Structural synaptic signatures of Alzheimer's disease and dementia with Lewy bodies in the male brain

The impact of Alzheimer’s Disease (AD) and Dementia with Lewy Bodies (DLB) on synaptic organisation remains poorly understood. Here, we found that in humans, DLB and AD were associated with increased synaptic levels of glutamate transporter vGlut1 and active zone protein Bassoon clustering respectively; these effects were only observed in male brain samples. These findings demonstrate disease- and sex-specific presynaptic structural remodelling in age-related neurodegenerative disorders.

Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) are the two major forms of neurodegenerative dementia, characterised by distinct pathological hallmarks involving aggregation of misfolded proteins. One proposed key pathophysiological mechanism in AD and DLB, as well as in other neurodegenerative disorders, is synaptic dysfunction. [1][2][3][4][5][6][7][8] The details of human synaptic pathology in AD and DLB besides synaptic loss are, however, poorly characterised, and the underlying processes remain unknown.
Our previous studies identified distinct patterns of aberrant protein expression in AD and DLB, 5,9 raising the possibility that manifestation of AD and DLB may differ at the local synaptic level. To directly address this, we employed immunostaining, confocal and superresolution microscopy to visualise synaptic structure in post mortem human brain isolates ( Figure S1). Samples represented Brodmann Area 9 (BA9), a key part of the prefrontal cortex implicated in neurodegeneration, 5 from a cohort of 32 control subjects and cases with severe AD and DLB.
We first assessed synaptic structure in 24 cases (Table S1. We then performed double immunostaining for Bsn and vGlut1, a key component of synaptic vesicles (SV) that has been implicated in AD, as well as in Parkinson's disease (PD), which shares many neuropathological features of DLB. Synaptic vGlut1 levels were elevated in both AD and DLB male samples compared with the controls, but the increase in AD samples was not statistically significant. In contrast, synaptic vGlut1 levels were significantly increased in male DLB samples compared with male controls or female DLB samples ( Figures 1A,B and S3d). Therefore, we conclude that vGlut1 is specifically enriched in synapses from male subjects with DLB.
Recent evidence shows that neuronal activity regulates nanoscale clustering of the AZ 10 ; we therefore reasoned that aberrant synaptic activity in AD may lead to nanoscale changes in AZ architecture and investigated this in an extended cohort including 24 control and AD female and male cases (Table S1.2, Figure S4). We assessed Bsn clustering using confocal microscopy and a ratiometric assay developed by us previously 10 ( Figure S5a). Clustering in AD samples was

Key Points
• Confocal and super-resolution microscopy reveals alterations in synaptic organisation associated with neurodegeneration in post mortem human brain samples.
• Different types of dementia are associated with distinct changes in synaptic structure.
• The observed changes are restricted to male brain samples.
significantly increased in male but not in female brains ( Figure 1C-F), indicative of an increased density of Bsn packing within the AZ. Clustering did not significantly correlate with age ( Figure S4d), suggesting that the difference between control and AD samples was not due to the difference in median age ( Figure S4a). Our findings show that two major forms of dementia are associated with distinct changes in presynaptic organisation, providing direct evidence for disease-specific synaptic defects in neurodegeneration ( Figure 1K); furthermore, we present evidence for sex-specific effects in dementia at the level of synaptic structure, in line with association between biological sex and clinical manifestation of dementia. 12,13 Some data from female samples showed similar trends to those observed in male samples, yet failed to reach the threshold of significance ( Figure 1E,F), possibly reflecting the higher incidence of 'pure' dementia in men vs a mixed LBD/AD pathology reported in women. 14 It remains to be determined whether the observed changes represent a direct effect of pathology or a form of compensatory synaptic plasticity. 3 In order to enhance the exploratory power of the approach described in this paper, further investigation will require larger case cohorts combined with testing for multiple synaptic markers, with particular relevance for quantitative analysis of nanoscale synaptic structure. Thus, our initial observations reported here pave the way for deeper investigation of synaptic architecture in neurodegeneration, with the long-term potential for development of targeted diagnostics and therapies for dementia. 3 Last but not least, in-depth structural analysis of human synaptic dysregulation will allow for knowledge-based validation of animal models, 15

CONFLICTS OF INTEREST
None declared.

ETHICS STATEMENT
Freshly frozen brain samples were provided by the London Neurodegenerative Diseases Brain Bank (KCL), part of the UK Brain Banks Network. The project was carried out under the ethical approval of the tissue bank (18/WA/0206 REC for Wales). Written informed consent from the donors and/or their relatives as appropriate was obtained by the London Neurodegenerative Diseases Brain Bank.

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/nan.12852.

DATA AVAILABILITY STATEMENT
For the materials and methods, see the supporting information. The datasets generated and analysed during the current study are available from the corresponding author on reasonable request.