A self-aligned functionalization approach to order neuronal networks

Subscriber access provided by Kaohsiung Medical University

Biological and Environmental Phenomena at the Interface

A self-aligned functionalization approach to order neuronal networks at the single cell level. Adrien Casanova, Charline Blatché, Cecile Ferre, Hélène Martin, Daniel GONZALEZ-DUNIA, Liviu Nicu, and Guilhem Larrieu Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b00529 • Publication Date (Web): 12 May 2018 Downloaded from http://pubs.acs.org on May 14, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

A self-aligned functionalization approach to order neuronal networks at the single cell level.

Adrien CASANOVA1, Marie-Charline BLATCHE1, Cécile A. FERRE2, Hélène MARTIN2, Daniel GONZALEZDUNIA2, Liviu NICU1, Guilhem LARRIEU1*. 1

2

LAAS-CNRS, Université de Toulouse, CNRS, Toulouse, France INSERM UMR 1043, CNRS UMR 5282, Université de Toulouse, France. *[email protected]

Abstract

Despite significant progress, our knowledge of the functioning of the central nervous system still remains scarce to date. A better understanding of its behavior, in either normal or diseased conditions, goes through an increased knowledge of basic mechanisms involved in neuronal function, including at the single cell level. This has motivated significant efforts for the development of miniaturized sensing devices to monitor neuronal activity with high spatial and signal resolution. One of the main challenges remaining to be addressed in this domain is, however, the ability to create in vitro spatially-ordered neuronal networks at low density with a precise control of the cell location, to ensure proper monitoring of the activity of a defined set of neurons. Here, we present a novel self-aligned chemical functionalization method, based on repellant surface with patterned attractive areas, which permits the elaboration of low density neuronal network down to individual cells with a high control of the soma location and axonal growth. This approach is compatible with CMOS line technology at wafer scale and allows performing the cell culture on packaged chip outside microelectronics facilities. Rat cortical neurons were cultured on such patterned surfaces for over one month and displayed a very high degree of organization in large networks. Indeed, more than 90% of the network nodes were settled by a soma and 100% of the connecting lines were occupied by a neurite, with a very good selectivity (low parasitic cell connections). After optimization, networks composed of 75% of unicellular nodes were obtained, together with a control at the micron scale of the location of the somas. Finally, we demonstrated that the dendritic neuronal growth was guided by the surface functionalization, even when micrometer scale topologies were encountered and we succeeded to control the extension growth

1 Environment ACS Paragon Plus

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

along 1D-aligned nanostructures with sub-micrometrical scale precision. This novel approach now opens the way for precise monitoring of neuronal network activity at the single cell level.

Introduction

Due to the complexity of the brain structure and the difficulty of access linked to the presence of many non-neural cells and the vasculature, it is very difficult to isolate the analysis of neuronal function in vivo. In contrast, a very sharp level of analysis can be achieved by in vitro systems, notably when working at the single cell level. A better understanding of neuronal communication at the level of small neuronal networks actually represents a fundamental step towards a better description of overall neuronal function [1]. Although such systems are simplified in comparison with in vivo neural networks, it was shown that isolated neurons for which axonal growth have been oriented remarkably retain their physiological, physical and dynamic transports properties [2]. Consequently, finding new methods at the single-cell level aiming at spatially tracking propagation of action potentials fired by a defined set of neurons is a very active field of research [3], [4]. In this context, the miniaturization of electronic components, together with the emergence of nano-biotechnology coupled with the mastering of ordered cultures hold great promise to get further insight on neuronal function at the single-cell level. Several types of nanodevices have been proposed to increase the recording resolution of neuronal activity [3]–[9]. One of the main issues of interfacing these nanodevices with ordered cultures is, however, the difficulty to maintain neurons alive when cultured at low density [10], [11] and to limit the potential parasitic impact of glial cells during recordings. Although some works succeeded to culture hippocampal or cortical neurons at low density for several weeks [12], [13], the patterning of low density neuronal networks down to single neurons with a perfect control of the localization, together with its interfacing with recording capabilities has not been satisfactorily achieved to date (Table 1). To create spatially-ordered neuronal networks by controlling the location of somas and their extensions, two main approaches have been proposed. First, the positioning of somas and the orientation of dendritic growth can be forced by surface topographic structuration, where mechanical constraints impose the organization of the cellular network. The surface structuration can be done by SU8 channels and holes that trap somas and guide neurite growth [14], [15], or by nanostructures patterned on the chip surface such as nanopillars [16], mushrooms [17], [18], microtubes [19], [20] or scaffolds [21]. Although these topographic strategies indeed allow an efficient control of network architecture, they suffer from fabrication compatibilities when coupled with the micro-patterning of recording devices.


Page 2 of 18

Page 3 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

The second approach to create ordered neuronal networks is based on chemical surface functionalization. In most cases, attractive areas are created by adhesion molecules such as Laminin, poly-Ornithine, poly-Ethylenimine or poly-Lysine, which favor cellular adhesion on the substrate. In some cases, they can be associated with repulsive areas coated with repellant functionalization (PEG or hydrophobic treatment), thereby allowing the control of neuronal growth [3], [22]–[28]. This approach has been widely used in the 90's and early 2000's and was based on microlithography [22], [29] or micro-contact printing [25], [30]–[35] and allowed localized neuronal attachment and control of extension growth on large micropatterns. However, the recent emergence of nano-bio devices highlighted a remaining major and still unresolved challenge, i.e., to improve both the cell location and the rate of spots occupied by a single soma in order to efficiently couple such ordered neuronal networks with nanodevices for electrical activity monitoring. [4], [5], [7], [23], [36], [37]. Indeed, micro-contact printing presented intrinsic limitations due to its alignment precision (500 nm). Likewise, conventional microlithography based on resist structuration suffered from the interdependence between microfabrication and biological analysis (Figure 1a), thereby introducing serious drawbacks, such as degradation of the adhesion molecule during the resist removal or the limited time allowed between the different procedures.

To address these different issues, we propose here a new robust, “self-aligned” chemical functionalization approach at wafer scale which aims at precisely locating neuronal bodies and guiding the growth of their extensions. This strategy, which is easier and less constraining to process than the conventional ones, allows creating ordered neuronal networks with a very high rate of perfectly located single cells and a precise location of axonal growth with a sub-micrometer scale precision, as well as improved survival rate. Moreover, the complete dissociation between microfabrication and biological processes allows performing the biological assays out of the clean room facilities, further improving the quality and flexibility of functionalization steps.

Results and discussion

To create ordered neuronal networks and to selectively localize neuronal attachment, the classical approaches usually perform conventional photolithography in order to pattern hydrophobic (repulsive) and hydrophilic (attractive) zones. From a technological point of view, the photolithography step is performed after the hydrophobic treatment. However, due to the weak resist adhesion on such modified surfaces, the hydrophobic properties of the surface must be subsequently degraded to enhance the adherence of the resist to the substrate [23], leading to a


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 18

decrease of the contrast between zones with and without neuronal growth. In contrast to these conventional approaches that uses a topographic resist patterning to localize the adhesion molecule, our “self-aligned” functionalization (SAF) approach is based on a very high hydrophilic/hydrophobic surface contrast that promotes a very precise control of the positioning of adhesion molecules. In a first step, the micro-patterns are defined by photolithography on a silicon dioxide surface thermally grown on a (100) silicon substrate. The substrate surface is coated with a photoresist and insolated with i-line photo-repeater (Stepper CANON FPA 3000i4) to allow a large-scale fabrication (6-inch wafer) and a high pattern resolution (Critical Dimension: 350 nm). Next, the hydrophobic treatment is performed using a vapor deposition technique with a Surface Prepared Deposition (SPD) equipment (Memsstar), to create a hydrophobic self-assembled monolayer (SAM) on areas not covered by the resist. The resist can then be stripped by washes in acetone and piranha solution (H2O2/H2SO4, 1:1) to obtain a very high contrast between hydrophilic and hydrophobic areas. Finally, the molecules allowing cell adhesion are selectively coated on the hydrophilic micro-patterns previously covered by the resist. Thanks to the SAF process, the selective coating with the adhesion molecules is based on high hydrophobic/hydrophilic contrast, contrary to conventional approaches [22], [23] where the coating is controlled by the structuration of the resist. By avoiding resist adhesion issues on hydrophobic surfaces, a very high level of repellant treatment can be used outside the patterned network. Moreover, a complete dissociation between microfabrication and biological steps is obtained, allowing to perform the coating step with adhesion molecules only when needed for the neuronal culture (Figure 1a), without any additional step in clean room facilities. One of the key steps of this procedure is to ensure that the hydrophobic treatment of the substrate is as efficient as the topological structuration induced by a resist masking in order to ensure a subsequent selective coating of the adhesion molecules on hydrophilic surface. Commonly, hydrophobic treatment is achieved using a liquid-phase SAM, grown with a Chlorosilane-based precursor [22], [3], [23], [28]. However, this approach presents some drawbacks such as complex process control, possible risks of contamination of the substrate during wet chemical process or run to run reliability issues with large variability of SAM coating efficiency [38], [39]. As a reference, we have reproduced a liquid phase process similar to what was described in [3], [23], leading to a measured contact angle of 60° +/- 1°. In addition, the stability of the passivation was very weak, as we observed a very quick degradation upon acetone or piranha washes (Figure 1b). To overcome these

issues,

we

used

a

vapor

deposition

treatment

(SPD)

with

1H,1H,2H,2H-

Perfluorodecyltrichlorosilane (FDTS). This vapor phase approach demonstrated a very high hydrophobic surface (contact angle of 110° +/- 1°) at large scale (6 inch-wafer) with a long-lasting stability (more than two months). It equally exhibited high chemical stability, with no degradation


Page 5 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

even after washes in acetone or piranha solutions (Figure 1b), while avoiding issues previously mentioned such as lack of reliability, contamination issues or process difficulties [39], [40].

Figure 1: (a) Comparison between the conventional chemical treatments used to order neuronal networks (1) and our new self-aligned functionalization approach (SAF) (2) which allows a complete dissociation between microfabrication and biological assays. (b) Measurement of time and chemical stability for the hydrophobic liquid phase treatment and the vapor phase treatment with the SPD device. (c) Fluorescent images of several patterned surfaces (either large gear wheels patterns, left, and narrow 1 µm-wide lines, right). treated with our new SAF procedure and coated for 5 h with FITC-labeled Poly-L-Lysine. Scale bars are, respectively, 300 µm and 100 µm.

To validate this approach, we visualized the deposition of fluorescein isothiocyanate (FITC)-labeled Poly-L-Lysine after 5 hours of coating to ensure the uniformity of deposition of this adhesive molecule (See Supporting Information). As expected, Poly-L-Lysine only stuck on hydrophilic surfaces. We observed high selectivity and resolution, with a control of the location of the adhesion molecule at the micron scale, even when using complex designs (Figure 1c). Next, we investigated the growth of neuronal cultures on such patterned surfaces. We used cortical neurons prepared from embryonic Sprague-Dawley rats at gestational day 18. Our protocol aimed at minimizing the number of glial cells present in the culture (elimination by density gradient and culture in serum-free medium), since they would divide after plating and thus perturb the creation of well-defined neuronal networks. We designed a micro-pattern network, composed of large spots for soma settling, which were connected with lines to guide neurite growth (Figure 2a). For the biological assays, to favor cell adhesion, the chips were functionalized using an overnight coating based on two complementary adhesion


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

molecules (Laminin and poly-Ornithine) [41], [42]. We used Laminin because it is an efficient adhesion molecule, representing one of the main extracellular matrix proteins found in the brain and because many studies demonstrated its benefit to promote neurite elongation and differentiation [43], [44]. Because Laminin cannot graft directly on a SiO2 surface, we first applied a poly-Ornithine coating [45]. Neurons were plated immediately after dissection and maintained in serum-free Neurobasal medium (see details in supplementary section) with 5 % of CO2 and at 37 °C [46]–[50]. A few hours after plating, we observed cell adhesion by sedimentation on the substrate. When neurons were located on hydrophobic surfaces free of adhesion molecules, they either migrated to hydrophilic areas or died by apoptosis. In contrast, neurons readily settled onto coated hydrophilic surfaces and started to extend projections. As a result, well-ordered neuronal networks were successfully created, with somas being positioned on the nodes of the network and neurites connecting each node with a very good level of organization in large networks, even when culture was performed at very low density (Figure2 b-e). After 3 days in vitro (DIV), more than 90 % of network nodes (400 µm² spots) were settled by at least one soma and almost 100 % of the connecting lines were occupied by a neurite, with a very good selectivity (low parasitic cell connections). As shown in Figure 2d when using a pattern similar to the laboratory (LAAS) logo, more complex designs demonstrated the high ability of the SAF approach to guide extension growth along the functionalized lines. We next sought to improve the survival of our ordered neuronal networks cultured at low density, even after neurons had maturated (10 DIV). To this aim, different approaches have been described, such as covering the neuronal culture with a separate glass coverslip supporting a culture of glial cells, to be used as feeding layer [13], [23], [51]. Rather, we chose an easier-to-implement approach, based on the use of conditioned medium derived from glial cells [52]–[55]. To evaluate the impact of the culture medium on the metabolic activity of neurons, we used a colorimetric assay, i.e., MTT assay (shown in Supplementary Information) [56], [57]. We first plated cortical neurons on several flat surfaces (i.e., silicon, silicon oxide and aluminum) and cultivated them in serum-free Neurobasal medium and we used a reference well of neurons grown directly on plastic without any chip. Although the measurement of the metabolic activity proved the biocompatibility of all materials at 3 DIV (>80 % compared to the reference), it also revealed a significant decrease (of at least 70 %) of neuronal metabolic activity from 3 DIV to 9 DIV. In sharp contrast, when medium was replaced by medium conditioned by glial cells, the mitochondrial activity of neurons raised drastically (+730 % at 9 DIV compared to the reference at 3 DIV). Based on these findings, we thereafter systematically replaced the culture medium after 3 DIV by a conditioned medium derived from glial cells, in order to achieve long term survival of low density neuronal networks. The Figure 2f shows a fluorescent


Page 6 of 18

Page 7 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

picture of a mature 15 DIV neuronal network after a viability assay (Calcein AM/Ethidium homodimer) further proving that cells indeed remained healthy, even when cultured at low density.

Figure 2: (a-c) Fluorescent images of several patterned surfaces treated with our new SAF procedure and coated for 5 h with FITC-labeled Poly-L-Lysine. (b-d) Confocal images of rat cortical neurons cultured for three days with a perfect selectivity of on patterned areas shown in (a) and (c). Nuclei are labeled with DAPI (blue) and dendrites and axons with MAP-2 (red). (e) SEM images of rat cortical neurons grown for three days in an ordered network, with somas being positioned on the nodes of this network and neurites growing along functionalized lines. (f) Fluorescent image of an ordered neuronal network after a viability assay (Calcein AM/ EthD-1) performed after 15 days of culture, showing that neurons are still alive and healthy. Scale bars are 100 µm for all pictures.

We next assessed the impact of two different parameters in order to improve both the positioning of somas down to a single cell and the organization of the neuronal network. All statistics presented in this paper were performed using three independent sets of cultures, overall including more than 2000 spots. First, we evaluated the influence of the cellular concentration using three different densities (from 500 cells/mm² to 200 cells/mm²), on the same spot size (square of 400 µm²) and under the same culture conditions. It appeared that a lower occupation rate was observed when the cell concentration decreased, dropping from 92 % for concentrations of 500 cells/mm², down to 60 % for 200 cells/mm². However, the lowest concentration was also found to be more favorable to maximize the unicellular character. Indeed, the rate of occupied spots bearing a single cell increased from 17.5 % to 41 % when the cell concentration was lowered from 500 cells/mm² to 200 cells/mm² (Figure 3a). Nevertheless, when the cellular density became too low (< 200 cells/mm²), we noted that it severely compromised the quality of neuronal growth and no network could thus be formed.


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In parallel, we analyzed the influence of the design patterns. We reasoned that a reduction of the spot size could improve the positioning of somas as well as increase the unicellular rate. Figure 3 b presents the occupation rate for four different sizes of spot (S= 400 µm2, S= 314 µm2, S= 154 µm2 and S = 78 µm2). Even if the occupation rate logically decreased with the spot size, when considering the occupied spots, we noted that the rate spots occupied by only one neuron strongly increased, from 41 % with 400 µm²-spots to 74 % with 78 µm²-spots. Finally, we combined the two previous findings and sought to pattern selective low density neuronal networks characterized by a maximal single cell occupancy rate at a defined place by using both the smallest size of spots (78 µm²) and a low cell density (200 cells by mm²). This allowed the generation of very well-ordered networks, with almost 75 % of occupied spots bearing a single cell (Figure 3c and e) and where all somas were interconnected. As a matter of fact, the location of the neurons could be perfectly mastered, because the chosen size spot is in the range of the size of a soma size. We next applied a similar strategy to control growth of neuronal extensions, using several widths of lines (patterned from 10 µm to 1 µm) connecting the soma spots, in order to impose the growth track of the neurites. As shown in Figure 3f, a 10 µm-wide line displayed multiple neurites distributed on the same track. In contrast, when the line width was in the same scale (~ 1 µm) than the neurite itself (Figure 3e), we could evidence that a single extention was growing on each functionalized line. As a result, our self-aligned functionalization approach allowed creating neuronal networks with a very high degree of organization, a very precise control of both the location of the somas and growth of neuronal projections, at the single cell level.


Page 8 of 18

Page 9 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 3: (a) Graph describing the impact of the neuronal concentration on the occupation rate of the spots of the network, as well as on the rate of spots bearing a single soma. (b) Graph showing the impact of the spot size on the number of somas by spot for a density of 200 cells/mm². (c-d) Comparison between two sizes of spots 400 µm² c) and 78 µm² d) – performed by immunofluorescence and confocal microscopy, illustrating nuclei labeled with DAPI (blue), dendrites labeled with MAP-2 (red) and axons labeled with TAU (green). (e-f) Immunofluorescence confocal images of two widths of connecting lines (e- 1 µm and f- 10 µm) and nuclei labeled with DAPI (blue), dendrites labeled with MAP-2 (red) and axons labeled with TAU (green). Scale bars are 20 µm for all pictures.

Finally, we evaluated the behavior of SAF on structured surfaces in order to further demonstrate its potentialities (i)

from a phenomenological point of view, i.e., to test whether surface functionalization could still impose the direction of neurite growth even in the presence of topological obstacles) and,

(ii)

from a technological point of view, i.e., to assess whether it was possible to precisely align such ordered networks on nanodevices in order to record neuronal activity (planar nanoFET [3], vertical nanoprobes [4], [58]).

To this aim, we cultured neurons on chemical functionalized patterns aligned on micrometer scale topologies ( Figure 4a). Interestingly, we demonstrated that the growth of neuronal extensions was still guided by the chemical functionalization even when micrometer walls were encountered ( 9 Environment ACS Paragon Plus

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4b and c), proving that our surface chemical treatment is predominant over topographic structuration. Because our approach uses standard CMOS protocols at wafer scale, it could beneficiate both from processes with nanometer scale resolution in structure patterning, but also in terms of multilayer mask alignment. In that scope, the SAF was aligned on multiple vertical 1D nanostructures (diameter of 200 nm and height of 3 µm) using a standard i-line photo-repeater, as shown in Figure 4d. The 1 µm-wide functionalized line was able to guide growth of neurites along these structures, with a perfect control of location at micrometrical scale, thereby promoting a large interaction between nanowires and neurites ( Figure 4e).

Figure 4: (a-b-c) SEM images of rat cortical neurons cultured on functionalized surface with location of somas perfectly controlled on the nodes of the network and neurites growing along lines connecting nodes, even when micrometer scale topologies (walls or bottlenecks) are encountered. (d) SEM images of three aligned 200 nm diameter nanowires with the functionalized line (in blue) and (e) very high control of neurite growth along nanowires at micrometrical scale. f) SEM image of a rat cortical cell (False color green) with the soma perfectly located on 3D vertical nanowires with its axonal extension perfectly following the functionalization above a planar transistor (False color purple) and a single vertical nanowire (False color brown). Scale bars are 10 µm for all pictures except for b-e nanowire-zoom (1 µm).


Page 10 of 18

Page 11 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Table 1 displays a comparison of the self-aligned functionalization technique with other approaches from the literature [23], [25], [29], [34], [35] to pattern neuronal network on electronic device. This synthetic benchmark highlights the benefits of the self-aligned functionalization strategy at large scale with a precise control (sub-micrometer scale) of cell location and dendrite extension at the single cell level. This approach now opens the opportunity to be implemented with new 3D nanodevices [4], [6], [7], [59], [60] for neuronal interfacing, which have emerged in the last years and require high alignment precision (deca-nanometer scale) on tridimensional nanostructures in comparison with the conventional methods. It also opens the way for new perspectives regarding the in vitro monitoring of neuronal activity, since it allows both the refinement of the analysis down to single cells connected in networks, together with the capacity to stimulate and record the cell somas, as well as to track the signal along the axon, as highlighted by the Figure 4 f.

Table 1: Comparison of the SAF technique with other approaches from the literature for neural patterning on electronic devices.

Conclusions In summary, we have designed an innovative and robust self-aligned functionalization strategy, allowing to order neuronal networks at the single cell level with unparalleled precision. This approach presents improved time and chemical stabilities, together with a complete dissociation


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

between microfabrication and biological assays. Wide neuronal networks at low density were obtained by this method, with a very good selectivity (low parasitic cell connection). By tuning the conditions of cell growth and the design of the patterns, we achieved the construction of neural network with a very high rate of unicellular spots (75 % on occupied spots) connected by single neurite extensions. Decreasing the spot size in our patterns improved our control of the location of the somas, resulting in neuronal networks at the single cell level with a micrometrical control of positioning. Finally, we proved that our SAF approach could efficiently guide neurite outgrowth even when micrometer scale topologies were encountered, allowing to precisely control neuronal extension along aligned 1D nanostructures with sub-micrometrical scale precision.

Methods Process of self-aligned functionalization to obtain hydrophobic areas with FDTS and hydrophilic areas coated with an adhesion molecule

Silicon substrate (100) p type (2-10 Ω.cm) was cleaned by oxygen plasma in a TEPLA machine (800 W, 5 min). Then, a photoresist ECI 3012 1.1µm was spin-coated on the substrate and exposed with a CANON stepper to define the pattern. Resist was post-exposure-baked at 110 °C during 90 s, developed with MF-CD-26, rinsed with deionized water (DIW) and blown dry with N2. Substrate was then placed in the SPD machine to create a hydrophobic 1H,1H,2H,2H-Perfluorodecyltrichlorosilane (FDTS) self-assembled monolayer (SAM) only on free areas. After several hours, needed to stabilize the SAM, the resist was stripped by successive washes in acetone and piranha solution (H2O2/H2SO4 1:1) to achieve the microfabrication. The day before plating the neurons, samples were decontaminated with 70 % ethanol during 15 min and soaked overnight with 0.3 mg/mL polyornithine. This coating was rinsed two times with sterile water and replaced by laminin during 2 h. Laminin was rinsed using Neurobasal culture medium just before plating the neurons on the functionalized chips.

Fabrication of the nanostructures

Nanostructures were fabricated using a top-down approach, combining a photolithography step followed by a reactive ion etching to transfer the desired nanopatterns onto the silicon substrate. To


Page 12 of 18

Page 13 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

ensure a large scale and cost-efficient process, a stepper FPA CANON 3000i4 (i-line: 365nm) was used, allowing to reach a critical dimension close to 400 nm. Photolithography was performed on an ECI-coated (thickness = 1.1 µm) silicon substrate with a dose of 750 mJ/cm² and a numerical aperture of 0.63. Then, the sample was post-exposure-baked at 110 °C during 60 s, developed in an ECI developer (MF CD 26) during 35 s and rinsed in DIW using an Automated Resist Processing System (EVG 120). Resist nano-patterns were obtained with a diameter of 400 nm and this diameter could be reduced and tuned using a short plasma O2 step. Then, the vertical resist nanopillars were transferred onto the substrate by RIE based on SF6 (45 sccm), C4F8 (30 sccm) and O2 (2 sccm) to ensure an anisotropic etching. The residual ECI layer was finally removed by plasma O2 (800 W, 15 min) and washes in piranha solution (H202/H2SO4: 1:1) during 5 min. rinsed in DIW and dried using a spin coater under N2 flow.

Culture of rat cortical neurons

Primary cortical neurons were prepared from Sprague-Dawley rat embryos at gestational day 18 using a previously described procedure [48], [61] with the following modifications: after dissection, the cortex tissue was dissociated by incubation for 15 min at 37 °C in phosphate buffer saline (PBS) containing 10 U/ml Papain (Worthington), followed by gentle dissociation in PBS containing 1.5 mg/ml BSA and 1.5 mg/ml Trypsin inhibitor (from chicken egg, Sigma). After dissociation, cell homogenates were subjected to a centrifugation step at 3000 g for 10 minutes through a 4 % BSA cushion (diluted in Neurobasal medium), to minimize presence of glial cells in the final pellet. Cultures were maintained in serum-free Neurobasal medium (Invitrogen) supplemented with 0.5 mM glutamine, 1 % penicillin/streptomycin and 2 % B-27 supplement (Invitrogen). Cells were plated on the chips at a concentration between 200 cells / mm² and 500 cells / mm² and cultivated in a humidified incubator at 37 °C and 5 % CO2.

Viability assay

This viability assay (Live/Dead, Thermofisher Scientific) allows discriminating live from dead cells. Indeed, the polyanionic dye Calcein-AM is retained within live cells and produces a green fluorescence, whereas Ethidium homodimer-1 (EthD-1) only enters cells with damaged membranes and produces a red fluorescence in dead ones. Rat cortical neurons were plated on chemically functionalized samples in serum-free Neurobasal medium. After 3 days in vitro (DIV), this medium was replaced by medium conditioned by glial cells. At 15 DIV, each sample was rinsed with 1 mL of


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Dulbecco’s phosphate-buffered saline (D-PBS) before been covered by the LIVE/DEAD reagent (Calcein AM 1 µM: EthD-1 1 µM in serum-free Neurobasal medium minus phenol red). Then, samples were incubated for 20 min at 37 °C and observed with a fluorescence microscope.

Preparation of conditioned medium by glial cells

After dissection, each cortex hemisphere was cut into small pieces, placed into a 50 mL Falcon tube and digested with papain (4.3 %) at 37 °C for 8 min. Then, papain was inactivated with Lo-Ovo (Ovomucoid solution) and tissues were dissociated by vigorous pipetting using a plastic pipette (10 to 20 times). The resulting mixture was filtered with a 70 µm cell strainer and subjected to a centrifugation step at 1100 g for 10 minutes. The cell pellet was resuspended in DMEM medium supplemented with 10 % of fetal calf serum (FCS) and 1 % penicillin/streptomycin and plated in a cell culture flask. After ten days, this medium was replaced by serum-free Neurobasal medium (Invitrogen) supplemented with 0.5 mM glutamine, 1 % penicillin/streptomycin and 2 % B-27 supplement (Invitrogen). Three days later, this conditioned medium was harvested to be used for neuronal culture.

Immunofluorescence and confocal microscopy analysis

Samples were rinsed with PBS, fixed with 4 % paraformaldehyde for 30 min and rinsed again twice with PBS. The cells were incubated for 4 min in permeabilization buffer: 0.1 % Triton X100 in PBS and rinsed with PBS/0.02 % Tween 20. Then, they were incubated for 15 min in blocking solution: 2 % BSA in PBS/0.02 % Tween 20. The primary antibodies MAP2 and TAU were applied for 1 hour at room temperature in a humidified chamber (Anti-TAU (1/500), Anti-MAP2 (1/250), Millipore), and rinsed three times for 3 min with PBS/0.02 %Tween 20 while shaking. The secondary antibodies conjugated with Alexafluor-488 or Alexafluor-594 (Millipore) were applied for 1 hour at room temperature in a humidified chamber and rinsed four times for 3 min with PBS/0.02 % Tween 20 while shaking. The cells were incubated for 4 min in counterstaining solution: DAPI 0.02 mg/mL in PBS Tween, and briefly rinsed in PBS. The samples were placed on mounting medium (Molecular Probes) and we used a coverslip to seal the samples with nail polish.

Preparation for SEM imaging


Page 14 of 18

Page 15 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

For the SEM imaging, cultures were rinsed with PBS and fixed with 5 % glutaraldehyde in water during 4 h. Then glutaraldehyde was rinsed with PBS and samples were dehydrated by successive incubations in graded ethanol washes (25 %, 50 %, 80 % and 100 %) each time during 8 min. Samples were dried and images made with a FEI Helios 600i SEM.

Acknowledgments This work was supported by the Occitanie Region through the RBIO project (Nano MitoX), by the Idex UNITI (Transversally program), by the CNRS (PEPS INSIS) and by the French RENATECH network (French national nanofabrication platform).

Supporting Information Available MTT assays Impact of coating duration on the uniformity of deposition of the adhesion molecule

References [1]

S. Salinas, G. Schiavo, and E. J. Kremer, “A hitchhiker’s guide to the nervous system: the complex journey of viruses and toxins,” Nature Reviews Microbiology, pp. 645–655, 2010. [2] G. Shahaf and S. Marom, “Learning in networks of cortical neurons,” J. Neurosci., vol. 21, no. 22, pp. 8782–8788, 2001. [3] F. Patolsky et al., “Detection, Stimulation, and Inhibition of Neuronal Signals with High-Density Nanowire Transistor Arrays,” Sci. Transl. Med., vol. 313, no. 5790, pp. 1100–1104, Aug. 2006. [4] J. T. Robinson, M. Jorgolli, A. K. Shalek, M.-H. Yoon, R. S. Gertner, and H. Park, “Vertical nanowire electrode arrays as a scalable platform for intracellular interfacing to neuronal circuits,” Nat. Nanotechnol., vol. 7, no. 3, pp. 180–184, Jan. 2012. [5] Q. Qing, Z. Jiang, L. Xu, R. Gao, L. Mai, and C. M. Lieber, “Free-standing kinked nanowire transistor probes for targeted intracellular recording in three dimensions,” Nat. Nanotechnol., vol. 9, no. 2, pp. 142–147, Dec. 2013. [6] F. Santoro et al., “Interfacing Electrogenic Cells with 3D Nanoelectrodes: Position, Shape, and Size Matter,” ACS Nano, vol. 8, no. 7, pp. 6713–6723, Jul. 2014. [7] C. Xie, Z. Lin, L. Hanson, Y. Cui, and B. Cui, “Intracellular recording of action potentials by nanopillar electroporation,” Nat. Nanotechnol., vol. 7, no. 3, pp. 185–190, Feb. 2012. [8] R. Gao et al., “Outside Looking In: Nanotube Transistor Intracellular Sensors,” Nano Lett., vol. 12, no. 6, pp. 3329–3333, Jun. 2012. [9] H. J. Lee, Y. Son, J. Kim, C. J. Lee, E.-S. Yoon, and I.-J. Cho, “A multichannel neural probe with embedded microfluidic channels for simultaneous in vivo neural recording and drug delivery,” Lab Chip, vol. 15, no. 6, pp. 1590–1597, 2015. [10] D. Kleinfeld, K. H. Kahler, and P. E. Hockberger, “Controlled outgrowth of dissociated neurons on patterned substrates,” J. Neurosci., vol. 8, no. 11, pp. 4098–4120, 1988. [11] K. Goslin and G. Banker, “Culturing Nerve Cells,” in Neurons: Methods and Applications for the Cell Biologist, Academic Press., 1991, pp. 251–281.


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

[12] G. J. Brewer and C. W. Cotman, “Survival at low and growth of hippocampal neurons in defined medium density: advantages of a sandwich culture technique or low oxygen,” Brain Res., vol. 494, pp. 65–74, 1989. [13] G. J. Brewer, J. R. Torricelli, E. K. Evege, and P. J. Price, “Optimized survival of hippocampal neurons in B27-supplemented Neurobasal, a new serum-free medium combination.,” J. Neurosci. Res., vol. 35, no. 5, pp. 567–576, Aug. 1993. [14] F. Larramendy, M. C. Blatche, L. Mazenq, A. Laborde, P. Temple-Boyer, and O. Paul, “Microchannel-connected SU-8 honeycombs by single-step projection photolithography for positioning cells on silicon oxide nanopillar arrays,” J. Micromechanics Microengineering, vol. 25, no. 4, p. 045010, Apr. 2015. [15] W. Li et al., “NeuroArray: A Universal Interface for Patterning and Interrogating Neural Circuitry with Single Cell Resolution,” Sci. Rep., vol. 4, Apr. 2014. [16] C. Xie, L. Hanson, W. Xie, Z. Lin, B. Cui, and Y. Cui, “Noninvasive Neuron Pinning with Nanopillar Arrays,” Nano Lett., vol. 10, no. 10, pp. 4020–4024, Oct. 2010. [17] F. Santoro, J. Schnitker, G. Panaitov, and A. Offenhäusser, “On Chip Guidance and Recording of Cardiomyocytes with 3D Mushroom-Shaped Electrodes,” Nano Lett., vol. 13, no. 11, pp. 5379– 5384, Nov. 2013. [18] G. Panaitov, S. Thiery, B. Hofmann, and A. Offenhäusser, “Fabrication of gold micro-spine structures for improvement of cell/device adhesion,” Microelectron. Eng., vol. 88, no. 8, pp. 1840–1844, Aug. 2011. [19] C. S. Bausch et al., “Guided neuronal growth on arrays of biofunctionalized GaAs/InGaAs semiconductor microtubes,” Appl. Phys. Lett., vol. 103, no. 17, p. 173705, 2013. [20] P. Froeter et al., “Toward Intelligent Synthetic Neural Circuits: Directing and Accelerating Neuron Cell Growth by Self-Rolled-Up Silicon Nitride Microtube Array,” ACS Nano, vol. 8, no. 11, pp. 11108–11117, Nov. 2014. [21] F. Cavallo, Y. Huang, E. W. Dent, J. C. Williams, and M. G. Lagally, “Neurite Guidance and ThreeDimensional Confinement via Compliant Semiconductor Scaffolds,” ACS Nano, vol. 8, no. 12, pp. 12219–12227, Dec. 2014. [22] C. Wyart, C. Ybert, L. Bourdieu, C. Herr, C. Prinz, and D. Chatenay, “Constrained synaptic connectivity in functional mammalian neuronal networks grown on patterned surfaces,” J. Neurosci. Methods, vol. 117, no. 2, pp. 123–131, 2002. [23] M. Kwiat et al., “Highly Ordered Large-Scale Neuronal Networks of Individual Cells – Toward Single Cell to 3D Nanowire Intracellular Interfaces,” ACS Appl. Mater. Interfaces, vol. 4, no. 7, pp. 3542–3549, Jul. 2012. [24] B. Liu, J. Ma, E. Gao, Y. He, F. Cui, and Q. Xu, “Development of an artificial neuronal network with post-mitotic rat fetal hippocampal cells by polyethylenimine,” Biosens. Bioelectron., vol. 23, no. 8, pp. 1221–1228, Mar. 2008. [25] S. B. Jun et al., “Low-density neuronal networks cultured using patterned poly-l-lysine on microelectrode arrays,” J. Neurosci. Methods, vol. 160, no. 2, pp. 317–326, Mar. 2007. [26] W. C. Chang and D. W. Sretavan, “Novel High-Resolution Micropatterning for Neuron Culture Using Polylysine Adsorption on a Cell Repellant, Plasma-Polymerized Background,” Langmuir, vol. 24, no. 22, pp. 13048–13057, Nov. 2008. [27] M. A. Hughes, P. M. Brennan, A. S. Bunting, K. Cameron, A. F. Murray, and M. J. Shipston, “Patterning human neuronal networks on photolithographically engineered silicon dioxide substrates functionalized with glial analogues: Patterning human neuronal networks,” J. Biomed. Mater. Res. A, vol. 102, no. 5, pp. 1350–1360, May 2014. [28] C. Tomba, C. Braïni, B. Wu, N. S. Gov, and C. Villard, “Tuning the adhesive geometry of neurons: length and polarity control,” Soft Matter, vol. 10, no. 14, p. 2381, 2014. [29] M. Suzuki et al., “Neuronal cell patterning on a multi-electrode array for a network analysis platform,” Biomaterials, vol. 34, no. 21, pp. 5210–5217, Jul. 2013. [30] B. C. Wheeler, J. M. Corey, G. J. Brewer, and D. W. Branch, “Microcontact printing for precise control of nerve cell growth in culture,” J. Biomech. Eng., vol. 121, no. 1, pp. 73–78, 1999. 16 Environment ACS Paragon Plus

Page 16 of 18

Page 17 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

[31] P. M. S. John et al., “Preferential glial cell attachment to microcontact printed surfaces,” J. Neurosci. Methods, vol. 75, no. 2, pp. 171–177, 1997. [32] B. C. Wheeler and G. J. Brewer, “Designing Neural Networks in Culture,” Proc. IEEE, vol. 98, no. 3, pp. 398–406, Mar. 2010. [33] A. Offenhäusser et al., “Microcontact printing of proteins for neuronal cell guidance,” Soft Matter, vol. 3, no. 3, pp. 290–298, 2007. [34] Y. Nam, D. W. Branch, and B. C. Wheeler, “Epoxy-silane linking of biomolecules is simple and effective for patterning neuronal cultures,” Biosens. Bioelectron., vol. 22, no. 5, pp. 589–597, Dec. 2006. [35] E. Marconi et al., “Emergent Functional Properties of Neuronal Networks with Controlled Topology,” PLoS ONE, vol. 7, no. 4, p. e34648, Apr. 2012. [36] T.-M. Fu et al., “Sub-10-nm intracellular bioelectronic probes from nanowire-nanotube heterostructures,” Proc. Natl. Acad. Sci., vol. 111, no. 4, pp. 1259–1264, Jan. 2014. [37] M. Kwiat and F. Patolsky, “2 Interfacing Biomolecules, Cells and Tissues with Nanowire-based Electrical Devices,” in Applications of Electrochemistry and Nanotechnology in Biology and Medicine II, vol. 53, N. Eliaz, Ed. Boston, MA: Springer US, 2012, pp. 67–104. [38] R. Maboudian, W. R. Ashurst, and C. Carraro, “Tribological challenges in micromechanical systems,” Tribol. Lett., vol. 12, no. 2, pp. 95–100, 2002. [39] W. R. Ashurst, C. Carraro, R. Maboudian, and W. Frey, “Wafer level anti-stiction coatings for MEMS,” Sens. Actuators Phys., vol. 104, no. 3, pp. 213–221, May 2003. [40] W. R. Ashurst, C. Carraro, and R. Maboudian, “Vapor Phase Anti-Stiction Coatings for MEMS,” IEEE Trans. Device Mater. Reliab., vol. 3, no. 4, Décembre 2003. [41] D. Edgar, R. Timpl, and H. Thoenen, “The heparin-binding domain of laminin is responsible for its effects on neurite outgrowth and neuronal survival,” EMBO J., vol. 3, no. 7, pp. 1463–1468, 1984. [42] G. Bugnicourt, J. Brocard, A. Nicolas, and C. Villard, “Nanoscale Surface Topography Reshapes Neuronal Growth in Culture,” Langmuir, vol. 30, no. 15, pp. 4441–4449, Apr. 2014. [43] L. Luckenbill-Edds, “Laminin and the mechanism of neuronal outgrowth,” Brain Res. Rev., vol. 23, pp. 1–27, 1997. [44] H. S. Koh, T. Yong, C. K. Chan, and S. Ramakrishna, “Enhancement of neurite outgrowth using nano-structured scaffolds coupled with laminin,” Biomaterials, vol. 29, no. 26, pp. 3574–3582, Sep. 2008. [45] L. Kam, W. Shain, J. N. Turner, and R. Bizios, “Axonal outgrowth of hippocampal neurons on micro-scale networks of polylysine-conjugated laminin,” Biomaterials, vol. 22, pp. 1049–1054, 2000. [46] M. A. Dichter, “Rat cortical neurons in cell culture: culture methods, cell morphology, electrophysiology, and synapse formation,” Brain Res., vol. 149, pp. 279–293, 1978. [47] J. J. Bajramovic, S. Munter, S. Syan, U. Nehrbass, M. Brahic, and D. Gonzalez-Dunia, “Borna Disease Virus Glycoprotein Is Required for Viral Dissemination in Neurons,” J. Virol., vol. 77, no. 22, pp. 12222–12231, Nov. 2003. [48] E. Suberbielle et al., “Proteomic Analysis Reveals Selective Impediment of Neuronal Remodeling upon Borna Disease Virus Infection,” J. Virol., vol. 82, no. 24, pp. 12265–12279, Dec. 2008. [49] G. Chevalier et al., “Neurons are MHC Class I-Dependent Targets for CD8 T Cells upon Neurotropic Viral Infection,” PLoS Pathog., vol. 7, no. 11, p. e1002393, Nov. 2011. [50] E. M. Bonnaud et al., “Borna Disease Virus Phosphoprotein Modulates Epigenetic Signaling in Neurons To Control Viral Replication,” J. Virol., vol. 89, no. 11, pp. 5996–6008, Jun. 2015. [51] S. Shimizu, A. Abt, and O. Meucci, “Bilaminar Co-culture of Primary Rat Cortical Neurons and Glia,” J. Vis. Exp., no. 57, Nov. 2011. [52] A. S. Asch, L. L. K. Leung, J. Shapiro, and R. L. Nachman, “Human brain glial cells synthesize thrombospondin,” Proc. Natl. Acad. Sci., vol. 83, no. 9, pp. 2904–2908, 1986. [53] K. S. Christopherson et al., “Thrombospondins Are Astrocyte-Secreted Proteins that Promote CNS Synaptogenesis,” Cell, vol. 120, no. 3, pp. 421–433, Feb. 2005. 17 Environment ACS Paragon Plus

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

[54] B. Conde Guerri, E. Sinues Porta, M. Arrazola Schlamilch, F. Comunas Gonzalez, and V. C. Maldonado, “Effects of glia-conditioned medium on primary cultures of central neurons,” 1989. [55] J. Engele, D. Schubert, and M. Bohn, “Conditioned media derived from glial cell lines promote survival and differentiation of dopaminergic neurons in vitro: role of mesencephalic glia.,” J. Neurosci. Res., vol. 30, no. 2, pp. 359–371, 1991. [56] D. Gerlier and N. Thomasset, “Use of MTT colorimetric assay to measure cell activation,” J. Immunol. Methods, vol. 94, no. 1–2, pp. 57–63, 1986. [57] M. Ferrari, M. C. Fornasiero, and A. M. Isetta, “MTT colorimetric assay for testing macrophage cytotoxic activity in vitro,” J. Immunol. Methods, vol. 131, no. 2, pp. 165–172, 1990. [58] R. Liu et al., “High Density Individually Addressable Nanowire Arrays Record Intracellular Activity from Primary Rodent and Human Stem Cell Derived Neurons,” Nano Lett., vol. 17, no. 5, pp. 2757–2764, May 2017. [59] Q. Qing, Z. Jiang, L. Xu, R. Gao, L. Mai, and C. M. Lieber, “Free-standing kinked nanowire transistor probes for targeted intracellular recording in three dimensions,” Nat. Nanotechnol., vol. 9, no. 2, pp. 142–147, Dec. 2013. [60] J. Abbott et al., “CMOS nanoelectrode array for all-electrical intracellular electrophysiological imaging,” Nat. Nanotechnol., vol. 12, no. 5, pp. 460–466, Feb. 2017. [61] C. M. A. Prat et al., “Mutation of the Protein Kinase C Site in Borna Disease Virus Phosphoprotein Abrogates Viral Interference with Neuronal Signaling and Restores Normal Synaptic Activity,” PLoS Pathog., vol. 5, no. 5, p. e1000425, May 2009.

Table of contents


Page 18 of 18

A self-aligned functionalization approach to order neuronal networks

Recommend Documents