Evaluation of the oxidation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) to toxic pyridinium cations by monoamine oxidase (MAO) enzymes and its use to search for new MAO inhibitors and protective agents

Abstract

Monoamine oxidase (MAO) enzymes catalyze the oxidative deamination of amines and neurotransmitters and inhibitors of MAO are useful as neuroprotectants. This work evaluates the human MAO-catalyzed oxidation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), a dopaminergic neurotoxin, to the directly-acting neurotoxic metabolites, 1-methyl-4-phenyl-2,3-dihydropyridinium (MPDP⁺) and 1-methyl-4-phenylpyridinium (MPP⁺) measured by High-Performance Liquid Chromatography (HPLC), and this approach is subsequently used as a new method for screening of MAO inhibitors and protective agents. Oxidation of MPTP by human MAO-B was more efficient than by MAO-A. R-Deprenyl, a known neuroprotectant, norharman (β-carboline), 5-nitroindazole and menadione (vitamin K3) inhibited MAO-B and reduced the formation of toxic pyridinium cations. Clorgyline and the β-carbolines, harman and norharman, inhibited the oxidation of MPTP by MAO-A. Cigarette smoke, as well as the naturally occurring β-carbolines (norharman and harman) isolated from smoke and coffee inhibited the oxidation of MPTP by MAO-B and/or MAO-A, suggesting protective effects against MPTP. The results show the suitability of the approach used to search for new MAO inhibitors with eventual neuroprotective activity.

Keywords::

• Monoamine oxidase (MAO)
• 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)
• pyridinium cation
• MPP+
• neurotoxins
• β-carboline
• 5-nitroindazole; menadione (2-methyl-1,4-naphtoquinone)
• tobacco smoke
• Parkinson’s disease

Introduction

Monoamine oxidase (MAO) is a flavoenzyme located at the outer membrane of mitochondria in the human brain and peripheral tissues. It catalyzes the oxidative deamination of xenobiotic amines and neurotransmitters, including dopamine, serotonin, norephinephrine, tyramine, tryptamine and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) neurotoxin¹. MAO appears as two isozymes, MAO-A and B, distinguished by substrate and inhibitor selectivities and plays an important role in the central nervous system and peripheral organs. MAO-A is involved in psychiatric conditions and depression and MAO-B in neurological disorders and diseases¹. The oxidation of biogenic amines and neurotransmitters by MAO results in the production of hydrogen peroxide (H₂O₂), oxygen radicals and aldehydes that represent risk factors for cell oxidative injury^2–4. Inhibition of MAO may protect against oxidative stress and thereof the identification of novel MAO inhibitors is a subject of interest in drug discovery^1,5,6.

Parkinson’s disease (PD) is a neurodegenerative disease associated with a progressive loss of dopaminergic neurons in the substantia nigra. The ultimate causes of PD are unknown, but the participation of environmental toxins is likely⁷. In this regard, MPTP has been of particular interest as a parkinsonism-inducing neurotoxin⁸. MPTP produces neurotoxicity in humans and monkeys, and it is currently used to generate experimental PD in animal models⁹. It crosses the blood-brain barrier and is bioactivated by mitochondrial-MAO enzymes to toxic pyridinium metabolites^8,10,11 (Figure 1), which are selectively uptaken into dopaminergic cells via dopamine activated transporter (DAT) producing inhibition of complex I in mitochondria, energy depletion and cell death¹⁰. MPDP⁺/MPP⁺ species induce oxidative stress through the generation of free radicals, including OH˙, superoxide, and NO˙, which are implicated in dopaminergic neurotoxicity^7,12,13. These biochemical features may be partly attenuated in presence of MAO inhibitors^1,4 and antioxidants¹⁴. MPTP is deactivated by Cytochrome P450 (CYP) enzymes through N-demethylation and aromatic hydroxilation and the metabolic balance between MAO and CYP influences its toxic outcome¹⁵.

Figure  1.  Bioactivation of MPTP neurotoxin by MAO enzymes with the formation of reactive oxygen species (ROS) (H₂O₂) involved in oxidative damage and pyridinium cations (MPDP⁺ and MPP⁺), which are directly-acting toxicants producing neurotoxicity. Inhibition of MAO enzymes is a target for neuroprotection.

The activity of MAO-B increases with age¹, and its elevation in the brain may result in an increased risk of neurodegeneration⁶. Thus, inhibition of MAO may afford neuroprotection through a lower production of reactive oxygen species (ROS) and aldehydes, as well as a diminished activation of toxins such as MPTP and/or related substances^2,6,16,17. Therefore, studying MAO enzymes and finding new MAO inhibitors as purported neuroprotectants is a subject of current interest in drug discovery^1,18–26. MAO enzymes are usually monitored by measuring the absorbance and/or fluorescence of the oxidation products generated from key amines (kynuramine, tryptamine, tyramine, etc)^27–33. The purpose of this research was to evaluate the oxidation of the neurotoxin MPTP by human MAO isozymes (human MAO-A and B) through the chromatographic analysis of the toxic pyridinium species (MPDP+ and MPP+) generated, and subsequently use this approach as a new tool to search for new MAO inhibitors (MAO-A and B) and eventual protective agents.

Material and methods

Recombinant human monoamine oxidase A and B were obtained from Gentest BD biosciences (Woburn, MA, USA). Enzymes were expressed in insect cells from MAO-A and MAO-B cDNA using a baculovirus expression system and were prepared as membrane protein fractions. Insect cell supersomes lacking MAO enzymes were used for controls. Pooled human liver mitochondria subcellular fraction (HLM) was obtained from Xenotech (Tebu-bio). 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) hydrochloride (caution: MPTP is a neurotoxin and should be handled with appropriate precautions), 1-methyl-4-phenyl-2,3-dihydropyridinium (MPDP⁺) perchlorate, 1-methyl-4-phenylpyridinium (MPP⁺) iodide, kynuramine, 4-hydroxyquinoline, R-deprenyl, clorgyline, 5-nitroindazole, menadione (2-methyl-1,4-napthoquinone), harman and norharman were from Sigma. HPLC grade acetonitrile, methanol and dimethyl sulfoxide (DMSO) were from Scharlau (Spain) and dichloromethane from Merck. Cigarrette and coffee extracts, as well as the isolation of β-carbolines from those samples were obtained as above^34,35 and used for inhibition studies of the oxidation of MPTP to neurotoxicant pyridinium species.

Oxidation of MPTP by monoamine oxidase A and B

Oxidation of MPTP by MAO enzymes was used to measure MAO activity and was determined following the dehydrogenation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) to give 1-methyl-4-phenyl-2,3-dihydropyridinium (MPDP⁺) and 1-methyl-4-phenylpyridinium (MPP⁺) that were identified and quantified by HPLC-DAD and electrospray mass spectrometry (ESI). MAO activity was also determined by deamination of kynuramine to give 4-hydroxyquinoline^11,15,30 analyzed by RP-HPLC-DAD (MS). For that, a reaction mixture (0.2 mL, final volume) containing membrane protein fractions of the enzyme (recombinant human MAO-A or -B) (0.01–0.05 mg/mL) in 75 mM potassium phosphate buffer (pH 7.4) were added with MPTP (0–2 mM) or kynuramine (0–0.25 mM) as substrates and incubated at 37°C for 40 min. The enzymatic reaction was stopped by addition of 2N NaOH (75 μL) and 70% perchloric acid (25 μL), subsequently centrifuged (9000 rpm, 5°C), and 20 μL of the supernatant injected into the HPLC-DAD (MS) to determine 1-methyl-4-phenyl-2,3-dihydropyridinium (MPDP⁺) and 1-methyl-4-phenylpyridinium (MPP⁺). On the other hand, reaction mixtures (0.2 mL, final volume) containing human liver mitochondria (0.2 mg/mL protein) in 75 mM potassium phosphate buffer (pH 7.4) were added with MPTP neurotoxin (0.3 mM), and incubated at 37°C for 40 min. The reaction was stopped and analyzed as above to determine 1-methyl-4-phenyl-2,3-dihydropyridinium (MPDP⁺) and 1-methyl-4-phenylpyridinium (MPP⁺) cations.

Inhibition of MPTP oxidation

To perform inhibition assays, aliquots containing inhibitory substances (0–50 µM final concentration), extracts obtained from cigarette smoke and the corresponding isolated β-carbolines were added to reaction mixtures (0.2 mL) containing MPTP and MAO, as mentioned previously. The reaction mixture in 75 mM potassium phosphate buffer (pH 7.4) was incubated at 37°C, 40 min and stopped with 2 N NaOH and perchloric acid as above. Enzyme kinetics and mechanism of inhibition were assessed by analyzing the corresponding Michaelis-Menten curves by fitting reaction velocity versus substrate concentration (0–600 μM MPTP) to non-linear regression analysis (Graphpad Prism 4.0), and also by double reciprocal Lineweaver-Burk plots obtained at different concentration of substrates and inhibitors. Reaction velocity (v) was determined as nmol of products (MPDP⁺, MPP⁺ or 4-hydroxyquinoline) per min and mg protein. Incubation reactions were performed at least in duplicate and data obtained from different experiments.

RP-HPLC chromatographic analysis and mass spectrometry

The analysis of the enzymatic reaction products: MPDP⁺, MPP⁺ and 4-hydroxyquinoline was performed by RP-HPLC with uv-DAD and fluorescence detection using an HPLC 1050 (Hewlett Packard) with a Diode Array Detector (DAD) and a 1046A-fluorescence detector. A 150 mm × 3.9 mm, 4 μm, Nova-pak C18 column (Waters, Milford, MA, USA) was used for chromatographic separation. Chromatographic conditions were: 50 mM ammonium phosphate buffer (pH 3) (buffer A) and 20% of A in acetonitrile (buffer B). Gradient programmed from 0% (100% A) to 32% B in 8 min, and 90% B at 15 min. The flow rate was 1 mL/min, the column temperature was 40°C and the injection volume was 20 μL. Absorbance detection was set at 355 nm for analysis of MPDP⁺, 280 nm for analysis of MPP⁺ and 320 nm for analysis of 4-hydroxyquinoline. A response curve of area versus concentration was constructed to calculate the concentration of each compound. Identification of reaction products was done by UV (DAD spectra) and fluorescence. Confirmation of the identity of the reaction products (4-hydroxyquinoline, MPDP⁺ and MPP⁺) was performed with HPLC-electrospray(ESI)-mass spectrometry as previously^11,15,30.

Results and discussion

MPTP produces neurotoxicity through bioactivation following its oxidation to neurotoxic pyridinium cations (Figure 1). Incubation of MPTP with human MAO enzymes generated the corresponding pyridinium cations that were subsequently identified and quantified by HPLC-DAD (MS) (Figure 2). MPDP⁺ was determined at 355 nm and MPP⁺ at 280 nm, and these metabolites were identified by HPLC-MS (ESI) affording ions at m/z 172 and 170, respectively. By using this chromatographic method, it was found that both MAO-A and -B isozymes, were able to oxidize MPTP to MPDP⁺ (with detectable amounts of MPP⁺) in a concentration-dependent manner (Figure 3A). Reproducibility of the method, including incubation and analysis was good (RSD of 4.0% for MAO-A and 3.53% for MAO-B; 50 μM MPTP and 0.05 mg/mL protein). Michaelis-Menten kinetics corresponding to the oxidation of MPTP (to MPDP⁺ species) provided values of V_max and K_m of 9.2 ± 0.2 nmol/min mg prot and 125 ± 7 μM, respectively for MAO-B and 5.2 ± 0.06 nmol/min mg prot. and 40.5 ± 2.2 μM for MAO-A (Figure 3B). The V_max for the oxidation of MPTP with MAO-B was much higher than for the oxidation with MAO-A, so that a higher oxidation of MPTP is expected for MAO-B, particularly at higher concentrations of substrate. MAO-A exhibited a higher affinity for MPTP (lower K_m) and its reaction rate reached saturation at low concentrations of substrate. In a subsequent experiment that was designed to compare MAO-A and -B, a relative oxidation rate using two different substrates of MAO (i.e. MPTP and kynuramine) was measured by comparison of their corresponding products, MPDP⁺ and 4-hydroxyquinoline, under the same experimental conditions. Those products were determined by HPLC by using the method described here (MPDP⁺) and a previous one (4-hydroxyquinoline)³⁰. Again, MAO-B had a much higher relative activity on MPTP than MAO-A (Figure 4). Thus, the conversion of MPTP to MPDP⁺ compared with that of kynuramine to 4-hydroxyquinoline was 2.5-fold higher for MAO-B compared to MAO-A. On the other hand, incubation experiments by using human liver mitochondria (HLM), which contain MAO-A and−B, instead of the recombinant enzymes, showed that in presence of R-deprenyl (1 μM) (a potent MAO-B inhibitor), the oxidation of MPTP is highly inhibited (> 80%), whereas clorgyline (1 μM) (MAO-A inhibitor) produced little inhibition. These results clearly suggest that MAO-B is the key isozyme involved in the oxidation of MPTP in good agreement with previous results^19,33,36. Nevertheless, results obtained here also suggest that MAO-A might somehow accomplish the oxidation of MPTP under certain circumstances, and particularly, at low concentration of the toxin. Indeed, previous studies have shown that MPTP and MPTP analogs which are also neurotoxic substances might be oxidized with the participation of MAO-A^37,38.

Figure  2. . HPLC chromatogram of the enzymatic oxidation of MPTP neurotoxin by human MAO-B (A) and MAO-A (B). MPDP⁺ is determined at 355 nm, MPP⁺ at 280 nm, and MPTP at 254 nm.

Figure  3.  (A) Concentration of pyridinium cations (MPDP⁺ and MPP⁺) (μM) determined by HPLC and produced from MPTP oxidation in presence of increasing concentrations of recombinant human MAO-B (MPDP⁺, ▪;MPP⁺, o) and MAO-A (MPDP⁺, •; MPP⁺ Δ). Incubations (37°C, 40 min) and MPTP (200 μM). (B) Michaelis-Menten kinetics for the formation of pyridinium cations determined by HPLC, and corresponding to the oxidation of MPTP catalyzed by MAO-B (MPDP⁺, ▪;MPP⁺, Δ) and MAO-A (MPDP⁺, ▴; MPP⁺, x). Incubations (37°C, 40 min) contained MAO enzymes (0.05 mg/mL protein fraction).

Figure  4. . Activity ratio of MAO-B and -A isozymes under the same conditions determined by using the products from two different substrates: kynuramine (formation of 4-hydroxyquinoline) following a previous method (30), and MPTP (formation of MPDP⁺) analyzed by HPLC, as described here. Separate incubations (37°C, 40 min) with enzymes (0.05 mg/mL protein fraction) and 150 μM kynuramine or 150 μM MPTP were carried out.

A good correlation usually exists between the oxidation of MPTP (or MPTP analogs) with the pyridinium cations generated and the neurotoxic effects observed in vivo³⁷, and also, between the ability to protect against the neurotoxic action of tetrahydropyridines (i.e. MPTP and its analogs) and the reduction of pyridinium cations exerted by MAO inhibitors (Figure 1). Indeed, the bioactivation of MPTP proneurotoxin to directly-acting neurotoxic pyridinium species relies on the participation of MAO enzymes. Therefore, an incubation of MPTP with human MAO (or alternatively with human mitochondria) along with the subsequent chromatographic analysis of the toxic pyridinium cations generated might be useful for a preliminary screening of eventual protective agents and MAO inhibitors (Figure 5). Thus, the reduction of pyridinium species produced by MAO-B in presence of R-deprenyl, a known neuroprotectant used in PD therapy, showed the suitability of this approach (Figure 6). Also, norharman, a naturally-occurring β-carboline alkaloid, reported to be a competitive inhibitor of MAO-B and possible neuroprotectant analog^34,35,39,40, inhibited and decreased the oxidation of MPTP to neurotoxic species (MPDP⁺ and MPP⁺) (Figure 6) with an IC₅₀ of 2.9 ± 0.1 μM and a K_i of 0.85 μM calculated for competitive inhibition⁴¹. In the same assay, harman (1-methyl-β-carboline) did not show significant inhibition up to 20 μM, indicating that this β-carboline is a poor inhibitor of MAO-B. Inhibition of MAO-B and protection against MPTP oxidation was also observed for other agents such as 5-nitroindazole and the vitamin-K analogue menadione, suggesting possible actions of these compounds as protective agents (Figure 7)^19,33. On the other hand, clorgyline that is an inhibitor of MAO-A, highly inhibited the oxidation of MPTP by MAO-A (Figure 8) and this oxidation was also inhibited by the β-carboline harman with IC₅₀ of 2.8 ± 0.25 μM and an experimental K_i of 0.19 μM as measured at different concentrations of MPTP. In the same assay, norharman appeared to be a weak inhibitor over the oxidation of MPTP by MAO-A (i.e. 30% inhibition, 20 μM).

Figure  5. . Inhibitors of the oxidation of MPTP neurotoxin to toxic pyridinium cations catalyzed by MAO-A and -B.

Figure  6.  Inhibition of the human MAO-B (0.05 mg/mL protein fracion) catalyzed-oxidation of MPTP (300 µM) to give pyridinium cations in the presence of R-deprenyl (A) or the β-carboline norharman (B) (MPDP⁺, ▴;MPP⁺, •). Incubations carried out at 37°C for 40 min.

Figure  7.  Reduction of pyridinium cation (MPDP+) produced in the oxidation of MPTP catalyzed by MAO-B in the presence of several protective agents: control (absence of inhibitor), norharman (5 μM), 5-nitroindazole (5 μM), menadione (5 μM) and deprenyl (1 μM). Incubations (37°C, 40 min) were carried out with MPTP (300 μM) and MAO-B (0.05 mg/mL protein fraction) in the presence or absence of agent.

Figure  8.  Inhibition of the human MAO-A (0.05 mg/mL recombinant protein) catalyzed oxidation of MPTP neurotoxin (300 μM) to give pyridinium cations in the presence of clorgyline (A) or the β-carboline harman (B) (MPDP+, ▪; MPP+, ▴). Incubations carried out at 37°C for 40 min.

Tobacco smoke inhibits MAO both in vitro and in vivo, and this inhibition might result in neuroprotection and antidepressant effects in smokers^34,42,43. Moreover, smoke contains MAO inhibitors such as the β-carbolines norharman and harman that were previously isolated and characterized³⁴. By using the method described here, it was observed that smoke extracts decreased the formation of toxic pyridinium cations from MPTP and MAO-B (Figure 9), suggesting the occurrence of MAO inhibitors and protective substances against MPTP in cigarette smoke. A correlation between MPDP⁺ and MPP⁺ was not observed in this case since smoke seemed to accelerate the conversion of MPDP⁺ to MPP⁺. This is expected to occur when MPDP⁺ is incubated in presence of prooxidant species or oxidative systems¹¹, and smoke is a known source of reactive oxygen species (ROS)⁴⁴. Taking both pyridinium species together (MPDP⁺ and MPP⁺), the oxidation of MPTP was highly reduced in the presence of smoke extract. In addition, the naturally-occurring β-carboline norharman that was isolated from tobacco smoke as previously³⁴, highly reduced the formation of pyridinium species by MAO-B (Figure 9B). Similar results were obtained using norharman isolated from coffee extracts³⁵ (Figure 9C). On the other hand, harman isolated from tobacco smoke or coffee inhibited MAO-A (harman at 1.5 μM into the assay reduced the formation of MPDP⁺ + MPP⁺ by about 36%). These results suggest that those naturally occurring β-carbolines (norharman and harman) could be protective substances against the MPTP neurotoxin bioactivation by MAO-A and B.

Figure  9.  Inhibition of the human MAO-B-catalyzed oxidation of MPTP to toxic pyridinium cations in the presence of cigarette smoke extract (A): MPDP+ + MPP+ (□), MPDP+ (⧫), MPP+ (•), and the inhibition (%) in the presence of the β-carboline norharman isolated from cigarette smoke (B) or coffee (C) as reported previously34,35. Incubations carried out at 37°C for 40 min with MPTP (300 μM) and MAO-B (0.05 mg/mL).

The results obtained above show the usefulness of the determination of pyridinium cations generated from MPTP as a tool for screening of eventual protective agents and MAO inhibitors. Results regarding inhibition of β-carbolines, deprenyl or clorgyline were comparable to those obtained by using kynuramine or other substrates of MAO^{1,30,34,35,40}. The procedure employed has the advantage of using human MAO isozymes with a true neurotoxin as substrate (i.e. MPTP), although it requires the chromatographic analysis of the pyridinium products. Remarkably, MPTP is being currently used to induce parkinsonism in animal models and also to study neuroprotection in vivo^9,18,20. MAO isozymes are good targets for antidepressants (MAO-A inhibitors) and neuroprotective drugs (MAO-B inhibitors)^1,45. The oxidation of biogenic amines and neurotransmitters by MAO produces reactive oxygen species (ROS)^2,16, and the elevation of MAO-B is associated with an increased susceptibility to neurodegeneration⁶. In addition, MAO may oxidize neurotoxins like MPTP and analogs to directly-acting toxic pyridinium metabolites (MPDP+ and MPP+)^{10,11,15,17,37,46}(Figure 1). Therefore, a convenient use of MAO-inhibiting substances may result in protection against oxidative stress and toxicants^1,2,16,17. In this regard, the oxidation of MPTP by MAO could be successfully used for preliminary screening of MAO inhibitors and potential protective agents. As shown, MAO-B is the main isozyme involved in the bioactivation of MPTP, although MAO-A might play a role in the metabolism of MPTP and analogs^1,37,38. In this research, MAO inhibitors and potential neuroprotectants such as deprenyl, highly reduced the oxidation of MPTP to pyridinium cations. Oxidation was also decreased by naturally-occurring β-carbolines. Norharman inhibited the oxidation by MAO-B and both norharman and harman inhibited the oxidation by MAO-A. Environmental samples such as cigarette smoke decreased the oxidation of MPTP by MAO-B and those isolated β-carbolines from smoke or coffee inhibited the oxidation of MPTP neurotoxin by MAO, suggesting potential protective actions of these substances. β-Carboline alkaloids are antioxidants and inhibitors of MAO^34,47,48, and interestingly the norharman analog, 9-methylnorharman (9-methyl-β-carboline) has been recently proposed as a novel antiparkinsonian and restorative agent in neurons³⁹. Other compounds such as 5-nitroindazole and menadione (vitamin K3) exerted the same inhibitory action on MPTP activation and could have further utility as protective agents^19,33.

In summary, the oxidation of the MPTP neurotoxin by MAO enzymes (MAO-A and−B) along with the analysis of the toxic pyridinium species generated was evaluated here, and successfully used as a new tool for screening of MAO inhibitors and eventual protective substances. Deprenyl, clorgyline, β-carbolines (norharman and harman), 5-nitroindazole, menadione and tobacco smoke showed inhibitory properties on human MAO-B and/or MAO-A, and highly decreased the bioactivation (oxidation) of MPTP proneurotoxin to give pyridinium cations. By using human MAO isozymes and the neurotoxin MPTP as a substrate to assessing for MAO inhibitors and eventual protective agents, the procedure used here allows to directly observe the effects of the agents on the bioactivation of this neurotoxin, while reducing the expected variability for other substrates and MAO sources^1,46,49,50.

Acknowledgements

The author thanks the project AGL-2010-18448 (MICINN, Spain) and the intramural CSIC 200470E658 for financial support, and Carolina Chaparro, Juan Galisteo, and Hugo Guillén for technical assistance.

Declaration of interest

This work was financially supported by project AGL-2010-18448 (MICINN, Spain) and the intramural CSIC 200470E658. The author declares that there are no conflicts of interest.