The bacterial culture was prepared in a Mueller-Hinton (MH) broth (Merck). MH medium containing: beef extract (8.8% w/w), casein hydrolysate (77% w/w), starch (6.6% w/w), and agar (7.5%w/w).

21 g of the MH medium was dissolved in 1 L of distilled water; the resulting mixture was sterilized in an autoclave for 15 min at 121°C.

Prior to electrochemical experiments, 100 ml of the sterilized mixture, and aliquot of 1 ml of the bacterial culture of P. aeruginosa (ATCC 15442) was added, at 37°C with shaking for 24 h.

The potentiodynamic measurements were carried out with an Autolab PGSTAT30 potentiostat/galvanostat. All materials, both liquid and solid, were sterilized before use. The electrolyte was autoclaved at 120°C for 20 min, while the electrodes and the electrochemical The cells were washed with ethanol (70% v/v) three times and exposed to ultraviolet light (254 nm, 20 W, Hg vapor lamp) for 15 min. A conventional three electrode electrochemical cell configuration was used for cyclic and square wave voltammetry measurements. A platinum counter electrode (platinum wire coil of 8 cm² area) was used, Ag/AgCl, (reference electrode 3.5 M) was the reference, and the working electrode was pyrolytic graphite with an area of 0.2 cm². The working electrode was modified with CoPc, which was adsorbed on the surface to sense hydrogen peroxide. The adsorption process was carried out at room temperature by placing an aliquot of 10^–4 M CoPc solution in dimethylformamide in contact with the electrode’s surface for 5 min. After surface modification, the working electrode was washed with ethanol to remove any excess of CoPc. Electrochemical measurements of OCP in sterilized (without bacteria) and inoculated media (previously sterilized) with bacteria P. aeruginosa.

After placing the corresponding electrodes on the sterilized media without bacteria and inoculated with the bacterium P. aeruginosa, the electrochemical cell was hermetically sealed during the whole experiment.

Bacterial numbers were determined by spectrophotometric measurements of absorbance at 600 nm and calculated as colony-forming units per milliliter (CFU/ml) by McFarland Method (0.5 OD / 5 × 10⁵ CFU/mL) (Baeza et al., 2013; Gulppi et al., 2019). The inoculated medium was used in exponential phase (OD 0.5) after 5 h of incubation at 37°C. All electrochemical measurements were repeated five times, and the measured values and plotted correspond to the averages.

Bacteria were first imaged on an aluminum surface. Atomic force microscopy (AFM) images of cells deposited on AA 2024 and AA 6063 aluminum alloys were obtained in the air using the Nanoscope III instrument (Digital Instruments) in the tapping mode, using J scanner. Olympus AC240TS silicon nitride cantilever, with a resonance frequency of 70–100 kHz, a radius of curvature of 20 nm, and stiffness of 5 N/m, was used at a scanning speed of 1 Hz. The cells were cultured in a Muller-Hinton broth in Falcon tubes containing alloy sheets previously washed with phosphate-buffered saline (PBS) at 37°C for 24 h. After the exposure, the aluminum alloy surfaces exhibited relatively well-developed biofilms. Before imaging, samples were rinsed with copious milli-Q water and air-dried in a clean environment maintained at 30 % relative humidity.

Silicon nitride cantilevers with a sphere of SiO₂ of 5 nm in diameter at one end (purchased from Novoscan PTGS) were immersed for 2 min in drops of 1% (v/v) Polyethylenimine solution (PEI). This procedure was performed to evaluate cell adhesion. After removing excess of the solution, the probes were dried with N₂ gas and stored at 4°C in a darkened non-frost- free refrigerator until use. For immobilizing bacteria on the colloidal probe, harvested as described above, pelleted cells were manually transferred to the PEI coated microsphere, using a micromanipulator (Narishige MHW-3), under a Nikon inverted microscope. A Nanoscope III AFM, operated in contact mode, was used to measure the adhesion force between the colloidal sphere previously coated with the bacteria and the two aluminum alloy surfaces coated with the bacteria. Scheme 1 schematizes the colloidal probe configuration.

Adhesion measurements were carried out preparing phosphate buffer solutions at different pH (5,5 to 7,5) and then placing the bacterial inoculum. The tensile force measurement was carried out by activating AFM with a scan frequency of 0.3 Hz and zero scan size. The ramp’s size was approximately 300 nm, and the loading force was applied by setting the maximum deflection value (maximum force on bacteria was approximately 10 nN). Care was taken to minimize lateral force and keep the vertical compression force small on approach to limit the cell probe’s damage. Several force curves were automatically acquired at different locations on the alloy surface by moving the cantilever laterally by an incremental step distance of 50 nm. In this way, sufficient statistical data were obtained.

Various sectors of the sample were measured to ensure representative statistics on the macroscopic surface. From the cantilever deflection curves, the maximum tensile force was obtained by multiplying the maximum cantilever deflection by the cantilever stiffness. The thermal fluctuation method was used to obtain cantilever stiffness.

The work of adhesion, W_adh, is evaluated as the total energy necessary to completely separate the bacterium body from the aluminum surface. Schema 1 schematizes how the work of adhesion is evaluated through the measurement of the force of adhesion, F, which is integrated over the distance, z, between the cell and the alloy surface. In the schema 1, the interval of distances [L₀, L_adh] accounts for the whole range of interaction, for which the cell is under tension, and F_adh is the maximum force of adhesion. Thus, in schema 1, the hatched zone represents the work of adhesion.

It should be noticed that this definition of work of adhesion is suitable in the present study since it considers that the main source of adhesion comes from the pilus stretching and its unzipping [ (Knight et al., 1998; Pratt and Kolter, 1998; Cavalcanti-Adam et al., 2006)]. In addition, it is worth mentioning that our definition differs from that provided by the theory of Johnson, Kendall, and Robert (Butt et al., 2005), which applies when the cell body deforms elastically due to the force of adhesion.

Figure 1 shows the open circuit potential (OCP)-time response in Mueller-Hinton (MH) culture broth in the absence and presence of Pseudomonas aeruginosa in the presence of a solution previously saturated with oxygen, purging the solution for 2 hours.

The OCP measurement in the microorganism’s presence started immediately after adding a 1 ml of the culture inoculum containing P. aeruginosa to 100 ml MH culture broth. The OCP decreases markedly in the first 2 h, because of the gradual decrease of oxygen concentration in the electrolyte. P. aeruginosa has five terminal oxidases for aerobic respiration (Comolli and Donohue, 2002; Comolli and Donohue, 2004). These terminal oxidases are expected to have their specific affinity for oxygen, since P. aeruginosa uses oxygen as an electron receptor in its metabolic processes. (Arai, 2011).

Furthermore, from Figure 1, it can be seen that after approximately 2 h, the potential value increases again sharply to a maximum of -150 mV, exhibiting OCP peculiar behavior (see the graphic inset in Figure 1) that is discussed below. After approximately 240 min, the OCP remained relatively stable during the rest of the measurement. The bacterial culture solution reached pseudo-thermodynamic equilibrium with the environment (open system) at a potential of approximately -173 mV (vs. Ag/Ag/Cl 3.5 M).

Figure 2 shows the OCP-time response in MH medium inoculated with P. aeruginosa at different pHs. At pH 6.5, the OCP-time response is similar to that at pH 7.0 shown in Figure 1; the potential falls sharply at the beginning, then increases steeply to a maximum to form a peak. This peculiar behavior, as previously mentioned, is associated with converting hydrogen peroxide to oxygen and is not observed at the other pHs, suggesting that catalase’s action is favored at neutral pH, in agreement with previous work (Gulppi et al., 2019). The interrelation between the generation of hydrogen peroxide by bacteria and the generation of oxygen resulting from catalase action has been demonstrated for two positive catalase microorganisms, Escherichia coli and Staphylococcus aureus (Gulppi et al., 2019). The experiment consisted of simultaneously sensing the OCP and the medium’s oxygen concentration, using a Clark probe, and following the procedure described by Zlayev et al. (Zlatev et al., 2013). Oxygen concentration and OCP in the culture media are closely associated with microbial culture growth (Zlatev et al., 2013; Gulppi et al., 2019), independent of the bacterial type strain.

Previous research (Baeza et al., 2013; Gulppi et al., 2019) by the authors has shown that enzyme activity can be determined in aqueous solutions in the presence of microorganisms, mainly bacteria. In these studies, the E / I response has been measured in voltammetric scans, using a working electrode made of pyrolytic graphite modified with surface-confined CoPc. During cathodic polarization, CoPc catalyzes the reduction of oxygen in two consecutive steps of 2 electrons each, according to the following reactions (written for acid solution): O 2 + 2 H + + 2 e - → H 2 O 2 (1) H 2 O 2 + 2 H + + 2 e - → 2 H 2 O (2) On the other hand, monofunctional catalases that contain a heme group (Fita and Rossmann, 1985) exhibit a common two-step mechanism for the degradation of H₂O₂. In a first step, a hydrogen peroxide molecule oxidizes heme (in catalases that contain heme) to an oxyferryl (involving the oxidation of Fe(III) to Fe(IV)). An oxidation equivalent of iron is removed, and another of the porphyrin ring, to generate the cationic porphyrin radical: (Eq. 3). Enz ( Por - F e III ) + H 2 O 2 → Compound   I ( Po r + - F e IV = O ) + H 2 O (3)

A second hydrogen peroxide molecule is utilized as a reductant of compound (I) to regenerate the resting-state enzyme, water, and oxygen (Eq. 4). Compound   I ( Po r + -F e IV = O ) + H 2 O 2 → Enz ( Por - F e III ) + H 2 O + O 2 (4) Despite this common reaction mechanism, enzymatic activities are very different among the various monofunctional catalase genes present in the bacterium P. aeruginosa. These are nominally differentiated as katA, katB, and katE. The main catalase katA is the most inducible against the damaging effect of hydrogen peroxide and is present in P. aeruginosa (Ochsner et al., 2000; Klotz and Loewen, 2003). We cannot refer to that type of catalase that releases the microorganism, but we can study the enzymatic activity associated with the enzyme released by P. aeruginosa. (Chelikani et al., 2004; Mishra and Imlay, 2012).

Figure 3A shows the cyclic voltammogram obtained on a bare pyrolytic graphite electrode bare (OPG) (black line) and modified with CoPc (red and blue lines) in the oxygen-free MH culture medium that was purged with nitrogen for 2 h. The red line corresponds to the cyclic voltammogram obtained when hydrogen peroxide is added in the absence of bacteria. A single peak is observed at -0.64 V, corresponding to the reduction of hydrogen peroxide, as shown in Equation 2. When hydrogen peroxide is added to the MH medium inoculated with bacteria and in the stationary phase of growth, the voltammogram (blue line) reveals a current peak at approximately -0.08 V. This is associated with the reduction of a significant concentration of molecular oxygen (Eq. 1) resulted from the enzymatic activity of catalase (McLeod et al., 1922; Hassett et al., 1997; Baeza et al., 2013; Gulppi et al., 2019). Thus, as previously mentioned, the catalytic activity is assessed by deliberately adding hydrogen peroxide to the culture in the stationary growth phase.

Further, the square wave voltammetry (SWV) curves in Figure 3B were obtained after adding 30 µL of H₂O₂ to the MH culture medium with bacteria under shaking (400 rpm) (blue line) and the bacteria-free broth (red line). In the blue curve, a current peak representing the reduction of oxygen is evident at about -0.08 V, whereas the current peak representing a reduction of hydrogen peroxide to water is smaller than that in the red curve, consistent with the cyclic voltammogram shown in Figure 3A. Comparing Figure 3A with Figure 3B, it is evident that the SWV is much more sensitive for detecting catalase enzyme activity.

Figure 4A shows the voltammograms obtained using a CoPc-modified OPG electrode in a Mueller Hinton medium in the presence of P. aeruginosa at different pHs and saturated with oxygen. From the cyclic voltammograms, it is observed that at pH 7.5 (black curve), the first cathodic current peak, which is associated with the reduction of oxygen to hydrogen peroxide, appears at a half-wave potential of 0.18 V, being more positive than those observed at more acidic pHs. On the other hand, from the SWV curves shown in Figure 4B, it is evident that at pH 7.5, the first cathodic current peak appears at more positive potentials (black curve). The behavior described above was very reproducible, suggesting that pH 7.5 favors the electrocatalytic process to reduce oxygen to hydrogen peroxide. At other pH values, the voltammograms are similar, and the differences are mainly in the intensities of the current peaks, the more acidic the environment, the lower the current peaks.

The shift of the first current peak to more positive potentials at ca. 175 mV, and related to the ORR, suggests that catalase activity at pH 7.5 is influenced by the heme group’s orientation and availability. The latter is the active center where hydrogen peroxide is reduced to water according to the oxidative stress described in Equation 3. pH 7.5 seems to be the optimum pH for electron transfer between catalase and hydrogen peroxide. The highest catalase activity at this pH was visibly observed by the abundant generation of oxygen at the electrode surface, which agrees with the O₂ generation reaction described in Equation 4. Oxygen can be reduced to hydrogen peroxide by reducing O₂ on the CoPc modified OPG electrode, according to Equation 1. This form of the autocatalytic mechanism was first studied by Busalman et al. (Busalmen et al., 2002). Considering this mechanism, the voltammetric response at more acidic pH revealed in Figure 4 is possibly the result of a metal center blocking process, or in other words, of the impediment for hydrogen peroxide to enter the metal center for its reduction to water (Equations 3, 4), which is possibly reflected in a reduction of enzyme (katA) activity (Hassett et al., 1997). This recycling mechanism with O₂ production promotes oxygen reduction to hydrogen peroxide, which was evident by cyclic voltammetry and square wave voltammetry, CV, and SWV.