SUMMARY. A high activity glutathione S-transferase TI-1 (GSTTI-1) towards dichloromethane was isolated from human liver cytosol and purified to homogenity ...
Vol. 39, No. 1, May 1996
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL Pages 21-29
PURIFICATION, CHARACTERIZATION AND TISSUE DISTRIBUTION OF HUMAN CLASS THETA GLUTATHIONE S-TRANSFERASE TI-I. Erkki Juronen ~, Gunnar Tasa, Mart Uuskiila, Margus Pooga ~* and Aavo-Valdur Mikelsaar University of Tartu, Institute of General and Molecular Pathology, Department of Human Biology and Genetics, Veski str. 34, EE2400 Tartu, ESTONIA and **University of Tartu, Institute of Molecular and Cellular Biology, Riia str. 23, EE2400 Tartu, ESTONIA
Received December 17, 1995 Received after revision February9, 1996 SUMMARY
A high activity glutathione S-transferase TI-1 (GSTTI-1) towards dichloromethane was isolated from human liver cytosol and purified to homogenity in 18.5% yield with a purification .factor of 4400-fold. The GSTTI-1 was also isolated from erythrocytes, but the enzyme activity decreased rapidly in the final stages of purification. The purified GSTT 1-1-s were homo-dimeric enzymes with a subunit M, value 25,300 and pl 6.64, as confirmed by SDS-PAGE, IEF and Western blot analysis. The N-terminal amino acid sequences ofGSTTI-1 from liver and red blood cells, analyzed up to the 12th amino acid, were identical. Immunoblot analysis revealed that GSTT1-1 was also present in lung, kidney, brain, skeletal muscle, heart, small intestine and spleen, but not in lymphocytes. Key words: Glutathione S-transferase T 1-1; purification; tissue distribution
INTRODUCTION The glutathione S-transferases (GSTs; EC 2.5.1.18) are a group of dimeric detoxification enzymes catalysing the conjugation of GSH with a broad spectrum of electrophiles (1). The subunits of the GST family are divided into four multigene classes, Alpha, Mu, Pi and Theta according to the basis of sequence identity (2,3). One polymorphic gene locus of the glutathione S-transferase Mu family (GSTM1) has attracted interest because it is present in approximately 60% of the individuals investigated (4,5) and its product (GSTMI-1) can detoxify potential man-made carcinogens (6). The absence of this enzyme seems to be a possible contributor to an increased risk of developing certain types of malignancies caused by cigarette smoking or exposure to toxic chemicals (7,8). Differently from the other multigene families of GST, relatively little is known about the human class Theta enzymes. This GST family is responsible for the GSH-dependent detoxification of monohalomethanes in human erythrocytes and liver (9-11). Approximately 60-70% of the human population are able to carry out this metabolic reaction ("conjugators"), whereas the remainder are 1039-9712/96/010021-09505.00/0
21
Copyright © 1996 by Academic Press Australia. All rights of reproduction in any form reserved.
Vol. 39, No. 1, 1996
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
unable to perform the conjugation ("non-conjugators") (9). Further characterization of these phenotypes revealed that GSH conjugation of the industrial chemicals dichloromethane (DCM) and ethylene oxide (EO), could only be catalysed by blood samples from the "conjugator" population (12,13). The monohalomethanes, EO, DCM and other man-made alkylhalides are widely used industrial methylating agents: fumigants, pesticides and solvents, and any polymorphic locus in humans that may be involved in their metabolism would be of epidemiological interest. It is not yet established how many class Theta GSTs exist in the human. Up to now, two Theta-class transferases, namely GSTT 1-1 and GSTT2-2 have been isolated from human liver (3,14). A cDNA encoding GSTTI-1 has been isolated, and the GSTT1 gene was shown to be absent in 40% of the population (11). It was concluded that GSTT1 must encode the liver-derived enzyme GST Theta (GSTTI-1) as the deduced amino acid sequences were identical with the N-terminal sequence of human liver GSTTI-1, reported by Meyer et al. (3). In addition, by correlating the GSTT1 genotype with DCM-conjugating activity present in the erythrocytes, it was deduced that it was the same locus that was also expressed in red blood cells. Until recently the properties and tissue distribution of GSTT 1-1 were poorly described, and the identity of GSTTI-1 from liver and erythrocytes only just supposed, and GSTTI-1 itself had not been isolated in its fully active form. In the present study we succeeded in isolating of active GSTTI-1 from liver and erythrocytes, describing its properties and establishing tissue distribution by Western blot analysis with monoclonal antibodies.
MATERIALS AND METYIODS
Tissues Human tissue samples were obtained less than 8 h post mortem and stored at -80°C until use. Macroscopic examination was performed at autopsy to establish that there were no abnormalities in the samples collected. Erythrocytes from blood donors were prepared as described earlier (15). Enzyme assay The GSTTI-1 activity determination towards DCM ~was performed essentially as described earlier (16). Protein determination Protein concentrations were determined by the Bradford technique, with mouse IgG as a standard (17). Protein sequencing Automated amino acid sequencing of purified GSTT 1-1 was performed on an Applied Biosystems model 477A pulsed-liquid protein sequencer equipped with model 120A PTH analyzer using standard programs as described elsewhere (I 8).
22
Vol. 39, No. 1, 1996
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
Electrophoresis and isoelectric focusing 15% SDS-PAGE was performed using the discontinuous buffer system of Laemmli (1.9) and 8-25% gradient SDS-PAGE as described previously (20). Analytical isoelectric focusing (tEF) was carried out using an LKB 2117 Multiphore II equipment in 2 mm 7.5% (w/v) polyacrylamide gel (pH range 3.5-10) as described by the manufacturer (Pharmacia LKB Biotechnology AB, Bromma, Sweden). The protein MW standards were obtained from Sigma, St.Louis, MO and the protein pI standards used to calibrate the IEF gel came from Pharmacia, Sweden. Coomassie Blue and silver-stained gels were scanned by Model SL-2D/1D UV/VIS densitometer (Biomed Instruments, Inc., Fullerton, CA) and the data analyzed using the manufacturer software.
Immnnoblotting Tissue cytosols were run in 8-25% SDS-PAGE gradient gel and transferred onto a nitrocellulose filter (45 gm, Schleicher and Schall, Germany) using a semidry LKB Novablot electrophoretic transfer apparatus (21). The nonspecific binding sites were blocked with 0.15 M phosphate buffered saline, pH 7.4, containing 0.05% (v/v) Tween-20 (PBS-Tween) and 0.05% (w/v) casein for 30 min. The blots were then incubated overnight with anti-GSTTl-1 monoclonal antibody IA2 hybridoma supernatant (22) at a 1:5 dilution in PBS-Tween, followed by a 90 min incubation with peroxidase conjugated goat anti-mouse IgG antibody (LabAs Ltd., Tartu, Estonia) at a 1:1000 dilution. Each incubation step was followed by washing 4x5 min with PBS-Tween. Staining was performed using mixed 4-chloro-l-naphtol and 3,3'-diaminobenzidine (Sigma, St.Louis, MO) chromogenic substrate solution in PBS until bands were detected (23). Purification of GSTTI-1 All steps of purification were performed at 4°C. A section of human liver (approx. 500 g) was homogenized in 2 vol. (v/w) of ice-cold 20 mM-sodium phosphate buffer, pH 6.4, containing 2 mM EDT& 2 mM 2-mercaptoethanol and 25 p.M phenylmethanesulphonyl fluoride (PMSF). The soluble supernatant obtained by centrifugation at 30 000 g for 40 min was immediately mixed with an equal volume of CM-Sephadex C-50 (Pharmacia, Uppsala, Sweden), preswollen in the homogenization buffer. The lysate and ionexchanger were mixed one hour on ice with gentle stirring and unbound proteins were recovered by filtration. The pH of lysate was adjusted to 7.2 with 2 M NaOH and applied to a column (2.5 cm x 50 cm) of DEAE-Sepharose Fast Flow (Pharmacia, Uppsala, Sweden) equilibrated with 20 mM sodium phosphate buffer, pH 7.2, containing 2 mM EDTA, 2 mM 2mercaptoethanol and 25 gM PMSF. The column was run with the same buffer at a flow rate of 180 ml/h. Unbound fi'action, having the activity towards DCM was collected and ammonium sulfate added to final concentration of 20% (w/v). The sample was clarified by centrifugation and applied (in 4 portions) at ambient temperature to a Pharmacia HR 10/10 column, containing thiophilic sorbent (Tgel). The resin was prepared as described previously (24) and equilibrated with 100 mM Tris/HC1 buffer, pH 8.0, containing 2 mM 2-mercaptoethanol and 20% (w/v) of ammonium sulfate. The column was developed with a decreasing step-wise ammonium sulfate gradient in equalibration buffer, formed in two stages consisting of 15% (w/v) ammonium sulfate over 15 min and followed by 10% (w/v) ammonium sulfate over 10 min at a flow rate of 2 ml/min. Peaks with activity towards DCM, eluated at 10% (w/v) ammonium sulfate, were pooled and dialysed against three changes, 4 litres of each, of 10 mM-sodium phosphate buffer, pH 6.9, containing 2 mM 2-mercaptoethanol and 25 ~M PMSF for 48 h at 4°C. The dialysate was subjected in three cycles to dye-ligand chromatography in the same buffer on Matrex Gel Orange A (Amicon, Lexington, MS, U.S.A.) in Pharmacia HR 16/10 column. Peaks with DCM activity, eluated at 90 mM KC1 in the running buffer, were pooled and dialysed extensively for 4 h against 25 mM ethanolamine/acetic acid buffer, pH 9.6, containing 2 mM 2-mercaptoethanol. The dialysate was loaded to a PBE94 (Pharmacia, Uppsala, Sweden) chromatofocusing column and the pH gradient was developed with Polybuffer 96/acetic acid, pH 6.0 at a flow rate of 15 ml/h. Active fractions were pooled, added to final concentration 20% (w/v) of
23
Vol. 39, No. 1, 1996
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
ammonium sulfate and concentrated on the 2 ml of T-gel. The concentrated sample was dialysed extensively for 4 h against 20 mM ethanolamine/acetic acid buffer, pH 9.8, containing 2 mM 2-mercaptoethanol and loaded onto a Q-Sepharose Fast Flow HR 10/10 (Pharmacia, Uppsala, Sweden) column, pre-equilibrated with dialysing buffer. The column was developed with a linear gradient of 0-400 mM NaC1 over 60 min in starting buffer at a flow rate of 2 ml/min. Peak with activity towards DCM was concentrated on the T-gel and stored at -40°C in small portions. The GSTTI-1 from erythrocytes was purified in the same way, with the only exception that 10 ml ofpreswollen CM-Sephadex C-50 was used per ml of lysed erythrocytes. RESULTS AND DISCUSSION
Purification of GSTTI-1 from liver and erythrocytes
To date, the existence of a protocol that allows the purification of active GSTTI-1 has not been documented. Meyer et al. (1991) have isolated a human hepatic Theta-class transferase GST O (3). Unfortunately, the obtained enzyme lose activity during the purification process. The present investigation describe the new protocol for GSTT1-1 purification. Six liquid chromatography steps were used to isolate a GSTTI-1 enzyme with a high activity towards DCM from human liver cytosol. The GSTT 1-1 from liver and red blood cells behaved .identically by the purification process, but the activity of the enzyme from erythrocytes decreased rapidly in the final stages of purification. Like other rat and human class Theta GSTs (3,14), GSTTI-1 lost much of its specific activity towards DCM in soluble supematants with storage at 4°C. To prevent this dramatic loss of activity, samples were mixed immediately, without dialysis, with preswollen CM-Sephadex C-50 and loaded without delay onto a DEAE-Sepharose column. Neither CM-Sephadex nor DEAE-Sepharose could bind the GST activity towards DCM Differently from the GST purification schemes published earlier (3,14), we dropped the hydroxyapatite chromatography step, and the supernatant was rapidly concentrated and partially purified using salt-promoted thiophilic adsorption on the T-gel. In this step material for a large-scale preparation of GSTTI-1 with little loss &enzyme activity could be obtained by freezing peaks with activity towards DCM. The fourth GSTTI-1 purification step was chromatography on Matrex Gel Orange A. The Orange A chromatography step revealed a major peak of activity towards DCM eluated approximately with 0.1 M KCI, as described elsewhere (3). The sample with DCMGSH conjugating activity was subjected to final isolation steps on chromatofocusing column, followed by anion-exchange chromatography on a Q-Sepharose column. The highest activity peaks towards DCM were found in the PBE94 column at pH 7.3 and in the Q-Sepharose column in the single peak between 200 and 220 mM NaCI on the salt gradient.
24
Vol. 39, No. 1, 1996
BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL
Characterization of human GSTTI-I
By the purification scheme, described above, the hepatic GSTTI-1 was recovered in 18.5% yield and with a purification fold of 4400 times (Table 1). Our data indicate that GSTTI-1 represents approximately 0.02% of the soluble liver protein, the value being much higher than it was suggested earlier (0.003%) (3). The activi.ty ofGSTTI-1, purified from erythrocytes decreased rapidly during the last stages of the purification making the calculation of the yield in terms of activity impossible. The isolated enzyme migrated as a single band in SDS-PAGE and was judged to be at least 99% pure (Fig. 1). SDS-PAGE analysis of the hepatic GSTT 1-1 preparations indicated that the Mr value of the subunit was 25,300, estimated with a comparative electrophoretic study on the same gel plate with Mr marker proteins as well as the GSTMla/GSTMlb subunit protein, isolated from human liver cytosol (25). SDS-PAGE analysis of GSTT1-1 isolated from erythrocytes and Western blot analysis of red blood cells cytosols, presented an Mr value for GSTT1-1 in erythrocytes that was essentially the same as the M~value for hepatic GSTTI-1 (Fig. 3)~ As shown for the GSTT2-2 (26), the SDS-PAGE estimated GSTT 1-1 subunit molecular mass was slightly lower than the calculated molecular mass (27,200) (11). The Mr values described for GSTTI-1 subunits were quite similar to those of GSTT2-2 (Mr 25, I00) and GST 7~(Mr 24,800) but their electrophoretic mobility was slightly higher than those for GST c~(Mr 26,000), GSTMla (Mr 26,700) and GSTMlb (Mr 26,600) (4,14).
Table 1. Purification of human GSTTI-I from liver
For experimental details see the text. GSTTI-1 activity was measured towards DCM at 37°C and protein concentration by the method of Bradford (16). Abbreviation: ND, not detected values. Fraction
Total protein (mg)
Total activity (nM/min)
Specificactivity (nM/min/mg)
Recovery (%)
Liver cytosol
23060
8510
0.37
100
DEAE-Sepharose
3760
8400
2.23
98
T-gel
605
4801
7.94
56
Orange A
84
2192
26.09
25
Chromatofocusing
7.5
ND
ND
ND
Q-Sepharose
0.98
1577
1609
18.5
25
Vol. 39, No. 1, 1 9 9 6
BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL
kD - 36,0 - 29,0
wt
- 24,0 - 20,0
!!!i:! i!!i!iii!!i:iiii - 14,2 1
2
Figure 1. S D S - P A G E of the purified G S T M I - 1 and G S T T I - 1 from liver, silver-stained. The
human liver GSTTI-1 was purified and analysed by 15% SDS-PAGE as described in the text. Purified human GSTMI-1 was included in the analysis to provide protein standards Lane 1: GSTMI-1 (Mr 26,700), lane 2: purified GSTTI-1.
Isoelectric focusing of the purified GSTTI-1 and liver cytosols revealed the presence of the GSTTI-1 band at pI 6.64 (Fig2). The small ladder of more alkaline bands is the result of freezingthawing of purified enzyme preparation. The GSTT 1-1 from liver and erythrocyte cytosols had the same isoelectric point as shown by isoelectric focusing, followed by immunoblotting Unlike the GSTMI-1, where GSTMla-la, GSTMla-lb and GSTMlb-lb isoenzymes could be detected in IEF (27), no differences in pI values were observed in 10 different immunoblotted liver and red blood cell cytosols, indicating only one isoform of GSTT 1-1. GSTTI-1 from liver and erythrocytes had a non-blocked N-terminus and identical N-terminal amino acid sequences, determined with an automatic amino acid sequencer by Edman degradation method. The sequence of Gly-Leu-Glu-Leu-Tyr-Leu-Asp-Leu-Leu-Ser-Gln-Proof the first 12 amino acids corresponds to the sequence of N-tern~ni in the hepatic GSTTI-1 described in literature (3). All the data obtained - similar behaviour during purification, the same subunit molecular weight, isoelectric point, N-terminal amino acid sequence and similar immunochemical properties (see below) show that the polymorphic class Theta GSTTI-1 in liver and erythrocytes is one and the same enzyme, encoded by one and the same locus - GSTT1.
Tissue distribution of GSTTI-1
Immunoblot analysis of various tissue cytosols, carried out using 8-25% gradient SDS-PAGE and anti-GSTTl-1 monoclonal antibody 1A2, suggested that all tissues investigated, expressed a detectable amount of GSTT 1-1 (Fig. 3). GSTT 1-1 was not detected in the immunoblotted lysate of
26
Vol. 39,
No.
1, 1996
BIOCHEMISTRY and M O L E C U L A R
A
B
BIOLOGY
INTERNATIONAL
pl -7.35 -6.85 - 6.55
-5.85
- 5.20
1
2
3
4
5
6
7
Figure 2. Isoelectric focusing of human hepatic GSTTI-1. A. IEF of purified enzymes was
performed in a broad range gel, pH 35-10, in 7.5% (w/v) polyacrylamide and silver-stained. Lane 1: purified GSTTI-1; lane 2: protein pI calibration standards; lane 3: GSTMla-la isoenzyme; lane 4: GSTMlb-lb isoenzyme. B. Immunoblotting of liver cytosols after IEF. Lanes 5 and 7: cytosols with activity towards DCM; lane 6: cytosol without activity towards DCM. kD
- 66,0 - 45,0 - 36,0 - 29,0 ~
m. .
.
.
.
24,0 "- 20,0 - 14,2
1 2 3 4 5 6 7 8 9 1 0
Figure 3. Immunoblot analysis of GSTTI-I in the cytosols of various human tissues. Protein
samples (300 pg per lane) were resolved by 8-25% gradient SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was sequentially incubated with 0.05% casein, antiGSTT1-1 monoclonal antibody 1A2 supernatant, goat anti-mouse IgG pero~idase conjugate and substrate solution as described in the text. Lane 1: cytosolic proteins of liver; lane 2: cytosolic proteins of erythrocytes; lane 3: cytosolic proteins of lymphocytes; lane 4: cytosolic proteins of lung; lane 5: cytosolic proteins &kidney; lane 6: cytosolic proteins of brain; lane 7: cytosolic proteins of skeletal muscle; lane 8: cytosolic proteins of heart; lane 9: cytosolic proteins of small intestine; lane 10: cytosolic proteins of spleen.
27
Vol. 39, No. 1, 1996
BIOCHEMISTRYand MOLECULAR BIOLOGY INTERNATIONAL
isolated human lymphocytes (data not shown). Lysates from erythrocytes and spleen proved to contain lower concentrations of these enzyme than the other lysates. Immunohlot analysis of various tissue cytosols suggested that like to other polymorphic glutathione S-transferase GSTM 1-1, class Theta GSTTI-1 is widely expressed in different human tissues. GSTMI-1, however, is not present in erythrocytecells but existed in lymphocytes (28). The importance of polymorphic GSTs in conferring resistance to chemicals and chemotherapeutic drugs has resulted in these enzymes attracting significant attention. To date, the contribution of class Theta GSTs to chemicalsand drugs has not been investigated. Such studies will be assisted by knowledge of the enzymaticand other properties of class Theta GSTs. The precise role of this family of GSTs in protection against chemical insult awaits futher study.
ACKNOWLEDGMENTS
This work was supported by a grant (No. 1358) from Estonian Science Foundation.
REFERENCES
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)
Ketterer, B. (1988) Mutat. Res. 202, 343-361. Mannervik,B., Alin, P., Guthenberg, C., Jensson, H., Tahir, M., Warholm, M. and Jornwall, H (1985) Proc. Natl. Acad. Sci. U.S.A. 82, 7202-7206. Meyer, D.J., Coles, B., Pemble, S.E, Gilmore, K., Fraser, G.M. and Ketterer, B. (1991) Biochem. J. 274, 409-4t4. Hussey,AJ., Stockman, P.K., Beckett, GJ. and Hayes, JD. (1986) Biochim. Biophys. Acta 874, 1-12. Seidegard,J.,Guthenberg, C., Pero, R.W. and Mannervik, B. (1987) Biochem. J. 246, 783785. Warholm, M., Guthenberg, C. and Mannervik, B. (1983) Biochemistry 22, 3610-3617. Seidegard, J., Pero, R.W. and Miller, DG. (1986) Carcinogenesis 7, 751-753. Daly,A.K.~Thomas, D.J., Cooper, J., Pearson, W.R., Neal, D.E and Idle, J.R. (1993) Brit. Med. J. 307, 481-482. Peter, H., Deutschmann, S., Reichel, C. and Hallier, E. (1989) Arch. Toxicol. 63,351-355. Hallier,E., Langhof, T., Dannapel, D., Leutbecher, H., Schroeder, K.R., Goergens, H.W., Muller, A. and Bolt, H.M. (1993) Arch. Toxicol. 67, 173-178. Pemble, S., Schroeder, K.R, Spencer, S.R., Meyer, D.J., Hallier, E., Bolt, H.M., Ketterer, B. and Taylor, J.B. (1994) Biochem. J. 300, 271-276. Thier, R., Foest, U., Deutschmann, S., Schroeder, KR., Westphal, G., Hallier, E. and Peter, H. (1991) Arch. Toxicol. Suppl. 14, 254-258. Foest,U., Hallier, E, Ottenwalder, H., Bolt, HM. and Peter, H. (1991) Hum. Exp. Toxicol. 10, 25-31. Hussey, A.J. and Hayes, J.D. (1992) Biochem. J. 286, 929-935. Norppa,H., Hirvonen~A., J~ventaus, H, Uusk~la, M., Tasa, G., Ojaj~rvi, A. and Sorsa, M. (1995) Carcinogenesis 16, 1261-1264.
28
Vol. 39, No. 1, 1996
BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL
(16)
Bogaards, J.J.P., van Ommen, B. and van Bladeren, P.J. (1993) Biochem. Pharmacol. 45, 2166-2169. (17) Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. (18) Matsudaira, P.J. (1987) Biol. Chem. 262, 10035-10038. (19) Laemmli, U.K. (1970) Nature 277, 680-685. (20) Fling,S.P. and Gregerson, D.S. (1986) Anal. Biochem. 155, 83-88. (21) Kyhse-Andersen, J. (1984) J. Biochem. Biophys. Methods 10, 203-209. (22) Juronen,E., Tasa, G., Uuskgla, M., Pooga, M. and Mikelsaar, A.-V. (1996) Hybridoma 15, 77-82. (23) Young, P.R. (1989) J. Immunol. Methods 121,295-296. (24) Juronen, E., Parik, J. and Toomik, P. (1991) J. Immunol. Methods 136, 103-109. (25) Juronen, E., Tasa, G., Uusk~la, M., Parik, J. and Mikelsaar, A.-V. (1994) Hybridoma 13, 477-484. (26) Tan, K.L., Webb, G.C., Baker, R.T. and Board, P.G. (1995) Genomics 25, 381-387. (27) Harada,S., Abei, M., Tanaka, N., Agarwal, D.P. and Goedde, H.W. (1987) Hum Genet 75, 322-325. (28) Laisney,V., Nguyen Van Cong, Gross, M.S. and Frezal, J. (1984) Hum Genet 68, 221-227.
29
