Amygdala kindling increased fear-response, but did not impair spatial memory in rats

Boldness suppresses hoarding behavior in food hoarding season and reduces over-wintering survival in a social rodent

The "pace-of-life" syndrome (POLS) framework can encompass multiple personality axes that drive important functional behaviors (e.g., foraging behavior) and that co-vary with multiple life history traits. Food hoarding is an adaptive behavior important for an animal's ability to adapt to seasonal fluctuations in food availability. However, the empirical evidence for the relationships between animal personality and hoarding behavior remains unclear, including its fitness consequences in the POLS framework. In this study, the Mongolian gerbil (Meriones unguiculatus), a social rodent, was used as a model system to investigate how boldness or shyness is associated with food hoarding strategies during the food hoarding season and their impact on over-winter survival and reproduction at both individual and group levels. The results of this study showed that, compared with shy gerbils, bold gerbils had a lower effort foraging strategy during the food hoarding season and exhibited lower over-winter survival rates. However, bold-shy personality differences had no effect on over-winter reproduction. These findings suggest that the personality is a crucial factor influencing the foraging strategy during the food hoarding season in Mongolian gerbils. Personality may be related to energy states or the reaction to environmental changes (e.g., predation risk and food availability) in bold or shy social animals. These results reflect animal life history trade-offs between current versus future reproduction and reproduction versus self-maintenance, thereby helping Mongolian gerbils in adapting to seasonal fluctuations in their habitat.

Anxious Profile Influences Behavioral and Immunohistological Findings in the Pilocarpine Model of Epilepsy

Anxiety and epilepsy have a complex bidirectional relationship, where a depressive/anxious condition is a factor that can trigger seizures which in turn can aggravate the depressive/anxious condition. In addition, brain structures such as the hippocampus and amygdala might have a critical relevance in both epilepsy and anxiety. The aim of the present work was to investigate the influence of different anxious profiles to epileptogenesis. Initially, animals were screened through the elevated plus-maze anxiety test, and then seizure development was evaluated using the pilocarpine model of epilepsy. There were no differences in the susceptibility to status epilepticus, mortality rate or frequency of spontaneous recurrent seizures between animals characterized as anxious as compared to the non-anxious animals. Next, we evaluated immunohistological patterns related to seizures and anxiety in various related brain areas. Despite a decrease in the density of neuropeptide Y and parvalbumin expression in epileptic animals, those presenting greater neuropeptide Y immunoreactivity in various brain regions, also showed higher spontaneous recurrent seizures frequency. Differences on the anxious profile showed to interfere with some of these findings in some regions. In addition, animals that were injected with pilocarpine, but did not develop status epilepticus, had behavioral and neuroanatomical alterations as compared to control animals, indicating its importance as an additional tool for investigating the heterogeneity of the epileptogenic response after an initial insult. This study allowed to better understand the association between anxiety and temporal lobe epilepsy and might allow for therapeutic targets to be developed to minimize the negative impacts associated with it.

Modelling Anxiety Disorders Following Chemical Exposures

The effects of kindling and inverse benzodiazepine receptor agonist -carbolines on animal models of anxiety are briefly reviewed in relation to affective disorder associated with chemical exposure. Recent experimental results are described. In the present study, cats were given the inverse benzodiazepine receptor agonist, FG-7142, a powerful anxiogenic compound in humans and animals. Neural transmission in pathways involved in defensive behavior in the cat was monitored using evoked potential techniques. Change in these pathways was related to behavioral changes induced by the drug. It was found that a single dose offG-7142 lastingly increased defensive response to rodents for at least 40 days after drug administration. Behavioral change was specific to defensive response, since approach-attack behavior remained unchanged, replicating previous studies. The benzodiazepine receptor antagonist, Flumazenil, reversed the increase in defensiveness in a drug-dependent manner, replicating previous findings. Increased defensiveness was paralleled by a delayed onset potentiation of neural transmission between the amygdala and the medial hypothalamus of the left hemisphere. Potentiation in the left hemisphere was transient, decaying between 6 and 12 days after the drug. There was a longer lasting potentiation (LTP) of activity evoked in the left and right amygdalo-periacqueductal gray pathways and in the right amygdalo-medial hypothalamic pathway. Potentiation in these pathways appeared at the time of behavioral change. Potentiation of the right amygdalo-periacqueductal gray and right amygdalo-medial hypothalamic pathways persisted until the end of the experiment. In contrast, potentiation of the left amygdalo-periacqueductal gray pathway faded by 40 days after the drug. Flumazenil decreased potentiation only in the right amygdalo-periacqueductal gray pathway. These data strongly suggest that lasting affective change is mediated by lasting changes in particular efferents of the amygdala of the right hemisphere. Behavioral and physiological effects offG-7142 were blocked by the N-methyl-D-Aspartate (NMDA) receptor blocker, AP7. The data suggest that failure of neural inhibition induced by FG-7142 engages NMDA receptor processes to produce lasting potentiation of transmission in neural circuits that mediate defensive response with behavioral consequences. Since FG-7142 interferes with GABA mediated neural inhibition and is proconvulsant, its action might mimic the action of other environmental chemicals with similar properties, such as chlorinated hydrocarbon insecticides. The relationship of the present data to the literature on the neural and behavioral effects of insecticide exposure is discussed. The significance of these findings for multiple chemical sensitivity disorder is also briefly discussed.

The effect of left and right long-term amygdala kindling on interictal emotionality and Fos expression

Clinical observations have often reported that patients with seizures arising from limbic structures on the right side of the brain have a higher incidence of emotional disturbances, such as fear and anxiety, than those who have seizures lateralized to limbic structures on the left side. However, there have been some inconsistent reports regarding the presence of these laterality effects. The use of animal models of epilepsy can help circumvent many of the methodological and ethical issues that arise from human clinical studies. In the present study, we examined the unique contribution of left- or right-sided long-term kindling of the amygdala on the development of interictal emotional disturbances. Following kindling to 99 electrical stimulations, male kindled and control rats were examined on a series of behavioral tests - open-field exploration, elevated plus maze, forced swim, and social interaction. Our results revealed that long-term amygdala kindling, irrespective of the hemisphere stimulated, increased general behavioral hyperactivity and fearful behavior. Interestingly, rats that were kindled from the left amygdala showed greater social avoidance and defensive behaviors during interactions with another kindled conspecific. To examine the brain structures that support long-term kindling, we also examined the expression of the immediate early gene product Fos 1 h after rats received their last electrical stimulation. Compared with control rats, kindled rats had increased Fos expression in several brain regions (e.g., piriform, frontal motor cortex, perirhinal cortex) involved in the generation and development of epilepsy. However, decreased Fos expression was also observed in several subregions of the hippocampus and amygdala that are known to be important fear behavior and memory. These findings suggest that both left and right amygdala kindling produce similar changes in emotional behavior and support the idea that the development of kindled fear may result from reduced activation of specific hippocampal and amygdaloid circuits.

Environmental enrichment protects against stress-induced anxiety: Role of glucocorticoid receptor, ERK, and CREB signaling in the basolateral amygdala

Environmental enrichment (EE) is an experimental animal model that enhances an animal's opportunity to interact with sensory, motor, and social stimuli, compared to standard laboratory conditions. A prominent benefit of EE is the reduction of stress-induced anxiety. The relationship between stress and the onset of anxiety-like behavior has been widely investigated in experimental research, showing a clear correlation with structural changes in the hippocampus and basolateral amygdala (BLA). However, the mechanisms by which EE exerts its protective roles in stress and anxiety remain unclear, and it is not known whether EE reduces the effects of acute stress on animal behavior shortly following the cessation of stress. We found that EE can prevent the emergence of anxiety-like symptoms in rats measured immediately after acute restraint stress (1 h) and this effect is not due to changes in systemic release of corticosterone. Rather, we found that stress promotes a rapid increase in the nuclear translocation of glucocorticoid receptor (GR) in the BLA, an effect prevented by previous EE exposure. Furthermore, we observed a reduction of ERK (a MAPK protein) and CREB activity in the BLA promoted by both EE and acute stress. Finally, we found that EE decreases the expression of the immediate-early gene EGR-1 in the BLA, indicating a possible reduction of neuronal activity in this region. Hyperactivity of BLA neurons has been reported to accompany anxiety-like behavior and changes in this process may be one of the mechanism by which EE exerts its protective effects against stress-induced anxiety.

Limbic but not non-limbic kindling impairs conditioned fear and promotes plasticity of NPY and its Y2 receptor

Epileptic seizures negatively affect cognition. However, the mechanisms that contribute to cognitive impairments after seizures are largely unknown. Here, we examined the effects of long-term kindling (i.e., 99 stimulations) of limbic (basolateral amygdala, dorsal hippocampus) and non-limbic (caudate nucleus) brain sites on conditioned fear and hippocampal plasticity. We first showed that kindling had no effect on acquisition of a hippocampal-dependent trace fear-conditioning task but limbic kindling impaired the retrieval of these fear memories. To determine the relationship between memory and hippocampal neuronal activity, we examined the expression of Fos protein 90 min after memory retrieval (i.e., 4 days after the last kindling stimulation). We found that limbic kindling, but not non-limbic kindling, decreased Fos expression in the granule cell layer, hilus, CA3 pyramidal cell layer, and CA1 pyramidal cell layer. Next, to investigate a mechanism that could contribute to dampen hippocampal neuronal activity in limbic-kindled rats, we focused on the endogenous anticonvulsant neuropeptide Y (NPY), which is expressed in a subset of GABAergic interneurons and can prevent glutamate release through interactions with its Y2 receptor. We found that limbic kindling significantly decreased the number of NPY-immunoreactive cells in several hippocampal subfields despite minimal staining of the neurodegenerative marker Fluoro-Jade B. However, we also noted that limbic kindling enhanced NPY immunoreactivity throughout the mossy fiber pathway. In these same regions, we observed limbic kindling-induced de novo expression of the NPY Y2 receptor. These novel findings demonstrate the site-specific effects of kindling on cognition and NPY plasticity, and they provide evidence that altered hippocampal NPY after limbic seizures coincides with dampened neural activity and cognitive impairments.

Amygdala kindling disrupts trace and delay fear conditioning with parallel changes in Fos protein expression throughout the limbic brain

Amygdala kindling is well known to increase unconditioned fear and anxiety. However, relatively little is known about whether this form of kindling causes functional changes within the neural circuitry that mediates fear learning and the retrieval of fear memories. To address this issue, we examined the effect of short- (i.e., 30 stimulations) and long-term (i.e., 99 stimulations) amygdala kindling in rats on trace and delay fear conditioning, which are aversive learning tasks that rely predominantly on the hippocampus and amygdala, respectively. After memory retrieval, we analyzed the pattern of neural activity with Fos, the protein product of the immediate early gene c-fos. We found that kindling had no effect on acquisition of the trace fear conditioning task but it did selectively impair retrieval of this fear memory. In contrast, kindling disrupted both acquisition and retrieval of fear memory in the delay fear conditioning task. We also found that kindling-induced impairments in memory retrieval were accompanied by decreased Fos expression in several subregions of the hippocampus, parahippocampus, and amygdala. Interestingly, decreased freezing in the trace conditioning task was significantly correlated with dampened Fos expression in hippocampal and parahippocampal regions whereas decreased freezing in the delay conditioning task was significantly correlated with dampened Fos expression in hippocampal, parahippocampal, and amygdaloid circuits. Overall, these results suggest that amygdala kindling promotes functional changes in brain regions involved in specific types of fear learning and memory.

Anxiogenic-like profile of Wistar adult rats based on the pilocarpine model: an animal model for trait anxiety?

RATIONALE

There is extensive evidence indicating the influence of seizures on emotional responses observed in human and animals, but so far few studies are focusing on the behavioral profile of animals that do not have seizures despite being treated with convulsant agents.

OBJECTIVES

We aimed to establish the behavioral profile, biochemical, and electrographic features of rats submitted to the pilocarpine model of temporal lobe epilepsy

METHODS

Rats treated with pilocarpine (20 to 350 mg/kg, i.p.) that did not develop status epilepticus or spontaneous recurrent seizures were evaluated 1 month later in the elevated plus maze (EPM), T-maze (ETM), open-field (OF), and step-down avoidance tests. Electroencephalographic (EEG), glutamate uptake, and hippocampal neuronal death assays were also performed

RESULTS

Pilocarpine (150 or 350 mg/kg) promoted anxiogenic-like effects in rats evaluated in the EPM, ETM, and OF tests, whereas only the highest dose evoked spike-wave discharges during EEG recordings. Hippocampal theta rhythm was increased by pilocarpine 150 or 350 mg/kg and only the highest dose reduced the L-[(3)H]-glutamate uptake and cell viability on hippocampal slices.

CONCLUSIONS

Subconvulsant doses of pilocarpine promote long-lasting alterations on neural circuitry, reflected by an increased theta activity in the hippocampus and an anxiety-like profile of rats evaluated 1 month after the treatment which is independent of seizure occurrence and is not related to changes in glutamate uptake or hippocampal damage. These results prompt us to suggest that a systemic administration of subconvulsant doses of pilocarpine could be useful as a new tool to model trait anxiety in rats.

Epilepsy and forgetfulness: one impairment, multiple mechanisms

Impaired Single Cell Firing and Long-term Potentiation Parallels Memory Impairment Following Recurrent Seizures. Zhou JL, Shatskikh TN, Liu X, Holmes GL. Eur J Neurosci 2007;25(12):3667–3677. Patients with epilepsy are at substantial risk for memory impairment. Animal studies have paralleled these clinical observations, demonstrating impaired hippocampal function as measured by spatial memory in rodents subjected to seizures. However, the mechanism of seizure-induced hippocampal impairment is unclear. Here we investigated the effects of recurrent seizures on water-maze performance, a behavioural measure of learning and memory, long-term potentiation (LTP; considered a test of synaptic plasticity and memory) and place-cell firing patterns, a single-cell indicator of spatial memory. LTP and CA1 place-cell activity were examined in separate groups of freely moving rats, before and after 10 flurothyl-induced seizures. Water maze performance was examined in a third group of rats, five with previously induced seizures and five controls. Recurrent flurothyl seizures were associated with marked impairment in LTP and a reduction in the frequency of the peak theta power. Compared to baseline recordings, place-cell firing patterns following recurrent seizures were significantly less precise, had lower firing rates and were less stable. Impaired place-cell firing was seen as early as after two seizures and persisted at least 72h after the last seizure. Water-maze performance was also significantly impaired in animals that underwent recurrent seizures. No cell loss or synaptic reorganization was observed in the hippocampus or in several other cortical areas that are vulnerable to seizures. These results demonstrate that relatively brief excitatory events, not producing visible cell damage, can nevertheless cause long-lasting changes in hippocampal physiology, observable as impairments in place-cell function, LTP and spatial memory.

Cortical kindling induces elevated levels of AMPA and GABA receptor subunit mRNA within the amygdala/piriform region and is associated with behavioral changes in the rat

Cortical kindling causes alterations within the motor cortex and results in long-standing motor deficits. Less attention has been directed to other regions that also participate in the epileptiform activity. We examined if cortical kindling could induce changes in excitatory and inhibitory receptor subunit mRNA in the amygdala/piriform regions and if such changes are associated with behavioral deficits. After cortical kindling, amygdala/piriform regions were dissected to analyze mRNA levels of NMDA, AMPA, and GABA receptor subunits using reverse transcription polymerase chain reaction, or rats were subjected to a series of behavioral tests. Kindled rats had significantly greater amounts of GluR1 and GluR2 AMPA receptor mRNA, and alpha1 and alpha2 GABA receptor subunit mRNA, compared with sham controls, which was associated with greater anxiety-like behaviors in the elevated plus maze and reduced freezing behaviors in the fear conditioning task. In summary, cortical kindling produces dynamic receptor subunit changes in regions in addition to the seizure focus.

Which clinical and experimental data link temporal lobe epilepsy with depression?

The association of temporal lobe epilepsy with depression and other neuropsychiatric disorders has been known since the early beginnings of neurology and psychiatry. However, only recently have in vivo and ex vivo techniques such as Positron Emission Tomography, Magnetic Resonance Imaging and Magnetic Resonance Spectroscopy in combination with refined animal models and behavioral tests made it possible to identify an emerging pattern of common pathophysiological mechanisms. We now have growing evidence that in both disorders altered interaction of serotonergic and noradrenergic neurons with glutamatergic systems is associated with abnormal neuronal circuits and hyperexcitability. Neuronal hyperexcitability can possibly evoke seizure activity as well as disturbed emotions. Moreover, decreased synaptic levels of neurotransmitters and high glucocorticoid levels influence intracellular signaling pathways such as cAMP, causing disturbances of brain-derived and other neurotrophic factors. These may be associated with hippocampal atrophy seen on Magnetic Resonance Imaging and memory impairment as well as altered fear processing and transient hypertrophy of the amygdala. Positron Emission Tomography studies additionally suggest hypometabolism of glucose in temporal and frontal lobes. Last, but not least, in temporal lobe epilepsy and depression astrocytes play a role that reaches far beyond their involvement in hippocampal sclerosis and ultimately, therapeutic regulation of glial-neuronal interactions may be a target for future research. All these mechanisms are strongly intertwined and probably bidirectional such that the structural and functional alterations from one disease increase the risk for developing the other. This review provides an integrative update of the most relevant experimental and clinical data on temporal lobe epilepsy and its association with depression.

Phenytoin normalizes exaggerated fear behavior in p-chlorophenylalanine (PCPA)-treated rats

Temporal lobe epilepsy may be associated with emotional difficulties such as depression and anxiety. Because the amygdala is involved in both epilepsy and emotion, common neural mechanisms in this temporal lobe structure may underlie the emotional disturbances observed in people with epilepsy. The neurotransmitter serotonin (5-hydroxytryptamine, or 5-HT) is implicated in many psychopathologies, and 5-HT also modulates amygdala excitability. Therefore, the present study uses the fear-potentiated startle (FPS) paradigm to investigate the effect of neuronal excitability on fear behavior in rats treated with p-chlorophenylalanine (PCPA) to chronically inhibit 5-HT synthesis. PCPA treatment selectively enhanced FPS in individually housed rats. The exaggerated FPS response was reduced to control level by the anticonvulsant phenytoin at 10mg/kg, and phenytoin at 30mg/kg further decreased FPS behavior. These data suggest that a subseizure state of neuronal excitability mediated by low 5-HT in brain fear circuits may be associated with pathological fear behavior.

Open-field behaviors and water-maze learning in the F substrain of Ihara epileptic rats

PURPOSE

Genetically epileptic model rats, Ihara epileptic rat (IER/F substrain), have neuropathologic abnormalities and develop generalized convulsive seizures when they reach the age of approximately 5 months. Because the neuromorphologic abnormalities are centered in the hippocampus, we expected to observe spatial cognitive deficits. The present study aimed to evaluate emotionality and learning ability of the F substrain of IER.

METHODS

To determine whether deficits are caused by inborn neuropathologic abnormalities or by repeated generalized convulsions, we tested nine 6- to 12-week-old IER/F rats that had not yet experienced seizures (experiment 1) and nine 7- to 9-month-old IER/F rats that had repeatedly experienced seizures (experiment 2) with identical tasks: an open-field test and the Morris water-maze place and cue tasks.

RESULTS

Both groups of IER/Fs showed behaviors that were different from those of control rats in the open-field test, and extensive learning impairments were seen in both the place task, which requires spatial cognition, and the cue task, which does not require spatial cognition but requires simple association learning. Their impaired performance of the cue task indicates that their deficiency was not limited to spatial cognition.

CONCLUSIONS

Because young IER/F rats without seizure experiences also showed severe learning impairments, genetically programmed microdysgenesis in the hippocampus was suspected as a cause of the severe learning deficits of IER/Fs.

Behavioral differences between subgroups of rats with high and low threshold to clonic convulsions induced by DMCM, a benzodiazepine inverse agonist

In epileptic patients, there is a high incidence of psychiatric comorbidities, such as anxiety. Gamma-aminobutyric acid (GABA) ionotropic receptor GABA(A)/benzodiazepine allosteric site is involved in both epilepsy and anxiety. This involvement is based on the fact that benzodiazepine allosteric site agonists are anticonvulsant and anxiolytic drugs; on the other hand, benzodiazepine inverse agonists are potent convulsant and anxiogenic drugs. The aim of this work was to determine if subgroups of rats selected according to their susceptibility to clonic convulsions induced by a convulsant dose 50% (CD50) of DMCM, a benzodiazepine inverse agonist, would differ in behavioral tests commonly used to measure anxiety (elevated plus-maze, open field) and depression (forced swimming test). In the first experiment, subgroups of adult male Wistar rats were selected after a single dose of DMCM and in the second experiment they were selected after two injections of DMCM given after an interval of 1 week. Those rats presenting full clonic convulsions were termed Low Threshold rats to DMCM-induced clonic convulsions (LTR) and those not having clonic convulsions High Threshold rats to DMCM-induced clonic convulsions (HTR). In both experiments, only those rats presenting full clonic convulsions induced by DMCM and those not showing any signs of motor disturbances were used in the behavioral tests. The results showed that the LTR subgroup selected after two injections of a CD50 of DMCM spent a significantly lower time in the open arms of the elevated plus-maze and in the off the walls area of the open field; moreover, this group also presented a higher number of rearings in the open field. There were no significant differences between HTR and LTR subgroups in the forced swimming test. LTR and HTR subgroups selected after only one injection of DMCM did not differ in the three behavioral tests. To verify if the behavioral differences between HTR and LTR subgroups of rats selected after two injections of DMCM were due to the clonic convulsion, another experiment was carried out in which subgroups of rats susceptible and nonsusceptible to clonic convulsions induced by a CD50 of picrotoxin, a GABA(A) receptor channel blocker, were selected and submitted to the elevated plus-maze and open field tests. The results obtained did not show any significant differences between these two subgroups in the elevated plus-maze and open field tests. In another approach to determine the relation between fear/anxiety and susceptibility to clonic convulsions, subgroups of rats were selected in the elevated plus-maze as more or less fearful/anxious. The CD50 for clonic convulsions induced by DMCM was determined for each of these two subgroups. The results showed a significantly lower CD50 for the more fearful/anxious subgroup, which means a higher susceptibility to clonic convulsions induced by DMCM. The present findings show a relation between susceptibility to clonic convulsions and fear/anxiety and vice versa which may be due to differences in the assembly of GABA(A)/allosteric benzodiazepine site receptors in regions of the brain.

Stress-induced enhancement of fear learning: an animal model of posttraumatic stress disorder

Fear is an adaptive response that initiates defensive behavior to protect animals and humans from danger. However, anxiety disorders, such as Posttraumatic Stress Disorder (PTSD), can occur when fear is inappropriately regulated. Fear conditioning can be used to study aspects of PTSD, and we have developed a model in which pre-exposure to a stressor of repeated footshock enhances conditional fear responding to a single context-shock pairing. The experiments in this chapter address interpretations of this effect including generalization and summation or fear, inflation, and altered pain sensitivity. The results of these experiments lead to the conclusion that pre-exposure to shock sensitizes conditional fear responding to similar less intense stressors. This sensitization effect resists exposure therapy (extinction) and amnestic (NMDA antagonist) treatment. The pattern predicts why in PTSD patients, mild stressors cause reactions more appropriate for the original traumatic stressor and why new fears are so readily formed in these patients. This model can facilitate the study of neurobiological mechanisms underlying sensitization of responses observed in PTSD.

Conditioned effects of kindling three different sites in the hippocampal complex of the rat

Rats received kindling stimulations to the perirhinal cortex (PRh), ventral hippocampus (VH), or dorsal hippocampus (DH) in 1 environment and an equivalent number of sham stimulations in a 2nd environment. The PRh-kindled rats displayed rapid kindling and a swift emergence of conditioned interictal defensiveness. In contrast, the VH- and DH-kindled rats displayed much slower kindling and slow or no conditioning, respectively. No effects of conditioning on the convulsions, comparable with those associated with amygdala kindling, were observed. These results establish the generality of some of the previously reported kindling-related conditioned effects, confirm the site specificity of some of these effects, and suggest that the convulsions, rather than the stimulations, function as the unconditioned stimuli for the conditioning of interictal behavior.

Anticipating the Attack: Temporal Conditioning During Amygdala Kindling in Rats

The present study showed that amygdala-kindled rats use short-interval timing superimposed on phase or ordinal timing to predict when a convulsion will occur. In 2 experiments, rats received 1 stimulation and 1 sham stimulation each day, always at the same times (conditioned stimulus [CS]+ and CS- times, respectively) and 150 s after rats had been placed in the testing chamber (the preadministration interval). As kindling progressed, the rats displayed more defensive behavior at the CS+ time than at the CS- time. Then, a stimulation-free peak-procedure test was conducted: At the CS+ time, but not at the CS- time, defensive behavior increased progressively as the 150-s preadministration interval elapsed, and then it gradually declined.

Amygdala kindling increases fear responses and decreases glucocorticoid receptor mRNA expression in hippocampal regions

Amygdala kindling dramatically increases fearful behavior in rats. Because kindling-induced fear increases in magnitude as rats receive more stimulations, kindling provides an excellent model for studying the nature and neural mechanisms of fear sensitization. In the present experiment, we studied whether the development of kindling-induced fear is related to changes in glucocorticoid receptor (GR) mRNA expression in various brain regions. Rats received 20, 60 or 100 amygdala kindling stimulations or 100 sham stimulations. One day after the final stimulation, their fearful behavior was assessed in an unfamiliar open field. Then, the rats were sacrificed and their brains were processed for in situ hybridization of GR mRNA expression. We found that compared with the sham-stimulated rats, the rats that received 60 or 100 kindling stimulations were significantly more fearful in the open field and also had significantly less GR mRNA expression in the dentate gyrus and CA1 subfield of the hippocampus. Importantly, the changes in fearful behavior were significantly correlated with the changes in GR mRNA expression. These results suggest that alterations in GR mRNA expression in hippocampal regions may play a role in the development of kindling-induced fear.

Kindling-induced emotional behavior in male and female rats

Modeling fear in animals is a critical approach for identifying the neural mechanisms involved in human disorders such as generalized anxiety and panic. Amygdala kindling has proven useful in this regard because it produces dramatic increases in fearful behavior. The purpose of this experiment was to compare the behavioral effects of kindling in male and female rats. Compared with the sham-stimulated rats, the kindled male and female rats showed similar increases in fearful behavior, with some sex differences in fear-related open-field activity. They also showed decreased immobility in the forced-swim test and increased sucrose consumption. These results suggest that kindling-induced fear is generally similar in male and female rats and that kindling does not appear to induce depression-like behavior.

Aversive conditioning following repeated withdrawal from ethanol and epileptic kindling

Repeated withdrawal from ethanol, a procedure which resembles amygdala kindling in increasing seizure sensitivity, impairs the acquisition of fear conditioning (Stephens et al., 2001, Eur. J. Neurosci.,14, 2023-31). In contrast, rats previously kindled by repeated electrical stimulation of basolateral amygdala, or repeated administration of pentylenetetrazol, showed increased suppression of operant responding during the presentation of a stimulus conditioned to footshock when conditioning took place several weeks following the kindling experience. Neither form of kindling nor repeated ethanol withdrawal altered taste aversion conditioning, though rats treated chronically with ethanol and given a single withdrawal experience showed enhanced taste aversion conditioning. These results suggest that, despite evidence suggesting a common neuronal mechanism underlying seizure sensitivity following these types of kindling, they differ in their effects on fear conditioning.

Conditioning of a flavor aversion in rats by amygdala kindling

Rats received 30 stimulations and 30 sham stimulations (the lead was attached to the subjects but no current was delivered) to the left basolateral amygdala in a quasirandom sequence. Stimulations were preceded by the presentation of 1 flavored solution conditional stimulus (CS+); sham stimulations were preceded by the presentation of another flavored solution, CS-. As kindled motor seizures developed, the rats began to consume significantly less of the CS+ than the CS-. Moreover, at the end of the experiment, the rats consumed significantly less of the CS+ than the CS- during a 20-min conditioned flavor preference test in which both solutions were available simultaneously. These findings confirm and extend the recent report that interictal changes in defensive behavior can be conditioned by amygdalar kindling.

Social attraction between rats in open field: long-term consequences of kindled seizures

Kindling of the amygdaloid complex in rats results in an enhanced emotionality frequently expressed by an elevated anxiety and defensive attitude toward other animals. Defensive attitude may have important consequences in social context and if tested in a large space it may eventually lead to social withdrawal. To test this hypothesis, rats were subjected to daily kindling sessions and their behavior was compared to implanted-sham and intact rats. Blood was collected after selected kindling trials for assessment of corticosterone response. Behavioral tests started 1 month after the last kindling trial and consisted of two open field sessions. A solitary rat was tested in the 1st session and pair of rats was tested simultaneously in the second session. Results showed that kindling changed the balance between exploration and occupation of a home base (HB) in the open field, in favor of higher preference of the home base occupancy. These results were apparent only during the social session leading to the conclusion that rats preferred to stay in the home base to maximize the proximity to a partner rat. This was supported by the observation that by increasing the occupancy of the HB, the kindled rats accomplished the longest concurrent presence with the partner rat in the common HB. We discuss the level of inter-rats aggression as a factor defining whether the anxious kindled rats will respond with increased or decreased social attraction in the open field test.

Hippocampal involvement in the expression of kindling-induced fear in rats

Kindling dramatically increases fearful behavior in rats. Because kindling-induced fear increases in magnitude as rats receive more stimulations, kindling provides a superb opportunity to study the nature and neural mechanisms of fear sensitization. Interestingly, these changes in behavior are accompanied by increased binding to inhibitory receptors and decreased binding to excitatory receptors in the CA1 and dentate gyrus regions of the hippocampus. This led us to hypothesize that kindling-induced fear may result from an increased inhibitory tone within hippocampal circuits. To test this hypothesis, we investigated FOS protein immunoreactivity in hippocampal and amygdalar regions of kindled rats that were exposed to an unfamiliar open field. We found that FOS immunoreactivity was significantly decreased in the CA1 region, dentate gyrus, and perirhinal cortex of kindled rats compared to sham-stimulated rats. These results support our hypothesis that kindling-induced fear may be produced by inhibition within hippocampal circuits. They also suggest that neural changes within the hippocampus may be important for the sensitization of fear.

Does the kindling model of epilepsy contribute to our understanding of multiple chemical sensitivity?

Multiple chemical sensitivity (MCS) is a phenomenon whereby individuals report an increased sensitivity to low levels of chemicals in the environment. Kindling is a model of synaptic plasticity whereby repeated low-level electrical stimulation to a number of brain sites leads to permanent increases in seizure susceptibility. Stimulation that is initially subthreshold for subclinical seizure provocation comes, over time, to elicit full-blown motor seizures. Kindling can also be induced by chemical stimulation, and repeated exposures to some pesticides have been shown to induce signs of behavioral seizure, facilitate subsequent electrical kindling, and induce subclinical electrographic signs of hyperexcitability in the amygdala. Many of the symptoms of MCS suggest that CNS limbic pathways involved in anxiety are altered in individuals reporting MCS. Limbic structures are among the most susceptible to kindling-induced seizures, and persistent cognitive and emotional sequelae have been associated with temporal lobe epilepsy (TLE) in humans and kindling in animals. Thus, a number of parallels exist between kindling and MCS phenomena, leading to initial speculations that MCS may occur via a kindling-like mechanism. However, kindling requires the activation of electrographic seizure discharge and has thus been primarily examined as a model for TLE. Events leading to the initial evocation of a subclinical electrographic seizure have been much less well studied. It is perhaps these events that may serve as a more appropriate model for the enhanced chemical responsiveness characteristic of MCS. Alternatively, kindling may be useful as a tool to selectively increase sensitivity in subcomponents of the neural fear circuit to address questions relating the role of anxiety in the development and expression of MCS.

Long-term amygdala kindling in rats as a model for the study of interictal emotionality in temporal lobe epilepsy

Temporal lobe epileptics often experience profound interictal (i.e. between seizure) emotional disturbances, such as fear, anxiety, and depression. Although the presence of this interictal emotionality has been well documented, little progress has been made in identifying its precise nature and cause because it is not amenable to experimental analysis in clinical populations. Accordingly, there is much to gain by studying the fundamental nature and neural basis of interictal emotionality using animal models. Kindling is a widely studied animal model of temporal lobe epilepsy in which daily electrical stimulation of certain brain regions results in the gradual progression and intensification of limbic motor seizures. Several investigators have found that partial and short-term kindling produce robust changes in emotional behavior in both cats and rats. Recently, our laboratory has developed a new model to study interictal emotionality using long-term kindling in rats. These long-term kindled rats display profound changes in fearful and defensive behavior which last for at least two months after the final stimulation. We are now beginning to use this model to study the neural mechanisms underlying the development and expression of interictal emotionality.

Neuroplasticity in specific limbic system circuits may mediate specific kindling induced changes in animal affect-implications for understanding anxiety associated with epilepsy

In two complementary experiments, we studied the effects of low frequency stimulation (LFS) of the amygdala on behavioral effects of kindling in rats and cats. These studies tested the hypothesis that kindling induced long term potentiation (KLTP) in amygdala circuits underlies interictal behavioral change. Since LFS can depotentiate LTP, it was predicted that LFS should both depotentiate KLTP and reverse behavioral effects of kindling. In cats, the effects of LFS on KLTP of amygdala efferents was studied, and related to behavioral effects. Partial ventral hippocampal kindling in cats and right amygdala kindling in rodents lastingly increased defensive response to rats in cats, and anxiety-like behavior (ALB) in the elevated plus-maze in rats. In addition, partial kindling reduced predatory attack behavior in cats independent of its effects on defensive response. Partial kindling also induced KLTP of amygdala efferent transmission to ventromedial hypothalamus (VMH) and periaqueductal gray (PAG) in left and right hemispheres. Depotentiation of amygdala efferent KLTP by bilateral amygdala LFS selectively reduced KLTP in right amygdala efferents. At the same time, defensive behavior, but not attack behavior, was returned to levels seen prior to partial kindling. Defensiveness returned to post kindling levels between 44 and 76days after LFS. At the same time, LTP was restored in the right Amygdalo-PAG pathway only. These findings suggest that lasting change in affect produced by kindling depends on LTP of right amygdala efferent transmission to PAG, replicating studies of the effects of FG-7142 on brain and behavior in the cat. The findings suggest further that the spectrum of behavioral changes produced by partial kindling are dependent on changes in a variety of neural circuits, and that amygdala efferent transmission changes are responsible for changes in defensive behavior, but not attack behavior. Effects of LFS were not due to damage, as thresholds to evoke amygdala efferent response were unchanged. Other data suggest KLTP and depotentiation in right Amygdalo-PAG may reflect changes in glutamate receptor density/synapse number. Kindling effects on rat ALB persisted for at least 1month. Bilateral but not unilateral amygdala LFS reversed the effects of kindling on risk assessment in the plus maze for at least 3weeks. Bilateral LFS also reversed the effects of kindling on open arm exploration, but effects were shorter lived, appearing at 1day but not 3weeks after kindling and LFS. These findings are consistent with other studies which suggest that amygdala neuroplasticity in separable amygdala circuits mediates lasting changes in open arm avoidance and risk assessment. Taken together, the findings of both studies support the hypothesis that a form of LTP of specific amygdala circuits underlies lasting changes in affect produced by limbic kindling. Clinical implications of these findings are discussed.

The mnemonic effects of kindling

Kindling produces enduring changes in the brain that are evident in not only enhanced susceptibility to seizure-evoking stimuli but also alterations in non-epileptic behaviors or functions. The present review examines the effects of kindling on one class of non-epileptic functions, learning and memory, and explores the dependence of these effects on variables such as the site of kindling, extent of kindling, and interval between kindling and testing. Current research shows that kindling is capable of altering performance on a variety of tasks including those that require spatial cognition, aversive conditioning, and object-related cognition and that non-mnemonic effects are unlikely, in at least some cases, to underlie these effects. Consideration of the conditions under which these effects are observed indicates a distinct relation between specific mnemonic effects and both the site and extent of kindling. Continued characterization of the mnemonic effects of kindling should provide a theoretical framework to guide discovery of their underlying mechanisms, which, in turn, may lead to rational therapy for mnemonic dysfunction associated with epilepsy and insights into the mechanisms of learning and memory.

Persistence of the interictal emotionality produced by long-term amygdala kindling in rats

Long-term amygdala kindling in rats results in large and reliable increases in emotional behaviour that model the interictal emotionality often observed in temporal lobe epileptics [Kalynchuk L. E. et al. (1997) Biol. Psychiat. 41, 438-451; Pinel J. P. J. et al. (1977) Science 197, 1088-1089]. These experiments investigated the persistence of these kindling-induced increases in emotional behaviour after the cessation of the kindling stimulations. In Experiment 1, rats received 99 amygdala or sham stimulations. Then, they were tested on three tests of emotionality (i.e. activity in an unfamiliar open field, resistance to capture from the open field, and activity in an elevated-plus maze) either one day, one week, or one month after the final stimulation. The rats tested one day after the last stimulation displayed substantial decreases in open-field activity, increases in resistance to capture and increases in open-arm activity on the elevated-plus maze; these effects decreased, but not to control levels, in the rats tested one month after the final stimulation. In Experiment 2, rats received 99 amygdala or sham stimulations, and their resistance to capture was assessed one day later. Then, after a 60-day stimulation-free period, the rats received another zero, one, 10, or 30 amygdala stimulations and their resistance to capture was reassessed one day later. The high levels of resistance to capture observed in the rats tested one day after the 99 stimulations declined significantly during the 60-day stimulation-free period, but it remained significantly above control levels. However, the administration of 30 additional stimulations reinstated asymptotic levels of resistance to capture. These results provide the first systematic evidence that kindling-induced increases in emotional behaviour persist at significant levels for at least two months following the termination of kindling stimulations. Thus, they suggest that the neural changes underlying the genesis of interictal emotionality may be closely related to those mediating epileptogenesis itself.

Long-term kindling and interictal emotionality in rats: effect of stimulation site

Long-term amygdala kindling in rats produces increases in emotionality (Kalynchuk et al., Biol. Psychiatry, 41 (1997) 438-451). The present experiment was conducted to investigate whether this hyperemotionality is specific to amygdala kindling or whether it can be produced by kindling other structures. Rats received 99 convulsive or sham stimulations of either the amygdala, the hippocampus, or the caudate nucleus. One day after the stimulation phase, each rat's open-field activity and resistance to capture were assessed; the following day, each rat was tested on an elevated plus maze. The site of stimulation had a significant effect on the results of each of these tests. The amygdala-kindled and hippocampal-kindled rats explored less in the open field, were more resistant to capture from the open field, and engaged in a greater percentage of open-arm activity in the elevated plus maze than did the caudate-kindled rats or the sham-stimulated controls. The caudate-kindled rats were more active in the open field than their sham-stimulated controls, but they did not significantly differ from them in terms of the other measures. These results suggest that kindling-induced emotionality is produced by limbic kindling but not nonlimbic kindling.

Effects of extended electrical kindling on exploratory behavior and spatial learning

Short-term electrical kindling, a widely used experimental model of epilepsy, appears to have little effect on behavior. The effects of extended kindling are largely unknown. Rats implanted with kindling electrodes in amygdala (AM) or perforant path (PP) received 300 kindling trials over approximately 7 months, and were tested in the Morris watermaze after a 7-10 day recovery period. Kindled animals were impaired during the initial training on hidden-platform acquisition, but not in retention of platform location. No deficits were found in acquiring a new hidden-platform location, latency to reach a visible-platform, or in swim speed. Open-field activity showed a sustained increase when tested during kindling, but only a transient increase when tested following suspension of kindling. Similar results were obtained for both AM and PP kindled animals. Hence, long-term kindling of both of these sites produced behavioral changes that were transient in nature. Further, these results also indicate that propagation of seizure activity from remote sites can alter hippocampally-mediated or related behavior.

Anxiogenic-like consequences in animal models of complex partial seizures

Several kinds of psychiatric symptoms (anxiety, depression, schizophrenia) have been associated with epilepsies, and clinical data suggest that patients with seizures involving limbic structures are the most prone to develop behavioural disorders between the seizures (i.e. interictally). Studying the neurobiological mechanisms that underlie these symptoms is difficult in humans because of different interfering factors (e.g. psychosocial difficulties, pharmacological side-effects, lesions), which can be avoided in animal models. Using repetitive electrical stimulations (kindling) or local applications of a neuroexcitotoxin in limbic structures (mainly the amygdala and hippocampus), several authors have reported lasting changes of emotional reactivity in cats and rats. These changes appear as anxiety-related reactions expressed as a hyperdefensiveness in the cat, or a reduction of spontaneous exploration in tests predictive of anxiogenic effects in the rat. Some neuroplasticity processes known to develop during epileptogenesis (neuronal-hyperexcitability, modulation of GABA/benzodiazepine transmission) may participate in these lasting changes of behaviour, especially in structures involved in the control of fear-promoted reactions (amygdala, periaqueductal grey matter). In addition, endogenous control systems may also play a critical role in the occurrence of interictal behavioural disorders.

Changes in emotional behavior produced by long-term amygdala kindling in rats

The effects of long-term amygdala kindling on emotional behavior were investigated. In Experiment 1, rats received 99 basolateral amygdala, central amygdala, or sham stimulations. The rats in both kindled groups displayed more resistance to capture from an open field and more open-arm activity on an elevated plus maze than did the sham control rats. In Experiment 2, rats received either 20, 60, or 100 amygdala stimulations or sham stimulations. Compared to the sham controls, the kindled rats explored less during the first 30s in a novel open field, avoided the central area of the open field, resisted being captured from the open field, and engaged in more open-arm activity on the elevated plus maze. The magnitude of these effects was greatest in the 100-stim rats and least in the 20-stim rats. Together, these results suggest that long-term amygdala kindling in rats is a useful model for studying the emotionality associated with temporal lobe epilepsy.

Amygdala kindling does not change emotional responding as measured by the acoustic startle response in the rat

Human patients with limbic epilepsy are often prone to anxiety and depression. The present study investigated the effect of seizures on emotional responding in the kindling model of complex-partial seizures. Male Wistar rats received electrodes into the right basolateral amygdala and were subsequently kindled until fully generalized seizures could be elicited. These rats did not show a change in the magnitude of the acoustic startle response (ASR) compared to the response amplitude before kindling, or compared to unimplanted controls and to only partially kindled rats. Since the ASR amplitude is a sensitive measure for anxiety or fear, these findings suggest that amygdala kindling does not induce a state of anxiety or fear. However, when analyzing the time course of the ASR within a session, the normally occurring habituation of the ASR was absent in electrode-implanted rats suggesting that the physiological changes induced by electrode implantation interfere with response habituation. Kindling even tended to sensitize the ASR in electrode-implanted rats. Finally, the effect of carbamazepine which is both anticonvulsant and antipsychotic was tested. Carbamazepine significantly reduced the ASR in control and partially kindled rats while this effect was less pronounced in fully kindled rats. It is concluded that amygdala kindling does not change emotional responding as measured by the ASR in rats.

Amygdala kindling in the rat: anxiogenic-like consequences

Patients with complex partial seizures of temporal lobe origin may experience behavioural disorders like depressive, anxiety-related or schizophrenic-like symptoms between seizures, i.e. interictally. The neural mechanisms underlying these enduring interictal disorders remain to be investigated. The aim of the present study was to examine the behavioural consequences of kindling of the basolateral nuclei of the amygdala, an animal model of limbic complex partial seizures. Animals having experienced 15 stage 5 seizures were compared to non-kindled controls in different behavioural tests performed at least seven days after the last seizure. Kindled animals showed a significant reduction of exploration of open arms in the elevated plus-maze test. In the social interactions test, they showed a decrease of non-social behaviour and an increase of immobility. No modifications were observed in kindled animals when tested in the open field, the sucrose preference or the forced swimming test. The reduction of open arm exploration in the elevated plus-maze was reversed by a pretreatment with chlordiazepoxide (2 mg/kg i.p.), a benzodiazepine anxiolytic. Finally, a similar reduction of open arm exploration was observed when animals were kindled only until a stage 3 seizure occurred. These data, along with previous studies, suggest that kindling of the amygdala has anxiogenic consequences and provide an animal model to study the neuroplasticity phenomena underlying enduring interictal disorders in humans.

Differential impairments of spatial memory and social behavior in two models of limbic epilepsy

To explore memory impairments in temporal lobe epilepsy, we used two experimental models in the rats: (a) kainate-induced status epilepticus (SE) resulting in excitotoxic damage and in later spontaneous seizures; and (b) amygdala kindling, known to induce no lesions (or only minor) and neuronal reorganization. Long-term effects of these models on memory were investigated with a spatial learning task in a radial-arm maze, and a social interaction test that implies degree of short-term memory. An histological analysis was made to determine neuronal damage or loss caused by epileptic activity in brain regions that could be related to memory functions. Kainate-induced epilepsy produced large memory deficits in animals tested 5 months after the injection. The rats showed severe lesions in amygdala and hippocampus and piriform and entorhinal cortex. Spatial memory was strongly diminished. The social memory test was severely impaired, probably due to the extent of amygdala injury, which is known to disturb social behavior. On the contrary, kindled rats showed no evident lesion in any brain region and displayed performances as good as those of controls in both tests. These experiments demonstrated that memory deficits appear to be related to the severity of neuronal damage in limbic areas, and the ability to develop seizures (permanence) is not solely responsible for these memory disturbances.

Enhanced susceptibility of audiogenic seizures in Fyn-kinase deficient mice

Mice with a mutation in fyn genes were examined for their susceptibility to acoustically primed audiogenic seizures. Homozygous mutant (fynz/fynz) mice were significantly more likely to have seizures and to show the stronger seizure syndrome (clonus). These results indicate that the susceptibility of acoustically primed audiogenic seizures is enhanced in the Fyn kinase deficient mice.

Increased fearfulness of Fyn tyrosine kinase deficient mice

Fyn-deficient mice were produced by inserting the beta-galactosidase gene (lacZ) into the fyn gene locus. The homozygously Fyn kinase-deficient (fynz/fynz) mice exhibited stronger light aversion in the light-dark choice test and higher fear-response scores in the novelty preference and passive avoidance tests than did the heterozygously Fyn-deficient (+/fynz) mice. These results indicate that fynz/fynz mice are hyperresponsive to fear-inducing stimuli.

Evidence for a GABAergic interface between cortical afferents and brainstem projection neurons in the rat central extended amygdala

The synaptic circuitry of the intrinsic GABAergic system of the central extended amygdala (CEA) in relation to efferent neurons and cortical afferents was examined in the present study. Neurons in the CEA projecting to the dorsal vagal complex and the parabrachial complex were identified by the retrograde transport of wheat germ agglutinin-horseradish peroxidase (WGA-HRP). Postembedding GABA-immunocytochemistry revealed that GABA-immunoreactive (GABA-IR) terminals formed largely symmetrical synaptic contacts with the perikarya and proximal dendritic processes of almost all WGA-HRP-labeled neurons in the CEA. To determine the relationship between cortical afferents and CEA GABAergic neurons, WGA-HRP was used to anterogradely label afferents from the insular cortex in combination with postembedding immunogold detection of GABA. Cortical afferents formed asymmetrical synaptic contacts predominantly on small dendrites and dendritic spines. Many of the dendrites postsynaptic to cortical terminals in the central nucleus were immunoreactive for GABA although only relatively few spines were GABA-IR. Combining pre-embedding GAD-immunocytochemistry with cortical lesions resulted in approximately 40% of degenerating terminals of insular cortical origin in the central nucleus in contact with small, GAD-IR dendrites and spines. The present results demonstrate that the neurons providing the major CEA outputs to the brainstem receive an extensive GABAergic innervation, strongly supporting our proposal that CEA efferent neurons are under strong tonic inhibition by intrinsic GABAergic neurons. Further, our finding that the major cortical input to the central nucleus preferentially innervates intrinsic GABAergic neurons suggests that these neurons in the CEA may serve as an interface between the principal inputs and outputs of this forebrain region.

The effect of kindling of different nuclei in the left and right amygdala on anxiety in the rat

The effects on rodent anxiety of kindling in the medial or basolateral amygdaloid nuclei in each hemisphere were examined. Anxiety was measured using the hole board and elevated plus maze tests. The animals were kindled in medial or basolateral amygdalas, of either the left or right hemisphere. Controls had electrodes implanted in comparable areas, but were not kindled. Analysis of electrode location showed that some animals were kindled in amygdaloid nuclei other than medial or basolateral amygdala. These animals were labelled outliers. Kindling of the medial/basolateral amygdala in the left hemisphere decreased anxiety for at least 1 week after the last kindled seizure. Right hemisphere medial/basolateral kindling tended to increase anxiety. Outlier-kindled rats were less anxious than their controls regardless of hemisphere 1 week after their last kindled seizure. Clear anxiogenic effects were not likely seen in the right hemisphere in this study because of the electrode locations. The degree of anxiety following kindling was correlated with electrode location in the anterior-posterior plane. More anterior foci in the amygdala were associated with more anxiety. More posterior amygdala foci were associated with less anxiety. These findings point to the importance of kindled focus in the amygdala for behavioral effect. Future research should carefully control the location of kindled foci when investigating effects of amygdala kindling on anxiety and other behaviors.

Amygdala kindling, anxiety, and corticotrophin releasing factor (CRF)

Wistar rats were kindled electrically in the anterior or posterior medial amygdala of the right hemisphere. One week after the fourth stage 5 seizure, anxiety was assessed in the elevated plus maze test. Anxiety levels of rats kindled in posterior medial amygdala were reduced relative to implanted controls, but not relative to unoperated controls. Kindling of the anterior medial amygdala increased anxiety relative to implanted and unoperated controls. The different effects of kindling on behavior were unrelated to any parameter of kindling. The stress of an ICV injection of saline increased anxiety in unkindled controls but reduced anxiety in anterior medial amygdala-kindled rats. Injection stress effects on behavior were blocked by 50 micrograms of alpha-helical CRF (the CRF receptor blocker). These findings suggest that CRF released by the stress of the injection procedure mediates the behavioral effects in both kindled and control rats. In contrast, injection of CRF (2 micrograms, ICV) has no greater effect than ICV saline in anterior medial amygdala kindled rats, whereas it was anxiogenic in unkindled rats. ICV vehicle and CRF reduce kindling-induced anxiety equally. These findings suggest that CRF released during the injection procedure saturates available CRF receptors. Finally, kindling did not alter basal plasma corticosterone levels. These and other findings suggest that the anxiety-modulating actions of CRF are at central CRF receptors.