Disposable Stochastic Sensor Based on Deposition of a Nanolayer of Silver on Silk for Molecular Recognition of Specific Biomarkers

All chemicals were of analytical grade. Carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9), tumor suppressor p53, α-cyclodextrin (α-CD) were purchased from Sigma Aldrich, and paraffin oil (d₄, ²⁰ 0.86 g × cm⁻¹) was purchased from Fluka. The CEA, CA19-9 and p53 solutions were prepared using the serial dilution method and buffered using the phosphate buffer solution (PBS, pH = 7.5).

The electrochemical cell comprised three electrodes: the reference electrode - Ag/AgCl nanolayers deposited on silk, the auxiliary electrode which is a platinum nanolayer deposited on silk and the working electrode, represented by the disposable stochastic sensor designed in this paper (Scheme 1). Stochastic measurements were performed using an AUTOLAB/PGSTAT 302 N (Metrohm) connected to a computer with GPES software, to record and interpret the measurements.

Zoom In Zoom Out Reset image size

Thermionic vacuum arc method, a cold plasma system, was used for the application onto commercially available silk textile of Silver nanofilm nanocoatings. The experimental setup of the plasma source was displayed in Fig. 1.

Zoom In Zoom Out Reset image size

The ignition principle of this plasma source was based on obtaining silver vapors and ionizing them using accelerated electrons emitted from a hot cathode. The electrons were progressively accelerated by increasing the voltage at the anode (tungsten crucible) containing the silver bulk material (99.9% purity), up to a point where a stable silver plasma was obtained. The method seems to be suitable for deposition these coatings onto temperatures sensitive organic substrates materials. This method made possible to produce nanoarchitectured films coatings at very low temperature (room temperature), under high vacuum of about 10⁻⁶ Torr, high degree of uniformity and no buffer gas was needed.

A piece of synthetic silk with the size of 210 × 297 mm was used as substrate. Before the deposition, the textile substrate was washed with deionized water and dried under nitrogen flow followed by thermal treatment in vacuum chamber for 2 h to increase the surface functionalities.

For Ag deposition, the silk textile substrate was introduced in the deposition chamber which was later pumped down to ∼3.0 × 10⁻⁶ mBar by a turbo-molecular pump. The electrical parameters used for the silver plasma were: 54 A filament current, 1.6 A plasma current and 250 V plasma voltage. The silk substrate was placed face down at 27 cm distance from the plasma plume to obtain the optimum plasma coatings. The film thickness was controlled by the deposition time and it lasted for 7 min. Digital photos of coated and uncoated silver nanofilm onto synthetic silk were shown in Fig. 2.

Zoom In Zoom Out Reset image size

Synthetic silk used as a substrate for Ag nanofilm coatings in this study is illustrated by optical microscope images with different magnifications, showing differences in surface texture (Fig. 3).

Zoom In Zoom Out Reset image size

Scanning electron microscopy was performed in order to characterize the surface morphologies of uncoated and coated synthetic silk. SEM images of 250x and 63x magnification of uncoated synthetic silk are shown in Figusre 4a and 4b while in Figs. 4c and 4d, 500x and 56x magnification SEM images of silver coated synthetic silk were presented. The images shown that the synthetic silk uncoated was smooth, clean, and silver nanofilm coated surface displaied a uniform and smooth coverage.

Zoom In Zoom Out Reset image size

On the same piece of silk, left and right were deposited nanolayers of Ag/AgCl (to serve as reference electrode in the cell), and of Platinum (to serve as auxilliary electrode) (Scheme 1).

The silk coated with Silver nanofilm part was soaked in a solution of 10⁻³ mol l⁻¹ α-cyclodextrin for 1 h. After 1 h, the silk was dried at room temperature. The width of the working electrode was 1 mm. By its shape, the α-cyclodextrin can ensure the necessary channels for stochastic sensing.

The stochastic mode is based on the channel conductivity. The signatures of the three analites (t_off values) were used for their molecular recognition (by identification of their signatures in the diagrams, Scheme 2, Fig. 5) while the t_on corresponding values found in the diagrams recorded for the

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

biological samples (Fig. 5) were used for the quantitative determination of the analites accordingly with the equation: ^1–4

$$\begin{eqnarray*}&&1/{{\boldsymbol{t}}}_{{\boldsymbol{on}}}={\boldsymbol{a}}+{\boldsymbol{b}}\times {\boldsymbol{Conc}}{.}_{{\boldsymbol{analyte}}}\end{eqnarray*}$$

All measurements were done using the chronoamperometric technique, at 125 mV vs Ag/AgCl, at 25 °C. The equations of calibrations were obtained, after screening for each biomarker solutions containing different concentrations of biomarker; linear regression method was used for determination of the equation of calibration.

Whole blood and tissue samples (Ethics committee approval number 65573/14.12.2018 awarded by the University Emergency Hospital from Bucharest) were collected from patients with confirmed glioblastoma. The samples were analyzed as collected from patients, without any pretreatment. Informed consent was obtained from all patients. A drop of whole blood samples was placed on the disposable sensor and measured; the tissue sample was placed on top of the disposable sensor, and measured.

The response of the disposable stochastic sensors was based on channel conductivity. ^1–4 The sensing mechanism takes place in two steps: (1) molecular recognition step (qualitative analysis takes place)—since entering the channel, the analyte blocks the channel and the current drops to zero for a certain time (until the entire molecule enter the channel) called the signature of the analyte (found in the diagrams as t_off); (2) analysis step (quantitative analysis takes place)—the analyte binds the internal channel wall and undergo redox process for a certain time–found in the diagrams as t_on, which is correlated with the concentration of the analyte (Scheme 1, Fig. 5). The biomarkers are going one by one in the channel (Scheme 2, Fig. 5).

Different signatures (t_off values) recorded for the biomarkers (Table I) allow their simultaneous molecular recognition. The disposable stochastic sensor based on silk material gave high values of sensitivity and low limits of determination for the assay of CA19-9, CEA and p53. The linear concentration ranges covered both healthy and GBM confirmed patients. Different strips were used for measurements—the sensitivity recorded varying with less than 1.00%; this result proved the reproducibility of the design of the disposable stochastic sensor.

Comparing the proposed disposable sensor with 3D sensors based on graphene materials ²⁰ reported previously, one should conclude that: the sensitivity of the proposed disposable sensor is higher than of those based on graphene materials; the sensor is also cost-effective—its production cost is ten times lower than the cost of any of the graphene-based 3D sensors; no contamination from previous samples can be claimed, because it is used just for measurement of one sample.

The selectivity of the sensor is given by the signatures (t_off values) of different biomarkers—recorded on the molecular recognition step. The signature of the analyte and possible interferent depends on many factors such as: unfolding capacity (if there are 2D/3D structures), size of the molecule, conformation of the molecule, velocity of going inside the channel; therefore, the signature can act as molecular recognition element, contributing to qualitative analysis of mixtures. Different signatures obtained for biomarkers like CEA19-9, p53, CA19-9, KRAS, maspin, D-aspartic acid, D-glutamine proved the selectivity of the proposed disposable stochastic sensor (Table II). The t_off values recorded in Table II were determined in the synthetic mixture containing all biomarkers; despite the concentration of the biomarkers, the t_off values did not change.

Whole blood samples, and tissue samples were obtained from GBM confirmed patients. Examples of diagrams obtained for the screening of whole blood and tissue samples were shown in Fig. 5. The molecular recognition of p53, CEA, and CA19-9 was performed first using the signature of these biomarkers using the disposable stochastic sensor (Table I). The t_on values determined from diagrams were used for the quantitative determination of these biomarkers in the biological samples, accordingly with the procedure described above in the stochastic mode. The results obtained for the assay of the three biomarkers using the disposable stochastic sensor and the results obtained using the ELISA—the standard method (Table III) shown a high correlation between the results obtained the proposed method and the standard method.

Accordingly, the new material synthesised for the design of the disposable stochastic sensor can be reliable used for molecular recognition and also for the reliable determination of biomarkers such as CEA, p53, and CA19-9 in biological samples such as whole blood and tissue samples.