Inhibitory effects of agmatine on monoamine oxidase (MAO) activity: Reconciling the discrepancies

Agmatine is an endogenous polyamine derived in mammals from arginine by the arginine decarboxylase enzyme. Several physiological functions have been demonstrated for this polyamine (1), including the release of insulin, catecholamine and luteinizing hormone from beta cells, adrenal gland and hypothalamus, respectively, and the ability to inhibit glutamatergic transmission and nitric oxide production. Agmatine reduced the effect of naloxone-precipitated abstinence syndrome of morphine dependent rats (2), and displayed antiproliferative properties in smooth muscle cells and neuroprotection in neurons (3, 4, 5). Substantial evidence supports agmatine as a co-neurotransmitter given its presence in synaptic vesicles and its potassium and calcium -dependent release from neurons (6, 7). Agmatine has been shown to bind to the imidazoline binding site I₂ localized in mitocondrial outer membrane of liver, kidney and brain (8). Among the various I₂ binding proteins identified there is the monoamino oxidases (MAO) (9, 10). The I₂ binding site displays a high sequence homology with a specific domain present on the monoamino oxidase-B (MAO-B), and shows the same molecular weight (11, 12, 13). Moreover, transfection of MAO into yeast results in coherent coexpression of I₂ sites. Altogether, these data indicate that the I₂ binding site is located on the MAO molecule and constirutes allosteric sites of these enzymes (14). Agmatine can thus represent an endogenous modulator of the MAO enzyme. There is however no dedicated study that conclusively shows the ability of agmatine to modulate the MAO activity, however the same information to this regard can be found in studies carried out to evaluate the modulation of MAO by imidazoline ligands.

A dose dependent inhibition of MAO activity by agmatine, with an IC₅₀ of 168 μM, was reported in liver homogenate by Raasch and co-workers using a kynuramine-based assay (14). Furthermore agmatine has been shown interact with the catalytic sites of human recombinant MAO-A, and inhibit 50% of the oxidation activity at 1 mM (15). Contrasting observations were however observed in other studies where agmatine was found to weakly inhibit the MAO activity in rat brain (16, 17) and liver homogenates (18). Although the lack of inhibitory effect of agmatine on MAO activity found by Holt and Becker (16) could be due to the low agmatine concentration used (50 μM), this explanation cannot apply to the negative results obtained by Su et al., 2001, and by Ozaita et al., 1997 (17,18), where the agmatine concentration was 1 and 10 mM, respectively. The reason of these conflicting results is thus still an undetermined, although the possibility exists that they can be due to differences in experimental conditions used in the studies mentioned above, namely the incubation time, pH and substrate used. In this study we verified whether the inhibitory action of agmatine on MAO activity is dependent on the experimental conditions and thus reconciling the different conclusions reached by the above citied authors.

Following the addition of MAO substrate tyramine, a significant MAO activity could be observed in rat liver homogenates. This was visualized by peroxidase-coupled assay following hydrogen peroxide production by MAO as indicated by the increase of peak of absorbance at 492 nm respect to no-tyramine condition (Fig. 1A and S1A).

Figure 1

The MAO activity was fully prevented by incubation with the MAO inhibitor tranylcypromine (10 μM; Fig. 1A inset), with an IC₅₀ of 756 nM (Fig. 1SB). By contrast, application of 1 mM agmatine did not induce any inhibitory effect on the MAO activity (red trace in Fig. 1A and inset). These data agree with several previous studies (16, 17, 18), but are at variance with others (14, 15).

However, in both these studies, where an inhibitory effect of agmatine on MAO activity was observed (14, 15), kynuramine was used as substrate to assay the MAO activity. To verify the dependency on the substrate of agmatine action on MAO activity we repeated the test described above using the kynuramine-based assay (Fig. S2A). Incubation of liver homogenates with kynuramine resulted in an absorbance increase to 330 nm (as expected from its conversion into 4-idroxychinoline) that could be fully suppressed by saturating concentrations of tranylcypromine (Fig. 1B). We also found using the peroxidase-coupled assay, agmatine both at 100 μM and 1 mM still failed to inhibit the MAO activity (Fig. 1B and S2B). Under these conditions idazoxan, a ligand for the imidazoline I2 binding site (15, 18) dose-dependently inhibited the MAO activity (IC₅₀ and h of 365 μM and 1.2 respectively, Fig. S2B).

We then verified whether incubation time and pH, could influence agmatine inhibition action on MAO activity. We first changed the agmatine (1 mM) incubation time, by increasing it to 60 minutes, while keeping the pH constant at 7.4.

Under these conditions no differences could be observed in the MAO activity, as compared to level measured with 5 min incubation time (Fig. 2A). No difference in the agmatine modulation of MAO activity was also found when we changed the pH, and carried out the experiments at a pH of 9.2, while keeping the incubation time of agmatine constant at 5 minutes (Fig. 2B). To this regard it needs to be said that at pH 9.2 we found a significant increase in MAO activity (inset Fig. 2B), as consequence of deprotonation of amine (kynuramine) substrate (14). By contrast, when the agmatine (1 mM) incubation time was increased to 60 min, and simultaneously the pH was increased to 9.2, we observed a marked inhibition of the MAO activity by agmatine, which we failed to observe previously when we separately changed only one of these two conditions, namely incubation time and pH (Fig. 2C).

Figure 2

We then investigated some features of agmatine inhibition on MAO activity under these experimental conditions (60 minutes incubation time and pH fixed at 9.2). Agmatine was able to inhibit the MAO activity in a dose-dependent manner (Fig. 3A) with IC₅₀ and h (Hill’s coifficent) of 1.3 mM and 0.35 respectively, values similar to those reported by Raash et al. (14). The relatively long incubation time required for the agmatine inhibition of MAO activity raised the possibility that the accumulation of a agmatine metabolite was needed for the inhibitory effect. The two main agmatine metabolites known are putrescine, catalyzed by agmatinase, and 4-guanoglutaraldeide catalyzed by diamino oxidase (DAO) (14). Since putrescine has already been shown to be ineffective on MAO activity (14), under conditions similar to ours, we then verified the requirement of 4-guanoglutaraldeide for the agmatine dependent inhibition of MAO activity. For this purpose we added the DAO inhibitor aminoguanidine to agmatine in MAO mixture reaction (16). As shown in Fig. 3B, aminoguanidine at 1 mM did not abolish the inhibitory effect of agmatine on liver MAO activity, suggesting that an enzymatic conversion of agmatine by DAO is not needed for its inhibitory effects.

Figure 3

Our results indicate that the apparently contrasting conclusions previously reached with regard to the inhibitory efficacy of agmatine on MAO activity can be reconciled by taking into account the different experimental conditions used, namely differences in the agmatine incubation time and the pH of the milieu. More specifically, a significant inhibitory effect of MAO activity by agmatine is observed when both relatively long incubation times and high pH are used. A possible explanation for the experimental condition dependence in the agmatine MAO activity inhibition could be found in the change of conformational state of enzyme induced by experimental reaction conditions.

Note that a similar incubation time and pH dependency has been known for a long time, for example, in the studies involving hydrazine reagents as MAO inhibitors (23). In addition, MAO inhibitors such as β-carboline can inhibit the MAO activity depending on the incubation time (24), while oxazolidinones display a remarkable increased inhibitory potency in alkaline melieu (15).

Agmatine administration produced antidepressant-like behaviour in animal (25) and in human case studies (26) with a involvement of non-serotoninergic meccanism but with imidazolines receptors. Since MAO enzyme is an important target in anti-depressive drugs and associated to I2 imidazolines receptors, the inhibitory effect of agmatine in MAO activity here investigated might contribute to a greater understandment of the role of agmatine in the neurological disorders. However, further studies are need to conclusively define the inhibitory mechanism of agmatine on MAO activity and its involvement in pathophysiological states (27).