Neonatal isolation stress alters bidirectional long-term synaptic plasticity in amygdalo-hippocampal synapses in freely behaving adult rats

Abstract

The basolateral amygdala (BLA) is known to be involved in emotional and stress responses, while the dentate gyrus (DG), a subfield of the hippocampus, is implicated in learning and memory. Together, the BLA–DG neuronal pathway is thought to link memory with emotional and physiological stress responses. To assess whether neonatal isolation, a known early life stressor, has enduring effects on bidirectional neuroplasticity in adulthood, changes in long-term potentiation (LTP) and long-term depression (LTD) of BLA–DG synapses were recorded in neonatally isolated and non-handled freely behaving adult male rats. Rats isolated (ISO) from their mother and each other for 1 h daily from postnatal days 2–9 were allowed to mature to adulthood at which time they were chronically implanted with stimulating electrodes in the BLA and recording electrodes in the DG via stereotaxic surgery. A second group of rats which received no isolation treatment and which were not handled (NH) during the neonatal period underwent the same surgical procedures and served as the control group. Following a 1-week postsurgical recovery period, either LTP (100-pulse, 5-Hz theta-burst stimulation [TBS]) or LTD (900-pulse, 1-Hz low-frequency stimulation [LFS]) was induced in the DG of both groups. ISO rats showed significantly enhanced levels of both LTP and LTD compared to NH counterparts. These results indicate that neonatal isolation stress alters bidirectional neural plasticity in BLA–DG synapses, which may help to clarify the development of neural mechanisms linking emotional and stress responses in the amygdala with memory consolidation and information processing in the hippocampus.

Introduction

The hippocampus is a limbic structure closely implicated in learning and memory function. It has also been found to be highly susceptible to activity-dependent alterations in synaptic strength. Both long-term potentiation (an increase in synaptic efficacy – LTP) and long-term depression (a decrease in synaptic efficacy – LTD) are forms of activity-dependent bidirectional neural plasticity which have been proposed as candidate mechanisms underlying higher order brain functions, including learning and memory, formation and modification of cognitive maps, and the ability of organisms to cope with novel or stressful experiences (Bliss and Collingridge, 1993, Bear et al., 1987, Malenka and Nicoll, 1999, Abraham et al., 2001, Abraham and Williams, 2003). Early life experiences, whether a single event or chronically repeating incidents, are known to promote long-lasting and sometimes permanent changes in hippocampal function (Cirulli et al., 2003, Cordero et al., 2003, Pryce and Feldon, 2003, Radley and Morrison, 2005, Akers et al., 2006). Of particular interest are reports that stress experienced early in life during critical periods of neuronal development can lead to alterations in assembly of neural circuits involved in emotion and the regulation of the stress response and that these effects endure into adulthood (Cahill and McGaugh, 1998, Kalinichev et al., 2002, Card et al., 2005). For example, brief stressors, such as early handling (EH) of neonatal rats (daily 15-min separation sessions), tend to promote beneficial qualities in brain function, such as decreased emotionality (Holmes et al., 2005, Kim et al., 2006) and increased hippocampal synaptic plasticity (Liu et al., 1997, Fenoglio et al., 2006, Kim et al., 2006) that persist into adulthood, while longer periods of maternal separation (MS) tend to have aversive effects on emotionality (Avishai-Eliner et al., 2002, Holmes et al., 2005), hippocampal LTP and memory (Roceri et al., 2002). On the other hand, the effects of neonatal isolation (ISO) exhibit similar characteristics to both brief handling and longer bouts of MS, in that ISO subjects show an enhancement in hippocampal LTP, impairment in context-induced fear conditioning, inhibitory avoidance and object recognition and increased emotionality (Kehoe and Bronzino, 1999, Kosten et al., 2005, Kosten et al., 2007).

Emotional responses to stress are mediated by the amygdala, which includes the basolateral (BLA), lateral, central and basomedial nuclei (Isaacson, 1982). The BLA, in particular, plays a crucial role in modulating synaptic output during times of stress and is thought to be involved in fear conditioning (LeDoux, 2000, Blair et al., 2001, McGaugh et al., 2002, Richter-Levin, 2004). Specifically, during times of stress, stress hormones trigger the release of neuromodulators from the BLA; these neuromodulators can influence synaptic plasticity in many areas of the brain, including the hippocampus (Isaacson, 1982, McGaugh et al., 2002). Interestingly enough, the hippocampus itself plays an important role in terminating the stress response by providing negative feedback to the hypothalamus via activation of hippocampal corticosterone receptors (Swanson and Simmons, 1989, Liu et al., 1997, McGaugh, 2004, Avital et al., 2006, Fenoglio et al., 2006).

The synergistic way in which the BLA and the hippocampus works to mediate the stress response, the connection between these two brain regions and the ability of this pathway to support neuroplasticity have been the focus of many investigations (Akirav and Richter-Levin, 1999, Akirav and Richter-Levin, 2002, Frey et al., 2001, Roozendaal et al., 2003). For example, previous studies indicating that BLA lesions impair active-avoidance responses, inhibit memory consolidation of fear conditioning responses and attenuate hippocampal LTP support the idea that the BLA influences the formation of emotional memories (Horvath, 1963, Ikegaya et al., 1994, Korz and Frey, 2005). In addition, infusion of certain pharmacological agents in the BLA has been shown to enhance hippocampal-dependent memory, while infusion of other agents impairs it (McGaugh, 2004, Kim et al., 2005, Rodriguez-Manzanares et al., 2005).

Other investigators have also reported that priming of the BLA results in biphasic modulation of synaptic plasticity in the DG, such that prestimulating the BLA just prior to perforant path tetanization results in an enhancement of DG–LTP, whereas prestimulating the BLA in a longer time interval results in a reduction of DG-LTP. (Akirav and Richter-Levin, 1999, Akirav and Richter-Levin, 2002). Moreover, electrophysiological observations in BLA–DG synapses have indicated that single pulse stimulation of the BLA results in recording of evoked field potentials in the DG (Ikegaya et al., 1996, Abe et al., 2003, Nakao et al., 2004). The DG field potential disappears if the stimulating electrode is positioned outside the BLA or in other nuclei of the amygdala; and raising the recording electrode to the dendritic layer of the DG results in the reversal of polarity of the recorded waveform. Paired pulse analysis of these evoked potentials indicates that the BLA-DG connection is monosynaptic (Ikegaya et al., 1996). Furthermore, Ikegaya and colleagues (1996) have reported that BLA–DG-evoked field potentials are not affected by tetracaine inactivation of the perforant path and that the BLA–DG synapse displays LTP independently of the PP–DG pathway, which suggests that neural inputs from the BLA to the DG are not mediated by the perforant pathway (Ikegaya et al., 1996). In other studies, activation of the BLA by N-methyl-D-aspartate upregulates the expression of c-fos in the hippocampus (McGaugh et al., 1993). Together, these findings add to growing evidence that a direct functional link exists between the BLA and the hippocampus, which is independent of the entorhinal cortex.

Therefore, understanding the role of early life stress on hippocampal plasticity is crucial especially since approximately 80% of the adult complement of dentate granule cell population develops during the first 3 weeks of life (Bayer, 1980, Bayer, 1982, Kaplan and Bell, 1983, Boss et al., 1985, Boss et al., 1987, Crespo et al., 1986, Gage et al., 1998, Kempermann and Gage, 2000). Moreover, these late-generated neurons do not merely replace dying nerve cells but contribute to a progressive increase in the total population of dentate neurons making efferent and afferent connections within the previously established hippocampal network and thus may strategically enhance the ability of this network to process information of increasing complexity as the organism matures (Crespo et al., 1986, Gage, 2002, Gould et al., 1999, Kempermann, 2002, Snyder et al., 2001). Therefore, any severe early life experiences, such as neonatal isolation stress, would be expected not only to impact the normal maturation of hippocampal circuits involved in information processing and memory consolidation but also to affect the ability of animals later in life to effectively respond to a variety of routine as well as novel stimuli from their environment.

Despite the body of evidence that the BLA modulates hippocampal plasticity under normal physiological conditions as well as during times of stress, little is known concerning how stress experienced early in life may affect bidirectional plasticity in the BLA–DG synapse later in adulthood. In the present study, experiments were designed to determine whether neonatal isolation stress has enduring effects on the induction and maintenance of bidirectional long-term synaptic plasticity in the BLA–DG synapse in the freely behaving adult rat brain.