Antagonism of lateral amygdala alpha1-adrenergic receptors facilitates fear conditioning and long-term potentiation

Norepinephrine receptors have been studied in emotion, memory, and attention. However, the role of alpha1-adrenergic receptors in fear conditioning, a major model of emotional learning, is poorly understood. We examined the effect of terazosin, an alpha1-adrenergic receptor antagonist, on cued fear conditioning. Systemic or intra-lateral amygdala terazosin delivered before conditioning enhanced short- and long-term memory. Terazosin delivered after conditioning did not affect consolidation. In vitro, terazosin impaired lateral amygdala inhibitory postsynaptic currents leading to facilitation of excitatory postsynaptic currents and long-term potentiation. Since alpha1 blockers are prescribed for hypertension and post-traumatic stress disorder, these results may have important clinical implications.Although norepinephrine (NE) has been widely studied as an important modulator of memory and emotion, comparatively little is known about the role of NE in amygdala-dependent Pavlovian fear conditioning, a major model for understanding the neural basis of fear learning and memory. In fear conditioning, an emotionally neutral conditioned stimulus (CS; i.e., tone) is temporally paired with an aversive unconditioned stimulus (US; i.e., footshock). After very few pairings, a lasting, robust CS–US association is acquired, and the CS elicits stereotypical defensive responses, including behavioral freezing (Blanchard and Blanchard 1969; Bolles and Fanselow 1980).The lateral nucleus of the amygdala (LA) is a key structure underlying fear conditioning (LeDoux 2000). Convergence of CS and US information in LA is believed to play an important role in initiating synaptic plasticity. Long-term potentiation (LTP)-like changes in LA CS processing are critical for fear memory storage (LeDoux 2000; Blair et al. 2001; Maren 2001; Walker and Davis 2002). LA receives auditory CS inputs from the thalamus and cortex and connects directly and indirectly with the central nucleus of the amygdala to control expression of Pavlovian fear responses.Of the noradrenergic receptor subtypes, alpha1 receptors have received the least attention in fear conditioning. LA receives NE-containing inputs from the locus coeruleus that fire tonically and phasically in response to aversive stimuli like footshock (Pitkänen 2000; Tanaka et al. 2000; Aston-Jones and Cohen 2005). Alpha1-adrenergic receptors are expressed in LA, most likely on both excitatory and inhibitory neurons (Jones et al. 1985; Domyancic and Morilak 1997). Alpha1 receptor activation stimulates GABA-mediated miniature inhibitory postsynaptic currents in LA (Braga et al. 2004), suggesting that alpha1 receptors contribute to inhibition in fear conditioning pathways. Several elegant experiments recently demonstrated that LA inhibition gates synaptic plasticity necessary for fear conditioning, and this inhibitory gate can be influenced by neuromodulators including NE (Stutzmann and LeDoux 1999; Shumyatsky et al. 2002; Bissière et al. 2003; Shaban et al. 2006; Shin et al. 2006; Tully et al. 2007). However, the role of alpha1 receptor activity in gating amygdala LTP and fear learning has never been examined.We hypothesized that alpha1 blockers would facilitate fear learning, possibly by impairing LA inhibition and facilitating LA LTP. To test this hypothesis, we injected rats with terazosin, a selective alpha1-adrenergic receptor antagonist, systemically or directly into LA before or after fear conditioning. We examined in vitro the effect of terazosin on LA pyramidal cell inhibitory postsynaptic currents (IPSCs) and excitatory postsynaptic currents (EPSCs) in response to fiber stimulation of the thalamic CS input pathway to LA, as well as the effect of terazosin on LA LTP in this same pathway. We found that intra-LA terazosin facilitated fear conditioning when injected before but not after training. We also found that terazosin impaired IPSCs in LA pyramidal cells, leading to facilitated EPSCs and LTP.Behavioral experiments were conducted on adult, male Sprague–Dawley rats (Hilltop Laboratory Animals) weighing approximately 300 g upon arrival. Rats were individually housed, maintained on a 12/12 h light/dark schedule, and allowed free access to food and water. Testing was conducted during the light phase. All procedures and experiments were approved by NYU''s Animal Care and Use Committee.For systemic injections, terazosin (20 mg/kg; Sigma) was dissolved in saline and injected intraperitoneally (i.p.) 30 min prior to conditioning in 1 mL/kg volume. For bilateral infusions, terazosin (125 ng/0.25 µL) was dissolved in aCSF and infused into the LA at 0.1 µL/min 30 min prior to or immediately after fear conditioning. Bilateral guide cannulae (22 gauge; Plastics One) aimed at LA (−3.3 mm anterior, 5.2 mm lateral, −7 mm dorsal to bregma) were surgically implanted as previously described (Sotres-Bayon et al. 2009). Rats were given at least 7 d to recover from surgery before testing. For infusions, dummy cannulae were removed, and infusion cannulae (28 gauge, +1 mm beyond guides) were inserted into guides. Infusion cannulae were connected to a 1.0 μL Hamilton syringe via polyethylene tubing. Infusion rate was controlled by a pump (PHD22/2000; Harvard Apparatus), and infusion cannulae were left in place for an additional 60 sec to allow diffusion of the solution away from the cannula tip, then dummy cannulae were replaced. Upon completion of the experiment, rats were euthanized, brains removed, and cannulae placements verified histologically as previously described (Sotres-Bayon et al. 2009).Two contexts (A and B) were used for all testing as previously described (Schiller et al. 2008). The grid floors in Context B were covered with black Plexiglas inserts to reduce generalization. No odors were used and chambers were cleaned between sessions. CSs were 30 sec, 5 kHz, 80 dB tones, and USs were 1 sec, 0.8 mA scrambled electric footshocks. Experiments consisted of two phases separated by 48 h: (1) fear conditioning in Context A and (2) long-term memory (LTM) test in Context B. On Day 1, rats were placed in Context A, allowed 5 min to acclimate, and then received three CS–US pairings separated by variable 5 min ITIs. On Day 3, rats were placed in Context B and allowed 5 min to acclimate before receiving one CS-alone presentation.The index of fear in behavioral experiments was “freezing,” the absence of all non-respiratory movement (Blanchard and Blanchard 1971; Fanselow 1980). Following testing, freezing was manually scored from DVDs by a scorer blind to group specification. Graphs represent group means ± SEM. Statistical analysis was conducted with GraphPad Prism.Whole-cell electrophysiological recordings were obtained from LA pyramidal cells using in vitro coronal slices from rats aged P21–P30 d as described in Cunha et al. (2010). Terazosin was bath-applied for 10 min to achieve stable responses before testing. The cells were voltage-clamped using an Axopatch 200B amplifier at −35 mV for recording EPSCs and IPSCs. Synaptic responses were evoked with sharpened tungsten bipolar stimulating electrodes. Internal capsule fibers containing thalamic afferents were stimulated for paired-pulse facilitation (PPF) (ISI = 50 msec; 0.1 Hz) using a photoelectric stimulus isolation unit with a constant current output. Cells were rejected if access resistance (8–26 MΩ) changed more than 15%. Signals were filtered at 2 kHz and digitized (Digidata 1440 A; Axon Instruments), and peak amplitude, 10%–90% rise time, and IPSC decay time constants were analyzed offline using pCLAMP10.2 software (Axon Instruments).Brain slices for LTP experiments were prepared from rats aged 3–5 wk as in Johnson et al. (2008) and maintained on an interface chamber at 31°C. Glass recording electrodes (filled with aCSF, 5 MΩ resistance) were guided to LA neurons. Bipolar stainless steel stimulating electrodes (75 kΩ) were positioned medial to LA in internal capsule fibers. Orthodromic synaptic potentials were evoked via an isolated current generator (Digitimer; 100 μsec pulses of 0.3–0.7 mA). Evoked field potentials were recorded with an Axoclamp 2B amplifier and Axon WCP software (Axon Instruments). Data were analyzed offline using WCP PeakFit (Axon Instruments). LTP was measured as a change in evoked field potential amplitude.Baseline responses were monitored at 0.05 Hz for 30 min with a stimulus intensity of 40%–50% of maximum fEPSP before LTP induction. Terazosin (10 µM) was superfused for 15 min, and then LTP was elicited by three tetanus trains (100 Hz × 1 sec, 3 min ITI) with the same intensity and pulse duration as the baseline stimuli. In one experiment, picrotoxin (PTX; 75 µM) was present in the perfusion solution to block fast GABAergic signaling.     

Context fear learning specifically activates distinct populations of neurons in amygdala and hypothalamus

The identity and distribution of neurons that are involved in any learning or memory event is not known. In previous studies, we identified a discrete population of neurons in the lateral amygdala that show learning-specific activation of a c-fos-regulated transgene following context fear conditioning. Here, we have extended these studies to look throughout the amygdala for learning-specific activation. We identified two further neuronal populations, in the amygdalo-striatal transition area and medial amygdala, that show learning-specific activation. We also identified a population of hypothalamic neurons that show strong learning-specific activation. In addition, we asked whether these neurons are activated following recall of fear-conditioning memory. None of the populations of neurons we identified showed significant memory-recall-related activation. These findings suggest that a series of discrete populations of neurons are involved in fear learning in amygdala and hypothalamus. The lack of reactivation during memory recall suggests that these neurons either do not undergo substantial change following recall, or that c-fos is not involved in any such activation and change.     

Food-deprivation-induced behavioral arousal: mediation by hypothalamus and amygdala.

Electrolytic lesions of the basomedial hypothalamus eliminated food-deprivation-induced stabilimeter activity in rats that were prevented from becoming obese. Knife cuts lateral to the basomedial area (separating the medial and lateral hypothalamus) potentiated this activity, as did transections posterior to the basomedial region. Anterior transections (between anterior and medial hypothalamus), however, eliminated the effect. Lesions of the stria terminalis and amygdala likewise abolished deprivation-induced locomotor activity, but elevated ab-lib activity to a level comparable with that after deprivation in intact animals. Animals with combined basomedial-stria terminalis lesions behaved like animals with basomedial lesions. These results suggest that food-deprivation-induced locomotor activity in stabilimeter cages is due to a disinhibition of the basomedial hypothalamus by the amygdala via the stria terminalis.     

Enhanced inhibitory avoidance learning produced by post-trial injections of substance P into the basal forebrain

The effects of injections of the neuropeptide substance P or the GABA agonist muscimol on performance of a step-down inhibitory avoidance task were examined. Immediately after the training trial, rats with chronically implanted cannulas were injected with 100 or 10 ng of substance P or 500 or 50 ng of muscimol into the region of the nucleus basalis magnocellularis. Control groups included vehicle-injected rats, a sham-operated group, a substance P 5-h delay group, and a substance P no-footshock group. Rats injected with 100 ng of substance P exhibited longer step-down latencies when tested 24 h later than did vehicle-injected rats. The retention latencies for rats in the substance P 5-h delay group did not differ from those of vehicle-injected animals, indicating that proactive effects on performance were not responsible for the effect. In contrast to injections of SP, injections of 500 or 50 ng of muscimol disrupted performance. However, in the absence of a delayed-injection control group, proactive effects cannot be ruled out.     

Post-trial injection of atropine into the caudate nucleus interferes with long-term but not with short-term retention of passive avoidance

One-trial passive avoidance training was given to Wistar rats and retention of the task was measured 30 min and 24 h later. Atropine (60 micrograms) was injected into the anterior caudate nucleus 2 min after training. Excellent retention was evident 30 min after training, whereas a significant deficit in memory was found when retention was tested 24 h after training. These results suggest that blockade of cholinergic activity of the caudate nucleus induced shortly after training interferes with the consolidation of long-term memory but not with short-term memory processes.     

Effects of lesions of the amygdala, preoptic area, and hypothalamus on estradiol-induced activity in the female rat.

Lesions of the anterior hypothalamic nucleus and the medial preoptic area sharply attenuated enhancement of wheel running by estradiol benzoate in ovariectomized female rats. Lesions of the corticomedial amygdala had no effect on this behavior. The hormonal effects on activity were largely independent of any changes in body weight. Results of this first experiment indicated that the anterior hypothalamic and medial preoptic areas are critically involved in the induction of activity by estradiol. However, this experiment provided no support for suggestions that the corticomedial amygdala inhibits those structures that mediate the estrogenic induction of activity. In the second experiment, food deprivation was used to stimulate activity. Results of this experiment suggested that the reduction in the ability of estradiol to induce activity following anterior hypothalamic and medial preoptic lesions does not reflect a general inability to become more active.     

Muscimol,AP5, or scopolamine infused into perirhinal cortex impairs two-choice visual discrimination learning in rats

The perirhinal cortex (PRh) has been strongly implicated in object recognition memory and visual stimulus representation. Studies of object recognition have revealed evidence for the involvement of several neurotransmitter subsystems, including those involving NMDA (N-methyl-d-aspartic acid) and muscarinic cholinergic receptors. In the present study, we assessed the possible involvement of PRh and related receptor subsystems in two-choice visual discrimination learning by Lister Hooded rats tested in touchscreen-equipped operant boxes. In Experiment 1, daily pre-training inactivation of PRh with the GABA_A receptor agonist muscimol (0.5 μg/hemisphere) significantly impaired acquisition of the two-choice visual discrimination. In Experiment 2, daily pre-training blockade of either NMDA or muscarinic receptors in PRh with AP5 (5.9 μg/hemisphere) or scopolamine (10 μg/hemisphere), respectively, impaired task acquisition. These results parallel the findings from object recognition studies and suggest a generality of neurotransmitter receptor involvement underlying the role of PRh in both object recognition memory and visual discrimination learning. The involvement of PRh in both types of tasks may be related to its role in complex visual stimulus representation.     

Amnesia by post-training infusion of glutamate receptor antagonists into the amygdala, hippocampus, and entorhinal cortex.

The blockers of glutamate receptors, aminophosphonovaleric acid (AP5) (5.0 micrograms) and cyano-nitroquinoxaline-dione (CNQX) (0.5 microgram), were infused bilaterally into the amygdala, dorsal hippocampus, or entorhinal cortex of rats through indwelling cannulae 0, 90, 180, or 360 min after step-down inhibitory avoidance training. Animals were tested for retention 24 h after training. In the amygdala or hippocampus, AP5 was amnestic when given 0 min after training and CNQX was amnestic when given 0, 90, or 180 min after training. In the entorhinal cortex, AP5 was amnestic when given 90 or 180 min after training and CNQX had no effect. The results suggest that a phenomenon sensitive first to AP5 and then to CNQX in the amygdala and hippocampus, probably long-term potentiation (LTP), is crucial to post-training memory processing. LTP in these two structures could underlie their role in memory consolidation and could explain the late involvement of the entorhinal cortex in post-training memory processing.     

Chronic sodium azide treatment impairs learning of the Morris water maze task.

A reduction in the activity of cytochrome oxidase, a respiratory chain enzyme, has been recently identified in mitochondria from blood platelets and postmortem brain tissue from Alzheimer's disease (AD) patients. We have developed an animal model of this deficit in rats by chronic subcutaneous infusion of sodium azide, a selective inhibitor of cytochrome oxidase, delivered via Alzet 2ML4 osmotic minipumps. In previous work, azide-treated rats were impaired in an appetitively motivated spatial learning task, the radial arm maze. In the present investigation, we tested male Sprague-Dawley rats (350-400 g), which were tonically infused with azide or saline, on an aversively motivated spatial task, the Morris water maze. Azide-treated rats were impaired on both acquisition and retention of this task, without showing evidence of a motor impairment. Thus, the present results are consistent with previous findings showing that chronic azide treatment produces a learning and memory deficit. These findings strengthen the hypothesis that azide treatment in rats produces a useful animal model of some aspects of AD.     

Effects of systemic glucose injection on the development of amygdala kindling in rats.

The effects of systemic glucose administration on the development of electric amygdaloid kindling seizures were examined in Wistar rats. Daily intraperitoneal injections of glucose 100 mg/kg, 20 min prior to each amygdaloid stimulation did affect the rate of kindling development in comparison with the saline-injected group. The number of stimulations required to reach stage 5 was lower in animals treated with glucose. The facilitation of kindling development found in glucose-treated rats is related to a specific decrease in the number of stimulations needed to evolve from stage 2 to stage 3. These results indicate that systemic glucose administration facilitates amygdaloid kindling in rats.     

Inactivation of the basolateral amygdala impairs the retrieval of recent and remote taste-potentiated odor aversion memory

Memory reorganization as a time-dependent process can be investigated using various learning tasks such as the taste-potentiated odor aversion (TPOA). In this paradigm rats acquire a strong aversion to an olfactory cue presented simultaneously with a gustatory cue. Together these cues are paired with a delayed visceral illness. The basolateral amygdaloid nucleus (BLA) plays a key role in TPOA acquisition but its involvement in retrieval remains unclear. We investigated the involvement of the BLA in either recent or remote retrieval of TPOA. In each case, the number of licks observed in response to the presentation of either the odor or the taste was used to assess retrieval. Before the retrieval test, rats received a bilateral infusion of lidocaine to inactivate the BLA. We observed that both recent and remote TPOA retrieval tests induced by the odor presentation were disrupted in the lidocaine-injected rats. By contrast, the BLA inactivation had no effect upon the aversion towards the taste cue regardless of the time of retrieval. The present study provides evidence that BLA functioning is necessary for retrieval of aversive odor memory, even with a long post-acquisition delay.