Testosterone sulfotransferase: Evidence in the guinea pig that this reaction is carried out by 3α-hydroxysteroid sulfotransferase

Article abstract of DOI:10.1016/S0039-128X(99)00027-6

During the course of isolating, characterizing, and cloning estrogen and 3-hydroxysteroid sulfotransferases from the guinea pig adrenal gland, it was noted that cytosolic preparations from this tissue would also sulfonate testosterone. Therefore, we set out to isolate and clone the enzyme that performs this reaction. Testosterone sulfotransferase (TST) was isolated from the guinea pig adrenal by using the standard procedures of ion exchange, affinity, and high-performance liquid chromatography. When purified, TST was examined by liquid-phase nondenaturing isoelectric focusing, it was found that the TST activity profile completely overlapped with the activity profile of the 3α-hydroxysteroid sulfotransferase (3αHST) isoform, but not the 3β-hydroxysteroid sulfotransferase (3βHST) isoform. This finding was further investigated by overexpressing the cDNAs for 3αHST and 3βHST in Escherichia coli and examining the expressed proteins for TST activity. This experiment confirmed that 3αHST does indeed function as a TST. In addition, 3αHST was also found to sulfonate estradiol but not estrone, a finding that further suggested that 3αHST may function as a general 17β-hydroxysteroid sulfotransferase. Copyright (C) 1999.

Full text of DOI:10.1016/S0039-128X(99)00027-6

Steroids 64 (1999) 510–517
Testosterone sulfotransferase: Evidence in the guinea pig that this
reaction is carried out by 3␣-hydroxysteroid sulfotransferase
Byoung C. Park, Young C. Lee, Charles A. Strott*
Section on Steroid Regulation, Endocrinology and Reproduction Research Branch, National Institute of Child Health and Human Development,
Bldg 49, Rm 6A36, National Institutes of Health, Bethesda, MD 20892-4510, USA
Received 4 November 1998; received in revised form 21 December 1998; accepted 28 January 1999
Abstract
During the course of isolating, characterizing, and cloning estrogen and 3-hydroxysteroid sulfotransferases from the guinea pig adrenal
gland, it was noted that cytosolic preparations from this tissue would also sulfonate testosterone. Therefore, we set out to isolate and clone
the enzyme that performs this reaction. Testosterone sulfotransferase (TST) was isolated from the guinea pig adrenal by using the standard
procedures of ion exchange, affinity, and high-performance liquid chromatography. When purified, TST was examined by liquid-phase
nondenaturing isoelectric focusing, it was found that the TST activity profile completely overlapped with the activity profile of the
3␣-hydroxysteroid sulfotransferase (3␣HST) isoform, but not the 3␤-hydroxysteroid sulfotransferase (3␤HST) isoform. This finding was
further investigated by overexpressing the cDNAs for 3␣HST and 3␤HST in Escherichia coli and examining the expressed proteins for TST
activity. This experiment confirmed that 3␣HST does indeed function as a TST. In addition, 3␣HST was also found to sulfonate estradiol
but not estrone, a finding that further suggested that 3␣HST may function as a general 17␤-hydroxysteroid sulfotransferase. Published by
Elsevier Science Inc.
Keywords: 3␣-/3␤-Hydroxysteroid; 17␤-Hydroxysteroid; Sulfotransferase; Androsterone; Testosterone; Estradiol
1. Introduction
tion of individual steroid sulfotransferases had been sparse,
primarily because of a general lack of adequate enzyme
purification. This situation has now greatly changed as a
result of the development of more powerful tools for iso-
lating and resolving proteins and the strength of recombi-
nant DNA technology. Furthermore, with the availability of
bacterial and yeast overexpression systems, these enzymes
can be produced in large quantities and purified for bio-
chemical characterization and structural analysis.
The sulfonation of steroids has a profound influence on
their biological activity [8]. For this reason, the sulfonation
of estrogens has been extensively studied, particularly in
uterine and mammary target tissues (for review, see Ref.
[3]); specific estrogen sulfotransferases have been cloned
and expressed from cow [9], rat [10], guinea pig [11],
human [12], and mouse [13]. In contrast to the case with
estrogens, the sulfonation of androgens has not received as
much attention, although this process is thought to play a
major role in age-related changes in androgen sensitivity in
the rat liver [14]. The sulfoconjugation of testosterone is
known to occur in several tissues including the testis [15–
18], adrenal [19], and liver [20–23]. Although a specific
The sulfonation or sulfoconjugation of steroids is a major
metabolic route in the biotransformation of these biologi-
cally important compounds [1]. Steroid sulfonation, which
occurs widely across species [2], is catalyzed by a family of
enzymes termed sulfotransferases [3], enzymes typically
isolated from the cytosolic fraction of tissue preparations
[4]. Moreover, the enzymes that sulfoconjugate other rela-
tively small endogenous compounds such as cate-
cholamines, iodothyronines, ascorbate, cholesterol, vitamin
D, and bile acids, as well as drugs and xenobiotics, are also
present in the cell sap [5,6]. This is in contrast to the
subclass of sulfotransferases that catalyzes the sulfonation
of macromolecules, i.e. proteins, proteoglycans, and glycos-
aminoglycans as well as sulfolipids, which are associated
with cell membranes and the Golgi complex [7]. Until
recently, reliable biochemical and structural characteriza-
* Corresponding author. Tel.: ϩ1-301-496-3025; fax: ϩ1-301-496-
7435.
E-mail address: chastro@box-c.nih.gov (C.A. Strott)
0039-128X/99/$ – see front matter Published by Elsevier Science Inc.
PII: S0039-128X(99)00027-6

B.C. Park et al. / Steroids 64 (1999) 510–517
511
testosterone sulfotransferase (TST) has not been isolated or
cloned, hydroxysteroid sulfotransferases (HSTs) have been
cloned and expressed from several species that will sulfo-
conjugate testosterone, e.g. rat [24], rabbit [25], and human
[26,27]. These enzymes appear to have a broad substrate
specificity and will sulfonate a wide range of 3␤- and
3␣-hydroxylated steroid substrates, e.g. 3␤-hydroxy-5-an-
drosten-17-one (dehydroepiandrosterone), 3␤-hydroxy-5-
pregnen-20-one (pregnenolone), 3␣-hydroxy-5␣-androstan-
17-one (androsterone), and 3␣-hydroxy-5␣-pregnan-20-one
(allopregnanolone); in addition, they will also sulfonate
17␤-hydroxy-4-androsten-3-one (testosterone) and the es-
trogens, 3-hydroxy-1,3,5 [10] estratrien-17-one (estrone)
and 1,3,5 [10] estratriene-3, 17␤-diol (estradiol).
During the course of isolating and cloning several steroid
sulfotransferases from the guinea pig adrenal gland [11,28,
29], we noted the presence of TST activity in cytosolic
preparations from this tissue. Therefore, we set out to isolate
and characterize TST from the guinea pig adrenal; this
report describes our experience in doing so. During the
purification scheme, which included ion exchange, 3Ј-phos-
phoadenosine 5Ј-phosphate (PAP) affinity chromatography,
and gel-permeation high-performance liquid chromatogra-
phy (HPLC), as well as during analysis with liquid-phase
nondenaturing isoelectric focusing, it was noted that the
TST activity profile overlapped completely with the activity
profile of 3␣-hydroxysteroid sulfotransferase (3␣HST) but
not the 3␤-hydroxysteroid sulfotransferase (3␤HST) iso-
form. As a result of this finding, we overexpressed the
3␣HST and 3␤HST isoforms in Escherichia coli as fusion
proteins and purified them by affinity column chromatog-
raphy for biochemical analysis. This study confirmed that
the 3␣HST isoform, but not the 3␤HST isoform, was capa-
ble of sulfonating testosterone. Furthermore, 3␣HST was
capable of sulfonating estradiol but not estrone, suggesting
that the 3␣HST isoform may serve as a general 17␤-hy-
droxysteroid sulfotransferase.
New England Biolabs (Beverly, MA, USA). Unless other-
wise stated, all other chemicals and reagents were purchased
from Sigma and used as obtained from the supplier.
2.2. Animals and tissue preparation
Male National Institutes of Health inbred strain 2 guinea
pigs, obtained from the National Cancer Institute (Freder-
ick, MD, USA) were maintained on guinea pig chow with
added greens and water ad libitum in a controlled light/dark
environment at constant temperature and humidity. The
experimental protocol and all animal procedures used in this
investigation were approved by the National Institute of
Child Health and Human Development Animal Care and
Use Committee.
Animals were anesthetized with CO2 gas and decapi-
tated. The adrenal glands were quickly removed and placed
in homogenization buffer (10 mM Tris-HCl, pH 7.5, con-
taining 1 mM EDTA, and 1.5 mM dithiothreitol) on ice. The
adrenals were cleaned of fat and fibrous material, weighed,
and minced. Samples were kept on ice and homogenized in
2 vol (wt/vol) of the homogenization buffer by using a
glass/Teflon apparatus (five strokes). The cytosolic fraction
was prepared by centrifuging homogenates at 105 000 ϫ g
for 90 min at 4°C, after which the floating lipid layer was
carefully suctioned off, and the supernatant decanted.
2.3. Enzyme purification
Adrenal cytosol was applied to a 20 ϫ 1.6-cm column of
DE52 anion-exchange resin (Whatman Labsales, Hillsboro,
OR, USA) equilibrated with 20 mM Tris-HCl, pH 7.4, and
1 mM EDTA (TE). The column was washed extensively
with TE until the 280-nm absorbance returned to baseline.
Adsorbed proteins were eluted with a linear gradient of
NaCl in TE (0–300 mM, 1000 ml total volume) at a flow
rate of 50 ml/h, and 1 ml fractions were collected. Fractions
containing the highest TST specific activity were pooled
and concentrated by Centriprep 30 (Amicon, Beverly, MA,
USA). The concentrate was applied to a 16 ϫ 1.6-cm
PAP-agarose (Sigma) affinity column equilibrated with TE.
After washing with TE, adsorbed proteins were eluted with
10 ml of the 1 mM PAPS solution (see Section 2.1) fol-
lowed by a chase with TE buffer. Fractions containing the
highest TST activity were pooled and concentrated by Cen-
triprep 30. The concentrate was applied to an HPLC Super-
dex 200 HR 10/30 column (Pharmacia, Piscataway, NJ,
USA) equilibrated and run with TE containing 100 mM
NaCl. Chromatography was performed at a flow rate of 1
ml/min, and 1-ml fractions were collected. The most active
TST fractions were pooled and concentrated as described
above and reapplied to the same HPLC column under the
same conditions; this step was then repeated once more.
2. Materials and methods
2.1. Materials
Tritium-labeled steroids were purchased from DuPont/
NEN (Boston, MA, USA). Crystalline steroids (pregnenolo-
ne, dehydroepiandrosterone, androsterone, allopreg-
nanolone, estradiol, estrone, and testosterone) were
obtained from either Steraloids, Inc (Wilton, NH, USA) or
Sigma (St. Louis, MO, USA). Steroid stock solutions
(0.1–10 mM) were prepared in absolute ethanol.
3Ј-Phosphoadenosine 5Ј-phosphosulfate (PAPS), 3Ј-
phosphoadenosine 5Ј-phosphate (PAP), and dithiothreitol
were purchased from Sigma. A PAPS stock solution (1 mM)
was made up in buffer consisting of 100 mM Tris-HCl (pH
7.7) and 5 mM magnesium acetate, divided into aliquots and
stored at Ϫ70°C. Restriction enzymes were obtained from

512
B.C. Park et al. / Steroids 64 (1999) 510–517
2.4. Bacterial overexpression and purification of 3␣HST
and 3␤HST
[31]. For immunochemical analysis, proteins were electro-
phoretically transferred to nitrocellulose membranes
(Schleicher and Schuell, Keene, NH, USA) as previously
detailed [32]. Western blotting and colorimetric visualiza-
tion of immunoreactive proteins were performed exactly as
described previously [33].
The cDNAs for guinea pig 3␣HST [28] and 3␤HST [29]
were amplified and inserted into the pProEx-HTc expres-
sion vector (Life Technologies, Grand Island, NY, USA) at
the SalI--NotI restriction sites, and competent E. coli DH5␣
cells were transformed with these constructs. A his-tag and
a tobacco etch virus protease site are located in front of the
HST coding region. The E. coli cells were placed in Luria–
Bertani (LB) medium and incubated at 30°C for 3 h. After
induction with 1 mM isopropylthiogalactoside (IPTG), in-
cubations were continued for another 4 h at 30°C, after
which the cells were centrifuged at 3000 ϫ g for 10 min at
4°C. Cell pellets were lysed at 4°C with 1 vol (w/v) TE
buffer (pH 7.5) containing 10 mM imidazole and 1 mg/ml
lysozyme. Lysates were centrifuged at 105 000 ϫ g for 1 h
at 4°C, and the supernatants were applied to nickel-affinity
columns (Qiagen, Valencia, CA, USA) and washed with 20
mM imidazole in TE (pH 7.5). Bound HSTs were eluted
with 200 mM imidazole in TE, diluted with excess TE, and
concentrated with Centriprep 30. To remove his-tags, puri-
fied fusion proteins (0.1 mg) were incubated with 500 U of
recombinant tabacco etch virus (Life Technologies) for 16 h
at 4°C, and the preparations were again applied to nickel-
affinity columns for chromatography.
2.7. Protein determination
Protein concentrations were determined by the Bradford
method with bovine serum albumin as a standard [34].
2.8. Sulfotransferase assay
Steroid sulfotransferase assays were performed in a so-
lution composed of 100 mM Tris-HCl (pH 7.7), 5 mM
magnesium acetate, 0.1 mM PAPS, and varying concentra-
3
tions of H-labeled steroids (40 nM to 50 ␮M). The total
reaction volume was generally 100 ␮l. For the routine assay
of column fractions, incubations were performed at room
temperature, whereas kinetic and specific activity analyses
were performed at 37°C. Reactions were stopped by the
addition of 100 ␮l of 0.1 N NaOH, followed by 800 ␮l of
dichloromethane. Unconjugated steroids were extracted by
vigorous Vortex mixing followed by centrifugation at
13 000 rpm for 5 min. The amount of radioactivity in an
aliquot (typically 100 ␮l) of the aqueous phase containing
sulfoconjugated steroid was determined by liquid scintilla-
tion. To correct for extraction efficiency, background blanks
(labeled substrate added after the addition of 0.1 N NaOH to
2.5. Liquid-phase nondenaturing isoelectric focusing
A Rotofor preparative electrofocusing cell (Bio-Rad
Laboratories, Hercules, CA, USA) was used to separate
purified HST proteins by charge in an ampholyte-generated
pH gradient as initially reported by Driscoll et al. [30]. The
aqueous focusing medium consisted of (vol/vol) 25% glyc-
erol, 3% ampholine, pH 5.0 to 7.0 (Pharmacia), and 2%
Bio-Lyte 3/10 (Bio Rad). A purified HST preparation was
thoroughly mixed with 50 ml of ice-cold focusing solution
before loading into the Rotofor chamber. The anode and
cathode electrolytes were 0.1 M H3PO4 and 0.1 M NaOH,
respectively. During the operation, the chamber was cooled
by a recirculating water bath set at 1°C. The focusing was
performed at 12-W constant power for 4 to 6 h. After the
voltage plateaued and remained constant for Ϸ60 min, the
fractions were collected. The pH of each fraction was mea-
sured by using a digital Model 3890 pH meter (Beckman
Instruments), and an aliquot of each fraction was removed
for enzymatic and sodium dodecyl sulfate-polyacrylamide
gel electrophoresis (SDS-PAGE) analyses. The fractions
were made 100 mM in Tris-HCl by adding the appropriate
volume of 1 M Tris-HCl (pH 7.7) and concentrated by
Centriprep 30.
Fig. 1. Sulfotransferase activity profiles after liquid-phase nondenaturing
isoelectric focusing. Adrenal cytosol was prepared, and the testosterone
sulfotransferase activity was fractionated by ion-exchange chromatogra-
phy; the peak activity fractions were pooled, concentrated, and applied to
a Rotofor apparatus as described in Section 2. Sulfotransferase activity in
50 ␮l of the eluted fractions was assayed essentially as described in Section
2, by using testosterone (stippled columns), androsterone (hatched col-
umns), and pregnenolone (solid columns) as substrates; assays were per-
formed at 37°C for 10 min at steroid concentrations of 2 ␮M.
2.6. SDS-PAGE and Western blot analyses
Proteins were resolved in 15% polyacrylamide gels as
previously reported and visualized by Coomassie staining

B.C. Park et al. / Steroids 64 (1999) 510–517
513
Fig. 3. Sulfotransferase activity profiles after liquid-phase nondenaturing
isoelectric focusing of the material eluting in the high-performance liquid
chromatography fraction containing the peak sulfotransferase activities
noted in Fig. 2. Sulfotransferase activity in 50 ␮L of the eluted fractions
was assayed essentially as described in Section 2, by using testosterone
(stippled columns) and androsterone (hatched columns) as substrates; as-
says were performed at 37°C for 10 min at steroid concentrations of 1 ␮M.
Fig. 2. Proteins and sulfotransferase activity in fractions eluting from a
Superdex 200 HR 10/30 high-performance liquid chromatography (HPLC)
column. Testosterone sulfotransferase activity purified by ion-exchange
and 3Ј-phosphoadenosine 5Ј-phosphate (PAP) affinity chromatography
was subjected to HPLC as described in Section 2. HPLC fractions (5 ␮l)
were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophore-
sis and stained with Coomassie blue (top) and assayed for sulfotransferase
activity (bottom) essentially as described in Section 2, by using testoster-
one (solid columns) and androsterone (open columns) as substrates; assays
were performed at 37°C for 10 min at steroid concentrations of 4 ␮M.
Molecular mass markers refer to 3␣-hydroxysteroid sulfotransferase (32
kDa) and 3␤-hydroxysteroid sulfotransferase (33 kDa).
using testosterone, androsterone, and pregnenolone as
substrates (Fig. 1). The highest 3␣HST activity eluted at
a pH of 5.5 to 6.0, whereas 3␤HST activity eluted at a pH
of Ϸ6.5 to 7.0. This result is consistent with our previous
finding on the resolution of these two enzymes by using
this procedure [30]; of interest, however, is that TST
specific activity coeluted with 3␣HST activity and was
clearly separated from 3␤HST activity (Fig. 1).
TST was further isolated by PAP affinity chromatogra-
phy and gel-permeation HPLC. Purified TST activity elut-
ing from the Superdex 200 HR 10/30 HPLC column be-
haved like a protein with an apparent Mr of 130 000,
indicating that, in solution, TST behaves as a multimeric
complex. This behavior is similar to the behavior of 3␣HST
and 3␤HST when analyzed by gel-permeation chromatog-
raphy [30]. When HPLC-purified TST was analyzed by
SDS-PAGE, two protein bands with apparent Mr values of
32 000 and 33 000 were noted [Fig. 2 (top)]. The 32 000 Mr
protein corresponds to 3␣HST, whereas the 33 000 Mr pro-
tein corresponds to 3␤HST [30]. Again, of interest, the TST
activity profile eluting from the HPLC column completely
coincided with the activity profile for 3␣HST [Fig. 2 (bot-
tom)]. The latter preparation was further analyzed by liquid-
phase nondenaturing isoelectric focusing, again revealing
that the two activity profiles completely overlap (Fig. 3).
the enzyme mixture) were routinely processed in parallel
with the complete reactions, and the values obtained were
substracted. The rate of sulfonation was calculated from the
corrected amount of 3H-labeled sulfoconjugated steroid that
partitioned into the aqueous phase. Kinetic data were ana-
lyzed by using the Sigma Plot program (version 4.0) (SPSS,
Chicago, IL, USA).
3. Results
3.1. Isolation of guinea pig adrenal TST activity
Fractions eluting from an anion-exchange column, af-
ter the application of adrenal cytosol, that contained the
highest TST activity were also found to contain high
activities for 3␣HST (androsterone) and 3␤HST (preg-
nenolone). Because we had previously shown that
3␣HST and 3␤HST could be clearly separated by liquid-
phase nondenaturing isoelectric focusing, this procedure
was used to examine the partially purified TST prepara-
tion. Thus, fractions containing the TST peak off the
ion-exchange column were pooled, concentrated, and an
aliquot was applied to the Rotofor apparatus. The eluted
fractions were assayed for sulfotransferase activity by
3.2. Bacterial expression and purification of 3␣HST and
3␤HST
To further investigate the overlapping activity profiles of
TST and 3␣HST, the cDNA clones of 3␣HST [28] and

514
B.C. Park et al. / Steroids 64 (1999) 510–517
Fig. 4. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and
Western blot analyses of 3␣-hydroxysteroid sulfotransferase (lanes 1–4)
and 3␤-hydroxysteroid sulfotransferase (lanes 5–8) overexpressed in E.
coli and purified as described in Section 2. Lanes 1 and 5, crude E. coli
extracts of expressed proteins (15 ␮g); lanes 2 and 6, affinity-purified
expressed proteins (1 ␮g); lanes 3 and 7, affinity-purified expressed pro-
teins after treatment with tobacco etch virus (TEV) protease (1 ␮g); lanes
4 and 8, flow through of affinity column after treatment of expressed
proteins with recombinant TEV protease (1 ␮g). Prestained protein mo-
lecular weight markers are shown in the center of the figure.
Fig. 5. Purified overexpressed 3␣-hydroxysteroid sulfotransferase
(3␣HST) and 3␤-hydroxysteroid sulfotransferase (3␤HST) were analyzed
(0.1 ␮g each) for sulfotransferase specificity by using androsterone (o),
allopregnanolone (Œ), pregnenolone (F), and dehydroepiandrosterone (f)
as substrates (1 ␮M) essentially as described in Section 2.
3␤HST [29] were overexpressed in E. coli as fusion pro-
teins. The expressed proteins were nickle column-purified
before and after removal of the his-tags with recombinant
tabacco etch virus protease digestion. The purified proteins
were subjected to SDS-PAGE and Western analysis by
using an anti-guinea pig 3␣/3␤HST polyclonal antibody
preparation [30,35]. Based on immunoreactivity, the West-
ern analysis clearly demonstrated expression of the 3␣HST
and 3␤HST proteins (Fig. 4).
(Fig. 6). The positional specificity of estradiol sulfonation
was essentially established by the failure of 3␣HST to
sulfonate estrone (Fig. 6). The two enzyme preparations
were further analyzed kinetically with the same substrates
(Table 1). From using 3␣HST, the Vmax for allopreg-
nanolone and androsterone was shown to be 10 to 30 times
greater than that for estradiol and testosterone, whereas the
Km values for all four steroids were more or less similar,
with estradiol having the lowest Km of the group. When the
V
max/Km ratios were examined for 3␣HST and 3␤HST,
3.3. Steroid specificity of the bacterially overexpressed
and purified 3␣HST and 3␤HST
androsterone was used by 3␣HST Ϸ25 times more effi-
ciently than either estradiol or testosterone, and preg-
nenolone was used by 3␤HST about five times more effi-
ciently than dehydroepiandrosterone (Table 1).
Overexpressed and purified 3␣HST and 3␤HST (com-
pare Fig. 4, respectively, lanes 4 and 8) were assayed by
using androsterone, allopregnanolone, pregnenolone, and
dehydroepiandrosterone as substrates (Fig. 5). Consistent
with our previous findings, 3␣HST used only 3␣-hydroxy-
lated steroids as substrates, whereas 3␤HST primarily used
3␤-hydroxylated substrates. When the overexpressed and
purified enzyme preparations were assayed by using testos-
terone as a substrate, the 3␣HST isoform but not the 3␤HST
isoform exhibited TST activity (Fig. 6), confirming the
previous finding with partially purified TST activity (com-
pare with Fig. 3). We were curious as to whether 3␣HST
would also sulfonate the 17␤-hydroxy group of estradiol.
This was thus examined and found to indeed be the case
4. Discussion
Originally, the finding that testosterone was sulfonated
by guinea pig adrenocortical cytosol led us to believe that a
specific sulfotransferase acting on testosterone was present
in this tissue. It was not surprising, however, that during
purification it was difficult to separate the TST activity from
other neutral steroid sulfotransferases, because these en-
zymes are similar in size and will bind to a PAP affinity
column. Nevertheless, it was thought that it might be pos-

B.C. Park et al. / Steroids 64 (1999) 510–517
515
charge. If this was true, then isolating TST would indeed be
a daunting undertaking. Therefore, to determine whether
this was the case, it was decided to overexpress 3␣HST and
3␤HST and examine the purified enzymes for TST activity.
By doing so, it is now clear that 3␣HST can function as a
TST. This finding suggests that, in the guinea pig, a specific
sulfotransferase acting on testosterone may not exist, al-
though it does not necessarily rule the possibility out. Even
more interesting was the finding that 3␣HST will not only
sulfonate testosterone, it will also sulfonate estradiol. It is
noteworthy that the fact that 3␣HST would not sulfonate
estrone indicated that estradiol was sulfonated only at the
17␤-hydroxy position. Thus, these findings are interpreted
to mean that 3␣HST can function as general 17␤-hydroxy-
steroid sulfotransferase. This conclusion must, however, be
considered tentative, because 17␣-hydroxylated steroids,
particularly epitestosterone and epiestradiol, have not as yet
been examined as substrates; therefore, stereospecificity for
the 17␤-hydroxy configuration remains to be fully estab-
lished.
It is known that estradiol can be sulfonated at both the
3-phenol and 17␤-hydroxy positions [1]. Cloned guinea pig
estrogen sulfotransferase, however, will only sulfonate es-
tradiol at the 3-phenol position and will not transfer a
sulfonate group to the 17␤-hydroxy position of estradiol
[36]. It now appears that estradiol-17␤-yl sulfonate, at least
in the guinea pig, may be formed by 3␣HST. That 3␣HST
but not 3␤HST will sulfonate steroids at the 17␤-hydroxy
position, although intriguing, is not easily understood (the
two enzymes are 87% identical [29]). Nevertheless, there is
precedent for this unexpected behavior of 3␣HST. For in-
stance, the steroid metabolizing enzyme 3␣-hydroxysteroid
dehydrogenase (3␣HSD), type 2, which has been cloned
and characterized, has been reported to also function as a
17␤-hydroxysteroid dehydrogenase (17␤HSD) [37]. As
commented, the ability of 3␣HSD to be positionally selec-
tive and stereospecific on the one hand and indiscriminate
on the other invokes the subject of steroid pocket architec-
ture [37]. It was suggested that, for 3␣/17␤HSD, steroid
substrates would have to bind backward to maintain stereo-
chemistry of hydride transfer; ie, the D-ring would enter the
active site first instead of the A-ring and be upside down
with the ␤-face of the steroid in the ␣-face orientation [37].
Conceivably, 3␣HST can be thought to function in a similar
manner. Furthermore, this suggests that the 3␤HST isoform
can function in a comparable manner, ie, the 3␤HST iso-
form will sulfonate 17␣-hydroxysteroids (this provocative
idea is being pursued).
Fig. 6. Purified overexpressed 3␣-hydroxysteroid sulfotransferase
(3␣HST) and 3␤-hydroxysteroid sulfotransferase (3␤HST) were analyzed
(0.1 ␮g each) for sulfotransferase specificity, by using estradiol (*), estrone
(ƒ), and testosterone (छ) as substrates (0.5 ␮M), essentially as described
in Section 2.
sible to resolve TST from 3␣HST and 3␤HST by charge
with the Rotofor apparatus, because this procedure so
clearly separates 3␣HST and 3␤HST, enzymes that are
otherwise quite similar in size and rather difficult to sepa-
rate. At first, it was assumed that the complete overlap in
activities for TST and 3␣HST probably meant that these
two enzymes were not only similar in size but also in
Table 1
Kinetic analysis of overexpressed and purified 3␣- and 3␤-
hydroxysteroid sulfotransferasea
Preg DHEA Allopreg Andro Testo Estrad
3␣HST
Vmax (fmol/min) ND
ND
ND
ND
ND
390
190
2.1
800
260
3.1
27
18
Km (nM)
max/Km
ND
ND
ND
200
0.13
4.4
125
0.14
4.7
V
Rel.%
68.2
100
3␤HST
Vmax (fmol/min) 310 212
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
The finding that 3␣HST also functions as TST in the
guinea pig suggests the idea of tissue expression of 3␣HST
and 3␤HST. To investigate this issue, cytosol was prepared
from multiple guinea pig tissues and assayed for 3␣HST-
and 3␤HST-specific activities (Fig. 7). In the adrenal,
3␣HST specific activity was about six times that of 3␤HST
specific activity. In contrast, the 3␤HST-specific activity
clearly exceeded that for 3␣HST in liver (by Ϸ75%) and
Km (nM)
max/Km
Rel.%
250 788
1.24 0.27
100 21.8
V
a Preg, pregnenolone; DHEA, dehydroepiandrosterone; Allopreg, allo-
pregnanolone; Andro, androsterone; Testo, testosterone; Estrad, estradiol;
3␣HST, 3␣-hydroxysteroid sulfotransferase; 3␤HST, 3␤-hydroxysteroid
sulfotransferase; ND, not detectable; Rel.%, percentage of highest Vmax
Km.
/

516
B.C. Park et al. / Steroids 64 (1999) 510–517
[12] Aksoy IA, Wood TC, Weinshilboum R. Human liver estrogen sulfo-
transferase: identification by cDNA cloning and expression. Biochem
Biophys Res Commun 1994;200:1621–9.
[13] Song W-C, Moore R, McLachlan JA, Negishi M. Molecular charac-
terization of a testis-specific estrogen eulfotransferase and aberrant
liver expression on obese and diabetogenic C57BL/KsJ-db/db mice.
Endocrinology 1995;136:2477–84.
[14] Chatterjee B, Roy AK. Changes in hepatic androgen sensitivity and
gene expression during aging. J Steroid Biochem Mol Biol 1990;37:
437–45.
[15] Laatikainen T, Laitinen EA, Vihko R. Secretion of neutral steroid
sulfates by the human testis. J Clin Endocrinol 1969;29:219–24.
[16] Laatikainen T, Laitinen EA, Vihko R. Secretion of free and sulphate
conjugated neutral steroids by the human testis: effect of administra-
tion of human chorionic gonadotropin. J Clin Endocrinol 1971;32:
59–64.
[17] Ruokonen A, Laatikainen T, Laitinen EA, Vihko R. Free and sulfate-
conjugated neutral steroids in human testis tissue. Biochemistry 1972;
11:1411–6.
[18] Vihko R, Ruokonen A. Steroid sulphates in human adult testicular
steroid synthesis. J Steroid Biochem 1975;6:353–6.
Fig. 7. Sulfotransferase activities in the cytosol of various guinea pig
tissues. Cytosol was prepared and sulfotransferase activity was assayed
(0.2 mg) by using testosterone (2 ␮M, stippled columns), androsterone (4
␮M, hatched columns), and pregnenolone (8 ␮M, solid columns) as sub-
strates, essentially as described in Section 2.
[19] Jaffe RB, Payne AH. Gonadal steroid sulfates and sulfatase. IV.
Comparative studies on steroid sulfokinase in the human fetal testis
and adrenal. J Clin Endocrinol Metab 1971;33:592–6.
[20] Payne AH. Testicular steroid sulfotransferases: comparison to liver
and adrenal steroid sulfotransferases of the mature rat. Endocrinology
1980;106:1365–70.
[21] Lyon ES, Jakoby WB. The identity of alcohol sulfotransferases with
hydroxysteroid sulfotransferases. Arch Biochem Biophys 1980;202:
474–81.
small intestine (by about fivefold); there was a small but
detectable amount of 3␣HST activity in the prostate. In
several other tissues, including the testis, 3␣HST and
3␤HST activities were essentially undetectable with this
procedure (Fig. 7).
[22] Marcus CJ, Sekura RD, Jakoby WB. A hydroxysteroid sulfotrans-
ferase from rat liver. Anal Biochem 1980;107:296–304.
[23] Falany CN, Vazquez ME, Kalb JM. Purification and characterization
of human dehydroepiandrosterone sulphotransferase. Biochem J
1989;260:641–6.
References
[1] Bernstein S, Solomon S. Chemical and biological aspects of steroid
conjugation. New York: Springer-Verlag, 1970.
[24] Ogura K, Satsukawa M, Okuda H, Akira H, Watabe T. Major hy-
droxysteroid sulfotransferase STa in rat liver cytosol may consist of
two microheterogeneous subunits. Chem Biol Interact 1994;92:129–
44.
[25] Yoshinari K, Nagata K, Shiraga T, Iwasaki K, Hata T, Ogino K, Ueda
R, Fujita K, Shimada M, Yamazoe Y. Molecular cloning, expression,
and enzymatic characterization of rabbit hydroxysteroid sulfotrans-
ferase AST-RB2 (ST2A8). J Biochem 1998;123:740–6.
[26] Falany CN, Wheeler J, Oh TS, Falany JL. Steroid sulfation by
expressed human cytosolic sulfotransferases. J Steroid Biochem Mol
Biol 1994;48:369–75.
[27] Falany CN, Comer KA, Dooley TP, Glatt H. Human dehydroepi-
androsterone sulfotransferase. Ann NY Acad Sci 1995;774:59–72.
[28] Lee YC, Park C-S, Strott CA. Molecular cloning of a chiral-specific
3␣- hydroxysteroid sulfotransferase. J Biol Chem 1994;269:15838–
45.
[29] Luu NX, Driscoll WJ, Strott CA. Molecular cloning and expression of
a guinea pig 3-hydroxysteroid sulfotransferase distinct from chiral-
specific 3␤- hydroxysteroid sulfotransferase. Biochem Biophys Res
Commun 1995;217:1078–86.
[30] Driscoll WJ, Martin BM, Chen H-C, Strott CA. Isolation of two
distinct 3- hydroxysteroid sulfotransferases from the guinea pig ad-
renal. J Biol Chem 1993;268:23496–503.
[31] Laemmli UK. Cleavage of structural proteins during the assembly of
the head of bacteriophage T4. Nature (London) 1970;227:680–5.
[32] Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of
proteins from polyacrylamide gels to nitrocellulose sheets: proce-
dure and some applications. Proc Natl Acad Sci USA 1979;76:
4350–4.
[2] Hobkirk R. Steroid sulfotransferases and steroid sulfatases: charac-
teristics and biological roles. Can J Biochem Cell Biol 1985;63:1127–
44.
[3] Strott CA. Steroid sulfotransferases. Endocr Rev 1996;17:670–97.
[4] Roy AB. Enzymological aspects of steroid conjugation. In: Bernstein
S, Solomon S, editors. Chemical and biological aspects of steroid
conjugation. New York: Springer-Verlag, 1970. pp. 74–130.
[5] Roy AB. Sulfotransferases. In: Mulder GJ, editor. Sulfation of
drugs and related compounds. Boca Raton: CRC Press, 1981. pp.
83–130.
[6] Mulder GJ, Jakoby WB. Sulfation. In: Mulder GJ, editor. Conjugation
reactions in drug metabolism: an integrated approach. New York:
Taylor and Francis, 1990. pp. 107–61.
[7] Huxtable RJ. Biochemistry of sulfur. New York: Plenum, 1986.
[8] Roy AK. Regulation of steroid hormone action in target cells by
specific hormone-inactivating enzymes. Proc Soc Exp Biol Med
1992;199:265–72.
[9] Nash AR, Glenn WK, Moore SS, Kerr J, Thompson AR, Thompson
EOP. Oestrogen sulfotransferase: molecular cloning and sequencing
of cDNA for bovine placental enzyme. Aust J Biol Sci 1988;41:507–
16.
[10] Demyan WF, Song CS, Kim DS, Her S, Gallwitz W, Rao TR,
Slomczynska M, Chatterjee B, Roy AK. Estrogen sulfotransferase of
the rat liver: complementary DNA cloning and age- and sex-specific
regulation of messenger RNA. Mol Endocrinol 1992;6:589–97.
[11] Oeda T, Lee YC, Driscoll WJ, Strott CA. Molecular cloning and
expression of a full-length complementary DNA encoding the guinea
pig adrenocortical estrogen sulfotransferase. Mol Endocrinol 1992;6:
1216–26.

B.C. Park et al. / Steroids 64 (1999) 510–517
517
[33] Lee YC, Driscoll WJ, Strott CA. Charge isoforms of the adrenocortical
pregnenolone-binding protein: influence of phosphorylation on isofor-
mation and binding activity. Proc Natl Acad Sci USA 1990;87:2003–7.
[34] Bradford M. A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein-dye
binding. Anal Biochem 1976;72:248–54.
[35] Whitnall MH, Driscoll WJ, Lee YC, Strott CA. Estrogen and hydroxy-
steroid sulfotransferases in guinea pig adrenal cortex: cellular and
subcellular distributions. Endocrinology 1993;133:2284–91.
[36] Tomizuka T, Oeda T, Tamura Y, Yoshida S, Strott CA. Character-
ization of guinea pig estrogen sulfotransferase expressed by Chinese
Hamster Ovary cell-K1 stable transfectants. Endocrinology 1994;
135:938–43.
[37] Lin H-K, Jez JM, Schlegel BP, Peehl DM, Pachter JA, Penning TM.
Expression and characterization of recombinant type 2 3␣-hydroxy-
steroid dehydrogenase (HSD) from human prostate: demonstration of
bifunctional 3␣/17␤-HSD activity and cellular distribution. Mol En-
docrinol 1997;11:1971–84.