Results and Discussions
The electrochemical miniaturised silicon chip used in the present study is reported in Fig.1. It integrates four types of layout of EC-cells configured with three planar microelectrodes (WE, CE and CE) featured by different sizes and shapes (Fig 1a). Table 1 reports both the physical dimensions and the eras of each electrode/cell-type. The silicon chip has been integrated with plastic PC ring to create a reaction sampling chamber of 25 µL for each EC cell and mounted on a customised electronic board for driving the electrical signals. The final whole portable EC system was connected to PC to automatically managed the reading process (Fig. 1b-c).
The first part of our study was focused on the evaluation of the EC-cells performances of the 4 layouts present in the EC chip. With this purpose, we carried out electrochemical measurements using Osmium based redox probe ([Os(bpy)2DPPZ]Cl2) at concentration ranging from 0.1µM to 100 µM in KCl 20mM. Figure 2 reports the square wave (SW) intensity signals recorded at the oxidation peak of Os complex of 0,76V. The data show that four cells exhibit different performances with current intensities of EC1> EC2> EC4>EC3 (Fig. 2(a)), with a linear behavior in the concentration range of 0 - 50 µM of Os complex (Fig. 2(c)). This finding can be correlated with both the area of WE and the WE-CE distance cell parameters. Actually, as it is illustrated in Fig 2 (c) and (d), EC1 cell present the highest WE area and highest WE-CE distance and it has been found to the best performer (Fig 2(a)). In case of EC2, even if it has the same WE area of EC1, however it is featured by a shorter WE-CE distance leading to current intensities lower than EC1. For EC4 and EC3 the worst performances can be correlated to both the lower WE area and the shorter WE-CE distance (both the lowest for EC3). The results were also confirmed using other redox probe, such as FeII(CN)6 (data not showed).
Based on the above reported results we selected layout 1 of the EC chip for the heavy metal detection through the whole-cell sensing element using the engineeredEscherichia coli . The principle of the sensing method is schemed in Fig. 3. More in details, thanks to the genetic modification carried out on E. coli , when AsIII metal goes into the bacterium cell, the ArsR protein is released from its DNA binding site (located to the downstream of ars operon promoter sequence Pars) and bind the AsIII metal. This process has been designed taking inspiration to the intrinsic arsenate resistance mechanism of bacteria (Cervantes et al.,1994). When ArsR-AsIII binding event occurs, Pars is free to increase the expression level oflac Z gene, resulting in the synthesis of β-galactosidase (Stocker et al., 2003; Cortés-Salazar et al., 2013). The enzyme, then, cleaves the 4-aminophenyl β-D-galactopyranoside (PAPG) and produces the 4-aminophenol (PAP). Once released by the E. coli cells, PAP interacts with the WE inside the EC device and trig a redox exchange producing a current signal that is, then, detected and quantified by CV analysis to give the final output. Depending on the amount of AsIII present in the sample, E. coli produces a correspondent concentration of PAP that is electrochemically detected. Fig. 4(a) reports CV results obtained by using concentration of AsIII from 0 to 100ppb. As shown, PAP oxidation peaks are centred from +1.0 V to +1.2 V. By going from the lowest concentration of 2.5 ppb (red line) up to the highest 100 ppb (orange line) the peak intensities increased over the background (0 ppb sample) (blue dashed arrow). It can be observed a voltage shifts of the PAP oxidation peak as illustrated in the inset of Fig. 4(a) (green dashed arrow). Theses shifts correspond to + 50mV for 2.5ppb, +150 mV for 10ppb, + 300 mV for 50ppb and +350mV for 100ppb. They can be explained by the gradual adsorption of the redox product of PAP over the platinum surface of WE, affecting the electrolytic process (Chandrashekar et al., 2011; Wudarska et al., 2015; Salavagione et al., 2004; Amanullah et al., 2010).
It can be also observed that the E. coli strain exhibit a proper broad peak at about 1V (Fig. 4(a) curve 0 ppb). This is probably lead to the biological matrix of the cellular line. The reduction peak for dissolved oxygen was observed in the range-0.5 V -1.1 V.
The analysis of CV signal intensities is reported in Fig. 4(b-c). Data showed a linear current increase of PAP peaks when AsIII concentration is from 2.5 ppb to 50 ppb and a slope change from 50 ppb to 100 ppb (Fig. 4(b)). It is noteworthy that the sensor shows a detectable signal at 2.5ppb of AsIII metal below the WHO threshold (10 ppb).
The linear trend was deeper investigated using AsIII concentrations of 2.5 ppb, 5ppb, 10ppb and 50 ppb. The current intensities of CV curves as function of the AsIII concentration are plotted and linearly interpolated in Fig. 4(c). The analysis of the curve show sensitivity of 0.122 µA ppb-1 LoD of 1.5 ppb and a LoQ of 5 ppb. These values are lower than the acceptable limit of arsenite in water as defined by the WHO (LoD 10 ppb), suggesting the possibility to quantify the pollutant amount already before its dangerous accumulation in water. It is noteworthy that, compared to our previous system using the whole-cell detection for AsIII (Cortés-Salazar et al., 2013), the proposed platform has a sensitivity about 40 times higher (0.12 µA/ppb versus 0.0028 µA/ppb). This finding could be reasonable explained by best performance of our silicon-based EC-cells which planar electrodes assured a higher current intensity signal, as confirmed by the comparison of the intensity current values recorded by two systems. In particular water samples containing AsIII in concentration 10 ppb, interrogated with the proposed platform gives a current intensity value of 1.5 µA while the previous system gives a very low current value of 55 nA. It is noteworthy, this highly sensitive integrated detection was obtained without any additional reagent consumption, since the enzyme is directly produced by the sensing cells, giving an extreme precise quantification of AsIII in a small sample of water (just 20 µL), with a linear trend revealed at low concentration range, well below the 10 ppb guideline value.
The selectivity of the miniaturized biosensor was studied by testing the whole AsIII-E. coli cell portable sensor towards HgII and CdII metals. Results are summarised in Fig. 5. When HgII and CdII metals are present in the sampling water solution, no PAP signals were detected at in the range +1.0 V -1.2 V, while the reduction peaks for dissolved oxygen were still present the range-0.5 V -1.1 V (Fig 5(a-b)). Figure 5c reports the plots of the delta currents of both HgII (red line) and CdII (blue line) compared with those showed by AsIII (green line; data extracted from Fig 4b). The results confirm the specific genetic recognition operating inside the E. coli cells with the activity of Pars promoter enhanced only the by AsIII species inducing the arsR-mediated expression of β-galactosidase and the production and release of the PAP redox mediator. The selectivity is guaranteed without the need of any chemical reagent and enzymatic or nanostructural functionalizations, that can degenerate after long usages and affect the quality and cost of screening.
In order to assess the multiplex capability of our portable platform, we also included an additional test using a genetically modified E. Coli strain specific for the HgII species. This strain was modified in order to trig the β-galactosidase expression, thus the PAP production, once interacted with HgII species revealed in water. We preliminary investigated this assay using the following amount of HgII: 0, 0.25ppb, 1ppb, 2.5ppb and 10 ppb. The recorded I delta currents values are plotted in fig. 6(a). A linerar behaviour was observed in the range 0-2.5 ppb luading to a sensitivity value of 2.1 µA/ppb, LOD value of 0.1 ppb and LoQ value of 0.34 ppb (inset of Fig 5a). These starting data are optimal considering the 1ppb WHO tolerance value for total HgII in water. A good specificity was also observed since no signal was recorded in presence of unspecific AsIII and Cd II species (Fig 6(A)).
Optimization tests are in progress to improve the sensing performances of the system towards HgII species.
Conclusions
In this work we propose an innovative miniaturized electrochemical biosensing platform for the specific and high-sensitive quantification of metal ions in water sample. The platform synergically interface a biosensing module based on whole-cell using engineered Escherichia coli and a miniaturised electrochemical silicon chip integrating 4 type of EC cells configured with three planar microelectrodes and a portable EC-reader. The electrochemical characterization of the silicon chip using Os complex redox probe show that that EC1 cell show the best EC performances in terms of current intensity respect to the other ECs configuration (EC2-EC4). This was attributed to the highest WE area and WE-CE distance of EC1 that was chosen for the sensing test towards the heavy metals. The sensing mechanism relies on the selective recognition from a specific engineered E. coli towards a given metals producing the 4-aminophenol (PAP) redox active mediator detected through a cyclic voltammetry analysis. The miniaturized biosensor is able to operate a portable, robust and high-sensitivity detection showing for AsIII a sensitivity of 0.122 µA ppb-1, LoD of 1.5 ppb and a LoQ of 5 ppb. It is noteworthy that the LoD value is one order of magnitude below of the value indicated to WHO to be dangerous (10 μg/L). The system was proved to be fully versatile being effective in the detection of Hg(II) as well. A preliminary study on Hg(II) showed sensitivity value of 2.11 µA/ppb a LOD value of 0.1 ppb and LoQ value of 0.34 ppb. Also in this case, the detected LOD was ten time lower than that indicated by WHO (1 ppb).
The platform miniaturization together with the limited fluidic movement make of the proposed assay a potential portable and easy-to-use system for unspecialized personnel for outdoor application. These results pave the way for advanced sensing strategies suitable for the environmental monitoring and the public safety.