Comparative study of aflatoxin B₁ adsorption by Thai bentonite and commercial toxin binders at different temperatures in vitro

S Wongtangtintan, L Neeratanaphan, P Ruchuwararak, S Suksangawong, U Tengjaroenkul¹, P Sukon and B Tengjaroenkul

Khon Kaen University, Khon Kaen 4002, Thailand
btengjar@kku.ac.th
¹ Chiang Mai University, Chiang Mai 50200, Thailand

Abstract

Thai bentonite (TB) and two commercial toxin binders including commercial bentonite (CB) and activated carbon (AC) were investigated for their adsorption capacities of aflatoxin B₁ (AFB₁) at different temperatures in vitro. Each sample of 5 mg/l AFB₁ solution was shaken at 25^o, 37^o and 45^oC for 24 hours, and the supernatants of centrifuged samples were analyzed for concentrations of AFB₁ using a UV spectrophotometer at a wavelength of 362 nm. The results indicated that TB is capable of sequestering AFB₁ from aqueous solutions and had significantly greater adsorption capacities than CB and AC (p<0.05). The adsorption capacities were then calculated and applied to isotherm equations. The linearized Langmuir and linearized Freundlich adsorption isotherm equations indicated that TB was the best toxin binder for adsorbing AFB₁ as demonstrated by its significantly higher estimated maximum binding capacity (p<0.05), the distribution coefficient and the heterogeneity factor.  The mean adsorptions of AFB₁ on all binders were highest at 25^oC, whereas adsorbed significantly decreased when temperatures were increased from 25^oC to 45^oC (p<0.05). Furthermore, the results revealed that the Freundlich model presented a better fitted to the experimental data than the Langmuir model. This implied that the adsorption behavior of AFB₁ on these toxin binders represented multilayer/multiple site adsorption on the binders’ surfaces. The results support the conclusion that TB adsorbs AFB₁ in vitro more efficiently than other commercial toxin binders, especially at 25^oC.

Keywords: aflatoxin, binder, charcoal, clay, fungal toxin, model

Introduction

Aflatoxins (AF) are toxic substances produced by Aspergillus flavus and A. parasiticus. Aflatoxin B₁ (AFB₁) is widely known as the most liver toxic among AF (Hueber et al 2004, Godfrey et al 2013). One approach to detoxify AF is to combine adsorbent to animal feed to inhibit the bioavailability of the toxin absorbed through the alimentary tract (Basalan et al 2006). This approach is practical, especially when when used with contaminated feed (Hueber et al 2004, Pimpukdee et al 2004, Kossolova et al 2009). Several reports have shown that adsorbents, such as activated charcoal, aluminosilicate, bentonite and zeolite effectively adsorb AFB₁ in vivo (Rosa et al 2001, Khajarern et al 2003, Pasha et al 2007, Manafi 2011, Khadem et al 2012, Rao and Chopra 2012, Sadeghi et al 2012, Neeff et al 2013).

Previous reports have applied methods of isothermal analysis to characterize the adsorption of AF onto the surfaces of toxin binders (Grant and Phillips 1998, Pimpukdee et al 2000, Hinz 2001). These methods provide evidence of the molecular mechanisms involved with different binders and allow for comparisons of similarities and differences. Binders having distinctive molecular structures bind AFB₁ differently (Phillips 1999). Reports comparing the binding capacity of Thai clays with commercial binders for adsorption capacities of AFB₁ has been limited. Thus, the main objective of this study was to investigate the adsorption capacities and affinities of Thai bentonite from Lumphun Province and to compare it to other commercial toxin binders at different temperatures using Langmuir and Freundlich isotherm modeling approaches (Tengjaroenkul et al 2011).

Materials and Methods

Chemicals

Standard AFB₁ was purchased from Sigma Chemical Co. (St. Louis, USA) and two commercial binders were obtained from Thai suppliers. Thai bentonite (TB) was collected from Lumphun Province having the highest adsorptive capacity (4.68x10^-3 mol/kg) among 20 clay samples from different regions in Thailand (Tengjaroenkul et al 2011). All binders were sieved to achieve particle sizes less than 60 µm. Highly purified water was prepared by processing deionized water through a Milli-Q^uf+ system.

Construction of the aflatoxin B₁ calibration curves

The standard curves for AFB₁ determination were constructed by using a series of AFB₁ solutions having the concentrations of 0.50, 1.00, 1.50, 2.00, 2.50, 3.00, 3.50, 4.00, 5.00, 6.50, and 8.00 ppm, respectively. Each standard solution was measured for absorbance at 362 nm by UV-VIS spectrophotometer, and subsequently the calibration graph was obtained by plotting the absorbance versus AFB₁ concentration.

Study of adsorption isotherms

Three different toxin binders including the TB as well as the commercial bentonite (CB) and activated carbon (AC) from the Thai supplier were used in this study. A 10.0 mg of toxin binder was weighed out in a clean borosilicate test tube. To each tube, the 4 ml of AFB₁ solution with each concentration of 0.50, 1.00, 1.50, 2.00, 2.50, 3.00, 3.50, 4.00, 5.00, 6.50, and 8.00 ppm were added to the toxin binder, giving a final concentration of 0.25% w/v. The mixtures were shaken at 200 x g for 24 hours at 25°C (Innova 4060 Shaker, New Brunswick Scientific, USA). Toxin control containing 4 ml of each AFB₁ concentration without toxin binder were also prepared. The control and samples were centrifuged (Beckman Coulter, USA) at 12,000 x g for 30 min, to separate the toxin binder from the supernatant. The adsorbed amount of AFB₁ onto each toxin binder was investigated from the concentration of AFB₁ left in the solution determined by using UV-VIS spectrophotometer (Perkin Elmer, USA) at a wavelength of 362 nm (Grant and Phillips 1998, Pimpukdee et al 2000). All experiments were performed in triplicate. The data were applied to fit isotherm equations to characterize the adsorption behavior of AFB₁ onto each toxin binder.

Later, the experimental data were transferred to the Microsoft Excel^® program and fitted with the linearized Langmuir and linearized Freundlich models, and to compare these to the generalized Langmuir model (GLM), the generalized Freundlich model (GFM), and the modified Freundlich model (MFM) (Mayura et al 1998, Pimpukdee et al 2000, Doll et al 2004, Tengjaroenkul et al 2011) using 2D Table Curve program (Jandel Scientific, USA), which could characterize the adsorption behavior of AFB₁ on each toxin binder.

Results

Construction of the AFB₁ calibration curves

In order to determine the amounts of AFB₁ concentrations left in all solutions, the calibration curve of the standard AFB₁ solutions was established by plotting the standard AFB₁ concentration (mol/L)(x-axis) against the absorbance (y-axis). In this work, the standard AFB₁ solutions with concentrations ranging from 0.50 to 8.00 ppm were employed to construct the calibration curve. Each standard AFB₁ solution was measured by using UV-VIS spectrophotometry for absorbance at a wavelength of 362 nm. Figure.1 presented the calibration curve and the adsorption data of AFB₁ adsorption at 25^oC.

Figure 1: Calibration curve of the standard AFB1 solution at 25oC.

Study of adsorption isotherms

In this work, toxin binders were examined for adsorption of AFB₁ in vitro. Adsorption capacities were obtained by plotting the adsorbed AFB₁ versus the concentration of AFB₁ at equilibrium. The data representing the mean adsorption of AFB₁ onto the binders from three replicate experiments at 25^oC are shown in Figure 2.

Figure 2: Mean adsorption capacities for AFB1 on: (a) TB, (b) CB and (c) AC, at 25oC.

The plots in Figure 2 present the adsorption levels of AFB₁ on the different binders as a function of increasing aqueous AFB₁ concentrations after equilibrium for 24 hr. The extent of adsorption increased with the initial concentration increasing from 0.50 to 8.00 ppm. Overall, the adsorption behaviors of AFB₁ were similar, with differences in the amounts adsorbed. The linearized Freundlich and linearized Langmuir plots for AFB₁ adsorption onto the TB, CB and AC can be seen in Figure 3 and 4, respectively. The comparison of fitting characteristics achieved for the two linearized isotherm models for three toxin binders at 25^oC are also shown in Table 1.

Figure 3: The linearized Langmuir plots for AFB1 adsorption on: (a) TB, (b) CB and (c) AC, at 25oC.

Figure 4: The linearized Freundlich plots for AFB1 adsorption on: (a) TB, (b) CB and (c) AC, at 25oC.

Table 1 reveals that the Freundlich model of three binders resulted in correlation coefficients (r²) around 0.9 or more, whereas the r² of the Langmuir model of all binders were lower than 0.75. The adsorption capacities of the TB, CB and AC were around 1.82x10¹⁶, 1.35x10⁹ and 9.27x10⁷ mol/kg, respectively, and the affinity constants for the commercial toxin binder, the TB, CB and AC were 3.39, 2.12 and 1.63, respectively. From the values of adsorption capacity and affinity constants, it was found that the TB was most effective for adsorbing AFB₁ from aqueous solutions, followed by the CB and AC, respectively, and had significantly greater adsorption capacities than CB and AC (p<0.05).

In order to understand the nature of adsorption occurring on the surface of the toxin binder, adsorption isotherm at different temperatures were applied in this study. The experiments were carried out at three different temperatures: 25^o, 37^o and 45^oC. The data from isothermal analysis were calculated and plotted as the amount of AFB₁ adsorbed (q) versus the AFB₁ concentration (C_e). Adsorption isotherms of AFB₁ on the TB, CB and AC at 25^o, 37^o and 45^oC are illustrated in Figure 5. Furthermore, it was found that the adsorbed amount of AFB₁ significantly decreased when temperature was increased from 25^oC to 45^oC (p<0.05). The highest degree of adsorption of AFB₁ obtained at 25^oC.

Figure 5: Adsorption isotherms of AFB1 for binders: (a) TB, (b) CB and (c) AC, at 25o, 37o and 45oC. TB: Thai bentonite, CB: Commercial bentonite, AC: Commercial activated carbon.

The isotherms of the linearized Langmuir and Freundlich models were fitted to the experimental data to obtain the values of the adsorption capacity and affinity constants for all toxin binders. These two approaches at different temperatures are summarized and presented in Table 2.

The results in Table 2 demonstrated that the adsorption capacity for AFB₁ increased as the temperature decreased. The high correlation coefficients (r²) of all toxin binders in the linearized models were much greater in the Freundlich model than the Langmuir model.

The adsorption ability of each toxin binder for AFB₁ was significantly different when considering the adsorption capacity and affinity constants (p<0.05)(Table 2). The TB had greatest adsorption ability for AFB₁, and had both the highest adsorption capacity and affinity constants for AFB₁. The CB demonstrated good adsorption capability, whereas, AC adsorbed less AFB₁ from the aqueous solution.

Experimental data were also transferred to the 2D Table Curve program. The isotherm equations in Table 3 to 4 were entered as user-defined functions. Each function has limits and beginning values or first approximations for the variable parameters. The values of the estimated maximum capacity (Q_max) and the distribution coefficient (K_d) were obtained from the double-logarithmic plot of these isotherms. The plots normally present a break in the curve. The value on the x-axis is an estimation of K_d^-1 and Q_max is an estimate from the y-axis where the curve breaks. These values were entered into user-defined functions

The best-fit isotherms for AFB₁ adsorption onto three toxin binders at 25^oC are presented in Figure 6.

Figure 6: The best-fit isotherms for the adsorption of AFB1 onto: (a) TB, (b) CB and (c) AC. TB: Thai bentonite, CB: Commercial bentonite, AC: Commercial activated carbon.

From the best-fit isotherms, it was demonstrated that the estimated maximum capacities for the TB, CB and AC were 6.75x10^-2, 4.52x10^-3 and 3.32x10^-2 mol/kg, respectively, and the distribution coefficients for each were 1.43x10¹¹, 2.52x10⁵ and 2.46x10⁶, respectively. The heterogeneity factor for the TB, CB and AC were 1.21, 2.82 and 3.19, respectively (Table 5). Among the studied toxin binders, TB was the best toxin binder for the adsorption of AFB₁ from aqueous solutions as demonstrated by the significantly higher estimated maximum capacity (p<0.05), the distribution coefficient and the heterogeneity factors (Table 5).

Discussion

Study of absorption isotherms

The linearized Langmuir and linearized Freundlich adsorption isotherm data representing the mean adsorption of AFB₁ onto the TB, CB and AC at 25^oC increased as the AFB₁ concentration increased. However, AFB₁ seems to prefer activated carbon for the fastest adsorption. The linearized Langmuir and Freundlich isotherm models were applied to characterize the adsorption behavior, and estimate values of the adsorption capacity and affinity constants for each toxin binder. The Langmuir model is a simple model based on monolayer adsorption onto a surface, while the Freundlich model is an empirical equation used for the description of multilayer adsorption with interaction between adsorbed molecules and heterogeneous surfaces with a uniform energy distribution and reversible adsorption (Hinz 2001, Gimbert et al 2008). The Langmuir constants were calculated from the plots of C_e/q versus C_e. The Freundlich constants, the adsorption capacity (K_f) and the affinity constants (n^f), were obtained from the linear plot of log q against log C_e. The intercept of the line is an indicator of the adsorption capacity, and the slope is an indication of the affinity constant. The value of Freundlich’s constant can be correlated and used to predict the adsorption capability of the binder.

Table 1 reveals that the Freundlich model represented a better fit for the experimental data than the Langmuir model, as indicating from the high values of correlation coefficients (r²). This implied that the adsorption behavior of AFB₁ on these toxin binders were multilayer/multiple site adsorptions on heterogeneous surfaces. The adsorption was not limited to a monomolecular layer, but could continue to adhere to the multi-molecular layers of the toxin binder surface (Ruthven 1984).

It was found that TB was highly effective for the adsorption AFB₁ from aqueous solutions as it had the highest adsorption capacity and affinity constants, followed by the CB and AC, respectively. TB is a clay-based material consisting mainly of montmorillonite and it demonstrated the highest cation exchange capacity (CEC) and conductivity. In general, montmorillonite clays are very soft phyllosilicate minerals that typically form microscopic crystals (Gimbert et al 2008). They can swell considerably more than other clays with the addition of water. TB also has the property of adsorbing organic substances either on its external surfaces or within it inter-laminar spaces by the interaction with or substitution for the exchange cations presented in these spaces (Ramos and Hernandez 1996, Abdel-Whalab et al 2002). Therefore, this binder is highly effective for the adsorption of AFB₁ from aqueous solutions, and it is commonly used as an anticaking agent in animal feeds throughout the world (Yang 2003, Afriyie-Gyawu et al 2005). Similarly, CB demonstrated good adsorption capability for AFB₁ in vitro and this commercial bentonite is also a clay material of hydrated aluminosilicate, primary composed of montmorillonite as its major constituent. Therefore, bentonite has received considerable recognition as a toxin binder similar to other commercial toxin binders.

AC also adsorbed AFB₁ from the aqueous solutions because of its activated carbon surface that is non-polar or only slightly polar, which results in high affinity for non-polar adsorbates, such as organics (Yang 2003). It also has a large surface area and adsorbs compounds primarily by hydrogen bonding (Huwig et al 2001, Lemke et al 2001^a,b). In aqueous solutions, it can adsorb most of mycotoxins efficiently. AC is a relatively unspecific toxin binder that adsorbs not only mycotoxins, but also essential nutrients and it has been shown that high doses of AC are beneficial in acute poisoning situations.

Table 5 presents values of estimated maximum capacity, the distribution coefficient and the heterogeneity factor, obtaining from the best-fit isotherm models. From these results, it shows that the generalized Freundlich model is the best fit for the results of most of the toxin binders, as demonstrated by the higher correlation coefficients at 25^oC. The generalized Freundlich model is a basic equation used mostly to explain multilayer adsorption with interaction between adsorbed molecules at low concentrations, whereas the modified Freundlich model is a reduced form used when the equilibrium concentration is very low. These models are sufficiently flexible to give a good representation of the data and appears to reduce the error of the estimated parameters. Consequently, the Freundlich models are better representation of the experimental data than the linearized Langmuir models.

Conclusions

• In conclusion, TB adsorbs AFB₁ in vitro more efficiently than the other commercial toxin binders, particularly at 25^oC and the in vitro analytical method is useful for accurately predicting the efficacy of toxin binders.

Acknowledgements

Financial support was provided by the Higher Education Research Promotion and National Research University Project of Thailand, Office of the Higher Education Commission;   the Research Group on Toxic Substances in Livestock and Aquatic Animals, Khon Kaen University, and Faculty of Veterinary Medicine, Khon Kaen University. The authors thank Dr. Frank F. Mallory for reviewing the manuscript.

References

Abdel-Wahhab M A, Nada S A and Khalil F A 2002 Physiological and toxicological responses in rats fed aflatoxin-contaminated diet with or without sorbent materials. Animal Feed Science and Technology, 97: 209-219

Afriyie-Gyawu E, Mackie J, Dash B, Wiles M, Taylor J, Huebner H, Tang L, Guan H, Wang J S and Phillips T D 2005 Chronic toxicological evaluation of dietary Novasil clay in Spraqudawley rats. Food Additives and Contaminants, 22: 259-269

Basalan M, Gungor T, Aydogan I, Hismiogullari S E, Erat S and Erdem E 2006 Effects of feeding mycotoxin binder (HSCAS) at later ages on gastrointestinal environment and metabolism in broilers. Archiva Zootechinca, 9(1): 5-9

Döll S, Dänicke S,  Valenta H and  Flachowsky G 2004 In vitro studies on the evaluation of mycotoxin detoxifying agents for their efficacy on deoxynivalenol and zearalenone. Archives of Animal Nutrition, 58(4): 311-24

Gimbert F, Morin-Crini N, Renault F, Badot P M, Crini G 2008 Adsorption isotherm models for dye removal by cationized starch-based material in a single component system: Error analysis. Journal of Hazardous Materials, 157: 34-46 

Godfrey S B, David K, Lubega A, Ogwal-Okeng J, Anokbonggo W W and. Kyegombe D B 2013 Review of the Biological and Health Effects of Aflatoxins on Body Organs and Body Systems. In: Aflatoxins-Recent Advances and Future Prospects. MR Abyaneh (ed). Croatia: Intech, 239-265 

Grant P G and Phillips T D 1998 Isothermal Adsorption of Aflatoxin B₁ on HSCAS Clay. Journal of Agricultural Food and Chemistry, 46: 599–605

Hinz C 2001 Description of sorption data with isotherm equations. Geoderma, 99: 225-243

Huebner H J, Herrera P and Phillips T D 2004 Clay-based interventions for the control of chemical and microbial hazards in food and water. In: Preharvest and Postharvest Food Safety-Contemporary Issues and Future Directions. RC Beier, SD Pilai and TD Phillips (eds). Iowa, USA: IFT Press and Blackwell Publishing, 389-402

Huwig A, Freimund S, Kappeli O and Dutler H 2001 Mycotoxin detoxication of animal feed by different adsorbents. Toxicological Letter, 122: 179–188

Khadem A A, Sharift S D, Barati M and Borji M 2012 Evaluation of the effectiveness of yeast, zeolite and active charcoal as aflatoxin absorbents in broiler diets. Global Veterinaria, 8(4): 426-432

Khajarern J, Khaharern S, Moon T H and Lee J H 2003 Effects of dietary supplementation fermented chitin-chitosan (Fermkito) on toxicity of mycotoxin in ducks. Asian-Australian Journal of Animal Science, 16(5): 706-713, from http://jas.fass.org/ 

Kossolova A, Stroka J, Breidbach A, Kroeger K, Ambrosio M, Bouten K and Ulberth F 2009 Evaluation of the effect of mycotoxin binders in animal feed on the analytical performance of standardized methods for the determination of mycotoxins in feed. JRC Scientific and Technical Reports, The European Commission, 5-12

^aLemke S L, Ottinger S E, Mayura K, Ake C L, Pimpukdee K, Wang N and Phillips T D 2001 Development of a multi-tiered approach to the in vitro prescreening of clay-based development of enterosorbents. Animal Feed Science and Technology, 93: 17-29

^bLemke S, Mayura K, Reeves W R, Wang N, Fickey C and Phillips T D 2001 Investigation of Organophilic Montmorillonite Clay Inclusion in Zearalenone-Contaminated Diets Using the Mouse Uterine Weight Bioassay. Journal of Toxicological and Environmental Health, 62: 243-258

Manafi M 2011 Evaluation of different mycotoxin binders on aflatoxin B₁ (Aspergillus parasiticus) produced on rice (Oriza sativa) on fertility, hatchability, embryonic, mortality, residues in egg and semen quality. Advance in Environmental and Biology, 5(13): 3818-3825

Mayura K, Abdel-Wahab M A, Mckenzi K S, Sarr B A, Edwards J F, Naguib K and Phillips T D 1998 Prevention of maternal and developmental toxicity in rats via dietary inclusion of common aflatoxin sorbents: potential for hidden risks. Toxicolgical Science, 41(2): 175-182

Neeff D V, Ledoux D R, Rottinghaus G E, Bermudez A J, Dakovic A, Murarolli R A and Oliveira C A F 2013 In vitro and in vivo efficacy of a hydrated sodium calcium aluminosilicate to bind and reduce aflatoxin residues in tissues of broiler chicks fed aflatoxin B1. Poultry Science, 92(1): 131-137  

Pasha T N, Farooq M U, Khattak F M, Jabbar M A and Khan A D 2007 Effectiveness of sodium bentonite and two commercial products as aflatoxin absorbents in diets for broiler chickens. Animal Feed Science and Technology, 132(1): 103-110

Pimpukdee K, Ake C, Lemke S L, Mayura K and Phillips T D 2000 High affinity sorption of aflatoxin B₁ by hectorite clay. Toxicological Science, 54(1): 143

Pimpukdee K, Kubena L F, Bailey C A, Huebner H J, Afriyie-Gyawu E and Phillips T D 2004 Aflatoxin induced toxicities and depletion of hepatic vitamin A in young broiler chicks: Protection of chicks in the presence of low levels of novasil plus in the diet. Poultry Science, 83(5): 737-744

Phillips T D 1999 Dietary clay in the chemoprevention of aflatoxin-induced disease. Toxicological Science, 52(1): 118-126

Ramos A J and Hernández E 1996 In vitro aflatoxin adsorption by means of a montmorillonite silicate. A study of adsorption isotherms. Animal Feed Science Technology, 62: 263-269

Rao S B N and Chopra R C 2012 Influence of sodium bentonite and activated charcoal on aflatoxin M1 excretion in milk of goats. Small Ruminant Research, 41(3): 203-213

Rosa C A R, Miazzo R, Magnoli C, Salvano M, Chiacchiera S S M, Ferrero S, Saenz M, Carvalho E C Q and Dalcero A 2001 Evaluation of the efficacy of bentonite from the south of Argentina to ameliorate the toxic effects of aflatoxin in broilers. Poultry Science, 80: 139-144

Ruthven D M 1984 Principles of adsorption and adsorption proceses. John Wiley & Sons, New York.

Sadeghi A A, Ahmadi-Mazhin H, Shawrang P and Mohammadi-Sangcheshmeh A 2012 Effects of MOS and heat activated sodium bentonite as aflatoxin absorbents on antibody titers against Newcastle disease and infectious bursal disease viruses in broiler chickens. World Application Science Journal, 18(1): 127-129

Tengjaroenkul B, Tengjaroenku U and Wongtangtintharn S 2011 Study on efficacy of natural Thai clays for adsorption of aflatoxin B1. Livestock Research for Rural Development, 23 (11): Article# 232, from http://www.lrrd.org/

Yang R T 2003 Adsorbents: Fundamentals and Applications. Wiley Interscience, Hoboken, New Jersey, USA.

Received 10 August 2015; Accepted 29 February 2016; Published 1 April 2016

Go to top