- Email: [email protected]
Neurotransmitters regulating defensive rage behavior in the cat
Neuroscience and Biobehavioral Reviews, Vol. 21, No. 6, pp. 733–742, 1997 q 1997 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0149-7634/97/$32.00 + .00
Pergamon
PII: S0149-7634(96)00056-5
Neurotransmitters Regulating Defensive Rage Behavior in the Cat ALLAN SIEGEL, a ,* KRISTIE L. SCHUBERT*† AND MAJID B. SHAIKH* *Departments of Neurosciences and Psychiatry, New Jersey Medical School, Newark, NJ 07103, USA †Department of Biology, Kean College, Union, NJ 07083, USA
SIEGEL, A., K. L. SCHUBERT AND M. B. SHAIKH. Neurotransmitters regulating defensive rage behavior in the cat. NEUROSCI BIOBEHAV REV 21(6) 733–742, 1997.—This review summarizes recent findings of our laboratory that have been directed at: (1) identifying the neural circuits underlying the expression and modulation of defensive rage behavior in the cat and the neurotransmitters associated with these pathways; and (2) determining which components of the circuitry are affected by alcohol administration and which significantly alter the rage mechanism. The experiments described herein incorporated a number of converging methods, which include brain stimulation, behavioral pharmacology, immunocytochemistry, retrograde tract tracing and receptor binding. For behavioral pharmacological studies, monopolar electrodes and cannula-electrodes were implanted into selected regions along the limbic–midbrain axis for electrical stimulation and local microinfusion of drugs. The findings demonstrated: (1) a direct pathway from the anterior medial hypothalamus to the dorsal periaqueductal gray (PAG) over which this response is mediated. This pathway utilizes excitatory amino acids that act upon NMDA receptors within the midbrain PAG; (2) that the region of the dorsal PAG, from which defensive rage could be elicited, receives other inputs from the basal amygdala that facilitate this response by acting upon NMDA receptors; (3) a pathway from the medial amygdala to the medial hypothalamus that also facilitates defensive rage and whose functions are mediated by substance P receptors within the medial hypothalamus; (4) that the PAG also receives enkephalinergic inputs from the central nucleus of amygdala, which act upon m receptors, and which powerfully suppress defensive rage; and (5) that recent findings reveal that ethanol administration facilitates defensive rage by virtue of its interactions with the medial hypothalamus, its descending projection to the PAG, and possibly with NMDA receptors within this pathway. q 1997 Elsevier Science Ltd. Defensive rage behavior Hypothalamus Periaqueductal gray Amygdala Excitatory amino acids Enkephalins Receptor binding Immunocytochemistry Ethanol
Substance P
midbrain PAG. Such structures are said to be ‘modulators’ of defensive rage behavior. Limbic structures may either suppress or facilitate the occurrence of defensive rage, but it is important to note that the rage response cannot be elicited by electrical stimulation of any of these limbic structures.
STUDIES conducted over the past three decades have established that the neural substrates for defensive rage behavior in the cat include the hypothalamus, midbrain periaqueductal gray (PAG) as well as limbic nuclei such as the amygdala, hippocampal formation and septal area (18– 21,23–26,29,30). The general functions of each of these regions with respect to rage and aggression have now been identified. Two different classes of neuronal regions can be distinguished. The first, consisting primarily of the medial hypothalamus and midbrain PAG, represent structures from which defensive rage can be elicited by electrical stimulation. Such structures are said to be the ‘effector’ regions for the expression of defensive rage behavior. The second class of structures consists primarily of the limbic system. The major structures of this group include the amygdala, hippocampal formation, septal area, prefrontal cortex, cingulate cortex and related regions of the basal forebrain, and influence the rage mechanism by virtue of their actions upon either the medial hypothalamus or
1.
DEFENSIVE RAGE BEHAVIOR
This form of aggressive behavior occurs under natural conditions when the cat is threatened by another animal of the same or different species. The nature of the external threat could come from different sources. A typical example is the presence of another cat that is competing for its territory (11). The response is characterized by various sympathetic signs, such as pupillary dilatation, piloerection, ear retraction, salivation, unsheathing of the claws, hissing and paw strike (8,11,15,30). As noted above, this response can be elicited by electrical stimulation of the medial
a Correspondence should be addressed to: Allan Siegel, Ph.D., Department of Neurosciences, Medical Science Building, Rm H-512, New Jersey Medical School, 185 South Orange Avenue, Newark, NJ 07103, USA.
733
734
SIEGEL ET AL.
hypothalamus or the midbrain PAG. The section below summarizes the regions from which defensive rage can be elicited as well as the primary descending pathways over which this response is mediated. 2. REGIONS ASSOCIATED WITH THE EXPRESSION OF DEFENSIVE RAGE AND THEIR DESCENDING PATHWAYS
2.1. Defensive rage sites Defensive rage behavior can be elicited under laboratory conditions by electrical stimulation principally of the medial hypothalamus or midbrain PAG. Within the medial hypothalamus, sites from which defensive rage behavior can be elicited extend throughout its rostro-caudal extent. The most notable sites at which this response can be elicited include the region of the ventromedial nucleus, the dorsomedial hypothalamic area, and the anterior medial hypothalamus. Defensive rage behavior cannot be elicited from regions of hypothalamus other than its medial aspect. Within the midbrain PAG, defensive rage can be easily elicited from is dorsal aspect, and in particular, at the level
FIG. 1. Maps indicate sites within the: (A) midbrain periaqueductal gray and (B) medial hypothalamus from which defensive rage behavior was elicited by electrical stimulation in the studies described in this review. CI, internal capsule; CS, superior colliculus; Fx, fornix; HL, lateral hypothalamus; HVM, ventromedial hypothalamus; PAG, midbrain periaqueductal gray; Ped, cerebral peduncle; RE, nucleus reuniens; RN, red nucleus; SN, substantia nigra; TO, optic tract.
of the rostral half of the neuropil (2,18,19,21,22). At more caudal levels of the PAG, defensive rage is not readily seen. In its place, one can identify sites that produce howling as well as flight behavior (2). That defensive rage is limited to the dorsal aspect of the PAG is further evidenced by the fact that stimulation of its ventral aspect can only produce elements of predatory attack behavior, a form of aggression that is characterized by the initial stalking followed by the biting of the back of a rat’s neck. Maps indicating the loci of sites within the hypothalamus and PAG, from which defensive rage is typically elicited by electrical stimulation, are shown in Fig. 1. 2.2. Descending pathways mediating defensive rage behavior The data described above point to the importance of the medial hypothalamus and dorsal PAG as the key regions of the brain involved in the integration of defensive rage behavior. A question of central importance concerns the descending neuroanatomical circuits mediating the expression of defensive rage behavior from the hypothalamus. A series of studies conducted in our laboratory utilized anterograde and retrograde tracing methods to answer this question (7,12,17,18). With respect to the projection from the hypothalamus to the PAG, the initial studies of Fuchs et al. (7) utilized the combined methods of tritiated amino acid radioautography and 14C-2-deoxyglucose autoradiography to trace the pathways from sites in the medial hypothalamus to the brainstem. The findings from this study indicated that neurons arising from the ventromedial nucleus associated with defensive rage behavior generally project in a rostral direction to the anterior medial hypothalamus from which this response can also be elicited. Cells located in the region that extends rostrally from the dorsomedial hypothalamus to the anteromedial–preoptic region give rise to long, descending axons that project to that part of the dorsal aspect of the PAG which is also associated with the expression of defensive rage. This finding was basically replicated recently with retrograde tracing methods following the placement of Fluoro-Gold into defensive rage sites within the PAG (12,17). These studies showed that the larger majority of descending neurons that project to identifiable rage sites in the PAG arise from cells along a band extending rostrally and somewhat dorsally from the ventromedial nucleus with relatively few neurons located in the ventromedial nucleus, per se. This pathway is summarized in Fig. 2. It should be noted that it has recently been shown that the PAG is organized in a series of columns that extend along its longitudinal axis (2). One such functional column mediates defensive rage and threat behavior and is situated in the dorsal aspect of the rostral two-thirds of the PAG. As indicated above, significant inputs to the PAG from the medial hypothalamus in association with defensive rage behavior terminate within the rostral half of this column (7). Thus, it would seem that the rostral half of this column is critical for the integration of descending fibers from the medial hypothalamus required for the expression of defensive rage behavior. The efferent pathways of the PAG that mediate defensive rage behavior project primarily in a descending direction,
NEUROTRANSMITTERS AND DEFENSIVE RAGE BEHAVIOR
FIG. 2. Diagram illustrates the primary projections of the medial hypothalamus associated with defensive rage behavior (i.e. referred to in this figure and Fig. 3 as ‘affective defense’) in the cat. Note that the projections of the medial hypothalamus to the dorsal aspect of the midbrain periaqueductal gray (called central gray in this figure and in Fig. 3) arise principally from the rostral half of the hypothalamus. This illustration also indicates that fibers associated with defensive rage behavior that arise from the ventromedial nucleus of the hypothalamus generally project to more rostral levels of the medial hypothalamus. (From Siegel and Pott (25).)
735
FIG. 3. Diagram illustrates the principal outputs of the dorsal aspect of the PAG from which defensive rage can be elicited. Fibers arising from the periaqueductal gray in association with defensive rage behavior are distributed to a number of regions in the lower brainstem that are important for the expression of this response. One such region includes the motor nucleus of the trigeminal nerve, which is essential for jaw opening and for the occurrence of ‘hissing’ and other forms of vocalization that are integral to the defensive rage response. A second projection includes the distribution of fibers to autonomic regions of the lower brainstem, which help to mediate the sympathetic components of this response. In addition, a third, ascending component supplies the posterior medial hypothalamus. This projection likely serves as a ‘Positive feedback’ to the hypothalamus, thus increasing the duration of the defensive rage response. (Adapted from Shaikh et al. (18).)
3. FUNCTIONS OF THE PATHWAY FROM THE MEDIAL HYPOTHALAMUS although some ascending fibers of significance have also been identified (18). The primary descending projections innervate parts of the lower brainstem tegmentum, locus ceruleus, and motor and sensory nuclei of the trigeminal complex. In this manner, the descending projections of the PAG innervate both autonomic and somatomotor nuclei of the lower brainstem, which provide the substrate by which an integrated defensive rage response can be elicited from the PAG. In fact, the PAG represents the most caudal region of the brainstem where such integration can take place. The ascending projection of the PAG is also of interest. Axons from the dorsal PAG supply parts of the posterior half of the medial hypothalamus from which defensive rage could also be elicited. This connection provides the basis for a positive feedback circuit between the medial hypothalamus and the dorsal PAG. Such a circuit is of potential important survival value to the animal in that it provides the basis for a response of relatively long duration that is required during combat. A diagram summarizing the primary projections of the dorsal PAG from which defensive rage can be elicited is shown in Fig. 3.
TO THE PAG ARE MEDIATED BY EXCITATORY AMINO ACIDS
As noted above, the pathway from the medial hypothalamus to the dorsal PAG links two regions of the brain that are central to the integration of defensive rage behavior. Therefore, a primary objective of our research efforts was to attempt to identify the likely neurotransmitter associated with this pathway. The strategy was to employ a variety of converging methodological approaches. These included brain stimulation methods coupled with pharmacological manipulations, immunocytochemistry coupled with retrograde tracing procedures, and receptor binding. Recent studies conducted in the rat demonstrated the presence of high concentrations of glutamate and aspartate within the PAG (1,3). Therefore, we sought to test the hypothesis that excitatory amino acids (EAA) mediate the rage mechanism over the pathway from the medial hypothalamus to the PAG. 3.1. Evidence derived from neuroanatomical– immunocytochemical studies In these studies, the following procedures were implemented. Cannula electrodes were implanted into sites within
736
the PAG from which defensive rage behavior was elicited by electrical or chemical (i.e. homocysteic acid) stimulation. Then, Fluoro-Gold (4–8%) was microinjected into the PAG and the animals were sacrificed 7–14 days postinjection. Alternate sections were processed for immunocy-
SIEGEL ET AL.
tochemical labeling of glutamate or aspartate neurons within the forebrain and brainstem (12,17). The objective of this approach was to be able to determine whether or not neurons in the medial hypothalamus that are labeled for Fluoro-Gold are also labeled for glutamate or aspartate as well. Such data would support the basic hypothesis that the projection from the medial hypothalamus to the PAG that mediates defensive rage behavior is mediated by EAA. The results of these studies are summarized in Fig. 4. The loci of individual microinjections of Fluoro-Gold placed into the dorsal PAG are depicted in Fig. 4(a). As shown in Fig. 4(b), cells labeled for both glutamate (or aspartate) and Fluoro-Gold were located throughout the rostral half of the medial hypothalamus, extending from the level of the ventromedial nucleus to the preoptic region. The largest quantities of double-labeled neurons were situated in the dorsomedial and anteromedial hypothalamus, rostral to the level of the ventromedial nucleus. A few double-labeled neurons were also identified in the ventromedial nucleus. Fig. 4(b) also reveals that many other cells were labeled for either Fluoro-Gold or glutamate (or aspartate) but not for both, indicating that such neurons either project to the PAG but presumably do not utilize EAA as a neurotransmitter, or that they may utilize glutamate or aspartate but do not project to the dorsal PAG. Additional evidence supporting the view that neurons expressing aspartate project to the dorsal PAG is provided in Fig. 4(c) as well, which indicates the presence of labeled aspartatergic axons and preterminals within the dorsal PAG. Collectively, these findings provide direct support for the hypothesis of a monosynaptic pathway from the rostral half of the medial hypothalamus to the PAG whose behavioral functions are mediated by EAA. However, it should be pointed out that we have only examined the inputs from the medial hypothalamus to the functional column of cells situated within the rostral half of the dorsal PAG. It remains to be determined whether the inputs to more caudal aspects of the dorsal PAG column associated with threat and defensive behavior are also mediated by EAA that act upon NMDA receptors. 3.2. Evidence from behavioral pharmacological experiments The hypothesis of a pathway from the medial hypothalamus to the dorsal PAG which utilizes EAA to mediate the expression of defensive rage behavior was tested directly in the following experiments (12,17). A cannula electrode was initially implanted into a site within the dorsal PAG from
FIG. 4. (a) Section illustrates the loci of Fluoro-Gold injections at sites within the PAG from which stimulation produced defensive rage behavior. (b) Maps indicate the presence of neurons that are single-labeled for glutamate (or aspartate) (closed circles), Fluoro-Gold (open circles), or doublelabeled for glutamate (or aspartate) and Fluoro-Gold (closed triangles). Note that the highest concentration of double-labeled neurons is located in the dorsomedial and anteromedial hypothalamus, the regions where defensive rage behavior are typically elicited. Abbreviations: alld, dorsal hypothalamus; Fx, fornix; HL, lateral hypothalamus; Hp, posterior hypothalamus. (c) Presence of aspartate labeled axons and preterminals within the dorsal aspect of the PAG—the region from which stimulation typically elicits defensive rage behavior. HVM, ventromedial hypothalamic nucleus; PAG, midbrain periaqueductal gray; RE, nucleus reuniens; RN, red nucleus. (Maps represent combined data from studies of Lu et al. (12) and unpublished observation from our laboratory.)
NEUROTRANSMITTERS AND DEFENSIVE RAGE BEHAVIOR 100
2.0 0.5 0.1 4.4 BL
80
Change in latency (%)
which defensive rage could be elicited and where drugs could be infused at a later time. Then, a stimulating electrode was implanted into a site within the medial hypothalamus from which defensive rage could also be elicited. The experimental paradigm involved the application of paired trials of (a) a single stimulation of the PAG from which defensive rage was elicited and (b) a dual stimulation of the PAG and medial hypothalamus (at current levels below threshold for the elicitation of defensive rage behavior). The effects of dual stimulation were then compared to that of single stimulation of the PAG alone. When significant modulation of the hypothalamus was determined, the current was maintained constant. In order to test the basic hypothesis, we first demonstrated that, prior to drug infusion, dual stimulation involving the medial hypothalamus significantly facilitated the occurrence of defensive rage behavior elicited from the PAG. Then, the procedures of dual stimulation were repeated following the administration of different doses of: (a) a nonselective EAA receptor antagonist (kynurenic acid); and (b) a selective NMDA antagonist [dl-2-amino-7-phosphonoheptanoic acid; (AP-7)]; as well as a single dose of the non-NMDA antagonist CNQX, which was equivalent to the highest dose of AP-7 administered. Following the delivery of either the non-selective antagonist, kynurenic acid, or the selective NMDA antagonist, AP7, the facilitative effects of medial hypothalamic stimulation were blocked in a dose- and time-dependent manner (Figs 5 and 6). In contrast, infusion of the non-NMDA antagonist CNQX did not modify the facilitative effects of dual stimulation. These results provide direct evidence that the pathway mediating defensive rage behavior from the medial hypothalamus to the PAG utilizes EAA that act through NMDA receptors. Most recently, the importance of NMDA receptors within the dorsal PAG in mediating defensive rage behavior was tested in yet another way (17). In this experiment, infusion of the AP-7 into sites within the PAG from which defensive rage could be elicited blocked defensive rage behavior that was elicited by single stimulation of the medial hypothalamus. Thus, when taken together, the results of these experiments involving two different experimental paradigms provide converging evidence in support of the role played by NMDA receptors within the dorsal PAG in the regulation of defensive rage behavior.
737
60 40 20 0 -20 -40 -60 -80 0
60
120
180
FIG. 5. Graphs indicate that microinjections of kynurenic acid (kyn) into PAG sites from which defensive rage was elicited block the facilitatory effects of medial hypothalamic stimulation upon this response elicited from the PAG. A higher dose of atropine (atr) did not alter the facilitatory effects of medial hypothalamic stimulation. For this and Fig. 6: BL, theoretical baseline indicating the position on the graph where dual stimulation of the medial hypothalamus had no effect in altering PAG response latencies as determined from single stimulation of the PAG. Drug doses are in nanomolar values. Vertical bars indicate SEM for this and subsequent figures. (From Lu et al. (12).)
receptors by virtue of the presence of positive NMDA-R1 immunoreactivity in the dorsal aspect of the PAG. As noted above, the dorsal PAG receives the inputs from cell groups in the medial hypothalamus associated with the expression of defensive rage. Therefore, it is of interest to point out that further analysis revealed that such immunoreactivity was not present in the ventral PAG, which does not receive afferents from the medial hypothalamus. 4.
AMYGDALOID MODULATION OF DEFENSIVE RAGE BEHAVIOR
As indicated previously, the regions considered above— the medial hypothalamus and midbrain PAG—constitute effector regions for the expression of defensive rage behavior. Other regions also play an important role in the regulation of defensive rage behavior. They do so by modulating the defensive rage mechanism at the level of the hypothalamus or PAG. The following review briefly summarizes the results of recent studies concerning the
Change in latency (%)
In the previous section, pharmacological methods were employed to demonstrate the role of NMDA receptors in the PAG in regulating defensive rage behavior. In this section, we employed receptor histochemical procedures to demonstrate the presence of NMDA receptors within the dorsal aspect of the PAG. In brief, brain sections were reincubated with 10% normal goat serum in phosphate-buffered saline and then reacted with a purified monoclonal antibody against the NMDA-R1 receptor subtype at a 1:1000 dilution in phosphate buffer. Sections were then washed and processed with a species-specific Vectastain ABC Elite immunoperoxidase kit (Burlingame, CA), after which sections were stained with DAB in phosphate-buffered saline. The results, shown in Fig. 7, demonstrate the existence of glutamate
1440
Postinjection time (min)
50
3.3. Further evidence for the presence of NMDA receptors in the dorsal PAG
kyn kyn kyn atr
2.0 nM AP7 0.5 nM AP7 0.1 nM AP7 4nM CNQX BL
40 30 20 10 0 -10 -20 0
45
90
135
180
Postinjection time (min) FIG. 6. Graphs indicate that infusion of AP-7 into sites within the PAG from which defensive rage had been elicited blocks the facilitatory effects of medial hypothalamic stimulation upon this response. Note that a higher dose of CNQX had little effect upon medial hypothalamic facilitation of this response (From Lu et al. (12).)
738
mechanisms underlying amygdaloid modulation of defensive rage behavior. The basic paradigm utilized to identify the modulating properties, the origins and distributions of the pathways, and likely neurotransmitters of the amygdaloid regions in question, was identical to the dual stimulation procedure described earlier for analyzing the modulating effects of medial hypothalamic stimulation upon PAG-elicited defensive rage. To date, three regions of the amygdala have been studied. Two of them—the medial nucleus and the basal complex—facilitate defensive rage, whereas a third region, the central (and adjoining lateral) nucleus suppresses defensive rage. 4.1. Regions that facilitate defensive rage behavior 4.1.1. The medial amygdala For testing the modulating effects of this structure, a cannula electrode was placed into a site within the medial hypothalamus from which defensive rage was elicited, and a stimulating electrode was lowered into the medial amygdala. Response latencies following paired trials of single (i.e. medial hypothalamus alone) and dual (i.e. medial nucleus þ medial hypothalamus) stimulation were compared both prior to and after drug administration (22). Prior to drug administration, dual stimulation produced a 30–40% decrease in response latencies relative to single stimulation of the medial hypothalamus alone. Then, the specific NK 1, non-peptide, substance P antagonist, CP 96,345, was microinjected into sites within the medial hypothalamus from which defensive rage was elicited in doses of 0.05–2.5 nmol/0.25 ml. The role of substance P was directly tested in this study for the following reasons: (1) substance P-positive neurons have been shown to be present in the medial amygdala in rodents (13,14,27,28); (2) the medial amygdala is known to project directly to the medial hypothalamus via the stria terminalis (6,10,30); and (3) substance P terminals have been identified in the medial hypothalamus (4). Drug administration significantly decreased medial amygdaloid facilitation in a dose- and time-dependent manner. As shown in Fig. 8, the highest dose used, 2.5 nmol of CP 96,345, decreased response facilitation by approximately 40% (p , 0.001). In a related study, Fluoro-Gold was microinjected into
SIEGEL ET AL.
medial hypothalamic sites from which defensive rage had been elicited. Following survival times of 5–6 days, animals were sacrificed, and alternate sections were processed for immunocytochemical labeling of substance P. Substantial quantities of neurons in the medial nucleus of amygdala were found to be double-labeled for both substance P and Fluoro-Gold (see Fig. 9). This finding provides additional support to the view that medial amygdaloid facilitation of defensive rage is mediated through substance P receptors within the ventromedial hypothalamus. 4.1.2. The basal complex The same paradigms as described above for characterizing medial amygdaloid facilitation of defensive rage behavior were applied to the analysis of basal amygdaloid modulation of this response. The basal amygdaloid complex is known to facilitate defensive rage behavior (5) and projects its axons to the PAG (9). Moreover, as described earlier in this paper, numerous glutamatergic and aspartatergic nerve terminals and NMDA receptors have been shown to be present in this neuropil as well. Therefore, the basic hypothesis tested in this study was that basal amygdaloid stimulation facilitates defensive rage behavior at the level of the PAG by acting through NMDA receptors. Dual stimulation of the basal amygdala and PAG (from which defensive rage was elicited) facilitated the occurrence of defensive rage behavior. Following administration of the selective NMDA antagonist, AP-7, into the PAG, basal amygdaloid facilitation of defensive rage was significantly suppressed by 19% over the first 60 min post-injection period (Fig. 10) (20). However, this effect was clearly less potent than that observed for other systems described above. This finding would suggest that while EAA contributes to the facilitative effects of basal amygdaloid stimulation, it cannot account for the entire facilitative effect. Other neurotransmitter mechanisms are thus likely involved. However, at the present time, their identity remains unknown. Additional evidence supporting a role for EAA mediation of the facilitative effects of basal amygdaloid stimulation was also obtained from a combined retrograde labeling–immunocytochemical analysis. The results of this study revealed the presence of fairly dense quantities of cells double labeled for Fluoro-Gold (following microinjections of this tracer into the PAG) and glutamate within the parvocellular as well as magnocellular divisions of the basal complex of amygdala (see Fig. 9). 4.2. Amygdaloid suppression of defensive rage behavior—role of the central nucleus
FIG. 7. Photomicrographs show that NMDA-R1 immunopositive receptors are distributed within the dorsal PAG—a region from which defensive rage behavior is typically elicited.
Previously, it has been demonstrated that stimulation of the central nucleus of amygdala can suppress defensive rage behavior (5). It is also known that the central nucleus projects directly to the PAG (10) and contains high concentrations of neurons that immunoreact with enkephalins (16). Thus, the hypothesis tested in this experiment is that central amygdaloid suppression of defensive rage behavior is mediated through a mechanism involving enkephalinergic receptors at the level of the PAG. Again, the same general paradigm was applied to the analysis of central amygdaloid suppression of defensive rage as was utilized for the study of other amygdaloid nuclei described above. In brief, central amygdaloid stimulation
NEUROTRANSMITTERS AND DEFENSIVE RAGE BEHAVIOR 25
2.5 nmole 0.5 0.05 baseline
-84
Change in response facilitatioin (%)
Change in response facilitatioin (%)
-110
-58
-32
-110
0
0
5
60
120
739
2.0 nmole 0.2 nmole baseline
-15
-5
5
180
Postinjection time (min) FIG. 8. Graphs indicate that microinfusion of CP-96,345 into sites within the medial hypothalamus from which defensive rage was elicited blocks the facilitatory effects of medial amygdaloid stimulation upon this response. (From Shaikh et al. (22).)
initially suppressed defensive rage elicited from PAG (19). Following microinjections of either naloxone (27.3 nmol) or the selective m opioid antagonist, b-FNA (0.2 nmol), the suppressive effects of central amygdaloid stimulation were blocked over a 30-min post-injection period (Fig. 11). Further support for the present ‘enkephalinergic’ hypothesis
15 0
5
60
120
180
Postinjection time (min) FIG. 10. Graphs indicate that microinfusion of AP-7 into sites within the PAG, from which defensive rage was elicited, produces a partial blockade of the facilitatory effects of basal amygdaloid stimulation upon this response. (From Shaikh et al. (20).)
was obtained from a combined neuroanatomical–immunocytochemical study. Following microinjections of FluoroGold into sites within the PAG from which defensive rage had been elicited, neurons labeled for both Fluoro-Gold and Met-enkephalin within the central nucleus and adjoining medial aspect of the lateral nucleus were identified (see Fig. 9). 4.3. Application of our findings to the study of the possible mechanisms underlying how certain psychoactive substances may modulate aggressive behavior—a case in point: the potentiating role of ethanol in defensive rage behavior
FIG. 9. Maps indicate the loci of double-labeled cells containing both Fluoro-Gold and: substance P (closed circles), glutamate (closed stars), or Met-enkephalin (closed triangles) in association with microinjections placed into the medial hypothalamus (with respect to substance P) and the PAG with respect to glutamate and Met-enkephalin. AB, basal nuclear complex of amygdala; AC, central nucleus of amygdala; AL, lateral nucleus of amygdala; AM, medial nucleus of amygdala; OT, optic tract. (Adapted from Shaikh et al. (19,20,22).)
From the studies described above, we now have a new understanding of some of the basic neural substrates and neurochemical mechanisms mediating the expression and control of defensive rage behavior. One of the benefits of having obtained such data is that they could provide a basis by which we could understand better the neural mechanisms underlying how certain psychoactive substances potentiate violent behavior in humans. Our basic assumption here is that the enhancing actions of a given psychoactive drug on aggression most likely occur as a result of its effects upon components of the limbic–hypothalamic–midbrain circuitry described in this paper. Accordingly, if this assumption is a correct one, then it becomes possible to design specific experiments to test various hypotheses concerning the neural bases by which psychoactive substances may enhance aggressive behavior. One such psychoactive substance that we believe to be appropriate for this kind of analysis is ethanol, which is known to have significant enhancing effects upon aggression and violence in humans (31). In fact, we have utilized such an approach to study the effects and possible mechanisms underlying ethanol’s potentiating properties upon defensive rage behavior in the cat. In the experiments described below, we have provided preliminary evidence of a likely mechanism by which alcohol potentiates defensive rage behavior.
740
SIEGEL ET AL. 1000
pre-injection post-injection (B–FNA 0.2 nM)
800 600 400 200 0 -200
0
20
40
60
80
100
120
140
Change in response
1000
pre-injection post-injection (B–FNA 0.05 nM)
800 600 400 200 0 -200
0
20
40
60
80
100
120
140
1000
pre-injection post-injection (ICI 174,864 0.7 nM)
800 600 400 200 0 -200
0
20
40
60
80
100
120
140
Time after amygdaloid stimulation (min) FIG. 11. Effects of microinfusion of b-FNA into the PAG upon central amygdaloid suppression of PAG-elicited defensive rage behavior. Open circles indicate the time course for central amygdaloid-induced suppression of defensive rage behavior as determined by the percentage change in response latencies from baseline values obtained prior to amygdaloid stimulation. These response latencies were determined from single stimulation of the PAG following stimulation of the central nucleus. Closed circles indicate the effects of infusion of drug. Note that administration of 0.2 nmol of FNA completely blocked the suppressive effects of central amygdaloid stimulation (see upper panel), while administration of a lower dose, 0.05 nmol, resulted in a partial blockade of central amygdaloid-induced suppression. Microinfusion of the selective d-antagonist ICI 174,864 did not block the suppressive effects of central amygdaloid stimulation. *p , 0.05. (From Shaikh et al. (19).)
In the first part of this study, we observed that ethanol administration (0.1–1.0 g/kg) significantly facilitated the occurrence of defensive rage behavior elicited from the medial hypothalamus in a dose- and time-dependent manner (17) (see Fig. 12). In fact, the potency of the effects of ethanol reduced response latencies by as much as 50% over a 60-min period at 1.0 g/kg, after which time response latencies approached baseline levels recorded prior to alcohol delivery. Central questions raised by this observation include: (1) are the effects of ethanol upon defensive rage mediated over
the pathway from the medial hypothalamus to the PAG and (2) is such a notion testable? We sought to test this hypothesis in the following manner. Initially, baseline response latency values for defensive rage elicited from the medial hypothalamus were determined. Then, ethanol (1.0 g/kg) was administered ip and its facilitating effects, noted above, were identified. This entire procedure was then repeated with the exception that AP-7 (2.0 nmol, ic) was microinjected into the PAG following ethanol administration. The results indicated that AP-7 administration totally blocked the facilitative effects of ethanol upon defensive rage behavior (Fig. 13). In fact, the blocking effects of AP-7 upon defensive rage behavior in the absence of ethanol administration were virtually identical to those observed when both ethanol and AP-7 were delivered to the cat. This observation suggests that the facilitating effects of ethanol upon defensive rage behavior are, indeed, mediated at some (unknown) point along the pathway from the medial hypothalamus to the PAG. That such effects may involve the interaction of ethanol upon NMDA receptors in the PAG is also possible. However, the fact that NMDA receptor blockade within the PAG can eliminate the facilitative effects of ethanol upon defensive rage does not eliminate alternative interpretations of the data. For example, it is possible that medial hypothalamic neurons release EAA on to interneurons within the PAG, which in turn synapse upon other neurons that mediate defensive rage behavior by projecting to motor and autonomic cells groups of the lower brainstem. It is conceivable that ethanol’s effects upon defensive rage may not be related to its interactions with NMDA receptors. Instead, it may be related to its interactions with the receptor mechanism associated with the neurons with which the interneuron makes a synapse. Nevertheless, the fact that NMDA receptor blockade can eliminate the facilitative effects of hypothalamically elicited defensive rage behavior provides, for the first time, evidence of the overall circuitry upon which the potentiating effects of alcohol are likely mediated. Ongoing studies in our laboratory are currently employing a variety of related methodologies in order to replicate and further characterize this phenomenon. 5. SUMMARY The studies described in this review have begun to identify the circuitry underlying the expression and modulation of defensive rage behavior in the cat. A summary diagram of these pathways and their functional relationships is shown in Fig. 14. One pathway is associated with the expression of defensive rage behavior. The projection originates mainly from the rostral half of the medial hypothalamus and projects directly to the PAG. Its functions are mediated by excitatory amino acids that act through NMDA receptors. Related data were also presented, suggesting that ethanol’s potentiating effects upon defensive rage are mediated through this descending pathway. Other pathways considered in this review relate to three different regions of amygdala that modulate defensive rage behavior. Two of the regions facilitate defensive rage behavior. These include the medial nucleus and basal complex. The facilitatory effects of the medial nucleus result from its excitatory effects upon neurons of the ventromedial hypothalamus from which defensive rage behavior is elicited. These
NEUROTRANSMITTERS AND DEFENSIVE RAGE BEHAVIOR
741
Decrease in response latency (%)
60
1.0 g/kg 0.5 g/kg 0.01 g/kg
46
32
A CE M Y G ME D A L A BA
ENK
SP
+ MH
EAA
+
– Output
CG +
EAA
18
4
-10
FIG. 12. Bar graph indicating that peripheral administration (i.p.) of ethanol significantly potentiated the occurrence of defensive rage behavior (n ¼ 5) as shown by a decrease in response latency in a dose- and time-dependent manner. (Adapted from Schubert et al. (17).)
FIG. 14. Schematic diagram summarizes the principal neuroanatomical pathways and their likely neurotransmitters from the amygdala and medial hypothalamus that regulate defensive rage behavior in the cat. In this scheme, defensive rage behavior is elicited from either the medial hypothalamus (MH) or midbrain periaqueductal gray (CG). The projection from the MH to the PAG utilizes excitatory amino acids (EAA) that act upon NMDA receptors within the PAG to facilitate the occurrence of PAG-elicited defensive rage behavior. In the amygdala, two regions that facilitate defensive rage are indicated. These include: (1) the basal complex (BA), which projects to the PAG and whose functions are mediated through NMDA receptors; and (2) the medial amygdala (ME), which projects directly to the MH and whose functions are mediated by substance P (SP), which acts through NK receptors. Inhibition is associated with the central nucleus (CE), which projects directly to the PAG and whose functions are mediated by enkephalins that act upon m opioid receptors. (From Siegel and Schubert (26).)
Decrease in response latency (%)
60
1.0g/kg ETOH 45
30
15
0
5
60
120
180
Increase in response latency (%)
90
2.0 nM AP-7
64
AP-7 & ETOH 48
ACKNOWLEDGEMENT
This research was supported by NIH Grant NS07941-26.
32
REFERENCES
16
0
effects are mediated by substance P, which acts through NK 1 receptors. The facilitatory effects of the basal amygdala result, in part, from its excitatory effects upon neurons of the dorsal PAG from which defensive rage behavior could also be elicited. Such facilitation is mediated by excitatory amino acids that act upon NMDA receptors within the PAG. Suppressive actions of the amygdala upon defensive rage behavior are associated with the central amygdaloid nucleus. Its effects are manifest via an enkephalinergic pathway that supplies the dorsal PAG and acts through m receptors within the PAG. Finally, we further considered the application of these findings to the analysis of the potentiating effects of ethanol upon defensive rage. In these experiments, we demonstrated for the first time that the potentiating effects of ethanol upon defensive rage behavior are mediated, at least in part, over the descending pathway from the medial hypothalamus to the dorsal PAG.
5
60
120
180
Postinjection time (min) FIG. 13. Effects of microinjections of AP-7 upon ethanol-induced potentiation of defensive rage behavior elicited from the medial hypothalamus. Bar graphs indicate that microinjections of AP-7 into the PAG blocked the potentiating effects of ethanol administration upon defensive rage behavior to an extent similar to that observed when AP-7 was microinjected into the PAG in the absence of ethanol administration (lower panel). Solid bars (shown in upper panel) depict the potentiating effects of ethanol upon defensive rage behavior in the absence of AP-7 treatment (n ¼ 6). (From Schubert et al. (17).)
1. Banay-Schwartz, M.; Lajtha, A.; Palkovits, M. Changes with aging in the levels of amino acids in rat CNS structural elements. I. Glutamate and related amino acids. Neurochem. Res. 14:555–562; 1989. 2. Bandler, R. J.; Depaulis, A. Midbrain periaqueductal gray control of defensive behavior in the cat and the rat. In: Depaulis, A.; Bandler, R. J. eds. The midbrain periaqueductal gray matter. New York: Plenum Press; 1991:175–198. 3. Beitz, A. J. Possible origin of glutamatergic projections to the midbrain periaqueductal gray and deep layer of the superior colliculus of the rat. Brain Res. Bull. 23:25–35; 1989. 4. Brownstein, M. J.; Mroz, E. A.; Kizer, J. S.; Palkovits, M.; Leeman, S. E. Regional distribution of substance P in the brain of the rat. Brain Res. 116:299–305; 1976. 5. Brutus, M.; Shaikh, M. B.; Siegel, A.; Edinger, H. Effects of experimental temporal lobe seizures upon hypothalamically elicited aggressive behavior in the cat. Brain Res. 366:53–63; 1986. 6. De Olmos, J. S.; Ingram, W. R. The projection field of the stria
742
7. 8. 9. 10. 11. 12. 13.
14.
15. 16.
17.
18.
SIEGEL ET AL. terminalis in the rat brain. An experimental study. J. Comp. Neurol. 146:303–334; 1972. Fuchs, S. A. G.; Edinger, H. M.; Siegel, A. The organization of the hypothalamic pathways mediating affective defense behavior in the cat. Brain Res. 330:77–92; 1985. Hess, W. R.; Brugger, M. Das subkortikale Zentrum der affektiven abwehrreaktion. Helv. Physiol. Pharmac. Acta 1:33–52; 1943. Hopkins, D. A.; Holstege, G. Amygdaloid projections to the mesencephalon, pons and medulla oblongata in the cat. Exp. Brain Res. 32:529–547; 1978. Krettek, J. E.; Price, J. L. Amygdaloid projections to subcortical structures within the basal forebrain and brainstem in the rat and cat. J. Comp. Neurol. 178:225–254; 1978. Leyhausen, P. Cat behavior. The predatory and social behavior of domestic and wild cats. New York: Garland STPM Press; 1979. Lu, C.-L.; Shaikh, M. B.; Siegel, A. Role of NMDA receptors in hypothalamic facilitation of feline defensive rage elicited from the midbrain periaqueductal gray. Brain Res. 581:123–132; 1992. Neal, C. R.; Newman, S. W. Prodynorphin peptide distribution in the forebrain of the syrian hamster and rat: a comparative study with antisera against dynorphin A, dynorphin B, and the c-terminus of the prodynorphin precursor molecule. J. Comp. Neurol. 288:353–386; 1989. Neal, C. R.; Swann, J. M.; Newman, S. W. The colocalization of substance P and prodynorphin immunoreactivity in neurons of the medial preoptic area, bed nucleus of the stria terminalis and medial nucleus of the amygdala of the Syrian hamster. Brain Res. 496:1–13; 1989. Ranson, S. W.; Magoun, H. W. The hypothalamus. ERG Physiol. 41:56–163; 1939. Rao, R.; Yamano, M.; Shiosaka, S.; Shinohara, S.; Tohyama, M. Origin of leucine-enkephalin fibers and their two main afferent pathways in the bed nucleus of the stria terminalis in the rat. Exp. Brain Res. 65:411–420; 1987. Schubert, K.; Shaikh, M. B.; Pohorecky, L.; Han, Y.; Siegel, A. Differential effects of ethanol upon feline rage and predatory attack behavior: an underlying neural mechanism. Alc. Clin. Exp. Res. 20:882–889; 1996. Shaikh, M. B.; Barrett, J. A.; Siegel, A. The pathways mediating affective defense and quiet biting attack behavior from the midbrain central gray of the cat: an autoradiographic study. Brain Res. 437: 9–25; 1987.
19. Shaikh, M. B.; Lu, C. L.; Siegel, A. An enkephalinergic mechanism involved in amygdaloid suppression of affective defense behavior elicited from the midbrain periaqueductal gray in the cat. Brain Res. 559:109–117; 1991. 20. Shaikh, M. B.; Schubert, K.; Siegel, A. Basal amygdaloid facilitation of midbrain periaqueductal gray elicited defensive rage behavior in the cat is mediated through NMDA receptors. Brain Res. 635:187–195; 1994. 21. Shaikh, M. B.; Shaikh, A. B.; Siegel, A. Opioid peptides within the midbrain periaqueductal gray suppress affective defense behavior in the cat. Peptides 9:999–1004; 1988. 22. Shaikh, M. B.; Steinberg, A.; Siegel, A. Evidence that substance P is utilized in medial amygdaloid facilitation of defensive rage behavior in the cat. Brain Res. 625:283–294; 1993. 23. Siegel, A.; Brutus, M. Neural substrates of aggression and rage in the cat. In: Epstein, A. N.; Morrison, AR. eds. Progress in psychobiology and physiological psychology. San Diego: Academic Press; 1990:135–233. 24. Siegel, A.; Flynn, J. P. Differential effects of electrical stimulation and lesions of the hippocampus and adjacent regions upon attack behavior in cats. Brain Res. 7:252–267; 1968. 25. Siegel, A.; Pott, C. B. Neural substrate of aggression and flight in the cat. Progr. Neurobiol. 31:261–283; 1988. 26. Siegel, A.; Schubert, K. Neurotransmitters regulating feline aggressive behavior. Rev. Neurosci. 6:47–61; 1995. 27. Swann, J. M.; Macchione, N. Photoperiodic regulation of substance P immunoreactivity in the mating behavior pathway of the male golden hamster. Brain Res. 590:29–38; 1992. 28. Swann, J. M.; Newman, S. W. Testosterone regulates substance P within neurons of the medial nucleus of the amygdala, the bed nucleus of the stria terminalis and the medial preoptic area of the male golden hamster. Brain Res. 590:18–28; 1992. 29. Wasman, M.; Flynn, J. P. Directed attack elicited from hypothalamus. Arch. Neurol. 6:220–227; 1962. 30. Watson, R. E.; Troiano, R.; Poulakos, J. J.; Weiner, S.; Block, C. H.; Siegel, A. [14C]2-Deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. I. The amygdala. Brain Res. Rev. 5:1–44; 1983. 31. Zeichner, A.; Allen, J. D.; Giancola, P. R.; Lating, J. M. Alcohol and aggression: effects of personal threat on human aggression and affective arousal. Alc. Clin. Exp. Res. 18:657–663; 1994.
Pergamon
PII: S0149-7634(96)00056-5
Neurotransmitters Regulating Defensive Rage Behavior in the Cat ALLAN SIEGEL, a ,* KRISTIE L. SCHUBERT*† AND MAJID B. SHAIKH* *Departments of Neurosciences and Psychiatry, New Jersey Medical School, Newark, NJ 07103, USA †Department of Biology, Kean College, Union, NJ 07083, USA
SIEGEL, A., K. L. SCHUBERT AND M. B. SHAIKH. Neurotransmitters regulating defensive rage behavior in the cat. NEUROSCI BIOBEHAV REV 21(6) 733–742, 1997.—This review summarizes recent findings of our laboratory that have been directed at: (1) identifying the neural circuits underlying the expression and modulation of defensive rage behavior in the cat and the neurotransmitters associated with these pathways; and (2) determining which components of the circuitry are affected by alcohol administration and which significantly alter the rage mechanism. The experiments described herein incorporated a number of converging methods, which include brain stimulation, behavioral pharmacology, immunocytochemistry, retrograde tract tracing and receptor binding. For behavioral pharmacological studies, monopolar electrodes and cannula-electrodes were implanted into selected regions along the limbic–midbrain axis for electrical stimulation and local microinfusion of drugs. The findings demonstrated: (1) a direct pathway from the anterior medial hypothalamus to the dorsal periaqueductal gray (PAG) over which this response is mediated. This pathway utilizes excitatory amino acids that act upon NMDA receptors within the midbrain PAG; (2) that the region of the dorsal PAG, from which defensive rage could be elicited, receives other inputs from the basal amygdala that facilitate this response by acting upon NMDA receptors; (3) a pathway from the medial amygdala to the medial hypothalamus that also facilitates defensive rage and whose functions are mediated by substance P receptors within the medial hypothalamus; (4) that the PAG also receives enkephalinergic inputs from the central nucleus of amygdala, which act upon m receptors, and which powerfully suppress defensive rage; and (5) that recent findings reveal that ethanol administration facilitates defensive rage by virtue of its interactions with the medial hypothalamus, its descending projection to the PAG, and possibly with NMDA receptors within this pathway. q 1997 Elsevier Science Ltd. Defensive rage behavior Hypothalamus Periaqueductal gray Amygdala Excitatory amino acids Enkephalins Receptor binding Immunocytochemistry Ethanol
Substance P
midbrain PAG. Such structures are said to be ‘modulators’ of defensive rage behavior. Limbic structures may either suppress or facilitate the occurrence of defensive rage, but it is important to note that the rage response cannot be elicited by electrical stimulation of any of these limbic structures.
STUDIES conducted over the past three decades have established that the neural substrates for defensive rage behavior in the cat include the hypothalamus, midbrain periaqueductal gray (PAG) as well as limbic nuclei such as the amygdala, hippocampal formation and septal area (18– 21,23–26,29,30). The general functions of each of these regions with respect to rage and aggression have now been identified. Two different classes of neuronal regions can be distinguished. The first, consisting primarily of the medial hypothalamus and midbrain PAG, represent structures from which defensive rage can be elicited by electrical stimulation. Such structures are said to be the ‘effector’ regions for the expression of defensive rage behavior. The second class of structures consists primarily of the limbic system. The major structures of this group include the amygdala, hippocampal formation, septal area, prefrontal cortex, cingulate cortex and related regions of the basal forebrain, and influence the rage mechanism by virtue of their actions upon either the medial hypothalamus or
1.
DEFENSIVE RAGE BEHAVIOR
This form of aggressive behavior occurs under natural conditions when the cat is threatened by another animal of the same or different species. The nature of the external threat could come from different sources. A typical example is the presence of another cat that is competing for its territory (11). The response is characterized by various sympathetic signs, such as pupillary dilatation, piloerection, ear retraction, salivation, unsheathing of the claws, hissing and paw strike (8,11,15,30). As noted above, this response can be elicited by electrical stimulation of the medial
a Correspondence should be addressed to: Allan Siegel, Ph.D., Department of Neurosciences, Medical Science Building, Rm H-512, New Jersey Medical School, 185 South Orange Avenue, Newark, NJ 07103, USA.
733
734
SIEGEL ET AL.
hypothalamus or the midbrain PAG. The section below summarizes the regions from which defensive rage can be elicited as well as the primary descending pathways over which this response is mediated. 2. REGIONS ASSOCIATED WITH THE EXPRESSION OF DEFENSIVE RAGE AND THEIR DESCENDING PATHWAYS
2.1. Defensive rage sites Defensive rage behavior can be elicited under laboratory conditions by electrical stimulation principally of the medial hypothalamus or midbrain PAG. Within the medial hypothalamus, sites from which defensive rage behavior can be elicited extend throughout its rostro-caudal extent. The most notable sites at which this response can be elicited include the region of the ventromedial nucleus, the dorsomedial hypothalamic area, and the anterior medial hypothalamus. Defensive rage behavior cannot be elicited from regions of hypothalamus other than its medial aspect. Within the midbrain PAG, defensive rage can be easily elicited from is dorsal aspect, and in particular, at the level
FIG. 1. Maps indicate sites within the: (A) midbrain periaqueductal gray and (B) medial hypothalamus from which defensive rage behavior was elicited by electrical stimulation in the studies described in this review. CI, internal capsule; CS, superior colliculus; Fx, fornix; HL, lateral hypothalamus; HVM, ventromedial hypothalamus; PAG, midbrain periaqueductal gray; Ped, cerebral peduncle; RE, nucleus reuniens; RN, red nucleus; SN, substantia nigra; TO, optic tract.
of the rostral half of the neuropil (2,18,19,21,22). At more caudal levels of the PAG, defensive rage is not readily seen. In its place, one can identify sites that produce howling as well as flight behavior (2). That defensive rage is limited to the dorsal aspect of the PAG is further evidenced by the fact that stimulation of its ventral aspect can only produce elements of predatory attack behavior, a form of aggression that is characterized by the initial stalking followed by the biting of the back of a rat’s neck. Maps indicating the loci of sites within the hypothalamus and PAG, from which defensive rage is typically elicited by electrical stimulation, are shown in Fig. 1. 2.2. Descending pathways mediating defensive rage behavior The data described above point to the importance of the medial hypothalamus and dorsal PAG as the key regions of the brain involved in the integration of defensive rage behavior. A question of central importance concerns the descending neuroanatomical circuits mediating the expression of defensive rage behavior from the hypothalamus. A series of studies conducted in our laboratory utilized anterograde and retrograde tracing methods to answer this question (7,12,17,18). With respect to the projection from the hypothalamus to the PAG, the initial studies of Fuchs et al. (7) utilized the combined methods of tritiated amino acid radioautography and 14C-2-deoxyglucose autoradiography to trace the pathways from sites in the medial hypothalamus to the brainstem. The findings from this study indicated that neurons arising from the ventromedial nucleus associated with defensive rage behavior generally project in a rostral direction to the anterior medial hypothalamus from which this response can also be elicited. Cells located in the region that extends rostrally from the dorsomedial hypothalamus to the anteromedial–preoptic region give rise to long, descending axons that project to that part of the dorsal aspect of the PAG which is also associated with the expression of defensive rage. This finding was basically replicated recently with retrograde tracing methods following the placement of Fluoro-Gold into defensive rage sites within the PAG (12,17). These studies showed that the larger majority of descending neurons that project to identifiable rage sites in the PAG arise from cells along a band extending rostrally and somewhat dorsally from the ventromedial nucleus with relatively few neurons located in the ventromedial nucleus, per se. This pathway is summarized in Fig. 2. It should be noted that it has recently been shown that the PAG is organized in a series of columns that extend along its longitudinal axis (2). One such functional column mediates defensive rage and threat behavior and is situated in the dorsal aspect of the rostral two-thirds of the PAG. As indicated above, significant inputs to the PAG from the medial hypothalamus in association with defensive rage behavior terminate within the rostral half of this column (7). Thus, it would seem that the rostral half of this column is critical for the integration of descending fibers from the medial hypothalamus required for the expression of defensive rage behavior. The efferent pathways of the PAG that mediate defensive rage behavior project primarily in a descending direction,
NEUROTRANSMITTERS AND DEFENSIVE RAGE BEHAVIOR
FIG. 2. Diagram illustrates the primary projections of the medial hypothalamus associated with defensive rage behavior (i.e. referred to in this figure and Fig. 3 as ‘affective defense’) in the cat. Note that the projections of the medial hypothalamus to the dorsal aspect of the midbrain periaqueductal gray (called central gray in this figure and in Fig. 3) arise principally from the rostral half of the hypothalamus. This illustration also indicates that fibers associated with defensive rage behavior that arise from the ventromedial nucleus of the hypothalamus generally project to more rostral levels of the medial hypothalamus. (From Siegel and Pott (25).)
735
FIG. 3. Diagram illustrates the principal outputs of the dorsal aspect of the PAG from which defensive rage can be elicited. Fibers arising from the periaqueductal gray in association with defensive rage behavior are distributed to a number of regions in the lower brainstem that are important for the expression of this response. One such region includes the motor nucleus of the trigeminal nerve, which is essential for jaw opening and for the occurrence of ‘hissing’ and other forms of vocalization that are integral to the defensive rage response. A second projection includes the distribution of fibers to autonomic regions of the lower brainstem, which help to mediate the sympathetic components of this response. In addition, a third, ascending component supplies the posterior medial hypothalamus. This projection likely serves as a ‘Positive feedback’ to the hypothalamus, thus increasing the duration of the defensive rage response. (Adapted from Shaikh et al. (18).)
3. FUNCTIONS OF THE PATHWAY FROM THE MEDIAL HYPOTHALAMUS although some ascending fibers of significance have also been identified (18). The primary descending projections innervate parts of the lower brainstem tegmentum, locus ceruleus, and motor and sensory nuclei of the trigeminal complex. In this manner, the descending projections of the PAG innervate both autonomic and somatomotor nuclei of the lower brainstem, which provide the substrate by which an integrated defensive rage response can be elicited from the PAG. In fact, the PAG represents the most caudal region of the brainstem where such integration can take place. The ascending projection of the PAG is also of interest. Axons from the dorsal PAG supply parts of the posterior half of the medial hypothalamus from which defensive rage could also be elicited. This connection provides the basis for a positive feedback circuit between the medial hypothalamus and the dorsal PAG. Such a circuit is of potential important survival value to the animal in that it provides the basis for a response of relatively long duration that is required during combat. A diagram summarizing the primary projections of the dorsal PAG from which defensive rage can be elicited is shown in Fig. 3.
TO THE PAG ARE MEDIATED BY EXCITATORY AMINO ACIDS
As noted above, the pathway from the medial hypothalamus to the dorsal PAG links two regions of the brain that are central to the integration of defensive rage behavior. Therefore, a primary objective of our research efforts was to attempt to identify the likely neurotransmitter associated with this pathway. The strategy was to employ a variety of converging methodological approaches. These included brain stimulation methods coupled with pharmacological manipulations, immunocytochemistry coupled with retrograde tracing procedures, and receptor binding. Recent studies conducted in the rat demonstrated the presence of high concentrations of glutamate and aspartate within the PAG (1,3). Therefore, we sought to test the hypothesis that excitatory amino acids (EAA) mediate the rage mechanism over the pathway from the medial hypothalamus to the PAG. 3.1. Evidence derived from neuroanatomical– immunocytochemical studies In these studies, the following procedures were implemented. Cannula electrodes were implanted into sites within
736
the PAG from which defensive rage behavior was elicited by electrical or chemical (i.e. homocysteic acid) stimulation. Then, Fluoro-Gold (4–8%) was microinjected into the PAG and the animals were sacrificed 7–14 days postinjection. Alternate sections were processed for immunocy-
SIEGEL ET AL.
tochemical labeling of glutamate or aspartate neurons within the forebrain and brainstem (12,17). The objective of this approach was to be able to determine whether or not neurons in the medial hypothalamus that are labeled for Fluoro-Gold are also labeled for glutamate or aspartate as well. Such data would support the basic hypothesis that the projection from the medial hypothalamus to the PAG that mediates defensive rage behavior is mediated by EAA. The results of these studies are summarized in Fig. 4. The loci of individual microinjections of Fluoro-Gold placed into the dorsal PAG are depicted in Fig. 4(a). As shown in Fig. 4(b), cells labeled for both glutamate (or aspartate) and Fluoro-Gold were located throughout the rostral half of the medial hypothalamus, extending from the level of the ventromedial nucleus to the preoptic region. The largest quantities of double-labeled neurons were situated in the dorsomedial and anteromedial hypothalamus, rostral to the level of the ventromedial nucleus. A few double-labeled neurons were also identified in the ventromedial nucleus. Fig. 4(b) also reveals that many other cells were labeled for either Fluoro-Gold or glutamate (or aspartate) but not for both, indicating that such neurons either project to the PAG but presumably do not utilize EAA as a neurotransmitter, or that they may utilize glutamate or aspartate but do not project to the dorsal PAG. Additional evidence supporting the view that neurons expressing aspartate project to the dorsal PAG is provided in Fig. 4(c) as well, which indicates the presence of labeled aspartatergic axons and preterminals within the dorsal PAG. Collectively, these findings provide direct support for the hypothesis of a monosynaptic pathway from the rostral half of the medial hypothalamus to the PAG whose behavioral functions are mediated by EAA. However, it should be pointed out that we have only examined the inputs from the medial hypothalamus to the functional column of cells situated within the rostral half of the dorsal PAG. It remains to be determined whether the inputs to more caudal aspects of the dorsal PAG column associated with threat and defensive behavior are also mediated by EAA that act upon NMDA receptors. 3.2. Evidence from behavioral pharmacological experiments The hypothesis of a pathway from the medial hypothalamus to the dorsal PAG which utilizes EAA to mediate the expression of defensive rage behavior was tested directly in the following experiments (12,17). A cannula electrode was initially implanted into a site within the dorsal PAG from
FIG. 4. (a) Section illustrates the loci of Fluoro-Gold injections at sites within the PAG from which stimulation produced defensive rage behavior. (b) Maps indicate the presence of neurons that are single-labeled for glutamate (or aspartate) (closed circles), Fluoro-Gold (open circles), or doublelabeled for glutamate (or aspartate) and Fluoro-Gold (closed triangles). Note that the highest concentration of double-labeled neurons is located in the dorsomedial and anteromedial hypothalamus, the regions where defensive rage behavior are typically elicited. Abbreviations: alld, dorsal hypothalamus; Fx, fornix; HL, lateral hypothalamus; Hp, posterior hypothalamus. (c) Presence of aspartate labeled axons and preterminals within the dorsal aspect of the PAG—the region from which stimulation typically elicits defensive rage behavior. HVM, ventromedial hypothalamic nucleus; PAG, midbrain periaqueductal gray; RE, nucleus reuniens; RN, red nucleus. (Maps represent combined data from studies of Lu et al. (12) and unpublished observation from our laboratory.)
NEUROTRANSMITTERS AND DEFENSIVE RAGE BEHAVIOR 100
2.0 0.5 0.1 4.4 BL
80
Change in latency (%)
which defensive rage could be elicited and where drugs could be infused at a later time. Then, a stimulating electrode was implanted into a site within the medial hypothalamus from which defensive rage could also be elicited. The experimental paradigm involved the application of paired trials of (a) a single stimulation of the PAG from which defensive rage was elicited and (b) a dual stimulation of the PAG and medial hypothalamus (at current levels below threshold for the elicitation of defensive rage behavior). The effects of dual stimulation were then compared to that of single stimulation of the PAG alone. When significant modulation of the hypothalamus was determined, the current was maintained constant. In order to test the basic hypothesis, we first demonstrated that, prior to drug infusion, dual stimulation involving the medial hypothalamus significantly facilitated the occurrence of defensive rage behavior elicited from the PAG. Then, the procedures of dual stimulation were repeated following the administration of different doses of: (a) a nonselective EAA receptor antagonist (kynurenic acid); and (b) a selective NMDA antagonist [dl-2-amino-7-phosphonoheptanoic acid; (AP-7)]; as well as a single dose of the non-NMDA antagonist CNQX, which was equivalent to the highest dose of AP-7 administered. Following the delivery of either the non-selective antagonist, kynurenic acid, or the selective NMDA antagonist, AP7, the facilitative effects of medial hypothalamic stimulation were blocked in a dose- and time-dependent manner (Figs 5 and 6). In contrast, infusion of the non-NMDA antagonist CNQX did not modify the facilitative effects of dual stimulation. These results provide direct evidence that the pathway mediating defensive rage behavior from the medial hypothalamus to the PAG utilizes EAA that act through NMDA receptors. Most recently, the importance of NMDA receptors within the dorsal PAG in mediating defensive rage behavior was tested in yet another way (17). In this experiment, infusion of the AP-7 into sites within the PAG from which defensive rage could be elicited blocked defensive rage behavior that was elicited by single stimulation of the medial hypothalamus. Thus, when taken together, the results of these experiments involving two different experimental paradigms provide converging evidence in support of the role played by NMDA receptors within the dorsal PAG in the regulation of defensive rage behavior.
737
60 40 20 0 -20 -40 -60 -80 0
60
120
180
FIG. 5. Graphs indicate that microinjections of kynurenic acid (kyn) into PAG sites from which defensive rage was elicited block the facilitatory effects of medial hypothalamic stimulation upon this response elicited from the PAG. A higher dose of atropine (atr) did not alter the facilitatory effects of medial hypothalamic stimulation. For this and Fig. 6: BL, theoretical baseline indicating the position on the graph where dual stimulation of the medial hypothalamus had no effect in altering PAG response latencies as determined from single stimulation of the PAG. Drug doses are in nanomolar values. Vertical bars indicate SEM for this and subsequent figures. (From Lu et al. (12).)
receptors by virtue of the presence of positive NMDA-R1 immunoreactivity in the dorsal aspect of the PAG. As noted above, the dorsal PAG receives the inputs from cell groups in the medial hypothalamus associated with the expression of defensive rage. Therefore, it is of interest to point out that further analysis revealed that such immunoreactivity was not present in the ventral PAG, which does not receive afferents from the medial hypothalamus. 4.
AMYGDALOID MODULATION OF DEFENSIVE RAGE BEHAVIOR
As indicated previously, the regions considered above— the medial hypothalamus and midbrain PAG—constitute effector regions for the expression of defensive rage behavior. Other regions also play an important role in the regulation of defensive rage behavior. They do so by modulating the defensive rage mechanism at the level of the hypothalamus or PAG. The following review briefly summarizes the results of recent studies concerning the
Change in latency (%)
In the previous section, pharmacological methods were employed to demonstrate the role of NMDA receptors in the PAG in regulating defensive rage behavior. In this section, we employed receptor histochemical procedures to demonstrate the presence of NMDA receptors within the dorsal aspect of the PAG. In brief, brain sections were reincubated with 10% normal goat serum in phosphate-buffered saline and then reacted with a purified monoclonal antibody against the NMDA-R1 receptor subtype at a 1:1000 dilution in phosphate buffer. Sections were then washed and processed with a species-specific Vectastain ABC Elite immunoperoxidase kit (Burlingame, CA), after which sections were stained with DAB in phosphate-buffered saline. The results, shown in Fig. 7, demonstrate the existence of glutamate
1440
Postinjection time (min)
50
3.3. Further evidence for the presence of NMDA receptors in the dorsal PAG
kyn kyn kyn atr
2.0 nM AP7 0.5 nM AP7 0.1 nM AP7 4nM CNQX BL
40 30 20 10 0 -10 -20 0
45
90
135
180
Postinjection time (min) FIG. 6. Graphs indicate that infusion of AP-7 into sites within the PAG from which defensive rage had been elicited blocks the facilitatory effects of medial hypothalamic stimulation upon this response. Note that a higher dose of CNQX had little effect upon medial hypothalamic facilitation of this response (From Lu et al. (12).)
738
mechanisms underlying amygdaloid modulation of defensive rage behavior. The basic paradigm utilized to identify the modulating properties, the origins and distributions of the pathways, and likely neurotransmitters of the amygdaloid regions in question, was identical to the dual stimulation procedure described earlier for analyzing the modulating effects of medial hypothalamic stimulation upon PAG-elicited defensive rage. To date, three regions of the amygdala have been studied. Two of them—the medial nucleus and the basal complex—facilitate defensive rage, whereas a third region, the central (and adjoining lateral) nucleus suppresses defensive rage. 4.1. Regions that facilitate defensive rage behavior 4.1.1. The medial amygdala For testing the modulating effects of this structure, a cannula electrode was placed into a site within the medial hypothalamus from which defensive rage was elicited, and a stimulating electrode was lowered into the medial amygdala. Response latencies following paired trials of single (i.e. medial hypothalamus alone) and dual (i.e. medial nucleus þ medial hypothalamus) stimulation were compared both prior to and after drug administration (22). Prior to drug administration, dual stimulation produced a 30–40% decrease in response latencies relative to single stimulation of the medial hypothalamus alone. Then, the specific NK 1, non-peptide, substance P antagonist, CP 96,345, was microinjected into sites within the medial hypothalamus from which defensive rage was elicited in doses of 0.05–2.5 nmol/0.25 ml. The role of substance P was directly tested in this study for the following reasons: (1) substance P-positive neurons have been shown to be present in the medial amygdala in rodents (13,14,27,28); (2) the medial amygdala is known to project directly to the medial hypothalamus via the stria terminalis (6,10,30); and (3) substance P terminals have been identified in the medial hypothalamus (4). Drug administration significantly decreased medial amygdaloid facilitation in a dose- and time-dependent manner. As shown in Fig. 8, the highest dose used, 2.5 nmol of CP 96,345, decreased response facilitation by approximately 40% (p , 0.001). In a related study, Fluoro-Gold was microinjected into
SIEGEL ET AL.
medial hypothalamic sites from which defensive rage had been elicited. Following survival times of 5–6 days, animals were sacrificed, and alternate sections were processed for immunocytochemical labeling of substance P. Substantial quantities of neurons in the medial nucleus of amygdala were found to be double-labeled for both substance P and Fluoro-Gold (see Fig. 9). This finding provides additional support to the view that medial amygdaloid facilitation of defensive rage is mediated through substance P receptors within the ventromedial hypothalamus. 4.1.2. The basal complex The same paradigms as described above for characterizing medial amygdaloid facilitation of defensive rage behavior were applied to the analysis of basal amygdaloid modulation of this response. The basal amygdaloid complex is known to facilitate defensive rage behavior (5) and projects its axons to the PAG (9). Moreover, as described earlier in this paper, numerous glutamatergic and aspartatergic nerve terminals and NMDA receptors have been shown to be present in this neuropil as well. Therefore, the basic hypothesis tested in this study was that basal amygdaloid stimulation facilitates defensive rage behavior at the level of the PAG by acting through NMDA receptors. Dual stimulation of the basal amygdala and PAG (from which defensive rage was elicited) facilitated the occurrence of defensive rage behavior. Following administration of the selective NMDA antagonist, AP-7, into the PAG, basal amygdaloid facilitation of defensive rage was significantly suppressed by 19% over the first 60 min post-injection period (Fig. 10) (20). However, this effect was clearly less potent than that observed for other systems described above. This finding would suggest that while EAA contributes to the facilitative effects of basal amygdaloid stimulation, it cannot account for the entire facilitative effect. Other neurotransmitter mechanisms are thus likely involved. However, at the present time, their identity remains unknown. Additional evidence supporting a role for EAA mediation of the facilitative effects of basal amygdaloid stimulation was also obtained from a combined retrograde labeling–immunocytochemical analysis. The results of this study revealed the presence of fairly dense quantities of cells double labeled for Fluoro-Gold (following microinjections of this tracer into the PAG) and glutamate within the parvocellular as well as magnocellular divisions of the basal complex of amygdala (see Fig. 9). 4.2. Amygdaloid suppression of defensive rage behavior—role of the central nucleus
FIG. 7. Photomicrographs show that NMDA-R1 immunopositive receptors are distributed within the dorsal PAG—a region from which defensive rage behavior is typically elicited.
Previously, it has been demonstrated that stimulation of the central nucleus of amygdala can suppress defensive rage behavior (5). It is also known that the central nucleus projects directly to the PAG (10) and contains high concentrations of neurons that immunoreact with enkephalins (16). Thus, the hypothesis tested in this experiment is that central amygdaloid suppression of defensive rage behavior is mediated through a mechanism involving enkephalinergic receptors at the level of the PAG. Again, the same general paradigm was applied to the analysis of central amygdaloid suppression of defensive rage as was utilized for the study of other amygdaloid nuclei described above. In brief, central amygdaloid stimulation
NEUROTRANSMITTERS AND DEFENSIVE RAGE BEHAVIOR 25
2.5 nmole 0.5 0.05 baseline
-84
Change in response facilitatioin (%)
Change in response facilitatioin (%)
-110
-58
-32
-110
0
0
5
60
120
739
2.0 nmole 0.2 nmole baseline
-15
-5
5
180
Postinjection time (min) FIG. 8. Graphs indicate that microinfusion of CP-96,345 into sites within the medial hypothalamus from which defensive rage was elicited blocks the facilitatory effects of medial amygdaloid stimulation upon this response. (From Shaikh et al. (22).)
initially suppressed defensive rage elicited from PAG (19). Following microinjections of either naloxone (27.3 nmol) or the selective m opioid antagonist, b-FNA (0.2 nmol), the suppressive effects of central amygdaloid stimulation were blocked over a 30-min post-injection period (Fig. 11). Further support for the present ‘enkephalinergic’ hypothesis
15 0
5
60
120
180
Postinjection time (min) FIG. 10. Graphs indicate that microinfusion of AP-7 into sites within the PAG, from which defensive rage was elicited, produces a partial blockade of the facilitatory effects of basal amygdaloid stimulation upon this response. (From Shaikh et al. (20).)
was obtained from a combined neuroanatomical–immunocytochemical study. Following microinjections of FluoroGold into sites within the PAG from which defensive rage had been elicited, neurons labeled for both Fluoro-Gold and Met-enkephalin within the central nucleus and adjoining medial aspect of the lateral nucleus were identified (see Fig. 9). 4.3. Application of our findings to the study of the possible mechanisms underlying how certain psychoactive substances may modulate aggressive behavior—a case in point: the potentiating role of ethanol in defensive rage behavior
FIG. 9. Maps indicate the loci of double-labeled cells containing both Fluoro-Gold and: substance P (closed circles), glutamate (closed stars), or Met-enkephalin (closed triangles) in association with microinjections placed into the medial hypothalamus (with respect to substance P) and the PAG with respect to glutamate and Met-enkephalin. AB, basal nuclear complex of amygdala; AC, central nucleus of amygdala; AL, lateral nucleus of amygdala; AM, medial nucleus of amygdala; OT, optic tract. (Adapted from Shaikh et al. (19,20,22).)
From the studies described above, we now have a new understanding of some of the basic neural substrates and neurochemical mechanisms mediating the expression and control of defensive rage behavior. One of the benefits of having obtained such data is that they could provide a basis by which we could understand better the neural mechanisms underlying how certain psychoactive substances potentiate violent behavior in humans. Our basic assumption here is that the enhancing actions of a given psychoactive drug on aggression most likely occur as a result of its effects upon components of the limbic–hypothalamic–midbrain circuitry described in this paper. Accordingly, if this assumption is a correct one, then it becomes possible to design specific experiments to test various hypotheses concerning the neural bases by which psychoactive substances may enhance aggressive behavior. One such psychoactive substance that we believe to be appropriate for this kind of analysis is ethanol, which is known to have significant enhancing effects upon aggression and violence in humans (31). In fact, we have utilized such an approach to study the effects and possible mechanisms underlying ethanol’s potentiating properties upon defensive rage behavior in the cat. In the experiments described below, we have provided preliminary evidence of a likely mechanism by which alcohol potentiates defensive rage behavior.
740
SIEGEL ET AL. 1000
pre-injection post-injection (B–FNA 0.2 nM)
800 600 400 200 0 -200
0
20
40
60
80
100
120
140
Change in response
1000
pre-injection post-injection (B–FNA 0.05 nM)
800 600 400 200 0 -200
0
20
40
60
80
100
120
140
1000
pre-injection post-injection (ICI 174,864 0.7 nM)
800 600 400 200 0 -200
0
20
40
60
80
100
120
140
Time after amygdaloid stimulation (min) FIG. 11. Effects of microinfusion of b-FNA into the PAG upon central amygdaloid suppression of PAG-elicited defensive rage behavior. Open circles indicate the time course for central amygdaloid-induced suppression of defensive rage behavior as determined by the percentage change in response latencies from baseline values obtained prior to amygdaloid stimulation. These response latencies were determined from single stimulation of the PAG following stimulation of the central nucleus. Closed circles indicate the effects of infusion of drug. Note that administration of 0.2 nmol of FNA completely blocked the suppressive effects of central amygdaloid stimulation (see upper panel), while administration of a lower dose, 0.05 nmol, resulted in a partial blockade of central amygdaloid-induced suppression. Microinfusion of the selective d-antagonist ICI 174,864 did not block the suppressive effects of central amygdaloid stimulation. *p , 0.05. (From Shaikh et al. (19).)
In the first part of this study, we observed that ethanol administration (0.1–1.0 g/kg) significantly facilitated the occurrence of defensive rage behavior elicited from the medial hypothalamus in a dose- and time-dependent manner (17) (see Fig. 12). In fact, the potency of the effects of ethanol reduced response latencies by as much as 50% over a 60-min period at 1.0 g/kg, after which time response latencies approached baseline levels recorded prior to alcohol delivery. Central questions raised by this observation include: (1) are the effects of ethanol upon defensive rage mediated over
the pathway from the medial hypothalamus to the PAG and (2) is such a notion testable? We sought to test this hypothesis in the following manner. Initially, baseline response latency values for defensive rage elicited from the medial hypothalamus were determined. Then, ethanol (1.0 g/kg) was administered ip and its facilitating effects, noted above, were identified. This entire procedure was then repeated with the exception that AP-7 (2.0 nmol, ic) was microinjected into the PAG following ethanol administration. The results indicated that AP-7 administration totally blocked the facilitative effects of ethanol upon defensive rage behavior (Fig. 13). In fact, the blocking effects of AP-7 upon defensive rage behavior in the absence of ethanol administration were virtually identical to those observed when both ethanol and AP-7 were delivered to the cat. This observation suggests that the facilitating effects of ethanol upon defensive rage behavior are, indeed, mediated at some (unknown) point along the pathway from the medial hypothalamus to the PAG. That such effects may involve the interaction of ethanol upon NMDA receptors in the PAG is also possible. However, the fact that NMDA receptor blockade within the PAG can eliminate the facilitative effects of ethanol upon defensive rage does not eliminate alternative interpretations of the data. For example, it is possible that medial hypothalamic neurons release EAA on to interneurons within the PAG, which in turn synapse upon other neurons that mediate defensive rage behavior by projecting to motor and autonomic cells groups of the lower brainstem. It is conceivable that ethanol’s effects upon defensive rage may not be related to its interactions with NMDA receptors. Instead, it may be related to its interactions with the receptor mechanism associated with the neurons with which the interneuron makes a synapse. Nevertheless, the fact that NMDA receptor blockade can eliminate the facilitative effects of hypothalamically elicited defensive rage behavior provides, for the first time, evidence of the overall circuitry upon which the potentiating effects of alcohol are likely mediated. Ongoing studies in our laboratory are currently employing a variety of related methodologies in order to replicate and further characterize this phenomenon. 5. SUMMARY The studies described in this review have begun to identify the circuitry underlying the expression and modulation of defensive rage behavior in the cat. A summary diagram of these pathways and their functional relationships is shown in Fig. 14. One pathway is associated with the expression of defensive rage behavior. The projection originates mainly from the rostral half of the medial hypothalamus and projects directly to the PAG. Its functions are mediated by excitatory amino acids that act through NMDA receptors. Related data were also presented, suggesting that ethanol’s potentiating effects upon defensive rage are mediated through this descending pathway. Other pathways considered in this review relate to three different regions of amygdala that modulate defensive rage behavior. Two of the regions facilitate defensive rage behavior. These include the medial nucleus and basal complex. The facilitatory effects of the medial nucleus result from its excitatory effects upon neurons of the ventromedial hypothalamus from which defensive rage behavior is elicited. These
NEUROTRANSMITTERS AND DEFENSIVE RAGE BEHAVIOR
741
Decrease in response latency (%)
60
1.0 g/kg 0.5 g/kg 0.01 g/kg
46
32
A CE M Y G ME D A L A BA
ENK
SP
+ MH
EAA
+
– Output
CG +
EAA
18
4
-10
FIG. 12. Bar graph indicating that peripheral administration (i.p.) of ethanol significantly potentiated the occurrence of defensive rage behavior (n ¼ 5) as shown by a decrease in response latency in a dose- and time-dependent manner. (Adapted from Schubert et al. (17).)
FIG. 14. Schematic diagram summarizes the principal neuroanatomical pathways and their likely neurotransmitters from the amygdala and medial hypothalamus that regulate defensive rage behavior in the cat. In this scheme, defensive rage behavior is elicited from either the medial hypothalamus (MH) or midbrain periaqueductal gray (CG). The projection from the MH to the PAG utilizes excitatory amino acids (EAA) that act upon NMDA receptors within the PAG to facilitate the occurrence of PAG-elicited defensive rage behavior. In the amygdala, two regions that facilitate defensive rage are indicated. These include: (1) the basal complex (BA), which projects to the PAG and whose functions are mediated through NMDA receptors; and (2) the medial amygdala (ME), which projects directly to the MH and whose functions are mediated by substance P (SP), which acts through NK receptors. Inhibition is associated with the central nucleus (CE), which projects directly to the PAG and whose functions are mediated by enkephalins that act upon m opioid receptors. (From Siegel and Schubert (26).)
Decrease in response latency (%)
60
1.0g/kg ETOH 45
30
15
0
5
60
120
180
Increase in response latency (%)
90
2.0 nM AP-7
64
AP-7 & ETOH 48
ACKNOWLEDGEMENT
This research was supported by NIH Grant NS07941-26.
32
REFERENCES
16
0
effects are mediated by substance P, which acts through NK 1 receptors. The facilitatory effects of the basal amygdala result, in part, from its excitatory effects upon neurons of the dorsal PAG from which defensive rage behavior could also be elicited. Such facilitation is mediated by excitatory amino acids that act upon NMDA receptors within the PAG. Suppressive actions of the amygdala upon defensive rage behavior are associated with the central amygdaloid nucleus. Its effects are manifest via an enkephalinergic pathway that supplies the dorsal PAG and acts through m receptors within the PAG. Finally, we further considered the application of these findings to the analysis of the potentiating effects of ethanol upon defensive rage. In these experiments, we demonstrated for the first time that the potentiating effects of ethanol upon defensive rage behavior are mediated, at least in part, over the descending pathway from the medial hypothalamus to the dorsal PAG.
5
60
120
180
Postinjection time (min) FIG. 13. Effects of microinjections of AP-7 upon ethanol-induced potentiation of defensive rage behavior elicited from the medial hypothalamus. Bar graphs indicate that microinjections of AP-7 into the PAG blocked the potentiating effects of ethanol administration upon defensive rage behavior to an extent similar to that observed when AP-7 was microinjected into the PAG in the absence of ethanol administration (lower panel). Solid bars (shown in upper panel) depict the potentiating effects of ethanol upon defensive rage behavior in the absence of AP-7 treatment (n ¼ 6). (From Schubert et al. (17).)
1. Banay-Schwartz, M.; Lajtha, A.; Palkovits, M. Changes with aging in the levels of amino acids in rat CNS structural elements. I. Glutamate and related amino acids. Neurochem. Res. 14:555–562; 1989. 2. Bandler, R. J.; Depaulis, A. Midbrain periaqueductal gray control of defensive behavior in the cat and the rat. In: Depaulis, A.; Bandler, R. J. eds. The midbrain periaqueductal gray matter. New York: Plenum Press; 1991:175–198. 3. Beitz, A. J. Possible origin of glutamatergic projections to the midbrain periaqueductal gray and deep layer of the superior colliculus of the rat. Brain Res. Bull. 23:25–35; 1989. 4. Brownstein, M. J.; Mroz, E. A.; Kizer, J. S.; Palkovits, M.; Leeman, S. E. Regional distribution of substance P in the brain of the rat. Brain Res. 116:299–305; 1976. 5. Brutus, M.; Shaikh, M. B.; Siegel, A.; Edinger, H. Effects of experimental temporal lobe seizures upon hypothalamically elicited aggressive behavior in the cat. Brain Res. 366:53–63; 1986. 6. De Olmos, J. S.; Ingram, W. R. The projection field of the stria
742
7. 8. 9. 10. 11. 12. 13.
14.
15. 16.
17.
18.
SIEGEL ET AL. terminalis in the rat brain. An experimental study. J. Comp. Neurol. 146:303–334; 1972. Fuchs, S. A. G.; Edinger, H. M.; Siegel, A. The organization of the hypothalamic pathways mediating affective defense behavior in the cat. Brain Res. 330:77–92; 1985. Hess, W. R.; Brugger, M. Das subkortikale Zentrum der affektiven abwehrreaktion. Helv. Physiol. Pharmac. Acta 1:33–52; 1943. Hopkins, D. A.; Holstege, G. Amygdaloid projections to the mesencephalon, pons and medulla oblongata in the cat. Exp. Brain Res. 32:529–547; 1978. Krettek, J. E.; Price, J. L. Amygdaloid projections to subcortical structures within the basal forebrain and brainstem in the rat and cat. J. Comp. Neurol. 178:225–254; 1978. Leyhausen, P. Cat behavior. The predatory and social behavior of domestic and wild cats. New York: Garland STPM Press; 1979. Lu, C.-L.; Shaikh, M. B.; Siegel, A. Role of NMDA receptors in hypothalamic facilitation of feline defensive rage elicited from the midbrain periaqueductal gray. Brain Res. 581:123–132; 1992. Neal, C. R.; Newman, S. W. Prodynorphin peptide distribution in the forebrain of the syrian hamster and rat: a comparative study with antisera against dynorphin A, dynorphin B, and the c-terminus of the prodynorphin precursor molecule. J. Comp. Neurol. 288:353–386; 1989. Neal, C. R.; Swann, J. M.; Newman, S. W. The colocalization of substance P and prodynorphin immunoreactivity in neurons of the medial preoptic area, bed nucleus of the stria terminalis and medial nucleus of the amygdala of the Syrian hamster. Brain Res. 496:1–13; 1989. Ranson, S. W.; Magoun, H. W. The hypothalamus. ERG Physiol. 41:56–163; 1939. Rao, R.; Yamano, M.; Shiosaka, S.; Shinohara, S.; Tohyama, M. Origin of leucine-enkephalin fibers and their two main afferent pathways in the bed nucleus of the stria terminalis in the rat. Exp. Brain Res. 65:411–420; 1987. Schubert, K.; Shaikh, M. B.; Pohorecky, L.; Han, Y.; Siegel, A. Differential effects of ethanol upon feline rage and predatory attack behavior: an underlying neural mechanism. Alc. Clin. Exp. Res. 20:882–889; 1996. Shaikh, M. B.; Barrett, J. A.; Siegel, A. The pathways mediating affective defense and quiet biting attack behavior from the midbrain central gray of the cat: an autoradiographic study. Brain Res. 437: 9–25; 1987.
19. Shaikh, M. B.; Lu, C. L.; Siegel, A. An enkephalinergic mechanism involved in amygdaloid suppression of affective defense behavior elicited from the midbrain periaqueductal gray in the cat. Brain Res. 559:109–117; 1991. 20. Shaikh, M. B.; Schubert, K.; Siegel, A. Basal amygdaloid facilitation of midbrain periaqueductal gray elicited defensive rage behavior in the cat is mediated through NMDA receptors. Brain Res. 635:187–195; 1994. 21. Shaikh, M. B.; Shaikh, A. B.; Siegel, A. Opioid peptides within the midbrain periaqueductal gray suppress affective defense behavior in the cat. Peptides 9:999–1004; 1988. 22. Shaikh, M. B.; Steinberg, A.; Siegel, A. Evidence that substance P is utilized in medial amygdaloid facilitation of defensive rage behavior in the cat. Brain Res. 625:283–294; 1993. 23. Siegel, A.; Brutus, M. Neural substrates of aggression and rage in the cat. In: Epstein, A. N.; Morrison, AR. eds. Progress in psychobiology and physiological psychology. San Diego: Academic Press; 1990:135–233. 24. Siegel, A.; Flynn, J. P. Differential effects of electrical stimulation and lesions of the hippocampus and adjacent regions upon attack behavior in cats. Brain Res. 7:252–267; 1968. 25. Siegel, A.; Pott, C. B. Neural substrate of aggression and flight in the cat. Progr. Neurobiol. 31:261–283; 1988. 26. Siegel, A.; Schubert, K. Neurotransmitters regulating feline aggressive behavior. Rev. Neurosci. 6:47–61; 1995. 27. Swann, J. M.; Macchione, N. Photoperiodic regulation of substance P immunoreactivity in the mating behavior pathway of the male golden hamster. Brain Res. 590:29–38; 1992. 28. Swann, J. M.; Newman, S. W. Testosterone regulates substance P within neurons of the medial nucleus of the amygdala, the bed nucleus of the stria terminalis and the medial preoptic area of the male golden hamster. Brain Res. 590:18–28; 1992. 29. Wasman, M.; Flynn, J. P. Directed attack elicited from hypothalamus. Arch. Neurol. 6:220–227; 1962. 30. Watson, R. E.; Troiano, R.; Poulakos, J. J.; Weiner, S.; Block, C. H.; Siegel, A. [14C]2-Deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. I. The amygdala. Brain Res. Rev. 5:1–44; 1983. 31. Zeichner, A.; Allen, J. D.; Giancola, P. R.; Lating, J. M. Alcohol and aggression: effects of personal threat on human aggression and affective arousal. Alc. Clin. Exp. Res. 18:657–663; 1994.