Why Monoamine Oxidase B Preferably Metabolizes N-Methylhistamine over Histamine: Evidence from the Multiscale Simulation of the Rate-Limiting Step

Abstract

Histamine levels in the human brain are controlled by rather peculiar metabolic pathways. In the first step, histamine is enzymatically methylated at its imidazole N^τ atom, and the produced N-methylhistamine undergoes an oxidative deamination catalyzed by monoamine oxidase B (MAO-B), as is common with other monoaminergic neurotransmitters and neuromodulators of the central nervous system. The fact that histamine requires such a conversion prior to oxidative deamination is intriguing since MAO-B is known to be relatively promiscuous towards monoaminergic substrates; its in-vitro oxidation of N-methylhistamine is about 10 times faster than that for histamine, yet this rather subtle difference appears to be governing the decomposition pathway. This work clarifies the MAO-B selectivity toward histamine and N-methylhistamine by multiscale simulations of the rate-limiting hydride abstraction step for both compounds in the gas phase, in aqueous solution, and in the enzyme, using the established empirical valence bond methodology, assisted by gas-phase density functional theory (DFT) calculations. The computed barriers are in very good agreement with experimental kinetic data, especially for relative trends among systems, thereby reproducing the observed MAO-B selectivity. Simulations clearly demonstrate that solvation effects govern the reactivity, both in aqueous solution as well as in the enzyme although with an opposing effect on the free energy barrier. In the aqueous solution, the transition-state structure involving histamine is better solvated than its methylated analog, leading to a lower barrier for histamine oxidation. In the enzyme, the higher hydrophobicity of N-methylhistamine results in a decreased number of water molecules at the active side, leading to decreased dielectric shielding of the preorganized catalytic electrostatic environment provided by the enzyme. This renders the catalytic environment more efficient for N-methylhistamine, giving rise to a lower barrier relative to histamine. In addition, the transition state involving N-methylhistamine appears to be stabilized by the surrounding nonpolar residues to a larger extent than with unsubstituted histamine, contributing to a lower barrier with the former.

Keywords: DFT calculations; N-methylhistamine; QM/MM; activation free energy; empirical valence bond; histamine; metabolic pathway; monoamine oxidase B; multiscale molecular simulations; rate constant; selectivity.

Scheme 1

Rate limiting step of histamine (R = H) and N-methylhistamine (R = CH₃) oxidation catalyzed by the MAO-B enzyme, proceeding by the hydride transfer mechanism.

Figure 1

Free energy profiles for the calibrated EVB gas-phase simulations for histamine (HIS/blue) and N-methylhistamine (NMH/red), each given in a batch of 10 replicas. ε is the generalized reaction coordinate defined by the difference between the potential surfaces of the reactant and product states. The tunable EVB parameters were determined by fitting such that the activation barrier and reaction free energy averaged over the replicas match the values obtained by DFT calculations (Table 1).

Figure 2

EVB free energy profiles of histamine (HIS/blue) and N-methylhistamine (NMH/red) oxidation in aqueous solution, both acquired from 10 simulation replicas. The average barrier and reaction free energy are listed in Table 2. ε is the generalized reaction coordinate defined by the difference between the potential surfaces of the reactant and product state. Note that the displayed profiles pertain to the reaction step and do not include the contribution associated with deprotonation of the substrate. The corresponding deprotonation corrections are 3.20 and 2.95 kcal/mol for HIS and NMH, respectively. The total free energy barriers are listed in Table 2.

Figure 3

Running average number of water molecules within 7 Å of the C4 atom of the imidazole ring (marked with purple star) of histamine (HIS/blue) and N-methylhistamine (NMH/red) computed over reaction simulation.

Figure 4

EVB free energy profiles for histamine (HIS/blue) and N-methylhistamine (NMH/red) oxidation in monoamine oxidase B (MAO-B), both acquired from 10 simulation replicas. The average barrier and reaction free energy are listed in Table 2. ε is the generalized reaction coordinate defined by the difference between the potential surfaces of the reactant and product state. Note that the displayed profiles pertain to the reaction step and do not include the contribution associated with deprotonation of the substrate. The corresponding deprotonation corrections are 3.20 and 2.95 kcal/mol for HIS and NMH, respectively. The total free energy barriers are listed in Table 2.

Figure 5

Average number (in 0.25 ns blocks) of hydrophobic residues within 5 Å of the common ring nitrogen atom (N^τ) of both substrates (histamine: HIS/blue; N-methylhistamine: NMH/red) during simulation of the reaction.

Figure 6

N-methylhistamine (NMH) in the active site of monoamine oxidase B during reaction simulation, with surrounding hydrophobic residues and Tyr326 aromatic residue indicated.

Figure 7

Structure of the solvated monoamine oxidase B (MAO-B) with the reacting N-methylhistamine (NMH) and flavin adenine dinucleotide (FAD) prosthetic group in a spherical simulation cell centered at the N5 atom of FAD.