Polyclonal antibodies (developed in rabbit) against Staphylococcus aureus were obtained from BIOtech RDP (Sfax, Tunisia). 3-Mercaptopropionic acid (MPA), phosphate buffered saline (PBS), N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), potassium ferrocyanide (K₄Fe(CN)₆), potassium ferricyanide (K₃Fe(CN)₆), ethanolamine, sulfuric acid, hydrogen peroxide (30%) were purchased from Sigma Aldrich, France. Ethanol was obtained from Fluka (purity >99%). All solutions were prepared with ultra-pure Milli-Q water.

In the present study, S. aureus reference strain (ATCC25923) was used. The bacteria strain was cultured in a tryptic soy broth (TSB, (Difco)) or on TSB agar plates for 24 h at 37 °C. A high titer of bacteria suspension was prepared as follows: liquid culture media was inoculated by 100 μL of the pre-culture solution and then cultivated at 37 °C for 18–24 h. The bacterial culture was centrifuged for 5 min at 6,400 rpm and was washed twice. Finally, the culture was resuspended in sterile PBS. The determination for the rate of viable cells and bacterial concentration was performed with the spread-plate technique. The optical density (OD) of the bacterial culture was measured for the determination of bacterial growth stationary phase. The cultures of S. aureus strain were grown to late log phase (OD₆₀₀ = 0.6). At the growth stationary phase, bacteria concentration does not change with time, and, subsequently, all the operations for the determination of the bacteria concentration can be performed with accuracy.

Electrochemical measurements were performed using a potentiostat-galvanostat Voltalab 40 with a standard three-electrode configuration. Platinum plate was used as the auxiliary electrode, saturated calomel electrode (SCE) as the reference electrode, and gold plate as the working electrode. Two electrochemical techniques were applied in this work:

- Cyclic voltammetry was performed in 8 mM ferro/ferricyanide PBS solution at a scan rate of 100 mV/s.

- EIS is based upon inducing a disturbance in an electrochemical reaction from its steady state by applying a small excitation sinusoidal signal to the system (amplitude 10 mV; frequency range 100 mHz to 100 kHz).

- The SAMs were characterized through CAM, using a GBX Scientific Instruments (Romans-France). All measurements were made by depositing 1 µL of Milli-Q water at 24 ± 2 °C. Multiple values were recorded for each analysis. The precision of the micro-regulator was 0.33 µL for a 1 µL drop. The results were processed using Windrop++ software.

The Au electrodes were ultrasonically cleaned in acetone for 10 min, followed by 5 min in a mixture of piranha solution (H₂O₂/H₂SO₄, 3/7, v/v). The electrodes were thoroughly rinsed with ultrapure water, followed by absolute ethanol, and then dried in a flow of nitrogen. Subsequently, the electrodes were incubated in a MPA (10 mM) solution made-up with ethanol for 12 h to formulate the SAMs. After the formation, the electrodes were rinsed in ethanol. Then, the terminal carboxylic acid (–COOH) groups were activated in a solution of NHS/EDC (0.1 M/0.1 M) for 1 h at room temperature. After rinsing with PBS, the electrodes were incubated for 1 h in a 0.3 mg/mL solution of anti-S. aureus antibodies. The terminal amine groups on the antibody enable covalent bonding to occur through the activated carboxylic functions from MPA functionalized NHS/EDC. Finally, after rinsing with PBS, the electrodes were incubated for 20 min in ethanolamine. This blocks the remaining acidic functionalities. The mechanism for this activation can be seen in Figure 1.

We studied the sensor response towards different concentrations of the S. aureus bacteria. For this reason, the electrodes were incubated in different aliquots corresponding to different bacterial concentrations in CFU/mL.

After thorough cleansing and immersion into MPA ethanolic solution, the gold electrode was characterized by water contact angles. The results are shown in Figure 2. Here, we observe that the contact angle decreases from 48° to 7° after SAMs deposition. This value is in agreement with results already published [15]. In this instance, when the carboxylic acid-terminated MPA monolayer is previously rinsed in ethanol, then, the observed contact angle is found to be <10°.

For electrochemical measurements, the CV was recorded before and after the deposition of the MPA monolayer. In Figure 3, the Fe(II)/Fe(III) redox peaks have decreased after the functionalization of the MPA monolayer. This can be attributed to the decrease of the electron transfer rate that was created by the compactness of the formulated SAMs. After antibody binding, the redox peaks decrease even more due to the decrease of the electron transfer rate. This was due to an increase of the biolayer thickness that was developed on the gold surface.

The Faradaic impedance measurements are in good agreement with CV measurements as the diameter of semicircles for the Nyquist diagrams increase significantly after the different stages for the elaboration of the immunosensor (Figure 4).

The impedance measurement for the direct detection of bacterial cells is generally known to have improved significance by analysis with a Bode plot. This technique is suitable for studying the direct relationship of impedance with the frequency [16]. Here, the binding ability of bacteria and the bioreceptor was then verified through the EIS technique. In Figure 5, we show (A) the Bode plot, and (B) the Nyquist diagram that were obtained after incubation in an increasing concentration of the target bacteria. Here, the charge transfer resistance (R_CT) increases gradually as the bacteria concentration increases after consecutive incubations. This was made from 10 CFU/mL to 10⁷ CFU/mL.

In order to express the characterization of modified electrode/electrolyte interface after the immobilization of biomolecules and bacteria binding, EIS is often analyzed using an equivalent circuit. The equivalent circuit is applied to fit the experimental data and extract the necessary information about the electrical parameters responsible for the impedimetric change. The impedance spectra obtained in the presence of different concentrations of bacteria ranging from 10 to 10⁷ CFU/mL were modeled by the equivalent electrical circuit shown in Figure 6(A). The circuit is a combination of:

Series resistance of solution and of ohmic contacts (Rs).

The impedance immunosensor shows a linear relationship between the R_CT variation and the logarithmic value of S. aureus concentrations. These were found within the range of 10 to 10⁶ CFU/mL with a correlation coefficient of 0.994. The variation of R_CT was calculated against the value of R_CT in the absence of bacteria (Figure 6(B)). The reproducibility of the immunosensor was validated by repeating the experiment with four different immunosensors prepared in the same conditions. The results showed that the response of the immunosensor has an average standard deviation of 8%. This verifies that a good reproducibility of the system was achievable for the detection of S. aureus.

A comparative study was made to ensure that the response of the immunosensor was a direct cause of the immobilized antibody and not due to adsorption of bacteria on the surface modified SAMs. Here, different electrodes were functionalized in the same conditions as our immunosensor, but without the antibody incubation step. The results show that there is not a significant variation of the response for the modified electrode (relative variation of R_CT) after incubation of the electrode in varying staphylococcus concentrations (Figure 7). The fixation of the antibody is a logical approach for the high detection of these pathogenic bacteria.

The characteristics of relevant and previous immunosensors for S. aureus detection in terms of the limit of detection (LOD) and of the dynamic range are given in Table 1. In this work, the LOD obtained for the developed immunosensor is of the lower value and the dynamic range is of the larger value.

E. coli and Staphylococcus epidermidis (S. epidermidis) were detected in order to determine the specificity of our impedimetric immunosensor. These two bacteria are the most abundant strains found in real samples. The same protocol was followed for the surface functionalization using anti-S. aureus. Here, the antibody was immobilized and the electrodes were exposed to the suspensions of E. coli and S. epidermidis. In a similar manner to the detection of S. aureus, the concentrations of the two bacteria were increased in order to observe if there was a significant effect on the immunosensor by EIS. In Figure 8, the positive-to-negative response ratios of the immunosensor for non-specific targets (E. coli and S. epidermidis) are given. The results show a higher selectivity for E. coli compared to S. epidermidis.