabino Canteras Department of Physiology and Biophysics, Institute of Biomedical Sciences, University of São Paulo, Av. Prof. Lineu Prestes, 1524, CEP 05508-900 São
Paulo, SP, Brazil
Received 19 April 2002; received in revised form 25 June 2002; accepted 25 June 2002
Abstract
Considering the periaqueductal gray’s (PAG) general roles in mediating motivational responses, in the present study, we
compared the Fos expression pattern in the PAG induced by innate behaviors underlain by opposite motivational drivers, in rats,
namely, insect predation and defensive behavior evoked by the confrontation with a live predator (a cat). Exposure to the predator
was associated with a striking Fos expression in the PAG, where, at rostral levels, an intense Fos expression was found largely
distributed in the dorsomedial and dorsolateral regions, whereas, at caudal levels, Fos-labeled cells tended to be mostly found in the
lateral and ventrolateral columns, as well as in the dorsal raphe nucleus. Quite the opposite, insect predation was associated with
increased Fos expression predominantly in the rostral two thirds of the lateral PAG, where the majority of the Fos-immunoreactive
cells were found at the oculomotor nucleus levels. Remarkably, both exposure to the cat and insect predation upregulated Fos
expression in the supraoculomotor region and the laterodorsal tegmental nucleus. Overall, the present results clearly suggest that the
PAG activation pattern appears to reflect, at least partly, the animal’s motivational status. It is well established that the PAG is
critical for the expression of defensive responses, and, considering the present findings, it will be important to investigate how the
PAG contributes to the expression of the predatory behavior, as well.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Periaqueductal gray; Defensive behavior; Predatory behavior; Foraging behavior; Aggressive behavior; Emotion
1. Introduction
The periaqueductal gray (PAG) is known to play an
important role in the modulation of nociceptive sensory
transmission [10,43], regulation of the cadiovascular
system [20,41], and in the expression of a variety of
behaviors, including defensive [21], sexual [52], maternal
[40], and feeding behaviors [57,61]. Throughout the past
decades, however, two themes have largely dominated
Abbreviations: AQ, cerebral aqueduct; DRr, ce, ca, lw-dorsal raphe
nucleus, rostral part, central part, caudal part, lateral wing; EW,
Edinger /Westphal nucleus; LDT, laterodorsal tegmental nucleus; ND,
nucleus of Darkschewitsch; III, oculomotor nucleus; PAGdm, dl, l, vlperiaqueductal gray, dorsomedial column, dorsolateral column, lateral
column, ventrolateral column; SOM, supraoculomotor region; IV,
trochlear nucleus.
Corresponding author. Tel.: /55-11-3091-7628; fax:/55-11-30917285
E-mail address: [email protected] (N.S. Canteras).
research on PAG, namely, inhibition of nociception,
and the integration of behavioral responses to threatening or stressful stimuli (see [4]).
The pioneering studies of Hunsperger [34] led to the
widely accepted view that the PAG is critical for the
expression of defensive behavior. It was first found that
lesions centered in the PAG significantly impair spontaneous defensive behavior in cats confronted with a dog,
resulting in passive animals that were easily handled,
and rarely, if ever, showed defensive responses [34].
Subsequent studies largely corroborated and significantly extended this idea, and showed that PAG
stimulation may induce a pattern of somatomotor and
autonomic responses reminiscent of the behavior of
animals facing natural threats [2,25,35]. In fact, we had
shown that exposure to a live predator was associated
with a striking activation of the PAG, where Fos
expression was particularly abundant in the rostral
two thirds of the PAG in the dorsomedial and dorso-
0166-4328/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 6 - 4 3 2 8 ( 0 2 ) 0 0 1 9 7 - 3
18
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
lateral regions; moreover, in the caudal PAG, a more
widespread activation was observed [15].
Considering the PAG’s general roles in mediating
motivational responses, it would be important, at this
point, to examine how the PAG participates in an
opposite behavioral circumstance, where the animal,
instead of confronting a live predator, plays the
predator’s role, and is actually engaged in prey hunting.
Similar to the defensive behavior, predatory hunting
has been regarded as an innate behavioral response
seemingly critical for the animals’ survival [26]. Previous
studies have shown that pathways involved in the
expression of defensive and predatory behaviors may
diverge at brainstem levels. Thus, while defensive
responses appear to be abolished by central gray lesions,
predatory responses are said to be eliminated by lesions
of the ventral midbrain tegmentum [45], and reduced by
lesions of the mesencephalic [8] and pontine [9] reticular
formation. Conversely, predatory attack can be elicited
by electrical stimulation of the ventral midbrain [5] and
pontine tegmentum [9], and appears to be facilitated by
stimulation of the midbrain reticular formation [55].
Notably, Karli and colleagues, using the mouse-killing behavioral paradigm, showed that, in rats, extensive
PAG lesions facilitated predatory attack, whereas stimulation of the dorsal PAG, known to elicit defensive
behavior, inhibited the expression of interspecific aggressive responses [22]. These results suggest that neural
systems involved in eliciting defensive responses may
inhibit the outcome of predatory attack; however, they
do not rule out the possibility that certain PAG regions
may indeed facilitate the occurrence of the predatory
attack itself or other behaviors likely to be expressed in
the predatory hunting context, such as the behavioral
responses involved in the prey searching.
In the present work, in order to start addressing this
issue, we first examined the pattern of Fos immunoreactivity in the PAG of rats that predated cockroaches.
Cockroach predation was presently chosen as the
behavioral paradigm to explore predatory behavior in
the rat, since it is vividly expressed by all individuals, a
fact confirmed for all the different strains tested [46].
Moreover, roaches are relatively innoucuous and easily
overcome, and do not seem to induce appreciable
defensive reactions in rats. In contrast to insect predation, however, the mouse-killing behavioral paradigm,
widely used in the past to study the neural basis of
predatory behavior in rats, presents serious limitations
constraining its use in the present study. Thus, mousekilling behavior is likely to be expressed by only a
limited number of individuals (see [58]), and the
confrontation with a live mouse appears to induce overt
defensive reactions in a significant number of rats (WR
Schmidek, personal communication).
The results on the PAG activation pattern in animals
that performed predatory hunting were further com-
pared with those from another series of experiments
where the distribution of Fos immunoreactivity was
analyzed in the PAG of rats confronted with a live
predator.
Overall, the present results clearly indicate that insect
predation and exposure to a natural predator induce an
opposite activation pattern of the PAG, perhaps reflecting the diverse motivational drives underlying each one
of these responses.
2. Materials and methods
Adult male Wistar rats (n /15), weighing about 250 g
and obtained from the local breeding facilities, were
used in the present study. The animals were kept under
controlled temperature (23 8C) and illumination (12-h
cycle) in the animal quarters, and had free access to
water and standard laboratory diet (Purina). Conditions
of animal housing and all experimental procedures were
conducted under institutional guidelines, and in accordance with NIH guidelines on animal care.
One week before the experimental procedures, animals were individually housed into a plexiglass cage
(50 /35 /16 cm) with a wire-meshed cover and front
wall, and were handled repeatedly by the same investigator that conducted the behavioral tests. Five animals
were then exposed to a live predator by means of the
placement of an adult female cat (2.5 kg) in close contact
with the home cage for 10 min. Another group of
animals (n /5) was induced to hunt by a simultaneous
introduction, into the home cage, of five mature intact
cockroaches (Periplaneta americana ) raised for this
purpose in our laboratory. A third group of rats (n/
5) served as controls; they were housed and handled in
the same way as the animals of the other experimental
groups, but were neither induced to predate nor exposed
to the cat. All the behavioral tests were performed
between 17:00 and 18:00 h, and recorded with a video
camera.
Ninety minutes after the behavioral tests, each animal
was deeply anesthetized with sodium pentobarbital (40
mg/kg, i.p.) and perfused transcardially with a solution
of 4.0% para -formaldehyde in 0.1 M phosphate buffer
at pH 7.4; the brains were removed and left overnight in
a solution of 20% sucrose in 0.1 M phosphate buffer at
4 8C. The brains were then frozen, and four series of 30
mm sections were cut with a sliding microtome in the
frontal plane and collected from the caudal thalamus to
the rostral levels of the pontine central gray. One
complete series of sections was processed for immunohistochemistry with anti-Fos serum raised in rabbit (Ab5, Oncogene Science, lot # D09803) at a dilution of
1:10 000. The primary antiserum was localized using a
variation of the avidin /biotin complex system (ABC;
[33]). In brief, sections were incubated for 90 min at
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
room temperature in a solution of biotinylated goat
anti-rabbit IgG (Vector Laboratories), and then placed
in the mixed avidin-biotin-horseradish peroxidase
(HRP) complex solution (ABC Elite Kit; Vector Laboratories) for the same period of time. The peroxidase
complex was visualized by a 10-min exposure to a
chromogen solution containing 0.02% 3,3? diaminobenzidine tetrahydrochloride (DAB) with 0.3% nickel /
ammonium sulfate in 0.05 M Tris /buffer (pH 7.6),
followed by incubation for 10 min in chromogen
solution with hydrogen peroxide (1:3000) to produce a
blue /black product. The reaction was stopped by
extensive washing in potassium phosphate-buffered
saline (KPBS; pH 7.4). Sections were mounted on
gelatin-coated slides, and then dehydrated and coverslipped with DPX.
In order to delineate the PAG columns, another series
of sections was processed for immunohistochemistry
with a monoclonal antiserum directed against brain
nitric oxide synthase raised in mouse (Sigma) at a
dilution of 1:1000. The antigen/antibody complex was
localized by using a variation of the avidin-biotin
complex system (ABC; [28]), with a commercially
available kit (ABC Elite Kit; Vector Laboratories).
The sections were mounted on gelatin-coated slides
and then treated with osmium tetroxide to enhance
visibility of the reaction product. Slides were then
dehydrated and coverslipped with DPX. A third series
was stained with thionin to serve as a reference series for
cytoarchitectonic purposes. The PAG columns (dorsomedial, dorsolateral, lateral, and ventrolateral) were
delineated using the guidelines described by Carrive et
al. [19].
Fos-like immunoreactive cells were detected with the
X10 objective of a Nikon Eclipse E600 microscope
equipped with a camera lucida. For a cell to be
considered as expressing Fos-like immunoreactivity,
the nucleus of the neurons had to be of appropriate
size (ranging approximately from 8 to 15 mm) and shape
(oval or round), show the characteristic blue /black
staining of oxidized DAB-Ni, and be distinct from the
background at magnification of X10. For each animal,
at six distinct rostrocaudal levels of the PAG, Fospositive cells were plotted and counted from two
adjacent sections (120 mm apart). For each experimental
group (five animals in each group), the counts were then
divided by the number of counted sections to provide a
mean cell count per slice for each PAG column in each
of the six rostrocaudal segments. The data were
analyzed by a multivariate analysis of variance (oneway MANOVA, where we treated Fos-like immunoreactive cell counts in each different PAG region as
dependent variables, and the experimental groups as the
between-groups independent variable), followed by
planned comparisons. The significant level was set at
5%. All the values are expressed as mean9/S.E.M.
19
3. Results
The simultaneous introduction of five cockroaches
into the animals’ home cage induced a marked predatory behavior. Thus, at first, the animals sniffed
vigorously around the cage, and, as the prey object
was located, the predator rushed toward the roaches and
tried to seize them. The prey capture was assisted by
pinning the prey to the substrate with the forepaws, or
grasping the prey with the forepaws either simultaneously or shortly after the killing bite was administered
toward the prey’s head. The killed roaches were then
taken to one corner of the cage, where the rats
started eating them voraciously. All the animals observed in the present study took less than 40 min
(28.649/3.39 min) to consume the five roaches placed
into their home cages.
In order to describe our results, we followed the PAG
parcellation originally proposed by Bandler et al. [1] and
recently revised by Carrive et al. [19]. This parcellation is
based on anatomical, neurochemical, and functional
data, and divides the PAG into dorsomedial, dorsolateral, lateral, and ventrolateral columns, in addition to a
ventromedial portion containing the supraoculomotor
region, as well as the oculomotor, dorsal raphe, and
laterodorsal tegmental nuclei, usually considered on
functional and anatomical grounds as separable from
the rest of the PAG. The animals that hunted the
roaches displayed a similar pattern of Fos immunoreactivity along the rostrocaudal axis of the PAG (Fig.
1A /F, Fig. 5D). Thus, compared with the other
experimental groups, insect predation induced a significantly larger increase in Fos expression in the rostral
two thirds of the lateral column of the PAG (all F1,12 ]/
37.198, all P B/0.001) (Fig. 4), where the majority of the
labeled cells were found at the oculomotor nucleus levels
(Fig. 1A /D, Figs. 4 and 5D). Interestingly, at these
levels, Fos-immunoreactive cells tended to be clustered
in the outer half of the lateral column (Fig. 1B, C, Fig.
5D), which appears to correspond, at least in part, to a
region known to present a distinct acetylcholinesterasepositive neuropil [51]. Moreover, compared with the
control group, the animals that preformed predatory
hunting presented a significant increase in the number of
Fos-immunostained cells in the supraoculomotor region
(all F1,12 ]/5,835, all P B/0.05) (Fig. 1B, C, Fig. 3B, C,
Fig. 5A, D) and in the laterodorsal tegmental nucleus
(F1,12 ]/100,148, P B/0.001) (Fig. 1F, Fig. 3F), whereas,
in the other PAG sites, the number of Fos-immunoreactive cells found for the animals that hunted the
roaches did not significantly differ from the one found
for the control group (Fig. 4).
In the group of animals that confronted the predator,
during the 10 min of direct exposure, all the five animals
remained expressing a robust freezing response most of
the time (8.189/0.55 min), and, compared with the other
20
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
Fig. 1. (A /F) Camera lucida drawings of sections taken through the periaqueductal gray, arranged from rostral (A) to caudal (F), showing the
distribution of Fos-immunoreactive cells (black dots) in a rat that predated roaches. The approximate distance from the interaural line is indicated on
the top of each figure. For abbreviations, see list. Scale bar/1 mm.
experimental groups, presented, in several specific PAG
sites, a marked increase in Fos immunoreactivity (Fig.
2A /F, Fig. 5B). Thus, at the level of nucleus of
Darkschewitsch, Fos expression was particularly in-
creased in the dorsomedial column, whereas the lateral
PAG contained relatively sparser Fos staining (Fig. 2A).
Proceeding caudally, at the oculomotor nucleus levels, a
striking Fos expression was found in the dorsomedial
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
21
Fig. 2. (A /F) Camera lucida drawings of sections taken through the periaqueductal gray, arranged from rostral (A) to caudal (F), showing the
distribution of Fos-immunoreactive cells (black dots) in a rat exposed to the predator. The approximate distance from the interaural line is indicated
on the top of each figure. For abbreviations, see list. Scale bar/1 mm.
and dorsolateral PAG, whereas, in sharp contrast, the
lateral PAG contained a much smaller number of
stained cells (Fig. 2B, C, Fig. 5B). Moreover, at these
levels, compared with the control group, a distinct
increase in Fos expression was also noted in the
supraoculomotor region (all F1,12 ]/21.682, all P B/
0.001) (Fig. 2B, C, Fig. 3B, C, Fig. 5A, B). At the
caudal half of the PAG, the dorsomedial and dorsolateral columns of the PAG continued expressing a
remarkable number of Fos-stained cells, which, however, tended to show a clear decline toward the caudal
end of the PAG (Fig. 2D/F, Fig. 4). Conversely, at
these levels, in the lateral PAG, Fos expression increased
toward the caudal end (Fig. 2D/F, Fig. 4), where an
impressive number of Fos-labeled cells was observed
(Fig. 2F). Notably, according to the parcellation proposed by Ruiz-Torner et al. [51], at the caudal end of the
PAG, the ventrolateral column encompasses that which
22
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
Fig. 3. (A /F) Camera lucida drawings of sections taken through the periaqueductal gray, arranged from rostral (A) to caudal (F), showing the
distribution of Fos-immunoreactive cells (black dots) in a control rat. The approximate distance from the interaural line is indicated on the top of
each figure. For abbreviations, see list. Scale bar/1 mm.
we have presently considered as being the lateral region.
Exposure to the predator also induced a robust Fos
expression in the rest of the ventrolateral column, which
presented a larger number of stained cells toward the
caudal pole of the PAG (Fig. 2D /F, Fig. 4). In addition,
compared with the control group, at caudal levels of the
PAG, a distinct increase in Fos immunoreactivity was
also found in the dorsal raphe nucleus (all F1,12 ]/
148.939, all P B/0.001) (Fig. 2D/F, Fig. 3D/F), where
the region of the lateral wing contained a particularly
high density of immunoreactive cells (Fig. 2E). Finally,
at the caudal end of the PAG, compared with the
control group, exposure to the predator also induced a
significant Fos expression increase in the laterodorsal
tegmental nucleus (F1,12 ]/269.71, P B/0.001) (Fig. 2F,
Fig. 3F).
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
23
Fig. 4. Frequency histograms show the mean number of Fos-immunoreactive cells per section in all experimental groups (control, insect predation,
and exposure to the predator) for the dorsomedial (dm), dorsolateral (dl), lateral (l) and ventrolateral (vl) regions of the PAG. Levels A /F
correspond to the rostrocaudal levels depicted in Figs. 1 /3. Data are expressed as mean9/S.E.M.
4. Discussion
Before discussing our results regarding the pattern of
Fos immunoreactivity in the PAG, we should consider
some of this technique’s potential limitations. Although
Fos protein expression has been used as a sensitive
cellular marker for neuronal activation induced by a
variety of stimuli [42], it is important to keep in mind
that the absence of neuronal Fos expression cannot be
interpreted as lack of influence on neuronal activity,
especially considering that Fos expression is not induced
by the opening of ionotropic channels that do not
increase the intracellular levels of second messengers,
such as the classical chloride channel [56]. Moreover,
absence of Fos staining may also result from an
antibody dilution, inadequate to detect small changes
in some neuronal populations. Therefore, in the present
study, several antibody concentrations were initially
screened to determine a dilution resulting in robust
Fos staining and low levels of background labeling.
We have presently shown that different behavioral
responses, underlain by opposite motivational drives,
such as insect predation and defensive behavior during
the confrontation with a live predator, are accompanied
by an opposite pattern of mobilization of the PAG
columns.
The present results confirm and, in many ways,
extend the conclusions drawn by our previous study
on the pattern of Fos immunoreactivity in the PAG
after the encounter with a live predator [15]. Although
the number of Fos-stained cells in the PAG regions of
the present experiments was much higher when compared with the figures observed in that previous study, a
similar profile in the distribution of Fos- immunoreactive cells was presently found along the rostro-caudal
extent of the dorsomedial, dorsolateral, lateral, and
ventrolateral PAG columns. Thus, similarly to what we
had previously reported for animals exposed to a live
predator [15], we have presently found, at rostral levels,
an intense Fos expression largely distributed in the
dorsomedial and dorsolateral regions, whereas, toward
the caudal end of the PAG, Fos-labeled cells tended to
be more distributed in the ventrolateral and lateral
columns.
Interestingly, in the PAG, there is a striking overlap
between the distribution of Fos expression induced by
confrontation with a live predator and the sites of
projection from the dorsal premammillary nucleus plus
the dorsomedial part of the ventromedial hypothalamic
nucleus [16,17], which are known to represent integral
elements of the medial hypothalamic defensive system
[18]. Of particular relevance, this medial hypothalamic
system receives inputs from specific amygdalar and
septohippocampal domains likely to be involved in
processing information related to environmental threats,
and, in turn, appears to be critical for the adequate
expression of innate fear responses [18].
According to the present data, the rostral two-thirds
of the dorsolateral column is the most mobilized PAG
site during the encounter with the predator. It is well
established that stimulation of the dorsal PAG evokes
vigorous somatomotor responses, such as running and
24
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
Fig. 5. A, B, and D-Photomicrographs of transverse Fos-stained sections of the PAG, at the level of the oculomotor nucleus, from a control rat (A),
from a rat exposed to a cat (B), and from a rat that predated roaches (D). C-Photomicrograph of brain nitric oxide synthase immunostained section,
adjacent to the one shown in D, used as reference for delineating the PAG columns. Left side of the brain. For abbreviations, see list. Scale bars/200
mm.
jumping, as well as reactive immobility (freezing),
accompanied by sympathoexcitation, non-opioid analgesia, and cardiorespiratory changes, characterized
by blood pressure and heart rate increase, as well as
vasodilatation in hindlimb muscles, along with an
increase in the rate and depth of respiration
[3,20,24,41,43].
Apart from receiving inputs from elements of the
medial hypothalamic defensive system, the dorsolateral
PAG, at the level of the oculomotor and troclear nuclei,
receives substantial projections from the medial regions
of the intermediate and deep layer of the superior
colliculus, as well as from the cuneiform nucleus, which
are known to be critical for transmitting information
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
related to visual threats, such as rapidly expanding or
looming shadows in the upper visual field [47]. Moreover, the dorsolateral PAG is also densely innervated by
elements of the orbitomedial prefrontal cortex, including the caudal parts of the infralimbic and prelimbic
areas, dorsal and ventral parts of the anterior cingulate
cortex, as well as the perirhinal cortex, thought to be
involved in mediating active emotional coping response
to psychological stressors [27]. Actually, in addition to
exposure to a live predator, a number of other situations
which may threaten the animals (e.g. exposure to the
plus maze and handling) are also likely to induce a
prominent Fos expression upregulation in the rostral
part of the dorsolateral PAG (NS Canteras, personal
observations). Importantly, however, this PAG region
does not seem to be particularly activated by any
physical stressor, such as cutaneous and visceral pain
(see [35]).
Exposure to predator also induced an impressive Fos
expression in the ventrolateral PAG. However, in
contrast to what has just been said about the dorsal
PAG, stimulation of the ventrolateral PAG evokes
hyporeactive immobility, sympathoinhibition, blood
pressure and heart rate decrease, and opioid-dependent
analgesia [3,20,24,41,43], typically found in the passive
emotional coping responses likely to occur during the
recuperative phase of a number of stressful situations.
Since our behavioral paradigm prevented direct contact
between prey and predator, there were no physical
harms to the animals. As far as we can tell, after the
encounter, the rats did not seem to present passive
emotional coping responses, such as the ‘hyporeactive
immobility’ normally seen after the defeat, but instead,
continued expressing a strong ‘hypereactive immobility’
(freezing) along with a clear exophthalmos. Therefore, it
is quite questionable whether the increased Fos expression in the ventrolateral PAG, observed in the present
experiments, should be correlated with passive emotional coping responses. Here, it is noteworthy that a
similar increased Fos expression in the ventrolateral
PAG has been reported in rats under pentobarbital
anesthesia, and which had received chemical stimulation
in the dorsoloteral PAG [53]. Overall, the evidence
supports the idea that the increased Fos expression in
the ventrolateral PAG, observed in our experiments,
should be viewed as integrating the general pattern of
PAG mobilization in response to the presence of a live
predator. Although quite speculative, it is plausible to
believe that exposure to the predator may induce
increased Fos expression in inhibitory interneurons of
the ventrolateral PAG. Indeed, many authors have
proposed that, during exposure to environmental
threats, the dorsal PAG may inhibit the ventrolateral
PAG, and some support for this comes from in vitro
studies with PAG slices (see [7]).
25
According to the present results, exposure to predator
also induced increased Fos expression in the ventromedial region of the PAG, where a substantial number of
immunoreactive cells were found in the supraoculomotor region, as well as in the dorsal raphe and laterodorsal tegmental nuclei. In contrast to the present
findings, however, in our previous study examining the
pattern of Fos expression in the PAG of rats exposed to
a natural predator [15], we have found only a sparse
number of Fos-immunoreactive cells in the ventromedial PAG. Since the experimental paradigms used for
exposing the prey to the predator were very similar in
both studies, we believe that this discrepancy may well
be related to the higher sensitivity of the anti-Fos serum
used in the present experiments. Corroborating this idea
is the fact that, in the present analysis, after confrontation with the predator, the number of Fos-immunoreactive cells found in the dorsomedial and dorsolateral
PAG was approximately three times higher than previously reported in our former study.
Several behavioral situations associated with high
levels of alertness (e.g. physically aversive foot shock
[39], swim stress, and insect predation) are likely to
upregulate Fos expression in the supraoculomotor
region and the laterodorsal tegmental nucleus. Although
a great deal still remains to be learned about the
connectivity and putative functional roles of the supraoculomotor region, a number of evidence associates
this region with the oculomotor control [59]. Conversely,
the laterodorsal tegmental nucleus, which provides
extensive cholinergic projections to the forebrain [12],
is thought to be particularly involved in modulating
global attentive states in response to novel stimuli [37].
As far as the dorsal raphe nucleus is concerned, it is
noteworthy that the nucleus also upregulates Fos
expression in response to other stressful situations
[6,31,35,38,54]. To our knowledge, the particular roles
played by the dorsal raphe nucleus during the encounter
with a predator have never been previously addressed.
However, for the present context, it is important to
know that this nucleus appears to mediate, at least
partly, the behavioral depression occurring in response
to uncontrollable stressors (i.e. inescapable tailshock)
[29,30], which may somehow be compared with our
experimental paradigm where the animals had no
chance of escaping from the predator.
Quite strikingly, an opposite pattern of PAG activation was seen in animals that performed insect predation, thus, corroborating the idea that predatory
behavior appears to be favored whenever the neural
circuitry underlying defensive responses is not particularly activated (see Section 1). In sharp contrast to what
was found for the animals confronted with a live
predator, predatory hunting is associated with increased
Fos expression predominantly in the rostral two thirds
of the lateral PAG, as well as, to a lesser degree, in the
26
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
supraoculomotor region and the laterodorsal tegmental
nucleus. In these experiments, the majority of the Fosimmunoreactive cells were found at the levels of the
oculomotor nucleus, where the labeled neurons tended
to be clearly clustered in the outer half of the lateral
PAG.
The present observations corroborate the idea that
insect hunting performed by rats retains the basic
behavioral patterns observed in other forms of predatory behavior in mammals [26]. During insect predation,
the animals first tried to locate the prey by sniffing
vigorously around the cage, reflecting, perhaps, a vivid
foraging behavior; and they eventually caught the
roaches, performing the predatory attack itself. Unfortunately, at the moment, it is not possible to ascertain if
the pattern of PAG activation observed during the
predatory hunting is associated with both foraging and
predatory attack, or just one of these behavioral
responses. According to preliminary data from our
laboratory, increased Fos expression in this same region
of the lateral PAG was also found after foraging
behavior of animals that were allowed neither to predate
insects nor to consume regular chow.
In contrast to the PAG regions mobilized in response
to confrontation with the predator, the rostral twothirds of the lateral PAG do not receive inputs from
elements of the medial hypothalamic defensive system.
However, similar to several other PAG sites, the portion
of lateral PAG mobilized during insect predation is also
substantially targeted by a number of medial prefrontal
fields (e.g. prelimbic and anterior cigulate areas) [27],
which, at the top of the perception-action cycle, are
thought to play a critical role in the behavioral planning
[28,36]. Moreover, this region of the lateral PAG is also
known to receive moderate inputs from the medial part
of the central amygdalar nucleus [49], where we have
also observed increased Fos expression in response to
predatory hunting [23]. Here, it is noteworthy that the
central amygdalar nucleus appears to play a key role in
the integration of the hedonic value of the aliments [44];
and the nucleus is also known to be involved in the
acquisition of conditioned approach responses to palatable food [32], reflecting, perhaps, its potential role in
controlling foraging behavior. Furthermore, previous
anterograde tract-tracing studies have shown that this
particular region of the lateral PAG, mobilized during
insect predation, also receives massive inputs from the
retrochiasmatic area [48] and the anterior part of the
ventromedial hypothalamic nucleus [16]. While the
putative functional roles of this latter hypothalamic
region remain yet to be investigated, there is experimental evidence indicating that the retrochiasmatic area
is intimately related to the hypothalamic circuitry that
regulates feeding [60]. Therefore, in addition to the
medial prefrontal cortex, the region of the lateral PAG
activated during predatory hunting also appears to be
influenced by prosencephalic sites seemingly involved in
regulating foraging and feeding behaviors. Although
feeding and predatory behaviors are likely to be
subserved by distinct neural systems (see [5]), it has
been shown that food-deprivation facilitates the expression of predatory behavior [58].
The region of the lateral PAG here described as
upregulating Fos expression in response to predatory
hunting also receives substantial somatossensory input
from the cervical levels of the spinal cord and the
laminar spinal trigeminal nucleus, thought to convey
nociceptive and thermoreceptive information, respectively, from the region of the forelimbs and face [11].
Consistent with the anatomy, functional studies have
indicated that physical stressors, such as cutaneuos pain,
evoke Fos expression within the lateral PAG [35,39].
Considering the experimental paradigm used for insect
predation in our study, it is quite unlikely that the
increased Fos expression in the lateral PAG, observed in
the present experiments, is correlated with any physical
injury occurred during confrontation with the roaches.
Nevertheless, it is important to note that stimulation of
the lateral PAG has been classically associated with
increased somatomotor activity, sympathoexcitation,
and non-opioid-mediated analgesia [3,20,24,41,43],
which, in the context of the predatory hunting, may
represent a very adaptive defensive mechanism against
potential harms that may occur during the confrontation with the prey.
The efferent projections of this particular region of
the lateral PAG mobilized during predatory hunting
have yet to be critically analyzed. However, taken the
rostral region of the lateral PAG as a whole, previous
anatomical studies have shown that it provides extensive
descending projections to a number of brainstem sites
critically involved in controlling somatomotor and
autonomic responses [14]. Moreover, the rostral region
of the lateral PAG has also been shown to project
extensively to the lateral hypothalamic area, perhaps
reflecting its potential roles in the control of generalized
arousal and sensorimotor integration [13]. Notably, this
hypothalamic projection also targets a particular region
in the lateral hypothalamus classically known to elicit
predatory attack when electrically stimulated (see [50]).
In conclusion, the PAG integrates a vast array of
limbic information, likely to reflect the animal’s motivational status; and we have presently shown that behaviors associated with opposite motivational drives, such
as defensive and predatory behaviors, evoke a quite
distinct PAG activation pattern. On the behavioral side,
it is well established that the PAG is critical for the
expression of defensive responses, and, considering the
present findings, it will be important to investigate how
the PAG contributes to the expression of the predatory
behavior, as well. Therefore, in the future, we will be
able to determine to what extent the PAG influences the
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
switch of goal-oriented behaviors in response to different motivational drives.
Acknowledgements
This work was supported by FAPESP grant 92/050968 and CNPq fellowship 300562/93-4 to N.S.C., FAPESP
fellowship 00/14637-0 to E.R.R.B., and FAPESP fellowship 97/10490-0 to E.C. The authors wish to express
appreciation to Ana M.P. Campos for expert technical
assistance, and to Dr Marcus Vinicius C. Baldo for
preparing the statistical analysis.
[14]
[15]
[16]
[17]
[18]
References
[1] Bandler R, Carrive P, Depaulis A. Emerging principles of
organization of the midbrain periaqueductal gray matter. In:
Depaulis A, Bandler R, editors. The Midbrain Periaqueductal
Gray Matter: Functional, Anatomical, and Neurochemical Organization. New York: Plenum Press, 1991:1 /8.
[2] Bandler R, Depaulis A, Vergnes M. Identification of midbrain
neurones mediating defensive behavior in the rat by microinjections of excitatory amino acids. Behav Brain Res 1985;15:107 /19.
[3] Bandler R, Depaulis A. Midbrain periaqueductal gray control of
defensive behavior in the cat and rat. In: Depaulis A, Bandler R,
editors. The Midbrain Periaqueductal Gray Matter: Functional,
Anatomical, and Neurochemical Organization. New York: Plenum Press, 1991:175 /98.
[4] Bandler R, Shipley MT. Columnar organization of the midbrain
periaqueductal gray: modules for emotional expression. Trends
Neurosci 1994;17:379 /89.
[5] Bandler RJ, Chi CC, Flynn JP. Biting attack elicited by
stimulation of the ventral midbrain tegmentum of cats. Science
1972;177:364 /6.
[6] Beck C, Fibiger H. Conditioned fear-induced changes in behavior
and in the expression of the intermediate early gene c-fos with and
without diazepam pretreatment. J Neurosci 1995;15:709 /20.
[7] Behbehani MM. Functional characteristics of the midbrain
periaqueductal gray. Prog Neurobiol 1995;46:575 /605.
[8] Berntson GG. Blockade and release of hipothalamically and
naturally elicited aggressive behaviors in cats following midbrain
lesions. J Comp Physiol Psychol 1972;81:541 /54.
[9] Berntson GG. Attack, grooming and threat elicited by stimulation
of the pontine tegmentum in cats. Physiol Behav 1973;11:81 /7.
[10] Besson JM, Fardin V, Oliveras JL. Analgesia produced by
stimulation of the periaqueductal gray matter: true antinoceptive
effects versus stress effects. In: Depaulis A, Bandler R, editors.
The Midbrain Periaqueductal Gray Matter: Functional, Anatomical, and Neurochemical Organization. New York: Plenum
Press, 1991:121 /38.
[11] Blomqvist A, Craig AD. Organization of spinal and trigeminal
input to the PAG. In: Depaulis A, Bandler R, editors. The
Midbrain Periaqueductal Gray Matter: Functional, Anatomical,
and Neurochemical Organization. New York: Plenum Press,
1991:345 /63.
[12] Bolton RF, Cornwall J, Phillipson OT. Collateral axons of
cholinergic pontine neurones projecting to midline, mediodorsal,
and parafascicular thalamic nuclei in the rat. J Chem Neuroanat
1993;6:101 /14.
[13] Cameron AA, Khan IA, Westlund KN, Cliffer KD, Willis WD.
The efferent projections of the periaqueductal gray in rat: a
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
27
Phaseolus vulgaris-leucoagglutinin study. I. Ascending projections.
J Comp Neurol 1995;351:585 /601.
Cameron AA, Khan IA, Westlund KN, Willis WD. The efferent
projections of the periaqueductal gray in rat: a Phaseolus vulgarisleucoagglutinin study. II. Descending projections. J Comp Neurol
1995;351:585 /601.
Canteras NS, Goto M. Fos-like immunoreactivity in the periaqueductal gray of rats exposed to a natural predator. Neuroreport
1999;10:413 /8.
Canteras NS, Simerly RB, Swanson LW. Organization of
projections from the ventromedial nucleus of the hypothalamus:
a Phaseolus vulgaris-leucoagglutinin study in the rat. J Comp
Neurol 1994;348:41 /79.
Canteras NS, Swanson LW. The dorsal premammillary nucleus:
an unusual component of the mammillary body. Proc Natl Acad
Sci USA 1992;89:10089 /93.
Canteras NS. The medial hypothalamic defensive system: hodological organization and functional implications. Pharmacol
Biochem Behav 2002;71:481 /91.
Carrive P, Leung P, Harris J, Paxinos G. Conditioned fear to
context is associated with increased fos expression in the caudal
ventrolateral region of the midbrain periaqueductal gray. Neuroscience 1997;78:165 /77.
Carrive P. Functional organization of PAG neurons controlling
regional vascular beds. In: Depaulis A, Bandler R, editors. The
Midbrain Periaqueductal Gray Matter: Functional, Anatomical,
and Neurochemical Organization. New York: Plenum Press,
1991:67 /100.
Carrive P. The periaqueductal gray and defensive behavior:
functional representation and neural organization. Behav Brain
Res 1993;58:27 /47.
Chaurand JP, Vergnes M, Karli P. Substance grise centrale du
mésencéphale et comportement d’agression interspécifique du rat.
Physiol Behav 1972;9:475 /81.
Comoli E, Canteras NS. Characterization of the neural systems
mobilized during predatory behavior in rats. Soc Neurosci Abstr
1998;24:678.
Depaulis A, Keay KA, Bandler R. Longitudinal neuronal
organization of defensive reactions in the midbrain periaqueductal gray region in the rat. Exp Brain Res 1992;90:307 /18.
Di Scala G, Schmitt P, Karli P. Flight induced by infusion of
bicuculline methiodide into periventricular structures. Brain Res
1984;309:199 /208.
Eisenberg JF, Leyhausen P. The phylogenesis of predatory
behavior in mammals. Z Tierpsychol 1972;30:59 /93.
Floyd NS, Price JL, Ferry AT, Keay KA, Bandler R. Orbitomedial prefrontal cortical projections to distinct longitudinal columns of the periaqueductal gray in the rat. J Comp Neurol
2000;422:556 /78.
Fuster JM. Executive frontal functions. Exp Brain Res
2000;133:66 /77.
Grahn RE, Maswood S, McQueen MB, Watkins LR, Maier SF.
Opioid-dependent effects of inescapable shock on escape behavior
and conditioned fear responding are mediated by the dorsal raphe
nucleus. Behav Brain Res 1999;99:153 /67.
Grahn RE, Watkins LR, Maier SF. Impaired escape performance
and enhanced conditioned fear in rats following exposure to an
uncontrollable stressor are mediate by glutamate and nitric oxide
in the dorsal raphe nucleus. Behav Brain Res 2000;111:33 /41.
Grahn RE, Will MJ, Hammack SE, Maswood S, McQueen MB,
Watkins LR, Maier SF. Activation of serotonin-immunoreactive
cells in the dorsal raphe nucleus in rats exposed to an uncontrollable stressor. Brain Res 1999;826:35 /43.
Hitchcott PK, Phillips GD. Double dissociation of the behavioral
effects of R (/) 7-OH-DPAT infusions in the central and
basolateral amygdala nuclei upon Pavlovian and instrumental
28
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
E. Comoli et al. / Behavioural Brain Research 138 (2003) 17 /28
conditioned
appetitive
behaviors.
Psychopharmacology
1998;140:458 /69.
Hsu SM, Raine L. Protein A, avidin and biotin in immunohistochemistry. J Histochem Cytochem 1981;29:1349 /53.
Hunsperger RW. Affektreaktionen auf elektrische Reizung im
Hirnstamm der Katze Helv. Physiol Pharmacol Acta 1956;14:70 /
92.
Keay KA, Bandler R. Parallel circuits mediating distinct emotional coping reactions to different types of stress. Neurosci
Biobehav Rev 2001;25:669 /78.
Kolb B. Functions of the frontal cortex of the rat: a comparative
review. Brain Res Rev 1984;8:65 /98.
Koyama Y, Jodo E, Kayama Y. Sensory responsiveness of
‘broad-spike’ neurons in the laterodorsal tegmental nucleus, locus
coeruleus and dorsal raphe of awake rats: Implications for
cholinergic and monoaminergic neuron-specific responses. Neuroscience 1994;63:1021 /31.
Krukoff T, Khalili P. Stress-induced activation of nitric oxideproducing neurons in the rat brain. J Comp Neurol
1997;377:509 /19.
Li HY, Sawchenko PE. Hypothalamic effector neurons and
extended circuitries activated in ‘neurogenic’ stress: a comparisson
of footshock effects exerted acutely, chronically, and in animals
with controlled glucocorticoid levels. J Comp Neurol
1998;393:244 /66.
Lonstein JS, Stern JM. Role of the midbrain periaqueductal gray
in maternal nurturance and aggression: c-fos and electrolytic
lesion studies in lactating rats. J Neurosci 1997;17:3364 /78.
Lovick TA. Interactions between descending pathways from the
dorsal and ventrolateral periaqueductal gray matter in the rat. In:
Depaulis A, Bandler R, editors. The Midbrain Periaqueductal
Gray Matter: Functional, Anatomical, and Neurochemical Organization. New York: Plenum Press, 1991:101 /20.
Morgan JI, Curran T. Stimulus-transcription coupling in the
nervous system: involvement of inducible proto-oncogenes fos
and jun. Annu Rev Neurosci 1991;14:421 /51.
Morgan MM. Differences in antinociception evoked from dorsal
and ventral regions of the caudal periaqueductal gray matter. In:
Depaulis A, Bandler R, editors. The Midbrain Periaqueductal
Gray Matter: Functional, Anatomical, and Neurochemical Organization. New York: Plenum Press, 1991:139 /50.
Pomonis JD, Jewett DC, Kotz CM, Briggs JE, Billington CJ,
Levine AS. Sucrose consumption increases naloxone-induced cFos immunoreactivity in limbic forebrain. Am J Physiol
2000;278:R712 /9.
Proshansky E, Bandler RJ, Flynn JP. Elimination of hypothalamically elicited biting attack by unilateral lesion of the ventral
midbrain tegmentum of cats. Brain Res 1974;77:309 /13.
Rebouças RCR, Schmidek WR. Handling and isolation in three
strains of rats affect open field, exploration, hoarding and
predation. Physiol Behav 1997;62:1159 /64.
Redgrave P, Dean P. Does the PAG learn about emergencies from
the superior colliculus. In: Depaulis A, Bandler R, editors. The
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
midbrain periaqueductal gray matter: functional, anatomical, and
neurochemical organization. New York: Plenum Press, 1991:199 /
209.
Ribeiro-Barbosa ER, Skorupa AL, Cipolla-Neto J, Canteras NS.
Projections of the basal retrochiasmatic area: a neural site
involved in the photic control of pineal metabolism. Brain Res
1999;839:35 /40.
Rizvi TA, Ennis M, Behbehani MM, Shipley MT. Connections
between the central nucleus of the amygdala and the midbrain
periaqueductal gray: topography and reciprocity. J Comp Neurol
1991;303:121 /31.
Roberts WW. Are hypothalamic motivational mechanisms functionally and anatomically specific. Brain Behav Evol 1969;2:317 /
42.
Ruiz-Torner A, Olucha-Bordonau F, Valverde-Navarro AA,
Martı́nez-Soriano F. The chemical architecture of the rat’s
periaqueductal gray based on acetylcholinesterase histochemistry:
a quantitative and qualitative study. J Chem Neuroanat
2001;21:295 /312.
Sakuma Y, Paff DW. Facilitation of female reproductive behavior
from mesencephalic central gray in the rat. Am J Physiol
1979;237:R278 /84.
Sandkühler J, Herdegen T. Distinct patterns of activated neurons
throughout the rat midbrain periaqueductal gray induced by
chemical stimulation within its subdivisions. J Comp Neurol
1995;357:546 /53.
Senba M, Matsunaga K, Tohyama M, Noguchi K. Stress-induced
c-fos expression in the rat brain: activation mechanism of
sympathetic pathway. Brain Res Bull 1993;31:329 /44.
Sheard MH, Flynn JP. Facilitation of attack behavior by
stimulation of the midbrain of cats. Brain Res 1967;4:324 /33.
Sheng M, Greenberg ME. The regulation and function of c-fos
and other immediate early genes in the nervous system. Neuron
1990;4:477 /85.
Van Erp AMM, Kruk MR, Meelis W, Veening JG. Periaqueductal gray lesions do not affect grooming induced electrically in
the hypothalamic paraventricular area in the rat. Behav Brain Res
1993;59:95 /101.
Vergnes M. Déclenchement de réactions d’agression interspécifique après lésion amygdalienne chez le rat. Physiol Behav
1975;14:271 /6.
Vertes RP, Martin GF. Autoradiographic analysis of ascending
projections from the pontine and mesencephalic reticular formation and the median raphe nucleus in the rat. J Comp Neurol
1988;275:511 /41.
Watts AG, Sanchez-Watts G, Kelly AB. Distinct patterns of
neuropeptide gene expression in the lateral hypothalamic area and
arcuate nucleus are associated with dehydration-induced anorexia. J Neurosci 1999;19:6111 /21.
Weiss GF, Leibowitz SF. Efferent projections from the paraventricular nucleus mediating alpha2-noradrenergic feeding. Brain
Res 1985;347:225 /38.