Alkaline butanol extraction of bile salt and steroid sulfate esters: Application to the assay of sulfotransferases

ANALYTICAL

BIOCHEMISTRY

133,470-475

(1983)

Alkaline Butanol Extraction of Bile Salt and Steroid Sulfate Esters: Application to the Assay of Sulfotransferases STEPHENBARNES,**~" RODNEY~ALDROP,~ AND ANNS. NEIGHBORS? *Departments of Biochemistry and Pharmacology and fComprehensive Cancer Center, University of Alabama in Birmingham,

Birmingham, Alabama 35294

Received April 28, 1983 A butan- l-01 solvent-extraction procedure has been evaluated for the assay of 3’-phosphoadenosine-5’-phosphosulfate:sulfotransferase activity with various bile salt and steroid substrates. Although butanol extracted the sulfate esters of steroids and bile salts from aqueous solution at neutral pH, extraction at basic pH gave optimum recovery which was independent of protein in the sample. Greater than 99.9% of unmatted 3’-phosphoadenosine-5’-phospho[3’S]sulfate remained in the aqueous phase. The data for sulfotransferase activities obtained with this solventextraction assay were not si&icantly different from those obtained with a standard thin-layer chromatography method. Solvent extraction has enabled multiple, rapid assaysof several steroid and bile salt sulfotransferases during chromatographic purification of these enzymes from tissue fractions. KEY WORDS: radioassay; sulfotransferase; bile salt; steroid; liver; solvent extraction.

The formation of bile salt and steroid sulfate esters is an important metabolic pathway in health and disease (1,2). The enzymes responsible for this metabolic step utilize an active sulfate intermediate, 3’-phosphoadenosine-S-phosphosulfate (PAPS),2 for sulfate ester formation. Unfortunately, the reaction cannot be followed by spectrophotometric measurements of the reactants or products. Instead, it has been necessary to utilize radiolabeled substrates; the extent of the reaction is determined by the separation of the sulfated and unsulfated species. Such separations have ’ To whom correspondence should be addressed: Department of Pharmacology, R 123 Volker Hall, University of Alabama in Birmingham, Birmingham, Ala. 35294. 2 Abbreviations used: PAPS, 3’-phosphoadenosine-5’phosphosulfate; TLC, thin-layer chromatography; glycolithocholate and glycochenodeoxycholate, trivial names for the N-glycine conjugates of 3a-hydroxy-5@zholan-24oate and 3a,7adihydroxy-5,%cholan-24-oate, respectively; bile salt sulfotransferases, a term used to describe a group of sulfotransferascs which catalyze the formation of bile salt sulfates. One of these enzymes has been previously partiahy purifred (7), taurolithocholate sulfotmnsferase (EC 2.8.2.14). 0003-2697183 $3.00 Copyright 0 1983 by Academic Press. Inc. All rights of reproduction in any form reserwd.

470

been carried out by thin-layer chromatography (3,4), high-pressure liquid chromatography (5), or solid-phase methods (6,7). None of these procedures was found to be appropriate for purification studies of hepatic sulfotransferases, where speed and simplicity are essential. In contrast, solvent-extraction methods are usually rapid (8,9). In this study we sought to develop a single solvent-extraction procedure which would be suitable for the rapid, quantitative assay of several bile salt and steroid sulfotransferases. In the course of these investigations, we examined the effect of varying amounts of protein and the effect of the extraction pH on the recovery of the sulfate esters. MATERIALS

AND

METHODS

Unlabeled bile salts and steroids and dicyclohexylcarbodiimide were purchased from Sigma Chemical Company, St. Louis, Missouri. [ 1,2-3H]Glycochenodeoxycholate (2.3 Ci/mmol), [2,4,6,7-3H]estradio1 (90 Ci/mmol), [ 1,2-3H]dehydroepiandrosterone (40.2 Ci/

SOLVENT-EXTRACTION

ASSAYS FOR SULFOTRANSFERASES

mmol), and 3’-phosphoadenosine-5’-phospho[35S]stiate (1.5 Ci/mmol) were purchased from New England Nuclear Corporation, Boston, Massachusetts. Unlabeled PAPS was obtained from P-L Biochemicals, Milwaukee, Wisconsin. The purity of 35S-labeled and unlabeled PAPS was checked by paper electrophoresis as described earlier ( 10). Silica gel Gcoated glass thin-layer plates were obtained from Fisher Chemical Company, St. Louis, Missouri. Synthesis of sulfate esters. The method of Mumma (11) was used for synthesis of sulfate esters, as previously described ( 12). After initial separation by TLC, the sulfate ester was eluted from the TLC plate with methanol:ammonium hydroxide (49: 1, v/v). The solvent solution was cooled to 0°C and the methanol evaporated under a stream of N2. The aqueous residue was dissolved in water and freezedried. The purity of the radioactive sulfate esters was checked by TLC in chlorofornnmethanol:acetic acid:water (65:24: 15:5) and in chlorofornnmethanokammonium hydroxide (8 M)

471

cm-wide track on a 20 X 20-cm glass plate coated with a 0.25-mm-thick layer of silica gel G. After development with chloroform:methanol:acetic acid:water (6524: 15:5), the appropriate standards, which had been spotted in adjacent lanes, were located by spraying with Usui’s reagent (13) and heating at 100°C for 5- 10 min. Zones corresponding to the sulfate ester and the unreacted labeled substrate were scraped into plastic scintillation vials, 5 ml NE-260 scintillant (Nuclear Enterprises, San Carlos, Calif.) was added, and the radioactivity was determined. In preliminary experiments, each track was divided into 0.5-cm zones to locate the metabolic product formed from each substrate; this corresponded with the location of the standard sulfate ester in each case. Solvent-extraction method. Assays were terminated by boiling, and after cooling were diluted with 0.4 ml of ice-cold 1 M ammonium hydroxide and thoroughly mixed. The precipitated protein was removed by centrifugation at 3000g for 5 min. The supernatant was transferred to a 10 X 75-mm glass test (60:30:4). tube and the 35S-labeled products were exAssay of sulfatransferases. Enzyme activity tracted with 1 ml butanol by vigorous mixing. was assayed in a total volume of 100 ~1 con- After centrifugation at 3000g for 5 min to taining bile salt or steroid (100 PM), PAPS (90 separate the phases, the butanol layer was pM), 0.2 PCi labeled substrate (either ‘H-lacarefully transferred to a 10 X 75-mm test beled bile salt or steroid, or [35S]PAPS), MgC12 tube (carryover of small amounts of the (5 mM), 100 mM sodium phosphate buffer, aqueous phase was permitted at this step). It pH 7.0, and hepatic or breast tumor cytosol was backwashed with 0.45 ml of 1 M am(50-250 pg protein). Blank incubations were monium hydroxide containing 100 mM socarried out using 200 pg bovine serum al- dium phosphate buffer (pH 7.0) presaturated bumin (Cohn Fraction V) (Sigma Chemical with butanol, final pH 10.8. An aliquot (0.8 Co.) in place of cytosol. Incubations were car- ml) of the butanol layer was removed for mearied out at 37°C for 0 to 2 h. Reaction was surement of radioactivity. Correction for the terminated by placing the tubes in a boilingvariable quenching between samples was perwater bath for 2 min to precipitate protein. formed using a channels-ratio method. Thin-layer chromatographic analysis. To Effect of protein on recovery of product. validate the technique, reactants from assays Since bile salt and steroid sulfates are extenwere analyzed by TLC separation. Assays in sively bound to serum proteins in blood ( 14), which 3H-labeled substrates were used were it was anticipated that binding might influence processed as described previously (4, 10). In recovery in the solvent-extraction assays. incubations using [35S]PAPS, the precipitated Known amounts of labeled bile salt or steroid protein was removed by centrifugation and sulfates were added to incubation mixtures 20 ~1 of the supernatant was applied to a 1S- containing varying amounts of liver or tumor

472

BARNES, WALDROP, TABLE 1 RECOVERYOFSULFATEESTERSOFBILE SALTSANDSTEROIDS Percentage

Sulfate ester Glycolithocholate-3cY-[“S]sulfate Glycochenodeoxycholate-7a[“S]sulfate 3/3-Hydroxy-5-cholenoate-36[“S]sulfate Dehydroepiandrosterone-3@ [“S]sulfate [2,4,6,73H]Estradiol-3-sulfate

recovery” 83.71 ? 1.82 (6) 84.73 f 0.96 (6) 92.89 f 0.70 (6) 91.08 +_0.25 (6) 88.39 f 0.86 (5)

Note. A 200 pmol(O.025 pCi) amount of each of the above sulfates was added to the incubation buffer which was then boiled, diluted with 1 M ammonium hydroxide, and extracted with butanol. The butanol phase was backwashed with ammonium hydroxide:phosphate buffer, pH 10.8. See Methods for full details of the extraction procedure. u Mean +- SD, number of replicates in parentheses.

cytosol. Each was immediately tracted as above.

boiled and ex-

RESULTS

Recovery of Substrates in Butanol At pH values ranging from 7.0 to 11 .O and a butanol:water ratio of 2: 1, less than 0.1% of

AND NEIGHBORS

the radioactivity (30-40 cpm) from [35S]PAPS entered the butanol phase. In contrast, although recoveries of bile salt and steroid sulfates at pH 11 ranged from 85 to 92% using the standard method (Table I), low recoveries were measured under other previously used conditions (9,15). Termination of incubation by boiling at pH 7.0 resulted in losses caused by binding of bile salt and steroid sulfates to precipitated protein. These losses ranged from 11.1% for glycochenodeoxycholate-7-sulfate to 38.4% for glycolithocholate-3-sulfate (Table 2). These losses were dependent on the amount of protein and therefore varied between different samples (Fig. 1). However, addition of ammonium hydroxide (final concentration of 1 M) to the boiled incubates before centrifugation reduced losses to 3 to 4.5%, which were independent of added protein over the range 50-250 pg (Table 2). Recoveries of bile salt and steroid sulfates were also affected by the nature of the aqueous phase used for backwashing the butanol phase. Water removed 14.7% of glycochenodeoxycholate-7-sulfate from the butanol phase. Ammonium hydroxide (1 M) alone gave even greater losses (up to 26%). It was found that an aqueous mixture of 1 M ammonium hy-

TABLE 2 EFFUX OFALKALINIZATIONONTHELOSSESOFSULFATEDSUBSTRATESINTHEBUTANOL-EXTRACTIONASSAY Percentage loss to precipitated protein Sulfated substrate” GCDC GLC 3/9A5 DHEA Estradiol

Percentage loss in final aqueous phase

No NH,

NH, added

Water

1 M NH,OH

11.1 38.4 31.4 9.83 14.6

2.89 4.4 1 4.11 3.48 3.25

14.7 3.37 3.19 2.18

26.0 4.22 -

N&OH:phosphate 5.21 0.74 3.47 1.42 1.95

Note. A 250-pmol aliquot of 3’S-labeled sulfated substrate was boiled with 250 pg protein in 0.1 ml 100 mht sodium phosphate:5 mM MgSO, buffer, pH 7.0. After cooling, it was diluted with either 0.4 ml water or 0.4 ml 1 M ammonium hydroxide, vortexed vigorously, and centrifuged. The supematant was extracted with butanol(1 ml) and mixed, and the phases were separated. The butanol phase was backwashed with 0.45 ml of either water, 1 M ammonium hydroxide, or 1 M ammonium hydroxide: 100 mM sodium phosphate buffer, pH 7.0 (final pH 10.8). Each result above is the mean of a minimum of three replicates. ’ These compounds are listed in Table 1.

SOLVENT-EXTRACTION

‘. ‘L--*-0 ---.. --*-----. ;:2 ‘O:‘A 5 60 2 A< A ; Lz

40-

20 1 o-l, 0

I 0.05

I 0.1 PROTEIN

1 0.15 ADDED

I 0.2

I 0.25

( MG)

FIG. 1. Effect of added protein on the recovery of estmdiol-3-sulfate (0, 0) and glycolithocholate-3-sulfate (A, A) using butanol extraction from (O-250 pg) sodium phosphate buffer, pH 7.0 (open symbols) and from sodium phosphate buffer, pH 7.0: 1 M NH,OH ( 1:4 v/v), final pH 11.1 (closed symbols).

droxide: 100 mM sodium phosphate buffer, pH 7.0 (4: 1) (final pH 10.8) minimized losses during washing. Optimization of Incubation Added Protein

40 IJG

80 PROTEIN

100

120

of These Assays

Varying amounts of sulfotransferase activity were obtained from the fractionation of female rat liver cytosol on a 90 X 1.6-cm column of Sephacryl S-200 (Pharmacia, Piscataway, N. J.) and from various breast tumor cytosols. Duplicate samples were analyzed by the bu-

Time and

60

tion was determined. With a 2-h incubanon time, lo-70 pg of hepatic cytosolic protein gave a linear increase in radioactivity (Fig. 2). This was observed for all four substrates irrespective of how effective each substrate was for the sulfotransferases. Alter 2 h the amount of sulfate esters formed stabilized; at that time paper electrophoresis of boiled incubates showed that all the 35Slabel had either been transferred to the sulfated substrate or was inorganic sulfate. In time-course experiments using 100 pg hepatic cytosolic protein, there was a linear increase in radioactivity for at least 60 min (Fig. 2). Reproducibility of the assay was excellent for each substrate, the standard deviation ranging from 0.23 to 3.36% of the mean sulfotransferase activity with 25 Kg added hepatic cytosolic protein and from 0.39 to 1.30% with 100 pg protein (Table 3). Application

Linearity of the sulfotransferase activity with incubation time and protein concentra-

20

473

ASSAYS FOR SULFOTRANSFERASES

I 140

45 30 45 60

90 MIN

120

180

FIG. 2. Optimization of sulfotransfemse assay.(A) The rate of sulfate ester formation was examined for a 2-h incubation with varying amounts (lo-140 crp) of female rat hepatic cytosolic protein. (B) The rate of sulfate ester formation was examined over a 3-h period when substrates were incubated with 100 pg hepatic cytosolic protein. All substrates were at 100 PM: glycochenodeoxycholate (W); glycolithocholate (0); 3@-hydroxy-5cholenoate (Cl); and dehydroepiandrosterone (0).

474

BARNES, WALDROP,

AND NEIGHBORS

tanol extraction technique and the TLC method. Excellent correlations were obtained, and the enzyme activities observed with the solvent-extraction method, when corrected for known losses (see Table l), gave 1:l correspondence with the results obtained by the TLC method (see Fig. 3). DISCUSSION

w

A solvent-extraction method for measuring the enzymatic production of bile salt sulfate esters has been previously described by Loof and his collaborators (9,15). They used [35S]PAPS as the radiolabel and glycolithocholate as the bile salt substrate. Enzyme activity was determined by the radioactivity which was extracted into butanol at pH 7.0. This was surprising since it implied that glycolithocholate 3-sulfate was soluble in butanol at a pH (7.0) above the pK, of both the sulfate and carboxyl groups. However, we have confirmed this result and have shown that this type of extraction can be applied to sulfate esters of two other bile salts, namely glycochenodeoxycholate and 3B-hydroxy-5-cholenoate, as well as to two other steroids, estradiol and dehydroepiandrosterone. We found that the butanol-extraction method at pH 7.0 as described by Loof (9,15)

TABLE 3 REPRODUCIBILITY

Substrate

OF THE BUTANOL Protein added bg)

ASSAY

MWII activity (pm01 h-‘)

(Ef ItlfZUl)

Glycochenodeoxycholate

25 100

75.2 300.5

1.48 0.95

Glycolithocholate

25 100

261. I 1133.8

1.70 1.30

3&Hydroxy-kholenoate

25 100

303.6 1293.2

0.23 0.39

Dehydroepiandrosterone

25 100

525.0 1955.3

3.36 0.92

Note. Female rat hepatic cytosol was used in this experiment. The bile salt/steroid substrate concentrations were 100 pM. Each value in the table comes from five replicates.

.

,

,

.

,

,

(

15 30 45 60 75 90 105 120 SULFATION (PMOL/HR) TLC

RATE

METHOD

FIG. 3.Comparison of &radio1 sulfotransferase activities from fractionated rat liver cytosol and from human breast tumor cytosol, measured by thin-layer chromatography (3H-labeled substrate) and butanol solvent extraction ([35S]PAPS).

was sensitive to protein binding, and there were substantial losses of steroid and bile salt sulfates during protein precipitation and extraction. These defects have been overcome by raising the pH to 10.8 by the inclusion of ammonium hydroxide, which prevents losses due to protein binding. In those cases where the protein concentration is very low, it may be unnecessary to precipitate the protein by boiling, but the addition of ammonium hydroxide would be essential to prevent protein binding of the sulfate ester products. Interestingly, the addition of ammonium hydroxide did not prevent losses into the aqueous phase during backwashing of the butanol phase. The concomitant presence of 100 KiM sodium phosphate was necessary; the small difference in pH (11.1 for 1 M ammonium hydroxide, 10.8 for the buffered solution) does not provide an explanation for this. It is possible that there is a salting-out effect from sodium phosphate, or alternately an ionpair complex is formed which is soluble in the butanol. The high partition coefficient for the butanol phase under alkaline conditions has been previously reported for neutral steroid sulfates (16), but not bile salt sulfates. This alkaline butanol-extraction procedure has proved to be a simple, rapid, accurate,

SOLVENT-EXTRACTION

475

ASSAYS FOR SULFOTRANSFERASES

and quantitative procedure for the measurement of cytosolic bile salt and steroid sulfotransferase activities. It has the advantage that the same technique can be used for a variety of substrates and does not require any specialized reagents or equipment. In this laboratory, a technician can carry out 100 assays each day. ACKNOWLEDGMENTS This study was supported by Federal Grant AM-255 11 from the National Institute of Arthritis, Metabolism and Digestive Diseases. The authors wish to acknowledge the constructive criticism of this manuscript given by Dr. Norman Radin and Dr. Jerry G. Spenney.

REFERENCES 1. Cowen, A. E., Korman, M. G., Hofmann, A. F., and Cass, 0. W. (1975) Gastroenterology 69, 59-66. 2. Hadd, H. E., and Blickenstaff, R. T. (1969) in Conjugates of Steroid Hormones pp. 293-318, Academic Press, New York.

3. Chen, L-J., Bolt, R. J., and Admirand, W. H. (1977) Biochim.

Biophys.

Ada

480, 2 19-227.

4. Barnes, S., Burhol, P. G., Zander, R., Haggstrom, G., Settine, R. L., and Hirschowitz, B. I. (1979) J. Lipid

Rex

20, 952-959.

5. Kirkpatrick, R. R., Lack, L., and Kihenberg, P. G. (1980) J. Biol. Chem. 255, 10157-10159. 6. Chen, L-J. (1980) Anal. Biochem. 105, 170-173. 7. Sekura, R. D., Marcus, C. J., Lyon, E. S., and Jakoby, W. B. (1979) Anal. Biochem. 95, 82-86. 8. Singer, S. S., Giera, D., Johnson, J., and Sylvester, S. (1976)

Endocrinology

98, 963-974.

9. Loof, L., and Wengle, B. (1978) Biochim. Acta 530,45 10.

Biophys.

l-460.

Barnes, S., and Spenney, J. G. (1982) Biochim. phys. Acta 704, 353-360. Mumma, R. 0. (1966) Lipids

Bio-

11. 1, 221-223. 12. Barnes, S., Gollan, J. L., and Billing, B. H. (1977) Biochem. J. 166,65-73. 13. Usui, T. (1963) J. Biochem. (Tokyo) 54,283-286. 14. Rudman, D., and Kendall, F. E. (1957) .J.Clin. Invest. 36, 538-542. 15. Loof, L., and Hjerten, S. (1980) B&him. Biophys. Acta 617, 192-204. 16. Jayle, M.-F., and Crepy, 0. (1953) Ann. Endocrinol. 14, 74-77.