Search

Alzheimer's disease (AD) is characterised by memory loss, insidious cognitive decline, profound neurodegeneration, and the extracellular accumulation of amyloid-beta (Aβ) peptide in senile plaques and intracellular accumulation of tau in neurofibrillary tangles. Synaptic dysfunction and loss is the strongest pathological correlate of cognitive decline in AD with increasing evidence implicating neuropathological forms of both amyloid-beta and tau protein in this process. A large amount of evidence suggests that oligomeric forms of Aβ, associated with senile plaques, are toxic to synapses but the precise localisation of Aβ and which forms are synaptotoxic remain unknown. Using the high-resolution technique, array tomography, this thesis characterised the synaptic localisation of different forms of Aβ oligomers in a mouse model of amyloidopathy. These results show that different oligomeric Aβ species are present in both presynapses and postsynapses. This study highlights the potential of array tomography for rapid testing of aggregation state specific Aβ antibodies in brain tissue. Following these results, the presence of tau at synapses was examined. Despite the knowledge that tau spreads through defined synaptic circuits, it is currently unknown whether synapse loss occurs before the accumulation of tau or as a consequence. To address this, array tomography was used to examine a mouse model in which mutant P301L human tau is expressed primarily in the entorhinal cortex (rTgTauEC). It has previously been shown that rTgTauEC mice exhibit neuronal loss in the entorhinal cortex and synapse density loss in the middle molecular layer (MML) of the dentate gyrus at 24 months of age. The density of tau-expressing and total presynapses, and the spread of tau into the postsynapse in the MML of 3-6, 9, and 18 month old mice were examined. No loss of synapse density was observed in the MML up to 18 months of age, even in axons expressing tau. Despite the maintenance of synapse density, we see spread of human tau from presynaptic terminals to postsynaptic compartments in the MML at very early ages. This indicates that the spread of tau through neural circuits is not due to the degeneration of axon terminals and is an early feature of the disease process. Following examination of both synaptic amyloid-beta and tau in separate models, this thesis then examined how these two proteins may be synergistically working together to drive synaptic pathology. To investigate this a novel mouse model was used in which amyloid-beta deposits are present in combination with non-mutated human tau expression (APP/PS1 + hTau). These results suggested that the addition of human tau expression does not increase plaque associated synapse loss, neither does it increase the proportion of synapses colocalising with amyloid-beta. Similarly the presence of human tau at individual postsynapses was not enhanced in the presence of oligomeric Aβ. Surprisingly, intact long-term recognition memory was observed in APP/PS1 + hTau mice. However a hyperactive phenotype was detected in these mice that could be prevented upon tau suppression. This suggests a synergistic relationship may exist in the presentation of this phenotype. Finally in the last part of this thesis, synapses from post-mortem human Alzheimer's disease and age-matched controls were investigated. It has previously been suggested that both amyloid-beta and tau can interfere with mitochondrial transport to the synapse and mitochondrial function. For this reason the presence of synaptic mitochondria at both the presynapse and postsynapse was determined in order to investigate any alteration in the diseased state. A reduction in the proportion of presynapses with multiple mitochondria present was detected in anterior/posterior transverse temporal cortex (BA41/42). This was not observed in dorsolateral prefrontal cortex (BA46), suggesting either a selective vulnerability of the former brain region or a selective resistance of the latter brain region, to mitochondrial depletion at the synapse. The findings presented in this thesis demonstrate that when investigated in isolation, pathological forms of amyloid-beta are present at a subset of synapses where they may contribute to toxicity, whilst the spread of tau protein is an early feature of the disease process and occurs prior to overt synapse loss. This thesis also explores the proposed synergistic relationship between amyloid-beta and tau using a novel mouse model and human post-mortem brain tissue. Since these two proteins both have been implicated in synaptic dysfunction, investigating Aβ and tau in new mouse models and human brain tissue will be instrumental in furthering our understanding of mechanisms and features of synaptotoxicity that could be important therapeutic targets.

This thesis examines the functional connectivity of the hippocampus with the rest of the brain, with a focus on the neocortex. The hypothesis explored, in an animal model, is whether the frequency-dependent behaviour of certain brain connectivity relationships applies to hippocampal-neocortical connections. To encompass the temporal and spatial resolution necessary to do this, two main techniques are used in combination in most of the experimental work hereby presented: (1) electrophysiological recordings of local field potentials (LFPs), and (2) functional activity recordings of blood oxygenation level dependent (BOLD) signal using functional magnetic resonance imaging (fMRI). The main hypothesis is that the frequency-dependent behaviour of specific hippocampal synapses imposes the rules of extra-hippocampal activity propagation and hippocampal-neocortical interactions. The main discovery is that short and long-term plasticity modulates network activation, a finding suggesting a possible mechanism that could mediate the encoding and consolidation of memory traces. Chapters 1 to 3 introduce the vast literature review in which this project lies, and the general methods utilised. Chapter 4 (first experimental chapter) describes, using electrophysiology in rats, the evoked response of the main hippocampal output (CA1 neurons) when its major input (CA3 pyramidal cells) is activated at frequencies that in subsequent experiments were used to build brain-wide functional maps. CA1 spiking activity is found to be optimal in maintaining the amplitude of the population spike (PS) at beta frequencies (10-20 Hz), whereas lower (< 10 Hz) and higher (> 20 Hz) frequencies are normally less effective. Chapter 5 describes, using fMRI, how these intra-hippocampal activity patterns relate to long-range activity propagation in fMRI experiments. Hippocampal activation exhibits a linear monotonic increase with evoked frequency, whilst a network of selected structures is activated preferentially when beta frequencies are applied (mainly neocortical structures like the prefrontal and parietal cortices, motor and sensory cortices, and some subcortical structures like the nucleus accumbens and the striatum). This data is highly correlated with the PS recorded in CA1 and with multi-unit activity (MUA) and single-unit activity (SUA) simultaneously recorded in the medial prefrontal cortex (mPFC), one of the structures receiving propagated activity at beta frequencies, as described in Chapter 6. As mPFC also receives hippocampal input at a restricted beta frequency range stimulation of the dorsal hippocampus, Chapter 7 describes the use of a combined electrophysiology/fMRI approach to identify the pathway responsible for activity propagation. We performed microsurgery lesions to investigate the pathway responsible for the polysynaptic propagation of activity. Findings indicate that the septo-temporal longitudinal pathway is the one leading information transfer from dorsal to ventral hippocampus in the rat, and from there directly to the ventral subiculum, apparently by-passing entorhinal cortex. Last, in Chapter 8 the effect of durable modifications of synaptic weights by long-term potentiation (LTP) in the previously described frequency-dependent activity propagation is also described and contextualized in the memory trace consolidation framework, both electrophysiologically (Chapter 5) and with fMRI (Chapter 6). LTP is a long-lasting change in synaptic weights that, at the CA3-CA1 synapse, is capable of modifying hippocampal-neocortical connections such as to open the opportunity for higher frequency patterns (> 40 Hz) to propagate to neocortical structures. These results suggest that, by means of frequency-coding, the hippocampus normally regulates propagation of selected information to the neocortex, but that at specific moments (e.g. when the hippocampus undergoes LTP) this regulation broadens to permit high-frequency information to pass through and affect neural activity in the cortex. It is a beautifully simple mechanism that merits further detailed examination in a multi-disciplinary manner as outlined in Chapters 9 and 10.