This thesis aims to examine the role of the hippocampus in declarative memory through the development of animal behavioural models of episodic memory for laboratory rats. Episodic memory- memory for unique events or episodes- is part of the declarative memory system thought to be mediated by the medial temporal lobe area of the brain in humans. One commonly used test of episodic memory in human subjects is paired associate learning. The first part of this thesis describes the adaptation of this human test for use with laboratory rats. Using their natural foraging tendency, rats were trained to search for different flavours of food at different locations within a large enclosure. When cued with a piece of food of a particular flavour in a separate box, rats learned to return to the place where that flavour of food had previously been found. This paradigm was used to investigate the role of the hippocampus in paired-associate learning using temporary pharmacological inactivation and permanent neurotoxic lesion techniques. The hippocampus has also been strongly implicated in spatial navigation, learning and memory in rats and humans. In the experiments described previously, attempts were therefore made to demonstrate that the results were not confounded by a simple deficit in spatial navigation. An alternative approach to studying episodic memory in the laboratory rat is to use the criteria established by Tulving in 1972 to describe episodic memory. He stated that episodic memory should encompass the memory for an event and the spatiotemporal context in which it occurred, i.e. the ”what”, ”where” and ”when” of an event. He later updated these criteria to include demonstration of autonoetic consciousness- most easily described as a sense of self awareness. Since this is difficult or impossible to demonstrate in animals, the term ”episodic-like” memory was coined (Clayton & Dickinson 1998) to describe the flexible use of information about the spatiotemporal aspects of an event by non-human species. Since it has been difficult to demonstrate the use of time (when) in rats (Bird et al; 2003, Babb & Crystal 2006a), Eacott & Norman (2004) suggested that the ”when” component could be replaced by context; i.e. another element specific to a particular event that they labelled ”which”. The next part of this thesis describes the use of the task published by Eacott & Norman to test episodic-like memory in the laboratory rat. Using the innate spontaneous behaviour of rats to explore novel aspects of their environment, they were exposed to multiple unique events. These consisted of various three-dimensional objects being presented in different locations within different contexts. Their memory for manipulations of the environment was then tested by presenting them with an event in which one combination of object, location and context was different from combinations which had previously been encountered. Due to their tendency to explore novel aspects of their environment, normal rats spent the majority of their time exploring the object that was in a novel location relative to the context in which it was presented. This successfully demonstrated integrated memory for what, where and which- similar to that previously defined by Tulving. The rats also showed that they could use this information flexibly because every trial involved unique combinations of objects, locations and contexts so there was no inadvertent semantic rule-learning involved. Permanent neurotoxic lesions of the hippocampus were used to determine the extent to which this structure is involved in memory for the what, where and which of an event. The experimental results presented in this thesis demonstrate an indisputable role for the hippocampus in a variety of tasks designed to parallel episodic memory in humans. The next steps in this line of research should involve characterisation of the roles of the various subregions of the hippocampus in episodic-like and paired associate memory.

Cerebral ischemia is a highly debilitating condition where shortage of oxygen and glucose leads to profuse cell death. Insufficient blood supply to the brain leads to cerebral ischemia and increase in extracellular lactate concentrations. Rise in lactate concentration and the leak potassium channel TREK1 have been independently associated with cerebral ischemia. Lactate is a neuroprotective metabolite whose concentrations increase to 15-30 mM during ischemia and TREK1 is a neuroprotective potassium channel which is upregulated during ischemia. Recent literature suggests lactate to be neuroprotective and TREK1 knockout mice show an increased sensitivity to brain and spinal cord ischemia, however the connecting link between the two is missing. We hypothesized that lactate might interact with TREK1 channels and mediate neuroprotection. The aim of this study was to investigate the effect of lactate on activity and expression of TREK1 channels and evaluate the role of lactate-TREK1 interaction in conferring neuroprotection in the ischemia-prone hippocampus Ischemic concentrations (15-30 mM) of lactate at pH 7.4 increased whole cell TREK1 current in CA1 stratum radiator astrocytes and caused membrane hyperpolarization. We confirmed the intracellular action of lactate on TREK1 in hippocampal slices using mono carboxylate transporter blockers. The intracellular effect of lactate on TREK1 channels is specific since other mono carboxylates such as pyruvate at pH 7.4 failed to increase TREK1 current. We used immunostaining, western blot and electrophysiology to show that 15-30 mM of lactate increased functional TREK1 protein expression by 1.5-3 fold in hippocampal astrocytes. Next, we performed quantitative PCR to investigate if the increase in TREK1 protein expression was due to increased transcription and found that lactate stimulated TREK1 mRNA transcription to increase TREK1 protein. Lactate mediated increase in TREK1 expression was dependent on protein kinase A as inhibitors of protein kinase A abolished the increase in TREK1 mRNA and protein. The role of lactate-TREK1 interaction in neuroprotection was subsequently investigated using an in vitro oxygen glucose deprivation model of ischemia. Addition of 30 mM lactate to oxygen glucose deprived slices reduced neuronal death in the hippocampal CA1 pyramidal layer. However, 30 mM lactate failed to reduce cell death in rat hippocampal slices treated with TREK1 channel blockers signifying the requirement of active TREK1 channels for lactate mediated neuroprotection. However, lactate in the presence of protein kinase inhibitor failed to reduce cell death. This might be related to the role of protein kinase A in upregulation of TREK1 channels. We also estimated CA1 pyramidal neuronal TREK1 channel expression and found both lactate and oxygen glucose deprivation to decrease TREK1 channel expression that was surprisingly opposite to the effects on astrocytes. As TREK1 channel activation and upregulation decreases neuronal excitability, a decrease in neuronal TREK1 channel expression in response to lactate is expected to cause higher neuronal death and fails to explain lactate mediated neuroprotection. Since, lactate upregulated TREK1 channel expression and functional activity in CA1 stratum radiate astrocytes, we reasoned that the lactate mediated neuroprotection might be via astrocytic TREK1 channels requiring viable functional astrocytes. This was tested by disrupting astrocyte function using gliotoxin, and estimating cell death in oxygen glucose deprived hippocampal slices. Lactate failed to reduce cell death in presence of gliotoxin signifying the importance of viable astrocytes for lactate mediated neuroprotection. The above effects were specific to lactate as pyruvate failed to increase TREK1 expression and reduce cell death. TREK1 channels contribute to neuroprotection by enhancing potassium buffering and glutamate clearance capacity of astrocytes. We propose that lactate promotes neuronal survival in hippocampus by increasing TREK1 channel expression and activity in astrocytes during ischemia. This pathway serves as an alternate mechanism of neuroprotection.