Beta amyloid aggregates induce sensitised TLR4 signalling causing long-term potentiation deficit and rat neuronal cell death

ARTICLE https://doi.org/10.1038/s42003-020-0792-9 OPEN 1234567890():,; Beta amyloid aggregates induce sensitised TLR4 signalling causing long-term potentiation deficit and rat neuronal cell death Craig Hughes1,11, Minee L. Choi2,3,11, Jee-Hyun Yi4,9,11, Seung-Chan Kim4,5, Anna Drews1,10, Peter St. George-Hyslop 6, Clare Bryant 7, Sonia Gandhi2,3, Kwangwook Cho4,5 & David Klenerman1,8 ✉ The molecular events causing memory loss and neuronal cell death in Alzheimer’s disease (AD) over time are still unknown. Here we found that picomolar concentrations of soluble oligomers of synthetic beta amyloid (Aβ42) aggregates incubated with BV2 cells or rat astrocytes caused a sensitised response of Toll-like receptor 4 (TLR4) with time, leading to increased production of TNF-α. Aβ aggregates caused long term potentiation (LTP) deficit in hippocampal slices and predominantly neuronal cell death in co-cultures of astrocytes and neurons, which was blocked by TLR4 antagonists. Soluble Aβ aggregates cause LTP deficit and neuronal death via an autocrine/paracrine mechanism due to TLR4 signalling. These findings suggest that the TLR4-mediated inflammatory response may be a key pathophysiological process in AD. 1 Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW, UK. 2 Department of Clinical and Movement Neurosciences, UCL Queen Square Institute of Neurology, London WC1N 3BG, UK. 3 The Francis Crick Institute, 1 Midland Road, London NW1 1AT, UK. 4 Centre for Synaptic Plasticity, Faculty of Health Sciences, University of Bristol, Whitson Street, Bristol BS1 3NY, UK. 5 UK-Dementia Research Institute at King’s College London, King’s College, Department of Basic and Clinical Neuroscience, Institute of Psychiatry, Psychology and Neuroscience, London SE5 9NU, UK. 6 Cambridge Institute for Medical Research, University of Cambridge, Cambridge Biomedical Campus The Keith Peters Building Hills Road, Cambridge CB2 0XY, UK. 7 Department of Veterinary Medicine, University of Cambridge, Madingley Road, Cambridge CB3 0ES, UK. 8 UK Dementia Research Institute, University of Cambridge, Cambridge CB2 0XY, UK. 9Present address: Center for Synaptic Brain Dysfunctions, Institute for Basic Science, Daejeon 34126, Republic of Korea. 10Present address: The German Center for Neurodegenerative Diseases (DZNE), Sigmund-Freud-Str. 27, Venusberg-Campus, Gebäude 99, 53127 Bonn, Germany. 12These authors contributed equally: Craig Hughes, Minee L. Choi, Jee-Hyun Yi. ✉email: COMMUNICATIONS BIOLOGY | (2020)3:79 | https://doi.org/10.1038/s42003-020-0792-9 | www.nature.com/commsbio 1 ARTICLE M COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-020-0792-9 emory loss is a very common symptom of Alzheimer’s disease (AD), however the molecular basis by which memory loss occurs is not understood1. This means it is currently challenging to develop treatments for AD. A synaptic correlate of memory is long-term potentiation (LTP). LTP is widely considered one of the major cellular mechanisms that underlies learning and memory. It has been found that soluble beta-amyloid (Aβ) aggregates from a variety of sources including soaked brain2, brain homogenate, concentrated CSF and synthetic aggregates3,4 can cause LTP deficit in brain slices. Significant efforts have been made to identify the nature of the aggregates that affect LTP deficit, so they can be targeted for potential therapy. Antibodies that bind the N-terminus of Aβ4 4, knock-out of PrP or the use of PrP antibodies5 have all been shown to be effective in preventing aggregate-induced LTP deficit. These results show that soluble Aβ aggregates initiate LTP deficit, but the mechanism by which this occurs and whether it is a result of the direct interaction of aggregates with synapses or occurs by a different mechanism has not been established to date. In vivo, Aβ can be post-translationally modified and interact with other proteins present, so that the aggregates present are heterogeneous in both size and composition. In contrast, synthetic aggregates made by aggregating Aβ42 in the test-tube are only heterogeneous in size not composition and still capable of causing LTP deficit4. In most experiments the aggregate concentration is not measured but only the total Aβ monomer concentration is known. This means that while it has been observed that brainderived aggregates are more effective at causing LTP deficit than synthetic Aβ aggregates this could simply occur because the concentration of aggregates is higher in the preparations used. It is not possible to determine which type of aggregate is more effective at causing LTP without knowing the aggregate concentration. Aβ aggregates can trigger the production of a number of proinflammatory cytokines, including TNF-α, from astrocytes and microglia6,7, and the media from conditioned astrocytes is toxic to neurons8 suggesting that neuronal cell death can occur via an inflammatory mechanism. One of the routes that proinflammatory cytokines are produced occurs via Toll-like receptors, pattern recognition molecules that recognize damaged molecules, particularly TLR2 and TLR49,10. Our recent work shows that synthetic Aβ aggregates exist in a range of different sizes and structures with the longer protofibrils being the inflammatory species and signal via TLR411. There is a crystal structure of TLR3, which is in the same family as TLR4, bound to an RNA dimer which is about 2 nm in diameter12. TLR3 signalling occurs when the RNA dimer is longer than 15 nm13. This suggests that long protofibrillar Aβ aggregates, which have a comparable diameter, initiate TLR4 signalling by forming a similar structure with a TLR4 dimers bound along the protofibril, providing a plausible explanation of both why they are the inflammatory species and how they initiate TLR4 signalling. However, to date, this experiment and many other experiments on aggregate induced inflammation have been performed at high aggregate concentrations in short time periods, typically 24 h. Therefore, there are important questions about the relevance of the results obtained at these high aggregate doses to AD. In particular, it is not clear how the response is altered at more relevant physiological concentrations of aggregates applied over longer times or if TLR4 signalling occurs at all. To address this issue, we first explored the response of BV2 microglial to extended doses of low concentrations of soluble aggregates, close to physiological levels, finding that this leads to sensitized response to these aggregates due to TLR4 signalling. We then explored if this aggregateinduced inflammatory response could lead to LTP deficit and neuronal cell death, cellular correlates of the symptoms associated with the development of AD, by performing experiments in the presence and absence of TLR4 antagonists. Results Pro-inflammatory response of macrophages to beta -amyloid. Experiments were firstly performed using synthetic oligomers of Aβ1-42, made by aggregating monom (...truncated)