NMDA systems in the amygdala and piriform cortex and nicotinic effects on memory function

Abstract

Both nicotinic cholinergic and NMDA glutaminergic systems are important for memory function. Nicotine has been found repeatedly to significantly improve working memory performance in the radial-arm maze. The NMDA antagonist dizocilpine has been found to impair working memory performance. There is neuropharmacological evidence that these two systems are functionally related. Nicotine is potent at releasing many transmitters including glutamate. The current study was conducted to examine the interaction of nicotinic and NMDA systems within the amygdala with regard to working and reference memory. Rats were trained on a working/reference procedure on a 16-arm radial maze. After acquisition, local infusion cannulae were implanted bilaterally into the amygdala and piriform cortex using stereotaxic techniques. Then 20 min prior to running the rats on the radial-arm maze, they were injected subcutaneously with (−) nicotine ditartrate at doses of 0 and 0.4 mg/kg. Following this, the rats received local infusions of (+) dizocilpine maleate (MK-801) at doses of 0, 2, 6 and 18 μg per side into the lateral amygdala or piriform cortex 10 min prior to running on the radial-arm maze. Each of the eight nicotine and dizocilpine combinations was administered to each rat in a counterbalanced order. After completion of the drug sessions the rats were sacrificed, and using histological methods the cannulae placements were verified. Acute amygdalar infusions of the NMDA glutamate receptor antagonist dizocilpine induced dose-related working and reference memory deficits in the radial-arm maze. Systemic nicotine was not seen to reverse these effects. Dizocilpine infusions into the adjacent piriform cortex did not impair memory function, supporting the specificity of dizocilpine effects in the amygdala. Latency effects were seen with both drugs in both areas. Latencies were decreased with both systemic nicotine and dizocilpine in both the lateral amygdala and the piriform cortex. This study demonstrated the importance of NMDA glutamate systems in the amygdala for appetitively-motivated spatial memory performance.

Introduction

Numerous studies have shown that both N-methyl-d-aspartate (NMDA) glutaminergic and nicotinic cholinergic receptor systems are important for memory function (for reviews see Refs. [7], [10], [11]). We have found that nicotinic–NMDA interactions are important inasmuch as systemic nicotine administration can substantially attenuate the memory impairment caused by systemic administration of the NMDA antagonist dizocilpine (MK-801) [16]. This nicotine-induced reversal of the memory impairment caused by dizocilpine may be related to the fact that nicotine stimulates glutamate release [23], [41]. The anatomic substrate of this interaction is not yet known. Determining the anatomic sites for nicotinic–NMDA interactions and memory function will help advance our understanding of the complex neural basis of cognitive function and will help in the development of novel avenues of treatment for cognitive dysfunction.

The NMDA antagonist dizocilpine causes substantial impairment of spatial learning and memory [36], [37]. Cole et al. [2] found that both 0.1 and 0.2 mg/kg of dizocilpine caused significant deficits in an operant delayed matching to position task in a delay-dependent manner. Ward et al. [39] showed that dizocilpine can produce a dose-dependent disruption of radial-arm maze memory performance. Acquisition of working and reference memory performance on an eight-arm radial maze was impaired by 0.0625 mg/kg of dizocilpine [37], but retention of performance was not affected. We have found working and reference memory impairments in the 16-arm radial maze caused by dizocilpine at doses as low as 25 μg/kg [16].

NMDA systems in the amygdala are important components for glutamate involvement in cognitive function. Pre-training intra-amygdala infusion of AP5 or AP7, other NMDA receptor antagonists, hinders retention of a condition startle response [26]. The immediate post-training intra-amygdala infusion of AP5 hinders retention of step-through inhibitory avoidance [19], of a mixture of active and inhibitory avoidance called continuous avoidance [9], and of a step-down inhibitory avoidance and a water finding task.

Nicotinic agonists have been found in rodent and non-human primate studies to improve performance on a variety of memory tasks. The 16-arm radial maze has been used to show the relative specificity of nicotine-induced improvement in working, but not reference, memory [14], [15]. In a complementary fashion, nicotinic antagonists such as mecamylamine have been shown to impair working memory function. A variety of studies have shown that acute treatment with nicotine or nicotinic agonists can improve working memory function in the radial-arm maze in rats [3], [13], [15]. An acute dose of 0.2 mg/kg nicotine injected subcutaneously 20 min before testing significantly improved working memory with the typical inverted U-shaped function for memory-enhancing drugs. This effect was specific to working memory (changing memory specific to each session). Reference memory (unchanging memory constant across sessions) errors were not significantly affected. It has also been found that acute treatment with other selective nicotinic agonists significantly improves memory performance [12].

Nicotine has been shown to attenuate the amnestic effects of dizocilpine. Dizocilpine (0.1 mg/kg) caused substantial deficits in both working and reference memory on the 16-arm radial maze [16]. This dose of dizocilpine eliminated the memory improvement caused by 0.2 mg/kg nicotine. However a higher dose of nicotine, 0.4 mg/kg, which is usually too high to cause a significant improvement in working memory performance, did cause a significant reduction in dizocilpine-induced deficits in both working and reference memory.

One likely point of nicotinic–NMDA interaction regarding memory is the hippocampus. There is a significant nicotinic interaction with hippocampal NMDA systems. However, this interaction is quite different from what we have seen on a systemic level. We found that local ventral hippocampal dizocilpine infusions reversed the systemic effect of nicotine on memory such that systemic nicotine increased working memory error rates on the radial-arm maze [18]. Interestingly, this reversal of the systemic nicotine effect was seen with ventral hippocampal dizocilpine doses that were not seen by themselves to impair memory function.

The amygdala is another area likely to be important for nicotinic–NMDA interactions and memory. The amygdala is known to play an important role in memory function [22]. However, little is known about the role of nicotinic systems in the amygdala regarding memory. There have been several studies that suggest cholinergic activity as the mechanism by which the amygdala modulates memory [24], [28], [34]. Both muscarinic cholinergic and nicotinic cholinergic systems have been shown to be involved [24], [28]. While the effects of cholinergic drugs on aversive tasks appear to involve the amygdala [34], cholinergic drug effects on spatial tasks do not seem to depend on the amygdala [34]. Several different studies have shown the amygdala to be responsive to nicotine exposure. Fos activity was increased upon nicotinic stimulation [35], [40].

The amygdala has been generally thought to be primarily involved in the processing of emotional memories [21], [29], [33], however, there is evidence that the amygdala plays a broader role in learning and memory function. The amygdala is also involved in non-aversively motivated memory. While most tests of amygdalar function in learning and memory have involved training with aversive stimuli, there is also evidence that the amygdala is important for learning tasks that use appetitive motivation when the reinforcement is of high affective value [6], [8], [31]. In some studies, such as delayed non-matching-to-sample and the three panel runway appetitive task, basolateral amygdala lesions in rats led to working memory impairments [28], [30]. Ohno et al. have shown working but not reference memory to be impaired by basolateral amygdala lesions [28]. Systemic treatment with indirect cholinergic agonists or direct muscarinic agonists improved the memory deficit. Aversive tasks have a different neural basis in the amygdala. Riekkinen and colleagues reported that the passive avoidance deficits caused by amygdaloid lesions were not reversible by nicotine or muscarinic treatment [34]. Not all studies have found impairments due to amygdala lesions. In some studies, such as spatial alternation, a lesion did not lead to any significant changes in performance [38].

The basolateral amygdala has high concentrations of NMDA glutamate receptors, and receives afferent glutaminergic projections from the cortex and thalamus [27]. Recent findings suggest that the NMDA receptor system in the amygdala is involved in aversively motivated learning. Microinfusion of an NMDA receptor antagonist, d,l-2-amino-5-phosphate (d,l-AP5), into the basolateral amygdala impairs one-trial inhibitory avoidance learning [19], Pavlovian conditioning and extinction of fear-potentiated startle [4], [26], blocking both acquisition and expression of contextual fear conditioning [5], [20]. Most importantly pre- and post-training intra-amygdala administration of NMDA receptor antagonists has been shown to impair memory storage [9], [19], [26].

The purpose of this study is to investigate the interactions between nicotinic and glutaminergic systems with respect to memory function. The amygdala may be a critical site for the significant interactions found between nicotine and the NMDA antagonist dizocilpine. The effect of dizocilpine infusions into the lateral amygdala or piriform cortex combined with systemic nicotine on the working and reference memory in rats was assessed using a 16-arm radial maze. The object of this study is to shed light on the question of how nicotinic systems interact with amygdalar glutamate systems in the neural foundation of memory function and to help in the development of new treatments for cognitive dysfunction.