Periodic Reporting for period 1 - NEUREKA (A smart, hybrid neural-computo device for drug discovery)
Reporting period: 2019-12-01 to 2021-05-31
To help reverse this trend, we will bolster one of the weakest links in the drug discovery pipeline: the functional screening of compounds in vitro, via cellular assays. While cellular assays are insuperable in giving experimental access to both the molecular and the cellular function of brain circuits, they are limited by the loss of those cell-to-cell interactions governing native tissues. This is particularly detrimental for modelling brain pathology in vitro, where cultured neurons are isolated and deprived of native in-brain synaptic inputs that are fundamental for driving neuronal and network processing. Current technologies have little-if any- control over the network excitability, plasticity or connectivity; furthermore, they are limited to single cell or sparse observational readouts (e.g. by patch-clamp, optical imaging or Micro-Electrode Arrays, MEA). For these reasons, assessing how molecular deficits (e.g involving synaptic plasticity) and treatments modify the function of neuronal circuits remains a prohibitive challenge. We propose a visionary and innovative hybrid technology, whereby nanoelectrodes and sophisticated computational models of neuronal circuits are combined to readout and manipulate the activity and connectivity of cultured neuronal networks of Alzheimer’s disease, with subcellular accuracy. For the first time ever, a biophysical model will drive dendritic stimulation of Alzheimer’s cultured neurons, enabling us to reproduce the pathological network connectivity and excitability, in addition to molecular features of the disease. Our system will also allow us to drive the network to disease states that reveal deficits, so as to optimize the assessment of drug effects. By increasing success rates in drug discovery, this technology is envisioned to reduce costs and minimize animal use prior to pre-clinical/clinical trials.
Regarding the development of the in-vitro platform for neuron-chip interfacing through nanowires, we performed a thorough material study of the PtSi layer, including the surface oxidation for a proper integration of the coating layers. We also fabricated the 60-channels NWA and established the procedure for electrical characterization of such a system in recording (impedance) and stimulation. We also developed a low temperature NWA process compatible with an integration on CMOS circuit.
We have also developed an initial version of the AlzModel software, the computational model of the neuronal circuit mimicking Alzheimer's disease, which will be used for initial testing of the 60-channel nanowire array. The computational model of the Synaptor, which will serve as the interface between the nanoelectrodes and the cultured neurons, has been developed. This model must now be validated with electrophysiological data. Towards this goal, we performed in vitro measurements from rat hippocampus (E19 embryonic) neuronal cultures on nanowire array prototypes and on the high-density planar multi-electrode array provided by our industrial partner.
Finally, we made some progress towards integrating and test the NEUREKA system at the hardware-software levels. Specifically, we implemented the first integrated NEUREKA system between the 60-channel NMA and the biophysical AlzModel, through HW/SW intercommunication platforms. We also developed a software package for automatic detection and functional characterization of subcellular axonal physiological signals, across hundreds of neurons within a neuronal network. Lastly, we have established protocols for stable and reproducible culturing of human iPSC-derived glutamatergic neurons and human astrocytes, that will be used to culture disease neurons once the NEUREKA platform is ready.
The project will also contribute with new knowledge and technological discoveries across numerous disciplines: new materials optimized for nanowire conductivity; novel devices (nanowire arrays), able to drive neurons in unprecedented resolution, with dendritic stimulation and recording; powerful, new software for reading-out electrical signals, processing and driving model neurons in speeds ensuring on-line control; new biophysical models of AD, realistically capturing the pathology in large scale networks and seemingly interfacing with MEA. Large-scale neural data (simultaneous intracellular electrical signals from thousands of neurons), for better analysis and modelling. Within the first 18 months of the project, the scientific progress that has been made beyond the state of the art concerns primarily the integration of the AlzModel with the nanowire array and the ability to stimulate the nanowires using the activity of model neurons.
Importantly, NEUREKA will have a positive impact on the training of young, skilled researchers by supporting their research projects. Already, within the 18 months of the project, a total of 11 young students and postdoctoral fellows are hired on the project and more will follow soon.
By its completion, NEUREKA will also have a significant impact on market creation. It will lead the way for exploiting the lack of high-throughput devices to measure neurons’ intracellular signals and the demand for new and accurate methods to screen hundreds of available human samples, generated by iPSC technology. In the long run, our SME (MaxWell) aims to develop NEUREKA in a commercial product: the first multi-well drug screening platform that enables direct access to a full dynamic range of intracellular electrical potentials from thousands of neurons simultaneously. Our pharma advisors will consider using NEUREKA for drug screening in various neurodegenerative diseases, thus greatly expanding its commercial worth and impact on industries well beyond the participating SME.