AuNPs-functionalized PANABA-MWCNTs nanocomposite-based impedimetric immunosensor for 2,4-dichlorophenoxy acetic acid detection

Author’s Accepted Manuscript AuNPs-functionalized PANABA-MWCNTs Nanocomposite-based Impedimetric Immunosensor for 2,4-Dichlorophenoxy Acetic Acid Detection Giovanni Fusco, Francesca Gallo, Cristina Tortolini, Paolo Bollella, Federica Ietto, Antonella De Mico, Andrea D’Annibale, Riccarda Antiochia, Gabriele Favero, Franco Mazzei

PII: DOI: Reference:

www.elsevier.com/locate/bios

S0956-5663(16)31012-0 http://dx.doi.org/10.1016/j.bios.2016.10.016 BIOS9238

To appear in: Biosensors and Bioelectronic Received date: 16 June 2016 Revised date: 4 October 2016 Accepted date: 5 October 2016 Cite this article as: Giovanni Fusco, Francesca Gallo, Cristina Tortolini, Paolo Bollella, Federica Ietto, Antonella De Mico, Andrea D’Annibale, Riccarda Antiochia, Gabriele Favero and Franco Mazzei, AuNPs-functionalized PANABA-MWCNTs Nanocomposite-based Impedimetric Immunosensor for 2,4-Dichlorophenoxy Acetic Acid Detection, Biosensors and Bioelectronic, http://dx.doi.org/10.1016/j.bios.2016.10.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AuNPs-functionalized PANABA-MWCNTs Nanocomposite-based Impedimetric Immunosensor for 2,4-Dichlorophenoxy Acetic Acid Detection Giovanni Fuscoa,b, Francesca Galloa, Cristina Tortolinia, Paolo Bollellaa, Federica Iettod, Antonella De Micob,c, Andrea D’Annibaleb, Riccarda Antiochiaa, Gabriele Faveroa*, Franco Mazzeia a

Department of Chemistry and Drug Technologies, Sapienza University of Rome, Italy b c

d

Department of Chemistry, Sapienza University of Rome, Italy

Institute of Molecular Biology and Pathology - National Research Council, Italy

Department of technological innovations and safety of plants, products and anthropic settlements, National Institute for Insurance against Accidents at Work, Rome, Italy

*

Corresponding author: [email protected]

Abstract In this work, we developed an impedimetric label-free immunosensor for the detection of 2,4Dichlorophenoxy Acetic Acid (2,4-D) herbicide either in standard solution and spiked real samples. For this purpose, we prepared by electropolymerization a conductive polymer poly-(aniline-co-3-aminobenzoic acid) (PANABA) then we immobilized anti-2,4-D antibody onto a nanocomposite AuNPs-PANABA-MWCNTs employing the carboxylic moieties as anchor sites. The nanocomposite was synthesized by electrochemical polymerization of aniline and 3-aminobenzoic acid, in the presence of a dispersion of gold nanoparticles, onto a multi-walled carbon nanotubes-based screen printed electrode. Aniline-based copolymer, modified with the nanomaterials, allowed to enhance the electrode conductivity thus obtaining a more sensitive antigen detection. The impedimetric measurements were carried out by electrochemical impedance spectroscopy (EIS) in faradic condition by using Fe(CN)63-/4- as redox probe. The developed impedimetric immunosensor displayed a wide linearity range towards 2,4-D (1-100 ppb), good repeatability (RSD 6%), stability and a LOD (0.3 ppb) lower than herbicide emission limits.

Keywords:

Nanocomposite,

Polyaniline,

Au

nanoparticles,

Impedimetric

immunosensor,

2,4-

dichlorophenoxy acetic acid, Screen Printed Electrodes

1. Introduction One of the best promising application fields of the biosensors technology is the screening of pollutants in real sample matrices, due to the selectivity towards the analyte and the rapid, reproducible and possibly quantitative or semi-quantitative response. In this context, our group has developed some inhibition-based on

1

enzymatic electrochemical biosensors for the detection of environmental contaminants (Tortolini et al. 2015; Bollella et al. 2016; Tortolini et al. 2016). Impedimetric immunosensors have recently received particular attention because they allow a label-free detection based on electrochemical transduction; also, the sensitivity of the electrochemical impedance spectroscopy (EIS) can reach very low LOD (Holford et al. 2013; Yang et al. 2014; Billah et al. 2012). Conductive polymers were extensively used to increase the conductivity of the sensing element to develop new biosensor platforms (Bartlett and Birkin. 1993; Wallace et al. 1999; Ahuja et al. 2007; Gerard et al. 2002). Polyaniline, in particular, was often employed (Antiochia et al. 2014; Teixeira et al. 2016; Radhapyari et al. 2013) thanks to its conductivity ascribed to the form of emeraldine salt. In fact, unlike other known electrically conductive polymers, polyaniline can exist, depend on the degree of oxidation, in various forms: leucoemeraldine, emeraldine and pernigraniline, which represent the fully reduced form, the semi-oxidized form and the fully oxidized form of the polymer, respectively. The only conductive form is the emeraldine salt. The nitrogen atom is involved in the system of conjugated double bonds, and this entails that the conductivity of the polymer is closely related to both the degree of protonation and the degree of oxidation (MacDiarmid and Epstein, 1989) Among the various techniques used for the production of such a polymer, the most used is the oxidative coupling in a strictly acid environment. The acidic environment is required since, with the increase of pH, the degree of polymerization is significantly reduced and also a remarkable decrease of the conductivity of the polymer itself is observed (Focke et al. 1987). It is likable to use the electrochemical polymerization instead of traditional methods because the polyaniline is synthesized to be directly deposited on the electrode. Also, through a careful modulation of electrochemical parameters during the experiment, the thickness and the conductivity of the polymer film can be easily controlled. The electrochemical polymerization occurs in a 3-step process: i) oxidation at the anode of the soluble monomer and oligomer formation in the diffusion layer; ii) oligomer deposition through the nucleation process; iii) propagation of the chain by solid state polymerization (Heinze et al. 2010). Unfortunately, despite its unique properties, the application of polyaniline in biochemical systems is limited by the loss of activity occurring at a neutral-basic pHs. This problem can be solved with the introduction of different functional groups within the polyaniline chain. In this regard, co-polymerizations of meta or ortho basic (amino) or acidic (carboxylic or sulphonic acid) group-substituted aniline allowed the generation of polymers which have shown electrochemical activity even at relatively high pH (Sarauli et al. 2013; Scherbahn et al. 2014; Chen et al. 2013; Marmisolle et al. 2015). The antistatic and high fouling resistance properties and the possibility to introduce different functional groups in the polymer chain make polyaniline an attractive to be used as a component of a nanocomposite with enhanced electrochemical features in impedimetric biosensors. In this context, nanomaterials like gold nanoparticles (AuNPs) and multi-walled carbon nanotubes (MWCNTs) were widely employed to improve the conductivity performances of the surface within which they were embedded (Saha et al. 2012; Justino et al. 2013) and, finally, could be used in a such a polyaniline based nanocomposite to further lower the charge

2

transfer resistance of the electrode surface and then to improve the biosensor features in terms of sensitivity and LOD. 2,4-Dichlorophenoxy acetic acid (2,4-D) is an auxinic herbicide with grown regulator activity that has been widely used for controlling broadleaf weeds in cereal grain crops (Veldstra et al. 1963). On account of its carcinogenic, teratogenic and estrogenic activity, the presence of residues of 2,4-D in agricultural products and environment can be extremely harmful to both humans and animals (Garabrant and Philbert, 2002; Wu et al. 2005). United States Environmental Protection Agency has set an enforceable regulation for 2,4-D, called a maximum contaminant level (MCL), at 0.07 mg/L or 70 ppb , while European Community has established that concentrations of pesticides in drinking water may not exceed 0.1 µg/L or 0.1 ppb for a single pesticide and 0.5 µg/L or 0.5 ppb for total pesticides. Hence, a reliable and rapid technique for its determination is essential to ensure both environmental and food safety. In this work the copolymerization of aniline (ANI) and 3-amino benzoic acid (3-ABA) has been employed for two specific reasons: i) the two molecules have a strictly correlated electrochemical profile and ii) to bind 2,4-D antibody using EDC-NHS chemistry of 3-ABA carboxylic group). The so obtained electrochemical interface binding 2,4-D antibody was used to determine 2,4-D either in standard and real sample solutions.

2. Material and methods 2.1 Materials 2,4-Dichlorophenoxyacetic acid antibody (Ab, 0.5 mg/mL, orb10004) was purchased from Biorbyt (Cambridge UK);

2,4-Dichlorophenoxyacetic acid (2,4-D or Ag, 98%, D70724), Gold nanoparticles

(AuNPs, 5 nm diameter, 5.5×1013 particles/mL, 741949), aniline (ANI, ≥99.5%), 3-Aminobenzoic acid (3ABA, 98%), hydrochloric acid (37% wt), sodium hydroxide, potassium chloride, sodium phosphate dibasic, sodium phosphate monobasic, potassium ferricyanide(III), potassium hexacyanoferrate(II) trihydrate, N-(3Dimethylaminopropyl)-N′-Ethylcarbodiimide

hydrochloride

(EDC),

N-Hydroxysuccinimide

(NHS),

ethanolamine (≥98%), all from Sigma-Aldrich (St. Louis, MO, USA) were used as received. All solutions have been prepared with high purity deionized water (Resistance: 18.2 MΩxcm at 25 °C; TOC  <  10 μg/L) obtained from a system MilliQ-UV, Millipore (France).

2.2 Apparatus The electrochemical experiments (CV, EIS) were performed in a thermostated glass cell (5 ml) using a three electrode system constituted by a Graphite (G) or Multi-Walled Carbon Nanotubes (MWCNTs) screen printed (SPE) as working electrode (Dropsens, Oviedo Spain, DRP110 or DRP110CNT respectively, 4 mm of diameter in each case), a SCE as reference electrode (Type 303-SCG/12, Amel, MI, Italy) and a glassy carbon rod as counter electrode (Cat. 6.1241.020, Metrohm, Switzerland). An Autolab PGSTAT204 potentiostat controlled by the NOVA program (Metrohm Autolab B.V., Utrecht, The Netherlands) was used to acquire amperometric and impedimetric experiments, performed at 25 °C and the potentials were referred to SCE electrode (0.244 V vs. NHE). High-Resolution Field Emission Scanning Electron Microscopy (HR

3

FESEM, Zeiss Auriga Microscopy) equipped with Microanalysis EDS ≤ 123 Mn-Kα eV (Bruker) was used to characterized the morphological features of the modified electrodes. The infrared spectra were collected with a Bruker Optic Alpha-R portable interferometer with an external reflectance head covering a circular area of about 5 mm of diameter. The investigated spectral range was 4000-375 cm-1 with a resolution of 4 cm-1 and 250 scans or more.

2.3 Preparation of Ab/AuNPs-PANABA/SPE immunosensor The working electrode of an SPE (either Graphite or MWCNTs) was modified by cyclic voltammetry (CV) electropolymerization in a solution of ANI and HCl as dopant, to obtain AuNPs-PANABA-MWCNTs-based nanocomposite. At the optimal experimental conditions of 0.1 M ANI and 0.5 M HCl, different aliquots of AuNPs (5.5×1013 particles/ml) solution were added to obtain the AuNPs-PANI composite material. Once established the best electropolymerization conditions regarding the best combination of ANI, HCl and AuNPs, 3-ABA was mixed to the electropolymerizing solution thus obtaining AuNPs-PANABA copolymer. In any case, the electropolymerization was carried out by scanning potential between -0.35 V and 0.75 V for 5 cycles at 100 mV/s under an inert atmosphere. Then, the copolymer carboxylic groups (belonging to 3ABA) were activated with an aqueous solution of 0.5 M EDC and 0.1 M NHS for 15 min. At this point, 2,4D Antibody was linked at the surface by spreading 10 μL of Ab solution (0.1 mg/mL in 10 mM PBS pH 7.4) onto the working electrode and let react for 60 min. Finally, the unreacted carboxylic functionalities were deactivated with 1 M ethanolamine solution (pH 8.5). After every modification step, the surface was rinsed with deionized water and dried under N2 flow. The schematic diagram showing the fabrication process of the impedimetric label-free immunosensor is shown in figure SM1.

2.4 Electrochemical characterization Electrochemical features (electroactive area (Ael) and roughness factor (ρ) calculated as ratio between Ael and geometric area (12.5 mm2), electron transfer rate constant (k0), resistance to charge transfer (RCT)) of the PANABA nanocomposite-modified electrode were measured every each step of modification by CV, scanning potential between -0.45 V and 0.75 V in 1.1 mM K3Fe(CN)6, 0.1 M KCl, 0.1 M PBS pH 7 solution at different scan rates and by Electrochemical Impedance Spectroscopy (EIS) in 5 mM Fe(CN)63-/4-, 0.1 M KCl, 0.1 M PBS pH 7 solution, applying a sinusoidal AC voltage pf 10 mV (peak-to-peak voltage) in amplitude at the frequency range 0.05-10000 Hz (for graphite-based electrodes) or 1-1000 Hz (for MWCNTs-based electrodes) at the open circuit potential (OCP). In order to measure Ael, ferricyanide peak currents were correlated to the square root of the scan rates according to the Randles-Sevcik equation (Bard and Faulkner, 2000), meanwhile, the electron transfer rates were evaluated according to Nicholson-KlinglerKochi method (Lavagnini et al. 2004), as shown in figure SM2. EIS experimental data were fitted with a modified Randles [RS([RCTW]Q)] circuit in which active electrolyte resistance (RS) is connected in series with the parallel combination of a constant phase element (Q), that replaces the double-layer capacitance for nanostructured and rough surface, and a charge transfer resistance (RCT) at the electrode surface of a

4

ferricyanide redox probe, in series with a Warburg element (W), indicating the diffusion of the species at low frequencies. Once the electrode was modified with bulky un-conductive elements (such as antibody and finally antigen), the generation of the faradic current by the redox probe was hindered and RCT can thus be used for 2,4-D detection: the lower conductivity of the electrode surface, the higher measured RCT. 2.5 2,4-D detection and calibration curve realization The detection of antigen was performed by dipping the modified electrode in a 2,4-D standard solution and then, after rinsed the bound immunosensor in the buffer, by registering the relative Niquist spectrum in 5 mM Fe(CN)63-/4-, 0.1 M KCl, 0.1 M PBS pH 7 solution. The corresponding RCT increase was compared to the Ab/AuNPs-PANABA/MWCNTs-SPE one as blank. The incubating time between Ab and Ag solution was changed to establish the optimum conditions which were used successively to build the calibration curve by varying the 2,4-D concentration (the greater is the [2,4-D], the greater is the increase percentage of RCT). In each case, the impedimetric signal was evaluated normalizing the difference between the R CT values before and after the 2,4-D detection relative to the Ab/AuNPs-PANABA/MWCNTs-SPE one.

2.6 Real samples analysis Spiked samples were prepared by adding known concentration of 2,4-D in tap real waters and finally evaluating the RCT variation in 5 mM Fe(CN)63-/4-, 0.1 M KCl, 0.1 M PBS pH 7 solution.

3. Results and Discussion 3.1 Realization of AuNPs-PANI nanocomposite To maximize the conductivity of the immobilization platform for the immuno-analytical detection of 2,4-D, ANI was electropolymerized on different commercial SPEs by varying either ANI : HCl ratio and their concentrations at a fixed optimal concentration ratio. The conductivity of the electrodic material was evaluated comparing electroactive area (Ael), electron transfer rate (k0) constant and RCT. The results are summarized in Table 1; each detection is the mean value of at least three independent measurements. The increase of electrode conductivity allows enhancing the sensitivity (see below). In the case of PANI/Graphite electrodes, it can be observed as both the ANI : HCl ratio and overall the concentration affected the electrochemical performances of the surface. In particular, the best results in term of electroactive area, electron transfer rate constant and resistance to charge transfer have been obtained by using for the electropolymerisation a 0.1 M ANI and 0.5 M HCl solution. Moreover, by comparing the same modification procedure on electrodes constituted by different materials, a further increase of electrochemical parameters was observed when using MWCNTs instead of Graphite. As a consequence, the PANI/MWCNTs-SPE obtained by electropolymerization of a 0.1 M ANI : 0.5 M HCl solution was adopted in the following experiments. In this case, three different concentrations of AuNPs were added by modifying solution before aniline electropolymerization; then the RCT was detected; the results are reported in Table 2 indicating that the concentration of 0.2% AuNPs led to the highest conductivity value.

5

3.2 Morphological characterization of AuNPs-PANABA nanocomposite and antibody immobilization. AuNPs-PANI/MWCNTs electrode was not able to covalently link biological molecules, while antibodies could only be unspecifically adsorbed onto the surface and, after binding, the antigen was gradually released in solution with increasing measurement time (Figure SM3). In order to confer to the AuNPs-PANI nanocomposite the ability to covalently link a protein that could be used as biochemical recognition element in biosensors development (Tanne et al. 2014), 3-ABA was added as co-monomer to the solution used throughout the electropolymerization in view to employ its carboxylic group in an EDC-NHS coupling reaction with the protein. The presence of carboxylic group in meta with respect to the amine group does not hinder the polymerization reaction (which occurs in ortho and para), as confirmed by Salavagione et al. (2004) that have proved a difference of around 560 times in the polymerization reactivity between 3-ABA and 2-ABA. The CV profile of AuNPs-PANABA nanocomposite electropolymerization displayed the typical peaks of emeraldine salt electrochemistry, as reported in the literature (Pournaghi-Azar and Habibi, 2007, see Figure SM4), thus indicating that the presence of 3-ABA did not disrupt the electrochemistry of PANI. Also, the influence of 3-ABA percentage in the electropolymerizing solution has been systematically studied and the influence on RCT value on a naked MWCNTs surface has been reported in Figure SM5. As it can be observed, the increase of 3-ABA in the electropolymerizing solution decreased the surface conductivity so that, for a 3-ABA percentage between 25% and 65%, the RCT value was almost doubled if compared to that one in absence of 3-ABA while for greater percentages it attained the same conductivity of uncovered MWCNTs if not even lower. The solid phase infrared spectra of MWCNTs, AuNPs-PANI/MWCNTs and AuNPs-PANABA/MWCNTs confirmed the electrodeposition of the polymers (Trchová et al. 2011), also in presence of 3-ABA (Fig SM6). The spectrum of MWCNTs bare electrode is characterized only by a signal convolution in the aromatic zone (800-400 cm-1), while the stretching of N-H bond (3439 cm-1 and 3367 cm-1), that of C=N double bond (2597cm-1 and 2567 cm-1) and that of C-N bond (1312 cm-1 and 1246 cm-1) are present with PANI and PANABA. In particular, the C=O double bond stretching at 1691 cm-1 is a proof of the ANI and 3-ABA copolymerization in our described procedure, as evidenced before by impedimetric analysis. A further characterization of the obtained nanocomposite was carried out by microscopic investigation either by SEM (Figure SM7). In Figure SM7(b) the presence of both organic polymer and homogeneously dispersed AuNPs is clearly evident on the surface with respect to (a) where pristine MWCNTs are visible. In addition, the elemental analysis shows the presence of gold inside the electropolymerised material (Fig SM8) and of nitrogen in the copolymer, both absent in the bare MWCNTs electrode. Hence, 3-ABA was added to the aniline solution to be polymerized maintaining constant (i.e. 0.1 M) the total concentration of ANI + 3-ABA but changing the ratio between each other; to assess the most suitable value thereof, the RCT increase in the Fe(CN)63-/4- solution was evaluated after the immobilization of the antibody and, therefore, of the antigen (2,4-D). In this respect, the greater was the increase percentage of RCT, the

6

greater was the number of antibody molecules immobilized on the surface and, in theory, the greater was the number of molecules of 2,4-D which can be anchored and then detected by the biosensor. In Figure 1, the EIS spectra obtained for naked MWCNTs surface (red), then AuNPs-PANABA/MWCNTs nanocomposite before (yellow) and after Ab covalent linking (blue) and finally upon interaction with 50 ppb 2,4-D (green), are reported; in Table 3 the data also obtained with different ANI:3-ABA ratios are reported for comparison. Once having considered data above, it can be observed as the highest useful analytical signal was obtained for ANI:3-ABA 1:2; the RCT increase due to Ab link is inversely related to ANI:3-ABA ratio (first column) and the further interaction with 50 ppb 2,4-D produced a huge increase in the case of 1:2 ratio (77%) while it was much smaller for both 2:1 and 1:1 ratios (second column). It is also reasonable to consider that the 1:5 ratio, entailing a greater number of carboxylic groups available for Ab coupling, led to a further increase of the immobilized amount of antibody as indicated by the increased RCT value (115% vs. 106%) with respect to 1:2 ratio. On the other hand, this resulted in a worsening of analytical performances since the useful signal dropped from 77% to 30% likely because of the steric hindrance of immobilized Ab molecules.

3.3 Calibration curve towards PANI and real samples analysis The incubation time that has to be adopted either for the calibration curve and the real samples analysis was optimized by evaluating the analytical response at increasing time. The results reported in Table SM1 indicate that the longer was the incubation time from 0 to 120 min, the higher was the RCT increase; a longer incubation time up to 1 hour did not produce a significant signal change. We opted to operate with 10 min of incubation time since this seemed to be a good compromise between analytical sensitivity (it allowed to accomplish a LOD equal to 0.3 ppb, see below) and time of analysis. Anyway, it should be pointed out that when higher sensitivity or lower LOD were needed, the incubation time could be increased thus relying on better analytical performances. In Figure 2 the experimental response obtained by Ab/AuNPs-PANABA/MWCNTs-SPE for increasing concentrations of 2,4-D in the range 1-100 ppb, is reported clearly by evidencing as the higher was the concentration of 2,4-D with which the immunosensor has been incubated, the greater was the measured RCT value due to the hindering effect of Ab-Ag complex towards the redox probe in reaching the electrodic surface. The detected values of RCT in the whole considered 2,4-D concentration range, obtained with either Ab/AuNPs-PANABA/MWCNTs-SPE (full circles) and Ab/MWCNTs-SPE (empty circles) are plotted in Figure 3. As can be observed, the electrode modification with the nanocomposite, enhancing the surface conductivity, clearly improved the immunosensor performances, leading to a wider dynamic response to antigen (1-100 ppb instead of 1-50 ppb) and a higher sensitivity. This could be awarded, also, to the higher number of Ab molecules immobilized on the surface thanks to the presence of carboxylic groups of 3-ABA. The calibration curve equation was log y = 0.94 + 0.58 log x, with R = 0.993 (Figure 3 inset reports the logarithmic graph, according to the equation above); LOD, as the concentration corresponding to three times the noise, is 0.3 ppb. Measurements were carried out at least six times, with a RSD of about 6%. Our results

7

are quite consistent, in term of both linear range and LOD, with the previous works (listed in Table SM2) in the literature and in some cases even better taking also into account the different transducers employed. As far as the electrochemical transduction is concerned, excellent analytical performances were obtained by Dequaire by using nanomagnetic beads for preconcentration in a complex ELISA-like competitive assay (Dequaire et al. 1999). In this case, good sensitivity was accomplished thanks to the inherent amplification provided by the linked enzyme. In our case, comparable analytical performances were obtained using a direct transduction method without the need of either preconcentration and amplification. Finally, the proposed immunosensor was tested in real sample analysis (tap water); the results obtained on fortified samples are reported in Table SM3, indicating good values of recovery ranging from 82 to 120% thus envisaging the possibility to apply the proposed immunosensor to real sample analysis.

4. Conclusions In this work, we developed an impedimetric label-free immunosensor for the detection of 2,4-D. We optimized the experimental conditions for the nanocomposite-based AuNPs-PANABA-MWCNTs modified electrode as follows: 0.033 M ANI, 0.066 M 3-ABA and 0.5 M HCl, plus 0.2% AuNPs (5.5×1013 particles/ml solution). Aniline-based copolymer, modified with the gold nanoparticles, allowed to enhance the electrode conductivity thus obtaining a wider dynamic range and a more sensitive 2,4-D detection. The developed impedimetric immunosensor displayed a wide linearity range towards 2,4-D (1-100 ppb), good repeatability (RSD 6%), stability and a LOD (0.3 ppb) lower than herbicide emission limits.

References Ahuja, T., Mir, I.A., Kumar, D., Rajesh. 2007. Biomaterials 28, 791-805. Antiochia, R., Mazzei, F., Gorton, L., Leech, D., Favero, G. 2014. Electroanalysis 26, 1623-1630. Bartlett, P.N., Birkin, P.R. 1993. Synt. Met. 61, 15-21. Billah, M.M., Hays, H.C.W., Hodges, C.S., Ponnambalam, S., Vohra, R., Millner, P.A. 2012. Sensor. Actuat. B Chem., 173, 361-366. Bollella, P., Fusco, G., Tortolini, C., Sanzò, G., Antiochia, R., Favero, G., Mazzei, F. 2016. Anal. Bioanal. Chem. 408, 3203-3211. CE. URL: http://ec.europa.eu/eurostat/statistics-explained/index.php/Agri-environmental_indicator__pesticide_pollution_of_water. Chen, C., Ding, G., Zhou, D., Lu, X. 2013. Electrochim. Acta 97, 112-119. Dequaire, M., Degrand, C., Limoges, B. 1999. Anal. Chem. 71, 2571-2577. EPA. URL: contaminants.

https://www.epa.gov/ground-water-and-drinking-water/table-regulated-drinking-water-

8

Focke, W.W., Wnek, G.E., Wei, Y. 1987. J. Phys. Chem. 91, 5813-5818. Garabrant, D.H., Philbert, M.A. 2002. Crit. Rev. Toxicol. 32, 233-257. Gerard, M., Chaubey, A., Malhotra, B.D. 2002. Biosens. Bioelectron. 17, 345-359. Hall, J.C., Deschamps, R.J.A., Krieg, K.K. 1989. J. Agric. Food Chem. 37, 981-984. Haupt K., Dzgoev, A., Mosbach, K. 1998. Anal. Chem. 70, 628-631. Heinze, J., Frontana-Uribe, B.A., Ludwigs, S. 2010. Chem. Rev. 110, 4724-4771. Holford, T.R.J., Holmes, J.L., Collyer, S.D., Davis, F., Higson, S.P.J. 2013. Biosens. Bioelectron. 44, 198203. Justino, C.I.L., Rocha-Santos, T.A.P., Cardoso, S., Duarte, A.C. 2013. Trends Anal. Chem. 47, 27-36. Kashyap, S.M., Pandya, G.H., Kondawar, V.K., Gabhane, S.S. 2005. J. Chromatogr. Sci. 43, 81-86. Lavagnini, I., Antiochia, R., Magno, F. 2004. Electroanal. 16, 505-506. MacDiarmid, A.G., Epstein, A.J. 1989. Farad. Discuss. 88, 317-332. Marmisolle, W.A., Gregurec, D., Moya, S., Azzaroni, O. 2015. ChemElectroChem 2, 2011-2019. Navràtilovà, I. and Sklàdal, P. 2004. Bioelectrochemistry 62, 11 – 18. Pournaghi-Azar, M.H., Habibi, B. 2007. Electrochim. Acta 52, 4222-4230. Radhapyari, K., Kotoky, P., Das, M.R., Khan, R. 2013. Talanta 111, 47-53. Saha, K., Agasti, S.S., Kim, C., Li, X., Rotello V.M. 2012. Chem. Rev. 112, 2739-2779. Salavagione, H.J., Acevedo, D.F., Miras, M.C., Motheo, A.J., Barbero, C.A. 2004 J. Polym. Sci. A Polym. Chem. 42, 5587—5599. Sarauli, D., Xu, C., Dietzel, B., Schulz, B., Lisdat, F. 2013. Acta Biomaterialia 9, 8290-8298. Scherbahn, V., Putze, M.T., Dietzel, B., Heinlein, T., Schneider, J.J., Lisdat, F. 2014. Biosens. Bioelectron. 61, 631-638. Tanne, J., Kracher, D., Dietzel, B., Schulz, B., Ludwig, R., Lisdat, F., Scheller, F.W., Bier, F.F. 2014. Biosensors 4, 370-386. Teixeira, S.R., Lloyd, C., Yao, S., Gazze, A.S., Whitaker, I.S., Francis, L., Conlan, R.S., Azzopardi, E. 2016. Biosens. Bioelectron. 18, 395-402. Tortolini, C., Bollella, P., Antiochia, R., Favero, G., Mazzei, F. 2016. Sensor. Actuat. B Chem. 224, 552558. Tortolini, C., Bollella, P., Antonelli, M.L., Antiochia, R., Mazzei, F., Favero, G. 2015. Biosens. Bioelectron. 67, 524-531. Trchová, M. and Stejskal J. 2011. Pure Appl. Chem. 83, 1803-1817.

9

Veldstra, H. 1963. Plant Hormones, Section b. Synthetic Auxins. In: Stotz, M., Florkin, E.H. Comprehensive Bioelectrochemistry: Water-Soluble Vitamin, Hormones, Antibiotics, vol 9 Elsevier Publishing Company, Amsterdam, London, New York, pp. 127-150. Vinayaka, A.C., Basheer, S., Thakur, M.S. 2009. Biosens. Bioelectron. 24, 1615-1620. Wallace, G.G., Smyth, M., Zhao, H. 1999. Trends Anal. Chem. 18, 245-251. Wu, J.M., Ee, K.H., Lee, H.K. 2005. J. Chromatogr. A 1082, 121-127.

-Z" ()

Yang, F., Han, J., Zhuo, Y., Yang, Z., Chai, Y., Yuan, R. 2014. Biosens. Bioelectron. 55, 360-365.

3.5 10

2

3 10

2

2.5 10

2

2 10

2

1.5 10

2

1 10

2

5 10

1

0 4 10

2

5 10

2

6 10

2

7 10

2

8 10

2

9 10

2

1 10

3

1.1 10

3

Z' ()

Figure 1. EIS at MWCNTs-SPE (red), AuNPs-PANABA/MWCNTs-SPE (yellow), Ab/ AuNPsPANABA/MWCNTs-SPE before (blue) and after incubation with 50 ppb 2,4-D (green) in 5 mM Fe(CN)63-/4, 0.1 M KCl, 0.1 M PBS pH 7 solution, applying a sinusoidal AC voltage of 10 mV (peak-to-peak voltage) in amplitude at the frequency range 1-1000 Hz at OCP. Experimental data were fitted with [R([RW]Q)] circuit (inset).

10

-Z" ()

8 10

2

7 10

2

6 10

2

5 10

2

4 10

2

3 10

2

2 10

2

1 10

2

0 ppb

100 ppb

0 4 10

2

6 10

2

8 10

2

1 10

3

1.2 10

3

1.4 10

3

Z' ()

Figure 2. EIS at Ab/AuNPs-PANABA/MWCNTs-SPE before and after incubation with different 2,4-D concentration (1-100 ppb) in 5 mM Fe(CN)63-/4-, 0.1 M KCl, 0.1 M PBS pH 7 solution, applying a sinusoidal AC voltage of 10 mV (peak-to-peak voltage) in amplitude at the frequency range 1-1000 Hz at OCP. Experimental data were fitted with [R([RW]Q)] circuit (inset).

11

160

120 100 1000

Calibration Curve

80 100

60 40

10

%

RCT (2,4-D - Ab) / RCT (Ab) / a.u.

140

20 1 .1

0 0

20

40

60

80

1

10

100

100

120

1000

140

[2,4-D] / ppb Figure 3. Experimental data obtained for the determination of 2,4-D by Ab/AuNPs-PANABA/ MWCNTsSPE (full) and by Ab/MWCNTs-SPE (empty). Inset, calibration curve for Ab/AuNPs-PANABA/ MWCNTsSPE is shown. Experiments were performed in 5 mM Fe(CN)63-/4-, 0.1 M KCl, 0.1 M PBS pH 7 solution, applying a sinusoidal AC voltage of 10 mV (peak-to-peak voltage) in amplitude at the frequency range 11000 Hz at OCP with increasing concentration of 2,4-D.

Table 1. Comparison between modified electrode electrochemical features by varying either the ANI : HCl ratio and their concentrations in the electropolymerization reaction and then the carbon electrode material. ANI : HCl (M : M) Ael (mm2) ko (cm/s) ρ RCT decrease -3 Graphite (blank) 3.4 ± 0.3 (0.44 ± 0.03) 10 0.27 1:2 (0.1:0.2) 1:5 (0.1:0.5) PANI/Graphite

MWCNTs (blank) PANI/MWCNTs

1:10 (0.1:1.0)

5.0 ± 0.4 6.3 ± 0.3 4.9 ± 0.4

(0.82 ± 0.03) 10-3 0.40

38%

(2.20 ± 0.02) 10

-3

0.50

50%

(0.70 ± 0.03) 10

-3

0.40

40%

-3

1:5 (0.05:0.25)

5.8 ± 0.6

(1.50 ± 0.02) 10

0.46

25%

1:5 (0.2:1.0)

5.1 ± 0.5

(1.40 ± 0.03) 10-3 0.40

20%

1:5 (0.1:0.5)

5.5 ± 0.6 9.9 ± 0.8

12

(1.30 ± 0.02) 10

-3

0.44

-

(2.80 ± 0.07) 10

-3

0.80

32%

Table 2. Correlation between AuNPs percentage (v/v) in the electropolymerising solution and the RCT decrease. AuNPs (%) RCT decrease 0 32% 0.1 50% 0.2 75% 0.5 40%

Table 3. Correlation between ANI:3-ABA ratio and the immunosensor features in terms of RCT increasing due to different Ab loading and then to different amount of 50 ppb 2,4-D binding. RCT increase due to RCT increase due to ANI : 3-ABA Ab loading 50 ppb 2,4-D interaction 5:1 20% 18% 2:1

50%

15%

1:1

72%

26%

1:2

106%

77%

1:5

115%

30%

Highlights   

Impedimetric label-free immunosensor for the detection of 2,4-Dichlorophenoxy Acetic Acid (2,4-D) herbicide Electropolymerization a conductive polymer poly-(aniline-co-3-aminobenzoic acid) (PANABA) in the presence of a dispersion of gold nanoparticles (AuNPs) Electrochemical impedance spectroscopy (EIS) in faradic condition to characterize the nanocomposite material and to perform immunochemical analysis

13