APPARATUS AND METHODS FOR FLUIDIC MICROACTUATION OF

ELECTRODES

CROSS-REFERENCE TO RELATED PATENT APPLICATIONS This application claims priority to U.S. Provisional Patent Application Serial No.

62/583,787 filed November 9, 2017, the entire contents of which are incorporated herein by reference.

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

The invention was made with government support under Grant No. R21EY028397 awarded by the National Institutes of Health, Grant No. FA9550-15-1-0370 awarded by the United States Air Force / Air Force Office of Scientific Research and Grant No. CBET-1351692 awarded by the National Science Foundation. The government has certain rights in the invention.

BACKGROUND This invention was also supported by Grant No. C-1668 awarded by The Welch

Foundation.

A. FIELD

This disclosure relates to apparatus and methods for implanting and positioning flexible electrodes in tissue. More particularly, embodiments of the invention relate to apparatus and methods for implanting and positioning flexible electrodes in tissue using fluidic control.

B. RELATED ART

Chronically implanted electrodes that can measure individual neuron activity with millisecond resolution are key tools for understanding the neural processes underlying cognition, motion, sensation and neurological diseases. Traditionally, these recordings are performed with micromachined silicon or metal microelectrodes implanted below the surface of the brain. Over the past few years these intracortical microelectrodes (IMEs) have led to breakthrough discoveries and technological innovations including control of robotic end effectors [1,2] restoration of the cortical control of the upper limbs [3] discovery of fundamental mechanisms underlying cognitive processes [4,5] and identification of unique neural firing patterns associated with epileptic activity [6-8].

Despite the rapid progress and the tremendous potential for IMEs there is a clear need for new flexible electrodes, and methods to insert and position these electrodes in neural tissue. The approximately l08-fold stiffness mismatch between traditional electrode materials and the soft host brain tissue causes acute and chronic injury that results in extensive neuronal death, formation of gliotic encapsulation and isolation of the recording sites from the neuronal bodies [9,10]. Moreover, relative micromotion between the tissue and the implant can exacerbate the strain-induced inflammatory reaction and cause the recording site to drift from its original position [11,12]. Engineered flexible substrates and ultrasmall microwires approximating the diameter of individual cells have been shown to significantly mitigate the chronic injury and improve signal quality and longevity of neural recordings [13-17]. Recently, the inventors demonstrated that flexible carbon nanotube fiber (CNTf) microelectrodes are effective for neuromodulation and chronic recording applications, while eliciting minimal foreign body reaction when compared to metal implants [18]. One drawback of these novel compliant materials is the need of temporary stiffening strategies to overcome the buckling force upon penetration in the cortex [19]. For example, flexible electrodes can be stiffened through attachment to a rigid probe during insertion [16,20,18] or by overcoating with a hydrogel sheath [21-23] that dissolves minutes after implantation. However, the increased rigidity and footprint of the stiffener-electrode assembly can aggravate the acute and chronic injury, damaging or destroying nearby neurons and breaching the blood brain barrier [18,24-25]. Recent work [17,26] demonstrated that macroporous 3-D mesh electrodes with cellular scale feature sizes (approximately 20 pm) and mechanical flexibility 107 times greater than conventional silicon greatly improve the integration of neural electrodes with brain tissue and are capable of intraoperative and chronic recordings in vivo. However, to implant these devices using traditional techniques a rigid syringe must be inserted into the cortex and the recording sites cannot be repositioned after implantation.

Neural electrodes implanted in the brain are a valuable tool for applications that require recording and stimulating the activity of individual neurons at high spatial and temporal resolution, such as brain-machine interfaces and neuroprosthetic devices. The size and rigidity of these implants, however, triggers a foreign body reaction in the host tissue that limits the longevity of the recordings for chronic applications. While flexible electrodes reduce the foreign body response, there are currently no known technologies to insert and reposition these electrodes without stiffeners that produce acute damage.

Historically, rigid planar silicon microelectrodes have reliably measured multiple single unit activities in acute preparations in which the neurons are recorded on the day of electrode insertion. In fact, several successful companies have commercialized stiff electrode arrays (e.g. NeuroNexus, Blackrock Micro). However, chronic recordings using these rigid electrodes have proven much less reliable. Evidence suggests that a combination of acute and chronic damage limits the number individual neurons that can be recorded and the duration that these recordings are stable [27,28]. To overcome these challenges, researchers have turned to thin, flexible neural electrodes. A major advantage of these flexible neural electrodes is improved signal quality and longevity produced by reducing the chronic inflammation induced when stiff electrodes fail to bend and flex with the micromovement of the brain [13-16].

SUMMARY

Briefly, the present disclosure provides apparatus, systems and methods utilizing microfluidic-actuation technology that allows insertion and positioning of flexible electrodes, without the need of stiffening support. Embodiments of the present disclosure can significantly reduce the trauma to the soft brain tissue during implantation and improve biocompatibility, longevity and therapeutic efficacy of neural electrodes.

Exemplary embodiments include an apparatus for inserting a microelectrode in tissue, the apparatus comprising: a substrate comprising; a channel proximal to the substrate, wherein the channel comprises a first end and a second end; a fluid inlet in fluid communication with the channel; a first outlet in fluid communication with the channel; a second outlet in fluid communication with the channel; and a microelectrode disposed within the channel.

Certain embodiments further comprise a valve configured to control fluid flow in the channel. In particular embodiments the microelectrode is a flexible microelectrode. In some embodiments the channel is formed in the substrate, and in specific embodiments the channel is a separate component from the substrate. In certain embodiments the channel is a nano-fabricated channel.

In particular embodiments the valve is configured to allow the fluid flow to the second end of the channel when the valve is open and wherein the valve is configured to restrict the fluid flow to the second end of the channel when the valve is closed. In some embodiments the microelectrode has a thickness between 1 and 50 pm, or more particularly between 1 and 25 pm. In specific embodiments the microelectrode has a diameter between 5 and 50 pm, or more particularly between 10 and 25 pm. In certain embodiments the microelectrode is a flexible carbon nanotube fiber, and in particular embodiments the microelectrode comprises a rectangular cross-section. In particular embodiments the microelectrode is a nanofabricated flexible polymer electrode.

In specific embodiments the substrate comprises a plurality of channels, and each channel in the plurality of channels comprises an electrode. In some embodiments the channel has a first cross-sectional area at the first end; the channel has a second cross-sectional area at the second end; the microelectrode has a third cross-sectional area; the first cross-sectional area is between 3 to 100 times larger than the third cross-sectional area; and the second cross-sectional area is 1.5 to 10 times larger than the third cross-sectional area. In certain embodiments the fluid inlet is proximal to the first end of the channel, and the second outlet is proximal to the second end of the channel.

Particular embodiments further comprise a third outlet, where: the first outlet and the third outlet are each fluid vent outlets; the first outlet is in fluid communication with the channel via a first vent channel; and the third outlet is in fluid communication with the channel via a second vent channel. In some embodiments the channel comprises a first portion with a first diameter proximal to the first end; the channel has a second portion with a second diameter proximal to the second end; the apparatus comprises a first vent channel in fluid communication with the second outlet and the first portion of the channel; and the apparatus comprises a second vent channel in fluid communication with the third outlet and the first portion of the channel.

Specific embodiments further comprise a fluid under pressure directed to the fluid inlet, where: at least 90 percent of the fluid by volume flows from the fluid inlet to the first outlet and the third outlet; and less than 10 percent of the fluid by volume flows from the fluid inlet to the second outlet. In certain embodiments at least 97 percent of the fluid by volume flows from the fluid inlet to the first outlet and the third outlet; and less than 3 percent of the fluid by volume flows from the fluid inlet to the second outlet. In particular embodiments the fluid flows at a rate of 1-1,000 pL through the channel. In some embodiments the microelectrode is directed out of the second outlet when a fluid flow is directed from the first end of the channel to the second end of the channel. In specific embodiments the microelectrode is directed out of the second outlet by a drag force between the fluid flow and the flexible microelectrode. In certain embodiments the substrate comprises two layers of polydimethylsiloxane (PDMS). In particular embodiments the substrate comprises glass, silicon, polymethylmethacrylate or polycarbonate. In some embodiments the substrate comprises polydimethylsiloxane bonded to glass or silicon.

Specific embodiments further comprise a plurality of channels and a plurality of microelectrodes, where: the channel is a first channel in the plurality of channels; the microelectrode is a first microelectrode in the plurality of microelectrodes; and each microelectrode in the plurality of microelectrodes is disposed within a channel of the plurality of channels.

Certain embodiments further comprise a fluid control system configured to direct fluid flow to the fluid inlet. In particular embodiments the fluid control system is configured to open and close the valve to control fluid flow. In some embodiments the fluid control system is configured to open the valve to allow fluid flow through the channel and direct the microelectrode out of the second outlet of the channel; and the fluid control system is configured to close the valve to restrict fluid flow through the channel and stop movement of the flexible microelectrode relative to the channel. In specific embodiments the fluid control system is configured to direct the microelectrode at least 1 mm past the second outlet of the channel. In certain embodiments the channel is formed via a photolithography process.

Particular embodiments include a method of inserting a microelectrode in a target material, the method comprising: placing an apparatus comprising a microelectrode and a channel proximal to the target, wherein the channel has inlet and an outlet and wherein the outlet is adjacent the target material; directing a fluid from the inlet of the channel to the outlet of the channel; and directing the microelectrode from the outlet of the channel into the target material via a drag force between the fluid and the microelectrode. Some embodiments further comprise venting fluid from the channel. In specific embodiments at least 90 percent of the fluid by volume is vented from the channel, or more particularly at least 97 percent of the fluid by volume is vented from the channel.

Certain embodiments further comprise controlling a flow of the fluid via a valve. Particular embodiments further comprise controlling the valve with a fluid control system. In some embodiments the target material is brain tissue, and in specific embodiments the target material is peripheral nerve tissue. In certain embodiments the channel is formed via a photolithography process. In particular embodiments the apparatus comprises a plurality of channels and a plurality of microelectrodes and each channel contains a microelectrode; each channel has inlet and an outlet, wherein the outlet of each channel is adjacent the target material; and the method further comprises: directing the fluid from the inlet of each channel to the outlet of each channel; and directing the microelectrode contained in each channel from the outlet of each channel into the target material via a drag force between the fluid and the microelectrode contained in each channel.

Any embodiment of any of the present methods, composition, kit, and systems may consist of or consist essentially of - rather than comprise/include/contain/have - the described steps and/or features. Thus, in any of the claims, the term“consisting of’ or“consisting essentially of’ may be substituted for any of the open-ended linking verbs recited above, in order to change the scope of a given claim from what it would otherwise be using the open-ended linking verb.

The use of the term“or” in the claims is used to mean“and/or” unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive, although the disclosure supports a definition that refers to only alternatives and“and/or.”

Throughout this application, the term“about” is used to indicate that a value includes the standard deviation of error for the device or method being employed to determine the value. Following long-standing patent law, the words“a” and“an,” when used in conjunction with the word “comprising” in the claims or specification, denotes one or more, unless specifically noted.

Other objects, features and advantages of the present invention will become apparent from the following detailed description. It should be understood, however, that the detailed description and the specific examples, while indicating certain embodiments of the invention, are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

The following drawings form part of the present specification and are included to further demonstrate certain aspects of the present invention. The invention may be better understood by reference to one or more of these drawings in combination with the detailed description of specific embodiments presented herein. The patent or application file may contain at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

FIG. 1 is a partial perspective view of an apparatus for insertion and positioning of flexible electrodes according to an exemplary embodiment of the present disclosure.

FIG. 2 is an assembled perspective view of the embodiment of FIG. 1.

FIG. 3 illustrates a microscopic view of the channel and flexible electrode of the embodiment of FIG. 1.

FIG. 4 illustrates graphs and photographs of microfluidic actuation of electrodes according to an exemplary embodiment of the present disclosure.

FIG. 5 illustrates the embodiment of FIG. 1 during use inserting an electrode into a first test subject.

FIG. 6 illustrates the embodiment of FIG. 1 during use inserting an electrode into a second test subject.

FIG. 7 illustrates electrodes according to an exemplary embodiment showing stable single unit recordings 45 days after implantation.

FIG. 8 illustrates assembly techniques and simulated fluid flows according to an exemplary embodiment of the present disclosure. FIG. 9 illustrates assembly techniques according to an exemplary embodiment of the present disclosure

FIG. 10 illustrates a device configured for use in assembly of an apparatus for insertion and positioning of flexible electrodes according to an exemplary embodiment of the present disclosure.

FIG. 11 illustrates a schematic of parallel channels sourced by independent fluidic input ports used to selectively position multichannel electrodes by modulating fluidic input.

FIG. 12 illustrates a device holder with stereotactic arm configured for use with the embodiment of FIG. 1.

FIG. 13 illustrates the device holder of FIG. 12 during use to insert a flexible electrode.

FIG. 14 illustrates targeted neuromodulation using magnetic fields, including magnetoelectric thin films and magnetogenetics according to exemplary embodiments of the present disclosure.

DETAILED DESCRIPTION OF THE INVENTION

Exemplary embodiments of the present disclosure can implant and position neural electrodes that are both small and flexible without using external supports or stiffening agents. Instead, exemplary embodiments use viscous fluid flow to maintain tension in the electrode - effectively stiffening the electrode without increasing the cross-sectional area of the electrode. Microfluidic vent channels in the apparatus can divert a significant portion ( e.g . more than 98 percent) of this fluid away from the point of electrode insertion. Embodiments combine features of acute experiments - electrodes positioned in ideal locations using electro-physiological feedback - with the chronic stability endowed by the flexibility of the electrodes. In addition, literature suggests that reduced chronic inflammatory responses due to minimal insertion trauma will improve the number of neurons that can be recorded and their signal quality [13,29].

Referring initially to FIG. 1-3, an apparatus 100 comprises a substrate 110 comprising a channel 120 with a first end 121 and a second end 122. FIG. 1 illustrates a first perspective view of the apparatus during assembly, while FIG. 2 illustrates the assembled apparatus. FIG. 3 illustrates a microscopic view of channel 120 with a flexible electrode 130 disposed within channel 130. As shown in FIG. 2, apparatus 100 comprises a fluid inlet 140 and multiple fluid outlets in fluid communication with channel 120. It is understood the flow parameters shown in FIG. 2 ( e.g . the percentages of fluid flow in each outlet) are merely exemplary and other embodiments may comprise different percentages of fluid flow in the different channels. In certain embodiments, flexible electrode 130 may be a planar electrode with a thickness between 1 and 50 pm. In other embodiments, flexible electrode 130 may be a cylindrical electrode with a diameter between 1 and 50 pm. In particular embodiments, flexible electrode 130 may be a flexible carbon nanotube fiber (CNTf), while in other embodiments flexible microelectrode 130 may comprise a planar micro-fabricated or nano-fabricated electrode with a rectangular cross section.

In this embodiment, apparatus 100 comprises a first outlet 141, a second outlet 142 and a third outlet 143 in fluid communication with channel 120. It is understood that other embodiments may comprise a different number of outlets. As explained in more detail below, in this embodiment first and third outlets 141 and 143 are configured as vent outlets, while second outlet 142 is configured to direct flexible microelectrode 130 from channel 120 when a fluid flow is directed from first end 121 to second end 122 of channel 120. First outlet 141 is in fluid communication with channel 120 via a first vent channel 145, while third outlet 143 is in fluid communication with third outlet 143 via a second vent channel 146.

In the embodiment shown, apparatus 100 further comprises a first valve 150 and a second valve 152 configured to control fluid flow in channel 120. While the illustrated embodiment includes first and second valves 151 and 152, it is understood that other embodiments may not comprise valves to control fluid flow, or may comprise a different number of valves. As best shown in FIG. 1, channel 120 comprises portions with different diameters. In this embodiment, channel 120 comprises a first portion 123 proximal to first end 121 that has a larger diameter than a second portion 125 that is proximal to second end 122.

During operation of apparatus 100, flexible microelectrode 130 is initially disposed within channel 120 and subsequently directed from second outlet 122 of channel 120 via fluid low within channel 120. In a particular embodiment, flexible microelectrode 130 is initially disposed in first portion 123 and fluid is directed from fluid inlet 140 near first end 121 to second outlet 142 near second end 122. As fluid flows from fluid inlet 140 to second outlet 142, a drag force between the fluid flow and flexible microelectrode 130 directs flexible microelectrode 130 from second outlet 142. In certain embodiments, second outlet 142 can be placed proximal to a target material ( e.g . brain or other tissue) so that flexible microelectrode 130 is directed from second outlet 142 into the target material.

The use of controlled fluid flow to position flexible microelectrode 130 into the target material provides significant advantages over typical electrode insertion methods. For example, apparatus 100 can eliminate the need for external supports or stiffening agents used to assist in the insertion of flexible electrodes. By controlling the pressure of the fluid with first and second valves 151 and 152, a user can microactuate and precisely implant flexible microelectrode 130 with a precise resolution (e.g. less than 50 pm), in vitro and in vivo. Furthermore, using the flow control system, the flow and insertion parameters can be optimized, while minimizing the amount of fluid injected into the target tissue during insertion or actuation of flexible microelectrode 130.

As previously mentioned, in certain embodiments flexible microelectrode 130 may comprise a planar micro-fabricated or nano-fabricated electrode with a rectangular cross section. Such planar geometries can allow for multi-site or multi-electrode geometries, which enable scaling in a way that may not be possible with carbon nanotube electrodes. In addition, such multi-channel electrode system can include an integrated connector (rather than separate wiring to the electrode) that couples to the electrode during insertion.

Particular embodiments of the system may include a pump to circulate liquid through apparatus 100 (e.g. from outlets 141 and 143 to fluid inlet 140).

Further description and explanation of the operating principles can also be found in the discussion of the example and results that follow.

V. Examples

The following examples are included to demonstrate preferred embodiments of the disclosure. It should be appreciated by those of skill in the art that the techniques disclosed in the examples which follow represent techniques discovered by the inventor to function well in the practice of the disclosure, and thus can be considered to constitute preferred modes for its practice. However, those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the disclosure.

Fluidic Microactuation of Electrodes A major advantage of exemplary embodiments of fluidic microdrives compared to traditional syringe injection is the fact that the inventors can minimize the volume of fluid injected into the brain and thereby reduce any changes in intracranial pressure that could produce trauma. By creating a large hydraulic resistance through the exit channel, the inventors can divert the fluid flow into microfluidic vent channel that safely transport the fluid away from the injection site. To quantify the efficiency of this approach the inventors calculated the percentage of fluid that is successfully diverted into the vent ports.

Because the hydraulic resistance depends inversely as the third power of the channel cross sectional area [30,31] the inventors created three main sections of exemplary embodiments of the device where the inventors adjusted the channel width to control the fluidic resistances: a wide, low-resistance upstream channel connected to the flow input port (100-500 pm width, 50 pm height, 5-10 mm length) converging into a high resistance outlet channel for CNTf delivery (30 pm width, 45 pm height, 1 mm length) and two low-resistance side ports (200 pm width, 50 pm height, 400 pm length).

Computational analysis of the flow field in the device shows that the area of maximum flow velocity is concentrated in the converging nozzle and that the venting ports divert 98.5% of the input volume, while only 1.5% of the fluid flows downstream to the exit channel (as shown in FIG. 2). Note that the inventors computed the relative fluid flow rates in the absence of the electrode, thus the 1.5% of fluid exiting the device represents an upper bound. When the electrode enters the exit channel, the effective fluidic resistance of this channel will increase and the inventors expect an even smaller percentage of fluid to exit the device. The hydraulic design and the control of the on-chip valves enable actuation and positioning of the CNTf by simply tuning the flow parameters.

FIG. 4 illustrates microfluidic actuation of CNTf microelectrodes in vitro. Panel (A) of FIG. 4 illustrates the distance traveled by the CNTf in the channel at varying flow rates. In the graphs, lines indicate mean values, while shaded areas represent standard deviation. FIG. 4 Panel (B) illustrates stepwise electrode insertion (blue trace) is controlled by opening the flow control valve for 100 ms intervals (red trace). The average fiber displacement during the open valve period is 16.4 ± 6.4 pm. Positions labeled Cl to C4 refer to the images shown in Panel (C), which provides microscope images of the CNTf position in a brain phantom at times corresponding to Panel (B). Dashed lines, spaced approximately 15 pm, indicate the position of the fiber end.

The continuous forward and reverse motion and the velocity of the CNTf in the channel in has been shown in response to changes in the flow direction and input flow rate respectively. Moreover, by varying the flow rate and actuating the on-chip valves, the inventors could reliably control the position, velocity and depth of insertion in brain phantoms, as shown in FIG. 4 Panels (B) and (C). The precise actuation of flexible microelectrodes is particularly desirable for in vivo implants, where the position of the microelectrode can be dynamically adjusted to reach the target brain structure and optimize the proximity to neural bodies to improve recording quality and compensate for electrode drift.

To test exemplary embodiments of the fluidic actuated electrodes in vivo the inventors used Hydra as a model organism because they are easily cultured in the laboratory and are known to generate compound action potentials that correspond to body contractions (aka. Contraction bursts (CB)) [32]. To interrogate Hydra, the inventors modified exemplary embodiments of the microfluidic system to control the position of the animal body. In exemplary embodiments of the experimental setup depicted in FIG. 5 Panel (A), the working Ag/AgCl electrode is connected to the microfluidic device and makes electrical contact to the CNTf through the conductive Dextran solution filling the channels. The reference Ag/AgCl electrode is placed in the Hydra media bath.

Because the impedance of the CNTf-fluid interface is measured to be 180 times lower than the impedance of the Dextran solution (and the impedance of the CNTf is negligible), the inventors expect that the signals recorded using the Ag/AgCl working electrode suffer less than 1% attenuation due to fluidic shorts to ground (SI Figure S6). Throughout the experiment, the inventors held the Hydra approximately 3 mm downstream from the microdrive exit channel by applying moderate negative pressure to trap channels located in a separate PDMS block a shown in FIG. 5 Panel (B). The pressure applied by these channels was sufficient to prevent translation, while allowing the Hydra to contract, elongate and nod.

By controlling the fluid flow and the on chip valve actuation system, the inventors were able to bring the CNTf in contact with Hydra (n=3) and record compound action potentials. When the CNTf electrode was retracted into the fluidic device the inventors recorded no electrophysiological activity even during whole body contractions as shown in FIG. 5 Panel (C); however, when the inventors used the microfluidic device to position the CNTf approximately 50 pm from the Hydra body the inventors detected small-amplitude spikes (-120 pV) that corresponded with body contractions as shown in FIG. 5 Panel (D) and (E). No activity was observed during body elongation and nodding.

Next, the inventors inserted the CNTf into the Hydra and recorded spikes with much larger amplitudes (4.5 - 6 mV, as shown in FIG. 5 Panel (F) and (G)) correlated with body contractions. These spikes are consistent with previously reported CBs recorded via conventional electrophysiology using suction pipettes [33], confirming that exemplary embodiments of the fluidic actuated CNTfs act as microactuated flexible electrodes for neural recording. In addition to large amplitude spikes, the inventors also recorded low-amplitude spikes (50 - 150 pV, as shown in FIG. 5 Panel (G) that did not correlate with body contractions. These low-amplitude potentials are likely due to other behaviors such as elongation or tentacle contractions, which are known to correspond to small amplitude spikes recorded using suction pipettes [33].

To illustrate exemplary embodiments of the ability to reposition the CNTf to record from different areas the inventors used exemplary embodiments of the fluidic microdrive to retract and re-insert the CNTf. When the inventors retracted the CNTf the inventors once again recorded low-amplitude peaks correlated with contractions, as shown in FIG. 5 Panels (H) and (I). After re-insertion, the inventors recorded high- amplitude signals correlated with contractions together with the low amplitude activity as shown in FIG. 5 Panel (K) independent from contraction. The inventors noted, however, a change on the waveform of the contraction peaks. This variation in the waveform is likely due to a difference in the Hydra position, which agrees with previous reports that recorded CB waveforms depend on the electrode location within the animal body [33]

To show that exemplary embodiments of the fluidic actuation technology can also position flexible electrodes within the mammalian CNS, the inventors performed proof-of- principle experiments in thalamocortical brain slices of mice (13-21 days old). Although the mechanical properties of brain slices may be different from the intact brain, these ex vivo slice preparations provided the advantage of imaging the electrode position to determine the placement accuracy of the fluidic actuation technology. To minimize damage to the neural tissue, the inventors etched the ends of CNTfs to create a sharp tip (30 degrees) using a Focused Ion Beam (FIB) mill. Using the fluidic microdrives, the inventors successfully inserted CNTfs into cortex (up to approximately 1 mm deep, as shown in FIG. 6 Panels (A) and (D)). In addition, the inventors were also able to insert these flexible electrodes into deep brain structures such as specific nuclei of the thalamus, at a distance of approximately 4 mm from the cortical surface (FIG. 6 Panels (A) and (E)). Following CNTf insertion into the cortex (FIG. 6 Panel (B)), the inventors were able to detect spontaneous neuronal activity. The automated spike detection and clustering algorithm isolated spikes with amplitudes ranging from 50 pV to 500 pV.

A major advantage offered by exemplary embodiments of the disclosed technology is the ability to reposition the flexible electrode within the brain. To demonstrate this capability the inventors employed the CNTf to record spatially confined neuronal activity in the thalamic reticular nucleus (TRN), a shell-like structure that had a thickness of 168.3 ± 62.1 pm in the slice preparation (n = 4 slices). Neurons in the TRN are the target of cholinergic synaptic afferents from the basal forebrain and the brainstem. Previous studies have shown that stimulation of these afferents leads to the fast and reliable activation of both nicotinic and muscarinic acetylcholine receptors (nAChRs and mAChRs,) [34] and the generation of short-latency action potentials specifically in TRN neurons, but not in neighboring thalamic nuclei [34]. To selectively activate cholinergic afferents using optogenetics, the inventors used brain slices obtained from ChAT- ChR2-EYFP transgenic mice (FIG. 6 Panel (D)), which express channelrhodopsin-2 (ChR2) specifically in cholinergic neurons.

When the inventors positioned the CNTf at the TRN/internal capsule (IC) boundary (FIG. 6 Panel (E), top), the inventors did not detect neuronal activity in response to optical stimulation (FIG. 6 Panel (F), top). By opening the microfluidic flow control valves for 100 ms intervals, the inventors then gently pushed the CNTf into the TRN (FIG. 6 Panel (E, middle)). In this region, the inventors observed action potential activity approximately 30 ms following laser stimulation (FIG. 6 Panel (F, middle)). This latency is comparable to previous studies [35]. Next, the inventors pushed the CNTf deeper into the ventrobasal (VB) thalamus below the TRN (FIG. 6 Panel (E, bottom)). In this configuration, no responses were recorded by the electrode (FIG. 6 Panel (F, bottom)), indicating that electrical activity was only detected by the CNTf tip. To confirm that laser-evoked activity in the TRN was evoked by the activation of postsynaptic nAChRs, the inventors bath-applied the specific nAChR antagonist dihydro-P-erythroidine hydrobromide (ϋHbE), which completely eliminated TRN activity. Taken together, these experiments show that exemplary embodiments of the platform can not only insert bare flexible electrodes into the cortex, but can also drive these electrodes deep within the mammalian brain and accurately position them in specific brain regions.

Fabrication of CNTf electrodes

CNT fibers were fabricated from Meijo EC1.5-P CNTs (Meijo Nano Carbon Co., Ltd.) using a wet spinning process previously reported.14 Recordings from Hydra were obtained using CNTf coated with a dielectric double layer of A1203 and Hf02 by Atomic Layer Deposition (ALD). To obtain a conformal coating, CNTf with 23 pm in diameter were suspended between two glass slides separated by a few millimeters to prevent most of the fiber length from being in contact with a substrate. First, the fibers were exposed to an oxygen plasma cleaning process for a few minutes to clean residues and facilitate dielectric nucleation. Next, a standard ALD deposition (Cambridge Ultratech Savvanah S200) was performed. A1203 layer (50 nm) was grown using the precursors trimethylaluminum (TMA) and water and Hf02 layer (25 nm) used tetrakis[dimethylamido]hafnium (TDMAH) and water. The process chamber was at l50°C and 1 Torr pressure with N2 flowing at 20 seem as carrier gas. Fibers with current leak < 100 nA through the coating were selected for device fabrication.

Recordings from brain slices were obtained using Parylene coated fibers. Parylene C coating was done following standard protocols on a SCS Labcoter® 2 Parylene Deposition System. A donut shape acrylic holder was specially designed to load the fiber in the coating chamber allowing most of their length to be suspended. Scanning electron microscope (SEM) inspection indicates conformal and crack-free coating.

Leakage Current Tests

The insulation coating thickness was measured with an ellipsometer (Filmetrics). The coating morphology was observed using both optical and electron microscopy to check for any defects. Furthermore, integrity of insulation coatings was assessed by measuring DC leakage currents between insulated CNTfs and a large carbon wire counter electrode in saline at room temperature [36]. A Gamry Reference 600 potentiostat (Gamry Instruments) was used to apply DC voltage bias of 1 V and the leakage current between the insulated CNTf and the counter electrode was measured by passing the working electrode over them with a motorized arm. Coatings with current values lower than 100 nA with respect to open circuit current were considered good insulation. FIB Patterning CNTf

To facilitate insertion and reduce damage in brain slices, parylene coated CNTf with 23 pm in diameter were sharpened using standard FIB (FEI Helios NanoLab 660 DualBeam) protocols. A 30kV gallium beam (65nA) was focused on the surface of the fiber to cut it at a 30° angle with respect to its axis.

Dextran rheology

Dextran viscosity measurements were carried out at room temperature with Advanced Rheometric Expansion System (ARES, Rheometrics Scientific, now TA Instrument) in a Couette geometry. A solution of 40 % w/w Dextran in DI was loaded in a titanium cup (diameter = 16.5 303 mm), which rotates relative to a titanium bob in the center of the cup. Viscosity was measured as a function of shear rate (test duration: 20min, shear rate: 0.1 to 100 s-l)

Device Fabrication

Double layer PDMS microfluidic devices were fabricated following standard soft- lithography technique. Briefly, Si wafers (University Wafer) were spin coated with SU-8 2050 (Microchem), followed by a photolithography process which defined valve and flow channels on two separate wafers. Next, a thin layer (approximately 70 pm) of flexible (20:1 ElastomenCross- linker, w/w ratio) PDMS (RTV615, Fischer Scientific Company LLC) was spin-coated on the flow channel patterned wafer. At the same time, a thick (approximately 2 mm, 5:1 ElastomenCross-linker, w/w ratio) PDMS was poured on the valve layer patterned wafer. Both wafers were baked in an oven at 90°C: 15 min for the flow layer and 45 min for the valve layer. After alignment of the flow and valve control layers, the devices were baked for additional 24 hours at 90°C. Coated CNTf with diameters from 12 to 25 pm were manually loaded in the PDMS device using carbon-tip tweezers. To seal the microfluidic device with the fiber, an oxygen plasma cleaning step (Harrick plasma PDC-001) on the PDMS and glass was performed, allowing for a strong and leak free covalent bound between their exposed surfaces.

Brain Phantom Insertion Tests

Agar gel was used as brain phantom for insertion tests. The gel was prepared by first mixing bacteriological agar (Sigma-Aldrich Corporation) with DI water on a 0.6% w/w concentration. The solution was then heated in a microwave until boiling and finally left at room temperature to gelate at least 2 hours. For imaging purpose, a drop of red food colorant was added to the solution. Microfluidics devices were loaded with ALD coated CNTf with 12 mih in diameter as described above. The glass slide was vertically mounted on a micromanipulator and gently brought in contact with the agar gel. To insert the fiber in agar, manual pressure was applied to the syringe filled with a blue Dextran solution (40% w/w in DI with blue food colorant) connected to the fiber input port).

Manual CNTf insertion attempted by moving a similar fiber vertically towards the agar gel using a self-closing tweezer assembled on a micromanipulator. The CNTf was held ~4 mm away from the edge and the micromanipulator was manually operated to move the fiber at a speed of 75 pm/s.

Brain Phantom Actuation Test

Microfluidic devices loaded with CNTf 23 pm in diameter and sharpened by FIB were fabricated as described above and horizontally attached in a petri dish. Valve channels were filled with DI water and connected to a valve controller. A Matlab (Mathworks Inc.) script was used to switch between Closed and Open valve states.

A cubic block of agar gel (0.6 % w/w in DI prepared as described above) approximately 1 cm3 was gently placed in contact with the exit channel of the device. To drive the fiber, manual pressure was applied to the syringe filled with Dextran solution (40% w/w in DI) connected to the fiber input port. A video of the fiber movement in Agar was collected at approximately 8.9 fps using Hamamatsu Orca 03-G camera mounted on an inverted Nikon scope. A Matlab script was used to isolate the fiber from the background and determine the position of the distal end.

Hydra Culture

Hydra (H. littoralis obtained from Carolina Biological Supply Company) were cultivated in plastic vials at room temperature in standard Hydra culture medium (protocol adapted from Steele lab). The animals were fed once every two days with freshly hatched brine shrimps (Artemia naupli). Before electrophysiology experiments Hydras were food deprived for 24 hours.

Device Impedance

We measured a total input impedance of 261 ± 16 kQ (20mV, lkHz) for fluidic microdrives containing a high conductive Dextran solution (40% w/w in 20x PBS) and 25-pm- diameter Parylene-C coated CNTfs using a Gamry Reference 600 potentiostat (Gamry Instruments). The resistivity of the Dextran solution was estimated to be approximately 4.5 W-m based on the known length and cross sectional area of the fluidic channels and a measurement of the input impedance without a CNTf electrode. To estimate the impedance to ground through the conductive Dextran solution during CNTf recording (ZDex, SI Figure S6), the inventors considered that a 25 pm-diameter CNTf occupies approximately 90% of the cross-sectional area of the exit channel. Based on the reduced channel cross-section and the resistivity of the Dextran solution the inventors estimated Zi _CX to be approximately 45 MW.

Hydra Electrophysiology

Glass slide containing both CNTf microfluidic channels and Hydra trap was fixed to a plastic Petri dish filled with Hydra media. Hydra (n=3) with length between 1.5 and 2.0 mm were first partially immobilized by the Hydra trap. This was accomplished by positioning the animal using a glass pipette next to the trap channels, followed by application of negative pressure at the trap port. Next, the CNTf actuated towards the Hydra by applying manual pressure to a syringe connected to the fiber input port and filled with a conductive dextran solution (40% w/w in lx PBS). Valves were actuated (opened and closed) to guarantee precise positioning of the fiber inside or next to the animal.

The working Ag/AgCl electrode was placed in contact with the conductive Dextran solution in the microfluidic channel while the reference Ag/AgCl electrode was placed in the Hydra medium. Data were collected by using an AM System model 1800 amplifier and a Digidata 1550 digitizer (Molecular Devices) at 10 kHz sampling rate with 0.5 Hz low-cut filter and 20 kHz high374 cut filter.

Fluid dynamic analysis

Flow simulations were performed using Comsol (Comsol Inc.). Stationary Navier-Stokes equation was solved in laminar flow regime using a finite element method applied to a 3D model of the device layout. Boundary conditions were set to constant pressure of 10 atm at the inlet port and 1 atm at the outlet port and no- slip on the sidewalls.

Brain Slice Preparation

Young male and female mice (P13-21) were used from two lines: C57BL/6J wild-type and bacterial artificial chromosome (BAC)-transgenic mice expressing channelrhodopsin-2 (ChR2) under the control of the choline acetyltransferase (ChAT) promoter (ChAT-ChR2- EYFP). Thalamocortical slices (400 pm) were prepared as described previously [37]. Animals were anesthesized with isofluorane and decapitated, following procedures in accordance with NIH guidelines and approved by the UTHealth animal welfare committee. Slices (450 pm) were cut in an ice-cold solution consisting of (in mM): 234 sucrose, 2.5 KC1, 1.25 NaH2P04, 10 MgS04, 26 NaHC03, 10 glucose, and 0.5 CaCl2, saturated with 95% 02-5% C02, using a vibratome (Leica VT1200S) at speeds of 0.2 mm/s and a blade vibration amplitude of 0.8 mm. Slices were incubated at 34°C for 40 min in artificial cerebrospinal fluid (ACSF) containing (in mM): 126 NaCl, 26 NaHC03, 2.5 KC1, 1.25 NaH2P04, 10 glucose, 2 CaCl2, and 2 MgCl2. Slices were then kept at room temperature prior to recordings.

Brain Slice Electrophysiology

CNTf microfluidic devices were sealed at the edge of a 500 pm thick glass slide and mounted on a 60 mm diameter plastic petri dish. Tubes for ACSF perfusion were placed on opposing sides of the dish. Freshly dissected 450 pm brain slices were placed on a 150 pm thick glass slide pre-coated with poly-L-lysine (Sigma Aldrich). The brain slice was manually positioned as close as possible to the device with cortex facing the fiber exit channel. All recordings were carried out at near-physiological temperatures (32-34 °C) CNTf was inserted in the brain slice by applying manual pressure to a syringe connected to the fiber input port and filled with a highly conductive Dextran solution (40% w/w in 20x PBS). Valves were actuated (opened for 100ms and closed for 900ms) to guarantee precise positioning and actuation of the fiber inside the brain slice.

The working Ag/AgCl electrode was placed in contact with the conductive Dextran solution in the fiber channel while the reference Ag/AgCl electrode was placed in contact with the ACSF bath in the dish. Data were collected by using an AM System model 1800 amplifier and a Digidata 1550 digitizer (Molecular Devices) at a 50 kHz sampling rate with 0.5 Hz lowpass filter and 20 kHz highpass filter.

Thickness of TRN was measured on n=4 brain slices used for experiments. The mean thickness was calculated by averaging the values of 3 regions in a given microscopy image. Optical stimulation was performed using a DL447 laser (Crystalaser) at a wavelength of 440- 447 nm. Single laser pulses of 5 ms duration and 5mW power were used to activate cholinergic afferents in the TRN.

Brain slice data analysis Electrophysiological recordings were analyzed and processed off-line with a custom Matlab (Mathworks Inc.) script. To isolate spikes, the raw data stream was first digitally filtered between 300 Hz and 6 KHz. Spike detection was performed on the bandpass filtered data based on an amplitude threshold automatically set as 4.5 times the estimated standard deviation of the noise [38]. Principal components of the waveforms were calculated on a 3 ms window isolated from the threshold-crossing spikes followed by unsupervised clustering and sorting.

Multichannel Electrodes

A primary goal of this example is to demonstrate a generalizable strategy for minimally invasive insertion and actuation of scalable multichannel flexible electrodes. The major impact of this work is the concept that as electrode technology improves (increased recording densities, wireless communication, etc.), fluidic Microdrive technology can be reconfigured to support these new form factors to provide a minimally invasive delivery method. As a proof-of-principle, the inventors will develop and test fluidic microdrives compatible with state-of-the-art, flexible multichannel polymer electrodes fabricated at Lawrence Livermore National Laboratory (LLNL). By tailoring the fluidic microdrives for proven flexible electrodes already used by several labs, the inventors can take advantage of existing control data, integrate new microfabricated designs, and disseminate our technology to an existing community. In current practice, the arrays are implanted while glued to a 30-50 pm silicon stiffener [39] (Vanessa Tolosa, personal communication). Lollowing positioning, the glue is dissolved and the stiffener is removed. With this approach, stable neural ensembles have been recorded for months. LIG. 7 illustrates example micrograph electrodes comprising PDOT-coated platinum electrodes on a flexible polyimide substrate (110 pm wide and 15 pm thick) showing stable single unit recordings 45 days after implantation using a stiffener. The inventors expect fluidic microdrive insertion will: (1) improve electrode placement accuracy, and (2) reduce acute damage and increase the number of single units in chronic recordings. Note that, presumably due to penetration injury, the stable neurons appear only some days post-implantation. The inventors expect that the reduction in acute injury enabled by the fluidic microdrives may: (1) allow single unit activity to be recorded more immediately following implantation; and (2) result in larger stable ensembles.

The inventors will create fluidic microdrives compatible with the most common form factor of multichannel electrodes ( e.g . thin and flat), by designing and testing microfluidic channels specifically designed for these high-aspect ratio flexible electrodes. Key modifications the inventors will make to the CNTf fluidic microdrives described above are to: (1) create short and wide fluidic channels with proper fluidic resistances; and (2) promote fluidic actuation of electrodes which have a greater surface to volume ratio. With respect to first challenge the inventors will use the same PDMS replica molding strategy used for the preliminary data to create a microfluidic channel design optimized for electrodes approximately 110 pm wide and 15 pm thick. To allow for easy electrode loading and high flow rates in the drive channel, the inventors will design the main fluidic drive channel to be 500 pm wide and 40 pm tall. Should higher flow rates be needed, the inventors can increase the channel width to at least 1000 pm before needing PDMS support pillars to prevent channel collapse. The inventors’ previous experiments showed that similar channels widths did not collapse during CNTf fluidic actuation. To ensure that only a limited amount of the drive fluid (40% high-molecular-weight dextran in PBS) exits the device above the brain, the inventors will use analytical calculations of the fluidic resistances based on the channel geometries [30,31], followed by COMSOL finite element analysis simulations. Because the preliminary data showing that the inventors can implant CNTfs harmlessly into brain slices after the inventors attenuate the fluid flow rate to less than 2% in the exit channel, the inventors will design the multichannel electrode devices to reduce fluid flow rates to less than 2% out of the exit channel in the high— aspect ratio devices. Preliminary COMSOL simulation shows that an exit channel width of 120 pm and length of 2 mm attenuates the exit fluid rate to 0.5 %, which is expected to minimize any pressure differential applied to the brain (see FIG. 8)

To validate and characterize insertion and actuation the inventors will use an agar tissue phantom to approximate the mechanical properties of the brain (as described in our preliminary data). In this in vitro test environment, the inventors will measure the maximum penetration depth and the accuracy of microactuation.

Once the inventors have fabricated and validated fluidic microdrive for thin, flat electrodes the inventors will investigate the acute and chronic histological responses to fluidically implanted electrode "blanks" (i.e. flexible polyimide probes without expensive metal layers). Experiments will be performed in vivo, using stiffener assisted insertion as a control.

Development of on-demand microdrive disassembly Rapid, on-demand disassembly of the fluidic microdrive will allow the inventors to record during implantation and remove the microdrive without affecting the electrode placement. The primary strategy to create on-demand disassembly will be to use a water dissolvable sacrificial layer (e.g. dextran) that can be removed after electrode implantation. First, a glass substrate will be spin coated with a thin (0.5 pm) layer of dextran and cured at 80 C for 10 min. Subsequent atomic layer deposition (ALD) of a pinhole free A1203 thin film of 25 nm will act as a water barrier preventing the dextran from being dissolved during electrode insertion. The inventors will then complete the fluidic microdrive assembly on top of this sacrificial substrate by depositing 25 nm of Si02 for bonding with the PDMS flow channels. Following electrode implantation, the inventors can quickly delaminate the device by first scratching through the thin A1203 and Si02 films that protect the dextran, and then rinsing the device with buffer saline or ACSF. Preliminary experiments show that the inventors can successfully plasma bond a PDMS block to a silicon wafer coated with dextran and Si02. After scratching, the inventors can separate the PDMS from the substrate by washing in buffered saline for less than 2 minutes using a bulb pipette (see FIG. 9).

The thin A1203 and Si02 films fracture during the peeling process allowing the inventors to gently remove the probe from the microfluidic channels. The preliminary experiments with the CNTf show that modest force applied to the electrode following implantation does not significantly alter the electrode placement. In fact, the inventors observed in recordings from brain slices that the fiber could be pulled with enough force to move the brain slice without changing the implantation depth of the fiber. The inventors will assess the ability to accurately position electrodes into the hippocampus (and maintain this position following microdrive disassembly).

Should the inventors encounter difficulties with hermetic sealing of the sacrificial layer the inventors will investigate alternative conformal water barriers like parylene. Alternatively, the inventors will investigate a“clamshell” enclosure (fabricated using a 3D printer and/or machined aluminum) that will hold the PDMS layer against a glass substrate with mechanical force. To disassemble the microdrive, the inventors will remove a quick— release locking mechanism that will remove the pressure holding the microdrive together (see FIG. 10).

Fluidic microdrives for implantation and actuation of multiple multichannel electrodes Microfabricated flexible electrodes provide a path towards large-scale neural recordings via increased electrode density. However, reaching thousands of electrodes will require dense insertion of multiple devices. To this end, the inventors will create fluidic microdrives capable of implanting and actuating multiple independent multichannel electrodes. By creating parallel microdrive channels sourced by independent fluidic input ports the inventors will be able selectively position each multichannel electrode by modulating each fluidic input (see FIG. 11). To create this multi— electrode fluidic microdrive the inventors will add a second flow channel layer above the electrode flow channel layers that will serve as a fluidic input layer. This multi-layer microfluidic device will be fabricated by using a thermal curing process to permanently bond the two PDMS layers together. Briefly, the bottom flow layer will be fabricated by spin coating a flexible PDMS (20:1 elastomencuring agent ratio) solution on a SU- 8 mold, followed by a partial curing annealing process in at 90°C. The top layer will be made by casting a more rigid PDMS (5:1 ratio) solution on a separate Si wafer with an SU-8 mold. Then, the partially cured mold will be pealed from the wafer, aligned to the rigid PDMS layer and annealed overnight for a final curing step. This process will allow the two PDMS layers to crosslink, forming a single PDMS block. In the region where these layers connect, the inventors will fabricate small vias using an excimer laser. This multi-layer PDMS block can then be used with the same rapid disassembly procedure described above.

Prototype multi-electrode fluidic microdrives will feature 8 channels with a 125 pm spacing between electrode shanks. The inventors will validate implantation and independent actuation using electrode blanks and the in vitro brain phantoms described above.

Should the inventors experience difficulties independently actuating the multichannel electrodes, the inventors will: (1) investigate microfluidic valve configurations to independently clamp specific electrodes in place; or (2) develop a device that actuates all electrodes in parallel. Should the inventors experience difficulty loading the multiple shanks into the fluidic Microdrive, the inventors will: (1) investigate methods to use shuttles to guide the electrodes into the device where and then separate the shuttles from the probes prior to implantation; or (2) bond the PDMS fluidic layer directly to the silicon substrate used to manufacture the probes. Once in place the probes will be released during a wet etch step into the bonded fluidic microdrive.

The inventors anticipate that microfluidic insertion of ultraflexible electrodes will reduce the biological responses to the penetration injury that is incurred due to insertion of large and rigid stiffeners, leading a greater number of high— quality single unit recordings. To evaluate the improvements afforded by the microfluidic insertion technology, the inventors will carry out in vivo experiments as described herein.

Quantifying acute and long term tissue changes following microfluidic insertion

As has previously been extensively described [13,20] penetration injury damages vasculature, tears and compresses tissue, and kills neurons and glia. In order to evaluate the impact of microfluidic insertion on these acute phenomena and the chronic consequences, the inventors will implant test-blank devices in four cohorts of N=8 rats, and histologically evaluate biomarkers near the electrodes at 1 day, 1 week, 4 weeks, and 12 weeks post implantation. In each animal the inventors will use microfluidic insertion to implant a device into one hemisphere and a removable stiffener (i.e., the current strategy with LLNL probes) in the opposite hemisphere, counterbalancing the order of implantation. The flexibility of the electrodes requires that the dura be sliced or removed in order for devices to penetrate (even with stiffeners, Loren Frank, personal communication). In sections taken perpendicular to the path of the implant, the inventors will quantify the radial density of cell nuclei (DAPI), neurons/filament (NeuN, Neurofilament) and glia (microglia/activation: OX— 42 and Ibal, astrocytes: GFAP, macrophages: CD68, CCR7, and CD206), as well as blood brain barrier leakage (IgG and laminin) and neural precursors (nestin and doublecortin) [18,20]. The inventors anticipate that microfluidic insertion will radically reduce acute tissue changes and significantly improve long term electrode performance.

The inventors do not expect that the order of implantation technique should change results. If the inventors find evidence of this, the inventors will carry out additional experiments with one technique per animal. It is possible that microfluidically-inserted electrodes will travel in non-straight paths. In order to quantify responses along potentially curved paths, the inventors will reserve a subset of animals for sectioning parallel to the electrode path.

Acute and chronic electrophysiology following microfluidic insertion

As described above, the inventors’ partners at LLNL have been collaborating with the Frank Lab at UCSF to use temporary stiffeners to enable their electrodes to be inserted into the medial prefrontal cortex (mPFC) of rats. Given this existing data exists for comparison, the inventors will implant flexible, l6-site electrode arrays into the mPFC of one cohort of N=4 animals. Following the example of our pathological study, the inventors will implant bilaterally, with stiffener-assisted insertion in one hemisphere and microfluidic insertion in the other. The inventors will begin recording immediately post-surgery from both electrode arrays, and quantify the number of detectable neurons per electrode, cluster quality, and SNR3. The inventors and others have noted that the acute responses to insertion stiffeners (dissolvable [20] or removed following insertion [18]) can lead to noticeable tissue differences up to 12 weeks post implantation. Hence, the inventors anticipate that microfluidic insertion will result in significantly more neurons being detectable in the first 1-2 months post implantation. Even asymptotically, the number of neurons observable within 25-50 pm of a stiffener— implanted electrode is typically reduced by half [20]. Thus, the inventors further hypothesize that the inventors may chronically record twice as many neurons following microfluidic insertion.

Precise positioning

Many nuclei in the brain are small and thus very difficult to target for chronic recording. Recording in area CA1 of the hippocampus requires targeting a 100-150 pm layer of pyramidal cell bodies. Historically, tissue inflation during implantation has made it impossible to position electrodes within the pyramidal layer during surgery and achieve stable, long term recordings. Microfluidic actuation enables positioning of electrodes with approximately 20 pm precision. To test whether the lack of acute damage also results in stable positioning, the inventors will position the devices (an 8x2 grid spanning 280 pm (FIG. 7)) such that the central electrodes are positioned within the CA1 cell layer. In a cohort of N=4 animals, the inventors will implant electrodes bilaterally, using a removable stiffener in one hemisphere and microfluidic insertion in the other. The inventors anticipate that microfluidic insertion but not stiffener— assisted insertion, will result in stable CA1 cell layer placement.

It is possible that the process of delaminating the microfluidic chamber may cause the electrodes to move. In this case, the inventors will explore modifying alternative electrode designs LLNL has produced with attachment points, which can be anchored to the skull. Device Holder and Insertion

An example of a device holder with stereotactic arm is shown in FIG. 12. As shown in this embodiment, device holder comprises a housing with two separate halves that can be coupled to a stereotactic arm and an apparatus/device as described herein. In certain embodiments, the device holder may be manufactured using 3D printing.

FIG. 13 illustrates the device holder of FIG. 12 during use to insert a flexible electrode into brain tissue of a rat. Close-up views are shown of approximately 0.5 mm retraction and 1.0 mm retraction.

Additional analysis will be performed regarding minimally invasive implantation and actuation of bare flexible electrodes into cortex and deep brain regions. Large scale analysis will include multiple electrodes and multi-electrode shanks. In addition, wireless power and data will be analyzed, as well as optical and magnetic interfaces to develop injectable flexible electrodes positioned precisely within the central nervous system (CNS) and peripheral nervous system (PNS) with minimal damage.

FIG. 14 illustrates targeted neuromodulation using magnetic fields, including magnetoelectric thin films and magnetogenetics.

An optical alternative may include an implanted device that could show neuronal GCaMP (a genetically encoded calcium indicator).

All of the methods disclosed and claimed herein can be made and executed without undue experimentation in light of the present disclosure. While the compositions and methods of this invention have been described in terms of preferred embodiments, it will be apparent to those of skill in the art that variations may be applied to the methods and in the steps or in the sequence of steps of the method described herein without departing from the concept, spirit and scope of the invention. More specifically, it will be apparent that certain agents which are both chemically and physiologically related may be substituted for the agents described herein while the same or similar results would be achieved. All such similar substitutes and modifications apparent to those skilled in the art are deemed to be within the spirit, scope and concept of the invention as defined by the appended claims. V. References

The following references, to the extent that they provide exemplary procedural or other details supplementary to those set forth herein, are specifically incorporated herein by reference.

1. Hochberg, L. R. et al. Neuronal ensemble control of prosthetic devices by a human with tetraplegia. Nature 442, 164-171 (2006).

2. Collinger, J. L. et al. High-performance neuroprosthetic control by an individual with tetraplegia. Lancet 381, 557-564 (2013).

3. Shaikhouni, A. et al. Restoring cortical control of functional movement in a human with quadriplegia. Nature 1-13 (2016). doi:l0.l038/naturel7435

4. Chan, A. M. et al. Speech- specific tuning of neurons in human superior temporal gyrus. Cereb. Cortex 24, 2679-2693 (2014).

5. Quiroga, R. Q., Reddy, L., Kreiman, G., Koch, C. & Fried, I. Invariant visual representation by single neurons in the human brain. Nature 435, 1102-1107 (2005).

6. Truccolo, W. et al. Single-neuron dynamics in human focal epilepsy. Nat. Meth.

14, 635- 641 (2011).

7. Merricks, E. M. et al. Single unit action potentials in humans and the effect of seizure activity. Brain 138, 2891-2906 (2015).

8. Stead, M. et al. Microseizures and the spatiotemporal scales of human partial epilepsy. Brain 133, 2789-2797 (2010).

9. Polikov, V. S., Tresco, P. A. & Reichert, W. M. Response of brain tissue to chronically implanted neural electrodes. J NeuroSci Methods 148, 1-18 (2005).

10. Winslow, B. D. & Tresco, P. A. Quantitative analysis of the tissue response to chronically implanted microwire electrodes in rat cortex. Biomaterials 31, 1558— 1567 (2010).

11. Gilletti, A. & Muthuswamy, J. Brain micromotion around implants in the rodent somatosensory cortex. J Neural Eng 3, 189-195 (2006).

12. Subbaroyan, J., Martin, D. C. & Kipke, D. R. A finite-element model of the

mechanical effects of implantable microelectrodes in the cerebral cortex. J Neural Eng 2, 103-113 (2005).

13. Kozai, T. D. Y. et al. Ultrasmall implantable composite microelectrodes with bioactive surfaces for chronic neural interfaces. Nat. Mater. 11, 1065-1073 (2012).

14. Shen, W. et al. Extracellular matrix -based intracortical microelectrodes: Toward a microfabricated neural interface based on natural materials. Microsyst. Nanoeng.

1, 15010 (2015).

15. Canales, A. et al. Multifunctional fibers for simultaneous optical, electrical and chemical interrogation of neural circuits in vivo. Nat. Biotechnol. 33, 277-284 (2015).

16. Kim, T.-I. et al. Injectable, cellular- scale optoelectronics with applications for wireless 445 optogenetics. Science 340, 211-216 (2013). 17. Fu, T.-M. el al. Stable long-term chronic brain mapping at the single-neuron level. Nat. Meth. 13, 875-882 (2016).

18. Vitale, F., Summerson, S. R., Aazhang, B., Kemere, C. & Pasquali, M. Neural stimulation and recording with bidirectional, soft carbon nanotube fiber microelectrodes. ACS Nano 9, 4465-4474 (2015).

19. Goldstein, S. R. & Salcman, M. Mechanical factors in the design of chronic

recording intracortical microelectrodes. IEEE Transactions on Biomedical BME- 20, 260-269 (1973).

20. Kozai, T. D. Y. et al. Chronic tissue response to carboxymethyl cellulose based dissolvable insertion needle for ultra-small neural probes. Biomaterials 35, 9255- 9268 (2014).

21. Lewitus, D., Smith, K. L., Shain, W. & Kohn, J. Ultrafast resorbing polymers for use as carriers for cortical neural probes. Acta Biomaterialia 7, 2483-2491 (2011).

22. Lecomte, A. et al. Silk and PEG as means to stiffen a parylene probe for insertion in the 453 brain: toward a double time-scale tool for local drug delivery. Journal of Micromechanics and Microengineering 25, 125003 (2015).

23. Hassler, C. et al. Intracortical polyimide electrodes with a bioresorbable coating.

Biomedical Microdevices 18, 81 (2016).

24. Potter, K. A., Buck, A. C., Self, W. K. & Capadona, J. R. Stab injury and device implantation within the brain results in inversely multiphasic neuroinflammatory and neurodegenerative responses. J Neural Eng 9, 046020 (2012).

25. Saxena, T. et al. The impact of chronic blood-brain barrier breach on intracortical electrode function. Biomaterials 34, 4703-4713 (2013).

26. Liu, J. et al. Syringe-injectable electronics. Nature Nanotech 10, 629-636 (2015).

27. Prodanov, D. & Delbeke, J. Mechanical and biological interactions of implants with the brain and their impact on implant design. Frontiers in Neuroscience 10, (2016).

28. Lee, J. EL, Kim, H. S., Kim, J. H. & Lee, S.— H. Soft implantable microelectrodes for future medicine: prosthetics, neural signal recording and neuromodulation.

Lab Chip 16, 959-976 (2016).

29. Guitchounts, G., Markowitz, J. E., Liberti, W. A. & Gardner, T. J. A carbon— fiber electrode array for long-term neural recording. J. Neural Eng. 10, 46016 (2013).

30. Bird, R. B., Stewart, W. E. & Lightfoot, E. N. Transport phenomena. (JohnWiley & Sons).

31. Gad-el-Hak, M. MEMS. (CRC Press, 2005).

32. Bode, H. R. Head regeneration in Hydra. Dev. Dyn. 226, 225-236 (2003).

33. Ruggieri, R. D., Pierobon, P. & Kass-Simon, G. Pacemaker activity in hydra is modulated by glycine receptor ligands. Comparative Biochemistry and

Physiology Part A: Molecular & Integrative Physiology 138, 193-202 (2004).

34. Sun, Y. G. et al. Biphasic cholinergic synaptic transmission controls action

potential activity in thalamic reticular nucleus neurons. Journal of Neuroscience 33, 2048-2059 (2013). Pita-Almenar, J. D., Yu, D., Lu, H. C. & Beierlein, M. Mechanisms underlying desynchronization of cholinergic -evoked thalamic network activity. Journal of Neuroscience 34, 14463-14474 (2014).

Cogan, S. F., Edell, D. J., Guzelian, A. A., Ping Liu, Y. & Edell, R. Plasma- enhanced chemical 516 vapor deposited silicon carbide as an implantable dielectric coating. J Biomed Mater Res A 67A, 856-867 (2003).

Agmon, A. & Connors, B. W. Thalamocortical responses of mouse

somatosensory (barrel) cortex in vitro. Neuroscience 41, 365-379 (1991).

Quiroga, R. Q., Nadasdy, Z. & Ben-Shaul, Y. Unsupervised spike detection and sorting with wavelets and superparamagnetic clustering. Neural Comput 16,

1661-1687 (2004).

Tooker, A. et al. Towards a large-scale recording system: demonstration of polymer— based penetrating array for chronic neural recording. Conf. Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Annu. Conf. 2014, 6830-6833 (2014).