During the past two decades, there has been an explosion of knowledge regarding neurotransmitter-receptor systems that modulate aggressive behavior in animals and humans. However, many studies have focused exclusively on the effects of a specific neuro-transmitter or receptor subtype on one or more aspects of violent or aggressive behavior. Integration of such fine-scale neurochemical data with the large-scale neurocognitive network data acquired in neuroanato-mical and neurophysiological studies has been limited. Most experimental and clinical reports on the "neu-rochemistry of aggression'' likely describe the effects of neuromodulators at peripheral, brain stem, hypotha-lamic, and diffusely projecting hemispheric sites, which can reduce or raise the individual's overall predisposition to aggression. Still lacking are studies that evaluate the multisynaptic integration of parallel-processed streams of complex sensory and limbic information that link amygdala, orbitofrontal, and other higher cortical centers. It is quite likely that these networks are subserved by diverse messenger systems and not exclusively controlled by a single neurotrans-mitter. We also have insufficient understanding of the complex interactive effects that neurotransmitters and neurohormones exert on one another or the specific receptor subtypes mediating a particular response. However, because neurochemical studies provide the basis for pharmacological interventions in aggressive patients, data from such studies are clinically invaluable.
The hypothalamus, amygdala, and frontal lobe are richly innervated by monoaminergic neurotransmit-ters, acetylcholine, and neuropeptides. Neurotrans-mitter systems strongly linked to mediation of aggressive behaviors in animal and human studies include serotonin, acetylcholine, norepinephrine, do-pamine, g-aminobutyric acid (GABA), and testosterone and other androgens as well as nitric oxide, opioids, and glutamate. Serotonergic systems appear to be particularly important and have been the subject of intense experimental and clinical investigation.
Diverse animal and human studies suggest that serotonin is a critical modulator of aggressive behavior. Experimental work in animal models of aggressive conduct supports a critical role of serotoninergic systems in hostile behavior. Eltoprazine, a 5-HT1 agonist, reduces attack behavior in several species. In the rat brain, eltoprazine binding is greatest in the dorsal subiculum, substantia nigra, ventral pallidum, and globus pallidus. Other 5-HT1A agonists, including buspirone and gepirone, reduce isolation-induced aggression in mice without causing sedation or incoordination. It has been proposed that serotonin increases the ability of an organism to arrange for and tolerate delay. Decreased serotonin leads to an increase of behaviors that are usually suppressed. Studies of isolation-induced aggression, shock-induced fighting, and muricidal and filicidal behavior have demonstrated an inverse relationship between serotonin activity and aggression in rats and mice. Recent studies have begun to delineate more complex species-specific and receptor-specific effects of serotonin on aggression. One recent rat study found that agonists at 5-HT1A, 5-HT1B, and 5-HT2 receptors all reduced offensive aggression, but only 5-HT2 agonists reduced defensive aggression.
Findings of studies of naturally behaving animal populations are consonant with the findings from experimentally induced aggression paradigms. Domesticated silver foxes, who easily tolerated human contact, had a higher level of midbrain and hypotha-lamic serotonin than wild silver foxes bred in captivity. The domesticated foxes also had a reduced density of 5-HT1A receptor binding in the hypothalamus. Rhesus monkeys with the highest blood levels of serotonin were socially dominant, whereas animals with decreased whole blood serotonin tended to be ostracized. Aggressive behavior was also associated with high levels of cortisol, suggesting greater stress. Adolescent male rhesus macaque monkeys show an inverse correlation between cerebrospinal fluid (CSF) levels of the serotonin metabolite 5-hydroxyindoleacetic acid (5-HIAA) and risk taking and aggression in the wild.
Serotonin interacts with other neurotransmitter and neurohumoral systems in modulating impulsivity and aggression. For example, one group of investigators examined the effects of testosterone and serotonin administration on dominance and aggression in rats. Male rats given testosterone became dominant. Qui-pazine, a serotonin agonist, blocked aggression in both naturally dominant and testosterone-induced dominant rats. Nonspecific serotonin antagonists blocked aggression only in testosterone-induced dominant males. This study demonstrates pharmacoselectivity among different forms of aggression, a property that may be desirable for the development of pharmacolo-gic treatments.
Data from human clinical studies are consistent with data from the experimental and observational animal literature, suggesting a critical role for serotonergic systems in human aggression. One major domain of investigation has focused on serotonergic markers in patients who have attempted or committed violent suicide. Caution must be exercised when interpreting these studies. The psychologic and biologic substrates of aggression against the self, manifest in suicide, are likely to differ from the underpinnings of aggression directed against others. Nonetheless, there are important behavioral affiliations between self-directed and outwardly directed violence and the two actions often cosegregate. Lifetime externally directed aggression is more frequent in suicide attempters than in others. Accordingly, findings in suicide patients offer important insight into general neurobiologic substrates of violent behavior.
Many postmortem studies have found decreased levels of serotonin and of presynaptic serotonin binding sites, such as the serotonin transporter site, in subcortical and neocortical brain regions in patients completing violent suicide. Recent studies suggest these abnormalities are more prominent in the orbital prefrontal cortex than in the dorsolateral prefrontal cortex. Similarly, postsynaptic serotonin receptors, such as 5-HT1A and 5-HT2A receptors, are generally elevated in the frontal cortex of violent suicides, especially in orbitofrontal sectors. These findings are consistent with a regulatory increase in postsynaptic serotonin receptors in response to decreased presy-naptic serotonergic activity.
Cerebrospinal fluid levels of 5-HIAA are decreased in patients attempting suicide and in violent criminal offenders. A stronger correlation has been noted between CSF 5-HIAA and suicidal behavior than suicidal ideation alone, suggesting that low CSF 5-HIAA levels are a marker not simply of depression and suicidal risk but also of a tendency to aggressive and impulsive behavior. Several studies in criminal and interpersonally violent psychiatric populations support this view. Lowered CSF 5-HIAA levels were found in samples of impulsive arsonists and impulsive murderers. In a group of soldiers with behavior problems, CSF 5-HIAA was negatively correlated with aggressive behavior, and CSF 5-HIAA was reduced in a group of borderline patients with aggressive and suicidal behavior.
Neuroendocrine challenge studies have been utilized to probe serotonergic function in aggression. Serotonin administration in normals causes a release of prolactin. Suicidal depressed patients and patients with personality disorders who exhibit impulsive and aggressive behavior have a blunted prolactin response to fenfluramine, a releaser of presynaptic serotonin stores, and to m-chlorophenylpiperazine, a 5-HT2 agonist. This work suggests that suicidal and impulsive/aggressive patients have serotonergic hypoactiv-ity. On the other hand, patients with depression and suicidality demonstrated an increased release of cortisol to 5-hydroxytryptamine, suggesting hypersensitiv-ity of other postsynaptic serotonin receptors in this group. Other workers have not found a significant correlation between neuroendocrine challenge abnormalities and suicidality.
Additional support for the influence of central serotonergic activity on aggressive propensities comes from studies of normal volunteers administered selective serotonin reuptake inhibitors. The resulting increase in central serotonergic activity correlated with decreases in hostility and in negative affect generally and an increase in affiliative behaviors.
At least 14 different receptors for serotonin exist in the human brain. Recent investigations ofthe relationship between serotonin and aggression have begun to more precisely dissect serotonergic systems by employing molecular probes of specific serotonergic receptor subtypes. Buspirone, a 5-HT1A agonist, produced a normal prolactin release when given intravenously to healthy male volunteers. This effect was blocked by the nonselective 5-HT receptor antagonist metergoline and by pindolol, a b-adrenergic and 5-HT1 antagonist, in a dose-related fashion. Prolactin response to buspirone was inversely correlated with levels of "irritability" in patients with personality disorders, suggesting that decreased sensitivity of the 5-HT1A receptor may be responsible for components of impulsive-aggressive behavior in patients with personality disorders.
Recently, genetic studies have provided further evidence of an important role of serotonin in aggression regulation. A polymorphism in the gene for tryptophan hydroxylase, the rate-limiting enzyme in the biosynthesis of serotonin, has been associated with suicidal behavior in impulsive alcoholic criminals and in individuals with major depression. Amino acid substitutions in the 5-HT7 receptor gene have been reported among alcoholics with violent behavior.
Some of the earliest work on the neurochemistry of aggression focused on acetylcholine. Electrical stimu lation of the lateral hypothalamus in rats leads to predatory attack on mice in animals that previously tolerated mice in their cage without attacking them. The attack terminates as soon as the electrical stimulation is discontinued. Applying carbachol, a cholinergic agonist, to the lateral hypothalamus provokes the sterotypic aggressive response, which can be blocked by atropine and facilitated by acetylcholines-terases. This cholinergic-induced predatory response is target specific—directed only at the animal's usual prey—and without affective display. Electrical stimulation of the lateral or dorsal amygdala facilitates predatory attack through its connections to the lateral hypothalamus. Applying carbachol to the amygdala also induces a predatory response. Aggressive behavior following human exposure to cholinesterase inhibitors has been observed in several clinical case reports. Despite well-documented early animal experimentation, cholinergic mediation of aggression and its clinical implications have been understudied in recent years. For example, the muscarinic receptor subtypes mediating hypothalamic aggression and cholinergic regulation of aggression in the frontal cortex have not been characterized in detail.
Catecholamine systems are associated with aggressive behavior in several animal models and clinical populations. Peripherally administered norepinephrine (NE) enhances shock induced fighting in rats. a2 receptor agonists increase rat aggressive behavior, whereas clonidine decreases rodent aggressive behavior acutely. b-Adrenergic blocking decreases aggressive behavior in laboratory animals. Several human studies have found a correlation between increased CSF or frontal cortex NE or its metabolite 3-methoxy-4-hydroxyphe-nylglycol and aggressive behavior. It has been proposed that central noradrenergic tracts originating in the locus coeruleus innervate a behavioral inhibitory system that projects widely to the hippocampus, septal region, and frontal lobes. Modulatory disturbances in central norepinephrine would then lead to impulsivity and episodic violence. Long-term b-adrenergic blockade with agents such as propranolol is a well-established, effective therapy to reduce aggressive responding in diverse neuropsychiatric patient groups with violent behaviors.
L-Dopa can induce aggressive behavior in rodents and humans. Apomorphine, a potent dopamine agonist, can induce fighting in rats. Dopamine antagonists tend to reduce aggression but usually at doses that also slow motor and cognitive performance. A few studies have shown reduced levels of a dopamine metabolite, homovanillic acid, in suicidal patients.
Recent genetic studies support an important role of catecholamine systems in human aggression. The genetic loci for MAO and catechol-O-methyltransfer-ase (COMT), two enzymes critical in the catabolism of catecholamines, are located on the X chromosome. In a large human kindred, impulsive violent behavior and mental retardation among several males co-segregated with a point mutation in the MAO type A gene that produced enzyme deficiency and presumably increased central catecholaminergic activity. Male knockout mice lacking the MAO-A gene also show aggressive behavior. COMT gene alleles include a common polymorphism that produces three- or four-fold variation in enzyme activity. The allele coding for the less active form of the enzyme (and resulting increased central catecholaminergic activity) has been associated with violent behavior in two studies of schizophrenic and schizoaffective patients and in male knockout mice.
D. /-Aminobutyric Acid
Several lines of evidence suggest that GABA inhibits aggression in animals and humans. GABA injected into the olfactory bulbs in rats inhibits mouse killing, whereas GABA antagonists can induce muricidal behavior. Benzodiazepines and other agents that facilitate GABA can decrease isolation-induced fighting in mice and attenuate aggression caused by limbic lesions. In humans, despite their tranquilizing and antiaggressive effect in the vast majority of patients, benzodiazepines can rarely lead to a transient increase in aggressive behavior ("paradoxical rage'').
Testosterone is an important mediator of aggressive responding in diverse mammalian species. In rats, dominant males have higher levels of testosterone than submissive males. Cortisol increases in both groups, but cortisol is higher in the submissive group, suggesting a greater level of stress. In vervet monkeys, increases in serum and salivary testosterone levels correlated with the number of aggressive encounters. Moyer suggested that androgens increased intermale and irritable, but not predatory, sexual, fear-induced, and maternal forms of aggression. An interaction between androgens and other neuromodulators such as the monoamine neurotransmitters appears to govern aggressive responding. Testosterone-induced dominance in rats is reduced after treatment with 5-HT1A, -1B, and -2A/2C receptor agonists.
The association of androgens with aggression suggested by the simple observation that males enact aggressive behaviors more frequently than females in most mammalian species, including humans, is supported by observations of convergent hormonal-behavioral relationships in female spotted hyenas. The spotted hyena is one of the most aggressive animals in the wild. Spotted hyenas also have a very organized and highly nurturant clan society. Male and female hyenas are approximatley equal in size, and female genitalia are masculinized. The colonies are dominated by the females in a tightly ranked hierarchy. Females are more aggressive than males, and adult males are usually not able to feed from a kill while the dominant females are eating. The females' large body habitus, androgenous genitalia, and aggressive behavior are related to the high circulating levels of androstenedione. The role of androgens in mediating aggression thus transcends sexual lines in this species.
In humans, numerous studies support an important link between circulating androgens and aggressive behavior. Elevated testosterone levels in adolescent boys correlate with low frustration tolerance and impatience. Boys with increased testosterone are more likely to respond aggressively when provoked. Increased levels of free testosterone have been measured in the saliva of incarcerated violent criminals. Victorious collegiate wrestlers show a greater increase in their serum testosterone than do their defeated counterparts. Violent behaviors have been reported in individuals taking anabolic steroids for body-building programs. Male alcoholics who abused other people had higher levels of testosterone and lower levels of cortisol than those who did not. In a treatment study, inhibiting gonadal function with a GnRH antagonist reduced outward-directed aggression. A meta-analysis of reported studies demonstrated a strong positive correlation between testosterone levels and observer-rated aggressiveness.
It is important to note some common weaknesses in the data currently available on the neurochemistry of aggression. Most studies on neurotransmitter and neurohormonal effects on aggression have been conducted on male animals and men. Endocrine and neurochemical factors influencing aggression in females have not been fully evaluated. Caution must be exercised when generalizing findings across species, particularly when comparing responses between humans, other primates, and other mammalian orders. Many aggression-related neurotransmitters are conserved across species, but there are likely important variations in receptor subsystems. Precision in defining and measuring aggression, impulsivity, and irritability in many animal models and humans is difficult to achieve. Also, many studies fail to recognize and fully clarify state-trait distinctions. Neuroendocrine challenge studies are subject to wide variability, depending on agent dosage, route of administration, and outcome measure.
Nonetheless, progress in basic investigations of the neurochemical and neuroendocrine mediators of aggression sets the stage for advances in the pharma-cotherapeutics of violent disorders. Moreover, once a better understanding of the individual neurochemical and neuroendocrine factors contributing to aggression is attained, interactions among the multiple systems operating convergently and divergently at hierarchial sites in the neuraxis that regulate hostile behavior may be more fully explored.
Was this article helpful?