Monitoring the heme iron state in horseradish peroxidase to detect ultratrace amounts of hydrogen peroxide in alcohols

Despite the importance of hydrogen peroxide (H2O2) in initiating oxidative damage and its connection to various diseases, the detection of low concentrations of H2O2 (<10 μM) is still limited using current methods, particularly in non-aqueous systems. One of the most common methods is based on examining the color change of a reducing substrate upon oxidation using UV/Vis spectrophotometry, fluorophotometry and/or paper test strips. In this study, we show that this method encounters low efficiency and sensitivity for detection of ultratrace amounts of H2O2 in non-aqueous media. Thus, we have developed a simple, fast, accurate and inexpensive method based on UV/Vis spectrophotometry to detect H2O2 in non-aqueous systems, such as alcohols. In this regard, we demonstrate that monitoring the Soret and Q-band regions of high-valent iron-oxo (ferryl heme) intermediates in horseradish peroxidase (HRP) is well suited to detect ultratrace amounts of H2O2 impurities in alcohols in the range of 0.001–1000 μM using UV/Vis spectrophotometry. We monitor the optical spectra of HRP solution for the red shift in the Soret and Q-band regions upon the addition of alcohols with H2O2 impurity. We also monitor the reversibility of this shift to the original wavelength over time to check the spontaneous decay of ferryl intermediates to the ferric state. Thus, we have found that the ferryl intermediates of HRP can be used for the detection of H2O2 in alcohols at μg L−1 levels through via UV/Vis spectrophotometric method.


Introduction
Hydrogen peroxide (H 2 O 2 ) is an essential oxygen metabolite in living systems and serves as a messenger in cellular signal transduction. 1 The overproduction of H 2 O 2 from the mitochondrial electron transport chain results in oxidative stress, causing functional decline in organ systems. 2 Such oxidative stress over time is also connected to various diseases, including cancer, 3 cardiovascular disorders, 4 and Alzheimer's disease and related neurodegenerative diseases. 5 Moreover, H 2 O 2 and its derivatives are strong oxidizing agents employed in many industrial and medical processes, such as the synthesis of organic compounds and disinfection. 6,7 The signicant impact of H 2 O 2 on a variety of oxidative damage mechanisms, 8,9 environmental hazards, 7,10 and human health, 11,12 as well as its application in biosensing, 13 provide motivation to develop a sensitive and selective diagnostic method for detecting and quantifying H 2 O 2 , particularly at low concentrations.
Over the past several decades, many H 2 O 2 sensing techniques have been devised based on spectrophotometry, 14,15 uorescence, 12,16 chemiluminescence, 17 enzymatic, and electrochemical methods. [18][19][20][21][22] One of the most extensively used enzymatic systems for H 2 O 2 sensing is horseradish peroxidase-H 2 O 2 system (HRP-H 2 O 2 ). [21][22][23][24][25][26] The broad application of horseradish peroxidase (HRP) in H 2 O 2 sensing is due to its ability to translate catalysis into an electrochemical signal, as well as its stability and commercial availability. 27 HRP is able to catalyze the heterolytic cleavage of the peroxidic bond in H 2 O 2 and form a high-valent iron-oxo (ferryl heme) intermediate of the enzyme (compound I). [28][29][30][31] In compound I, the iron at the heme center has been oxidized from Fe III to Fe IV ]O, and the porphyrin or an amino acid in the side chain of HRP is oxidized to a radical. 32 Thus, compound I can oxidize two molecules of a reducing substrate, such as ABTS (2,2 0 -azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)), through two consecutive single electron reactions to form compound II, before nally being reduced back to the Fe III state. [32][33][34][35][36][37][38][39] Most HRP-based methods, however, are limited by some serious disadvantages, such as environmental instability, complex fabrication design, tedious immobilization procedures, and high cost. 15,18 For example, 3,3 0 ,5,5 0tetramethylbenzidine/HRP-based method to detect H 2 O 2 is mainly based on the use of 3,3 0 ,5,5 0 -tetramethylbenzidine (TMB), which is the most commonly used chromogen for HRP. TMB performs as a reducing organic substrate and is oxidized by ferryl intermediates formed upon the reaction of HRP and the H 2 O 2 impurity in the media. 40 Moreover, most of these methods are only efficient for H 2 O 2 detection in aqueous media and encounter low efficiency and sensitivity in organic solvents, such as alcohols. As previously reported, the partial oxidation of primary or secondary alcohols due to autoxidation results in the production of H 2 O 2 , which produces an aldehyde or ketone as a coproduct. 41 Since alcohols are commonly used in various chemical and enzymatic reactions, the presence of unreported amounts of H 2 O 2 can interfere with reaction cascades. 42 Therefore, providing a selective, rapid, convenient, and low cost analytical method for detection of H 2 O 2 in alcohols is of great interest.
Here, we report a simple spectrophotometric method to detect H 2 O 2 in alcohols at mg L À1 levels through the direct detection of the ferryl intermediate of HRP. In this manner, we are able to detect ultratrace amounts of H 2 O 2 in alcohols, such as ethanol, glycerol, 2-chloroethanol, and isopropanol. In this method, we monitor the red shi in the Soret and Q-band regions of the HRP's optical spectrum upon the addition of alcohols with H 2 O 2 impurity to the HRP aqueous solution at pH 6.0. The red shi of the Soret band from 402 nm to 418 nm, and the appearance of two Q-bands at 527 nm and 557 nm are indicative of the formation of the ferryl intermediates of HRP, which can be formed only in the presence of H 2 O 2 impurity. [43][44][45] Thus, we consider the change in the Soret band of Fe(III) and formation of Fe(IV) ferryl intermediates as conrmation of the presence of H 2 O 2 impurity. Furthermore, we monitor the reversibility of the red shis over time to their original wavelengths as an indication of the spontaneous decay of the ferryl intermediates to the ferric state, distinguishing these red shis from possible solvatochromic shis. Our method is based on the use of HRP by itself, and does not need any reducing substrate such as ABTS and TMB. We monitor the formation of ferryl intermediates upon the reaction of HRP and H 2 O 2 impurity in the media. Using this method, we can efficiently detect mg L À1 levels of H 2 O 2 (0.001-1000 mM) in alcohols, where it is barely possible to detect this amount of H 2 O 2 using other common methods, such as hydrogen peroxide test strips. We characterize the ferryl intermediates and their decay to ferric heme upon the addition of alcohols to HRP using UV/Vis spectrophotometry and conrm their presence using electron paramagnetic resonance (EPR) and cyclic voltammetry (CV). This demonstration suggests the importance of monitoring the ferryl intermediates for the detection of H 2 O 2 impurity in alcohols at mg L À1 levels using a simple, cost-effective, and accurate method.

Results and discussion
Common hydrogen peroxide test strips were used to detect H 2 O 2 impurity in a 100% ethanol sample, the minimum detectability of these strips is reported as 30 mM H 2 O 2 ( Fig. 1a(i)). Fig. 1a(ii) illustrates that the strips showed no detectable hydrogen peroxide in ethanol. We prepared different concentrations of H 2 O 2 in both water and pure ethanol, and compared the color of hydrogen peroxide test strips aer exposure to these solutions (Fig. 1a, iii). The results showed the test strips barely detected H 2 O 2 in ethanol at low concentrations (<100 mM), and cannot accurately quantify the concentrations of H 2 O 2 (<1 mM) (Fig. 1a, iii). Thus, we aim to develop an accurate method to detect and quantify the H 2 O 2 impurity in alcohols.
Studies have shown that the optical spectra of the resting state of HRP (Fe III ) has a Soret band of 402 nm and a Q-band of 497 nm, while the ferryl heme intermediate of HRP (Fe IV ]O) has a Soret of 418 nm and two Q-bands of 527 nm and 557 nm (Fig. 1b). [43][44][45] We hypothesized that these indicative peaks could be used to detect the H 2 O 2 impurity of alcohols and developed a simple spectrophotometric method utilizing HRP solutions to detect and quantify H 2 O 2 based on the formation of ferryl intermediates (Fig. 1c). 44 We Thus, we suggest using the 1 mM HRP solution for detection of high and low H 2 O 2 impurity in ethanol in the range of 0.001-1000 mM. The lower concentration of HRP is better for lower concentrations of hydrogen peroxide because less HRP is in resting ferric state.
The red shis correlated to the ferryl intermediates can be distinguished from solvatochromic shis. First, it is noteworthy that the addition of higher concentrations of ethanol containing H 2 O 2 impurity do not increase the red shi to more than 418 nm (Fig. 3a). Second, the ferryl intermediates of HRP are not thermodynamically stable and decay spontaneously (Fig. 3b). 32,46 Thus, the red shis of the ferryl intermediates would be reversible. These two characteristics of the red shis, correlated to the ferryl intermediates, can distinguish them from solvatochromic shis. Even at lower concentrations of H 2 O 2 in ethanol, the red shi can reach 418 nm upon the titration of the HRP solution (either 1 mM or 10 mM) with more ethanol, which adds more H 2 O 2 to the system ( Fig. 3c and d). However, over time ($2 h) the Soret band moved back to 402 nm, showing that the ferryl intermediates decay to the iron(III) state. Treating HRP solutions with different quantities of ethanol, all of which contain the 80 mM H 2 O 2 impurity, we found that the time for the ferryl intermediates of HRP to decay to 402 nm increases from 2 h for 10% v/v ethanol to 2.5 h for 20% v/v ethanol in the reaction mixture ( Fig. 3e and f). This increase in the decay time is because of the addition of more H 2 O 2 in the reaction mixture along with the ethanol, which makes it possible for each molecule of HRP to produce ferryl intermediates for a longer time. Thus, the ferryl intermediates are accumulating in the reaction mixture for a longer period of time, and subsequently, the shis take longer to return to their original wavelength. To investigate the utility of this method beyond ethanol, we explored the detection of H 2 O 2 impurity in several different alcohols. The UV/Vis spectra show that the Soret band of HRP red shis to 418 nm upon the titration of the HRP solution with ethanol, glycerol, isopropanol, and 2-chloroethan-1-ol, with a shoulder at 350 nm, as well as Q-bands at 527 nm and 557 nm (Fig. S3a-d †). However, methanol, 2,2-dichloroethan-1-ol, 2,2,2trichloroethan-1-ol, 2,2,2-triuoroethan-1-ol, 2-mercaptoethan-1-ol, or ethylene glycol do not show any red shi in the Soret band or Q-band of HRP (Fig. S4 †). This suggests that they do not contain any detectable H 2 O 2 impurity. Interestingly, when ferryl intermediates decay to the ferric state over time, HRP can show the same red shi from 402 nm to 418 nm upon the second addition of these alcohols ( Fig. S5a and b †). Thus, the HRP solution can be used as a recyclable method to detect H 2 O 2 impurities. Moreover, the circular dichroism data shows that HRP conformation does not change at all during the ethanol treatment at both 1 mM and 10 mM HRP solution concentrations ( Fig. S5c and d †).
Previous studies with HRP and H 2 O 2 have shown that the ferryl intermediates of HRP formed through the reaction of ferric state of HRP with H 2 O 2 are EPR-silent and generate only a broad EPR signal characteristic of the oxyferryl porphyrin pcation radical (g ¼ 2.0). 38,44,47,48 EPR spectra for HRP solutions treated with ethanol (Fig. 4) and glycerol (Fig. S6 †) containing H 2 O 2 impurity conrm the formation of ferryl intermediates, which are also shown in the UV/Vis spectra shown earlier ( Fig. S3a and b †). The HRP solution with pure ethanol does not show any formation of ferryl intermediates, based on the results from UV/Vis spectrophotometry (Fig. 2b). The initial formation of an EPR-silent intermediate upon the reaction of ferric HRP with alcohol molecules over a short timescale (30 s) shows that the ferric signals in native HRP (control) (Fig. 4a and S5a †) disappear in the high-spin region of heme while a new resonance is observed in the region of the low spin heme (g ¼ 2.0) arising from a radical cation located on the heme (Fig. 4b and S5b †). The EPR spectra of the intermediates formed upon the addition of impure ethanol and glycerol to HRP solutions are also in agreement with previous reports of EPR analysis of ferryl intermediates upon the addition of H 2 O 2 to HRP, ascorbate peroxidase, and cytochrome C peroxidase. 44,47,48 Examining the EPR spectra for one hour, we found that the features of the high spin ferric heme (Fe III ) begin to reappear over time, while the organic radical signal decays in the low-spin region (Fig. 4c-e, and S6c-e †). Thus, we also conrmed the decay of the ferryl intermediates through the reversibility of the EPR signals at both high-spin and low-spin regions.
We also conrmed the formation of the ferryl intermediate (Fe IV ]O) and its decay to Fe III using cyclic voltammetry. The cyclic voltammogram of the HRP solution shows the appearance of a reduction peak associated with the ferryl intermediate upon the addition of ethanol known to contain 80 mM H 2 O 2 (Fig. 5a), which is the same as the reduction peak that appeared upon the addition of H 2 O 2 aqueous solution to HRP solutions (Fig. 5b). The intensity of this reduction peak increases upon the addition of more ethanol percentage (v/v) containing the known 80 mM H 2 O 2 impurity (Fig. 5c). The decay of the ferryl intermediate over time is also shown by the decreasing peak intensity over 10 minutes (Fig. 5d). This result is in agreement with the decay of the Fe IV ]O intermediate demonstrated in the UV/Vis spectra (Fig. 3e-f) and EPR spectra (Fig. 4a-e), and the formation of ferryl intermediates upon the reaction of heme and H 2 O 2 . 46 We also quantify the detected H 2 O 2 impurity in alcohols using ABTS as a reducing organic substrate in the aqueous solution of HRP (pH 6.0). ABTS can donate electrons to ferryl intermediates of HRP (Fig. 6a, i). ABTS, which has been widely used in the literature to measure HRP activity, changes from colorless to blue-green in color upon oxidation, and the intensity of this color can be easily measured by UV/Vis spectrophotometry (Fig. 6a, ii). Measuring the color intensity of ABTS (2 mM and 20 mM) 30 minutes aer the reaction with HRP solutions (0.1 mM, 1 mM, and 10 mM) and ethanol containing different concentrations of H 2 O 2 impurity, we optimized the   (Fig. S7a, i †). Thus, we suggest using 0.1 mM HRP in a solution of 20 mM ABTS to quantify the H 2 O 2 impurity in alcohols in the ranges of 0-1000 mM.
Since the slope of the graphs for ABTS absorption versus H 2 O 2 concentration in the range of 0-1000 mM follows a step gradient, we propose three formulas to measure the accurate level of H 2 O 2 impurity in alcohols in the range of 0-1 mM, 1-100 mM, and 100-1000 mM, individually (Table S1 †). These formulas are derived from the linear ts based on the UV/Vis absorbance of ABTS at 420 nm for three ranges: 0-0.3 a. u., 0.3-1 a. u., and 1-4 a. u. Using these formulas based on the absorbance of ABTS, we successfully measured the H 2 O 2 impurity in a few common primary and secondary alcohols (Table S2 †).
In conclusion, we demonstrate that the ferryl intermediates of HRP can be used for the detection of H 2 O 2 in alcohols at mg L À1 levels using UV/Vis spectrophotometry. The red shi in the Soret band in the optical spectra of the HRP solution from 402 nm up to 418 nm upon the addition of alcohols was measured and it was shown to be reversible over time. Using this method, we can efficiently detect mg L À1 levels of H 2 O 2 impurity in alcohols, where it is barely possible using other common methods such as hydrogen peroxide test strips. The EPR spectra and CV results conrm the formation and spontaneous decay of ferryl intermediates upon the reaction of ferric state of HRP with H 2 O 2 . We successfully detected an adventitious amount of H 2 O 2 in alcohols, such as ethanol, glycerol, 2chloroethanol, and isopropanol. This demonstration suggests a simple, cost-effective, and accurate method for the detection of ultratrace amount of H 2 O 2 impurity in alcohols using UV/Vis spectrophotometry, which enables the use of this method in biomedical, biological and chemical applications.
Detection of H 2 O 2 through the analysis of the ferric state of HRP and its ferryl intermediate using UV/Vis spectrophotometry Aqueous solutions of HRP at three different concentrations (1 mM, and 10 mM) were prepared in 0.1 M potassium phosphate buffer pH 6.0. Hydrogen peroxide (30% w/w) was initially diluted in ethanol to 0.01 M and then further diluted in ethanol to reaction concentrations through serial dilutions (0.001 mM, 0.01 mM, 0.1 mM, 1 mM, 10 mM, 100 mM, 1000 mM, and 10 4 mM). The HRP solution (500 mL) was treated with 500 mL of each ethanol/H 2 O 2 solution to form ferryl intermediates. The UV/Vis absorption spectra of the ferryl intermediates were collected using a UV-Vis Spectrophotometer UV-2600, Shimadzu Scien-tic Instruments/Marlborough, MA in the range of 200-800 nm. The spectra were investigated at the Soret and Q-band regions.
Analysis of compound I and II formed from the ferric state of HRP upon the addition of H 2 O 2 using UV/Vis spectrophotometry UV/Vis absorption spectra of the native HRP solution (10 mM) and the HRP solution treated with aqueous solutions of H 2 O 2 (100 mM) were obtained using UV/Vis spectrophotometry (UV-Vis Spectrophotometer UV-2600, Shimadzu Scientic Instruments/Marlborough, MA) in the scan range of 200-800 nm. The spectra were recorded immediately upon the addition of 500 mL of 100 mM H 2 O 2 to 500 mL of 10 mM HRP solution in 0.1 M potassium phosphate buffer (pH 6.0).

Analysis of the ferric state of HRP and its ferryl intermediate using cyclic voltammetry
The CV data was collected using a BASi EC Epsilon potentiostat. The reference electrode was a silver wire immersed in a saturated solution of KCl, the counter electrode consisted of a platinum wire coil with 10 cm length, and the working electrode was a glassy carbon electrode of 0.3 mm diameter. The working electrode was polished to a mirror-like nish on a pad with 0.3 mm alumina and deionized water, and then sonicated for 30 s in deionized water. The platinum counter electrode was burned with a butane ame for 30 s. The reference electrode solution was made fresh for every measurement and at the end of each experiment, a small amount of potassium ferricyanide was added as an internal reference. Anhydrous nitrogen gas was purged through the HRP solution for at least 10 minutes prior to analysis.