Wonjune Kim

BackgroundMagnetic control of cell activity has applications from non-invasive neurostimulation to remote activation of cell-based therapies. To date, no magnetically gated channel has been identified, limiting magnetic approaches to strong alternating fields (kHz to MHz) or using co-factors such as nanoparticles.ObjectiveWe aimed to determine whether slowly varying magnetic fields can elicit a response in excitable cells without exogenous co-factors, characterize the intrinsic magnetic sensitivity and response time to stimulation, and identify the mechanism of action.MethodSpiking HEK cells were exposed to a magnetic field reaching 500 mT within 200 ms using a permanent magnet. Calcium imaging was used to quantify responses across >7,500 micropatterned independent cell clusters (150-300 cells each). Pharmacological blockers and co-culture with MIN6 beta-pancreatic cells were used to probe mechanism and functional output.ResultsThe response is rare, approximately 1 in 50 cell clusters, but is fast and reproducible, generating rapid, spike-like calcium transients within 200 ms of magnetic field stimulation consistent with electrically driven depolarization.However, with a magnetic field 10,000 times slower than in transcranial magnetic stimulation (TMS) and an induced electric field an order of magnitude lower than necessary for membrane depolarisation, we explored alternative biophysical mechanisms.Gadolinium and pharmacological inhibition of TRP channels attenuated the response, while Piezo channel inhibition had no effect, implicating mechanosensitive TRP channels in the magnetic response. When cells form gap junctions, the magnetically-induced calcium transient can propagate to neighboring cells, enabling tissue-level responses. Co-culture of spiking HEK cells with beta-pancreatic MIN6 cells showed that this co-culture responds to magnetic fields by increasing insulin production.ConclusionOur calculations suggest that induced electric fields, eddy currents, heating, and ROS-mediated pathways are quantitatively or kinetically unable to explain the observed response. Instead, our calculation suggests that this response depends on mechanoreception pathways activated by the magnetic torque of TRP-associated lipid rafts, a hypothesis supported by our data. Together, these results open a path towards the fast magnetic control of biological activity without needing synthetic co-factors. By better understanding this mechanism and enriching for magneto-sensitivity it may be possible to adapt this approach to the rapidly expanding tool kit for wireless cell activity regulation.

Magnetic control of cell activity has applications ranging from non-invasive neurostimulation to remote activation of cell-based therapies. Unlike other methods of regulating cell activity like heat and light, which are based on known receptors or proteins, no magnetically gated channel has been identified to date. As a result, effective approaches for magnetic control of cell activity are based on strong alternating magnetic fields able to induce electric fields or materials that convert magnetic energy into electrical, thermal, or mechanical energy to stimulate cells. In our investigations of magnetic cell responses, we found that a spiking HEK cell line with no other co-factors responds to a magnetic field that reaches a maximum of 500 mT within 200 ms using a permanent magnet. The response is rare, approximately 1 in 50 cells, but is fast and reproducible, generating an action potential within 200 ms of magnetic field stimulation. The magnetic field stimulation is over 10,000 times slower than the magnetic fields used in transcranial magnetic stimulation (TMS) and the induced electric field is more than an order of magnitude lower than necessary for neuromodulation, suggesting that induced electric currents do not drive the cell response. Instead, our calculation suggests that this response depends on mechanoreception pathways activated by the magnetic torque of TRP-associated lipid rafts. Despite the relatively rare response to magnetic stimulation, when cells form gap junctions, the magnetic stimulation can propagate to nearby cells, causing tissue-level responses. As an example, we co-cultured spiking HEK cells with beta-pancreatic MIN6 cells and found that this co-culture responds to magnetic fields by increasing insulin production. Together, these results point toward a method for the magnetic control of biological activity without the need for a material co-factor such as synthetic nanoparticles. By better understanding this mechanism and enriching for magneto-sensitivity it may be possible to adapt this approach to the rapidly expanding tool kit for wireless cell activity regulation.

To maximize the capabilities of minimally invasive implantable bioelectronic devices, we must deliver large amounts of power to small implants; however, as devices are made smaller, it becomes more difficult to transfer large amounts of power without a wired connection. Indeed, recent work has explored creative wireless power transfer (WPT) approaches to maximize power density [the amount of power transferred divided by receiver footprint area (length × width)]. Here, we analyzed a model for WPT using magnetoelectric (ME) materials that convert an alternating magnetic field into an alternating voltage. With this model, we identify the parameters that impact WPT efficiency and optimize the power density. We find that improvements in adhesion between the laminated ME layers, clamping, and selection of material thicknesses lead to a power density of 3.1 mW/mm2, which is over four times larger than previously reported for mm-sized wireless bioelectronic implants at a depth of 1 cm or more in tissue. This improved power density allows us to deliver 31 and 56 mW to 10 and 27-mm2 ME receivers, respectively. This total power delivery is over five times larger than similarly sized bioelectronic devices powered by radiofrequency electromagnetic waves, inductive coupling, ultrasound, light, capacitive coupling, or previously reported magnetoelectrics. This increased power density opens the door to more power-intensive bioelectronic applications that have previously been inaccessible using mm-sized battery-free devices.

To maximize the capabilities of minimally invasive implantable bioelectronic devices, we must deliver large amounts of power to small implants; however, as devices are made smaller, it becomes more difficult to transfer large amounts of power without a wired connection. Indeed, recent work has explored creative wireless power transfer (WPT) approaches to maximize power density (the amount of power transferred divided by receiver footprint area (length × width)). Here, we analyzed a model for WPT using magnetoelectric (ME) materials that convert an alternating magnetic field into an alternating voltage. With this model, we identify the parameters that impact WPT efficiency and optimize the power density. We find that improvements in adhesion between the laminated ME layers, clamping, and selection of material thicknesses lead to a power density of 3.1 mW/mm ² , which is over 4 times larger than previously reported for mm-sized wireless bioelectronic implants at a depth of 1 cm or more in tissue. This improved power density allows us to deliver 31 mW and 56 mW to 10-mm ² and 27-mm ² ME receivers, respectively. This total power delivery is over 5 times larger than similarly sized bioelectronic devices powered by radiofrequency electromagnetic waves, inductive coupling, ultrasound, light, capacitive coupling, or previously reported magnetoelectrics. This increased power density opens the door to more power-intensive bioelectronic applications that have previously been inaccessible using mm-sized battery-free devices.

Extracellular vesicles (EVs) have emerged as novel biomarkers and therapeutic material. However, the small size (~200 nm) of EVs makes efficient separation challenging. Here, a physical/chemical stress-free separation of EVs based on diffusion through a nanoporous membrane chip is presented. A polycarbonate membrane with 200 nm pores, positioned between two chambers, functions as the size-selective filter. Using the chip, EVs from cell culture media and human serum were separated. The separated EVs were analyzed by nanoparticle tracking analysis (NTA), scanning electron microscopy, and immunoblotting. The experimental results proved the selective separation of EVs in cell culture media and human serum. Moreover, the diffusion-based separation showed a high yield of EVs in human serum compared to ultracentrifuge-based separation. The EV recovery rate analyzed from NTA data was 42% for cell culture media samples. We expect the developed method to be a potential tool for EV separation for diagnosis and therapy because it does not require complicated processes such as immune, chemical reaction, and external force and is scalable by increasing the nanoporous membrane size.