Design of Cultured Neuron Networks in vitro with Predefined Connectivity Using Asymmetric Microfluidic Channels

Neuronal circuits on a chip for biological network monitoring

Cultured neuronal networks (CNNs) are a robust model to closely investigate neuronal circuits' formation and monitor their structural properties evolution. Typically, neurons are cultured in plastic plates or, more recently, in microfluidic platforms with potentially a wide variety of neuroscience applications. As a biological protocol, cell culture integration with a microfluidic system provides benefits such as accurate control of cell seeding area, culture medium renewal, or lower exposure to contamination. The objective of this report is to present a novel neuronal network on a chip device, including a chamber, fabricated from PDMS, vinyl and glass connected to a microfluidic platform to perfuse the continuous flow of culture medium. Network growth is compared in chips and traditional Petri dishes to validate the microfluidic chip performance. The network assessment is performed by computing relevant topological measures like the number of connected neurons, the clustering coefficient, and the shortest path between any pair of neurons throughout the culture's life. The results demonstrate that neuronal circuits on a chip have a more stable network structure and lifespan than developing in conventional settings, and therefore this setup is an advantageous alternative to current culture methods. This technology could lead to challenging applications such as batch drug testing of in vitro cell culture models. From the engineering perspective, a device's advantage is the chance to develop custom designs more efficiently than other microfluidic systems.

Experimental Platform to Study Spiking Pattern Propagation in Modular Networks In Vitro

The structured organization of connectivity in neural networks is associated with highly efficient information propagation and processing in the brain, in contrast with disordered homogeneous network architectures. Using microfluidic methods, we engineered modular networks of cultures using dissociated cells with unidirectional synaptic connections formed by asymmetric microchannels. The complexity of the microchannel geometry defined the strength of the synaptic connectivity and the properties of spiking activity propagation. In this study, we developed an experimental platform to study the effects of synaptic plasticity on a network level with predefined locations of unidirectionally connected cellular assemblies using multisite extracellular electrophysiology.

Novel Concept of Micro Patterned Micro Titer Plates Fabricated via UV-NIL for Automated Neuronal Cell Assay Read-Out

The UV-nanoimprint lithography(UV-NIL) fabrication of a novel network of micron-sized channels, forming an open channel microfluidic system is described. Details about the complete manufacturing process, from mastering to fabrication in small batches and in high throughput with up to 1200 micro titer plates per hour is presented. Deep insight into the evaluation of a suitable UV-curable material, mr-UVCur26SF is given, presenting cytotoxic evaluation, cell compatibility tests and finally a neuronal assay. The results indicate how the given pattern, in combination with the resist, paves the way to faster, cheaper, and more reliable drug screening.

Electrophysiology Read-Out Tools for Brain-on-Chip Biotechnology

Brain-on-Chip (BoC) biotechnology is emerging as a promising tool for biomedical and pharmaceutical research applied to the neurosciences. At the convergence between lab-on-chip and cell biology, BoC couples in vitro three-dimensional brain-like systems to an engineered microfluidics platform designed to provide an in vivo-like extrinsic microenvironment with the aim of replicating tissue- or organ-level physiological functions. BoC therefore offers the advantage of an in vitro reproduction of brain structures that is more faithful to the native correlate than what is obtained with conventional cell culture techniques. As brain function ultimately results in the generation of electrical signals, electrophysiology techniques are paramount for studying brain activity in health and disease. However, as BoC is still in its infancy, the availability of combined BoC-electrophysiology platforms is still limited. Here, we summarize the available biological substrates for BoC, starting with a historical perspective. We then describe the available tools enabling BoC electrophysiology studies, detailing their fabrication process and technical features, along with their advantages and limitations. We discuss the current and future applications of BoC electrophysiology, also expanding to complementary approaches. We conclude with an evaluation of the potential translational applications and prospective technology developments.

Comprehensive Evaluation of Stable Neuronal Cell Adhesion and Culture on One-Step Modified Polydimethylsiloxane Using Functionalized Pluronic

Polydimethylsiloxane (PDMS) is a popular and property-advantageous material for developing biomedical microsystems and advancing cell microengineering. The requirement of constructing a robust cell-adhesive PDMS interface drives the exploration of simple, straightforward, and applicable surface modification methods. Here, a comprehensive evaluation of highly stable neuronal cell adhesion and culture on the PDMS surface modified in one step using functionalized Pluronic is presented. According to multiple comparative tests, this modification is sufficiently verified to enable more significant cell adhesion and spreading in both quantity and stability, higher neuronal differentiation and viability/growth, more complete formation of the neuronal network, and stabler neuronal cell culture than the common coating tools on the PDMS substrate. The comparable and even superior cellular effects of this modification on PDMS to the standard coating of polystyrene for in vitro neurological research are demonstrated. Long-term microfluidic neuron culture with stable adhesion and high differentiation on the modified PDMS interface is accomplished, too. The achievement provides a detailed experimental demonstration of this simple and effective modification for strengthening neuronal cell culture on the PDMS substrate, which is useful for potential applications in the fields of neurobiology, neuron microengineering, and brain-on-a-chip.

Impact of Physical Obstacles on the Structural and Effective Connectivity of in silico Neuronal Circuits

Scaffolds and patterned substrates are among the most successful strategies to dictate the connectivity between neurons in culture. Here, we used numerical simulations to investigate the capacity of physical obstacles placed on a flat substrate to shape structural connectivity, and in turn collective dynamics and effective connectivity, in biologically-realistic neuronal networks. We considered μ-sized obstacles placed in mm-sized networks. Three main obstacle shapes were explored, namely crosses, circles and triangles of isosceles profile. They occupied either a small area fraction of the substrate or populated it entirely in a periodic manner. From the point of view of structure, all obstacles promoted short length-scale connections, shifted the in- and out-degree distributions toward lower values, and increased the modularity of the networks. The capacity of obstacles to shape distinct structural traits depended on their density and the ratio between axonal length and substrate diameter. For high densities, different features were triggered depending on obstacle shape, with crosses trapping axons in their vicinity and triangles funneling axons along the reverse direction of their tip. From the point of view of dynamics, obstacles reduced the capacity of networks to spontaneously activate, with triangles in turn strongly dictating the direction of activity propagation. Effective connectivity networks, inferred using transfer entropy, exhibited distinct modular traits, indicating that the presence of obstacles facilitated the formation of local effective microcircuits. Our study illustrates the potential of physical constraints to shape structural blueprints and remodel collective activity, and may guide investigations aimed at mimicking organizational traits of biological neuronal circuits.

Potential of Microfluidics and Lab-on-Chip Platforms to Improve Understanding of "prion-like" Protein Assembly and Behavior

Human aging is accompanied by a relevant increase in age-associated chronic pathologies, including neurodegenerative and metabolic diseases. The appearance and evolution of numerous neurodegenerative diseases is paralleled by the appearance of intracellular and extracellular accumulation of misfolded proteins in affected brains. In addition, recent evidence suggests that most of these amyloid proteins can behave and propagate among neural cells similarly to infective prions. In order to improve understanding of the seeding and spreading processes of these "prion-like" amyloids, microfluidics and 3D lab-on-chip approaches have been developed as highly valuable tools. These techniques allow us to monitor changes in cellular and molecular processes responsible for amyloid seeding and cell spreading and their parallel effects in neural physiology. Their compatibility with new optical and biochemical techniques and their relative availability have increased interest in them and in their use in numerous laboratories. In addition, recent advances in stem cell research in combination with microfluidic platforms have opened new humanized in vitro models for myriad neurodegenerative diseases affecting different cellular targets of the vascular, muscular, and nervous systems, and glial cells. These new platforms help reduce the use of animal experimentation. They are more reproducible and represent a potential alternative to classical approaches to understanding neurodegeneration. In this review, we summarize recent progress in neurobiological research in "prion-like" protein using microfluidic and 3D lab-on-chip approaches. These approaches are driven by various fields, including chemistry, biochemistry, and cell biology, and they serve to facilitate the development of more precise human brain models for basic mechanistic studies of cell-to-cell interactions and drug discovery.

Integrating Biosensors in Organs-on-Chip Devices: A Perspective on Current Strategies to Monitor Microphysiological Systems

Organs-on-chip (OoC), often referred to as microphysiological systems (MPS), are advanced in vitro tools able to replicate essential functions of human organs. Owing to their unprecedented ability to recapitulate key features of the native cellular environments, they represent promising tools for tissue engineering and drug screening applications. The achievement of proper functionalities within OoC is crucial; to this purpose, several parameters (e.g., chemical, physical) need to be assessed. Currently, most approaches rely on off-chip analysis and imaging techniques. However, the urgent demand for continuous, noninvasive, and real-time monitoring of tissue constructs requires the direct integration of biosensors. In this review, we focus on recent strategies to miniaturize and embed biosensing systems into organs-on-chip platforms. Biosensors for monitoring biological models with metabolic activities, models with tissue barrier functions, as well as models with electromechanical properties will be described and critically evaluated. In addition, multisensor integration within multiorgan platforms will be further reviewed and discussed.

3D-Printed Soft Lithography for Complex Compartmentalized Microfluidic Neural Devices

Compartmentalized microfluidic platforms are an invaluable tool in neuroscience research. However, harnessing the full potential of this technology remains hindered by the lack of a simple fabrication approach for the creation of intricate device architectures with high-aspect ratio features. Here, a hybrid additive manufacturing approach is presented for the fabrication of open-well compartmentalized neural devices that provides larger freedom of device design, removes the need for manual postprocessing, and allows an increase in the biocompatibility of the system. Suitability of the method for multimaterial integration allows to tailor the device architecture for the long-term maintenance of healthy human stem-cell derived neurons and astrocytes, spanning at least 40 days. Leveraging fast-prototyping capabilities at both micro and macroscale, a proof-of-principle human in vitro model of the nigrostriatal pathway is created. By presenting a route for novel materials and unique architectures in microfluidic systems, the method provides new possibilities in biological research beyond neuroscience applications.

Microfluidic Multielectrode Arrays for Spatially Localized Drug Delivery and Electrical Recordings of Primary Neuronal Cultures

Neuropathological models and neurological disease progression and treatments have always been of great interest in biomedical research because of their impact on society. The application of in vitro microfluidic devices to neuroscience-related disciplines provided several advancements in therapeutics or neuronal modeling thanks to the ability to control the cellular microenvironment at spatiotemporal level. Recently, the introduction of three-dimensional nanostructures has allowed high performance in both in vitro recording of electrogenic cells and drug delivery using minimally invasive devices. Independently, both delivery and recording have let to pioneering solutions in neurobiology. However, their combination on a single chip would provide further fundamental improvements in drug screening systems and would offer comprehensive insights into pathologies and diseases progression. Therefore, it is crucial to develop platforms able to monitor progressive changes in electrophysiological behavior in the electrogenic cellular network, induced by spatially localized injection of biochemical agents. In this work, we show the application of a microfluidic multielectrode array (MEA) platform to record spontaneous and chemically stimulated activity in primary neuronal networks. By means of spatially localized caffeine injection via microfluidic nanochannels, the device demonstrated its capability of combined localized drug delivery and cell signaling recording. The platform could detect activity of the neural network at multiple sites while delivering molecules into just a few selected cells, thereby examining the effect of biochemical agents on the desired portion of cell culture.

Neurohybrid Memristive CMOS-Integrated Systems for Biosensors and Neuroprosthetics

Here we provide a perspective concept of neurohybrid memristive chip based on the combination of living neural networks cultivated in microfluidic/microelectrode system, metal-oxide memristive devices or arrays integrated with mixed-signal CMOS layer to control the analog memristive circuits, process the decoded information, and arrange a feedback stimulation of biological culture as parts of a bidirectional neurointerface. Our main focus is on the state-of-the-art approaches for cultivation and spatial ordering of the network of dissociated hippocampal neuron cells, fabrication of a large-scale cross-bar array of memristive devices tailored using device engineering, resistive state programming, or non-linear dynamics, as well as hardware implementation of spiking neural networks (SNNs) based on the arrays of memristive devices and integrated CMOS electronics. The concept represents an example of a brain-on-chip system belonging to a more general class of memristive neurohybrid systems for a new-generation robotics, artificial intelligence, and personalized medicine, discussed in the framework of the proposed roadmap for the next decade period.

Large-scale acoustic-driven neuronal patterning and directed outgrowth

Acoustic manipulation is an emerging non-invasive method enabling precise spatial control of cells in their native environment. Applying this method for organizing neurons is invaluable for neural tissue engineering applications. Here, we used surface and bulk standing acoustic waves for large-scale patterning of Dorsal Root Ganglia neurons and PC12 cells forming neuronal cluster networks, organized biomimetically. We showed that by changing parameters such as voltage intensity or cell concentration we were able to affect cluster properties. We examined the effects of acoustic arrangement on cells atop 3D hydrogels for up to 6 days and showed that assembled cells spontaneously grew branches in a directed manner towards adjacent clusters, infiltrating the matrix. These findings have great relevance for tissue engineering applications as well as for mimicking architectures and properties of native tissues.

Exploring the form- And time-dependent effect of low-frequency electromagnetic fields on maintenance of hippocampal long-term potentiation

Low-frequency electromagnetic field (LF-EMF) stimulation is an emerging neuromodulation tool that is attracting more attention because of its non-invasive and well-controlled characteristics. However, the effect of different LF-EMF features including the forms and the time of addition on neuronal activity has not been completely understood. In this study, we used multi-electrode array (MEA) systems to develop a flexible in vitro magnetic stimulation device with plug-and-play features that allows for real-time delivery of LF-EMFs to biological tissues. Crucially, the method enables different forms of LF-EMF to be added at any time to a long-term potentiation (LTP) experiment without interrupting the process of LTP induction. We demonstrated that the slope of field excitatory postsynaptic potentials (fEPSPs) decreased significantly under post or priming uninterrupted sine LF-EMFs. The fEPSPs slope would continue to decline significantly when LF-EMFs were added two times with a 20-min interval. Paired-pulse ratio (PPR) was analyzed and the results reflected that LF-EMFs induced LTP was expressed postsynaptically. The results of pharmacological experiments indicated that AMPA receptor activity was involved in the process of LTP loss caused by post-LF-EMFs. Moreover, the effect of priming sine or Quadripulse stimulation (QPS)-patterned LF-EMFs depended on the time interval between the end of LF-EMF and the beginning of baseline recording. Interestingly, the effect of sine LF-EMFs on LTP would not disappear within 120 min, while the impact of QPS-patterned LF-EMFs on LTP might disappear after 90 min. These results indicated that LF-EMF might have a form- and time-dependent effect on LTP.

Spatial Properties of STDP in a Self-Learning Spiking Neural Network Enable Controlling a Mobile Robot

Development of spiking neural networks (SNNs) controlling mobile robots is one of the modern challenges in computational neuroscience and artificial intelligence. Such networks, being replicas of biological ones, are expected to have a higher computational potential than traditional artificial neural networks (ANNs). The critical problem is in the design of robust learning algorithms aimed at building a “living computer” based on SNNs. Here, we propose a simple SNN equipped with a Hebbian rule in the form of spike-timing-dependent plasticity (STDP). The SNN implements associative learning by exploiting the spatial properties of STDP. We show that a LEGO robot controlled by the SNN can exhibit classical and operant conditioning. Competition of spike-conducting pathways in the SNN plays a fundamental role in establishing associations of neural connections. It replaces the irrelevant associations by new ones in response to a change in stimuli. Thus, the robot gets the ability to relearn when the environment changes. The proposed SNN and the stimulation protocol can be further enhanced and tested in developing neuronal cultures, and also admit the use of memristive devices for hardware implementation.

Instrumented Microphysiological Systems for Real-Time Measurement and Manipulation of Cellular Electrochemical Processes

Recent advancements in electronic materials and subsequent surface modifications have facilitated real-time measurements of cellular processes far beyond traditional passive recordings of neurons and muscle cells. Specifically, the functionalization of conductive materials with ligand-binding aptamers has permitted the utilization of traditional electronic materials for bioelectronic sensing. Further, microfabrication techniques have better allowed microfluidic devices to recapitulate the physiological and pathological conditions of complex tissues and organs in vitro or microphysiological systems (MPS). The convergence of these models with advances in biological/biomedical microelectromechanical systems (BioMEMS) instrumentation has rapidly bolstered a wide array of bioelectronic platforms for real-time cellular analytics. In this review, we provide an overview of the sensing techniques that are relevant to MPS development and highlight the different organ systems to integrate instrumentation for measurement and manipulation of cellular function. Special attention is given to how instrumented MPS can disrupt the drug development and fundamental mechanistic discovery processes.

Layer-By-Layer: The Case for 3D Bioprinting Neurons to Create Patient-Specific Epilepsy Models

The ability to create three-dimensional (3D) models of brain tissue from patient-derived cells, would open new possibilities in studying the neuropathology of disorders such as epilepsy and schizophrenia. While organoid culture has provided impressive examples of patient-specific models, the generation of organised 3D structures remains a challenge. 3D bioprinting is a rapidly developing technology where living cells, encapsulated in suitable bioink matrices, are printed to form 3D structures. 3D bioprinting may provide the capability to organise neuronal populations in 3D, through layer-by-layer deposition, and thereby recapitulate the complexity of neural tissue. However, printing neuron cells raises particular challenges since the biomaterial environment must be of appropriate softness to allow for the neurite extension, properties which are anathema to building self-supporting 3D structures. Here, we review the topic of 3D bioprinting of neurons, including critical discussions of hardware and bio-ink formulation requirements.

Toward the formation of neural circuits in human brain organoids

Because of the ability to recapitulate normal developmental processes, brain organoids derived from pluripotent stem cells are an important experimental resource to investigate the development and pathogenesis of human brains. Although brain organoids are used in research on diseases such as microcephaly, it has traditionally been difficult to analyze diseases that affect neuronal networks between distant brain regions, as effective brain organoids containing multiple brain regions with defined connectivity have yet to be established. In this review, we discuss strategies to construct such organoids and provide a review on recent progress on brain organoids.

Polarity Induced in Human Stem Cell Derived Motoneurons on Patterned Self-Assembled Monolayers

The control of polarized human neurite/axon development at the single neuron level is critical in geographically directing signal propagation in engineered neural networks, for both in vitro and in vivo applications. While there is an increasing need to exert control over axonal growth for the successful development and establishment of integrative and functional in vitro systems, controlled, polarized distribution of either human-derived neurons or motoneurons in vitro has yet to be reported. In this study, we established the polarized distribution of stem cell derived human motoneurons, using a patterned surface, and maintained the cells in a serum-free system. A surface pattern with defined polarity was developed using self-assembled monolayers (SAMs). A cell permissive SAM, DETA (trimethoxysilyl propyldiethylenetri-amine), combined with photolithography and a nonpermissive fluorinated silane, 13F (tridecafluoro-1,1,2,2-tetrahydroctyl-1-dimethylchloro-silane), generated a surface where neurons only adhered to the designed attachment sites and did so with preferred orientation. In addition, 75% of the cells attached to the patterns were motoneurons compared to their percentage in the standard unpatterned surface which was used as a control condition (20%), demonstrating the preference of these human motoneurons in adhering to the patterns. The ability to dictate the distribution and polarity of human motoneurons will be essential to the engineering of human-based functional in vitro systems in which the control of signal propagation is necessary but more importantly for cell implantation studies. Such systems will greatly benefit the study of motor function as well as aid the development of high-throughput systems for drug screening and test beds for use in preclinical studies related to conditions such as spinal cord injury, ALS, and muscular dystrophy.

Asymmetric confinement for defining outgrowth directionality

Directional connectivity is required to develop accurate in vitro models of the nervous system. This research investigated the interaction of murine neuronal outgrowths with asymmetric microstructured geometries to provide insights into the mechanisms governing unidirectional outgrowth bias. The structures were designed using edge-guidance and critical turning angle principles to study different prohibitive to permissive edge-guidance ratios. The different structures enable outgrowth in the permissive direction, while reducing outgrowth in the prohibitive direction. Outgrowth bias was probabilistic in nature, requiring multiple structures for effective unidirectional bias in primary hippocampal cultures at 14 days in vitro. Arrowhead structures with acute posterior corners were optimal, enabling 100% unidirectional outgrowth bias by virtue of re-routing and delay effects.

Advances in ex vivo models and lab-on-a-chip devices for neural tissue engineering

The technologies related to ex vivo models and lab-on-a-chip devices for studying the regeneration of brain, spinal cord, and peripheral nerve tissues are essential tools for neural tissue engineering and regenerative medicine research. The need for ex vivo systems, lab-on-a-chip technologies and disease models for neural tissue engineering applications are emerging to overcome the shortages and drawbacks of traditional in vitro systems and animal models. Ex vivo models have evolved from traditional 2D cell culture models to 3D tissue-engineered scaffold systems, bioreactors, and recently organoid test beds. In addition to ex vivo model systems, we discuss lab-on-a-chip devices and technologies specifically for neural tissue engineering applications. Finally, we review current commercial products that mimic diseased and normal neural tissues, and discuss the future directions in this field.

Compartmentalized Devices as Tools for Investigation of Human Brain Network Dynamics

Neuropsychiatric disorders have traditionally been difficult to study due to the complexity of the human brain and limited availability of human tissue. Induced pluripotent stem (iPS) cells provide a promising avenue to further our understanding of human disease mechanisms, but traditional 2D cell cultures can only provide a limited view of the neural circuits. To better model complex brain neurocircuitry, compartmentalized culturing systems and 3D organoids have been developed. Early compartmentalized devices demonstrated how neuronal cell bodies can be isolated both physically and chemically from neurites. Soft lithographic approaches have advanced this approach and offer the tools to construct novel model platforms, enabling circuit-level studies of disease, which can accelerate mechanistic studies and drug candidate screening. In this review, we describe some of the common technologies used to develop such systems and discuss how these lithographic techniques have been used to advance our understanding of neuropsychiatric disease. Finally, we address other in vitro model platforms such as 3D culture systems and organoids and compare these models with compartmentalized models. We ask important questions regarding how we can further harness iPS cells in these engineered culture systems for the development of improved in vitro models. Developmental Dynamics 248:65-77, 2019. © 2018 Wiley Periodicals, Inc.

Selective intracellular delivery and intracellular recordings combined in MEA biosensors

Biological studies on in vitro cell cultures are of fundamental importance to investigate cell response to external stimuli, such as new drugs for the treatment of specific pathologies, or to study communication between electrogenic cells. Although three-dimensional (3D) nanostructures brought tremendous improvements on biosensors used for various biological in vitro studies, including drug delivery and electrical recording, there is still a lack of multifunctional capabilities that could help gain deeper insights in several bio-related research fields. In this work, the electrical recording of large cell ensembles and the intracellular delivery of few selected cells are combined on the same device by integrating microfluidic channels on the bottom of a multi-electrode array decorated with 3D hollow nanostructures. The novel platform allows the recording of intracellular-like action potentials from large ensembles of cardiomyocytes derived from human induced pluripotent stem cells (hiPSC) and from the HL-1 line, while different molecules are selectively delivered into single/few targeted cells. The proposed approach shows high potential for enabling new comprehensive studies that can relate drug effects to network level cell communication processes.

An integrated microfluidic/microelectrode array for the study of activity-dependent intracellular dynamics in neuronal networks

In the central nervous system, neurons are organized in specific neural networks with distinct electrical patterns, input integration capacities, and intracellular dynamics. In order to better understand how neurons process information, it is crucial to keep the complex organization of brain circuits. However, performing subcellular investigations with high spatial and temporal resolution in vivo is technically challenging, especially in fine structures, such as axonal projections. Here, we present an on-a-chip system that combines a microfluidic platform with a dedicated matrix of electrodes to study activity-dependent dynamics in the physiological context of brain circuits. Because this system is compatible with high-resolution video-microscopy, it is possible to simultaneously record intracellular dynamics and electrical activity in presynaptic axonal projections and in their postsynaptic neuronal targets. Similarly, specific patterns of electrical activity can be applied to both compartments in order to investigate how intrinsic and network activities influence intracellular dynamics. The fluidic isolation of each compartment further allows the selective application of drugs at identified sites to study activity-dependent synaptic transmission. This integrated microfluidic/microelectrode array (microMEA) platform is a valuable tool for studying various intracellular and synaptic dynamics in response to neuronal activity in a physiologically relevant context that resembles in vivo brain circuits.

Reconstruction of directed neuronal networks in a microfluidic device with asymmetric microchannels

Microfluidic devices for controlling neuronal connectivity in vitro are extremely useful tools for deciphering pathological and physiological processes occurring in neuronal networks. These devices allow the connection between different neuronal populations located into separate culture chambers through axon-selective microchannels. In order to implement specific features of brain connectivity such as directionality, it is necessary to control axonal growth orientation in these devices. Among the various strategies proposed to achieve this goal, one of the most promising and easily reproducible is the use of asymmetric microchannels. We present here a general protocol and several guidelines for the design, production and testing of a new paradigm of asymmetric microchannels geometries based on a "return to sender" strategy. In this method, axons are either allowed to travel between the emitting and receiving chambers within straight microchannels (forward direction), or are rerouted toward their initial location through curved microchannels (reverse direction). We introduce variations of these "arches" microchannels and evaluate their respective axonal filtering capacities. Importantly, one of these variants presents an almost complete filtration of axonal growth in the non-permissive direction while allowing robust axonal invasion in the other one, with a selectivity ratio as high as 99.7%.