- Email: [email protected]
Determination of Metallothionein by High-Performance Liquid Chromatography with Fluorescence Detection Using an Isocratic Solvent System
ANALYTICAL BIOCHEMISTRY ARTICLE NO.
258, 168 –175 (1998)
AB982578
Determination of Metallothionein by High-Performance Liquid Chromatography with Fluorescence Detection Using an Isocratic Solvent System Shinichi Miyairi, Seiji Shibata, and Akira Naganuma1 Department of Molecular and Biochemical Toxicology, Faculty of Pharmaceutical Sciences, Tohoku University, Aoba-ku, Sendai 980-77, Japan
Received October 29, 1997
Metallothionein (MT) is a low-molecular-weight protein which plays a role in detoxification of heavy metals and protection against oxidative stress. A sensitive and convenient determination method for MT is necessary to clarify its physiological roles. High-performance liquid chromatography (HPLC) with an isocratic solvent system for MT was, therefore, developed utilizing an unique fluorescence labeling reagent, ammonium 7-fluorobenz2-oxa-1,3-diazole-4-sulfonate (SBD-F). The HPLC system with two separation columns, Shodex RSpak RP18-413 column (a styrene divinylbenzene polymer gel-packed column) and Puresil C18 column (an ODS gel-packed column), connected in tandem successfully separated SBD-labeled MT from biological interference. The SBDlabeled MT was stable and could be stored for at least 1 week without any changes in fluorescence intensity. Although Hg-MT was not detectable, this method is applicable to determination of major MTs such as Zn-MT, Cd-MT, and Cu-MT using commercially available rabbit MT as a standard. Determination of the MT concentration in cells was possible in aliquots of only 1 3 104 cultured cells. The present method using a tandem column HPLC system with isocratic elution might be useful for monitoring the concentration of MT in cultured cells as well as in animal tissues. © 1998 Academic Press
Metallothionein (MT)2 is characterized prominently to be a unique low-molecular-weight protein by its high cysteine content and high affinity to heavy metal ions 1 To whom correspondence should be addressed. Fax: 181-22-2176869. E-mail: [email protected]. 2 Abbreviations used: SBD-F, ammonium 7-fluorobenzo-2-oxa-1,3diazole-4-sulfonate; NPC, N-propionylcystamine; MT, metallothionein; ICP, inductively coupled plasma; mp, melting point; TBP, trin-butylphosphine; IS, internal standard; BCS, bathophenanthroline disulfonate.
168
such as cadmium, zinc, copper, and mercury (1). MT has been suggested to play important roles in zinc and copper homeostasis, detoxification of heavy metals, and protection against oxidative stress (1). Recently, neurophysiological and neuropathological studies have also been focused on clarifying the physiological importance of MT (2, 3). To clarify the physiological and pathophysiological roles of MT, it is necessary to monitor the frequent changes in concentration of this protein in tissues or cells. Since MT itself has no reliable functional groups for specific and sensitive detection (4), the MT concentration has been determined by indirect methods ordinarily based on the concentration of coordinated metal ions in the MT molecule. Among such methods, the most convenient and commonly used method is a metal saturation assay with radioactive (5) or nonradioactive metal (6). In this assay, the apparent MT concentration obtained was frequently influenced by the stability of MT–metal complex and the other biological components with affinity for the metal ion used for monitoring. Recently, high-performance liquid chromatography (HPLC) methods in combination with a metal detector such as atomic absorption spectroscopy (7), inductively coupled plasma (ICP)-atomic emission spectroscopy (8), or ICP-mass spectrometry (9) have been developed. However, these methods require specialized equipment and MT is still determined indirectly. Accordingly, the development of a convenient and reliable method for direct determination of MT is urgently needed. Among various methods for determination of the biological significance of MT, HPLC with fluorescence detection appears to be the most promising with respect to convenience, resolution, sensitivity, and versatility. Furthermore, it is generally accepted that HPLC with an isocratic solvent system is preferable for the determination of a specific component in numerous specimens rather than that using a gradient 0003-2697/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
CHROMATOGRAPHIC DETERMINATION OF METALLOTHIONEIN
169
tometer, and a Chromatocorder 21 data processor (Toso, Tokyo, Japan). Separation columns used were a Shodex RSpak RP18-413 column (150 3 4.6 mm internal diameter (i.d.); Showa Denko, Tokyo, Japan), GolfPak HR (150 3 4.6 mm i.d.; Millipore-Waters, Milford, MA), a Puresil C18 column (5 mm, 150 3 4.6 and 250 3 4.6 mm i.d.; Millipore-Waters), and Asahipak GS220HQ and GS-520HQ (300 3 7.6 mm i.d.; Asahikasei Kogyo, Kawasaki, Japan). 1H NMR spectra were recorded on a Varian FT-NMR R-220 spectrometer at 300 MHz using tetramethylsilane as an internal standard. Mass spectra were obtained by a JASCO JMSA X-500 mass spectrometer. Melting point (mp) values were measured on an electric micro hot stage apparatus and are given uncorrected. Animals and Cells
FIG. 1. Chromatographic profiles of SBD derivatives prepared from mouse liver and purified mouse MT-2 in HPLC with a gradient solvent system. Samples were prepared from the livers of mice treated with CdSO4 (A and B) or from purified mouse MT-2 (C). Chromatography was performed with a Puresil C18 column with a linear gradient from 10 to 50% CH3CN in 20 mM potassium phosphate buffer (pH 7.5) for 25 min (A and C), or a Golfpak HR column with a linear gradient from 12 to 60% CH3CN in 20 mM potassium phosphate buffer (pH 7.5) for 25 min (B). Fluorescence intensities of SBD derivatives were monitored with excitation at 384 nm and emission at 510 nm. Arrow indicates SBD–MT.
solvent system. The present paper deals with the development of a novel and convenient method for direct determination of MT by HPLC with fluorescence detection using an isocratic solvent system. MATERIALS AND METHODS
Materials Rat liver metallothionein I and II (MT-1 and MT-2) were generous gifts from Dr. M. Sato of Fukushima Medical College. Rabbit liver MT-1 and MT-2 were obtained from Sigma (St. Louis, MO). Mouse liver MT-1 and MT-2 were prepared by the method of Bu¨hler and Ka¨gi (10). Ammonium 7-fluorobenz-2-oxa-1,3-diazole-4-sulfonate (SBD-F) was purchased from Dojindo Laboratories (Kumamoto, Japan). Tri-n-butylphosphine (TBP) and cystamine z HCl were obtained from Nacalai Tesque (Kyoto, Japan). Apparatus The apparatus used for HPLC was a CCPM-II solvent delivery system equipped with an SD-8023 degasser, an SC-8020 system controller, an AS-8020 automatic injector, a FS-8020 fluorescence spectropho-
ICR/Slc mice (10 –13 weeks old, male) were given ZnSO4 (200 mmol/kg of weight), CdSO4 (5 mmol/kg of weight), or vehicle (saline) intraperitoneally once a day for 1 or 5 days. The liver or kidney collected 24 h after the last administration was subjected to reaction with SBD-F after homogenization as described below. Mouse L cells, HepG2 cells, and HeLa cells were cultured at 37°C for 24 h under a 5% CO2 atmosphere in Dulbecco’s modified Eagle’s medium in the presence of fetal calf serum (10% (v/v)), kanamycin sulfate (60 mg/ml), l-glutamine (316 mg/ml), and NaHCO3 (1 mg/ ml). Then, the cells were incubated for an additional 24 h with or without ZnCl2. The cells were recovered by centrifugation at 250 g for 5 min at 4°C after being scraped from the dish in phosphate-buffered saline. Sample Preparation Tissue was homogenized with 150 mM KCl solution with a Polytron homogenizer. The cells were lysed by ultrasonication in KCl solution containing an internal standard (IS) for quantitative determination. Homogenates and lysates were heated in boiling water for 5 min followed by centrifugation at 1000 g for 10 min at 4°C. The supernatant was subjected to derivatization reaction for determination of MT with SBD-F. The protein concentration was determined by the dye-binding assay method using bovine serum albumin as a standard (11). Derivatization of Metallothionein with SBD-F Samples (150 ml) were mixed with 1 M borate buffer (pH 10.5) containing 5 mM ethylenediaminetetraacetic acid (EDTA) (350 ml), 20% TBP/isopropanol (10 ml) and 0.5% SBD-F/water (40 ml). After heating for 30 min at 50°C, the reaction was terminated by the addition of 4 M HCl (50 ml). The resultant mixture was subjected to HPLC following dilution with water. Fluorescence in-
170
MIYAIRI, SHIBATA, AND NAGANUMA
FIG. 2. HPLC with isocratic solvent system chromatographic profiles of SBD derivatives prepared from mouse liver. Samples were prepared from the livers of mice treated with saline (1) or ZnSO4 (2). Separation columns used were a Shodex RSpak RP18-413 column (A), a Puresil C18 column (B), or a tandem column system with Shodex RSpak RP18-413 column and Puresil C18 column (C). Mobile phase, 20 mM potassium phosphate buffer (pH 7.5): acetonitrile:methanol (80:18:2); flow rate, 0.73 ml/min. Fluorescence intensities of SBD derivatives were monitored with excitation at 384 nm and emission at 510 nm. Arrow indicates SBD–MT.
tensity of MT derivatized with SBD-F was monitored with excitation at 384 nm and emission at 510 nm.
stand on ice with occasional shaking for 2 h. The supernatant was prepared by centrifugation at 1000 g for 10 min at 4°C after heating in boiling water for 5 min.
Preparation of Metal Ion-Substituted MT ZnSO4 (150 mmol/kg of weight) was administered intraperitoneally to mice once a day for 2 days. The liver was removed 24 h after the last administration of ZnSO4 and homogenized in 20 mM borate buffer (pH 8.0) (0.1 g of wet tissue/ml). For substitution of metal ions in the MT molecule the homogenate was mixed with an equal volume of CdCl2, CuCl2, or HgCl2 solution (0.5 mM) and then the mixture was allowed to
Preparation of 2,29-Dithio-bis[N-propionylethylamine] (N,N9-dipropionylcystamine) N,N9-Dipropionylcystamine was prepared for IS. A mixture of cystamine HCl (20 g) and propionic anhydride (40 ml) in pyridine (100 ml) was stirred at room temperature for 3 h with occasional heating at 70°C. The reaction was terminated by addition of water (30 ml). After the solvent was removed under reduced
FIG. 3. Effects of EDTA (A) or TBP (B) concentration on the peak area of SBD–MT. Samples were prepared from livers of mice treated with ZnSO4. Derivatization reaction with SBD-F was performed at 50°C for 30 min in borate buffer containing various concentrations of EDTA when TBP (10 ml of 10% solution) was added (A) or at 50°C for 30 min in borate buffer containing 5 mM of EDTA when various concentrations of TBP (10 ml) were added (B).
CHROMATOGRAPHIC DETERMINATION OF METALLOTHIONEIN
FIG. 4. Relationship between reaction period for SBD derivatization and peak area of SBD–MT. Samples were prepared from livers of mice treated with ZnSO4. Derivatization reaction with SBD-F was performed at 50°C in borate buffer containing 5 mM of EDTA when TBP (10 ml of 10% solution) was added.
pressure, the product was extracted with ethyl acetate. The organic layer was washed sequentially with NaOH (5%), HCl (5%), NaHCO3, (5%) and water, dried over anhydrous Na2SO4, and evaporated. The residue was crystallized from ethyl acetate to give N,N9-dipropionylcystamine (12 g) as colorless needles [mp 116°C. 1 H NMR (CDCl3)d: 1.10(6H, t, J 5 7.6 Hz, 2 3 CH3CH2CO), 2.19(4H, q, J 5 7.6 Hz, 2 3 CH3CH2CO), 2.77(4H, t, J 5 6.5 Hz, 2 3 NHCH2CH2S), 3.51(4H, q, J 5 6.3 Hz, 2 3 NHCH2CH2S), 6.26 (2H, bs, 2 3 NH). Anal. Calcd for C10H20N2O2S2: C, 45.42; H, 7.62; N, 10.60. Found: C, 45.63; H, 7.51; N, 10.53. MS (electron impact): 264 [M]1 (abbreviations used: t, triplet; q, quartet; bs, broad singlet)].
171
FIG. 5. Changes in peak areas and the peak area ratios of SBD–MT to SBD–NPC(IS) during storage at 4°C. Samples were prepared from livers of mice treated with ZnSO4, derivatized with SBD-F in the presence of NPC as an internal standard (IS), stored at 4°C in a refrigerator, and subjected to HPLC every day. Symbols: u, peak area of SBD–MT; ■, peak area of NPC(IS); Œ, the peak area ratios of SBD–MT to NPC (IS).
closely eluted with SBD–MT. Similar results were obtained by size exclusion column chromatography with an isocratic solvent system when Asahipak GS 220HQ and 520HQ columns were used (data not shown). In the reversed-phase column chromatography with an isocratic solvent system, although SBD–MT was eluted as a sharp peak, retention time of SBD-derivatized rabbit MT could not be controlled by changing the concentration of organic solvent in the mobile phase under various experimental condition examined with
RESULTS AND DISCUSSION
Development of HPLC System for Determination of MT after Derivatization with SBD-F MT in the liver homogenate from mice given ZnSO4 or CdSO4, a potent inducer of MT synthesis, were derivatized with SBD-F (12) for fluorescence labeling. To separate the SBD–MT from the other SBD-labeled substances using HPLC, a number of gradient elution systems were examined. However, we obtained no favorable separation of SBD–MT from the other SBDlabeled components in the liver. For example, the elution profiles of SBD–MT and other SBD-labeled components in the liver of CdSO4-treated mice from reversed-phase columns, the Puresil C18 column (ODS-packed column), or the Golf pak HR column (styrene divinylbenzene polymer gel-packed column) are shown in Fig. 1. The peaks corresponding to SBD–MT appeared with small but broad peaks which were not observed in the chromatogram of standard SBD–MT prepared from mouse MT-2, indicating that some inseparable interfering components were coeluted and/or
FIG. 6. Relation of amount of MT and the ratio of peak height of MT to NPC(IS). Rabbit MT-1 or MT-2 was derivatized with SBD-F in the presence of NPC as an internal standard (IS) and subjected to HPLC. Symbols: ³, rabbit MT-1 (r 5 0.997); r, rabbit MT-2 (r 5 0.997).
172
MIYAIRI, SHIBATA, AND NAGANUMA TABLE 1
The Recovery Rate of Metallothionein (MT) from Mouse Liver Homogenate MT added (mg)
Expected (mg)
Found (mg)
Recovery rate (%)
30 10 3.0 1.0
32.14 12.14 5.14 3.14
32.06 11.86 4.87 2.93
99.8 97.7 94.7 93.3
Note. Rabbit MT-2 was added to mouse liver homogenate, and then the MT concentration was determined by the tandem column HPLC method after derivatization with SBD-F. The concentration of MT in the mouse liver homogenate was 2.14 mg/ml.
several columns. Among the reversed-phase columns examined, the Shodex RSpak RP18-413 column (a styrene divinylbenzene polymer gel packed column) gave partial separation of SBD–MT from other peaks of SBD derivatives when liver samples from ZnSO4-treated mice were applied after derivatization with SBD-F (Fig. 2A). The most promising separation of SBD–MT peak was obtained by HPLC using a Puresil C18 column (250 3 4.6 mm i.d.) (Fig. 2B). However, the peak areas of SBD–MT eluted from the Puresil C18 column were significantly larger than those from the Shodex RSpak RP18-413 column. The difference in peak areas of SBD–MT in the chromatograms of the control mouse liver was large as about 6 to 1. These observations suggested that SBD derivatives of hepatic substance(s) other than MT were eluted together with SBD–MT from the Puresil C18 column. To utilize the favorable characteristics of these columns simultaneously, we connected a Shodex RSpak column and a Puresil C18 column to the HPLC system in tandem. In this system, a peak corresponding to SBD–MT was satisfactory separated from other peaks as shown in Fig. 2C. In the chromatograms of preparations from livers of control
and ZnSO4-treated mice, peak areas of SBD–MT eluted from the tandem column system were 23 3 103 and 502 3 103, respectively, while the peak areas were 139 3 103 and 743 3 103 from the Puresil C18 column and 24 3 103 and 660 3 103 from the Shodex RSpak column. These results suggest that the peak of SBD–MT from the tandem column system was eluted without any interference. Optimization of SBD Derivatization Reaction of MT To examine the effects of EDTA, a chelating agent, on derivatization of MT with SBD-F, heated liver homogenate from mice given CdSO4 was reacted with SBD-F in 1 M borate buffer (pH 10.5) containing EDTA at various concentrations. While EDTA did not affect the derivatization of glutathione with SBD-F (data not shown), fluorescence intensity of the peak area corresponding to SBD–MT was increased by derivatization of MT with SBD-F in the presence of EDTA at concentrations in excess of 1 mM in the reaction buffer (350 ml of 1 M borate buffer, pH 10.5) (Fig. 3A). EDTA may be effective for removing coordinated metal ions such as zinc and cadmium from MT. These results indicate that it may be necessary for thiol groups in MT to be in the free form in the derivatization reaction with SBD-F, and addition of a chelating agent is essential for the derivatization of MT. The effects of TBP, a reducing agent, on the reaction of MT with SBD-F were also examined. The peak area of SBD–MT was increased by about seven-fold by the addition of TBP solution (10 ml) at concentrations of 10% or higher to the reaction mixture (Fig. 3B). MT in tissue samples seems to be easily oxidized during sample preparation or derivatization reaction. Figure 4 shows the effects of reaction time on derivatization of MT with BSD-F. The peak area of SBD–MT increased with reaction time up to 20 min and was constant up to 40 min when the
FIG. 7. Apparent MT concentration determined in the liver homogenate containing a specific metal binding form of MT derivatized with SBD-F. Samples were prepared from liver homogenate containing Zn–, Cd–, Cu–, or Hg–MT. Derivatization reaction with SBD-F was performed at 50°C for 30 min in borate buffer containing EDTA (5 mM) (A) or EDTA (5 mM) and BCS (1 mM)(B) (n 5 3).
CHROMATOGRAPHIC DETERMINATION OF METALLOTHIONEIN
173
When the peak height ratio of MT to internal standard was plotted against the concentration of MT, the calibration curve showed good linearity in the range of 70 ng/ml to 50 mg/ml for rabbit MT-1 or MT-2 (Fig. 6). The detection limit of SBD–MT was a few hundred picograms per injection. Recovery of rabbit MT-2 from mouse liver homogenate was satisfactory (more than 93%) in the range of 1 to 30 mg/ml (Table 1). The accuracy of the obtained values clarified by the intraand interassay coefficients of variation was 5.1% (n 5 7) and 3.9% (n 5 3), respectively. Effects of Metal Ion Coordinated in MT on Derivatization with SBD-F FIG. 8. Determination of MT in mouse liver and kidney determined by the HPLC system. Samples were prepared from liver (A) and kidney (B) of mouse treated for 5 days with vehicle (■) (n 5 3), ZnSO4 (u) (n 5 5) or CdSO4 (o) (n 5 5).
supernatant of heat-denatured liver homogenate of mice treated with ZnSO4 was subjected to the reaction. According to these results, it was concluded that MT in biological samples (150 ml) was sufficiently converted into SBD–MT in a mixture of 1 M borate buffer (pH 10.5) containing 5 mM EDTA (350 ml), 20% TBP in isopropanol (10 ml), and 0.5% SBD-F in water (40 ml) for 30 min at 50°C.
Removal of metal ions from the MT molecule by chelating agents is essential for derivatization of MT with SBD-F as described above. The efficiency of removal of metal ions from MT may depend on the affinity of the chelating agent to the metal ion and its concentration. Metal ions such as Zn21, Cu21, Cd21, and Hg21 can bind to MT in vivo, and relative magnitude of affinity of the metals to MT is in the order of Zn21 , Cd21 , Cu21 , Hg21 (13). To examine the effect of EDTA on removal of metals from MT, liver homogenates containing Cd–MT, Cu–MT, or Hg–MT with the same concentration of MT were prepared by replacement of the metal ions from liver homogenate of ZnSO4-treated mice which predominantly contained Zn–MT. These homogenates were employed for deri-
Evaluation of the HPLC System To employ this HPLC method for quantitative determination of MT, N,N9-dipropionylcystamine, the oxidized dimer of N-propionylcystamine (NPC), was newly prepared for an appropriate internal standard. This compound reacts with SBD-F as well as MT when thiol groups are generated by TBP. Using N,N9-dipropionylcystamine as an IS, the tandem column HPLC method established here was evaluated as follows. Figure 5 shows the stability of SBD-MT during storage at 4°C. The peak area ratio of SBD-MT to SBD– NPC(IS) was constant and maintained for at least for 5 days, while the apparent peak areas of both SBD derivatives increased with time in storage, suggesting that SBD-MT is stable chemically and can be stored for several days before analysis. Some isoforms of MT have been identified in mammalian tissues (1). Retention times of SBD derivatives of isoforms of MT (MT-1 and MT-2) from rabbit, rat, and mouse were identical in the tandem column method. Fluorescence intensities of these MT isoforms after derivatization with SBD-F were also almost equivalent (data not shown). These results suggest that total amount of MT can be determined by this tandem column system, and commercially available rabbit MT can be used as a standard sample for determination of MT in several animal species.
FIG. 9. Chromatograms of SBD–MT prepared from mouse L cells cultured with or without ZnCl2. Samples were prepared from mouse L cells cultured with (150 mM) (B) or without ZnCl2 (A). The tandem column system with a Shodex RSpak RP18-413 column and a Puresil C18 column was used. Mobile phase, 20 mM potassium phosphate buffer (pH 7.5):acetonitrile:methanol (80:18:2); flow rate, 0.7 ml/min. MT and IS in the figure represented retention times of SBD–MT and SBD–NPC used as an internal standard (IS).
174
MIYAIRI, SHIBATA, AND NAGANUMA
FIG. 10. Determination of cellular MT levels in HepG2, HeLa, and mouse L cells. Samples were prepared from HepG2 cells (A), HeLa cells (B), or mouse L cells (C) after 24-h incubation with various concentrations of ZnCl2.
vatization with SBD-F in the presence of EDTA (5 mM) and for determination of MT by the present HPLC procedure. After heating these homogenate, excess amounts of added metal ions (Cd21, Cu21, or Hg21) were precipitated with denatured proteins and most of each metal ion in the supernatant was bound to MT. The concentration of MT determined by the HPLC method in the heated supernatant containing Zn–MT was comparable with that in the heated-supernatant containing Cd–MT. However, the apparent concentrations of MT in the heated supernatant containing Cu–MT or Hg–MT were considerably lower than those in Zn–MT or Cd–MT (Fig. 7A). When the derivatization reaction was carried out in borate buffer containing 1 mM bathophenanthroline disulfonate (BCS) in addition to 5 mM EDTA, the MT concentration determined by the HPLC method in the heated supernatant containing Cu–MT was almost identical to that containing Zn–MT (Fig. 7B). This indicated that utilization of BCS in combination with EDTA as a chelating agent is effective for the derivatization of Cu–MT with SBD-F. Long–Evans Cinnamon rats spontaneously develop hepatoma and accumulate copper ions in the liver as a complex with MT (14). Animal models of Menkes disease (15) also accumulate excessive concentrations of copper in their kidneys. Derivatization of MT with SBD-F in the presence of BCS and EDTA might be useful for the determination of Cu–MT in tissues or cultured cells from these animals using the present HPLC method.
Determination of MT in Mouse Tissues and Cultured Cells Mice (n 5 5) were intraperitoneally administered ZnSO4 (200 mmol/kg of weight), CdSO4 (5 mmol/kg of weight), or vehicle (saline) (0.1 ml) once a day for 5 days. The liver and kidneys collected 24 h after the last injection were subjected to HPLC analysis for the determination of MT concentration. The basal MT levels
of liver and kidney were 15 and 38 mg/g of wet tissue, respectively. The MT levels of these tissues were elevated to 655 and 261 mg/g by treatment with ZnSO4, and 1163 and 298 mg/g by treatment with CdSO4, respectively (Fig. 8). MT concentrations in HeLa, HepG2, and mouse L cells cultured with or without ZnCl2 were also determined by the present HPLC method. Reliable chromatograms were obtained for the basal and induced levels of MT in L cells (Fig. 9) as well as the other two cell lines. The basal levels of MT in HepG2, HeLa, and L cells were 4.7, 2.3, and 0.3 mg/mg of protein, respectively. These MT levels were increased dose-dependently by treatment with ZnCl2, and the highest concentrations of MT observed were 132 mg/mg of protein in HepG2 cells (at 150 mM of ZnCl2), 22 mg/mg in HeLa cells (at 100 mM of ZnCl2), and 10 mg/mg in L cells (at 150 mM of ZnCl2) (Fig. 10). In the present HPLC method, aliquots of only about 1 3 104 cells were needed for determination of concentration of MT. In conclusion, a sensitive, reliable, and convenient method for the determination of MT using a tandem column HPLC system with isocratic elution was established in the present study. Pretreatment of samples and the derivatization reaction in this method are quite simple, and the fluorescence intensity of the resultant SBD-derivatized MT did not decrease as a function of time over the week. Although Hg–MT was not detectable by the present method, only a few studies have been performed on Hg–MT. The tandem column HPLC method presented here might be useful for the determination of most major MTs such as Zn–MT, Cd– MT, and Cu-MT. ACKNOWLEDGMENTS This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan and a grant for drug development from the Japan Health Science Foundation.
CHROMATOGRAPHIC DETERMINATION OF METALLOTHIONEIN
REFERENCES 1. Ka¨gi, J. H. R. (1993) in Metallothionein III (Suzuki, K. T., Imura, N., and Kimura, M., Eds.), pp. 29 –55, Birkha¨user Verlag, Basel, Switzerland. 2. Uchida, Y., Takio, K., Titani, K., Ihara, Y., and Tomonaga, M. (1991) Neuron 7, 337–347. 3. Vitarella, D., Conklin, D. R., Kimelberg, H. K., and Aschner, M. (1996) Brain Res. 738, 213–221. 4. Ka¨gi, J. H. R. (1991) Methods Enzymol. 205, 613– 626. 5. Patrierno, S. R., Dellis, N. R., Evans, R. M., and Costa, M. (1983) Life Sci. 32, 1629 –1636. 6. Onosaka, S., and Cherian, M. G. (1981) Toxicology 22, 91–101. 7. Lehman, L. D., and Klaassen, C. D. (1986) Anal. Biochem. 153, 305–314.
175
8. Suzuki, K. T. (1991) Methods Enzymol. 205, 198 –205. 9. Sunaga, H., Kobayashi, E., Shimojo, N., and Suzuki, K. T. (1987) Anal. Biochem. 160, 160 –168. 10. Bu¨hler, R., and Ka¨gi, J. H. R. (1974) FEBS Lett. 39, 229 – 234. 11. Bradford, M. M. (1976) Anal. Biochem. 72, 248 –254. 12. Imai, K., Toyo’oka, T., and Watanabe, Y. (1983) Anal. Biochem. 128, 471– 473. 13. Webb, M. (1979) in The Chemistry, Biochemistry and Biology of Cadmium (Webb, M., Ed.), pp. 195–266, Elsevier/North Holland, Amsterdam. 14. Sugawara, N., Sugawara, C., Sato, M., and Mori, M. (1993) Trace Elem. Exp. Med. 6, 15–21. 15. Hunt, D. M. (1974) Nature 249, 852– 854.
258, 168 –175 (1998)
AB982578
Determination of Metallothionein by High-Performance Liquid Chromatography with Fluorescence Detection Using an Isocratic Solvent System Shinichi Miyairi, Seiji Shibata, and Akira Naganuma1 Department of Molecular and Biochemical Toxicology, Faculty of Pharmaceutical Sciences, Tohoku University, Aoba-ku, Sendai 980-77, Japan
Received October 29, 1997
Metallothionein (MT) is a low-molecular-weight protein which plays a role in detoxification of heavy metals and protection against oxidative stress. A sensitive and convenient determination method for MT is necessary to clarify its physiological roles. High-performance liquid chromatography (HPLC) with an isocratic solvent system for MT was, therefore, developed utilizing an unique fluorescence labeling reagent, ammonium 7-fluorobenz2-oxa-1,3-diazole-4-sulfonate (SBD-F). The HPLC system with two separation columns, Shodex RSpak RP18-413 column (a styrene divinylbenzene polymer gel-packed column) and Puresil C18 column (an ODS gel-packed column), connected in tandem successfully separated SBD-labeled MT from biological interference. The SBDlabeled MT was stable and could be stored for at least 1 week without any changes in fluorescence intensity. Although Hg-MT was not detectable, this method is applicable to determination of major MTs such as Zn-MT, Cd-MT, and Cu-MT using commercially available rabbit MT as a standard. Determination of the MT concentration in cells was possible in aliquots of only 1 3 104 cultured cells. The present method using a tandem column HPLC system with isocratic elution might be useful for monitoring the concentration of MT in cultured cells as well as in animal tissues. © 1998 Academic Press
Metallothionein (MT)2 is characterized prominently to be a unique low-molecular-weight protein by its high cysteine content and high affinity to heavy metal ions 1 To whom correspondence should be addressed. Fax: 181-22-2176869. E-mail: [email protected]. 2 Abbreviations used: SBD-F, ammonium 7-fluorobenzo-2-oxa-1,3diazole-4-sulfonate; NPC, N-propionylcystamine; MT, metallothionein; ICP, inductively coupled plasma; mp, melting point; TBP, trin-butylphosphine; IS, internal standard; BCS, bathophenanthroline disulfonate.
168
such as cadmium, zinc, copper, and mercury (1). MT has been suggested to play important roles in zinc and copper homeostasis, detoxification of heavy metals, and protection against oxidative stress (1). Recently, neurophysiological and neuropathological studies have also been focused on clarifying the physiological importance of MT (2, 3). To clarify the physiological and pathophysiological roles of MT, it is necessary to monitor the frequent changes in concentration of this protein in tissues or cells. Since MT itself has no reliable functional groups for specific and sensitive detection (4), the MT concentration has been determined by indirect methods ordinarily based on the concentration of coordinated metal ions in the MT molecule. Among such methods, the most convenient and commonly used method is a metal saturation assay with radioactive (5) or nonradioactive metal (6). In this assay, the apparent MT concentration obtained was frequently influenced by the stability of MT–metal complex and the other biological components with affinity for the metal ion used for monitoring. Recently, high-performance liquid chromatography (HPLC) methods in combination with a metal detector such as atomic absorption spectroscopy (7), inductively coupled plasma (ICP)-atomic emission spectroscopy (8), or ICP-mass spectrometry (9) have been developed. However, these methods require specialized equipment and MT is still determined indirectly. Accordingly, the development of a convenient and reliable method for direct determination of MT is urgently needed. Among various methods for determination of the biological significance of MT, HPLC with fluorescence detection appears to be the most promising with respect to convenience, resolution, sensitivity, and versatility. Furthermore, it is generally accepted that HPLC with an isocratic solvent system is preferable for the determination of a specific component in numerous specimens rather than that using a gradient 0003-2697/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
CHROMATOGRAPHIC DETERMINATION OF METALLOTHIONEIN
169
tometer, and a Chromatocorder 21 data processor (Toso, Tokyo, Japan). Separation columns used were a Shodex RSpak RP18-413 column (150 3 4.6 mm internal diameter (i.d.); Showa Denko, Tokyo, Japan), GolfPak HR (150 3 4.6 mm i.d.; Millipore-Waters, Milford, MA), a Puresil C18 column (5 mm, 150 3 4.6 and 250 3 4.6 mm i.d.; Millipore-Waters), and Asahipak GS220HQ and GS-520HQ (300 3 7.6 mm i.d.; Asahikasei Kogyo, Kawasaki, Japan). 1H NMR spectra were recorded on a Varian FT-NMR R-220 spectrometer at 300 MHz using tetramethylsilane as an internal standard. Mass spectra were obtained by a JASCO JMSA X-500 mass spectrometer. Melting point (mp) values were measured on an electric micro hot stage apparatus and are given uncorrected. Animals and Cells
FIG. 1. Chromatographic profiles of SBD derivatives prepared from mouse liver and purified mouse MT-2 in HPLC with a gradient solvent system. Samples were prepared from the livers of mice treated with CdSO4 (A and B) or from purified mouse MT-2 (C). Chromatography was performed with a Puresil C18 column with a linear gradient from 10 to 50% CH3CN in 20 mM potassium phosphate buffer (pH 7.5) for 25 min (A and C), or a Golfpak HR column with a linear gradient from 12 to 60% CH3CN in 20 mM potassium phosphate buffer (pH 7.5) for 25 min (B). Fluorescence intensities of SBD derivatives were monitored with excitation at 384 nm and emission at 510 nm. Arrow indicates SBD–MT.
solvent system. The present paper deals with the development of a novel and convenient method for direct determination of MT by HPLC with fluorescence detection using an isocratic solvent system. MATERIALS AND METHODS
Materials Rat liver metallothionein I and II (MT-1 and MT-2) were generous gifts from Dr. M. Sato of Fukushima Medical College. Rabbit liver MT-1 and MT-2 were obtained from Sigma (St. Louis, MO). Mouse liver MT-1 and MT-2 were prepared by the method of Bu¨hler and Ka¨gi (10). Ammonium 7-fluorobenz-2-oxa-1,3-diazole-4-sulfonate (SBD-F) was purchased from Dojindo Laboratories (Kumamoto, Japan). Tri-n-butylphosphine (TBP) and cystamine z HCl were obtained from Nacalai Tesque (Kyoto, Japan). Apparatus The apparatus used for HPLC was a CCPM-II solvent delivery system equipped with an SD-8023 degasser, an SC-8020 system controller, an AS-8020 automatic injector, a FS-8020 fluorescence spectropho-
ICR/Slc mice (10 –13 weeks old, male) were given ZnSO4 (200 mmol/kg of weight), CdSO4 (5 mmol/kg of weight), or vehicle (saline) intraperitoneally once a day for 1 or 5 days. The liver or kidney collected 24 h after the last administration was subjected to reaction with SBD-F after homogenization as described below. Mouse L cells, HepG2 cells, and HeLa cells were cultured at 37°C for 24 h under a 5% CO2 atmosphere in Dulbecco’s modified Eagle’s medium in the presence of fetal calf serum (10% (v/v)), kanamycin sulfate (60 mg/ml), l-glutamine (316 mg/ml), and NaHCO3 (1 mg/ ml). Then, the cells were incubated for an additional 24 h with or without ZnCl2. The cells were recovered by centrifugation at 250 g for 5 min at 4°C after being scraped from the dish in phosphate-buffered saline. Sample Preparation Tissue was homogenized with 150 mM KCl solution with a Polytron homogenizer. The cells were lysed by ultrasonication in KCl solution containing an internal standard (IS) for quantitative determination. Homogenates and lysates were heated in boiling water for 5 min followed by centrifugation at 1000 g for 10 min at 4°C. The supernatant was subjected to derivatization reaction for determination of MT with SBD-F. The protein concentration was determined by the dye-binding assay method using bovine serum albumin as a standard (11). Derivatization of Metallothionein with SBD-F Samples (150 ml) were mixed with 1 M borate buffer (pH 10.5) containing 5 mM ethylenediaminetetraacetic acid (EDTA) (350 ml), 20% TBP/isopropanol (10 ml) and 0.5% SBD-F/water (40 ml). After heating for 30 min at 50°C, the reaction was terminated by the addition of 4 M HCl (50 ml). The resultant mixture was subjected to HPLC following dilution with water. Fluorescence in-
170
MIYAIRI, SHIBATA, AND NAGANUMA
FIG. 2. HPLC with isocratic solvent system chromatographic profiles of SBD derivatives prepared from mouse liver. Samples were prepared from the livers of mice treated with saline (1) or ZnSO4 (2). Separation columns used were a Shodex RSpak RP18-413 column (A), a Puresil C18 column (B), or a tandem column system with Shodex RSpak RP18-413 column and Puresil C18 column (C). Mobile phase, 20 mM potassium phosphate buffer (pH 7.5): acetonitrile:methanol (80:18:2); flow rate, 0.73 ml/min. Fluorescence intensities of SBD derivatives were monitored with excitation at 384 nm and emission at 510 nm. Arrow indicates SBD–MT.
tensity of MT derivatized with SBD-F was monitored with excitation at 384 nm and emission at 510 nm.
stand on ice with occasional shaking for 2 h. The supernatant was prepared by centrifugation at 1000 g for 10 min at 4°C after heating in boiling water for 5 min.
Preparation of Metal Ion-Substituted MT ZnSO4 (150 mmol/kg of weight) was administered intraperitoneally to mice once a day for 2 days. The liver was removed 24 h after the last administration of ZnSO4 and homogenized in 20 mM borate buffer (pH 8.0) (0.1 g of wet tissue/ml). For substitution of metal ions in the MT molecule the homogenate was mixed with an equal volume of CdCl2, CuCl2, or HgCl2 solution (0.5 mM) and then the mixture was allowed to
Preparation of 2,29-Dithio-bis[N-propionylethylamine] (N,N9-dipropionylcystamine) N,N9-Dipropionylcystamine was prepared for IS. A mixture of cystamine HCl (20 g) and propionic anhydride (40 ml) in pyridine (100 ml) was stirred at room temperature for 3 h with occasional heating at 70°C. The reaction was terminated by addition of water (30 ml). After the solvent was removed under reduced
FIG. 3. Effects of EDTA (A) or TBP (B) concentration on the peak area of SBD–MT. Samples were prepared from livers of mice treated with ZnSO4. Derivatization reaction with SBD-F was performed at 50°C for 30 min in borate buffer containing various concentrations of EDTA when TBP (10 ml of 10% solution) was added (A) or at 50°C for 30 min in borate buffer containing 5 mM of EDTA when various concentrations of TBP (10 ml) were added (B).
CHROMATOGRAPHIC DETERMINATION OF METALLOTHIONEIN
FIG. 4. Relationship between reaction period for SBD derivatization and peak area of SBD–MT. Samples were prepared from livers of mice treated with ZnSO4. Derivatization reaction with SBD-F was performed at 50°C in borate buffer containing 5 mM of EDTA when TBP (10 ml of 10% solution) was added.
pressure, the product was extracted with ethyl acetate. The organic layer was washed sequentially with NaOH (5%), HCl (5%), NaHCO3, (5%) and water, dried over anhydrous Na2SO4, and evaporated. The residue was crystallized from ethyl acetate to give N,N9-dipropionylcystamine (12 g) as colorless needles [mp 116°C. 1 H NMR (CDCl3)d: 1.10(6H, t, J 5 7.6 Hz, 2 3 CH3CH2CO), 2.19(4H, q, J 5 7.6 Hz, 2 3 CH3CH2CO), 2.77(4H, t, J 5 6.5 Hz, 2 3 NHCH2CH2S), 3.51(4H, q, J 5 6.3 Hz, 2 3 NHCH2CH2S), 6.26 (2H, bs, 2 3 NH). Anal. Calcd for C10H20N2O2S2: C, 45.42; H, 7.62; N, 10.60. Found: C, 45.63; H, 7.51; N, 10.53. MS (electron impact): 264 [M]1 (abbreviations used: t, triplet; q, quartet; bs, broad singlet)].
171
FIG. 5. Changes in peak areas and the peak area ratios of SBD–MT to SBD–NPC(IS) during storage at 4°C. Samples were prepared from livers of mice treated with ZnSO4, derivatized with SBD-F in the presence of NPC as an internal standard (IS), stored at 4°C in a refrigerator, and subjected to HPLC every day. Symbols: u, peak area of SBD–MT; ■, peak area of NPC(IS); Œ, the peak area ratios of SBD–MT to NPC (IS).
closely eluted with SBD–MT. Similar results were obtained by size exclusion column chromatography with an isocratic solvent system when Asahipak GS 220HQ and 520HQ columns were used (data not shown). In the reversed-phase column chromatography with an isocratic solvent system, although SBD–MT was eluted as a sharp peak, retention time of SBD-derivatized rabbit MT could not be controlled by changing the concentration of organic solvent in the mobile phase under various experimental condition examined with
RESULTS AND DISCUSSION
Development of HPLC System for Determination of MT after Derivatization with SBD-F MT in the liver homogenate from mice given ZnSO4 or CdSO4, a potent inducer of MT synthesis, were derivatized with SBD-F (12) for fluorescence labeling. To separate the SBD–MT from the other SBD-labeled substances using HPLC, a number of gradient elution systems were examined. However, we obtained no favorable separation of SBD–MT from the other SBDlabeled components in the liver. For example, the elution profiles of SBD–MT and other SBD-labeled components in the liver of CdSO4-treated mice from reversed-phase columns, the Puresil C18 column (ODS-packed column), or the Golf pak HR column (styrene divinylbenzene polymer gel-packed column) are shown in Fig. 1. The peaks corresponding to SBD–MT appeared with small but broad peaks which were not observed in the chromatogram of standard SBD–MT prepared from mouse MT-2, indicating that some inseparable interfering components were coeluted and/or
FIG. 6. Relation of amount of MT and the ratio of peak height of MT to NPC(IS). Rabbit MT-1 or MT-2 was derivatized with SBD-F in the presence of NPC as an internal standard (IS) and subjected to HPLC. Symbols: ³, rabbit MT-1 (r 5 0.997); r, rabbit MT-2 (r 5 0.997).
172
MIYAIRI, SHIBATA, AND NAGANUMA TABLE 1
The Recovery Rate of Metallothionein (MT) from Mouse Liver Homogenate MT added (mg)
Expected (mg)
Found (mg)
Recovery rate (%)
30 10 3.0 1.0
32.14 12.14 5.14 3.14
32.06 11.86 4.87 2.93
99.8 97.7 94.7 93.3
Note. Rabbit MT-2 was added to mouse liver homogenate, and then the MT concentration was determined by the tandem column HPLC method after derivatization with SBD-F. The concentration of MT in the mouse liver homogenate was 2.14 mg/ml.
several columns. Among the reversed-phase columns examined, the Shodex RSpak RP18-413 column (a styrene divinylbenzene polymer gel packed column) gave partial separation of SBD–MT from other peaks of SBD derivatives when liver samples from ZnSO4-treated mice were applied after derivatization with SBD-F (Fig. 2A). The most promising separation of SBD–MT peak was obtained by HPLC using a Puresil C18 column (250 3 4.6 mm i.d.) (Fig. 2B). However, the peak areas of SBD–MT eluted from the Puresil C18 column were significantly larger than those from the Shodex RSpak RP18-413 column. The difference in peak areas of SBD–MT in the chromatograms of the control mouse liver was large as about 6 to 1. These observations suggested that SBD derivatives of hepatic substance(s) other than MT were eluted together with SBD–MT from the Puresil C18 column. To utilize the favorable characteristics of these columns simultaneously, we connected a Shodex RSpak column and a Puresil C18 column to the HPLC system in tandem. In this system, a peak corresponding to SBD–MT was satisfactory separated from other peaks as shown in Fig. 2C. In the chromatograms of preparations from livers of control
and ZnSO4-treated mice, peak areas of SBD–MT eluted from the tandem column system were 23 3 103 and 502 3 103, respectively, while the peak areas were 139 3 103 and 743 3 103 from the Puresil C18 column and 24 3 103 and 660 3 103 from the Shodex RSpak column. These results suggest that the peak of SBD–MT from the tandem column system was eluted without any interference. Optimization of SBD Derivatization Reaction of MT To examine the effects of EDTA, a chelating agent, on derivatization of MT with SBD-F, heated liver homogenate from mice given CdSO4 was reacted with SBD-F in 1 M borate buffer (pH 10.5) containing EDTA at various concentrations. While EDTA did not affect the derivatization of glutathione with SBD-F (data not shown), fluorescence intensity of the peak area corresponding to SBD–MT was increased by derivatization of MT with SBD-F in the presence of EDTA at concentrations in excess of 1 mM in the reaction buffer (350 ml of 1 M borate buffer, pH 10.5) (Fig. 3A). EDTA may be effective for removing coordinated metal ions such as zinc and cadmium from MT. These results indicate that it may be necessary for thiol groups in MT to be in the free form in the derivatization reaction with SBD-F, and addition of a chelating agent is essential for the derivatization of MT. The effects of TBP, a reducing agent, on the reaction of MT with SBD-F were also examined. The peak area of SBD–MT was increased by about seven-fold by the addition of TBP solution (10 ml) at concentrations of 10% or higher to the reaction mixture (Fig. 3B). MT in tissue samples seems to be easily oxidized during sample preparation or derivatization reaction. Figure 4 shows the effects of reaction time on derivatization of MT with BSD-F. The peak area of SBD–MT increased with reaction time up to 20 min and was constant up to 40 min when the
FIG. 7. Apparent MT concentration determined in the liver homogenate containing a specific metal binding form of MT derivatized with SBD-F. Samples were prepared from liver homogenate containing Zn–, Cd–, Cu–, or Hg–MT. Derivatization reaction with SBD-F was performed at 50°C for 30 min in borate buffer containing EDTA (5 mM) (A) or EDTA (5 mM) and BCS (1 mM)(B) (n 5 3).
CHROMATOGRAPHIC DETERMINATION OF METALLOTHIONEIN
173
When the peak height ratio of MT to internal standard was plotted against the concentration of MT, the calibration curve showed good linearity in the range of 70 ng/ml to 50 mg/ml for rabbit MT-1 or MT-2 (Fig. 6). The detection limit of SBD–MT was a few hundred picograms per injection. Recovery of rabbit MT-2 from mouse liver homogenate was satisfactory (more than 93%) in the range of 1 to 30 mg/ml (Table 1). The accuracy of the obtained values clarified by the intraand interassay coefficients of variation was 5.1% (n 5 7) and 3.9% (n 5 3), respectively. Effects of Metal Ion Coordinated in MT on Derivatization with SBD-F FIG. 8. Determination of MT in mouse liver and kidney determined by the HPLC system. Samples were prepared from liver (A) and kidney (B) of mouse treated for 5 days with vehicle (■) (n 5 3), ZnSO4 (u) (n 5 5) or CdSO4 (o) (n 5 5).
supernatant of heat-denatured liver homogenate of mice treated with ZnSO4 was subjected to the reaction. According to these results, it was concluded that MT in biological samples (150 ml) was sufficiently converted into SBD–MT in a mixture of 1 M borate buffer (pH 10.5) containing 5 mM EDTA (350 ml), 20% TBP in isopropanol (10 ml), and 0.5% SBD-F in water (40 ml) for 30 min at 50°C.
Removal of metal ions from the MT molecule by chelating agents is essential for derivatization of MT with SBD-F as described above. The efficiency of removal of metal ions from MT may depend on the affinity of the chelating agent to the metal ion and its concentration. Metal ions such as Zn21, Cu21, Cd21, and Hg21 can bind to MT in vivo, and relative magnitude of affinity of the metals to MT is in the order of Zn21 , Cd21 , Cu21 , Hg21 (13). To examine the effect of EDTA on removal of metals from MT, liver homogenates containing Cd–MT, Cu–MT, or Hg–MT with the same concentration of MT were prepared by replacement of the metal ions from liver homogenate of ZnSO4-treated mice which predominantly contained Zn–MT. These homogenates were employed for deri-
Evaluation of the HPLC System To employ this HPLC method for quantitative determination of MT, N,N9-dipropionylcystamine, the oxidized dimer of N-propionylcystamine (NPC), was newly prepared for an appropriate internal standard. This compound reacts with SBD-F as well as MT when thiol groups are generated by TBP. Using N,N9-dipropionylcystamine as an IS, the tandem column HPLC method established here was evaluated as follows. Figure 5 shows the stability of SBD-MT during storage at 4°C. The peak area ratio of SBD-MT to SBD– NPC(IS) was constant and maintained for at least for 5 days, while the apparent peak areas of both SBD derivatives increased with time in storage, suggesting that SBD-MT is stable chemically and can be stored for several days before analysis. Some isoforms of MT have been identified in mammalian tissues (1). Retention times of SBD derivatives of isoforms of MT (MT-1 and MT-2) from rabbit, rat, and mouse were identical in the tandem column method. Fluorescence intensities of these MT isoforms after derivatization with SBD-F were also almost equivalent (data not shown). These results suggest that total amount of MT can be determined by this tandem column system, and commercially available rabbit MT can be used as a standard sample for determination of MT in several animal species.
FIG. 9. Chromatograms of SBD–MT prepared from mouse L cells cultured with or without ZnCl2. Samples were prepared from mouse L cells cultured with (150 mM) (B) or without ZnCl2 (A). The tandem column system with a Shodex RSpak RP18-413 column and a Puresil C18 column was used. Mobile phase, 20 mM potassium phosphate buffer (pH 7.5):acetonitrile:methanol (80:18:2); flow rate, 0.7 ml/min. MT and IS in the figure represented retention times of SBD–MT and SBD–NPC used as an internal standard (IS).
174
MIYAIRI, SHIBATA, AND NAGANUMA
FIG. 10. Determination of cellular MT levels in HepG2, HeLa, and mouse L cells. Samples were prepared from HepG2 cells (A), HeLa cells (B), or mouse L cells (C) after 24-h incubation with various concentrations of ZnCl2.
vatization with SBD-F in the presence of EDTA (5 mM) and for determination of MT by the present HPLC procedure. After heating these homogenate, excess amounts of added metal ions (Cd21, Cu21, or Hg21) were precipitated with denatured proteins and most of each metal ion in the supernatant was bound to MT. The concentration of MT determined by the HPLC method in the heated supernatant containing Zn–MT was comparable with that in the heated-supernatant containing Cd–MT. However, the apparent concentrations of MT in the heated supernatant containing Cu–MT or Hg–MT were considerably lower than those in Zn–MT or Cd–MT (Fig. 7A). When the derivatization reaction was carried out in borate buffer containing 1 mM bathophenanthroline disulfonate (BCS) in addition to 5 mM EDTA, the MT concentration determined by the HPLC method in the heated supernatant containing Cu–MT was almost identical to that containing Zn–MT (Fig. 7B). This indicated that utilization of BCS in combination with EDTA as a chelating agent is effective for the derivatization of Cu–MT with SBD-F. Long–Evans Cinnamon rats spontaneously develop hepatoma and accumulate copper ions in the liver as a complex with MT (14). Animal models of Menkes disease (15) also accumulate excessive concentrations of copper in their kidneys. Derivatization of MT with SBD-F in the presence of BCS and EDTA might be useful for the determination of Cu–MT in tissues or cultured cells from these animals using the present HPLC method.
Determination of MT in Mouse Tissues and Cultured Cells Mice (n 5 5) were intraperitoneally administered ZnSO4 (200 mmol/kg of weight), CdSO4 (5 mmol/kg of weight), or vehicle (saline) (0.1 ml) once a day for 5 days. The liver and kidneys collected 24 h after the last injection were subjected to HPLC analysis for the determination of MT concentration. The basal MT levels
of liver and kidney were 15 and 38 mg/g of wet tissue, respectively. The MT levels of these tissues were elevated to 655 and 261 mg/g by treatment with ZnSO4, and 1163 and 298 mg/g by treatment with CdSO4, respectively (Fig. 8). MT concentrations in HeLa, HepG2, and mouse L cells cultured with or without ZnCl2 were also determined by the present HPLC method. Reliable chromatograms were obtained for the basal and induced levels of MT in L cells (Fig. 9) as well as the other two cell lines. The basal levels of MT in HepG2, HeLa, and L cells were 4.7, 2.3, and 0.3 mg/mg of protein, respectively. These MT levels were increased dose-dependently by treatment with ZnCl2, and the highest concentrations of MT observed were 132 mg/mg of protein in HepG2 cells (at 150 mM of ZnCl2), 22 mg/mg in HeLa cells (at 100 mM of ZnCl2), and 10 mg/mg in L cells (at 150 mM of ZnCl2) (Fig. 10). In the present HPLC method, aliquots of only about 1 3 104 cells were needed for determination of concentration of MT. In conclusion, a sensitive, reliable, and convenient method for the determination of MT using a tandem column HPLC system with isocratic elution was established in the present study. Pretreatment of samples and the derivatization reaction in this method are quite simple, and the fluorescence intensity of the resultant SBD-derivatized MT did not decrease as a function of time over the week. Although Hg–MT was not detectable by the present method, only a few studies have been performed on Hg–MT. The tandem column HPLC method presented here might be useful for the determination of most major MTs such as Zn–MT, Cd– MT, and Cu-MT. ACKNOWLEDGMENTS This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan and a grant for drug development from the Japan Health Science Foundation.
CHROMATOGRAPHIC DETERMINATION OF METALLOTHIONEIN
REFERENCES 1. Ka¨gi, J. H. R. (1993) in Metallothionein III (Suzuki, K. T., Imura, N., and Kimura, M., Eds.), pp. 29 –55, Birkha¨user Verlag, Basel, Switzerland. 2. Uchida, Y., Takio, K., Titani, K., Ihara, Y., and Tomonaga, M. (1991) Neuron 7, 337–347. 3. Vitarella, D., Conklin, D. R., Kimelberg, H. K., and Aschner, M. (1996) Brain Res. 738, 213–221. 4. Ka¨gi, J. H. R. (1991) Methods Enzymol. 205, 613– 626. 5. Patrierno, S. R., Dellis, N. R., Evans, R. M., and Costa, M. (1983) Life Sci. 32, 1629 –1636. 6. Onosaka, S., and Cherian, M. G. (1981) Toxicology 22, 91–101. 7. Lehman, L. D., and Klaassen, C. D. (1986) Anal. Biochem. 153, 305–314.
175
8. Suzuki, K. T. (1991) Methods Enzymol. 205, 198 –205. 9. Sunaga, H., Kobayashi, E., Shimojo, N., and Suzuki, K. T. (1987) Anal. Biochem. 160, 160 –168. 10. Bu¨hler, R., and Ka¨gi, J. H. R. (1974) FEBS Lett. 39, 229 – 234. 11. Bradford, M. M. (1976) Anal. Biochem. 72, 248 –254. 12. Imai, K., Toyo’oka, T., and Watanabe, Y. (1983) Anal. Biochem. 128, 471– 473. 13. Webb, M. (1979) in The Chemistry, Biochemistry and Biology of Cadmium (Webb, M., Ed.), pp. 195–266, Elsevier/North Holland, Amsterdam. 14. Sugawara, N., Sugawara, C., Sato, M., and Mori, M. (1993) Trace Elem. Exp. Med. 6, 15–21. 15. Hunt, D. M. (1974) Nature 249, 852– 854.