Devices that electrically regulate the deep brain have made important breakthroughs in the management of neurological and mental diseases. Such devices are usually centimeter-level and require surgical implantation and wired power supply, which increases the risk of bleeding, infection and damage in daily activities. The use of smaller remotely powered materials may result in less invasive neuromodulation. Here, we introduced injectable magnetoelectric nanoelectrodes, which can wirelessly transmit electrical signals to the brain in response to an external magnetic field. This regulation mechanism does not require genetic modification of nerve tissue, allowing animals to move freely during the stimulation process and use non-resonant carrier frequencies. Using these nanoelectrodes, we demonstrated neuronal regulation of deep brain targets in vitro and in vivo. We also show that local subthalamic modulation promotes modulation in other areas connected by basal ganglia circuits, which leads to behavioral changes in mice. Magnetoelectric materials provide a universal platform technology that can be used for less invasive deep neuromodulation of the brain.
Electrical communication and regulation with the central nervous system (CNS) are essential to our current understanding of neurobiology and the diagnosis and treatment of neurological diseases. By sensing and/or regulating neuroelectric activity, the main treatment method of CNS has achieved significant medical breakthroughs. For more than 30 years, deep brain stimulation (DBS) has provided patients with symptomatic relief of Parkinson’s disease and other diseases using electrodes connected to deep brain targets (1). Recently, the closed-loop control of epidural electrical stimulation has enabled patients with spinal cord injury to walk (2). This type of equipment allows patients to move freely, thereby achieving daily activities and long-term patient use.
In recent years, in order to make neural interventions less invasive, more durable and safer, the function of neural devices has been developed [for a review, please refer to (3)]. The main challenge for this type of equipment is power supply. When using a wired power supply, for DBS equipment, patients need to replace the surgical battery every 3 to 5 years (4). Instead, magnetic induction (5), photoelectric signals (6-8), acoustic power of piezoelectric materials (9-14), magnetic heating (15), and piezoelectric power of light-emitting diodes (5) are used to generate remotely powered Neural equipment. LEDs (16, 17) or magnetoelectric materials (18) replace wired batteries.
Similar to traditional DBS electrodes, centimeter-level equipment requires surgery and implantation of hardware outside the CNS, which may cause cerebral hemorrhage, infection and damage during daily activities (4). As a result, some neurodevice technologies have shifted to smaller (nanometer to millimeter) devices that can potentially be implanted completely into the CNS by injection. However, the smaller size makes it more difficult to power the nerve device. So far, the tissue penetration depth of remotely powered devices using magnetic induction (5) or photoelectric signals (6, 7) has been limited, reaching the maximum of 1 cm and 6 mm, respectively (19). Ultrasonic-powered piezoelectric devices are perhaps the most promising of these technologies, and recently demonstrated the use of sub-mm3 devices to record in multiple locations through 5 cm tissue phantom materials (10). However, the use of piezoelectric devices for modulation is currently only demonstrated in the use of millimeter-level devices or in the nervous system outside the body (12-14). Since power transmission is usually done at the mechanical resonance frequency of this type of equipment, a basic trade-off arises, that is, it is possible to power smaller and smaller devices with higher resonance frequencies at shallower tissue depths (20, 21). Therefore, the power supply based on resonance coupling creates obstacles for the use of injectable-sized devices to adjust deep brain targets.
In order to circumvent the challenge of signal transmission, other strategies also use genetic neuron modification and magnetocaloric nanoparticles (15) or piezoelectric powered LEDs (16, 17) to trigger ion channels to open. However, the reliance of such technologies on genetic modification creates regulatory barriers to transforming them into patients. Clinically, transcranial magnetic stimulation (TMS) can be used for wireless adjustment of neural activity without the need for implanted equipment (22). However, TMS can only regulate cortical tissue (23) and has a compromise of depth focus area (24, 25), so DBS cannot be performed through TMS at present.
In order to transmit wireless signals to the injectable device, we use magnetoelectric nanoelectrodes, which couple magnetic and electrical signals (Figure 1, A and B). Techniques for neuromodulation using magnetoelectric materials have been explored previously. Centimeter-level and millimeter-level magnetoelectric equipment has been used in DBS, which uses equipment mounted on the skull and connected to the deep part of the brain (18). In another study, magnetoelectric nanoparticles (MENPs) were used to achieve neuromodulation, although no repetitive or statistical analysis was available to verify in vivo efficacy (26). However, these studies prove the promise of magnetoelectric materials for neural devices.
The schematic diagram shows the two-phase magnetoelectricity (A) in a material made of strain-coupled magnetostriction and piezoelectric materials. This schematic diagram illustrates the principle of using a large DC magnetic field covered with an AC magnetic field to generate the best magnetoelectric output (B). Diagram of MENP administration method in vivo. MENP was bi-injected into the subthalamic area of mice, and MENP was wirelessly stimulated using AC and DC magnetic fields (C). Transmission electron microscope (TEM) (D) and transmission electron energy loss spectroscopy (TEM-EELS) images (E) show MENP morphology and BaTiO3/CoFe2O4 phases (green and red respectively), and each is measured by quantitative elemental analysis in mole percentage Material (E). MENP was analyzed by X-ray powder diffraction (XRD) to confirm the perovskite crystal structure of BaTiO3 (green) and the spinel crystal structure of CoFe2O4 (red) (F). au, arbitrary unit. Dynamic light scattering (DLS) is used to characterize the hydrodynamic properties of MENP in cell culture media and artificial cerebrospinal fluid (aCSF) (G). Measure the magnetization of MENP in the range of -1 to 1 T, and oscillate in the range of 0.205 to 0.235 T (inset) (H). mu, electromagnetic unit. The input-output magnetoelectric coefficient (αME) of the particles in the sintered particles was measured as a function (I) of the DC bias field in MENPs and MSNPs. A 220 mT DC magnetic field was used to simultaneously change the AC magnetic field amplitude to measure the voltage normalized to the thickness of MENP particles (J). A 220 mT DC magnetic field (K) was used to measure the AC magnetic field frequency dependence of αME. The graph shows a single point with an average of ±SD (n = 3) (G), and a single point (J and K) fitted to a linear correlation.
Here, we report wireless DBS in mice using injectable magnetoelectric nanoelectrodes. They were implanted in the subthalamic area by stereotactic infusion and were powered by an external magnetic field at a non-resonant carrier frequency and freely moving mice (Figure 1C). In particular, we made two-phase MENP using magnetostrictive CoFe2O4 nanoparticles (MSNP) coated with piezoelectric BaTiO3. The two materials are strain-coupled through the sol-gel growth of BaTiO3 on CoFe2O4 nanoparticles. Wireless particle stimulation is achieved by applying a magnetic field, which generates strain in CoFe2O4, which in turn applies strain to BaTiO3, resulting in charge separation (Figure 1A). Below, we demonstrate the use of an applied magnetic field to wirelessly generate an electric field across MENPs. Then, we showed that the magnetic stimulation of MENPs can wirelessly regulate neuronal activity in vitro and in vivo. Finally, we demonstrated the therapeutic potential of this technology by modulating the activity of the motor cortex and non-motor thalamus and the ability to change animal behavior.
A two-phase MENP was synthesized using a protocol similar to Corral-Flores et al. (27). The morphology of the nanoparticles (Figure 1, D and E), the ratio of magnetostriction to piezoelectric material (Figure 1E) and the crystal structure (Figure 1F) are characterized. We observed that the perovskite and spinel crystal structures contained 36.1±0.6% BaTiO3 and 63.9±0.6% CoFe2O4 in two-phase MENP, respectively. The hydrodynamic properties of MENP were also characterized by dynamic light scattering (DLS) in cell culture media and artificial cerebrospinal fluid (aCSF) solutions. In the medium and aCSF, the average particle size was 224±17 nm and 277±18 nm, respectively, and the zeta potential was -8.6±0.5 mV and -6.7±0.5 mV, respectively (Figure 1G). The magnetization of MENP was measured in the range of -1 to 1 T (Figure 1H) and 205 to 235 mT oscillation (Figure 1H, inset).
Next, we measured the electrical output of MENP with an applied magnetic field to characterize its magnetoelectric response. By connecting electrodes and measuring the output voltage through a lock-in amplifier, MENPs are measured as sintered polarized balls (Figure S1). Although this method does not allow us to measure the magnetoelectric effect on the nanoscale, it can verify whether our material is magnetoelectric and has previously been used to evaluate the magnetoelectricity of core-shell particles (28-30). The pellet containing only MSNP was used as a negative control. In order to optimize our ME output, we applied a smaller AC magnetic field and a larger DC bias magnetic field along the same axis (Figure 1B). This direction is used to align the magnetic domains, magnetostriction axis, and piezoelectric polarization axis to sum the magnetoelectric output along our measurement axis. Applying a sinusoidal magnetic field to the magnetoelectric material will output a sinusoidal electric field whose frequency and duration match the input magnetic field. Therefore, we can use a lock-in amplifier (29) to measure this output. The magnetoelectric coefficient (αME) quantifies the relationship between the input AC magnetic field and the output voltage, and changes nonlinearly with the DC field, which is a typical magnetoelectric material (31). In MENP clumps, αME reaches a maximum of 86 V m-1 T-1 at 200 and 225 mT, while MSNPαME does not depend on the DC field (Figure 1I). Using a DC magnetic field in the maximum αME range (220 mT), we measured the linear relationship between AC magnetic field strength and voltage (normalized to particle thickness; R2 was 99.8 and 99.7% at AC frequencies of 140 and 280 Hz, respectively ; Figure 1J) is also a typical magnetoelectric material (31).
We found that in the entire range of the test (35 to 385 Hz), αME has low dependence on the frequency of the AC magnetic field (respectively 1.4 and 1.3% for R2 with AC amplitude of 2 and 3 mT), which covers the DBS frequency Range of clinical effects (Figure 1K) [reviewed in (32)]. The attenuation of this frequency range in the tissue is also very small, thereby improving the potential signal penetration depth (20, 21).
Using intracellular Ca2 + signaling in differentiated human SH-SY5Y cells, the effect of wireless MENP signaling on neuronal cell activity was tested in vitro in real time. Twenty minutes before the test, NP, MSNP and piezoelectric nanoparticles (PENP) were not used as controls, and 100 μg/ml MENP was suspended in the imaging medium. Before selecting the concentration, through lactate dehydrogenase (LDH) analysis and metabolic activity analysis ([3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)- 2-(4-sulfophenyl)-2H-tetrazolium (MTS) (Figure S2).
The magnetic stimulation parameters are no magnetic field, 225 mT (within the maximum αME range) DC magnetic field, 6 mT, 140 Hz AC magnetic field, or use a custom coil system to use DC and AC magnetic fields together (Figure S3). Select AC stimulation parameters to match the frequency commonly used in clinical DBS, and maximize the magnetoelectric output by increasing (6 mT) the amplitude of the AC magnetic field. It is expected that the DC or AC magnetic field alone will not produce enough magnetoelectric effects to regulate neuronal activity, so it is used as a control.
We found that when both AC and DC magnetic fields (20.1±2.3%) were used to stimulate MENPs, the percentage of cells exhibiting Ca2 + transients increased significantly relative to the basal activity (2.8±2.6%) (Figure 2, A to C, And movie S1). This increase was not observed when cells were exposed to AC and DC magnetic stimulation (1.0±1.7%), MSNP (1.4±1.3%) or PENP (1.4±1.2%) alone, which supports our hypothetical measured activity The increase is attributed to MENP response. Although MENP has certain effects on neuronal activity (7.2±5.0%, 5.2±6.0% or 3.8±5.0%, respectively, no electric field, only AC electric field or only DC electric field), there is no significant difference in this effect from any Other negative control groups (Figure 2, B and C, Table S1 and Movie S1).
Before magnetic stimulation, cells were treated with MENPs, and NPs, MSNPs or PENPs were not used as controls. Magnetic stimulation is 220 mT DC (DC), 6 mT and 140 Hz AC (AC), or DC and AC magnetic fields (AC + DC) along the same axis. Use Fluo4 dye to measure neuronal activity in real time through intracellular Ca2+ imaging and track the cell fluorescence of each cell over time. The image showing the total Ca2+ activity of the selected experimental group over time. The calibration bar represents ΔF/Fo(A). (B) summarizes the percentage of cells that show intracellular Ca2+ transients in (A), and (C) marks significantly different group comparisons in yellow. Before treatment with MENP and AC and DC magnetic stimulation, the cells were treated with TTX, Cd2 + or EGTA, and then the total Ca2 + activity was displayed over time, next to the graph depicting the inhibitory activity of each drug (D). The Ca2+ transients of the drug-treated cells are summarized. Without the drug, MENP and AC+DC field-treated cells are shown as faded plot points for comparison (E). The curve shows the traces of Ca2 + activity in a single cell (A and D) over time. The bar graph at each point represents the mean ± SD (B and E) (n = 3 to 6); Tukey's post-test ( C) or Dunnett's post-test (no drug) as a control ANOVA (E); *** P <0.001, no significant difference between the unlabeled group.
To support our hypothesis that the Ca2+ activity we measured is related to electrophysiological cell activity, we stimulated MENP with AC and DC magnetic fields, but first treated with a voltage-gated Na+ channel blocker [tetradotoxin (TTX)] cell. Voltage-gated Ca2+ channel blocker (Cd2+) or extracellular Ca2+ chelator (EGTA) (the schematic in Figure 2D shows the drug activity). In the presence of each drug, the cells were unable to produce any Ca2+ transients (Figure 2, D and E).
Then, we tried to evaluate the feasibility of MENP-based neuromodulation in vivo. MENPs were injected bilaterally into the subthalamic area (C57Bl/6J) of naive mice at a dose of 100μg per animal (100μg/ml total amount is 1μl), and it was tolerable after dose toxicity assessment (Figure S4). We found that there was no significant difference in glial fibrillary acidic protein (GFAP) or Iba-1 staining around the injection site with MENP concentrations of 25, 50, or 100 mg/ml (Figure S4, C and D). We also evaluated the injection site at 48 hours, 2 weeks and 4 weeks after the 100 mg/ml injection of MENP, and no qualitative signs of particle loss or changes in tissue response were observed. In the third phase of our animal study, animals euthanized 7 weeks after injection also showed no qualitative signs of MENP migration or loss. We analyzed the injection area of these mice after death and found that MENP occupies a volume of 0.0088±0.0023 mm3 (Figure 3D).
After DC magnetic stimulation (A) or AC and DC magnetic stimulation (B), local staining of c-Fos protein at the MENP injection site showed increased c-Fos expression (C) and c-Fos positive tissue volume (D) increased the latter . When stimulating MENP with AC and DC magnetic fields (H) instead of DC magnetic fields (G), quantification of c-Fos expression in the motor cortex (E) and limbic thalamus (F) showed increased expression. TH, Thalamus; cp, stalk; STh, subthalamic nucleus, HPF; hippocampus formation; cc, call body; PV, paraventricular nucleus. Time-lapse image showing mouse movement (I) in the CatWalk video recording system. The dynamic motion parameters measured by CatWalk recording showed that after AC and DC stimulation and DC stimulation, in MENP-treated mice, the mouse speed, limb duty cycle and limb step length (J) changed significantly, and There was no significant change in MSNP-treated mice. The static movement parameters of the rat movement measured by CatWalk recording, such as regularity index, maximum movement change and front paw support base, in the two nanoparticle groups (K), AC and DC have no significant changes compared with DC only . In any nanoparticle group (K), AC and DC compared with DC-only magnetic stimulation, the latency of Rotarod decline did not change significantly. Scale bars, 250 (overview) and 50μm (illustration) (A, B, G and H). The graph shows a single point, and the bar graph represents the mean ± SD (C to F, J and K) [n = 8 mice (C); n = 6 to 7 mice (D to F); n = 8 Up to 9 mice (J and K), the duty cycle and stride length of each limb value]; unpaired t-test (C to F) or paired t-test (J and K), ** P <0.01 and * ** P <0.001. ns, not important.
During the magnetic stimulation, the mice are awake and unconstrained in our internal electromagnetic coil equipment (Figure S5). As a control group, mice were treated with MENP and DC magnetic fields only, which meant that they were placed in a magnetic device while the AC coil was kept closed. We use immunohistochemistry to measure the expression of c-Fos protein, a widely used marker of cell activity, to assess changes in local neural activity (33). We found that when animals were treated with MENPs and AC and DC electric fields (38.5±8.0 cells), there were significantly more c-Fos positive cells in the nanoparticle injection area than only the DC electric field (4.25±3.0 cells) (Figure 3, A to C). In addition, when animals were treated with MENPs and AC and DC fields, the tissue volume containing c-Fos-positive tissue was significantly larger (0.0349±0.0089 mm3), while only the DC field (0.0098±0.0031 mm3) was used. There were no significant differences in tissue regions containing MENP (0.0090±0.0014 and 0.0085±0.0031 mm3, respectively) (Figure 3D).
Next, we wanted to determine whether the local subthalamic neuromodulation caused by MENPs is sufficient to cause the regulation of other areas of the cortex-basal ganglia-thalamic cortical circuit. We found that after stimulation with MENPs and AC/DC magnetic field, the expression of c-Fos protein in the motor cortex and non-motor thalamus increased significantly (1046.4±232.4 and 348.4±137.7 cells/mm2, respectively), while only the DC magnetic field ( 424.8± are 214.9 and 19.9±27.6 divisions/mm2, respectively) (Figure 3, E to H). We did not observe overall changes in c-Fos protein expression, such as in the hippocampal CA1 area (Figure 3, G and H).
To determine whether the induced neuromodulation affects animal behavior, we tested mice with a rotating tripod test and an automated CatWalk XT gait analysis system. Mice were injected with MENP or MSNP as controls. Compare the AC and DC magnetic stimulation behaviors of each mouse with only the DC magnetic stimulation behavior. During the CatWalk test, static parameters related to gait and balance, such as regularity index, running maximum change and support base, were not significantly different after AC and DC stimulation in any nanoparticle group (MSNP: 97.3 ±2.5 and 97.9±1.8%, 28.3±7.6 and 31.9±11.0%, 1.2±0.1 and 1.2±0.1 cm, DC and AC and DC stimulation respectively) (MENPs: 97.8±1.0 and 98.1±1.3%, 23.7±3.3 and 23.6±8.9% were DC and AC and DC stimulation was 1.2±0.1 cm and 1.2±0.1 cm, respectively) (Figure 3, J and K and movie S2). The Rotarod test also showed that there was no significant difference in the drop latency of any nanoparticle group (MSNPs: 170.4±72.9 vs 170.9±73.2 s; MENPs: 202.6±40.4 vs 184.0±40.5 s, DC vs AC and DC stimulation) (Figure 3K). In general, there were no significant differences in gait and balance parameters between the two groups. This is essential for determining that the magnetic stimulation of MENP will not affect the animal’s movement.
On the contrary, when analyzing the speed-related parameters and dynamic parameters of the CatWalk test, these parameters indicate high-speed gait. We found significant differences in the behavior of MENP-treated animals, but no such differences were observed in MSNP-treated animals (Figure 2) . 3. J and K, and movie S2). Under AC and DC stimulation, the average speed, duty cycle of each limb and step length of each limb of the mice treated with MENP changed significantly (51.1±10.9 vs. 33.6±4.8 cm/s, 48.1 ±3.0 vs. 49.9±3.5%, and 8.1±0.5 vs. 7.3±0.7 cm, respectively, DC vs. AC and DC stimulation), but not applicable to MSNP-treated mice (28.3±5.0 vs. 29.4±3.8 cm/s, 51.9 ±3.9 to 51.1±3.2% and 6.5±0.4 to 6.4±0.6 cm, DC to AC and DC stimulation).
With the increase in potential applications of neural devices, new technologies that make neural interventions safer, longer lasting, and less invasive have aroused people’s interest in small wireless devices. The remote power supply of nerve equipment can not only make the equipment smaller, but also does not require huge equipment or replacement of surgical batteries. Recently, some remotely powered devices have appeared, which can achieve less invasive neuromodulation and even reach the depths of the brain (3, 12-18). One of the most promising methods does not rely on genetically modified nerve tissue, but directly generates electrical signals to achieve neuromodulation (12-14, 17, 18). However, it is not yet possible to sufficiently reduce the size of such devices so that they can be fully implanted in the brain while still achieving deep brain neuromodulation.
In this work, we use MENPs as nanoelectrodes, with the purpose of wirelessly regulating neuronal activity through remote power supply through a magnetic field. We characterized the magnetization of MENP, especially observed the magnetic field range that oscillated in the magnetic field range used in the in vitro and in vivo experiments, and did not find hysteresis (Figure 1H, inset). It is important to show that MENP does not generate heat during in vitro and in vivo experiments, thereby eliminating heat as a source of off-target neuromodulation.
We characterize the magnetoelectric response of nanoelectrodes as sintered particles, especially the electrical output when the input magnetic field changes. Although only an AC magnetic field can initiate the magnetoelectric response, a large AC magnetic field is required to reach the maximum αME, which will require a powerful coil system and other components to achieve active cooling. Instead, we use permanent magnets to apply a larger DC magnetic field and cover a smaller AC magnetic field to maximize αME (Figure 1, B and C, and Figure S3 and S5). When changing each input component (ie, DC field size, AC field size and AC field frequency), we determined that MENP nanoelectrodes output the electrical response characteristics of magnetoelectric materials (Figure 1, I and J). We have not found that this phenomenon occurs with only magnetic MSNP.
In this study, an important finding corresponding to previous work in the field of magnetoelectric materials is that the magnetoelectric output has a low dependence on the frequency of the input carrier AC field (Figure 1K). When αME increases sharply near the mechanical resonance frequency of the magnetoelectric material, αME otherwise remains relatively constant (34). In this study, our carrier magnetic signal is far away from the resonance frequency range of nanomaterials (140 Hz vs. GHz range). Previous neural device technologies based on piezoelectric and magnetoelectric materials usually rely on carrier frequencies that provide resonant coupling for remote power supply (12, 13, 18). However, compared to the carrier frequency and possible tissue penetration depth, this fundamentally produces an inverse correlation between device size. As a result, with injectable-sized devices, such devices cannot show neuronal modulation in deep brain tissue.
Since the αME generated by the signal irrelevant to the resonance coupling of the magnetoelectric material is low, we need to determine whether sending a signal to the MENP far from the resonance will generate enough output electrical signals to regulate neuronal activity. It has been shown that the electric field generated by resonant coupling with magnetoelectric materials is much higher than the threshold required for neuromodulation (18). Therefore, we assume that using magnetoelectric materials and signal conduction independent of resonance coupling, we will be able to use nanoscale materials to regulate brain activity.
We first evaluated this in neuronal cells in vitro, and measured intracellular Ca2+ as the second messenger of electrophysiological activities. As mentioned earlier, both a large DC magnetic field and a small AC magnetic field are necessary to generate a magnetoelectric output (Figure 1, I and J). We use AC and DC magnetic fields as controls to control potential side effects caused by magnetic fields, respectively. alone. We also use PENPs and MSNPs as material controls, because piezoelectric materials and magnetostrictive materials do not produce electrical output to magnetic input alone, so any increase in cell activity caused by these materials will send out signal modulation signals due to external effects . The percentages of cells in all control combinations tested showed no significant difference in the percentage of cells with Ca2+ transients, while the AC and DC magnetic stimulation of MENP significantly increased the cells with transients (Figure 2, A to C).
To support our hypothesis that the measured Ca2+ transients are caused by the electrophysiological activity of the cells, we treated the cells with drugs to independently block voltage-gated Na+ channels and voltage-gated Ca2+ channels Or the activity of extracellular Ca2+ sources. By using these drugs, the magnetically stimulated MENPs were significantly reduced in neuronal Ca2+ activity compared with the stimulation of unblocked drugs (Figure 2, D and E). This confirms the dependence of our measured Ca2+ transients on voltage-gated ion channels and extracellular Ca2+ sources, and supports the relationship between our measured Ca2+ activity and cell electrophysiological activity.
Next, an in vivo study was conducted to evaluate the feasibility of MENP-based DBS. For this goal, naive mice received bilateral injections of MENP in their subthalamic area. AC and DC magnetic fields were used to stimulate the injected MENP, and the DC field was only used as a control. The basic principle of selecting the area is that the basal ganglia and subthalamic area are the common target areas of DBS Bank (35). In addition, these aspects have been intensively studied in the field of DBS and neuromodulatory neurological diseases (36, 37). These structures are connected to the higher area (through the cortex of the thalamus) and the lower area (brain stem) through partially parallel projections and partially integrated projections. These predictions are mainly responsible for motor control and other functions such as motor learning, association function and emotion. According to the classic basal ganglia model, information flows through the basal ganglia back to the cortex in two ways, while the new model shows that parallel circuits support the classic functions of the basal ganglia and participate in the junction and marginal areas (38, 39). Therefore, the cortical basal ganglia-thalamic cortex circuit provides a useful tool to reliably study the influence of neuromodulation on a variety of behavioral functions.
Stereotactic injection was used to implant MENP into the subthalamic area. Using a dose of 100 μg per animal, we determined that the volume of MENPs can be compared with traditional rodent DBS electrodes (0.3 mm diameter). Since both traditional DBS electrodes and MENP can displace tissue during implantation, the coverage area can control potential sham surgery effects. After different concentrations of MENP were injected, the tissue dose toxicity assessment allowed us to confirm this concentration because the inflammation markers did not show significant changes compared to other tested concentrations (Figure S4, C and D). The qualitative assessment of the particle dose showed that there was no change in the volume of MENP or the tissue response to MENP within 4 weeks. Seven weeks after the injection, the injected MENP was still present at the injection site (Figure 3A).
We used an antibody against c-Fos, a widely used marker of cell viability, to evaluate the local neuronal activity in the MENP injection area (33). Quantification of stained sections showed that in animals treated with AC and DC fields, the number of c-Fos-positive cells was significantly increased in the MENP injection area compared with DC field alone (Figure 3, A to C). We also found that under AC and DC magnetic field stimulation, the volume of the tissue containing c-Fos positive cells surrounding MENP was much larger than that of DC magnetic field alone (Figure 3D). These data support our hypothesis that we can use the magnetoelectric response of MENP to magnetic stimulation to wirelessly regulate local brain activity.
To determine whether the local neuronal activity induced by MENP is sufficient to drive neuronal activity in the thalamic cortical pathway, we evaluated c-Fos protein expression in other areas of the brain. We found that after stimulation with MENPs and AC/DC magnetic fields, the expression of c-Fos protein in the motor cortex of the thalamus and the paraventricular nucleus (PV) was significantly higher than that of the DC magnetic field alone (Figure 3, E to H). We observed selective rather than global c-Fos protein expression in the brains of stimulated animals. In summary, these data support our hypothesis that the increase in c-Fos protein expression is due to the cortical hypothalamus stimulating the cortical basal ganglia-thalamic cortical circuit, rather than the non-specific global regulation of neural activity by the magnetic field.
Next, we tested these mice in a rotating tripod test and an automated CatWalk gait analysis system to determine whether the neuronal modulation induced in the thalamic cortical pathway affects the animal’s movement. During the CatWalk test, the static parameters related to gait and balance of measuring motor function were not significantly different after AC and DC stimulation of the two nanoparticle groups (Figure 3, I to K). The Rotarod test also showed that there was no significant difference in the drop latency of any nanoparticle group (Figure 3K). Although we expected no improvement in motor function, we only tested naive mice, but these results are to prove that we did not observe an adverse effect on the animal’s gait and balance due to neuromodulation via MENP. The finding that there is no general behavioral change also corresponds to our c-Fos expression finding, where we only find selective expression changes.
When analyzing the dynamic parameters of the CatWalk test (indicating the speed of the animal), we found that the behavioral parameters of MENP-treated animals (but not MSNP-treated animals) changed significantly (Figure 3, I to K). Specifically, we see an increase in speed, followed by an increase in stride length, and a decrease in duty cycle (ie, the percentage of each stride spent between gait and swing decreases Up). The aforementioned selective behavioral response is interestingly consistent with the selective c-Fos expression in PV. The current literature provides sufficient evidence that PV projects information from the brainstem and subthalamic areas to the nucleus accumbens and amygdala and cortical areas related to these subcortical areas (40). The selective activation of PV is known to produce arousal states, which can cause fear, anxiety, reward regulation, and defensive behavior (40). The expression of the asynchronous state adapts to the behavioral requirements during exercise. In this regard, alternate gaits such as walking and exploring occur at a lower movement speed, while the synchronized gait during flight occurs at a faster movement speed (41). Although the exact neuronal substrates behind them have not yet been elucidated, recent evidence suggests that low-speed and high-speed gaits originate from different midbrain regions (42). Considering that static and dynamic CatWalk gait parameters are more likely to be affected in low-speed and high-speed motions, respectively, it can be considered that dynamic gait parameters have changed mainly due to the increase in running speed.
Based on this evidence, we believe that the measured animal speed changes are due to wireless subthalamic stimulation via MENP. The combined results of c-Fos protein immunohistochemistry and animal behavior testing support the following conclusion: Magnetically stimulated MENP wirelessly modulates neurons in the deep parts of the brain, thereby affecting brain behavior. Based on these data, we suggest that MENPs can cause specific behavioral changes related to selective perturbation of the thalamic cortical circuit.
Although we find animal behavior changes related to anxiety, in the future, it will be important to evaluate the therapeutic effects of wireless modulation in animal disease models. A successful DBS would not expect healthy animals to show benefits in motor function. Therefore, in this study, we can only evaluate the movement of animals to make sure that we are not adversely affecting movement. In the future, it will be necessary to study DBS through magnetoelectric nanoelectrodes in the Parkinson’s model to measure the benefits of motor function.
A key finding of this work is that the resonant coupling of the neural device’s unrelated remote power supply can generate enough electrical activity to regulate brain activity. This eliminates the relationship between device size and potential power supply depth, allowing nanoscale materials to modulate deep brain tissue. The model of transmitting carrier signals through tissues to magnetoelectric devices will benefit future device designs and clarify the limitations of tissue penetration to human-scale deep brain targets. In addition, future work will be necessary to understand how MENP propagates the carrier frequency to the stimulation signal received by neurons, because the timing of stimulation is the key to the effect of DBS treatment (43, 44). The exact mechanism of neuronal regulation is still an open question. Therefore, future work will be necessary to learn more about the various input parameters that can be modulated (for example, nanoparticle concentration, magnetic stimulation amplitude, stimulation frequency and duration). Since it has been shown that the electric field gradient along the axon is the key determinant of activation (45, 46), we hypothesized that the mechanism of action may be related to the high gradient along very small nanoparticles. However, the nanoscale field gradients associated with neuronal activation have never been evaluated. We need to better understand the nano-scale magnetoelectric response and how MENP and its field interact to further speculate. Regarding the lifespan of MENPs, although we know that they only need to stay at the injection site and can regulate neurons for up to 7 weeks after injection, the transformation of this technology to patients still requires long-term compatibility, immune response, and nanoparticle functions. . However, this work represents an important proof of concept for remotely powering nanoscale neural devices.
The results of this paper demonstrate the potential of magnetoelectric materials as nanoelectrodes for radio modulation of deep brain targets. We have shown that we can stimulate MENP with a magnetic field to remotely generate MENP electrical polarization. We have shown evidence that MENP’s non-resonant magnetism locally modulates neuronal activity in vitro and in vivo. We have also shown that this regulation is sufficient to change the behavior of animals and regulate other areas of the cortical basal ganglia-thalamic cortical circuit. Future work will be the key to optimizing magnetoelectric-based neural devices and understanding the capabilities and limitations of the technology. Magnetoelectric nanoelectrodes have shown promise for new technologies in wireless nerve devices.
The purpose of this study is to evaluate the potential of MENPs to wirelessly modulate neuronal activity through the magnetoelectric response to an applied magnetic field. By applying a magnetic field to MENPs (and control nanoparticles) and measuring (i) their output electrical signal conduction, (ii) their ability to regulate neuronal cell activity in culture, (iii) they regulate brain activity in mice Ability to achieve this goal. , And (iv) the effect of this modulation on mouse behavior. Although a continuous and measurable magnetoelectric effect can only be expected when a large DC magnetic field covered with a small AC magnetic field is applied, AC and DC magnetic fields are used alone as experimental controls. All experiments in this study were conducted through controlled laboratory experiments. The sample size of each experiment can be determined independently without formal power analysis. Where appropriate, references used to determine sample size are cited. The method of each experiment and the sample size are listed in the title of the graph showing the results. The following method also describes how the sample size corresponds to sampling and experiment repetition. Endpoints vary from experiment to experiment and are listed below. Exclusion criteria for animal safety are listed in the instructions for toxicity analysis. However, due to these criteria, no animals were excluded from the study. Animal behavior testing is done by experimenters and data analysts who do not know the identity of animals. Listed below are information about the cell lines, animals and antibodies used. The following lists the ethical supervision and approval of animal research.
MENP is synthesized in a manner similar to Corral-Flores et al. (27). Suspend CoFe2O4 nanoparticles (30 nm; Sigma-Aldrich) in deionized water (dH2O) at a concentration of 10 mg/ml to 80°C while stirring. 30% by weight of oleic acid relative to CoFe 2 O 4 was added to the suspension, and the temperature was increased to 90°C for 30 minutes, and then lowered to 60°C. The volume of octane and dH2O was added to the suspension in a ratio of 1:1 to separate the oleic acid-coated CoFe2O4 particles into the organic layer. Then the organic layer was washed 3 times with dH 2 O. Dissolve barium acetate and titanium butoxide in glacial acetic acid containing stearic acid (the final concentration is 0.01%) so that the final molar ratio of BaTiO3 to CoFe2O4 is 1:3. The solution was stirred and heated to 90°C, and CoFe 2 O 4 solution and 2-methoxyethanol were added to the final volume concentration of 30%. The solution was dried, calcined at 700°C for 2 hours, and then ground with a mortar and pestle. MSNP control particles are unmodified, commercially available CoFe2O4 nanoparticles used as MENP cores. The PENP control particles are commercially available BaTiO 3 nanoparticles (50 nm; Sigma-Aldrich). In order to select particles with better colloidal stability, all the nanoparticles were suspended in dH2O and centrifuged at 10g for 1 minute, and the particles in the supernatant were retained for further experiments.
X-ray diffraction (XRD) analysis of MENPs was performed on the Bruker D8 Advance powder diffractometer, using the Bragg-Brentano beam path, Cu radiation generated at 40 kV / 40 mA. Use 0.5° divergence slits, 2° and 4° anti-scatter slits, and Soller slits. The VÅNTEC-1 one-dimensional (1D) detector is used to receive the output beam. Use the International Diffraction Data Center database to identify peaks. MathWorks MATLAB software is used for baseline correction of the spectrum through the msbackadj function.
MENP element analysis was performed by inductively coupled plasma (ICP)-optical emission spectrometry using Spectro Ciros spectrometer (Kleve, Germany). MENP was first dissolved in an aqueous solution of 3% HNO3 and 1% HF, and then the sample was loaded into the spectrometer. Use Spectro ICP Analyzer software to analyze the data to detect Ba, Ti, Co and Fe spectra. The data is expressed as the average ± SD of each element measured in BaTiO3 and CoFe2O4.
MENP is prepared for transmission electron microscopy (TEM) analysis by casting an aqueous suspension droplet onto a C-coated Cu TEM grid and air drying. TEM and TEM electron energy loss spectroscopy (EELS) images were obtained using Zeiss subelectron voltammography. The data was acquired in TEM mode at 200 kV. For EELS, we acquired energy filtered TEM spectral images from 30 to 120 eV with a step size of 3 eV and a bin of 4x. After data collection, the EELS signals from Ba (N4, 5 edge, 90 eV) and Fe (M2, 3 edge, 54 eV) are extracted and used in the element map.
The hydrodynamic diameter and zeta potential of MENP were measured by DLS using Wyatt Mobius DLS instrument, and analyzed by Wyatt DYNAMICS software. During the measurement, MENPs were diluted to a concentration of 100 μg/ml in our cell culture differentiation medium (see below) or aCSF solution (47). Data from three independent experiments were analyzed.
A Microsense EZ Vibrating Sample Magnetometer (VSM) was used to measure the magnetic properties of MENP. MENPs were measured with 6 mg MENPs powder and fixed on the VSM probe with wax. The magnetization is measured in the range of -1 to 1 T, and the magnetization is measured in the range of 205 to 235 mT.
In order to perform ME measurement of pellets, 0.65 g MENPs were mechanically pressed into pellets with a diameter of 8 mm using a pressure of 6 tons/cm 2, and then sintered at 1150° C. for 12 hours. The MSNP pellets were prepared in the same way, but only CoFe 2 O 4 nanoparticles were used. The round surface of the particles is coated with conductive silver glue to fix the copper plate (Figure S1A). The pellets were heated to 140°C, polarized at a thickness of 1 kV/mm for 5 minutes, and then allowed to cool to room temperature while maintaining the applied voltage. The pellets are then connected to the charge amplifier. The particle and charge amplifier are encapsulated in a Faraday shield and connected externally to a lock-in amplifier for voltage measurement (Figure S1, B to H).
In order to measure the magnetoelectric response of the ball, a charge amplifier is used to eliminate the influence of stray capacitance on the piezoelectric charge measurement. The battery-powered amplifier is built on a standard FR4 printed circuit board and placed in a Faraday shield. The charge amplifier uses an operational amplifier circuit based on Texas Instruments OPA340 (Figure S1, F to H). The amplifier has a high-pass characteristic of −3-dB frequency of -3 Hz, and the calculated gain of the circuit in the passband is 200 mV/pC.
The Microsense EZ VSM is used as a DC magnetic field source and has been modified to accommodate another smaller Helmholtz coil. It is powered by a signal generator (35 to 385 Hz sine wave) connected to a linear voltage amplifier (Hewlett Packard) to provide current to the smaller coil, thereby generating an AC magnetic field in the sample plane. Position the particles so that the AC and DC magnetic fields are parallel to the central axis of the particles (Figure S1E). Before the experiment, a Gauss meter was used to measure the AC magnetic field strength. Before all measurements, the pellets were demagnetized.
SH-SY5Y cells were purchased from Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ) (American Type Culture Collection CRL-2266). Maintain the culture at 37°C, 5.0% CO2 with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin in Dulbecco’s modified Eagle medium (DMEM)/F12 (Gibco). The medium is changed every 3 to 4 days. Before performing experimental plating, cover the wells with Ca2 + / Mg2 + laminin (5 μg/ml) in phosphate buffered saline (PBS) for 1 hour at 37°C. For Ca2+ signal transduction experiments, cells were seeded on four-well IBIDIμ-slides treated with cell culture at a concentration of 20,000/cm2. For toxicity analysis, cells were seeded on a 96-well plate with cell culture at a concentration of 20,000/cm2. Before all experiments, the experimental cultures were differentiated in DMEM/F12 medium containing 1% FBS, 1% penicillin/streptomycin, and 10 μM retinoic acid (Sigma-Aldrich) for 4 days.
MENP was suspended in the experimental cell culture medium at a concentration of 0, 50, 100, 200, or 300 μg/ml, and then added to the cells. At 24 hours after MENP administration, the toxicity was assessed by CyQUANT LDH analysis kit (Thermo Fisher Scientific) and CellTiter 96 AQueous One MTS analysis. Use BioTek Synergy 2 microplate reader (Figure S2) to read the measurement results. Each experiment was tested in four wells, and the average of these values was recorded to provide a single data point. Analyze data from three independent experiments.
A magnetic stimulation device was designed to fit the Zeiss Axio Observer A1 microscope, and a four-hole IBIDIμ-slide was fixed (Figure S3). Three permanent NdFeB magnets (N42, diameter 6 cm, height 5 mm; super magnet) on both sides of the cell provide a DC magnetic field to generate a magnetic field of 225 mT in the center of the cell culture hole. The electromagnetic coil is used to provide an AC magnetic field along the same axis. The AC signal is generated by National Instruments DAQ USB X series equipment, controlled by LabVIEW software, and amplified by a Class D audio amplifier. For all experiments using AC magnetic stimulation, the AC field component is a 140 Hz 6 mT sine wave applied in a time window of 10 to 30 s during the time-lapse recording. The strength of AC and DC magnetic fields are verified with a magnetometer.
Load 1μMFluo4-AM dye (Thermo Fisher Scientific) in the cells in the Live Cell Imaging Solution (LCIS; Invitrogen) at 37°C for 30 minutes. An experimental suspension without NP, MENP, PENP or MSNP was prepared at 100 μg/ml in LCIS. After loading Fluo4, the cells were washed 3 times with LCIS, and then the particle suspension was added. The cells with particles were incubated at 37°C for 20 minutes to deesterify Fluo4, and then moved to a Zeiss Axio Observer A1 microscope equipped with an in vitro coil system. For experiments using inhibitory drugs, Fluo4 was loaded as described above, and the drug with MENP was added to LCIS after washing. For EGTA, PBS was used instead of LCIS, and PBS was added in the Fluo4 loading step. The concentration of TTX added was 100 nM, the concentration of CdCl2 (Cd2 +) added was 100 μM, and the concentration of EGTA added was 5 mM, which was previously determined to be an inhibitory but non-toxic concentration (14).
Use a 470 nm LED with a 484/25 nm excitation filter to excite Fluo4 and observe through a 519/30 nm emission filter. The time-lapse images were taken at ×10 magnification under 50 milliseconds of illumination, 240 s every 1 s, and recorded with Zeiss Axiocam 503 SLR camera (2.8 megapixels). Each group collects data from three to six independent experiments.
Use ImageJ software to analyze time-lapse video. In short, stack the first 10 images of each time interval into a single image for region of interest (ROI) selection. After performing brightness normalization, blurring, background subtraction, and threshold settings, use the “Analyze Particles” function (all settings remain the same for all time periods) to select an ROI from the image. Then overlay these ROIs on the completely unmodified time-lapse sequence, and record the average gray value of each ROI in each frame. These values are then used to calculate the Ca2 + transient amplitude as ΔF/Fo. Using MathWorks MATLAB software, using linear baseline correction and peak finder functions, you can calculate cells that can actively display Ca2 + transients. The image (A and D) in Figure 2 is generated by creating the maximum Z stack of the entire video.
The experiment was conducted on 65 male naive mice (C57Bl/6J; Jackson laboratory). In the opposite 12-hour day/night cycle (light on at 7 pm), the mice are placed under controlled conditions (21°±2°C, humidity 40% to 60%) for social rearing until surgery until. The mice were given food and water ad libitum. At the time of surgery, the mice were 3 months old. Experiments were carried out in accordance with the Directive 2010/63/EU concerning animal experiments and reached an agreement with the Animal Experiment and Ethics Committee of Maastricht University in Maastricht, the Netherlands.
Half an hour before surgery, buprenorphine (0.1 mg/kg) was injected subcutaneously as an analgesic. Inhalation anesthesia was induced with isoflurane (Abbot Laboratories, Maidenhead, UK) and maintained at 4% and 1.5% to 3%, respectively. After sufficient induction of anesthesia, the mouse was placed in a small animal stereotaxic frame (Stoelting, Dublin, Ireland) and fixed with an ear stick with zygoma earmuffs (Kopf, Los Angeles, USA) and a mouse gas anesthesia head frame (Stoelting). Dublin, Ireland). To maintain body temperature at 37°C throughout the procedure, place the mouse on the thermostat pad. Use eye lubricant to prevent dry eyes. 1% lidocaine (Streuli Pharma, Uznach, Switzerland) was injected subcutaneously on the incision side for local anesthesia.
Continuously drill burrs above the subthalamic area [anterior-posterior (AP): −2.06 mm, inside-outside (ML): ±1.50 mm, dorsal-abdominal (DV): −4.50], a total of 2μl with the microinjection device Nanoject II (Drummond Scientific) injection of MENPs or MSNPs. The infusion rate is 100 nl/min. After the injection, the syringe needle stays in the brain for another 10 minutes before being slowly withdrawn.
Use a customized coil system for all in-vivo magnetic stimulation, which will allow the mice to move freely during the experiment. When designing an animal experiment device, a DC magnetic field of 220 mT and a 140 Hz AC magnetic field of 6 mT should be provided along the same axis in the center of the animal room. The image and design of the in-body coil system is shown in Figure 2. S5. The structure was 3D printed with acrylonitrile-butadiene-styrene using uPrint SE Plus 3D printer. On each side of the animal chamber, six NdFeB disk magnets (N42, diameter 6 cm, height 5 mm; supermagnet) provide a DC magnetic field. For safety, the permanent magnet is covered by a protective cover, and the bottom of the animal support is 3D printed with a fixing option to improve durability. The AC magnetic field is provided by two coils on both sides of the animal room. A 1mm thick copper wire is wound on a 3D printed plastic coil frame with 360 turns per turn. The corresponding coil pair resistance is 4.94 ohm, and the coil pair inductance is 24.5 mH. Use Voltcraft 8210 signal generator to provide 140 Hz sine wave, and use QSC-GX7 power amplifier to amplify it. Then connect them to the AC coil. The strength of AC and DC magnetic fields are verified with a magnetometer. For all AC and DC stimulation experiments, the mice were stimulated for 180 s with the coil turned on. For the stimulation experiment with only direct current, the mouse was placed in the animal room for 180 seconds and the coil was kept closed.
The first stage: toxicity assessment. We first adjusted the optimal concentration of MENP. Three doses were tested, including 25, 50, and 100 mg/ml. Mice were randomly assigned to 25, 50, or 100 mg/ml test groups (n = 8) and received stereotactic injection of MENPs (Figure S4A). Monitor the animals for signs of subdural or epidural hemorrhage, neurological symptoms of the injection, welfare (weight, responsiveness and water consumption) and discomfort/pain. Due to non-compliance with these standards, no animals were eliminated from the experiment. On the fourteenth day after the operation, the mice were euthanized for brain immunohistochemistry (IHC) analysis as described below. Five brains were randomly selected for IHC. As the tissue broke during the treatment, the section belonging to one mouse was excluded. The brain slices were processed with antibodies against astrocytes and microglia (Figure S4, B and C). Another series of brain slices were stained using standard hematoxylin and eosin (H&E) to assess tissue damage at the injection site (Figure S4D).
The second stage: the persistence of nanoparticles at the injection site and the expression of c-Fos protein. The mice were randomly divided into three test groups (n = 8) and received stereotactic injection of MENPs (100 mg/ml). We tested the elution of MENP at different time points (including 48 hours, 2 weeks, and 4 weeks) (Figure S4, E and F). At the end of each time point, the mice were intracardiac perfused, and the brains were removed for IHC and H&E analysis. To assess the expression of c-Fos protein, half of the mice in each group received 180 s magnetic stimulation two hours before perfusion. As a control group, the other half of the mice were placed in a coil with no current flowing through the coil, and they were only exposed to the DC magnetic field of a permanent magnet. Because the tissue was broken or damaged during the tissue processing, the sections of the three mice were excluded and IHC analysis was not possible. In previous work, the IHC quantification of at least 5 subjects in each group was sufficient for effective analysis (48, 49).
The third stage: behavioral testing. To assess the effect of MENP-induced neuronal regulation on brain tissue, two groups of animals were tested and behavioral responses were evaluated. The mice were randomly divided into two groups and received stereotactic injection of MENPs (n = 9) or MSNPs (n = 8; 100 mg/ml). After the recovery period of 1 week after the operation, the animals were stimulated in a magnetic field and behavioral tests were performed. Specifically, the animal is stimulated with AC and DC magnetic fields (in the in-vivo coil system with the coil open) or only with the DC magnetic field (in the in-vivo coil system with the coil closed). After stimulation with AC and DC magnetic fields and only with DC magnetic field stimulation, the behavioral parameters measured between the same mice were compared (Figure S4G).
CatWalk test. An automated gait analysis system CatWalk XT (Noldus 7.1, Wageningen, Netherlands) will evaluate athletic behavior. CatWalk includes an enclosed walkway with glass panels and a high-speed video camera (Figure 3I). Use CatWalk analysis software to evaluate and record gait performance. Before testing each subject, the glass plates were washed and dried to minimize the spread of olfactory cues and to prevent the animals from stopping to smell or explore something while running. Usually, a successful test record consists of an average of five uninterrupted runs, with comparable running speeds, with a maximum change of 30%. In the process of behavioral testing and data analysis, experimenters and data analysts are blind to the identity of the animals. The following 20 static and dynamic parameters for evaluating individual paw function and gait patterns are analyzed: posture, average strength, printing area, printing length, printing width, average swing, swing speed, stride, maximum contact maximum intensity, maximum intensity , Minimum intensity, step, duty cycle, regularity index, support basis for forelimbs, support basis for hind limbs, support for three limbs, speed and rhythm.
Rotarod test. An accelerated rotating tripod with a grooved rotating beam (3 cm) raised above the platform (Ugo Basile Biological Research Apparatus, Italy, model 47650, model 47650) was used to measure coordination. Record the waiting time for falling from the rotating rod. The data is expressed as the average of three experiments. The mice received four 300-s tests per day for three consecutive days (days 1 to 3), with an interval of about 15 minutes. The mice are forced to run on the drum, the speed of which starts from 4 rpm and accelerates to 40 rpm in 300 s. Within 300 s of the entire task, the mouse remaining on the beam was removed from the rotating tripod and awarded the highest score. In the process of behavioral testing and data analysis, experimenters and data analysts are blind to the identity of the animals.
The mice were deeply anesthetized with pentobarbital and perfused with tyrosine buffer per heart, and then perfused with ice-cold 4% paraformaldehyde fixative in 0.1 M phosphate buffer. The brain was extracted from the skull, fixed in 4% paraformaldehyde overnight, and then immersed in sucrose for cryoprotection (placed in 20% sucrose at 5°C for 24 hours). Cut coronal brain slices (20μm) on a cryostat and store at -80°C.
The tissue sections were incubated with polyclonal rabbit antibody (1:1000; Santa Cruz Biotechnology Inc.; sc-253), GFAP (1:1000; Dako; Z-033429) or Iba-1 (1) against c-Fos protein Overnight: 1000; Wako; 016-26461). c-Fos IHC uses a biotinylated donkey anti-rabbit secondary antibody (1:400; Jackson ImmunoResearch Laboratories Inc.; 711065152) and avidin-biotin peroxidase complex (1:800; Elite ABC- kit, Vector Laboratories; PK-6100). The staining was observed through the combination of 3,3′-diaminobenzidine and NiCl2 fortification. GFAP and Iba-1 were visualized using immunofluorescence and donkey anti-rabbit Alexa 488 (1:100; Invitrogen; A-21206).
The stained photos of the motor cortex and thalamus at the three rod-shaped coccyx anatomical levels (AP: -0.58, -0.94, and -1.22) from the front three were taken at 10x magnification. We used Cell P software (Olympus Soft Imaging Solutions, Münster, Germany) connected to an Olympus DP70 digital camera connected to an Olympus AX 70 microscope (Olympus, Zoeterwoude, Netherlands). In the image of the region of interest, use ImageJ software [version 1.52; National Institutes of Health (NIH), Bethesda]. Manually count the cells immunopositive for c-Fos, and correct the average cell number of the surface area, and express it as the number of cells per square millimeter. When the intensity of cell staining is significantly higher than the surrounding background, the cell is considered positive. The average of the three parts was used for the statistical analysis of each subject. For the subthalamic area (infusion site), a digital photo was taken at an anatomical anterior reg (-2.06), and all c-Fos positive cells within 1 mm2 of the injection site were counted.
The stained photographs of the motor cortex and thalamus slices taken from the three rod-shaped coccyx anatomical levels of the anterior reg (AP: -1.70, -2.06 and -2.30) are magnified 10 times. We used Cell P software (Olympus Soft Imaging Solutions, Münster, Germany) connected to an Olympus DP70 digital camera connected to an Olympus AX 70 microscope (Olympus, Zoeterwoude, Netherlands). In the image of the region of interest, the fluorescence density was measured using ImageJ software (version 1.52; NIH, Bethesda, USA). The average of the three parts was used for the statistical analysis of each subject.
Three-dimensional volume measurement was performed at the MENP injection site in the subthalamic area and the c-Fos expression area around it. We used a stereo computer microscope system (Stereo Investigator, Microbrightfield Bioscience, version 10, Williston, VT, USA). In short, after the areas stained with c-Fos on the video image displayed on the monitor delineate these areas, the volume is calculated according to the principle of Cavalieri (Cavalieri, 1635) (50). Calculate the volume of each part by multiplying the surface by the section thickness and the number of slices in each series. Finally, all these parts are added together to calculate the total volume of the MENP injection site and the c-Fos expression area.
Unless otherwise stated, the data are presented as individual values, and the bar graph represents the mean ± SD. Linear regression was used to determine the dependence of the AC magnetic field strength and frequency on the MENP voltage output, and the coefficient of determination was expressed as R2. A one-way analysis of variance (ANOVA) and Tukey’s later test were used to compare all groups to analyze the Ca2+ transient activity in vitro and c-Fos expression in vivo (Figure S4). Using untreated cells as a control, the in vitro analysis of Ca2+ signaling using inhibitors was analyzed using Dunnett’s post hoc one-way ANOVA analysis. The unpaired t test was used to analyze the expression of c-Fos protein in brain tissue. The paired t test was used to analyze the behavioral parameter changes of the same mice after being stimulated by DC magnetic fields or AC and DC magnetic fields. In the process of behavioral testing, behavioral data analysis and IHC slice quantification, experimenters and data analysts are blind to the identity of the animals. In all cases, P <0.05 was considered statistically significant.
For supplementary materials for this article, please see http://advances.sciencemag.org/cgi/content/full/7/3/eabc4189/DC1
This is an open access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License, which allows the use, distribution and reproduction in any medium, as long as the final use is not for commercial gain and the premise is that the original work is correct. Reference.
Note: We only ask you to provide your email address so that the person you recommend to the page knows that you want them to see the email and that it is not spam. We will not capture any email addresses.
This question is used to test whether you are a visitor and prevent automatic spam submission.
KL Kozielski, A. Jahanshahi, HB Gilbert, Y. Yu, Ö. Irene Francisco Francisco Alosami Y Temel Siti
The wireless power supply of the magnetoelectric nanoelectrodes can be used for deep brain stimulation of freely moving and non-transgenic mice.
KL Kozielski, A. Jahanshahi, HB Gilbert, Y. Yu, Ö. Irene Francisco Francisco Alosami Y Temel Siti
The wireless power supply of the magnetoelectric nanoelectrodes can be used for deep brain stimulation of freely moving and non-transgenic mice.
©2021 American Association for the Advancement of Science. all rights reserved. AAAS is a partner of HINARI, AGORA, OARE, CHORUS, CLOCKSS, CrossRef and COUNTER. ScienceAdvances ISSN 2375-2548.
Post time: Jan-15-2021