- Email: [email protected]
Investigation of complexation of immobilized metallothionein with Zn(II) and Cd(II) ions using piezoelectric crystals
Biosensors and Bioelectronics 18 (2003) 1039 /1046 www.elsevier.com/locate/bios
Investigation of complexation of immobilized metallothionein with Zn(II) and Cd(II) ions using piezoelectric crystals Reza Saber, Erhan Pis¸kin * Department of Chemical Engineering and Bioengineering Division, Hacettepe University, 06532 Beytepe, Ankara, Turkey Received 9 May 2002;space; received in revised form 3 September 2002; accepted 19 September 2002
Abstract The aim of this study is to investigate complexation of metallothionein (MT) with cadmium and zinc ions. An oligopeptide (i.e. Lys-Cys-Thr-Cys-Cys-Ala), a fragment of MT was covalently immobilized onto piezoelectric crystals, which were first treated with ethylene diamine plasma in a glow-discharge apparatus, and then were chemically reacted with glutaraldehyde. Complexation of the immobilized MT with Zn(II) and Cd(II) ions in aqueous media was followed by recording the changes of the frequency shifts of the piezoelectric quartz crystals. The amount of Cd(II) ions interacted with the immobilized MT molecules was the highest at pH 7.4, and decreased with an increase in the pH of the medium, in parallel to the decrease in the amount of immobilized MT. The number of Zn(II) ions interacted with the immobilized MT molecules was higher than the number of Cd(II) ions when the adsorption was from solutions containing a single-metal ion with the same ion concentrations. In consecutive adsorption studies, we observed that the type of metal ions used in the first interaction is important. These experiments showed also that there is an exchange between the metal ions, and competition provokes adsorption of both ions due to synergistic /antagonistic effects. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Piezoelectric quartz crystals; Glow-discharge; Plasma polymerization; Metallothionein fragment; Complexation with Zn(II) and Cd(II) ions
1. Introduction Metallothioneins (MTs) are a group of cysteine-rich, low molecular weight (B/6000D), and heavy metal binding proteins (Kagi et al., 1974; Chaoui et al., 1997; Armitage et al., 1982; Dabrio and Rodriguez, 2000; Harlyk et al., 1997). Their physiological roles include storage of heavy metal ions (especially Zn(II) and Cd(II)) and detoxification of heavy metals (e.g. Cd(II)). MTs are characterized by the lack of aromatic amino acids and by their selective capability to bind metallic ions. They contain 61 amino acids of which 20 are cysteinyl residues. Heavy metal ions are bound to
* Corresponding author. Tel.: /90-312-2977473; fax: /90-3122992124 E-mail address: [email protected] (E. Pis¸kin).
these cysteine residues through thiolate bonds by tetrahedral coordination. In the related literature, special attention has been paid to the study of complexing properties of MTs with heavy metal ions. The peptidic fragment ‘Lys-Cys-Thr-Cys-Cys-Ala [56 /61] MT 1’ of the mouse liver MT has been used as a model in several investigations that were aimed to better understand these complexing properties of MTs. In recent studies, electrochemical techniques have been used to investigate the complexing properties of this MT fragment, especially with zinc and cadmium ions (Nieto and Rodriguez, 1996; Mutlu et al., 1999; Nieto and Rodriguez, 1999). In this study, for the first time in the related literature, we used immobilized form (not free) of MT fragments to investigate complexation with Zn(II) and Cd(II) ions in aqueous media under different conditions by following frequency shifts of piezoelectric crystals.
0956-5663/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 5 6 - 5 6 6 3 ( 0 2 ) 0 0 2 1 7 - 8
1040
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
2. Material and methods 2.1. Materials The MT fragment was purchased from Sigma Chem. Corp. (Catalog No. L 4512, USA). Ethylene diamine (EDA) and glutaraldehyde (GA) were purchased from Fluka, Switzerland and from Aldrich (UK), respectively. All other reagents (Na2HPO4, NaH2PO4, Na2B4O7.10H2O, Tris) were purchased from Merck (Germany) and/or BDH (UK). 2.2. Metallothionein immobilization For immobilization of the MT fragments, the piezoelectric crystals were first treated in a glow-discharge apparatus, as described previously (Roesijadi, 1996). EDA was used as the active monomer to create aminolike functional groups on the quartz crystal surface. Briefly, the glow-discharge reactor was evacuated to 10 1 mbar. The monomer, EDA, was allowed to flow though the reactor at a flow rate of 35 ml/min. The quartz crystals were exposed to the EDA plasma for 3 min at glow-discharge power of 15 W. It should be noted that these conditions were selected after preliminary studies. In the first step of MT immobilization onto the crystal surfaces, the EDA-plasma-treated quartz crystals werincubated with a 2.5% GA solution (in a phosphate buffer, pH 7.4) at 4 8C for 16 h. These conditions were selected based on preliminary studies (Saber, 2002). The physically adsorbed GA molecules were removed by washing with distilled water several times. Following the washing step, the crystals were dried in air (at room temperature for about 2 h, until constant frequency measurements was obtained) and the frequencies of the air-dried crystals were noted. GA molecule was selected as it is a dialdehyde, containing two aldehyde groups such that one aldehyde group on each GA covalently interacts with an amine group on the EDA treated crystal, while the other (aldehyde group) remains free for the next treatment step. In the final step, the GA-attached quartz crystals were dipped into the MT solutions containing 0.30 mg MT in 1 ml buffer at different pH (7.4, 8.4, 9.4, 10.4). The incubation medium was mixed continuously on a rotary mixer (Ko¨ttermann Labortechnik, Germany) with a constant turning rate of 120 rpm, for 4 h at room temperature. Physically adsorbed MT molecules were removed by first washing the samples with the incubation buffer followed by incubation for 20 min in 0.15 M Tris buffer, and then with distilled water for additional 15 min. The frequency shifts were measured with the microbalance system described below at a constant temperature of 25 8C. It should be noted that, at this final step, the amine groups of the MT molecules react
with the free aldehyde groups on the piezoelectric crystals obtained in the previous step and are rendered covalently immobilized. 2.3. Microbalance system The microbalance system consists of an AT-cut 10 MHz quartz crystal of a 8.80 /0.16 mm2 wafer which is placed between two silver electrodes (diameter: 5 mm) and is mounted on a ceramic holder with a plug (AEC, Taiwan). The oscillation electronic circuit is a typical Collpits oscillator, which has a buffer amplifier. 15 V DC is applied to the oscillator circuit to drive the crystal and the frequency is measured with a Hewlett Packard frequency counter (Model No: HP 53181A, 225 MHz, USA). The drift observed of this system was less than 1 Hz in the measurements in air, after an initial stabilization period, which is about 5 min in air. 2.4. Complexation with heavy metal ions The MT-immobilized crystals were incubated with the aqueous solutions (20 ml) containing different amounts (1 /7 ppm in the first set and 10 /50 ppm in the second set) of zinc or cadmium nitrate at pH 7.4 (phosphate buffer) for 1 h on a rotary mixer. In order to study the competitive adsorption, the solutions containing mixtures of these two ions with different Cd(II)/Zn(II) ratios (1/7, 2/3, 3/5, 1/1, 5/3, 3/1, 7/1) were also used. Ten crystals were tested in each experiment. After 1 hinteraction, the crystals were taken out, washed several times with buffer and with distilled water, dried and the frequency shifts were measured in air.
3. Results and discussion 3.1. Metallothionein immobilization In this part of the study, MT molecules were first immobilized onto the GA-attached piezoelectric crystals. Immobilizations were achieved at different pH (7.4 and 8.4 in phosphate buffer, and 9.4 and 10.4 in tetraborate buffer). The initial concentration and immobilization time were 30 mg/ml and 4 h, respectively. In order to find the effectiveness of the immobilized MT molecules, these crystals were then incubated with Cd(II) solutions with an initial concentration of 50 ppm, at pH 7.4 and for 1 h. The extent of MT immobilization and the MT-Cd(II) complexation reaction was followed by measuring the frequency shifts with the micro-balance system after each step (Fig. 1). As seen here, the maximum efficiency of MT immobilization was achieved at pH 7.4 while at higher pH immobilizations were significantly lower. Probably, the three-dimensional structure of the MT molecules at pH
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
1041
Fig. 1. Effects of pH on MT immobilization and complexation of the immobilized MT with Cd(II) ions.
7.4 was more suitable for the interaction with the aldehyde groups on the piezoelectric crystals. It is interesting to note that this is the physiological pH, in which, most probably, MT molecules are in their native form, even in aqueous solutions that we have applied in the study. The amount of Cd(II) ions interacted with the immobilized MT molecules was also the highest at pH 7.4, and decreased with an increase in the pH of the medium, in parallel with the decrease in the amount of MT immobilized, as expected. We attempted to calculate the number of MT molecules immobilized onto the crystals, and also the number of metal ions interacted with these immobilized MT molecules from the Saurbrey equation by using the frequency shifts measured which are given below (Saber, 2002). However, the data were not stable (rather fluctuating) enough to make any conclusion, and therefore we decided that, this equation is not valid in all cases (such as in our case here), and these data should not be included here. Df 2:26 10 6 f 2 Dm=A here: Df is change in frequency, f is fundamental frequency of the crystal, Dm is the mass adsorbed on the surface, A is the surface area of the electrode. 3.2. Measurement of metal ions concentrations In this group of experiments we investigated the interactions of both Cd(II) and Zn(II) ions from their aqueous solution (containing single ions) at different concentrations and also from solutions containing both ions at different Cd(II)/Zn(II) ratios. Here the MTimmobilized crystals prepared at pH 7.4 were used. Fig. 2 shows the interaction of Cd(II) and Zn(II) ions of the
MT carrying crystals from their single-metal ion-solutions and from their mixtures. As observed from these data, the frequency shifts increased with an increase in the initial concentrations of either Cd(II) or Zn(II). It should be noted that frequency shifts are proportional to the deposited mass onto the crystal surface. Therefore, the increase in the frequency shift corresponds to the increase in the amount of metal ions interacted with the immobilized MT molecules (or simply the amount of metal ions adsorbed on the crystals). It is apparent from the frequency shifts in Fig. 2 that the MT carrying crystals adsorbed more Cd(II) ions than Zn(II) ions. However, this is not correct, as shown in Fig. 3, where the frequency shifts were divided by the atomic weights of the respective ions and then the graphs were redrawn. Thus, by compensating for the difference in the mass of the ions, the number of Zn(II) ions interacted with the immobilized MT molecules is higher than the number of Cd(II) ions when the adsorption was from solutions containing a single-metal ion with the same ion concentrations. In order to further investigate the adsorption behavior of Cd(II) and Zn(II) ions, we ran a second set of experiment where we first incubated solutions containing the desired amount of Zn(II) ions (or Cd(II) ions), measured the frequency shift, and then the same crystals were incubated with a solution containing Cd(II) ions (or Zn(II) ions) at the same concentration. These tests were repeated for different initial concentrations of metal ions, namely 10, 20, 30, 40, or 50 ppm. It should be noted that these concentrations were higher than those studied in the previous set of experiments. Figs. 4 and 5 show the frequency shifts after each interactions, and also total changes as a function of the initial concentration of metal ions. The only difference in these
1042
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
Fig. 2. Change in frequency shifts in the adsorption of Cd(II) and Zn(II) ions from their single-metal ion solutions containing different amounts of ions or their mixtures.
two figures is the type of metal ions used in the first and second interactions (as indicated also in the graphs). The first point that can be extracted from these two figures is that the amount of metal ions interacted with the immobilized MT molecules increases with the initial concentration of the metal ions in the solution, as expected. The more interesting point is the effect of the type of metal ions used in the first interaction. When we first treated the crystals carrying the MT molecules with Cd(II) ions and then Zn(II) ions, we reached higher total adsorption. It should also be noted that there
might be some kind of exchange of the metal ions between the aqueous phase (in the second incubation) and the crystals (already treated in the first incubation). Hence, Fig. 4 shows that after Zn(II) binding in the first interaction, the Cd(II) additions provoke the release of some or most of the complexed Zn(II) and enhances the complexation of Cd(II) with no modification of the coordination of Zn(II) with MT molecules which is in agreement with other researches (Stillman, 1995). However, we do not have sufficient data to determine the validity of this point. Nietto and Rodriguez used square
Fig. 3. Change in frequency shifts/atomic weight in the adsorption of Cd(II) and Zn(II) ions from their single-metal ion solutions containing different amounts of ions.
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
1043
Fig. 4. Change in frequency shifts in the adsorption of Cd(II) and Zn(II) ions from their single-metal ion solutions containing different amounts of ions. The crystals were incubated with first the Zn(II) ions and then with the Cd(II) ions.
wave voltammetry to study interaction of Cd(II) and Zn(II) ions with MT molecules in solution, and reported that the Cd(II) initially complexed by MT was not released when Zn(II) was added because the Cd(II) peak was not detected (Stillman, 1995). On the contrary, the Zn(II) initially complexed with MT was released by successive additions of Cd(II) for an initial proportion of Zn(II)/MT /1.0/1.0 but not for an initial proportion of Zn(II)/MT /0.5/1.0. It is well known that metal ions like Cd, Cu, and Hg, have greater affinities for ligands than Zn and would thus be expected to displace Zn from other Zn binding sites through a metal /metal exchange
reaction (Dabrio and Rodriguez, 2000). Fig. 5 shows total higher frequency shifts compared to Fig. 4 and also after the second interaction with Zn(II) which is in agreement with not releasing of initially complexed Cd(II)-MT, but also provoking of Zn(II) complexed and higher frequency shift. In order to obtain the competitive adsorption of Cd(II) and Zn(II) ions, we prepared aqueous solutions containing these two ions having different Cd(II)/Zn(II) ratios. In all cases the total amount of metal ions was remained constant at 10 ppm in, and only the ratio of ions was changed. As seen in Fig. 6, when the ratio is 1
Fig. 5. Change in frequency shifts in the adsorption of Cd(II) and Zn(II) ions from their single-metal ion solutions containing different amounts of ions. The crystals were incubated first with the Cd(II) ions and then with the Zn(II) ions.
1044
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
Fig. 6. Change in frequency shifts in the adsorption of Cd(II) and Zn(II) ions from their solutions with different Cd(II)/Zn(II) ratios.
(which means that we used 5 ppm Cd(II) and 5 ppm Zn(II)) the total amount of ions adsorbed is very close to the values obtained from their single ion solutions (1589/34, 1309/30, 1389/28 Hz for Cd(II), Zn(II), and ‘50% Cd(II) and 50% Zn(II)’ mixture, respectively). However, when this ratio is lower or higher than 1 (especially higher than 1) the total amount increase quite significantly. It seems that competition triggers the interaction of the metal ions with the immobilized MT molecules on the crystals. These crystals show preference for the Cd(II) ions over Zn(II) ions (as noted from the higher adsorption at higher Cd(II)/Zn(II) ratios). This result is in agreement with data obtained with square wave voltammetry (Stillman, 1995), and apparently indicates that competition between the ions triggers the adsorption. The increase in metal binding in competitive adsorption may be due to synergistic / antagonistic effects (Stillman, 1995; Yamaguchi et al., 1997; Nieto and Rodriguez, 1999). In the following experiments, the MT carrying crystals were incubated with the aqueous solutions containing the mixture of Cd(II) and Zn(II) with different Cd(II)/Zn(II) ratios (I interaction), and the frequency shifts were measured. Then, these crystals were reincubated with the solutions containing the metal ions with different Zn(II)/Cd(II) ratios (II interaction). The responses for these two consecutive interactions are given in Fig. 7 and also in Table 1 (for convenience). We already discussed the competitive adsorption of metal ions with the immobilized MT molecules when we used different amounts of Cd(II) and Zn(II) ions at the first interaction, in the previous paragraph. For the second interaction, again interestingly, while the Zn(II)/Cd(II) ratio in the solution increases, the frequency shift
decreases. This may be an indication of Cd(II) release into the solution, which was adsorbed in the first step of the competitive adsorption. The frequency shifts after the second step interaction do not only show the adsorption of the metal ions, but also indicate the release of the adsorbed metal ions in the previous interaction that can occur in parallel with the adsorption of the metal ions onto the active surface. Hence, increasing the ratio of Zn(II)/Cd(II) in the second step provoked some release of the complexed Cd(II) ions in the previous step. As observed in the Fig. 7 (and also in Table 1) for the Zn(II)/Cd(II) ratio of 7/1 in the second interaction, the frequency shift was measured in the positive direction. This means that Zn(II) ions at this ratio may cause desorption of most of the Cd(II) ions (adsorbed previously) to be released. This result supports the above conclusion that adsorption and desorption onto the surface occur in parallel processes at different levels related to the ratios of the metal ions.
4. Conclusion A quartz crystal balance system was used to study interaction of MT molecules with Cd(II) and Zn(II). The MT molecules were in the immobilized form on the piezoelectric crystals, which allowed us to observe complexation reactions with a very simple system (apparatus), by just following the frequency shifts with a microbalance system. The results obtained were in agreement with results presented by others, which were observed in aqueous media. The main limitation of our approach was the quantification of the amount of metal ions adsorbed by the immobilized MT molecules. We
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
1045
Fig. 7. Competitive adsorption of Cd(II) and Zn(II) ions by MT carrying crystals from their mixtures after the I and II interactions.
Table 1 Competitive adsorption of Cd(II) and Zn(II) ions by MT carrying crystals from their mixtures after the I and II interactions. I interaction
II interaction
Type of metal ions
Concentration of metal ions (ppm)
Frequency shift (Hz)
Type of metal ions
Concentration of metal ions (ppm)
Frequency shift (Hz)
Cd(II) Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II)
10 10 1/7 1/3 3/5 1 1 2/3 3 7
158934 130930 280952 154924 98913 138928 230925 238948 3709100
Zn(II) Cd(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II)
10 10 7 3 1 2/3 10 3/5 1/3 1/7
493953 415945 385931 354929 315965 171956 112922 234929 160940
were not able to calculate the number of metal ions interacted with each MT molecule on the crystal surface, mainly due to limitations of the Saurbrey equation in this case. The highest Cd(II) binding capacity of the MTs was achieved at pH 7.4. The sensitivities (Df/DC ) of the sensors for Cd(II) and Zn(II) ions appear to be higher for Cd(II) than Zn(II) for adsorption from a singlemetal solution in the range of 1.25 /10 ppm. However, normalizing the data based on the atomic weights showed is just the opposite, and the immobilized MT molecules exhibit more preference to the Zn(II) ions compared with Cd(II) ions. Interestingly, in the consecutive adsorption studies, we observed that the type of metal ions used in the first interaction is important. When we first treated the crystals carrying the MT molecules with Cd(II) ions and then Zn(II) ions, higher total adsorption was reached. These experiments
showed also the exchange of the metal ions, namely exchange of the previously absorbed ones with the new ions in solution. In the experiments that we used mixtures of these two metal ions consecutively, we observed competition that was provoked adsorption of both ions due to synergistic /antagonistic effects, as was also indicated by others (Stillman, 1995; Yamaguchi et al., 1997; Nieto and Rodriguez, 1999). These competition causes also adsorption and desorption onto the surfaces which occur as parallel processes at different levels related to the ratios of the metal ions.
References Armitage, I.M., Otvos, J.D., Briggs, R.W., Boulanger, A.Y., 1982. Structure elucidation of the metal-binding sites in metallothionein by 113Cd NMR. Federation Proc. 41 (13), 2974 /2980.
1046
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
Chaoui, A., Ghorbal, M.H., Ferjani, E.E., 1997. Effects of cadmiumzinc interactions on hydroponically grown bean (Phaseolus vulgaris L.). Plant Sci. 126, 21 /28. Dabrio, M., Rodriguez, A.R., 2000. Study of complexing properties of the a-metallothionein domain with cadmium and/or zinc, using differential pulse polarography. Anal. Chem. Acta 424, 77 /90. Harlyk, C., Bordin, G., Nieto, O., Rodriguez, A.R., 1997. Square wave voltammetry data analysis by multivariate curve resolution: application to the mixed-metal system Cd-Zn-{Lys-Cys-Thr-CysCys-Ala [56 /61]. Electroanalysis 9 (8), 608 /613. Kagi, J.H., Himmelhoch, S.R., Whanger, P.D., Bethune, J.L., Valle, B.L., 1974. Equine hepatic and renal metallothioneins. Purification, molecular weight, amino acid composition, and metal content. The J. Biol. Chem. 249 (11), 3537 /3542. Mutlu, S., Saber, R., Koc¸um, I.C., Piskin, E., 1999. An immunosensor: Immobilization of anti-HBs antibody on glow discharge treated piezoelectric quartz crystal for HBs-Ag detection. Anal. Lett. 32 (2), 317 /334.
Nieto, O., Rodriguez, A.R., 1996. Complexing properties of the metallothionein fragment Lys-Cys-Thr-Cys-Cys-Ala [56 /61] MT 1 with zinc using square wave voltammetry. Bioelectrochem. Bioenergetics 40, 215 /222. Nieto, O., Rodriguez, A.R., 1999. Complexing properties of the metallothionein fragment Lys-Cys-Thr-Cys-Cys-Ala [56 /61] MT 1 with zinc and cadmium using square wave voltammetry. Electroanal. Bioenergetics 11 (3), 175 /182. Roesijadi, G., 1996. Metallothionein and its role in toxic metal regulation. Comp. Biochem. Physiol. 113 (C(2)), 117 /123. Saber, R., Mutlu, S., Piskin, E., 2002. Glow discharge treated piezoelectric quartz crystals as immunosensors for HSA detection. Biosensors Bioelectron. 17, 727 /734. Stillman, M.J., 1995. Metallothioneins. Coordination Chem. Rev. 144, 461 /511. Yamaguchi, R., Kurasaki, M., Kojima, Y., 1997. A mutant metallothionein which inverse fragment composition exhibits high cadmium-binding ability. Biochem. Biol. Int. 41 (1), 49 /56.
Investigation of complexation of immobilized metallothionein with Zn(II) and Cd(II) ions using piezoelectric crystals Reza Saber, Erhan Pis¸kin * Department of Chemical Engineering and Bioengineering Division, Hacettepe University, 06532 Beytepe, Ankara, Turkey Received 9 May 2002;space; received in revised form 3 September 2002; accepted 19 September 2002
Abstract The aim of this study is to investigate complexation of metallothionein (MT) with cadmium and zinc ions. An oligopeptide (i.e. Lys-Cys-Thr-Cys-Cys-Ala), a fragment of MT was covalently immobilized onto piezoelectric crystals, which were first treated with ethylene diamine plasma in a glow-discharge apparatus, and then were chemically reacted with glutaraldehyde. Complexation of the immobilized MT with Zn(II) and Cd(II) ions in aqueous media was followed by recording the changes of the frequency shifts of the piezoelectric quartz crystals. The amount of Cd(II) ions interacted with the immobilized MT molecules was the highest at pH 7.4, and decreased with an increase in the pH of the medium, in parallel to the decrease in the amount of immobilized MT. The number of Zn(II) ions interacted with the immobilized MT molecules was higher than the number of Cd(II) ions when the adsorption was from solutions containing a single-metal ion with the same ion concentrations. In consecutive adsorption studies, we observed that the type of metal ions used in the first interaction is important. These experiments showed also that there is an exchange between the metal ions, and competition provokes adsorption of both ions due to synergistic /antagonistic effects. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Piezoelectric quartz crystals; Glow-discharge; Plasma polymerization; Metallothionein fragment; Complexation with Zn(II) and Cd(II) ions
1. Introduction Metallothioneins (MTs) are a group of cysteine-rich, low molecular weight (B/6000D), and heavy metal binding proteins (Kagi et al., 1974; Chaoui et al., 1997; Armitage et al., 1982; Dabrio and Rodriguez, 2000; Harlyk et al., 1997). Their physiological roles include storage of heavy metal ions (especially Zn(II) and Cd(II)) and detoxification of heavy metals (e.g. Cd(II)). MTs are characterized by the lack of aromatic amino acids and by their selective capability to bind metallic ions. They contain 61 amino acids of which 20 are cysteinyl residues. Heavy metal ions are bound to
* Corresponding author. Tel.: /90-312-2977473; fax: /90-3122992124 E-mail address: [email protected] (E. Pis¸kin).
these cysteine residues through thiolate bonds by tetrahedral coordination. In the related literature, special attention has been paid to the study of complexing properties of MTs with heavy metal ions. The peptidic fragment ‘Lys-Cys-Thr-Cys-Cys-Ala [56 /61] MT 1’ of the mouse liver MT has been used as a model in several investigations that were aimed to better understand these complexing properties of MTs. In recent studies, electrochemical techniques have been used to investigate the complexing properties of this MT fragment, especially with zinc and cadmium ions (Nieto and Rodriguez, 1996; Mutlu et al., 1999; Nieto and Rodriguez, 1999). In this study, for the first time in the related literature, we used immobilized form (not free) of MT fragments to investigate complexation with Zn(II) and Cd(II) ions in aqueous media under different conditions by following frequency shifts of piezoelectric crystals.
0956-5663/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 5 6 - 5 6 6 3 ( 0 2 ) 0 0 2 1 7 - 8
1040
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
2. Material and methods 2.1. Materials The MT fragment was purchased from Sigma Chem. Corp. (Catalog No. L 4512, USA). Ethylene diamine (EDA) and glutaraldehyde (GA) were purchased from Fluka, Switzerland and from Aldrich (UK), respectively. All other reagents (Na2HPO4, NaH2PO4, Na2B4O7.10H2O, Tris) were purchased from Merck (Germany) and/or BDH (UK). 2.2. Metallothionein immobilization For immobilization of the MT fragments, the piezoelectric crystals were first treated in a glow-discharge apparatus, as described previously (Roesijadi, 1996). EDA was used as the active monomer to create aminolike functional groups on the quartz crystal surface. Briefly, the glow-discharge reactor was evacuated to 10 1 mbar. The monomer, EDA, was allowed to flow though the reactor at a flow rate of 35 ml/min. The quartz crystals were exposed to the EDA plasma for 3 min at glow-discharge power of 15 W. It should be noted that these conditions were selected after preliminary studies. In the first step of MT immobilization onto the crystal surfaces, the EDA-plasma-treated quartz crystals werincubated with a 2.5% GA solution (in a phosphate buffer, pH 7.4) at 4 8C for 16 h. These conditions were selected based on preliminary studies (Saber, 2002). The physically adsorbed GA molecules were removed by washing with distilled water several times. Following the washing step, the crystals were dried in air (at room temperature for about 2 h, until constant frequency measurements was obtained) and the frequencies of the air-dried crystals were noted. GA molecule was selected as it is a dialdehyde, containing two aldehyde groups such that one aldehyde group on each GA covalently interacts with an amine group on the EDA treated crystal, while the other (aldehyde group) remains free for the next treatment step. In the final step, the GA-attached quartz crystals were dipped into the MT solutions containing 0.30 mg MT in 1 ml buffer at different pH (7.4, 8.4, 9.4, 10.4). The incubation medium was mixed continuously on a rotary mixer (Ko¨ttermann Labortechnik, Germany) with a constant turning rate of 120 rpm, for 4 h at room temperature. Physically adsorbed MT molecules were removed by first washing the samples with the incubation buffer followed by incubation for 20 min in 0.15 M Tris buffer, and then with distilled water for additional 15 min. The frequency shifts were measured with the microbalance system described below at a constant temperature of 25 8C. It should be noted that, at this final step, the amine groups of the MT molecules react
with the free aldehyde groups on the piezoelectric crystals obtained in the previous step and are rendered covalently immobilized. 2.3. Microbalance system The microbalance system consists of an AT-cut 10 MHz quartz crystal of a 8.80 /0.16 mm2 wafer which is placed between two silver electrodes (diameter: 5 mm) and is mounted on a ceramic holder with a plug (AEC, Taiwan). The oscillation electronic circuit is a typical Collpits oscillator, which has a buffer amplifier. 15 V DC is applied to the oscillator circuit to drive the crystal and the frequency is measured with a Hewlett Packard frequency counter (Model No: HP 53181A, 225 MHz, USA). The drift observed of this system was less than 1 Hz in the measurements in air, after an initial stabilization period, which is about 5 min in air. 2.4. Complexation with heavy metal ions The MT-immobilized crystals were incubated with the aqueous solutions (20 ml) containing different amounts (1 /7 ppm in the first set and 10 /50 ppm in the second set) of zinc or cadmium nitrate at pH 7.4 (phosphate buffer) for 1 h on a rotary mixer. In order to study the competitive adsorption, the solutions containing mixtures of these two ions with different Cd(II)/Zn(II) ratios (1/7, 2/3, 3/5, 1/1, 5/3, 3/1, 7/1) were also used. Ten crystals were tested in each experiment. After 1 hinteraction, the crystals were taken out, washed several times with buffer and with distilled water, dried and the frequency shifts were measured in air.
3. Results and discussion 3.1. Metallothionein immobilization In this part of the study, MT molecules were first immobilized onto the GA-attached piezoelectric crystals. Immobilizations were achieved at different pH (7.4 and 8.4 in phosphate buffer, and 9.4 and 10.4 in tetraborate buffer). The initial concentration and immobilization time were 30 mg/ml and 4 h, respectively. In order to find the effectiveness of the immobilized MT molecules, these crystals were then incubated with Cd(II) solutions with an initial concentration of 50 ppm, at pH 7.4 and for 1 h. The extent of MT immobilization and the MT-Cd(II) complexation reaction was followed by measuring the frequency shifts with the micro-balance system after each step (Fig. 1). As seen here, the maximum efficiency of MT immobilization was achieved at pH 7.4 while at higher pH immobilizations were significantly lower. Probably, the three-dimensional structure of the MT molecules at pH
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
1041
Fig. 1. Effects of pH on MT immobilization and complexation of the immobilized MT with Cd(II) ions.
7.4 was more suitable for the interaction with the aldehyde groups on the piezoelectric crystals. It is interesting to note that this is the physiological pH, in which, most probably, MT molecules are in their native form, even in aqueous solutions that we have applied in the study. The amount of Cd(II) ions interacted with the immobilized MT molecules was also the highest at pH 7.4, and decreased with an increase in the pH of the medium, in parallel with the decrease in the amount of MT immobilized, as expected. We attempted to calculate the number of MT molecules immobilized onto the crystals, and also the number of metal ions interacted with these immobilized MT molecules from the Saurbrey equation by using the frequency shifts measured which are given below (Saber, 2002). However, the data were not stable (rather fluctuating) enough to make any conclusion, and therefore we decided that, this equation is not valid in all cases (such as in our case here), and these data should not be included here. Df 2:26 10 6 f 2 Dm=A here: Df is change in frequency, f is fundamental frequency of the crystal, Dm is the mass adsorbed on the surface, A is the surface area of the electrode. 3.2. Measurement of metal ions concentrations In this group of experiments we investigated the interactions of both Cd(II) and Zn(II) ions from their aqueous solution (containing single ions) at different concentrations and also from solutions containing both ions at different Cd(II)/Zn(II) ratios. Here the MTimmobilized crystals prepared at pH 7.4 were used. Fig. 2 shows the interaction of Cd(II) and Zn(II) ions of the
MT carrying crystals from their single-metal ion-solutions and from their mixtures. As observed from these data, the frequency shifts increased with an increase in the initial concentrations of either Cd(II) or Zn(II). It should be noted that frequency shifts are proportional to the deposited mass onto the crystal surface. Therefore, the increase in the frequency shift corresponds to the increase in the amount of metal ions interacted with the immobilized MT molecules (or simply the amount of metal ions adsorbed on the crystals). It is apparent from the frequency shifts in Fig. 2 that the MT carrying crystals adsorbed more Cd(II) ions than Zn(II) ions. However, this is not correct, as shown in Fig. 3, where the frequency shifts were divided by the atomic weights of the respective ions and then the graphs were redrawn. Thus, by compensating for the difference in the mass of the ions, the number of Zn(II) ions interacted with the immobilized MT molecules is higher than the number of Cd(II) ions when the adsorption was from solutions containing a single-metal ion with the same ion concentrations. In order to further investigate the adsorption behavior of Cd(II) and Zn(II) ions, we ran a second set of experiment where we first incubated solutions containing the desired amount of Zn(II) ions (or Cd(II) ions), measured the frequency shift, and then the same crystals were incubated with a solution containing Cd(II) ions (or Zn(II) ions) at the same concentration. These tests were repeated for different initial concentrations of metal ions, namely 10, 20, 30, 40, or 50 ppm. It should be noted that these concentrations were higher than those studied in the previous set of experiments. Figs. 4 and 5 show the frequency shifts after each interactions, and also total changes as a function of the initial concentration of metal ions. The only difference in these
1042
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
Fig. 2. Change in frequency shifts in the adsorption of Cd(II) and Zn(II) ions from their single-metal ion solutions containing different amounts of ions or their mixtures.
two figures is the type of metal ions used in the first and second interactions (as indicated also in the graphs). The first point that can be extracted from these two figures is that the amount of metal ions interacted with the immobilized MT molecules increases with the initial concentration of the metal ions in the solution, as expected. The more interesting point is the effect of the type of metal ions used in the first interaction. When we first treated the crystals carrying the MT molecules with Cd(II) ions and then Zn(II) ions, we reached higher total adsorption. It should also be noted that there
might be some kind of exchange of the metal ions between the aqueous phase (in the second incubation) and the crystals (already treated in the first incubation). Hence, Fig. 4 shows that after Zn(II) binding in the first interaction, the Cd(II) additions provoke the release of some or most of the complexed Zn(II) and enhances the complexation of Cd(II) with no modification of the coordination of Zn(II) with MT molecules which is in agreement with other researches (Stillman, 1995). However, we do not have sufficient data to determine the validity of this point. Nietto and Rodriguez used square
Fig. 3. Change in frequency shifts/atomic weight in the adsorption of Cd(II) and Zn(II) ions from their single-metal ion solutions containing different amounts of ions.
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
1043
Fig. 4. Change in frequency shifts in the adsorption of Cd(II) and Zn(II) ions from their single-metal ion solutions containing different amounts of ions. The crystals were incubated with first the Zn(II) ions and then with the Cd(II) ions.
wave voltammetry to study interaction of Cd(II) and Zn(II) ions with MT molecules in solution, and reported that the Cd(II) initially complexed by MT was not released when Zn(II) was added because the Cd(II) peak was not detected (Stillman, 1995). On the contrary, the Zn(II) initially complexed with MT was released by successive additions of Cd(II) for an initial proportion of Zn(II)/MT /1.0/1.0 but not for an initial proportion of Zn(II)/MT /0.5/1.0. It is well known that metal ions like Cd, Cu, and Hg, have greater affinities for ligands than Zn and would thus be expected to displace Zn from other Zn binding sites through a metal /metal exchange
reaction (Dabrio and Rodriguez, 2000). Fig. 5 shows total higher frequency shifts compared to Fig. 4 and also after the second interaction with Zn(II) which is in agreement with not releasing of initially complexed Cd(II)-MT, but also provoking of Zn(II) complexed and higher frequency shift. In order to obtain the competitive adsorption of Cd(II) and Zn(II) ions, we prepared aqueous solutions containing these two ions having different Cd(II)/Zn(II) ratios. In all cases the total amount of metal ions was remained constant at 10 ppm in, and only the ratio of ions was changed. As seen in Fig. 6, when the ratio is 1
Fig. 5. Change in frequency shifts in the adsorption of Cd(II) and Zn(II) ions from their single-metal ion solutions containing different amounts of ions. The crystals were incubated first with the Cd(II) ions and then with the Zn(II) ions.
1044
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
Fig. 6. Change in frequency shifts in the adsorption of Cd(II) and Zn(II) ions from their solutions with different Cd(II)/Zn(II) ratios.
(which means that we used 5 ppm Cd(II) and 5 ppm Zn(II)) the total amount of ions adsorbed is very close to the values obtained from their single ion solutions (1589/34, 1309/30, 1389/28 Hz for Cd(II), Zn(II), and ‘50% Cd(II) and 50% Zn(II)’ mixture, respectively). However, when this ratio is lower or higher than 1 (especially higher than 1) the total amount increase quite significantly. It seems that competition triggers the interaction of the metal ions with the immobilized MT molecules on the crystals. These crystals show preference for the Cd(II) ions over Zn(II) ions (as noted from the higher adsorption at higher Cd(II)/Zn(II) ratios). This result is in agreement with data obtained with square wave voltammetry (Stillman, 1995), and apparently indicates that competition between the ions triggers the adsorption. The increase in metal binding in competitive adsorption may be due to synergistic / antagonistic effects (Stillman, 1995; Yamaguchi et al., 1997; Nieto and Rodriguez, 1999). In the following experiments, the MT carrying crystals were incubated with the aqueous solutions containing the mixture of Cd(II) and Zn(II) with different Cd(II)/Zn(II) ratios (I interaction), and the frequency shifts were measured. Then, these crystals were reincubated with the solutions containing the metal ions with different Zn(II)/Cd(II) ratios (II interaction). The responses for these two consecutive interactions are given in Fig. 7 and also in Table 1 (for convenience). We already discussed the competitive adsorption of metal ions with the immobilized MT molecules when we used different amounts of Cd(II) and Zn(II) ions at the first interaction, in the previous paragraph. For the second interaction, again interestingly, while the Zn(II)/Cd(II) ratio in the solution increases, the frequency shift
decreases. This may be an indication of Cd(II) release into the solution, which was adsorbed in the first step of the competitive adsorption. The frequency shifts after the second step interaction do not only show the adsorption of the metal ions, but also indicate the release of the adsorbed metal ions in the previous interaction that can occur in parallel with the adsorption of the metal ions onto the active surface. Hence, increasing the ratio of Zn(II)/Cd(II) in the second step provoked some release of the complexed Cd(II) ions in the previous step. As observed in the Fig. 7 (and also in Table 1) for the Zn(II)/Cd(II) ratio of 7/1 in the second interaction, the frequency shift was measured in the positive direction. This means that Zn(II) ions at this ratio may cause desorption of most of the Cd(II) ions (adsorbed previously) to be released. This result supports the above conclusion that adsorption and desorption onto the surface occur in parallel processes at different levels related to the ratios of the metal ions.
4. Conclusion A quartz crystal balance system was used to study interaction of MT molecules with Cd(II) and Zn(II). The MT molecules were in the immobilized form on the piezoelectric crystals, which allowed us to observe complexation reactions with a very simple system (apparatus), by just following the frequency shifts with a microbalance system. The results obtained were in agreement with results presented by others, which were observed in aqueous media. The main limitation of our approach was the quantification of the amount of metal ions adsorbed by the immobilized MT molecules. We
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
1045
Fig. 7. Competitive adsorption of Cd(II) and Zn(II) ions by MT carrying crystals from their mixtures after the I and II interactions.
Table 1 Competitive adsorption of Cd(II) and Zn(II) ions by MT carrying crystals from their mixtures after the I and II interactions. I interaction
II interaction
Type of metal ions
Concentration of metal ions (ppm)
Frequency shift (Hz)
Type of metal ions
Concentration of metal ions (ppm)
Frequency shift (Hz)
Cd(II) Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II)
10 10 1/7 1/3 3/5 1 1 2/3 3 7
158934 130930 280952 154924 98913 138928 230925 238948 3709100
Zn(II) Cd(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II) Cd(II)/Zn(II)
10 10 7 3 1 2/3 10 3/5 1/3 1/7
493953 415945 385931 354929 315965 171956 112922 234929 160940
were not able to calculate the number of metal ions interacted with each MT molecule on the crystal surface, mainly due to limitations of the Saurbrey equation in this case. The highest Cd(II) binding capacity of the MTs was achieved at pH 7.4. The sensitivities (Df/DC ) of the sensors for Cd(II) and Zn(II) ions appear to be higher for Cd(II) than Zn(II) for adsorption from a singlemetal solution in the range of 1.25 /10 ppm. However, normalizing the data based on the atomic weights showed is just the opposite, and the immobilized MT molecules exhibit more preference to the Zn(II) ions compared with Cd(II) ions. Interestingly, in the consecutive adsorption studies, we observed that the type of metal ions used in the first interaction is important. When we first treated the crystals carrying the MT molecules with Cd(II) ions and then Zn(II) ions, higher total adsorption was reached. These experiments
showed also the exchange of the metal ions, namely exchange of the previously absorbed ones with the new ions in solution. In the experiments that we used mixtures of these two metal ions consecutively, we observed competition that was provoked adsorption of both ions due to synergistic /antagonistic effects, as was also indicated by others (Stillman, 1995; Yamaguchi et al., 1997; Nieto and Rodriguez, 1999). These competition causes also adsorption and desorption onto the surfaces which occur as parallel processes at different levels related to the ratios of the metal ions.
References Armitage, I.M., Otvos, J.D., Briggs, R.W., Boulanger, A.Y., 1982. Structure elucidation of the metal-binding sites in metallothionein by 113Cd NMR. Federation Proc. 41 (13), 2974 /2980.
1046
R. Saber, E. Pis¸kin / Biosensors and Bioelectronics 18 (2003) 1039 /1046
Chaoui, A., Ghorbal, M.H., Ferjani, E.E., 1997. Effects of cadmiumzinc interactions on hydroponically grown bean (Phaseolus vulgaris L.). Plant Sci. 126, 21 /28. Dabrio, M., Rodriguez, A.R., 2000. Study of complexing properties of the a-metallothionein domain with cadmium and/or zinc, using differential pulse polarography. Anal. Chem. Acta 424, 77 /90. Harlyk, C., Bordin, G., Nieto, O., Rodriguez, A.R., 1997. Square wave voltammetry data analysis by multivariate curve resolution: application to the mixed-metal system Cd-Zn-{Lys-Cys-Thr-CysCys-Ala [56 /61]. Electroanalysis 9 (8), 608 /613. Kagi, J.H., Himmelhoch, S.R., Whanger, P.D., Bethune, J.L., Valle, B.L., 1974. Equine hepatic and renal metallothioneins. Purification, molecular weight, amino acid composition, and metal content. The J. Biol. Chem. 249 (11), 3537 /3542. Mutlu, S., Saber, R., Koc¸um, I.C., Piskin, E., 1999. An immunosensor: Immobilization of anti-HBs antibody on glow discharge treated piezoelectric quartz crystal for HBs-Ag detection. Anal. Lett. 32 (2), 317 /334.
Nieto, O., Rodriguez, A.R., 1996. Complexing properties of the metallothionein fragment Lys-Cys-Thr-Cys-Cys-Ala [56 /61] MT 1 with zinc using square wave voltammetry. Bioelectrochem. Bioenergetics 40, 215 /222. Nieto, O., Rodriguez, A.R., 1999. Complexing properties of the metallothionein fragment Lys-Cys-Thr-Cys-Cys-Ala [56 /61] MT 1 with zinc and cadmium using square wave voltammetry. Electroanal. Bioenergetics 11 (3), 175 /182. Roesijadi, G., 1996. Metallothionein and its role in toxic metal regulation. Comp. Biochem. Physiol. 113 (C(2)), 117 /123. Saber, R., Mutlu, S., Piskin, E., 2002. Glow discharge treated piezoelectric quartz crystals as immunosensors for HSA detection. Biosensors Bioelectron. 17, 727 /734. Stillman, M.J., 1995. Metallothioneins. Coordination Chem. Rev. 144, 461 /511. Yamaguchi, R., Kurasaki, M., Kojima, Y., 1997. A mutant metallothionein which inverse fragment composition exhibits high cadmium-binding ability. Biochem. Biol. Int. 41 (1), 49 /56.