Early Detection of Alzheimer's Disease From Cortical and Hippocampal Local Field Potentials Using an Ensembled Machine Learning Model.

Early diagnosis of Alzheimer's disease (AD) is a very challenging problem and has been attempted through data-driven methods in recent years. However, considering the inherent complexity in decoding higher cognitive functions from spontaneous neuronal signals, these data-driven methods benefit from the incorporation of multimodal data. This work proposes an ensembled machine learning model with explainability (EXML) to detect subtle patterns in cortical and hippocampal local field potential signals (LFPs) that can be considered as a potential marker for AD in the early stage of the disease. The LFPs acquired from healthy and two types of AD animal models (n = 10 each) using linear multielectrode probes were endorsed by electrocardiogram and respiration signals for their veracity. Feature sets were generated from LFPs in temporal, spatial and spectral domains and were fed into selected machine-learning models for each domain. Using late fusion, the EXML model achieved an overall accuracy of 99.4%. This provided insights into the amyloid plaque deposition process as early as 3 months of the disease onset by identifying the subtle patterns in the network activities. Lastly, the individual and ensemble models were found to be robust when evaluated by randomly masking channels to mimic the presence of artefacts.

Classification of Whisker Deflections From Evoked Responses in the Somatosensory Barrel Cortex With Spiking Neural Networks.

Spike-based neuromorphic hardware has great potential for low-energy brain-machine interfaces, leading to a novel paradigm for neuroprosthetics where spiking neurons in silicon read out and control activity of brain circuits. Neuromorphic processors can receive rich information about brain activity from both spikes and local field potentials (LFPs) recorded by implanted neural probes. However, it was unclear whether spiking neural networks (SNNs) implemented on such devices can effectively process that information. Here, we demonstrate that SNNs can be trained to classify whisker deflections of different amplitudes from evoked responses in a single barrel of the rat somatosensory cortex. We show that the classification performance is comparable or even superior to state-of-the-art machine learning approaches. We find that SNNs are rather insensitive to recorded signal type: both multi-unit spiking activity and LFPs yield similar results, where LFPs from cortical layers III and IV seem better suited than those of deep layers. In addition, no hand-crafted features need to be extracted from the data-multi-unit activity can directly be fed into these networks and a simple event-encoding of LFPs is sufficient for good performance. Furthermore, we find that the performance of SNNs is insensitive to the network state-their performance is similar during UP and DOWN states.

Accelerated Aging Characterizes the Early Stage of Alzheimer's Disease.

For Alzheimer's disease (AD), aging is the main risk factor, but whether cognitive impairments due to aging resemble early AD deficits is not yet defined. When working with mouse models of AD, the situation is just as complicated, because only a few studies track the progression of the disease at different ages, and most ignore how the aging process affects control mice. In this work, we addressed this problem by comparing the aging process of PS2APP (AD) and wild-type (WT) mice at the level of spontaneous brain electrical activity under anesthesia. Using local field potential recordings, obtained with a linear probe that traverses the posterior parietal cortex and the entire hippocampus, we analyzed how multiple electrical parameters are modified by aging in AD and WT mice. With this approach, we highlighted AD specific features that appear in young AD mice prior to plaque deposition or that are delayed at 12 and 16 months of age. Furthermore, we identified aging characteristics present in WT mice but also occurring prematurely in young AD mice. In short, we found that reduction in the relative power of slow oscillations (SO) and Low/High power imbalance are linked to an AD phenotype at its onset. The loss of SO connectivity and cortico-hippocampal coupling between SO and higher frequencies as well as the increase in UP-state and burst durations are found in young AD and old WT mice. We show evidence that the aging process is accelerated by the mutant PS2 itself and discuss such changes in relation to amyloidosis and gliosis.

Neuronal Avalanches Across the Rat Somatosensory Barrel Cortex and the Effect of Single Whisker Stimulation.

Since its first experimental signatures, the so called "critical brain hypothesis" has been extensively studied. Yet, its actual foundations remain elusive. According to a widely accepted teleological reasoning, the brain would be poised to a critical state to optimize the mapping of the noisy and ever changing real-world inputs, thus suggesting that primary sensory cortical areas should be critical. We investigated whether a single barrel column of the somatosensory cortex of the anesthetized rat displays a critical behavior. Neuronal avalanches were recorded across all cortical layers in terms of both multi-unit activities and population local field potentials, and their behavior during spontaneous activity compared to the one evoked by a controlled single whisker deflection. By applying a maximum likelihood statistical method based on timeseries undersampling to fit the avalanches distributions, we show that neuronal avalanches are power law distributed for both multi-unit activities and local field potentials during spontaneous activity, with exponents that are spread along a scaling line. Instead, after the tactile stimulus, activity switches to a transient across-layers synchronization mode that appears to dominate the cortical representation of the single sensory input.

Familial Alzheimer's disease presenilin-2 mutants affect Ca2+ homeostasis and brain network excitability.

Alzheimer's disease (AD) is the most frequent cause of dementia in the elderly. Few cases are familial (FAD), due to autosomal dominant mutations in presenilin-1 (PS1), presenilin-2 (PS2) or amyloid precursor protein (APP). The three proteins are involved in the generation of amyloid-beta (Aß) peptides, providing genetic support to the hypothesis of Aß pathogenicity. However, clinical trials focused on the Aß pathway failed in their attempt to modify disease progression, suggesting the existence of additional pathogenic mechanisms. Ca2+ dysregulation is a feature of cerebral aging, with an increased frequency and anticipated age of onset in several forms of neurodegeneration, including AD. Interestingly, FAD-linked PS1 and PS2 mutants alter multiple key cellular pathways, including Ca2+ signaling. By generating novel tools for measuring Ca2+ in living cells, and combining different approaches, we showed that FAD-linked PS2 mutants significantly alter cell Ca2+ signaling and brain network activity, as summarized below.

Presenilin-2 and Calcium Handling: Molecules, Organelles, Cells and Brain Networks.

Presenilin-2 (PS2) is one of the three proteins that are dominantly mutated in familial Alzheimer's disease (FAD). It forms the catalytic core of the γ-secretase complex-a function shared with its homolog presenilin-1 (PS1)-the enzyme ultimately responsible of amyloid-ß (Aß) formation. Besides its enzymatic activity, PS2 is a multifunctional protein, being specifically involved, independently of γ-secretase activity, in the modulation of several cellular processes, such as Ca2+ signalling, mitochondrial function, inter-organelle communication, and autophagy. As for the former, evidence has accumulated that supports the involvement of PS2 at different levels, ranging from organelle Ca2+ handling to Ca2+ entry through plasma membrane channels. Thus FAD-linked PS2 mutations impact on multiple aspects of cell and tissue physiology, including bioenergetics and brain network excitability. In this contribution, we summarize the main findings on PS2, primarily as a modulator of Ca2+ homeostasis, with particular emphasis on the role of its mutations in the pathogenesis of FAD. Identification of cell pathways and molecules that are specifically targeted by PS2 mutants, as well as of common targets shared with PS1 mutants, will be fundamental to disentangle the complexity of memory loss and brain degeneration that occurs in Alzheimer's disease (AD).

ORAI2 Down-Regulation Potentiates SOCE and Decreases Aß42 Accumulation in Human Neuroglioma Cells.

Senile plaques, the hallmarks of Alzheimer's Disease (AD), are generated by the deposition of amyloid-beta (Aß), the proteolytic product of amyloid precursor protein (APP), by ß and γ-secretase. A large body of evidence points towards a role for Ca2+ imbalances in the pathophysiology of both sporadic and familial forms of AD (FAD). A reduction in store-operated Ca2+ entry (SOCE) is shared by numerous FAD-linked mutations, and SOCE is involved in Aß accumulation in different model cells. In neurons, both the role and components of SOCE remain quite obscure, whereas in astrocytes, SOCE controls their Ca2+-based excitability and communication to neurons. Glial cells are also directly involved in Aß production and clearance. Here, we focus on the role of ORAI2, a key SOCE component, in modulating SOCE in the human neuroglioma cell line H4. We show that ORAI2 overexpression reduces both SOCE level and stores Ca2+ content, while ORAI2 downregulation significantly increases SOCE amplitude without affecting store Ca2+ handling. In Aß-secreting H4-APPswe cells, SOCE inhibition by BTP2 and SOCE augmentation by ORAI2 downregulation respectively increases and decreases Aß42 accumulation. Based on these findings, we suggest ORAI2 downregulation as a potential tool to rescue defective SOCE in AD, while preventing plaque formation.

Dampened Slow Oscillation Connectivity Anticipates Amyloid Deposition in the PS2APP Mouse Model of Alzheimer's Disease.

To fight Alzheimer's disease (AD), we should know when, where, and how brain network dysfunctions initiate. In AD mouse models, relevant information can be derived from brain electrical activity. With a multi-site linear probe, we recorded local field potentials simultaneously at the posterior-parietal cortex and hippocampus of wild-type and double transgenic AD mice, under anesthesia. We focused on PS2APP (B6.152H) mice carrying both presenilin-2 (PS2) and amyloid precursor protein (APP) mutations, at three and six months of age, before and after plaque deposition respectively. To highlight defects linked to either the PS2 or APP mutation, we included in the analysis age-matched PS2.30H and APP-Swedish mice, carrying each of the mutations individually. Our study also included PSEN2-/- mice. At three months, only predeposition B6.152H mice show a reduction in the functional connectivity of slow oscillations (SO) and in the power ratio between SO and delta waves. At six months, plaque-seeding B6.152H mice undergo a worsening of the low/high frequency power imbalance and show a massive loss of cortico-hippocampal phase-amplitude coupling (PAC) between SO and higher frequencies, a feature shared with amyloid-free PS2.30H mice. We conclude that the PS2 mutation is sufficient to impair SO PAC and accelerate network dysfunctions in amyloid-accumulating mice.

Cx32 hemichannel opening by cytosolic Ca2+ is inhibited by the R220X mutation that causes Charcot-Marie-Tooth disease.

Mutations of the GJB1 gene encoding connexin 32 (Cx32) cause the X-linked form of Charcot-Marie-Tooth disease (CMTX1), a demyelinating peripheral neuropathy for which there is no cure. A growing body of evidence indicates that ATP release through Cx32 hemichannels in Schwann cells could be critical for nerve myelination, but it is unknown if CMTX1 mutations alter the cytosolic Ca2+-dependent gating mechanism that controls Cx32 hemichannel opening and ATP release. The current study uncovered that loss of the C-terminus in Cx32 (R220X mutation), which causes a severe CMTX1 phenotype, inhibits hemichannel opening during a canonical IP3-mediated increase in cytosolic Ca2+ in HeLa cells. Interestingly, the gating function of R220X hemichannels was completely restored by both the intracellular and extracellular application of a peptide that mimics the Cx32 cytoplasmic loop. All-atom molecular dynamics simulations suggest that loss of the C-terminus in the mutant hemichannel triggers abnormal fluctuations of the cytoplasmic loop which are prevented by binding to the mimetic peptide. Experiments that stimulated R220X hemichannel opening by cell depolarization displayed reduced voltage sensitivity with respect to wild-type hemichannels which was explained by loss of subconductance states at the single channel level. Finally, experiments of intercellular diffusion mediated by wild-type or R220X gap junction channels revealed similar unitary permeabilities to ions, signalling molecules (cAMP) or larger solutes (Lucifer yellow). Taken together, our findings support the hypothesis that paracrine signalling alteration due to Cx32 hemichannel dysfunction underlies CMTX1 pathogenesis and suggest a candidate molecule for novel studies investigating a therapeutic approach.

Ionic currents in hair cells dissociated from frog semicircular canals after preconditioning under microgravity conditions.

The effects of microgravity on the biophysical properties of frog labyrinthine hair cells have been examined by analyzing calcium and potassium currents in isolated cells by the patch-clamp technique. The entire, anesthetized frog was exposed to vector-free gravity in a random positioning machine (RPM) and the functional modification induced on single hair cells, dissected from the crista ampullaris, were subsequently studied in vitro. The major targets of microgravity exposure were the calcium/potassium current system and the kinetic mechanism of the fast transient potassium current, I(A). The amplitude of I(Ca) was significantly reduced in microgravity-conditioned cells. The delayed current, I(KD) (a complex of I(KV) and I(KCa)), was drastically reduced, mostly in its I(KCa) component. Microgravity also affected I(KD) kinetics by shifting the steady-state inactivation curve toward negative potentials and increasing the sensitivity of inactivation removal to voltage. As concerns the I(A), the I-V and steady-state inactivation curves were indistinguishable under normogravity or microgravity conditions; conversely, I(A) decay systematically displayed a two-exponential time course and longer time constants in microgravity, thus potentially providing a larger K(+) charge; furthermore, I(A) inactivation removal at -70 mV was slowed down. Stimulation in the RPM machine under normogravity conditions resulted in minor effects on I(KD) and, occasionally, incomplete I(A) inactivation at -40 mV. Reduced calcium influx and increased K(+) repolarizing charge, to variable extents depending on the history of membrane potential, constitute a likely cause for the failure in the afferent mEPSP discharge at the cytoneural junction observed in the intact labyrinth after microgravity conditioning.