Peter Haslehurst, Damian M. Cummings, Francesca Cacucci, Tiffanie A. Benway, Gaukhar Baidildinova, Owain T. James, Jill C. Richardson, Wenfei Liu, Fabia Moroni, Frances A. Edwards, Muzammil A. Nahaboo Solim, Dervis A. Salih, Evelyn Medawar, Taylor A. Hanan, and Laurenz Muessig
BackgroundProgression of Alzheimer’s disease is thought initially to depend on rising amyloidβ and its synaptic interactions. Transgenic mice (TASTPM; APPSwe/PSEN1M146V) show altered synaptic transmission, compatible with increased physiological function of amyloidβ, before plaques are detected. Recently, the importance of microglia has become apparent in the human disease. Similarly, TASTPM show a close association of plaque load with upregulated microglial genes.MethodsCA1 Synaptic transmission and plasticity were investigated using in vitro electrophysiology. Migroglial relationship to plaques was examined with immunohistochemistry. Behaviour was assessed with a forced-alternation T-maze, open field, light/dark box and elevated plus maze.FindingsThe most striking finding is the increase in microglial numbers in TASTPM, which, like synaptic changes, begins before plaques are detected. Further increases and a reactive phenotype occur later, concurrent with development of larger plaques. Long-term potentiation is initially enhanced at pre-plaque stages but decrements with the initial appearance of plaques. Finally, despite altered plasticity, TASTPM have little cognitive deficit, even with a heavy plaque load, although they show altered non-cognitive behaviours.InterpretationThe pre-plaque synaptic changes and microglial proliferation are presumably related to low, non-toxic amyloidβ levels in the general neuropil and not directly associated with plaques. However, as plaques grow, microglia proliferate further, clustering around plaques and becoming phagocytic. Like in humans, even when plaque load is heavy, without development of neurofibrillary tangles and neurodegeneration, these alterations do not result in cognitive deficits. Behaviours are seen that could be consistent with pre-diagnosis changes in the human condition.Research in contextEvidence before this studyThere is a large body of research examining many aspects of phenotypes associated with mouse models of Alzheimer’s disease – a PubMed search for the terms Alzheimer* AND mouse returns in excess of 21000 articles. However, there are few systematic articles pulling together pathological, functional (electrophysiological), and behavioural analyses across the life-span of such models. There is also a number of conflicting outcomes, for example reports of impaired versus enhanced synaptic plasticity; cognitive impairments or not.Recently, the importance of microglia in Alzheimer’s disease has come to the fore in human Genome Wide Association Studies (GWAS), with variants of a number of microglial genes identified as risk-factors for developing the disease. Interestingly, we have recently reported that Trem2 and other genes identified as risk-factors in humans are strongly up regulated in close association to plaque development in the mouse model used in this study. Moreover, this previous study predicted two of the most recently identified genes that were identified in GWAS since the publication of our paper.We have previously used this model to identify the earliest synaptic changes and shown changes in release of glutamate, the primary excitatory neurotransmitter in the brain, to occur even before plaques are detectable.Added value of this studyBy studying this transgenic mouse model of Alzheimer’s disease, throughout the development of plaques, from prior to detection through to heavy plaque loads, we have been able to identify a clear time course of key phenotypic changes associated with early disease. In particular, this study identifies the very early changes in microglia and can separate the time course of the microglial phenotype. In addition, we detail the changes in synaptic plasticity over time and importantly identify that, like in humans in the absence of Tau tangles or neurodegeneration, considerable synaptic changes can occur and a heavy plaque load without resulting in substantial cognitive loss.Implications of all the available evidenceOur data indicate that rising amyloid beta prior to detectable plaque deposition results in changes in synaptic function that likely reflects an enhanced physiological effect of amyloid beta. At this stage, microglia proliferate but do not activate. Once plaques begin to appear, microglia migrate to surround the plaque and become phagocytic, likely targeting dystrophic synapses and neurites caused by the cloud of highly-toxic amyloid beta around the plaque. Similarly to humans, who have plaques but no tangles and have yet to develop substantial neurodegeneration, cognitive deficits are not seen, even with a heavy plaque load; behavioural changes are limited to anxiety-like effects.This investigation of the parallel time-course of events highlights the probability that, if progression of disease can be reversed or slowed early enough, before Tau tangles and substantial neurodegeneration occur, the symptoms of cognitive decline could be very largely avoided. Moreover, it suggests that the substantial increases in microglia number and upregulation of their specific gene expression in association with plaques, is not associated with cognitive loss and may indeed be protective.