healthmedicinet

Illustrated by cases from his own practice, Chris Butler (University of Oxford, UK)
discussed the typical but clinically heterogeneous features of AD and the tools that
can be used to detect them. Nick Fox (UCL) showed how imaging techniques are accurate
in revealing hippocampal atrophy up to 5 years before the onset of symptoms with an
annual reduction of 3–4 % as opposed to 0.3 % in healthy people. Imaging in posterior
cortical atrophy showed widespread cortical amyloid deposition but more selective
tau distribution, suggesting that tau is more closely linked to hypometabolism and
symptomatology. Henrik Zetterberg (University of Gothenburg, Sweden and UCL, UK) showed
that, in AD, CSF A?42 levels correlated with the presence of amyloid positron emission
tomography and with postmortem amyloid cortical plaque load. Zetterberg also presented
the C-terminus of the dendritic protein neurogranin as a potential biomarker for synaptic
degeneration across neurodegenerative diseases and showed particular correlation with
cognitive decline. The 2015 Nobel laureate John O’Keefe (UCL, UK) described his seminal
work on spatial cells in hippocampal formation, now being translated into AD diagnostic
markers. O’Keefe described how the 4 Mountains Test, a topographical memory task,
was highly sensitive to focal hippocampal pathology and had high diagnostic sensitivity
and specificity in differentiating mild cognitive impairment with and without CSF
biomarkers.

At a molecular level, James Duce (University of Leeds, UK) presented evidence that
iron disruption in AD could be modulated by the processing of amyloid precursor protein
(APP) on the cell surface whereby the amyloidogenic pathway that requires BACE1 and
leads to A? generation also causes intraneuronal iron retention and increases vulnerability
to reactive oxygen species neurotoxicity; a common concomitant in AD. Rob Andrew (Nigel
Hooper’s laboratory, University of Manchester, UK) investigated the interactomes of
the APP695 and APP751 isoforms to clarify the mechanisms underpinning their differential
processing to A?. Despite a significant overlap, the role of APP695 in nuclear signalling
and mitochondrial functions was consolidated and higher levels were reported at the
plasma membrane than for APP751. Bart De Strooper (KU Leuven, Belgium and UCL, UK)
described the identification of G-protein coupled receptor 3 (GPR3) as a mediator
of A? generation through interaction with ?-arrestin 2. Subsequent work showed that
GPR3 modulation, unlike other ?-secretase modulators, did not impact on notch signalling.
Human brain samples showed increased GPR3 expression, and evidence from five mouse
models showed that lower GPR3 expression reduced plaque burden, soluble A? and behavioural
deficits.

During the session on tau, Karen Duff (Columbia University, NY, USA) showed that rTgTauEC,
a model with predominant expression of abnormal tau in the entorhinal cortex, has
reduced place and grid cell firing and that firing patterns across days were less
specific, coherent and stable compared with control aged mice. Using these mice crossed
with APP/PS1 mice, Amy Pooler (Nestlé Institute of Health Sciences, Switzerland; previously
KCL, UK) found that at 16 months concurrent amyloid deposition in the cortex led to
an increase in the speed of tau propagation to distal areas, and an increase in tau-induced
neuronal loss. Duff showed data suggesting that tau accumulated on spine-like structures
on the axon and was taken up by cells after a lag phase. In-vitro data presented by
Tara Spires-Jones (University of Edinburgh, UK) provided further evidence that tau
was required for A?-induced synapse loss, possibly mediated by caspase 3. Spires-Jones
also showed that the plaque size, dystrophic neurites per plaque and the amount of
soluble A? in synaptoneurosomes were increased in APP/PS1/hTau mice compared with
APP/PS1 mice, correlating with data presented by Pooler. It was suggested that amyloid
accumulation could trigger neuronal activity and therefore release of tau. In the
P301S mouse model, Maria Grazia Spillantini (University of Cambridge, UK) determined
that several proteins were differentially expressed between 1 and 5 months, as tau
pathology and neuronal death progressed. Spillantini also showed that astrocytes and
astrocyte culture medium from these mice were toxic to neurones, with a reduction
of synaptic protein function.

Dominic Walsh (Harvard University, MA, USA) kick-started the annual debate ‘Are oligomers
the real target?’ arguing for the motion of A? oligomers as the toxic species. He
suggested that A?, acting as an initiator of AD, would not have to show direct correlation
with clinical severity. Walsh showed that certain APPTg mice exhibit cognitive deficits
prior to appreciable amyloid deposition and benefit from acute treatment with anti-A?
monoclonal antibodies, and that human brain-derived, non-fibrillar A? has been shown
to have disease-relevant activity. De Strooper, arguing against the motion, disputed
evidence of A? toxicity, citing the gap in concentrations and time courses observed
in in-vitro models versus the human disease. He further argued that the large variety
of receptors reported to bind A? and mediate its effects cannot be specific effects
of A?. The lively debate highlighted important barriers to overcome, including lack
of experimental standardisation and integration across laboratories as well as lack
of basic knowledge of oligomer biochemistry and the relationship between A? and cognitive
decline.