Molecular Recognition of C-Reactive Protein, Adiponectin and Zn²⁺ in Serum Samples

C-reactive protein, zinc chloride, human adiponectin, protoporphyrin IX (PIX), diamond powder and graphite powder were purchased from Sigma Aldrich, and the paraffin oil was purchased from Fluka (Buchs, Switzerland). 2,6-bis (E)-2-(thiophen-3-yl)-4-4-(4,6,8-trimethylazulen-1-1yl) pyridine was synthesized in the house.³⁰ Graphene powders modified with Pt nanoparticles were synthesized by National Institute for Research and Development of Isotopic and Molecular Technologies and Development, Cluj-Napoca, Romania. All materials used were of analytical purity.

C-Reactive Protein solutions were prepared from a stock solution with a concentration of 0.1 mg/mL using the method of successive dilutions over a range of concentrations from 100 mg/mL to 16.4 fg/mL. The stock solution of adiponectin was 25mg/mL and using the method of serial dilutions, solutions up to 2.5fg/mL were prepared. For the determination of Zinc ions in the human serum samples was used the zinc chloride; the concentration of the stock solution was 136mg/mL, and by serial dilutions method solutions up to 13.6fg/mL were prepared. All solutions were prepared in phosphate buffer pH 7.40.

All measurements were performed with a potentiostat AUTOLAB/PGSTAT 302N (Metrohm) and recorded with the GPES software. An electrochemical cell of three-electrode system was used. The reference electrode was Ag/AgCl in 0.1mol/L KCl, the auxiliary electrode was a platinum wire and the working electrode was the stochastic microsensor. Measurements were made at a constant potential of 125 mV using stochastic method.

Diamond, graphite and platinum nanoparticles modified-graphene powders were used in the design of three stochastic microsensors. Each powder was homogenized with paraffin oil and then an electroactive material was added. A solution of 2,6-bis (E)-2-(thiophen-3-yl)-4-4-(4,6,8-trimethylazulen-1-1yl) pyridine (PyD) has been added to the diamond (PyD/DP) and graphite pastes (PyD/GP), while a solution of protoporphyrin IX was added to the graphene paste (PIX-Gr-Pt). Three plastic tips have been filled separately with the modified pastes and a silver wire was inserted in order to achieve the electrical contact. Before each measurement, the microsensors surface was washed with deionized water and then dried. When these microsensors are not used, they are stored at room temperature in a dark and dry place.

Stochastic mode was used for the simultaneously assay of C-reactive protein, human adiponectin and zinc ions in the serum samples. All measurements were performed at room temperature (25°C). Chronoamperometry technique was employed at 125mV. Calibration measurements were performed for each microsensor and for each of the analyte solutions. The molecular recognition of C-reactive protein, human adiponectin and zinc ions was done based on their signature (t_off value, Figure 1). The quantification of the analytes was done using the t_on values. For each microsensor the calibration equation was statistically evaluated using linear regression method: 1/t_on = a + b x Concentration. The unkmown concentrations were calculated by introducting the t_on values obtained in the diagrams of serum samples in the equations of calibration. All measurements were performed at 25°C.

Zoom In Zoom Out Reset image size

Serum samples were collected from a group of 10 patients investigated in the National Institute of Diabetes, Nutrition and Metabolic Diseases "Pr. Dr. N. Paulescu" from Bucharest, with the informed consent from all subjects, and accordingly with the Ethics Committee approval number 75/2015. The apparatus cell was filled with the serum sample and the stochastic measurements were performed. The unknown concentrations were determined from the calibration equations as described in the recommended procedure section.

The response of the stochastic microsensors is based on the conductivity of the channel. When a potential of 125mV vs Ag/AgCl was applied, the molecules are driven to the channel, and they getting one by one in the channel accordingly with their size, unfolding speed, speed of passing through the channel, geometry and stereochemistry; while they enter the channel the intensity of the current is droping to zero, the time spent to enter the channel being known as the signature of the molecule/biomarker (t_off value). Inside the channel the molecule was undergoing processes of binding accordingly with the reactions shown below, as well as redox processes:

where Ch is the channel, i is the membrane-solution interface, CRP is the C-reactive protein and AdipoQ is the adiponectin.

The response characteristics of the proposed stochastic microsensors were shown in Table II. The signatures of CRP, adiponectin and Zn²⁺ are different when the same stochastic microsensor was used proven that the three biomarkers can be simultaneous determined in the serum samples.

For the determination of CRP, the largest linear concentration range (1.64 × 10⁻⁸–0.41mg mL⁻¹), as well as the lowest limit of determination (1.64 × 10⁻⁸mg mL⁻¹) and the highest sensitivity (3.92 × 10⁴ s⁻¹/mg mL⁻¹) were recorded when the stochastic microsensor based on PIX and Gr-Pt paste was used. For the determination of adiponectin, the largest linear concentration range (2.50 × 10⁻⁸–0.25μg mL⁻¹) was recorded when the stochastic microsensors based on PyD/GP and PyD/DP were used while the higher sensitivity (1.58 × 10⁴s/μg mL⁻¹) was obtained for the stochastic microsensor based on PyD/DP; the same limit of determination was determined for all three proposed stochastic microsensors (2.50 × 10⁻⁸μg mL⁻¹). For the determination of Zn²⁺, the largest linear concentration range (1.36 × 10⁻¹⁰ – 1.00μg mL⁻¹), the lowest limit of determination (1.36 × 10⁻¹⁰μg mL⁻¹) as well as the highest sensitivity (4.41 × 10⁶s/μg mL⁻¹) were recorded for the stochastic microsensor based on PyD/GP.

The linear concentration ranges recorded using the proposed stochastic microsensors were largest than those shown in Table I, proposed to date for the same biomarkers. Also, the limits of determination reported in this paper for each microsensor, and each biomarker are lowest than those reported in Table I.

The stochastic microsensors were used for determination of CRP, adiponectin, and Zn²⁺ for more than ten months, when the RSD (%) values of the sensitivities did varied with 1.25%. Five stochastic microsensors of each type were design using the proposed design method; for each microsensors there were determined for three months the response characteristics; a difference on the sensitivity of 0.80% was recorded for each type of microsensor, when compared the values recorded for each microsensor during this period; these proved that the proposed microsensor had a reproducible design.

The selectivity of the stochastic microsensors is given by the values of t_off (signatures) of the possible interferences. First of all, the three biomarkers: CRP, adiponectin and Zn²⁺ did not interfere each with the other, because they had different signatures (Table II). Other substances like dopamine, epinephrine, Mg²⁺, Ca²⁺ had different signatures than CRP, adiponectin, and Zn²⁺ proving that they did not interfere in their determinations (Table III).

The response characteristics of the proposed stochastic microsensors as well as their selectivity, made possible their utilization for the screening of serum samples for CRP, adiponectin, and Zn²⁺. The first step in the stochastic sensors'validation for the utilization in the determination of CRP, adiponectin and Zn²⁺ in serum samples was the recovery test, performed as following: to 5 serum samples, different amounts of CRP, adiponectin and Zn²⁺ were added. Measurements with the stochastic microsensors were performed before and after the addition of known amounts of CRP, adiponectin and Zn²⁺. The recovered amounts were compared with the added amounts, and the % recovery was determined (Table IV). As it can be seen in Table IV, very high recoveries were determined using the proposed stochastic microsensors, proving that the microsensors can be used realiably for the simultaneous assay of CRP, adiponectin and Zn²⁺ in serum samples.

The second step in the validation process was to determine the amounts of CRP, adiponectin and Zn²⁺ in ten samples of diabetic patients (Table V, Figure 1), and to compare the values obtained using standard methods (ELISA for CRP, and adiponectin, and ICP/MS for Zn) with those obtained using the stochastic microsensors. Table V shown high accuracy (very good correlations between the data were obtained) and reliability (very low values of RSD, % were recorded) of the measurements, proving that the proposed stochastic microsensors can be reliable used for the screening of serum samples in order to perform reliable qualiattive and quantitative analysis of CRP, adiponectin and Zn²⁺. The main advantages of the proposed method versus the methods proposed to date, are: the serum samples can be used without any further processing; the three biomarkers can be determined simultaneously – no need to use different kits, or expensive methods; the proposed method is very sensitive, accurate, and precise.