XENOBIOTICA,
1991, VOL. 21,
NO.
2, 171-177
Human liver sulphotransferase and UDP-glucuronosyltransferase: structure-activity relationship for phenolic substrates A. T E M E L L I N I t , M. FRANCHIt, L. GIULIANIS and G. M. PACIFICItQ
t Department of Experimental Biomedicine, Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
1Department of Surgery, Medical School, University of Pisa, 56100 Pisa, Italy
Received 14 February 1990; accepted 10 August 1990
1. Human liver sulphotransferase and UDP-glucuronosyltransferase were studied with phenol, methyl-, ethyl-, propyl-, butyl-, phenyl-, nitro-, amino-phenols and hydroxybenzoic acids as substrates. 2. T h e Michaelis-Menten constants (K,) and the maximum velocities of reaction (V,,,,,) of sulphotransferase and UDP-glucuronosyltransferase for each substrate were measured. 3. T h e K , values for sulphotransferase varied over 5000-fold whereas they varied over 25-fold for UDP-glucuronosyltransferase. 4. Sulphotransferase and UDP-glucuronosyltransferase have different structureactivity relationships with phenolic substrates.
Introduction UDP-glucuronosyltransferase and sulphotransferase are two important conjugation enzymes, each consisting of different forms (Boutin 1987, Burchell and Coughtrie 1989, Jakoby et al. 1984). Overlapping substrate specificities of UDPglucuronosyltransferase and sulphotransferase have been described for a limited number of substrates (Mulder and Meerman 1978, Mulder 1982, Pacifici et al. 1990), but it is unknown whether such overlapping substrate specificities may be extended to a broad spectrum of substrates. A multitude of drugs and environmental pollutants have a phenolic structure. This prompted us to investigate the structure-activity relationship of sulphotransferase and glucuronyltransferase with phenolic molecules. T h e majority of the compounds studied were found to be substrates for both UDPglucuronosyltransferase and sulphotransferase, but some served as substrate for only one of the two enzymes. We also observed that the Vmaxto K, ratio, which gives an indication of enzyme efficiency, has a more marked substrate dependence for sulphotransferase than for UDP-glucuronosyltransferase. Thus, these two enzymes may contribute to the conjugation of the same substrate, but their contribution is substrate-dependent.
Materials and methods Chemicals Labelled adenosine 3‘-pho~phate-5’-phospho[~~S]sulphate (PAPS) (specific activity 2.5 Ci/mmol) and I4C-uridine 5’-diphosphoglucuronic acid (UDPGA) (specific activity 304 mCi/mmol) were obtained from New England Nuclear Co. (Florence, Italy). Unlabelled UDPGA, bovine serum albumin (BSA),
5 Address correspondence to: Gian Maria Pacifici, Associate Professor in Pharmacology, Department of Experimental Biomedicine, Medical School, Via Roma 55, 56100 Pisa, Italy. 0049-8254/91 $3.00 0 1991 ‘Taylor & Francis Ltd
172
A. Temellini et al.
DI,-dithiothreitol ( D T T ) and Tr is (tris-(hydroxymethy1)aminomethane) were obtained from Sigma Chemical Co. (St Louis MO, USA). Sulphate or glucuronate acceptor substrates were from Aldrich (Milan, Italy) and their chemical purity ranged between 98% and 99% for all the substrates except rn- and p-phenylphenol, whose purity was 90% and 97%, respectively.
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
Biological material Wedge human liver specimens were obtained at laparotomy from two males 50 and 60 years old, and 11 females (age between 23 and 65 years) undergoing cholecystectomy. Liver specimens with normal cell architecture only were used. Biopsies were homogenized in 5 vol. 0.25 M sucrose by means of a glassTeflon homogenizer. T h e homogenates were centrifuged at 12 OOOg for 15 min and the supernatants were centrifuged again at 105 OOOg for 1 h. T h e ensuing supernatants were used as the cytosolic fractions, and the pellets were resuspended in 0.1 M Tril-HCI (pH7.4) containing 30% glycerol and studied as the microsomal fractions. Assays Sulphotransjerase. T h e incubation mixture for sulphotransferase was as previously described by Anderson and Weinshilboum (1980) and modified by Campbell et al. (1987a). Briefly, the incubation consisted of lOmM potassium pyrophosphate buffer ( p H 6.5),0 . 4 35S-PAPS ~ ~ mixture (final vol. 150~1) (about 400 000 dpm), the sulphate acceptor substrate, dissolved in dimethyl sulphoxide, 1 4 m D ~TT, 0.0645% BSA and loop1 of diluted cytosolic fraction to give a final protein concentration of 1 pg/ml. Liver cytosols were diluted with 5 mM potassium pyrophosphate buffer ( p H 7.4) containing 0,625 mg/ml albumin and 1.54 mg/ml DTT to give a final concentration of 1.5 pg/ml, divided into 1 ml aliquots and stored at -80°C. T h e reaction was started by addition of PA PS and carried out at 37°C for 30min and stopped by addition of a mixture of zinc sulphate and barium acetate as described elsewhere (Campbell e t al. 1987a). After centrifugation at 1500g. aliquots ( 4 0 0 ~ 1of ) the clear supernatant were transferred to scintillation vials containing 3 ml of scintillation fluid (Scintillator 299TM,from Packard, Milan, Italy). l' he linearity for protein concentration and incubation time were determined for phenol and each of p-substituted phenols, and occasionally for other substrates. Reaction was linear at least u p to 5pg/ml cytosolic protein and 40 min of incubation with all substrates studied. Final protein concentration and incubation time were fixed to 1 pg/ml and 30min, respectively, and such conditions were kept constant throughout the study. Each assay had two blank samples which were the same as the active samples except that the substrate was replaced by dimethyl sulphoxide. Four liver samples, whose donors were one man 60 years old and three women 38.50 and 5 1 years old, were used throughout the study. T h e appropriate range of substrate concentration was first studied bylOfold substrate dilutions from 1 mM to 0.01 p ~ T.h e enzyme kinetics were then studied with six concentrations of substrate, each being double the next, and assayed in duplicate. T h e enzyme kinetic constants, the Michaelis-Menten constant (K,) and the maximum reaction velocity (V,,,) were determined by Eadie-Hofstee plots using a computer program and the given data are the averages of four livers. Each curve was fitted by a straight line whose correlation was greater than 0.98. C;lururonosy[transferase. T h e incubation mixture for UDP-glucuronosyltransferase was as previously described by Bansal and Gessner (1980). Incubations were carried out at 37°C for 1 0 mi n in a final volume o f loop1 containing 0.1 M 'I'ris-maleate buffer (pH7.4), 5 mM MgCI,, 2 mM 14C-UDPGA (SOOOOdpm), the appropriate concentration of the substrate which was added in 2p1 of dimethyl sulphoxide, and an aliquot of the microsomal protein to give a final protein concentration of 2.5 mg/ml. No activator of U D P ~lucuronosyltransferasewas added to the incubation mixture. Reactions were started by addition of ) the incubates were applied to U D P G A and stopped by addition of 200pl of ethanol. Aliquots ( 4 0 ~ 1 of t.1.c. plates L K 5 D F (Whatman, Milan, Italy) and chromatographed in n-butanol-acetone-acetic acid30% ammonia-water (70: 50: 18 : 1.5 : 60 by vol.). T.1.c. plates were visualized under U.V.light and the spots corresponding to the glucuronides were removed and placed in scintillation vials containing 10 ml of scintillant fluid (Scintillator 299TM,Packard, Milan, Italy). Each sample was assayed in duplicate and for each incuhation two blanks were used, having the same composition as the active samples except that the substrate was replaced by 2 p l of dimethyl sulphoxide. Labelled U D P G A was originally provided in ethanol-water (7 : 3 v / v ) and to keep the final concentration of ethanol in incubation mixture below 5%, as it inactivates UDP-glucuronosyltransferase, the ethanolic solution of U D P G A was evaporated at room temperature under a flow of N,, and the original volume reconstituted with water. UDI'GA solutions were divided into aliquots and stored at -20°C for not longer than 2 weeks. Optimal incubation time and protein concentration were determined with p-substituted phenols and occasionally with other substrates, and were found to be 10min and 2.5 mg/ml protein, respectively. Because of the relatively high amount of microsornal protein required for each assay 11 liver biopsies were used in the study. T h e liver donors were nine women and two men, aged between 23 and 65 years. for each substrate were measured in three livers by construction of T h e kinetic parameters K,,, and VmaK the Eadie-Hofstee plots. T h e enzyme obeyed Michaelis--Menten kinetics with each substrate studied, and the regression coefficient of each plot was greater than 0.97.
Human conjugases
173
The activities of sulphotransferaseand UDP-glucuronosyltransferasewere measured on the basis of the specific radioactivity of PAPS or UDPGA after correction for blanks. The radioactivity in active samples was at least twice as high as in the blank samples. Protein concentration was measured as described by Lowry et al. (1951).
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
Results Table 1 summarizes the kinetic parameters of the human hepatic sulphotransferase obtained with different phenolic substrates. Both K , and V,,, are dependent on (i) the chemical nature of the substituent and also on (ii) its position in the phenol molecule. The highest V,,, and the lowest K , were obtained withp-nitrophenol and m-phenylphenol, respectively. T h e lowest V,,, and the highest K, were obtained with para-aminophenol and ortho-tert.-butyl phenyl, respectively. T h e carboxyl group placed at any position of the phenol molecule inhibits sulphation. K , varied between 0.06 and 344pM, i.e. over a 5000-fold range. V,,, varied between 0 0 3 and 0.66 nmol/minmg-', i.e. over a 20-fold range. Table 2 gives the kinetic parameters of the human hepatic UDPglucuronosyltransferase obtained with different substrates. Both V,,, and K , vary with the chemical nature of the substrate. T h e highest V,,, and the lowest K , were obtained with p-nitrophenol and p-tert.-butylphenol, respectively. T h e lowest V,,, and the highest K, were obtained with phenol and o-ethylphenol, respectively. T h e tert.-butyl or the phenyl group placed in the ortho-position and the amino group placed in any position of the phenol molecule inhibit the glucuronidation of the , over 25-fold. V,,, ranged substrate. K , ranged between 0 0 3 and 0 * 7 5 m ~i.e. between 1.18 and 7.60 nmol/min mg-' i.e. over 6-fold. Hydroxybenzoic acids were substrates for UDP-glucuronosyltransferase but it was difficult to determine the kinetics of this enzymic reaction with this substrate. We thus measured the UDP-glucuronosyltransferase activity with a fixed concentration of hydroxybenzoic acid. ratios for sulphotransferase and Table 3 summarizes the V,,,/K,,, UDP-glucuronosyltransferase. T h e lowest and the highest ratios for sulphotransferase were found with tert.-butylphenol (0.0002) and with m-phenylphenol (2.9) ratios thus range over 14 000-fold with phenolic substrates. respectively. Vma,/Km ratios for UDP-glucuronosyltransferase varied between 0.004 T h e Vmax/Km (0-ethylphenol) and 0-16 (p-tert.-butylphenol), thus over a 40-fold range. Table 1. Kinetic parameters of the sulphotransferasein human liver: data given are means 2 SD of four livers. Phenol: K,,, 16.4f3.0; V,,, 0.56&013 Position of the substituent meta
ortho
para
Substituent
Km
Vmx
Km
VInm
Km
Vmax
Methyl Ethyl Propyl tert .-Butyl Phenyl Nitro Amino Carboxyl
1.11 k0.42 1.27f0.35 2.06 & 0.22 344k 21 5 0.28 f0.03 0 9 7 & 0.02 574& 3.27 n.m.
0.32 & 0.1 6 0.22 & 0.1 1 0.23 rf: 0.1 6 0.09 +_ 0.1 1 019&014 0.27 k 0.1 2 0.64 k0.19
6.91 f2.55 4.79 k 1.24
028+016 0 2 5 kO.11
0 9 3 kO.50 0.06 f0.02 027k0.10 23O;tSl n.m.
0.56k035 019k0.14 0.61 0.27 0.62 f0.23
1.58 k0.3 1 12.6 f5.92 3 1.0 & 5.64 110_+ 32.5 10.6 f2 2 5 0.82 k 0.15 33.8 k 18.2 n.m.
0.31 k 0.19 0 2 3 kO.08 0.28 kO.15 0.58 k0.42 014+ 0.09 0.66 k 0.2 1 0.03 f0 0 2
K , , , = ~ MV,a,=nmolmin-L ; mg-'; n.m.=not measurable.
A. Temellini et a!.
174 Table 2.
Kinetic parameters of glucuronyltransferase in human liver: data given are meansf SD of three livers. Phenol:
K,,, 0.051004; Vmsx1.18k0.90 Position of the substituent
ortho
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
Substituent Methyl Ethyl Propyl tert.-Butyl Phenyl Nitro Amino Carboxylf
meta
K m
Vmax
K m
Vmax
K m
Vmax
0.48+0.45 0.75 f0.47 0.22 f0.1 6
2.651099 253 f 1.29 1.93 k0.04
0.24+004 0.32 0.16
4.81 k0.46 4.84 10.83
n.m. n.m.
n.m. n.m.
0.32 10.19
5.69k 1.67
0.08+0.04 005fO.02 005+002
4.15+0.01 5.44k1.55 5.55k0.84
0.17k001 0.32 + 0 1 3 0.10 0.01 003 10.01 0.10f0.04 010+0.04
6.50 k 1.79 7.18 f 0.87 3.78 10.94 4.08f0.53 491 k 1.54 7.60 & 2.55
n.m.
n.m.
n.m.
n.m.
n.m. n.m.
n.m.
+
0.25 & 0.03
0.63 k0.25
K,=mM;
para
V,,,=nmoImin~'mg-l; n.m.=not measurable.
t Final substrate concentration was 0.5 VM. Table 3.
V,,JK,,, ratios for sulphotransferase and glucuronyltransferase. Phenol Glucuronyltransferase
Sulphotransferase
0.04 0.01
0.03 10.009
Position of the substituent
Position of the substituent
ortho
Substituent
meta
ortho
para ~-
Methyl Ethyl Propyl tert.-Butyl Phenyl Nitro Amino
0.18 f0.14 0.17k0.06 0.12 f 0.1 1 00002 0~0001 0.70 0.51 0.28 f 0.12 0.14 k 0.07
+ +
0.04 k 0,035 0.05 k0.01 1.09f084 2.88 f 1.51 2.44 f0 9 0 0.003 +0.001
,.I he ratio VmaJKmis expressed as -.-.nmol
0.22 k0.17
002 k 0006 0.10 k0.007 0005 f 0.005 001 _+0.01 045k0.35 0002 If: 0.002
0.01 10006 0.004+0003 002 f 0.005
meta
0.02 f0.002 002f0.006
0.12 f 0.1 1 010+0~01 002f0.008
para
~~
0.13+0.05
0.04 & 0.009 003&0008 0.04 & 0.01 0 16 5 0 0 8 0.05 k 0 0 2 0.12f0.08
1
m i n x m g PM
Discussion T h e present results show that the Michaelis-Menten constants (K,) and the maximum velocities of reaction (V,,,) of human liver UDP-glucuronosyltransferase and sulphotransferase measured with phenolic substrates are influenced by (i) the position of the substituent in the phenol molecule and also by (ii) its chemical nature (figure 1). Sulphotransferase and UDP-glucuronosyltransferase show independent structure-activity relationships. T h e K,,, of sulphotransferase with phenol is one order of magnitude higher than with methyl- or ethyl- or propyl-ortho-substituted phenols. T h u s one of those substituents adjacent to the functional moiety increases the substrate affinity for the enzyme. In contrast, a bulky group such as the tert.-butyl in the ortho position inhibits the sulphation of the substrate. However, the bulkiness of the ortho-substituent does not appear to be of fundamental importance since orthophenylphenol is a good substrate for sulphotransferase. From these data it appears
175
Human conjugases
Phenol
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
M e t hylphenol
Ethylphend
@
6 \
CH3
OC"*
Ropylphenol
tert-Bu t ylphenol CH3
80 T
Phenylphenol
Nitrophenol
Arninophenol
tlydroxybenzoic acid
Figure 1.
b
N
H
2
@-
Cheniical structures of the substrates for sulphotransferase and UDP-glucuronosyltransferase.
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
176
A. Temellini et al.
that a planar substituent, such as the phenyl group, in the ortho position does not hinder the sulphation of the substrates. Interestingly, sulphotransferase usually has higher K , with p-substituted than ortho-substituted phenols and consistent results were obtained with partially purified phenol sulphotransferase (Campbell et al. 1987b). Such a finding is opposite to that observed for UDP-glucuronosyltransferase. Polar substituents greatly influence the sulphotransferase activity, nitrophenol being a good substrate, aminophenol being a poor substrate and hydroxybenzoic acid not being a substrate. From these data it is evident that sulphotransferase requires substrates whose molecules have neither basic nor acidic substituents. Sulphation of different phenols differs more for the Michaelis-Menten constant than for the maximum velocity of reaction, the former varying over a 5000-fold range, and the latter varying over a 20-fold range. Thus the V,,, to K , ratio varies greatly with different substrates. UDP-glucuronosyltransferase seems to be less influenced than sulphotransferase by the chemical nature of the substrate, since the K , for the former varies only over a 25-fold range. In vivo, glucuronidation generally has low affinity and high capacity, whereas sulphation has high affinity and low capacity, therefore the ratio of these conjugations in vivo will depend on substrate concentration relative to the respective K , values. Phenol is the substrate with the lowest K , and V,,,. An alkyl group in the ortho position increases the K , for UDP-glucuronosyltransferase which is in contrast with the behaviour of sulphotransferase described. T h e presence of a methyl or ethyl group in any position of the phenol molecule increases the rate of glucuronidation. para-Substituted phenols have a higher V,,, than the orthosubstituted ones. T h e presence of a bulky substituent, such as tert.-butyl or phenyl, adjacent to the functional moiety inhibits the substrate glucuronidation. It is of interest that ortho tert.-butylphenol is the compound with the lowest Vmax/Km ratio for sulphotransferase, whereas it is not glucuronidated in vktro b y human liver microsomes. If these data apply to the in vivo situation we might expect that orthotert.-butylphenol may be glucuronidated and sulphated at lower rates than meta- or para-tert.-butylphenol. Aminophenols and hydroxybenzoic acids are poor substrates for human liver UDP-glucuronosyltransferase. In rat liver the rate of glucuronidation of ortho-aminophenol is pH-dependent, reaching the lowest activity at neutral pH (Howland and Bohm 1977). Our assay conditions may not be the optimal ones for this substrate. Biphenyl, a food and environment contaminent with toxic properties, is metabolized by mixed-function oxygenase to para-hydroxybiphenyl with smaller amounts of ortho-hydroxybiphenyl and meta-hydroxybiphenyl (Billings and McMahon 1977, Creaven et al. 1965, Haugen 1981). As suggested by the V,,,, glucuronidation may be more important than sulphation and consistent results were obtained with different methodology (Powis et al. 1987). ortho-Hydroxybiphenyl is a toxic molecule widely used to protect edible crops against fungal growth. This compound may enter the body by intestinal absorption and concentrate in tissues. It is of interest that it is not glucuronidated in vitro by human liver microsomes as glucuronidation usually results in the inactivation of harmful molecules. The Vmax/Km ratios, rather than V,,,, gives a better idea of the efficiency of the enzyme when the substrate concentration is much lower than K,, as may occur in vivo with food or environmental contaminants. T h e Vmax/Km ratios vary over a 14000-fold range for sulphotransferase and a 40-fold range for UDP-glucuronosyltransferase.
Human conjugases
177
Under particular circumstances, hepatic glucuronidation prevails over hepatic sulphation, thus the former seems more important than the latter in the conjugation of phenols. This consideration is also corroborated by the lower dependence of UDP-glucuronosyltransferase on substrate molecular structure. In extrahepatic tissues the picture may be different, with a prevalence of sulphation over glucuronidation (Powis et al. 1987, Pacifici et al. 1988). More work is necessary to ascertain whether the substrate dependence pattern of K , and V,,, of sulphotransferase observed in the liver is consistent in different tissues.
Acknowledgement Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
This work was supported by the Italian Association for Cancer Research.
References ANDERSON, R. J., and WEINSHILBOUM, R. M., 1980, Phenol sulphotransferase in human tissue: radiochemical enzymatic assay and biochemical properties. Clinica et Chimica Acta, 103, 79-90. BANSAL, S. K., and GESSNER, T., 1980, A unified method for the assay of uridine diphosphoglucuronyltransferase activities toward various aglycones using uridine diphospho (U-I4C) glucuronic acid. Analytical Biochemistry, 109, 32 1-329, BILLINGS, R. E., and MCMAHON, R. E., 1977, Microsomal biphenyl hydroxylation: the formation of 3-hydroxybiphenyl and biphenyl catechol. Molecular Pharmacology, 14, 145-1 54. J. A,, 1987, Indirect evidence of UDP-glucuronosyltransferase heterogeneity: how can it help BOUTIN, purification? Drug Metabolism Review, 18, 517-551. BURCHELL, B., and COUGHTRIE, M. W. H., 1989, UDP-glucuronosyltransferases. Pharmacology and Therapeutics, 43, 261-289. R. M., 1987a, Human liver phenol CAMPBELL, N. R. C., VAN LOON,J. A,, and WEINSHILBOUM, sulphotransferase: assay condition, biochemical properties and partial purification of isozymes of the thermostable form. Biochemical Pharmacology, 36, 1435-1446. K.S.,AMES,M. M., HANSCH, C., and WEINSHILBOUM, CAMPBELL, N. R. C., VANLOON,J. A., SUNDARAM, R., 1987 b, Human and rat liver phenol sulphotransferase: structureactivity relationships for phenolic substrates. Molecular Pharmacology, 32, 81 3-819. D. V., and WILLIAMS, R., 1965, A fluorimetric study of the hydroxylation of CREAVEN, P. J., PARKE, biphenyl in oitro by liver preparations of various species. Biochemical Journal, 96,879-885. ~ I A U G ED. N A,, , 1981, Biphenyl metabolism by rat liver microsomes. Drug Metabolism and Disposition, 9, 21 2-21 8. HOWI.AND, R. D., and BOHM,L . D., 1977, Possible multiple binding sites for o-aminophenol on uridine diphosphate plucuronyltransferase. Biochemical Journal, 163, 125-1 31. S., 1984, Sulphotransferase active with JAKOBY, W. B., DUFFEL, M. W., LYON,E. S., and RAMASWAMY, xenobiotics-comments on mechanism. In Progress in Drug Metabolism, edited by J. W. Bridges and L. F. Chasseaud (London and Philadelphia: Taylor & Francis), pp. 11-100. LOWRY, 0. H., ROSEBROUGH, N. J., FARR,A. L., and RANDALL, R. J., 1951, Protein measurement with Folin phenol reagent. Journal of Biological Chemistry, 193, 265-275. MULDER, G. J., 1982, Conjugation of phenols. In Metabolic Basis of Detoxication, Metabolism of Functional Groups, edited by W. B. Jakoby, J. R. Bend and J. Caldwell (New York and London: Academic Press), pp. 248-269. J. H., 1978, Glucuronidation and sulphation in vivo and in uitro: selective MULDER, G. J., and MEERMAN, inhibition of sulphation by drugs and deficiency of inorganic sulphate. In Conjugation Reactions in Drug Biotransformation, edited by A. Aito (Amsterdam: Elsevier-North-Holland), pp. 389-397. PACIFICI, G. M., FRANCHI, M., BENCINI, C., REPETTI, F., DI LASCIO, N., and MURARO, G . B., 1988, 'I'issue distribution of drug-metabolizing enzymes in humans. Xenobiotica, 18, 849-856. PACIFICI, G. M., FRANCHI, M., GIULIANI, L., and RANE,A , , 1990, Development of the glucuronyltransferase and sulphotransferase towards 2-naphthol in human fetus. Developmental Pharmacology and Therapeutics, 14, 108-1 14. K. S., 1987, A high-performance liquid POWIS,G., MOORE,D. J., WII-KE,T. J., and SANTONE, chromatography assay for measuring integrated biphenyl metabolism by intact cells: its use with rat liver and human liver and kidney. Analytical Biochemistry, 167, 191-198.
1991, VOL. 21,
NO.
2, 171-177
Human liver sulphotransferase and UDP-glucuronosyltransferase: structure-activity relationship for phenolic substrates A. T E M E L L I N I t , M. FRANCHIt, L. GIULIANIS and G. M. PACIFICItQ
t Department of Experimental Biomedicine, Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
1Department of Surgery, Medical School, University of Pisa, 56100 Pisa, Italy
Received 14 February 1990; accepted 10 August 1990
1. Human liver sulphotransferase and UDP-glucuronosyltransferase were studied with phenol, methyl-, ethyl-, propyl-, butyl-, phenyl-, nitro-, amino-phenols and hydroxybenzoic acids as substrates. 2. T h e Michaelis-Menten constants (K,) and the maximum velocities of reaction (V,,,,,) of sulphotransferase and UDP-glucuronosyltransferase for each substrate were measured. 3. T h e K , values for sulphotransferase varied over 5000-fold whereas they varied over 25-fold for UDP-glucuronosyltransferase. 4. Sulphotransferase and UDP-glucuronosyltransferase have different structureactivity relationships with phenolic substrates.
Introduction UDP-glucuronosyltransferase and sulphotransferase are two important conjugation enzymes, each consisting of different forms (Boutin 1987, Burchell and Coughtrie 1989, Jakoby et al. 1984). Overlapping substrate specificities of UDPglucuronosyltransferase and sulphotransferase have been described for a limited number of substrates (Mulder and Meerman 1978, Mulder 1982, Pacifici et al. 1990), but it is unknown whether such overlapping substrate specificities may be extended to a broad spectrum of substrates. A multitude of drugs and environmental pollutants have a phenolic structure. This prompted us to investigate the structure-activity relationship of sulphotransferase and glucuronyltransferase with phenolic molecules. T h e majority of the compounds studied were found to be substrates for both UDPglucuronosyltransferase and sulphotransferase, but some served as substrate for only one of the two enzymes. We also observed that the Vmaxto K, ratio, which gives an indication of enzyme efficiency, has a more marked substrate dependence for sulphotransferase than for UDP-glucuronosyltransferase. Thus, these two enzymes may contribute to the conjugation of the same substrate, but their contribution is substrate-dependent.
Materials and methods Chemicals Labelled adenosine 3‘-pho~phate-5’-phospho[~~S]sulphate (PAPS) (specific activity 2.5 Ci/mmol) and I4C-uridine 5’-diphosphoglucuronic acid (UDPGA) (specific activity 304 mCi/mmol) were obtained from New England Nuclear Co. (Florence, Italy). Unlabelled UDPGA, bovine serum albumin (BSA),
5 Address correspondence to: Gian Maria Pacifici, Associate Professor in Pharmacology, Department of Experimental Biomedicine, Medical School, Via Roma 55, 56100 Pisa, Italy. 0049-8254/91 $3.00 0 1991 ‘Taylor & Francis Ltd
172
A. Temellini et al.
DI,-dithiothreitol ( D T T ) and Tr is (tris-(hydroxymethy1)aminomethane) were obtained from Sigma Chemical Co. (St Louis MO, USA). Sulphate or glucuronate acceptor substrates were from Aldrich (Milan, Italy) and their chemical purity ranged between 98% and 99% for all the substrates except rn- and p-phenylphenol, whose purity was 90% and 97%, respectively.
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
Biological material Wedge human liver specimens were obtained at laparotomy from two males 50 and 60 years old, and 11 females (age between 23 and 65 years) undergoing cholecystectomy. Liver specimens with normal cell architecture only were used. Biopsies were homogenized in 5 vol. 0.25 M sucrose by means of a glassTeflon homogenizer. T h e homogenates were centrifuged at 12 OOOg for 15 min and the supernatants were centrifuged again at 105 OOOg for 1 h. T h e ensuing supernatants were used as the cytosolic fractions, and the pellets were resuspended in 0.1 M Tril-HCI (pH7.4) containing 30% glycerol and studied as the microsomal fractions. Assays Sulphotransjerase. T h e incubation mixture for sulphotransferase was as previously described by Anderson and Weinshilboum (1980) and modified by Campbell et al. (1987a). Briefly, the incubation consisted of lOmM potassium pyrophosphate buffer ( p H 6.5),0 . 4 35S-PAPS ~ ~ mixture (final vol. 150~1) (about 400 000 dpm), the sulphate acceptor substrate, dissolved in dimethyl sulphoxide, 1 4 m D ~TT, 0.0645% BSA and loop1 of diluted cytosolic fraction to give a final protein concentration of 1 pg/ml. Liver cytosols were diluted with 5 mM potassium pyrophosphate buffer ( p H 7.4) containing 0,625 mg/ml albumin and 1.54 mg/ml DTT to give a final concentration of 1.5 pg/ml, divided into 1 ml aliquots and stored at -80°C. T h e reaction was started by addition of PA PS and carried out at 37°C for 30min and stopped by addition of a mixture of zinc sulphate and barium acetate as described elsewhere (Campbell e t al. 1987a). After centrifugation at 1500g. aliquots ( 4 0 0 ~ 1of ) the clear supernatant were transferred to scintillation vials containing 3 ml of scintillation fluid (Scintillator 299TM,from Packard, Milan, Italy). l' he linearity for protein concentration and incubation time were determined for phenol and each of p-substituted phenols, and occasionally for other substrates. Reaction was linear at least u p to 5pg/ml cytosolic protein and 40 min of incubation with all substrates studied. Final protein concentration and incubation time were fixed to 1 pg/ml and 30min, respectively, and such conditions were kept constant throughout the study. Each assay had two blank samples which were the same as the active samples except that the substrate was replaced by dimethyl sulphoxide. Four liver samples, whose donors were one man 60 years old and three women 38.50 and 5 1 years old, were used throughout the study. T h e appropriate range of substrate concentration was first studied bylOfold substrate dilutions from 1 mM to 0.01 p ~ T.h e enzyme kinetics were then studied with six concentrations of substrate, each being double the next, and assayed in duplicate. T h e enzyme kinetic constants, the Michaelis-Menten constant (K,) and the maximum reaction velocity (V,,,) were determined by Eadie-Hofstee plots using a computer program and the given data are the averages of four livers. Each curve was fitted by a straight line whose correlation was greater than 0.98. C;lururonosy[transferase. T h e incubation mixture for UDP-glucuronosyltransferase was as previously described by Bansal and Gessner (1980). Incubations were carried out at 37°C for 1 0 mi n in a final volume o f loop1 containing 0.1 M 'I'ris-maleate buffer (pH7.4), 5 mM MgCI,, 2 mM 14C-UDPGA (SOOOOdpm), the appropriate concentration of the substrate which was added in 2p1 of dimethyl sulphoxide, and an aliquot of the microsomal protein to give a final protein concentration of 2.5 mg/ml. No activator of U D P ~lucuronosyltransferasewas added to the incubation mixture. Reactions were started by addition of ) the incubates were applied to U D P G A and stopped by addition of 200pl of ethanol. Aliquots ( 4 0 ~ 1 of t.1.c. plates L K 5 D F (Whatman, Milan, Italy) and chromatographed in n-butanol-acetone-acetic acid30% ammonia-water (70: 50: 18 : 1.5 : 60 by vol.). T.1.c. plates were visualized under U.V.light and the spots corresponding to the glucuronides were removed and placed in scintillation vials containing 10 ml of scintillant fluid (Scintillator 299TM,Packard, Milan, Italy). Each sample was assayed in duplicate and for each incuhation two blanks were used, having the same composition as the active samples except that the substrate was replaced by 2 p l of dimethyl sulphoxide. Labelled U D P G A was originally provided in ethanol-water (7 : 3 v / v ) and to keep the final concentration of ethanol in incubation mixture below 5%, as it inactivates UDP-glucuronosyltransferase, the ethanolic solution of U D P G A was evaporated at room temperature under a flow of N,, and the original volume reconstituted with water. UDI'GA solutions were divided into aliquots and stored at -20°C for not longer than 2 weeks. Optimal incubation time and protein concentration were determined with p-substituted phenols and occasionally with other substrates, and were found to be 10min and 2.5 mg/ml protein, respectively. Because of the relatively high amount of microsornal protein required for each assay 11 liver biopsies were used in the study. T h e liver donors were nine women and two men, aged between 23 and 65 years. for each substrate were measured in three livers by construction of T h e kinetic parameters K,,, and VmaK the Eadie-Hofstee plots. T h e enzyme obeyed Michaelis--Menten kinetics with each substrate studied, and the regression coefficient of each plot was greater than 0.97.
Human conjugases
173
The activities of sulphotransferaseand UDP-glucuronosyltransferasewere measured on the basis of the specific radioactivity of PAPS or UDPGA after correction for blanks. The radioactivity in active samples was at least twice as high as in the blank samples. Protein concentration was measured as described by Lowry et al. (1951).
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
Results Table 1 summarizes the kinetic parameters of the human hepatic sulphotransferase obtained with different phenolic substrates. Both K , and V,,, are dependent on (i) the chemical nature of the substituent and also on (ii) its position in the phenol molecule. The highest V,,, and the lowest K , were obtained withp-nitrophenol and m-phenylphenol, respectively. T h e lowest V,,, and the highest K, were obtained with para-aminophenol and ortho-tert.-butyl phenyl, respectively. T h e carboxyl group placed at any position of the phenol molecule inhibits sulphation. K , varied between 0.06 and 344pM, i.e. over a 5000-fold range. V,,, varied between 0 0 3 and 0.66 nmol/minmg-', i.e. over a 20-fold range. Table 2 gives the kinetic parameters of the human hepatic UDPglucuronosyltransferase obtained with different substrates. Both V,,, and K , vary with the chemical nature of the substrate. T h e highest V,,, and the lowest K , were obtained with p-nitrophenol and p-tert.-butylphenol, respectively. T h e lowest V,,, and the highest K, were obtained with phenol and o-ethylphenol, respectively. T h e tert.-butyl or the phenyl group placed in the ortho-position and the amino group placed in any position of the phenol molecule inhibit the glucuronidation of the , over 25-fold. V,,, ranged substrate. K , ranged between 0 0 3 and 0 * 7 5 m ~i.e. between 1.18 and 7.60 nmol/min mg-' i.e. over 6-fold. Hydroxybenzoic acids were substrates for UDP-glucuronosyltransferase but it was difficult to determine the kinetics of this enzymic reaction with this substrate. We thus measured the UDP-glucuronosyltransferase activity with a fixed concentration of hydroxybenzoic acid. ratios for sulphotransferase and Table 3 summarizes the V,,,/K,,, UDP-glucuronosyltransferase. T h e lowest and the highest ratios for sulphotransferase were found with tert.-butylphenol (0.0002) and with m-phenylphenol (2.9) ratios thus range over 14 000-fold with phenolic substrates. respectively. Vma,/Km ratios for UDP-glucuronosyltransferase varied between 0.004 T h e Vmax/Km (0-ethylphenol) and 0-16 (p-tert.-butylphenol), thus over a 40-fold range. Table 1. Kinetic parameters of the sulphotransferasein human liver: data given are means 2 SD of four livers. Phenol: K,,, 16.4f3.0; V,,, 0.56&013 Position of the substituent meta
ortho
para
Substituent
Km
Vmx
Km
VInm
Km
Vmax
Methyl Ethyl Propyl tert .-Butyl Phenyl Nitro Amino Carboxyl
1.11 k0.42 1.27f0.35 2.06 & 0.22 344k 21 5 0.28 f0.03 0 9 7 & 0.02 574& 3.27 n.m.
0.32 & 0.1 6 0.22 & 0.1 1 0.23 rf: 0.1 6 0.09 +_ 0.1 1 019&014 0.27 k 0.1 2 0.64 k0.19
6.91 f2.55 4.79 k 1.24
028+016 0 2 5 kO.11
0 9 3 kO.50 0.06 f0.02 027k0.10 23O;tSl n.m.
0.56k035 019k0.14 0.61 0.27 0.62 f0.23
1.58 k0.3 1 12.6 f5.92 3 1.0 & 5.64 110_+ 32.5 10.6 f2 2 5 0.82 k 0.15 33.8 k 18.2 n.m.
0.31 k 0.19 0 2 3 kO.08 0.28 kO.15 0.58 k0.42 014+ 0.09 0.66 k 0.2 1 0.03 f0 0 2
K , , , = ~ MV,a,=nmolmin-L ; mg-'; n.m.=not measurable.
A. Temellini et a!.
174 Table 2.
Kinetic parameters of glucuronyltransferase in human liver: data given are meansf SD of three livers. Phenol:
K,,, 0.051004; Vmsx1.18k0.90 Position of the substituent
ortho
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
Substituent Methyl Ethyl Propyl tert.-Butyl Phenyl Nitro Amino Carboxylf
meta
K m
Vmax
K m
Vmax
K m
Vmax
0.48+0.45 0.75 f0.47 0.22 f0.1 6
2.651099 253 f 1.29 1.93 k0.04
0.24+004 0.32 0.16
4.81 k0.46 4.84 10.83
n.m. n.m.
n.m. n.m.
0.32 10.19
5.69k 1.67
0.08+0.04 005fO.02 005+002
4.15+0.01 5.44k1.55 5.55k0.84
0.17k001 0.32 + 0 1 3 0.10 0.01 003 10.01 0.10f0.04 010+0.04
6.50 k 1.79 7.18 f 0.87 3.78 10.94 4.08f0.53 491 k 1.54 7.60 & 2.55
n.m.
n.m.
n.m.
n.m.
n.m. n.m.
n.m.
+
0.25 & 0.03
0.63 k0.25
K,=mM;
para
V,,,=nmoImin~'mg-l; n.m.=not measurable.
t Final substrate concentration was 0.5 VM. Table 3.
V,,JK,,, ratios for sulphotransferase and glucuronyltransferase. Phenol Glucuronyltransferase
Sulphotransferase
0.04 0.01
0.03 10.009
Position of the substituent
Position of the substituent
ortho
Substituent
meta
ortho
para ~-
Methyl Ethyl Propyl tert.-Butyl Phenyl Nitro Amino
0.18 f0.14 0.17k0.06 0.12 f 0.1 1 00002 0~0001 0.70 0.51 0.28 f 0.12 0.14 k 0.07
+ +
0.04 k 0,035 0.05 k0.01 1.09f084 2.88 f 1.51 2.44 f0 9 0 0.003 +0.001
,.I he ratio VmaJKmis expressed as -.-.nmol
0.22 k0.17
002 k 0006 0.10 k0.007 0005 f 0.005 001 _+0.01 045k0.35 0002 If: 0.002
0.01 10006 0.004+0003 002 f 0.005
meta
0.02 f0.002 002f0.006
0.12 f 0.1 1 010+0~01 002f0.008
para
~~
0.13+0.05
0.04 & 0.009 003&0008 0.04 & 0.01 0 16 5 0 0 8 0.05 k 0 0 2 0.12f0.08
1
m i n x m g PM
Discussion T h e present results show that the Michaelis-Menten constants (K,) and the maximum velocities of reaction (V,,,) of human liver UDP-glucuronosyltransferase and sulphotransferase measured with phenolic substrates are influenced by (i) the position of the substituent in the phenol molecule and also by (ii) its chemical nature (figure 1). Sulphotransferase and UDP-glucuronosyltransferase show independent structure-activity relationships. T h e K,,, of sulphotransferase with phenol is one order of magnitude higher than with methyl- or ethyl- or propyl-ortho-substituted phenols. T h u s one of those substituents adjacent to the functional moiety increases the substrate affinity for the enzyme. In contrast, a bulky group such as the tert.-butyl in the ortho position inhibits the sulphation of the substrate. However, the bulkiness of the ortho-substituent does not appear to be of fundamental importance since orthophenylphenol is a good substrate for sulphotransferase. From these data it appears
175
Human conjugases
Phenol
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
M e t hylphenol
Ethylphend
@
6 \
CH3
OC"*
Ropylphenol
tert-Bu t ylphenol CH3
80 T
Phenylphenol
Nitrophenol
Arninophenol
tlydroxybenzoic acid
Figure 1.
b
N
H
2
@-
Cheniical structures of the substrates for sulphotransferase and UDP-glucuronosyltransferase.
Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
176
A. Temellini et al.
that a planar substituent, such as the phenyl group, in the ortho position does not hinder the sulphation of the substrates. Interestingly, sulphotransferase usually has higher K , with p-substituted than ortho-substituted phenols and consistent results were obtained with partially purified phenol sulphotransferase (Campbell et al. 1987b). Such a finding is opposite to that observed for UDP-glucuronosyltransferase. Polar substituents greatly influence the sulphotransferase activity, nitrophenol being a good substrate, aminophenol being a poor substrate and hydroxybenzoic acid not being a substrate. From these data it is evident that sulphotransferase requires substrates whose molecules have neither basic nor acidic substituents. Sulphation of different phenols differs more for the Michaelis-Menten constant than for the maximum velocity of reaction, the former varying over a 5000-fold range, and the latter varying over a 20-fold range. Thus the V,,, to K , ratio varies greatly with different substrates. UDP-glucuronosyltransferase seems to be less influenced than sulphotransferase by the chemical nature of the substrate, since the K , for the former varies only over a 25-fold range. In vivo, glucuronidation generally has low affinity and high capacity, whereas sulphation has high affinity and low capacity, therefore the ratio of these conjugations in vivo will depend on substrate concentration relative to the respective K , values. Phenol is the substrate with the lowest K , and V,,,. An alkyl group in the ortho position increases the K , for UDP-glucuronosyltransferase which is in contrast with the behaviour of sulphotransferase described. T h e presence of a methyl or ethyl group in any position of the phenol molecule increases the rate of glucuronidation. para-Substituted phenols have a higher V,,, than the orthosubstituted ones. T h e presence of a bulky substituent, such as tert.-butyl or phenyl, adjacent to the functional moiety inhibits the substrate glucuronidation. It is of interest that ortho tert.-butylphenol is the compound with the lowest Vmax/Km ratio for sulphotransferase, whereas it is not glucuronidated in vktro b y human liver microsomes. If these data apply to the in vivo situation we might expect that orthotert.-butylphenol may be glucuronidated and sulphated at lower rates than meta- or para-tert.-butylphenol. Aminophenols and hydroxybenzoic acids are poor substrates for human liver UDP-glucuronosyltransferase. In rat liver the rate of glucuronidation of ortho-aminophenol is pH-dependent, reaching the lowest activity at neutral pH (Howland and Bohm 1977). Our assay conditions may not be the optimal ones for this substrate. Biphenyl, a food and environment contaminent with toxic properties, is metabolized by mixed-function oxygenase to para-hydroxybiphenyl with smaller amounts of ortho-hydroxybiphenyl and meta-hydroxybiphenyl (Billings and McMahon 1977, Creaven et al. 1965, Haugen 1981). As suggested by the V,,,, glucuronidation may be more important than sulphation and consistent results were obtained with different methodology (Powis et al. 1987). ortho-Hydroxybiphenyl is a toxic molecule widely used to protect edible crops against fungal growth. This compound may enter the body by intestinal absorption and concentrate in tissues. It is of interest that it is not glucuronidated in vitro by human liver microsomes as glucuronidation usually results in the inactivation of harmful molecules. The Vmax/Km ratios, rather than V,,,, gives a better idea of the efficiency of the enzyme when the substrate concentration is much lower than K,, as may occur in vivo with food or environmental contaminants. T h e Vmax/Km ratios vary over a 14000-fold range for sulphotransferase and a 40-fold range for UDP-glucuronosyltransferase.
Human conjugases
177
Under particular circumstances, hepatic glucuronidation prevails over hepatic sulphation, thus the former seems more important than the latter in the conjugation of phenols. This consideration is also corroborated by the lower dependence of UDP-glucuronosyltransferase on substrate molecular structure. In extrahepatic tissues the picture may be different, with a prevalence of sulphation over glucuronidation (Powis et al. 1987, Pacifici et al. 1988). More work is necessary to ascertain whether the substrate dependence pattern of K , and V,,, of sulphotransferase observed in the liver is consistent in different tissues.
Acknowledgement Xenobiotica Downloaded from informahealthcare.com by Nyu Medical Center on 04/28/15 For personal use only.
This work was supported by the Italian Association for Cancer Research.
References ANDERSON, R. J., and WEINSHILBOUM, R. M., 1980, Phenol sulphotransferase in human tissue: radiochemical enzymatic assay and biochemical properties. Clinica et Chimica Acta, 103, 79-90. BANSAL, S. K., and GESSNER, T., 1980, A unified method for the assay of uridine diphosphoglucuronyltransferase activities toward various aglycones using uridine diphospho (U-I4C) glucuronic acid. Analytical Biochemistry, 109, 32 1-329, BILLINGS, R. E., and MCMAHON, R. E., 1977, Microsomal biphenyl hydroxylation: the formation of 3-hydroxybiphenyl and biphenyl catechol. Molecular Pharmacology, 14, 145-1 54. J. A,, 1987, Indirect evidence of UDP-glucuronosyltransferase heterogeneity: how can it help BOUTIN, purification? Drug Metabolism Review, 18, 517-551. BURCHELL, B., and COUGHTRIE, M. W. H., 1989, UDP-glucuronosyltransferases. Pharmacology and Therapeutics, 43, 261-289. R. M., 1987a, Human liver phenol CAMPBELL, N. R. C., VAN LOON,J. A,, and WEINSHILBOUM, sulphotransferase: assay condition, biochemical properties and partial purification of isozymes of the thermostable form. Biochemical Pharmacology, 36, 1435-1446. K.S.,AMES,M. M., HANSCH, C., and WEINSHILBOUM, CAMPBELL, N. R. C., VANLOON,J. A., SUNDARAM, R., 1987 b, Human and rat liver phenol sulphotransferase: structureactivity relationships for phenolic substrates. Molecular Pharmacology, 32, 81 3-819. D. V., and WILLIAMS, R., 1965, A fluorimetric study of the hydroxylation of CREAVEN, P. J., PARKE, biphenyl in oitro by liver preparations of various species. Biochemical Journal, 96,879-885. ~ I A U G ED. N A,, , 1981, Biphenyl metabolism by rat liver microsomes. Drug Metabolism and Disposition, 9, 21 2-21 8. HOWI.AND, R. D., and BOHM,L . D., 1977, Possible multiple binding sites for o-aminophenol on uridine diphosphate plucuronyltransferase. Biochemical Journal, 163, 125-1 31. S., 1984, Sulphotransferase active with JAKOBY, W. B., DUFFEL, M. W., LYON,E. S., and RAMASWAMY, xenobiotics-comments on mechanism. In Progress in Drug Metabolism, edited by J. W. Bridges and L. F. Chasseaud (London and Philadelphia: Taylor & Francis), pp. 11-100. LOWRY, 0. H., ROSEBROUGH, N. J., FARR,A. L., and RANDALL, R. J., 1951, Protein measurement with Folin phenol reagent. Journal of Biological Chemistry, 193, 265-275. MULDER, G. J., 1982, Conjugation of phenols. In Metabolic Basis of Detoxication, Metabolism of Functional Groups, edited by W. B. Jakoby, J. R. Bend and J. Caldwell (New York and London: Academic Press), pp. 248-269. J. H., 1978, Glucuronidation and sulphation in vivo and in uitro: selective MULDER, G. J., and MEERMAN, inhibition of sulphation by drugs and deficiency of inorganic sulphate. In Conjugation Reactions in Drug Biotransformation, edited by A. Aito (Amsterdam: Elsevier-North-Holland), pp. 389-397. PACIFICI, G. M., FRANCHI, M., BENCINI, C., REPETTI, F., DI LASCIO, N., and MURARO, G . B., 1988, 'I'issue distribution of drug-metabolizing enzymes in humans. Xenobiotica, 18, 849-856. PACIFICI, G. M., FRANCHI, M., GIULIANI, L., and RANE,A , , 1990, Development of the glucuronyltransferase and sulphotransferase towards 2-naphthol in human fetus. Developmental Pharmacology and Therapeutics, 14, 108-1 14. K. S., 1987, A high-performance liquid POWIS,G., MOORE,D. J., WII-KE,T. J., and SANTONE, chromatography assay for measuring integrated biphenyl metabolism by intact cells: its use with rat liver and human liver and kidney. Analytical Biochemistry, 167, 191-198.
