Power tools for Alzheimer’s disease – an electrochemical preamp for Aβ


  • Henrik Zetterberg,

    1. Institute of Neuroscience and Physiology, Department of Psychiatry and Neurochemistry, the Sahlgrenska Academy at the University of Gothenburg, Mölndal, Sweden
    2. UCL Institute of Neurology, London, UK
    Search for more papers by this author
  • Per Hammarström

    1. IFM-Department of Chemistry, Linköping University, Linköping, Sweden
    Search for more papers by this author

  • Read the full article ‘Microbiosensor for Alzheimer’s disease diagnostics: detection of amyloid beta biomarkers’ on page 374.

Address correspondence and reprint requests to Henrik Zetterberg, MD, PhD, Institute of Neuroscience and Physiology, Department of Psychiatry and Neurochemistry, The Sahlgrenska Academy at the University of Gothenburg, S-431 80 Mölndal, Sweden. E-mail: henrik.zetterberg@gu.se

Abbreviations used

Alzheimer’s disease

amyloid β

Three neuropathological changes are firmly associated with Alzheimer’s disease (AD): aggregation of amyloid β (Aβ) in senile plaques, accumulation of hyperphosphorylated tau proteins in neurofibrillary tangles and loss of synapses and neurons (Blennow et al. 2006). Biomarker studies suggest that AD may be considered a sequence of neuropathological changes with a spectrum of presentations from pre-symptomatic to full-blown dementia. The approximate duration of the pathological process is at least 10–15 years of which the symptomatic phase comprises 25–50%. The data clearly show that amyloid pathology precedes neurodegeneration and tau pathology (Buchhave et al. 2012). More specifically, low CSF levels of the 42 amino acid isoform of Aβ (Aβ42) and/or retention of amyloid PET tracers in the brain parenchyma, both reflecting amyloid neuropathology, appear already in pre-clinical disease stages. In contrast, release of intraneuronal tau proteins to the CSF, brain atrophy visualized by neuroimaging, and cognitive decline are down-stream phenomena, separated by 5–10 years from the first biomarker sign of brain amyloid accumulation (Jack et al. 2010).

Experiments in cell and animal models have revealed profound and rapid toxicity of low-n oligomers of Aβ to synapses and neurons (Walsh and Selkoe 2007). How can the human brain resist amyloid pathology for many years before the toxicity becomes apparent? One way of interpreting the data from longitudinal AD biomarker studies in humans is that amyloid build-up and toxicity occur in two distinct mechanistic phases. First, amyloid pathology propagates via a prion-like mechanism of corruptive protein templating, in which aggregated Aβ itself is the seed (Rosen et al. 2012). Then, once saturated (after many years), the plaques enter a new phase and either start to leak diffusible forms or catalyze nascent formation of modified or oligomerized Aβ in a toxic pathway that rapidly injures neurons and produces tau pathology and clinical manifestations of AD. This tentative model is very attractive in the light of recent studies on mammalian prion diseases (Sandberg et al. 2011).

However, another major question remains. What starts the amyloidogenic cascade in sporadic AD? To understand this, we need to know more about the dynamics of the mechanisms that govern Aβ homeostasis in the brain. Most of the Aβ in the brain appears to be produced in response to neuronal activity in the form of 37–40 amino acid long Aβ fragments that are less aggregation-prone as compared with Aβ42 (Cirrito et al. 2005). Recent data from stable isotope kinetic labelling experiments using lumbar CSF suggest that the clearance of Aβ is rapid (around 8% per hour in the absence of amyloid pathology) (Bateman et al. 2006). It is reasonable to hypothesize that factors that increase Aβ42 production, shift the balance between Aβ42 and the less aggregation-prone Aβ species, and/or inhibit enzymes or chaperones that help clearing Aβ, may induce an amyloidogenic cascade that eventually turns autocatalytic. However, to study such factors we need methods that enable real-time, sensitive and specific measurements of Aβ in living tissues.

In this issue of Journal of Neurochemistry, Prabhulkar et al. (2012) present an immunosensor for Aβ40 and Aβ42 built to fit in brain microdialysis experiments. The novel microsensor is based on three 5-μm wide carbon electrodes embedded in epoxy where the biorecognition is performed on the electrode tip in separated triplex format employing Aβ40, Aβ42 and Aβ17-28 (all Aβ peptides) specific antibodies. The microsensor, subsequent to capture of Aβ, uses electrochemistry to oxidize Tyr10 and can hereby, through the generated current, quantify the amount of captured Aβ isoforms on each of the sensor tips. The approach is demonstrated to distinguish Aβ40 and Aβ42 isoforms in spiked CSF samples. Previous experiments assessing Aβ levels in CSF or brain interstitial fluid have relied upon subsequent measurement of the peptides using standard immunoassays. Although the current study was demonstrated on ex vivo samples, the Prabhulkar et al. work demonstrates a reproducible and rather sensitive (20–140 nM) device with a reagentless detection process with the potential for direct in vivo monitoring of Aβ, where the read out is completed over the course of 7 min. If the described microelectrode is integrated within a microdialysis device, this is a major leap forward to enable researchers to monitor Aβ homeostasis in brain interstitial fluid in response to different stressors or amyloid-targeting drug candidates in animal models. It would also allow for rapid assessment of Aβ in spinal catheter studies in humans and also off-line determination of Aβ concentrations should the device prove stable enough. One obvious concern for such a device based on antibody recognition would be degradation of the tip during sampling in a complex milieu afforded by brain interstitial fluid or the CSF. If this can be overcome through stabilization of the antibodies or by shielding of the device, it is likely to be an important invention in the AD field. If this stable device can be developed, it would be highly interesting to monitor Aβ40/42 concentrations in real time at different sites in the brain, ventricles and along the spine of an experimental animal to observe the gradient of Aβ40/42 on-line both during aging and post-treatment by β- or γ-secretase inhibitors and modulators.


HZ was supported by the Swedish Research Council (K2010-63P-21562-01-4, K2011-61X-20401-05-6), Swedish State Support for Clinical Research and the EU Joint Programme – Neurodegenerative Disease Research (BIOMARKAPD). PH was supported by the Swedish research council (NT-2011-5804) and the EU-FP-7 Health (LUPAS). The authors have no conflict of interest to declare.