Active Site Binding Is Not Sufficient for Reductive Deiodination by Iodotyrosine Deiodinase

Article abstract of DOI:10.1021/acs.biochem.6b01308

The minimal requirements for substrate recognition and turnover by iodotyrosine deiodinase were examined to learn the basis for its catalytic specificity. This enzyme is crucial for iodide homeostasis and the generation of thyroid hormone in chordates. 2-

Full text of DOI:10.1021/acs.biochem.6b01308

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN
Article
Active Site Binding is not Sufficient for Reductive
Deiodination by Iodotyrosine Deiodinase
Nattha Ingavat, Jennifer M. Kavran, Zuodong Sun, and Steven E. Rokita
Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.6b01308 • Publication Date (Web): 03 Feb 2017
Downloaded from http://pubs.acs.org on February 7, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 37
Biochemistry
1
2
3
For revision for Biochemistry
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Active Site Binding is not Sufficient for Reductive
Deiodination by Iodotyrosine Deiodinase
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
‡,§
†
,†
Nattha Ingavat, Jennifer M. Kavran, Zuodong Sun, and Steven E. Rokita*
†
Department of Chemistry, Johns Hopkins University, 3400 North Charles Street,
Baltimore, Maryland 21218 United States
‡
Department of Biochemistry and Molecular Biology, Johns Hopkins Bloomberg School of Public
Health, 615 North Wolfe Street Baltimore, Maryland 21205 United States
§
Department of Biophysics and Biophysical Chemistry, Johns Hopkins School of Medicine, 925
North Wolfe Street Baltimore, Maryland, 21205 United States
1
ACS Paragon Plus Environment

Biochemistry
Page 2 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract:
The minimal requirements for substrate recognition and turnover by iodotyrosine deiodinase were
examined to learn the basis for its catalytic specificity. This enzyme is crucial for iodide
homeostasis and the generation of thyroid hormone in chordates. 2ꢀIodophenol binds only very
weakly to the human enzyme and is dehalogenated with a k /K that is more than 4ꢀorders of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
cat
m
magnitude lower than that for iodotyrosine. This discrimination likely protects against a futile cycle
of iodinating and deiodinating precursors of thyroid hormone biosynthesis. Surprisingly, a very
similar catalytic selectivity was expressed by a bacterial homolog from Haliscomenobacter
hydrossis. In this example, discrimination was not based on affinity since 4ꢀcyanoꢀ2ꢀiodophenol
bound to the bacterial deiodinase with a K lower than that of iodotyrosine and yet was not
d
detectably deiodinated. Other phenols including 2ꢀiodophenol were deiodinated but only very
inefficiently. Crystal structures of the bacterial enzyme with and without bound iodotyrosine are
nearly superimposable and quite similar to the corresponding structures of the human enzyme.
Likewise, the bacterial enzyme is activated for single electron transfer after binding to the substrate
analog fluorotyrosine as previously observed with the human enzyme. A coꢀcrystal structure of
bacterial deiodinase and 2ꢀiodophenol indicates that this ligand stacks on the active site FMN in a
orientation analogous to that of bound iodotyrosine. However, 2ꢀiodophenol association is not
sufficient to activate the FMN chemistry required for catalysis, and thus the bacterial enzyme
appears to share a similar specificity for halotyrosines even though their physiological roles are
likely very different from those in humans.
2
ACS Paragon Plus Environment

Page 3 of 37
Biochemistry
1
2
3
Iodotyrosine deiodinase (IYD) was first identified in nature by its ability to salvage iodide from
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
iodotyrosine. Iodide is an essential micronutrient for chordates if only for its iodination of
thyroglobulin that serves as a precursor and template for generating the thyroid hormones thyroxine
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
and triiodothyronine as well as their byproduct iodotyrosine (IꢀTyr). Chronic iodide deficiency is a
3
result of a defective IYD, and more generally, insufficient or excess iodide can cause severe
4
,5
disorders that affect human development and health. Iodide generation by IYD also provides a
rare example of reductive dehalogenation in aerobic organisms and is even more unusual for its
6
dependence on flavin mononucleotide (FMN) (Scheme 1). This enzyme also catalyzes
7,8
dehalogenation of chloroꢀ and bromotyrosine but appears limited to processing simple tyrosine
derivatives. Thyroxine (Oꢀ(4ꢀhydroxyꢀ3,5ꢀdiiodophenoxy)ꢀ3,5ꢀdiiodoꢀLꢀtyrosine) is also a tyrosine
9
derivative but is not dehalogenated by IYD and may not template a productive orientation of its
10,11
active site residues.
Similarly, iodinated tyrosine residues on the surface of thyroglobulin are not
predicted to serve as substrates for IYD due to their lack of a zwitterion for coordination with
1
0,11
IYD.
Scheme 1. Iodotyrosine deiodinase (IYD) reduces IꢀTyr as it oxidizes its bound FMNhq.
The selectivity of IYD for IꢀTyr is essential to prevent a futile cycle of iodinating and
deiodinating thyroxine and its precursors that could lead to thyroid disease. Sterics alone may
exclude large compounds containing halogens from the active site of IYD but in the absence of
1
0,11
substrate, its bound FMN is highly accessible
and sensitive to inhibition by many polycyclic
12
haloaromatics such as triclosan, polychlorinated biphenyls and polybrominated diphenyl ethers. If
proximity to the reduced form of FMN (FMN ) were sufficient for dehalogenation than perhaps the
hq
3
ACS Paragon Plus Environment

Biochemistry
Page 4 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
halophenol component alone might act as a minimal recognition unit. In coꢀcrystals of mammalian
IYD and IꢀTyr, the iodophenol component of IꢀTyr stacks directly over the FMN and the aryl iodide
bond is positioned above the region of FMN that is associated with electron and hydride transfer
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(C4a, N5). Aromatic stacking between substrates and FMN is not uncommon and was previously
1
3,14
observed for nitroreductases that are in the same superfamily as IYD.
A similar substrateꢀFMN
1
5
stacking was also evident for an azoreductase from Pseudomonas. In addition, stacking between
FMN and aromatic side chains of the surrounding protein is common and can dramatically influence
1
6
the redox potential of FMN.
Iodophenol and a few of its derivatives have now been examined with IYD to measure the
intrinsic ability of this minimal unit to associate and react with the FMN. In addition, a potential
correlation between ligand recognition and turnover has been examined as a basis for substrate
selection during reductive dehalogenation by IYD. Studies below reveal that binding of 2ꢀ
iodophenol (2IP) to human IYD is very weak relative to that of IꢀTyr. The origin of this specificity
was subsequently challenged by using an alternative IYD from a bacterium, Halisscomenobacter
hydrossis, for which the physiological substrate and metabolic role has not yet been defined. This
IYD (hhIYD) demonstrated less stringency in substrate binding. Xꢀray crystal structures for hhIYD
in the alternative presence of IꢀTyr and 2IP are now described below and reveal that aromatic
stacking between substrate and FMN is not sufficient to support efficient turnover.
■
MATERIALS AND METHODS
General Materials. Reagents of the highest grade available were used without purification. 3ꢀ
Iodoꢀ
Lꢀtyrosine (IꢀTyr), 2ꢀiodophenol (2IP), mꢀcresol and formic acid (88%) were obtained from
Acros Organics. 3ꢀFluoroꢀ
L
ꢀtyrosine (FꢀTyr) was purchased from Astatech, Inc. 4ꢀHydroxybenzoic
acid was obtained from Alfa Aesar. 4ꢀCyanoꢀ2ꢀiodophenol (2IPCN) was purchased from
4
ACS Paragon Plus Environment

Page 5 of 37
Biochemistry
1
2
3
Pharmabridge, Inc. 4ꢀAminoꢀ2ꢀiodophenol (4A2IP), xanthine and xanthine oxidase were obtained
from SigmaꢀAldrich.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4ꢀHydroxyꢀ3ꢀiodobenzoic Acid (2IPCOOH). 4ꢀHydroxybenzoic acid (830 mg, 6.0 mmol) was
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
o
dissolved in conc. NH OH (aq) (29% w/w, 100 ml) and cooled to 4 C while stirring. Iodine (1.5 g,
4
6.0 mmol) in EtOH (18 ml) was added dropwise to this solution over 1 h and the resulting mixture
o
was stirred for another 2.5 h at 4 C. The mixture was rotoevaporated to a reduced volume of 40 ml
and then acidified to pH 2 by addition of HCl (1 N). The mixture was further rotoevaporated to
dryness and the solid residue was dissolved in ethyl acetate (20 ml), washed twice with H O (20 ml)
2
and rotoevaporated to dryness again. The crude product was dissolved in 20% aq. acetonitrile with
0
.44% formic acid (v/v) and purified by reverseꢀphase HPLC (Econosphere C18 column; 250 x 10
mm, 10 ꢁm) using a gradient of 20% to 30% acetonitrile in aqueous 0.44 % (v/v) formic acid over
1
2
8
5 min (5 ml/min). Fractions containing the product (t = 7 min) were lyophilized to yield
R
1
IPCOOH as white solid (360 mg, 23 % yield). H NMR (400 MHz, acetoneꢀd ) δ 7.05 (d, 1H, J =
6
13
.4 Hz), 7.92 (dd, 1H, J = 2.0, 8.5 Hz), 8.40 (d, 1H, J = 2.0 Hz) (Figure S1); C NMR (101 MHz,
1
13
acetoneꢀd ) δ 82.9, 114.5, 123.8, 131.6, 141.2, 160.7, 165.4 (Figure S2). Both H and C NMR
6
1
7
spectra are consistent with the literature.
IYD Preparation. Human IYD without its Nꢀterminal transmembrane sequence (residues 2ꢀ31)
1
1
was expressed as a SUMO fusion as described previously. IYD from H. hydrossis (hhIYD) does
not contain a membrane binding sequence and was expressed as its fullꢀlength native sequence
without a SUMO fusion. The hhIYD gene was amplified from a previous construct containing a
1
8
SUMO fusion using primers 5’ꢀAATTAATCATATGAAGCAAAAGCCTGCTTꢀ3’ and 5’ꢀ
AATTAATCTCGAGCTAGTGATGGTGATGꢀ3’. This provided a 5’ꢀflanking NdeI site and a 3’ꢀ
flanking XhoI site that were used to subclone the gene into a pETꢀ24a(+) vector. The resulting
hhIYD gene contained a Cꢀterminal His tag. Both human IYD and hhIYD were expressed and
6
5
ACS Paragon Plus Environment

Biochemistry
Page 6 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2+
11,18
purified (Ni affinity and size exclusion chromatography) as described previously.
induction conditions for both IYDs were modified by lowering the isopropyl βꢀ
thiogalactopyranoside concentration from 400 ꢁM to 20 ꢁM, decreasing the induction temperature
Only the
D
ꢀ1ꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
o
o
from 18 C to 16 C and increasing the induction time from 4 h to 12ꢀ16 h.
General Methods. Protein concentrations for crystallization studies were determined by A280
ꢀ1
ꢀ1
ꢀ1
ꢀ1
(
ε280 = 37,930 M cm for human IYD and ε = 31,400 M cm for hhIYD as estimated by
280
1
9
ExPASy ProtParam tool) after correcting for the contribution of its bound FMN (A /A , 1.9 ±
280 450
0
.1 for human IYD and 1.72 ± 0.04 for hhIYD). Enzyme concentrations for kinetic and binding
ꢀ1
ꢀ1
studies were determined by A450 derived from the bound FMN (ε , 13,000 ± 300 M cm for
450
ꢀ
1
ꢀ1
human IYD and 13,600 ± 300 M cm for hhIYD) as determined from the native enzyme and free
ꢀ1
ꢀ1 20
FMN (ε450 = 12,500 M cm ) released after heat denaturation and removal of the protein by
2
+
centrifugation and Ni affinity chromatography. Dissociation constants (K ) for the ligands of IYD
d
were determined by their ability to quench the fluorescence of the oxidized FMN (FMN ) (λ 450,
ox
ex
1
1,21
λem 523 nm) in the active site of IYD as previously described.
Catalytic Deiodination. Catalytic turnover of IYD was observed by quantifying the product of
22
deiodination after separation by reverseꢀphase HPLC as previously described. Briefly, iodinated
substrates were incubated with hhIYD in 100 mM potassium phosphate pH 7.4 and reaction was
initiated by addition of 0.5% dithionite in 5% sodium bicarbonate. This mixture was incubated at 25
o
C for the specified time and then quenched with 88% formic acid. Internal standards of either mꢀ
cresol or benzoic acid were used to quantify formation of each product. See Figure S5ꢀS7 for more
details. Studies with human IYD followed the same procedures but contained an additional 200 mM
KCl to stabilize the enzyme.
Anaerobic reduction of hhIYD by xanthine and xanthine oxidase. Solutions of hhIYD (20
ꢀ
M), potassium phosphate (100 mM, pH 7.4), sodium chloride (500 mM), glycerol (10% v/v),
6
ACS Paragon Plus Environment

Page 7 of 37
Biochemistry
1
2
3
xanthine (1 mM) and methyl viologen (13 µM) were purged of molecular oxygen by gently bubbling
with Ar for 30 min. Reduction was initiated by addition of xanthine oxidase (140 µg/mL) as
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1,23,24
previously described.
Enzyme reduction was repeated in the alternative presence of FꢀTyr
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(500 µM) and 2IP (800 µM) to study the effect of ligand on the redox properties of hhIYD.
Crystallization. Crystallization conditions were determined using sparse matrix screening with
commercially available screens and a Phoenix robot. Crystals of hhIYD without ligand were grown
by mixing 1 µl of protein (13 mg/ml, in 50 mM sodium phosphate pH 7.4, 100 mM NaCl, 0.5 mM
TCEP and 10 % glycerol) with 0.8 µl of well solution containing 80 mM TrisꢀHCl pH 8.5, 100 mM
MgCl , 24% (w/v) PEG 4000 and 20% glycerol. To obtain coꢀcrystals of hhIYD and IꢀTyr, hhIYD
2
1
3 mg/ml (final), in 25 mM HEPES pH 7.4, 0.5 mM TCEP and 10 % glycerol was incubated with 2
o
mM (final) of IꢀTyr overnight at 4 C. Crystals were grown by mixing 0.8 µl of hhIYD•IꢀTyr with 1
µl of well solution containing 100 mM HEPES pH 7.5, 150 mM MgCl , 25% (w/v) PEG 3350, 2
2
mM IꢀTyr and 15% glycerol. To obtain a complex of hhIYD with 2IP, hhIYD (15 mg/ml, final) in
2
5 mM HEPES pH 7.4, 0.5 mM TCEP and 10 % glycerol was incubated with 2 mM (final) of 2IP
o
overnight at 4 C. Crystals were grown by mixing 1 µl of hhIYD•2IP with 0.8 µl of well solution
containing 100 mM TrisꢀHCl pH 9, 200 mM MgCl , 2 mM 2IP, 17% (w/v) PEG 8000, 17% glycerol
2
o
and 10% (v/v) ethylene glycol. Crystals were grown at 20 C by hanging drop vapor diffusion and
appeared within 3 days. Crystals were frozen in liquid nitrogen directly from the mother liquor prior
to data collection.
Data Collection and Structure Determination. Diffraction data were collected at the 12ꢀ1 and
7
ꢀ1 beamlines at the Stanford Synchrotron Radiation Lightsource (SSRL) and processed with either
25
26
XDS/Aimless (for PDB 5KO7 and 5KO8) or HKL2000/Scalepack (for PDB 5KRD). Phases
27
were determined by molecular replacement with PHASER using the mouse IYD structure (PDB
1
0
3
GB5) as the search model. All three structures were built and refined by iterative rounds of
7
ACS Paragon Plus Environment

Biochemistry
Page 8 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
28
29
modelꢀbuilding in COOT and refinement in PHENIX. All structural figures were drawn with
PyMOL (Schrödinger, LLC).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
■
RESULTS
Expression and Purification of IYD. Expression of mammalian IYD had been problematic
1
1,30
until development of a His ꢀtagged SUMOꢀfusion.
This construct reproducibly provides soluble
6
and active enzyme in good yield from Escherichia coli. SUMO fusions were hence used when
18
testing the activity of IYD homologs from other organism including H. hydrossis. In this and other
examples, purification was often complicated by coꢀelution of IYD and the hydrolyzed SUMO. In
attempt to optimize IYD purity for crystallization trials, the native sequence of hhIYD was expressed
without a SUMO fusion. Previously, IYD from Drosophila melanogaster was successfully prepared
2
2
under these conditions, and hhIYD provided similar success. Human IYD from the IYDꢀSUMO
fusion and native hhIYD lacking an attached SUMO were both isolated in yields of 15ꢀ24 mg/L of
culture with purity of > 98% as evident from SDSꢀPAGE, coomassie staining and ImageQuantTL
analysis (Figure S3). The size of the IYD polypeptides indicated by denaturing gel electrophoresis
(
30 kDa and 25 kDa for human IYD and hhIYD, respectively) are consistent with the estimates
3
1
generated by ExPASy. Native IYD exists in solution as a homodimer and preparations contained
ratios of 1.9 – 2.1 FMN per dimer as measured by A450 and A . This ratio indicates approximately
280
1
8
full loading of the cofactor based on the theoretical maximum of 2.0.
Substrate and Ligand Binding to IYD. Binding of tyrosine and phenol derivatives to the
7
active site of IYD can be monitored by their ability to quench the fluorescence of its FMN . To
ox
assess the contribution of the aromatic component of IꢀTyr on binding, affinities of IꢀTyr and 2IP
were initially compared using human IYD. Previously, the K of IꢀTyr was observed in the subꢀꢀM
d
8
ACS Paragon Plus Environment

Page 9 of 37
Biochemistry
1
2
3
range and a similar value of 0.09 ꢀM was measured in this study (Figure 1, Table 1). In contrast,
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2IP expressed only a weak affinity for human IYD as evident by an increase in K by more than 4ꢀ
d
orders of magnitude. Thus, the human enzyme requires the zwitterion of IꢀTyr for tight binding and
is likely limited to substrates containing such a zwitterion in vivo. This selectivity is also essential
for proper thyroid function.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Quenching of FMN fluorescence as
ox
a monitor for substrate binding to IYD. Human
IYD (4.3 ꢁM and 200 mM KCl) and hhIYD (2.5
ꢁM) in 100 mM potassium phosphate pH 7.4
o
were stirred at 25 C for 30 min prior to addition
of the indicated ligand. The K values were
d
determined from the best fit to quenching of
FMN fluorescence using λ and λ of 450
ox
ex
em
nm and 523 nm, respectively, as described
1
1
previously.
The necessity of a similar selectivity in lower organisms and particularly bacteria is not apparent
since the generation and function of halotyrosines in these organisms have not yet been
3
2,33
established.
The role of the aromatic component of IꢀTyr in the absence of its zwitterion was
therefore interrogated with hhIYD as well. Using the same fluoresence assay, IꢀTyr was found to
bind hhIYD with a ~90ꢀfold weaker affinity than human IYD (Figure 1, Table 1). In contrast, 2IP
exhibited a ~20ꢀfold increase in affinity for hhIYD relative to human IYD. Thus, the halophenol
9
ACS Paragon Plus Environment

Biochemistry
Page 10 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
component contributes more to the binding of IꢀTyr with hhIYD than with human IYD. This
difference had the potential to suggest a broader substrate profile for hhIYD versus human IYD.
Table 1. Binding affinity and kinetic parameters for human IYD and hhIYD.
c
ꢀ2 ꢀ1 d
d
ꢀ1 ꢀ1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
IYD
Substrate
R
pK (phenol)
K (ꢁM)
d
kcat (x 10 s )
K (mM)
m
k /K (M s )
cat m
a
a
b
4
Human
IꢀTyr
2IP
ꢀ
8.4, 8.3
0.09 ± 0.02
1,410 ± 70
10.2 ± 0.4
0.0073 ± 0.0008
4.1 ± 0.4
(1.4 ± 2) x 10
1.0 ± 0.1
a
b
H
8.4, 8.5
0.40 ± 0.02
hhIYD
a
b
b
4
IꢀTyr
ꢀ
8.4, 8.3
8.2 ± 0.5
67 ± 2
27 ± 2
0.0012 ± 0.0002
4.1 ± 0.8
(23 ± 4) x 10
a
2
4
2
2
IP
H
8.4, 8.5
0.29 ± 0.03
0.7 ± 0.2
a
f
A2IP
NH2 9.3
96 ± 3
–
–
4.1 ± 0.3
ꢀ
a
a
IPCOOH COO 8.6
24.1 ± 0.7
1.44 ± 0.05
0.31 ± 0.01
7.1 ± 0.6
0.44 ± 0.04
ꢀ3e
IPCN
CN
6.7
< 2.1 x 10
–
–
a
pK values of the phenolic protons were calculated with the ACE and JChem acidity and basicity
a
b
calculator (https://epoch.uky.edu/ace/public/pKa.jsp). pK values were also determined
a
5
6ꢀ58
c
d
experimentally as published previously
Determined from data of Figures 1 and S4. Determined
e
from data of Figure S7. Estimated from the detection threshold of 4ꢀcyanophenol (Figure S6B).
f
Estimated from Lineweaver–Burk analysis. Errors derive from least squares fitting.
The promiscuity of halophenol binding to hhIYD was further challenged by three additional
derivatives alternatively adding a charge (4ꢀhydroxyꢀ3ꢀiodobenzoate, 2IPCOOH), an electron
withdrawing group (4ꢀcyanoꢀ2ꢀiodophenol, 2IPCN) or an electron donating group (4ꢀaminoꢀ2ꢀ
iodophenol, 4A2IP) to the parent iodophenol. Previous studies with human IYD demonstrated
11
preferential binding to the phenolate form of diiodotyrosine (I ꢀTyr) and hence the shift in acidity
2
caused by the cyano and amino substituents were used to test if a similar preference would also
extend to the binding of simple phenols to hhIYD. The calculated pK for 4A2IP is only slightly
a
higher than that for 2IP but, as anticipated, the K value observed for 4A2IP is 40% larger than that
d
for 2IP (Table 1). However, this difference may not be significant since much of the fluorescence
1
0
ACS Paragon Plus Environment

Page 11 of 37
Biochemistry
1
2
3
quenching data of 2IP and 4A2IP are experimentally indistinguishable (Figure S4). Differences in
the pK and K values of 2IP and 2IPCN are considerably more pronounced. The acidity of 2IPCN
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
d
is greater by 2 pK units than that of 2IP and the K value of 2IPCN is almost 50ꢀfold lower than that
a
d
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
of 2IP (Table 1 and Figure S4). 2IPCN affinity for hhIYD is even greater than that of IꢀTyr. This
could imply that zwitterion recognition may not be crucial for hhIYD catalysis but results below
highlight differences in the criteria for binding and turnover. Both the acidity of 2IPCOOH and its
affinity for hhIYD are intermediate between 2IP and 2IPCN (Table 1), and the presence of an
additional negative charge of the carboxylate does not significantly affect recognition by the active
site of hhIYD.
Catalytic Deiodination. Enzyme turnover of each potential substrate was monitored by
formation of its deiodinated product after separation by reverseꢀphase HPLC (Figures S5 and S6)
and the kinetic constants were determined by measuring initial rates as a function of substrate
concentration (Figure S7). As typical for IYD, dithionite was used as the reductant since the
3
4
physiological system for reduction has yet to be identified. The k value for deiodination of IꢀTyr
cat
ꢀ
by human IYD (Table 1) is nearly identical to that recently determined for Drosophila IYD (12 x 10
2
ꢀ1 22
s ) and in both examples their K values are more than 20ꢀfold (Drosophila) and 80ꢀfold (human)
m
greater than their K values. The steadyꢀstate parameter k /K for turnover of IꢀTyr by human IYD
d
cat
m
is also within twoꢀfold of its second order k for IꢀTyr dependent oxidation of FMN as measured
ox
hq
8
by rapid kinetics. Similarly, the k /K value for IꢀTyr is only 50% larger than that measured for I ꢀ
cat
m
2
125
11
Tyr using a complementary Iꢀiodide release assay. The lack of the zwitterion in 2IP greatly
compromises its turnover by human IYD. The kcat for 2IP is 26ꢀfold less than that for IꢀTyr and,
even more significantly, the K for 2IP is 560ꢀfold greater than that for IꢀTyr (Table 1). This large
m
increase in K is still less than the 16000ꢀfold increase in the K for 2IP versus IꢀTyr (Table 1).
m
d
1
1
ACS Paragon Plus Environment

Biochemistry
Page 12 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Despite the lower affinity of IꢀTyr for hhIYD relative to human IYD, hhIYD expresses a higher
catalytic efficiency as indicated by the 16ꢀfold increase in its k /K value (Table 1). Deiodination
cat
m
1
1,18
of IꢀTyr is also more efficient than that of I ꢀTyr for hhIYD as well as for human IYD.
The loss
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
in efficiency for deiodination of 2IP lacking the zwitterion of IꢀTyr is even more dramatic for
hhIYD. The decrease in k /K from IꢀTyr to 2IP is 33,000ꢀfold and twice as large as that for human
cat
m
IYD. The physiological substrate for hhIYD has not yet been identified, but clearly this enzyme is
not a general dehalogenase for halophenols. The low rates of 2IP turnover by both IYDs are still
faster than the intrinsic reactivity between the free cofactor FMNhq and 2IP. Under standard
conditions in the presence of FMN (2.5 ꢀM), 2IP (5 mM) and dithionite (0.5% w/v), no phenol was
detected above the threshold of ~ 3 nmol after a 2 h incubation (Figure S8). However, the enhanced
binding of 2IP to hhIYD versus human IYD did not result in a significant gain of catalytic
efficiency.
Turnover of the 2IP derivatives was also investigated to learn whether the active site affinity
of these minimal structures might still influence catalysis. Kinetics were again monitored by
formation of the deiodinated products after separation by reverseꢀphase HPLC (Figures S5 – S7).
2IPCOOH exhibited a kcat that was experimentally indistinguishable from that of 2IP and the k /Km
cat
values for both compounds were both quite minimal (Table 1). However, the K for 2IPCOOH was
m
7
5% larger than that for 2IP and contrary to the hierarchy of their K values. The activity of 4A2IP
d
was also not consistent with its relative binding affinity for hhIYD since its k /K value was almost
cat
m
6
ꢀfold greater than the parent 2IP (Table 1). Poor solubility of 4A2IP restricted its further
characterization. Likewise, the lack of detectable product provided only a lower estimate for
turnover of 2IPCN. Speculation on the electronic demands of dehalogenation from the data on
4
A2IP and 2IPCN is tempting but should be deferred until a comprehensive structureꢀactivity survey
1
2
ACS Paragon Plus Environment

Page 13 of 37
Biochemistry
1
2
3
can be completed. Still, the inability for IYD to deiodinate 2IPCN was initially surprising since
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
IPCN binds the resting enzyme even more tightly than IꢀTyr (Table 1).
Crystallographic Studies of hhIYD. The structures of hhIYD in the absence of substrate and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
bound to either IꢀTyr or 2IP were compared to their mammalian counterparts to examine the
structural basis for a lack of correlation between halophenol binding and catalysis. Crystals of
hhIYD, hhIYD•IꢀTyr and hhIYD•2IP diffracted Xꢀrays to 2.3 Å, 2.2 Å, and 2.1 Å, respectively, and
the phases of each were determined by molecular replacement using the mouse IYD structure as a
1
0
search model (Figure 2).
Data collection and refinement statistics are summarized in Table 2.
Unbiased electronꢀdensity indicated the position of substrates, and the signal from anomalous
difference maps was used to place the iodine atom of each ligand (Figure 3). Like its mammalian
homologs, the biological active form of hhIYD is an α dimer. The asymmetric unit contains the
2
biological dimer for hhIYD in the absence of ligand. The biological dimers are not contained within
the asymmetric unit for hhIYD in the presence of ligand (hhIYD•IꢀTyr and hhIYD•2IP) and were
instead generated by crystal symmetry. In each case the two subunits of each α dimer are nearly
2
identical and only one chain will be discussed unless otherwise indicated.
1
3
ACS Paragon Plus Environment

Biochemistry
Page 14 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 2. Structures of hhIYD and the environment surrounding its bound FMN. Cartoon
representation of the structure of the homodimers of (A) hhIYD (grey), (B) hhIYDꢂIꢀTyr (cyan) and
(C) hhIYDꢂ2IP (green). Individual polypeptides are indicated by their different shades of color and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
disordered regions are indicated by the dashed lines. The active site environment surrounding FMN
(carbon atoms in yellow) is illustrated for (D) hhIYD, (E) hhIYDꢂIꢀTyr and (F) hhIYDꢂ2IP.
Figure 3. Orientation of the substrateꢀFMN complex in
ox
the active site of hhIYD. Bound FMN is indicated by its
ox
yellow carbon atoms. (A) Two configurations of IꢀTyr
(carbons in cyan) are evident from the two positions of the
aryl iodide. (B) Only a single configuration of 2IP (carbons
in green) was detected. The unbiased F ꢀF difference maps
o
c
contoured at +3σ are shown by the green mesh. Anomalous
difference maps contoured at +3σ are shown by the orange
mesh.
1
4
ACS Paragon Plus Environment

Page 15 of 37
Biochemistry
1
2
3
Table 2. Data collection and refinement statistics
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Protein
hhIYD
FMN
hhIYD•IꢀTyr
hhIYD•2IP
FMN and 2IP
5KRD
Ligands
FMN and IꢀTyr
5KO8
Protein Data Bank code
5KO7
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Data collection
Space group
Unit cell parameters
a, b, c (Å)
α, β, γ ( )
Resolution (Å)
Wavelength (Å)
No. of unique reflections
No. of observed reflections
Completeness (%)
P 2 2 2
P 6 2 2
P 6 2 2
1
1
2
4
92.18, 110.88, 43.29
90, 90, 90
36.96–2.25 (2.32–2.25) 38.18–2.15 (2.22–2.15) 20.00–2.10 (2.17–2.10)
1.127
21,635 (1,685)
149,492
98.9 (88.4)
6.9 (5.1)
7.5 (145)
3.1 (69.4)
15.2 (1.1)
152.72, 152.72, 87.20 154.70, 154.70, 88.88
90, 90, 120 90, 90, 120
o
a
0.980
33,077 (2,795)
647,766
99.9 (100)
19.6 (20.4)
14.7 (375)
3.4 (84.0)
16.8 (1.1)
0.980
36,523 (3,607)
267,636
98.8 (99.5)
7.3 (7.2)
10.7 (100)
5.7 (100)
15.2 (1.0)
a
a
a
Redundancy
Rsym (%)
a,b
a,c
R
I/σI
(%)
p.i.m
a
Refinement
Resolution used in
refinement (Å)
Rwork (%)/Rfree (%)
No. of atoms
Protein
Heteroatoms
Solvent
Mean Bꢀfactor (Å )
Protein
Waters
FMN
IꢀTyr
2IP
RMSD from ideality
Bond lengths (Å)
Bond angles ( )
Ramachandran analysis
Favored (%)
Allowed (%)
35.4–2.3
21.6/25.4
38.2–2.2
19.1/23.2
6,925
20.0ꢀ2.1
17.7/22.1
6,401
d
6,336
100 (FMN)
62
174 (FMN and IꢀTyr) 116 (FMN and 2IP)
128
126
2
71.25
53.47
56.22
–
62.44
50.96
44.21
62.57
–
60.16
56.18
45.74
–
–
65.46
0.003
0.707
0.005
0.842
0.014
1.386
o
99.49
0.51
0
98.60
1.40
0
99.49
0.51
0
Outliers
a
The values in parentheses are for the highest resolution shell.
Rsym =ΣhklΣi |Ii ꢀ <I>| /ΣhklΣi Ii , where Ii is the intensity of an individual reflection and <I> is the
b
mean intensity obtained from multiple observations of symmetry related reflections.
c
Rp.i.m =Σhkl√((1/(nꢀ1)) Σi |Ii ꢀ <I>| / ΣhklΣi Ii, where Ii is the intensity of an individual reflection and <I> is the
mean intensity obtained from multiple observations of symmetry related reflections.
d
Rwork = Σhkl (|F | ꢀ | F | ) / Σhkl | Fobs | where Fobs is an observed amplitude and Fcalc a calculated amplitude;
obs
calc
Rfree is the same statistic calculated over a subset of the data that has not been used for refinement.
1
5
ACS Paragon Plus Environment

Biochemistry
Page 16 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Structure of hhIYD. The structure of hhIYD in the absence of substrate displays the common
3
5
features of the nitroꢀFMN reductase superfamily for which IYD is a member. Despite sharing only
4% sequence identity, human IYD and bacterial hhIYD have the same fold and superpose with an
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RMSD of 0.77 Å for 321 Cα atoms (Figure S9A). The two subunits of hhIYD form a dimer
containing an αꢀβ fold with domain swaps at the Nꢀ and Cꢀtermini. Each dimer has two active sites
that are formed by regions of each monomer, and each contain a FMN bound noncovalently (Figure
2A). In the absence of substrates, the structures of IYD from H. hydrossis, human, and mouse have
a disordered active site lid (Figure S9). In hhIYD, the loop adjacent to the active site (residues 133ꢀ
1
45) is ordered in the absence of substrate whereas in human IYD, this loop is only ordered upon Iꢀ
1
1
10
Tyr binding. Substrate binding is not even sufficient for fully ordering this loop in mouse IYD.
The enhanced stability of this loop in hhIYD may be rationalized by the rigidity introduced by
3
6
substitution of an Ala in human and mouse IYD for a Pro137 in hhIYD (Figure S10).
Additionally, Asp138 and Thr140 are within hydrogen bonding distances in hhIYD but are
10,11
substituted with Gln and Lys in human and mouse IYD, respectively.
Structure of hhIYD•IꢀTyr. Comparison of the structures of hhIYD in the presence and absence
of IꢀTyr revealed no large scale conformational changes (RMSD of 0.23 Å for 345 Cα atoms) upon
ligand binding except for an ordering of the active site lid (residues 95 ꢀ 112) (Figure 2B). Similar
conformational changes were observed in the structures of both mouse and human IYD bound with
1
0,11
substrates.
In each case, the active site lid is stabilized by interactions between the zwitterion of
IꢀTyr and a common set of residues that are illustrated by Glu91, Tyr95 and Lys116 in hhIYD
(Figure 2E). These residues are evident in all IYD homologs confirmed to date and have been used
18
to detect other likely deiodinases within the parent nitroꢀFMN reductase superfamily.
The
4
zwitterion of IꢀTyr also offers hydrogen bonding through its αꢀammonium group to the FMN O and
through its carboxylate group to the FMN N3. Other features that stabilize the substrateꢀenzyme
1
6
ACS Paragon Plus Environment

Page 17 of 37
Biochemistry
1
2
3
complex include hydrogen bonding between the phenolate of IꢀTyr and the amide backbone of Ala
64 and 2’ꢀHOꢀribityl of FMN and πꢀstacking between the aromatic region of IꢀTyr and the
isoalloxazine ring of FMN (Figure 2E). These interactions are also evident in substrate complexes
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0,11
formed by mouse and human IYD.
For hhIYD•IꢀTyr, the active site lid could be fully built for one active site but only partially built
for the other suggesting that the lid is not in a single fixed orientation when bound to IꢀTyr. The
electron density maps also revealed that IꢀTyr aligns in two orientations with respect to the iodo
1
0,11
substituent instead of the single orientation observed in the mammalian IYD structures.
Greater
density is evident for an alignment of the arylꢀI bond over the C4aꢀN5 bond of FMN and consistent
1
0,11
o
with that for mammalian IYD (Figure 3A).
The alternative conformation represents a ~ 180 flip
of the aryl ring. In this orientation, the arylꢀI bond resides above the pyrimidine ring of FMN and
toward Phe110 (3.3 Å). Phe110 likely stabilizes this alternative conformation by offering favorable
3
7
C−I···π interactions that are not available in the mammalian structures.
Structure of hhIYD•2IP. The structure of hhIYD bound with 2IP is similar to that bound with Iꢀ
Tyr but lacks an ordered active site lid. hhIYD•2IP also superposes with hhIYD lacking an active
site ligand with an RMSD of 0.30 Å for 337 Cα atoms (Figure S9D). Analysis of unbiased
difference maps for hhIYD bound with 2IP reveal electronꢀdensity for only the iodo group and
phenolate oxygen suggesting that the ring of 2IP is mobile (Figure 3B). The orientation of 2IP aligns
the arylꢀI bond over the C4aꢀN5 of FMN as observed previously for IꢀTyr bound to the mammalian
IYDs. By extrapolation from these structures, 2IP likely interacts with hhIYD through πꢀstacking
with the isoalloxazine ring of FMN and hydrogen bonding between the phenolate of 2IP and the
amide backbone of Ala64 and the 2’ꢀHOꢀribityl of FMN (Figure 2F). The distance between the iodo
substituent and N5 of FMN is relatively constant at ~ 4 Å for the structures containing either IꢀTyr or
1
7
ACS Paragon Plus Environment

Biochemistry
Page 18 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
IP (Figures 2E and 2F). In contrast to IꢀTyr, no zwitterion is provided by 2IP to coordinate the
FMN and impose order in the active site lid.
Reduction of FMN in the Active Site of hhIYD. The diversity of FMNꢀdependent catalysis in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
38ꢀ41
nature is based in part on its ability to function as an one and two electron acceptor and donor.
For example, reduction of nitroaromatic compounds by enzymes in the nitroꢀFMN reductase
superfamily is promoted by stabilizing a two electron pathway and destabilizing alternative one
4
2ꢀ44
electron pathways.
process at the expense of competing two electron processes.
demonstrated that the one electron reduced FMN semiquinone (FMN ) accumulated in the presence
Conversely, deiodination of IꢀTyr is promoted by stabilizing an one electron
1
1,44
Previous studies on human IYD
sq
1
1
of either a halotyrosine substrate (IꢀTyr) or its inert analogue fluorotyrosine (FꢀTyr). The ability of
hhIYD to support a similar response to FꢀTyr was confirmed by anaerobic reduction under standard
23,24
conditions using xanthine/xanthine oxidase.
Most importantly, this analysis was also repeated in
the presence of 2IP to learn the effect of base stacking on the stability of FMNsq.
In the absence of an active site ligand, the FMN of hhIYD (λmax = 446 nm and shoulder near
ox
4
70 nm) was reduced to FMN without detection of an intermediate FMN (λ = 590 nm) (Figure
hq sq max
S11). Next, binding of FꢀTyr to hhIYD was confirmed for its use as an inert mimic of IꢀTyr to
induce closure of the active site lid. By measuring quenching of FMN fluorescence in hhIYD as
ox
described above, FꢀTyr was shown to bind with a K of 46 ± 2 µM (Figure S12). More than a 10ꢀ
d
fold excess of FꢀTyr (500 ꢀM) was then added to hhIYD prior its reduction. Under these conditions,
accumulation of the FMNsq intermediate was clearly detected by the initial increase and then
1
1
decrease in A590 (Figure 4A) in a manner similar to that observed with human IYD.
1
8
ACS Paragon Plus Environment

Page 19 of 37
Biochemistry
1
2
3
4
5
Figure 4. Anaerobic reduction of FMN bound to hhIYD
by xanthine and xanthine oxidase as monitored by
absorbance spectroscopy. Sample spectra of hhIYD (20
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ
M) over 40 min indicate the reduction of FMN by loss of
absorbance (~ 450 nm) in the presence of (A) FꢀTyr (500
M) and (B) 2IP (800 ꢁM). Absorbance (~590 nm)
ꢀ
increases (black) and then decreases (grey) only during
reduction in (A).
The use of 2ꢀfluorophenol as an inert analogue of 2IP
was impractical since it bound only very weakly to hhIYD
(
K = 20 ± 1 mM) (Figure S12). Instead, 2IP was studied
d
directly for possible stabilization of the FMN intermediate.
sq
This was possible due to the very low efficiency of 2IP turnover. From its kcat value, less than one
equivalent could be dehalogenated in the time necessary for the redox titration. Consequently, 2IP
(
800 ꢀM) was added to hhIYD in greater than a 10ꢀfold excess above its K value (Table 1) prior to
d
initiating reduction. During reduction, 2IP did not detectably promote an accumulation of the
FMN intermediate in contrast to the effect of FꢀTyr (Figure 4B). This inability of 2IP to stabilize
sq
FMN to a detectable extent is also consistent with its very slow turnover.
sq
■
DISCUSSION
The simple halophenol 2IP binds to the active site of hhIYD in a manner equivalent to the
halophenol component of the substrate IꢀTyr. Crystallographic analysis reveals that both ligands
1
9
ACS Paragon Plus Environment

Biochemistry
Page 20 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
stack on the isoalloxazine ring of FMN, and the hhIYD structures are mostly superposable with the
corresponding structures of human and mouse IYD. In addition, the arylꢀI bond is positioned over
4
1
the C4aꢀN5 bond of FMN, a highly active region participating in redox catalysis.
Association
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
between 2IP and hhIYD is likely stabilized by hydrophobic and van der Waals interactions, πꢀ
stacking with FMN, and hydrogen bonding between the phenolate form of this ligand, the peptide
backbone and the 2’ꢀHO group of the ribityl portion of FMN. These same interactions are common
1
0,11
to all substrates and active site ligands of IYD.
However, 2IP binds only very weakly to human
IYD. The differences in affinity between human IYD and hhIYD might arise from the relative
dynamics and preꢀorganization of their active sites in the absence of ligand. Although both enzymes
exhibit disorder in the helical region of the active site lid (hhIYD residues 95ꢀ112) in the absence of
substrate, the adjacent loop (hhIYD residues 133ꢀ145) is ordered in hhIYD (Figure S9) and may
enhance the population of conformers that readily accommodate the halophenol ligand. This affinity
may be beneficial for the physiological function of IYD in bacteria but would be disadvantageous
for mammalian IYD which must carefully distinguish between its halotyrosine substrate and other
halophenol derivatives that are created by thyroid hormone metabolism.
Binding and orientation of the iodophenol component in the active site of IYD is not
sufficient to drive reductive dehalogenation. This is best demonstrated by the derivative 2IPCN that
binds to hhIYD with a greater affinity than IꢀTyr but does not participate in catalysis (Table 1). This
characteristic is not shared by other enzymes in the same structural superfamily for which reactivity
is often dictated by proximity. For example, NADH directly reduces FMN bound to nitroreductase
NfsB (E. coli) and nicotinic acid has been observed to stack directly over the isoalloxazine ring of
1
3
FMN in a coꢀcrystal with nitroreductase NfsB (E. coli.). Similar orientations of substrates stacking
over the isoalloxazine ring have been observed in coꢀcrystals of NFsB with dinitrobenzamide
45
46
prodrugs and dicoumarol and coꢀcrystals of flavin reductase (Vibrio fischeri) with dicoumarol.
2
0
ACS Paragon Plus Environment

Page 21 of 37
Biochemistry
1
2
3
Even the regiospecificity of dinitroaromatic reduction can be manipulated by altering the orientation
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
7
of the two nitro groups with respect to their stacking over the FMN of NfsB. In contrast, no redox
exchange occurs between the electron acceptor iodophenol and the electron donor FMN even when
hq
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
stacked together in the active site of IYD.
Substrate specificity for both mammalian and H. hydrossis IYD is consequently a function of
selective activation of the catalytic chemistry required for dehalogenation. The redox properties of
4
8
flavin are often modulated by πꢀstacking as measured previously in both models systems and
1
6,49ꢀ51
numerous flavoproteins.
Such πꢀinteractions are not sufficient for IYD catalysis since both
active (IꢀTyr) and inactive (2IP) compounds demonstrate similar abilities to stack above the active
site FMN (Figure 3). Instead, the major determinant for dehalogenation is likely based on the extent
11,44
to which FMN can participate in single electron transfer.
Major activity classes within the nitroꢀ
FMN reductase superfamily can be distinguished by their alternative ability to promote one and two
electron processes of FMN. An obligate two electron transfer is catalyzed by the nitroreductases in
42,43
this superfamily and competing one electron chemistry to form FMN is actively suppressed.
sq
This may in part be related to the hydrogen bonding between the amide backbone and FMN N5.
Human IYD promotes FMN formation once substrate binds to the active site and induces a
sq
conformational shift allowing Thr239 to approach for hydrogen bonding between its side chain
1
1
hydroxyl group and FMN N5. This type of hydrogen bonding is often associated with activation of
5
2ꢀ54
one electron pathways but it can also influence hydride transfer as well.
For human IYD, the
interactions between the sideꢀchain of Thr239 and FMN N5 have a profound effect on catalysis and
4
4
act as a switch between one and two electron processes.
Crystal structures of hhIYD in the absence and presence of IꢀTyr illustrate a substrateꢀ
dependent closure of the active site lid similar to that characterized previously for human IYD
1
1
(
Figure 2). However, a key hydrogen bond induced by closure of the active site lid in mammalian
2
1
ACS Paragon Plus Environment

Biochemistry
Page 22 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
IYD differs from that observed for hhIYD. In human IYD bound with IꢀTyr, Thr239 is within
hydrogen bonding range of FMN N5 (3.1Å) but is 4.9 Å away in the absence of IꢀTyr (Figure 5A).
In hhIYD, the equivalent residue, Thr173, does not move into close proximity with the FMN N5
upon binding of IꢀTyr (Figure 5B). Instead, Thr173 exists in multiple conformations and its oxygen
remains at an intermediate distance from the N5 position of FMN (3.6ꢀ4.0 Å). This intermediate
distance is also observed between Thr173 and FMN N5 in the structure of hhIYD bound with 2IP
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Figure 5C). The limited motion of the loop containing Thr173 in hhIYD may result from
constraints conferred by Pro176 that is replaced by Asn 242 in human IYD (Figure 5). Still, Thr173
is key to single electron transfer by IYD. A mutation of T173A in hhIYD suppresses FMNsq
4
4
formation and IꢀTyr dehalogenation while concurrently promoting nitroaromatic reduction.
Additionally, binding of FꢀTyr to hhIYD clearly affects its ability to stabilize FMNsq during
reductive titration (Figure 4). Similar stabilization is not evident after association of hhIYD with
2
IP. Thus, the zwitterion dominants catalytic activation of IYD and suggests that halotyrosines are
the likely substrates for bacterial as well as mammalian IYD. The details of substrate dependent
activation of hhIYD and the expected hydrogen bonding between Thr173 and FMN N5 may be
clarified after a crystal structure of the reduced FMN form of IYD•FꢀTyr can determined.
hq
2
2
ACS Paragon Plus Environment

Page 23 of 37
Biochemistry
1
2
3
Figure 5. Conformational dependence for hydrogen bonding between a Thr and the N5 position of
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
FMN . (A) Cartoon representation of the superposition of human IYD in the absence (blue) and
ox
presence of IꢀTyr (orange). (B) Equivalent superposition of hhIYD in the absence (grey) and
presence of IꢀTyr (cyan). (C) Equivalent superposition of hhIYD in the absence (grey) and presence
of 2IP (green).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
■
CONCLUSION
The specificity of human IYD is easily rationalized by its physiological role to salvage iodide
from IꢀTyr while avoiding dehalogenation of other iodophenolꢀcontaining materials associated with
thyroid hormone biosynthesis and regulation. The iodophenol component neither binds to human
IYD nor participates in catalytic turnover efficiently. The constraints for the bacteria homolog
hhIYD were not anticipated since its natural substrates and metabolic function have yet to be
established. Although hhIYD does not discriminate as well as human IYD for binding halotyrosine
versus halophenol derivatives, their catalytic selectivity remains similar. Both require the zwitterion
of halotyrosine to stabilize the formation of FMN directly or indirectly and facilitate the single
sq
44
electron transfer process that appears necessary for reductive dehalogenation. The basal level of
iodophenol deiodination is very low and not likely relevant in nature. Thus, active site stacking of
substrate and FMN is not sufficient for reductive deiodination despite the close proximity of the
reactants. Since iodide is not a known requirement for bacteria, hhIYD has the potential to consume
halotyrosines if present in their environment. hhIYD may also contribute to secondary metabolism
in analogy to a recently discovered reductive debrominase that is required for the biosynthesis of a
5
5
marine natural product.
■
ASSOCIATED CONTENT
Supporting Information
2
3
ACS Paragon Plus Environment

Biochemistry
Page 24 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The supporting information is available free of charge on the ACS Publications website at DOI: xxx.
Protein purification, NMR data for 2IPCOOH, binding, kinetics, structural images and reduction of
hhIYD (PDF)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Accession codes
The Xꢀray coordinates and structure factors have been deposited at the Protein Data Bank (PDB)
under accession codes 5KO7, 5KO8 and 5KRD for hhIYD, hhIYD•IꢀTyr and hhIYD•2IP structures,
respectively.
■
AUTHOR INFORMATION
Corresponding Author
*Email: rokita@jhu.edu.
Funding
This work was supported in part by the National Institutes of Health (NIH) grant AI119540. Use of
the Stanford Synchrotron Radiation Lightsource (SLAC) at the National Accelerator Laboratory is
supported by the U.S. Department of Energy, Office of Basic Energy Sciences under Contract No.
DEꢀAC02ꢀ76SF00515. The SSRL Structural Molecular Biology Program is supported by the DOE
Office of Biological and Environmental Research, and by the National Institute of General Medical
Sciences (including P41GM103393).
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
We thank Jimin Hu and Jennifer Buss for providing plasmids encoding the human IYD and hhIYD
genes.
2
4
ACS Paragon Plus Environment

Page 25 of 37
Biochemistry
1
2
3
4
5
■
ABBREVIATIONS
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
A2IP, 4ꢀaminoꢀ2ꢀiodophenol; I ꢀTyr, 3,5ꢀdiiodoꢀ
L
ꢀtyrosine; FMN, flavin mononucleotide; FMNhq,
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
reduced FMN hydroquinone; FMN , oxidized FMN; FMN , oneꢀelectron reduced FMN
ox
sq
semiquinone; FꢀTyr, 3ꢀfluoroꢀ
L
ꢀtyrosine; 2FP, 2ꢀfluorophenol; 2IP, 2ꢀiodophenol; 2IPCN, 4ꢀcyanoꢀ
ꢀtyrosine; IYD,
2
ꢀiodophenol; 2IPCOOH, 4ꢀhydroxyꢀ3ꢀiodobenzoic acid; IꢀTyr, 3ꢀiodoꢀ
L
iodotyrosine deiodinase; hhIYD, IYD from bacterium Haliscomenobacter hydrossis (DSM 1100);
RMSD, root mean squared deviation; TCEP, tris(2ꢀcarboxyethyl)phosphine hydrochloride.
■
REFERENCES
Roche, J., Michel, O., Michel, R., Gorbman, A. and Lissitzky, S. (1953) Sur la
1.
Deshalogénation Enzymatique des Iodotyrosines par le Corps Thyroïde et sur son Rôle
Physiologique. II. Biochim. Biophys. Acta 12, 570ꢀ576.
2
.
Cavalieri, R. R. (1977) Iodine Metabolism and Thyroid Physiology: Current Concepts.
Thyroid 7, 177ꢀ181.
3.
4.
Moreno, J. C., Klootwijk, W., van Toor, H., Pinto, G., D'Alessandro, M., Lèger, A., Goudie,
D., Polak, M., Grüters, A., and Visser, T. J., (2008). Mutations in the Iodotryosine
Deiodinase Gene and Hypothyroidism. N. Engl. J. Med. 358, 1811ꢀ1818.
Delange, F. (2002) Iodine Deficiency in Europe and its Consequences: an Update. Eur. J. N.
Med. Mol. I. 29, S404–S416.
5
.
.
Zimmermann. M B. (2009) Iodine Deficiency. Endocr. Rev. 30, 376–408.
Rokita, S. E. Flavoprotein Dehalogenases. In Handbook of Flavoproteins; Hille, R., Miller,
S. M. and Palfey, B., Eds.; DeGruyter: Berlin, 2013; Vol. 1, pp 337ꢀ350.
6
2
5
ACS Paragon Plus Environment

Biochemistry
Page 26 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
.
McTamney, P. M., and Rokita, S. E. (2009) A Mammalian Reductive Deiodinase has Broad
Power to Dehalogenate Chlorinated and Brominated Substrates. J. Am. Chem. Soc. 131,
1
4212–14213.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
.
.
Bobyk, K. D., Ballou, D. P., and Rokita, S. E. (2015) PreꢀSteadyꢀState Characterization of
Iodotyrosine Deiodinase from Human Thyroid. Biochemistry 54, 4487–4494.
SolisꢀS., J. C., Villalobos, P. and ValverdeꢀR., C. (2004) Comparative Kinetic
Characterization of Rat Thyroid Iodotyrosine Dehalogenase and Iodothyronine Deiodinase
Type 1. J. Endocrinol. 181, 385ꢀ392.
9
10.
11.
12.
13.
14.
Thomas, S. R., McTamney, P. M., Adler, J. M., LaRondeꢀLeBlanc, N., and Rokita, S. E.
(2009) Crystal Structure of Iodotyrosine Deiodinase, a Novel Flavoprotein Responsible for
Iodide Salvage in Thyroid Glands. J. Biol. Chem. 284, 19659ꢀ19667.
Hu, J., Chuenchor, W., and Rokita, S. E. (2015) A Switch Between Oneꢀ and TwoꢀElectron
Chemistry of the Human Flavoprotein Iodotyrosine Deiodinase is Controlled by Substrate. J.
Biol. Chem. 290, 590ꢀ600.
Shimizu, R., Yamaguchi, M., Uramaru, N., Kuroki, H., Ohta, S., Kitamura, S., and Sugihara,
K. (2013) Structure–Activity Relationships of 44 Halogenated Compounds for Iodotyrosine
DeiodinaseꢀInhibitory Activity. Toxicology 314, 22ꢀ29.
Lovering, A. L., Hyde, E. I., Searle, P. F., and White, S. A. (2001) The Structure of
Escherichia coli Nitroreductase Complexes with Nicontinic Acid: Three Crystal Forms at
1
.7 Å, 1.8 Å and 2.4 Å Resolution. J. Mol. Biol. 309, 203ꢀ213.
Johansson, E., Parkinson, G. N., Denney, E. J. and Neidle, S. (2003) Studies on the
Nitroreductase ProdrugꢀActivating System. Crystal Structures of Complexes with the
Inhibitor Dicoumarol and Nitrobenzamide Prodrugs and of the Enzyme Active Form. J. Med.
Chem. 46, 4009ꢀ4020.
2
6
ACS Paragon Plus Environment

Page 27 of 37
Biochemistry
1
2
3
1
5.
6.
Wang, C.ꢀJ., Hagemeier, C., Rahman, N., Lowe, E., Noble, M., Coughtrie, M., Sim, E., and
Westwood, I. (2007) Molecular Cloning, Characterization and LigandꢀBound Structure of an
Azoreductase from Pseudomonas aruginosa. J. Mol. Biol. 373, 1213ꢀ1228.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
Lostao, A., GómezꢀMoreno, C., Mayhew, S. G., and Sancho, J. (1997) Differential
Stabilization of the Three FMN Redox Forms by Tyrosine 94 and Tryptophan 57 in
Flavodoxin from Anabeaena and its Influence on the Redox Potentials. Biochemistry 36,
1
4334ꢀ14344.
17.
Edgar, K. J., and Falling, S. N. An Efficient and Selective Method for the Preparation of
Iodophenols. J. Org. Chem. 55, 5287–5291.
18.
19.
20.
Phatarphekar, A., Buss, J. M., and Rokita, S. E. (2014) Iodotyrosine Deiodinase: a Unique
Flavoprotein Present in Organisms of Diverse Phyla. Mol. BioSyst. 10, 86ꢀ92.
Pace, C. N., Vajdos, F., Fee, L, Grimsley, G., and Gray, T. (1995) How to Measure and
Predict the Molar Absorption Coefficient of a Protein. Protein Sci. 4, 2411–2423.
Koziol, J. (1971) Fluorometric Analyses of Riboflavin and its Coenzymes. Methods.
Enzymol. 18, 235–285.
21.
22.
McTamney, P. M., and Rokita, S. E. (2009) A Mammalian Reductive Deiodinase has broad
Power to Dehalogenate Chlorinated and Brominated Substrates. J. Am. Chem. Soc. 131,
1
4212–14213.
Phatarphekar, A., and Rokita, S. E. (2016) Functional Analysis of Iodotyrosine Deiodinase
from Drosophila melanogaster. Protein. Sci. 25, 2187ꢀ2195.
2
3.
Massey, V. A Simple Method for the Determination of Redox Potentials. In Flavins
Flavoproteins Proc. Int. Symp; B. Curti, S. Ronchi and G. Zanetti, Eds.; Gruyter & Co.:
Berlin, 1991, pp 59ꢀ66.
2
7
ACS Paragon Plus Environment

Biochemistry
Page 28 of 37
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
4.
5.
van den Heuvel, R. H. H., Fraaije, M. W., and Van Berkel, W. J. H. (2002) Redox Properties
of VanillylꢀAlcohol Oxidase. Methods Enzymol. 353, 177ꢀ186.
2
Kabsch, W. (1993) Automatic Processing of Rotation Diffraction Data from Crystals of
Initially Unknown Symmetry and Cell Constants, J. Appl. Cryst. 26, 795–800.
Otwinowski, Z., and Minor, W. (1997) [20] Processing of Xꢀray Diffraction Data Collected
in Oscillation Mode, In Methods in Enzymology, pp 307–326, Academic Press.
McCoy, A. J., GrosseꢀKunstleve, R. W., Adams, P. D., Winn, M. D., Storonia, L. C., and
Reada, R. J (2007) Phaser Crystallographic Software, J. Appl. Cryst. 40, 658–674.
Emsley, P., and Cowtan, K. (2004) Coot: ModelꢀBuilding Tools for Molecular Graphics,
Acta. Cryst. 60, 2126–2132.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
26.
2
7.
8.
2
29.
Adams, P. D., Afonine, P. V., Bunkóczi, G., Chen, V. B., Davis I. W., Echols, N., Headd, J.
J., Hung, L.ꢀW., Kapral, G. J., GrosseꢀKunstleve, R. W., McCoy, A. J., Moriarty, N. W.,
Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C., and Zwart,
P. H. (2010) PHENIX: a Comprehensive PythonꢀBased System for Macromolecular
Structure Solution, Acta Cryst. 66, 213–221.
30.
31.
Buss, J. M., McTamney, P. M., and Rokita, S. E. (2012) Expression of a Soluble Form of
Iodotyrosine Deiodinase for Active Site Characterization by Engineering the Native
Membrane Protein from Mus musculus, Protein. Sci. 21, 351ꢀ361.
Gasteiger, E., Hoogland, C., Gattiker, A., Duvaud, S., Wilkins, M. R., Appel, R. D., and
Bairoch, A. (2005) Protein Identification and Analysis Tools on the ExPASy Server, In The
Proteomics Protocols Handbook (Walker, J. M., Ed.), pp 571–607 Humana Press.
Heyland, A., Hodin, J., and Reitzel, A. M. (2005) Hormone Signaling in Evolution and
Development: a Nonꢀmodel System Approach, BioEssays 27, 64ꢀ75.
3
2.
2
8
ACS Paragon Plus Environment

Page 29 of 37
Biochemistry
1
2
3
3
3.
4.
Miller, A. E. M., and Heyland, A. (2013) Iodine Accumulation in Sea Urchin Larvae is
Dependent on Peroxide, J. Exp. Biol. 216, 915ꢀ926.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
Rokita, S. E., Adler, J. M., McTamney, P. M., and Watson, J. A., Jr. (2010) Efficient Use and
Recycling of the Micronutrient Iodide in Mammals, Biochimie 92, 1227ꢀ1235.
Friedman, J. E., Watson, J. A., Jr., Lam, D. W.ꢀH., and Rokita, S. E. (2006) Iodotyrosine
Deiodinase is the First Mammalian Member of the NADH Oxidase / Flavin Reductase
Superfamily, J. Biol. Chem. 281, 2812ꢀ2819.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
35.
36.
Li, Y., Reilly, P. J., and Ford, C. (1997) Effect of Introducing Proline Residues on the
Stability of Aspergillus awamori, Protein Eng. 10, 1199–1204.
37.
38.
39.
Gilday, L. C., Robinson, S. W., Barendt, T. A., Langton, M. J., Mullaney, B. R., and Beer, P.
D. (2015) Halogen Bonding in Supramolecular Chemistry, Chem. Rev. 115, 7118–7195.
Massey, V. (2000) The Chemical and Biological Versatility of Riboflavin, Biochem. Soc.
Trans. 28, 283ꢀ296.
Fraaije, M. W., and Mattevi, A. (2000) Flavoenzymes: Diverse Catalysts with Recurrent
Features, Trends Biochem. Sci. 25, 126ꢀ132.
40.
Mansoorabadi, S. O., Thibodeaux, C. J., and Liu, H.ꢀw. (2007) The Diverse Roles of Flavin
CoenyzmesꢀꢀNature's Most Versatile Thespians, J. Org. Chem. 72, 6329ꢀ6342.
Fagan, R. L., and Palfey, B. A. (2010) FlavinꢀDependent Enzymes, In Ch. 3 in
Comprehensive Natural Products II (Begley, T. P., Ed.), pp 37ꢀ114, Elsevier, Oxford.
Haynes, C. A., Koder, R. L., Miller, A.ꢀF., and Rodgers, D. W. (2002) Structures of
Nitroreductase in Three States, J. Biol. Chem. 277, 11513ꢀ11520.
41.
42.
43.
Koder, R. L., Haynes, C. A., Rodgers, M. E., Rodgers, D. W., and Miller, A.ꢀF. (2002)
Flavin Thermodynamics Explain the Oxygen Insensitivity of Enteric Nitroreductase,
Biochemistry 41, 14197ꢀ14205.
2
9
ACS Paragon Plus Environment