Deiodination of thyroid hormones by iodothyronine deiodinase mimics: Does an increase in the reactivity alter the regioselectivity?

Article abstract of DOI:10.1021/ja201657s

Organoselenium compounds as functional mimics of iodothyronine deiodinase are described. The naphthyl-based compounds having two selenol groups are remarkably efficient in the inner-ring deiodination of thyroxine. The introduction of a basic amino group in close proximity to one of the selenol moieties enhances the deiodination. This study suggests that an increase in the nucleophilic reactivity of the conserved Cys residue at the active site of deiodinases is very important for effective deiodination.

Full text of DOI:10.1021/ja201657s

COMMUNICATION
pubs.acs.org/JACS
Deiodination of Thyroid Hormones by Iodothyronine Deiodinase
Mimics: Does an Increase in the Reactivity Alter the Regioselectivity?
Debasish Manna and Govindasamy Mugesh*
Department of Inorganic and Physical Chemistry, Indian Institute of Science, Bangalore 560 012, India
S Supporting Information
b
Scheme 1. Regioselective Deiodinations of Thyroid
Hormones Catalyzed by Three Iodothyronine Deiodinases
(ID-1, ID-2, and ID-3)
ABSTRACT: Organoselenium compounds as functional
mimics of iodothyronine deiodinase are described. The
naphthyl-based compounds having two selenol groups are
remarkably efficient in the inner-ring deiodination of thyroxine.
The introduction of a basic amino group in close proximity to
one of the selenol moieties enhances the deiodination. This
study suggests that an increase in the nucleophilic reactivity of
the conserved Cys residue at the active site of deiodinases is
very important for effective deiodination.
odothyronine deiodinases (IDs) are mammalian selenoenzymes
1
I
that catalyze the regioselective deiodination of thyroid hormones.
0
Outer-ring (5 ) deiodination of thyroxine (T4) by the type-1 and -2
deiodinases (ID-1 and ID-2) produces the biologically active hormone
0
0
3
,5,3 -triiodothyronine (T3). These two enzymes also convert 3,3 ,
0
0 0
-triiodothyronine (reverse T3 or rT3) to 3,3 -diiodothyronine (3,3 -
2
5
T2) by outer-ring deiodination (Scheme 1). The type-3 deiodinase
(
ID-3) catalyzes the convertion of T4 to rT3 by an inner-ring
2,3
deiodination pathway. Furthermore, ID-3 plays an important role
in the maintenance of serum T3 concentration by catalyzing the
0
conversion of T3 to 3,3 -T2, which has been shown to be a key step in
2ꢀ4
the protection of tissues from excess thyroid hormone.
Initial attempts to develop simple selenium compounds as5
mimics of the deiodinases have led to only limited success.
Recently, Goto et al. reported that selenol 1 bearing sterically
6
indicating that activation of the selenol moiety by the amino
group is probably not important for inner-ring deiodination.
7
hindered substitutents can convert N-butyrylthyroxine methyl
ester to the corresponding triiodo derivative by outer-ring
deiodination. In this case, nucleophilic attack of selenol 1 at
one of the outer-ring iodines has been shown to produce the
corresponding selenenyl iodide. We have reported that selenol 2
bearing a thiol group in close proximity to selenium mimics the
The decrease in the deiodinase activity upon introduction of a
basic amino group in close proximity to the selenol was un-
expected, as one or more histidine residues (His) appear to play
crucial roles in the deiodination reactions. K €o hrle and others
have shown that one of the His residues at the active site of ID-1
forms an imidazoliumꢀselenolate ion pair, and therefore, the Sec
0
ID-3 activity by converting T4 and T3 to rT3 and 3,3 -T2,
7
8
respectively, under physiologically relevant conditions. We have
residue is strongly activated toward nucleophilic attack. In fact,
also shown that replacement of the selenol moiety in 2 by a thiol
group (compound 3) significantly reduces the deiodinase activ-
ity. Similarly, compound 4 bearing a basic N-ethylamino group
exhibits much lower deiodinase activity in comparison with 2,
the His residues corresponding to positions 158 and 174 in
2a
6
human ID-1 are conserved in ID-2 and ID-3. Goto et al.
reported the interesting observation that outer-ring deiodination
of N-butyrylthyroxine methyl ester by compound 1 takes place
only in the presence of triethylamine. Herein we report that the
introduction of an additional selenol moiety remarkably en-
hances the deiodinase activity. We also show for the first time
that cooperative effects of a thiolꢀselenol pair or two selenol
groups play a key role in the deiodination.
To understand whether the replacement of the thiol group in
compound 2 by a selenol can alter the regioselectivity of the
Received: February 22, 2011
Published: June 07, 2011
r 2011 American Chemical Society
9980
dx.doi.org/10.1021/ja201657s
|J. Am. Chem. Soc. 2011, 133, 9980–9983

Journal of the American Chemical Society
COMMUNICATION
Figure 2. Comparison of the rates of deiodination of T4 by ID mimics
2
, 3, and 5ꢀ10. The reactions were carried out in phosphate buffer (pH
7.5) at 37 °C. The final assay mixture contained 600 μM ID mimic, 300
Figure 1. HPLC chromatograms for the deiodination of T4 by com-
pound 5. Inset: deiodination (%) as a function of time obtained by
determining the amount of T4 at various time intervals.
μM T4, and 12 mM DTT.
different pH values (Figure S17). These experiments indicated
that compound 5 exhibits greater activity at alkaline pH, which
is consistent with the pH profile of the enzyme. Interestingly,
a rapid increase in the activity was observed at pH >7.0,
probably as a result of stepwise deprotonation of the two thiol
moieties present in DTT, leading to the formation of reactive
deiodination, we studied the deiodination of T4, rT3, and T3 by
compound 5 in the presence of dithiothreitol (DTT). When T4
was treated with 5, a rapid decrease in the peak due to T4 was
observed, with a new peak appearing at 8.5 min. A comparison of
the HPLC chromatograms obtained at various time intervals
1
0
0
with those of authentic rT3, T3, and 3,3 -T2 along with mass
thiolate.
spectral analyses indicated that compound 5 removes iodine
exclusively from the inner ring of T4 to produce rT3 as the only
deiodinated product (Figure 1). When pure rT3 was treated with
The facile deiodination of T4 and T3 by compound 5 led us to
investigate the deiodinase activity of compounds 6ꢀ10 having
thiol and/or selenol moieties. A comparison of the rates of
deiodination for various compounds (Figure 2) indicated that
compound 5 having two selenol moieties is more active than
compounds 2 and 3, which contain a thiolꢀselenol pair and two
thiol moieties, respectively. While the replacement of one of the
thiol groups in compound 3 by a selenol moiety led to only
12-fold enhancement in the activity, a remarkable 91-fold increase in
the deiodinase activity was observed when both thiol groups were
replaced by selenol moieties. Interestingly, compound 7, which has a
SePh substituent instead of the SeH group, exhibited very weak
deiodinase activity. In contrast, compound 6, which contains a thiol
moiety, has been shown to be inactive in both T4 and T3 assays even
at higher concentrations. Kuiper et al. showed that the selenocys-
teine (Sec) residue in the catalytic center of ID-3 is crucial for inner-
ring deiodination, as the substitution of Sec by a cysteine (Cys)
significantly reduces the catalytic efficiency. They showed that the
substrate turnover numbers for the deiodinations of T4 and T3 by
the Cys mutant were 6- and 2-fold lower, respectively, than for the
wild-type enzyme. However, it is not known whether the re-
placement of the conserved Cys residue by a Sec residue can
enhance the deiodinase activity of the ID-3 enzyme. The present
study suggests that such a modification may lead to the generation of
enzymes with higher deiodinase activities.
0
5
, no 3,3 -T2 formation was observed, indicating that compound
5
does not remove iodine from the outer ring of rT3. In contrast,
0
treatment of T3 with compound 5 readily produced 3,3 -T2
(Figure S15 in the Supporting Information), suggesting that
compound 5 is selective toward the inner-ring iodines of T4 and
T3. Although the deiodination of T4 or T3 by compound 5
occurred even in the absence of DTT, the concentration of DTT
appeared to have a significant effect on the reaction rate. While
the deiodination was enhanced initially with an increase in the
concentration of DTT, the activity was reduced at higher
concentrations of DTT (Figure S16). Furthermore, DTT did
7
9
0
not affect the ratio of 5 versus 5 deiodination, as the formation of
T3 was not observed during the deiodination of T4 even at very
high concentrations of DTT. This is in agreement with the effect
9
of DTT on ID-3-mediated deiodination reactions.
It has been shown that the regioselectivity of the deiodination
of T4 and T3 by ID-1 is pH-dependent. While ID-1 catalyzes the
outer-ring deiodination of T4 to produce T3 at physiological pH,
this enzyme can remove iodine from the inner ring of T3 to
The investigations of the deiodinase activities of the amino-
substituted compounds 8ꢀ10 led to some interesting observa-
tions. While the introduction of an N-methylamino group into
compound 2 significantly reduced the deiodinase activity, com-
pound 10 having an N-methylamino group was found to be more
active than the corresponding unsubstituted compound, 5. A
comparison of the deiodinase activities indicated that compound
10 is almost 100 times more active than 9 in the conversion of T4
to rT3 (Figure 2). Similarly, the introduction of an N-methyl-
amino group into compound 3 significantly increased the deio-
dinase activity, whereas dithiol 8 having an N-methylamino
0
produce 3,3 -T2 at alkaline pH. In contrast, ID-3 and its Cys
mutant do not catalyze the outer-ring deiodination of T4, T3, or
rT3 over a wide pH range. Furthermore, the activity of ID-3 is
significantly enhanced under alkaline conditions. To under-
stand whether the pH profile for the inner-ring deiodination of
T3 by compound 5 is comparable to that of ID-3, the
deiodination experiments were carried out in buffers having
9
981
dx.doi.org/10.1021/ja201657s |J. Am. Chem. Soc. 2011, 133, 9980–9983

Journal of the American Chemical Society
COMMUNICATION
Scheme 2. Reactivities of Thiol and Selenol Moieties in
Proteins and Compounds 5 and 7 toward Iodoacetic Acid
(
IAA)
Figure 3. Abstraction of a proton from the thiol or selenol by an
N-methylamino group in compounds 8ꢀ10.
Table 1. Catalytic Parameters for the Monodeiodinations of
a
T4 by Compounds 2, 3, 5, and 8ꢀ10
compd Vmax (μM min
ꢀ
1
ꢀ3
ꢀ1
ꢀ1
ꢀ1
m
) K (mM) 10 kcat (min ) η (M min )
2
3
5
8
9
0
0.78 ( 0.01
0.76 ( 0.01
1.68 ( 0.07
0.74 ( 0.02
0.65 ( 0.01
0.13 ( 0.01
0.17 ( 0.01
1.57 ( 0.02
1.52 ( 0.03
11.64 ( 0.51
8.96 ( 0.57
12.49 ( 0.83
2.28 ( 0.06
15.34 ( 1.28
19.71 ( 1.85
b
2.58 ( 0.22 32.20 ( 0.56
0.65 ( 0.04
0.09 ( 0.01
1.47 ( 0.04
1.30 ( 0.01
b
1
0.66 ( 0.05
0.67 ( 0.11 13.10 ( 0.99
a
The parameter values were obtained from LineweaverꢀBurk plots. The
reactions were carried out in phosphate buffer (pH 7.5) at 37 °C using
0
.5ꢀ1.0 mM of T4, 0.5 mM of DTT, and 500 μM 2, 3, 8, or 9 or 50 μM 5
b
or 10. A relatively lower concentration of the catalyst (50 μM) was
employed for the assays, as no saturation kinetics was observed at a
concentration of 500 μM under identical experimental conditions.
value obtained for compound 10 (0.67 ( 0.11 mM) is much
lower than that of 5 (2.58 ( 0.22 mM), indicating that the
deprotonation of the selenol moiety in 10 by the secondary
amino group facilitates the interaction of this compound with T4.
It has been shown that one of the histidine residues (His174) at the
substituent was found to be ∼1.6 times more active than 3,
indicating that the basic amino group may deprotonate one of the
thiol groups to produce a more reactive thiolate. Interestingly,
the increase in the activity of dithiol 3 upon introduction of an
amino group was almost identical to that of diselenol 5. These
observations strongly suggest that the relative nucleophilicity
of the two functional groups (thiolꢀthiol, thiolꢀselenol, and
8
active site of ID-1 is critically involved in substrate binding. In this
particular case, the His174Asn mutant exhibited a much higher Km
8b
(
20.9 ( 2.85 μM) than the wild-type enzyme (0.22 ( 0.04 μM).
A comparison of the catalytic efficiencies (η) indicated that
compound 10 having two selenol moieties is almost 9 times more
active than the corresponding thiol 8, although a relatively lower
concentration of 10 was used for the determination of the Vmax and
11
selenolꢀselenol) plays an important role in the deiodination.
7
7
In the Se NMR spectrum of compound 9, the signal for this
compound (δ = 107 ppm) was shifted upfield relative to that for
K values. This is in agreement with the observations of Kuiper
m
9
77
2
(δ = 162 ppm). Similarly, the Se NMR signal for one of the
et al. that the relative substrate turnover number for wild-type ID-3
is 6-fold higher than that of the Cys mutant.
selenol groups in compound 10 (δ = 76 ppm) was shifted upfield
significantly relative to that of the other selenol in the same
compound (δ = 214 ppm) or that in compound 5 (δ = 156
ppm). These data suggest that the secondary amino group in
compounds 8ꢀ10 can increase the nucleophilic reactivity by
deprotonating the adjacent thiol or selenol moiety (Figure 3).
When two thiols (compound 8) or two selenols (compound 10)
are present, the increase in the reactivity of one of the thiol or
selenol groups may enhance the cooperative binding of these
groups to T4. In compound 9, the drastic increase in the reac-
tivity of the selenol relative to that of the thiol in the same com-
pound maydisfavorthecooperative interactions of the thiolꢀselenol
pair with the substrate. This may account for the differences
observed in the deiodinase activity upon introduction a basic amino
group in compounds 2, 3, and 5.
It has been shown that ID-1 activity is irreversibly inhibited by
gold thioglucose (GTG), 6-n-propyl-2-thiouracil (PTU), and
1,2
iodoacetic acid (IAA). While PTU reacts with the selenenyl
iodide (ID-1-SeI) intermediate to form a selenenyl sulfide as a
dead-end product, GTG and IAA may react with the selenol
group of the enzyme. Interestingly, ID-2 and ID-3 are very less
1,2
sensitive to the effect of these inhibitors. Although all three
deiodinases contain a Sec residue at the catalytic center, the
reason for the difference in their sensitivities to these inhibitors is
still unclear. It is known that IAA is an irreversible inhibitor of
12
several Cys peptidases, the Sec-containing glutathione perox-
13
14
idase (GPx), and thioredoxin reductase (TrxR). The catalytic
Cys and Sec residues in these enzymes react with IAA to produce
the corresponding alkylated derivatives (Scheme 2). When
compound 5 was treated with IAA, the reaction did not produce
the expected Se-carboxymethylated compound 11 but instead pro-
duced diselenide 12 with the elimination of acetic acid (Scheme 2).
On the other hand, compound 7 having a SePh substitutent instead
of the selenol group underwent facile carboxymethylation by IAA to
produce compound 13 as the major product. The deiodination of
IAA by compound 5 was found to be much faster than that of T4 or
To further understand the effect of the T4 concentration on
the rate of deiodination, we carried out detailed kinetic experi-
ments. The catalytic parameters [maximum velocity (Vmax),
Michaelis constant (K ), turnover number (k ), and catalytic
m
cat
efficiency (η)] for the deiodinations of T4 in the presence of
compounds 2, 3, 5, and 8ꢀ10 were obtained from the corre-
sponding LineweaverꢀBurk plots (Table 1). These data indicate
that the introduction of a basic amino group enhances the
binding of the thiolꢀselenol moiety to the substrate. The Km
15
T3. These observations suggest that IAA may undergo rapid
9
982
dx.doi.org/10.1021/ja201657s |J. Am. Chem. Soc. 2011, 133, 9980–9983

Journal of the American Chemical Society
COMMUNICATION
deiodination in the presence of ID-3, which may account for the
insensitivity of this enzyme toward IAA.
(8) (a) K €o hrle, J.; Hesch, R. D. Horm. Metab. Res., Suppl. Ser. 1984,
4, 42–55. (b) Berry, M. J. J. Biol. Chem. 1992, 267, 18055–18059.
1
(9) Kuiper, G. G. J. M.; Klootwijk, W.; Visser, T. J. Endocrinology
In conclusion, we have shown that naphthyl-based com-
pounds having two selenol groups are remarkably more efficient
mimics of iodothyronine deiodinase than ones having two thiol
groups or a thiolꢀselenol pair. An increase in the reactivity of the
selenol by introducing a basic amino group or increasing the pH
of the reaction medium does not change the regioselectivity of
deiodination. The cooperative effects of nucleophilic thiol and
selenol groups play an important role in the inner-ring deiodina-
tion of thyroxine. This study suggests that an increase in the
nucleophilic reactivity of the conserved Cys residue at the active
site of ID-3 is very important for effective deiodination, as an
activated Cys may cooperate well with the Sec residue in
polarizing the CꢀI bonds in T4 and T3. The insensitivity of
ID-3 toward the ID-1 inhibitor iodoacetic acid (IAA) can be
ascribed to the facile inactivation of IAA by ID-3-mediated
deiodination.
2
003, 144, 2505–2513.
(10) It has been shown that ∼50% of the DTT molecules undergo
monodeprotonation at pH 9.0. For details, see: Mottley, C.; Mason, R. P.
J. Biol. Chem. 2001, 276, 42677–42683.
(11) Nucleophilic attack of a thiol or selenol at the positively
charged iodine is consistent with the mechanism previously proposed
by Goto et al.
(12) (a) Chave, K. J.; Galivan, J.; Ryan, T. J. Biochem. J. 1999,
343, 551–555. (b) Bruno, M. A.; Pardo, M. F.; Caffini, N. O.; L ꢀo pez,
L. M. I. J. Protein Chem. 2003, 22, 127–134.
6
(
13) Mauri, P.; Benazzi, L.; Floh ꢀe , L.; Maiorino, M.; Pietta, P. G.;
Pilawa, S.; Roveri, A.; Ursini, F. Biol. Chem. 2003, 384, 575–588.
14) Zhong, L.; Arn ꢀe r, E. S. J.; Holmgren, A. Proc. Natl. Acad. Sci. U.S.A.
000, 97, 5854–5859.
15) The deiodination of IAA could be conveniently followed by H
(
2
1
(
NMR spectroscopy. When IAA was added to compound 5, rapid and
quantitative conversion of IAA to acetic acid was observed. Similarly,
iodoacetamide, bromoacetic acid, and bromoacetamide were dehaloge-
nated by 5, although the debromination was found to be much slower
than the deiodination.
’
ASSOCIATED CONTENT
S
Supporting Information. Details of experimental proce-
b
dures, chemical syntheses, and assay conditions. This material is
available free of charge via the Internet at http://pubs.acs.org.
’
AUTHOR INFORMATION
Corresponding Author
mugesh@ipc.iisc.ernet.in
’
ACKNOWLEDGMENT
This study was supported by the Department of Science and
Technology (DST), New Delhi. G.M. acknowledges the DST for
the award of a Swarnajayanti Fellowship, and D.M. thanks the
Indian Institute of Science for a fellowship. We thank Prof. Josef
K €o hrle for helpful discussions.
’
REFERENCES
(1) (a) Behne, D.; Kyriakopoulos, A.; Meinhold, H.; K €o hrle,
J. Biochem. Biophys. Res. Commun. 1990, 173, 1143–1149. (b) Leonard,
J. L.; Visser, T. J. In Thyroid Hormone Metabolism; Hennemann, G., Ed.;
Marcel Dekker: New York, 1986; p 189. (c) Berry, M. J.; Banu, L.;
Larsen, P. R. Nature 1991, 349, 438–440. (d) Larsen, P. R.; Berry, M. J.
Annu. Rev. Nutr. 1995, 15, 323. (e) Leonard, J. L.; K €o hrle, J. In The
Thyroid; Braverman, L. E., Utiger, R. D., Eds.; Lippincott-Raven:
Philadelphia, 1996; p 144. (f) St. Germain, D. L.; Galton, V. A. Thyroid
1
997, 7, 655.
2) (a) Bianco, A. C.; Salvatore, D.; Gereben, B.; Berry, M. J.; Larsen,
P. R. Endocr. Rev. 2002, 23, 38–89. (b) K €o hrle, J. Methods Enzymol. 2002,
47, 125–167. (c) Kuiper, G. G. J. M.; Kester, M. H. A.; Peeters, R. P.;
Visser, T. J. Thyroid 2005, 15, 787–798.
3) (a) Visser, T. J.; Schoenmakers, C. H. H. Acta Med. Austriaca
(
3
(
1
992, 19, 18–21. (b) K €o hrle, J. Mol. Cell. Endocrinol. 1999, 151,
103–119.
(
(
4) Bianco, A. C.; Kim, B. W. J. Clin. Invest. 2006, 116, 2571–2579.
5) (a) Beck, C.; Jensen, S. B.; Reglinski, J. Bioorg. Med. Chem. Lett.
1
994, 4, 1353–1356. (b) Vasil’ev, A. A.; Engman, L. J. Org. Chem. 1998,
63, 3911–3917.
(
6) Goto, K.; Sonoda, D.; Shimada, K.; Sase, S.; Kawashima, T.
Angew. Chem., Int. Ed. 2010, 49, 545–547.
7) Manna, D.; Mugesh, G. Angew. Chem., Int. Ed. 2010, 49, 9246
249.
(
9
9
983
dx.doi.org/10.1021/ja201657s |J. Am. Chem. Soc. 2011, 133, 9980–9983