Nitrite and nitroso compounds can serve as specific catalase inhibitors

Abstract

Objective: We present evidence that nitrite and nitrosothiols, nitrosoamines and non-heme dinitrosyl iron complexes can reversibly inhibit catalase with equal effectiveness.

Methods: Catalase activity was evaluated by the permanganatometric and calorimetric assays.

Results: This inhibition is not the result of chemical transformations of these compounds to a single inhibitor, as well as it is not the result of NO release from these substances (as NO traps have no effect on the extent of inhibition). It was found that chloride and bromide in concentration above 80 mM and thiocyanate in concentration above 20 μM enhance catalase inhibition by nitrite and the nitroso compounds more than 100 times. The inhibition degree in this case is comparable with that induced by azide.

Discussion: We propose that the direct catalase inhibitor is a positively charged NO-group. This group acquires a positive charge in the active center of enzyme by interaction of nitrite or nitroso compounds with some enzyme groups. Halides and thiocyanate protect the NO⁺ group from hydration and thus increase its inhibition effect. It is probable that a comparatively low chloride concentration in many cells is the main factor to protect catalase from inhibition by nitrite and nitroso compounds.

Introduction

The phenomenon of catalase inhibition by nitric oxide (NO) is well known.^1,2 It was shown that the rate of hydrogen peroxide decomposition by catalase decreased in the presence of nitrite (the product of NO oxidation). This effect was explained as the result of competition between nitrite and hydrogen peroxide for catalase compound I or the result of catalase inhibition by trace amounts of nitric oxide.^3,4 But ability of nitrite to produce compounds with ferric hemoproteins was demonstrated by means of spectrophotometric and by ESR⁵ assays.

We found that nitrite ability to reduce the rate of H₂O₂ decomposition by catalase can be increased 100 fold in the presence of chloride, bromide or thiocyanate.^6–8 Later this phenomenon was confirmed by other researchers.⁹ The enzymatic sensors for nitrite detection based on the nitrite interaction with catalase were offered.^10,11

Later it was found that not only nitrite, but S nitrosothiols (RSNO) and dinitrosyl iron complexes (DNIC) possess a similar ability to decrease the rate of H₂O₂ decomposition by catalase. Halide ions and thiocyanate also increase the effect.¹²

In this case two questions arise: (1) Why this so different substances possess a similar property to decrease the catalase ability to destruct H₂O₂? (2) Catalase is known to be an intracellular enzyme. The concentration of nitroso compounds in the cell can reach the 10th of micromoles.¹³ In this case the catalase activity may be decreased by 30% and more.¹² How the enzyme can normally function in these conditions?

The purpose of this paper is to elucidate the mechanism of the catalase activity inhibition by nitrite and nitroso compounds as well as the nature of an inhibitor.

Materials and methods

Materials

The following chemicals were used in the experiments: sodium phosphate monobasic, sodium chloride, potassium bromide, potassium thiocyanate, hydrogen peroxide, iron(II) sulfate, potassium nitrite, o-phenanthroline (o-Ph), sodium azide, catalase from bovine liver, N-nitrosoethanolamine, N-nitroso-di-n-propylamine, hemoglobin (Sigma, USA), sulfanilic acid, N-(1-naphthyl)-ethylenediamine dihydrochloride (NED), 1,4-dithiothreitol (DTT), glutathione, cysteine (Merck, Germany).

Dinitrosyl iron complexes with glutathione (DNIC/GSH) were synthesized by reaction of the nitric oxide with 5 mM solution of ferrous sulfate in distilled water followed by the addition (in an atmosphere of NO) of glutathione in 15 mM Hepes, pH 7.4 (at a Fe²⁺:glutathione ratio 1:2), according to the protocol described earlier.¹⁴ Initially in NO atmosphere the DNIC yield was 100%.¹⁴ The concentration of forming DNIC was close to the concentration of the ferrous iron initially present in the medium. The DNIC concentration was controlled by spectrophotometric assay assuming ε₃₁₀ = 3000 M^–1 cm^–1.¹²

S-nitrosoglutathione (GSNO) and S-nitrosocysteine (CysNO) were prepared by incubation of 0.1 M nitrite with 0.1 M glutathione/cysteine and 0.1 M HCl (1:1:2, v/v) for 15 minutes followed by the dilution with 40 mM sodium phosphate buffer, pH 6.0, to a required concentration. Nitrosothiols concentration were evaluated by spectrophotometric assay assuming ε₃₄₀ = 930 M^–1 cm^–1.¹⁵

Methods

Nitrite concentration was measured by the Griess assay.¹⁶

Determination of catalase activity

Catalase activity was evaluated by the permanganatometric and calorimetric assays described earlier.^6,12,17 The permanganatometric assay is based on the determination of H₂O₂ concentration in the samples taken from the reaction medium. The catalase reaction was initiated by H₂O₂ addition to the reaction medium in a final concentration of 10 mM. The enzyme activity was determined by measurement of the H₂O₂ concentration decrease in 10 seconds after the beginning of experiment.⁶

The permanganatometric assay do not allow to monitor kinetics of the reaction, because we cannot correctly monitor the initial linear segment of the kinetics which is necessary for a precise evaluation of the enzymatic activity. The turbidity of the sample, its color, the presence of various reducing agents bleaching permanganate limit the ability of this assay. The calorimetric assay has no these shortcomings. This assay is based on monitoring of the heat production of H₂O₂ decomposition in the catalase-mediated reaction.^6,12,17 The heat effect of H₂O₂ decomposition to water and oxygen is very high (23.6 kcal M⁻¹ of H₂O₂). The heat effects of other processes that can take place in this system are negligible as compared to that of catalyzed by catalase.¹⁷ We created an experimental setup on the basis of a Dithermanal calorimeter (Hungary). The setup contained two cells – for control and test samples, containing highly sensitive thermistors (to 0.001°C) immersed into a continuously stirred medium. Both thermistors were switched into the electric bridge measuring circuits. The difference of electric potential between bridges was increased by amplifier and then recorded by the computer terminal.

The process was initiated by the addition of 10 mM H₂O₂ to the test cell.¹⁷ The experimental setup enabled to monitor the continuous kinetics of hydrogen peroxide destruction by catalase.¹⁷ We had proven experimentally linear relationship between initial peroxide concentration and the value of temperature increasing in the concentration range from 2 to 20 mM. The kinetics of heat production completely corresponded to the kinetics of H₂O₂ decomposition in the reaction medium.¹⁷ The kinetic measurement time was from 20 to 60 seconds. The transient response of the device is 2–3 seconds.

The catalase activity was estimated by measuring the slope of the initial segment of the kinetics. Its value is directly proportional to the active enzyme concentration in the range of 2–15 nM.^6,12,17 The calorimetric method is less inertial than manometry and allows to distinguish different pathways of H₂O₂ destruction since the values of their enthalpy are different.^6,12,17

The results were statistically processed by BioStat software. The data were presented as the mean ± standard deviation.

Results

Comparison of inhibiting effects of nitrite, S-nitrosothiols and DNIC

The presence of nitrite, DNIC/GSH, and S-nitrosoglutathione induce the decrease of the rate of H₂O₂ decomposition by catalase (Table 1). The value of the decrease remained unchanged after subsequent additions of H₂O₂. The concentration of nitrite did not change after H₂O₂ additions (Table 1). These observations mean that decrease is result of the enzyme inhibition but no competition of some substance with peroxide for catalase.

Table 1. The effect of nitrite, DNIC/GSH and GSNO on catalase activity as measured by the permanganatometric (Perm.) and calorimetric (Cal.) assays.

	Activity (% to control)
	1st addition of Н2О2		2nd addition of Н2О2		3rd addition of Н2О2
Substance	Perm.	Cal.	Perm.	Cal.	Perm.	Cal.
KNO2	19.0 ± 1.8	24.2 ± 3.2	18.3 ± 1.2	23.1 ± 2.9	18.0 ± 1.5	23.2 ± 2.9
DNIC/GSH	58.2 ± 2.5	64.5 ± 5.2	58.1 ± 2.8	61.2 ± 3.2	57.9 ± 3.2	62.8 ± 3.8
GSNO	52.1 ± 3.2	58.3 ± 4.1	52.3 ± 3.8	56.3 ± 3.1	51.8 ± 3.1	57.5 ± 4.1

Note: Reaction medium contained: 40 mM phosphate buffer, 150 mM NaCl, 9.0 nM catalase, рН 6.5 and also tested substances in concentration 2.0 µM. The catalase reaction was initiated by addition of 10.0 mM Н₂О₂ into the reaction medium. Every subsequent addition of 10.0 mM Н₂О₂ was made after a complete destruction of previously added Н₂О₂. The completeness of H₂O₂ destruction was tested by permanganatometric assay. Catalase activity of the sample with a similar content but without nitroso compounds was taken as 100%.

According to Griess test, the nitrite concentration was not changed reliably after three additions of hydrogen peroxide.

There were no significant changes in catalase-inhibiting efficiency of GSNO, DNIC/GSH within 1 hour after their synthesis and of nitrite within 1 hour after the reagent dissolving in water.

The reciprocal of the inhibitor concentration induce 30% decrease of enzyme activity was taken as a measure of the inhibition effectiveness. According to the data presented in Fig. 1A and B, nitrite, GSNO, and DNIC/GSH demonstrate almost the same inhibition effectiveness in the range of 0–35% decreasing of the enzyme activity. The further increasing of the substance concentration showed an increased inhibition rate for nitrite but no significant increase for DNIC and GSNO (Fig. 1A and B).

Figure 1.

The dependence of catalase activity (А, %) on the concentration of nitrite and nitroso compounds. The reaction medium contained 40 mM phosphate buffer, 150 mM NaCl, 9.0 nM catalase, рН 6.0 and also: (A) 1 – nitrite, 2 – GSNO, 3 – DNIC/GSH. (B) Samples 1′, 2′ and 3′ are similar to 1, 2 and 3, but NaCl was removed. Replacing of 150 mM NaCl for KBr (in the same concentration) or 30 µM KSCN gave no difference in nitrite and the nitroso compound effect. The catalase activity of the samples without nitrite and nitroso compounds was taken as 100%.

Table 2 presents the dependences of inhibition effectiveness on pH for all these substances. The dependences are very similar within the pH range from 6.0 to 7.2 and show the increase of inhibition at pH decreasing.

Table 2. The concentration of nitrite, nitroso compounds and azide (µM) induce a 30% decrease of catalase activity at different pH. Influence of chloride*.

	рН**
Substance	6.0	6.5	7.2
Nitrite	0.13 ± 0.01 (15.3 ± 1.5)	0.61 ± 0.04 (85.4 ± 4.5)	7.3 ± 0.4 (740.0 ± 35.3)
DNIC/GSH	0.13 ± 0.01 (16.2 ± 1.8)	0.61 ± 0.03 (94.2 ± 5.4)	7.3 ± 0.5 (785.3 ± 40.5)
GSNO	0.14 ± 0.01 (16.4 ± 1.4)	0.59 ± 0.04 (91.3 ± 4.8)	7.2 ± 0.4 (790.0 ± 45.0)
Azide	0.04 ± 0.001 (0.04 ± 0.001)	0.06 ± 0.002 (0.06 ± 0.003)	0.10 ± 0.004 (0.10 ± 0.003)

*The reaction medium in all the samples contained 40 mM phosphate buffer, 150 mM NaCl, 9.0 nM catalase. The catalase reaction was initiated by the addition of 10.0 mM Н₂О₂ into the reaction medium.

**Data for the samples without NaCl are presented in brackets.

The next specific characteristic of catalase inhibition by nitrite and nitroso compounds is dependence of its inhibition efficiency on halide and thiocyanate concentration (Fig. 1A and B, Fig. 2A–C, Table 2). The presence of chloride or bromide in a concentration above 80 mM or thiocyanate in a concentration above 20 μM increased the inhibition efficiency by two orders of magnitude (Fig. 2A–C). The dependences of inhibition degree on halide concentration are very similar for nitrite, GSNO and DNIC/GSH and almost do not depend on pH in the range of 6.0–7.2 (Table 2, Fig. 2А) and on the inhibitor concentration (Fig. 2B and C). In the presence of halides, the inhibiting effect of nitroso compounds at pH 6.0 is close to that of azide. But the inhibiting effect of azide did not depend on the presence of halide (Table 2).

Figure 2.

The effect of halide ions and thiocyanate on catalase activity (A, %) in the presence of nitrite, DNIC/GSH and S-nitrosoglutathione. (А) 1 – KNO₂, 2 – DNIC/GSH, 3 – S-nitrosoglutathione, all in a concentration of 0.15 µM, рН 6.0. 4–6 are similar to 1–3, but pH was shifted to 7.2 and the concentration of nitrite and nitroso compounds was 9.0 µM. (B) 1–250 µM KNO₂ + NaCl; 2–12.5 µM KNO₂ + NaCl; 3 and 4 are similar to 1 and 2, but contain KBr instead of NaCl; 5–8 – 250 µM KNO₂ + KH₂PO₄, Na₂SO₄, CH₃COOK, C₆O₇H₅K₃ instead of NaCl, accordingly, рН 7.2. (C) KSCN + KNO₂ (µM): 1–12.5; 2–37.5; 3–250.0, рН 7.2. The reaction medium in all samples contained 40 mM phosphate buffer, 9.0 nM catalase. The catalase activity of the sample without nitroso compounds was taken as 100%.

Reversibility of inhibition

DNIC/GSH loses the ability to inhibit catalase in the presence of hemoglobin (NO(NO⁺) trap) and o-phenanthroline (iron chelator). This effect is independent on the time of o-phenanthroline addition: before the first or subsequent additions of H₂O₂ (Fig. 3A and B). In the absence of hemoglobin or iron chelators, the loss of ability to inhibit was not observed (Table 3).

Figure 3.

The effect of hemoglobin and o-phenanthroline on the ability of DNIC/GSH (А, B) and nitrite (C) to decrease catalase activity. The following reagents – DNIC/GSH – 5.0 µМ; KNO₂ – 0.15 µМ; o-phenanthroline (о-ph) – 0.5 mM were added to the reaction medium 1 minute before initiating the catalase reaction. It was initiated by the addition of 10.0 mM Н₂О₂ into the reaction medium. Every subsequent addition of 10.0 mM Н₂О₂ was made after a complete destruction of previously added Н₂О₂. The reaction medium in all cases contained 40 mM phosphate buffer, 150 mM NaCl, 9.0 nM catalase, 200 µМ hemoglobin, pH 6.0.

Table 3. Specific features of catalase inhibition by nitrite and some nitroso compounds.

	Added reagents
Tested compounds	С	GSH	HbО2	o-ph	HbО2 + о-ph	о-ph + HbО2
Nitrite, N-nitroso – ethanolamine, N-nitroso-di-n-propilamine	+	+	+	+	+	+
Same + Fe + Glu	+	+	+	+	+	+
DNIC/GSH	+	+	+	+	−	+
GSNO, CysNO	+	+	+	+	+	+
Same + Fe + Glu	+	+	+	+	−	+

Note: ‘Plus’ and ‘minus’ – the inhibiting effect is present (decrease of activity by 35–40% as compared to the initial one) or absent (no change of the enzyme activity), respectively. Concentration of all tested compounds in the reaction mixture – 0.5 µM; FeSO₄ – 100 µM; GSH (glutathione) – 100 µM; o-phenanthroline (o-ph) – 0.5 mM; HbO₂ – 100 µM. The reaction medium in all cases contained: 40 mM phosphate buffer, 150 mM NaCl, 9.0 nM catalase, pH 6.0.

To 5.0 µM solutions of the tested compounds in 40 mM phosphate buffer, pH 6.0, we added the reagents in the indicated sequence with an interval of 1 minutes. After 5 minutes the incubated substances were transferred into the reaction medium and therefore diluted by 10 times. Catalase activity in the reaction medium was measured calorimetrically.

Glutathione, cysteine and DТТ in a concentration of 10 mM did not affect catalase inhibition.

In the absence of NaCl the effectiveness of the inhibition by all compounds tested was lower by two orders of magnitude and all effects of added reagents were kept.

Unlike DNIC/GSH, nitrite and RSNO didn't lose ability to inhibit in the presence of hemoglobin and o-phenanthroline (Fig. 3C, Table 3). But the ability of RSNO to inhibit catalase was lost when ferrous iron was added in reaction medium prior to hemoglobin and o-phenanthroline. Nitrite and nitroso amines maintained their inhibiting ability in those conditions (Table 3).

Discussion

The cause of the catalase activity decrease in the presence of nitrite and nitroso compounds

The decrease of the catalase activity in the presence of nitrite, RSNO, and DNIC/GSH was demonstrated by the permanganatometry and by calorimetric method (Table 1). The inhibition efficiency showed no changes after several subsequent H₂O₂ additions (Table 1). Therefore, the inhibitor was not accumulated and consumed in the catalase–hydrogen peroxide system.

The enthalpy of H₂O₂ decomposition did not change in the presence of nitrite and the nitroso compounds.^8,12 This is another confirmation that the essence of the reaction did not change because the peroxidase process has another enthalpy than the catalase one.^8,12

The loss of DNIC/GSH inhibiting effect in the system containing NO(NO⁺) trap and iron chelator (Fig. 3A–C, Table 3) may be the result of the complex destruction by the iron chelator and binding of NO by hemoglobin. RSNO can be transformed into DNIC after ferrous cation addition¹⁴ and acquire all its properties (Table 3). Nitrite ineffectively produces iron nitrosyl complexes in a neutral or near neutral medium, as compared with RSNO.¹⁸ This is apparently the cause of maintenance of its inhibiting effect (Fig. 3C, Table 3).

Therefore, catalase enzymatic activity is recovered after the DNIC or RSNO chemical transformations. Thus, these nitroso compounds induce the reversible enzyme's inhibition. The reversibility of catalase inhibition by nitrite was shown earlier by dilution of the reaction medium⁶ and by nitrite oxidation by lactoperoxidase.⁸ Nitrite oxidation by lactoperoxidase in system catalase – lactoperoxidase – nitrite recovered the catalase activity.⁸

The essence of direct inhibitor

An almost equal efficiency of the inhibition by nitrite, GSNO and DNIC/GSH, enhancement of the inhibition efficiency by chloride, bromide and thiocyanate, and a similar dependence of the inhibition degree on pH allow to suggest a similar inhibition mechanism in all cases. Two explanations can be given for this similarity. First, all these substances produce a common intermediate that is inhibitor. Second, all these substances contain the same group responsible for inhibition.

All the studied compounds retained the inhibition activity in the presence of hemoglobin and low molecular weight thiols (Table 3). This fact proves that the catalase inhibition is not based on NO (NO⁺) release from the substances.

The disappearance of DNIC/GSH inhibiting effect in the system containing NO trap and iron chelator and the absence of this phenomenon in the cases of nitrite and nitrosoamines (Fig. 3A–C, Table 3) testify to the absence of chemical transformation of all these substances to a single product.

All the substances listed in Table 3 contain a nitroso group. It can be supposed that the inhibiting effect is based on the presence of this group.

The next question is how to explain the enhancement of the inhibition effect by halides and thiocyanate. It cannot be explained as the effect of ionic strength because the salts containing no halides do not induce this enhancement (Fig. 2B). Halide ions and thiocyanate are known to potentiate nitrosation by producing complexes such as NOCl, NOBr, or NOSCN.^19,20 Thus, NO group must have a positive charge to bind with a negatively charged group of halides or thiocyanate. NO groups of S-nitrosothiols and DNIC/SH have some positive charge.¹⁸ Nitrous acid (HNO₂) in acid medium can produce a nitrosation agent.^19,20

Similar inhibiting efficiency of DNIC (two NO groups) and RSNO (one NO group) have two possible explanations. First one is that only one NO group from quite large molecule can interact with corresponding unit of the enzyme. Second one is related to the suggestion that not all NO groups of a complex can bear positive charge. Some schemes presented earlier show that there is only one positively charged NO group in DNIC molecule.¹⁸

Nevertheless, the analysis of the curves presented in Fig. 2A–C makes us doubt that the inhibitor is a nitroso halide complex produced in the reaction medium. All experimental dependences of the enzyme activity on the halide concentration reach the plateau at the same concentrations of halides independent on the concentration of the inhibitor (80 mM for chloride or bromide and 20 μM for thiocyanate) and pH (Fig. 2A–C). In the case of thiocyanate, the plateau is reached at its concentration much less than that of nitrite (Fig. 2C). Therefore, the plateau is not the result of complete nitrite transformation to NOSCN.

We can suppose that some NO⁺..E complex is produced (E means a catalase subunit). According to this supposition, the existence of the plateau on the curves in Fig. 2A–C is the result of a complete transformation of NO⁺..Е complexes to E.. NOCl (E..NOBr, E..NOSCN).

The mechanism of inhibition

Since inhibition is reversible, the nitroso group should be removed from the enzyme by some NO(NO⁺) traps in the reaction medium or produce nitrous acid and nitrite after hydration.

But these traps (DTT, glutathione, cysteine, hemoglobin) did not affect the inhibition (Table 3). DNIC and RSNO were not transformed into nitrite in the catalase–peroxide system (Fig. 3A–C). It can be supposed that catalase inhibition is the result of transnitrosation between a nitroso compound and some apoenzyme structure. Nitrite can be transformed into nitrous acid by interaction with some protonated apoenzyme groups. Nitrous acid produces a nitrosating agent

$$\mathrm{E}{\mathrm{H}}^{+}+\mathrm{ON}{\mathrm{O}}^{-}\leftrightarrow \mathrm{N}{\mathrm{O}}^{+}..\mathrm{E}+\mathrm{O}{\mathrm{H}}^{-}$$ (1)

In the case of DNIC and RSNO (L⁻NO⁺) the equation can be presented as follows:

$$\mathrm{E}{\mathrm{H}}^{+}+{\mathrm{L}}^{-}\mathrm{N}{\mathrm{O}}^{+}\leftrightarrow \mathrm{N}{\mathrm{O}}^{+}..\mathrm{E}+{\mathrm{L}}^{-}{\mathrm{H}}^{+}$$ (2)

The dynamic equilibrium between an enzyme (protein macromolecule) and a low molecular weight nitroso compound can be established as in the system described by Vanin et al.¹⁸ The lifetime of NO⁺..Е compound can depend on pH, because NO⁺ hydration is less possible at lower pH. Increase of the inhibition efficiency by chloride, bromide and thiocyanate can be the result of protection of NO group from hydration by production of E.. NOCl (E..NOBr, E..NOSCN). The latter process is apparently not pH dependent.

Fluoride ion makes no effect on the catalase inhibition by nitrite and nitroso compounds. Moreover, it decreases the effect of other halides and thiocyanate during inhibition.^6,8 It is proposed that fluoride competes with them, because higher concentrations of them are necessary to reach the same extent of inhibition in the presence of fluoride.^6,8 NO-F complex may be supposed to be more stable than that with other halides. F⁻ anion has the smallest radius among other halides, and this fluoride's effect may be the consequence of the complex stability.

According to Mori et al., ²¹ diesel exhaust particles contain a substance inhibiting catalase with similar dependence on chloride, bromide and thiocyanate. We suppose this substance (substances) can be nitrite or some other nitroso compounds produced during oxidation of nitric oxide contained in the exhaust.

The observation that large molecules (GSNO, DNIC/GSH) cannot inhibit catalase activity by 100% (see Fig. 1) can be explained in the following manner. A catalase molecule can catalyze H₂O₂ destruction if all four subunits are present in the enzyme molecule. ²² Addition of large molecules such as DNIC and S-nitrosothiols to one of the catalase subunit may weaken their contacts with the other subunits. This effect can have mechanistic or electrostatic essence.¹² Nitrite (small molecule) induces complete inhibition also in the presence of DNIC and GSNO.¹² If a complete inactivation of the enzyme requires to inhibit more than one subunit, it is probably that the addition of the big molecule to one or two subunits impedes the following inhibition.

Specificity of the inhibitory effect of nitrite and nitroso compounds on catalase

Heme-containing peroxidases are significantly more resistant to inhibition by nitrite and nitroso compounds than catalase. Halide ions have no effect on their activity in the presence of nitrite.^6–8 We suppose that this difference is due to the features of apoenzyme structure. These peroxidases and catalase contain ferri-heme in the active center.²²

It is probable that a comparatively low chloride in many cells^21,23 may be the main factor protecting catalase from inhibition by nitroso compounds.

Disclaimer statements

Contributors All authors contributed equally.

Funding None.

Conflict of interest None.

Ethics approval Not applicable.