An important cellular function performed by neurons is the manufacture of chemical neu-rotransmitters. Neurotransmitters are produced in the cell body of specialized neurons by the endoplasmic reticulum, which is dispersed throughout most of the cytoplasm of the neuron. After manufacture, neurotrans-mitters are stored in vesicles produced by another cell structure called the Golgi apparatus. The vesicles containing the neurotransmitter are subsequently transported down the axon along microtubules and stored in the presynaptic terminal. This process is called axonal transport and includes both a slow and fast variety. Fast transport moves chemical transmitters quickly down the axon at a rate of 10 to 20 mm a day, whereas slow transport may move substances at a much slower rate of only about 1 mm per day. Axon transport takes place in both directions—both away from and back toward the cell body (Thompson, 1993).
As previously discussed, communication between neurons takes place at small gaps between neurons called synapses. Different chemical transmitters are involved, each possessing specific functions at different levels of neural organization. Peripheral neurons innervating skeletal muscle fibers act via the release of acetylcholine (ACh). The secretion of ACh into the synaptic cleft stimulates adjacent postsynaptic receptor sites to open ionic channels, resulting in the depolarization of the affected cell. The stimulative effects of ACh continue as long as it remains in the synapse. To open the synapse for additional transmissions, the receptor cell releases acetyl-cholinesterase (AChE), an enzyme that degrades ACh into acetate and choline. An interesting aspect of ACh in the body is that it exhibits an excitatory or an inhibitory effect depending on the muscle receptors involved. Skeletal muscles are excited by ACh, whereas the heart muscle is inhibited by it. Curare, a compound used experimentally to inhibit voluntary muscle activity, blocks the receptor sites for ACh in the skeletal muscles (resulting in paralysis) but has no effect on the heart muscle. Atropine, on the other hand, blocks the inhibitory effects of ACh on the heart muscle but has no discernible effect on skeletal muscles. Nicotine acts on skeletal muscle receptor cells in much the same ways as ACh. Sites sensitive to the excitatory effects of nicotine and ACh are referred to as nicotinic receptors. Muscarine (a poison derived from mushrooms) has an inhibitory effect much like that of ACh on the activity of the heart. As a result, ACh receptor sites that serve to slow the heart rate are called mus-carinic receptors.
Synaptic transmission within the brain is also mediated by neurotransmitters synthesized from various amino acids derived from dietary protein. Excitatory transmissions are conducted by glutamate, whereas GABA is responsible for inhibitory transmission across neural synapses. Unlike ACh, glutamate and GABA are not broken down by enzymatic actions within the synaptic cleft but are reabsorbed by the presynaptic terminal through a reuptake process called pinocytosis. During the reuptake process, the presynaptic membrane enfolds around the transmitter mole cule, drawing it back into the axon. Glutamate and GABA balance and check each other through a complex excitatory-inhibitory process of neural homeostasis. A complete loss of GABA in the brain would result in uncontrolled excitation and convulsions.
GABA has been implicated in the control of phobias and generalized anxiety disorders. Intense fear and anxiety problems in dogs are frequently treated with various benzodi-azepine preparations. Such anxiolytics appear to affect benzodiazepine-GABA receptors concentrated along fear circuits communicating between the amygdala and hypothalamus. Benzodiazepine receptors are closely associated with GABA receptors. Medications such as diazepam (Valium) appear to work by modifying the binding of GABA to its receptor, thereby amplifying receptor activity and reducing fear and anxiety by inhibiting activity in fear circuits (Panksepp, 1998). Mur-phree (1974) tested the effects of several common psychotropic drugs on the extreme anxiety reactions of genetically fearful pointers. Of the various drugs tested, which included phenobarbital, chlorpromazine, amphetamine, and alcohol, Murphree determined that the benzodiazepines were "far superior." Nervous dogs treated with benzodiazepines learned a bar-pressing response more quickly and performed the response at a higher rate than dogs not treated. Since benzodiazepines have specific receptor sites mediating their effect on fear and anxiety, it has been speculated that the brain itself produces anxiolytic substances much like the analgesic opioids (endorphins) are produced in response to pain. Like morphine, benzodi-azepines are potentially highly addictive.
Catecholamines: Dopamine and Norepinephrine
Another group of important neural transmitters are the catecholamines. Tyrosine (an amino acid) is converted through various chemical actions from l-dopa (l-3,4-dihy-droxyphenylalanine) to dopamine, NE, and lastly epinephrine. Each of these chemical changes requires the action of a specific enzyme. Some neurons possess the necessary enzymes needed to produce dopamine, whereas others have an additional enzyme for the synthesis of NE (Fig. 3.5). Although epi-nephrine is not produced in the brain, its production is under hypothalamic influence via the adrenal medulla.
Most dopamine is produced and distributed through three brain systems: (1) The ni-grostriatal system involves dopamine-produc-ing neurons originating in the substantia nigra of the midbrain, with axons projecting into the basal ganglia (a forebrain area involved in coordinated movement). (2) The mesolimbic system originates in dopamine-producing cells within the ventral tegmental area (located adjacent to the substantia nigra). Mesolimbic axons project to various regions via the MFB, including the amygdala, lateral septum, hypothalamus, hippocampus, and nucleus accumbens. (3) The mesocortical system also originates in the medial tegmen-tal area, with axons projecting to the limbic cortex (cingulate and entorhinal areas), pre-frontal cortex, and hippocampus. in addition, a fourth dopamine system communicates between the hypothalamus and the pituitary gland. Both mesolimbic and meso-cortical dopamine circuits have been implicated in the development of serious cognitive and behavioral disorders, such as schizophrenia (Kandel, 1991). It has been theorized that an affected person's brain contains either too much dopamine or too many receptor sites for dopamine activity. Phenothiazines are a class of major tranquilizers that bind with these receptor sites, thereby preventing dopamine from doing so. Chlorpromazine (Thorazine) is a commonly prescribed an-tipsychotic drug that functions specifically as a dopamine antagonist. On the other hand, depletion of dopamine can also result in serious problems, as observed in Parkinson's disease, which involves the second dopamine circuit (nigrostriatal) originating in the sub-stantia nigra, with projections terminating in the basal ganglia. Parkinson's disease results from the depletion of dopamine and the destruction of dopamine-producing neurons. The disease is associated with several motor deficiencies, including repetitive movement, tremors, and loss of coordinated movement. Parkinson's disease is treated with the dopamine precursor l-dopa. Dopaminergic circuits have been implicated in the development of compulsive disorders in dogs. Finally, dopamine plays a central role in the mediation of classical and instrumental learning. Reward experiences occurring as the result of either negative or positive reinforcement appear to be dopamine dependent. The reinforcement effects derived from appetitive stimuli, as well as those occurring as the result of the successful avoidance of aversive stimulation, are both interfered with when dopamine activity is blocked (Carlson, 1994).
NE circuits in the brain originate in neurons belonging to the locus coeruleus located in the brain stem. Axonal fibers extending from these NE-producing neurons project into all major structures of the brain. These
TYROSINE P Tyrosine Hydroxylase
DOPA ^ Dopa Decarboxylase
DOPAMINE (DA) -f- Dopamine ß-Hydroxylase
Fig. 3.5. Synthesis of catecholamines from dietary tyrosine.
diffuse projections contribute to the ARAS, providing a steady level of arousal or wakeful-ness within these divergent circuits and systems. NE axons often form synaptic terminals in a very different way than the basic pattern previously described. Instead of the conventional synapse, the NE axons form swollen protuberances along their surfaces. At each of these protuberances, NE is released as the action potential moving along the axon reaches these swellings. NE is reabsorbed through a reuptake mechanism. Among its many functions, NE is an excitatory transmitter of the ANS, stimulating increased heart rate and respiration during sympathetic arousal.
An important neurotransmitter in the neural economy of dogs is serotonin or 5-hydroxy-tryptamine (5-HT), which is especially important for the control of sleep cycles and has been implicated in the neurochemistry of stress, depression, and aggression. Specialized neurons manufacture serotonin from nutritional tryptophan (Fig. 3.6). Serotonin is stored in vesicles located in the presynaptic axon, and under appropriate stimulation, these serotonin-containing vesicles are released into the synaptic cleft. Serotonin molecules bind to specific serotonergic-receptor sites located on the postsynaptic neuron. Like other monoamines already discussed, serotonin is not broken down in the synapse like ACh but is recaptured through a reuptake mechanism. Excess amounts of serotonin are broken down by monoamine oxidase (MAO) within the presynaptic terminal. Serotonin-
producing neurons are located in the raphe nuclei located in the medulla, with projections into various parts of the brain. The raphe nuclei send serotonin-containing fibers to sleep-wake regulatory centers in the hypothalamus (suprachiasmatic nucleus), to the amygdala, hippocampus, septum, basal ganglia, and cerebral cortex. Besides controlling sleep-wake cycles, serotonin projections terminating in the limbic system play an important role in inhibiting anger and aggression. Further, serotonin directly attenuates the subjective experience of pain occurring during highly emotional displays involving anger or aggression, thereby mitigating against the effectiveness of physical punishment in the control of emotionally charged (affective) aggression.
Depression is often treated with drugs that either inhibit the reuptake of serotonin and NE or block the action of MAO—an enzyme that chemically breaks down the neurotrans-mitter. MAO inhibitors prevent the enzymatic breakdown of serotonin and other monoamines reabsorbed into the presynaptic terminal, thus making more of these substances available for use. Antidepressants like fluoxetine (Prozac) function to keep more serotonin in the synaptic cleft by selectively inhibiting its reuptake. Other antidepressants (tricyclics) like imipramine (Tofranil) and amitriptyline (Elavil) inhibit the reuptake of both serotonin and NE. The benefits of tri-cyclic medications on depression have led to theories implicating low levels of serotonin and NE in its development. Iorio and colleagues (1983) isolated a group of "depressed" beagles and tested various anxiolytic and psychotropic drugs on them. That re-
TRYPTOPHAN-► Tryptophan Hydroxylase
5-HYDROXYTRYPTOPHAN (5-HTP)-► 5-HTP Decarboxylase
SEROTONIN (5-Hydroxytryptamine, 5-HT)
Fig. 3.6. Synthesis of serotonin from dietary tryptophan.
search group found a significant improvement in 50% of the dogs exposed to imipramine, amitriptyline, and isocarboxazid (an MAO inhibitor). Interestingly, the dogs tested all exhibited a 2-week (10- to 17-day) delay from the onset of treatment to the appearance of signs of improvement. None of the dogs showed immediate improvement under tricyclic treatment, and all (except one) returned to baseline levels of depression when medication was withdrawn after 28 days.
More recently, Rapoport and colleagues (1992) demonstrated a connection between serotonergic activity and acral lick dermatitis (ALD), a compulsive disorder in dogs. A total of 42 dogs exhibiting compulsive licking were exposed to controlled trials involving various drugs, including clomipramine (Anafranil) and fluoxetine (Prozac). The results of the study showed that clomipramine (a tricyclic antidepressant) and fluoxetine [a selective serotonin reuptake inhibitor (SSRI)] were both effective against ALD, whereas the other medications tested were not beneficial. The authors speculate from research carried out by Jacobs and coworkers (1990) in cats that a specific serotonin subsystem in the dorsal raphe may be inappropriately activated by chewing and licking, thus implicating it as a potential neural site for ALD.
Clomipramine has also been shown to be an effective medication for the treatment of fear and generalized anxiety in companion dogs not responsive to behavior therapy (de-sensitization and counterconditioning) or previous treatment with anxiolytics (di-azepam) or other tricyclics lacking strong serotonin reuptake-blocking effects (Stein et al., 1994). The study involved five dogs of various ages and breeds presenting with symptoms of fear and generalized anxiety. All the dogs exhibited improvement (three of them much improved to very much improved) within 2 to 3 weeks under the influence of clomipramine. A previous study carried out by Tancer and colleagues (1990) evaluated the effects of imipramine (Tofranil, a related tricyclic drug) on 1 7 genetically nervous pointers but without much success. Imipramine is commonly prescribed for the control of panic disorder in humans. In the case of the nervous pointers, however, little sustained improvement was observed in the dogs treated orally with 50 mg given twice daily.
Several studies have implicated monoamines in the regulation of aggressive behavior (Siegel and Edinger, 1981). For example, quiet or predatory aggression is significantly reduced in animals by increasing the levels of NE in the hypothalamus and the medial nucleus of the amygdala. On the other hand, increased levels of NE stimulate affective hostility involving intruder-induced or pain-induced aggression. Eichelman and colleagues (1981) have reviewed the relevant literature regarding the biochemistry and pharmacology of aggression. Included was a series of studies by Reis (1972) demonstrating that electrically induced rage via the amygdala in cats results in the depletion of NE reserves in both forebrain and brain stem areas. Other studies of decerebrate cats have shown that electrical evocation of sham aggression results in a depletion of NE in the brain stem in proportion to the magnitude and duration of the rage evoked. This depletion is followed by a sharp increase of NE metabolism as evidenced by rising levels of tyrosine hydroxy-lase activity (Leventhal and Brodie, 1981). Tyrosine hydroxylase is the rate-limiting factor in the production of both dopamine and NE. The amount of this enzyme in the neuron determines how much NE it can produce. Sham rage is entirely suppressed in cases where catecholamine reserves are completely depleted and synthesis is chemically blocked. Lithium, a drug that reduces brain NE, attenuates shock-induced aggression, but this effect is confounded by a possible involvement of increased serotonin availability also caused by lithium.
Arons and Shoemaker (1992) studied the distribution of catecholamines (dopamine and NE) and beta-endorphin in different brain regions of three behaviorally distinct breeds: the Border collie, shar planinetz, and the Siberian husky. These breeds exhibit different predatory responses toward mice serving as prey, with the husky showing the most predatory and consummatory behavior and the shar showing the least predatory and con-summatory behavior toward mice. They found significant differences in the relative concentrations of some of the neural transmitters measured, suggesting that breed-specific behavioral differences may be related to underlying neurochemical differences obtained through selective breeding. For example, in the lateral hypothalamus, a site associated with quiet attack, shars exhibited a significantly lower concentration of dopamine than found in collies or huskies. Despite the evident breed differences in cate-cholamine concentrations, the authors note that complex behavior patterns like predation are probably governed by a complex interaction of many neurotransmitter systems. One particular neurotransmitter differentiation between the breeds studied seemed especially suggestive. An important trait difference between collies, shars, and huskies is their general activity and exploratory levels. Collies and huskies tend to be more active and interactive with their immediate environment than are shars. NE is frequently associated with arousal and general activity levels. Consequently, it is not surprising to find that collies and huskies exhibit a 40% to 60% higher level of NE than shars in important NE areas of the brain (e.g., the locus coeruleus, brain stem, and diencephalic areas). NE levels may provide an important biological marker correlated with general activity and exploratory levels in different breeds of dogs.
Although cholinergic pathways in the brain are not as well studied as monoaminer-gic pathways, some studies have shown a linkage between ACh and aggressive behavior. Injections of ACh placed in the ventricular system (fluid-filled areas inside the brain) result in affective aggression and rage in cats. Further, direct cholinomimetic stimulation (carbachol) of the amygdala also results in aggressive behavior in cats. Cholinergic agonists injected into the lateral hypothalamus of nonkilling animals induces quiet attack behavior. This predatory response is blocked by the cholinergic antagonist atropine (Eichel-man, 1987). Dopaminergic and beta-adrener-gic blockers do not suppress cholinergic-in-duced aggression (Leventhal and Brodie,
Increasing evidence suggests that the in-doleamine serotonin plays an inhibitory role over the exhibition of both predatory (quiet attack behavior) and affective aggression. Depletion of serotonin increases affective aggression in rats and quiet attack behavior in cats, whereas increased serotonin production reduces affective aggression in rats and reduces fighting behavior among isolated (usually more aggressive) mice. Recently, Reisner and colleagues (1996) demonstrated that the cerebrospinal fluid (CSF) of dogs exhibiting dominance-related aggression contains lower levels of serotonergic and dopaminergic metabolites than found in normal (nonaggressive) controls. Among the dominant-aggressive dogs studied, those that reportedly attacked without warning were found to have significantly lower concentrations of 5-hy-droxyindoleacetic acid (5-HIAA) and ho-movanillic acid (HVA) in their CSF than those dogs that gave warning before biting. The investigators suggest that this difference between dogs that warn and those that do not may indicate an impairment of a seroton-ergic-mediated impulse control mechanism modulating such aggressive displays. Also, dogs studied that had a history of biting hard (puncturing or lacerating the skin) tended to have lower concentrations of 5-HIAA and HVA than did dogs not delivering damaging bites. Interestingly, Popova and colleagues (1991) found significant differences in the serotonergic activity of human friendly versus human aggressive/defensive silver foxes. Foxes selected for tame behavior have greater amounts of serotonin and related by-products in their brain tissue, suggesting increased serotonergic activity. Popova and coworkers speculated that increased serotonergic activity may play an instrumental role in the process of domestication, serving to reduce aggressive tendencies and replacing them with more prosocial and tame ones. They found a similar pattern of increased serotonergic activity in tame versus wild Norway rats.
The influence of serotonin on aggressive behavior appears to be linked to the strong inhibitory effect that the neurotransmitter has over emotional processes and impulsive behavior. Stein and coworkers reported find ings indicating that a decrease in serotonergic activity results in "an inability to adopt passive or waiting attitudes, or to accept situations that necessitate or create strong inhibitory tendencies" (1993:10). Reducing the availability of serotonin by blocking its synthesis or available receptor sites negatively affects the suppressive effects of punishment, whereas the restoration of normal serotonin levels reverses this disinhibitory effect. Olivier and colleagues (1987) demonstrated strong inhibitory effects of serotonin-enhancing drugs on the frequency of various forms of aggression exhibited by mice and rats, including intermale aggression (mice), resident-intruder aggression (rats), isolation-induced aggression (mice), maternal aggression (rats), and mouse-killing behavior in rats. Especially strong inhibitory effects were observed in animals medicated with the serenics fluprazine and eltoprazine (serotonin agonists). Eltop-razine, in particular, exhibited very promising characteristics for the control of aggressive behavior. it not only inhibited a wide spectrum of aggressive behaviors, but seemed initially to be highly specific with few collateral side effects on other behavioral systems. Unfortunately, subsequent research seems to indicate that the aggression-reducing effects may be due to anxiogenic side effects. Dod-man (1998) found that although eltoprazine did reduce aggression, it also appeared to elevate anxious behavior in the two dogs treated. Other research (Kemble et al., 1991) seems to support the conclusion that serenics elevate social anxiety, thus making their use highly questionable in the control of aggressive behavior.
The apparent connection between enhanced serotonin activity and the inhibition of aggressive behavior has led to the widespread use of SSRIs and tricyclic antidepressant medications for the control of canine aggression problems, especially dominance-related aggression (Dodman et al., 1996a). Another drug found to show some promise for the control of dominance aggression in dogs is lithium (Reisner, 1994). Physiologically, lithium decreases NE turnover and inhibits tyrosine hydroxylase activity, thus affecting dopamine production. in addition, lithium produces an increase in blood levels of tryptophan; increases serotonin production in the brain, while at the same time inhibiting its metabolism; and, in general, enhances the aggression-inhibiting functions of the serotonergic system (Leventhal and Brodie, 1981).
The brain's production of serotonin depends on nutritionally derived tryptophan. Trypto-phan, like other precursor amino acids used in the manufacture of neurotransmitters, reaches the brain by passing through the blood-brain barrier. Research first carried out at the Massachusetts Institute of Technology under R. J. Wurtman has demonstrated that diets rich in protein tend to deplete brain tryptophan levels. This is a somewhat paradoxical finding, since tryptophan is a protein-forming amino acid and should be made more available to the brain as blood protein levels increase (Young, 1986). Even more paradoxical is a related finding that diets high in carbohydrates actually increase available tryptophan for serotonin synthesis, even if the food itself contains only modest amounts of tryptophan. The explanation for these apparent discrepancies involves two parts. (1) Naturally occurring tryptophan represents only a small proportion of the various amino acids making up protein (approximately 1% to 1.6%). The other larger and more prevalent amino acids all compete with tryptophan for a limited number of transport channels passing through the blood-brain barrier. The result of the foregoing biochemical scenario is that tryptophan is blocked out and the brain may be quickly depleted of available stores of the amino acid needed for the steady production of serotonin. (2) A more complicated metabolic process is needed to explain how a high-carbohydrate diet raises brain levels of tryptophan. Diets containing a proportionately higher level of carbohydrates than protein (at least 1 part protein to 5 to 6 parts carbohydrate) stimulates the secretion of insulin. An important effect of insulin production is its diversion of large neutral amino acids (other than tryptophan) into muscle tissue. Because of its unique molecular structure differentiating it from other amino acids, tryptophan is not similarly affected by the secretion of insulin. The outcome is that the proportion of plasma tryptophan is greatly increased, thus obtaining an advantage over other amino acids competing for transport through the blood-brain brain. As a result, the brain's production of serotonin is significantly increased.
For the increased movement of tryptophan to occur, the diet must be kept both low in protein and high in carbohydrates. In rats, a diet with protein levels exceeding 18% is sufficient to block the tryptophan effect (Spring, 1986). The exact percentages for dogs have not been determined but are assumed to be very similar (Dodman et al., 1996b). Unfortunately, these estimates have not been confirmed through appropriate physiological studies.
A common protein source in dog foods is corn. Corn, however, is unusually low in tryptophan and may represent some risk to animals sensitive to serotonergic underactiv-ity. Lytle and colleagues (1975), who studied the effects of a restricted corn diet on pain thresholds in rats, found that a diet restricted to corn as the primary source of protein results in a significant reduction of plasma and brain levels of tryptophan, with a subsequent decrease in the production of brain serotonin. Serotonin has an important analgesic effect on pain. Animals fed a restricted corn diet exhibit a lower threshold for pain (measured by the magnitude of a flinch or jump response to electric shock) than controls on a balanced amino acid diet of casein. Test subjects fed a tryptophan-rich diet or receiving an injection of tryptophan soon recovered from the hyperalgesic effect induced by the corn diet.
The foregoing studies are suggestive for the management of pain and aggressive behavior in dogs. Ballarini (1990) proposed that dietary protein be routinely adjusted as part of a comprehensive treatment program involving aggression in dogs. A study carried out by Dodman and coworkers (1996b) showed a promising linkage between reduced dietary protein and some forms of aggressive behavior in dogs. Dogs exhibiting territorial aggression with a strong component of fear-fulness responded beneficially to a reduced protein diet (17%), while territorial aggressors of the dominant type showed no significant change. The study, however, is not without possible flaws, perhaps accounting for its failure to show a stronger response than reported. Three problematic areas stand out: (1) protein levels were not kept sufficiently low, (2) carbohydrate levels may not have been high enough to induce an increased passage of tryptophan across the blood-brain barrier, or (3) the dogs may not have been exposed to the diet for a sufficient time. Behavioral effects from the diet were measured only after a relatively short period (2 weeks), so perhaps added benefits might be expected from a longer-term exposure (6 to 8 weeks). Also, Aronson has noted that, in addition to the diet's beneficial effect on fear-related territorial aggression, it is "possible that a more radical reduction in dietary protein levels would produce a reduction of dominance aggression and hyperactivity as well" (1998:80).
An important area of basic research is obviously wanting: a determination of the relative protein/carbohydrate proportions and percentages needed to induce (or block) tryp-tophan influx in dogs. Before any conclusions can be drawn with regard to the effect of low-protein diets on impulsive agonistic behavior in dogs, such questions will need to be explored and answered in detail. Furthermore, no study to date has directly implicated dietary tryptophan depletion in the causation of canine aggression or hyperactiv-ity, except by way of extrapolation from studies involving other animal species. Therefore, another important area of future research is determination of the effect of tryptophan depletion and supplementation on canine behavior. In a prototype study conducted by Chamberlain and colleagues (1987) in vervet monkeys, the monkeys were fed an identical diet except for the relative content of nutritional tryptophan. Three groups of monkeys were differentially fed diets containing normal tryptophan levels, high tryptophan levels, and low tryptophan levels. Although little benefit was seen with the provision of a higher percentage of tryptophan in the diet, a strong correlation was observed in terms of two parameters of aggression and the low-tryptophan diet. Monkeys fed a relatively low-tryptophan diet exhibited an increase of competitive aggression over food (dominance aggression) and spontaneous agonistic displays among themselves. The researchers also found a significant link between tryptophan depletion and an increase in general motor activity. Interestingly, in both cases, the observed behavioral effects of tryptophan depletion were restricted to male monkeys.
The level of tryptophan in the blood serum of assaultive alcoholics is at a lower than normal ratio to other amino acids, suggesting a possible connection between serotonin depletion in the brain and the exhibition of impulsive aggression among alcoholics. Morand and colleagues (1983) performed a pilot study with human patients to determine the effects of tryptophan on chronically aggressive schizophrenics. The study involved supplemental tryptophan at dosages of 4 to 8 grams a day. There was an approximately 30% reduction in the incidence of aggressive behavior while the patients received the tryptophan supplementation, but the response of patients was variable, with some becoming even more depressed and disorganized. Christensen (1996) wrote a critical review of the literature on the relationship between diet and behavior, providing a concise and objective summary of the current state of research in this important area.
Vasopressin has received considerable experimental attention, especially with respect to its influence over scent marking, dominance behavior, and affective aggression. Also known as antidiuretic hormone, vasopressin is a pep-tide hormone that controls water retention by the kidneys. In addition to this peripheral role, the hormone also appears to play a central neuromodulatory function over the expression of aggressive behavior. C. F. Ferris (University of Massachusetts Medical
School), who has studied the effects of argi-nine vasopressin (AVP) in golden hamsters for several years, found that the vasopressin-ergic system in the hypothalamus mediates the expression of several agonistic behavior patterns: flank marking (an AVP-dependent behavior), offensive aggression, and the formation of dominant-subordinate relationships (Ferris et al., 1986; Ferris and Potegal, 1988).
AVP receptors overlay androgen and estrogen receptors, suggesting that sex hormones and AVP may interact in the expression of aggressive behavior. In fact, the aggression-facilitating effect of AVP appears to depend on the presence of testosterone. Delville and coworkers (1996), for example, found that the hamster's behavioral response to microinjections of AVP varies depending on the presence or absence of testosterone. They showed that latency of attack is reduced by AVP microinjections into the ventrolateral hypothalamus (VLH), but only if the subjects are pre-treated with testosterone prior to injection. Although AVP regulates the onset and latency of aggression via the VLH, it does so without concurrently affecting the behavior's strength or number of bites delivered—a dimension of attack behavior that appears to be controlled by the selective activation of AVP receptors in the anterior hypothalamus (AH). This work suggests that the VLH and AH play different functional roles in the expression of aggressive behavior.
The regulation of aggressive behavior is more complicated than the interactions of testosterone and AVP acting directly on the hypothalamic vasopressinergic system. Besides AVP and sex hormones, researchers have discovered a robust interaction between AVP and serotonin in the hypothalamus (Ferris and Delville, 1994). Both the ventrolateral hypothalamus and anterior hypothalamus exhibit a high concentration of serotonin-bearing axon terminals and binding sites. Interestingly, fluoxetine (Prozac) injected peripherally inhibits AVP-induced offensive aggression and retards the onset of resident-intruder attacks, with fewer bites occurring during the attacks (Ferris and Delville, 1994; Ferris et al., 1997). These studies suggest that serotonin directly modulates AVP neurons in the hypothalamus, thereby antagonizing AVP-system-facilitated aggression.
There are many potential implications of this work for dogs. Until recently, progestins were commonly used for the control of unwanted aggression and marking behavior. The most frequently mentioned target site of progestin action is the hypothalamus, perhaps including the targeting and disruption of AVP activity. An antivasopressinergic link would appear logical, since progestins produce a diminution of both urine marking and aggressive behavior in treated animals. More recently, the veterinary use of fluoxe-tine has become increasingly popular for the control of unwanted behavior, especially dominance-related aggression and various compulsive disorders. Fluoxetine is rarely prescribed for intraspecific or territory-related aggression; given the findings of Ferris, though, perhaps such a wider use might prove very beneficial, especially in cases of refractory dog fighting and territory-related aggression. Lastly, serotonin-enhancing drugs may play a beneficial role in the control of household urine marking by dogs.
Was this article helpful?