Next generation CANCAN focusing for remote stimulation by nanosecond electric pulses

Abstract

Focusing electric pulse effects away from electrodes is a challenge because the electric field weakens with distance. Previously we introduced a remote focusing method based on bipolar cancellation, a phenomenon of low efficiency of bipolar nanosecond electric pulses (nsEP). Superpositioning two bipolar nsEP into a unipolar pulse canceled bipolar cancellation (“CANCAN” effect), enhancing bioeffects at a distance despite the electric field weakening. Here, we introduce the next generation (NG) CANCAN focusing with unipolar nsEP packets designed to produce bipolar waveforms near electrodes (suppressing electroporation) but not at the remote target. NG-CANCAN was tested in CHO cell monolayers using a quadrupole electrode array and labeling electroporated cells with YO-PRO-1 dye. We routinely achieved 1.5–2 times stronger electroporation in the center of the quadrupole than near electrodes, despite a 3–4-fold field attenuation. With the array lifted 1–2 mm above the monolayer (imitating a 3D treatment), the remote effect was enhanced up to 6-fold. We analyzed the role of nsEP number, amplitude, rotation, and inter-pulse delay, and showed how remote focusing is enhanced when re-created bipolar waveforms exhibit stronger cancellation. Advantages of NG-CANCAN include the exceptional versatility of designing pulse packets and easy remote focusing using an off-the-shelf 4-channel nsEP generator.

1. Introduction

Electrostimulation is a versatile clinical and research tool. Its applications span from neuromuscular and brain stimulation[1–4], defibrillation[5], modulation of immune and endocrine function[6–8], to electroporation[9–11], gene electrotransfer[12], and tissue and cancer ablation[8, 11, 13, 14].

There is a tremendous promise in the non-invasive, targeted electrostimulation[1–3, 15–18]. The challenge is avoiding stimulation near electrodes, where the electric field is the strongest, while stimulating at a distance by a (much) weaker electric field. We have recently introduced and validated a “cancellation of cancellation” (CANCAN) paradigm to focus electrostimulation remotely from stimulating electrodes [15, 19]. CANCAN stimulation takes advantage of the phenomenon of bipolar cancellation, which stands for the inhibition of effects when the stimulus polarity is reversed[20–28]. As a result, a single phase of a bipolar electric pulse can be more efficient at excitation, electroporation, and other electrostimulation effects than both phases applied together. Bipolar cancellation is typical for nanosecond-duration electric pulses (nsEP), although was reported outside of this range as well. A recent analysis of how bipolar cancellation depends on pulse parameters (phase duration and the ratio of the opposite polarity phases) has led us to conclude that cancellation for different cell types and endpoints may be caused by different mechanisms[24]. While these mechanisms have been only partially understood, CANCAN stimulation exemplifies the translation of the basic knowledge of nsEP effects and bipolar cancellation towards their medical and research applications.

For CANCAN stimulation[15], two pairs of electrodes forming a quadrupole array are energized from independent and electrically isolated, high-voltage nsEP generators. One pair of the electrodes delivers a triphasic pulse with the amplitude decreasing from the first to the last phase. The second pair of the electrodes delivers a biphasic pulse, identical and synchronous to phases 2 and 3 of the triphasic one. The superposition of the tri- and bipolar pulses in the center of the array compensates the 2^nd and the 3^rd phases, leaving the uncompensated 1^st phase as a unipolar pulse. High efficiency of this unipolar nsEP formed by the interference of two bipolar signals overweighs the inevitable weakening of the electric field with distance from the electrodes. As a result, effects such as electroporation could be made stronger in the center of the array than in the immediate vicinity of the stimulating electrodes.

This implementation of CANCAN required custom-designed, high-voltage bipolar nsEP generators[29] which fired synchronized pulses through electrically isolated output circuits. Such generators are bulky and offer little flexibility in the adjustment of pulse parameters (such as phase duration and amplitude) to maximize the remote effect and minimize it near the electrodes. In the present study, we propose and test an opposite paradigm of CANCAN remote stimulation: Instead of two bipolar nsEP which form a unipolar pulse remotely in the center of the array, we utilize packets of unipolar pulses which form bipolar electric field throughout the array but not in its center.

Fig. 1A illustrates how a bipolar electric field can be generated by applying unipolar pulses alternately to two electrodes. A cell positioned between the electrodes would have no means to “know” whether the bipolar electric field is produced by a bipolar pulse or by a pair of unipolar pulses. Figs. 1B and C illustrate the concept of how a packet of unipolar pulses can be employed for CANCAN stimulation targeting. In the first phase, a unipolar pulse is applied to one electrode while the other three electrodes serve for current return (ground). In the next phases, electrodes across one diagonal are energized and the other diagonal serves for current return. Concurrently, pulse amplitude is reduced with each phase. The unipolar pulse applied first creates a unipolar electric field throughout the array, including its center. Next, energizing either diagonal creates a bipolar electric field at the perimeter but not in the center. Fig. 1C shows the electric field signals experienced by cells at the respective locations within the array following the application of a pulse sequence (a “packet”) shown in Fig. 1B. The electric field vector at intermediate locations between the center and the perimeter changes in a complex, location-dependent way, at angles which are different from 90°. Same as with the original CANCAN, the electrostimulation efficiency of such complex multiphasic fields is difficult to predict and needs to be established by direct experiments.

Fig. 1.

The concept of remote targeting by interference of unipolar electric pulses. A: A schematic explaining two approaches how to generate a bipolar electric field between two electrodes. In the left panel, a bipolar pulse is applied to one electrode while the other electrode is ground. In the right panel, a unipolar pulse is applied alternately to two electrodes, and the non-energized electrode becomes ground. The electric field between the electrodes does not depend on the approach chosen. B: A sample sequence of unipolar pulses which creates a bipolar electric field at the perimeter of a quadrupole electrode array but not in its center. In Phase 1, a 4-kV pulse is applied to one electrode and 3 other electrodes are grounded. Arrows highlights the electric field direction and strength during in each phase. In Phases 2–4, two electrodes of one diagonal are energized by the same voltage while the other diagonal is grounded. The pulse amplitude is reduced with each phase. Note that the electric field in the center of the array is only generated during Phase 1, but not when a diagonal is energized. Pulse sequence and amplitudes exemplify stimulation with Packet 1 (Fig. 2B). C: The resulting waveforms at the perimeter and in the center of the electrode array. See text for more details.

This “next-generation” (NG) CANCAN yields the end result similar to the original CANCAN (bipolar electric field at the perimeter and unipolar in the center) but using a single-body 4-channel unipolar nsEP generator. A commercially available EPULSUS-FPM4–7 generator (Energy Pulse Systems, Lisbon, Portugal)[25, 26, 30] used in this study offers exceptional versatility in tuning and synchronizing signals from each channel, as well as options for “rotation” of preprogrammed pulse packets (see Methods for detail). We showed that the NG-CANCAN can be readily and efficiently implemented with this off-the shelf generator. We compared CANCAN targeting in two quadrupole arrays of different size and analyzed the role of several pulse parameters. Finally, we demonstrated that nsEP packets most efficient for remote focusing can be pre-designed based on the known cancellation efficiency of bipolar signals created by these packets.

2. Materials and methods

2.1. Cell lines and media

All experiments were performed in CHO-K1 cells (Chinese hamster ovary) purchased from the American Type Culture Collection (ATCC, Manassas, VA). Cells were propagated at 37 °C with 5% CO₂ in air, in Ham’s F12K medium (Mediatech Cellgro, Herndon, VA) (Atlanta Biologicals, Flowery Branch, GA), with 100 IU/mL penicillin and 0.1 μg/mL streptomycin (Gibco Laboratories, Gaithersburg, MD).

For measuring bipolar cancellation of electroporation in individual cells (Sections 2.2 and 3.1), they were seeded onto 10-mm diameter, #0 thickness glass coverslips (Ted Pella, Redding, CA) on the day before experiments. For all other experiments, cells were harvested 12–18 h before experiments, concentrated to 0.3–0.6 ×10⁶ cells/ml by spinning, and seeded at ~2 ml/dish onto 35 mm glass-bottom culture dishes (MatTek, Ashland, MA). Dishes with cell monolayers were removed from the incubator 10–15 min before nsEP exposure, and the medium was replaced with 2 ml of physiological solution containing (in mM): 140 NaCl, 5.4 KCl, 2 CaCl₂, 1.5 MgCl₂, 10 D-glucose, and 10 HEPES (pH 7.4). From this time on, cells were kept at room temperature in the physiological solution. Each dish was used in one experiment.

2.2. Measuring electroporation in individual cells by time-lapse imaging of dye uptake

These experiments were aimed at comparing the electroporation efficiency of unipolar nsEP and three types of bipolar, multi-phasic nsEP. All procedures except the nsEP generation were similar to those described before[31–34].

A coverslip with cells was placed into a glass-bottomed chamber (Warner Instruments, Hamden, CT) mounted on a stage of an Olympus IX81 inverted microscope configured with an FV1000 confocal laser scanning system (Olympus America, Center Valley, PA). The chamber was filled with the physiological solution supplemented with 1 μM of YO-PRO-1 dye (YP) (Thermo Fisher Scientific, Waltham, MA). This dye becomes brightly fluorescent upon entering cells with compromised membranes, and its fluorescence in cells is proportional to the membrane damage[19, 20, 27, 34–36]. YP fluorescence images (ex.: 488 nm/em.: 505–525 nm) were taken with a 40x, 0.95 NA dry objective every 10 s for 600 s.

nsEP were delivered to a small group of imaged cells by a pair of tungsten electrodes (100 μm diameter, 140 μm gap) placed precisely 50 μm above the coverslip with an MP 225 micromanipulator (Sutter Instruments, Novato, CA). Computational modeling of the electric field strength at cell location was accomplished with a finite element solver Sim4Life (ZMT Zurich MedTech AG, Zurich, Switzerland) as described recently [37]. Cells were exposed to nsEP (or sham exposed) at 57 s into the time lapse image sequence. Pulse delivery and imaging were synched and controlled using a Digidata board model 1322A and Clampex software (Molecular Devices, CA, USA).

A single unipolar pulse or a brief train of synched pulses (all 600 ns in duration) were delivered from two output channels of an EPULSUS-FPM4–7 generator. To generate a unipolar pulse, one channel was energized and the other served as ground. To produce bipolar electric field signals between the electrodes, the output channels were energized in alternation, as shown in Fig. 2, B–D for electrodes e1 and e2. These pulse protocols produce decremental bipolar waveshapes consisting, respectively, of 4 phases; 8 phases; and 8 phases with 5-μs delays between them. The phases diminished by a factor of 1/8^th per phase (each EPULSUS channel uses 8 Marx stages[30], and one stage per phase was disengaged to reduce the output). The electric field at the cell location during the first phase, as well as for a single unipolar pulse exposure, was 15 kV/cm.

Fig. 2.

The design of unipolar pulse packets for NG-CANCAN in a 4-electrode quadrupole array. A: A schematic of the four electrodes (e1-e4) installed orthogonal to a cell monolayer in a culture dish. B-D: the sequence of pulses applied to electrodes e1-e4 to form packets 1, 2, and 3, respectively. In all packets and for all electrodes, pulse duration is 600 ns. All packets start from e1 electrode and produce a 4-phasic (B) or 8-phasic (C and D) bipolar waveform near it. In B and C, there is be no delay between the phases of the bipolar signal; in D, all phases are be separated by 5-μs delays. The amplitude decrement is fixed at 1/8^th of the first phase for all subsequent phases (2/8^th for the 2^nd phase, 3/8^th for the third, etc.) See text for more details.

The experiments using different nsEP trains, single pulses, and sham exposures were performed in a random sequence. Stacks of cell images were quantified using the Fiji package of ImageJ software[38]. A region of interest (ROI) was drawn manually around each cell and the average pixel intensity (F) was measured within the ROI. The measurements were further averaged across six images taken prior to nsEP delivery, and this value was considered as a baseline (F0) for each cell. Emission intensity corrected for the background (F-F0) was used for graphs and statistical analyses.

2.3. nsEP delivery to cell monolayers, exposure control, and electric field simulation

Experiments aimed at testing CANCAN targeting were performed in cell monolayers. Unipolar nanosecond stimuli from the EPULSUS generator were delivered to the monolayer with a quadrupole electrode array (Fig. 2A). We utilized two quadrupole electrode arrays, with approximately a 2-fold size difference (Fig. 3, A,B and C,D). Both arrays were made with blunt, hollow, stainless-steel needles 1.5-mm in diameter (Integrated Dispensing Solutions, Agoura Hills, CA). They were positioned orthogonal to a cell monolayer and lowered using an MX130L micromanipulator (Siskiyou Corporation, Grants Pass, OR) until touching the glass bottom of the culture dish. In one set of experiments (section 3.5), electrodes were positioned at various distances (1–3 mm) above the glass.

Fig. 3.

Electric field simulation for the small (A and B) and large (C and D) quadrupole electrode arrays. Panels A and C show the electric field intensity when a 1 kV pulse is applied to the top left electrode, and other electrodes serve as ground (corresponds to Phase 1 in Fig. 1B). Panels B and D show the electric field distribution when 1 kV is applied to one electrode diagonal and the other diagonal is ground (corresponds to Phases 2–4 in Fig. 1B). Bottom panels show the same data, with the electric field vector direction added. Note that energizing a diagonal produces no electric field in the center of the array. See text for more details.

Four channels of the EPULSUS generator were connected to 4 electrodes (e1-e4, Fig. 2). Each channel was programmed to deliver a train of 600-ns pulses with 600-ns or longer intervals, to form a pulse “packet”. In this study, we tested 3 types of packets illustrated in Fig. 2, B–D.

Packet 1 (Fig. 2B) consisted of 7 pulses with 600-ns intervals (2 from electrodes e1, e2, and e4, and 1 from e3) to produce 4-phase bipolar waveforms. Pulse amplitude was reduced by 1/8^th (12.5%) with each phase (e.g., a 4-kV first pulse was followed by 3.5 kV, 3 kV, and 2 kV pulses, Fig. 1 B,C). The 12.5% decrement was the smallest step that could preprogrammed in the pulse generator (see 2.2 above).

Packet 2 (Fig. 2C) consisted of 15 pulses (4 from electrodes e1, e2, and e4, and 3 from e3) to produce 8-phase bipolar waveforms. Pulse amplitude was also reduced by 12.5% with each phase. As it turned out in the electroporation efficiency experiments (Section 3.1 below), the 8-phase bipolar signal was the best at bipolar cancellation, and this was the reason why it was chosen for most of the subsequent CANCAN targeting studies.

Packet 3 (Fig. 2D) was identical to Packet 2 in all aspects except for a 5-μs delay between pulses.

In most experiments, we applied multiple packets and utilized packet “rotation”, which greatly improved CANCAN targeting [15]. For the rotation, the electrode energized first in the packet was shifted by one position clockwise, e.g., from e1 to e2, then to e3, and to e4 (Fig. 2). The number of packets applied from each electrode and their rotation are designated in figure legends. For example, “1Px4” means that a single packet was applied once from each of 4 electrodes, making one full rotation. “10Px4” means applying 10 packets from each of 4 electrodes, to make one full rotation and deliver a total of 40 packets. “1Px40” means that 1 packet was “rotated” 40 times (made 10 full circles), to the same total of 40 packets. The packets, whether with rotation or not, were always delivered at a low repetition rate of 1 Hz to prevent temperature build-up[15].

Pulse shapes and amplitudes were continuously monitored with a TDS 3052 oscilloscope (Tektronix, Beaverton, OR) using either a DP20003 High Voltage Differential Probe (Shenzhen Micsig Instruments Co., Guangdong, China) or the pulse generator’s built-in monitor output.

Computational modeling of the electric field was accomplished using the Sim4Life software the same way as described in the previous paper[15]. Simulations were performed for the small and the large electrode arrays (Fig. 3A,B and C,D, respectively), by applying 1 kV to a single electrode (panels A and C) or to two electrodes in one of the diagonals (panels B and D). The resulting electric field strength is coded by color (top panels) and both by color and size of the electric field vectors (bottom panels). For example, applying 1 kV to one electrode produced the electric field of approximately 1.8 kV/cm and 0.8 kV/cm in center of the small and large arrays, respectively (Fig. 3, A and B). These electric field values scale linearly with the applied voltage, e.g., applying 2 kV will produce 3.6 and 1.6 kV/cm; 4 kV will produce 7.2 and 3.2 kV/cm, etc. The electric field strength was about 4-fold higher near the energized electrode (7 kV/cm and 3.5 kV/cm per 1 kV applied, for the small and large arrays, respectively) Note that energizing either electrode diagonal produced no electric field in the center of the array (panels B and D).

Throughout this paper, we will refer to different nsEP treatments by voltage of the first pulse, the type of packets, and the number and rotation of packets. This information, along with data in Figs. 2 and 3, should be sufficient for readers to calculate local electric field for specific treatment conditions and a location of interest, to facilitate comparison with other studies on electroporation by nsEP and bipolar cancellation.

2.4. Mapping cell membrane permeabilization in monolayers

Electroporation in cell monolayers was quantified by fluorescence imaging with YP dye, using the protocols described previously [15]. Cells were incubated in the physiological solution with 1 μM YP from the time immediately before nsEP exposure to 10 min after it, in the dark. YP was rinsed away with fresh physiological solution, and the dish was transferred to an IX83 microscope (Olympus America, Hamden, CT) configured with an automated MS-2000 scanning stage (ASI, Eugene, OR), X-Cite 110LED illuminator (Excelitas Technologies Corporation, Waltham, MA) and an ORCA-Flash4 sCMOS camera (Hamamatsu, Shizuoka, Japan). Multiple high-resolution images of adjacent regions between the electrodes were taken with a 10x, 0.38 NA objective and stitched together into one image. Stage re-positioning and synchronization with illumination and image acquisition were accomplished automatically with CellSens software (Olympus).

Fluorescence intensity was measured with MetaMorph 7.8.13 (Molecular Devices, Foster City, CA) as a function of the distance from the geometrical center of the electrode array. The measurements were performed in 92 or 224 regions of interest (ROI), placed along one or both diagonals, and covering a distance of +/− 1.65 mm or +/− 3.65 mm from the geometrical center for the small and large arrays, respectively. Areas immediately adjacent to the electrodes (<5% of the diagonal length) were excluded because of possible mechanical damage to the monolayer.

Graphs were prepared with Grapher 16 (Golden Software, Golden, CO). Fluorescence intensity is shown either in arbitrary units (a. u.) or in percent to the most peripheral (closest to the electrode) ROI. Fluorescence data are presented as a mean +/− s.e. for 3–5 independent experiments and fitted using LOESS regression. Two-sided Student’s t-test was employed to analyze the significance of differences; p<0.05 was considered statistically significant. We avoided using special symbols to mark the statistical significance in the graphs to preserve their clarity. Instead, the significance can be estimated from the gap between the error bars of the compared groups: A gap exceeding the length of the error bars indicates a significant difference at p<0.05 or better[39].

3. Results

3.1. Bipolar cancellation of electroporation in individual CHO cells

CANCAN targeting is based on the phenomenon of bipolar cancellation, which enables selective suppression of the electric field effects near electrodes. We hypothesized that bipolar signals which cause stronger cancellation will result in better CANCAN targeting, and that such signals can be pre-selected in simple experiments with cells exposed to the electric field between two electrodes. Small groups of cells on a coverslip were exposed to the electric field as described in Section 2.2, at 57 s into the imaging series (Fig. 4). Sham-exposed cells showed no gain in YP fluorescence during the 10 min of observation. In contrast, cells treated by electric pulses displayed gradual increase in YP signal, starting from the image frame taken immediately after nsEP exposure. A single 600-ns pulse at 15 kV/cm caused by far the strongest YP uptake, reaching 126±14 a.u. (n=25) by the end of recording. Bipolar signals started with exactly the same pulse (600 ns, 15 kV/cm) but caused significantly weaker effects (p<0.001). By the end of recording, YP signal reached only 20.8±2.0 a.u. for the 4-phasic signal; 6.9±1.6 a.u. for the 8-phasic signal; and 25.4±2.8 a.u. for the 8-phasic signal with 5-μs delays between phases. The impressive 20-fold bipolar cancellation achieved with the 8-phasic signal was the reason why Packet 2 (designed to produce the same signal near electrodes, Fig. 2C) was chosen for most CANCAN targeting studies described below.

Fig. 4.

YO-PRO-1 dye uptake in CHO cells electroporated by uni- and bipolar electric pulses. The pulse shapes are shown in insets next to the graphs. Pulses were delivered by one pair of electrodes (Section 2.2). For all electric pulse treatments, each phase (pulse) duration was 600 ns, and the field strength for the first phase was 15 kV/cm (decreasing thereafter by 1/8^th per phase). Vertical dashed line designates the time when electric pulses were applied. Mean data ± s.e. for 21–30 cells per group. Sham exposures were performed same way as the other treatments, but no pulses were delivered. Note that applying the 8-phasic bipolar signal with no inter-phase delays caused about 20-fold weaker electroporation than a single unipolar pulse (which was identical to the first phase of the bipolar signal).

3.2. Targeting of electroporation away from electrodes by NG-CANCAN

Fig. 5 illustrates how NG-CANCAN protocols enable focusing of electroporation to the center of a quadrupole electrode array. We applied Packet 2 (Fig. 2C) 40 times with 1-s intervals. The packets started with energizing e1 electrode (Fig. 5A, “40 Packets”). For control (Fig. 5A, “40 Trains”), we energized e1 electrode only with the same pulse trains (top panel in Fig. 2C), resulting in a unipolar electric field stimulation. The lack of cancellation caused a much stronger electroporation throughout the array, and the distribution of YP uptake expectedly followed the electric field strength distribution in Fig. 3C. In contrast, Packet 2 enhanced electroporation in the center of the array, despite gradual weakening of the electric field with distance from e1. However, the bipolar cancellation only partially offset the electric field weakening, and electroporation near e1 remained stronger than in the center.

Fig. 5.

Focusing electroporation to the center of a quadrupole electrode array by interference of unipolar pulses (NG-CANCAN). Packet 2 (see Fig. 2C) was applied starting from one electrode (A) or with electrode rotation (B). The first pulse in the packet was applied at 6.4 kV. A: Packet 2 was applied to the cell monolayer 40 times at 1 Hz, always starting from e1 electrode (“40 Packets”). For control, only e1 was energized with an identical train of pulses (top panel in Fig. 2C) to produce a unipolar electric field (“40 trains”). YP fluorescence intensity was measured along the diagonal between e1 and e3. Insets are sample YP fluorescence images within the quadrupole following either control (top) or CANCAN stimulation. Image brightness was adjusted for visual clarity. B: Same control (“1T × 40”) and CANCAN (“1P × 40”) treatments, but “rotating” the electrode which is energized first (e.g., 1^st packet started from e1, 2^nd from e2, 3^rd from e3, etc., to a total of 40 packets and 10 full rotations). YP emission data were averaged across both diagonals of the quadrupole. Brightness of the inset images was adjusted the same way as in panel A. C: Same data as in panel B, but plotted relative to measurements closest to electrodes (taken as 100%). Mean ± s.e., n= 4–5.

The limitation was addressed by rotating the electrode which is energized first (Fig. 5B). Rotation with Packet 2 “smeared” the electroporation effect at the perimeter and emphasized the remote effect in the center. This remote effect contrasted with a void of electroporation in the center when the pulse train was applied in the same manner instead of the packet. The striking difference in the spatial distribution of electroporation between the CANCAN and control treatments is appreciated best by comparing the normalized fluorescence data, when the emission intensity in the most peripheral (closest to the electrode) ROI is taken as 100% (Fig. 5C).

These proof-of-concept experiments show that remote targeting of electroporation can indeed be accomplished using packets of unipolar nsEP, and that packet rotation assists it. Further optimization of NG-CANCAN requires the knowledge of how targeting depends on different nsEP parameters and packet design.

3.3. Impact of the number of packets and nsEP amplitude on electroporation targeting with NG-CANCAN

The ratio of effects in the center and near the electrodes, such as shown in Fig. 5C, was used as a convenient index of targeting efficiency. The experiments were performed using Packet 2 (Fig. 2C) with rotation.

Increasing the total number of applied packets from 4 to 20 and 40 profoundly enhanced the electroporative YP dye uptake throughout the electrode array (Fig. 6A). The peak of electroporation in the center was clearly discernable for all tested treatments. Its brightness was almost twofold higher than at the periphery for a single packet was applied 4 times from different directions, or about the same as at the periphery for the rotation of 5 and 10 packets (Fig. 6B). Interestingly, the least affected area of the cell monolayer (between the center and the electrodes) showed about 2-fold lower fluorescence intensity than in the center for all the treatments. Also, the central peak appeared narrower (sharper) with increasing the number of packets.

Fig. 6.

Effect of the number of unipolar pulse packets on focusing electroporation to the center of a quadrupole electrode array. A: Packet 2 was applied 1, 5, or 10 times from each of the electrodes, to the total of 4, 20, or 40 packets, respectively (see Fig. 2C and Section 2.3). Insets show the respective sample images of YP fluorescence in the cell monolayer (brightness was adjusted for visual clarity). B: Same data plotted relative to measurements closest to electrodes (taken as 100%). Mean ± s.e., n= 3–4.

In contrast, changing nsEP amplitude affected the degree of electroporation but not its spatial distribution (Fig. 7). For example, when the first pulse amplitude was increased from 2.4 to 5.6 kV (and the amplitude of the other pulses in the packet was increased proportionally, see Fig. 2C), YP fluorescence increased 15.5 times at the most peripheral point; 16.2 times half-way to the center; and 16.8 times in the center (Fig. 7A). When the YP emission was normalized to the most peripheral ROI, the data for different nsEP amplitudes fully overlapped (Fig. 7B).

Fig. 7.

Effect of pulse voltage on CANCAN focusing of electroporation in the small (A and B) and large (C and D) electrode arrays. Packet 2 was applied once from each electrode to a total of 4 packets (A and B), or 10 times from each electrode to a total of 40 packets (C and D). Voltage of the first pulse in the packet is marked next to the graphs. B and D: same data as in A and C, respectively, plotted relative to measurements closest to electrodes (taken as 100%). Mean ± s.e., n=3.

The above measurements were performed for rotation of a single nsEP packet and using the smaller electrode array. The findings held true for rotating multiple packets and when using the larger array (Fig. 7C and D). Overall, our results show that NG-CANCAN targeting of electroporation does not depend on nsEP amplitude, at least within the studied amplitude range.

3.4. Bipolar cancellation determines CANCAN targeting

NG-CANCAN targeting hinges on producing a bipolar electric field and inhibition of nsEP effects near electrodes. Our experiments in Section 3.1 identified the 8-phasic bipolar signal as the most efficient at cancellation (~20-fold), and thereby as the most promising for CANCAN targeting. Here, we compared its targeting efficiency with the other two types of bipolar signals which were less efficient at cancellation (4–5-fold, Fig. 4). The three bipolar signals studied in Section 3.1 (4-phasic; 8-phasic; and 8-phasic with 5-μs interphase delays) match the bipolar electric field created near electrodes in a quadrupole array by packets 1, 2, and 3 respectively. However, cells in the center of the array should be affected equally by all packets because they experience just a single unipolar electric field pulse (Fig. 1B and C) and “will not know” how many pulses (phases) are applied afterwards.

Experimental measurements were in remarkable agreement with these expectations (Fig. 8). The different packets were applied at the same first pulse amplitude of 4 kV, and were all “rotated” one cycle. All packets produced the same YP uptake in the center (215–235 a.u.), but Packets 1 and 3 caused drastically more YP uptake near the electrodes (250–350 a.u. vs ~120 a.u. for packet 1, p<0.01, Fig. 8A). The superiority of Packet 2 for CANCAN targeting is highlighted by the drop of normalized YP emission in the center from 183% for Packet 2 to 83% and 74% for Packets 3 and 1, respectively (Fig. 8B). It was obviously the lower potency for cancellation of Packets 1 and 3 that resulted in weaker suppression of electroporation and larger YP uptake near electrodes.

Fig. 8.

CANCAN focusing of electroporation in a quadrupole array by different packets of unipolar pulses. Packets 1, 2, and 3 (Fig. 2, B–D) were applied once starting from each of the electrodes, to a total of 4 packets delivered. The 1^st pulse amplitude in all packets was 4 kV. Panels A and B show the same YP emission data in arbitrary units (A) and in % to the intensity in the most peripheral ROI, next to electrodes (B). Mean ± s.e., n=3–4.

3.5. NG-CANCAN targeting of electroporation in 3D

Most prospective applications of NG-CANCAN will require remote targeting in 3D tissue volumes, when at least some of target tissue is outside of the plane connecting electrode tips. We measured targeting efficiency in surrogate 3D experiments, when the electrode array was placed in the solution at a distance above the cell monolayer. The experiments were performed using the larger electrode array, with 8.4-mm diagonal (Fig. 3C, and D).

Introducing a gap between the monolayer and the electrodes drastically reduced electroporation near the electrodes, but only moderately in the center of the array (Fig. 9). Specifically, YP emission near the electrodes decreased 2.4, 6.4, and 9 times with 1-, 2- and 3-mm gaps (as compared to electrodes being in the same plane with cells); in the center, it decreased just 1.02, 1.4, and 3.8 times, respectively. Such disproportionally faster fading of the effect near electrodes boosted the CANCAN targeting efficiency (measured as a ratio of the effect in the center and at the periphery) to as much as 600% for the 2-mm gap (Fig. 9B). Consistent with these measurements, YP fluorescence images in Fig. 9C show that electroporation is restricted almost exclusively to the remote target (center) when electrodes are placed above the cell monolayer.

Fig. 9.

CANCAN focusing of electroporation in 3D. Tips of electrodes in a quadrupole array were either in contact with cell monolayer (0 mm distance) or were placed 1, 2, or 3 mm above it. Packet 2 was applied 10 times starting from each electrode, to a total of 40 packets delivered. The amplitude of the first pulse in the packet was 6.4 kV. A and B show YP emission for the indicated distances of the array from cell monolayer, in arbitrary units (A) and in % to measurements most distant from the center (B). Mean ± s.e., n=3–4. Panel C shows sample images of YP fluorescence in the monolayer when electrodes were placed at the indicated distances above it.

4. Discussion

We have introduced and successfully validated a novel way to focus effects remotely using the CANCAN paradigm. This method relies on the gradual change in stimulating waveform with distance, from “most bipolar” immediately next to stimulating electrodes to unipolar at the remote target. The “most bipolar” waveform has low potency to evoke bioeffects, which outweighs the high electric field strength near electrodes. In contrast, the unipolar pulse has the highest potency to evoke bioeffects, despite the electric field decline with distance. The original CANCAN approach utilized two electrically insulated bipolar waveforms, which overlapped into a unipolar pulse remotely [15, 19]. The NG-CANCAN approach is precisely the opposite of it, as it starts with unipolar pulses which produce bipolar waveshapes near electrodes but not at the remote target. While the original CANCAN was more a proof of concept than a research tool or a treatment application, the NG-CANCAN appears ready for both these tasks.

One reason for it is the simplicity and versatility of programming CANCAN stimulation using the EPULSUS generator. An endless variety of pulse packets can be designed and experimentally tested. Our study just touched upon a few of packet parameters while many others remain to be studied (pulse duration, amplitude decrement per phase, packet frequency and rotation rates, etc.) In contrast to the original CANCAN, where any change applies always to a pair of electrodes, NG-CANCAN enables tuning the pulse protocol for each of four electrodes separately and independently. This option goes beyond a simple advantage of selecting better targeting options and should allow deliberate focusing of bioeffects to a desired location off the center of the quadrupole array. Future experiments will need to demonstrate how tuning pulse amplitudes from each channel enables focusing of effects to a chosen remote location.

For the first time, we were able to confirm experimentally that the efficiency of CANCAN targeting can be predicted from simple experiments which measure electroporation between two electrodes. We tested three different types of bipolar signals, each capable of bipolar cancellation, but one being substantially better at it (i.e., produced less electroporation). Three packets for CANCAN targeting were designed to produce three types of bipolar signals near electrodes and we showed that, indeed, the strongest bipolar cancellation has led to the most efficient remote targeting (Figs. 4 and 8).

In our study, the efficiency of remote targeting was defined as a ratio of electroporation effects near the electrodes and remote from them, in the center of the array. However, there are obviously other parameters to be considered, such as the width (or “sharpness”) of the remote stimulation peak and the depth and width of the electroporation “void” between the strong effect near electrodes and the remote peak of the effect in the center of the quadrupole. Some CANCAN applications, such as tumor ablation, will benefit from a uniform effect within the array, without any “void”. Other potential applications, such as deep brain stimulation, will likely need as sharp remote effect as possible, with a deep “void” (no effect) elsewhere. While we have fair understanding of the effects expected near electrodes (where cells are exposed to a bipolar multiphasic signal) and at the target (where cells are exposed to a unipolar pulse), the effects at intermediate locations remain an uncharted territory. The problem is that the electric field vector at intermediate locations changes from the first phase to the next phase by an angle which is different from either 0 or 180°. We know that a vector change by 180° forms a bipolar pulse that will cause bipolar cancellation; however, we do not know if cancellation will still occur (or how much weaker it will be) when the field vector changes by, say, 120° or 90°. We are finalizing the first study that connects the efficiency of electroporation with the vector angle change between two phases of the pulse, and the results will be reported soon.

A critical question related to remote targeting is the scalability of results obtained in model studies with a small electrode array to physiologically relevant distances at animal and human scale. This was indeed one of the reasons why we tested two electrode arrays of about twofold different size (Fig. 3). The electric field distribution and strength achieved within each array at a given pulse voltage were obviously different, but the remote targeting, at least within studied limits, did not depend on either the electric pulse amplitude or array size (Fig. 7). We infer that findings from small electrode arrays can be extrapolated to larger inter-electrode distances.

Measurements of CANCAN targeting in the simulated 3D environment (Fig. 9) provide further encouragement for in vivo CANCAN applications. Remote effects can deliberately be targeted to a location off the plain containing electrode tips. Moreover, any effects immediately near electrodes can be prevented simply by introducing a spacer between the electrodes and the human body. Such a spacer will just slightly reduce the electric field at the remote target while preventing any stimulation near electrodes. An added advantage of such a spacer is that it could take away any heat produced by the high electric field near electrodes, thereby enabling high frequency bipolar stimulation protocols which showed the best cancellation efficiency[24].

The next studies will need to focus on optimizing CANCAN protocols for the highest remote stimulation efficiency and extending them from electroporation to neuromuscular stimulation and cytosolic Ca²⁺ mobilization effects.

nsEP interference enabled stronger electroporation remotely than near electrodes

Remote focusing was accomplished with packets or unipolar nanosecond pulses

Remote effects were 1.5–2 times stronger despite a 3–4-fold weaker electric field

The efficiency of remote focusing was determined by bipolar cancellation