Nanoporous electroporation needle for localized intracellular delivery in deep tissues

Abstract The exogenous control of intracellular drug delivery has been shown to improve the overall efficacy of therapies by reducing nonspecific off‐target toxicity. However, achieving a precise on‐demand dosage of a drug in deep tissues with minimal damage is still a challenge. In this study, we report an electric‐pulse‐driven nanopore‐electroporation (nEP) system for the localized intracellular delivery of a model agent in deep tissues. Compared with conventional bulk electroporation, in vitro nEP achieved better transfection efficiency (>60%) with a high cell recovery rate (>95%) under a nontoxic low electroporation condition (40 V). Furthermore, in vivo nEP using a nanopore needle electrode with a side drug‐releasing compartment offered better control over the dosage release, time, and location of propidium iodide, which was used as a model agent for intracellular delivery. In a pilot study using experimental animals, the nEP system exhibited two times higher transfection efficiency of propidium iodide in the thigh muscle tissue, while minimizing tissue damage (<20%) compared to that of bulk electroporation. This tissue‐penetrating nEP platform can provide localized, safe, and effective intracellular delivery of diverse therapeutics into deep tissues in a controlled manner.

immunogenicity. 14 In contrast, chemical methods (involving formation of nanosized complexes with lipids or polymers) are often limited by toxicity and quality control. 8 In addition, the delivery method for injecting chemical complexes can suffer from low delivery efficiency owing to endosomal entrapment that occurs during endocytosis. [15][16][17] Alternatively, physical approaches, including microinjection, biolistics, jet injection, ultrasound, and electroporation, have also been utilized to transiently modify cell membrane permeability and have gained considerable attention owing to their technical simplicity and operational flexibility. 18,19 Among these, electroporation (EP) has received significant attention because of its high efficacy in terms of the intracellular delivery of exogenous therapeutic agents for a broad range of cell types as well as its facile integration into medical devices and equipment. 20 The application of an electric field gradient (≈1 kV/cm) to cells typically generates transient membrane pores and increases the permeability of the cell membrane, thus allowing direct intracellular delivery of cell-impermeable materials. [21][22][23] Due to its rapid and immediate delivery characteristics, conventional bulk EP has been widely used in biomedical applications ranging from cellular dedifferentiation to electrochemotherapy. [24][25][26][27] However, bulk EP using plate or needle electrodes suffers from significant cell damage and random, low transfection efficiency, which may be attributed to the nonuniform generation of high-voltage pulses between the electrodes. 28,29 Alternatively, nanopore-electroporation (nEP) has been developed to achieve efficient intracellular drug delivery under lowvoltage conditions with improved cell viability. [30][31][32][33][34][35][36] By inducing a focused electric field through the nanopore, perforation can occur only in a small fraction of the cell membrane, thereby improving cell viability. In addition, since charged molecules can be directly injected into cells by nEP, the intracellular delivery efficiency of nEP is more than 10-fold higher than that of bulk EP systems. 31 Upon integration into microchannel devices, the nEP system could be used for singlecell transfection with high efficiency and precise delivery. 37 Compared with bulk EP, nEP generates localized and well-defined pores on the cell membrane to deliver charged gene-editing agents, such as plasmids, thus yielding high cell viability and transfection efficiency. 29 Notably, tissue nanotransfection chips fabricated by multiple semiconductor lithography techniques demonstrated promising results for in vivo tissue reprogramming following pulsed nEP. 30,38,39 However, the in vivo application of nEP systems with 2D planar designs is limited to the body surface (mainly the skin surface), as nEP is effective on cells that are in direct contact with the nanopores. Furthermore, for transdermal cargo delivery by nEP, pretreatments, such as microneedling 40 or exfoliation procedure 30,38,39 are required for having a F I G U R E 1 Schematic illustration of the in vivo nEP system. That consisting of a conventional metal needle and nanopore needle electrode with a side drug-releasing compartment, applied into the thigh muscle of a mouse. Application of an electric field through the nanopores of the needle electrode generates transient pores on the cell membrane. The charged molecules accelerated through electrical stimulation are released through the nanopores, thereby enabling intracellular delivery of charged target molecules through the temporarily perforated cell membrane.
closer interface between the planar nEP electrode and deeper layers of the skin.
A suitable platform to overcome the limitations of existing nEPbased intracellular delivery of functional molecules should (i) effectively permeabilize the cell membrane under a nontoxic working voltage, (ii) be applicable for deep tissues without the pretreatment of the target tissues, (iii) offer reliable and pulse-responsive intracellular delivery with minimal tissue damage, (iv) be a simple and scalable format for application in a large area, and (v) be able to integrate with the current EP system for easy translation.
Given this background, in this study, we attempted to design an in vivo nEP platform based on a nanopore needle electrode fabricated by the merging of micro-and nano-manufacturing technologies for low tissue damage and high transfection efficiency ( Figure 1). This in vivo nEP system controlled the release amount, release time, and release location of the target molecules in a better way for intracellular delivery upon electrical stimulation. We first developed an in vitro testing device for nEP to identify the optimal electric-field conditions for effective intracellular delivery. We then engineered the nEP system with 2-needle array electrodes consisting of a metal needle and a nanopore needle electrode with a side drug-releasing compartment for in vivo deep tissue application. This nanopore electrode can effectively focus the electric field, which can only be applied to cells around the nanopores. The constructed nEP system was investigated to assess the efficacy of intracellular delivery to deep tissues. The safety and efficiency of this nEP system were verified by applying it to the thigh muscle of a mouse, and then examining the extent of tissue damage as well as the transfected area with a cell-impermeable dye as a model agent.

| In vitro testing device for nEP
We first engineered an in vitro testing device to verify the principle of nEP and screen the electrical stimulation conditions before exploring the usage of nEP in in vivo experiments ( Figure 2). In principle, nEP using DC voltage pulses was selectively applied through a nanopore membrane only to cells in contact with the nanopores. Thus, (1) the area in which the cells are exposed to the electric field can be limited to the size of the nanopore, and (2) most of the DC voltage is applied only to the nanoporous membrane, which has a relatively higher resistance than the EP medium. In the case of conventional bulk EP, there is a clear trade-off between the EP efficiency and cell survival rate; when the voltage is increased to improve the efficiency, the cell survival rate markedly decreases. In contrast, nEP can decouple this relationship by adjusting the density and pore diameter of the nanopores and controlling the area of the electric field applied to single cells. In other words, even if a strong electric field is applied, cell damage can be minimized by reducing the area in which the field is applied to a single cell. In addition, electrolysis, a chronic issue of conventional bulk EP, can be considerably suppressed by increasing the electrical resistance of the overall EP system with the addition of the nanopore membrane, thereby decreasing the corresponding electric current.
Although the applied voltage can be largely focused on the nanopore membrane, the resistance of the relatively bulky EP medium cannot be neglected. Therefore, to apply uniform electric strength to the nanopores distributed in the nanopore membrane with a diameter of 25 mm, the nanopore membrane was placed between two circular plate electrodes with diameters of 21 mm (top) and 35 mm (bottom) ( Figure 2a). The nanopore membrane prepared by track-etching of the polycarbonate film had an average pore diameter of 100 nm with a narrow pore size distribution ( Figure S1) and a pore density of 1 Â 10 7 pores/cm 2 (Figure 2b). The cylindrical nanopores spanned the entire membrane thickness of 7 μm (Figure 2c). The device was fabricated in a multi-well format to improve screening throughput; the hydrophobic surface of the nanopore membrane was coated with a cell adhesion molecule, fibronectin, at a concentration of 20 μg/mL or higher to improve cell adhesion ( Figure 2d).

| Simulation study of bulk EP and nEP
We performed simulation studies to compare the electric field distributions in bulk EP and nEP devices. The bulk EP device is an electroporation cuvette composed of two parallel plate electrodes spaced 2 mm apart, whereas the nEP device consists of two circular electrodes spaced 2.7 mm apart. Since the magnitude of the electric field is inversely proportional to the distance between the electrodes, a smaller inter-electrode gap produces a higher electric field under the same voltage applied. To match the voltage per unit length between the two electrodes of each device, it was necessary to multiply the bulk EP voltage by 1.35 times. For example, 30 V in the bulk EP corresponds to 40.5 V in the nEP. As simulated by COMSOL Multiphysics, the electric field in the bulk EP was uniform between the plate electrodes, whereas the electric field in the nEP was locally focused on a thin membrane with a thickness of 7 μm (Figure 3a,b). This is due to the local increase in electrical resistance upon the addition of the nanopore membrane; the electric field strength at the nanopore entrance increased rapidly, exceeding the critical field strength for electroporation (3 kV/cm) up to 33.26 kV/cm at the applied voltage difference of 40 V (Figure 3c). 20,41,42 Under similar electric conditions, the electric field strength of the bulk EP was 0.15 kV/cm, which is markedly lower than the critical value and is expected to lead to insufficient electroporation.
These results indicated that the nanopore membrane can effectively focus the electric field through the nanopores, thereby enhancing the electric field strength to be sufficiently high for electroporation. Assuming that adherent cells are attached at a distance of approximately 100 nm from the surface, 43 Figure S2). 46 However, the abrupt change in conductivity near the nanopore can generate a higher electric field strength of 61.54 kV/cm under 40 V, supporting that nEP can be effective for in vivo electroporation of tissues.

| In vitro parametric study of nEP
To determine the optimal electroporation conditions, we tested the effects of electric pulse conditions on transfection efficiency, cell recovery, and cell viability. In electroporation, the applied voltage and voltage pulse width are important parameters that determine the rate and duration at which extracellular materials are delivered to the cells. potential drop across the nanopore membrane dramatically increased the electric field strength above the critical value, thereby ensuring effective electroporation (more than 55.3%) while retaining cell recovery (more than 84.9%) and viability (more than 94.6%).

| Characterization of nanopore needle electrode for in vivo nEP
To extend the nEP principle identified in the in vitro testing device for in vivo intracellular delivery in deep tissues, we designed 2-needle array electrodes consisting of a conventional metal (Au-coated stainless steel) needle electrode and a 3D-printed hollow needle with a side hole (to be covered with a nanoporous membrane), as illustrated in Figure 5a. The nanoporous membrane was bonded using an adhesive to seal the side hole of the nonconductive 3D-printed needle.
After filling the inner space of the nanopore needle with a solution containing delivery molecules, a Pt wire electrode was inserted into the nanopore needle while it was submerged in the solution. By applying electric pulses through the 2-needle array electrode, cells close to the nanoporous membrane could be changed to a selectively permeable state, and a target model agent (PI) in the solution could be delivered into the cells through the nanopores. Figure 5b shows the SEM image of a 3D-printed needle after bonding the nanoporous membrane to the side hole. The needle was made of polyacrylate-based photocurable resin using digital light processing (DLP)-based precision 3D printing. It had a conical tip with a diameter of less than 100 μm.
The side hole of the needle was tightly sealed by adhering it to the F I G U R E 3 Simulation results showing the electric field distributions in the bulk EP and nEP. Since the nanopores on the membrane can be considered as parallel resistors in an electric circuit, a single nanopore was used for electric field simulation. (a) The parallel plate electrodes in a homogeneous fluid generate a uniform electric field, while (b) the addition of the nanopore membrane results in electric field getting focused to the membrane. The applied voltages were 30 V and 40 V for bulk EP and nEP, respectively. (c) Electric field strength profiles along a direction away from a certain point in the bulk EP and nanopore entrance in the nEP.  (Figure 5f). Due to the limited loading volume of the PI solution in the nanopore needle, the release amount per electrical pulse decreased with an increase in the number of nEP. This electricresponsive release system would be advantageous for localized treatment to reduce off-target effects and personalized therapy that requires precise drug release according to the patient's condition. [47][48][49] In contrast, with the bulk EP system, it is difficult to achieve local, quantitative, and stimuli-responsive release because electrical stimulation must be applied after the drug solution is injected into the tissue.

| Efficacy and safety of in vivo nEP system for intracellular delivery in a pilot study
To confirm the transfection efficiency and safety of the nEP system based on a nanopore electrode, in vivo EP tests were performed following the insertion of the two-needle array electrode into the thigh muscle of a mouse (Figure 6a). Since 200 V is normally recommended for in vivo bulk EP. 50

| CONCLUSIONS
We have successfully designed and demonstrated a new nEP system that allows precise on-demand intracellular delivery of cell-

| Numerical simulation of bulk EP and nEP
Electric field simulations were performed using the finite-element COMSOL Multiphysics software (COMSOL, Inc.) to visualize the electric field distributions in the bulk EP and nEP devices and calculate the resulting electric field strengths. The 3D finite-element models for the devices were created in the same dimensions and were then solved using the AD/DC module of COMSOL Multiphysics. The electrical conductivity and relative permittivity were set to 1.4 S/m and 80, respectively, for PBS, and 0.3 S/m and 60,000, respectively, for the tissue. 46

| Fabrication of the nanopore electrode
The nanopore electrode was fabricated by wrapping a nanopore membrane on a side hole needle 3D printed using a high-resolution DLP 3D printer (Perfactory ® Micro Plus HD, EnvisionTEC, Germany) ( Figure 5b). The side-hole needle model was designed using Sketch up (Trimble), and the optimized STL model was sliced into 25 μm using Perfactory Rapid Prototyping (Envision TEC, Germany). A photocurable resin (HTM-140, EnvisionTEC, Germany) was used for 3D printing. To wrap the side of the 3D-printed needle with a nanopore membrane, a UV-curable adhesive was thinly applied on the 3Dprinted needle, and then the nanopore membrane was wrapped on the needle and UV-irradiated. The nanopore membrane was adhered to a 3 Â 0.5 mm side hole of the 3D-printed needle. The surface of the membrane wrapping the nanopore electrode was observed by field-emission scanning electron microscopy (FE-SEM; S-4700, Hitachi, Japan); pore blockage by the adhesive was not observed. The interior of the nanopore electrode was in the form of a column with a diameter of 500 μm, which could be filled with PI solutions. To apply a pulse to the nanopores, a Pt wire electrode was inserted into the interior of the nanopore electrode.

| Evaluation of the release of model agent from nanopore electrode in in vitro test
The amount of PI released from the nanopore electrode was mea-

| Animals
The animal protocol used in this study was reviewed and approved

| In vivo nEP test using nanopore electrode
To evaluate the transfection efficiency and safety of the nanopore electrode in in vivo experiments, each electrode was inserted into the shaved hindlimb thigh muscle of the mouse, and an electric pulse was applied. PI was used as the model agent. In the case of bulk EP, 10 μL of 1 mM PI was injected intramuscularly into the EP site before applying the pulses, and in the case of nEP, 1 μL of 10 mM PI was injected into the nanopore electrode. Both the pulse conditions commonly used for bulk EP (200 V, 20 ms, 4 times) and optimized for nEP (40 V, 2 ms, 99 times) were applied to each electrode. During electrical stimulation, the experimental animals were anesthetized using an anesthesia breathing system.

| Histopathological analysis
To evaluate the histopathological changes, the thigh muscle of each mouse was freshly excised, fixed in 10% neutral buffered formalin for 24 h, and embedded in paraffin. Paraffin-embedded specimens were sliced perpendicular to the electrode insertion direction into 5-μmthick sections. Deparaffinized skin sections were stained with H&E for analyzing tissue damage. Bright-stained sections were then examined with a digital fluorescence slide scanner (Axio Scan Z1, ZEISS, Germany) to assess histological changes, including damaged tissue.

| TUNEL assay
Apoptosis was evaluated using an in situ apoptosis detection kit (mk500, Takara, Japan). Proteolysis was performed using proteinase K for 15 min. Working Strength TdT enzyme was added to the proteolyzed slices (5 μL TdT enzyme: 45 μL reaction buffer). Then, the slices were incubated in a humidified chamber at 37 C for 90 min. Next, the slices were treated with fluorescein isothiocyanate-antibody (FITC-Ab) mixture in a humidified chamber at 37 C for 60 min. Finally, sections counterstained with Hoechst were observed using a digital fluorescence slide scanner. The areas with fluorescence were determined using the ImageJ 1.50e software (Bethesda).

This research was supported by Samsung Research Funding & Incubation
Center of Samsung Electronics (SRFC-IT1801-02) and Basic Science