Volume 54, Issue 3 p. 245-257
Review Article
Free Access

On the free radical scavenging activities of melatonin's metabolites, AFMK and AMK

Annia Galano

Corresponding Author

Annia Galano

Departamento de Química, Universidad Autónoma Metropolitana-Iztapalapa, DF, México

Address reprint requests to Annia Galano, Departamento de Química, Universidad Autónoma Metropolitana-Iztapalapa, San Rafael Atlixco 186, Col. Vicentina, Iztapalapa, CP 09340, México DF, México.

E-mail: [email protected]

Search for more papers by this author
Dun Xian Tan

Dun Xian Tan

Department of Cellular and Structural Biology, UT Health Science Center, San Antonio, TX, USA

Search for more papers by this author
Russel J. Reiter

Russel J. Reiter

Department of Cellular and Structural Biology, UT Health Science Center, San Antonio, TX, USA

Search for more papers by this author
First published: 24 August 2012
Citations: 627

Abstract

The reactions of N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK) and N1-acetyl-5-methoxykynuramine (AMK) with OH, OOH, and •OOCCl3 radicals have been studied using the density functional theory. Three mechanisms of reaction have been considered: radical adduct formation (RAF), hydrogen transfer (HT), and single electron transfer (SET). Their relative importance for the free radical scavenging activity of AFMK and AMK has been assessed. It was found that AFMK and AMK react with •OH at diffusion-limited rates, regardless of the polarity of the environment, which supports their excellent •OH radical scavenging activity. Both compounds were found to be also very efficient for scavenging •OOCCl3, but rather ineffective for scavenging •OOH. Regarding their relative activity, it was found that AFMK systematically is a poorer scavenger than AMK and melatonin. In aqueous solution, AMK was found to react faster than melatonin with all the studied free radicals, while in nonpolar environments, the relative efficiency of AMK and melatonin as free radical scavengers depends on the radical with which they are reacting. Under such conditions, melatonin is predicted to be a better •OOH and •OOCCl3 scavenger than AMK, while AMK is predicted to be slightly better than melatonin for scavenging •OH. Accordingly it seems that melatonin and its metabolite AMK constitute an efficient team of scavengers able of deactivating a wide variety of reactive oxygen species, under different conditions. Thus, the presented results support the continuous protection exerted by melatonin, through the free radical scavenging cascade.

Introduction

The beneficial effects of melatonin (N-acetyl-5-methoxytryptamine) on human health are well known and are frequently associated with the attenuation of oxidative damage 1-9. The protective effects of melatonin against the deleterious effects caused by oxidative stress (OS) are well documented 10-18. One of the most appealing properties of melatonin, which distinguishes it from most antioxidants, is that its metabolites also have the ability to scavenge reactive oxygen species (ROS) and reactive nitrogen species (RNS). The continuous protection exerted by melatonin and its metabolites, referred as the free radical scavenging cascade 19-21, makes melatonin highly effective, even at low concentrations, in protecting organisms from OS 21.

N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK) is one of the metabolites of melatonin and can be formed by both enzymatic or pseudoenzymatic 22-26 and nonenzymatic 27-33, 13 metabolic pathways. In turn, N[1]-acetyl-5-methoxykynuramine (AMK) is formed by deformylation of AFMK 19, 20, 27, 29, 34-36. Both metabolites have been found to exhibit protective effects against OS. AFMK reduces lipid peroxidation and oxidative DNA damage and prevents neuronal cell injuries caused by H2O2 and β-amyloid peptide 37-41. It has also been reported that AFMK protects against high-energy charged particle radiation-induced oxidative damage to the brain 42. Such neuroprotection was explained by its free radical scavenging function. This seems to be a logical assumption as AFMK is able to efficiently scavenge •OH radicals 37-39. According to the information garnered so far, it can be stated that AMK is a good and versatile free radical scavenger. It is able of deactivating a wide variety of ROS 43-45 and RNS 46-50 and also other oxidants 44, 51, 52.

Regarding their relative antioxidant capacity, AFMK has been reported to be a less effective protector than AMK and melatonin 14, 35, 53. In fact, AMK was found to be a potent singlet oxygen scavenger 43, while AFMK is remarkably inert in this regard. AMK was described as better a NO scavenger than melatonin 47 or AFMK 49. In addition, the efficiency of AMK for scavenging ROS and preventing protein oxidation has been reported to be higher than that of AFMK 44.Therefore, it seems that at least in general, their protective activities against OS follow the order AMK > melatonin > AFMK. However, there are no quantitative data for such a trend. There is also no data on whether the polarity of the environment affects this trend, or the dominant reaction mechanism involved in the free radical scavenging activity of AFMK and AMK. Moreover, values of the rate constants for the reaction of these two compounds with free radicals have not been reported. Consequently, it is the main purpose of this work to provide new information on these aspects. To that purpose, we have investigated the following reaction mechanisms: radical adduct formation (RAF), hydrogen transfer (HT), and single electron transfer (SET). In addition, it was previously demonstrated that the relative importance of different reaction mechanisms depends not only on the scavenger but also on the chemical nature of the radical they are reacting with 54-56. Therefore, different free radicals have been considered: •OH, •OOH, and •OOCCl3. Thermodynamic and kinetic data are provided, as well as a quantitative assessment of the contributions of the different mechanisms and channels of reaction to the overall reactivity of AFMK and AMK toward the aforementioned radicals. Comparisons with their precursor, melatonin, are also provided.

Computational details

Geometry optimizations and frequency calculations have been carried out using the M05-2X functional 57 and the 6−31 + G(d,p) basis set, in conjunction with the SMD continuum model 58 using benzene and water as solvents to mimic lipid and aqueous environments, respectively. The M05-2X functional has been recommended for kinetic calculations by their developers 57, and it has been also successfully used by independent authors for that purpose 59-61. It is also among the best performing functionals for calculating reaction energies involving free radicals 62. SMD is considered a universal solvation model, because of its applicability to any charged or uncharged solute in any solvent or liquid medium for which a few key descriptors are known 58.

Unrestricted calculations were used for open-shell systems, and local minima and transition states were identified by the number of imaginary frequencies (NIMAG = 0 or 1, respectively). In the case of the transition states, it was verified that the imaginary frequency corresponds to the expected motion along the reaction coordinate, by Intrinsic Reaction Coordinate calculations (IRC). All the electronic calculations were performed with the Gaussian 09 package of programs 63. Thermodynamic corrections at 298.15 K were included in the calculation of relative energies, which correspond to 1 m standard state. In addition, the solvent cage effects have been included according to the corrections proposed by Okuno 64, taking into account the free volume theory 65.

The rate constants (k) were calculated using the conventional transition state theory (TST) 66-68 and 1 m standard state as:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0001(1)
where kB and h are the Boltzmann and Planck constants, ΔG is the Gibbs free energy of activation, σ represents the reaction path degeneracy accounting for the number of equivalent reaction paths, and κ accounts for tunneling corrections.
The tunneling corrections, defined as the Boltzmann average of the ratio of the quantum and the classical probabilities, were calculated using the zero-curvature tunneling corrections (ZCT) 69. For the electron transfer (ET) reactions, the barriers were estimated using the Marcus theory 70, 71. It relies on the transition-state formalism and defines the SET activation barrier urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0002 as:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0003(2)
where urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0004 is the free energy of reaction and λ is a reorganization term. In this work, a very simple approximation has been made to calculate λ:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0005(3)
where ΔEET has been calculated as the nonadiabatic energy difference between reactants and vertical products. This approach is similar to that previously used by Nelsen et al 72, 73 for a large set of self-exchange reactions.
In addition, some of the calculated rate constants (k) are close to the diffusion limit, thus the apparent rate constant (kapp) cannot be directly obtained from TST calculations. In this work, the Collins–Kimball theory is used to that purpose 74:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0006(4)
where kact is the thermal rate constant, obtained from TST calculations, and kD is the steady-state Smoluchowski 75 rate constant for an irreversible bimolecular diffusion-controlled reaction:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0007(5)
where R denotes the reaction distance, NA is the Avogadro number, and DAB is the mutual diffusion coefficient of the reactants A (free radical) and B (scavenger). DAB has been calculated from DA and DB according to reference 76, and DA and DB have been estimated from the Stokes–Einstein approach 77, 78:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0008(6)
where kB is the Boltzmann constant, T is the temperature, η denotes the viscosity of the solvent, in our case water (η = 8.91 × 10−4 Pa s) and benzene (η = 6.04 × 10−4 Pa s), and a is the radius of the solute.

Results and discussion

To mimic any chemical species under physiological conditions, in particular in the aqueous phase, it is important to know which would be the prevailing acid/base form. To that end, it is crucial to know the pKa values of the studied species. Unfortunately, to our best knowledge, there are no reports for the pKa values of AFMK and AMK, while that of their precursor, melatonin has been reported to be 12.3 ± 0.1 79. Therefore, we have estimated the pKa values of AFMK and AMK, using the proton exchange method, also known as the isodesmic method, or the relative method 80, which is based on the reaction scheme:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0009
where HRef/Ref is the acid/base pair of a reference compound and should be structurally similar to the system of interest. In our case, we have chosen HRef = melatonin. Within this approach, the pKa is calculated as:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0010(7)
The pKa values estimated in this manner for AFMK and AMK are reported in Table 1, together with the molar fractions of their neutral (mfN) and anionic forms (mfA) at physiological pH. To calculate the molar fractions, we have obtained the acid constants (Ka) from the pKa values as:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0011(8)
Table 1. pKa and molar fractions of neutral (mfN) and anionic forms (mfA) of melatonin, N1-acetyl-N2-formyl-5-methoxykynuramine (AFMK), and N1-acetyl-5-methoxykynuramine (AMK) at physiological pH
pKa mfN mfA
Melatonin 12.3 [80] 1.00 0.00
AFMK 8.7 0.95 0.05
AMK 16.8 1.00 0.00
Then, using the definition of the equilibrium constant, for the deprotonation equilibrium (HA↔A + H+):
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0012(9)
The fraction of the anion can be easily obtained as:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0013(10)

where [H+] is calculated from pH. At physiological pH (7.4), [H+] = 3.98 × 10−8 m.

According to the values in Table 1, at physiological pH (7.4), AFMK and AMK, as well as melatonin, are expected to be mainly in their neutral form with populations larger than 95%. Accordingly, the neutral forms of AFMK and AMK are the ones used in this work (Scheme 1) for the study of the free radical scavenging activity in aqueous solution. They are also the ones used in nonpolar solution (lipid), as such medium does not promote the necessary solvation to stabilize ionic species, and therefore, the deprotonation is unlikely. To assess which nitrogen would be involved in the deprotonation, both possibilities were investigated, that is, from N1 and N2. For AFMK, it was found that the Gibbs free energy of the deprotonation from site N2 is 9.92 kcal/mol lower than that of the deprotonation from site N1. On the other hand, for AMK, the deprotonation from N1 was found to be 4.11 kcal/mol lower. Therefore, the deprotonations of neutral AFMK and AMK correspond to sites N2 and N1, respectively.

Details are in the caption following the image
Structures and site numbering. Blue and red labels represent radical adduct formation and hydrogen transfer reaction sites, respectively.

Regarding the reaction mechanisms, the free radical scavenging activity of AFMK and AMK can take place, through a variety of them, as it is the case for many other scavengers 81-87. As mentioned previously, those considered in this work are as follows: radical adduct formation (RAF), hydrogen transfer (HT), and single electron transfer (SET) in aqueous solution, and in nonpolar media, only RAF and HT. The SET mechanism has not been included in this case as nonpolar environments do not promote the necessary solvation of the intermediate ionic species yielded by this mechanism. However, just to prove this point, the energies of reaction for the SET process were calculated and found to be larger than 45 kcal/mol in all the cases (Table 2). The reaction sites for the HT and RAF reactions have been labeled in red and blue colors, respectively, in Scheme 1. Radical additions to sites C2 and C5, in AMK, and to sites C2, C5, and C6, in AFMK, were also tested. However, any attempt to locate the corresponding products invariably led to structures that are weak-bonded complexes rather than proper radical adducts. Therefore, these paths have been ruled out for yielding viable reaction products.

Table 2. Gibbs free energies of reaction (ΔG, kcal/mol), at 298.15 K, in benzene solution
path AFMK AMK
•OH •OOH •OOCCl3 •OH •OOH •OOCCl3
SET 77.78 97.43 62.97 62.30 81.95 47.49
HT-C1 19.89 12.99 7.25 20.17 12.71 6.98
HT-C3 24.00 8.88 3.14 24.16 8.72 2.98
HT-C4 26.06 6.82 1.09 25.55 7.33 1.59
HT-C6 22.38 10.50 4.76
HT-C13 19.65 13.23 7.49 21.12 11.76 6.02
HT-N1 9.03 23.85 18.12 9.27 23.61 17.87
HT-N2 18.86 14.02 8.29 26.09 6.79 1.05
RAF-C7 18.14 14.65 9.52 16.06 12.68 6.57
RAF-C8 13.31 17.66 15.08 14.82 13.12 12.81
RAF-C9 19.26 8.22 2.99 22.39 4.41 2.09
RAF-C10 17.07 10.91 6.72 20.60 8.84 1.68
RAF-C11 15.74 13.07 9.19 15.79 13.27 8.84
RAF-C12 13.50 12.93 7.60 21.29 6.88 2.10
  • The bold letters have been used to highlight the exergonic reaction paths.

The free radicals used to model the scavenging activity of AFMK and AMK are as follows: •OH, •OOH, and •OOCCl3. •OH has been chosen for being the most electrophilic 88, and reactive, of the oxygen-centered radicals with a half-life of ~10−9 s 89. In addition, AFMK and AMK, as well as their precursor melatonin, are known to efficiently scavenge this radical. Compared with •OH, peroxy radicals (•OOR) are much less reactive species capable of diffusing to distant cellular locations 90, with half-lives on the order of seconds 91. However, their reactivity can be highly influenced by the chemical nature of the R group. We have chosen •OOH because it is the simplest of the peroxyl radicals, and a good model for peroxyl radicals with R = nonhalogenated alkyl of alkenyl groups. Radical •OOCCl3 has been chosen because its reactivity is between those of •OH and •OOH and represents halogenated peroxyl radicals. It has been previously shown that melatonin is an excellent •OH radical scavenger, reacting at diffusion-limited rates, a very good •OOCCl3 scavenger, and rather ineffective for scavenging •OOH 55. Therefore, the selected set of free radicals is expected to embrace a wide range of reactivity.

For the reactions of both AFMK and AMK with •OH, all the HT and RAF reaction paths were found to be exergonic, regardless of the polarity of the environment, with the exception of the HT from site N1 in AFMK, in aqueous solution (Tables 2 and 3). Regarding the SET mechanism in aqueous solution, it was found to be exergonic by ~8 kcal/mol for the AMK + •OH reaction and slightly endergonic for the AFMK + •OH reaction. These results are in line with the high reactivity of the •OH radical and suggest that wide product distributions would arise from its reaction with AMK and AFMK.

Table 3. Gibbs free energies of reaction (ΔG, kcal/mol), at 298.15 K, in aqueous solution
path AFMK AMK
•OH •OOH •OOCCl3 •OH •OOH •OOCCl3
SET 5.69 29.18 8.61 8.05 15.45 5.13
HT-C1 24.40 8.51 1.46 21.64 11.27 4.21
HT-C3 28.00 4.91 2.14 27.68 5.24 1.82
HT-C4 31.96 0.96 6.10 30.74 2.18 4.88
HT-C6 21.75 11.16 4.10
HT-C13 21.25 11.66 4.61 23.04 9.87 2.81
HT-N1 2.48 35.40 28.34 8.84 24.07 17.02
HT-N2 22.58 10.34 3.28 28.82 4.09 2.97
RAF-C7 19.22 10.83 4.05 16.96 10.06 3.18
RAF-C8 14.15 8.78 2.06 13.76 15.45 8.75
RAF-C9 18.84 8.77 0.08 23.36 1.74 4.77
RAF-C10 18.07 11.25 4.20 20.41 7.99 0.67
RAF-C11 15.00 12.51 6.18 12.02 14.64 8.46
RAF-C12 14.57 11.60 4.59 20.23 7.58 0.63
  • The bold letters have been used to highlight the exergonic reaction paths.

For the AFMK + •OOCCl3 reaction, all modeled paths were found to be endergonic in benzene solution (Table 2), with the lowest values of Gibbs energies of reaction (ΔG) corresponding to paths C4 and C9 for HT and RAF processes, respectively. On the other hand, in aqueous solution, there are two HT paths that become exergonic (C3 and C4), while path RAF-C9 is almost isoergonic (Table 3). For the AMK + •OOCCl3 reaction in nonpolar media, there is an exergonic path (RAF-C9), and among the HT paths, those with the lowest ΔG values are HT-C4 and HT-N2. As is the case for AFMK, the number of viable reaction paths increases with the polarity of the environment. In aqueous solution, there are three viable HT paths (C3, C4, and N2) and two viable RAF paths (C9 and C12) for the AMK + •OOCCl3 reaction. In addition, the SET process is also viable, with ΔG = −5.13 kcal/mol.

All the reaction paths corresponding to the reactions of •OOH with both AFMK and AMK were found to be endergonic, regardless of the polarity of the environment, with the lowest endergonicity systematically corresponding to paths C4 and C9 for HT and RAF processes, respectively. This means that the metabolites of melatonin AFMK and AMK would not be able of efficiently scavenging •OOH radicals, which as mentioned before is also the case for their precursor.

For the kinetic study, we have not included the reaction paths described previously as endergonic because, even if they take place at a significant rate, they would be reversible and, therefore, the formed products will not be observed. However, they might still represent significant channels if their products rapidly react further. This would be particularly important if these later stages are sufficiently exergonic to provide a driving force and if their barriers of reactions are low. Because of the complexity of the biological systems, in which there are a wide variety of chemicals present, it is likely that it could be the case under such conditions. This would certainly be the case for the SET reactions in aqueous solution as they yield radical cations, which are prompt to easy and fast deprotonation. In addition, slightly endergonic processes can be important when there are no exergonic competing paths. This is the case of paths RAF-C9 and HT-C4 for the reactions with •OOH. Therefore, such processes have also been included in the kinetic calculations.

The fully optimized geometries of the transition states (TS) are shown in Figs 1-6. It was not possible to locate the TSs corresponding to path HT-N2 for AMK + •OH reaction in aqueous solution using full optimizations, even though it was located and characterized in a benzene solution (Fig. 2). Using partial optimizations with frozen N---H and H---OH bond distances, we obtain structures that present a single imaginary frequency corresponding to the desired transition vector. Unfreezing these two distances, during a saddle point optimization, invariably led to an increase in the H---OH distance and the corresponding decrease in the imaginary frequency and gradient, yielding the separated reactants. A relaxed scan, obtained by decreasing the H---OH distance, produces a similar result, that is, the energy decreases until the H atom is completely transferred. This means that the reaction is barrier less and strictly diffusion controlled. In other words, every encounter is effective in producing the conversion of reactants into products.

Details are in the caption following the image
Optimized geometries of the transition states corresponding to hydrogen transfer mechanism, involved in the reactions between N1-acetyl-N2-formyl-5-methoxykynuramine and OH radicals in water (benzene) solution.
Details are in the caption following the image
Optimized geometries of the transition states corresponding to hydrogen transfer mechanism, involved in the reactions between N1-acetyl-5-methoxykynuramine and OH radicals in water (benzene) solution.
Details are in the caption following the image
Optimized geometries of the transition states corresponding to radical adduct formation mechanism, involved in the reactions between N1-acetyl-N2-formyl-5-methoxykynuramine and OH radicals in water (benzene) solution.
Details are in the caption following the image
Optimized geometries of the transition states corresponding to radical adduct formation mechanism, involved in the reactions between N1-acetyl-5-methoxykynuramine and OH radicals in water (benzene) solution.
Details are in the caption following the image
Optimized geometries of the transition states corresponding to the reactions between N1-acetyl-N2-formyl-5-methoxykynuramine and peroxyl radicals (•OOH and •OOCCl3) in water (benzene) solution.
Details are in the caption following the image
Optimized geometries of the transition states corresponding to the reactions between N1-acetyl-5-methoxykynuramine and peroxyl radicals (•OOH and •OOCCl3) in water (benzene) solution.

Among the TSs for HT reactions from AFMK to •OH (Fig. 1), those involving sites C1, C3, and N2 present hydrogen bonding (HB)-like interactions. Based on the HB distances (rH), the strongest interaction corresponds to TS-C3, and it is stronger in aqueous solution than in benzene solution. For TS-C1 and TS-N2, on the other hand, the strength of the interaction decreases with the polarity of the environment. For the HT reactions from AMK to •OH (Fig. 2), the TSs presenting HB interactions are TS-C1, TS-C3, and TS-C4, with the strongest interactions corresponding to the later. In this case, the strength of the interaction slightly increases with the polarity of the environment for TS-C3 and TS-C4, while it has the opposite trend for TS-C1.

Regarding the TSs of the RAF reactions between •OH and AFMK (Fig. 3), those presenting HB interactions are TS-C8, TS-C9, and TS-C10. In this case, the TS with the strongest interaction depends on the polarity of the environment. In aqueous solution, it is TS-C8, while in benzene solution, it is TS-C9. In addition for TS-C8, the strength of the HB interaction increases with the polarity of the environment, while TS-C9 and TS-C10 shows the opposite behavior. For the RAF reactions between •OH and AMK (Fig. 4), the same transition states present HB (TS-C8, TS-C9, and TS-C10), but in this case, TS-C9 is the one with the strongest interaction, regardless of the polarity of the environment. In addition, according to the HB distance, the interaction in TS-C8 in benzene solution is expected to be negligible. The rH value is provided in this case only for comparison purposes. For TS-C9, the strength of the HB interaction decreases with the polarity of the environment, while for TS-C10, it increases.

The TSs corresponding to HT reactions from site C4 in AFMK and AMK to •OOH (Figs 5 and 6) both present HB interactions. In both cases, it is slightly weaker in benzene solution, with respect to aqueous solution. In addition, these two transition states present the strongest HB interaction compared with all the other TSs with this structural feature. The TSs of the •OOH additions (RAF) to site C9 in AFMK and AMK also present HB interactions, which are slightly stronger in nonpolar media. This kind of interactions is not present in any of the TSs involving reactions with •OOCCl3.

In general, the strength of the HB interactions in the transition states is stronger for those corresponding to HT processes than for the RAF ones. As this kind of stabilizing interactions lowers the energy of the TSs, it should be expected to contribute increasing the relative importance of HT over RAF in the •OH and •OOH scavenging activity of the studied compounds.

The Gibbs energies of activation (ΔG) for the reactions in benzene and aqueous solutions are reported in Tables 4 and 5, respectively. For the AFMK + •OH reactions, it was found that in nonpolar environments, paths HT-C3 and RAF-C9 are those with the lowest ΔG values among the HT and RAF reaction paths, respectively, with ΔG (HT-C3) about 1.2 kcal/mol lower than ΔG (RAF-C9). The finding that ΔG (HT-C3) is lower than ΔG (HT-C4) despite the larger exergonicity of the later can be explained by the strength of the HB interactions found in the corresponding TSs, which was discussed previously. In aqueous solution, HT-C3 remains as the HT path with the lowest ΔG, while for the RAF mechanisms, the ΔG values of paths C10, C12, and C9 are the lowest values and very similar to one another. For the AMK + •OH reactions, it was found that in nonpolar environments, path HT-C4 has the lowest ΔG value among the HT reaction paths (Table 4), while in aqueous solution, the lowest ΔG value among the HT reaction paths corresponds to HT-N2, followed by HT-C3 (Table 5). The ΔG values of the RAF reaction paths were also found to be influenced by the polarity of the environment. In benzene solution, the lowest value corresponds to path RAF-C12, followed by paths RAF-C9 and RAF-C10, in that order, while in aqueous solution, ΔG (RAF-C10) is significantly lower than ΔG (RAF-C12).

Table 4. Gibbs energies of activation (ΔG, kcal/mol), at 298.15 K, in benzene solution
path AFMK AMK
•OH •OOH •OOCCl3 •OH •OOH •OOCCl3
HT-C1 6.74 5.05
HT-C3 2.63 3.28
HT-C4 6.50 18.18 22.02 1.81 17.85 18.79
HT-C6 4.28
HT-C13 5.12 4.04
HT-N1 11.11 10.98
HT-N2 9.65 2.59
RAF-C7 5.74 6.01
RAF-C8 11.85 6.66
RAF-C9 3.83 19.53 14.89 1.08 16.80 9.86
RAF-C10 4.78 1.69
RAF-C11 5.07 3.66
RAF-C12 5.61 0.57
Table 5. Gibbs energies of activation (ΔG, kcal/mol), at 298.15 K, in aqueous solution
path AFMK AMK
•OH •OOH •OOCCl3 •OH •OOH •OOCCl3
SET 5.80 30.54 10.25 0.12 15.53 2.50
HT-C1 7.24 7.83
HT-C3 3.85 14.80 3.70 15.58
HT-C4 6.43 20.26 18.59 4.78 20.80 16.30
HT-C6 5.06
HT-C13 6.06 4.46
HT-N1 11.40 12.96
HT-N2 9.00 0.00 0.37
RAF-C7 4.07 1.54
RAF-C8 8.11 6.29
RAF-C9 3.31 18.22 9.57 0.94 15.08 2.32
RAF-C10 3.10 0.18
RAF-C11 4.87 3.26
RAF-C12 3.28 1.71 5.54

In general, the ΔG values of the RAF reaction paths were found to decrease with the polarity of the environment, while the ΔG values of the HT reaction paths increase, for the reactions of both AFMK and AMK with •OH. This indicates that the relative importance of HT and RAF mechanisms on the •OH scavenging activity of AFMK and AMK might be influenced by the polarity of the environment, which is suggested to promote RAF over HT for this particular radical. Another interesting trend is that in general, the ΔG values of the reaction paths involving AMK are lower than those of AFMK, regardless of the polarity of the environment and of the mechanism of reaction. This trend was also found for the reactions of these two melatonin's metabolites with the other studied radicals, which supports the higher reactivity of AMK. This point will be addressed in more detail in the discussion of the kinetics results.

For the AFMK + •OOH reactions, it was found that in benzene solution, ΔG (HT-C4) is lower than ΔG (RAF-C9), while in aqueous solution, the order is the opposite. For the reactions of AMK with the same radical, on the other hand, ΔG (RAF-C9) was found to be lower than ΔG (HT-C4), regardless of the polarity of the environment. This suggests that the relative importance of RAF and HT mechanisms on the •OOH scavenging activity of AFMK might be influenced by the polarity of the environment. On the other hand, RAF is expected to be systematically more important than HT for the •OOH scavenging activity of AMK.

Regarding the reactions of both AFMK and AMK with •OOCCl3, it was found that the ΔG values of RAF are significantly lower than those of HT, regardless of the polarity of the environment. However, in aqueous solution, the ΔG values of the SET processes are close to those of RAF, which suggests that this mechanism might also be relevant to the •OOCCl3 scavenging activity of the studied compounds.

The rate constants for the different reaction paths, in benzene and aqueous solutions, are reported in Tables 6 and 7, together with the overall rate coefficients, which have been calculated as the sum of the rate constants of each reaction path. For example, for the AMK + •OH reaction in aqueous solution:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0014
where:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0015
According to the values of koverall, AMK reacts systematically faster than AFMK, that is, regardless of the polarity of the environment and of the free radical that is being scavenged. This trend in reactivity for AMK and AFMK is in agreement with previously reported experimental evidence regarding their antioxidant activity 14, 35, 43, 44, 47, 53. In nonpolar environments, AMK was found to react 1.9, 2.3, and 7172.6 times faster than AFMK with •OH, •OOH, and •OOCCl3, respectively. In aqueous solution, AMK was found to react 1.7, 31.3, and 7982.9 times faster than AFMK with •OH, •OOH, and •OOCCl3, respectively. The large increase in reactivity toward the •OOCCl3 radical was identified to be because of RAF at site C9 and can be justified by the higher electron-donor character of the –NH2 group (AMK), compared with that of –NHCHO (AFMK).
Table 6. Rate constants of the different channels of reaction, and overall rate coefficient (per m/s), at 298.15 K, in benzene solution
path AFMK AMK
•OH •OOH •OOCCl3 •OH •OOH •OOCCl3
HT-C1 3.74E+08 1.70E+09
HT-C3 3.54E+09 3.48E+09
HT-C4 4.84E+08 4.52E+00 7.24E−02 3.63E+09 1.74E+00 6.59E+00
HT-C6 2.02E+09
HT-C13 2.13E+09 2.37E+09
HT-N1 4.92E+05 1.57E+06
HT-N2 7.15E+06 3.49E+09
RAF-C7 5.50E+08 5.73E+08
RAF-C8 2.54E+04 2.21E+08
RAF-C9 1.72E+09 5.97E-02 1.51E+02 2.81E+09 8.92E+00 1.08E+06
RAF-C10 1.28E+09 2.81E+09
RAF-C11 1.05E+09 2.62E+09
RAF-C12 6.35E+08 2.81E+09
Overall 1.38E+10 4.57E+00 1.51E+02 2.65E+10 1.07E+01 1.08E+06
Melatonin Overall [56]a 2.23E+10 3.11E+02 4.40E+08
  • a Calculated using the same methodology.
Table 7. Rate constants of the different channels of reaction, and overall rate coefficient (per M/s), at 298.15 K, in aqueous solution
path AFMK AMK
•OH •OOH •OOCCl3 •OH •OOH •OOCCl3
SET 3.33E+08 2.52E-10 1.89E+05 8.17E+09 2.58E+01 6.96E+09
HT-C1 3.25E+08 1.02E+08
HT-C3 2.17E+09 1.02E+03 2.32E+09 3.94E+02
HT-C4 5.08E+08 3.77E+00 1.69E+01 1.51E+09 7.61E−01 5.69E+02
HT-C6 1.12E+09
HT-C13 7.44E+08 1.42E+09
HT-N1 2.03E+05 6.00E+05
HT-N2 1.43E+09 2.48E+09 1.98E+09
RAF-C7 1.65E+09 1.90E+09
RAF-C8 1.40E+07 2.61E+08
RAF-C9 1.82E+09 5.51E−01 1.20E+06 1.91E+09 1.09E+02 1.51E+09
RAF-C10 1.84E+09 1.91E+09
RAF-C11 1.21E+09 1.84E+09
RAF-C12 1.82E+09 1.90E+09 6.33E+08
Overall 1.50E+10 4.32E+00 1.39E+06 2.57E+10 1.35E+02 9.11E+09
Melatonin Overall [56]a 1.85E+10 1.99E+01 1.40E+09
  • a Calculated using the same methodology.

Compared with their precursor melatonin, AFMK is predicted to be a weaker free radical scavenger. The koverall values for AFMK were found to be systematically lower than those of melatonin, for all the studied free radicals and regardless of the polarity of the environment. Accordingly, the efficiency of melatonin for scavenging free radicals is predicted to be reduced when it is metabolized to AFMK. On the other hand, the efficiency of AMK as free radical scavenger, relative to that of melatonin, was found to be influenced by the conditions under which the reactions take place. In aqueous solution, AMK was found to react 1.4, 6.8, and 7.9 times faster than melatonin with •OH, •OOH, and •OOCCl3, respectively. Therefore, the efficiency of melatonin for scavenging these radicals in aqueous solution is predicted to be increased when it is metabolized to AMK.

In nonpolar environments, the relative efficiency of AMK and melatonin as free radical scavengers depends on the radical they are reacting with. Under such conditions, AMK was found to react with •OH 1.2 times faster than melatonin, while melatonin reacts 29.2 and 405.6 times faster with •OOH and •OOCCl3. Therefore, in lipid media, melatonin is predicted to be a better •OOH and •OOCCl3 scavenger than AMK, while AMK is predicted to be slightly better than melatonin for scavenging •OH.

It should be noted, however, that despite their differences in reactivity, melatonin and its two metabolites AFMK and AMK, are all predicted to be excellent •OH scavengers, as according to the koverall values, their reactions with this dangerous ROS all take place at diffusion-limited rates (in the order of 1010 /m/s). This was found to be the case regardless of the polarity of the environment. Thus, this high efficiency for scavenging •OH is expected within different compartments of living organisms, that is, aqueous and lipid phases.

With respect to radical •OOCCl3, AFMK, AMK, and its precursor, all react faster in aqueous solution than in nonpolar environments, which is in part because of the SET mechanism that can only take place in the aqueous solution. In nonpolar media, melatonin is a very good •OOCCl3 scavenger (koverall = 4.4 × 108 /m/s), while AMK is moderately good (koverall = 1.1 × 106 /m/s) and AFMK rather poor (koverall = 1.5 × 102 /m/s). In an aqueous solution, AMK and melatonin are predicted to be excellent •OOCCl3 scavengers with rate coefficients close to the diffusion-limit regime (koverall = 9.1 × 109 /m/s for AMK and 1.4 × 109 /m/s for melatonin), while AFMK was found to be moderately good (koverall = 1.4 × 106 /m/s). According to these values, it can be stated that melatonin and its metabolites AFMK and AMK are efficient for scavenging radicals with high electron-withdrawing character, such as halogenated peroxyl radicals and •OH. For their reactions with •OOH, it was found that all the koverall values are lower than 103 /m/s, regardless of the polarity of the environment. According to these values and taking into account that the rate constants corresponding to the •OOH damage to polyunsaturated fatty acids are in the range 1.18—3.05 × 103 /m/s 92, it can be stated that AFMK, AMK, and also their precursor are rather ineffective for scavenging •OOH and probably other peroxyl radicals (•OOR) with R = nonhalogenated alkyl or alkenyl groups.

To investigate in detail the relative importance of the different mechanisms and paths of reaction to the overall activity of the studied compounds toward •OH, •OOH, and •OOCCl3, the branching ratios have been estimated. They represent the percent contribution of the different channels to the overall reaction and have been calculated as:
urn:x-wiley:07423098:media:jpi12010:jpi12010-math-0016(11)

where ki represents each reaction path.

Their values are reported for benzene and aqueous solution in Tables 8 and 9, respectively. For the reactions with •OH, a wide product distribution is expected, despite of the polarity of the environment, which is in line with the high reactivity of this radical. In nonpolar media, RAF and HT are both predicted to significantly contribute to the overall reactivity of AFMK and AMK toward •OH, albeit the contributions of HT are larger to some extent. In benzene solution, the contributions of the RAF and HT mechanism were found to be 38.0% and 62.0% for AFMK and 44.6% and 55.4% for AMK. In aqueous solution, on the other hand, the contributions of RAF become larger than those of HT, although both mechanisms remain important. It was found that the RAF contributions to the overall reactivity of AFMK and AMK in such medium are 55.7% and 37.8%, respectively, while those of the HT mechanism are 42.1% and 30.5%. In addition, in aqueous solution, SET becomes relevant for the •OH scavenging activity of AMK, with ΓSET = 31.8%, while for AFMK, the contribution of SET is low but significant (ΓSET = 2.2%).

Table 8. Branching ratios of the different channels of reaction, at 298.15 K, in benzene solution
path AFMK AMK
•OH •OOH •OOCCl3 •OH •OOH •OOCCl3
HT-C1 2.71 6.42
HT-C3 25.67 13.13
HT-C4 3.51 98.70 0.05 13.70 16.33 ~0.00
HT-C6 14.65
HT-C13 15.44 8.94
HT-N1 ~0.00 0.01
HT-N2 0.05 13.17
RAF-C7 3.99 2.16
RAF-C8 ~0.00 0.83
RAF-C9 12.50 1.30 99.95 10.59 83.67 ~100.00
RAF-C10 9.25 10.58
RAF-C11 7.62 9.88
RAF-C12 4.61 10.60
Table 9. Branching ratios of the different channels of reaction, at 298.15 K in aqueous solution
path AFMK AMK
•OH •OOH •OOCCl3 •OH •OOH •OOCCl3
SET 2.22 0.00 13.63 31.76 19.07 62.78
HT-C1 2.17 0.40
HT-C3 14.50 0.07 9.03 0.00
HT-C4 3.39 87.25 0.00 5.87 0.56 0.00
HT-C6 7.48
HT-C13 4.97 5.52
HT-N1 0.00 0.00
HT-N2 9.54 9.64 17.86
RAF-C7 11.02 1.54
RAF-C8 0.09 6.29
RAF-C9 12.13 12.75 86.29 0.94 80.37 13.66
RAF-C10 12.27 0.18
RAF-C11 8.06 3.26
RAF-C12 12.15 1.71 5.71

The reactions of both AFMK and AMK with •OOCCl3 were found to take place almost exclusively by the RAF mechanism (at site C9) in nonpolar media. In aqueous solution, the main reaction mechanism is also RAF (86.3%) for the •OOCCl3 scavenging activity of AFMK, but SET also plays a significant role (13.6%). For the •OOCCl3 scavenging activity of AMK, in aqueous solution, SET becomes the main mechanism of reaction, with contributions to the overall reactivity equal to 62.8%. In this case, the contributions of HT and RAF mechanisms are similar and equal to 17.9% and 19.4%, respectively.

Regarding the reactions with •OOH, in nonpolar media, the main mechanism of reaction for AFMK is HT, with contributions larger than 98%. For AMK, on the other hand, RAF is the main mechanism of reaction (83.7%), while HT accounts for 16.3% of the overall reactivity. In aqueous solution, HT still is the main mechanism involved in the •OOH scavenging activity of AFMK (87.3%), but the relative importance of RAF increases (12.7%). For AMK, in aqueous solution, SET becomes a significant mechanism (19.1%), while the importance of HT becomes almost negligible (0.6%), and RAF retains its preponderant role (80.4%).

According to the analysis of the branching ratios, the environment plays a role not only on the overall reactivity of AFMK and AMK toward free radicals, but also on the relative importance of the different mechanisms of reaction.

Conclusions

The pKa values of AFMK and AMK in aqueous solution were estimated, for the first time, and proposed to be 8.7 and 16.8, respectively. According to these values, both species are expected to be mainly in their neutral form, under physiological conditions (pH = 7.4).

The free radical scavenging activity of these two metabolites of melatonin has been investigated considering three mechanisms of reaction: radical adduct formation (RAF), hydrogen transfer (HT), single electron transfer (SET) and environments of different polarity. For that purpose, three different free radicals were used (•OH, •OOH, and •OOCCl3).

It was found that AFMK and AMK react with •OH at diffusion-limited rates (~1010 /m/s), regardless of the polarity of the environment, which supports their excellent •OH radical scavenging activity. These compounds were found to be also very efficient for scavenging •OOCCl3, but rather ineffective for scavenging •OOH.

It was found that AFMK is a poorer scavenger than AMK and melatonin, regardless of the polarity of the environment and of the free radical with which they are reacting. Accordingly, the efficiency of melatonin for scavenging free radicals is predicted to be reduced when it is metabolized to AFMK.

The relative efficiency of AMK, with respect to melatonin, was found to be influenced by the polarity of the environment. In aqueous solution, AMK was found to react faster than melatonin with all the studied free radicals. Therefore, the efficiency of melatonin for scavenging these radicals in aqueous solution is predicted to be increased when it is metabolized to AMK.

In nonpolar environments, the relative efficiency of AMK and melatonin as free radical scavengers depends on the radical with which they are reacting. Under such conditions, AMK was found to react faster than melatonin with •OH, while melatonin reacts faster with •OOH and •OOCCl3. Therefore, in lipid media, melatonin is predicted to be a better •OOH and •OOCCl3 scavenger than AMK, while AMK is predicted to be slightly better than melatonin for scavenging •OH.

Accordingly, it seems that melatonin and its metabolite AMK constitute an efficient team of scavengers capable of deactivating a wide variety of ROS, under different conditions. Moreover, these findings support the continuous protection exerted by melatonin, through the free radical scavenging cascade 12, 21.

Regarding the relative importance of the different mechanisms of reaction, it was found to be influenced by the polarity of the environment for the studied peroxy radicals, while for •OH, a wide product distribution is expected, regardless of the conditions under which the reactions take place. For this particular free radical, all the studied reaction mechanisms significantly contribute to the overall scavenging activity of AFMK and AMK.

In nonpolar environments, the protective effects of AMK take place mainly through the RAF, while in aqueous solution, the SET mechanism becomes particularly important for the reaction of AMK with the most electrophilic of the studied radicals (•OOCCl3); in fact in this case, SET is the main mechanism of reaction.

Acknowledgements

A. G. thanks Laboratorio de Visualización y Cómputo Paralelo at UAM – Iztapalapa for the access to its computer facilities and project SEP-CONACyT 167491.

      The full text of this article hosted at iucr.org is unavailable due to technical difficulties.