METHOD FOR DETECTING AN ANALYTE WITH THE HELP OF METAL NANOPARTICLES ON AN ELECTRODE

Abstract

A method for detecting at least one analyte by electrochemical detection, a working electrode of an analyte sensor and an analyte sensor for detecting at least one analyte in a sample by electrochemical detection. The method comprises contacting a fluid sample suspected to comprise the at least one analyte with the surface of an electrode comprising a binding agent capable of binding to the analyte; contacting the fluid sample with a detection agent comprising a further binding agent capable of binding to the analyte and a label, the label comprising a metal nanoparticle with a standard redox potential E° between 0 V and 1.2 V forming a detection complex on the surface of the electrode comprising the binding agent, the detection agent and the analyte precipitating at least a part of the label onto the electrode surface; and detecting the analyte by electrochemical detection.

Claims

1. A method for detecting at least one analyte by electrochemical detection, comprising: (i) contacting a fluid sample suspected to comprise the at least one analyte with the surface of an electrode comprising at least one binding agent capable of binding to the analyte; (ii) contacting the fluid sample suspected to comprise the at least one analyte with a least one detection agent, wherein the at least one detection agent comprises at least a further binding agent capable of binding to the analyte and a label, wherein the label comprises a metal nanoparticle with a standard redox potential E° between 0 V and 1.2 V; (iii) forming a detection complex on the surface of the electrode comprising at least the at least one binding agent, the at least one detection agent and the analyte; (iv) precipitating at least a part of the label onto the electrode surface; and (v) detecting the at least one analyte by electrochemical detection; wherein the method comprises a step of dissolving at least a part of the label by oxidation prior to precipitating at least a part of the label in step (iv), and wherein the step of dissolving at least a part of the label by oxidation comprises applying a voltage at the electrode suitable to oxidize the metal nanoparticle, wherein a positive potential higher than the oxidation potential of the respective metal nanoparticle is applied as this voltage, wherein the applied voltage results in direct oxidation of the metal nanoparticle.

2. The method according to claim 1, further comprising: a step of forming an analyte complex on the surface of the electrode comprising the at least one binding agent and the analyte, prior or simultaneously to step (iii) of forming of the detection complex.

3. The method according to claim 1, wherein the label comprises or consists of a metal nanoparticle of silver or alloys thereof; specifically a silver nanoparticle, or a metal nanoparticle comprising or consisting of an silver/gold alloy; more specifically a citrate-capped or citrate-stabilized silver nanoparticle.

4. The method of claim 3, wherein the positive potential is applied as a voltage in the range of 1.0 to 1.5 V.

5. The method according to claim 1, wherein the at least one binding agent is immobilized on the surface of the electrode exposable to the fluid sample, preferably via a linker.

6. The method according to claim 1, wherein the at least one binding agent and/or the at least one further binding agent are selected from antibodies and fragments thereof, nucleic acids, aptamers, peptide nucleic acids (PNAs), receptor or ligand proteins or peptides, and enzymes; preferably the binding agent and/or the further binding agent comprises an antibody, more preferably an IgG.

7. The method according to claim 1, wherein the detection agent is stored in a dry state and solubilized upon the addition of the fluid sample.

8. The method according to claim 1, wherein the electrochemical detection in step (v) comprises pulse voltammetry measurements, in particular differential pulse voltammetry (DPV) measurement.

9. The method according to claim 1, wherein the method comprises applying a negative potential following precipitating in step (iv).

10. The method according to claim 1, wherein the analyte is a protein, a carbohydrate, a polynucleotide or a lipid, preferably a protein, more preferably a protein present in a body fluid.

11. The method according to claim 1, wherein the precipitating in step (iv) comprises contacting a counter ion with the dissolved part of the label to allow for forming of an insoluble salt that precipitates from the solution.

12. A device for detecting at least one analyte by electrochemical detection, comprising: a) at least one working electrode, wherein the surface of the working electrode comprises at least a linker for coupling the at least one binding agent, and wherein the working electrode is configured to allow for the formation of a detection complex comprising at least one detection agent, the at least one binding agent and the analyte on the surface of the electrode exposable to a fluid sample, wherein the at least one detection agent comprises at least one further binding agent and a label, wherein the label comprises a metal nanoparticle with a standard redox potential E° between 0 V and 1.2 V; and operatively linked thereto; b) an analyzing unit configured for applying a positive potential higher than the oxidation potential of the respective metal nanoparticle as a voltage at the electrode suitable to oxidize the metal nanoparticle, and configured for measuring the detectable signal produced by the detection agent when detecting the at least one analyte; and optionally c) an evaluating unit.

13. The device according to claim 12, wherein the device is adapted to perform the method according to any one of claims 1-11.

14. The device according to claim 12, wherein the evaluating unit is configured to determine the amount of the analyte based on the detectable signal measured by the analyzing unit.

15. The device according to any of claim 12, wherein the amount of the analyte is determined by comparing the detectable signal measured by the analyzing unit to a reference.

16. The device according to any of claim 12, further comprising at least one reference and/or at least one counter electrode.

17. Use of a metal nanoparticle with a standard redox potential E° between 0 V and 1.2 V as a label in electrochemical detection of an analyte by pulse voltammetry measurement.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0127] FIG. 1 shows a schematic representation of electrochemical detection using AgNPs. The detection of AgNPs consists of four steps: In the first step, the AgNPs used as labels are located close to the WE surface. In a second step, the silver nanoparticles are oxidized in presence of KCl and AgCl precipitates immediately on the surface. In step three, the Ag.sup.+ ions are reduced upon formation of a silver layer. The oxidation in the last step is monitored by DPV.

[0128] FIGS. 2A-D depict a plot of the ratio of maximum absorbance in presence of 2 M NaCl to the maximum absorbance of the pure AgNP solution against the amount of probe AB used in the modification process (A). Non-blocked AB-AgNPs were used for this measurement. Error bars represent mean values+1σ and were calculated based on three parallel measurements (n=3). Change of hydrodynamic diameter (circle, left axis) and PdI (triangle, right axis) of blocked AB-AgNPs with changing amount of probe AB (B). Error bars represent mean values±1σ and were calculated based on three parallel measurements (n=3). Exemplary TEM image of the blocked AB-AgNP modified with 10 μg probe AB, scale bar represents 100 nm (C). Plot of peak area of the bioassay using a constant AG concentration of 100 ng.Math.mL.sup.−1 against the amount of probe AB used in the modification process of AgNPs (D). Error bars represent mean values+1σ and were calculated based on three parallel measurements on three different SPCEs (n=3).

[0129] FIG. 3 depicts the change of hydrodynamic diameter (circle, left axis) and PdI (triangle, right axis) of AB-AgNPs (modified with 10 μg AB, blocked with BSA) over 56 days after modification (left plot). Stability of the electrochemical signal of the bioassay using a constant AG concentration of 100 ng.Math.mL.sup.−1 over 56 days. Peak area was normalized to the signal right after the modification (right plot). Error bars represent mean values±1σ and were calculated based on three parallel measurements on three different SPCEs (n=3).

[0130] FIG. 4 depicts exemplary differential pulse voltammograms of the bioassay with AgNPs for differently concentrated antigen solutions (0-1000 ng.Math.mL.sup.−1, light to dark) normalized with respect to the baseline (left) and plot of corresponding peak area against logarithm of antigen concentration with logistic fit (solid line) and corresponding parameters (right). In the left plot, the x-axis depicts the applied potential [V], the y-axis the normalized current [μA], in the right plot the x-axis represents the antigen concentration [ng mL.sup.−1], the y-axis represents the peak areas [V μA]. The intersection of the dashed line with the measuring curve indicates the LOD. Standard deviations were calculated based on five parallel measurements on five different SPCEs, while outliers were removed after Q-test (confidence interval 95%). Error bars represent mean values±1σ (n≥4).

[0131] FIG. 5 shows a plot of peak area ([V μA], y-axis) against logarithm of antigen concentration ([ng mL.sup.−1], x-axis) in spiked human serum samples with logistic fit (solid line) and corresponding parameters. The intersection of the dashed line with the measuring curve indicates the LOD. Standard deviations were calculated based on five parallel measurements on five different SPCEs, while outliers were removed after Q-test (confidence interval 95%). Error bars represent mean values±1σ (n≥4 5).

[0132] FIG. 6 shows the DPV of the performance of the Bioassay with 1000 ng.Math.mL.sup.−1 analyte concentration with oxidation at 1.25 V for 30 s (black) versus oxidation at 0.8 V for 30 s (grey); the x-axis depicts the applied potential [V], the y-axis the current [μA].

[0133] FIG. 7 shows the correlation of peak area ([V μA], y-axis) and peak height ([μA], x-axis). Standard deviations of peak area and peak height were calculated based on five parallel measurements on five different SPCEs, while outliers were removed after Q-test (confidence interval 90%). Error bars represent mean values±1σ (n=5). Each data point correlates to one AG concentration.

[0134] FIG. 8 shows DPV measurements ([μA], y-axis) of various AgNP solutions ([pM], x-axis), dried on top of DropSens electrodes (left) and plot of peak area with AgNP concentration (right).

[0135] FIGS. 9A-B show the dose-response curve for the flow chips generation one with pre-incubation of the AG and AB-AgNP (A) and the flow chips with dried AB-AgNPs in trehalose matrix (B). The plot of peak area against logarithm of antigen concentration in PBS buffer (50 mM phosphate) with logistic fit (solid line) and corresponding parameters is depicted. The intersection of the dashed line with the measuring curve indicates the LOD. Standard deviations were calculated based on parallel measurements using different flow chips (A: flow chip generation one with pre-incubation of AG and AB-AgNP, B: flow chip generation two with dried AB-AgNP in trehalose matrix), while outliers were removed after Q-test (confidence interval 95%). Error bars represent mean values±1σ.

[0136] FIG. 10 depicts generation two measurements (with dried AB-AgNPs) in human serum samples, spiked with the respective AG concentration. Shown is the plot of peak area against logarithm of antigen concentration in human serum with logistic fit (solid line) and corresponding parameters. The intersection of the dashed line with the measuring curve indicates the LOD. Standard deviations were calculated based on parallel measurements using different flow chips (generation two with dried AB-AgNP in trehalose matrix), while outliers were removed after Q-test (confidence interval 95%). Error bars represent mean values±1σ.

EXAMPLES

[0137] The following examples serve to illustrate the invention. They must not be interpreted as limiting with regard to the scope of protection.

Example 1: Chemicals and Instruments

[0138] The biotinylated capture antibody (polyclonal proBNP sheep-IgG-biotin), antigen (NT-proBNP (1-76) amid) in buffer or human serum, probe antibody (monoclonal NTproBNP mouse-IgG) and antibody modified gold nanoparticles (monoclonal NTproBNP mouse-IgG-gold) were provided by Roche Diagnostics GmbH, Mannheim, Germany. Citrate capped silver nanospheres (d=50 nm, 0.022 mg.Math.mL.sup.−1) were purchased from nanoComposix (www.nanocomposix.com). Hydrochloric acid (HCl, 0.1 M, 1 M), sodium chloride (NaCl, p.a), disodium hydrogen phosphate (Na.sub.2HPO.sub.4 2H.sub.2O, p.a.) and potassium dihydrogen phosphate (KH.sub.2PO.sub.4, p.a.) were ordered from Merck (www.merckmillipore.com). Bovine serum albumin (BSA, >96%) and Tween 20 (>97%) were sup-plied from Sigma Aldrich (www.sigmaaldrich.com). Potassium chloride (KCl, p.a.) was obtained from Roth (www.carlroth.com). 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES, ≥99%) was acquired from VWR (de.vwr.com) and sodium hydroxide (NaOH, 1 M) was bought from Labochem international (www.labochem.de). HEPES buffer consisted of 10 mM HEPES and was adjusted to a pH of 7.4. HEPES blocking buffer was prepared by addition of 0.1% (w/v) BSA to this HEPES buff-er. Phosphate buffered saline (PBS) consisted of 137 mM NaCl, 2.7 mM KCl, 38 mM Na2HPO4.Math.2 H.sub.2O and 12 mM KH2PO.sub.4 with a pH of 7.4. For PBST washing buffer 0.05% (w/v) Tween 20 were added to this PBS.

[0139] All electrochemical measurements were performed using screen-printed carbon electrodes (SPCE) bare (DRP-110) or with streptavidin coated working electrode (DRP-110STR, both: Metrohm AG, www.dropsens.com) and an EmStat blue potentiostat with corresponding software (PalmSens, www.palmsens.com). For nanoparticle modification, the TermoMixer comfort (Eppendorf, online-shop.eppendorf.de) was used. A Plate reader Synergy Neo2 Hybrid Multi-Mode Reader from BioTek Instruments (BioTek Instruments Inc., www.biotek.de), a Malvern Zetasizer Nano-ZS (www.malvern.com) and a 120 kV Philips CM12 (www.fei.com) transmission electron microscope were employed for characterization of the modified AgNPs.

Example 2: Electrochemical Detection of Gold and Silver Nanoparticles

[0140] To test the electrochemical detection of both metallic nanoparticles (gold and silver), 10 μL of the respective NP solution (diluted in HEPES buffer) are dried on top of the working electrode of the SPCE (DRP-110). Immediately afterwards, 50 μL of 0.1 M HCl for the gold measurement or 0.1 M KCl for silver respectively, are added and the electrochemical measurement is started. The measurement procedure for gold nanoparticles consists of a pretreatment and the actual differential pulse voltammetry. First a voltage of 1.25 V is applied for 60 s, then DPV is performed from 1.25 V to 0 V with t.sub.puls=50 ms, E.sub.step=10 mV, E.sub.pulse=80 mV and scan rate=20 mV.Math.s.sup.−1. For silver nanoparticles an equivalent, but slightly different procedure is used. For pretreatment, a voltage of 1.25 V is applied for 60 s, then −0.8 V for 30 s. The DPV is recorded from −0.25 V to 0.6 V with same settings (Hao et al., 2011).

Example 3: Modification of Silver Nanoparticles

[0141] For the AgNP modification, a modified procedure of Szymanski et al. (2013) was utilized. A volume of 1 mL of AgNP solution (0.02 mg.Math.mL.sup.−1) is centrifuged for 10 min at 10,000 g. The supernatant is discarded and the pellet resuspended in 1 mL HEPES with different amount of probe AB. After incubation at room temperature with gentle mixing (350 rpm) for 2 h in the dark, the nanoparticles are centrifuged once again for 10 min at 10,000 g. The supernatant is discarded and the pellet resuspended in 1 mL HEPES or HEPES blocking buffer. For characterization of those particles, UV/Vis measurements are performed first. Four AgNP solutions, modified with 1, 5, 10 and 20 μg ABs in HEPES are adjusted to the same concentration (according to maximum absorbance). Then, 200 μL of each solution are pipetted into a transparent 96 well MTP (Greiner, shop.gbo.com) and the maximum absorbance at 340 nm is measured. After addition of 50 μL 2 M NaCl and mixing for 2 min, the maximum absorbance is measured once again. Four additional AgNP solutions are modified with 1, 5, 10 and 20 μg ABs in HEPES blocking buffer. These blocked AB-AgNPs are used for all further experiments. The characterization is completed by DLS measurements at 25° C. in disposable PMMA cuvettes (semi-micro), TEM measurements and the performance of the bioassay with 100 ng.Math.mL.sup.−1 AG concentration.

Example 4: Performance of the Bioassay

[0142] Prior to usage, the SPCEs (DRP-110STR) are washed three times with 50 μL PBS buffer. For the bioassay, 10 μL of the respective solution are dropped onto the working electrode. After incubation for 1 h at room temperature under water saturated atmosphere the solution is removed and the electrode washed three times with 50 μL PBST. After drying with nitrogen, the next solution is added. The exact sequence of solutions is shown in Table 1. After step 4, the electrodes are washed additionally with PBS and double distilled water. Directly prior to the electrochemical measurement, the electrode is dried with nitrogen and the three-electrode area covered with 50 μL of a 0.1 M HCl or KCl, respectively. The measurement is performed as described above.

TABLE-US-00001 TABLE 1 SEQUENCE OF STEPS WITH RESPECTIVE SOLUTIONS, ALL DILUTED IN PBS. Step Solution.sup.a 1 immobilization of capture antibody 25 μg .Math. mL.sup.−1 biotinylated capture AB 2 blocking 1% (w/v) BSA 3 antigen binding 0-3000 ng .Math. mL.sup.−1 AG 4 Probe antibody binding 7.1 ng .Math. mL.sup.−1 AuNP-tagged probeAB 20 μg .Math. mL.sup.−1 probe AB modified AgNP.sup.b .sup.aFirst three solutions were diluted in PBS, labelled probe ABs were diluted in HEPES. .sup.bDue to different concentration data given by the manufacturer, dilutions of AB-AuNP and AB-AgNP cannot be given in a consistent form. However, this dissimilar approach did not interfere with assay optimizations.

Example 5: Electrochemical Detection of Metallic Nanoparticles

[0143] For the electrochemical detection of both kinds of metallic nanoparticles, already published procedures were adjusted. The processes on the electrode surface are shown in FIG. 1. The gold nanoparticle detection was discussed in more detail by La Escosura-Muñiz et al., 2011. In the first step, the AuNPs are dried on the working electrode of the SPCE. After addition of hydrochloric acid, the particles dissolve upon oxidation at 1.25 V (step 2). Hereby, a gold chlorido complex forms near the electrode surface. In the third step, the reduction of bound Au.sup.3+ ions at 0.25 V is monitored by DPV. The hydrochloric acid is inevitable for the initial oxidation of the very stable AuNPs, because the complex is only formed at a pH around one.

[0144] For the silver nanoparticle detection (FIG. 1) an equivalent but slightly more complex approach was used (Hao et al., 2011). Here, the AgNPs dissolve by oxidation at 1.25 V (step 2) after drying (step 1). Due to the higher instability of silver compared to gold, already 0.8 V would be enough to oxidize the bare nanoparticles. However, the presence of proteins on the NP surface requires a higher potential to clean the NP surface and thus get a measureable DPV signal. The Ag.sup.+ ions then form an AgCl precipitate on the electrode surface with the present chloride, which hinders the diffusion away from the electrode surface (Ting et al., 2009). The third step consists of a reduction at −0.8 V to form a silver layer, which penetrates into the pores of the electrode material. In the last step, the oxidation around 0.025 V is monitored by DPV. This additional step renders the electrochemical detection considerably more sensitive and sharper peaks are obtained. The main advantage of silver over gold is that no addition of hydrochloric acid is needed due to its higher chemical instability. Since chloride ions are contained in biological samples, this enables an one-step analysis.

[0145] As proof of concept, differential pulse voltammetry was performed with both different mNPs. A concentration dependence of peak area and height was found (FIG. 7, FIGS. 4 and 8: left side). Due to marginal smaller errors, the peak area was used for all following data evaluation. The plot of peak are against NP concentration shows the same course for both metals and is shown exemplary for AgNPs in FIG. 8. First, the signal increases linearly with the amount of NP on the surface, while a constant value is reached after saturation of the electrode surface. This shows that the DPV detection method can be used for both, gold and silver nanoparticles.

Example 6: Performance of the Bioassay with Gold Nanoparticles

[0146] First, different techniques to immobilize the AG on the working electrode were tested. No gold signal at 0.25 V was obtained neither for a direct adsorption of the antigen (AG), nor for a covalent binding of the capture AB using EDC/NHS chemistry. Adsorption processes are highly dependent on the protein-electrode combination, which is used. In this case, the AG was washed away even after incubation overnight, since the interaction was too weak. The covalent AB immobilization itself worked according to impedance studies. The absence of a gold signal could be due to the blocking of the electrode by the pyrene butyric acid, which was used as anchor moiety, and the increased distance between NP label and electrode surface. Thus, a third approach exploiting streptavidin/biotin binding on purchased streptavidin-functionalized electrodes (DRP-110STR) was performed. The streptavidin is not coated on top of the electrode, but rather included in the conductive material. With this method, the AG was bound to the electrode and the presence of gold nanoparticles was measured at 0.25 V due to decreased NP-electrode distance and blocking of the electrode. In the next step, the capture AB concentration was varied and an optimal concentration of 25 μg.Math.mL.sup.−1 was found. The working electrode seems to be completely covered and further increasing the capture AB concentration did not change the signal. Moreover, different dilutions of the purchased AuNP-tagged probe AB solutions were used for the bioassay. Due to a drastically higher signal, the probe ABs were used in a 1:10 dilution. Using the AuNP solution without any dilution worsened the signal-to-noise (S/N) ratio due to an increase in background. With these optimized parameters, the bioassay was performed.

[0147] For better visualization of the concentration dependent peak heights, the voltammograms are normalized with respect to the baseline, which means the current difference to the current measured at 0.45 V is plotted. In the plot of the peak area against logarithm of AG concentration, a sigmoidal binding curve can be seen. The limit of detection (LOD) was calculated using the logistic fit parameter for the lower border A1 and the standard deviation of the blind SD (blind):

LOD=A1+3.Math.SD(blind)  (1)

[0148] A concentration value of 26 ng.Math.mL.sup.−1 was calculated based on the logistic fit. The mean error of all measurements is 17% and the dynamic range extends over nearly two orders of magnitude from 25 to 1000 ng.Math.mL.sup.−1. Overall, the gold bioassay is easy to perform and provides reliable results. However, the addition of hydrochloric acid is cumbersome considering a true POC application and the detection is not sensitive enough to use it for blood analysis, where NT-proBNP is contained in the pg.Math.mL.sup.−1 range. To solve both problems the metal used was changed to silver.

Example 7: Modification of Silver Nanoparticles

[0149] Purchased silver nanoparticles were modified with different amounts of probe AB to find the optimal loading density. UV/Vis analysis was performed to prove the passive adsorption of AB to AgNPs worked and see the stabilizing effect of the proteins (Hao et al., 2011).

[0150] In high ionic strength media, bare NPs tend to aggregate and in consequence, their color changes from yellow to transparent. For the different modifications (with 1 to 20 μg.Math.mL.sup.−1 AB), the maximum absorbance after addition of 2 M NaCl was divided by the original maximum value. Non-blocked particles had to be used for this study, since AB-AgNPs would not show any signal change after blocking. The ratio of both maxima, shown in FIG. 2 (A), reaches one when 10 μg AB are used in the modification and stays constant for higher amounts. This means, that a minimum modification concentration of 10 μg.Math.mL.sup.−1 is needed to preserve the NPs from agglomeration, i.e. to cover the NP surface completely. In the DLS measurements FIG. 2 (B) it can be seen that the hydrodynamic diameter and Polydispersity Index (PdI) decrease until a constant value of around 75 nm and 0.160 is reached for AB-AgNPs modified with 10 μg. A complete coverage of the surface with ABs decreases the overall size and increases uniformity, because less BSA adheres to the nanoparticle during blocking. The hydrodynamic diameter increases by about 25 nm compared to the bare NPs (52 nm, given by the manufacturer) due to protein uptake.

[0151] The TEM images FIG. 2 (C) show that the AgNP core size is not affected by the modification procedure and no aggregates are formed. The differently modified AB-AgNPs were then used in the bioassay with a constant AG concentration of 100 ng.Math.mL.sup.−1 FIG. 2 (D). With increasing amount of probe AB, the peak area decreases slightly, but the standard deviations show a huge improvement. This proposes an increased uniformity of the NPs with increasing surface coverage. In the following, AgNPs modified with 10 μg probe AB were used, since their surface is completely covered and they show excellent uniformity. Since silver nanoparticles are known to be quite unstable due to aggregation and oxidation, a stability study was performed next. Dynamic light scattering (DLS) measurements and the bioassay with a constant AG concentration were performed over eight weeks after modification (FIG. 3).

[0152] The hydrodynamic diameter and PdI decrease minimally within the first three days (FIG. 3 left). Afterwards, a constant value is reached. This is likely caused by a slow release of BSA loosely attached to the particle in the first days after modification until an equilibrium is reached. Application of the prepared and stored AB-AgNPs in the bioassay (FIG. 3 right) shows that they are stable as signals do not change within the margin of error observed for a period of 4 weeks. After eight weeks, the peak area drops to 50% of the original value. This indicates that the overall assay and signal enhancement strategy is rugged, but for final application, further studies are needed to increase the storage stability of the particles. The short-term stability of non-blocked AB-AgNPs were also tested. For short-term stability, blocking with BSA made no significant difference. However, to optimize long-term stability and avoid any unspecific binding in the final application, the blocked AB-AgNPs were used for all further measurements.

Example 8: Performance of the Bioassay with Silver Nanoparticles

[0153] The bioassay was performed with silver nanoparticle-labelled probe AB, while using the optimum parameters found with the gold assay. Exemplary differential pulse voltammograms in an AG concentration range from 0 to 1000 ng.Math.mL.sup.−1 are shown in FIG. 4 (left plot). These voltammograms are normalized with respect to the baseline using the current value at 0.05 V. The peaks have a considerably smaller full-width at half maximum (FWHM) compared to gold, which is due to the additional step in the detection. Moreover, the signal of the electrochemically more active silver arises at a ten-time smaller potential. This is always beneficial considering interferences in biological samples.

[0154] The plot of corresponding peak area against logarithm of AG concentration shows a sigmoidal curve (FIG. 4 right). Using the logistic fit and equation 1, a LOD of 4.0 ng.Math.mL.sup.−1 was calculated, which is around six-times lower than for the AuNP assay, while the reproducibility with a mean error of all measurements of 17% is comparable. Interestingly, the dynamic range between 4 and 100 ng.Math.mL.sup.−1 is narrowed. Considering the better analytical performance of the AgNP assay accompanied with the increased simplicity as no further solutions is needed, the AgNP bioassay highly valuable for a POC application.

[0155] Finally, demonstrating its true applicability, the AgNP bioassay was tested with real samples, specifically analyses were performed in human serum samples, spiked with different amount of NT-proBNP (FIG. 5). Most of the parameters are similar to those of the silver bioassay in buffer: the curve shape and with it, the dynamic range. The calculated LOD of 4.7 ng.Math.mL.sup.−1 is marginally higher due to background adsorption of serum proteins. The mean error of 15% is even slightly better. Thus, overall, the assay conditions and washing are suitable for application in real samples.

[0156] Two bioassays with metal nanoparticles for signal enhancement were investigated with respect to their applicability towards point-of-care sensing. In the case of AuNPs, these excel due to an excellent analyte concentration range, i.e. from 25 to 1000 ng.Math.mL.sup.−1 with very good signal to noise ratios. Moreover, it is well known that gold nanoparticles are easy to modify and stable over a longer period of time. However, due to their greater electrochemical activity, AgNPs provide a six-times more sensitive assay. Most importantly, the AgNP bioassay is significantly simpler and hence more adaptable to a POC setting as no further addition of any solution is necessary once it is used in a biological sample. Of importance here is also the long-term storage stability of the antibody-modified AgNPs.

[0157] In FIG. 6, DPV measurements demonstrate the performance of the Bioassay. The Bioassay was conducted using 1000 ng.Math.mL.sup.−1 analyte concentration with oxidation at 1.25 V for 30 s (black) versus oxidation at 0.8 V for 30 s (grey); the x-axis depicts the applied potential [V], the y-axis the current [μA].

Example 9: Microfluidic System

[0158] In here, we report the successful transfer of the AgNP assay into a microfluidic system. After optimization of the experimental setup and conduct and adjustment of the assay parameters, a very low LOD of 0.27 ng/mL with a mean error of 25% (n=5) was obtained (FIG. 9). This is an improvement by a factor of 15 in comparison to the assay performed on free electrodes (LOD of 4 ng/mL). The reproducibility may be improved but is a good representation of the many manually performed steps. The transfer of the original assay into a flow format bears significant implications on a future POCT as it demonstrates that reactions can all occur in flow and that the current AgNP signal generation system lends itself well for the envisioned future format. Most importantly, this flow system provides an LOD in a clinical relevant concentration range.

[0159] To obtain a flow cell an oval shape is cut out of double-sided adhesive tape (d=100 μm) by a laser cutter. After gluing it together with a PMMA top, a rectangular shape is cut out. These channel tops are stored for later usage. Next, the streptavidin coated DropSens electrode (DRP-110STR) was washed three times with 50 μL PBS and 10 μL of capture AB (polyclonal proBNP sheep-IgG-biotin) were added to the working electrode and incubated for 1 h at room temperature. After washing three times with 50 μL PBST the electrode was dried under nitrogen and glued together with the channel top to form the finished flow cell.

[0160] Two pumps and a T-piece allow an exchange of solution without getting bubbles into the closed system. The flow chip is placed in a chip holder with in- and outlet. The outlet tubing leads into a waste or a buffer reservoir depending on the mode the system is operated in. The two different modes are called injection and withdrawal mode and depend on the pumping direction. While washing buffer and KCl are injected via the syringes, blocking buffer and AG/AB-AgNPs are injected from the right side in withdrawal mode in order to avoid bubbles and ensure small solution consumption. The AG+AB-AgNP mix consists of 5 μL NT-proBNP (1-76) amid and 5 μL monoclonal NTproBNP mouse-IgG-silver, incubated for 2 h at room temperature before injection. After filling of the microfluidic channel with KCl, the flow chip is taken out of the chip holder, connected to a potentiostat and the electrochemical measurement is performed. This last step was done since the current chip holder does not allow connection to the potentiostat.

[0161] During assay development, the general microfluidic setup and flow scheme, as well as flow velocity, oxidation potential and pre-incubation time were varied. The optimized parameters were used to establish a dose-response curve with different AG concentrations (FIG. 9) with respect to further use of this microfluidic system with probe AB dried directly in the channel (generation two); the final injected concentration of AG is given on the x-axis.

[0162] FIG. 9 shows the dose-response curve for the flow chips generation one with pre-incubation of the AG and AB-AgNP (A) and the flow chips with dried AB-AgNPs in trehalose matrix (B; generation two).

[0163] For the flow chips generation one with pre-incubation of the AG and AB-AgNP (A) the microfluidic assay has a LOD of 0.27 ng/mL, which is a huge improvement compared to the free assay with 4.0 ng/mL. This sensitivity enhancement by a factor of 15 has two reasons. The AG/AB interaction is not limited by slow diffusion of the big nanoparticles due to small channel height, and the blind signal and standard deviation is decreased due to better and more intense washing. The mean error is 25%. There are two main weak points: due to an irregular baseline, peak picking is difficult and inconsistent pumping makes injection of constant volumes impossible. Still this is a proof of principle that the silver nanoparticle assay can be performed in one-step in flow and reaches a clinical relevant concentration range. As a side note, the signal intensity is lowered by around 50% compared to the assay on free electrodes, yet the signal-to-noise ratio is even better. This may be due to a reduction of working electrode area and change of oxidation potential during assay optimizations. Latter was necessary to inhibit bubble evolution in the channel.

[0164] For the flow chips with dried AB-AgNPs in trehalose matrix (FIG. 9 B; generation two), The calculated LOD is with 0.24 ng.Math.mL.sup.−1 very close to the 0.27 ng.Math.mL.sup.−1 for the microfluidic assay with pre-incubation of AG and AB-AgNP. Moreover, after optimizations of the procedure and flow chips, the relative error was decreased from 25% (n≥4) to 9% (n≥3) and the logistic fit is considerably better (R.sup.2=0.999).

[0165] The higher ground signal may probably be caused by residuals of dried NPs in the channel. The overall higher signal and clearer peaks enable a more reproducible data evaluation. This, in combination with the easier experimental conduct, leads to an excellent reproducibility.

[0166] The generation two measurements (with dried AB-AgNPs) were repeated in human serum samples, spiked with the respective AG concentration (FIG. 10). The LOD of 0.57 ng.Math.mL.sup.−1 is slightly higher than for the assay in buffer presumably because of higher errors of the blind due to protein adsorption. The mean standard deviation is excellent, merely 6% (n≥3).

CITED LITERATURE

[0167] Hao N, Li H, long Y, Zhang L, Zhao X, Xu D, Chen H-Y (2011) An electrochemical immunosensing method based on silver nanoparticles. Journal of Electroanalytical Chemistry 656(1-2):50-54. [0168] La Escosura-Muñiz A de, Parolo C, Maran F, Mekoçi A (2011) Size-dependent direct electrochemical detection of gold nanoparticles: application in magnetoimmunoassays. Nanoscale 3(8):3350-3356. [0169] Pollok N E, Rabin C, Walgama C T, Smith L, Richards I, Crooks R M (2020) Electrochemical Detection of NT-proBNP Using a Metalloimmunoassay on a Paper Electrode Platform. ACS Sens. 5: 853-860. [0170] Pumera M, Aldavert M, Mills C, Merkoçi A, Alegret S (2005) Direct voltammetric determination of gold nanoparticles using graphite-epoxy composite electrode. Electrochimica Acta 50(18):3702-3707. [0171] Szymanski M, Turner A F, Porter R (2010) Electrochemical Dissolution of Silver Nanoparticles and Its Application in Metalloimmunoassay. Electroanalysis 22(2):191-198. [0172] Szymanski M S, Porter R A (2013) Preparation and quality control of silver nanoparticle-antibody conjugate for use in electrochemical immunoassays. Journal of immunological methods 387(1-2):262-269. [0173] Ting B P, Zhang J, Gao Z, Ying J Y (2009) A DNA biosensor based on the detection of doxorubicin-conjugated Ag nanoparticle labels using solid-state voltammetry. Biosensors & bioelectronics 25(2):282-287.