Crystal structures and structure-activity relationships of imidazothiazole derivatives as IDO1 inhibitors

Article abstract of DOI:10.1021/ml500247w

Indoleamine 2,3-dioxygenase 1 (IDO1) is considered as a promising target for the treatment of several diseases, including neurological disorders and cancer. We report here the crystal structures of two IDO1/IDO1 inhibitor complexes, one of which shows that Amg-1 is directly bound to the heme iron of IDO1 with a clear induced fit. We also describe the identification and preliminary optimization of imidazothiazole derivatives as novel IDO1 inhibitors. Using our crystal structure information and structure-activity relationships (SAR) at the pocket-B of IDO1, we found a series of urea derivatives as potent IDO1 inhibitors and revealed that generation of an induced fit and the resulting interaction with Phe226 and Arg231 are essential for potent IDO1 inhibitory activity. The results of this study are very valuable for understanding the mechanism of IDO1 activation, which is very important for structure-based drug design (SBDD) to discover potent IDO1 inhibitors.

Full text of DOI:10.1021/ml500247w

Letter
pubs.acs.org/acsmedchemlett
Crystal Structures and Structure−Activity Relationships of
Imidazothiazole Derivatives as IDO1 Inhibitors
Shingo Tojo,*,† Tetsuya Kohno,† Tomoyuki Tanaka, Seiji Kamioka, Yosuke Ota, Takayuki Ishii,
Keiko Kamimoto, Shigehiro Asano, and Yoshiaki Isobe
Dainippon Sumitomo Pharma Co., Ltd., 3-1-98 Kasugade-naka, Konohana-ku, Osaka 554-0022, Japan
S
* Supporting Information
ABSTRACT: Indoleamine 2,3-dioxygenase 1 (IDO1) is considered as a promising target for the treatment of several diseases,
including neurological disorders and cancer. We report here the crystal structures of two IDO1/IDO1 inhibitor complexes, one
of which shows that Amg-1 is directly bound to the heme iron of IDO1 with a clear induced fit. We also describe the
identification and preliminary optimization of imidazothiazole derivatives as novel IDO1 inhibitors. Using our crystal structure
information and structure−activity relationships (SAR) at the pocket-B of IDO1, we found a series of urea derivatives as potent
IDO1 inhibitors and revealed that generation of an induced fit and the resulting interaction with Phe226 and Arg231 are essential
for potent IDO1 inhibitory activity. The results of this study are very valuable for understanding the mechanism of IDO1
activation, which is very important for structure-based drug design (SBDD) to discover potent IDO1 inhibitors.
KEYWORDS: Indoleamine 2,3-dioxygenagse 1, crystal structure, induced fit, structure-based drug discovery, imidazothiazole
-ray crystallography is a powerful tool for clarifying target
disease, and cerebral ischemia.7 In addition, overexpression of
Xprotein structures and plays a critical role in drug IDO1 has been reported in many tumor cells,8 and IDO1 is
considered as one of the key factors that contributes to cancer
immunosuppression in tumor microenvironment. IDO1 helps
create a tolerogenic state in both the tumor and its draining
lymph nodes by both direct suppression of T cells and
enhancement of local regulatory T cells.9 As IDO1 is strongly
induced by interferon γ, which is produced by cytotoxic T
lymphocytes,10 it presumably impairs the antitumor efficacy of
other immune therapy. Therefore, inhibition of IDO1 would be
very useful for the treatment of not only neurological diseases
but also cancer.
discovery, especially in structure-based drug design (SBDD),
which is one of the most useful methods for lead generation
and optimization. Detailed analysis of the crystal structure of a
target protein/ligand complex is essential for SBDD and its
success. Starting point of SBDD study is to obtain the suitable
crystals of target protein/ligand complexes. However, the
difficulty of it often depends on the target protein.1 For
example, heme-containing proteins present a challenge because
preparation of highly active recombinant protein is very
difficult.2
To date, several types of IDO1 inhibitors, including
tryptophan derivatives as represented by 1-methyltryptophan
(1-MT),11 4-phenyl imidazole (PI) analogues,12,13 the
thiazolotriazole compound Amg-1,14 and the hydroxylamidine
compound 1,15 have been reported (Figure 1). Compound 1
has the most potent IDO1 inhibitory activity (IC50 = 67 nM, ref
Indoleamine 2,3-dioxygenase 1 (IDO1) is a heme-containing
protein that catalyzes the oxidative cleavage of the C2−C3
double bond of the indole in tryptophan to provide N-formyl-
kynurenine. This reaction is known as the initial and rate-
limiting step in the kynurenine pathway of tryptophan
catabolism in mammals.3−6 The generated N-formyl-kynur-
enine is further metabolized to bioactive metabolites, including
kynurenine, kynurenic acid, 3-hydroxy-kynurenine, and quino-
linic acid, which are known to be involved in a number of
neurological disorders, such as Alzheimer’s disease, Parkinson’s
Received: June 10, 2014
Accepted: August 19, 2014
Published: August 21, 2014
© 2014 American Chemical Society
1119
dx.doi.org/10.1021/ml500247w | ACS Med. Chem. Lett. 2014, 5, 1119−1123

ACS Medicinal Chemistry Letters
Letter
Almost all studies of IDO1 inhibitors make use of the crystal
structure information on IDO1/PI complex or related plausible
docking models to optimize their compounds. However, many
of these studies discuss only the structure−activity relationships
(SAR) at the pocket-A and the interaction with the heme iron
without addressing SAR at the pocket-B. The reason should be
that PI is a fragment-like inhibitor and interacts directly with
heme iron and pocket-A, not pocket-B.
Figure 1. Structures of representative IDO1 inhibitors.
15) in currently reported IDO1 inhibitors. Even though a
number of IDO1 inhibitors have been reported, only the crystal
structure of IDO1/PI complex is currently available.16 This
indicates that preparation of suitable crystals of IDO1/IDO1
inhibitor complex would be very difficult, probably due to
additional technical issues distinctive of IDO1 as a heme-
containing protein.2
It is reported that PI interacts directly with IDO1 heme iron
at the imidazole nitrogen and that its phenyl group fits deeply
into one of IDO1 hydrophobic pockets (pocket-A). The other
hydrophobic pocket (pocket-B) is occupied by two molecules
of the buffer N-cyclohexyl-2-aminoethanesulfonic acid (CHES)
(Figure 2a; and the stereoview, see Figure S1 in the Supporting
In our discovery study for novel IDO1 inhibitors, we
considered that crystal structure information on IDO1/IDO1
inhibitor complex at pocket-B and elucidation of SAR at
pocket-B are important for finding potent IDO1 inhibitors.
Herein we report crystal structures of two IDO1/IDO1
inhibitor complexes, where the inhibitor interacts directly
with pocket-B. We also describe the identification and SAR of a
series of imidazothiazole derivatives as novel IDO1 inhibitors.
First, we sought to obtain the crystal structure of IDO1/
IDO1 inhibitor complex. For that, we initially prepared a highly
homogeneous recombinant IDO1 protein with a heme content
ratio of approximately 78% compared to fully native heme-
incorporated IDO1 protein.14 Incubation in 0.75 mM δ-
aminolaevulinic acid at 30 °C for 12 h and purification with a
strong reductant buffer containing 1 mM tris(2-carboxyethyl)-
phosphine (TCEP) were essential in achieving a heme content
ratio of more than 70%. Next, we turned our attention to
crystallization of IDO1/IDO1 inhibitor complex. Here, adding
the inhibitor while removing TCEP using a desalting column
was very important for crystallization. Investigation of some
currently reported IDO1 inhibitors led to the crystal structure
of IDO1/Amg-1 complex as shown in Figure 2b (for the omit
map and the stereoview, see Figure S2a,b in the Supporting
Information). Interestingly, contrary to a previous report,14
which suggests that Amg-1 does not interact with heme iron,
we found that the nitrogen of thiazolotriazole in Amg-1 was
directly bound to the heme iron and that tolyl group was placed
at pocket-A. Furthermore, IDO1/Amg-1 complex showed a
remarkable induced fit compared to IDO1/PI complex. Pocket-
A expanded as structure of the main chain Ala260-Ser263
changed, and a shift of both Phe226 and Arg231 resulted in
extension of pocket-B. The amide side chain of Amg-1 located
at the expanded pocket-B, and the methylenedioxyphenyl
moiety took a position adjacent to both residues. These
findings were considered as encouraging for SBDD since they
provided, for the first time, an induced fit for the crystal
structure of IDO1/IDO1 complex and allowed acquisition of
precise structure information, where IDO1 inhibitor directly
interacted with IDO1 protein at both pocket-A and pocket-B.
To confirm the potential of the basic scaffold of Amg-1, we
evaluated IDO1 inhibitory activity of the para-tolyl thiazolo-
triazole compound 2. Like PI, compound 2 showed weak IDO1
inhibitory activity (Figure 3). The crystal structures of Amg-1
and PI revealed that both 2 and PI would bind to IDO1 at the
same site and that 2 would directly interact with the heme iron.
We focused on the pKa value of the nitrogen atom bound to the
heme iron to design a novel IDO1 inhibitor since compound 2
calculated pKa value (1.09) was significantly lower than that of
PI (6.56). As it is generally known that the basicity of a ligand is
very important for forming a metal complex, we assumed that it
would be possible to improve IDO1 inhibitory activity of 2 by
control of the basicity of the nitrogen atom. Namely, we
thought that strong basicity of nitrogen atom would result in
strong binding to heme iron of IDO1; consequently, such
compound should have potent IDO1 inhibitory activity. Hence,
Figure 2. (a) Crystal structure of IDO1/PI (purple) complex (PDB:
2DOT). CHES is colored in blue. (b) Crystal structure of IDO1/Amg-
1 (green) complex (PDB: 4PK5). Main chain Ala260-Ser263 (yellow),
Phe226 (pink), and Arg231 (cyan) are shown in both panels.
Information). According to Shiro et al.,16 site-direct muta-
genesis analysis of IDO1 catalytic activity is ensured by two
amino acid residues: Phe226 and Arg231, both of which are
part of the residues forming pocket-B. They hypothesized that
both residues are directly involved in substrate recognition by
hydrophobic interactions.
1120
dx.doi.org/10.1021/ml500247w | ACS Med. Chem. Lett. 2014, 5, 1119−1123

ACS Medicinal Chemistry Letters
Letter
Hydrolysis of the ethyl ester 8a to give the carboxylic acid 12
and subsequent amide formation reaction with R2CH2CH2NH2
and WSC·HCl gave the amide compounds 13a−c. The urea
derivatives 17a−g were easily prepared in 4 steps from 8i.
Reduction of the ethyl ester 8i with DIBAL-H afforded the
alcohol 14 in 94%. The alcohol group of 14 was then converted
with DPPA and DBU19 to the azide compound 15, which was
reduced under Staudinger conditions20 to provide the amine
compound 16. Finally, treatment of 16 with isocyanate gave the
urea compounds 17a−g.
Figure 3. Design of the new imidazothiazole scaffold based on 2 and
PI pKa values. The nitrogen atom that binds to the heme iron is
highlighted in red. IC50 values are the mean of at least two
independent assays. pKa values were calculated by ACD/Laboratories
ver. 11.0.
To improve IDO1 inhibitory activity, we first examined a way
to optimize pocket-A using ethyl ester derivatives. As shown in
Table 1, the para-tolyl compound 8a was preferable to the
we designed the para-tolyl imidazothiazole compound 3 as a
novel scaffold by combining 2 and PI. The calculated pKa value
of the corresponding nitrogen atom of 3 was 7.04, which is
similar to the basicity of PI. As predicted, IDO1 inhibitory
activity of 3 was more potent than that of 2. These findings
indicated that 3 is a good starting compound for our discovery
study.
Table 1. IDO1 Inhibitory Activity of the Ethyl Ester
Derivatives 8a−j and 11
The imidazothiazole derivatives were prepared by the
following sequence (Scheme 1). First, the imidazothiazole
scaffold was constructed in 3 steps as previously described17
with some modifications. Treatment of the 4-bromothiazol-2-
amine 4 with N,N-dimethylformamide dimethyl acetal gave the
amidine compound 5. Salt formation in ethyl bromoacetate and
subsequent treatment of the salt 6 with DBU afforded the
imidazothiazole compound 7. Suzuki couplings18 with boronic
acids or pinacol boronates proceeded smoothly to provide 3-
substituted imidazothiazole derivertives 8a−h. The bromine
compound 8i, benzyl compound 8j, and fused ring derivative
1117 were prepared from the corresponding thiazol-2-amine
9i−j and 10 by a similar method. Reduction of the ethyl ester
8a with LiAlH4 and AlCl3 gave the methyl compound 3. The
amide derivatives 13a−c were readily prepared in 2 steps.
a
compd
R1
4-tolyl
IDO1 IC50 (μM)
8a
8b
8c
8d
8e
8f
18 3.7
>100
3-tolyl
2-tolyl
>100
phenyl
>100
1-cyclohexenyl
4-Et-C6H4
4-MeO-C6H4
4-Cl-C6H4
4-Br-C6H4
benzyl
>100
24 4.0
56 2.0
8.1 1.7
5.5 2.2
>100
8g
8h
8i
8j
11
>100
a
IC50 values are the mean of at least two independent assays.
a
Scheme 1. Synthesis of the Imidazothiazole Compounds
a
Reagents and conditions: (a) N,N-dimethylformamide dimethyl acetal, DMF, 80 °C; (b) ethyl bromoacetate, 80 °C; (c) DBU, DMF, 60 °C; (d)
R1B(OH)2 or R1BPin, Pd(PPh3)4, K2CO3, 1,4-dioxane-H2O (3:1), 100 °C; (e) LiAlH4, AlCl3, THF, reflux (15%); (f) 1 N NaOH, MeOH-THF
(99%); (g) R2CH2CH2NH2, WSC·HCl, HOBt, i-Pr2NEt, DMF; (h) DIBAL-H, THF, 0 °C (94%); (i) DPPA, DBU, DMF, 60 °C (75%); (j) PPh3,
H2O, THF, 45 °C (90%); (k) R3NCO, i-Pr2NEt, THF.
1121
dx.doi.org/10.1021/ml500247w | ACS Med. Chem. Lett. 2014, 5, 1119−1123

ACS Medicinal Chemistry Letters
Letter
meta-tolyl compound 8b or the ortho-tolyl compound 8c, while
the nonsubstituted phenyl compound 8d and the 1-cyclohexene
compound 8e were not effective. IDO1 inhibitory activity of
the ethyl compound 8f and the methoxy compound 8g was less
than that of 8a. Fortunately, introduction of a halogen atom
onto the para-position was favorable, allowing both the chlorine
compound 8h and the bromine compound 8i to be more
effective than the para-tolyl compound 8a. Both the benzyl
compound 8j and the fused ring type compound 11 were
inactive.
Next, we attempted to optimize pocket-B by converting the
ester moiety of 8a and prepared the amide derivatives (13a,
13b, and 13c), which have a side chain the same length as that
of Amg-1 (Table 2). However, these compounds showed
Consequently, 13b generated an induced fit and located
adjacent to Phe226, but not to Arg231.
These results indicated that as in Amg-1, a linearly rigid
linker had to be introduced into the side chain to allow
interaction with both the Phe226 and Arg231. Hence, we
designed to introduce a urea group as a linearly planar unit into
the linker moiety (Table 3). Introduction of the urea unit
Table 3. IDO1 Inhibitory Activity of the Urea Derivatives
17a−g
a
Table 2. IDO1 Inhibitory Activity of the Amide Derivatives
compd
R3
IDO1 IC50 (μM)
13a−c
17a
17b
17c
17d
17e
17f
phenyl
benzyl
0.22 0.03
0.75 0.01
1.8 0.40
cyclohexanyl
3,4-methylenedioxyphenyl
3-MeO-C6H4
0.12 0.04
0.16 0.03
0.19 0.05
0.077 0.015
4-MeO-C6H4
a
compd
R2
IDO1 IC50 (μM)
17g
4-CN-C6H4
a
13a
13b
13c
phenyl
3.5 1.0
1.9 0.5
3.6 1.2
IC50 values are the mean of at least two independent assays.
3-Cl-C6H4
3,4-methylenedioxyphenyl
predictably improved IDO1 inhibitory activity in a large
number of derivatives (17a, 17d, 17e, 17f, and 17g) with the
same length of side chain as Amg-1. The benzyl compound 17b
was less effective than the phenyl compound 17a, which
showed that the length of the linker is very important for an
adjacent location to Phe226 and Arg231. IDO1 inhibitory
activity of the cyclohexane compound 17c was much lower
than that of the phenyl compound 17a. This suggested that the
phenyl group interacts with Phe226 by a π−π interaction not a
hydrophobic interaction (Figure 5). Furthermore, the meta-
a
IC50 values are the mean of at least two independent assays.
almost the same IDO1 inhibitory activity as Amg-1 (IC50 = 3.0
μM, ref 14), even though their basic scaffold had a potential
superior to that of Amg-1. We therefore suspected that the side
chain of these compounds in pocket-B should not be located
exactly at the same position as the side chain of Amg-1. In fact,
the crystal structure of IDO1/13b complex supported this
hypothesis (Figure 4; for the omit map and the stereoview, see
Figure 5. Plausible interactions of urea derivatives with Phe226 (pink)
and Arg231 (cyan).
and/or para-substituted compounds (17d, 17e, 17f, and 17g)
were more effective than the unsubstituted compound 17a.
Particularly, the nitrile compound 17g was the most effective
compound with an IDO1 inhibitory activity comparable to that
of 1. These SARs support the idea that electrostatic interaction
of Arg231 with a functional group substituted at the meta- and/
or para-position of the phenyl group is necessary for potent
IDO1 inhibitory activity. These results demonstrate that
induced fit by extension of pocket-B and resulting interaction
with both Phe226 and Arg231 would be essential for potent
IDO1 inhibitory activity and are in accord with the results of
previous site-direct mutagenesis analysis.16 Furthermore, this is
the first case that clarified in detail the SAR at pocket-B.
In summary, to find novel potent IDO1 inhibitors, we
determined the crystal structures of two IDO1/IDO1 inhibitor
complexes, one of which showed that Amg-1 is directly bound
Figure 4. Crystal structure of IDO1/13b complex (green) (PDB:
4PK6). Main chain Ala260-Ser263 (yellow), Phe226 (pink), and
Arg231 (cyan) are shown.
Figure S3a,b in the Supporting Information). The nitrogen of
imidazothiazole in 13b was directly bound to the heme iron
and that tolyl group located at the expanded pocket-A in the
same manner as Amg-1. However, the side chain of 13b was
bent sharply at the methylene moiety since the shape of pocket-
B was dramatically changed by the shift of Phe226.
1122
dx.doi.org/10.1021/ml500247w | ACS Med. Chem. Lett. 2014, 5, 1119−1123

ACS Medicinal Chemistry Letters
Letter
(6) Forouhar, F.; Anderson, R. J. L.; Mowat, C. G.; Vorobiev, S. M.;
Hussain, A.; Abashidze, M.; Bruckmann, C.; Thackray, S. J.;
Seetharaman, J.; Tucker, T.; Xiao, R.; Ma, L.; Zhao, L.; Acton, T.
B.; Montelione, G. T.; Chapman, S. K.; Tong, L. Molecular insights
into substrate recognition and catalysis by tryptophan 2,3-dioxygenase.
Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 473−478.
to the heme iron of IDO1 with a clear induced fit. Using this
information, we identified the para-tolyl imidazothiazole
compound 3 as a novel IDO1 inhibitor. Additionally, on the
basis of crystal structure information on IDO1/13b complex,
we found a series of urea derivatives as potent IDO1 inhibitors.
Our novel structure information and SAR at the pocket-B
revealed that generation of an induced fit and the resulting
interaction with Phe226 and Arg231 are essential for potent
IDO1 inhibitory activity. We believe that this study is very
important for understanding the mechanism of IDO1
activation, which is very important for SBDD to find potent
IDO1 inhibitors. Further efforts to discover a promising
candidate are in progress.
́ ́
(7) Vecsei, L.; Szalardy, L.; Fulop, F.; Toldi, J. Kynurenines in the
̈
̈
CNS: recent advances and new questions. Nat. Rev. Drug Discovery
2013, 12, 64−82.
(8) Godin-Ethier, J.; Hanafi, L.; Piccirillo, C. A.; Lapointe, R.
Indoleamine 2,3-dioxygenase expression in human cancers: Clinical
and Immunologic perspectives. Clin. Cancer Res. 2011, 17, 6985−6991.
(9) Munn, D. H.; Mellor, A. L. Indoleamine 2,3-dioxygenase and
tumor-induced tolerance. J. Clin. Invest. 2007, 117, 1147−1154.
(10) Katz, J. B.; Muller, A. J.; Prendergast, G. C. Indoleamine 2,3-
dioxygenase in T-cell tolerance and tumoral immune escape. Immunol.
Rev. 2008, 222, 206−221.
ASSOCIATED CONTENT
■
S
* Supporting Information
(11) Cady, S. G.; Sono, M. 1-Methyl-DL-tryptophan, beta-(3-
benzofuranyl)-DL-alanine (the oxygen analog of tryptophan), and
beta-[3-benzo(b-)thienyl]-DL-alanine (the sulfur analog of tryptophan)
are competitive inhibitors for indoleamine 2,3-dioxygenase. Arch.
Biochem. Biophys. 1991, 291, 326−333.
Supplementary figures, synthetic procedures, characterization,
purification method of recombinant IDO1 protein, biological
assays, and X-ray crystallography. This material is available free
of charge via the Internet at http://pubs.acs.org.
(12) Kumar, S.; Jaller, D.; Patel, B.; LaLonde, J. M.; DuHadaway, J.
B.; Malachowski, W. P.; Prendergast, G. C.; Muller, A. J. Structure
based development of phenylimidazole-derived inhibitors of indole-
amine 2,3-dioxygenase. J. Med. Chem. 2008, 51, 4968−4977.
AUTHOR INFORMATION
■
Corresponding Author
(13) Rohrig, U. F.; Majjigapu, S. R.; Grosdidier, A.; Bron, S.;
̈
*(S.T.) Phone: +81-6-6466-5193. Fax: +81-6-6466-5297. E-
mail: shingo-tojo@ds-pharma.co.jp.
Stroobant, V.; Pilotte, L.; Colau, D.; Vogel, P.; Van den Eynde, B. J.;
Zoete, V.; Michielin, O. Rational design of 4-aryl-1,2,3-triazoles for
indoleamine 2,3-dioxygenase 1 inhibition. J. Med. Chem. 2010, 53,
1172−1189.
Author Contributions
†S.T. and T.K. contributed equally to this study. S.T. directed
medicinal chemistry. T.K. directed purification of IDO protein
and X-ray crystallography.
(14) Meininger, D.; Zalameda, L.; Liu, Y.; Stepan, L. P.; Borges, L.;
McCarter, J. D.; Sutherland, C. L. Purification and kinetic character-
ization of human indoleamine 2,3-dioxygenases 1 and 2 (IDO1 and
IDO2) and discovery of selective IDO1 inhibitors. Biochim. Biophys.
Acta 2011, 1814, 1947−1954.
Notes
The authors declare no competing financial interest.
(15) Yue, E. W.; Douty, B.; Wayland, B.; Bower, M.; Liu, X.; Leffet,
L.; Wang, Q.; Bowman, K. J.; Hansbury, M. J.; Liu, C.; Wei, M.; Li, Y.;
Wynn, R.; Burn, T. C.; Koblish, H. K.; Fridman, J. S.; Metcalf, B.;
Scherle, P. A.; Combs, A. P. Discovery of potent competitive inhibitors
of indoleamine 2,3-dioxygenase with in vivo pharmacodynamic activity
and efficacy in a mouse melanoma model. J. Med. Chem. 2009, 52,
7364−7367.
(16) Sugimoto, H.; Oda, S.; Otsuki, T.; Hino, T.; Yoshida, T.; Shiro,
Y. Crystal structure of human indoleamine 2,3-dioxygenase: Catalytic
mechanism of O2 incorporation by a heme-containing dioxygenase.
Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 2611−2616 The crystal
structure of IDO1/cyanid complex is also reported in this letter.
However, we recognize cyanid as purely a ligand of heme iron.
(17) Landreau, C.; Deniaud, D.; Evain, M.; Reliquet, A.; Meslin, J. C.
Efficient regioselective synthesis of triheterocyclic compounds:
imidazo[2,1-b]benzothiazoles, pyrimido[2,1-b]benzothiazolones and
pyrimido[2,1-b]benzothiazoles. J. Chem. Soc., Perkin Trans. 1 2002,
741−745.
ACKNOWLEDGMENTS
■
We are deeply grateful to Prof. Y. Shiro (RIKEN Harima
Institute/Spring-8 Japan) for his fruitful discussion and advice.
We also thank Ms. E. Koga and Dr. K. Kubota for their helpful
discussion, Ms. S. Sasabe for collecting data on biological
assays, and Ms. K. Yokokawa for her synthetic technical
assistance.
ABBREVIATIONS
■
WSC, 1-ethyl-3-(3′-dimethylaminopropyl)carbodiimide; HOBt,
1-hydroxybenzotriazole; DIBAL-H, diisobutylaluminum hy-
dride; DPPA, diphenylphosphoryl azide; DBU, 1,8-
diazabicyclo[5.4.0]undec-7-ene
REFERENCES
■
(18) Miyaura, N.; Yamada, K.; Suzuki, A. A new stereospecific cross-
coupling by the palladium-catalyzed reaction of 1-alkenylboranes with
1-alkenyl or 1-alkynyl halides. Tetrahedron Lett. 1979, 36, 3437.
(19) Thompson, A. S.; Humphrey, G. R.; DeMarco, A. M.; Mathre,
D. J.; Grabowski, E. J. J. Direct conversion of activated alcohols to
azides using diphenyl phosphorazidate. A practical alternative to
Mitsunobu conditions. J. Org. Chem. 1993, 58, 5886−5888.
(1) Grey, J.; Thompson, D. Challenges and opportunities for new
protein crystallization strategies in structure-based drug design. Expert
Opin. Drug Discovery 2010, 5, 1039−1045.
(2) Jung, Y.; Kwak, J.; Lee, Y. High-level production of heme-
containing holoproteins in Esherichia coli. Appl. Microbiol. Biotechnol.
2001, 55, 187−191.
(3) Takikawa, O.; Yoshida, R.; Kido, R.; Hayaishi, O. Tryptophan
degradation in mice initiated by indoleamine 2,3-dioxygenase. J. Biol.
Chem. 1986, 261, 3648−3653.
(4) Takikawa, O. Biochemical and medical aspects of the indole-
amine 2,3-dioxygenase initiated L-tryptophan metabolism. Biochem.
Biophys. Res. Commun. 2005, 338, 12−19.
̈
(20) Staudinger, H.; Meyer, J. Uber neue organische phosphorver-
bindungen III. Phosphinmethylenderivate und phosphinimine. Helv.
Chim. Acta 1919, 2, 635−646.
(5) Efimov, I.; Basran, J.; Thackray, S. J.; Handa, S.; Mowat, C. G.;
Raven, E. L. Structure and reaction mechanism in heme dioxygenases.
Biochemistry 2011, 50, 2717−2724.
1123
dx.doi.org/10.1021/ml500247w | ACS Med. Chem. Lett. 2014, 5, 1119−1123