Nanoelectrodes for determination of reactive oxygen and nitrogen species inside murine macrophages

Abstract

Reactive oxygen and nitrogen species (ROS and RNS) produced by macrophages are essential for protecting a human body against bacteria and viruses. Micrometer-sized electrodes coated with Pt black have previously been used for selective and sensitive detection of ROS and RNS in biological systems. To determine ROS and RNS inside macrophages, one needs smaller (i.e., nanometer-sized) sensors. In this article, the methodologies have been extended to the fabrication and characterization of Pt/Pt black nanoelectrodes. Electrodes with the metal surface flush with glass insulator, most suitable for quantitative voltammetric experiments, were fabricated by electrodeposition of Pt black inside an etched nanocavity under the atomic force microscope control. Despite a nanometer-scale radius, the true surface area of Pt electrodes was sufficiently large to yield stable and reproducible responses to ROS and RNS in vitro. The prepared nanoprobes were used to penetrate cells and detect ROS and RNS inside macrophages. Weak and very short leaks of ROS/RNS from the vacuoles into the cytoplasm were detected, which a macrophage is equipped to clean within a couple of seconds, while higher intensity oxidative bursts due to the emptying of vacuoles outside persist on the time scale of tens of seconds.

Macrophage cells are essential for the performance of the immune system. Their activation, either under normal biological conditions or by specific biochemical activators in vitro, results in the production of reactive oxygen and nitrogen species (ROS and RNS) and creation of a large number of vacuoles (phagosomes and phagolysosomes; see Fig. 1A and SI Appendix) (1–3). These vacuoles play an important role in phagocytosis—a mechanism used by the immune system to remove pathogens and cell debris. A cell (or debris) is engulfed into a vacuole and subjected to an intense oxidative burst (2), and the indigestible debris and excess ROS and RNS are subsequently evacuated from the macrophage (Fig. 1B).

Fig. 1.

Optical micrograph of a macrophage RAW 264.7 activated by interferon-γ and LPS evidencing the presence of phagolysosomes (white spots) (A) and schematic representation of an activated macrophage undergoing phagocytosis (B). I, capture of a cell or debris; II, internalization within phagosome; III, digestion by ROS/RNS within phagolysosomes; IV, expulsion of indigestible material and excess of ROS/RNS. Also shown in B are two configurations used for the detection of ROS/RNS by either a platinized nanoelectrode inside or a microelectrode outside the cell, as discussed in the text.

The changes in oxygen and hydrogen peroxide concentrations during the oxidative burst of a stimulated macrophage cell were detected previously using the scanning electrochemical microscope (4). Extensive studies with amperometric microelectrodes positioned in the cell proximity showed that the basal release is due to a cocktail composed of several ROS and RNS evolving from the primary production of and NO (5–8). However, the concept that ROS and RNS released inside phagolysosomes may diffuse across the vacuole membrane and leak in the cell cytoplasm remains controversial (9–12). In fact, NO and the trans-isomer of protonated peroxynitrite ion are capable of crossing biological membranes due to their lipophilicity (13, 14). This underscores the importance of probing for the intracellular presence of ROS and RNS in activated macrophages.

For electrochemical measurements inside an activated macrophage one needs nanometer-sized electrodes that can be inserted into a living cell without causing irreparable damage to its membrane. Also, the cell membrane must seal around the nanoelectrode to preserve as much as possible the integrity of intracellular mechanisms and prevent any direct liquid junction between extra- and intracellular compartments. Previously, quantitative electrochemical experiments were performed by inserting glass-sealed, polished Pt nanoelectrodes into cultured human breast cells (15). The cell membrane formed a tight seal around the penetrating nanotip that prevented the external solution from leaking inside the cell. This allowed the cell to remain alive for the entire time of experiment (> 10 min) with a nanoelectrode inside it. However, a polished Pt nanoelectrode is not suitable for the detection of ROS and RNS, which passivate its small surface and diminish the signal. Even micrometer-sized probes used for extracellular measurements of ROS and RNS had to be coated with Pt black to improve the stability of the response (16). Several approaches to fabricating micrometer and submicrometer-sized Pt electrodes with high surface area are available in the literature (16–19). For quantitative intracellular measurements, platinized probes have to be smaller (< 1 μm total diameter, including glass sheath), with porous Pt surface flush with the surrounding insulator.

The methodology for fabricating nanoelectrodes by electrodeposition of metals was reported recently (20). The electrodes were produced by electrodepositing metal into a nanocavity, which was formed by etching away a nm-thick layer of Pt from the glass-sealed, polished Pt nanoelectrode (21). The amount of deposited metal was controlled by monitoring the charge, and its excess was removed by polishing to yield a flat electrode. This strategy cannot be used to prepare platinized nanoelectrodes because the current efficiency in deposition of Pt black is relatively low and polishing was shown to diminish the electrode response. To overcome these problems, we developed unique methodology for fabricating platinized nanoelectrodes under atomic force microscope (AFM) control.

Results

Fabrication of the Platinized Nanoelectrodes.

An early attempt to produce a platinized nanoelectrode by electrodeposition is shown in the SI Appendix. The black curve in SI Appendix, Fig. S1A is a voltammogram of ferrocenemethanol (FcMeOH) at a polished Pt electrode with the radius, a = 60 nm, calculated from the diffusion limiting current. After etching, the diffusion current decreased to ∼50% of the original value (red curve in SI Appendix, Fig. S1A), which corresponds to the formation of a ∼40-nm-deep cavity, according to the available theory (21). After the platinization, the FcMeOH current increased almost ten times (green curve), indicating that the nanopore was significantly overfilled with Pt black. Most excess Pt black was removed by polishing, after which the limiting current (blue curve in SI Appendix, Fig. S1A) was only slightly higher than that obtained at the original polished electrode (black curve) in accordance with an SEM image (SI Appendix, Fig. S1B) showing the electrode radius of ∼80 nm.

Although seemingly successful, the above example shows major difficulties in platinization of nanoelectrodes. With no efficient control, large excess of Pt black was deposited. Although it could be removed, as shown in SI Appendix, Fig. S1, polished platinized electrodes typically exhibit very low responses to ROS/RNS. Finally, SEM imaging is not a convenient technique for monitoring the fabrication. It cannot be done in situ; has low z-axis resolution, which does not allow one to see whether the Pt electrode is recessed, flat, or protruding; and the imaged electrodes are no longer suitable for electrochemical experiments.

An alternative approach—electrodeposition of Pt black under the AFM control—is illustrated in Fig. 2. A noncontact topographic image of an etched Pt electrode (Fig. 2A) in solution before the platinization shows the effective radius, a ≈ 70 nm and the cavity depth of ≥15 nm (the triangular shape of the cross-section in Fig. 2A suggests that the tip did not reach the bottom of the cavity, and the actual depth could be larger). The deposition of Pt black was done by stepping the electrode potential to -100 mV versus Ag/AgCl, while the AFM tip, immersed in the platinization solution, was scanned in x direction above the electrode surface. Fig. 2B shows a stack of 60 consecutive topographic 1D scans obtained over a 60 s period. Initially, the deposition process was slow, and its rate increased with time, as the cavity depth decreased. The deposition was stopped by stepping the electrode potential to 0 mV after Pt black completely filled the cavity and slightly protruded (by ∼15 nm) from the glass sheath, as can be seen from the image of the same electrode obtained after the platinization (Fig. 2C). In another deposition experiment (SI Appendix, Fig. S2), the initial depth of the nanocavity was only ∼2 nm and the protrusion height after the deposition of Pt black was ∼3 nm.

Fig. 2.

Noncontact topographic images of an etched Pt nanoelectrode in solution before (A) and after (C) the deposition of Pt black, and time evolution of a line scan during the electrodeposition process (B). The red line in A corresponds to the shown cross-section. (B) The tip was scanned along the x axis with the scan rate of 1 Hz. The position of the line scan approximately corresponded to the red line in A.

The above methodology for fabricating Pt black nanoelectrodes, although powerful and reliable, is laborious and requires AFM instrumentation. A simpler approach to platinization makes use of the characteristic shape of the current transient. As noted above, the rate of the deposition process increases greatly when the cavity gets completely filled with metal. The corresponding sharp increase in current can be used to detect the completion of the platinization (Fig. 3). The etched electrode (Fig. 3A) was imaged in air, and then a current transient (Fig. 3B) was obtained in the platinization solution during the Pt black deposition into its cavity. The sharp increase in the slope of the current-time curve indicated that the nanocavity was filled, as can be seen in the image of the platinized electrode (Fig. 3C).

Fig. 3.

A noncontact AFM image of a 115-nm-radius etched electrode in air (A), a current transient of the Pt black deposition (B), and a topographic image of the same electrode after platinization (C).

In Vitro Detection of ROS/RNS.

The capacity of platinized nanoelectrodes for detecting in vivo the four typical electroactive ROS/RNS released by macrophages during oxidative bursts has been evaluated in vitro using aerated phosphate buffered saline (PBS) solutions of hydrogen peroxide (H₂O₂), peroxynitrite anion (ONOO^-), nitric oxide (NO), and nitrite anion () (Fig. 4A). and H₂O₂ species are stable at biological pH (7.4). Their voltammograms (red and black curves in Fig. 4A) are qualitatively similar to those recorded previously with micrometer-sized electrodes (6, 7), and the calibration curves obtained from the families of such voltammograms (Fig. 4B) are linear (Fig. 4 C and D). Conversely, ONOO^- is not stable at pH 7.4, and its voltammogram (blue curve in Fig. 4A) was obtained at pH 10 (22). Similarly, due to its rapid reaction with O₂, NO was generated in situ by controlled decomposition of DEANONOate (see experimental section) (23). The instability of these species was not, however, an issue for in vivo experiments where ROS/RNS are produced and detected on a much shorter experimental time scale (seconds) (6–8).

Fig. 4.

In vitro voltammetry of ROS/RNS species in aerated PBS. (A) Normalized voltammograms of oxidation of ONOO^- (blue curve; 10 mM, pH 10, see text; the foot of the wave is merged with that of the reduction of dioxygen), NO (green, ∼1 mM, see text), H₂O₂ (black; 1 mM, pH 7.4), and (red; 5 mM, pH 7.4). Voltammograms were recorded at different platinized nanoelectrodes with the average radius of 60 nm and normalized by their plateau currents. (B) Steady-state voltammograms of , 10 (blue), 5 (green), 1 (red), and 0.5 mM (black) at a platinized nanoelectrode. a = 40 nm. The scan rate was 20 mV/s. Calibration curves for (C) and H₂O₂ (D) obtained from diffusion limiting currents of steady-state voltammograms.

The importance of platinization is illustrated by SI Appendix, Fig. S3B, showing the voltammograms of oxidation at the same 65 nm Pt electrode before etching (black curve) and after the deposition of Pt black (red curve). Clearly, only platinized nanoelectrodes exhibit analytically useful response to ROS/RNS. The similarity of voltammograms obtained for FcMeOH oxidation (SI Appendix, Fig. S1A) and reduction (SI Appendix, Fig. S3A) at the same electrode before etching and after platinization indicates that the greatly enhanced response to after platinization is not due to the larger electrode radius or protrusion, but to electrocatalytic effects.

In Vivo Detection of ROS/RNS with Platinized Nanoelectrodes.

The next step involved the characterization of the global release of the above four ROS/RNS inside activated murine macrophages (RAW 264.7 line). In these experiments, we took advantage of the previous observation that the penetration of the cell membrane by a submicrometer tip activates it and induces fast oxidative burst release (6, 7). Hence, a platinized nanoelectrode was inserted inside a macrophage (Fig. 5A) with the purpose of eliciting its response. The intensities of oxidative bursts elicited by the nanoelectrode insertion were monitored outside the macrophage cells with a classical 10 μm platinized fiber electrode (polarized at 850 mV vs. Ag/AgCl) following a previously reported protocol (Fig. 5B; see also SI Appendix) (6, 7). The inserted nanoelectrode was also polarized at 850 mV vs. Ag/AgCl and used for monitoring the amount ROS/RNS released intracellularly (Fig. 5C).

Fig. 5.

Monitoring ROS/RNS release induced by the mechanical stimulation of a macrophage. (A) Optical microscopic micrograph of a nanoelectrode (a = 75 nm; 800 nm O.D.) inside a macrophage. (B) Amperometric detection of ROS/RNS outside the macrophage at the 10 μm platinized carbon fiber electrode polarized at 850 mV vs. Ag/AgCl; the oxidative burst was elicited by the penetration of a platinized nanoelectrode through the macrophage membrane. (C) Two typical amperometric current traces of the ROS/RNS release inside a macrophage induced by the insertion of a platinized nanoelectrode (Left, a = 50 nm, 700 nm O.D.; Right, a = 40 nm, 900 nm O.D.); E = 850 mV vs. Ag/AgCl.

The comparison of Fig. 5 B and C shows that the responses monitored outside and inside are completely different, the outside response being more intense and lasting longer time (Table 1). It was then essential for us to ensure that the cell membrane formed a tight seal around the nanoelectrode shaft to eliminate a possibility that the response monitored inside the cell resulted from traces of ROS/RNS released outside and leaking into the cell. For this purpose, a series of experiments were performed with macrophages bathed in PBS containing 1 mM Ru(NH₃)₆Cl₃ (Fig. 6). The complete absence of the reduction wave inside the cell, which was observed in ∼20% of these experiments, provided evidence for a perfect seal (15). The similarity of voltammograms obtained before (black curve in Fig. 6) and after (green curve) cell penetration indicates that the nanoelectrode capacity to respond to this species has not significantly diminished during the entire experiment. The typical current transients of ROS/RNS shown in Fig. 5C were recorded only in cells that have exhibited no solution leakage through the membrane/glass seal.

Table 1.

Average parameters of oxidative bursts produced by RAW 264.7 macrophages detected inside or outside the cells by different platinized electrodes

Detection	Electrode	Imax, pA	t1/2, s	Q, pC
Intracellular (n = 10)	platinized Pt nanoelectrode	11.7 ± 6.4	0.5 ± 0.1	7.7 ± 3.5
Extracellular (n = 8)	10 μm platinized carbon fiber	86 ± 23	5.8 ± 1.8	703 ± 200

Macrophages were stimulated mechanically by insertion of a platinized nanoelectrode (60-nm average radius). The potential of the detecting electrode was 850 mV vs. Ag/AgCl. All values are reported as a mean ± SEM; n, indicated in parentheses, is the number of experiments of each type (each one involving a different cell and a different electrode).

Fig. 6.

Voltammetric reduction of in solution and inside a macrophage. Voltammograms were obtained before cell penetration (black), inside the cell (red) and after the removal (green) of a platinized nanoelectrode (a = 60 nm) from the cell in a solution of PBS containing 1 mM Ru(NH₃)₆Cl₃. The scan rate was 20 mV/s. The inset shows a current-time trace corresponding to the nanoelectrode insertion into the cell and its subsequent retraction to the external solution; E = -400 mV vs. Ag/AgCl.

Discussion

Previously developed methodologies for nanoelectrode fabrication are not suitable for preparing platinized probes. Our approach based on the in situ AFM control of Pt black deposition was used to produce well-shaped platinized nanoprobes (Fig. 2 and SI Appendix, Fig. S2), which exhibited good amperometric responses to ROS and RNS. AFM images also showed the diameter of the insulating glass sheath, which determines the physical size of the probe essential for cell penetration. The response of platinized nanoprobes to ROS and RNS decreases on the time scale of several days; thus it is desirable to prepare new platinized electrodes right before in vivo experiments. To address this need (and also simplify the fabrication), one can detect the end point of Pt black deposition from the current transient instead of using AFM control. Noncontact AFM imaging of the nanoelectrode in air before and after platinization can be used to evaluate its geometry and applicability to intracellular measurements (Fig. 3). The latter method is faster and easier than AFM-controlled electrodeposition, but the former approach is more accurate and reliable.

The in vitro experiments with ROS/RNS evidenced that the platinized nanoelectrodes yield characteristic steady-state voltammograms for each of the four species composing oxidative bursts. The sensitivity of nanoelectrodes for ROS and RNS (e.g., ∼0.2 pA·mM^-1·nm^-1 for ; a = 60 nm, Fig. 4C) was about half that predicted theoretically or monitored for reduction and FcMeOH oxidation (SI Appendix, Figs. S1 and S3A). The same applies when comparing to the initial Pt disk nanoelectrodes (SI Appendix, Fig. S1A). These relative figures indicate that Pt black nanodeposits are somewhat less active than expected from their geometrical dimension presumably because their dendritic growth was constrained by nanometer dimensions of the underlying Pt surface (see discussion in SI Appendix). The difference in sensitivity was not critical for our purpose because these electrodes were not used to measure ROS/RNS concentrations, but to collect their fluxes produced either inside (nanoelectrodes) or outside (microelectrodes) stimulated macrophages. The important parameters here are the time durations (t_1/2) and the integral of the current-time responses [i.e., charge, Q, which reflects the quantity released (6–8)] recorded at the platinized electrode (Table 1).

In the absence of solution leakage through the membrane/glass seal, the nanoelectrode insertion into an isolated macrophage elicited a brief current spike and a small amount of charge showing that the collected ROS/RNS flux was of much lower intensity and shorter duration than that detected outside the stimulated cell at the same electrode potential (cf. Fig. 5 B and C and Table 1). The height of the current peak indicated that the cumulative ROS/RNS concentration released inside the cell peaked at submillimolar values before decaying rapidly.

Because all presented data was obtained with proper sealing of the cell membrane around the nanoelectrode shaft, it is clear that ROS/RNS detected inside the cell represent only species released inside the cell. Moreover, would there be unnoticed leakage from the outside, the kinetic features of the nanoelectrode responses would mimic those recorded outside the cell. On the contrary, two distinct components of the oxidative burst were produced by a mechanically activated macrophage: one small and brief (short t_1/2 and small Q; Fig. 5C and Table 1) inside the cell and another one, much longer and with a larger amount of charge-outside (Fig. 5B and Table 1). This observation is consistent with the view that in response to stimulation a macrophage produces small vacuoles whose membranes are equipped with NADPH-oxidase and NO-synthase enzymes that generate and NO, respectively, inside the vacuole (Fig. 1) (1, 12, 22). These two primary species react individually or cross-react to generate the oxidative burst cocktail composed of H₂O₂, , NO, and , which are detected outside the cell cytoplasmic membrane when the vacuoles fuse with it (6–8, 22). Among these species, NO and ONOOH can rapidly diffuse across biological membranes (13, 14) into the cell cytoplasm (hence giving rise also to , the product of peroxynitrite decomposition). In agreement with previous qualitative observations,(9–12) our experiments suggest that the transient oxidative bursts detected inside macrophages represent the leakage of these species from vacuoles before they can release their load outside the cell (Fig. 1). The short duration of oxidative bursts monitored inside an activated macrophage, as compared to the responses recorded outside (Table 1), indicates that this cell can rapidly eliminate ROS/RNS, thus avoiding damage to its inside.

In summary, the presented data supports the hypothesis of the ROS/RNS leakage from phagolysosomes. It also shows that a macrophage can avoid oxidative damage by rapidly reducing ROS/RNS concentration levels in its cytoplasm. One should notice that in our experiments the oxidative stress response was induced by mechanical stimulation of a macrophage in its rest state. However, if the efficiency of the ROS/RNS removal in activated macrophages performing phagocytosis is similar to that observed in this study, no cell damage should occur due to the leakage of ROS/RNS outside phagocytotic vacuoles.

Materials and Methods

Chemicals.

All aqueous solutions were prepared from deionized water (Milli-Q, Millipore Corp.). Phosphate buffered saline (PBS; pH 7.4; 0.137 M NaCl, 0.01 M Na₂HPO₄, and 0.003 M KCl) was prepared by dissolving tablets (Sigma) in water and used in experiments with ROS/RNS. Ferrocenemethanol from Aldrich was recrystallized twice from acetone. Platinization solution contained 0.087 g hexachloroplatinic acid (Aldrich) and 0.0014 g lead(II) acetate trihydrate (Alfa Aesar) in 1 mL of water to which 36 mL PBS was added. Etching solution was prepared by mixing 60% (by volume) water, 30% 5 M CaCl₂, and 10% HCl. DEANONOate and sodium peroxynitrite in alkaline solution were purchased from Cayman Chemical.

Cell Culture.

The murine macrophage RAW 264.7 (American Type Culture Collection) cell line was cultured at 37 °C under a 5% CO₂ atmosphere in Dulbecco’s modified Eagle’s medium (DMEM) containing 1.0 g L^-1 D-glucose and sodium pyruvate (Invitrogen). The medium was supplemented with 5% fetal bovine serum (Invitrogen) and 20 μg mL^-1 gentamicin (Sigma). Confluent monolayers of RAW 264.7 cells were resuspended through trypsinisation and plated in tissue culture Petri dishes (Nunc; 35-mm diameter) 24 h prior to electrochemical studies.

Preparation of Etched Pt Nanoelectrodes.

Disk-type, flat nanoelectrodes were prepared by pulling 25-μm-diameter annealed Pt wires into borosilicate glass capillaries with the help of a P-2000 laser pipette puller (Sutter Instrument Co.) and polished under video microscopic control as described previously (24). The RG (i.e., the ratio of glass radius to that of the Pt tip) varied from 5 to 10. The electrode radius was evaluated from steady-state voltammetry. The electrodes that exhibited good quality voltammetric response were etched with an alternating current of 1.5 V amplitude, 20 MHz frequency (Keithley 3940, multifunctional synthesizer), as described previously (21). The etched electrode was cleaned by sonication in water during 5 s and imaged by the AFM to determine its radius and the recess depth. The fabrication of 10 μm platinized carbon fiber microelectrodes is described in SI Appendix.

AFM Imaging and Deposition Control.

An XE-120 scanning probe microscope (Park Systems) was employed for imaging nanoelectrodes and for in situ control of Pt black deposition. PPP-NCHR AFM probes (Nanosensors) were used for noncontact imaging. The procedures for AFM imaging of nanoelectrodes either in air or in solution were developed recently (25). Briefly, a nanoelectrode was mounted vertically with its polished surface facing the AFM probe using a homemade sample holder, and the cantilever was positioned above it with the help of an optical microscope. In a noncontact mode, the tip was brought within a close proximity of the sample using the approach function, and then the nanoelectrode was moved laterally in 200-nm steps to bring the AFM probe to its apex. The travel direction was selected to effect z-axis retraction of the piezo actuator in a close-loop mode. This corresponded to sliding of the slanted tip surface along the edge of the glass insulating sheath of the electrode. When the piezo approached its upper limit, the z-stage motor was retracted by 1 μm to maintain the actuator within its range (12 μm).

Electrodeposition of Pt black into the etched cavity was carried out in a commercial liquid cell (Park Systems), which was mounted on the stage of the XE-120 scanning probe microscope. The etched working electrode was biased to -100 ± 30 mV vs. Ag/AgCl reference in the platinization solution using an EI-400 potentiostat (Ensman Instruments). Slightly different potential values were used depending on the initial recess depth to keep deposition time close to 1 min (the larger the recess depth the slower the deposition process). The cavity filling was controlled by line scanning above the central portion of the etched nanoelectrode and monitoring the cavity depth as a function of time.

Voltammetry and Electrochemical Experiments with Macrophages.

A two-electrode setup was used for voltammetric experiments with a nanometer-sized Pt working electrode and a commercial Ag/AgCl reference. Steady-state voltammograms of either aqueous ferrocenemethanol or ferrocene in acetonitrile were obtained for electrode characterization using a BAS 100B electrochemical workstation (Bioanalytical Systems). In vitro voltammetry of ROS/RNS was performed using either a BAS 100B or an EA162 Picostat (eDAQ) with an e-corder 401 system and EChem software.

Experiments with macrophages were performed at room temperature (22 ± 1 °C) on the stage of an inverted microscope (Axiovert 135, Zeiss) placed in a Faraday cage. For intracellular detection, the platinized nanoelectrode was moved slowly with a micromanipulator (MHW-103, Narishige) toward the cell until it touched the membrane. It was then further lowered by 500 nm to penetrate the cell. With the MHW-103 Narishige micromanipulator, this last movement could not be performed at a sufficiently slow rate as reported previously (15). This explains why only ∼20% of the insertions provided a perfect seal. In another ∼50% experiments the seal seemed to be good, but was not sufficiently tight, as evidenced by a small (a few percent of that recorded in the external solution) reduction wave obtained inside the cell. In the remaining ∼30% cases, the cell membrane was ruptured during the nanoelectrode insertion. The transients were recorded until the current attained a constant value corresponding to the baseline.

For extracellular detection of the ROS/RNS release induced by the nanoelectrode insertion, a 10-μm-diameter platinized carbon microelectrode (6–8) was initially positioned ∼30 μm above the cell and polarized for 3 min before each measurement. Then, the working microelectrode tip was precisely positioned with the micromanipulator at a fixed distance (5 μm) above the surface of the macrophage. At this point, the nanoelectrode was positioned with the second micromanipulator between the microelectrode platinized surface and the cell and inserted, as described above in the case of intracellular measurements. In both cases, the subsequent release of ROS/RNS was detected in real time by chronoamperometry (AMU130 amperometer, Radiometer Analytical) at +850 mV vs. Ag/AgCl; i.e., sufficiently positive for the oxidation of all H₂O₂, ONOO^-, NO, and species. The current transients were recorded using a Powerlab 4SP D/A converter with a Chart 4.2 interface (ADInstruments). The same equipment was used to insert a nanoelectrode into the cell and record current vs. time dependences (digitized at 10 kHz).

In all cases, the total charge (Q), the maximum current (I_max) and the half-time width (t_1/2) were extracted from the response that corresponds to the overall oxidation processes occurring at the measurement potential.