Molecular Modeling to Predict and to Verify Smectites' Optimal Mineralogy Properties As Aflatoxin Binders.

Aflatoxin occurrence in agricultural products is unavoidable due to heat, drought, humidity, insect stresses, or the combinations of these factors. Inactivation or removal of the toxins in food and feed or  during biofuel production as the last defensive measure is needed to prevent the harmful effects of aflatoxins to humans and animals. One  economically feasible and environmentally safe solution is to use smectites  as aflatoxin binders in the feed or food. To guide the selection or modification of smectites for this application, both experimental and theoretical evidences are need for the interactions between smectite and aflatoxins. Experimental observations have suggested the importance of size and polarity matching between aflatoxins and the adsorbing sites, but the optimal layer charge density, charge origin, exchange cation of the smectite cannot be evaluated effectively with experimental results due to the limitation of the availability of the smectites. The objective of this study is to use molecular modeling to predict and to verify smectite's , optimal structural properties  that determine the adsorption selectivity and capacity of the clays for aflatoxins.

Previous molecular dynamics (MD) simulations presented in Deng and Szczerba (2011) did not give fully satisfactory results for aflatoxin adsorption under moist condition. Infrared spectra suggested that aflatoxin B1 molecules are adsorbed on clay minerals through water bridges with interlayer smectite cations  under moisture condition (Deng et al., 2010). This was not fully confirmed by results of the MD simulations (Deng and Szczerba, 2011). The main reason was the fact that Aflatoxin B1 molecule has complicated molecular structure, for which standard force fields were not parameterized. The first attempt  of this study  was to parameterize force field for the aflatoxin B1 molecule. Using the CHARMM based force field and  CGenFF program, a less accurate set of  the force field parameters for aflatoxin B1 was generated. the procedure described by Vanommeslaeghe et al. (2010) were used to  further optimize atomic charges, bond and angle terms. The Lennard-Jones parameters of the original force field were kept unchanged. All the simulations used for parameterization of the force field parameters were performed in Gaussian 09 program under RHF/6-31*G+ level of theory.

Having force field properly optimized it was possible to perform simulations of the influence of clays' charge density, clays' exchange cation and clay's charge origin on  aflatoxin B1 sorption. To achieve this aim the following reaction was taken into account:

smectite/water + AfB1/water --> smectite/water/AfB1 + water

The amount of water for smectite/water substrate was taken to correspond to different hydration levels. Simulations for Na, Ca and Ba saturated montmorillonites with 0.4 charge per half of the unit cell were performed. Similar simulations for Ba saturated smectite with charge of 0.3, and 0.5 with different location of the charge (montmorillonite and beidellite) were performed.

                The results of the simulations indicate that aflatoxin spontaneously forms π – π stacking complexes in the interlayer spaces of smectites. Depending on the relative water content and aflatoxin content, monolayer, bilayer or disoriented aflatoxin structures are formed. Exchange cation type,  layer charge density and location  affected aflatoxin B1 adsorption on smectite  mainly through the influence on the interlayer water content, which in very large extent affects thermodynamics of aflatoxin sorption.