Satellite BPEsIn previous work27, we have shown diamond BPEs out-perform round or triangular BPEs with regards to linkage formation. An electrode point was mathematically modelled by two spherical electrodes connected by a wire35. Based upon this we hypothesise MW growth will be enhanced by incorporating smaller BPEs close to the larger BPEs which will lower cellular resistance providing directional cues for the pattering of microwire growth. To evaluate this, silver MW growth with smaller BPEs incorporated were investigated (Fig. 3). Single droplet BPEs were explored (30–50 mm), herein referred to as satellites. As demonstrated in Fig. 3A–D, satellite BPEs were printed around larger BPEs, and comparisons to larger BPEs were evaluated. This demonstrated that incorporation of satellite BPEs significantly increased the number of wires grown against BPEs alone (4.1 ± 0.73 to 7.2 ± 0.69, respectively (n = 20) as shown in Fig. 3E). This finding will impact the design of systems for the wireless growth of MWs and suggests that multiple-sized electrodes should be incorporated. We suggest this occurs due to a drop in cell resistance.Fig. 3Comparison of number of MWs grown using samples with and without satellite BPEs. Multiple geometries were used in pairs, without satellite BPEs (A, B) and with (C, D) satellite BPEs (scale bar = 250 μm). (E) Mean + /− SEM of number of MWs grown demonstrates a significant increase in MW growth in the presence of BPEs (P < 0.05) (n = 20).As per Eqs. 4 and 5, it is hypothesised this observed behaviour is due to smaller volumes possessing a higher charge density resulting in an enhanced electric field. Hence, electrodes with a pointed edge, rather than a smooth/rounded edge, are predicted to possess improved MW growth due to an enhanced electric field at their points. We have recently substantiated this with a nano-bipolar electrode which results in the need to apply smaller currents than predicted29. Whilst smaller BPEs concentrate charge to act as nucleation sites for MW growth to initiate, larger scale BPEs are still required to lower the overall impedance of the system and allow for lower potentials to be used36.$$\nabla \cdot E= \frac{\rho }{{\epsilon }_{0}}$$
(4)
where \(\nabla =\) Divergence, E = electric field, \(\rho =\) charge density and \({\epsilon }_{0}\) = permittivity$${\uprho } = \frac{{\text{q}}}{{\text{V}}} $$
(5)
where q = charge and V = volume of electrode.AC MW growthWe next sought to establish and optimised the MW growth procedure for later cellular studies. We have previously demonstrated that AC stimulation may improve cell viability in a bipolar electrode system27. With the introduction of a frequency component, we hypothesised that this could allow for alternative electrical inputs that are less harmful to cells. Tissue damage is often a result of water electrolysis and noxious by-product from chemical reactions; utilising AC stimulation may reduce this due to less charge building up. AC stimulation has been used previously to grow MWs, although these also required the use of harmful agents that would not be suitable for use with biology19,23. Small single drop BPEs (30–50 µm) were incorporated between two FEs, replicating previous DC studies27, using varying potential ranges (1–100 V) and frequency ranges (1 Hz–1 MHz), and stimulation was applied to attempt to elicit MW growth.As shown in Fig. 4, MW growth was not possible at 10 V for any frequency, similarly, MW growth was not possible at 1 kHz. The exact reason for this is unclear and requires further investigation. Figure 4 insets I–VI show varying frequencies producing observably different MWs. At frequencies > 100 Hz, broken MWs form that are 0.5–1 mm; these MWs form when using 100 V, or by applying 100 V for 2 s before lowering to 50 V. Higher Frequencies (10 kHz) initially produce MWs with a wider diameter, with large amounts of amorphous growth. Wire thickness gradually decreases becoming less amorphous to ~ 1 mm at 1 MHz. This proportionality of decreasing diameter at higher frequencies is comparable to AC MW growth using other materials22,23, although wire thickness is magnitudes larger. Ten and 100 kHz allowed for MWs to be grown at 50 V when applying 2 s of 100 V to initiate MW growth. Furthermore, 1 MHz induced growing MWs at 50 V, 100 kHz growing MWs between 60 and 70 V, and 10 kHz growing MWs between 80 and 90 V.Fig. 4AC MW growth attempted at varying frequencies and potentials. Characteristic wires for varying frequencies at 100 V (or 50 V at 1 MHz) can be seen in images I–VI.One of the critical advancements ultimately needed to facilitate in vivo MW growth is a method to non-invasively determine if growth has occurred. Visualising wire growth within 3D cell cultures is challenging, and these difficulties will likely be amplified in tissue samples and animal models. Therefore, a wireless monitoring technique for wire growth would be highly valuable.Electrochemical Impedance Spectroscopy (EIS) has demonstrated its effectiveness in probing nano-bipolar electrodes (nano-BPEs), making it a promising candidate for this application. Initial studies, conducted in the absence of cells, involved measuring the impedance BES before and after wire growth. A potentiostat was connected to two BPEs (illustrated in Fig. 5A), and EIS measurements were performed in galvanostatic mode during MW growth, resulting in Bode plots (Fig. 5B). EIS was used to assess the overall BES impedance before and after MW growth. Results (Fig. 6) consistently showed that MW growth led to a reduction in the BES’s impedance.Fig. 5Impedance measurements during wire growth, Set up (A) and example bode plot in galvanostatic mode (B).Fig. 6Example wire growth samples can be seen in A and B, scale bars show 100 μm. Impedance measurements before and after wire growth (C).The Bode plot depicted in Fig. 5B captures the impedance dynamics under galvanostatic conditions, where the current remains constant, across varying frequencies during the growth of microwires. Unlike cyclic voltammetry, where time is a direct variable, this plot represents impedance as a function of frequency. The directional indication on the graph refers to the progression of the impedance scan over time. Although the plot initially shows an higher impedance with frequency, the overall trend indicates a decrease in impedance as the microwire matrix develops and the system’s conductivity enhances leading to lower impedance at lower frequencies when analysed.Contrastingly, Fig. 6B presents a comparative view of impedance measurements taken before and after microwire growth. This figure demonstrates a decrease in impedance as a result of the formation of conductive paths within the system, reflecting a static comparison rather than the dynamic changes during microwire formation observed in Fig. 5B. This distinction is crucial for understanding the dynamic versus static nature of impedance changes associated with microwire growth and system conductance.3D cell culture MW growth proof of conceptWhilst others have applied 3D Kirigami electronics for interfacing with 3-dimonsional culture. These can still not be assembled in situ37. To utilise this method for wirelessly growing in-situ bioelectronics, it is necessary to control microwire growth in the presence of biology. As we have demonstrated, microwire growth is possible in a 2D space, however, the next stage requires microwire growth in a 3D cell culture model. Glioblastoma cells (U251) were chosen for future applications of in vivo grown bioelectronics in neural-based therapeutics. Spheroids were attached to BPEs between two FEs (1 mm apart) using a coating of poly-D-lysine. Average spheroid diameter ranged from \(\sim\) 100 to 400 μm, hence 2–3 spheroids were attached between FEs that were 1 mm apart. AC electrical stimulation was then used to attempt wire growth in the presence of the spheroids where; identical electrical inputs as above were used. As shown in Fig. 7, MW growth was possible at 50 V (0.5 kV/cm), 1 MHz; however, these were limited in length (50 mm) and required high potential, resulting in cell clumping. It was expected that wire growth was not possible at other inputs due to spheroids introducing high impedance to the BES.Fig. 7(A, B) Optical images of MW growth using 50 V at 1 MHz in the presence of 2D monolayers. Arrows highlight MWs. Scale bars represent 100 μm.As AC stimulation was not capable of growing MWs in the presence of cell spheroids, DC stimulation was pursued. As shown in Fig. 8, MW growth was successful when using 50 V (0.5 kV/cm) of DC applied to a single spheroid of approximately 800 µm in diameter. Wires were grown at 50 V using pulsed power supply in 10 pulses of 300 ms with a gap between pulses of 200 ms for until wires had been observed to have grown for approximately 60 s. It can be observed that MWs successfully grew around and underneath spheroids with a diameter of 7.89 ± 0.6 µm (n = 50), which is similar to electrical inputs without cells (Fig. 6). This suggests that the wire structure is similar to studies formed from AgNPs alone. Fig. 8Optical images of Ag MW growth in the presence of spheroids. Green dotted outlined shows the main bulk of the spheroid. (A) whole image of MW growth around a spheroid, (B–D) Higher magnification of spheroid perimeter with Trypan blue stain to demonstrate MW growth.To approximate the effect of MW growth on cell viability, trypan blue was applied to spheroids (Fig. 8B–D). This suggested that a small population of cells had broken away from the surface of the spheroid and were not viable; nevertheless the main cell population’s membranes appeared still intact.
Bipolar electrochemical growth of conductive microwires for cancer spheroid integration: a step forward in conductive biological circuitry
