(oxidation/reduction)

In order to prove that the proposed mechanism involves the catalytic site of the metal phthalocyanine, we compared our electrochemical signals with the metal-free phthalocyanine (the exclusion of the catalytic site). Experiments were performed with H2Pc (dihydrogen phthalocyanine), which is the phthalocyanine ring without any metal coordinated to the nitrogen ligands. The H2Pc was not responsive to peroxynitrite, underlining the importance of the metallic center in this catalytic process (data not shown).

The surface coverage was calculated using the equation Equation (6):

$$
\Gamma = \mathbf{Q}/n\mathbf{F}\mathbf{A}\tag{6}
$$

where Γ is the coverage of CoPC immobilized upon the desired electrode surface (mol cm<sup>−</sup>2), Q is the charge taken from the integration of the oxidation wave resulting from the Co1+/2+ couple recorded in a pH 7.4 phosphate buffer solution (PBS) at slow scan rates, n is the number of electrons taking place in the electrochemical process, F is the Faraday constant, and A is the geometrical electrode area (without recourse to any surface roughness corrections, and it was calculated to be 0.126 cm<sup>−</sup>2, as the diameter of the WE is 4 mm) [43]. Using this calculation method, there was a surface coverage of 7.0558 × 10−<sup>9</sup> mol cm<sup>−</sup><sup>2</sup> for the drop-casted CoPC SPEs, while the commercial (DRP 410, Drop Sense electrode) had a similar surface coverage (8.9643 × 10−<sup>9</sup> mol cm<sup>−</sup>2). These kinds of commercial electrodes are recommended for the detection of hydrogen peroxide at low potentials (0.4 V).

#### 3.1.2. Batch Optimization of the CoPc-Modified Electrodes for PON Detection

Bedoui et al. [44] mention that species such as ascorbic acid, nitrite and nitrate, uric acid, hydrogen peroxide, and others could interfere in blood or other biological samples with the signal that one may obtain for peroxynitrite. For this purpose, we studied a series of interfering species using Cyclic Voltammetry and a GCE-modified CoPc and compared the signal with the ones for PON at the same concentration, 100 μM (Figure 6). The biological concentrations of these interfering species are low, but we used the same concentration as for PON measurements, which usually is produced at a rate of up to 50–100 μM/min, but the steady-state reaction is in the nanomolar range, for hours [45]. Ascorbic acid gave rise to an oxidation peak around 0.4 V, hydrogen peroxide around 0.6 V, and PON was electro-catalyzed around 0.1 V (as already described above). Nitrate is not electrochemically active, and nitrite was not responsive within the chosen potential windows, meaning that using potentials around 0.1 V gives a very good selectivity toward PON.

**Figure 6.** Evaluation of the interfering species using CV for 100 μM peroxynitrite (PON—green): 100 μM ascorbic acid (AA—red), 100 μM hydrogen peroxide (H2O2—black), 100 μM nitrite (NO2—blue), using a glassy carbon electrode (GCE)/CoPc electrode. Scan rate 100 mV/s, electrolyte: PBS pH 9 0.1 M + 0.1 M KCl [1].

More sensitive techniques than CV (Cyclic Voltammetry) such as LSV (Linear Sweep Voltammetry) and DPV (Differential Pulse Voltammetry) were used for further characterization of the electrodes. LSV voltammograms (Figure 7a) in the potential window of −0.4 to 0.3 V were used to observe the electro-oxidation of PON over the SPCE/CoPc electrode at different PON concentrations. The oxidation potential shifts from 0.075 V (for 72 μM PON) to higher potentials (around 0.1 V for 145 μM PON).

**Figure 7.** (**a**) LSV using SPCE/CoPc for the detection of PON, using droplets of analyte over the electrode, −0.4 to 0.3 V (electrolyte: PBS pH 9 0.1M + 0.1 M KCl), using different PON concentrations. (**b**) DPV of the SPCE/CoPc in the presence (red, blue, green) and absence (black) of 125 μM of PON, using different negative starting potentials (−0.1 V, −0.2 V, and −0.3 V).

Differential Pulse Voltammetry (DPV) revealed that upon reduction of the CoPc films by starting the scans from lower potentials, the cumulative current (both cathodic and anodic) increases (Figure 7b). This suggests that the pre-treatment of the electrode could improve the response of the electrode for PON because of the reduction of Co2+ to Co1+. In addition, the shape of the DPV peak suggests that, besides diffusion, other processes occur (e.g., adsorption of product or reactant molecules on the surface of the electrode or even the coordination of the ONOO− molecule to the metallic catalytic center).

#### *3.2. FIA Optimization of the SPCE/CoPc Electrodes for PON Detection*

Initially, we did hypothesize that while the oxidation of Co1+ to Co2+ takes place, it can also be part of the redox process involving the reduction of peroxynitrite to nitrite. Starting from this hypothesis, an important issue was understanding how to overcome the apparent irreversible oxidation of PON from batch electrochemical analysis. The ability of flow injection analysis in understanding reaction mechanisms and complicated electrode

kinetics [46] served as an important tool to select between the oxidation potential of peroxynitrite and of other species, such as hydrogen peroxide, and the electrocatalytic reduction potential of peroxynitrite was unraveled and exploited.

FIA coupled with chronoamperometry gave us the opportunity to develop a very selective sensor for PON, using the gathered information from the batch electrochemistry. First, we wanted to establish the optimal potential, so we used chronoamperometry and changed the applied potentials (0.00, 0.10, and 0.25 V) on a SPCE/CoPc electrode. This study helped us to determine that the reduction of PON occurs below 0.1 V, as opposed to the oxidation of PON that occurs above 0.1 V (Figure 8a). Due to the several other advantages already described, a single line flow injection system was employed for further experiments using the revealed reduction potential for further optimization of the PON sensor.

**Figure 8.** (**a**) Chronoamperometry (CA) spectra of the flow injection analysis (FIA) using different potentials: 0.00 V (black), 0.10 V (green), and 0.25 V (red) [1]. (**b**) Interfering species study using the FIA equipment and chronoamperometry at 0.1 V, in PBS pH 9: 250μM nitrate, 250 μM nitrite, hydrogen peroxide 980 μM and 250 μM ascorbic acid (AA), as compared to 31, 38, and 45 μM PON signals.

Determination of the optimum flow rate was done using injections of 200 μM of PON (PBS pH 12) and amperometry at 0.1 V. The curve *flow rate* vs. *current* was fitted with R<sup>2</sup> = 0.9195 with a fourth-grade polynomial function. The optimal flow rate of 0.4 mL/min and a voltage of 0.1 V were used to further optimize the detection process, as the response time of this flow rate is fast, around 10 s. Interfering species were also evaluated using chronoamperometry at 20-fold higher concentration than PON for H2O2 and 5-fold higher concentration for the other species. As determined also by batch Cyclic Voltammetry, hydrogen peroxide gave rise to oxidation signals, but with very low sensitivity (Figure 8b) in contrast to PON, which gave rise to sensitive reduction signals. SPCE/CoPc electrodes are known to be used for the oxidation of hydrogen peroxide at 0.4 V.

As we have already shown for in batch electrochemistry (Figure 8b), the pre-treatment of electrodes with a reduction potential might be very important before the quantification of PON. We optimized the amount of time needed to reduce the CoPc film to obtain the best CA signal for PON. We used FIA amperometry for different time periods, 0, 20, 30, 60, 120, and 180 s and the reducing potential −0.3 V, which is the redox potential for Co2+/Co1+. We determined that applying a potential of −0.3 V for 60 s was the optimal procedure (data not shown).

We have performed the calibration (Figure 9a) using the flow injection system, at 0.1 V with the chronoamperometric method (Figure 9b), and we obtained a sensitivity of 6.31 nA μM−<sup>1</sup> (R2 = 0.9938), after the pre-treatment at −0.3 V for 60 s. The calculated LOQ = 2.41 μM, the calculated LOD = 0.72 μM, and the linear range is 3–180 μM. The reproducibility varied from 95% to 99% (50 μM PON) and the RSD for each calibration concentration (in triplicates) did not exceed 10%. The analytical parameters are very good if we consider several facts: (i) our unstable oxidative anion species are hard to detect, and (ii) screen-printed carbon electrodes are disposable electrodes.

**Figure 9.** (**a**) Calibration curve of the SPCE/CoPc electrode for PON, PBS pH 9. (**b**) Chronoamperogram measured using the FIA system and the SPCE/CoPc electrode, E = 0.1 V, flow rate = 0.4 mL/min [1].

By replacing the electrochemical polarization of the SPCE/CoPc electrode during the 60 s, at −0.3 V with the chemical oxidation of the CoPc film using sodium borohidrate (25 mM), for 20 min, the calibration was improved to Ired (nA) = 10.843·CPON (μM) − 36.484 (R2 = 0.9925). The new LOD was equal to 0.42 μM, and the LOQ was 1.4 μM.

The LODs of our method reached nanomolar level, the same level of PON under physiological conditions. Even though we did not study the interaction of PON in the absence or in the presence of myoglobin at physiological pH, our method has physiological relevance because it could be further used for this purpose by miniaturization of the electrodes with the same electrocatalytic bio-mimetic film. The micro-dimension of the surface-active area of the electrode offers enough sensitivity to study, for example, the formation of PON by cells (the cells being also in the micrometer range) involved in the oxidative burst (micromolar range of PON), in the redox signaling, or even in the steady state of PON (nanomolar range), as the literature suggests [47,48].

#### *3.3. UV-Vis and Determination of Synthesized PON for Kinetic Studies*

Molina et al. [30] describe the influence of buffer and pH over the stability of peroxynitrite solutions ("peroxynitrite" being the term widely accepted for peroxynitrite and peroxynitrate). The rate constants depend on pH, ionic strength, temperature, scavengers, and other parameters. Several mechanisms of PON decay have been already proposed in the literature. In acidic conditions, the isomerization of peroxynitrite occurs (mainly present as ONOOH, the form that decays rapidly to nitrate), independent of total peroxynitrite concentration. If pH is ≥7, nitrite is the main decomposition product of PON (the higher the pH value and concentration of PON, the higher the conversion yield to nitrite) [31]. As mentioned in Section 2.5, the Equation (1) describing the bimolecular decomposition of PON (as opposite to the mononuclear isomerization, as termed by IUPAC) is supposed to be predominant at pH = 9, especially when PON concentration is higher than 0.1 mM [32,49].

The reaction from Equation (1) was proposed to follow second-order or pseudo firstorder kinetics, depending on pH [31].

Molina et al. [30] proposed that the decomposition of PON to nitrite has more intermediate steps than the one in Equation (1) (disproportionation reaction followed by the formation of intermediary/adduct species and then followed by decomposition to nitrite). The formation of the adduct is very rapid in the range of 10<sup>4</sup> M−<sup>1</sup> at alkaline pH. The last direct decomposition step to nitrite is much slower than the other elementary steps [31]. So, this could be the rate-determining step in these series of proposed reactions involved. In addition, the disproportionation of PON at pH 9 seems to favor the equilibrium toward ONOO<sup>−</sup>, as knowing the pKa of ONOOH (6.8), one can calculate the concentration of ONOOH at pH 9. For a concentration of 150 μM ONOO− in PBS at pH 9, there are only 0.946 μM ONOOH, and for 50 μM ONOO− in PBS pH 9, there are 0.3154 μM

ONOOH. At pH 9, one can say that ONOOH concentration is insignificant (less than 1% of the ONOO− concentration) if pKa 6.8 is taken in consideration. Nevertheless, the pKa of ONOOH depends on the ionic strength and pH, so these calculated values could be further refined [49].

If ONOOH is present in small amounts at pH 9, one can assume that the isomerization is insignificant (as the only form to isomerize is ONOOH) and PON decay should follow the pseudo first-order reaction at this pH.

Although the mechanism of PON decay is not fully understood, we took in consideration the generally agreed model depicted in Equation (1) for pH 9, as the aim was to determine the accuracy of the FIA-EC method compared to the classical UV-Vis one in determining the kinetic parameters. We checked both possibilities of reaction order for Equation (1), i.e., the pseudo first-order and the second-order.

We calculated the decay (apparent) rate constant and half-lives of PON using two methods: Method A involves the assumption that both (the (pseudo) first and second order) integrated rate laws could be possible. One finds the rate constant value from the slope of the two graphs: *ln(CPON)* vs. *time*, where t1/2 = ln2/k (for the (pseudo) first-order reaction) and *1/CPON* vs. *time,* where t1/2 = 1/(C0PON·k) (for the second-order reaction). Using the slope of the most linear plot, one calculates the corresponding half-time.

The second method, Method B, also named the "half-life method", is based on the several half-time values from the *concentration* vs. *time* plot of the data, and it plots them after as *log(t1/2)* vs. *log(CPON*), where *n* = 1-slope (*n* being the (apparent) rate order). This method is more precise, as it is difficult to assess the linearity of a plot (as in Method A).

The idea was to compare the UV-Vis results with our proposed FIA-EC method using SPCE/CoPc electrodes. The UV-Vis spectrum of synthesized genuine PON is less complicated than other synthetic methods (such as using the nitric oxide and superoxide donor-based synthesis, SIN-1). The molar extinction coefficient of 1670 M−<sup>1</sup> cm<sup>−</sup><sup>1</sup> can be used to calculate the concentration of genuine ONOO− at 302 nm (Figure 10a) [33]. We evaluated the stability of our synthesized PON solution using UV-Vis spectrometry. As we have already described [1], because we performed our measurements at alkaline pH values, we had a significant amount of nitrite in the genuine PON solutions. The amount of nitrite is correlated with the absorbance at 355 nm. Calibration was performed at this wavelength using a Griess reagent-based protocol, (y = 0.0488·CNO2(μM) + 0.0076, R<sup>2</sup> = 0.9971, data not shown), the amount of nitrite was also assessed to be 72 ± 5 mM for a 117 mM PON solution (improved PON synthesis), at pH 9, in PBS 0.1 M. This amount of nitrite is confirmed in the literature [24]. Nitrite reaches a plateau between pH 9 and 10 [50].

The kinetic plots at 302 nm (*concentration* vs. *time,* Figure 10b for UV-Vis data and Figure 10d for FIA-EC data) were fitted with a single exponential curve in Origin 8.5 software, as suggested in the literature [2,24,30,34,51]. The *log*(*t*1/2) vs. *log(CPON)* was plotted, and the slope was equal to *slope* = 1 − *n*, where *n* is the reaction order (Figure 10c,e). As it can be observed, Method B can be applied for both UV-Vis, as well as for the FIA-EC described in Section 3.2, and it will be further descried and compared with Method A (Figure 10f).


**Table 2.** Calculated pseudo first-order decay rates constants for PON, in PBS pH 9 (0.1M), at 25 ◦C, determined using classical UV-Vis method at 302 nm and the FIA-EC method, using the SPCE/CoPc (Method A).

**Figure 10.** (**a**) UV-Vis spectra of 50 μM PON, pH 9, PBS 0.1 M. The illustration of the second method "half-lives method": plot of *CPON* vs. *time*, fitted with a single-exponential equation (y = y0·e<sup>−</sup>kx) for the data obtained using (**b**) the UV-Vis method and (**d**) our proposed FIA-EC method. Plot of *log(C0PON)* vs. *log*(*t*1/2) from (**c**) UV-Vis and from (**e**) FIA-EC, fitted linearly for the determination of apparent reaction order from the slope of the equation (Method B). The illustration of the first method (Method A): (**f**) plot of *ln(CPON)* vs. *time* according to pseudo first order, fitted linearly for both UV-Vis and our proposed FIA-EC method (linear correlation function, R<sup>2</sup> linear correlation coefficient and apparent kinetic constant is determined from these plots, data presented in Table 2). All spectra represent the data for 50 μM PON, pH 9, PBS 0.1 M.

#### *3.4. UV-Vis Determination of Different Forms of Myoglobin*

In meat extracts, different oxidation forms of myoglobin are present, as we depicted in Figure 1. The conversion of reduced myoglobin (MbFe2+OH2 or MbFe2+O2) to metmyoglobin (MbFe3+(OH2)) can be followed using UV-Vis due to PON scavenging activity (Figure 11): the concentration of the redMb solutions can be verified by measuring the absorbance at 417, 542, and/or 580 nm (<sup>ε</sup>417 = 128 mM−<sup>1</sup> cm<sup>−</sup>1, ε542 = 13.9 mM−<sup>1</sup> cm −1, and ε580 = 14.4 mM−<sup>1</sup> cm<sup>−</sup>1) [21], and the spectrum of metMb has a maximum of absorbance at 502 [<sup>ε</sup>502 = 10.2 mM−1cm−1] and 610 nm at pH 6.4, and the Soret band absorbance maximum is at 408 nM at pH 7.4 [52]. So, the scavenging effect could be identified using UV-Vis, but no quantification of PON decay can be done in a direct, rapid, sensitive, and selective manner.

**Figure 11.** (**a**) UV-Vis spectra for different reaction times of 75 mM NaBH4 and 25 μM metMb (PBS pH 9). (**b**) UV-Vis spectra for different incubation times (0, 10, 20, 30, 40, 120, and 180 s) of 100 μM PON with 10 μM redMb (PBS pH 9).

The quantification of the concentration of redMb was realized with measurements at 580 nm. The concentration of the stock meat extract was determined to be 480 μM, which corresponds to ca. 20 mg of myoglobin for 1 g of meat. Moreover, if we use the absorption at 525 nm (representing the isosbestic point for the absorption in visible range for the 3 forms of myoglobin) and a molar extinction coefficient of 7.6 mM−1cm−1, we obtained a value of 485 μM of myoglobin for the same stock solution. A more complex method for determining the content of myoglobin was described by Krzywicki, and improved by Tang, in 2004 [53].

The reactions of PON with myoglobin or meat were studied both with the electrochemical and spectrophotometric methods. Figure 12 describes the evolution of the Mb absorption peaks during Mb incubation with 50 and 150 μM PON. For PON 50 μM, almost no change was observed after 11 min of incubation, but the same incubation of metMb with 150 μM PON induced a more significant change, with decrease in absorbance at 542 and 580 nm, and the appearance of a band at 700 nm, corresponding to a qualitative evaluation of the catalytic reaction. The same incubation was studied with our FIA-EC method, also to obtain quantitative information regarding the catalyzed PON decay.

**Figure 12.** UV-Vis spectra of (**a**) 50 μM PON and (**b**) 150 μM PON incubated with 15 μM metMb, at different incubation periods.

#### *3.5. Studying the Reaction of Myoglobin with Peroxynitrite with FIA-EC*

The incubated solutions of 50 μM and 150 μM PON with 15 μM Mb were also investigated, in parallel, with our FIA-EC optimized method (Figure 13). It was expected that the kinetics between the two concentrations of PON to be different due to the ratio between PON and Mb, although the difference could be overcome by the kinetics of spontaneous PON decay independent of Mb, which presents opposite behavior. More precisely, increasing PON concentrations favor pseudo first-order reaction in relation to Mb and deviation to second-order reaction in the case of spontaneous PON decay. Moreover, approximatively at least eight equivalents of PON are necessary for the complete oxidation of myoglobin (depending on experimental conditions and pseudo first-order kinetics) [21]. The recovered current was converted into PON concentration with the calibration curve, and in the following sections, we discuss and compare the decay of PON using these data.

**Figure 13.** Chronoamperogram using the FIA-EC system for (**a**) 50 μM and (**b**) 150 μM PON, in the presence and in the absence of 15 μM metMb, at different incubation periods (0, 30, 60, 120, 240, 420, 660, 820 s). The decomposition of PON alone was studied at the same incubation periods to prove that the recovery of the current is due to the presence of PON in the metMb solution and not to the metMb itself (PBS pH 9, 0.1M, E = 0.1 V and flow rate 0.4 mL/min). The values obtained in these studies were further used for kinetic calculations.

#### *3.6. Studying the Reaction between Myoglobin from Meat Extracts and Peroxynitrite Using FIA-EC*

Meat extracts were first diluted 10-fold and analyzed with UV-Vis to determine the quantity of myoglobin. Using the isosbestic point at 525 nm, as described above, we determined 15 μM of myoglobin (independent of the oxidation form, Figure 14a). The incubation of the meat extracts with PON was studied both with UV-Vis and FIA-EC. Using the optimized calibration curve of the electrochemical method, we developed (Ired (nA) = 10.843·CPON (μM)–36.484, R<sup>2</sup> = 0.9925), quantified PON during its incubation with both Mb 15 μM and meat extracts diluted 10 times in PBS pH 9, and compared it to a normal decomposition rate of PON, at pH 9, without any scavenger (Figure 14b).

**Figure 14.** (**a**) UV-Vis spectra of meat diluted 10 times in PBS pH 9 in the presence of 50 μM PON and in absence, during 12 min. (**b**) The graphs of concentration (from FIA-EC) over time for 50 μM PON decay in the absence (black) or in the presence of 15 μM Mb (cyan) or meat extract (red).

It can be clearly observed that the decomposition of PON takes place at a faster rate for the PON samples incubated with both Mb and meat extracts.

Calculated apparent reaction orders, apparent rate constants, and half-lives determined using UV-Vis spectrophotometry and our FIA-EC method are presented in Tables 2–5 and further discussed in the following two sections. All measurements and calculations were done for PBS pH 9, 0.1 M, at 25 ◦C.

**Table 3.** The "half-life" method: Determined values of half-life for PON and calculation of rate orders. PBS pH 9 (0.1 M), at 25 ◦C, classical UV-Vis method at 302 nm and the FIA-EC method, using the SPCE/CoPc (Method B).


\* Rate order = n = 1- slope of the *log*(*C*0) vs. *log*(*t*1/2), the data were fitted with a single exponential function based on equation y = y0·e<sup>−</sup>kx. Estimations of the rate orders were done in the case of values lower than 10−4.

> **Table 4.** Observed first-order decay rates constants for PON, in PBS pH 9 (0.1M), at 25 ◦C, determined using the classical UV-Vis method at 302 nm and the FIA-EC method, using the SPCE/CoPc (Method B).


**Table 5.** Calculated second-order decay rates constants for PON, in PBS pH 9 (0.1M), at 25 ◦C, determined using classical UV-Vis method at 302 nm and the FIA-EC method, using the SPCE/CoPc (Method B).


3.6.1. Estimation of the Apparent Rate Decay Orders of PON in the Absence and Presence of Myoglobin

Using Method A (Table 2), for 50 μM PON, the R<sup>2</sup> values for the plot according to a first-order apparent kinetics (0.9985) were higher than for a second-order one (0.9246); as for 150 μM PON, the results were not conclusive (0.9872 for pseudo first order and 0.9877 for second order). Taking in consideration Equation (1) and the excess of ONOO− comparing to ONOOH, these results from Method A are relevant and indicate pseudo first-order kinetics.

Using Method B, the apparent decay of PON is also a pseudo first-order reaction for both 50 μM and 150 μM (see Table 3), with calculated order values of 1.0030 and 1.0054 for the UV-Vis method, in comparison with 1.0000 and 1.0001 for the FIA-EC method. The differences between these two methods were less than 3% (acceptable error values). The fitting curves for both 50 μM PON (Chi-Sqr = 3.9770, R<sup>2</sup> = 0.9865) and 150 μM PON Chi-Sqr = 1.9293, R<sup>2</sup> = 0.9990) were proper, as tolerance criteria were satisfied.

Our results are in accordance with the literature, describing that especially concentrations above 100 μM present deviations from (pseudo) first-order reactions as second-order in total peroxynitrite concentration at pH 9 [30,31]. These deviations are most pronounced at pH 9 [50].

Method B helped us to assess numerically the reaction order out of our data acquired with the FIA-EC method for the interaction of 50 and 150 μM PON with 15 μM Mb, and the calculated reaction orders were 1.0222 and 1.0001, suggesting also a pseudo first-order kinetics in both cases. Deviations in apparent second order from the single-exponential curve were higher in the case of 50 μM PON + 15 μM Mb, as expected (meaning that the model could be adjusted, as pseudo first-order kinetics were not fulfilled for PON:Mb, Chi-Sqr = 15.7763, R<sup>2</sup> = 0.9446). These deviations were smaller for 150 μM PON + 15 μM Mb (Chi-Sqr = 17.9047, R<sup>2</sup> = 0.9941), as pseudo first-order kinetics were fulfilled: PON: Mb was 10:1.

Using both methods (A and B), the only studied interaction that gave clear secondorder decay for PON was the meat diluted 10 times with 50 μM (R<sup>2</sup> = 0.9020 vs. R<sup>2</sup> = 0.9968 for (pseudo) first order and second order, respectively, using method A). The reaction order assessed with Method B is 1.94, which is very close to a second order. One explanation can be that in meat extracts, the interactions are more complicated/complex than with standard Mb, as other scavengers could be present, so an apparent second-order decay is easier to determine even using Method A (which is less precise than Method B). Method B will be used further on. Nevertheless, the fitting should be replaced with a more significant equation in this case (Chi-Sqr = 11.10, R<sup>2</sup> = 0.9452), because this high deviation from a single-exponential equation proves once again that a second-order decay might be involved. A double-exponential equation could be helpful, as suggested in reference [54].

Nevertheless, the single-exponential fitting of pseudo first-order situations (CPON = C0PON·e−kx) determines *kobs*, which allows us to detect the (apparent) second-order rate constants (kcat), for a larger concentration range of scavengers or other reactants involved (especially the concentration of the catalyzer, myoglobin, or other) [24,25]. The kcat obtained from a linear fit of *kobs* vs. *catalyst concentration* will refine our findings on second-order kinetics and improve the investigation of the scavenging effect of myoglobin over PON in meat extracts or other biological samples.

#### 3.6.2. Determination of Apparent Rate Constants and Half-Lives for the Decay of PON

Molina et al. [30] describe the observed half-life t1/2 = 9.4 ± 0.1 s for PBS 0.07 M, at pH 8, for PON 250 μM. We have determined that apparent k = 0.0086 s<sup>−</sup><sup>1</sup> with Method A and k = 0.0078 s<sup>−</sup><sup>1</sup> with Method B (the apparent first-order decomposition rate constant), for PBS 0.1 M, at pH 9, for a concentration of 150 μM (t1/2 = 252.12 ± 2.97). As it can be seen, in the chosen conditions, PON is more stable than in the conditions described by Molina et al., mainly because of more alkaline pH, more ionic strength in the buffer, and smaller concentrations of PON. Kissner et al. described this kind of behavior [31].

The decomposition of PON occurs faster at lower concentrations at pH 9, as it can be observed from the UV-Vis data in Table 3. The same conclusion can be drawn from our SPCE/CoPc developed method: PON 50 μM will decay faster (reaction rate of 0.426 × 10−<sup>6</sup> s<sup>−</sup>1, using t1/2 = 81.33 s) than PON 150 μM (reaction rate of 0.412 × 10−<sup>6</sup> s<sup>−</sup>1, using t1/2 = 252.12 s), where k = 0.693/t1/2 and r = k[CPON]. Reaction rates in the range of 10−<sup>6</sup> for 50 μM PON at pH 9 are described by Kissner et al. [31], and this is in very good accordance with the data we obtained.

The myoglobin-mediated decay of PON is described as a second-order rate interaction. Most probably, the scavenging effect gives rise to ferrylmyoglobin (MbFe4+=O) with a rate constant of 4.6 ± 0.2 × 10<sup>4</sup> M−<sup>1</sup> s<sup>−</sup>1, at pH 8.3, in the absence of CO2, that will further react with PON with a rate constant of 1.2 ± 0.2 × 10<sup>4</sup> M−<sup>1</sup> s<sup>−</sup><sup>1</sup> to form metMb [21]. The oxidation of redMb to metMb is not a stoichiometric reaction, as eight to 25 equivalents of PON are required for the oxidation reaction to be completed (depending on the absence or presence of CO2), as the natural decay of PON takes place at the same time. A 10-fold excess of PON is necessary for pseudo first-order conditions to be fulfilled. As the concentration of nitrite does not influence the interaction of myoglobin with peroxynitrite in a direct manner [21], our PON synthesized using the alkaline method is suitable for kinetic calculations (or otherwise saying, the decay of PON in the presence of myoglobin is a zero-order reaction in nitrite [2,21]).

The metMb scavenges PON at a lower rate than redMb or even than ferrylMb. MetMb catalyzes the isomerization of PON to nitrate, at pH 7, at 20 ◦C, with a rate of 29,000 ± 100 <sup>M</sup>−1s−1, and an iron Fe3+ core of metMb is involved in this process. The k values decrease with increasing pH; thus, the decay rate is expected to be smaller for pH 9 [24]. So, scavenging PON with metMb is less effective at pH 8 (or above) than at biological pH, with a PON decay rate constant k of 2700 ± 30 <sup>M</sup>−1s−1, for pH 8, 0.1 M PBS, for 100 μM of PON, in the absence of CO2, at 20 ◦C [24]. In our case, at pH 9, PBS 0.1 M, 25 ◦C, the apparent k value of the reaction of 15 μM myoglobin and 50 μM PON was calculated to be 311.87 ± 7.96 M−1s−<sup>1</sup> using Method B (where deviations from pseudo first order are second order). This value is lower than the value that Herold et al. obtained, which is probably due to the pH values and the temperature difference. As far as our knowledge goes, the kinetics of this exact decay conditions were not described in the literature before. Nevertheless, more PON or target molecule (myoglobin) concentrations are to be varied for refining better apparent second orders.

The scavenging was evaluated by comparing the decomposition rate of PON in the absence and in the presence of myoglobin (from both standard and meat extracts). The standard Mb decreased the half-life on PON from 81 to 65 s (for 50 μM) and from 250 to 88 s (for 150 μM), thus increasing the rate of decomposition in both cases. As in accordance with the UV-Vis spectra (Figure 12a) of metMb incubated with 50 μM of PON, the scavenging effect of 15 μM metMb with 50 μM PON is not very effective, and the calculated t1/2 values are in accordance with the UV-Vis data (with less 10% error values between the two methods). This is because the number of equivalents of PON (here around 3) were not sufficient to oxidize myoglobin. When we increase the number of equivalents to 10 (150 μM PON), the scavenging effect can be observed with both UV-Vis and FIA-EC.

The scavenging effect of the myoglobin from meat was stronger than we initially thought (as we expected the concentration of Mb to be similar to the standard Mb), with a t1/2 of 19.77 ± 0.10 s. Even if we estimated the concentration of myoglobin to be around 15 μM in the extracted meat (diluted 10 times in PBS pH 9), when we compare the decay constants of PON, we can observe that the standard Mb has a lower apparent k constant (311.87 ± 7.96 <sup>M</sup>−1s−1) than the meat extract (891.76 ± 220.54 <sup>M</sup>−1s−1); thus, it has a higher half-life. Nevertheless, this may come from different oxidation states of the myoglobin in meat extract, and the presence of other possible scavengers in meat is not excluded. Other studies including varying concentrations of PON, myoglobin, and/or other catalysts from meat or other biological samples are still to be performed further.
