Surface Epitope Coverage Affects Binding Characteristics of Bisphenol-A Functionalized Nanoparticles in a Competitive Inhibition Assay

1School of Chemical Engineering and ARC Training Centre for Advanced Technologies in Food Manufacture, University of New South Wales, Sydney, NSW 2052, Australia 2Key Laboratory of Food Nutrition and Safety, Ministry of Education, Tianjin Key Laboratory of Food Nutrition and Safety, Tianjin University of Science and Technology, Tianjin 300457, China 3School of Chemistry and Australian Centre for NanoMedicine, University of New South Wales, Sydney, NSW 2052, Australia 4GREMAN, François Rabelais University, Parc de Grandmont, 37200 Tours, France


Introduction
Due to the size-dependent optical and electronic properties of gold nanoparticles (AuNPs), the volume of literature on their application in sensing is immense [1][2][3].In electrochemical sensors, AuNPs have been used to enhance refractive index changes [4,5], increase the surface-volume ratio, provide high conductivity, and accelerate electron transfer to a redox active species [6][7][8].To fabricate AuNPs-based biosensors, AuNPs have been modified with short peptides or organic molecules [8,9].The organic molecules could be a ligand or an epitope to which an antibody could specifically bind.The modification of AuNPs with small organic molecules is a construct with considerable potential, particularly in immunosensing.Amplification of analytical signals that are made possible by gold nanoparticles is even more important for small molecule detection than macromolecule detection.The antibody binding of small molecules to a biosensing interface is notoriously hard to detect otherwise.To date, however, the vast majority of studies employing organic molecule functionalized AuNPs are motivated towards detecting protein [3,8,10] rather than small molecules such as pesticides, endocrine disruptors, and pharmaceutical or veterinary drugs.
The most relevant study where gold nanoparticles are used in immunosensing for detecting small molecules is 2 Journal of Nanomaterials by Taguchi et al. [11].In that study, an analogue of the endocrine disruptor, bisphenol-A (BPA), was used to modify AuNPs for the fabrication of a competitive surface plasmon resonance (SPR) based sensor.This competitive assay for small molecules had a limit of detection of 0.1 mM (equivalent to 22.8 mg L −1 ).The relevance of the Taguchi study for the present work is that AuNPs modified with small organic molecules are a central component in an immunoassay.We are interested in using organic modified nanoparticles in ELISA and as analogues of hapten-protein conjugates for the generation of antibodies.There have not been any studies in which gold nanoparticles modified with a small organic molecule epitope have been characterized using the standard tool of immunochemistry.The combination of ELISA and nanoparticles is potentially useful, not only as a rapid approach to characterizing the functionality of a nanoparticle as a sensing element [6,10], but also as an assay reagent.
Small molecule analytes are typically quantified by a competitive inhibition format in an immunoassay [12].Based on the law of mass action, assay sensitivity is proportional to the value of equilibrium constant (i.e., the amount of the antigen-antibody complex).In a competitive inhibition assay, there are two equilibrium constants to be considered; the equilibrium constant involving the antigen itself is denoted by   and the other involving the labelled competing antigen (often referred to as hapten) is denoted by   * , as illustrated in the equations below: Assay sensitivity is predominantly influenced by two factors: (1) the relative affinity of an antibody between the target molecule ([Ag]) and the competing hapten ([Ag * ]) which is labelled with an enzyme for measurement and (2) the epitope density of the competing hapten coupled to an enzyme as a label (  versus   * ).For the first factor, selection of an appropriate hapten structure with a right linker for the enzyme conjugation is critical in achieving the greatest possible sensitivity by controlling the relative affinity of the antibody between the target molecule and the competing molecule (  >   * ).For the second factor, the epitope density of the hapten-enzyme conjugate provides multivalence in order to enhance the binding affinity, particularly useful for the weak binding antibodies in a polyclonal population.This phenomenon has been well demonstrated in many immunoassays developed for small molecules such as endocrine disrupting chemicals, pesticides, and mycotoxins [13,14].The concentration of the competing hapten in an assay influences the assay affinity (i.e., avidity) in a competitive immunoassay.The concentration of the competing hapten in an assay is generally determined as the concentration of the conjugate concentration rather than the concentration of hapten.This is because the epitope density of the hapten-enzyme conjugate is heterogeneous and could not be readily determined.Therefore, batch-to-batch variation is typical, making quality control of assay reagents such as hapten-enzyme conjugate somewhat difficult.Nanoparticles can serve as a good platform for targeted functionalization with small molecule epitopes.The purpose of this paper is to demonstrate the application of AuNPs modified with small molecule epitopes for binding to antibodies in a competitive inhibition format.More importantly, the ability of the AuNPs to support multiple either homologous or heterologous epitopes means that more than one antibody can bind to the AuNPs in a manner reminiscent of a sandwich ELISA although the nature of the ELISA is a competitive inhibition.Given the inherent complexities of the interaction of surface epitopes with antibodies, however, which may be different then the biological conjugate (e.g., protein-hapten conjugates) in a competitive inhibition event, we sought to investigate factors that may affect the antibody-surface epitope interaction and the subsequent analytical performance in a competitive inhibition assay.A better understanding of quantitative relationship between structure and activity in terms of avidity relating to multivalent interaction in a competitive inhibition manner will provide information about the future design of targeted nanoparticles for immunosensing of small molecule analytes.In this study, the surface bound epitope is a thiolated bisphenol-A derivative and the antibodies used are anti-BPA antibodies we have described previously [15].

2.2.
Instrumentation.An ELISA plate reader (SpectroMax M2) was obtained from Molecular Devices (Sunnyvale, USA).Centrifugation was performed using a Sigma 1-14 Laboratory Table Top Microcentrifuge (16,163 ×g for 15 min).Nuclear magnetic resonance (NMR) measurements were performed on a Bruker Avance III 300 MHz.UV-Vis measurements were performed on a Varian Cary 50 Bio UV-Visible spectrophotometer.

Preparation of BPA Standard. Standard stock solution
(1 × 10 6 g L −1 ) of BPA was prepared by dissolving an appropriate amount of BPA in MeOH.Working standard solutions (0.51-10,000 g L −1 ) were prepared by diluting the stock solution in 10% EtOH/phosphate buffered saline (PBS) just before use.

i-ELISA.
The % conjugation of cysBPAv to AuNP was determined by measuring the amount of unconjugated cys-BPAv remaining in the supernatant by i-ELISA.The i-ELISA was conducted as follows.BPA-valerate-BSA was immobilized onto a microplate at 10 g mL −1 .The microwell plate was washed three times with washing solution (0.05% Tween-20) and blocked with 3% skim milk power in PBS (pH 7.4) for 1 h.After washing the plate as previously described, the cysBPAv standard solutions and Ab∝BPA-V2#4 antiserum were added to the respective wells and the mixture was incubated for 1 h.The plate was washed 4 times with washing solution, and the anti-rabbit IgG conjugate dissolved in PBS was added to all the wells and the mixture was incubated for 30 min.After washing the plate five times, substrate solution (TMB) was added to all the wells to develop the color.The color reaction was stopped after 20 min by adding 50 L of 1.25 mol L −1 sulphuric acid into each well and the absorbance values of the wells were measured at 450 nm.In this way, the more unconjugated cysBPAv are in solution, the less Ab∝BPA-V2#4 binds to the immobilized BPA-valerate-BSA and hence the lower absorbance after the color reaction is.

Two-Antibody Competitive Inhibition ELISA (ci-ELISA).
Microwells were immobilized with Ab∝BPA-V2#4 antibody at 10 g mL −1 .After washing the microwell plate three times with the washing solution, 1% soybean protein in PBS (SBP-PBS, pH 7.4) was incubated in the wells for 1 h.The plate was washed 3 times again.BPA standard solutions in 10% EtOH/PBS and cysBPAv-AuNPs in 10% EtOH/PBS were added to the respective wells and the mixture was incubated for 30 min.The plate was washed again, and Ab∝BPA-V2#4biotin in 10% EtOH/PBS was added and incubated for 1 h.
After washing the plate, avidin-HRP conjugate in PBS was added to the wells and incubated for 30 min.The plate was washed 5 times this time to remove any unbound reagents and nanoparticles; then substrate solution (TMB) was added to all the wells to develop the color.The absorbance values were measured at 450 nm after the color reaction was stopped after 20 min by adding 50 L of 1.25 mol L −1 sulphuric acid (Figure 1).

Characterization of cysBPAv-AuNPs
Using i-ELISA.The surface epitope coverage affecting the avidity of the Ab∝BPA-V2#4 antibody for the nanoparticles modified with the 5 different epitope coverage was studied using the indirect-ELISA at the nanoparticle molar concentrations in the range between 3.7 × 10 −10 and 4.7 × 10 −13 mol L −1 .As the Stokes-Einstein radius of IgG (56 Å) is relative to the average diameter of AuNP (5 nm), we made an assumption that only one cysBPAv-AuNP will bind to one antibody at each incubation step largely due to the steric constraints [15,17].Upon binding to the immobilized antibody, the remaining cysBPAv epitopes on the nanoparticle are potentially free to bind to the Ab∝BPA-V2#4-biotin conjugate as the detection antibody much like a sandwich-like configuration.The antibodyepitope binding was determined using a HRP-streptavidin conjugate that complexes with the biotinylated detection antibody.The use of avidin-biotin system in this assay proved to lower the nonspecific binding as well as allowing for signal amplification for more reliable measurement (data not shown).
The surface epitope coverage was estimated to span a range from 3.0 × 10 −10 to 15 × 10 −10 mol cm −2 (Table 1), representing an approximately fivefold change in the epitope molar concentrations.As illustrated in Figure 2, the absorbance increased with increasing epitope coverage on the nanoparticles.Nanoparticles with lower epitope coverage required higher molar particle concentrations to reach the same absorbance value as those with higher epitope coverage which required lower molar particle concentrations.For instance, the absorbance generated using 3 × 10 −10 mol L −1 cysBPAv-AuNP-163 (i.e., particle with lower epitope coverage) was the same as using 8 × 10 −11 mol L −1 cysBPAv-AuNP-801 (i.e., particle with higher epitope coverage).This indicated that the interaction between cysBPAv epitopes and antibodies particularly with the biotinylated antibody was facilitated by having multiple epitopes on the nanoparticles, which enhanced the binding strength (avidity).Based on the calculated epitope coverage, minimum of approximately 160 epitopes per particle (estimated to be roughly 20% surface coverage) was needed to generate a detectable and reproducible signal.Below this number, the antibody binding showed large deviation in the measurement (data not shown).
In short, these results show the importance of surface epitope coverage in providing more efficient multivalent interaction between antibody and surface epitopes in enhancing the assay avidity.Such findings are also observed in the previous studies using noncompetitive assays [18][19][20].

ci-ELISA under Different Molar Concentrations of cysBPAv-AuNP-265.
In order to study the relationship between surface epitope coverage (on nanoparticles) and multivalent interaction with an antibody, a competitive inhibition enzyme-linked immunosorbent assay (ci-ELISA) using two BPA-specific antibodies was devised.In this assay, the free BPA and cysBPAv-AuNP were allowed to compete with each other to bind to the immobilized antibody.The signal was generated via the second biotinylated antibody binding to the bound cysBPAv-AuNP.Avidin-HRP/TMB substrate system was added to generate the colorimetric signal.If free BPA was bound to an antibody, then this binding site was no longer free to bind to the cysBPAv-AuNP and hence does not generate color.CysBPAv-AuNP, on the other hand, had multivalent epitopes conjugated to the particle surface, some of which would still be available to bind to the biotinylated antibody.Therefore, the ci-ELISA devised in this manner could be used to quantify free BPA based on the degree of cysBPAv-AuNP competitively bound to the immobilized antibody relative to the free BPA.As for a competitive assay, the color development is inversely proportional to the concentration of free BPA.
The three factors selected for the study were the surface epitope coverage, the molar concentration of epitope in an assay, and the molar concentration of particle (i.e., cysBPAv-AuNPs).In an ELISA, both the epitope coverage of haptenprotein conjugates and concentration of epitope in an assay influence the linear detection range and an IC 50 .Thus, the epitope coverage of competing hapten-protein conjugate and its quantity in an assay are typically optimised to yield the lowest IC 50 .So it was relevant to study these factors with nanoparticles functionalized with small molecule epitopes.Nanoparticles serve as carriers of epitopes in this study; however, their physical structures may interfere with antibodyantigen interaction.Whether the diffusion of nanoparticle, which is dependent on their concentrations, influences the antibody binding to the surface epitope in a competitive assay has yet to be explored.
In the first experiment, the surface epitope coverage was fixed by using cysBPAv-AuNP-265 to investigate the effects of the molar concentration of nanoparticles and the molar concentration of epitopes in an assay on the competitive binding.The three molar concentrations of nanoparticles (6.6 × 10 −10 mol L −1 , 3.3 × 10 −10 mol L −1 , and 1.3 × 10 −10 mol L −1 ) selected were above the minimum particle concentration required for generating detectable signal in the indirect ELISA as determined above (1.2 × 10 −11 mol L −1 of cysBPAv-AuNP-265).
From the results (Figures 3(a) and 3(b) and Table 2), a typical sigmoidal curve of the competitive inhibition assay was observed, showing a decrease in absorbance with an increase in free BPA concentration.This indicated that the cysBPAv-AuNP behaved much like a multivalent hapten-protein conjugate in an immunoassay.As shown in Figure 3(b) and Table 2, the IC 50 value decreased eightfold (from 8 mol L −1 to 1.2 mol L −1 BPA) when the molar concentration of cysBPAv-AuNPs was reduced roughly 5-fold (from 6.6 × 10 −10 to 1.3 × 10 −10 mol L −1 ).When the molar concentration of cysBPAv-AuNPs decreased from 6.6 × 10 −10 mol L −1 to 1.3 × 10 −10 mol L −1 , the affinity constants (  ) for the respective molar concentration of surface epitopes were 0.13 ± 0.02, 0.23 ± 0.03, and 0.62 ± 0.09 L mol −1 , showing the binding affinity of antibody for BPA increased with the decrease of the surface epitopes.This result indicates that the molar concentration of nanoparticle and the molar concentration of epitope in an assay had a significant influence on the linear detection range and limit of detection in a competitive inhibition assay of this type.The assay sensitivity as measured by the slope of the standard curves was not significantly 1.9-30.20.9-18.60.2-5.6 * 20-80% of the % curve.BPA (mol L −1 ) influenced by the concentration of particles or epitopes.This is consistent with the assumption that AuNPs serve as a carrier for epitopes but do not participate in the binding process themselves.It was noted that, at lower epitope concentration or lower surface epitope coverage, less free BPA was required to compete with the nanoparticles for antibody binding sites (Figure 4 and Table 3).That is, the IC 50 decreased with a decrease in surface epitope coverage as did the linear detection range.Naturally, of course, this also means the molar concentration of the surface epitope also showed dose dependent effects.However, in contrast to the concentration of nanoparticles discussed in Section 3.4, the sensitivity was affected by the surface epitope coverage and reduced at higher surface epitope coverage as evidenced by the lower slope of the cysBPAv-AuNP-801 dose-response curve.The affinity constants of the antibody for cysBPAv-AuNP-190, cysBPAv-AuNP-396, and cysBPAv-AuNP-801 decreased with increasing surface epitopes (Figure 4 and Table 3).These results indicated that the binding affinity of antibody to the surface epitope was likely to be dominated by the concentration of particles.Both surface epitope coverage and the molar concentration of epitope in an assay influenced the apparent sensitivity.It was not possible, however, to distinguish which of the two factors influenced this parameter more strongly.

ci-ELISA with the Same Molar Concentration of Epitope in an
Assay of cysBPAv-AuNP-190, cysBPAv-AuNP-396, and cysBPAv-AuNP-801.Finally, in the third set of experiments, the molar concentration of epitope in an assay was fixed at 1.01 × 10 −7 mol L −1 in order to compare the effects of the surface epitope coverage and the molar concentration of particle in assay performance.The selected cysBPAv-AuNP, therefore, exhibited different particle molar concentrations.The cysBPAv-AuNP with the lowest surface epitope coverage had the highest molar concentration of particle in an assay: cysBPAv-AuNP-190 at particle concentration  As shown in Figure 5 and Table 4, the nanoparticles with different surface epitope coverage, particularly cysBPAv-AuNP-190 and cysBPAv-AuNP-396, exhibited similar maximum absorbance values (i.e., absorbance at saturation), sensitivities, and IC 50 values.The affinity constants of these nanoparticles at the same molar surface epitope concentration had no significant difference.The almost overlapping of the dose-response curves suggested that the molar concentration of nanoparticles probably had little effects on the competitive inhibition between free BPA and cysBPAv-AuNP for the antibody binding.This result again supports the assertion that the AuNPs are simply carriers for the epitopes and it is the epitope concentration and surface coverage that are the important variables for the assay sensitivity.

Conclusions
Interaction between an antibody and multivalent surface epitopes is a means to enhance antibody avidity for weak binding antigens.The effects of multivalent interaction in the competitive inhibition assay, however, have not previously been studied.In this study, we explored the interaction between an antibody and multivalent surface epitopes of a small molecule in a competitive inhibition assay.The competitive inhibition assay using a 96-microwell platform was devised using the two-antibody approach with the avidinbiotin signal enhancement system.Surface epitopes with varying epitope densities were synthesized by functionalizing AuNPs with various quantities of cysBPAv to yield surface coverage between 163 and 801 epitopes per nanoparticle.
The surface epitope coverage influenced the antibody avidity which was increased with higher surface epitope coverage.There was minimum surface epitope coverage which was required to give reliable antibody-epitope interaction for detection, and this value should be determined for each different platform.In this study, the minimum epitope coverage was estimated to be 163 epitopes per nanoparticle which was equivalent to 20% coverage.Higher surface epitopes provided stronger and more consistent interaction between antibody and surface epitopes.
The three different sets of competitive inhibition experiments conducted suggested that all three studied factors (i.e., surface epitope coverage, assay molar nanoparticle concentration, and assay molar concentration of epitope) affected the assay affinity, but each to different degree, and these factors were related with one another.The surface epitope coverage was important in increasing the avidity

Table 3 :
Assay parameters and calculated affinity constant derived from Figure4.