Research activities focus on the development of enabling technologies for the sampling, elicitation, and manipulation of neural activity and target at the hardware-embedded emulation and implementation of neural coding schemes. The investigation of electrical recording and stimulation paradigms in the context of network entrainment, learning studies and pattern recognition by in vitro neural networks is supported by the design of electrical and electro-optical manipulation, recording and stimulation devices, by the automation of cell culturing tasks, and by the steering of neural network connectivity with topological and biochemical cues.

We are looking at  response patterns in in vitro networks to chemical,  electrical and optical stimuli, and for  appropriate interpretation algorithms applied  thereon. Besides learning about fundamentals  of signal processing in brain-derived neural networks, one of our objectives is to design  a biohybrid pattern recognition system, which  could also be used as an alternative sensing  device to chemical, biochemical or biomimetic  sensors. The performance of such neurobionic entity might exceed that of  conventional computing and biosensing systems  for its parallel information processing, its ability to store  information and to adapt to changes in the input, and its biologically determined high response sensitivity paired with high selectivity to the type of stimulus.

Design of low-cost, metal-less, flexible all-polymeric microelectrode arrays (polyMEAs) for in vitro electrophysiology and in vivo neuroprosthetics and cardiography

Microelectrode electrophysiology has become a widespread technique for the extracellular recording of bioelectrical signals. To date, electrodes are made of metals or inorganic semiconductors, or hybrids thereof. We demonstrated that these traditional conductors can be completely substituted by soft, tissue-like electroconductive polymers.

The patented, highly flexible, stretchable, transparent, biocompatible and biostable all-polymer microelectrode arrays (polyMEAs) are based on replica-molded microchannel scaffolds with embedded organic conductors. These polymer electrodes with diameters down to 80 µm reliably transduce neural activity (action potentials and local field potentials) from cardiac and neuronal tissue in noninvasive extracellular in vitro and in vivo recordings.

The presented pathfinding concepts and validation results address the long-standing demand for more flexible and more biocompatible implants such as brain machine interfaces (BMIs). Similarly to the shift from silicon to organic electronics in consumer devices, they may be regarded as a milestone for bringing all-polymer electronics to future neuroprosthetics and biomedical devices.


polyMEAs_w200


Development and application of multi-trap optical tweezers and laser microdissection tools for manipulating neural network composition, modifying network interconnectivity, and characterizing biophysical parameters on a cellular and subcellular level

We designed an optical system that combines IR (1064 nm) holographic optical tweezers (HOTs) with a sub-nanosecond-pulsed UV (355 nm) laser microdissector (LMD) for the optical manipulation of single neurons and entire networks in vitro, on both transparent and non-transparent substrates. The phase-modulated laser beam illuminates the sample concurrently or independently from above or below assuring compatibility with different types of microelectrode array (MEA) and patch-clamp electrophysiology.

During the ablation of individual processes at single neuron and neural network level by the LMD, the release of tension in a partially ablated neurite can be quantified by simultaneous HOTs force spectroscopy down to the sub-pico-Newton range to precisely control and measure the inflicted damage. Its effect on network level can be monitored by changes in activity of neural sub-populations with subcellular resolution and high temporal resolution by concurrent calcium imaging and multichannel MEA electrophysiology.

In neural regeneration studies, the modular opto-electrophysiological platform serves as an excellent tool for testing potential pharmacological approaches aimed at the recovery of totally or partially damaged neuronal connections within a circuit.


HOT & LMD_w200


HOT&LMD photo_w200


Design of (µ-fluidic) perfusion chambers and control systems for cell culture automation on the lab bench for providing homogeneous and reproducible physiological conditions and microenvironments for simultaneous and uninterrupted optical and electrophysiological long-term screening studies

To date, a majority of in vitro MEA electrophysiology studies on neural cell cultures are performed in short-term ´snapshot´ experiments lasting from minutes to hours. Acquisition time span is mainly limited by a drift in pH and osmolarity. Therefore, part of the acquisition hardware has to be placed in a CO2 incubator if the continuous recording of neural activity in long-term studies is desired. To become independent from such incubator environment, we devised a simple perfusion concept that allows the uninterrupted (minutes to months) recording of neural activity from MEAs on a lab bench at ambient conditions. It is based on a small lid made of flexible, transparent and gas-permeable poly(dimethylsiloxane) (PDMS) with particular internal shape for the expulsion of gas bubbles, embedded polytetrafluoroethylene (PTFE) tubing and self-sealing septa. It was designed to neither obstruct the mounting of the MEA into the amplifier nor the light path in microscopy. Only temperature (simple T-controller) and pH (chemical buffering) need to be controlled to provide the right physiological conditions. The lid creates an otherwise self-sufficient and stabilized micro-environment that does not require any other incubation scheme. When installed on an inverted microscope, electrophysiology can be combined with time-lapse imaging of network morphology and is accessible to optical manipulation by HOTs and LMD.


PerfusionSetup_w200


Generation of biochemically and topologically µ-textured substrates by soft lithography, inkjet printing, µ-fluidics and laser-assisted substrate processing to guide interconnectivity and support cell adhesion, neural differentiation, and regeneration

In MEA electrophysiology, signal quality of extracellularly recorded signals strongly depends on the interface chemistry between cells and substrate. It not only determines how well the cells attach to the substrate and to the electrodes, but influences substrate biocompatibility, its biostability, cell differentiation as well as cell fate. If such interface chemistry is furthermore structured topologically, neural interconnectivity may be steered to result in pre-determined functional organization of the network. We resort to µ-stamping (for flat substrates) and piezo-inkjet printing (for any substrate topography) of common adhesion mediators (e.g., polyethyleneimine (PEI), poly-D/L-lysine (PDL or PLL)) and differentiation factors (e.g., laminin, fibronectin) supplemented by physical constraints such as microchannel scaffolds (primarily made of PDMS) to design highly adjustable neural network architectures.


MicrochannelInlaysCollage_w200


PEI Pattern Collage_w200


Development of tissue-analog 2D and 3D gel-supported cell culturing strategies for increasing the number of computational neural entities and a more tissue-like topology


We test various natural and artificial gelling agents for their compatibility with serum-free medium to provide a ECM-like scaffold to interact with, to protect cells from shear forces and medium evaporation, to stabilize the microenvironment around cells for efficient cell-cell communication, thereby increasing the cell density compared to classical monolayer 2D culture.

Cortical neurons embedded in guar gel

Emulation of the roundworm Caenorhabditis elegans


Biological neural systems are powerful, robust and highly adaptive computational entities that outperform conventional computers in almost all aspects of sensory-motor integration. Despite dramatic progress in information technology, there is a big performance discrepancy between artificial computational systems and brains in seemingly simple orientation and navigation tasks. In fact, no system exists that can faithfully reproduce the rich behavioural repertoire of the tiny worm Caenorhabditis elegans which features one of the simplest nervous systems in nature made of 302 neurons and about 8000 connections. The Si elegans project aims at providing this missing link.


Research funding by the following sources is greatly acknowledged:


European Union Research Funding Agency

European Union FP7 Research Funding Programme

Telethon Foundation Italy

Compagnia di San Paolo Funds

Italian Institute of Technology



http://www.si-elegans.eu/