The radical scavenging activity of glycozolidol in physiological environments: a quantum chemical study

Glycozolidol was isolated from the root of Glycosmis pentaphylla (6-hydroxy-2-methoxy-3-methylcarbazole, GLD). This molecule attracted considerable interest due to its beneficial biological activities that likely stem from its antioxidant activity; yet, the radical scavenging action of GLD has not been investigated thus far. In this study, DFT calculations were used to estimate the radical scavenging activity of GLD against a variety of biologically significant radical species in physiological environments. The findings demonstrated that GLD exerts significant antiradical activity in water at pH = 7.40 and in pentyl ethanoate (as a model of lipidic media) with koverall = 8.23 × 106 and 3.53 × 104 M−1 s−1, respectively. In aqueous solution, the sequential proton-loss electron transfer mechanism made the highest contribution to the activity, whereas in nonpolar solvents the formal hydrogen transfer mechanism dominated the activity. GLD is predicted to have strong antiradical activity against CH3O˙, CH3OO˙, CCl3OO˙, NO2, SO4˙−, DPPH and ABTS+˙ kapp ≈ 109 M−1 s−1 and kf ≈ 106 M−1 s−1. The results suggest that GLD is a good radical scavenger in physiological environments.


Introduction
Antioxidants attract considerable interest owing to their roles in important biological processes, and therefore their preferential inclusion in food and pharmaceutical products. 1,2 They protect against oxidative deterioration in the body and thus against oxidative stress-induced pathological processes such as cancer, aging, hypo sexuality, hyperlipidemia, cardiovascular disease, inammation, and many others. 1,3,4 It is estimated that more than 10 000 individual phytochemicals have been identied in plants, including polysaccharides, phenolics, triterpenoids, steroids, carotenoids, vitamins, essential oils, and alkaloids. Among the great structural diversity of phytochemicals, alkaloid components have attracted considerable attention for known or suspected activity in treating various diseases. 4,5 Thus, highly substituted carbazole alkaloids are active against e.g. neurodegenerative diseases, cancer, tuberculosis and Human Immunodeciency Virus infections. 6,7 Among these compounds, glycozolidol (6-hydroxy-2-methoxy-3-methylcarbazole) ( Fig. 1) was isolated from the roots of Glycosmis pentaphylla and since conrmed to exert antibacterial (against both Gram-positive and Gram-negative bacteria) and antifungal activities. 8,9 Glycozolidol belongs to a class of natural aromatic nitrogen heterocyclic alkaloids. Based on the oxygenation pattern of tricyclic carbazole alkaloids, glycozolidol could provide robust antioxidant activity due to the presence of amine and quinone groups, however, there are no reports on the mechanism and/or kinetics of the radical scavenging activity of glycozolidol. Herein, we explore the effects of solvent environments and molecular structures on the antioxidant activity and oxidation resistance of glycozolidol against a range of free radicals by using thermodynamic and kinetic calculations. the M06-2X functional and the 6-311++G(d,p) basis set. 11 The M06-2X functional is one of the most reliable methods for computing the thermodynamics and kinetics of radical reactions. 12,13 This method is oen used to evaluate the radical scavenging activity of natural and synthesized compounds 12,14-17 due to its reliable predictions when compared to experimental data. [18][19][20][21][22] The solvent effects of water and pentyl ethanoate were predicted using the SMD technique, 23 following a wellestablished practice in modelling the radical scavenging activity of antioxidants. 12,21,24 The proton affinity (PA), bond dissociation enthalpy (BDE), and ionization energy (IE) values were calculated as follows. 21 where H(e), H(H + ), H(Hc), H(GLD-H + c), H(GLD − ), H(GLDc), and H(GLD-H) are the relative enthalpies of the electron, proton, hydrogen atom, cation-radical, anion, radical and neutral molecule, respectively. The quantum mechanics-based test for overall free radical scavenging activity (QM-ORSA) protocol 20 was used to complete the kinetic calculations. The rate constants (k) were calculated using conventional transition state theory (TST) at 1 M standard state, 298.15 K, following eqn (4). 21,25-31 where s is the reaction symmetry number, 32,33 k stands for tunneling corrections that were calculated using Eckart barrier, 34 k B is the Boltzmann constant, h is the Planck constant, DG s is Gibbs free energy of activation. The Marcus theory was used to calculate the reaction barriers of single electron transfer (SET) reactions in media. 35,36 The eqn (5) and (6) were used to compute the Gibbs free energy change of reaction DG s for the SET reaction.
Where DG 0 SET represents the conventional Gibbs free energy change of the reaction and DE represents the nonadiabatic energy difference between reactants and vertical products for SET. 37,38 For rate constants around the diffusion limit, a modication was made. 20 Collins-Kimball theory was used to calculate the apparent rate constants (k app ) for an irreversible bimolecular diffusion-controlled reaction in solvents at 298.15 K, 39 D AB = D A + D B (denotes the mutual diffusion coefficient of A and B), 39,41 where D A or D B is obtained using the Stokes-Einstein formulation eqn (9). 42,43 h is the viscosity of the solvents (i.e. h(pentyl ethanoate) = 8.62 × 10 −4 Pa s, h(H 2 O) = 8.91 × 10 −4 Pa s) and a is the radius of the solute.
The existence of just one imaginary frequency served as a distinguishing feature across all of the transition stages. Intrinsic coordinate calculations were carried out to ensure that each transition state has an accurate connection to both the preand post-complexes.
To avoid over-penalizing entropy losses in solution, Okuno's corrections were applied, 44 which were then updated with the free volume theory utilizing the Benson correction. 20,[45][46][47] Values for pK a are computed in accordance with Galano et al.

The thermodynamic study
The GLD (GLD-H) can react with free radicals (Rc) via either of the three main pathways: -Formal hydrogen transfer (FHT) 21,49,50 -Sequential electron transfer-proton transfer (SETPT) 51,52 -Sequential proton loss electron transfer (SPLET) 52-55 In the initial step, the characteristic thermodynamic parameters (BDE, PA, IE) that dene the energy barrier of the rst step of each mechanism were calculated for all bonds in the HOOc radical scavenging activity of GLD in the gas phase (the standard medium for computational chemistry), in pentyl ethanoate for a lipid medium, and in water at pH 7.4 for an aqueous physiological environment. in the gas phase, pentyl ethanoate and water, respectively). The SPLET and SETPT mechanisms are not favorable in the gas phase or pentyl ethanoate solvent due to the higher PA and IE values compared with the BDE(O9-H). Thus it is expected that the O9-H bond will play a dominant role in the radical scavenging activity of GLD following the FHT reaction in all of the studied media; however, in the aqueous solution, the SPLET mechanism may have a major contribution in the radical scavenging activity if the spontaneous dissociation of the acidic protons eliminates the energy barrier of the rst step.
The Gibbs free energies of the GLD + HOOc reaction following the proton loss (PL-the rst step of SPLET), the FHT, and single electron transfer (SET-the rst step of SETPT) mechanisms were calculated in order to eliminate pathways that are not spontaneous thermodynamically, and rate constants were only calculated for the spontaneous reactions. The results are shown in Table 2. The results showed that the process was spontaneous only according to the FHT mechanism, especially at the O9-H bond (DG 0 = −7.3 to −3.6 kcal mol −1 ). The DG 0 values of the H-abstraction of the N-H bond were 3.0, 3.7 and 0.1 kcal mol −1 in the gas phase, pentyl ethanoate and water, respectively, while the lowest DG 0 for the C-H bonds was observed at the C14-H bond with 4.1, 4.8 and 1.7 kcal mol −1 in the gas phase, pentyl ethanoate and water, respectively. The FHT reactions of the other C-H bonds were not spontaneous (DG 0 = 10.6-53.1 kcal mol −1 ). Similarly, neither the proton loss nor SET reactions were spontaneous in any of the investigated solvents. Thus, in nonpolar environments the HOOc radical scavenging activity follows the FHT pathway, however in polar media such as water at pH = 7.40, the deprotonation of GLD needs also be addressed.

The kinetic study
3.2.1. Acid-base equilibrium. The conjugate base form oen exhibits much higher activity than the protonated form of acidic species in aqueous environments. 12,17 To identify the most probable radical scavenging activity, the protonation state of GLD was initially examined at physiological pH. The GLD structure permits protonation at the N-H and O9-H bonds in accordance with reactions (1) and (2); hence, the pK a values of GLD were determined based on the published literature 48 and are depicted in Fig. 2.
The computed pK a values for the amine were pK a1 = −2.56 (for the cation form of the N-H bond) and pK a2 = 10.64 (for the O9-H bond). In an aqueous solution with a pH of 7.4, GLD exists in two states: neutral (HA, 99.9%) and anion (A − , 0.1%). As a consequence of this, both states were considered during the kinetic analysis of the HOOc radical activity of GLD in water  Table 2 The computed DG 0 (in kcal mol −1 ) of the HOOc + GLD following the formal hydrogen transfer (FHT), proton loss (PL) and single electron transfer (SET) reactions in the studied media (G: the gas phase; P: pentyl ethanoate; W: water)  1 and 2) showed that the HOOc antiradical activity of GLD in the nonpolar media is dominated by the hydrogen transfer reaction of the N-H and O9-H bonds (DG 0 z 0 kcal mol −1 ). The H-abstraction of the C14-H bond (DG 0 = 1.7-4.8 kcal mol −1 ) should be omitted due to the lower HOO $ radical scavenging activity of C-H bonds compared with the N-H and O9-H bonds. 17,51 However, in the aqueous solution, the SET reaction of the anion state should be considered. 12,24,50 Therefore, the total rate constant (k overall ) of GLD antiradical activity against the HOOc radical in the gas phase, pentyl ethanoate and water can be calculated using eqn (17) and (18). Table 3 and Fig. 3 show the nal results.
Nonpolar environments: Water at physiological pH: The nuclear reorganization energy (l, in kcal mol −1 ). Fig. 3 The TS structures of the HOOc + GLD reaction following the FHT pathway; imaginary frequencies given in cm −1 , bond lengths inÅ and bond angles in°.

The antiradical activity of GLD against ordinary free radicals in aqueous solution following the SET reaction
Although HOOc scavenging activity is a valuable comparative metric, the antiradical activity against different radicals oen vary in a broad range. Therefore, the radical scavenging activity of GLD was subsequently modeled against a variety of common free radicals, including HOOc, CH 3 OOc, CCl 3 OOc, HOc, CH 3 Oc, CCl 3 Oc, NO, NO 2 , O 2 c − , SO 4 c − , N 3 c, ABTS + c and DPPH. The hydroperoxyl radical scavenging activity of GLD (G = 96.0%) is determined by the SET mechanism. Consequently, in this investigation, the antiradical activity against these radicals in water at pH = 7.4 was evaluated using the SET mechanism and the ndings are shown in Table 4.
GLD should have high activity against CH 3 Oc, CH 3 OOc, CCl 3 OOc, NO 2 , SO 4 c − , DPPH and ABTS + c radicals with k app z 10 9 M −1 s −1 and k f z10 6 M −1 s −1 , whereas NO, O 2 c − HOc, CCl 3 Oc, and N 3 c radicals cannot be eliminated by GLD under the studied conditions. By the SET reaction, GLD is less active against CCl 3 Oc, and N 3 c radicals than fraxin 55 or usnic acid, 56 but more effective against HOOc and CH 3 OOc radicals. For HOc, however, the prediction of low activity suggests that SET is not the correct mechanism; highly reactive radicals are known to follow alternative pathways (FHT or radical adduct formation to the neutral species) that for these radicals are essentially barrierless in aqueous media. Thus our results also highlight the limits of generalizations from one detailed study. 50

Conclusion
The antiradical activity of GLD against HOOc, CH 3 OOc, CCl 3 -OOc, HOc, CH 3 Oc, CCl 3 Oc, NO, NO 2 , O 2 c − , SO 4 c − , N 3 c, ABTSc + and DPPH was studied using DFT calculations. In the physiological environments GLD exhibited signicant antiradical activity. The overall rate constants for the antiradical trapping of GLD in water at pH = 7.40 and pentyl ethanoate were k overall = 8.23 × 10 6 and 3.53 × 10 4 M −1 s −1 , respectively. The SPLET mechanism made contributions to the activity in water at pH 7.40, however, the FHT mechanism characterized the activity in nonpolar solvents. Additionally, it was found that GLD has strong antiradical activity against CH 3 Oc, CH 3 OOc, CCl 3 OOc, NO 2 , SO 4 c − , DPPH and ABTS + c k app z 10 9 M −1 s −1 and k f z 10 6 M −1 s −1 . According to the computed results, GLD is more effective at trapping HOO $ than reference antioxidants like trolox and BHT in the physiological environment. The results suggest that GLD can join the long list of phytochemicals with good radical scavenging activity in physiological environments, emphasizing yet again the importance of varied plant sources in diets and in dietary supplements to maintain health and prevent disease.

Conflicts of interest
There are no conicts to declare. Table 4 Calculated kinetic data between GLD-O9-ANION (A − ) and the selected radicals (DG s , l in kcal mol −1 ; k D , k app and k f in M −1 s −1 ) a