A simple, sensitive and compact electrochemical

Journal of Electroanalytical Chemistry 758 (2015) 59–67

Contents lists available at ScienceDirect

Journal of Electroanalytical Chemistry journal homepage: www.elsevier.com/locate/jeac

A simple, sensitive and compact electrochemical ELISA for estradiol based on chitosan deposited platinum wire microelectrodes Tina T.-C. Tseng ⁎, Arwinda Gusviputri, Le Ngoc Quynh Hoa Department of Chemical Engineering, National Taiwan University of Science and Technology, Taipei, 10607, Taiwan

a r t i c l e

i n f o

Article history: Received 9 July 2015 Received in revised form 28 September 2015 Accepted 16 October 2015 Available online 17 October 2015 Keywords: Electrochemical ELISA Estradiol Chitosan Microelectrode 4-aminophenyl phosphate Platinum

a b s t r a c t A simple and inexpensive method for fabricating a sensitive and compact indirect sandwich type electrochemical ELISA for the detection of estradiol using a chitosan electrodeposited platinum (Pt) wire microelectrode was proposed. In this assay, anti-17β estradiol antibody produced in goat (goat anti-estradiol Ab) was used as the capture antibody which was immobilized on the chitosan coated Pt wire microelectrode, anti-17β estradiol antibody produced in mouse (mouse anti-estradiol Ab) was used as the detection antibody, and goat anti-mouse IgG (immunoglobulin G) conjugated with alkaline phosphatase (AP) was used as the secondary antibody. The effect of pre-coated layers (chitosan, the capture antibody, and the blocking reagent BSA) on the electron transfer resistance (Ret) at the surface of ELISA electrodes has been investigated and analyzed by the electrochemical impedance spectrum (EIS). 4-Aminophenyl phosphate (4-APP) was chosen as the AP substrate and the oxidation potential of the electroactive AP product, 4-aminophenol (4-AP), on the Pt electrode was determined to be + 0.14 V (vs. Ag/AgCl). The electrochemical ELISA was detected by constant potential amperometry at +0.14 V in the Tris buffer (pH 9.0). The limit of detection of this assay was 2.7 × 10−1 pg/mL with a wide detection range from 2.7 × 10−1 pg/mL up to 1.0 × 105 pg/mL. The assay specificity evaluated by testing the crossreactivity of the assay for progesterone and 17α-ethynylestradiol was found to be 0.033% and 3.4%, respectively. This assay has been tested with estradiol in spiked serum samples; however, further pretreatment of serum samples may be required to enhance precision and recovery. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Estradiol is a form of estrogen, a sex hormone mainly produced by ovaries and sometimes, by adrenal glands or by placenta during pregnancy, which plays an important role in the regulation of female productive cycles and the development of reproductive tissues [1]. The estradiol level usually fluctuates during the menstrual cycle; normally, the estradiol level in the premenopausal women is 15–350 pg/mL and will decrease to b10 pg/mL during postmenopausal period [2]. When the estradiol level is higher than normal, it may imply early puberty or may be associated with diseases, such as ovarian cancer, hyperthyroidism, or cirrhosis; on the other hand, if the estradiol level is lower than normal, it may imply menopause, ovarian failure, polycystic ovarian syndrome, or depleted estrogen production [3]. Therefore, the measurement of estradiol level using convenient, sensitive, and economic analytical techniques is required for assisting the evaluation and diagnosis of fertility problems, menstrual problems, or ovarian cancer [4]. Many analytical techniques have been developed to determine the estradiol level in the human serum, such as high performance liquid chromatography (HPLC) [5,6], surface plasmon resonance (SPR) [7–9], ⁎ Corresponding author. E-mail address: [email protected] (T.T.-C. Tseng).

http://dx.doi.org/10.1016/j.jelechem.2015.10.017 1572-6657/© 2015 Elsevier B.V. All rights reserved.

and gas chromatography–mass spectrometry (GC/MS) [10]. In general, these techniques have high sensitivity for detecting low level of estradiol; however, they usually require expensive and sophisticated instruments or complicated and tedious operation procedure. Direct immunoassay techniques like radioimmunoassay (RIA) [11–13], chemiluminescent immunoassay [14,15] and homogeneous enzyme immunoassays [16, 17] are usually used as more straightforward methods for the detection of estradiol. For conventional RIAs, samples are usually pretreated by extraction and chromatography steps; therefore, they are sensitive and reliable, but the use of radioisotopic labels poses potential health hazards [18]. Chemiluminescent immunoassays offer good sensitivity, but their signal intensity and detection range are usually low which can make the quantification process difficult [19]. Homogeneous enzyme immunoassays are simple and convenient because no separation steps are required; however, their limits of detection and sensitivities are relatively poor [20]. Enzyme-linked immunosorbent assay (ELISA) techniques have been developed for several decades since 1971 [21] and are widely used as analytical tools for the detection of hormones, proteins, peptides, or antibody with high sensitivity and strong specificity [22]. In ELISAs, antibody or antigen is attached to a solid support (e.g. plastic plate) for a specific antibody-antigen binding to recognize the analyte and the detection is accomplished by measuring the activity of an enzyme conjugated to the detection antibody that complexes with the

Indirect sandwich

4

5

Platinum wire (dia. = 50.8 μm)

10

5

Abbreviations: ABA (4-aminobenzoic acid); ACTH (adrenocorticotropin); AP (alkaline phosphatase); dia. (diameter); DPV (differential pulse voltammetry); FIA (flow injection analysis); HRP (horseradish peroxidase); Mab (monoclonal antibody); Pab (polyclonal antibody); SPCE (screen printed carbon electrodes); TMB (tetramethylbenzidine).

This work Progesterone; 17α-ethynyl estradiol Spiked serum 2.7 × 10−1/ 2.7 × 6.75 (2.5 hr pre-coating; 10−1–1.0 × 4.25 hr assay time)

Competitive SPCE (dia. = 4 mm)

3

Working electrode electrodeposited with chitosan followed by physical adsorption AP/4-APP of goat Pab

AP/1-naphthyl phosphate HRP-estradiol with H2O2 (hydroquinone as redox mediator) Working electrode coated with rabbit anti-mouse IgG for adsorbing mouse Mab Working electrode modified with grafted ABA followed by covalent binding of streptavidin and immobilization of biotinylated mouse Mab Competitive SPCE (dia. = 3 mm)

2

Constant potential amperometry at +0.14 V (vs. Ag/AgCl)

3.41 (2.58 hr pre-coating; 0.83 hr assay time) 0.77/ 1–250

Certified serum

[41]

Progesterone; testosterone; cortisol; [42] 17α-ethynylestradiol; ACTH; prolactin

Spiked serum N12.7 (10+ hr pre-coating; 2.7+ hr assay time) 50/ 25–500

No

[40] Testosterone; methyltestosterone; progesterone Non-extracted bovine serum N10.5 (10+ hr pre-coating; 0.5+ hr assay time) 15/ – AP/1-naphthyl phosphate Working electrode coated with anti-rabbit IgG for adsorbing rabbit Pab Graphite screen Competitive printed electrodes (dia. = 3 mm)

1

DPV scanning from 0 to +0.6 V (vs. silver screenprinted pseudo-reference) DPV scanning from −0.1 V to +0.7 V (vs. Ag/AgCl) Constant potential amperometry at +0.2 V (vs. silver pseudo reference electrode)

[39] No N15.5 (11+ hr pre-coating; 4.5+ hr assay time) Mouse Mab or rabbit Pab was attached to the anti-mouse IgG or anti-rabbit IgG HRP/TMB coated microtitre plate Glassy carbon disk, Competitive (dia. = 3 mm)

FIA coupled with constant potential amperometry at +0.1 V (vs. Ag/AgCl)

20/10–104 for Pab 100–104 for Mab

Bovine serum

Interference test Serum test Assay time (hr) Detection limit/range (pg/mL) Electrochemical methods Working electrode

Immobilization method

Enzyme label/substrate ELISA format No.

analyte [23]. For competitive ELISAs, they involve a competitive binding of the labeled (add-in) antigen to an antigen-specific antibody with the sample antigen and therefore, their output signal decreases as the concentration of the sample antigen increases. Although competitive ELISAs can be sensitive and specific [24], improper modification of the labeled analyte may change the analyte-antibody binding affinity and lead to inconsistent results [25]. For sandwich ELISAs, the analyte is captured by the immobilized primary antibody (i.e. capture antibody) and the detection antibody with enzyme label (for direct ELISAs) or without enzyme label (for indirect ELISAs) is used for attaching the captured antigen. The detectable assay signal can be generated from the enzyme labeled on the detection antibody (for direct ELISAs) or on the secondary antibody (for indirect ELISAs) [26]. Thus, the output signal of sandwich ELISAs increases as the concentration of the sample antigen increases [27]. Sandwich ELISAs usually have better specificity because antibodies against different epitopes of a target antigen are used [28]. A variety of enzyme labels have been used in ELISAs, such as glucose6-phosphate dehydrogenase, glucose oxidase, horseradish peroxidase (HRP), alkaline phosphatase (AP), and β-galactosidase; among them, HRP and AP are most commonly used since they can produce detectable products (e.g. 3,3′,5,5′-tetramethylbenzidine (TMB) for HRP and phenol for AP) with sensitive photometric signals [29]. Although photometric ELISAs have been widely used as diagnostic tools in medicine and laboratory, they have some limitations; for example, cumbersome instruments and large amount of reagents are usually required when performing the assay. Compared to conventional photometric ELISAs, electrochemical ELISAs can be performed using a compact and simple instrument (e.g. potentiostat) and the amount of reagents may be reduced by using a microelectrode; in addition, they can provide higher sensitivity and lower detection limit since the effect of interference caused by turbid or colored samples can be eliminated [30–32]. Similar to photometric ELISAs, electrochemical ELISAs also immobilize antibody or antigen on a solid support, but instead of using the plastic plate, an electrode is used. A variety of materials, such as gold, carbon, copper, and platinum have been used as the working electrode of sensors for the detection of estradiol [33–36] and among these electrodes, platinum is the most promising one due to its good electrochemical properties, inertness, and flexibility to be fabricated into many forms [37,38]. In Table 1, recent studies on electrochemical ELISAs for estradiol were summarized. Draisci et al. [39] demonstrated a competitive immunoassay for the detection of estradiol extracted from human serum which was carried out using a conventional ELISA plate. The assay solution was then injected into a flow system and the assay signal was detected by constant potential amperometry. Volpe et al. [40] used screen printed electrodes for the fabrication of disposable competitive ELISA immunosensors for estradiol. Anti-rabbit IgG was immobilized on the working electrode for adsorbing the detection antibody with controlled orientation. After the immobilization of antibody, competitive ELISA was performed; the sample antigen competed with alkaline phosphatase labeled estradiol for the binding site of the detection antibody. 1-Naphthylphosphate was used as the enzyme substrate and the signal of electroactive product (1-naphthol) was measured by differential pulse voltammetry (DPV) using a portable potentiostat. Pamberton et al. [41] and Ojeda et al. [42] also used screen printed electrodes and the similar method to develop a competitive electrochemical ELISA for estradiol. In Ojeda et al.'s study, estradiol-HRP was used to compete with the sample antigen for the binding site of the detection antibody. After the addition of H2O2 as the substrate, the assay signal generated from the enzyme product using hydroquinone as the redox mediator was measured by constant potential amperometry. In this work, we developed a sandwich-type electrochemical ELISA for estradiol using a chitosan coated platinum (Pt) microelectrode. The tiny micro-platinum wire working electrode enables concurrent incubation and washing of several ELISA electrodes in the same batch with small amounts of reagents and simultaneous measurements using a multichannel potentiostat. The electrode surface was modified with

Ref.

T.T.-C. Tseng et al. / Journal of Electroanalytical Chemistry 758 (2015) 59–67

Table 1 Electrochemical ELISAs for the detection of estradiol.

60


61

2. Experimental section

from Calbiochem (San Diego, CA, USA). β-Estradiol was purchased from Alfa Aesar (Ward Hill, MA, USA). N,N-Dimethylformamide (DMF, 73.09 g/mL) was purchased from Tedia (Fairfield, OH, USA). Human serum from a normal, healthy female donor was purchased from Valley Biomedical (Winchester, VA, USA). Ag/AgCl glassbodied reference electrodes with 3 M NaCl electrolyte and a platinum wire (diameter = 0.5 mm) auxiliary electrode were purchased from ALS Co., Ltd. (Tokyo, Japan). Perfluoroalkoxy (PFA) coated platinum wire (outer diameter = 4.0 × 10− 3 in.; inner diameter = 2.0 × 10− 3in.) used as the working electrode in this study was purchased from A-M System (Business Park Loop Sequim, WA, USA). The blocking buffer (pH = 7.4) was composed of 137 mM sodium chloride, 2.7 mM potassium chloride, 8.0 mM sodium phosphate (dibasic), 2.0 mM potassium phosphate (monobasic) and 1% BSA. The washing buffer (pH = 7.4) was composed 137 mM sodium chloride, 2.7 mM potassium chloride, 8.0 mM sodium phosphate (dibasic), 2.0 mM potassium phosphate (monobasic) and 0.1% Tween-20. The assay buffer (pH = 7.4) was composed of 15 mM sodium chloride, 10 mM sodium phosphate (dibasic), 0.1% Tween-20, and 1.0% BSA. 100 mM M Tris buffer (pH 9.0) was prepared by diluting 1.0 M Tris HCl buffer (pH 7.5) for 10 times and adjusting the pH with NaOH to pH 9.0. Electrochemical experiments for preparing ELISA electrodes (i.e. electrodeposition of chitosan), investigating the electrochemical behavior of the AP product (i.e. 4-AP) by performing the cyclic voltammetry, and testing the assay by performing the constant potential amperometry were conducted using a versatile multichannel potentiostat (model VSP300) equipped with the ‘p’ low current option driven by EC-LAB software (Bio-Logic, Knoxville, TN, USA). The threeelectrode configuration consists of a working electrode (i.e. Pt wire microelectrode with electrode surface area: 4.0 × 10−1 mm2), Pt wire auxiliary electrode, and a Ag/AgCl glass-bodied reference electrode.

2.1. Materials and instrumentation

2.2. Preparation of the ELISA electrode

Chitosan (from shrimp shells, ≥ 75% deacetylated, powder), 17αethynylestradiol (N 98% purity), progesterone (from plant source, 99% purity), glutaraldehyde (Grade I, 25% in H2O), bovine serum albumin (BSA, ~ 66 kDa, purity N98%, lyophilized powder), sodium chloride (NaCl, MW: 58.44 g/mol, purity N98%), sodium phosphate dibasic (Na2HPO4, MW: 141.96 g/mol, purity N99%), and Tween-20 (C58H114O26, viscous liquid) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Anti-17β estradiol antibody produced in goat (goat anti-estradiol Ab, polyclonal, whole antiserum, 100 μL, product no.: ab50660) and 4aminophenyl phosphate (4-APP) monosodium salt (N97% purity, powder), and goat anti-mouse IgG labeled with alkaline phosphatase (polyclonal, 1 mg/mL, liquid) were purchased from Abcam Biochemical (Cambridge, UK). Anti-17β estradiol 6 antibody produced in mouse (mouse anti-estradiol Ab, monoclonal, liquid, 5 mg/mL, 1 mL) was purchased from Life Span Bioscience (Seattle, WA, USA). Alkaline phosphatase (3.48 × 104 U/mL, from calf intestine, liquid) was purchased

The PFA coated Pt-wire (outer diameter = 4.0 × 10− 3 in.; inner diameter = 2.0 × 10− 3in.) was cut into a 1.5 cm piece of wire and then, both ends of the PEA coating was stripped and exposed 2.5 mm bare Pt wire on one side and ~ 1.0 mm bare Pt wire on the other side. The longer exposed side (2.5 mm side) was used as the working electrode (electrode surface area: 0.401 mm2). The shorter exposed side (1.0 mm side) of the Pt wire was connected to a piece of ~9.0 cm copper wire (AWG: 30) by soldering and then the soldering joint was wrapped with the epoxy glue. The other side of the copper wire was stripped and 1.5 cm of bare copper was exposed. Before any electrochemical experiments, the Pt wire microelectrode was cleaned with isopropyl alcohol by sonication and rinsed with DI water and then dried with argon. For making the chitosan coated Pt wire microelectrode, the chitosan solution was prepared by dissolving 10 mg chitosan in 25 mL DI water (m/v = 0.04%). To dissolve chitosan in the DI water, the solution pH was adjusted to pH = 3.0 with hydrogen chloride and then, the solution was sonicated until all chitosan flakes were dissolved. The chitosan solution was filtered using a syringe filter (pore size: 0.2 μm) followed by adjusting the solution pH to pH = 5.0 with sodium hydroxide. The electrodeposition of chitosan on the Pt wire microelectrode was achieved by performing cyclic voltammetry (CV) from − 1.0 to 0 V (vs. Ag/AgCl) for 20 cycles at the scan rate of 100 mV/s. The thickness of the chitosan coating was ~ 4 μm estimated from the SEM image of cross-section of a chitosan coated electrode (not shown). The resulting chitosan deposited Pt wire microelectrode was ready for the ELISA preparation and procedure.

electrodeposited chitosan as the matrix for immobilizing the capture antibody. The non-toxic chitosan provides good biocompatibility and high mechanical strength [43]; thus, it is an excellent material for fabricating biosensors [44–48]. Chitosan has the unique pH-dependent property: at lower pH, the amine groups of chitosan are protonated and soluble; oppositely, the amine groups are deprotonated and become insoluble at higher pH [49,50]. Thus, when electrodepositing chitosan, a high negative potential is applied to increase the local pH around the electrode and then, the soluble chitosan precipitates and deposited on the electrode. The capture antibody can be immobilized on the deposited chitosan by physical adsorption due to the trapping of the antibody in the chitosan matrix or the electrostatic interaction between chitosan and the antibody [51]. The secondary antibody linked with the enzyme label alkaline phosphatase (AP) was used. Alkaline phosphatase has been widely used in bioanalysis because of its high turnover number and broad substrate specificity [52]. AP catalyzes the hydrolysis of phosphate esters and produces inorganic phosphate and a phenolic group [53,54]. A variety of AP substrates has described and commercialized, such as phenyl phosphate, 4-aminophenyl phosphate (4-APP), α-naphthyl phosphate, and hydroquinone diphosphate (HQDP) and its electroactive products are phenol, 4-aminophenol (4-AP), α-naphthol, and hydroquinone (HQ), respectively [54]. Among these substrates, phenyl phosphate and 4-APP are usually used in electrochemical immunoassays. 4-APP has been shown to have a better electrochemical property than phenyl phosphate because it has reversible electrochemical behavior and its product 4-AP can be oxidized more easily than phenol (product of phenyl phosphate), so it will not foul the electrode even at high concentration [52]. The enzymatic reaction of 4-APP converted to 4-AP by the enzyme AP was described in Fig. 1.

2.3. Measuring oxidation potential of 4-AP on the Pt electrode

Fig. 1. The enzymatic reaction of 4-APP converted by alkaline phosphatase to generate 4-AP.

A Pt wire microelectrode immobilized with alkaline phosphatase (AP) was first prepared by crosslinking AP and BSA on the electrode using glutaraldehyde. The enzyme mixture for crosslinking contained

62


1.74 × 104 U/mL AP, 5.0 mg/mL BSA, and 0.125% glutaraldehyde and the mixture was coated manually on the electrode for 100 times using a 5.0 μL microsyringe. A testing solution containing 0 mM, 1 mM, and 4 mM 4-APP was prepared in 100 mM Tris buffer (pH 9.0). The CV tests from −0.2 to +0.3 V and from −0.2 to +0.7 V were performed for 10 cycles at the scan rate of 100 mV/s in the testing solution in a three electrode configuration consisting of a working electrode (Pt wire microelectrode immobilized with AP or the bare Pt electrode as the control), Pt wire auxiliary electrode, and a Ag/AgCl glass-bodied reference electrode. Then, the oxidation potential of 4-AP as the product of the enzymatic reaction of alkaline phosphatase was observed. 2.4. Assay preparation and procedure 2.4.1. Reagent preparation Purchased goat anti-estradiol Ab and mouse anti-estradiol Ab were aliquoted in the microcentrifuge tube and stored at −20 °C immediately upon arrival. Thawing of a frozen reagent was done rapidly before use. To be used in the array preparation and procedure, goat anti-estradiol Ab was diluted 5 times with the assay buffer and no dilution was required for mouse anti-estradiol Ab (5.0 mg/mL) which can be used directly. The secondary antibody (goat anti-mouse IgG labeled with AP) was diluted to 10 ng/μL with the assay buffer. β-estradiol calibrators were prepared by dissolving β-estradiol in DMF at 1.0 mg/mL followed by series diluting the solution with the assay buffer to provide estradiol calibrators with different concentrations (1.0 × 10− 1 pg/mL, 1.0 × 100 pg/mL, 1.0 × 101pg/mL, 1.0 × 102 pg/mL, 1.0 × 103 pg/mL, 1.0 × 104 pg/mL, and 1.0 × 105 pg/mL). All materials and reagents were equilibrated to room temperature prior to use. 2.4.2. Electrode coating The chitosan deposited Pt wire microelectrode was incubated in 5.0 μL diluted solution of goat anti-estradiol Ab (i.e. capture antibody) for 1 h at room temperature using a digital rotator. When incubation has been completed, the electrode was taken out and then washed and shaken three times with 20 μL washing buffer using the digital rotator for 5 min. To block unoccupied chitosan surface for avoiding nonspecific binding, the electrode was incubated in 20 μL blocking buffer for 1 h using the digital rotator. Then, the electrode was washed and shaken three times with 20 μL washing buffer using the digital rotator for 5 min. 2.4.3. Assay procedure The coated electrode was incubated in 10 μL sample solution of β-estradiol (i.e. calibrator or sample) for 1 h at room temperature using the digital rotator. When incubation has been completed, the electrode was taken out and then washed and shaken three times with 20 μL

washing buffer using the digital rotator for 5 min. Subsequently, the electrode was incubated in 5.0 μL solution of mouse anti-estradiol Ab (i.e. detection antibody) for 1 h at room temperature using the digital rotator. Then, the electrode was washed and shaken three times with 20 μL washing buffer using the digital rotator for 5 min. The electrode was incubated in 10 μL diluted solution of goat anti-mouse IgG labeled with AP (i.e. secondary antibody) for 1 h at room temperature using the digital rotator. Finally, the electrode was washed and shaken three times with 20 μL washing buffer using the digital rotator for 5 min. Once the ELISA electrode was dried, it was ready for the electrochemical tests. Electrochemical tests were performed by constant potential amperometry at 0.14 V vs. Ag/AgCl in 100 mM Tris buffer (pH 9.0) under stirring condition. Once the background current became stable (~10 min), the AP substrate (1 mM 4-APP) was added to the solution and the corresponding oxidation current from the AP product (4-AP) was recorded for 20 min. The average current signal over the 20 mininterval was used as the current response corresponding to certain estradiol concentration for establishing the calibration curve. The schematic diagram of the assay preparation and procedure was shown in Fig. 2. 2.5. Preparation of serum samples Purchased serum samples were aliquoted in the microcentrifuge tube and stored at − 20 °C immediately upon arrival. Thawing of a frozen serum sample was done rapidly before use. The thawed serum sample was aliquoted in small microcentrifuge tubes and was filtered using a syringe filter (pore size: 0.2 μm) before use. The baseline estradiol concentration was determined to be 5.21 × 101 pg/mL by chemiluminescence with Centaur Immunoassay System (Siemens, Munich, Germany) in the Union Clinical Laboratory (Taipei, Taiwan). In the assay procedure, 5 μL of the filtered serum sample without dilution was used for the sample incubation. For establishing one calibration curve, five serum samples spiked with different estradiol concentrations (total serum volume needed: 25 μL) and five ELISA electrodes were needed. The assay procedure for serum samples was similar to that of the buffer samples except the buffer samples were replaced by the serum samples during the incubation step for the analyte. 3. Results and discussion 3.1. The electrochemical impedance spectrum (EIS) of the pre-coated ELISA electrodes To investigate the effect of pre-coated layers (chitosan, the capture antibody, and the blocking reagent BSA) on the electron transfer resistance (Ret) at the surface of ELISA electrodes, the EIS of pre-coated ELISA electrodes in 5 mM K4Fe(CN)6, 5 mM K3Fe(CN)6, and 0.1 M KCl

Fig. 2. Procedure for preparing and performing the electrochemical ELISA for estradiol.


Fig. 3. EIS of the bare Pt electrode (■), the chitosan/Pt electrode (▼), the capture antibody/ chitosan/Pt electrode (▲), and the BSA/capture antibody/chitosan/Pt electrode (●).

was shown in Fig. 3. The EIS was measured at an open-circuit value of + 0.10 V between the frequency varied from 1.0 × 105 to 1.0 × 10−2 Hz and the ac excitation amplitude was 5 mV. The value of Ret at the electrode surface was determined by calculating the diameter of the semicircle in the Nyquist plot of EIS. The Ret of the bare Pt electrode was calculated to be 1489 Ω. The Ret of the chitosan coated electrode (chitosan/Pt) decreased to 1018 Ω; the decrease in Ret was probably due to the positively charged amine groups of chitosan which could promote the electron transfer between the negatively charged redox probe and the electrode surface [55,56]. After immobilizing the Fe(CN)3−/4− 6 capture antibody on the chitosan coated electrode, the Ret of the capture antibody/chitosan/Pt electrode increased to 1327 Ω. Subsequently, when the electrode was blocked with BSA, the Ret of the BSA/capture antibody/chitosan/Pt electrode further increased to 1422 Ω. The increase in Ret might be due to the exclusion between the negatively and the attached proteins. charged redox probe Fe(CN)3−/4− 6 3.2. Oxidation potential of 4-aminophenol on the platinum wire microelectrode In this study, alkaline phosphatase (AP) was chosen as the enzyme label on the secondary antibody. Since AP is able to catalyze a broad variety of phosphated compounds; therefore, the selection of a suitable enzyme substrate is important for generating an AP product which can be oxidized on the Pt electrode easily at a proper potential. Previous study proposed by Tang et al. [57] showed that 4-aminophenol (4-AP),

63

the product of AP using 4-aminophenyl phosphate (4-APP) as its substrate, could be oxidized on the carbon paste electrode easily at a low potential (+ 0.16 V vs. Ag/AgCl) in 50 mM carbonate buffer (pH 9.0) compared to phenol, another AP product using phenyl phosphate as its substrate, which was oxidized at + 0.55 V (vs. Ag/AgCl). Thus, when choosing 4-APP as the AP substrate, the current signal obtained by the oxidized product (4-AP) could be detected at a relatively low potential and therefore, it could prevent the fouling issues or avoid oxidation of the enzyme substrate 4-APP or other species on the electrode which could produce false positive current signals. In this experiment, we selected 4-APP as the AP substrate and the electrochemical behavior of 4-APP and 4-AP on the Pt electrode in the Tris buffer (pH 9.0) was investigated in order to determine the optimized oxidation potential of 4-AP on the Pt electrode. CV diagrams obtained from testing an AP immobilized Pt wire microelectrode in buffer solutions containing 0 mM, 1.0 mM, and 4.0 mM 4-APP were shown in Fig. 4. Two oxidation peaks at +0.14 V and +0.58 V were observed and these oxidation peaks increased as the concentration of 4-APP increased from 0 mM to 4.0 mM; based on the previous study shown by Tang et al. [57], the oxidation potential of the AP substrate, 4-APP, on the carbon paste electrode in the carbonate buffer (pH 9.0) was +0.58 V (vs. Ag/AgCl). Thus, the oxidation peak observed at + 0.58 V in Fig. 4(a) was suggested to be the oxidation peak of the AP substrate, 4-APP, on the Pt electrode rather than that of the product, 4-AP. It was also observed that the concentration of 4-APP presented in the solution would not affect the position of these oxidation peaks. In addition, constant potential amperometry at + 0.14 V was performed using the chitosan coated Pt wire microelectrode without AP (Chitosan-Pt) as well as the AP/chitosan Pt wire microelectrode (AP-chitosan-Pt) in Tris buffer (pH 9.0) before and after the addition of 4.0 mM 4-APP and the I–t curve was shown in Fig. 5. After the addition of 4-APP, an instant and obvious current step was observed on AP-chitosan-Pt; on the contrary, no current step was observed on Chitosan-Pt. It suggested that when the constant potential amperometry was performed at +0.14 V, little AP substrate (4-APP) would be oxidized on the Pt electrode and the enzyme product (4-AP) produced by AP could be oxidized on the Pt electrode at a relatively low potential (+0.14 V). Based on these results, the optimized oxidation potential of 4-AP for performing the electrochemical ELISA by constant potential amperometry was determined to be +0.14 V in this study. 3.3. Current responses of electrochemical ELISA for estradiol After the assay procedure, the ELISA electrodes were tested by constant potential amperometry at + 0.14 V in the analyzing solution

Fig. 4. CV diagrams obtained from testing an AP immobilized Pt wire microelectrode in the Tris buffer (pH 9.0) containing 0 mM (buffer), 1.0 mM, and 4.0 mM 4-APP; (a) −0.2 V to +0.7 V and (b) −0.2 V to +0.3 V.

64


Fig. 5. Current response (I) vs. time (t) at +0.14 V obtained from testing the chitosan coated Pt wire microelectrode (Chitosan-Pt) and the AP/chitosan Pt wire microelectrode (AP-chitosan-Pt) in Tris buffer (pH 9.0) before and after the addition of 4.0 mM 4-APP. The dashed arrow indicated the timing for the addition of 4-APP.

(100 mM Tris buffer at pH 9.0) under stirring condition. In Fig. 6, a representative plot of current responses of the electrochemical ELISA corresponding to different estradiol concentrations was demonstrated. These current responses (I) were recorded vs. time (t) continuously before and after the addition of the enzyme substrate 4-APP. When the substrate (1.0 mM 4-APP) was added to the analyzing solution, an instant and clear current step was observed. It was shown that the height of a current step increased as the concentration of estradiol presented in the assay sample increased. Since the concentration of 4-APP in the analyzing solution was fixed at 1.0 mM, the variation of the height of a current step was correlated to the concentration of the enzyme product 4-AP generated in the analyzing solution; that is, when more estradiol was captured during the assay procedure, more 4-AP could be produced since there was more enzyme labeled secondary antibody attaching to the ELISA electrode. Therefore, the calibration curve of the electrochemical ELISA for estradiol could be established by correlating the height of a current step (i.e. the current signal) to the concentration of estradiol presented in the assay sample. The height of a current step was calculated by subtracting the stable baseline current value before the addition of 4-APP from the stable current value after the addition of 4-APP. It was also observed that there was a current signal when the assay was tested with a sample free of estradiol (0 pg/mL) and this was probably due to the result of non-specific binding of the

Fig. 6. A representative plot of current responses of the electrochemical ELISA corresponding to different estradiol concentrations (0 pg/mL, 1.0 × 10−1 pg/mL, 1.0 × 100 pg/mL, 1.0 × 101pg/mL, 1.0 × 102 pg/mL, 1.0 × 103 pg/mL, 1.0 × 104 pg/mL, and 1.0 × 105 pg/mL). Electrochemical measurements were performed using chitosan coated Pt electrodes after the ELISA procedure by constant potential amperometry at +0.14 V vs. Ag/AgCl in 100 mM Tris buffer (pH 9.0) under stirring condition.

detection antibody or the secondary antibody onto the chitosan coated electrode. Even so, there was a clear discrimination between the current signal obtained from the estradiol sample at 1.0 × 10−1 pg/mL and that obtained from the estradiol sample at 0 pg/mL. The limit of detection of this assay was determined by interpolation at 3 standard deviations above the mean signal at background (0 pg/mL), using data from 3 standard curves. The effect of chitosan as the pre-coating material on the surface of the Pt wire microelectrode upon the immobilization of the capture antibody was investigated by comparing current signals corresponding to various concentrations of estradiol presented in assay samples measured on a chitosan coated Pt wire microelectrode with those measured on a bare Pt wire microelectrode. It was observed that when the ELISA was performed using the chitosan coated Pt electrode, the current signal was improved significantly. The current signals were improved 207%, 183%, and 197% corresponding to 1.0 × 10−1 pg/mL, 1.0 × 100 pg/mL, and 1.0 × 101 pg/mL estradiol presented in the assay sample, respectively. This result suggested that the current signal of the electrochemical ELISA was enhanced due to the modification of the Pt wire microelectrode with chitosan by which more capture antibody (i.e. goat anti-estradiol Ab) could be immobilized on the electrode and therefore more electroactive product 4-AP could be generated at the end. 3.4. Calibration of electrochemical ELISA for estradiol In Fig. 7, a typical calibration curve of the electrochemical ELISA for estradiol was demonstrated. It related the concentration of estradiol presented in the assay sample, ranging from 1.0 × 10−1 pg/mL to 1.0 × 105 pg/mL, to the current signal. The data showed that the current signal increased with increasing concentration of estradiol presented in the assay sample and then approached a saturation response as shown in Fig. 7(a) suggesting that almost all capture antibody was bound to the analyte estradiol when estradiol was at very high concentration in the assay sample. These calibration data could be linearized by plotting the current signal vs. the concentration of estradiol in log scale and fitting the data by the semi-log transformation using the equation (y = 11.77 log x + 40.12; R2 = 0.9866) as shown in Fig. 7(b). A linear trend was obtained up to 1.0 × 105 pg/mL estradiol, the highest concentration tested. The limit of detection of this assay was 2.7 × 10−1 pg/mL determined by interpolation at 3 standard deviations above the mean signal at background, using data from 3 standard curves. In summary, the detection range of this electrochemical ELISA for estradiol was from 2.7 × 10− 1 pg/mL to 1.0 × 105 pg/mL which covered levels of estradiol in women during the menstrual cycle (usually ~15– 350 pg/mL); therefore, the proposed electrochemical ELISA was suitable for detecting samples with low level of estradiol as well as samples with very high level of estradiol (e.g. samples from woman undergoing some infertility treatments). Compared to recent studies on electrochemical ELISA for estradiol listed in Table 1, our assay provided lowest detection limit (2.7 × 10− 1 pg/mL) and broadest detection range (2.7 × 10− 1 pg/mL to 1.0 × 105 pg/mL) when tested in buffer samples. The electrochemical indirect sandwich type ELISA for estradiol that we demonstrated has several key advantages compared to prior relevant studies listed in Table 1. Most of prior relavant studies relied on the competitive type ELISA [39–42]. In their studies, enzyme-estradiol conjugates were required for competing the binding site of the immobilized antibody; these conjugates included HRP-estradiol conjugate [39,42] and AP-estradiol conjugate [40,41]. However, the preparation of these conjugates, usually by the carbodiimide-mediated coupling reaction, was tedious and time consuming and they were expensive if obtained commercially. The assay we proposed was based on an indirect sandwich type ELISA and no modification of the antibody was required which made the preparation of the assay simpler. Instead of using milli-scale (diameter = 3–4 mm) carbon based electrodes as the ELISA working electrodes [39–42], the working electrode used in


65

Fig. 7. The calibration curve of electrochemical ELISA for estradiol plotted in (a) linear scale and (b) in semi-log scale with estradiol concentration ranging from 1.0 × 10−1 pg/mL to 1.0 × 105 pg/mL (N = 3). Electrochemical measurements were performed using chitosan coated Pt electrodes after the ELISA procedure by constant potential amperometry at +0.14 V vs. Ag/AgCl in 100 mM Tris buffer (pH 9.0) under stirring condition.

this study was a micron-scale (diameter = 50.8 μm) Pt wire and therefore, only little amount of sample volume (10 μL) and the reagent volume (5–10 μL) were required for the assay procedure. Most of prior relevant studies immobilized the detection antibody by oriented immobilization, including the passive adsorption of IgG for attaching the antibody [39–41] or binding between the immobilized streptavidin and the biotinylated anti-estradiol [42]. However, the passive adsorption of IgG usually required overnight incubation [39–41] and the immobilization of streptavidin involved the surface modification of the electrode and the carbodiimide-mediated coupling reaction; besides, its following assay steps were tedious [42]. In this study, the immobilization method involved the adsorption of the capture antibody on the chitosan electrodeposited Pt electrode which was very straightforward and simple. For those assays immobilizing the antibody based on passive adsorption of IgG, the total assay time was relatively long (N10 h) since the overnight incubation was required [39–41]; on the other hand, the assay demonstrated by Ojeda et al. had a superior total assay time (3.41 h) [42]. In the current study, the total assay time was 6.75 h (2.5 h for pre-coating steps and 4.25 h for the assay time); however, the incubation time of each assay steps may be further optimized and shortened and the assay efficiency may be enhanced by utilizing an automated assay system. 3.5. Assay specificity of electrochemical ELISA for estradiol To evaluate the specificity of the electrochemical ELISA for estradiol, tests of cross-reactivity tests were performed by comparing the assays response to some similar compounds, including progesterone and

17α-ethynylestradiol. Cross-reactivity for the noncompetitive sandwich assay was defined as the ratio of analyte (i.e. 17β-estradiol) concentration for which level of signal was computed to concentration of test compound corresponding to computed signal [58]. Usually, the response corresponding to the analyte concentration in the middle of the dose–response curve (i.e. calibration curve) was used; in this case, the cross-reactivity of the anti-17β estradiol antibody was evaluated at 1.0 × 102 pg/mL of 17β-estradiol. Based on this definition, the crossreactivity of progesterone and 17α-ethynylestradiol were found to be only 0.033% and 3.4%, respectively, which demonstrated excellent specificity of the electrochemical ELISA. 3.6. Calibration of electrochemical ELISA for serum estradiol The chitosan coated Pt wire microelectrodes were applied for performing the electrochemical ELISA for serum estradiol and the calibration curve was established. The serum sample with 5.21 × 101 pg/mL of baseline estradiol was spiked with β-estradiol to prepare serum estradiol calibrators at different concentrations: 5.21 × 101 pg/mL, 6.21 × 101 pg/mL, 1.05 × 103 pg/mL, 1.01 × 104 pg/mL, and 1.00 × 105 pg/mL. The calibration curve of the electrochemical ELISA for serum estradiol was demonstrated in Fig. 8. The calibration data showed that the current signal increased with increasing concentration of serum estradiol and then approached a saturation response at very high concentration of serum estradiol as shown in Fig. 8(a). These calibration data could also be linearized by plotting the current signal vs. the concentration of serum estradiol in log scale and fitting the data by the semi-log transformation using the equation (y = 18.41 log

Fig. 8. The calibration curve of electrochemical ELISA for serum estradiol plotted in (a) linear scale and (b) in semi-log scale (N = 3). Electrochemical measurements were performed using chitosan coated Pt electrodes after the ELISA procedure by constant potential amperometry at +0.14 V vs. Ag/AgCl in 100 mM Tris buffer (pH 9.0) under stirring condition.

66


x − 10.26; R2 = 0.9722) as shown in Fig. 8(b). A linear trend was obtained up to 1.00 × 105 pg/mL of serum estradiol, the highest spiked concentration tested. This result suggested that the electrochemical ELISA could also be used for detecting estradiol in serum. It was also observed that the slope and the intercept of the calibration curve obtained from the buffer samples and that obtained from the serum samples were different; the reason causing the difference is not very clear at present stage, but this might be due to combination effects of the binding of the sample estradiol to the endogenous serum proteins and the non-specific binding of the endogenous serum AP onto the chitosan coated electrode. The value of the current signal in this method depended on the amount of the enzyme label (AP) attached to the electrode. When the sample estradiol bound to the endogenous serum proteins, less AP was attached to the electrode; and therefore, lower current values were obtained which led to a lower intercept for the calibration curve obtained from the serum samples. On the other hand, it was also possible that the presence of endogenous AP in serum (b240 U/L for females and b270 U/L for males [59]) might increase the amount of AP attached to the chitosan coated electrode due to the non-specific binding, and thus gave false-positive responses for serum tests and contribute to the increased slope of the calibration curve obtained from the serum samples. Spiked serum samples were analyzed and precision values (R.S.D.%) were determined; satisfactory precision values were obtained ranging from 3.4 to 13.8% at higher estradiol concentrations. However, the precision values were 27.6 to 32.8% at lower estradiol concentrations (≤ 6.21 × 101 pg/mL) and the average recovery of three replicates of spiked serum samples at 5.21 × 101 pg/mL of estradiol was only 56%. The inferior precision and recovery at lower estradiol concentrations might be owing to the interferent effect from the untreated serum samples and this effect could be obvious at lower estradiol concentrations since a major portion of estradiol presented at low concentration in serum might bind nonspecifically to endogenous proteins. The further optimization on the pretreatment of the serum samples and the utilization of an assay automated system for obtaining better precision values and recovery are recommended. Additional work on real sample analysis is desired to further showing the feasibility of this assay.

4. Conclusions In this study, we proposed a simple and economic method to fabricate the ELISA electrode by electrodepositing chitosan on Pt wire microelectrodes for the immobilization of the capture antibody. The current signal of the assay was improved significantly when electrodes coated with chitosan were used. Compared to recent studies on electrochemical ELISA for estradiol, our assay provided lowest detection limit (2.7 × 10−1 pg/mL) and broadest detection range (2.7 × 10−1 pg/mL to 1.0 × 105 pg/mL) when tested in buffer samples. In addition, the fabrication method of our assay electrode was most straightforward and the electrode that we used was smallest (Pt wire microelectrode with diameter = 2.0 × 10−3 in.); thus, the electrode fabrication cost could be lowered and the amount of assay reagent used could be reduced as well. The assay has excellent specificity; the cross-reactivity of progesterone and 17α-ethynylestradiol were found to be only 0.033% and 3.4%, respectively. This electrochemical ELISA has been tested with estradiol in spiked serum samples; nevertheless, further pretreatment of serum samples is recommended to improve precision and recovery.

Acknowledgments Financial support by the Ministry of Science and Technology, Taiwan under grant numbers MOST 103-2221-E-011-135 and MOST 104-2221E-011-145 is gratefully acknowledged. The authors thank Dr. Chih-Ning Pao for his valuable suggestions on assay preparations.

References [1] Available online: U.S. National Library of Medicine, http://www.nlm.nih.gov/ medlineplus/ency/article/003711.htm. April 10th 2015. [2] M.W. Elmlinger, W. Kuhnel, M.B. Ranke, Reference ranges for serum concentrations of lutropin (LH), follitropin (FSH), estradiol (E2), prolactin, progesterone, sex hormone-binding globulin (SHBG), dehydroepiandrosterone sulfate (DHEAS), cortisol and ferritin in neonates, children and young adults, Clin. Chem. Lab. Med. 40 (2002) 1151–1160. [3] Available online: American Association for Clinical Chemistry, https://labtestsonline. org/understanding/analytes/estrogen/tab/test/ April 15th 2015. [4] W. Rosner, S.E. Hankinson, P.M. Sluss, H.W. Vesper, M.E. Wierman, Challenges to the measurement of estradiol: an endocrine society position statement, J. Clin. Endocrinol. Metab. 98 (2013) 1376–1387. [5] B. Yilmaz, Y. Kadioglu, Determination of 17β-estradiol in pharmaceutical preparation by UV spectrophotometry and high performance liquid chromatography methods, Arab. J. Chem. 2 (2013) 1–7. [6] Y. Yoon, P. Westerhoff, S.A. Snyder, M. Esparza, HPLC-fluorescence detection and adsorption of bisphenol A, 17 beta-estradiol, and 17 alpha-ethynyl estradiol on powdered activated carbon, Water Res. 37 (2003) 3530–3537. [7] J. Mitchell, Small molecule immunosensing using surface plasmon resonance, Sensors 10 (2010) 7323–7346. [8] M. Adamczyk, Y.-Y. Chen, J.C. Gebler, D.D. Johnson, P.G. Mattingly, J.A. Moore, et al., Evaluation of chemiluminescent estradiol conjugates by using a surface plasmon resonance detector, Steroids 65 (2000) 295–303. [9] H. Ou, Z. Luo, H. Jiang, H. Zhou, X. Wang, C. Song, Indirect inhibitive immunoassay for estradiol using surface plasmon resonance coupled to online in-tube SPME, Anal. Lett. 42 (2009) 2758–2773. [10] P.K. Tatrai, D. Bonds, L. Prokai, Simultaneous measurement of 17β-estradiol, 17αestradiol and estrone by GC-isotope dilution MS/MS, Chromatographia 71 (2010) 311–315. [11] J.O. Ström, A. Theodorsson, E. Theodorsson, Substantial discrepancies in 17βestradiol concentrations obtained with three different commercial direct radioimmunoassay kits in rat sera, Scand. J. Clin. Lab. Invest. 68 (2008) 806–813. [12] J. Geisler, D. Ekse, H. Helle, N.K. Duong, P.E. Lønning, An optimised, highly sensitive radioimmunoassay for the simultaneous measurement of estrone, estradiol and estrone sulfate in the ultra-low range in human plasma samples, J. Steroid Biochem. Mol. Biol. 109 (2008) 90–95. [13] A.P. Fonseca, J.M. Oliveira, V. Esteves, M. Cardoso, Using radioimmunoassay to investigate the uptake by amphibians of estrogens present in WWTP'S effluent, Pollut. Eff. Control 1 (2013) 1–4. [14] B.G. England, G.H. Parsons, R.M. Possle, D.S. McConnell, A.R. Midgley, Ultrasensitive semiautomated chemiluminescent immunoassay for estradiol, Clin. Chem. 48 (2002) 1584–1586. [15] J.D. Boever, A. Mares, G. Stans, E. Bosmans, F. Kohen, Comparison of chemiluminescent and chromogenic substrates of alkaline phosphatase in a direct immunoassay for plasma estradiol, Anal. Chim. Acta 303 (1994) 143–148. [16] M.L. Chiu, T.T.-C. Tseng, H.G. Monbouquette, A convenient, homogeneous enzymeimmunoassay for estradiol detection, Biotechnol. Appl. Biochem. 58 (2011) 75–82. [17] L. Kokko, N. Johansson, T. Lovgren, T. Soukka, Enzyme inhibitor screening using a homogeneous proximity-based immunoassay for estradiol, Biomol. Screen. 10 (2005) 348–354. [18] S.G. Korenman, R.H. Stevens, L.A. Carpenter, M. Robb, G.D. Niswender, B.M. Sherman, Estradiol radioimmunoassay without chromatography: procedure, validation and normal values, J. Clin. Endocrinol. Metab. 38 (2013) 718–720. [19] J. Osborn, A review of radioactive and non-radioactive-based techniques used in life science applications—part I: blotting techniques, Life Sci. News (Amersham Biosci.) (2000) 1–4. [20] S.S. Deshpande, Immunoassay Classification and Commercial Technologies, Enzyme Immunoassays: From Concept to Product Development, 1st ed.Chapman and Hall, New York, 1996. [21] R.M. Lequin, Enzyme immunoassay (EIA)/enzyme-linked immunosorbent assay (ELISA), Clin. Chem. 51 (2005) 2415–2418. [22] S. Pan, A. Ruedi, C. Ru, R. John, R.G. David, W.M. Martin, et al., Mass spectrometry based targeted protein quantification: methods and applications, J. Proteome Res. 8 (2009) 787–797. [23] M. Thompson, Immunoanalysis—part 2: basic principles of ELISA, Anal. Methods Comm. Tech. Briefs 45 (2010) 1–2. [24] S.D. Gan, K.R. Patel, Enzyme immunoassay and enzyme-linked immunosorbent assay, J. Investig. Dermatol. 133 (2013) 1–3. [25] K.L. Cox, D.P. Viswanath, A. Kriauciunas, M.B. Joseph, M.P. Chahrzad, S.P. Sitta, Immunoassay methods, in: G.S. Sittampalam, N.P. Coussens, H. Nelson, et al., (Eds.), Assay Guidance Manual, Eli Lilly & Company and the National Center for Advancing Translational Sciences, Bethesda, 2012. [26] F. Ricci, G. Adornetto, G. Palleschi, A review of experimental aspects of electrochemical immunosensors, Electrochim. Acta 84 (2012) 74–83. [27] Mercodia, Mercodia ELISA technology. Available online: http://www.mercodia.se/ learning-center/mercodia-elisa-technology/principle-of-technology.html May 1st 2015. [28] E. Osmekhina, A. Neubauer, K. Klinzing, J. Myllyharju, P. Neubauer, Sandwich ELISA for quantitative detection of human collagen prolyl 4-hydroxylase, Microb. Cell Factories 9 (2010) 1–11. [29] G.B. Wisdom, Enzyme-immunoassay, Clin. Chem. 22 (1976) 1243–1255. [30] A. Bhimji, A.A. Zaragoza, L.S. Live, S.O. Kelley, Electrochemical enzyme-linked immunosorbent assay featuring proximal reagent generation: detection of human immunodeficiency virus antibodies in clinical samples, Anal. Chem. 85 (2013) 6813–6819.

T.T.-C. Tseng et al. / Journal of Electroanalytical Chemistry 758 (2015) 59–67 [31] X. Liu, X. Wang, J. Zhang, H. Feng, X. Liu, D.K.Y. Wong, Detection of estradiol at an electrochemical immunosensor with a Cu UPD|DTBP–protein G scaffold, Biosens. Bioelectron. 35 (1) (2012) 56–62. [32] H. Sha, Y. Bai, S. Li, X. Wang, Y. Yin, Comparison between electrochemical ELISA and spectrophotometric ELISA for the detection of dentine sialophosphoprotein for root resorption, Am. J. Orthod. Dentofac. Orthop. 145 (2014) 36–40. [33] Y.S. Kim, H.S. Jung, T. Matsuura, H.Y. Lee, T. Kawai, M.B. Gu, Electrochemical detection of 17 beta-estradiol using DNA aptamer immobilized gold electrode chip, Biosens. Bioelectron. 22 (2007) 2525–2531. [34] X. Liu, D.K.Y. Wong, Electrocatalytic detection of estradiol at a carbon nanotube| Ni(Cyclam) composite electrode fabricated based on a two-factorial design, Anal. Chim. Acta 594 (2007) 184–191. [35] H. Kuramitz, M. Matsuda, J.H. Thomas, K. Sugawara, S. Tanaka, Electrochemical immunoassay at a 17 beta-estradiol self-assembled monolayer electrode using a redox marker, Analyst 128 (2003) 182–186. [36] S. Zhang, Y. Wang, Y. Zhang, T. Yan, L. Yan, Q. Wei, et al., An ultrasensitive electrochemical immunosensor for determination of estradiol using coralloid Cu2S nanostructures as labels, R. Soc. Chem. 5 (2014) 6512–6517. [37] R.S. Kelly, Analytical electrochemistry: the basic concepts, J. Anal. Sci. Digit. Libr. 1–3 (2009). [38] G. Li, P. Miao, Theoritical background of electrochemical analysis, Electrochemical Analysis of Proteins and Cells, Springer Science + Bussines Media, New York 2013, pp. 5–15. [39] R. Draisci, G. Volpe, D. Compagnone, Purificato, Q. Fd, G. Palleschi, Development of an electrochemical ELISA for the screening of 17 beta-estradiol and application to bovine serum, Analyst 125 (2000) 1419–1423. [40] G. Volpe, G. Fares, Q. Fd, R. Draisci, G. Ferretti, C. Marchiafava, et al., A disposable immunosensor for detection of 17 beta-estradiol in non-extracted bovine serum, Anal. Chim. Acta 572 (2006) 11–16. [41] R.M. Pemberton, T.T. Mottram, J.P. Hart, Development of a screen-printed carbon electrochemical immunosensor for picomolar concentrations of estradiol in human serum extracts, J. Biochem. Biophys. Methods 63 (2005) 201–212. [42] I. Ojeda, J. López-Montero, M. Moreno-Guzmán, B.C. Janegitz, A. González-Cortés, P. Yá ñez-sedeño, et al., Electrochemical immunosensor for rapid and sensitive determination of estradiol, Anal. Chim. Acta 743 (2012) 117–124. [43] L.F. Ang, L.Y. Por, M.F. Yam, Study on different molecular weights of chitosan as an immobilization matrix for a glucose biosensor, PLoS One 8 (8) (2013), e70597. [44] X. Wei, J. Cruz, W. Gorski, Integration of enzymes and electrodes: spectroscopic and electrochemical studies of chitosan-enzyme films, Anal. Chem. 74 (19) (2002) 5039–5046. [45] M. Zhang, C. Mullens, W. Gorski, Coimmobilization of dehydrogenases and their cofactors in electrochemical biosensors, Anal. Chem. 79 (6) (2007) 2446–2450.

67

[46] S. Karra, W. Gorski, Signal amplification in enzyme-based amperometric biosensors, Anal. Chem. 85 (21) (2013) 10573–10580. [47] T.T.-C. Tseng, J. Yao, W.-C. Chan, Selective enzyme immobilization on arrayed microelectrodes for the application of sensing neurotransmitters, Biochem. Eng. J. 78 (2013) 146–153. [48] T.T.-C. Tseng, C.-F. Chang, W.-C. Chan, Fabrication of implantable, enzymeimmobilized glutamate sensors for the monitoring of glutamate concentration changes in vitro and in vivo, Molecules 19 (2014) 7341–7355. [49] R. Fernandes, L.-Q. Wu, T. Chen, H. Yi, G.W. Rubloff, R. Ghodssi, et al., Electrochemically induced deposition of a polysaccharide hydrogel onto a patterned surface, Langmuir 19 (2003) 4058–4062. [50] H. Yi, L.-Q. Wu, W.E. Bentley, R. Ghodssi, G.W. Rubloff, J.N. Culver, et al., Biofabrication with chitosan, Biomacromolecules 6 (2005) 2881–2894. [51] S.T. Koev, P.H. Dykstra, X. Luo, G.W. Rubloff, W.E. Bentley, G.F. Payne, et al., Chitosan: an integrative biomaterial for lab-on-a-chip devices, Lab Chip 10 (22) (2010) 3026–3042. [52] W. Khalid, G. Göbel, D. Hühn, J.-M. Montenegro, P. Rivera-Gil, F. Lisdat, et al., Light triggered detection of aminophenyl phosphate with a quantum dot based enzyme electrode, J. Nanobiotechnol. 9 (2011) 1–10. [53] G. Volpe, D. Compagnone, R. Draisci, G. Palleschi, 3,3′,5,5′-tetramethylbenzidine as electrochemical substrate for horseradish peroxidase based enzyme immunoassays. A comparative study, Analyst 123 (1998) 1303–1307. [54] N.J. Ronkainen-Matsuno, H.B. Halsall, W.R. Heineman, Electrochemical immunoassay and immunosensor, in: J.M. van Emon (Ed.), Immunoassay and Other Bioanalytical Techniques, 1st ed.Taylor and Francis Group, Florida 2007, pp. 385–402. [55] M. Li, S. Huang, P. Zhu, L. Kong, B. Peng, H. Gao, A novel DNA biosensor based on ssDNA/Cytc/l-Cys/GNPs/Chits/GCE, Electrochim. Acta 54 (2009) 2284–2289. [56] F. Kuralay, T. Vural, C. Bayram, E.B. Denkbas, S. Abaci, Carbon nanotube-chitosan modified disposable pencil graphite electrode for vitamin B12 analysis, Colloids Surf. B: Biointerfaces 87 (2011) 18–22. [57] H.T. Tang, C.E. Lunte, H.B. Halsall, W.R. Heineman, p-Aminophenyl phosphate: an improved substrate for electrochmical enzyme immunoassay, Anal. Chim. Acta 214 (1988) 187–195. [58] M.N. Khan, P.D. Dass, J.H. Leete, R.F. Schuman, M. Gunsior, C. Sadhu, Development of ligand-binding assays for drug development support, in: M.N. Khan, J.W.A. Findlay (Eds.), Ligand-Binding Assays: Development, Validation, and Implementation in the Drug Development Arena, John Wiley & Sons, New Jersey 2010, pp. 50–52. [59] S. Turan, B. Topcu, I. Gökçe, T. Güran, Z. Atay, A. Omar, et al., Serum alkaline phosphatase levels in healthy children and evaluation of alkaline phosphatasez-scores in different types of rickets, J. Clin. Res. Pediatr. Endocrinol. 3 (1) (2011) 7–11.