Lack of enantioselectivity in the SULT1A3-catalyzed sulfoconjugation of normetanephrine enantiomers: An in vitro and computational study

Article abstract of DOI:10.1002/chir.22108

(1R)-Normetanephrine is the natural stereoisomeric substrate for sulfotransferase 1A3 (SULT1A3)-catalyzed sulfonation. Nothing appears known on the enantioselectivity of the reaction despite its potential significance in the metabolism of adrenergic amines and in clinical biochemistry. We confronted the kinetic parameters of the sulfoconjugation of synthetic (1R)-normetanephrine and (1S)-normetanephrine by recombinant human SULT1A3 to a docking model of each normetanephrine enantiomer with SULT1A3 and the 3'-phosphoadenosine-5'- phosphosulfate cofactor on the basis of molecular modeling and molecular dynamics simulations of the stability of the complexes. The K_M, V_max, and k_cat values for the sulfonation of (1R)-normetanephrine, (1S)-normetanephrine, and racemic normetanephrine were similar. In silico models were consistent with these findings as they showed that the binding modes of the two enantiomers were almost identical. In conclusion, SULT1A3 is not substrate-enantioselective toward normetanephrine, an unexpected finding explainable by a mutual adaptability between the ligands and SULT1A3 through an "induced-fit model" in the catalytic pocket. Copyright

Full text of DOI:10.1002/chir.22108

CHIRALITY (2012)
Lack of Enantioselectivity in the SULT1A3-catalyzed Sulfoconjugation of
Normetanephrine Enantiomers: An In Vitro and Computational Study
1
2
2
1
3
*
ERIC GROUZMANN, JEAN-BAPTISTE GUALTIEROTTI, SANDRINE GERBER-LEMAIRE, KARIM ABID, NOUREDDINE BRAKCH,
ALESSANDRO PEDRETTI,4 BERNARD TESTA,5 AND GIULIO VISTOLI4
1Service de Biomédecine, Lausanne University Hospital (CHUV), Lausanne, Switzerland
2Ecole Polytechnique Fédérale de Lausanne, ISIC, LSPN, Batochime, Lausanne, Switzerland
3Department of Internal Medicine, Service of Nephrology, Lausanne University Hospital (CHUV), Lausanne, Switzerland
4Dipartimento di Scienze Farmaceutiche “Pietro Pratesi”, Facoltà di Farmacia, Università degli Studi di Milano, Milano, Italy
5Department of Pharmacy, Lausanne University Hospital (CHUV), Lausanne, Switzerland
ABSTRACT
(1R)-Normetanephrine is the natural stereoisomeric substrate for sulfotransferase
1A3 (SULT1A3)-catalyzed sulfonation. Nothing appears known on the enantioselectivity of the
reaction despite its potential significance in the metabolism of adrenergic amines and in clinical
biochemistry. We confronted the kinetic parameters of the sulfoconjugation of synthetic
(1R)-normetanephrine and (1S)-normetanephrine by recombinant human SULT1A3 to a docking
model of each normetanephrine enantiomer with SULT1A3 and the 30-phosphoadenosine-50-
phosphosulfate cofactor on the basis of molecular modeling and molecular dynamics simula-
tions of the stability of the complexes. The KM, Vmax, and kcat values for the sulfonation of
(1R)-normetanephrine, (1S)-normetanephrine, and racemic normetanephrine were similar. In silico
models were consistent with these findings as they showed that the binding modes of the two
enantiomers were almost identical. In conclusion, SULT1A3 is not substrate-enantioselective
toward normetanephrine, an unexpected finding explainable by a mutual adaptability between
the ligands and SULT1A3 through an “induced-fit model” in the catalytic pocket. Chirality,
00:000-000, 2012. © 2012 Wiley Periodicals, Inc.
KEY WORDS: sulfotransferase 1A3; normetanephrine; enantioselectivity; in silico models;
enzyme kinetics; induced-fit model
INTRODUCTION
racemic sulfated calibrator we previously described does or
does not influence the results.5
The physiological effects of norepinephrine and epinephrine
are terminated by various conjugation pathways, notably
3-O-methoxylation to produce normetanephrine and meta-
nephrine followed by sulfonation at the para-hydroxyl
group (Fig. 1).1 Among the 13 known human cytosolic
sulfotransferases, sulfotransferase 1A3 (SULT1A3) (EC 2.8.2.1)
is the enzyme that catalyzes the transfer of a sulfonyl group
from the cofactor 30-phosphoadenosine-50-phosphosulfate
(PAPS) to the free remaining para-hydroxyl group on the
phenyl ring of metanephrine and normetanephrine.2
(1R)-Normetanephrine present in food or produced by our
body is sulfonated during enterohepatic cycling by SULT1A3
located in the gastrointestinal tract.3 Importantly, in the
context of this study, normetanephrine is used as an important
biomarker in the diagnosis and monitoring of patients suffering
from pheochromocytoma/paraganglioma, which are tumors
that secrete excessive amounts of catecholamines and free
metanephrines.4 In such bioanalyses, sulfoconjugated normeta-
nephrine must first be hydrolyzed. The sole commercially
available source of authentic sulfoconjugated normetanephrine
is the racemate, which is normally used for calibration and
quality control.5 However, this is a potential source of problems
because measurements involve an acid-catalyzed or enzymatic
hydrolysis step to hydrolyze the sulfoconjugate to the free
metabolite prior to analytical measurement by HPLC with
electrochemical detection. For these reasons, we considered
it significant to determine the substrate enantioselectivity
of the sulfonation of normetanephrine enantiomers by
SULT1A3 and find out whether their substitution by the
© 2012 Wiley Periodicals, Inc.
There are indeed several reports on the substrate enantio-
selectivity of SULT1A3, which for example shows a unique
sulfoconjugating activity toward (2S)-tyrosine and (2S)-(30,40-
dihydroxyphenyl)alanine (L-DOPA), with a high enantios-
electivity for their non-physiological (2R)-enantiomers.6 The
important role of Glu146 in the substrate specificity of
SULT1A3 was demonstrated,7,8 and the crystal structure of
SULT1A3–PAPS–ligand complexes confirmed that residues
Glu146 and Asp86 are crucial to the L-DOPA/tyrosine-
sulfonating activity of SULT1A3 and play also a role in the stereo-
selectivity of the reaction.6–8 There is also strong evidence that
the sulfonation of b-adrenoceptor agonists and antagonists is
enantioselective.9–14 Extrapolating from the existing literature,
we therefore expected the sulfoconjugation of synthetic (1R)-
normetanephrine and (1S)-normetanephrine to be notably
enantioselective and used recombinant human SULT1A3 to
challenge this hypothesis. Experimental investigations were
carried out to determine the enzymatic kinetic constants for each
enantiomer of normetanephrine. The results were compared
with computational models obtained by molecular modeling
Additional Supporting Information may be found in the online version of
this article.
*Correspondence to: Eric Grouzmann, Service de Biomédecine, Lausanne
University Hospital (CHUV), Avenue de Beaumont 29, CH-1011 Lausanne,
Switzerland. E-mail: Eric.Grouzmann@chuv.ch
Received for publication 11 June 2012; Accepted 24 July 2012
DOI: 10.1002/chir.22108
Published online in Wiley Online Library
(wileyonlinelibrary.com).

GROUZMANN ET AL.
Fig. 1. The SULT1A3-catalyzed reaction whose substrate enantioselectivity is investigated here.
(MM) and molecular dynamics (MD) to explain the observed
Enantioselective Synthesis of Normetanephrines
(and unexpected) lack of enantioselectivity of the reaction.
The chemical protocols used for the synthesis of normetanephrines
(compounds 8 and 9, Scheme 1) are available as Supplementary Material,
as is the procedure for determining enantiomeric excess and absolute config-
urations of intermediates 4 and 5. All commercially available reagents and sol-
vents (Fluka/Aldrich, Buchs, Switzerland and Acros, Wohlen, Switzerland)
were used without further purification. For reactions requiring anhydrous
conditions, dry solvents were obtained by filtration (Innovation Technology).
Unless stated otherwise, experiments were carried out under an argon
atmosphere. Reactions were monitored by thin-layer chromatography (Merck
MATERIAL AND METHODS
Chemicals and Reagents
Racemic normetanephrine (rac-normetanephrine) was purchased from
Sigma-Aldrich (St. Louis, Mo, USA). PAPS was provided by Calbiochem
(Laeufelfingen, Switzerland). Plasmid carrying the SULT1A3 cDNA was
a kind gift of Dr. Sakakibara, Department of Biochemistry, University of
Miyazaki, Miyazaki, Miyazaki 889-2192, Japan.
Scheme 1. The chemical synthesis of (1R)-normetanephrine (8) and (1S)-normetanephrine (9).
Chirality DOI 10.1002/chir

NON-STEREOSELECTIVITY OF NORMETANEPHRINE FOR SULT1A3
silica gel 60F254 plates), detection by UV light, KMnO4, or Pancaldi reagents
HPLC Method
[(NH4)6MoO4, Ce(SO4), H2SO4, H2O]. Purifications were performed by
flash chromatography on silica gel (Merck, Zoug, CH, Nꢀ 9385 silica gel 60,
240–400 mesh) and reverse phase HPLC. 1H-NMR spectra: Bruker ARX-
400, Bruker DPX-400 spectrometers at 400 MHz, and Bruker AVII-800 spec-
trometers at 800 MHz. Chemical shifts in ppm relative to the solvent’s residual
1H signal as internal reference were as follows: MeOD, 3.34 ppm; CDCl3,
7.27 ppm; C6D6, 7.30 ppm. 1H assignments were confirmed by 2D-COSY spec-
tra. Multiplicity reflects apparent patterns. Coupling constants J are in Hz (b
stands for broad). 13C-NMR spectra: same instrument as above at 101 MHz.
Reference values for solvents used as internal reference in ppm were as fol-
lows: MeOD, 49 ppm; CDCl3; 77ppm; C6D6, 128.5 ppm. Coupling constants
J are in Hz. 13C assignments were confirmed by 2D-HSQC spectra. IR spectra:
Perkin-Elmer Paragon 1000 FTIR spectrometer. Mass spectra: MALDI-TOF
spectrometer (Axima-CFR+, Kratos, Manchester, UK), ESI-Q spectrometer
(Finnigan SSQ 710C, Thermoquest, UK), and HRMS-ESI spectrometer
(Q-TOF Ultima spectrometer, Micromass, Manchester, UK).
The strategy envisaged for the enantioselective synthesis of (1R)-
normetanephrine and (1S)-normetanephrine relied on asymmetric
epoxidation or dihydroxylation to introduce the secondary alcohol of the
side-chain with high enantioselectivity. Starting from vanillin, protection of
the phenol as silyl ether followed by Wittig olefination of the aldehyde15
delivered intermediate 3 in high yield (Scheme 1). Treatment with
Jacobsen catalyst, in the presence of meta-chloroperbenzoic acid and
N-methylmorpholine-N-oxide,16–18 induced epoxidation of the olefin
followed by subsequent opening of the oxirane by the meta-chlorobenzoic
acid by-product. All our attempts to prevent opening of the epoxide during
this step did not meet with success.
The HPLC analyses were performed with an Alliance Instrument from
Waters coupled with the corresponding UV detector set to 280 nm. Separation
was carried out using a C18-reversed phase column (Macherey-Nagel,
Basel, Switzerland). The mobile phase consisted of 217 mM sodium
phosphate (NaH2PO4), 42.8mM citric acid, and 546 mM octanesulfonic
acid at pH 2.9 containing 2% acetonitrile. The flow rate was 0.9 ml/min;
25ml of the incubation mixture was injected, and areas under the curve
were determined for each substrate and metabolite (unconjugated and
sulfoconjugated normetanephrine) and converted into concentrations on
the basis of a calibration curve determined with synthetic free and sulfo-
nated compounds. The reaction components had the following order of
elution: PAPS (2.5min), PAP (2.9min), sulfonated normetanephrine
(3.6 min), and free normetanephrine (9.8min). KM and Vmax constants were
determined from initial velocity measurements plotted versus different
substrate concentrations using hyperbolic representations (GraphPad
Prism Software, San Diego, CA, USA). KM and Vmax values were expressed
as mM and pmol/min.
Docking Analysis by Molecular Modeling
The resolved structure of the human sulfotransferase SULT1A3 in
complex with dopamine and PAP was retrieved from the PDB database
(Id: 2A3R).6 After deleting water molecules and adding hydrogen atoms,
the structure was minimized keeping fixed the backbone atoms to
conserve the experimental folding. The resulting structure was updated
by manually modifying PAP into the PAPS cofactor whose atomic charges
were attributed by PM6 semi-empirical calculations using MOPAC. The
enzyme–cofactor complex was further minimized keeping fixed the
atoms outside a 15-Å radius sphere around the modified cofactor and then
used in the following docking analyses. The two enantiomers of normeta-
nephrine were built in their protonated forms, and their conformational
profile was investigated by a clustered MonteCarlo analysis as implemented
in the VEGA suite of programs to produce 1000 minimized conformers.24
A sphere of 12.0 Å radius was defined around the bound dopamine, so
encompassing the entire binding cavity. The resolution of the grid was
60 ꢁ 50 ꢁ 45 points with a grid spacing of 0.450 Å. The dopamine
molecule was then removed, and the lowest energy structures of the
two enantiomers of normetanephrine were inserted in turn into the
enzyme–cofactor complex within the 12.0-Å sphere using the AutoDock
4.0 software.25 The ligands were then docked into this grid with the
Lamarckian algorithm as implemented in AutoDock, and the flexible
bonds of the ligand were left free to rotate. The genetic-based algorithm
ran 20 simulations per substrate with 2,000,000 energy evaluations and
a maximum number of generations of 27,000. The crossover rate was
increased to 0.8, and the number of individuals in each population to 150.
All other parameters were left at the AutoDock default settings. The
obtained complexes were ranked considering both the AutoDock scores
and the distance between the substrate’s hydroxy group and cofactor.
The chosen complexes were finally minimized keeping fixed the atoms
outside a 15-Å radius sphere around the bound substrate and then used to
recalculate docking scores and in the subsequent MD simulations.
Gratifyingly, asymmetric dihydroxylation in the presence of AD-mix-a19,20
delivered diol 4 in good yield and 98% enantiomeric excess (the enantio-
meric excess and absolute configuration of the newly formed alcohol were
determined from the 1H-NMR spectra of the corresponding Mosher’s
esters21,22 and supplementary data section). The use of AD-mix-a led to the
formation of the other enantiomer 5 in 76% yield and 98% enantiomeric
excess. Selective tosylation of the primary alcohol was a delicate trans-
formation as migration of the tert-butyldimethylsilyl moiety from the phenol
to the secondary alcohol was observed as side process. Portion-wise addition
of tosyl chloride over 8 h was necessary to avoid this side reaction and
delivered tosylate 6 in 68% yield. Nucleophilic displacement with an excess
of sodium azide (10 eq) followed by Staudinger reduction of the resulting
azide and cleavage of the silyl ether delivered (R)-normetanephrine 8 in high
purity. The use of polymer bound triphenyl phosphine was necessary to
avoid decomposition during the purification procedure. The same pathway
was performed on diol 5 to afford (S)-normetanephrine 9 in 20% overall
yield (four steps).
Preparation of Recombinant Sulfotransferase 1A3
The SULT1A3 gene was cloned and expressed in Escherichia coli BL-21
strain (Promega, Walliselen, Switzerland) using the pGEX-2TK glutathione
S-transferase gene fusion system and purified using glutathione Sepharose
in conjunction with thrombin cleavage.23
Sulfotransferase Assay
Molecular Dynamics Simulations
(1R)-Normetanephrine, (1S)-normetanephrine, and rac-normetanephrine
were incubated along with the sulfotransferase and PAPS, the universal
sulfonyl group donor for SULT-catalyzed sulfonations. The reaction gives
rise to a sulfated product and adenosine 30,50-diphosphate (PAP). The
sulfotransferase assay was performed in 0.1 ml of 10mM sodium phosphate
buffer pH 6.8 containing 80mM PAPS and (1R)-normetanephrine,
(1S)-normetanephrine, or rac-normetanephrine at concentrations of 2.5, 5,
10, 20, 30, 40, 50, and 70 mM. The enzyme reaction was started by the
addition of 300 ng of recombinant SULT1A3, and the incubation was stopped
after 4 min at 37ꢀC by adding one volume of mobile phase prior to injection
on HPLC (50 ml). The rate of sulfonation was measured as a function of
normetanephrine concentration, and the kinetic parameters were deter-
mined by fitting the Michaelis–Menten equation from which were derived
the KM and Vmax constants. The reactions were performed in three separate
experiments for each compound.
The MD simulations involved SULT1A3 in complex with PAPS and the
two normetanephrine enantiomers as generated by docking simulations.
The complexes were firstly neutralized by adding nine Na+ ions whose
location was computed by the SODIUM software and then inserted into
a 50-Å radius sphere of water molecules. After a preliminary minimization
to optimize the relative position of solvent molecules, the systems underwent
5 nsec of all-atoms MD simulations with the following characteristics:
(1) spherical boundary conditions were introduced to stabilize the simulation
space; (2) Newton’s equation was integrated using the r-RESPA method
(every 4 fsec for long-range electrostatic forces, 2 fsec for short-range non-
bonded forces, and 1 fsec for bonded forces); (3) the temperature was
maintained at 300ꢂ 10 K by means of Langevin’s algorithm; (4) Lennard–
Jones interactions were calculated with a cutoff of 10Å, and the pair list
was updated every 20 iterations; (5) a frame was stored every 10psec,
yielding 500 frames; and (6) no constraints were applied to the systems.
Chirality DOI 10.1002/chir

GROUZMANN ET AL.
The simulations were carried out in two phases: an initial period of heating
both isomers is involved in ionic bonds with Asp86 and
Glu89. For (1R)-normetanephrine, it also seems that Glu146
contributes to the negatively charged cage that surrounds
and binds the protonated amino group. Furthermore, the
two isomers share a similar network of apolar contacts that
include p–p stacking of the substrate phenyl ring with
Phe24, Phe81, Phe142, and Tyr240 plus hydrophobic interac-
tions with Val84, Leu247, and Met248 (most of which not
displayed in Figure 3 for reasons of clarity).
The only noteworthy difference between the two complexes
lies in the orientation of the substrate’s hydroxyl group. In the
(1S)-normetanephrine complex (Fig. 3A), the hydroxyl group
forms a strong intramolecular H-bond with the protonated
amino group and seems to interact more weakly with Glu89.
In the (1R)-normetanephrine complex (Fig. 3B), however, the
hydroxyl group donates a H-bond to Glu146 and interacts
weakly with a few other polar residues in its vicinity.
Another interesting difference between the two complexes
lies in the position of some residues, most notably Asp86 and
Glu89. It thus appears that while the two substrates adopt a
comparable conformation in their respective complex, the
enzymatic binding site alters and adapts the position of some
of its key residues to maintain comparable weak bonds
with functional groups close to C(1), the center of chirality
in the substrate.
from 0 to 300 K over 6000 iterations (6psec, i.e., 1 K/20 iterations) and a
monitored phase of simulation of 5 nsec. Only the frames memorized during
this last phase were considered. All described minimizations were
performed using the conjugate gradient algorithm. The calculations
described here were carried out on a 16 CPU Tyan-VX50 system using
Namd2.6 with the force field CHARMM and Gasteiger’s atomic charges.26
RESULTS AND DISCUSSION
Enantioselective Synthesis of Normetanephrines
The challenges associated with the synthesis of optically
pure normetanephrines relied on the high degree of functio-
nalization of the aromatic core and the introduction of a chiral
center on the lateral side chain, with high enantioselectivity.
The use of vanillin as inexpensive starting material provided a
good template for the sequential introduction of the different
functional groups of normetanephrines, and Sharpless asym-
metric dihydroxylation afforded both good yields and high
enantioselectivities for the elaboration of the chiral secondary
alcohol. The synthetic pathways disclosed herein represent
reliable routes toward both enantiomers of normetanephrines
with high optical purities, in seven steps and 11–13% overall
yield from vanillin. The experimental procedures could be
easily scaled up to produce meaningful quantities of these
valuable biomarkers.
Various models have been proposed over more than a
century to account for chiral recognition in protein–ligand
complexes. The “Lock-and-Key Model” proposed in 1890 by
Emil Fischer is far too limited as it does not account for the
critical role played by ligand and target flexibility and mutual
adaptability in increasing affinity.29 An insightful model has
been proposed to depict this process of mutual adaptability
in biochemistry and pharmacology, namely the “induced-fit
model”.29–32 In a metaphoric view, we may interpret our
results by combining the “lock-and-key” and “induced fit”
models such that the “lock” adapts itself to two enantiomeric
“keys” of comparable conformation.
In summary, docking results indicate that the two enantiomers
stabilize very similar patterns of interactions, thus suggesting a
similar capacity to interact with SULT1A3. This is confirmed by
the almost identical docking scores of the two complexes,
namely ꢃ8.04kcal/mol for (1S)-normetanephrine versus
ꢃ8.11 for (1R)-normetanephrine. Further confirmation comes
from the low value (0.91 Å) of the root-mean-square deviation
(RMSD), a measure of the mean difference between the
positions of the heteroatoms in the two docked enantiomers.
Incubations of Normetanephrines with Sulfotransferase 1A3
The maximum concentration of normetanephrine used to
establish kinetic parameters was set at 50mM because product
inhibition was observed at higher normetanephrine concentra-
tions (Fig. 2). The KM values obtained for the production of the
sulfoconjugated (1R)-normetanephrine, (1S)-normetanephrine,
and rac-normetanephrine were similar: 1.90ꢂ 0.30, 1.82ꢂ 0.08,
and 1.97ꢂ 0.66mM, respectively (P = 0.87). The Vmax values for
the formation of the sulfoconjugates were 51.3ꢂ 1.5, 45.8ꢂ 2.8,
and 46.8ꢂ 1.2 pmol/min, respectively (P = 0.22). SULT1A3
exhibited similar specificity constants toward the two enan-
tiomers and the racemate of normetanephrine at 5.89, 5.26,
and 5.38 minꢃ1 (P = 0.22).
These experimental data suggest that the absolute configura-
tion at C(1) bearing the hydroxyl group does not play a signifi-
cant role in substrate affinity for SULT1A3 and that, at least
for normetanephrine, the enzyme also binds and metabolizes
the non-natural (1S)-enantiomer, in agreement with its broad
tolerance to different substrates.27 We are not aware of the
existence of experimental results assessing the enantio-
selectivity of SULT1A3 toward norepinephrine and epinephrine.
Molecular Dynamics of the Sulfotransferase 1A3–
Normetanephrine–30-Phosphoadenosine-50-
phosphosulfate Complexes
Molecular Docking of the Sulfotransferase
1A3–Normetanephrineꢃ30-Phosphoadenosine-50-
phosphosulfate Complexes
The computational methods used here are able to monitor
conformation changes but in no way can they simulate the
cleavage and formation of covalent bonds. This implies that
they cannot monitor the course of an enzymatic reaction.
Nevertheless, MD can assess the relative stability of docking
complexes by simulating simultaneously conformational
and translational movements.
With a view to further confirm the remarkable analogy
between the poses of the two normetanephrine enantiomers
in the SULT–normetanephrine–PAPS complexes, 5-nsec
MD runs were performed that allow the stability of both
complexes to be monitored for the duration of the simulations.
A preliminary analysis concerned the folding stability of the
Figure
3 compares the optimized poses of (1S)-
normetanephrine(Fig. 3A) and(1R)-normetanephrine(Fig. 3B)
in the catalytic cavity of the SULT1A3 enzyme. Except for the
orientation of the hydroxyl group, a very similar arrangement
of the two enantiomers is obvious in these figures. In other
words, the two enantiomers adopt a comparable conformation
in the optimized complexes.
Specifically, Figure 3 shows that the target para-hydroxyl
group is very close to the PAPS cofactor and to the catalytic
Lys106 and His108, a proximity conducive to the subsequent
transfer of the SO3 moiety.28 The protonated amino group of
Chirality DOI 10.1002/chir

NON-STEREOSELECTIVITY OF NORMETANEPHRINE FOR SULT1A3
Fig. 2. Michaelis–Menten (A, C, and E) and Lineweaver Burk (B, D, and F) plots of the rate of sulfoconjugation of (1R)-normetanephrine, (1S)-normetanephrine,
and rac-normetanephrine by recombinant human SULT1A3. The substrates were incubated at concentrations ranging from 5 to 70 mmol/l (x-axis). The products
were quantified by HPLC. The reactions were performed in three separate experiments for each compound, and the error bars represent the standard deviations.
simulated protein as assessed by RMSD values (computed
from backbone atoms only) and by the percentage of residues
falling in the allowed regions of the Ramachandran plot. In
the MD runs with either enantiomer, the enzyme protein
showed a remarkable stability because the RMSD values
remained constantly in a narrow range around 5 Å and the per-
centage of allowed residues did not vary significantly, remain-
ing around 70% during the whole simulations (plots not shown).
Such a folding stability suggests that the results described be-
low are mainly due to the dynamic response of the enzyme
and bound ligands and not to the random structural distortions.
The first major analysis concerned the stability of the cofac-
tor binding as monitored by the distance between the Arg130
and the 3-phosphate group in PAPS. This ion-pair belongs to
the network of ionic bonds (also involving Lys48, Lys106,
Lys197, Arg257, and Lys258) that play a pivotal role in PAPS
binding. Figure 4 clearly shows that PAPS remained stably
bound within the catalytic cavity during both simulations.
First, this result confirms the expected stability of the PAPS
binding mode as resolved by X-ray analysis.6 Furthermore,
Figure 4 confirms the stability for the whole architecture of
the catalytic site, thus indicating that the behavior of normeta-
nephrine enantiomers truthfully mirrors the stability of the
complexes and the absence of random distortions in the
catalytic cavity.
The second major MD analysis concerned the stability for
the computed poses of normetanephrine enantiomers as
assessed by the length of the ionic bond between Asp86
and the protonated amino group in the substrate. Figure 5
shows that the length of this ionic bond fluctuated inside a
narrow range (2.7 to 3.0 Å). Even more importantly, there
was no detectable difference between the two enantiomers,
a finding compatible with their similar enzyme affinity and
productive stability in the cavity.
Chirality DOI 10.1002/chir

GROUZMANN ET AL.
Fig. 5. Dynamic profile of the length of the ionic bond (in Angström)
between the Asp86 and the protonated substrate amino group in the
SULT1A3–normetanephrine–PAPS complexes as obtained by molecular
dynamics simulations. The gray line corresponds to the complex with (1S)-
normetanephrine, and the black line to the complex with (1R)-normetanephrine.
No meaningful difference exists between the two complexes, implying their
similar stability. See text for computational details.
results were obtained using the ad hoc synthetized normeta-
nephrine enantiomers and expressed human SULT1A3.
Although this finding appears compatible with the broad
substrate selectivity of SULT1A3,27 it remains difficult to
interpret. MM and MD simulations were therefore carried out
in an attempt to understand the biomolecular mechanisms
underlying this lack of enantioselectivity.
CONCLUSIONS
Taken together, docking computations and MD simulations
have confirmed the substantial equivalence of the two computed
binding modes in terms of both strength of stabilizing contacts
and dynamic behavior during the simulation time. In other
words, the computational results are compatible with and offer
a molecular interpretation to the experimental finding that
SULT1A3 conjugates both enantiomers of normetanephrine
with undistinguishable affinity (KM) and maximal rate of
reaction (Vmax).
Fig. 3. Docking simulations showing the main interactions stabilizing the
complexes between SULT1A3 and (1S)-normetanephrine (Figure 3A) or
(1R)-normetanephrine (Figure 3B). See text for computational details.
ACKNOWLEDGMENTS
We thank Ms. Adeline Herren for her technical assistance.
The authors are very grateful to Dr. Sakakibara for providing
the plasmid encoding the gene SULT1A3.
CONFLICT OF INTEREST
The authors have declared that there is no conflict of
interest.
LITERATURE CITED
1. Goldstein DS, Eisenhofer G, Kopin IJ. Sources and significance of plasma
levels of catechols and their metabolites in humans. J Pharmacol Exp
Ther 2003;305:800–811.
2. Testa B, Krämer SD. Metabolism. Volume 2: Conjugations, Conse-
quences of Metabolism, Influencing Factors. Weinheim, Germany:
Wiley-VCH, 2010;23–35.
3. Cappiello M, Giuliani L, Pacifici GM. Differential distribution of phenol
and catechol sulphotransferases in human liver and intestinal mucosa.
Pharmacology 1990;40:69–76.
Fig. 4. Dynamic profile of the distance (in Angström) between Arg130 and
the 3-phosphate group of PAPS in the SULT1A3–normetanephrine–PAPS
complexes as obtained by molecular dynamics simulations. The gray line
corresponds to the complex with (1S)-normetanephrine, and the black line
to the complex with (1R)-normetanephrine. No meaningful difference exists
between the two complexes, implying their similar stability. See text for
computational details.
The enzyme kinetics investigations reported herein yielded
an unexpected finding, namely the complete lack of substrate
enantioselectivity of the SULT1A3-catalyzed sulfoconjugation
of (1R)-normetanephrine and (1S)-normetanephrine. The
Chirality DOI 10.1002/chir
4. Manger WM, Gifford RW. 1996 In clinical and experimental
pheochromocytoma, 2nd ed. Cambridge: Blackwell Science.
5. Simonin J, Gerber-Lemaire S, Centeno C, Seghezzi C, Iglesias K, Abid K,
Grouzmann E. Synthetic calibrators for the analysis of total

NON-STEREOSELECTIVITY OF NORMETANEPHRINE FOR SULT1A3
metanephrines in urine: revisiting the conditions of hydrolysis. Clin Chim
Acta 2012;413:998–1003.
19. Kolb HC, Van Nieuwenhze MS, Sharpless KB. Catalytic asymmetric
dihydroxylation. Chem Rev 1994;94:2483–2547.
6. Lu JH, Li HT, Liu MC, Zhang JP, Li M, An XM, Chang WR. Crystal
structure of human sulfotransferase SULT1A3 in complex with dopamine
and 30-phosphoadenosine 50-phosphate. Biochem Biophys Res Commun
2005;335:417–423.
20. Sharpless KB, Amberg W, Bennani YL, Crispino GA, Hartung J, Jeong KS,
Kwong HL, Morikawa K, Wang ZM. The osmium-catalyzed asymmetric
dihydroxylation: a new ligand class and a process improvement. J Org Chem
1992;57:2768–2771.
7. Dajani R, Hood AM, Coughtrie MW. A single amino acid, glu146, governs
the substrate specificity of a human dopamine sulfotransferase, SULT1A3.
Mol Pharmacol 1998;54:942–948.
21. Dale JA, Dull DL, Mosher HS. a-Methoxy-a-trifluoromethylphenylacetic
acid, a versatile reagent for the determination of enantiomeric composi-
tion of alcohols and amines. J Org Chem 1969;34:2543–2549.
8. Brix LA, Barnett AC, Duggleby RG, Leggett B, McManus ME. Analysis of
the substrate specificity of human sulfotransferases SULT1A1 and
SULT1A3: site-directed mutagenesis and kinetic studies. Biochemistry
1999;38:10474–10479.
9. Eaton E, Walle UK, Wilson HM, Aberg G, Walle T. Stereoselective
sulphate conjugation of salbutamol by human lung and bronchial epithelial
cells. Br J Clin Pharmacol 1996;41:201–206.
22. Dale TA, Mosher HS. Nuclear magnetic resonance enantiomer reagents.
Configurational correlations via nuclear magnetic resonance chemical
shifts of diastereomeric mandelate, O-methylmandelate, and alpha-meth-
oxy-alpha-trifluoromethylphenylacetate (MTPA) esters. J Am Chem Soc
1973;95:512–519.
23. Sakakibara Y, Takami Y, Nakayama T, Suiko M, Liu MC. Localization
and functional analysis of the substrate specificity/catalytic domains of
human M-form and P-form phenol sulfotransferases.
1998;273:6242–6247.
24. Pedretti A, Villa L, Vistoli G. VEGA: a versatile program to convert, handle
and visualize molecular structure on Windows-based PCs. J Mol Graph
Model 2002;21:47–49.
J Biol Chem
10. Hartman AP, Wilson AA, Wilson HM, Aberg G, Falany CN, Walle T. Enan-
tioselective sulfation of b2-receptor agonists by the human intestine and
the recombinant M-form phenolsulfotransferase. Chirality 1988;10:800–803.
11. Pesola GR, Walle, T. Stereoselective sulfate conjugation of isoproterenol
in humans: comparison of hepatic, intestinal, and platelet activity.
Chirality 1993;5:602–609.
12. Walle T, Walle UK. Stereoselective sulfate conjugation of 4-hydroxypro-
pranolol and terbutaline by the human liver phenolsulfotransferases.
Drug Metab Dispos 1992;20:333–336.
13. Walle UK, Pesola GR, Walle T. Stereoselective sulfate conjugation of
salbutamol in humans: comparison of hepatic, intestinal and platelet
activity. Br J Clin Pharmacol 1993;35:413–418.
25. Morris GM, Goodsell DS, Halliday RS, Huey R, Hart WE, Belew RK, Olson
AJ. Automated docking using a Lamarckian genetic algorithm and empirical
binding free energy function. J Comput Chem 1998;19:1639–1662.
26. Phillips JC, Braun R, Wang W, Gumbart J, Tajkhorshid E, Villa E, Chipot
C, Skeel RD, Kalé L, Schulten K. Scalable molecular dynamics with
NAMD. J Comput Chem 2005;16:1781–1802.
27. Bidwell LM, McManus ME, Gaedigk A, Kakuta Y, Negishi M, Pedersen
L, Martin JL. Crystal structure of human catecholamine sulfotransferase.
J Mol Biol 1999;293:521–530.
28. Testa B, Krämer SD. The biochemistry of drug metabolism—an introduc-
tion. Part 4: reactions of conjugation and their enzymes. Chem Biodivers
2008;5:2171–2336.
14. Wilson AA, Wang J, Koch P, Walle T. Stereoselective sulphate conjugation of
fenoterol by human phenolsulphotransferases. Xenobiotica 1997;27:
1147–1154.
15. Tilley SD, Francis MB. Tyrosine-selective protein alkylation using
p-allylpalladium complexes. J Am Chem Soc 2006;128:1080–1081.
29. Koshland DE, Jr. The key-lock theory and the induced fit theory. Angew
Chem Int Ed Engl 1994;33:2375–2378.
30. Johnson KA. Role of induced fit in enzyme specificity: a molecular for-
ward/reverse switch. J Biol Chem 2008;283:26297–26301.
31. Post CB, Ray WJ Jr. Reexamination of induced fit as a determinant of
substrate specificity in enzymatic reactions. Biochemistry 1995;34:
15881–15885.
32. Sotriffer CA. Accounting for induced-fit effects in docking: what is possible
and what is not? Curr Top Med Chem 2011;11:179–191.
16. Jacobsen EN, Zhang W, Muci AR, Ecker JR, Deng LJ. Highly enantio-
selective epoxidation catalysts derived from 1,2-diaminocyclohexane.
J Am Chem Soc 1991;113:7063–7064.
17. Palucki M, McCormick GJ, Jacobsen EN. Low temperature asymmetric
epoxidation of unfunctionalized olefins catalyzed by (salen)Mn(III) com-
plexes. Tetrahedron Lett 1995;36:5457–5460.
18. Zhang W, Loebach JL, Wilson SR, Jacobsen EN. Enantioselective epoxidation
of unfunctionalized olefins catalyzed by salen manganese complexes.
J Am Chem Soc 1990;112:2801–2803.
Chirality DOI 10.1002/chir