Low-cost and easy-fabrication lightweight drivable electrode array for multiple-regions electrophysiological recording in free-moving mice

Figure 1 shows a schematic of the proposed drivable component of the drivable MEANS. The drivable component, a three-layer nested construct, described from outside to inside, consists of an ethylene-vinyl acetate copolymer (EVA) heat-shrinkable tube, non-closed loop ceramic bushing (inner diameter, ID 1.24 mm, OD 1.60 mm), and stainless ferrule (ID 0.63 mm, OD 1.25 mm) with a bulge and twining silver wire (OD 0.20 mm). The ceramic bushing is wrapped by the EVA heat-shrinkable tube, which behaves elastically. The role of ceramic bushing serves as a linear track for a stainless ferrule, and the EVA heat-shrinkable tube keeps the track clean. Moreover, half of the EVA heat-shrinkable tube surface will also be embedded in the dental cement during implantation. Thus, the EVA heat-shrinkable tube-ceramic bushing will be fixed in the brain, allowing the stainless ferrule to move in it. The silver wire is tightly twined around and affixed to the stainless ferrule by cyanoacrylate glue to adjust the step distance by controlling the ferrule insertion depth. Cyanoacrylate glue is selected for its ease of use and low ductility; its brittle bonds poorly resist impact loading, making it convenient to remove the silver wire. Thus, the length of the removed silver wire controls the migration distance of the stainless ferrule, with a lap of silver allowing the stainless ferrule to move approximately 0.2 mm. The bulge on the stainless ferrule aligns with the crevice of the ceramic bushing to prevent extra rotation during insertion, driving, and recording.

Zoom In Zoom Out Reset image size

The design of the drivable MEANS can be divided into three parts: the core drivable component, working component, and supporting component. The drivable component (figures 2(b)–(e)) has been illuminated in figure 1. The working component comprises a Omnetics connector (a) with reference, ground wires (g), and a bunch of microwires (h). The epoxy resin (i) and silicon tubes (f) comprise the supporting components. The working component (a, h) and supporting component (i, j) are fixed to the stainless ferrule to keep them moving synchronously. All connectors and microwire pins without an insulating layer were encapsulated by epoxy resin. The cross-sectional view of the inner structure (j) was described from outside to inside, consisting of (e), (d), (b), (f), and (h).

Zoom In Zoom Out Reset image size

In our design, the finished drivable MEANS had no space between the ceramic bushing and the silver wire (figure 3(a)). If a drive is needed, a piece of silver wire can be removed to form a space between the ceramic bushing and silver wire to allow the stainless ferrule to be driven by a force (figure 3(b)). Since the EVA heat-shrinkable tube/ceramic bushing structure (between red dotted lines) was fixed to the mouse heads by dental cement, the working component, supporting component, and stainless ferrule of our drivable MEANS were lowered by the drive. The drive stopped when the ceramic bushing and silver wire were touched again (figure 3(c)). The total drivable distance and drivable precision were adjustable as needed using silver wires of different diameters and turns (figure 3(c)).

Zoom In Zoom Out Reset image size

Consistent with the optrode introduced in previous studies, optical fibers were integrated into the supporting component. Thus, we arranged the silicon tubes to be distributed around the uncovered optical fiber and affixed them together into the drivable component (figure 4). To ensure that the illumination of the neurons was recorded, the tips of the microwires remained approximately 500 μm longer than the optical fiber.

Zoom In Zoom Out Reset image size

Formvar-coated nickel and chromium microwires (17.78 µm, HFV insulation, California Fine Wire Company) were cut into 10 cm pieces, and then both ends of the pieces were clamped together with a small foldback clip, which was cemented horizontally to a small magnetic rotor (figure 5(a)). The middle of the wire was then passed over a horizontally fixed glass rod (Φ 10 mm) and placed over the center of the magnetic stirrer. The magnetic rotor was rotated at 100 rpm to twine the wire into a double-stranded helix to fabricate a stereotrode. Further, hot air at 210 °C was applied evenly to the helix for approximately 4–6 s using a heat gun from three sides. Approximately 2 mm of the formvar coating was removed from the single-strand ring by brief exposure to a flame (figure 5(b)). The ring was cut into two channels. Therefore, one stereotrode was completed. The preparation process of the tetrodes was the same. The reference and ground wires were soldered to four pins on the outer side of the Omnetics connector (A79026-001, Omnetics, figure 5(c)).

Zoom In Zoom Out Reset image size

Eight primary parts are required for the fabrication of the drivable MEANS. At the time of the study, all parts of the drivable MEANS were available for purchase and customization. The ceramic bushing (LC Tube, OD = 1.60 mm, ID = 1.25 mm, length = 6.6 mm, Shanghai Fiblaser Technology Co.) and stainless ferrules (LC630-S, OD = 1.25 mm, ID = 630 μm, length = 6.3 mm, Shanghai Fiblaser Technology Co.) were customized. A heat-shrinkable tube (SALIPT S-902-600, SL-AA006, Sanlian Heat-shrinkable Materials Co., Ltd, China) was cut into pieces (length ≈ 6.6 mm), which were equal in length to the ceramic bushing. A silver wire (OD = 200 μm, 7440-22 n, Sinopharm Chemical Reagent Co., Ltd) was used to manufacture the drivable component, which resulted in a step resolution of ∼50 μm (1/4 lap). The silver wire was tightly wound around the stainless ferrule and fixed with glue (Deli Group Co., Ltd, 7144) (figure 6(a1)). The heat-shrinkable tube was then sleeved on the ceramic bushing. After assembling the heat-shrinkable tube, ceramic bushing, and stainless ferrule, hot air (100 °C–120 °C) produced by a heat gun (Bosch GHG16-50) was blown through a three-layer nested structure to compress them together using the contraction of the outermost heat-shrinkable tube (figure 6(a2)). Then, glue (Alterco SG-2) was applied as a dot into the gap between the heat-shrinkable tube and ceramic bushing to stick them together. The bulge in the stainless ferrule was aligned with the slit on the ceramic bushing, and the stainless ferrule was inserted into the ceramic bushing with 1/3 of it remaining outside (figures 6(a3) and (a4)).

Zoom In Zoom Out Reset image size

Silicon tubes (TSP100170, Polymicro Technologies) were cut to the same length, which was also the same as the fabricated drivable component (figure 6(b1)). We employed a 10 silicon tube as a bundle and inserted it into the drivable component, leaving ∼2 mm outside (figure 6(b1)). Epoxy resin (353ND, Epoxy Technology) was used to dot the juncture of the silicon tube bundle and stainless ferrule, which made the silicon tube bundle move synchronously with the stainless ferrule (figure 6(b2) and (b3)).

Ten pieces of prepared stereotrodes were independently inserted into the silicon tube (figure 6(c1)) and affixed to them (figure 6(c2)). Then, the single-stranded ring was cut into two ends, and each stereotrode was wrapped around two adjacent pins of a standard electrode connector (figure 6(c3)). Therefore, each drivable MEANS had 20 channels, and two drivable MEANS in simultaneous recordings had 40 channels. Conductive silver paint (Ted Pella 16021, PELCO® SEM-Gold/Silver Extender, 25 ml) was then used to reduce the resistance between the stereotrodes and the pins.

After assembly of the drivable and working components, the acrylic resin was used for packaging the electrode. Acrylic resin, the supporting component, encapsulates all of the pins of the connectors and single-stranded microwires, providing physical support and protection to the working components (figure 6(c4)). Finally, the tips of the stereotrodes were cut with tungsten scissors (F·S·T 14958-09), and the calculated length remained outside according to the target DV coordinates (figure 7).

Zoom In Zoom Out Reset image size

For the fabrication of the optrode, eight silicon tubes (with eight stereotrodes, 16 channels) were distributed around the uncovered optical fiber (OD = 200 μm) evenly and in parallel, which ensured that the tips of the fiber and microwires straightforwardly penetrated the target brain region (figure 4). The optical fiber was cut into the required length, and the unencapsulated part of the optical fiber was protected by a heat-shrinkable tube. To ensure consistent and sufficient light intensity near the recording sites, we rigidly attached stereotrode bundles to the fiber shaft (diameter, ∼200 μm) and cut them to extend 400–800 μm beyond the end of the optical fiber. The other fabrication processes were the same as those described above (figure 6).

All electrochemical experiments were performed using a potentiostat (Gamry Reference 600, USA). To reduce the impedance and improve the signal-to-noise ratio (SNR) of the electrodes, platinum was plated on the tips of the microwires. The electrodes were installed in a three-electrode cell with a saturated calomel electrode (SCE) as the reference electrode and a large-area platinum electrode as the counter electrode. PtCl₄ (1 g 13454-96-1 from Tianjin Jinbolan Fine Chemical Co., Ltd) was dissolved in 100 mmol l⁻¹ HCl solution to a concentration of 5 mmol l⁻¹ to prepare the electrodeposition solution. One week of incubation in the dark at 4 °C was required to stabilize the solution. Platinum nanoparticles were deposited on the tip of the microwires in the deposition solution via a simplified cathodic reaction at a potential of −0.2 V (vs. SCE). The electrodeposition process was described by the following equation:

$$\begin{equation*}{\text{P}}{{\text{t}}^{4 + }} + {\text{4}}{{\text{e}}^ - } = {\text{Pt}}. \end{equation*}$$

The impedance of the drivable MEANS was decreased to 300–800 kΩ prior to use. After chronic implantation, the drivable MEANS was taken from the mice after sacrifice, digested with 0.25% trypsin (25200072, Gibco) for 30 min, and washed with 0.1 M phosphate buffer saline (PBS) before measurement. The impedance of each channel was measured at 1 kHz in the PBS.

Adult male C57/BL6J mice (18–22 g; 6–8 weeks old) were provided by the Guangdong Medical Laboratory Animal Center (Guangdong Province, China). Adult Sst-IRES-Cre knock-in mice (Jackson Laboratory, #013044) expressing Cre recombinase in somatostatin-expressing neurons were bred, identified, and provided by the Shenzhen Institute of Advanced Technology under strict criteria (18–22 g; 6–8 weeks old). All animals were housed under controlled conditions (ambient temperature 24 ± 1 °C, humidity 50%–60%, 12 h light/dark cycle) with food and water provided ad libitum. All experiments were conducted in accordance with the protocols approved by the Ethics Committee for Animal Research at the Shenzhen Institute of Advanced Technology, part of the Chinese Academy of Sciences (SIAT-IACUC-20210 310NS-NTPZX-ZC-A1625-02).

Surgeries were performed under aseptic conditions in a fully equipped operating suite. Before surgery, male Sst-IRES-Cre knock-in mice (18–22 g; 6–8 weeks old) were fixed in a stereotaxic frame (68044, RWD Life Science Company), induced with 1.5%–2% isoflurane (R510-22-16, RWD Life Science Company) with oxygen flow at 1 l min⁻¹ and then maintained isoflurane at 1.25%–1.5% with a gas anesthesia machine (R540IE, RWD Life Science Company). A thin layer of erythromycin eye ointment (H32025968, BAIJINGYU Pharmaceutical) was daubed on the eyes, and the heads were shaved and disinfected with 75% alcohol. Throughout the operation procedure, the body temperature of the mice was maintained using a hot water bottle (∼38 °C).

After the animal's head was fixed in a standard stereotaxic frame, the cranium was exposed through a small midline scalp incision. A small craniotomy around the expected implant site was performed with a micro-burr (diameter = 0.6 mm, round tip) that was mounted on a high-speed dental drill. AAV-DIO-ChR2-mCherry virus aliquots were allowed to thaw on ice for at least 15 min and diluted with sterile PBS if required by careful mixing by tapping the tube. If necessary, the liquid was spun to the bottom of the tube using a small benchtop centrifuge. The virus was then carefully taken up into the needles by automatically raising the plunger of a 10  μl Hamilton syringe with a 34 gauge needle. Virus (500 nl) was injected into the dLS (anteroposterior (AP) −2.06 mm, mediolateral (ML) −1.35 mm, DV −2.0 mm) at 100 nl min⁻¹ followed by a 9 min pause. After recovery, the animals were housed for virus expression for at least three weeks.

After the recovery of optrode implantation surgery for at least one week, a 473 nm blue laser controlled by an analog input was used for SOM-ChR2 activation. The laser was applied to the implanted optrode array using an optical fiber (200 mm diameter) terminated with an optical fiber patch cable. Blue light was delivered at a power of 10 mW with a cycling stimulation mode (5 ms pulses at 60 Hz, 30 s on, 150 s off).

Male (9–12 weeks old) C57BL/6J mice or Sst-IRES-Cre knock-in mice (>25 g) were anesthetized using the methods mentioned earlier. After the skin and connective tissue over the skull was removed, four small craniotomies around the expected implant sites were performed with a micro-burr (diameter = 0.6 mm, round tip) that was mounted on a high-speed dental drill. Two stainless screws (OD = 0.6 mm) were tapped into the skull in front of the recording sites with the unpierced dura as an anchor for electrode fixation. Two other stainless screws were tapped into the skull post lambda for stabilization and served as the ground and reference sites. Two small skull windows (approximately 400 μm × 400 μm) were made over the dLS and dCA3 with a dental drill. The drilling procedure was periodically interrupted, and the skull surface was scrubbed with saline to prevent damage to the brain tissue. The dura was carefully removed from both craniotomy sites, and the exposed brain was kept moist with artificial cerebrospinal fluid (ACSF). The ASCF contained 119 mM NaCl, 2.5 mM KCl, 1.3 MgCl₂, 1.0 mM NaH₂PO₄ and NaHCO₃, 2.5 mM CaCl₂, and 11 mM glucose.

After the skin and connective tissue over the skull was removed, four small craniotomies around the expected implant sites were performed with a micro-burr (diameter = 0.6 mm, round tip) that was mounted on a high-speed dental drill. Two stainless screws (OD = 0.6 mm) were tapped into the skull in front of the recording sites with the unpierced dura as an anchor for electrode fixation. Two other stainless screws were tapped into the skull post lambda for stabilization and served as the ground and reference sites. Two small skull windows (approximately 400 μm × 400 μm) were made over the dLS and dCA3 with a dental drill. The drilling procedure was periodically interrupted, and the skull surface was scrubbed with saline to prevent damage to the brain tissue. The dura was carefully removed from both craniotomy sites, and the exposed brain was kept moist with ACSF. The ASCF contained 119 mM NaCl, 2.5 mM KCl 1.3 MgCl₂, 1.0 mM NaH₂PO₄ and NaHCO₃, 2.5 mM CaCl₂, and 11 mM glucose.

Drivable MEANS was fixed to the stereotaxic apparatus using two glass slides (Fisherbrand™ Plain Glass Microscope Slides, 12-549-3) and slowly lowered to the brain. In particular, the electrode array targeting the dLS was tilted 10° clockwise to avoid damaging the cerebral tissue surface and the subcortical vascular plexus, while the electrode array targeting the dCA3 was vertical (figure 7).

The tips of electrodes were directed toward the brain at the following stereotaxic coordinates: at AP +0.74 mm, ML −0.75 mm, and DV −2.80 mm for dLS recording, at AP −2.00 mm, ML −1.90 mm, and DV −2.00 mm for dCA3 recording, at AP −1.65 mm, ML −1.00 mm, and DV −4.80 mm for LH recording. The actual implantation coordinates for dLS were as following: at AP +0.74 mm, ML −0.2 to −0.5 mm and DV −2.60 to −3.1 mm (figure 7).

As the microwire tips reached the expected depth, the ground and reference wires were twined together and connected to the same bone screw to reduce the probability of being scratched by the mice. The ground and reference wires from the different electrodes were connected to different bone screws. Conductive silver paint was used to reduce the resistance between the contact points on the screws. The electrode array and bone screws were cemented in place with dental cement (figure 7). Notably, the dental cement encapsulated the screws and electrode with a hollow structure to reserve space for the drive.

After the procedure, saline was injected subcutaneously into the back of the animal to aid recovery. Buprenorphine (0.3 mg kg⁻¹) was administered twice daily for three days as a post-operative analgesic. The electrical adhesive tape was paved to encapsulate the entire electrode and dental cement. The animals were allowed to rest for a week in their home cages with free access to food and water. For optrode implantation in transgenic mice, all procedures were the same as described above.

We verified the placement of the implanted electrodes via histological analysis. The dLS and dCA3 showed clear traces of the microwire bundles and the final depths of the microwire tips in the target brain regions (figures 7(c) and (d)).

A free exploration paradigm was used to investigate the properties of the electrodes. The apparatus consisted of two test chambers (30 cm × 25 cm × 20 cm), which were connected by an 8 cm door to allow the mice to pass freely. One chamber was assembled with a box (10 cm × 10 cm) filled with chow, and another chamber was assembled with a box (4 cm × 4 cm) filled with water (figure 9(a)).

At the beginning of the test, the animal was gently placed on the door. Twenty minutes of free exploration was conducted, while food and water were freely accessible. The apparatus was cleaned with 20% ethanol, and then the food and water were replaced between each trial. Generally, behavioral tests and electrophysiological recordings were performed on days 7, 9, and 11. For long-term recording, tests were performed every three weeks.

Electrophysiological recordings were performed using a 64-channel neural acquisition processor (Plexon, Dallas, TX, USA) based on previous studies (Lu et al 2016, Wang et al 2020). Data in all recording channels were sampled at 40 kHz and bandpass-filtered at 300–5000 Hz. Synchronized mouse behavior was recorded using a digital video camera (Plexon, Dallas, TX, USA). The depth of each step was individually conducted before recording based on the number of recorded single-unit activities in each brain area. In general, a step of ∼50 μm was considered appropriate. During the chronic recording experiment, if no signal was recorded, the electrode was actuated manually at a distance of 50 μm at a time until a typical signal was observed.

Mice were anesthetized with isoflurane and transcardially perfused with cold 0.1 M PBS (pH 7.4), followed by a 4% solution of paraformaldehyde dissolved in 0.1 M PBS. A fixed solution of 300 ml was used per 100 g of body weight. After perfusion, the animal brain was removed from the skull and post-fixed in the same fixative at 4 °C for more than 24 h before being moved to 30% sucrose in 1× PBS solution to dehydrate for three days. Subsequently, the block tissue around the implant was paraffin-embedded, and horizontal sections (35 μm thick) were prepared from all brain samples using a freezing microtome (Leica CM 1950).

PBS (0.1 M) was used to wash the brain sections three times for 10 min each time. DAPI (4',6-diamidino-2-phenylindole, Beyotime, C1002) was used to label the cell bodies by staining for 10 min. The wash process was repeated. All staining and washing processes were performed on a shaker. Fluorescence images were obtained using an automated slide scanner (Olympus BX61VS). Histology was performed on all chronically implanted animals for verification of implantation precision.

Electrophysiological signal data analyses were performed using existing commercial software and software written in MATLAB and Neuroexplorer (Lu et al 2016, Wang et al 2020). Neural electrophysiological data for all recording channels were bandpass-filtered. Spike continuous signals were recorded at 40 kHz, and the local field potential (LFP) signals were recorded at 1 kHz. Spike activity was extracted from continuous spike signals using standard spike sorting routines (Offline Sorter, Plexon, USA). The spike continuous signals were high-pass filtered (300 Hz) with a Bessel filter for spike detection. The threshold for spike detection was set to −4.5 SD (standard deviations), and the spike waveforms were measured in a time window 1400 μs long and beginning 300 μs before threshold crossing. Principal component values were calculated for the unsorted aligned waveforms. A group of waveforms was considered to be generated from a single unit if it was distinct from the other clusters. The dead time was set to 1.4 ms.

To characterize putative GABAergic neurons and putative glutamatergic neurons, two features of the extracellular waveform, the peak-to-peak time, and the half-width of the spikes, were calculated. Neurons with narrow peaks were regarded as putative GABAergic neurons, whereas neurons with wide peaks were regarded as putative glutamatergic neurons (Lu et al 2016). Auto-correlogram histograms were created using Neuroexplorer. The firing rate was normalized by dividing by the average baseline (the firing rate 5 s before scramble onset) across all trials, separately for each unit.

The sorted data were analyzed using SNR, which provides a mature method to show the unit quality. The average SNR included all SNRs for all the excepted channels unable to detect single units. The SNR for a given cluster was calculated using the following formula (Suner et al 2005):

$$\begin{equation*}{\text{SNR = App}}/\left( {{\text{2}} \times {\text{S}}{{\text{D}}_\unicode{x03B5}}} \right){\text{,}}\end{equation*}$$

where App (amplitude) is the peak-to-peak voltage amplitude of the mean waveform of each unit, and SD is the standard deviation of noise in a single unit. App is the absolute value of the difference between the maximum and minimum voltages of the mean waveform, which is calculated by subtracting the mean waveform from each waveform, and noise is the collection of all residual values irrespective of their position in the matrix. SD is the standard deviation over all values of .

Statistical analyses were performed using MATLAB or Origin software. All data are reported as the mean ± SD measurement. Statistical significance was determined using a two-sample t-test. No statistical methods were used to predetermine sample sizes; however, our sample sizes were similar to those of previous studies (Lu et al 2012). For all the studies, individual animals were used as replicates. Replication studies confirmed our results in all the experiments. The data distribution was assumed to be normal; however, this was not formally tested. A 2-point moving average filtering was used on the cross-correlation to remove the variations.

A small, lightweight drivable MEANS was obtained. The overall dimensions of the fabricated drivable MEANS were 15 mm × 15 mm × 4 mm, and their weight was about 0.37 g. The weight of the completed optic fiber-integrated drivable MEANS increased slightly to 0.47 g. After the animals were sacrificed, the weight of the two electrode implants, including the weight of the dental cement, was estimated to be approximately 1.2 g. Bone screws and dental cement had a great increase in the weight of implants over the mouse head. The impedance at 1 kHz of the collected electrode was measured in PBS (figure 8(b)). The results show that electroplating can effectively reduce the impedance of the electrode, which does not change significantly after implantation. The appropriate impedance could be maintained during implantation for at least 150 days. Altogether, these results suggest that our electrodes show great potential for multiple-region electrophysiological recording in free-moving mice owing to their lightweight and impedance stability.

Zoom In Zoom Out Reset image size

To investigate the effect of surgery and implants on the locomotion of mice, we implanted two electrodes at the dCA3 and dLS simultaneously, as shown in figure 8. After recovery from surgery, a free exploration paradigm was performed (figure 9(a)). Experimental mice were able to freely explore every corner of the device (figure 9(b)). As the total traveled distance did not show a significant difference between the experimental and control mice (figure 9(c)), it was suggested that the surgery and implants exerted no noticeable changes in locomotion during recording. No apparent abnormal postures in food intake were observed in experimental animals before (figure 9(d)) or during recording (figure 9(e)). Together with our previous results, these data show that the drivable MEANS achieves a lightweight, stable drivable electrode for multiple-region electrophysiological recording in free-moving mice.

Zoom In Zoom Out Reset image size

To evaluate the efficiency of the ceramic drivable MEANS in chronic recording, we recorded hippocampal neurons in C57BL/6J mice (n = 4) that were either in their home cage or during free exploration. Electrophysiological recording data showed that most of the electrode channels could record spikes well (spikes were presented in ∼8 steretrodes, figure 10(a)). Superimposed waveforms from two different units recorded simultaneously in the same channel show them separated from noise with good clusters (figure 10(b)).

Zoom In Zoom Out Reset image size

To further demonstrate the efficiency of the chronic recording of our drivable MEANS, we recorded spontaneous neuron activity with the same electrode and achieved five-month recordings (figure 10(c)). Five months after implantation surgery, electrophysiological signals can also be recorded by the same electrode. Different waveforms come from different neurons recorded due to controlled displacements of the drivable MEANS. In addition, raw LFP traces and spectrograms show that during chronic recording, the recorded LFP showed no 50 Hz power line noise (figure 10(c)), and the SNR of the sorted units shows that the recorded spike signals are appropriate for further analysis in application (figures 10(d) and (e)). Altogether, these results suggest that our electrodes, which showed good stability and anti-interference performance, can measure neuronal activity during chronic electrophysiological recording in free-moving mice.

To demonstrate that our drivable MEANS can perform recording in multiple brain regions simultaneously to facilitate functional dissection of special neural circuits, we recorded the dorsal lateral septum (dLS) and its major downstream LH and monitored food-seeking behavior. We found that 68.97% of putative GABAergic neurons in the LS (LS^GABA) were excited when the mice entered the food zone (food-seeking behavior, shown in figures 11(a) and (b)). In the downstream LH, only the putative GABAergic neurons (LH^GABA) were inhibited, while putative glutamatergic neurons (LH^Glu) and 71.43% LH^GABA were excited (figures 11(c)–(e)). These results suggested that LS^GABA inhibited LH^GABA and then inhibited LH^Glu and other LH^GABA. To further confirm that the excited putative LS^GABA was disinhibited by LH^GABA, the cross-correlation during mice entering the food zone was analyzed (figure 11(f)). All the aforementioned results show that the LS^GABA-LH^GABA-LH^Glu/GABA circuit may be involved in food-seeking behavior. Thus, this vivid example shows that the drivable MEANS has great application prospects for functional dissection of neural circuits, especially those involving specially positioned brain regions.

Zoom In Zoom Out Reset image size

To investigate the potential of our design for integration with other technologies, we fabricated and implanted optic fiber-integrated drivable MEANS in the dLS of Sst-IRES-Cre knock-in transgenic mice. First, the ChR2-mCherry virus was injected and expressed in the dLS of six mice (figures 12(a) and (b)). Then, we implanted the optical drivable MEANS into the dLS to activate the SOM+ neurons and recorded the activity of dLS neurons simultaneously. The traces of the microwire bundles were confirmed by histological analysis (figure 12(c)). We then explored the integration of electrical recordings with optical stimulation in free-moving mice. We then directed 473 nm blue light pulses (10 mW mm⁻², 5 ms pulses at 60 Hz) into the implanted optrode array for optogenetic stimulation, and LFP traces and spikes continued to show light-evoked and frequency-dependent responses of SOM+ neurons (figures 12(d)–(f)). This result indicates that an optical fiber can be loaded onto the drivable MEANS to achieve synchronous optogenetic modulation and electrophysiological recording in vivo.

Zoom In Zoom Out Reset image size