Synthesis, Characterisation and Cytotoxicity of Gold Nanowires for Ultra-Sensitive Biosensor Development

With the long-term goal of developing an ultra-sensitive microcantilever-based nano biosensor for versatile biomarker detection, new controlled bioreceptor-analytes systems are being explored to overcome the disadvantages of conventional ones. Gold nanowires (AuNWs) have been used as a probe to overcome the tolerance problem that occurs in response to changes in environmental conditions. However, the cytotoxicity of AuNWs is still unclear. Here, we examined the cytotoxicity of AuNWs systems using both commercial and as-synthesised AuNWs. In vitro experiments show that commercial AuNWs with an average quoted length of 6 µm are highly toxic against Gram-negative Escherichia coli (E. coli) at 50 µg/ml. However, this toxicity is due to the presence of CTAB surfactant not by the nanostructure. Conversely, the as-synthesised AuNWs with an average length of 9.5 µm show non-cytotoxicity even at the maximum viable concentration (330µg/ml). These ndings may lead to the development of potentially life-saving cytotoxicity -free nano biosensors for an early diagnostic of potential diseases.


Introduction
A wide array of biomarkers secreted by cancerous cells can be found in blood samples [1][2][3][4][5][6]. With current emerging proteomics technologies, it is straightforward to obtain reliable sets of disease-speci c protein biomarkers. Early identi cation of disease blood protein signatures, is becoming the most promising strategy for effective cancer prevention [7][8][9][10]. Nevertheless, some protein biomarkers, such as those secreted by cell death in tumours microenvironment at a very early growth period, are produced only at ultra-low concentrations [11][12][13][14]. Low reproducibility highlights the di culty in detecting low concentrations of disease protein biomarkers. With concentration, the scales are six to seven orders of magnitude lower than the plasma protein (e.g. the reference range of albumin protein in blood is approximately 35-55 g/L [15,16]), making their detection challenging. Engineered nanomaterials-based biosensors with high sensitivity levels have been the focus of improved biomarker detection technology [17][18][19][20][21]. These advanced detection methods depend on the antibody and protein for recognition, identi cation and quanti cation of targeted cells.
There are several drawbacks on microcantilever (MC) biosensor, MC requires a high amount of analytes and should be sensitive enough to measure de ection in the range of few nanometres leads [22][23][24] to the integration of micro uid nano-biosensor layer for protein-protein interaction detection [25]. Hence, nanomaterial-based MC biosensors have good potential in advanced biosensor technology as they offer a high surface-to-volume ratio, so they are able to mediate faster with a higher kinetic electron transfer, allowing the merit of multiple sensing mechanisms in a single platform [26][27][28][29]. The size of nanomaterials can facilitate effective interaction depending on the nature of the chemical bonding with targeted biomolecules in opaque liquids such as blood or urine [30][31][32][33][34]. Furthermore, since the dimension of nanomaterials is comparable to the Debye length, their speci c surface properties affect the kinetics and electron transfer signi cantly [35][36][37]. Offering potential for electronic detection surface stress functionalisation as a result of the MC bending by a certain extent generated by analyte-ligand interactions in the biomarker layer.
Gold (Au) based one-dimensional (1-D) nanomaterial is a candidate for achieving an ultrasensitive MC biosensor device [38][39][40][41][42][43]. Speci cally, for exible chemo resistive sensors.Au nanowires (AuNWs) are attractive because they possess minimal cross-sectional area but have the ability to increase the oodcurrent along the axial current direction, resulting in higher conductance changes compared to the typical zero-dimensional (0-D) nanomaterials [44,45]. The chemoreceptive characteristics of AuNWs such as the capability to serve as biocompatible surface for the immobilisation of biomolecules in a correct orientation which enhances the binding a nity of antigen-antibody and the ability to support a swift electron-transfer which causes the ow of one-way electron between two electrodes are key considerations in designing the MC nano biosensor [46][47][48][49][50][51]. Chemoreceptive and also conductometric characteristics played by the AuNWs are both important for biosensor transduction mechanism.
On the other hand, there have been many concerns about the potential cytotoxicity of nanomaterials in general, which may arise upon medical application [52][53][54][55]. It has been proved that several quantum materials consisting of heavy metals may release ions that may cause cytotoxic effects. Although AuNWs are not considered toxic in most cases, there is a certain ambiguity about AuNWs toxicity impact regarding variability and threshold for speci c cell types [56][57][58]. Therefore, there is an urgent need to present new data that can assist in developing a mass-sensitive MC-based AuNWs biosensor. Important primary parameters that are involved in cytotoxicity assessments of AuNWs include the aspect ratio of nanowires, surface functionalisation method, cell type, administration of dosage and application protocols. In this work, we describe the synthesis of bio-friendly AuNWs for MC biosensor application. The objective is to ascertain the maximum dose of as-synthesised AuNWs, which is safe to avoid initial cytotoxicity whilst increasing the sensitivity of the nanowires array in comparison with commercial AuNWs suspended in CTAB. Bacteria cells are the most sensitive cells to detect toxicity arising from either free or nanostructured metallic ions. For toxicity detection, we use a set of antimicrobial approaches. The transduction mechanism of nanosensors depends on the binding of the protein of interest change which yields different conductivity output. The AuNWs play a vital role in the transduction mechanism. Hence it is essential to maximise the concentration of AuNWs to achieve maximum sensitivity. The investigation of the transduction mechanism is not within the scope of this study. The assembling nanowires on the top surface of the microscopy probe is also evaluated but not discuss further in the present study and only included in the Supplementary Materials section. Any biomarker should be able to be detected using this platform.
Commercial AuNWs (dispersion in H 2 O, contains cetyl trimethyl ammonium bromide, CTAB as a stabiliser) used was purchased from Merck Sigma-Aldrich, United Kingdom.

Microbial Cell Culture Preparation
The commercial Escherichia Coli (E.coli) DH5-Alpha strain (DH5α), an engineered non-pathogenic strain for routine subcloning procedures is used in the present work. It is quick and easy to grow and has been used extensively as a lab microbial model system. (ThermoFisher catalogue number 18263618). E.coli cells were grown in Luria-Bertani medium (LB), rich medium for cell growth. The cells were initially grown from an original culture sample, kept at -80°C. Cells were inoculated onto an LB agar plate and incubated at 37°C for 24 hours. At this point, a number of individual colonies had grown, each from a single cell. To reduce variation across the bacteria under study, a single colony was selected and incubated in LB to produce a working cell solution. This was used as the master cell culture for all the bacteria cell experiments. Optical cell density measurements were acquired using a Perkin Elmer Lambda Bio + Spectrophotometer (Germany). The spectrophotometric method measured the optical density at 600 nm based on an American Public Health Association (APHA) standard. The optical density (OD) of 0 is set for pure LB, and the spectrophotometer is calibrated so that the OD equal to 1.00 corresponds to a cell density of 109 ± 5 x108/ml. A linear relationship between OD and cell density is achieved over this range.

Disk Diffusion Assay
Cytotoxicity test is carried out through the agar disk diffusion assay method which has been used in many clinical microbiology laboratories for routine antimicrobial susceptibility testing. The standard measurement is based on the Clinical Laboratory Standard Institute (CLSI) and manufacturers' standard for each sample. The AuNWs colloidal solution (300 µg/ml) was suspended in distilled water and this portion deposited on the sterile paper disk that is 6 mm in diameter with 4 replicate disk per sample. The disk was placed on Gram-negative E. coli DH5α strain. Controls were prepared using pure hexane and/or distilled water.
E.coli DH5α were inoculated at a concentration of 107 to 108 cells/ml at stationary phase strain onto Luria-Bertani (LB) agar. The LB agar plates were then inoculated and incubated for 24 hours at 37°C to record the cytotoxicity effect. The size of the inhibition zone around the disc is control by the interchange between the AuNWs diffusion rate (Eq. 1), the exponential growth rate of the bacteria and the minimum inhibitory concentration which is usually higher than the breakpoint.
The C is referring to the AuNWs concentration at time (t) and distance (r) from the point source (disc centre) with the diffusively isotropic assumption in a 2-D plate form. The D is the corresponding diffusion coe cient and A is the amount of initial AuNWs used in the disc with h is the height of the agar in the plate. The concentration of AuNWs in the inhibition zone could drop below the minimum inhibitory concentration after overnight incubation. By this time, the diffusion of the nutrient in the plate occurs and gradually depleted by the bacteria that grew in the inhibition zone periphery. Overnight incubation made the cells cover almost 70% of the growing surface, except for the circles of growth surrounding disc which remained clear. The diameters of these circles are then measured and used as measurements of the inhibitory effects of the sample.

Colony Forming Unit Counting Method (CFU)
Methods for in vitro evaluation of antimicrobial activity, seek to assess the total number of E.Coli cells in culture media [59]. Colony-forming units (CFU) of E. coli DH5α were counted after plating triplicate serial dilutions of LB cultures on LB-Agar plates and incubating overnight at 37°C. Plates containing between 2 and 200 individually identi able colonies were counted. The relative number of colonies on a plate compared to the control gives a value which is a measurement of the toxicity.

Characterisation of As-Synthesised and Commercial AuNWs
AuNWs were resuspended and sonicated in ethanol medium. And let it dry on an aluminium substrate at room temperature. Field emission scanning electron microscopy (FESEM) imaging was conducted by a JEOL JSM 7500 F microscope equipped with a cold eld emission gun at an acceleration voltage of 15 kV. An in-lens detector collected secondary electrons and backscattered electrons signals and in-chamber multi-quadrant annular retractable solid-state detector, respectively. Element analysis was carried out using Zeiss Neon 40EsB (Carl Zeiss NTS GmbH, Oberkochen, Germany) at 20kV equipped with energy dispersive x-ray spectroscopy, EDX (Inca software, Oxford Instruments). Image analysis on FESEM data was carried out using ImageJ 1.50i. AuNWs density was determined using the plugin 'analyse particles' function of ImageJ on a total of 20 SEM images of AuNWs samples. The aggregated size of AuNWs was calculated from a total of 2000 randomly picked aggregated AuNWs from FESEM images.

Synthesis of Ultra-Thin AuNWs
AuNWs were synthesised based on the method described in the literature [60]. In the absence of heat, the synthesis reaction involved mixing of 100µl OA and 150 µl TIPS with about 3 to 5 mg of gold (III) chloride precursor solution, (HAuCl 4 99.9% trace metals basis, 30 wt.% in dilute HCl) in 2.5 ml nal volume of hexane without stirring. OA acts as a stabiliser and a template for 1-D growth with TIPS role as a highly reactive reducing agent. The observation of the reaction through the change in colour from light orange to red, followed by dark violet during the period of reaction left out at room temperature is demonstrated in Figure 1. The AuNWs products were centrifuged, washed with ethanol at least 3 times and nally redispersed in hexane for further characterisations. Nanowires produced, according to the published method, have diameters of about 2 nm, lengths of a few µm and self-assembled on deposition into organised structures. In this work, the resulting AuNWs were 3 nm in diameter with a standard deviation of 0.8 %. During the synthesis reaction, the potential of Au + reduction to Au is higher in the presence of OA [61][62][63][64]. The primary limiting factor for constant electron transfer rate would be the OA reagent. The relative concentration of the AuNWs obtained was > 50µg/ml. This method can also be applicable for the synthesis of other metal nanowires as long as the chemical combination is suitable and correct.

Morphological Properties of Synthesised Ultra-Thin AuNWs and Commercial AuNWs
The as-synthesised AuNWs samples obtained were characterised for their morphological properties by standard techniques. The morphology, distribution and purity of the as-synthesised AuNWs veri ed by FESEM are shown in Figure 2A. For a 24 h reaction, the resultant colloid colour is dark violet which indicates extended lament-like structures. The AuNWs tend to self-assemble into 1-D networks and forming closely packed parallel structures. The use of negatively charged TIPS in the synthetic reaction at room temperature controls the acceleration of the process [65,66]. The size distribution of parallel AuNWs bundles was di cult to measure. Higher magni cation of the assembled ultra-thin AuNWs was unachievable due to their highly sensitive towards the electron beam which resulted in melted nanowires within a few seconds of exposure. However, it is expected that the ultra-thin AuNWs possess diameters of approximately smaller than 3 nm with an aspect ratio (length-to-width ratio) above 1000 nm. In this case, the AuNWs tended to form a stack of parallel bundles which then self-assembled into 1-D network structures over macroscopic distances through a spontaneous directional aggregation that occurs during solvent evaporation. The directional aggregation is typically formed via oriented attachment in which AuNWs are permitted to fuse as the chemical potential between each chain is different. Therefore, the smoothing extension process to interconnect the nanoparticles of the nanowires takes place through diffusion. The AuNWs exhibit higher stability in a polar solvent which is favourable for the subsequent immobilisation of biomolecules. The ultra-thin self-assembled networks of individual crystalline AuNWs with a gap distance of ~2 nm can be of use to trap the analyte molecules and automatically locate them within the gap of closely packed parallel AuNWs making them as suitable substrate candidates for SERS studies due to the presence of closely packed 'hotspot'.
The primary objective of this work is to produce ultra-thin AuNWs with high aspect ratio and analyse the elements of samples with the quantitative and qualitative analyses properties. Figure 2B shows the element composition of obtained AuNWs samples using EDX. The EDX point measurements were carried out for an accurate estimation of Au amount present in each sample. For as-synthesised AuNWs, the amount of Au is at the level of 98% as the major element in the sample with the remaining amount is Si (substrate). The presence of high purity Au has resulted from homogeneous nucleation of metallic Au as spherical clusters and hence determined the growth of 1-D nanowires. This condition proves that in the presence of TIPS, Au seeds have rapid kinetics in the formation of nanowires.
On the other hand, the representative FESEM images of commercial AuNWs suspended in CTAB aqueous solution are shown in Figure 3. It has been observed that the network formed does not assemble into ordered nanostructures ( Figure 3A) as seen in the as-synthesised AuNWs sample beforehand. Instead, the commercial AuNWs are randomly self-gating forming junction connection. Some of the AuNWs showed thicker diameter with reduced lengths as demonstrated in Figures 3B and 3C. The AuNWs can be as long as 171 m ( Figure 3D). One can see that the diameters of the nanowires were averagely thin and approximately ~540 nm as indicated in Figure 3E. This type of trend is usually expected and frequently occurs in many chemical reactions [67][68][69][70]. This mainly caused by the surface diffusion of precursor that could not penetrate deeper into the bulk of AuNWs at a shorter reaction time which then caused the increased in diameter. The average length of nanowires, typically remained the same caused by the similar diffusion of intense high energy ions in the surrounding. The polydispersity distribution of the commercial AuNWs is shown in Figure 3F. The narrow length distribution range (up to ~ 30 m) indicates that the commercial AuNWs were stable and monodispersed. The wide distribution range is due to the more considerable length variation, which is up to ~ 171 m.
It shall be noted that the cloudy region encircling AuNWs is due to the strongly coupled capping action of CTAB upon the evaporation of the solvent, leaving the residue behind. Since cationic CTAB surfactant consists of long carbon atomic chain [71][72][73][74], it functions as space-lling of secondary material that lls up the gaps between the AuNWs and maintains the individual distribution of AuNWs. The FESEM micrographs distinctly manifest dissimilarity of CTAB grain with AuNWs, exhibiting large-sized ake grains (marked with the blue arrow in Figure 3). As stated in few reports [75][76][77], the tertiary ammonium ion cationic headgroup in the presence of bromide anion as a counterpart in nanowire solution results in higher binding a nity and leads to more stable bilayer on the nanowire surface. The addition of CTAB micelle helps to stabilise AuNWs when the equilibrium condition is achieved through the occupation some of the surface areas caused by the increased in the aggregation number of CTAB micelle. This stabilisation effect, while making them soluble, can be an inhibitor of subsequent self-assembling process.

Cytotoxicity Evaluation of As-Synthesised and Commercial AuNWs
The proof-of-concept on the cytotoxicity of the nanosensors-based nanowires requires long acquisition times. Some of the nanowires conventional synthesis protocols, such as the reduction of Au precursor in the presence of capping agent can introduce toxic materials from surface conditioning and chemistry [78,79]. These including surface charge and capping stabilisation without 'green' physical approach pro ted from natural nanostructure generation [80,81]. Additionally, the metal itself can produce toxicity and thus, being unfavourable for in-vivo applications. Taking into consideration the unique advantages of AuNWs, different capping interface, these types of AuNWs exhibits a signi cant effect in microbial viability. For this reason, we examined the potential cytotoxicity propensity for commercial and as-synthesised of AuNWs.
To further demonstrated the cytotoxicity tolerance in as-synthesised and commercial AuNWs samples, the commercial bacteria E.coli DH5α, an engineered non-pathogenic bacterial strain is extensively used as a lab cytotoxicity microbial model system. Two types of cytotoxicity test were carried out. The rst is the disc diffusion assay. This method is performed to detect cytotoxicity by inducing a gradient of concentration ( Figure 4A) around a disc (known as inhibitory effect) loaded with AuNWs. The inhibitory effect among the E.coli DH5α strains exhibited different outcomes. The ratio of the ring area (measure in mm) is directly proportional to the sample toxicity (white lines) as indicated in Figures 4B and 4C. The results show that the gradient of commercial AuNWs suspended in CTAB halts the growth of bacteria and creates an inhibition zone around the disc in both the 10 and 50 g/ml samples. Lower inhibition zone was found for 1000 g/ml Au ions. For these types of Au-based suspensions, the largest inhibition zone reached 3.9 0.2 mm, 2.9 0.3 mm and 1.2 0.1 mm for 50 g/ml commercial AuNWs-CTAB, 10 g/ml commercial AuNWs-CTAB and 1000 g/ml Au ions, respectively. Non-toxic samples with no inhibitory effect which refer to water ( Figure 4D) and hexane ( Figure 4E) samples. A similar observation was found for 400 g/ml Au ions and <330 g/ml as-synthesised AuNWs suspended in hexane. The six samples used to observe the average diameter of inhibition zones containing Au-based solutions is presented in Figure  4F.
The size of as-synthesised AuNWs prepared with growth solution contained AuNWs with a smaller length.
Since hexane is a non-polar solvent, it has no charge. Thus, the conjugated surface of AuNWs with negatively charge of the hydrophobic OA, yield a negative surface charge of the AuNWs system [82,83]. As a result, no electrostatic interactions due to absent of different charges in the solution between the cell surface and as -synthesised AuNWs suspended in hexane. As generally accepted, CTAB alone is highly susceptible and rapid killing strains. F. Disc diffusion measurements obtained using several concentrations of Au ions and both the as-synthesised and commercial AuNWs. The bars represent one standard error. toxic cationic surfactant [84][85][86] and the conjugate systems of AuNWs-CTAB are proven to still contribute to toxicity to culture cells especially when the elevated concentrations of CTAB are present (50 g/ml commercial AuNWs-CTAB). With increasing the amount of AuNWs-CTAB solution, the cytotoxicity of the sample is increased caused by the nonspeci c binding tendency of CTAB to negatively charged cell surfaces by electrostatic interactions [87,88]. Once the interaction with the cell occurs, it forms blebs and holes on the cell and leading to cell death evident by the increase in the diameter of the ring zone.
Moreover, the data also suggest that the abilities to induced cell death are also signi cant regardless of the aspect ratio of the AuNWs, in this case, the aspect ratio is constant for both 10 and 50 g/ml commercial AuNWs-CTAB solution. To minimise the cytotoxicity effects in AuNWs-CTAB, the interaction between the membrane cell and CTAB should be blocked by using protein types of coating on the bilayer surface of CTAB. Meanwhile, a similar trend of nonspeci c binding also observed when Au ions are incubated with the cell. The nonspeci c binding was evident by the increased of the ring zone when exposed to the concentration of the 1000 g/ml Au ions. This nonspeci c binding did not take place with 400 g/ml Au ions might be due to the reduced content of positively charge Au ions, which in turn, decreases electrostatic interaction between the Au ions and negatively charged cell membrane. These results a rm cytotoxicity potency of commercial AuNWs -CTAB and Au ions compared to as-synthesised AuNWs suspended in hexane.
For further validation of cytotoxicity detection on commercial and as-synthesised AuNWs, the CFU enumeration is carried out after 24 h incubation ( Figure 5A). As evident from Figure 5B, the CFU enumeration for 10 g/ml Au ion measured viability of 72%. At 30 g/ml Au ion approximates value were almost similar to the other one giving rise to 69%. Cell-viability assessed by CFU remained at 100% on the control (water) sample. Compared with 10 and 30 g/ml of Au ions, the cell incubated with a higher concentration of Au ions (100 g/ml) was observed to be 0% cell viability. In the case of 2 g/ml of commercial AuNWs suspended in CTAB, <300 g/ml as-synthesised AuNWs and hexane also, 0% were viable for all in CFU enumeration. Pure hexane, water and the ionic Au at three different concentrations (10, 30 and 100 g/ml) were used as negative controls. Assessment at 96 hours was likely to be optimal for AuNWs strain. Indeed, the effect of both commercial and as-synthesised AuNWs along with a high dosage of 100 g/ml Au-based ion seemed strong antibacterial agent in this assay. Considering the high amount concentration use for as-synthesised AuNWs, it is suggested that the AuNWs work markedly better, meaning presents less bacteria toxicity than Au ions (tenfold lower in concentration) as happened to other metallic materials.
Endowing the high antibacterial potency or toxicity effect based on the viability results for both types of AuNWs at those particular concentrations, another study is carried out to compare the inhibiting capability of commercial AuNWs-CTAB with AuNWs-sodium bicarbonate (NaHCO 3 ) towards the cell growth. The focused study is purposely done on the commercial AuNWs-CTAB since a relatively low amount of sample concentration used proved to affect the cell viability signi cantly. To further clarify the inhibitory antibacterial activity induced by the CTAB layer, commercial AuNWs were puri ed from CTAB capping layer and re-dispersed in 2 mM NaHCO 3 . Cells grown in LB media with AuNWs-CTAB and AuNWs NaHCO 3 with an approximate volume of 250 µl for both were observed. Water served as the standard control. The cell-nanowire-LB solutions were approximately 5µg/ml, incubated for 90 minutes and was serially diluted to 105 times of the initial concentration with the set to dispense for 100 µl onto plates. As shown in Figure 5C, CFU enumeration showed 4 % and 130% of capture e ciency for AuNWs-CTAB and AuNWs-NaHCO 3 solutions, respectively. The low CFU in the CTAB solution was found to be threefold lower than in NaHCO 3 , indicating that the toxicological effect of CTAB is relatively high, which signi cantly displayed susceptible inhibit growth of E.coli cells. NaHCO 3 had a low inhibitory effect.
The results demonstrated that surface modi cation is the decisive factor governing the cytotoxicity of AuNWs. CTAB proved the capability to induce E.coli DH5α cell death and destroy bacteria growth. As regards to the surface modi cation, the AuNWs capped with NaHCO 3 (well-known as baking soda), indicated minimal toxicity effect and had almost no effect on cell viability in an alkaline environment. The alkaline environment retarded the activity of the toxicity which inhibits the cellular uptake. When combining with NaHCO 3, the acidic environment that usually leads to cell death is halts and decreasing the risk of adverse effects and toxicity.
It is believed that the primary toxicity mechanism of the AuNWs-CTAB was via the activation of intracellular cell damage, which rigidly controls the per cent of cell survival. As the cell tries protecting itself from toxicity created by the molecule, other existing plasma proteins will adsorb on the surface of nanowires. The nanowires eventually penetrate the cells after adjusting the interfacial properties of the adsorbed protein shell. However, some of the adjustments are not always a success as not all modi ed AuNWs surface would able to enter the cells effectively. The proper conjugated AuNWs able to recognise the proteins on the cell membrane. Thus, adsorption of the protein to the surface of the nanowires mediate the direct penetration of nanowires. Typically, smaller AuNWs are frequently deemed as more toxic due to the ability to fully penetrate at intracellular locations (e.g., nucleus) which is not able to get through by larger nanostructures. Hence, since cytotoxicity was controlled by the physicochemical surface properties (surface charge number and dimensions) of AuNWs, it is crucial to employ riskless surface modi cation reagent to govern the biological effects of nanowires to ensure the safety of nanostructures used in medical applications.

Conclusions
As discussed, the synthesis of AuNWs has tailored ultra-thin nanowires that formed the irregular chainlike oriented assembly. These ultra-thin AuNWs in hexane solution are almost similar in average length and ten times smaller in diameter size as compared to commercial AuNWs suspended in CTAB solution.
However, they are unstable at relatively high temperatures, melting within a few seconds upon the heat exposure. The anionic polymer of OA was used as a capping layer to protect as-synthesised AuNWs from aggregation, which also render them to be biocompatible. The feasibility of the AuNWs reveals that the functionalised AuNWs in the present of OA was able to avoid cytotoxicity to the membrane cell. On the other hand, the commercial AuNWs in high crystallinity state are stable at elevated temperatures and the toxicity responds observed is mainly caused by the CTAB capping layer and not due to the aspect ratio of the AuNWs. In fact, the pristine AuNWs alone was generally non-toxic and proved to have low cytotoxicity when capped with NaHCO 3 upon eliminating the CTAB layer on the nanowires. These different surface modi cations have shown that the cytotoxicity of AuNWs is closely related to the original physicochemical properties of the capping layer based on the in vitro cell investigation results. Hence, the as-synthesised AuNws and the AuNWs-NaHCO 3 could be considered as the optimal options for integrating and implanting in MC biosensor devices, including high-throughput applications.