Redox Modulators Determine Luminol Luminescence Generated by

This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article Cite This: ACS Omega 2018, 3, 12295−12303

http://pubs.acs.org/journal/acsodf

Redox Modulators Determine Luminol Luminescence Generated by Porphyrin-Coordinated Iron and May Repress “Suicide Inactivation” Christoph Plieth* Zentrum für Biochemie und Molekularbiologie, Universität Kiel, Am Botanischen Garten 9, 24118 Kiel, Germany

ACS Omega 2018.3:12295-12303. Downloaded from pubs.acs.org by 5.62.154.150 on 10/08/18. For personal use only.

S Supporting Information *

ABSTRACT: Iron porphyrin catalysts of the luminol reaction (horseradish peroxidase, hemoglobin, cytochrome c, and hemin) interact with diverse reducing compounds. Here, it is demonstrated how the chemiluminescence yield is modulated by such interactions. The compounds accepted as substrates protect the catalyst against the “suicide inactivation” caused by high peroxide concentrations. The reducing agents not accepted by the catalyst inhibit light production either by generating a futile redox cycle of the luminophore or by irreversibly inactivating the catalytic center. In the case of a futile cycle, light emission resumes as soon as the reducing agents in the reaction are consumed, whereas with an irreversible inactivation, light emission does not recover. The characteristics of luminescence enhancement and quenching depending on interfering agents are also reported here. They reveal details about the relative redox potentials of the involved compounds. It is discussed how this should be considered when the luminol reaction is used for quantitative analyses and when unpurified samples with a broad compound matrix are to be assayed.

■

■

INTRODUCTION

RESULTS Luminol Protects Heme Catalysts from PeroxideInduced “Suicide Inactivation”. Luminol (LH−) delivers electrons to heme and thereby prevents the over-oxidation and disintegration of the catalytic center by H2O2. This protective effect of luminol can be verified when the dependency of luminescence on [H2O2] is studied at various luminol concentrations. The data (Figure 1) clearly show that the inhibition of the horseradish peroxidase (HRP)-catalyzed luminol reaction by H2O2 shifts toward higher [H2O2] when [LH−] is increased. The effective concentrations for 50% inhibition (IC50) were determined for each [LH−] by sigmoidal four-parametric logistic fit of the data (Supporting Information, Table S1.1). This shows that as more luminol is consumed, the “suicide inactivation” effect is shifted more toward higher [H2O2]. Doubling [LH−] roughly doubled IC50. Similar results were obtained with Cyt c (Supporting Information, Figure S1). Reducing Agents Quench Luminescence and Exert a Delay Effect on the Light-Yielding Reaction. Because luminol protects the catalysts from “suicide inactivation” (Figure 1; Supporting Information, Figure S1) and thereby exerts an antioxidative effect on the catalytic heme core, the effect of other AOs on heme destruction by high [H2O2] was tested. The most abundant AO found in vivo is glutathione (GSH). Therefore, the sensitivity of HRP to high [H2O2] was

Chemiluminescence based on the luminol reaction is used for many different applications (details in the Discussion section). The reaction mechanism is complex and needs alkaline pH to provide luminol monoanion (LH−) and involves diverse intermediates such as a radical form of luminol (L−•), azaquinone (AQ) and finally the light-emitting species aminophthalate (AP*). The reaction is most efficiently catalyzed by heme proteins. However, the light yield of the reaction decays under harsh oxidizing conditions ([H2O2] ≥ 1 mM). This “suicide inactivation” is an undesired effect when the luminol reaction is to be used with quantitative assays. Strategies have been reported to avoid “suicide inactivation” by improving the heme catalyst.1−9 However, the precise adjustment of the assay ingredients can generate a long-lasting glow-type luminescence of high intensity.10,11 This study is intended to demonstrate the diversity of compounds which are able to boost or otherwise influence the luminol reaction. It shows how important the balance of the two substrates (i.e., luminol and peroxide) is and how the light yield and luminescence kinetics are amplified or reduced by phenolic compounds, antioxidants (AOs), or reducing agents. Experiments are presented, which allow a ranking of reducing agents with respect to their redox potential relative to the substrates and catalysts of the luminol reaction. Finally, important considerations necessary for the design of sensitive and highly specific quantitative luminol-based assays are discussed. © 2018 American Chemical Society

Received: June 6, 2018 Accepted: August 14, 2018 Published: September 28, 2018 12295

DOI: 10.1021/acsomega.8b01261 ACS Omega 2018, 3, 12295−12303

ACS Omega

Article

Figure 1. Dependence of luminescence on [H2O2] and luminol concentration [LH−] with HRP as the catalyst and protection from “suicide inactivation”. (A) Luminescence, integrated over the first 15 min of the reaction, is plotted against [H2O2]. The reactions were started by injecting a mixture of the catalyst and luminol into a H2O2 solution to reach the following assay conditions: 100 mM Tris/HCl pH = 9; [HRP] = 2.5 μg/mL; [LH−] as indicated in the inset; and [H2O2] is given by the abscissa. (B) Data (ILY) were normalized by the mean of each data set to facilitate direct comparison. Data are the averages of n = 4 technical replicates. The error bars represent standard deviation (SD). SD is below the symbol size, where no error bar can be seen.

This would imply that GS• has a redox potential that is too low to abstract e− from LH−.

tested using GSH as described above. The results (Figure 2) reveal that GSH reduces the luminescence significantly but

#3 The catalyst is permanently inhibited by GSH and thus inactivated. Luminescence is thereby permanently quenched because of a decrease of the heme catalyst concentration. The first two mechanisms imply that GSH renders the catalyzed luminol reaction futile. GSH is consumed either by the catalyst (eqs 1 and 2) or by L−• (eq 3) and thus constitutes a factor that only transiently disturbs the luminol reaction, as long as GSH is present. The kinetics (Figure 3A,C) indeed reveal that the light yield is low in the presence of GSH but resumes as soon as all GSH is consumed. Both the extent of luminescence quenching during the delay period (Figure 4A, the quenching effect) and the duration of the delay period until the reaction resumes light emission depend on [GSH] (Supporting Information, Figure S2B). This effect was observed even at the nanomolar concentrations of GSH and confirms that the catalyst is not permanently affected (i.e., mechanism #3 does not apply). However, the delay effect on light production was not observed with all reducing agents and catalysts. This is because some reducing agents directly affect the catalyst and thereby decrease its light-yielding efficiency by mechanism #3 mentioned above. This is demonstrated with Cyt c (Figure 3C,D). Nicotinamide−adenine−dinucleotide disodium (NADH) is the most abundant redox cofactor in living cells and is able to affect Cyt c in the presence of peroxides.12 This results in permanent luminescence quenching by NADH (Figure 3D), and the kinetics do not exhibit the onset of delayed light yield. GSH in contrast seems not to affect Cyt c (Figure 3C). A quenching effect, be it produced by a futile cycle (mechanisms #1 and #2) or by the inactivation of the catalyst (mechanism #3), can be observed with any reducing agent and any catalyst that drives the luminol reaction. The curves (Figure 4) showing the initial degree of luminescence quenching with respect to the concentration of the reducing agent in the assay are possibly characteristic for the redox potential of the involved reducing agent relative to the potential of H2O2, catalyst, and luminol. Other tested reducing agents (ascorbate, dithiothreitol, uric acid, trolox, and sulfite) produce other characteristic inhibition curves with each catalyst (Supporting Information, Figure S3.1).

Figure 2. Dependence of luminescence on [H2O2] and GSH concentration with HRP as the catalyst. Assay conditions: 100 mM Tris/HCl pH = 9; [LH−] = 750 μM; [HRP] = 2 μg/mL; [GSH] is indicated in the inset; and the luminescence is plotted against [H2O2]. The luminescence data represent light yield (ILY) integrated over the first 15 min of the reaction. Data are the averages of n = 3 technical replicates. Error bars represent SD. SD is below the symbol size, where no error bar can be seen.

does not exert a protective effect on HRP. This can be seen directly because the peroxide-based inhibition does not shift toward higher [H2O2] (Figure 2) as was observed when [LH−] was varied (Figure 1). The IC50 values calculated from this data set (Figure 2) are presented in the Supporting Information (Table S1.2). There are three possible mechanisms which may explain the reduction of light yield (i.e., luminescence quenching) caused by GSH as shown in Figure 2: #1 The catalyst abstracts electrons from GSH to produce GS• radicals (eq 1), which lead to the oxidized form GSSG (eq 2). catalyst

GSH ⎯⎯⎯⎯⎯⎯→ GS• + H+ + e−

(1)

GS• + GS• → GSSG

(2)

#2 GSH is in competition with luminol radicals L−•, donating H+ and e− to L−•, thus recycling it back to LH−, and thereby preventing the light-yielding reaction (eq 3). 2GSH + 2L−• → GSSG + 2LH−

(3) 12296


ACS Omega

Article

Figure 3. Effect of GSH and NADH on the kinetics of the luminol reaction catalyzed by HRP and Cyt c. (A) Luminol luminescence catalyzed by HRP is quenched and delayed by GSH. It resumes when all AOs in the reaction are consumed. The lower the [GSH], the shorter is the luminescence delay time. (B) Same effect is observed with NADH as an AO. (C) GSH is also able to quench and delay luminescence when it is catalyzed by Cyt c. (D) However, when the experiment is performed with NADH, no delay effect can be seen, when Cyt c is the catalyst. Instead, luminescence is permanently quenched because Cyt c is reduced by NADH in the presence of H2O2, and thereby, the catalyst is withdrawn from the reaction. The reactions were started by injecting H2O2 into a mixture of the catalyst, luminol, and the reducing agent to reach the following assay conditions: Tris/HCl 100 mM pH = 9; [Cyt c] = 20 μg/mL; HRP 0.75 μg/mL; [LH−] = 750 μM;. [H2O2 ] = 1.1 mM; and reducing agents as indicated in the insets in micromolar.

Figure 4. Initial quenching effect of GSH and NADH on the luminol reaction catalyzed by diverse catalysts. Various concentrations of GSH and NADH were tested for their ability to reduce the light emission of the luminol reaction. The relative light yield integrated over the first minute of each reaction is plotted against the respective concentration of GSH (A) and NADH (B). Data are normalized by the luminescence obtained without AOs. The reactions were started by injecting H2O2 into a mixture of the catalyst, luminol, and the reducing agent to reach the following assay conditions: Tris/HCl 100 mM pH = 9; [LH−] = 750 μM; [H2O2] = 1.1 mM; and catalysts are indicated in the inset: [HRP] = 0.75 μg/mL; [Cyt c] = 15 μg/mL; [Hemin] = 1 μM; [Hb] = 20 μg/mL; and [FeEDDHA] = 200 μM; data are means of n = 3 technical replicates. Error bars represent SD. SD is below the symbol size where no error bar can be seen.

Enhancement of Luminescence by Aromatic Hydrogen (AH) Donors. The luminol reaction driven by a genuine peroxidase can be enhanced by a number of aromatic compounds.13−19 Such compounds are typical hydrogen donors (AH), which are preferably accepted by the peroxidase as substrates. However, at low concentrations, they do not inhibit the light-yielding reaction as AOs do but rather boost it by orders of magnitude (Figure 5) because their radical form (A•) abstracts hydrogen from luminol.18 In this way, they are able to work as redox mediators and thus amplify the turnover and oxidation power of the catalyst. It is important to note that the enhancing effect is only seen with HRP (green symbols, Figure 5). There is an optimal

concentration of enhancers giving the HRP reaction a maximum boost of light emission by a factor of 10−100 (Figure 5; Supporting Information, Figure S4.1). The luminol reaction driven by other heme catalysts (i.e., the so-called “pseudo-peroxidases”) cannot be enhanced this way. Enhancers at high concentrations inhibit the light reaction with all other catalysts tested (Figure 5). This effect is comparable with the luminescence quenching caused by AOs (Figure 4). Similar results to those in Figure 5 have been obtained with benzidine and coumarin as enhancers (Supporting Information, Figure S4.1). Enhancers Protect HRP from Peroxide-Induced “Suicide Inactivation”. Apart from boosting the reaction, 12297


ACS Omega

Article

sensitivity of the heme catalyst to H2O2, is demonstrated in Figure 1 and Supporting Information (Figure S1 and Table S1.1). Shielding is efficient as long as there is a balance between [H2O2] and [LH−] in the assay (i.e., [LH−] ≥ [H2O2]). A large excess of [H2O2] over [LH−] leads to the destruction of the heme catalyst. A similar “shielding effect” is also seen with the enhancers of the HRP-catalyzed reaction such as iodophenol (IP) or benzidine (Figure 6; Supporting Information, Figure S4.2 and Table S4.1). This suggests that, like luminol, the enhancers are also accepted by HRP as substrates. They exhibit a concentration optimum, providing a maximum enhancement of light yield (ILY). Here (Figure 5), the IP concentration for maximal enhancement of the HRPcatalyzed reaction is around 200 μM, which is in good agreement with the results reported by Bhandari et al.20 In contrast, Yakunin and Hallenbeck21 reported a maximal enhancement at [IP] = 4 mM, which might be caused by the much higher [H2O2] of 17.6 mM they used. For optimal enhancement and minimal “suicide inactivation”, luminol, enhancer, and peroxide concentrations must be well balanced. Consequently, at [H2O2] = 17.6 mM, an equivalent high enhancer concentration (4 mM) was necessary to prevent “suicide inactivation”.21 Luminol reactions catalyzed by heme compounds other than HRP (i.e., Cyt c, Hb, and Hemin) could not be enhanced by the phenolic compounds tested here. Their integrated light yield (ILY) decreases with increasing enhancer concentrations (Figure 5; Supporting Information S4.1) as it does with nonenhancing AOs (Figure 4; Supporting Information, Figure S3.1). In the following, it is discussed how this can reveal details about the redox properties of these catalysts. Luminescence Quenching and Delay Are Caused by AOs. Any reducing agent or AO can interfere with the peroxidative cycle and thereby inhibit the light-yielding luminol reaction to some extent (Figure 4; and Supporting Information, Figure S3.1). Generalizing the assumptions made for GSH above (eqs 1−3), different mechanisms of interference are imaginable (Figure 7), which depend on the relative redox potentials of the AO, L, and catalyst:

Figure 5. Effect of IP on the luminol reaction catalyzed by various catalysts. Various concentrations of IP were tested for their ability to boost the light-yielding reaction. The relative light yield of each reaction is plotted against the respective IP concentration. Only the HRP-catalyzed reaction is enhanced by a factor of about 40, when [IP] is around 200 μM. The efficiency of other catalysts cannot be enhanced by IP. The reactions were started by injecting H2O2 into a mixture of the catalyst, luminol, and IP to reach the following assay conditions: Tris/HCl 100 mM pH = 9; [LH−] = 750 μM; [H2O2] = 1.1 mM; and catalysts are indicated in the inset: [HRP] = 0.375 μg/ mL; [Cyt c] = 3.75 μg/mL; [Hemin] = 1.5 μM; [FeEDDHA] = 200 μM; and [Hb] = 20 μg/mL. Data are normalized by the luminescence obtained without IP. Data are means of n = 4 technical replicates. Error bars represent SD. SD is below the symbol size where no error bar can be seen.

the enhancers also protect the HRP from “suicide inactivation” (Figure 6). With an increasing enhancer concentration, the

#1 AOred serves as an immediate donor, delivering electrons to the catalyst and the resulting radicals (AO•) form an oxidized product AOox (Figure 7A). In this way, luminol is left unaltered as long as the peroxidative cycle is busy with AOred. This suggests that the catalyst prefers AOred and rejects LH− as a substrate, possibly because of the more negative redox potential of AOred relative to LH−. If AOred was GSH, then this scheme (Figure 7A) would resemble the activity of a GSH peroxidase.22 #2 The peroxidative cycle abstracts electrons preferably from LH− to produce L−•. However, L−• is scavenged by AOred and therefore cannot yield light (Figure 7B) because [AOred] is in excess over [L−•]. Instead, it is recycled in a futile cycle. This implies that AOred has a negative redox potential relative to L−• and that AO• is unable to exchange electrons with L•−.18 #3 AOred in combination with H2O2 inactivates the heme catalyst and thus eliminates it from the reaction cycle (Figure 7C). This may occur via a reduction of the protein by opening of disulfide bridges in combination with a reductive liberation of FeII or FeIII from the heme.12 Consequently, luminescence is quenched by the depletion of the heme catalyst with no recovery of light

Figure 6. Dependence of luminescence on [H2O2] and IP and the protection from “suicide inhibition”. [IP] is indicated in the inset, and luminescence is plotted against [H2O2] in order to demonstrate that the peroxide-induced inhibition of the reaction shifts toward higher [H2O2] with increasing [IP]. The reactions were started by injecting H2O2 into a mixture of HRP, luminol, and IP to reach the following assay conditions: Tris/HCl 100 mM pH = 9; [LH−] = 750 μM; [HRP] = 0.125 μg/mL; data are averages of n = 3 technical replicates. Error bars represent SD. SD is below the symbol size, where no error bar can be seen.

inhibitory effect shifts toward higher [H2O2]. The halfinhibitory H2O2 concentration (IC50) correlates and increases with increasing [IP] (Supporting Information, Table S4.1). This effect is similar to that seen when the luminol concentration is increased (Figure 1; Supporting Information, Table S1.1) and confirms the direct interaction of the enhancer with HRP.

■

DISCUSSION The “suicide inactivation” caused by H 2 O 2 and the antioxidative “shielding effect” of LH−, which lowers the 12298


ACS Omega

Article

Figure 7. Possible mechanisms explaining the interference of AOs with the luminol reaction catalyzed by the peroxidative cycle. The light yield of the peroxidase-catalyzed luminol reaction is reduced as long as a competitive AO (AOred) is present. AOred is turned into its oxidized form (AOox) by the peroxidative cycle and thus competes with luminol. Depending on its reduction power, AOred interferes directly and exclusively with the catalyst, leaving LH− unattached (A) or it recycles L−• back to LH− in a futile cycle (B). These scenarios (A,B) imply that light emission resumes as soon as all AOred is consumed and converted into AOox. No resumption of light emission occurs when AOred in collaboration with H2O2 inactivates the catalyst irreversibly (C).

course of the reaction (mechanisms #1 and #2; Figure 7A,B). It implies that light emission resumes when all AOred in the reaction cocktail is consumed (Figure 3A−C). However, when AOred inactivates the catalyst (mechanism #3; Figure 7C), then the reaction is quenched with no delayed resumption of light. This effect is seen when Cyt c-catalyzed luminescence is quenched by NADH (Figure 3D). It is also possible that different quenching mechanisms (Figure 7) are working simultaneously. This is evident when the luminescence quenching of the Cyt c-catalyzed reaction by GSH (pink symbols in Figure 4A) is compared with the inhibition of the HRP-catalyzed reaction (green symbols in Figure 4A). The inhibitory effect on Cyt c follows a simple sigmoid curve. Here, a single IC50 value of 41 μM could be deduced by fitting a four-parameter logistic curve to the data, which reveals an apparent single inhibition mechanism (Supporting Information, Figure S3.2A). The inhibition of HRP by GSH, in contrast, is of a biphasic sigmoidal shape, revealing two IC50 values (4.3 and 276 μM; Supporting Information, Figure S3.2B) and thus two different inhibition mechanisms occurring at different GSH concentrations. The first (IC50 = 4.3 μM) is possibly due to the role of GSH as an electron donor of the peroxidase cycle. The second (IC50 = 276 μM) may be caused by the reduction of disulfide bonds in the HRP and thus refolding of the protein. Cyt c lacks disulfide bonds, and no biphasic sigmoid is seen (Supporting Information, Figure S3.2A). In conclusion, the curves showing the degree of luminescence quenching depending on the concentration of the reducing agent (AOred) are characteristic for each catalyst (Figure 4; Supporting Information, Figures S3.1; S3.2). In connection with the time kinetics of the inhibited luminescence (Figure 3), they are able to unveil the details about both the quenching mechanisms and the relative redox potentials of the components involved (i.e., H2O2, catalyst, luminol, and AOred).

emission (Figure 3D), whereas a light emission recovery (Figure 3A−C; Supporting Information, Figure S2A) can be explained only by one of the first two mechanisms (Figure 7A,B). Although Figure 1 and Supporting Information (Figure S1) demonstrate that a reducing substrate (here LH−) can protect the heme core of the catalyst from being destroyed by H2O2, this is not generally the case for any other reducing agent (AOred). In the presence of GSH, for instance, the light yield of the reaction is drastically reduced (Figures 3A,C; 4A), but the inhibitory effect of H2O2 is not shifted toward higher [H2O2] (Figure 2; Supporting Information, Table S1.2) as is the case for luminol (Figure 1; Supporting Information, Table S1). Thus, it is reasonable to assume that mechanism #1 (Figure 7A) does not apply and that GSH does not directly interact with the catalyst but rather draws electrons from L−• to become GSSH according to mechanism #2 (eq 3 and Figure 7B). A similar mechanism of GSH oxidation has been proposed earlier for quinones other than luminol.23 When the extent of luminescence quenching is plotted against the concentration of the respective AOred (Figure 4), it is evident that there are different mechanisms by which the reducing agent affects the catalysis of the luminol reaction. GSH for instance starts to produce an effect on the Cyt ccatalyzed reaction at a concentration of about 10 μM (pink symbols in Figure 4A), whereas higher NADH concentrations (80 μM; pink symbols in Figure 4B) are needed to produce this effect. This is not only due to the different redox potentials of GSH and NADH but also depends on the inhibition mechanism (Figure 7), as can be seen in the kinetics (compare Figure 3C,D). In most cases, the addition of AOred to the luminol reaction leads to a delay of light emission (Figure 3A−C), and the delay time is dependent on the AOred concentration (Supporting Information, Figure S2B). This is in agreement with previous reports10,11,24−28 and reveals that AOred is oxidized in the 12299


ACS Omega

Article

Figure 8. Proposed “ping-pong” mechanisms of the enhanced catalytic peroxidative cycle and the subsequent luminol reaction. (A) “Normal” luminol reaction without an enhancer and (B) luminol reaction enhanced by a redox mediator (AH). This implies that AH works as an intermediate electron donor pool, which is favored by the catalyst as a substrate over LH−. The radical A• is able to join in electron exchange with LH− to produce L−• and thus multiplies the oxidation power and turnover of the catalyst. This drastically increases the light yield. Schemes are adapted from ref 10. Numbers represent equations eqs S1−S7 in the Supporting Information (Table S4.2).

Can the Catalytic Heme Core Be Protected against “Suicide Inactivation”? The inactivation of peroxidases by excess [H2O2] limits their use in many applications. Consequently, there have been many approaches to prevent “suicide inactivation”.1−9 A reasonable concentration of the reducing substrate is sufficient to limit the inactivation (Figure 1; Supporting Information, Table S1.1). The effect of LH− is also seen with Cyt c (Supporting Information, Figure S1). The oxidizing substrate H2O2 is of less harm as long as there is enough reducing substrate LH− delivering electrons to the catalyst, shielding it from over-oxidation. As mentioned above, the ratio of both substrates ([LH−]/[H2O2]) gives rise to the shielding effect, and a proper balance can achieve a long-lasting glow. Some phenolic hydrogen donors (AH) enhance the light emission when added to the luminol reaction. These so-called “enhancers”14,29 also shift the “suicide inactivation” of HRP to higher [H2O2] (Figure 6; Supporting Information, Table S4.1) as was observed with luminol (Figure 1; Supporting Information, Table S1.1). Seven reactions are crucial to understand the enhanced peroxidase luminol cycle (Supporting Information, Table S4.220,30). The enhancer mechanism is similar to the antioxidative mechanism #2 proposed above (Figure 7B), but with AH instead of LH− and LH− instead of AOred (Figure 8B). It implies that (i) AH has a more negative redox potential relative to cytochrome c peroxidase compound I (CMPI) and CMPII and (ii) LH− has a more negative redox potential relative to the oxidized product A•. This suggests the following ranking of the involved components in terms of their relative redox potentials: A• > L−• ≫ CMPI ≥ CMPII ≫ LH− > AH. The effect of enhancers on the light yield can be summarized by three intertwined mechanisms in Figure 8B:

The enhancing effect of phenolic compounds (Figure 5; Supporting Information, Figure S4.1) can only be observed with HRP. The luminol reaction catalyzed by the other heme compounds tested here could not be enhanced in this way. This confirms that HRP is a genuine peroxidase and is thus much more than just a catalytic heme embedded in an inert protein shell. Peroxidases such as HRP evolved in plants specifically to process a broad spectrum of phenolic compounds. “Pseudo”-peroxidases (e.g., Cyt c and Hb), in contrast, do not have this substrate spectrum and only resemble the catalytic action of hemin. In terms of peroxidase activity, the polypeptides of “pseudo”-peroxidases simply fulfill an “egg-cup” function to hold the catalytic porphyrincoordinated iron in place. The enhancing effect of phenol derivatives on the HRPcatalyzed reaction is prominent as long as [H2O2] < 10 mM (Figure 6; Supporting Information, Figure S4.2). At higher [H2O2], the differences between the luminescence from enhanced and nonenhanced reactions are less because of the “suicide inactivation”. The inactivation of the peroxidase activity is an inherent property of heme. Therefore, any attempts to make peroxidases less sensitive to peroxide inactivation by molecular and genetic engineering must focus on the development of means of arresting over-oxidation. This task, however, presents a dilemma. On the one hand, the heme must be shielded from too much peroxide. On the other hand, shielding must not prevent access of heme to the reducing substrate. Nevertheless, the report of the Cyt c triple mutant (N52I; Y67F; C102T) demonstrates that it is possible to solve this problem.31 Luminol Reaction and Assay DesignSpecific Considerations. Light from luminol reactions is widely used in many different applications (details in Supporting Information 5). Luminescence-based assays are nevertheless sometimes less accepted because of their alleged lack of robustness and reproducibility. This apparent disadvantage is due to the measuring principle with almost black background (i.e., large signal-to-background ratio) and the exceptional sensitivity of modern photodetectors. Even tiny variations in pH, [LH−], [H2O2], [HRP], or temperature will have an impact on the photon flux produced by the reaction. In addition, the presence of enhancers and the diversity of other compounds that are able to catalyze, boost, or otherwise influence the luminol reaction also give rise to problems. A principal goal, when optimizing luminometric assays, is to obtain a maximum period of long-lived light output (“glow”) at a minimum of effort and cost. This means that buffer

#1 The enzyme (HRP) accepts the enhancer (AH) as an immediate electron donor substrate more effectively than luminol (LH−) because AH has a more negative redox potential than LH− and because AH is more similar to the phenolic substrates encountered by the enzyme in vivo than LH−. #2 The electron transfer between A−• and LH− is much more efficient than between HRP and LH−, yielding much more light when AH is present. #3 AH suppresses the “suicide inactivation” and protects the heme core of HRP from being destroyed by H2O2 and thereby prevents a decay of light yield. 12300


ACS Omega

Article

#2 Nonsubstrate AOs or reducing agents lead to a reduction and/or a delay of luminescence (Figures 3 and 4; Supporting Information, Figures S2−S3.1).

conditions and concentrations of the catalyst, H2O2, and luminol must be appropriately balanced. In particular, the assay pH represents a dilemma because genuine peroxidases work best at pH ≤ 9 and the luminol reaction requires alkaline pH ≥ 9. This is detailed in Supporting Information 6. In addition, many more factors such as detergents,32 organic solvents such as dimethyl sulfoxide (DMSO),33,34 viscosity, ionic strength,20 halides,35 and the inadvertent presence of transition metals should also be considered in assay design and improvement of assay performance. Luminol assays with catalytic mechanisms based on free transition metals are used to quantify these elements36 or [H2O2].37 However, for [H2O2] quantification, a peroxidasebased chemistry including enhancers is preferred.38−40 For the quantification of AOs, two strategies are feasible.

#3 The HRP-catalyzed luminol reaction can be enhanced by diverse aromatic hydrogen donors (Figure 5; Supporting Information Figure S4.1). Such enhancers exhibit, like luminol, a protective antioxidative effect and prevent “suicide inactivation” (Figure 6; Supporting Information, Figure S4.2; Table S4.1).

■

1 On the one hand, the time of luminescence delay (Figure 3A−C) is a quantitative measure of antioxidative capacity in the sample (Supporting Information, Figure S2B). This requires a signal recording period which comprises the whole dark period after sample application until light emission resumes.10 Consequently, the recording period varies with the amount of AOs in the samples. This way of assaying AOs is not only time-consuming but also inconvenient to implement into automated plate reader assays. It requires appropriate sample dilution prior to assaying to meet the recording time defined with the assay.

#4 Only the luminol reaction catalyzed by the genuine peroxidase (HRP) can be enhanced. Other heme catalysts do not give rise to the enhancement of the light yield in the presence of aromatic hydrogen donors (Figure 5; Supporting Information, Figure S4.1).

CONCLUSIONS Because the luminol reaction is complex (details in Supporting Information S6) and influenced by many diverse parameters, the interpretation of luminol luminescence is difficult when the assay data are obtained from the samples containing a complex compound matrix. Thus, the design of luminometric assays requires precise knowledge about the chemistry, as well as considerations about the compound matrix in the samples to be assayed. The results presented here add to our knowledge on the intertwined interactions of substrates, redox modulators, and catalysts of the luminol reaction and thereby draw attention to factors that should be considered for optimal assay design.

2 On the other hand, the extent of luminescence quenching can be used to quantify AOs. This appears the more reasonable method because it requires only a defined short time period after sample addition and also allows the quantification of reducing agents which quench luminescence via catalyst reduction. In addition, HRP is probably not the optimal catalyst for AO assays when the samples contain enhancers (Figure 5; Supporting Information, Figure S4.1) as these compounds antagonize the quenching effect of other reducing analytes and thus lead to erroneous results. Consequently, catalysts insensitive to enhancers, such as Cyt c, Hb, or Hemin, are preferable for AO assays.25

■

EXPERIMENTAL SECTION Chemicals. Ascorbate (Sigma, #A7506), benzidine (Sigma #31614), calcium chloride (CaCl2·6H2O; Roth #T886), coumaric acid (Sigma #C9008), Cu/Zn-superoxide dismutase from bovine liver (Sigma #S8409), cytochrome c from equine heart (Sigma #C2506 and BioChemica/Fluka #30400), DMSO (Roth #4720), dithiothreitol (Roth#6908), ethylenediaminetetraacetic acid (Aldrich #E2,628-2), FeEDDHA (Duchefa #F0527), ferrous sulfate (FeIISO4·7H2O; Roth #P015 and Merck #3965), ferric nitrate (FeIIINO3·9H2O; Fluka #44949), GSH (Roth #6382), heminchlorid (Roth #7629), hemoglobin from bovine blood (Sigma #H2500), hydrochloric acid (HCl 34%; Roth #4625), hydrogen peroxide (H2O2 30%; Roth #8070 and Merck #1.08597), IP (Fluka #58020), luminol (Roth #4203), mannitol (Roth #4175), NADH (Roth #AE12), HRP (Sigma #P6140), potassium hydroxide (KOH; Roth #5658), sodium sulfite (Na2SO3; Fluka #71988), TRIS ultrapure (ICN Biomedicals #77861), trolox (KJ Ross-Petersen Aps; Denmark and Aldrich #23881), and uric acid (UA, Sigma #U2625). Stock Solutions and Buffers. 10× TRIS Stock: Tris/HCl 1 M; adjusted at 28 ± 2 °C to a desired pH; Tris working buffer: Tris/HCl 100 mM diluted 1:10 from TRIS Stock; KOH 5 M Stock in H2O; luminol stock 2 M in 5 M KOH; TriLu buffer: luminol in a working buffer diluted from luminol stock to the desired concentration; and a starter solution for the luminol reaction: H2O2 in Tris working buffer. Specific Material and Instrumentation. Luminescence was recorded using 96-well microtiter plates (MTPs) with white flat bottom (Greiner Bio-One #655075) readout by a plate reader Infinite M200 pro with injection unit (Tecan, Crailsheim, Germany).

Any luminol-based assay, working under well-defined laboratory conditions, does not necessarily prove valid when “real” samples are analyzed. “Real” samples are often neither processed nor sufficiently pure to meet the required assay conditions, and their complex compound matrix can interfere with the luminol reaction. Consequently, assay results based on the luminol reaction are notorious for misinterpretations when real samples are applied without circumspection. This is particularly problematical, when the luminol reaction is used in forensic chemistry, and the data obtained are used as evidence before a court.41−45 In such cases, several different methods46 have to be applied to substantiate the result. In summary, the following results can be listed: #1 The reducing substrate luminol delivers electrons to heme and thus prevents its disintegration by H2O2 (Figure 1; Supporting Information, Figure S1 and Table S1.1). This protective antioxidative effect is efficient as long as there is a balance between [H2O2], [LH−], and [catalyst] in the assay. “Suicide inactivation” occurs when [H2O2] exceeds [LH−] by orders of magnitude. 12301


ACS Omega

■

Methods. Luminescence recording was performed at 28 °C with the plate reader specified above. Prefilled MTPs were incubated for 1 h at 28 °C with sporadic shaking prior to recording. Reactions were started by dispensing the complementary assay compounds to prefilled wells. To study the luminol-derived light yield and its dependence on the hydrogen peroxide concentration and pH, four H2O2 dilution series (1:2) were established on an MTP. The dilution series were set up on two rows (150 μL each well). After incubation (1 h at 28 °C), the reactions were started by dispensing 50 μL of the catalyst dissolved in TriLu buffer. To unravel the effects of the reducing compounds, the MTPs were prefilled with the dilution series of the respective compound in TriLu buffer supplemented with the correct catalyst concentration (150 μL each well), and reactions were started by injecting 50 μL of H2O2 as the starter solution. Detailed assay conditions for each experiment are given in the figure legends.

■

ACKNOWLEDGMENTS

The author is grateful to Lee Shaw (UKSH; Kiel University) and Livia Saleh (IfAM; Kiel University) for critically reading the manuscript. The author thanks Sonja Vollbehr (BiMo; Kiel University) for technical assistance. BBE Moldaenke (Schwentinental) and WTSH (Kiel) generously provided the plate reader. Access to the core facilities of the BiMo/LMB of the CAU is gratefully acknowledged.

■

ABBREVIATIONS AO, antioxidant; AP, aminophthalate; AQ, azaquinone; CMP, compound; cps, counts per second; Cyt c, cytochrome c; ELY, light yield efficiency; FHW, Fenton−Haber−Weiss; Hb, hemoglobin; HRP, horseradish peroxidase; ILY, integrated light yield; LH−, luminol monoanion; MTP, microtiter plate; ROS, reactive oxygen species; SOD, superoxide dismutase; any molecular species “M” in square brackets (e.g., [M]), denotes the concentration of this substance

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.8b01261.

■

Article

REFERENCES

(1) Mao, L.; Luo, S.; Huang, Q.; Lu, J. Horseradish Peroxidase Inactivation: Heme Destruction and Influence of Polyethylene Glycol. Sci. Rep. 2013, 3, 3126. (2) Li, D.; Zhang, X.; Loni, Y.; Sunz, X. Inactivation of hemoglobin by hydrogen peroxide and protection by a reductant substrate. Life Sci. J. 2006, 3, 52−58. (3) Malomo, S. O.; Adeoye, R. I.; Babatunde, L.; Saheed, I. A.; Iniaghe, M. O.; Olorunniji, F. J. Suicide inactivation of horseradish peroxidase by excess hydrogen peroxide: The effects of reaction pH, buffer ion concentration, and redox mediation. Biokemistri 2011, 23, 124. (4) Arnao, M. B.; Acosta, M.; del Rio, J. A.; Varón, R.; GarcíaCánovas, F. A kinetic study on the suicide inactivation of peroxidase by hydrogen peroxide. Biochim. Biophys. Acta 1990, 1041, 43−47. (5) Arnao, M. B.; Acosta, M.; del Rio, J. A.; García-Cánovas, F. Inactivation of peroxidase by hydrogen peroxide and its protection by a reductant agent. Biochim. Biophys. Acta 1990, 1038, 85−89. (6) Mahmoudi, A.; Nazari, K.; Khosraneh, M.; Mohajerani, B.; Kelay, V.; Moosavi-Movahedi, A. A. Can amino acids protect horseradish peroxidase against its suicide-peroxide substrate? Enzyme Microb. Technol. 2008, 43, 329−335. (7) Nazari, K.; Mahmoudi, A.; Khodafarin, R.; Moosavi-Movahedi, A. A.; Mohebi, A. Stabilizing and suicide-peroxide protecting effect of Ni2+ on horseradish peroxidase. J. Iran. Chem. Soc. 2005, 2, 232−237. (8) Sefidbakht, Y.; Nazari, K.; Farivar, F.; Moosavi-Movahedi, Z.; Sheibani, N.; Moosavi-Movahedi, A. A. Microperoxidase-11/NH2FSM16 as a H2O2-resistant heterogeneous nanobiocatalyst: a suicideinactivation study. J. Iran. Chem. Soc. 2012, 9, 121−128. (9) Baynton, K. J.; Bewtra, J. K.; Biswas, N.; Taylor, K. E. Inactivation of horseradish peroxidase by phenol and hydrogen peroxide: a kinetic investigation. Biochim. Biophys. Acta 1994, 1206, 272−278. (10) Saleh, L.; Plieth, C. Total low-molecular-weight antioxidants as a summary parameter, quantified in biological samples by a chemiluminescence inhibition assay. Nat. Protoc. 2010, 5, 1627−1634. (11) Saleh, L.; Plieth, C. Fingerprinting antioxidative activities in plants. Plant Meth. 2009, 5, 2. (12) Misra, H. P.; Fridovich, I. A peroxide-dependent reduction of cytochrome c by NADH. Biochim. Biophys. Acta 1973, 292, 815−824. (13) Kricka, L. J.; Thorpe, G. H. G. Chemiluminescent and bioluminescent methods in analytical chemistry. A review. Analyst 1983, 108, 1274−1296. (14) Thorpe, G. H.; Kricka, L. J.; Moseley, S. B.; Whitehead, T. P. Phenols as enhancers of the chemiluminescent horseradish peroxidase-luminol-hydrogen peroxide reaction: application in lumines-

Half-inhibitory peroxide concentrations IC50 of the HRP-catalyzed luminol reaction and its dependence on the luminol concentration [LH−]; dependence of luminescence on [H2O2] and the luminol concentration [LH−] with Cytc as the catalyst; half-inhibitory peroxide concentrations IC50 of the HRP-catalyzed luminol reaction and its dependence on the GSH concentration; effect of GSH on the kinetics of the luminol reaction catalyzed by HRP; comparison of the initial quenching effects of reducing agents on the luminol reaction catalyzed by diverse catalysts; quenching effect of GSH on the luminol reaction catalyzed by Cytc and HRP; dependence of half-inhibitory H2O2 concentrations IC50 on the IP concentration [IP]; effect of benzidine and coumarin on the luminol reaction catalyzed by diverse catalysts; dependence of luminescence on [H2O2] in the presence of benzidine; reaction equations representing the mechanism of the enhanced catalytic peroxidative cycle and the luminol reaction catalyzed by it; luminolbased quantitative assays; summary of the luminol reaction; structures of luminol, intermediates, and the product of the reaction; light generation by the luminol reaction: the general principle; and catalytic peroxidative cycle and the luminol reaction catalyzed by it (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: #49 431 880 3888. Fax: #49 431 880 4929. ORCID

Christoph Plieth: 0000-0002-4230-7620 Author Contributions

C.P. conceived of the study, performed the experiments, evaluated the data, and wrote the article. Notes

The author declares no competing financial interest. 12302


ACS Omega

Article

cence-monitored enzyme immunoassays. Clin. Chem. 1985, 31, 1335−1341. (15) Kricka, L. J.; Thorpe, G. H. G.; Stott, R. A. W. Enhanced chemiluminescence enzyme immunoassay. Pure Appl. Chem. 1987, 59, 651−654. (16) Díaz, A. N.; Sánchez, F. G.; García, J. A. G. Chemical indicators as enhancers of the chemiluminescent luminol-H2O2-horseradish peroxidase reaction. J. Photochem. Photobiol., A 1995, 87, 99−103. (17) Yang, L.; Jin, M.; Du, P.; Chen, G.; Zhang, C.; Wang, J.; Jin, F.; Shao, H.; She, Y.; Wang, S.; Zheng, L.; Wang, J. Study on Enhancement Principle and Stabilization for the Luminol-H2O2HRP Chemiluminescence System. PLoS One 2015, 10, No. e0131193. (18) Díaz, A. N.; Sánchez, F. G.; Garcia, J. A. G. Phenol derivatives as enhancers and inhibitors of luminol-H2O2-horseradish peroxidase chemiluminescence. J. Biolumin. Chemilumin. 1998, 13, 75−84. (19) Easton, P. M.; Simmonds, A. C.; Rakishev, A.; Egorov, A. M.; Candeias, L. P. Quantitative Model of the Enhancement of Peroxidase-Induced Luminol Luminescence. J. Am. Chem. Soc. 1996, 118, 6619−6624. (20) Bhandari, A.; Kim, W.; Hohn, K. Luminol-based enhanced chemiluminescence assay for quantification of peroxidase and hydrogen peroxide in aqueous solutions: Effect of reagent pH and ionic strength. J. Environ. Eng. 2010, 136, 1147−1152. (21) Yakunin, A. F.; Hallenbeck, P. C. A Luminol/Iodophenol Chemiluminescent Detection System for Western Immunoblots. Anal. Biochem. 1998, 258, 146−149. (22) Brigelius-Flohé, R.; Maiorino, M. Glutathione peroxidases. Biochim. Biophys. Acta 2013, 1830, 3289−3303. (23) Butler, J.; Hoey, B. M. Reactions of glutathione and glutathione radicals with benzoquinones. Free Radic. Biol. Med. 1992, 12, 337− 345. (24) de Oliveira, S.; de Souza, G. A.; Eckert, C. R.; Silva, T. A.; Sobral, E. S.; Fávero, O. A.; Ferreira, M. J. P.; Romoff, P.; Baader, W. J. Evaluation of antiradical assays used in determining the antioxidant capacity of pure compounds and plant extracts. Quim. Nova 2014, 37, 497−503. (25) Bastos, E. L.; Romoff, P.; Eckert, C. R.; Baader, W. J. Evaluation of Antiradical Capacity by H2O2−Hemin-Induced Luminol Chemiluminescence. J. Agric. Food Chem. 2003, 51, 7481−7488. (26) Saha, T. K.; Karmaker, S.; Tamagake, K. Transient formation of the oxo-iron(IV) porphyrin radical cation during the reaction of iron(III) tetrakis-5,10,15,20-(N-methyl-4-pyridyl)porphyrin with hydrogen peroxide in aqueous solution. Luminescence 2003, 18, 162− 172. (27) Girotti, S.; Fini, F.; Bolelli, L.; Savini, L.; Sartini, E.; Arfelli, G. Chemiluminescent determination of antioxidant capacity of beverages. Ital. J. Food Sci. 2002, 2, 113−122. (28) Wong, J. K.; Salin, M. L. Quenching of peroxidized luminol chemiluminescence by reduced pyridine nucleotides. Photochem. Photobiol. 1981, 33, 737−740. (29) Chen, G.; Jin, M.; Du, P.; Zhang, C.; Cui, X.; Zhang, Y.; Wang, J.; Jin, F.; She, Y.; Shao, H.; Wang, S.; Zheng, L. A review of enhancers for chemiluminescence enzyme immunoassay. Food Agric. Immunol. 2017, 28, 315−327. (30) Tao, X.; Wang, W.; Wang, Z.; Cao, X.; Zhu, J.; Niu, L.; Wu, X.; Jiang, H.; Shen, J. Development of a highly sensitive chemiluminescence enzyme immunoassay using enhanced luminol as substrate. Luminescence 2013, 29, 301−306. (31) Villegas, J. A.; Mauk, A. G.; Vazquez-Duhalt, R. A cytochrome c variant resistant to heme degradation by hydrogen peroxide. Chem. Biol. 2000, 7, 237−244. (32) Motsenbocker, M. A.; Oda, K.; Ichimori, Y. Enhancers of the non-peroxidative metal porphyrin chemiluminescence reaction. J. Biolumin. Chemilumin. 1994, 9, 7−13. (33) Gorsuch, J. D.; Hercules, D. M. Studies on the chemiluminescence of luminol in dimethylsulfoxide and dimethylsulfoxide-water mixtures. Photochem. Photobiol. 1972, 15, 567−583.

(34) Ikariyama, Y.; Aizawa, M.; Suzuki, S. Solvent effects of dimethyl sulfoxide on the chemiluminescent reaction of luminol-H2O2 system catalyzed by horseradish peroxidase. J. Mol. Catal. 1985, 31, 39−48. (35) Bause, D. E.; Patterson, H. H. Enhancement of luminol chemiluminescence with halide ions. Anal. Chem. 1979, 51, 2288− 2289. (36) Högbom, M.; Ericsson, U. B.; Lam, R.; Bakali H, M. A.; Kuznetsova, E.; Nordlund, P.; Zamble, D. B. A high throughput method for the detection of metalloproteins on a microgram scale. Mol. Cell. Proteomics 2005, 4, 827−834. (37) Pérez, F. J.; Rubio, S. An improved chemiluminescence method for hydrogen peroxide determination in plant tissues. Plant Growth Regul. 2006, 48, 89−95. (38) Díaz, A. N.; Sanchez, F. G.; García, J. A. G. Hydrogen peroxide assay by using enhanced chemiluminescence of the luminol-H2O2horseradish peroxidase system: Comparative studies. Anal. Chim. Acta 1996, 327, 161−165. (39) Warm, E.; Laties, G. G. Quantification of hydrogen peroxide in plant extracts by the chemiluminescence reaction with luminol. Phytochemistry 1982, 21, 827−831. (40) Yang, X.; Guo, Y.; Mei, Z. Chemiluminescent determination of H 2 O 2 using 4-(1,2,4-triazol-1-yl)phenol as an enhancer based on the immobilization of horseradish peroxidase onto magnetic beads. Anal. Biochem. 2009, 393, 56−61. (41) Stoica, B. A.; Bunescu, S.; Neamtu, A.; Bulgaru-Iliescu, D.; Foia, L.; Botnariu, E. G. Improving Luminol Blood Detection in Forensics. J. Forensic Sci. 2016, 61, 1331−1336. (42) Castello, A.; Frances, F.; Verdu, F. Bleach interference in forensic luminol tests on porous surfaces: More about the drying time effect. Talanta 2009, 77, 1555−1557. (43) Nilsson, A. The Forensic Luminol Test for Blood: Unwanted Interference and the Effect on Subsequent Analysis; Linkö ping University: Linköping, Sweden, 2006; pp 1−12. (44) Creamer, J. I.; Quickenden, T. I.; Apanah, M. V.; Kerr, K. A.; Robertson, P. A comprehensive experimental study of industrial, domestic and environmental interferences with the forensic luminol test for blood. Luminescence 2003, 18, 193−198. (45) King, R.; Miskelly, G. The inhibition by amines and amino acids of bleach-induced luminol chemiluminescence during forensic screening for blood. Talanta 2005, 67, 345−353. (46) Webb, J. L.; Creamer, J. I.; Quickenden, T. I. A comparison of the presumptive luminol test for blood with four non-chemiluminescent forensic techniques. Luminescence 2006, 21, 214−220.

12303


Redox Modulators Determine Luminol Luminescence Generated by

Recommend Documents