Department of Molecular Pharmacology and Experimental Therapeutics,
Mayo Medical School-Mayo Clinic-Mayo Foundation, Rochester,
Minnesota
Sulfation is a major pathway in the biotransformation of many drugs
and other xenobiotic compounds. The sulfotransferase (SULT) enzymes
that catalyze these reactions use 3'-phosphoadenosine 5'-phosphosulfate
(PAPS) as a sulfate donor cosubstrate. The synthesis of PAPS from
inorganic sulfate and ATP is catalyzed by PAPS synthetase (PAPSS). We
previously cloned the genes for human PAPSS1 and PAPSS2 as a step
toward pharmacogenetic studies of these enzymes. We have now
developed a sensitive PAPSS radiochemical enzymatic assay for use in
genotype-phenotype correlation analyses. This coupled assay uses the
sulfation of 17
-[3H]estradiol catalyzed by
recombinant human SULT1E1 to measure PAPS, which has been generated by
PAPSS during the initial step of the assay. SULT1E1 proved to be ideal
for this application both because of its relative resistance to
inhibition by ATP, a substrate for the PAPSS-catalyzed step, and
because of its low Km values for both PAPS
(58 nM) and estradiol (29 nM). After optimal PAPSS assay conditions had
been established, substrate kinetic studies were performed with cytosol
preparations from human liver and cerebral cortex, two tissues with
very different expression patterns for PAPSS1 and PAPSS2 mRNA. Brain
and liver cytosol PAPSS activities had apparent
Km values for ATP of 0.26 and 0.62 mM, respectively, and for SO42
of 0.08 and 0.31 mM, respectively. PAPSS activity was then measured in 83 human liver
biopsy samples to determine the nature and extent of individual
variation in this enzyme activity. An 18-fold variation was observed.
This sensitive new radiochemical assay can now be used in
pharmacogenetic studies of PAPSS in humans.
 |
Introduction |
Sulfate conjugation is an important pathway in the metabolism of
many drugs, other xenobiotics, neurotransmitters, and hormones (Weinshilboum and Otterness, 1994
; Falany, 1997).
PAPS1
is the high-energy sulfate donor cosubstrate for the sulfotransferase (SULT) enzymes that catalyze these reactions (Weinshilboum and Otterness, 1994; Klaassen and Boles, 1997). In prokaryotes, PAPS is
synthesized from 2 moles of ATP and 1 mole of
SO42 by two separate enzymes,
ATP sulfurylase and adenosine 5'-phosphate kinase (Farooqui,
1980; Klaassen and Boles, 1997). However, in higher organisms these
reactions are catalyzed by a single bifunctional cytosolic enzyme, PAPS
synthetase (PAPSS) (Fig. 1, step A)
(Geller et al., 1987; Lyle et al., 1994). Two isoforms of PAPSS have
been identified in both humans and mice (Li et al., 1995; Girard et al., 1998; Kurima et al., 1998; ul Haque et al., 1998; Venkatachalam et
al., 1998). We recently cloned the genes for the two known PAPSS
isoforms in humans, PAPSS1 and PAPSS2, as a step
toward pharmacogenetic studies (Xu et al., 2000). Each of these genes consisted of 12 exons, with virtually identical exon-intron splice junction locations (Kurima et al., 1999; Xu et al., 2000). However, Northern blot analysis demonstrated different patterns of tissue expression. For example, PAPSS1 mRNA was highly expressed in human brain, whereas PAPSS2 mRNA was not, but the opposite was true of human
liver (Xu et al., 2000). Finally, although both isoforms have usually
been considered cytosolic enzymes, there has been a recent report that
PAPSS1 can be nuclear in its subcellular localization as can PAPSS2
when coexpressed with PAPSS1 (Besset et al., 2000).
Rare mutations of PAPSS2 that result in inactive enzyme have
been associated with congenital skeletal disorders in both humans and
mice (Kurima et al., 1998; ul Haque et al., 1998). However, no common
polymorphisms that alter the biochemical properties or levels of
activity of these enzymes have been described in humans. As the next
step toward pharmacogenetic studies of human PAPSS1 and PAPSS2, we have
developed a sensitive radiochemical enzymatic assay suitable for use in
genotype-phenotype correlation analyses. This assay uses the sulfation
of [3H]E2 by SULT1E1 (Fig. 1, step B) to
measure PAPS formed from ATP and inorganic sulfate by PAPSS. Although
assays based on a similar principle have been described previously
(Hazelton et al., 1985; Vargas, 1988; Wong et al., 1990), we found
their sensitivity inadequate for many purposes. The increased
sensitivity of our new assay resulted from the relative resistance of
human SULT1E1 to inhibition by ATP, the low
Km values of SULT1E1 for both E2 and PAPS,
and the commercial availability of high specific activity radioactively labeled E2. We have used this new assay both to determine selected biochemical properties of PAPSS in human tissue preparations and to
measure PAPSS activity in 83 human hepatic biopsy samples to study the
nature and extent of individual variation of this activity in an
important drug-metabolizing organ, the liver.
| |
Materials and Methods |
Chemicals and Reagents.
4-Nitrophenol, dopamine, E2, dehydroepiandrosterone (DHEA), ATP,
Na2SO4, PAPS (80% purity),
dithiothreitol (DTT), bovine serum albumin, and
NaClO3 were purchased from Sigma Chemical Co. (St Louis, MO). Purified PAPS (>95% purity) was obtained from Dr. S. Singer, Dayton University, Dayton, OH.
[35S]PAPS (2.52 Ci/mmol) and
[2,4,6,7-3H]E2 (72 Ci/mmol) were purchased from
New England Nuclear (Boston, MA).
Tissue Acquisition and Preparation.
Human liver (N = 83, 43 samples from men and 40 from
women) and temporal lobe cerebral cortical (N = 6, two
samples from men and four from women) surgical biopsy samples were
obtained from the operating rooms at St. Mary's and Rochester
Methodist hospitals in Rochester, MN. These tissue samples were removed
in the course of clinically indicated surgical procedures under
guidelines reviewed and approved by the Mayo Clinic Institutional
Review Board. Portions of liver or cerebral cortical "waste tissue"
distant from regions affected by disease and that appeared visually
normal were stored at 
80°C. Before enzyme assay, frozen tissue was
weighed and homogenized in 5 mM potassium phosphate buffer, pH 7.5. The
homogenates were centrifuged at 15,000g for 15 min, and the
resulting supernatant preparations were centrifuged at
100,000g for 1 h. All steps were performed at 4°C.
The final 100,000g "high-speed supernatant" (HSS)
preparations were stored at 80°C. Pooled samples that included equal volumes of HSS from 10 liver or six brain biopsy samples were
used to determine optimal conditions for the enzyme assay as well as
the biochemical properties of PAPSS activity in these two tissues.
Recombinant SULTs.
cDNAs for SULT1A1 (GenBank accession no. L19999), SULT1A3 (GenBank
accession no. U08032), SULT1E1 (GenBank accession no. U08098), and
SULT2A1 (GenBank accession no. U08024) were cloned into the mammalian
expression vectors p91023(B) or pCR3.1. Each of these expression
constructs was then used to transfect COS-1 cells with the DEAE-dextran
method (Luthman and Magnusson, 1983). After the transfected cells were
harvested, cell pellets were washed with 5 ml of phosphate-buffered
saline and were homogenized for 30 s in 2 ml of 5 mM potassium
phosphate buffer, pH 6.5. The homogenates were centrifuged for 15 min
at 15,000g at 4°C, and supernatants from that step were
centrifuged at 100,000g for 1 h at 4°C. The final HSS
preparations were used as a source of recombinant human SULTs.
SULT Assays.
The activities of recombinant human SULT1A1, SULT1A3, SULT1E1, and
SULT2A1 were determined with the method of Foldes and Meek (1973)
modified to measure the activities of each of these isoforms under
optimal conditions with isoform-specific "prototypic" substrates (Anderson and Weinshilboum, 1980; Campbell et al., 1987; Sundaram et
al., 1989; Otterness et al., 1992; Aksoy et al., 1994). All of these
SULT assays used 0.4 µM [35S]PAPS as a
sulfate donor cosubstrate. Blanks were samples that did not contain a
sulfate acceptor substrate. Reactions were terminated by the
precipitation of unreacted PAPS with Ba(OH)2. The
sulfate acceptor substrates used in these reactions were 4 µM
4-nitrophenol for SULT1A1, 60 µM dopamine for SULT1A3, 0.1 µM E2
for SULT1E1, and 5 µM DHEA for SULT2A1. These assays were used to
study the inhibition of recombinant human SULTs by ATP.
A different assay was used to measure apparent
Km values for SULT1E1. This assay used
radioactively labeled E2, the sulfate acceptor cosubstrate, rather than
radioactively labeled PAPS (Falany et al., 1995). Unconjugated E2 was
then removed by organic solvent extraction, as described subsequently,
and the radioactivity of the sulfate-conjugated steroid that remained
in the aqueous phase was measured.
Protein Assay.
Protein concentrations were measured by the dye-binding method of
Bradford (1976) with bovine serum albumin as a standard.
PAPSS Assay.
The first step in the PAPSS assay involved the generation of PAPS from
ATP and Na2SO4 (Fig. 1,
step A), followed by measurement of the PAPS that had been formed by
using it as a substrate for the sulfate conjugation of
[3H]E2 in a reaction catalyzed by recombinant
human SULT1E1 (Fig. 1, step B). Specifically, PAPS was generated in the
presence of 1 mM ATP, 4 mM
Na2SO4, 1 mM
MgCl2, and 2 mM DTT dissolved in 60 mM
glycine-NaOH buffer, pH 8.6. The reaction was initiated by the addition
to a 100-µl volume that contained the other reagents of a 50-µl
enzyme source that contained either 50 or 100 µg of tissue cytosol
protein. Blanks were samples that contained 50 µl of tissue cytosol
that had been heated at 100°C for 5 min. This final 150-µl reaction
mixture was incubated at 37°C for 20 min, and the PAPSS reaction was
terminated by heating at 100°C for 1 min. Subsequently, 150 µl of
50 mM potassium phosphate buffer, pH 5.5, was added to lower the pH to
7.0. This mixture was then centrifuged for 5 min at 16,000g
in an Eppendorf model 5415C desktop centrifuge (Brinkmann Instruments,
Westbury, NY) to precipitate protein. One hundred microliters of the
supernatant after centrifugation was added to 400 µl of 5 mM
potassium phosphate buffer, pH 6.5, to adjust the pH to that required
for the second stage reaction. A 50-µl aliquot of this 500-µl
volume was used as a source of PAPS during the second step in the assay.
That second step (Fig. 1, step B) involved a 20-min incubation at
37°C in a final volume of 160 µl that contained 10 mM potassium phosphate buffer, pH 6.5, 27 nM [3H]E2 (the
radioactively labeled sulfate acceptor substrate), 8 mM DTT, and 1.25 mM MgCl2. The 160-µl volume included 50 µl of recombinant human SULT1E1. The SULT1E1-catalyzed sulfate conjugation reaction was terminated by the addition of 1 ml of 10 mM KOH, and
organic solvent extraction was performed twice with 3 ml of chloroform.
Five hundred and fifty microliters of the aqueous phase remaining after
the two chloroform "washes" was aspirated and mixed with 5 ml of
Biosafe II to make it possible to measure the radioactivity of the
sulfate-conjugated [3H]E2 formed during the
second stage reaction. To quantitate PAPS, a series of known
concentrations that ranged from 7.8 to 62.5 nM, a range chosen on the
basis of PAPSS activity present in hepatic tissue samples, was
incubated under the same conditions as those present in the second step
of the assay. This standard curve was used to determine the
concentration of PAPS formed during the first step of the assay. In
addition, a single pooled human liver cytosol preparation was included
with each set of assays as a "positive control". We observed less
than a 5% day-to-day variation in PAPSS activity measured in this
control sample.
Data Analysis.
Apparent Km values were calculated with the
method of Wilkinson (1961) using a computer program written by Cleland
(1963). IC50 values for the inhibition of enzyme
activity were estimated from least-squares best fit polynomial
equations by using the GraphPad Inplot program, version 3.1 (GraphPad
Software Inc., San Diego, CA). Statistical analyses involved the use of
unpaired Student's t test, the Mann-Whitney U
test, and correlation analysis performed with the StatView program,
version 4.5 (Abacus Concepts, Berkeley, CA).
| |
Results |
These studies involved three sets of experiments. The first set
was designed to determine whether it would be possible to identify a
recombinant human SULT with optimal properties for use as a reagent in
a sensitive coupled radiochemical PAPSS assay. Once such a SULT had
been identified, assay conditions were optimized with human tissue
preparations as an enzyme source, and the assay was used to determine
selected biochemical properties of PAPSS activity in two human tissues
with differing patterns of PAPSS isoform mRNA expression
liver and
brain. Tissue preparations were used to perform these experiments
because tissue, not purified enzyme, will be used to perform future
genotype-phenotype correlation analyses. The third and final set of
experiments involved a determination of the nature and extent of
individual variation in level of PAPSS activity in an important human
drug-metabolizing organ, the liver.
Recombinant Human SULT Selection.
The strategy used in the final coupled assay is depicted schematically
in Fig. 1. For this strategy to succeed required the identification of
a SULT with optimal properties for use in the PAPS measurement step
(Fig. 1, step B). All SULTs in higher organisms use PAPS as a
cosubstrate (Klaassen and Boles, 1997). Therefore, any SULT-catalyzed
reaction could potentially be used for this purpose, and similar
coupled assays have been described previously, usually using cytosol
from organs that predominantly express phenol SULTs such as SULT1A1
(Hazelton et al., 1985; Vargas, 1988; Wong et al., 1990; Wong and Wong,
1994). Unfortunately, the sensitivity of previous coupled PAPSS assays
was limited. That was true in part because ATP, one of the cosubstrates
for PAPSS (Fig. 1, step A), inhibits all SULTs (Rens-Domiano and Roth,
1987). Therefore, we set out to identify a SULT that was both
relatively resistant to inhibition by ATP and that also had high
affinity for its substratesPAPS and the sulfate acceptor cosubstrate.
Low Km values for substrates would mean
that it would be possible to add only a small aliquot of the
PAPS-generating reaction to the second, coupled reaction. As a result,
the inhibitory effect of ATP could be further minimized. As described
subsequently, human SULT1E1 proved to have ideal properties for use in
a coupled PAPSS assay.
We began by comparing the inhibition of four different recombinant
human SULTs by ATP (Fig. 2). Human SULTs
belong to two large families: phenol SULTs and hydroxysteroid SULTs
(Weinshilboum et al., 1997). We selected three well characterized
members of the phenol SULT family: SULT1A1, SULT1A3, and SULT1E1, and
one hydroxysteroid SULT, SULT2A1, for study. The activities of these four recombinant human SULT isoforms were then assayed using prototypic substrates for each isoform in the presence of a series of
concentrations of ATP that ranged from 4 to 8000 µM (Fig. 2).
IC50 values for the inhibition of SULT1A1,
SULT1A3, SULT1E1, and SULT2A1 by ATP were 36, 205, 1300, and 2056 µM,
respectively. These results demonstrated that human SULT1E1 and SULT2A1
were much more resistant to ATP inhibition than were SULT1A1 and
SULT1A3. Therefore, SULT1E1 and SULT2A1, rather than the phenol SULTs
that have been used in previous coupled PAPSS assays (Hazelton et al.,
1985; Vargas, 1988; Wong et al., 1990; Wong and Wong, 1994), appeared
to be superior for this application. However, because a sensitive assay
for SULT1E1 had been described that was based on the use of
commercially available [3H]E2 as a sulfate
acceptor substrate (Falany et al., 1995), we began by evaluating
SULT1E1.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 2.
Inhibition of recombinant human SULT
isoforms by ATP.
The activities of recombinant human SULT1A1, 1A3, 1E1, and 2A1 were
measured under optimal assay conditions in the absence of ATP and in
the presence of a series of concentrations of ATP using the prototypic
substrates 4 µM 4-nitrophenol (SULT1A1), 60 µM dopamine (SULT1A3),
0.1 µM E2 (SULT1E1), and 5 µM DHEA (SULT2A1). Activities shown are
expressed as a percentage of control values.
|
|
The next step in the evaluation involved a determination of apparent
Km values of SULT1E1 for PAPS and E2.
Concentrations of [3H]E2 (18 Ci/mmol) that
ranged from 0.8 to 100 nM were used to determine the apparent
Km value of SULT1E1 for E2 in the presence of 0.4 µM PAPS. Figure 3A is a double
inverse plot of those data that demonstrates that SULT1E1, like most
human SULTs (Weinshilboum and Otterness, 1994), displays substrate
inhibition. The apparent Km value was 29 nM. A series of PAPS concentrations that ranged from 8 to 500 nM was
then tested in the presence of 27 nM [3H]E2,
and an apparent Km value for PAPS of 58 nM
was calculated. The high affinity of SULT1E1 for both E2 and PAPS, when
combined with the availability of high specific activity radioactively labeled E2 and the relative resistance of SULT1E1 to inhibition by ATP,
resulted in our selection of SULT1E1 for use in the second step of the
assay.
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 3.
Human SULT1E1 substrate kinetics.
SULT1E1 catalyzed sulfate conjugation of E2. Double reciprocal plots of
data obtained with E2 (A) or PAPS (B) as the varied substrate are
shown. Eight concentrations of E2 that varied from 0.8 to 100 nM were
used, as were seven concentrations of PAPS that ranged from 8 to 500 mM. See text for details.
|
|
PAPSS Assay Optimization and Enzyme Characterization.
The next series of experiments was designed to determine optimal
conditions for the coupled PAPSS assay and to use the assay to study
selected biochemical characteristics of PAPSS activity in human liver
and cerebral cortical preparations, tissue preparations like those that
will be used in future genotype-phenotype correlation analyses. When
the coupled assay was performed with both tissues as an enzyme source,
the reaction was linear for at least 20 min (data not shown). The
formation of PAPS by liver and cerebral cortex preparations was also
linear with regard to quantity of cytosol protein over a range from
12.5 to 100 µg of protein (data not shown). Finally, the optimal pH
for PAPS formation was approximately 8.6 with both cytosol preparations
as enzyme sources (data not shown). Therefore, either 50 or 100 µg of
cytosol protein, a 20-min incubation time, and a pH of 8.6 were the
assay conditions used to study selected biochemical properties of PAPSS
in human liver and brain cytosol. In addition, during each assay a
standard curve that used final concentrations of PAPS that ranged from
7.8 to 62.5 nM was also measured in the second stage reaction mixture. This PAPS was dissolved in a mixture that included all components of
the first-step assay since we found that failure to do so resulted, on
average, in an increase of approximately 5% in the apparent concentration of PAPS. We also compared commercially available PAPS
that was 80% pure with PAPS that was more than 95% pure as standards.
There was no significant difference between results obtained with the
two preparations as long as correction was made for the degree of
purity. However, all of our data were obtained using highly purified
PAPS (>95% purity) as a standard.
PAPSS1 is the major isoform expressed in human brain on the basis of
Northern blot analysis, whereas PAPSS2 is the predominant isoform
expressed in liver (Xu et al., 2000). Therefore, we used both brain and
liver cytosol as enzyme sources to study PAPSS substrate kinetics in
tissue preparations. It should be emphasized that this was done with
the understanding that kinetic constants obtained with tissue
preparations represent only estimates. However, an important future
application of this assay will involve genotype-phenotype correlation
analyses performed with tissue preparations. A series of ATP (0.125-8
mM) and Na2SO4 (0.125-32
mM) concentrations was used to estimate apparent
Km values for these two cosubstrates in
liver and brain preparations. When ATP was the varied substrate, the
concentration of Na2SO4
used was 4 mM, and when
Na2SO4 was the varied
substrate, the ATP concentration was 1 mM. The experimental basis for
the selection of these concentrations will be described subsequently.
Double inverse plots of the effect of varying ATP concentration for
both tissues are shown in Fig. 4, and
apparent Km values, values calculated using
only data that did not display substrate inhibition, are listed in
Table 1. The value for brain was
significantly lower than that for liver. However, both values are very
similar to those reported previously for human liver PAPSS (Wong et
al., 1990). The 1 mM ATP concentration used in the assay was selected
because it resulted in maximal activity under these assay conditions.
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 4.
Human PAPSS substrate studies.
Effect of ATP concentration on the PAPSS-catalyzed formation of PAPS
with either human liver (A) or cerebral cortical (B) cytosol as an
enzyme source. Double reciprocal plots for both tissues are shown for
data obtained with seven concentrations of ATP that varied from 0.125 to 8 mM. The concentration of Na2SO4 in all
assays was 4 mM. See text for details.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 1
PAPSS apparent Km values for ATP and SO42
measured with pooled human liver or brain cytosol as an enzyme source
Values are mean ± S.D.
|
|
Double inverse plots of data for liver and cerebral cortex with
Na2SO4 as the varied
substrate are shown in Fig. 5. These data
seemed to indicate the presence of both "high"- and
"low"-affinity components of the double inverse plots. Apparent
Km values for these two kinetic components
are listed in Table 1. In each case, the contribution of the
high-affinity component was subtracted before calculation of the
apparent Km value for the low-affinity component. To ensure that the biphasic kinetics did not result from the
effects of increasing concentrations of
Na2SO4 on the second step
of the assay (the step catalyzed by SULT1E1), an experiment was
performed during which 15.6 nM PAPS was added to the second stage
reaction in the presence of six different concentrations of
Na2SO4 that were the
equivalent of from 1 to 32 mM in the first stage reaction. These
concentrations of Na2SO4
had no effect on the second stage reaction.
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 5.
Human PAPSS substrate kinetic studies.
Effect of Na2SO4 concentration on the
PAPSS-catalyzed formation of PAPS with either human liver (A) or
cerebral cortical (B) cytosol as an enzyme source. Double reciprocal
plots for both tissues are shown for data obtained with a series of
concentrations of Na2SO4. The concentration of
ATP in all assays was 1 mM. See text for details.
|
|
The underlying molecular mechanism responsible for the biphasic
kinetics observed with
Na2SO4 remains unclear.
These curves could indicate either the existence of two separate
enzymes with very different Km values for
Na2SO4, or the presence of
substrate-dependent variation in the kinetic properties of a single
enzyme. However, the biphasic kinetics probably does not result from
the effects of the two known isoforms of PAPSS because the low-affinity
component, with an apparent Km of >20 mM,
could not be of physiological importance. Therefore, an
Na2SO4 concentration of 4 mM, one that would measure only the high-affinity component of the
curves shown in Fig. 5, was selected for use in the assay. It should
also be noted that apparent Km values for
the high-affinity components of the
Na2SO4 curves differed
significantly in the two tissues studied, with values of 0.31 ± 0.031 mM (mean ± S.D.) in the liver and 0.08 ± 0.004 mM in
brain (p < 0.0001 by Student's t
test). Use of 1 mM ATP and 4 mM
Na2SO4, a 20-min incubation
time, a pH of 8.6, and either 50 or 100 µg of cytosol protein
resulted in "signal-to-noise" ratios for the assay of approximately
7 for a pooled human liver cytosol sample and approximately 5 for a
pooled cerebral cortical cytosol preparation.
Since we had observed significant differences in apparent
Km values for
SO42 between the two tissues
studied, we also studied the inhibition of PAPSS activity in these two
tissues by chlorate, a competitive inhibitor of PAPSS with regard to
SO42 (Baeuerle and Huttner,
1986). When a series of concentrations of sodium chlorate was
tested, this compound, as anticipated, inhibited PAPSS activity in both
pooled human hepatic and cerebral cortical cytosol preparations (Fig.
6). Furthermore, chlorate inhibition of
PAPSS activity in brain and liver differed significantly, with
IC50 values of 6.57 ± 0.26 (mean ± S.D.) and 3.26 ± 0.13 mM for these two tissues, respectively
(p < 0.0001) (Fig. 6). The next studies,
experiments that represented a step toward future genotype-phenotype
analyses, involved use of this coupled assay to determine the nature
and extent of individual variation of PAPSS activity in an important
human drug-metabolizing organ, the liver.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 6.
Chlorate inhibition of PAPSS in human liver
or cerebral cortical cytosol.
PAPSS activity was measured in the absence of sodium chlorate and in
the presence of a series of concentrations of sodium chlorate. ATP (1 mM) and Na2SO4 (4 mM) were present in all
assays.
|
|
Hepatic PAPSS Activity.
PAPSS activity was measured in 83 surgical biopsy samples of hepatic
tissue obtained from 43 men and 40 women. These samples had been
collected over the past 3 years during clinically indicated surgical
procedures. The coefficient of variation for assay of the same sample
daily for 7 days was 4.3%, whereas the intra-assay coefficient of
variation for a single sample assayed 12 times was 1.8%. The average
level of PAPSS activity in these 83 samples was 8.43 ± 5.85 nmol/h/mg protein (mean ± S.D.), with an 18-fold interindividual
variation (Fig. 7). The frequency
distribution shown in Fig. 7 was skewed, with a small number of samples
having relatively high levels of enzyme activity, greater than 20 nmol/h/mg protein. We attempted to determine whether this variation in
level of PAPSS activity might be systematically related to the time of
tissue storage at 80°C. There was not a significant correlation between time of tissue storage and level of enzyme activity
(rs = 0.145, p = 0.196). We
next asked whether subject age or gender might be related to level of
hepatic PAPSS activity. There was also not a significant relationship
between subject age and PAPSS activity (rs = 0.130, p = 0.245). Finally, even though the average level of activity in hepatic tissue from men was higher than that for
women, 9.52 ± 6.33 nmol/h/mg protein (mean ± S.D.) versus 7.46 ± 5.20, this difference was not statistically significant (p = 0.19 by Mann-Whitney U test).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 7.
Human liver PAPSS activity.
Frequency distribution histogram of PAPSS activity in hepatic biopsy
samples from 83 patients (43 males and 40 females). See text for
details.
|
|
| |
Discussion |
We have developed a sensitive radiochemical enzymatic assay for
the measurement of PAPSS activity that is suitable for use in
genotype-phenotype correlation analyses as a step toward future pharmacogenetic studies. The principle underlying this coupled, two-step radiochemical assay is depicted schematically in Fig. 1. ATP
is a substrate for PAPSS, but it is also a potent competitive inhibitor
of SULTs (Rens-Domiano and Roth, 1987). Therefore, we initially
determined whether there might be significant differences among
recombinant human SULTs in their sensitivity to inhibition by ATP. We
found that SULT1E1 and SULT2A1 were much more resistant to ATP
inhibition than were the phenol SULTs studied (Fig. 2). In addition,
SULT1E1 had low Km values for both of its
cosubstrates, E2 and PAPS. As a result of these observations, and
because of the commercial availability of high specific activity
[3H]E2, SULT1E1 was selected for use in the
PAPS measurement step of the coupled assay. The final concentration of
ATP present in that step, after dilution of the reaction mixture from
the first step, was approximately 30 µM, a concentration that does
not inhibit SULT1E1 (Fig. 2). By taking this approach, we were able to
eliminate a significant problem that has been encountered in the course of previously described coupled PAPSS assays, since ATP that is "carried over" from the initial PAPS-generating step inhibited the
second stage reaction when phenol SULTs or tissue cytosol was used as
an enzyme source (Hazelton et al., 1985; Vargas, 1988; Wong et al.,
1990; Wong and Wong, 1994).
Human liver and brain have very different patterns of expression for
PAPSS isoforms on the basis of Northern blot analysis (Xu et al.,
2000). Liver displayed a high level of expression of PAPSS2 mRNA
relative to PAPSS1 mRNA, whereas the reverse was true of human brain.
We found that PAPSS activity in liver cytosol had slightly higher
apparent Km values for both ATP and
SO42 than did brain cytosol
(Table 1). In addition, substrate kinetic studies performed with
Na2SO4 as the varied
cosubstrate displayed biphasic behavior in both tissues (Fig. 5). It is
unclear whether the high- and low-affinity activities with regard to
SO42 result from the effects
of two separate enzymes or merely a single enzyme with kinetic
characteristics that vary with substrate concentration. However, it is
clear that the apparent low-affinity activity cannot be of
physiological significance since serum
SO42 concentrations are
estimated to be approximately 0.4 mM (Morris and Levy, 1983), well
below the apparent Km value for the
low-affinity activity. Therefore, we used an
SO42 concentration of 4 mM in
the assay. As a result, the low-affinity component with regard to
Na2SO4 would not contribute
significantly to the activity measured. It should be emphasized that
biochemical characteristics measured in tissue preparations must be
viewed with caution, and cannot substitute for values measured with
purified and/or recombinant enzyme. A further complicating factor that will have to be considered in future studies of PAPSS activity in
tissues is the recent report of the nuclear localization of PAPSS1
(Besset et al., 2000). The present studies were performed primarily to
optimize the assay, an assay that could be used, among other
applications, to measure PAPSS activity both in recombinant enzyme and
in tissue preparations. However, future experiments performed with
tissue preparations will have to take into account potential
differences in the subcellular localization of PAPSS isoforms.
Finally, in an attempt to begin the process of characterizing the
nature and extent of individual variation in PAPSS activity in human
tissues, we phenotyped 83 human liver surgical biopsy tissue samples
for level of activity. Levels of PAPSS activity in these 83 tissue
biopsy samples varied over an 18-fold range (Fig. 7). That variation
was not significantly related to either time of tissue storage or
subject age. The molecular mechanism or mechanisms responsible for
this wide variation in level of activity among individuals will be the
subject of future studies.
In summary, we have developed and optimized a sensitive, coupled
radiochemical enzymatic assay for the determination of PAPSS activity.
The selection of recombinant SULT1E1 for use in the PAPS measurement
step represented a critical factor for assay sensitivity. We then used
the assay to study selected biochemical characteristics of PAPSS
activity in two important human tissuesliver and brain. We also used
the assay to perform an initial study of the nature and extent of
individual variation in human hepatic PAPSS activity. All of these
experiments represent steps toward an eventual study of mechanisms,
including pharmacogenetic mechanisms, that might participate in the
regulation of PAPSS in humans.
We thank Carol Szumlanski for helpful advice and Luanne Wussow for
assistance with the preparation of this manuscript.
This study was supported in part by National Institutes of
Health Grants RO1 GM28157 (to R.M.W.), RO1 GM35720 (to R.M.W.), UO1
GM61388 (to R.M.W.), and a Merck Sharp & Dohme International Clinical
Pharmacology Fellowship (to Z.H.X.).
Abbreviations used are:
PAPS, 3'-phosphoadenosine 5'-phosphosulfate;
SULT, sulfotransferase;
PAPSS, 3'-phosphoadenosine 5'-phosphosulfate synthetase;
E2, estradiol;
DHEA, dehydroepiandrosterone;
DTT, dithiothreitol;
HSS, high-speed
supernatant.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
