Food Chemistry 172 (2015) 143–149
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Interaction of ochratoxin A with quaternary ammonium beta-cyclodextrin Miklós Poór a, Sándor Kunsági-Máté b,c, Lajos Szente d, Gergely Matisz b,c, Györgyi Secenji b,c, }szegi a,⇑ Zsuzsanna Czibulya b,c, Tamás Ko a
Institute of Laboratory Medicine, University of Pécs, Ifjúság útja 13., Pécs H-7624, Hungary Department of General and Physical Chemistry, University of Pécs, Ifjúság útja 6., Pécs H-7624, Hungary János Szentágothai Research Center, Ifjúság útja 20., Pécs H-7624, Hungary d CycloLab Cyclodextrin Research & Development Laboratory, Ltd., Illatos út 7., Budapest H-1097, Hungary b c
a r t i c l e
i n f o
Article history: Received 25 February 2014 Received in revised form 5 August 2014 Accepted 8 September 2014 Available online 16 September 2014 Keywords: Ochratoxin A Cyclodextrin Molecular inclusion Fluorescence spectroscopy Detoxification
a b s t r a c t Ochratoxin A (OTA) is a widely spread nephrotoxic food contaminant mycotoxin. Unfortunately, attenuation or prevention of the toxic effects of OTA is still an unresolved problem. Molecular inclusion of OTA by cyclodextrins (CDs) results in complexes with low stability. In the human organism, OTA exists mostly in the dianionic state (OTA2). Therefore, our major goal was to develop a chemically modified cyclodextrin which gives a more stable complex with OTA than the previously published derivatives and which shows stronger preference towards OTA2. In our fluorescence spectroscopic study we demonstrate that quaternary ammonium beta-cyclodextrin (QABCD) fulfils both of these requirements. The calculated stability constant of the QABCD–OTA2 complex was 28,840 M1 (about 200-fold higher than that of the b-CD–OTA2 complex). We hypothesize, that QABCD may be a suitable tool for the decontamination of different OTA-contaminated drinks; furthermore, for alleviation of the toxic effects of OTA, such complex formation may reduce its absorption from the intestine. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Mycotoxins are well known hazardous molecules, among which ochratoxin A (OTA) is a widespread food contaminant; OTA is a secondary metabolite of Aspergillus and Penicillium species (Pohland, Nesheim, & Friedman, 1992). OTA occurs in many foods (e.g. cereals, fruits, meat and dairy products) and drinks (e.g. milk, wine, beer and coffee), as well as in animal feeds (Duarte, Pena, & Lino, 2009; Pohland et al., 1992). Chemically, it consists of a dihydroisocoumarin moiety linked with L-phenylalanine (Fig. 1). OTA is mainly nephrotoxic but numerous cell and/or animal experiments also suggest its hepatotoxic, carcinogenic, immunotoxic and teratogenic effects (Ringot, Chango, Schneider, & Larondelle, 2006). In spite of the large body of evidences, unfortunately the exact mechanism of OTA toxicity has not yet been understood in all detail. Ochratoxin A has a very high oral bioavailability (about 93%), in addition, more than 99% of OTA is albumin-bound in the circulation of humans (Hagelberg, Hult, & Fuchs, 1989; Studer-Rohr, Schlatter, & Dietrich, 2000). This is responsible for the extremely
⇑ Corresponding author. Tel.: +36 72 536120; fax: +36 72 536121. } szegi). E-mail addresses: [email protected], [email protected] (T. Ko http://dx.doi.org/10.1016/j.foodchem.2014.09.034 0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.
long plasma half-life of the toxin, resulting in its long-term toxicity. Cyclodextrins (CDs) are one of the most studied molecular groups in the area of host–guest interactions; the most abundant CDs are a-, b-, and c-CDs, which are built up from six, seven and eight glucopyranoses, respectively (Szejtli, 1988). They possess a conical structure with a hydrophobic interior and a hydrophilic exterior space. The internal cavity can entrap a wide range of guest molecules but the complex stability and selectivity can be substantially increased by appropriate chemical modifications (Dodziuk, 2008). There is a wide range of utilizations of CDs, for example to enhance solubility of active components; furthermore, analytical, pharmacological or cosmetic applications are also known. Recent studies have revealed that certain cyclodextrins are able to increase the fluorescence signal of numerous mycotoxins; for this reason their analytical application is highly acknowledged (Aghamohammadi & Alizadeh, 2007; Maragos & Appell, 2007; Maragos et al., 2008; Zhou et al., 2012). Only a few studies have been published regarding the interaction of OTA with b-, and cCDs and their derivatives; on the other hand, the binding affinities of these complexes are relatively low, and the inclusion of OTA in several cases gives little or no increase in its fluorescence (Amadasi et al., 2007; Hashemi & Alizadeh, 2009; Verrone et al., 2007).
144
M. Poór et al. / Food Chemistry 172 (2015) 143–149
QABCD was investigated, using steady-state fluorescence spectroscopy and fluorescence polarization techniques. Furthermore, the influence of QABCD on OTA–magnesium and OTA–albumin interactions was investigated and molecular modelling studies were also performed. 2. Materials and methods 2.1. Reagents All reagents were of analytical grade. Ochratoxin A (OTA) was purchased from Sigma–Aldrich; 5000 lM stock solution was prepared in ethanol (Reanal, spectroscopic grade) and kept at 4 °C protected from light. b-Cyclodextrin (b-CD) and (2-hydroxy3-N,N,N-trimethylamino)propyl-b-cyclodextrin chloride (QABCD; average degree of substitution = 3.6; Fig. S1) were obtained from CycloLab Cyclodextrin Research & Development Laboratory, Ltd. Phosphate buffered saline (PBS) and 0.03 M ammonium acetate buffers (pH 5.2, 6.4, 7.4, 8.2, 9.0, respectively) were applied to examine the toxin-cyclodextrin interactions. 2.2. Instrumentation Fluorescence spectral analyses and fluorescence polarization measurements were performed using a Hitachi F-4500 fluorescence spectrophotometer and a Fluorolog s3 spectrofluorimetric system. All measurements were done at 25 °C in the presence of air. Fluorescence polarization (P) values were determined by applying the following equation:
P¼
Fig. 1. Chemical structures of OTA (above) and QABCD (below). [QABCD is a composite mixture of statistically substituted beta-cyclodextrin isomers; therefore the depicted formula is only a representative form.]
Besides analytical utilization of CDs, a further very interesting application of b-CD-polyurethane polymer is to remove OTA from wines (Appell & Jackson, 2012). Most of the previously studied CDs show little or no preference for monoanionic or dianionic forms of OTA (Hashemi & Alizadeh, 2009). Because, in physiological environments (pH, ionic strength), OTA exists principally in dianionic (OTA2) form (Poór, Kunsági-Máté, Czibulya, et al., 2013; Poór et al., 2012), the identification of new CD derivatives with strong preference for OTA2 would be of considerable medical importance. Since CDs are non-toxic molecules, we hypothesize that, if an appropriate complex stability were achieved, CDs would be suitable candidates for detoxification applications related to OTA, even in in vivo systems. In the present study, our main goal was to identify a chemically modified CD which gives a more stable complex with OTA than do the previously described derivatives. Moreover, an indispensible requirement was also to show a considerably higher affinity of modified CDs towards the dianionic than the monoanionic (OTA2 vs. OTA) form of OTA. (2-hydroxy-3-N,N,N-trimethylamino)propyl-b-cyclodextrin chloride (QABCD) that fulfilled both of these requirements. Quaternary ammonium cyclodextrins were previously described as very suitable agents during ophthalmic formulation because of their antiseptic and penetration/permeation enhancer properties (Kis, Schoch, & Szejtli, 2003; Schoch, Bizec, & Kis, 2007). In addition, further potential applications of QABCD have also been studied (Bunke & Jira, 1998; Sebestyén, BuváriBarcza, & Rohonczy, 2012; Wang, Cohen, Jicsinszky, & Douhal, 2012). Accordingly, the interaction between ochratoxin A and
ðIVV G IVH Þ ðIVV þ G IVH Þ
ð1Þ
where IVV and IVH are intensities measured at parallel and perpendicular orientations of the excitation and emission polarizer filters. G is the instrumental function of the system. For calculating degree of polarization, 15 measuring points were averaged. 2.3. Experimental procedures 2.3.1. Fluorescence properties of OTA in the absence and in the presence of QABCD Spectroscopic properties of nonionic and monoanionic OTA are very similar: both forms show excitation and emission wavelength maxima at 334 and 451 nm; on the other hand, dianionic OTA produces much higher fluorescence intensity and gives the corresponding maxima at 380 and 443 nm (Poór, Kunsági-Máté, Czibulya, et al., 2013; Poór et al., 2012). To investigate the influence of QABCD on the fluorescence spectra of OTA, 1 lM OTA and increasing cyclodextrin concentrations (25 lM–20 mM) were applied in pH 5.2 (for OTA) and in pH 9.0 (for OTA2) ammonium acetate buffers. 2.3.2. Investigation of QABCD–OTA interaction at pH 6.4, using fluorescence spectroscopy To investigate the potential preference of QABCD towards monoanionic or dianionic OTA, pH 6.4 ammonium acetate buffer was applied because, at pH 6.4, both the monoanionic and dianionic forms of OTA co-exist in aqueous solution (Fig. 2). Fluorescence spectra reflect elevation of OTA2 upon addition of QABCD; therefore the corresponding binding constants (logK) can be determined using fluorescence emission intensities of OTA2 at 380 nm excitation and 443 nm emission wavelengths. The HyperQuad 2006 program package was applied for evaluating binding constants (logK) from fluorescence emission spectra, using the previously published
145
M. Poór et al. / Food Chemistry 172 (2015) 143–149
Fig. 2. (a) Fluorescence excitation spectra of 1 lM OTA in the presence of increasing QABCD concentrations in 0.03 M ammonium acetate buffer (pH 6.4) [kexc = 380 nm; kem = 443 nm]. (b) Changes in emission spectra and molecular species distribution during formation of QABCD–OTA2 complexes as a function of QABCD concentration in ammonium acetate buffer (pH 6.4) [kexc = 380 nm; kem = 443 nm].
different pH values (pH 7.4, 8.2 and 9.0, respectively) in 0.03 M ammonium acetate buffers, using the fluorescence polarization technique (kexc = 380 nm; kem = 443 nm). Stability constants were quantified by a previously published method (Il’ichev, Perry, Rüker, Dockal, & Simon, 2002; Il’ichev, Perry, & Simon, 2002):
a¼
ðP P f Þ ½h ðPb PÞ þ P Pf
ð3Þ
where a is the bound fraction of the toxin, P is the polarization of the sample, Pf and Pb are fluorescence polarization values of completely free and bound OTA, respectively. Furthermore,
h¼
eb ub ef uf
ð4Þ
where eb and ef are the molar absorptivities of bound and free toxin,
Ub and Uf are the fluorescent quantum yields of the bound and free Fig. 3. Fluorescence polarization values of 1 lM OTA in the absence and in the presence of increasing QABCD concentrations at pH 9.0 in ammonium acetate buffer [kexc = 380 nm; kem = 443 nm].
1:1 stoichiometric method (Poór, Kunsági-Máté, Matisz, et al., 2013; Poór, Li, et al., 2013):
I ¼ I0 þ 0
ðIHG I0 Þ 2 ½H0
1 @½H0 þ ½G0 þ K
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi1 2 1 ½H0 þ ½G0 þ 4 ½H0 ½G0 A K
ð2Þ
where, I denotes the fluorescence emission intensity of OTA2 in the presence of QABCD at 443 nm; I0 denotes the fluorescence emission intensity of OTA2 in the absence of QABCD at 443 nm; IHG denotes the fluorescence emission intensity of pure QABCD–OTA2 complex at 443 nm, which is calculated by the Hyperquard2006; K denotes the binding constant; [H]0 and [G]0 denote the initial concentration of OTA2 and QABCD, respectively. 2.3.3. Investigation of QABCD–OTA interaction, using the fluorescence polarization technique Since fluorescence polarization values of OTA show significant increase during the interaction with QABCD (inclusion of OTA by the cyclodextrin molecule), this technique seems to be very suitable for following complex formation (Fig. 3). Therefore, binding constants of QABCD–OTA2 complexes were determined at several
OTA. For 1:1 stoichiometry (n = 1), the binding constant (K) can be estimated from the following equation:
K¼
a ½ð1 aÞ ðn C CD a C OTA Þ
ð5Þ
where CCD and COTA are the total concentrations of cyclodextrin and OTA, respectively. 2.3.4. Competition with magnesium ions In order to get more information on QABCD–OTA2 interaction and to confirm the calculated stability constant determined by the fluorescence polarization technique, the competition of QACBD with magnesium ions towards OTA was also investigated. PBS (pH 7.4) was used to mimic extracellular physiological conditions. Based on our previously described observation that OTA2 forms a complex with magnesium ions (Poór, Kunsági-Máté, Matisz, et al., 2013), competing ability of QABCD in the presence of Mg2+ and OTA was investigated. 1 lM OTA and 15 mM MgCl2 were applied in PBS (pH 7.4). Under these conditions, OTA exists only in magnesium-bound form. Then increasing amounts (0–50 mM) of QABCD were added to the system. Due to the spectral differences between Mg2+–OTA2 and OTA2, the spectra were decomposed into two peaks referring to the desorbed and magnesium-bound dianionic fractions. logK values have been determined by the HyperQuad program package, applying Eq. (2). 2.3.5. Effect of QABCD on OTA–albumin interaction Since albumin is the major carrier protein of OTA, we were curious about whether QABCD might have any influence on the
146
M. Poór et al. / Food Chemistry 172 (2015) 143–149
OTA–albumin interaction. In order to investigate the potential influence of QABCD, the previously described model system was applied (Poór, Kunsági-Máté, Czibulya, et al., 2013; Poór et al., 2012), where 1 lM OTA and saturating amount (1.7 lM) of human serum albumin (HSA) were mixed in PBS (pH 7.4). Fluorescence excitation and emission maxima of albumin-bound OTA (kexc = 393 nm; kem = 445 nm) were used for fluorescence polarization measurements. Because OTA forms a very stable complex with HSA, in this experiment higher QABCD concentrations were applied (12.5, 25, 50, 100 and 200 mM, respectively).
the toxin (tested with 0–20 mM QABCD in ammonium acetate buffer at pH 5.2 and at pH 9.0). This result shows that the inclusion of OTA and OTA2 by QABCD has no influence on the fluorescence behaviour of the toxin. The observed phenomenon is not surprising because recent studies highlighted that, in many cases, the inclusion of OTA by cyclodextrins is associated with little or no spectroscopic change (Amadasi et al., 2007; Hashemi & Alizadeh, 2009).
2.3.6. Molecular modelling studies Aiming to get a qualitative picture of the QABCD–OTA complex formation, molecular dynamics (MD) simulations were performed. The initial structure of the OTA molecule has been obtained from the ZINC database (Irwin & Shoichet, 2005), while the structure of the QABCD molecule was constructed by adding the three corresponding quaternary ammonium groups to the b-CD structure, the latter having been obtained from the Chemspider database. The Amber force fields, with the TIP3P explicit water model, as were implemented in the HyperChem 8.0 program (HyperChemÒ Professional 8.0, Hypercube, Inc., USA), have been used to perform the molecular dynamics simulations. The use of an explicit water model has been chosen by considering a previous detailed study related to b-CD molecules (Manunza, Deiana, Pintore, & Gessa, 1997), where it has been shown, that the cyclodextrin molecule in solution mainly exists in a highly distorted structure with a major influence by the solvent, in contrast to the crystal structure which is more ordered (Amadasi et al., 2007). Therefore, the QABCD molecule has been treated as flexible in the present study. Furthermore, it is clear that the explicit consideration of the water molecules is of importance for the electrostatic interaction in the case of the QABCD–OTA2 complex. The MD simulations were performed for a few hundreds of ps time and at 298 K in the NVT ensemble, starting from some selected and different initial structures. The obtained final structures, after reaching equilibria, were then analysed and compared.
In ammonium acetate buffer at pH 6.4, the monoanionic OTA predominates; on the other hand, the presence of QABCD (even at 25 lM concentration) results in significant increase of dianionic OTA (Fig. 2a). Since the inclusion of OTA by the QABCD molecule did not change its fluorescence spectra, these data suggest that QABCD forms a considerably more stable complex with OTA2 than with OTA. Fig 2 shows the spectral data and molecular species distribution as the function of QABCD concentration in ammonium acetate buffer (pH 6.4). Under these conditions, the stability constant of QABCD–OTA2 complex (Table 1) was determined by applying Eq. (2).
3. Results and discussion 3.1. Effects of QABCD on the fluorescence spectra of OTA Previous studies clearly show that fluorescence characteristics of OTA are highly influenced by the microenvironment. Depending on the pH, OTA could be present in nonionic, monoanionic and dianionic forms (spectral differences are detailed in 3.1.). The reported pKa values of the carboxyl and the phenolic hydroxyl groups of OTA in water are in the range of 4.2–4.4 and 7.0–7.3, respectively (Il’ichev, Perry, & Simon, 2002). Other very important parameters are the ionic strength, and the type of the ions. For example, both alkali and alkaline earth metal ions are able to form complexes with OTA2 (Poór, Kunsági-Máté, Matisz, et al., 2013). At pH 6.4, the monoanionic form is dominant but the dianionic form is also present. On the other hand, at pH 7.4 the dianionic OTA predominates. At pH 8.2, the amount of monoanionic form is very low and at pH 9.0 it is negligible. The addition of QABCD to 1 lM OTA solution did not modify the fluorescence spectra of
Table 1 Determined logK values (±SD) of QABCD–OTA2 complex at pH 6.4 (using Eq. (2)) and at pH 7.4, 8.2 and 9.0 (using Eqs. (3)–(5)) in ammonium acetate buffers, where K denotes the stability constant of QABCD–OTA2 complex in dm3/mol.
logK
pH 6.4
pH 7.4
pH 8.2
pH 9.0
3.65 ± 0.02
4.09 ± 0.04
4.30 ± 0.17
4.46 ± 0.09
3.2. Investigation of QABCD–OTA interaction at pH 6.4, using fluorescence spectroscopy
3.3. Determination of complex stabilities using fluorescence polarization The binding of OTA to the much larger cyclodextrin molecule results in substantially increased fluorescence polarization values because the inclusion of OTA by CD causes a significant decrease in its rotational freedom (Fig. 3). Using this principle, the reaction can be easily followed by the fluorescence polarization technique. The binding constants were determined by applying Eqs. (3)–(5) at pH 7.4, 8.2 and 9.0 (Table 1). In 0.03 M ammonium acetate buffers, the increasing pH results in significant elevation of logK values. The impact is understandable because, at pH 7.4 and 8.2, the monoanionic OTA is also present in the solution and it can compete with the dianionic form, despite the fact that the binding constant of QABCD–OTA complex is substantially lower (logK = 2.33 ± 0.19; determined in ammonium acetate buffer at pH 5.2 using Eqs. (3)–(5)). Because, at pH 9.0, OTA exists exclusively in the dianionic form, our data show the factual binding constant of QABCD–OTA2 complex; therefore we suggest that its logK value is 4.46 (±0.09). Thereafter, the binding constant of b-CD was compared to that of QABCD in the case of OTA2. Using fluorescence polarization technique, 2.16 ± 0.06 was obtained as the logK value of b-CD– OTA2 in ammonium acetate buffer at pH 9.0. Our results underline the interesting finding that QABCD shows about 200-fold higher complex stability with OTA2 than does the initial native b-CD. Furthermore, QABCD shows very strong preference towards the dianionic form of OTA (approximately 130-fold higher compared to OTA). In addition, the determined stability constant of the QABCD–OTA2 complex is more than 10-fold higher than that observed in the case of the previously described most stable CDderivative (heptakis-2,6-di-O-methyl-b-cyclodextrin; determined in an identical ammonium acetate buffer as in our experiment) (Hashemi & Alizadeh, 2009). Recent studies have highlighted that CDs may be suitable tools for removing certain targeted or undesirable molecules from aqueous medium, e.g. from milk or water extracts (Arora & Damodaran, 2011; Ratnasooriya & Rupasinghe, 2012; Tahir & Lee, 2013). This phenomenon was also confirmed by the study of Oláh, Cserháti, and Szejtli (1988), where the enhanced detoxification of industrial wastewaters by b-CD was demonstrated. In addition, Appell and Jackson (2012) further evidenced, that b-CD-polyurethane polymer is able to remove OTA from wines. The above listed examples indicate that the removal of certain toxins, pesticides or other toxic
M. Poór et al. / Food Chemistry 172 (2015) 143–149
147
Fig. 4. (a) Fluorescence emission spectra of 1 lM OTA in the presence of 15 mM Mg2+ and increasing amounts of QABCD in PBS (pH 7.4) [kexc = 375 nm; kem = 443 nm]. (b) Fluorescence polarization of OTA (1 lM) – HSA (1.7 lM) system in the absence and in the presence of increasing QABCD concentrations in PBS (pH 7.4) [kexc = 393 nm; kem = 445 nm].
compounds from different drinks is possible with native and chemically modified CDs, suggesting a promising detoxification application of the cyclodextrin molecular group in the future. Bearing in mind previous reports that OTA is absorbed mainly from the intestinal tract in nonionic and monoanionic forms (Galtier, 1978; Kumagai & Aibara, 1982; Roth et al., 1988) and considering our observations, we might assume that QABCD may be suitable for trapping OTA in dianionic form and in this way inhibiting the absorption of the toxin. This idea is further supported by
previous studies with cholestyramine: the cholestyramine-OTA complex shows a stability constant similar to that determined for the QABCD–OTA2 complex (Kerkadi et al., 1999). In vivo studies verify that cholestyramine inhibits the absorption of OTA from the intestines; therefore, it enhances toxin elimination and in this way protects against OTA toxicity (Kerkadi et al., 1999; Madhyastha, Frohlich, & Marquardt, 1992). Finally, our data show that, with ionic interaction, the inclusion of OTA by CDs can be dramatically increased. We suggest that, with
Fig. 5. The shape of the QABCD–OTA2 complex when the dianionic OTA is attached to the lower rim of the cyclodextrin molecule. The drawings were prepared with the Hyperchem 8.0 (HyperChem(™) Professional 8.0, Hypercube, Inc., Gainesville, USA) and the Molekel 5.4 (Ugo Varetto, MOLEKEL 5.4, Swiss National Supercomputing Centre, Lugano, Switzerland) programs.
148
M. Poór et al. / Food Chemistry 172 (2015) 143–149
further chemical modifications of QABCD, it may be possible to achieve even higher complex stability and selectivity. 3.4. Competition with magnesium ions In the next step, the OTA–Mg2+–QABCD system was investigated (in PBS, pH 7.4). The OTA–magnesium complex shows excitation and emission maxima at 375 and 427 nm (with a blue shift of the emission spectrum of OTA) and gives a considerably higher fluorescence signal than does free OTA2 (Poór, Kunsági-Máté, Matisz, et al., 2013). Under the applied conditions (see above), OTA is presented completely in the Mg2+–OTA2 complex. Fig. 4a demonstrates that QABCD is a very effective competitor against Mg2+ since the addition of QABCD results in a red shift of the fluorescence spectrum of OTA and also in a significant decrease of its intensity. Under these circumstances, logK = 1.50 ± 0.01 has been determined for the formation of QABCD–OTA2 at room temperature. The low logK value is probably due to the exchange reaction of Mg2+ and QABCD towards OTA2. The sum of this value and the previously determined data of Mg2+–OTA2 complex (logK = 3.00 ± 0.01 in PBS at pH 7.4) (Poór et al., 2014) is 4.50, which is in good correlation with the result obtained from fluorescence polarization-based experiments (logK = 4.46). 3.5. Effect of QABCD on OTA–albumin interaction Thereafter, the effect of QABCD on albumin binding of OTA was also investigated. OTA2 forms a very stable complex with HSA (logK = 7.41 at 25 °C in PBS) (Poór et al., 2012). HSA–OTA2 interaction is considered to be of high biological importance; therefore, in this experiment, PBS (pH 7.4) was used to mimic extracellular physiological conditions. Results are demonstrated in Fig. 4b where fluorescence polarization data of the OTA–HSA system were expressed in the presence of increasing QABCD concentrations. Since, under these conditions, both the QABCD and the albuminbound dianionic OTA2 are simultaneously detected (the fluorescence wavelength maxima of the two forms are very similar: 380 ? 443 nm and 393 ? 445 nm, respectively), the decrease of fluorescence polarization indicates the molecular displacement of OTA2 from the surface of HSA (Poór, Kunsági-Máté, Czibulya, et al., 2013; Poór et al., 2012). These data show that high QABCD concentrations are needed to achieve a competing ability against OTA because of the great difference in stabilities of HSA–OTA2 and QABCD–OTA2 complexes. 3.6. Molecular modelling studies The electrostatic potential map at a constant electron density (i.e. essentially the molecular shape and the electrostatic potential at 0.002 e/a30 electron density) of the structure obtained from the molecular dynamics simulations, where OTA2 takes its location at the lower rim of the cyclodextrin cavity, can be seen on Fig. 5. The determining interactions on the complex structure are mainly from the electrostatic interaction between the quaternary ammonium groups and the carboxyl group of OTA2 and the inclusion of the phenyl moiety into the apolar cavity of the cyclodextrin molecule. These results confirm our idea that the complex formation of OTA2 with cyclodextrins can be substantially strengthened by electrostatic interaction. Based on the molecular modelling studies, OTA and OTA2 ions do not behave in strongly different ways at complex formation with the QABCD cyclodextrin and they can attach to both the lower and the upper rim of the cyclodextrin cavity. However, these QABCD cyclodextrin molecules, with different positions and orientations of the quaternary ammonium groups, can show variable affinity towards OTA and its different anionic forms.
4. Conclusions Interaction of OTA with QABCD was studied under different circumstances, using fluorescence spectroscopy. Our results show that the positively charged part of the QABCD molecule results in about 200-fold higher complex stability with OTA2 than was observed in the case of b-CD. Furthermore, QABCD shows very high preference towards the dianionic form of OTA. Therefore we hypothesize that QABCD might be a suitable candidate for attenuating OTA toxicity. Acknowledgement Financial support of the University of Pécs, Medical School (PTE ÁOK-KA-2013/15) and the ‘‘Synthesis of supramolecular systems, examination of their physicochemical properties and their utilization for separation and sensor chemistry’’ project (SROP-4.2.2.A11/1/KONV-2012-0065) are gratefully acknowledged. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem. 2014.09.034. References Aghamohammadi, M., & Alizadeh, N. (2007). Fluorescence enhancement of the aflatoxin B1 by forming inclusion complexes with some cyclodextrins and molecular modeling study. Journal of Luminescence, 127, 575–582. Amadasi, A., Dall’Asta, C., Ingletto, G., Pela, R., Marchelli, R., & Cozzini, P. (2007). Explaining cyclodextrin-mycotoxin interactions using a ‘natural’ force field. Bioorganic & Medicinal Chemistry, 15, 4585–4594. Appell, M., & Jackson, M. A. (2012). Sorption of ochratoxin A from aqueous solutions using b-cyclodextrin-polyurethane polymer. Toxins, 4, 98–109. Arora, A., & Damodaran, S. (2011). Removal of soy protein-bound phospholipids by a combination of sonication, b-cyclodextrin, and phospholipase A2 treatments. Food Chemistry, 127, 1007–1013. Bunke, A., & Jira, T. (1998). Use of cationic cyclodextrin for enantioseparation by capillary electrophoresis. Journal of Chromatography A, 798, 275–280. Dodziuk, H. (2008). Cyclodextrins and their complexes. Germany, Weinheim: WileyVCH Verlag GmbH & Co. KGaA. Duarte, S. C., Pena, A., & Lino, C. M. (2009). Ochratoxin A non-conventional exposure sources – A review. Microchemical Journal, 93, 115–120. Galtier, P. (1978). Contribution of pharmacokinetic studies to mycotoxicology – Ochratoxin A. Veterinary Science Communications, 1, 349–358. Hagelberg, S., Hult, K., & Fuchs, R. (1989). Toxicokinetics of ochratoxin A in several species and its plasma-binding properties. Journal of Applied Toxicology, 9, 91–96. Hashemi, J., & Alizadeh, N. (2009). Investigation of solvent effect and cyclodextrins on fluorescence properties of ochratoxin A. Spectrochimica Acta Part A, 73, 121–126. Il’ichev, Y. V., Perry, J. L., & Simon, J. D. (2002a). Interaction of ochratoxin A with human serum albumin. Preferential binding of the dianion and pH effects. The Journal of Physical Chemistry B, 106, 452–459. Il’ichev, Y. V., Perry, J. L., Rüker, F., Dockal, M., & Simon, J. D. (2002b). Interaction of ochratoxin A with human serum albumin. Binding sites localized by competitive interactions with the native protein and its recombinant fragments. Chemico-Biological Interactions, 141, 275–293. Irwin, J. J., & Shoichet, B. K. (2005). ZINC – A free database of commercially available compounds for virtual screening. Journal of Chemical Information and Modeling, 45, 177–182. Kerkadi, A., Barriault, C., Marquardt, R. R., Frohlich, A. A., Yousef, I. M., Zhu, X. X., et al. (1999). Cholestyramine protection against ochratoxin A toxicity: Role of ochratoxin A sorption by the resin and bile acid enterohepatic circulation. Journal of Food Protection, 62, 1461–1465. Kis, G. L., Schoch, C., & Szejtli, J. (2003). Quaternary ammonium cyclodextrins as pharmaceutical penetration enhancers. PCT Int. Appl. WO 2003105867 A1 (Novartis), 24. Dec. 2003, pp. 35. Kumagai, S., & Aibara, K. (1982). Intestinal absorption and secretion of ochratoxin A in the rat. Toxicology and Applied Pharmacology, 64, 94–102. Madhyastha, M. S., Frohlich, A. A., & Marquardt, R. R. (1992). Effect of dietary cholestyramine on the elimination pattern of ochratoxin A in rats. Food and Chemical Toxicology, 30, 709–714. Manunza, B., Deiana, S., Pintore, M., & Gessa, C. (1997). Structure and internal motion of solvated beta-cyclodextrine: A molecular dynamics study. Journal of Molecular Structure: THEOCHEM, 419, 133–137.
M. Poór et al. / Food Chemistry 172 (2015) 143–149 Maragos, C. M., & Appell, M. (2007). Capillary electrophoresis of the mycotoxin zearalenone using cyclodextrin-enhanced fluorescence. Journal of Chromatography A, 1143, 252–257. Maragos, C. M., Appell, M., Lippolis, V., Visconti, A., Catucci, L., & Pascale, M. (2008). Use of cyclodextrins as modifiers of fluorescence in the detection of mycotoxins. Food Additives and Contaminants: Part A, 25, 164–171. Oláh, J., Cserháti, T., & Szejtli, J. (1988). Beta-cyclodextrin enhanced biological detoxification of industrial wastewaters. Water Research, 22, 1345–1351. Pohland, A. E., Nesheim, S., & Friedman, L. (1992). Ochratoxin A: A review. Pure and Applied Chemistry, 64, 1029–1046. }szegi, T. Poór, M., Kunsági-Máté, S., Bencsik, T., Petrik, J., Vladimir-Knezˇevic´, S., & Ko (2012). Flavonoid aglycones can compete with ochratoxin A for human serum albumin: A new possible mode of action. International Journal of Biological Macromolecules, 51, 279–283. Poór, M., Kunsági-Máté, S., Czibulya, Z., Li, Y., Peles-Lemli, B., Petrik, J., et al. (2013). Fluorescence spectroscopic investigation of competitive interactions between ochratoxin A and 13 drug molecules for binding to human serum albumin. Luminescence, 28, 726–733. Poór, M., Kunsági-Máté, S., Matisz, G., Li, Y., Czibulya, Z., Peles-Lemli, B., et al. (2013). Interaction of alkali and alkaline earth ions with ochratoxin A. Journal of Luminescence, 135, 276–280. }szegi, T. Poór, M., Li, Y., Kunsági-Máté, S., Petrik, J., Vladimir-Knezˇevic´, S., & Ko (2013). Molecular displacement of warfarin from human serum albumin by flavonoid aglycones. Journal of Luminescence, 142, 122–127. Poór, M., Kuzma, M., Matisz, G., Li, Y., Perjési, P., Kunsági-Máté, S., et al. (2014). Further aspects of ochratoxin A-cation interactions: complex formation with zinc ions and a novel analytical application of ochratoxin A-magnesium interaction in the HPLC-FLD system. Toxins, 6, 1295–1307. Ratnasooriya, C. C., & Rupasinghe, H. P. V. (2012). Extraction of phenolic compounds from grapes and their pomace using b-cyclodextrin. Food Chemistry, 134, 625–631.
149
Ringot, D., Chango, A., Schneider, Y. J., & Larondelle, Y. (2006). Toxicokinetics and toxicodynamics of ochratoxin A, an update. Chemico-Biological Interactions, 159, 18–46. Roth, A., Chakor, K., Creppy, E. E., Kane, A., Roschenthaler, R., & Dirheimer, G. (1988). Evidence for an enterohepatic circulation of ochratoxin A in mice. Toxicology, 48, 293–308. Schoch, C., Bizec, J.-C., & Kis, G. (2007). Cyclodextrin derivatives and cyclofructan as ocular permeation enhancers. Journal of Inclusion Phenomena and Macrocyclic Chemistry, 57, 391–394. Sebestyén, Z., Buvári-Barcza, Á., & Rohonczy, J. (2012). PH-dependent complex formation of amino acids with b-cyclodextrin and quaternary ammonium bcyclodextrin. Journal of Inclusion Phenomena and Macrocyclic Chemistry, 73, 199–210. Studer-Rohr, I., Schlatter, J., & Dietrich, D. R. (2000). Kinetic parameters and intraindividual fluctuations of Ochratoxin A plasma levels in human. Archives of Toxicology, 74, 499–510. Szejtli, J. (1988). Cyclodextrin Technology. Germany, Dordrecht: Kluwer. Tahir, M. N., & Lee, Y. (2013). Immobilisation of b-cyclodextrin on glass: characterisation and application for cholesterol reduction from milk. Food Chemistry, 139, 475–481. Verrone, R., Catucci, L., Cosma, P., Fini, P., Agostiano, A., Lippolis, V., et al. (2007). Effect of b-cyclodextrin on spectroscopic properties of ochratoxin A in aqueous solution. Journal of Inclusion Phenomena and Macrocyclic Chemistry, 57, 475–479. Wang, Y., Cohen, B., Jicsinszky, L., & Douhal, A. (2012). Femtosecond to second studies of a water-soluble porphyrin derivative in chemical and biological nanocavities. Langmuir, 28, 4363–4372. Zhou, Y., Chen, J., Dong, L., Lu, L., Chen, F., Hu, D., et al. (2012). A study of fluorescence properties of citrinin in b-cyclodextrin aqueous solution and different solvent. Journal of Luminescence, 132, 1437–1445.
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Interaction of ochratoxin A with quaternary ammonium beta-cyclodextrin Miklós Poór a, Sándor Kunsági-Máté b,c, Lajos Szente d, Gergely Matisz b,c, Györgyi Secenji b,c, }szegi a,⇑ Zsuzsanna Czibulya b,c, Tamás Ko a
Institute of Laboratory Medicine, University of Pécs, Ifjúság útja 13., Pécs H-7624, Hungary Department of General and Physical Chemistry, University of Pécs, Ifjúság útja 6., Pécs H-7624, Hungary János Szentágothai Research Center, Ifjúság útja 20., Pécs H-7624, Hungary d CycloLab Cyclodextrin Research & Development Laboratory, Ltd., Illatos út 7., Budapest H-1097, Hungary b c
a r t i c l e
i n f o
Article history: Received 25 February 2014 Received in revised form 5 August 2014 Accepted 8 September 2014 Available online 16 September 2014 Keywords: Ochratoxin A Cyclodextrin Molecular inclusion Fluorescence spectroscopy Detoxification
a b s t r a c t Ochratoxin A (OTA) is a widely spread nephrotoxic food contaminant mycotoxin. Unfortunately, attenuation or prevention of the toxic effects of OTA is still an unresolved problem. Molecular inclusion of OTA by cyclodextrins (CDs) results in complexes with low stability. In the human organism, OTA exists mostly in the dianionic state (OTA2). Therefore, our major goal was to develop a chemically modified cyclodextrin which gives a more stable complex with OTA than the previously published derivatives and which shows stronger preference towards OTA2. In our fluorescence spectroscopic study we demonstrate that quaternary ammonium beta-cyclodextrin (QABCD) fulfils both of these requirements. The calculated stability constant of the QABCD–OTA2 complex was 28,840 M1 (about 200-fold higher than that of the b-CD–OTA2 complex). We hypothesize, that QABCD may be a suitable tool for the decontamination of different OTA-contaminated drinks; furthermore, for alleviation of the toxic effects of OTA, such complex formation may reduce its absorption from the intestine. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Mycotoxins are well known hazardous molecules, among which ochratoxin A (OTA) is a widespread food contaminant; OTA is a secondary metabolite of Aspergillus and Penicillium species (Pohland, Nesheim, & Friedman, 1992). OTA occurs in many foods (e.g. cereals, fruits, meat and dairy products) and drinks (e.g. milk, wine, beer and coffee), as well as in animal feeds (Duarte, Pena, & Lino, 2009; Pohland et al., 1992). Chemically, it consists of a dihydroisocoumarin moiety linked with L-phenylalanine (Fig. 1). OTA is mainly nephrotoxic but numerous cell and/or animal experiments also suggest its hepatotoxic, carcinogenic, immunotoxic and teratogenic effects (Ringot, Chango, Schneider, & Larondelle, 2006). In spite of the large body of evidences, unfortunately the exact mechanism of OTA toxicity has not yet been understood in all detail. Ochratoxin A has a very high oral bioavailability (about 93%), in addition, more than 99% of OTA is albumin-bound in the circulation of humans (Hagelberg, Hult, & Fuchs, 1989; Studer-Rohr, Schlatter, & Dietrich, 2000). This is responsible for the extremely
⇑ Corresponding author. Tel.: +36 72 536120; fax: +36 72 536121. } szegi). E-mail addresses: [email protected], [email protected] (T. Ko http://dx.doi.org/10.1016/j.foodchem.2014.09.034 0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.
long plasma half-life of the toxin, resulting in its long-term toxicity. Cyclodextrins (CDs) are one of the most studied molecular groups in the area of host–guest interactions; the most abundant CDs are a-, b-, and c-CDs, which are built up from six, seven and eight glucopyranoses, respectively (Szejtli, 1988). They possess a conical structure with a hydrophobic interior and a hydrophilic exterior space. The internal cavity can entrap a wide range of guest molecules but the complex stability and selectivity can be substantially increased by appropriate chemical modifications (Dodziuk, 2008). There is a wide range of utilizations of CDs, for example to enhance solubility of active components; furthermore, analytical, pharmacological or cosmetic applications are also known. Recent studies have revealed that certain cyclodextrins are able to increase the fluorescence signal of numerous mycotoxins; for this reason their analytical application is highly acknowledged (Aghamohammadi & Alizadeh, 2007; Maragos & Appell, 2007; Maragos et al., 2008; Zhou et al., 2012). Only a few studies have been published regarding the interaction of OTA with b-, and cCDs and their derivatives; on the other hand, the binding affinities of these complexes are relatively low, and the inclusion of OTA in several cases gives little or no increase in its fluorescence (Amadasi et al., 2007; Hashemi & Alizadeh, 2009; Verrone et al., 2007).
144
M. Poór et al. / Food Chemistry 172 (2015) 143–149
QABCD was investigated, using steady-state fluorescence spectroscopy and fluorescence polarization techniques. Furthermore, the influence of QABCD on OTA–magnesium and OTA–albumin interactions was investigated and molecular modelling studies were also performed. 2. Materials and methods 2.1. Reagents All reagents were of analytical grade. Ochratoxin A (OTA) was purchased from Sigma–Aldrich; 5000 lM stock solution was prepared in ethanol (Reanal, spectroscopic grade) and kept at 4 °C protected from light. b-Cyclodextrin (b-CD) and (2-hydroxy3-N,N,N-trimethylamino)propyl-b-cyclodextrin chloride (QABCD; average degree of substitution = 3.6; Fig. S1) were obtained from CycloLab Cyclodextrin Research & Development Laboratory, Ltd. Phosphate buffered saline (PBS) and 0.03 M ammonium acetate buffers (pH 5.2, 6.4, 7.4, 8.2, 9.0, respectively) were applied to examine the toxin-cyclodextrin interactions. 2.2. Instrumentation Fluorescence spectral analyses and fluorescence polarization measurements were performed using a Hitachi F-4500 fluorescence spectrophotometer and a Fluorolog s3 spectrofluorimetric system. All measurements were done at 25 °C in the presence of air. Fluorescence polarization (P) values were determined by applying the following equation:
P¼
Fig. 1. Chemical structures of OTA (above) and QABCD (below). [QABCD is a composite mixture of statistically substituted beta-cyclodextrin isomers; therefore the depicted formula is only a representative form.]
Besides analytical utilization of CDs, a further very interesting application of b-CD-polyurethane polymer is to remove OTA from wines (Appell & Jackson, 2012). Most of the previously studied CDs show little or no preference for monoanionic or dianionic forms of OTA (Hashemi & Alizadeh, 2009). Because, in physiological environments (pH, ionic strength), OTA exists principally in dianionic (OTA2) form (Poór, Kunsági-Máté, Czibulya, et al., 2013; Poór et al., 2012), the identification of new CD derivatives with strong preference for OTA2 would be of considerable medical importance. Since CDs are non-toxic molecules, we hypothesize that, if an appropriate complex stability were achieved, CDs would be suitable candidates for detoxification applications related to OTA, even in in vivo systems. In the present study, our main goal was to identify a chemically modified CD which gives a more stable complex with OTA than do the previously described derivatives. Moreover, an indispensible requirement was also to show a considerably higher affinity of modified CDs towards the dianionic than the monoanionic (OTA2 vs. OTA) form of OTA. (2-hydroxy-3-N,N,N-trimethylamino)propyl-b-cyclodextrin chloride (QABCD) that fulfilled both of these requirements. Quaternary ammonium cyclodextrins were previously described as very suitable agents during ophthalmic formulation because of their antiseptic and penetration/permeation enhancer properties (Kis, Schoch, & Szejtli, 2003; Schoch, Bizec, & Kis, 2007). In addition, further potential applications of QABCD have also been studied (Bunke & Jira, 1998; Sebestyén, BuváriBarcza, & Rohonczy, 2012; Wang, Cohen, Jicsinszky, & Douhal, 2012). Accordingly, the interaction between ochratoxin A and
ðIVV G IVH Þ ðIVV þ G IVH Þ
ð1Þ
where IVV and IVH are intensities measured at parallel and perpendicular orientations of the excitation and emission polarizer filters. G is the instrumental function of the system. For calculating degree of polarization, 15 measuring points were averaged. 2.3. Experimental procedures 2.3.1. Fluorescence properties of OTA in the absence and in the presence of QABCD Spectroscopic properties of nonionic and monoanionic OTA are very similar: both forms show excitation and emission wavelength maxima at 334 and 451 nm; on the other hand, dianionic OTA produces much higher fluorescence intensity and gives the corresponding maxima at 380 and 443 nm (Poór, Kunsági-Máté, Czibulya, et al., 2013; Poór et al., 2012). To investigate the influence of QABCD on the fluorescence spectra of OTA, 1 lM OTA and increasing cyclodextrin concentrations (25 lM–20 mM) were applied in pH 5.2 (for OTA) and in pH 9.0 (for OTA2) ammonium acetate buffers. 2.3.2. Investigation of QABCD–OTA interaction at pH 6.4, using fluorescence spectroscopy To investigate the potential preference of QABCD towards monoanionic or dianionic OTA, pH 6.4 ammonium acetate buffer was applied because, at pH 6.4, both the monoanionic and dianionic forms of OTA co-exist in aqueous solution (Fig. 2). Fluorescence spectra reflect elevation of OTA2 upon addition of QABCD; therefore the corresponding binding constants (logK) can be determined using fluorescence emission intensities of OTA2 at 380 nm excitation and 443 nm emission wavelengths. The HyperQuad 2006 program package was applied for evaluating binding constants (logK) from fluorescence emission spectra, using the previously published
145
M. Poór et al. / Food Chemistry 172 (2015) 143–149
Fig. 2. (a) Fluorescence excitation spectra of 1 lM OTA in the presence of increasing QABCD concentrations in 0.03 M ammonium acetate buffer (pH 6.4) [kexc = 380 nm; kem = 443 nm]. (b) Changes in emission spectra and molecular species distribution during formation of QABCD–OTA2 complexes as a function of QABCD concentration in ammonium acetate buffer (pH 6.4) [kexc = 380 nm; kem = 443 nm].
different pH values (pH 7.4, 8.2 and 9.0, respectively) in 0.03 M ammonium acetate buffers, using the fluorescence polarization technique (kexc = 380 nm; kem = 443 nm). Stability constants were quantified by a previously published method (Il’ichev, Perry, Rüker, Dockal, & Simon, 2002; Il’ichev, Perry, & Simon, 2002):
a¼
ðP P f Þ ½h ðPb PÞ þ P Pf
ð3Þ
where a is the bound fraction of the toxin, P is the polarization of the sample, Pf and Pb are fluorescence polarization values of completely free and bound OTA, respectively. Furthermore,
h¼
eb ub ef uf
ð4Þ
where eb and ef are the molar absorptivities of bound and free toxin,
Ub and Uf are the fluorescent quantum yields of the bound and free Fig. 3. Fluorescence polarization values of 1 lM OTA in the absence and in the presence of increasing QABCD concentrations at pH 9.0 in ammonium acetate buffer [kexc = 380 nm; kem = 443 nm].
1:1 stoichiometric method (Poór, Kunsági-Máté, Matisz, et al., 2013; Poór, Li, et al., 2013):
I ¼ I0 þ 0
ðIHG I0 Þ 2 ½H0
1 @½H0 þ ½G0 þ K
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi1 2 1 ½H0 þ ½G0 þ 4 ½H0 ½G0 A K
ð2Þ
where, I denotes the fluorescence emission intensity of OTA2 in the presence of QABCD at 443 nm; I0 denotes the fluorescence emission intensity of OTA2 in the absence of QABCD at 443 nm; IHG denotes the fluorescence emission intensity of pure QABCD–OTA2 complex at 443 nm, which is calculated by the Hyperquard2006; K denotes the binding constant; [H]0 and [G]0 denote the initial concentration of OTA2 and QABCD, respectively. 2.3.3. Investigation of QABCD–OTA interaction, using the fluorescence polarization technique Since fluorescence polarization values of OTA show significant increase during the interaction with QABCD (inclusion of OTA by the cyclodextrin molecule), this technique seems to be very suitable for following complex formation (Fig. 3). Therefore, binding constants of QABCD–OTA2 complexes were determined at several
OTA. For 1:1 stoichiometry (n = 1), the binding constant (K) can be estimated from the following equation:
K¼
a ½ð1 aÞ ðn C CD a C OTA Þ
ð5Þ
where CCD and COTA are the total concentrations of cyclodextrin and OTA, respectively. 2.3.4. Competition with magnesium ions In order to get more information on QABCD–OTA2 interaction and to confirm the calculated stability constant determined by the fluorescence polarization technique, the competition of QACBD with magnesium ions towards OTA was also investigated. PBS (pH 7.4) was used to mimic extracellular physiological conditions. Based on our previously described observation that OTA2 forms a complex with magnesium ions (Poór, Kunsági-Máté, Matisz, et al., 2013), competing ability of QABCD in the presence of Mg2+ and OTA was investigated. 1 lM OTA and 15 mM MgCl2 were applied in PBS (pH 7.4). Under these conditions, OTA exists only in magnesium-bound form. Then increasing amounts (0–50 mM) of QABCD were added to the system. Due to the spectral differences between Mg2+–OTA2 and OTA2, the spectra were decomposed into two peaks referring to the desorbed and magnesium-bound dianionic fractions. logK values have been determined by the HyperQuad program package, applying Eq. (2). 2.3.5. Effect of QABCD on OTA–albumin interaction Since albumin is the major carrier protein of OTA, we were curious about whether QABCD might have any influence on the
146
M. Poór et al. / Food Chemistry 172 (2015) 143–149
OTA–albumin interaction. In order to investigate the potential influence of QABCD, the previously described model system was applied (Poór, Kunsági-Máté, Czibulya, et al., 2013; Poór et al., 2012), where 1 lM OTA and saturating amount (1.7 lM) of human serum albumin (HSA) were mixed in PBS (pH 7.4). Fluorescence excitation and emission maxima of albumin-bound OTA (kexc = 393 nm; kem = 445 nm) were used for fluorescence polarization measurements. Because OTA forms a very stable complex with HSA, in this experiment higher QABCD concentrations were applied (12.5, 25, 50, 100 and 200 mM, respectively).
the toxin (tested with 0–20 mM QABCD in ammonium acetate buffer at pH 5.2 and at pH 9.0). This result shows that the inclusion of OTA and OTA2 by QABCD has no influence on the fluorescence behaviour of the toxin. The observed phenomenon is not surprising because recent studies highlighted that, in many cases, the inclusion of OTA by cyclodextrins is associated with little or no spectroscopic change (Amadasi et al., 2007; Hashemi & Alizadeh, 2009).
2.3.6. Molecular modelling studies Aiming to get a qualitative picture of the QABCD–OTA complex formation, molecular dynamics (MD) simulations were performed. The initial structure of the OTA molecule has been obtained from the ZINC database (Irwin & Shoichet, 2005), while the structure of the QABCD molecule was constructed by adding the three corresponding quaternary ammonium groups to the b-CD structure, the latter having been obtained from the Chemspider database. The Amber force fields, with the TIP3P explicit water model, as were implemented in the HyperChem 8.0 program (HyperChemÒ Professional 8.0, Hypercube, Inc., USA), have been used to perform the molecular dynamics simulations. The use of an explicit water model has been chosen by considering a previous detailed study related to b-CD molecules (Manunza, Deiana, Pintore, & Gessa, 1997), where it has been shown, that the cyclodextrin molecule in solution mainly exists in a highly distorted structure with a major influence by the solvent, in contrast to the crystal structure which is more ordered (Amadasi et al., 2007). Therefore, the QABCD molecule has been treated as flexible in the present study. Furthermore, it is clear that the explicit consideration of the water molecules is of importance for the electrostatic interaction in the case of the QABCD–OTA2 complex. The MD simulations were performed for a few hundreds of ps time and at 298 K in the NVT ensemble, starting from some selected and different initial structures. The obtained final structures, after reaching equilibria, were then analysed and compared.
In ammonium acetate buffer at pH 6.4, the monoanionic OTA predominates; on the other hand, the presence of QABCD (even at 25 lM concentration) results in significant increase of dianionic OTA (Fig. 2a). Since the inclusion of OTA by the QABCD molecule did not change its fluorescence spectra, these data suggest that QABCD forms a considerably more stable complex with OTA2 than with OTA. Fig 2 shows the spectral data and molecular species distribution as the function of QABCD concentration in ammonium acetate buffer (pH 6.4). Under these conditions, the stability constant of QABCD–OTA2 complex (Table 1) was determined by applying Eq. (2).
3. Results and discussion 3.1. Effects of QABCD on the fluorescence spectra of OTA Previous studies clearly show that fluorescence characteristics of OTA are highly influenced by the microenvironment. Depending on the pH, OTA could be present in nonionic, monoanionic and dianionic forms (spectral differences are detailed in 3.1.). The reported pKa values of the carboxyl and the phenolic hydroxyl groups of OTA in water are in the range of 4.2–4.4 and 7.0–7.3, respectively (Il’ichev, Perry, & Simon, 2002). Other very important parameters are the ionic strength, and the type of the ions. For example, both alkali and alkaline earth metal ions are able to form complexes with OTA2 (Poór, Kunsági-Máté, Matisz, et al., 2013). At pH 6.4, the monoanionic form is dominant but the dianionic form is also present. On the other hand, at pH 7.4 the dianionic OTA predominates. At pH 8.2, the amount of monoanionic form is very low and at pH 9.0 it is negligible. The addition of QABCD to 1 lM OTA solution did not modify the fluorescence spectra of
Table 1 Determined logK values (±SD) of QABCD–OTA2 complex at pH 6.4 (using Eq. (2)) and at pH 7.4, 8.2 and 9.0 (using Eqs. (3)–(5)) in ammonium acetate buffers, where K denotes the stability constant of QABCD–OTA2 complex in dm3/mol.
logK
pH 6.4
pH 7.4
pH 8.2
pH 9.0
3.65 ± 0.02
4.09 ± 0.04
4.30 ± 0.17
4.46 ± 0.09
3.2. Investigation of QABCD–OTA interaction at pH 6.4, using fluorescence spectroscopy
3.3. Determination of complex stabilities using fluorescence polarization The binding of OTA to the much larger cyclodextrin molecule results in substantially increased fluorescence polarization values because the inclusion of OTA by CD causes a significant decrease in its rotational freedom (Fig. 3). Using this principle, the reaction can be easily followed by the fluorescence polarization technique. The binding constants were determined by applying Eqs. (3)–(5) at pH 7.4, 8.2 and 9.0 (Table 1). In 0.03 M ammonium acetate buffers, the increasing pH results in significant elevation of logK values. The impact is understandable because, at pH 7.4 and 8.2, the monoanionic OTA is also present in the solution and it can compete with the dianionic form, despite the fact that the binding constant of QABCD–OTA complex is substantially lower (logK = 2.33 ± 0.19; determined in ammonium acetate buffer at pH 5.2 using Eqs. (3)–(5)). Because, at pH 9.0, OTA exists exclusively in the dianionic form, our data show the factual binding constant of QABCD–OTA2 complex; therefore we suggest that its logK value is 4.46 (±0.09). Thereafter, the binding constant of b-CD was compared to that of QABCD in the case of OTA2. Using fluorescence polarization technique, 2.16 ± 0.06 was obtained as the logK value of b-CD– OTA2 in ammonium acetate buffer at pH 9.0. Our results underline the interesting finding that QABCD shows about 200-fold higher complex stability with OTA2 than does the initial native b-CD. Furthermore, QABCD shows very strong preference towards the dianionic form of OTA (approximately 130-fold higher compared to OTA). In addition, the determined stability constant of the QABCD–OTA2 complex is more than 10-fold higher than that observed in the case of the previously described most stable CDderivative (heptakis-2,6-di-O-methyl-b-cyclodextrin; determined in an identical ammonium acetate buffer as in our experiment) (Hashemi & Alizadeh, 2009). Recent studies have highlighted that CDs may be suitable tools for removing certain targeted or undesirable molecules from aqueous medium, e.g. from milk or water extracts (Arora & Damodaran, 2011; Ratnasooriya & Rupasinghe, 2012; Tahir & Lee, 2013). This phenomenon was also confirmed by the study of Oláh, Cserháti, and Szejtli (1988), where the enhanced detoxification of industrial wastewaters by b-CD was demonstrated. In addition, Appell and Jackson (2012) further evidenced, that b-CD-polyurethane polymer is able to remove OTA from wines. The above listed examples indicate that the removal of certain toxins, pesticides or other toxic
M. Poór et al. / Food Chemistry 172 (2015) 143–149
147
Fig. 4. (a) Fluorescence emission spectra of 1 lM OTA in the presence of 15 mM Mg2+ and increasing amounts of QABCD in PBS (pH 7.4) [kexc = 375 nm; kem = 443 nm]. (b) Fluorescence polarization of OTA (1 lM) – HSA (1.7 lM) system in the absence and in the presence of increasing QABCD concentrations in PBS (pH 7.4) [kexc = 393 nm; kem = 445 nm].
compounds from different drinks is possible with native and chemically modified CDs, suggesting a promising detoxification application of the cyclodextrin molecular group in the future. Bearing in mind previous reports that OTA is absorbed mainly from the intestinal tract in nonionic and monoanionic forms (Galtier, 1978; Kumagai & Aibara, 1982; Roth et al., 1988) and considering our observations, we might assume that QABCD may be suitable for trapping OTA in dianionic form and in this way inhibiting the absorption of the toxin. This idea is further supported by
previous studies with cholestyramine: the cholestyramine-OTA complex shows a stability constant similar to that determined for the QABCD–OTA2 complex (Kerkadi et al., 1999). In vivo studies verify that cholestyramine inhibits the absorption of OTA from the intestines; therefore, it enhances toxin elimination and in this way protects against OTA toxicity (Kerkadi et al., 1999; Madhyastha, Frohlich, & Marquardt, 1992). Finally, our data show that, with ionic interaction, the inclusion of OTA by CDs can be dramatically increased. We suggest that, with
Fig. 5. The shape of the QABCD–OTA2 complex when the dianionic OTA is attached to the lower rim of the cyclodextrin molecule. The drawings were prepared with the Hyperchem 8.0 (HyperChem(™) Professional 8.0, Hypercube, Inc., Gainesville, USA) and the Molekel 5.4 (Ugo Varetto, MOLEKEL 5.4, Swiss National Supercomputing Centre, Lugano, Switzerland) programs.
148
M. Poór et al. / Food Chemistry 172 (2015) 143–149
further chemical modifications of QABCD, it may be possible to achieve even higher complex stability and selectivity. 3.4. Competition with magnesium ions In the next step, the OTA–Mg2+–QABCD system was investigated (in PBS, pH 7.4). The OTA–magnesium complex shows excitation and emission maxima at 375 and 427 nm (with a blue shift of the emission spectrum of OTA) and gives a considerably higher fluorescence signal than does free OTA2 (Poór, Kunsági-Máté, Matisz, et al., 2013). Under the applied conditions (see above), OTA is presented completely in the Mg2+–OTA2 complex. Fig. 4a demonstrates that QABCD is a very effective competitor against Mg2+ since the addition of QABCD results in a red shift of the fluorescence spectrum of OTA and also in a significant decrease of its intensity. Under these circumstances, logK = 1.50 ± 0.01 has been determined for the formation of QABCD–OTA2 at room temperature. The low logK value is probably due to the exchange reaction of Mg2+ and QABCD towards OTA2. The sum of this value and the previously determined data of Mg2+–OTA2 complex (logK = 3.00 ± 0.01 in PBS at pH 7.4) (Poór et al., 2014) is 4.50, which is in good correlation with the result obtained from fluorescence polarization-based experiments (logK = 4.46). 3.5. Effect of QABCD on OTA–albumin interaction Thereafter, the effect of QABCD on albumin binding of OTA was also investigated. OTA2 forms a very stable complex with HSA (logK = 7.41 at 25 °C in PBS) (Poór et al., 2012). HSA–OTA2 interaction is considered to be of high biological importance; therefore, in this experiment, PBS (pH 7.4) was used to mimic extracellular physiological conditions. Results are demonstrated in Fig. 4b where fluorescence polarization data of the OTA–HSA system were expressed in the presence of increasing QABCD concentrations. Since, under these conditions, both the QABCD and the albuminbound dianionic OTA2 are simultaneously detected (the fluorescence wavelength maxima of the two forms are very similar: 380 ? 443 nm and 393 ? 445 nm, respectively), the decrease of fluorescence polarization indicates the molecular displacement of OTA2 from the surface of HSA (Poór, Kunsági-Máté, Czibulya, et al., 2013; Poór et al., 2012). These data show that high QABCD concentrations are needed to achieve a competing ability against OTA because of the great difference in stabilities of HSA–OTA2 and QABCD–OTA2 complexes. 3.6. Molecular modelling studies The electrostatic potential map at a constant electron density (i.e. essentially the molecular shape and the electrostatic potential at 0.002 e/a30 electron density) of the structure obtained from the molecular dynamics simulations, where OTA2 takes its location at the lower rim of the cyclodextrin cavity, can be seen on Fig. 5. The determining interactions on the complex structure are mainly from the electrostatic interaction between the quaternary ammonium groups and the carboxyl group of OTA2 and the inclusion of the phenyl moiety into the apolar cavity of the cyclodextrin molecule. These results confirm our idea that the complex formation of OTA2 with cyclodextrins can be substantially strengthened by electrostatic interaction. Based on the molecular modelling studies, OTA and OTA2 ions do not behave in strongly different ways at complex formation with the QABCD cyclodextrin and they can attach to both the lower and the upper rim of the cyclodextrin cavity. However, these QABCD cyclodextrin molecules, with different positions and orientations of the quaternary ammonium groups, can show variable affinity towards OTA and its different anionic forms.
4. Conclusions Interaction of OTA with QABCD was studied under different circumstances, using fluorescence spectroscopy. Our results show that the positively charged part of the QABCD molecule results in about 200-fold higher complex stability with OTA2 than was observed in the case of b-CD. Furthermore, QABCD shows very high preference towards the dianionic form of OTA. Therefore we hypothesize that QABCD might be a suitable candidate for attenuating OTA toxicity. Acknowledgement Financial support of the University of Pécs, Medical School (PTE ÁOK-KA-2013/15) and the ‘‘Synthesis of supramolecular systems, examination of their physicochemical properties and their utilization for separation and sensor chemistry’’ project (SROP-4.2.2.A11/1/KONV-2012-0065) are gratefully acknowledged. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem. 2014.09.034. References Aghamohammadi, M., & Alizadeh, N. (2007). Fluorescence enhancement of the aflatoxin B1 by forming inclusion complexes with some cyclodextrins and molecular modeling study. Journal of Luminescence, 127, 575–582. Amadasi, A., Dall’Asta, C., Ingletto, G., Pela, R., Marchelli, R., & Cozzini, P. (2007). Explaining cyclodextrin-mycotoxin interactions using a ‘natural’ force field. Bioorganic & Medicinal Chemistry, 15, 4585–4594. Appell, M., & Jackson, M. A. (2012). Sorption of ochratoxin A from aqueous solutions using b-cyclodextrin-polyurethane polymer. Toxins, 4, 98–109. Arora, A., & Damodaran, S. (2011). Removal of soy protein-bound phospholipids by a combination of sonication, b-cyclodextrin, and phospholipase A2 treatments. Food Chemistry, 127, 1007–1013. Bunke, A., & Jira, T. (1998). Use of cationic cyclodextrin for enantioseparation by capillary electrophoresis. Journal of Chromatography A, 798, 275–280. Dodziuk, H. (2008). Cyclodextrins and their complexes. Germany, Weinheim: WileyVCH Verlag GmbH & Co. KGaA. Duarte, S. C., Pena, A., & Lino, C. M. (2009). Ochratoxin A non-conventional exposure sources – A review. Microchemical Journal, 93, 115–120. Galtier, P. (1978). Contribution of pharmacokinetic studies to mycotoxicology – Ochratoxin A. Veterinary Science Communications, 1, 349–358. Hagelberg, S., Hult, K., & Fuchs, R. (1989). Toxicokinetics of ochratoxin A in several species and its plasma-binding properties. Journal of Applied Toxicology, 9, 91–96. Hashemi, J., & Alizadeh, N. (2009). Investigation of solvent effect and cyclodextrins on fluorescence properties of ochratoxin A. Spectrochimica Acta Part A, 73, 121–126. Il’ichev, Y. V., Perry, J. L., & Simon, J. D. (2002a). Interaction of ochratoxin A with human serum albumin. Preferential binding of the dianion and pH effects. The Journal of Physical Chemistry B, 106, 452–459. Il’ichev, Y. V., Perry, J. L., Rüker, F., Dockal, M., & Simon, J. D. (2002b). Interaction of ochratoxin A with human serum albumin. Binding sites localized by competitive interactions with the native protein and its recombinant fragments. Chemico-Biological Interactions, 141, 275–293. Irwin, J. J., & Shoichet, B. K. (2005). ZINC – A free database of commercially available compounds for virtual screening. Journal of Chemical Information and Modeling, 45, 177–182. Kerkadi, A., Barriault, C., Marquardt, R. R., Frohlich, A. A., Yousef, I. M., Zhu, X. X., et al. (1999). Cholestyramine protection against ochratoxin A toxicity: Role of ochratoxin A sorption by the resin and bile acid enterohepatic circulation. Journal of Food Protection, 62, 1461–1465. Kis, G. L., Schoch, C., & Szejtli, J. (2003). Quaternary ammonium cyclodextrins as pharmaceutical penetration enhancers. PCT Int. Appl. WO 2003105867 A1 (Novartis), 24. Dec. 2003, pp. 35. Kumagai, S., & Aibara, K. (1982). Intestinal absorption and secretion of ochratoxin A in the rat. Toxicology and Applied Pharmacology, 64, 94–102. Madhyastha, M. S., Frohlich, A. A., & Marquardt, R. R. (1992). Effect of dietary cholestyramine on the elimination pattern of ochratoxin A in rats. Food and Chemical Toxicology, 30, 709–714. Manunza, B., Deiana, S., Pintore, M., & Gessa, C. (1997). Structure and internal motion of solvated beta-cyclodextrine: A molecular dynamics study. Journal of Molecular Structure: THEOCHEM, 419, 133–137.
M. Poór et al. / Food Chemistry 172 (2015) 143–149 Maragos, C. M., & Appell, M. (2007). Capillary electrophoresis of the mycotoxin zearalenone using cyclodextrin-enhanced fluorescence. Journal of Chromatography A, 1143, 252–257. Maragos, C. M., Appell, M., Lippolis, V., Visconti, A., Catucci, L., & Pascale, M. (2008). Use of cyclodextrins as modifiers of fluorescence in the detection of mycotoxins. Food Additives and Contaminants: Part A, 25, 164–171. Oláh, J., Cserháti, T., & Szejtli, J. (1988). Beta-cyclodextrin enhanced biological detoxification of industrial wastewaters. Water Research, 22, 1345–1351. Pohland, A. E., Nesheim, S., & Friedman, L. (1992). Ochratoxin A: A review. Pure and Applied Chemistry, 64, 1029–1046. }szegi, T. Poór, M., Kunsági-Máté, S., Bencsik, T., Petrik, J., Vladimir-Knezˇevic´, S., & Ko (2012). Flavonoid aglycones can compete with ochratoxin A for human serum albumin: A new possible mode of action. International Journal of Biological Macromolecules, 51, 279–283. Poór, M., Kunsági-Máté, S., Czibulya, Z., Li, Y., Peles-Lemli, B., Petrik, J., et al. (2013). Fluorescence spectroscopic investigation of competitive interactions between ochratoxin A and 13 drug molecules for binding to human serum albumin. Luminescence, 28, 726–733. Poór, M., Kunsági-Máté, S., Matisz, G., Li, Y., Czibulya, Z., Peles-Lemli, B., et al. (2013). Interaction of alkali and alkaline earth ions with ochratoxin A. Journal of Luminescence, 135, 276–280. }szegi, T. Poór, M., Li, Y., Kunsági-Máté, S., Petrik, J., Vladimir-Knezˇevic´, S., & Ko (2013). Molecular displacement of warfarin from human serum albumin by flavonoid aglycones. Journal of Luminescence, 142, 122–127. Poór, M., Kuzma, M., Matisz, G., Li, Y., Perjési, P., Kunsági-Máté, S., et al. (2014). Further aspects of ochratoxin A-cation interactions: complex formation with zinc ions and a novel analytical application of ochratoxin A-magnesium interaction in the HPLC-FLD system. Toxins, 6, 1295–1307. Ratnasooriya, C. C., & Rupasinghe, H. P. V. (2012). Extraction of phenolic compounds from grapes and their pomace using b-cyclodextrin. Food Chemistry, 134, 625–631.
149
Ringot, D., Chango, A., Schneider, Y. J., & Larondelle, Y. (2006). Toxicokinetics and toxicodynamics of ochratoxin A, an update. Chemico-Biological Interactions, 159, 18–46. Roth, A., Chakor, K., Creppy, E. E., Kane, A., Roschenthaler, R., & Dirheimer, G. (1988). Evidence for an enterohepatic circulation of ochratoxin A in mice. Toxicology, 48, 293–308. Schoch, C., Bizec, J.-C., & Kis, G. (2007). Cyclodextrin derivatives and cyclofructan as ocular permeation enhancers. Journal of Inclusion Phenomena and Macrocyclic Chemistry, 57, 391–394. Sebestyén, Z., Buvári-Barcza, Á., & Rohonczy, J. (2012). PH-dependent complex formation of amino acids with b-cyclodextrin and quaternary ammonium bcyclodextrin. Journal of Inclusion Phenomena and Macrocyclic Chemistry, 73, 199–210. Studer-Rohr, I., Schlatter, J., & Dietrich, D. R. (2000). Kinetic parameters and intraindividual fluctuations of Ochratoxin A plasma levels in human. Archives of Toxicology, 74, 499–510. Szejtli, J. (1988). Cyclodextrin Technology. Germany, Dordrecht: Kluwer. Tahir, M. N., & Lee, Y. (2013). Immobilisation of b-cyclodextrin on glass: characterisation and application for cholesterol reduction from milk. Food Chemistry, 139, 475–481. Verrone, R., Catucci, L., Cosma, P., Fini, P., Agostiano, A., Lippolis, V., et al. (2007). Effect of b-cyclodextrin on spectroscopic properties of ochratoxin A in aqueous solution. Journal of Inclusion Phenomena and Macrocyclic Chemistry, 57, 475–479. Wang, Y., Cohen, B., Jicsinszky, L., & Douhal, A. (2012). Femtosecond to second studies of a water-soluble porphyrin derivative in chemical and biological nanocavities. Langmuir, 28, 4363–4372. Zhou, Y., Chen, J., Dong, L., Lu, L., Chen, F., Hu, D., et al. (2012). A study of fluorescence properties of citrinin in b-cyclodextrin aqueous solution and different solvent. Journal of Luminescence, 132, 1437–1445.
