Structure-Activity Relationships of cyclo(l -Tyrosyl- l -tyrosine) Derivatives Binding to Mycobacterium tuberculosis CYP121: Iodinated Analogues Promote Shift to High-Spin Adduct

Article abstract of DOI:10.1021/acs.jmedchem.9b01199

A series of analogues of cyclo(l-tyrosyl-l-tyrosine), the substrate of the Mycobacterium tuberculosis enzyme CYP121, have been synthesized and analyzed by UV-vis and electron paramagnetic resonance spectroscopy and by X-ray crystallography. The introduction of iodine substituents onto cyclo(l-tyrosyl-l-tyrosine) results in sub-μM binding affinity for the CYP121 enzyme and a complete shift to the high-spin state of the heme Fe^III. The introduction of halogens that are able to interact with heme groups is thus a feasible approach to the development of next-generation, tight binding inhibitors of the CYP121 enzyme, in the search for novel antitubercular compounds.

Full text of DOI:10.1021/acs.jmedchem.9b01199

Subscriber access provided by Bibliothèque de l'Université Paris-Sud
Article
Structure–activity relationships of cyclo(L-tyrosyl-L-tyrosine)
derivatives binding to Mycobacterium tuberculosis CYP121:
iodinated analogues promote shift to high-spin adduct
Sunnia Rajput, Kirsty J. McLean, Harshwardhan Poddar, Irwin Selvam, Gayathri
Nagalingam, James A Triccas, Colin Levy, Andrew W. Munro, and Craig A Hutton
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01199 • Publication Date (Web): 16 Oct 2019
Downloaded from pubs.acs.org on October 19, 2019
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 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 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.
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 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure–activity relationships of cyclo(L-tyrosyl-
L-tyrosine) derivatives binding to Mycobacterium
tuberculosis CYP121: iodinated analogues promote
shift to high-spin adduct
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
†
‡
‡
‡
Sunnia Rajput, Kirsty J. McLean, Harshwardhan Poddar, Irwin R. Selvam, Gayathri
¶
¶
‡
‡
*,†
Nagalingam, James A. Triccas, Colin W. Levy, Andrew W. Munro and Craig A. Hutton
†
‡
School of Chemistry and Bio21 Molecular Science and Biotechnology Institute, University of
Melbourne, 30 Flemington Rd, Parkville, VIC 3010, Australia
Centre for Synthetic Biology of Fine and Specialty Chemicals (SYNBIOCHEM), Manchester
Institute of Biotechnology, School of Chemistry, University of Manchester, 131 Princess Street,
Manchester M1 7DN, U.K.
¶
Department of Infectious Diseases and Immunology, Sydney Medical School, The University
of Sydney, Sydney, NSW 2006, Australia
chutton@unimelb.edu.au
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 56
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:
A series of analogues of cyclo(L-tyrosyl-L-tyrosine), the substrate of the Mycobacterium
tuberculosis enzyme CYP121, have been synthesized and analyzed by UV-Vis and EPR
spectroscopy and by X-ray crystallography. The introduction of iodine substituents onto cyclo(L-
tyrosyl-L-tyrosine) results in sub-M binding affinity for the CYP121 enzyme and a complete shift
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
III
to the high spin state of the heme Fe . The introduction of halogens that are able to interact with
heme groups is thus a feasible approach to the development of next-generation, tight binding
inhibitors of the CYP121 enzyme, in the search for novel anti-tubercular compounds.
ACS Paragon Plus Environment
2

Page 3 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
The successful development of effective antimicrobial agents has largely focused on the
1
-3
elucidation and targeting of biosynthetic pathways that are not present in humans. This approach
has been validated through numerous examples, including antibiotics that target the bacterial cell
wall (e.g. -lactams and vancomycin) and essential bacterial enzymes (e.g. sulfonamide drugs).
The knowledge necessary for the development of such antimicrobial agents includes a detailed
understanding of the molecular mechanisms of the enzymes present in such microbial biosynthetic
pathways.
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
The sequencing of the genome of Mycobacterium tuberculosis (Mtb)4-6 has unveiled several
essential genes from this pathogen that presumably encode for essential proteins. Amongst these
essential genes is one that encodes an unusual cytochrome P450 enzyme (CYP) known as
7
CYP121A1 (referred to as CYP121 hereafter). While most CYPs oxidize substrates through
8
incorporation of an oxygen atom, CYP121 catalyzes the oxidative coupling of two tyrosine
residues in the substrate cyclodityrosine (cYY, 1) to generate a C–C bond, forming the product
9
mycocyclosin 2 (Scheme 1). This CYP-catalyzed oxidative coupling is closely related to the
processes involved in biosynthesis of glycopeptide antibiotics such as vancomycin and
1
0
teicoplanin. CYP121 is not found in other microorganisms, nor humans, and is therefore a valid
target for development of specific agents against Mtb. In order to initiate such development, much
knowledge needs to be uncovered about this unusual enzyme.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 56
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
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
Scheme 1: CYP121 catalyzed biosynthesis of mycocyclosin 2 from cyclo(Tyr-Tyr) (cYY, 1).
CYP121 has been shown to bind tightly to azole antifungal drugs, suggesting this enzyme is a
possible target of such drugs.11,12 Azole drugs are known to affect antimicrobial activity through
inhibition of CYPs, though they are not highly selective for different CYP isoforms.11,13 Belin et
9
al. determined the substrate of CYP121 to be cyclodityrosine 1, and demonstrated that the enzyme
catalyzes oxidative C–C bond formation to generate mycocyclosin 2. The essential nature of the
gene encoding the CYP121 enzyme suggests mycocyclosin has a unique role in Mtb biology,
which has yet to be determined. Diketopiperazines (DKPs, cyclic dipeptides) such as
1
4
cyclodityrosine 1 are common secondary metabolites. However, oxidative coupling to generate
side-chain cross-links in such systems is highly unusual. The few structurally related natural
products that have been reported are the herqulines15-20 and piperazinomycin.21-23 The unusual
chemical process catalyzed by CYP121 and the unusual structural features of mycocyclosin24,25
suggest that selective inhibitors of this enzyme can be developed towards the search for novel anti-
tubercular therapies.
26
Preliminary structure-activity relationships have been reported by Belin and coworkers, who
showed the CYP121 can bind closely related substrate analogues including cYF and cYW with
comparable affinities to the natural substrate, but that these analogues do not exhibit significant
catalytic turnover. Only minor modifications to the substrate were tolerated with more significant
ACS Paragon Plus Environment
4

Page 5 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
variations to the substrate structure, including loss of one of the aromatic side chains, change in
stereochemistry or ‘ring opening’ to acyclic dipeptides, resulting in greatly diminished or complete
loss of binding affinity.
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
Small N-heterocycles have been discovered to bind to CYP121 with reasonable affinity.27-31
Several iterations of fragment-based discovery and intuitive design by Abell and co-workers have
unveiled the highest affinity compounds to date; anilinophenylpyrazoles which possess nM affinity
3
1
for CYP121 (K = 15 nM).
D
The extent of the structure–activity relationships of DKP-type substrate analogues has to date
been limited to readily accessible compounds prepared from commercially available amino acid
starting materials. We sought to probe structure–activity relationships of a wider range of substrate
analogues and to further define the binding requirements of CYP121 as an important step toward
developing potent inhibitors of this enzyme.
Design of substrate analogues
We describe herein the design and synthesis of a range of cYY analogues to further probe
structure–activity relationships of CYP121 binding compounds. Modifications at the 4-position of
the aromatic ring have been designed to interrogate the role of the tyrosyl phenolic groups on the
binding affinity to CYP121. Analogues that lack one or both phenolic groups (3, 4), or have one
or both phenolic groups blocked with O-methyl groups (5, 6), have been designed. Further, the
phenolic groups have been replaced with halogens (7, 8). Halogen (9–14) and methyl substituents
(15–18) have been introduced at positions ortho- or meta- to the phenols on the tyrosyl aromatic
rings, to probe the steric and electronic effects on binding to CYP121 (Figure 1).
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 56
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
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: Designed substrate analogues
RESULTS
Synthesis of substrate analogues
The synthesis of the substrate analogues 3, 4, 7–14 was conducted through a standard dipeptide
2
4
cyclization route (Scheme 2). N-Boc amino acids 19 were treated with an appropriate  -a min
methyl ester 20, N,N,N′,N′-tetramethyl-O-(1H-benzotriazol-1-yl)uronium hexafluorophosphate
(
HBTU, 1.1 equiv.) and N-hydroxybenzotriazole (HOBt, 1.1 equiv.) in DMF/acetonitrile to
generate the corresponding dipeptides 21 in high yields (87–96%). The dipeptides were treated
with excess formic acid to remove the Boc group, then cyclization was effected by heating in a
mixture of toluene and sec-butanol to obtain the corresponding diketopiperazines (3, 4, 7–14).
ACS Paragon Plus Environment
6

Page 7 of 56
Journal of Medicinal Chemistry
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
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
Scheme 2: Synthesis of diketopiperazines 3–6, 9–14.
Synthesis of the 4-O-methyl analogues 5 and 6 was performed at the dipeptide stage due to the
difficulties associated with the purification of diketopiperazines, which mainly emanate from their
highly insoluble nature. Thus, the tyrosyl–tyrosine dipeptide 21a was treated with methyl iodide
in the presence of anhydrous potassium carbonate to afford a mixture of doubly and singly
methylated dipeptides 21b and 21c in 60% and 30% yield, respectively (Scheme 3). The dipeptides
2
1b and 21c were converted to diketopiperazines 5 and 6 following the general procedure
described previously. The mono-methylated dipeptide 21b was isolated as a 1:1 mixture of
regioisomers, which were not separated as both isomers converge to the same diketopiperazine 5
after cyclization.
Scheme 3: Synthesis of methylated analogues 5 and 6.
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 56
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
As the tyrosine derivatives containing methyl substituents at the 2- and 3-positions of the
aromatic ring are not readily commercially available, a different route to compounds 15–18 was
required. These analogues were generated via a Knoevenagal-type condensation reaction 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
triacetyl tyrosyl–glycine DKP 23 with substituted benzaldehydes 24 (Scheme 4).32-34
Hydrogenation of the olefins 25 proceeds stereoselectively to generate the L,L-DKPs, with
concomitant hydrogenolysis of the O-benzyl protecting groups. Olefins 25a–c were susceptible to
hydrogenation at 6 bar in the presence of Pd/C over 16–24 hours. Subsequent deacetylation with
aqueous ammonia furnished cYY analogues 15–17. The 2,6-dimethyl olefin 25d required a higher
pressure (30 bar) to effect reduction, presumably due to increased steric hindrance. Under these
more forcing conditions, deacetylation of the intermediate occurred spontaneously to give 18
directly.
Scheme 4: Synthesis of DKPs 14–18 via Knoevenagel condensation/reduction.
The 3-fluoro analogue 14 could be prepared via this route (from 4-benzyloxy-3-
fluorobenzaldehyde 24e) using milder hydrogenation conditions (4–5 bar for 6 hours). However,
preparation of the other halogenated derivatives 11–13 was not possible due to dehalogenation
during the hydrogenation process.
ACS Paragon Plus Environment
8

Page 9 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Determination of binding affinities of substrate analogues to CYP121
Interactions of the natural substrate cYY 1 and analogues 3–18 with CYP121 were analyzed by
UV-Visible spectroscopy. The ligands were titrated from 0 μM to the point where saturation of the
signal was observed; i.e. there was no further change in absorbance. Typically, a decrease in the
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
III
intensity of the Soret peak at 416 nm (associated with the Fe low spin state) is observed together
III
with an increase in the intensity of the Soret peak at ~390 nm (associated with the Fe high spin
state) as the concentration of the ligand is increased (Figure 2). All analogues displayed a blue
shift of the Soret band (from 416 nm to ~390 nm) indicative of substrate-like or type-I binding.35
The change in absorbance was plotted against ligand concentration to generate hyperbolic curves
(Figure 3 and Figures S1–17) from which K values were derived (except for difluoro-analogue 5,
D
which exhibited a sigmoidal curve that was fitted to the Hill function, Fig.S6). The K values for
D
ligands 3–18 are listed in Table 1. Even at saturating concentrations of substrate 1 and most
substrate analogues, it was observed that there is a significant amount of enzyme remaining in the
low spin state (e.g. Figure 2A). However, iodinated compounds 9 and 11, and O-Me analogue 5,
resulted in a complete shift of the Soret band to the lower wavelength, high-spin state (e.g. Figure
2
B).
Figure 2: UV-visible spectrophotometric titration of CYP121 with the natural substrate 1 (A)
and the 3-iodo-analogue 9 (B).
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 56
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
For the natural substrate cYY 1, the K was determined to be 30±2 μM, which is in reasonable
D
9
agreement with the previously reported value of 19.4 μM. cYF 3 exhibited a K = 80.8±5.8 μM,
D
2
6
in reasonable accordance with the reported K value.
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
Figure 3: Plots of change in ligand-induced absorbance versus [ligand] for
representative examples of (A) weak (3), (B) moderate (5) and (C) tight-binding (9)
analogues.
Table 1: K values and percentage HS for binding of substrate 1 and analogues 3–18 to CYP121.
D
Compound
KD (μM)
% HS
Compound
KD (μM)
% HS
1
3
4
5
6
7
8
9
30 ± 2
52
5
11
12
13
14
15
16
17
18
0.28 ± 0.02
0.46 ± 0.02
1.5 ± 0.1
12.8 ± 0.6
1.1 ± 0.1
15.1 ± 0.3
39.8 ± 1.5
> 100
98
85
68
36
91
75
20
1
80.8 ± 5.8
2.4 ± 0.1
10.4 ± 0.5
29.4 ± 1.1
6.3 ± 0.3
0.72 ± 0.1
0.45 ± 0.04
3
61
35
1
3
100
ACS Paragon Plus Environment
10

Page 11 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
0
9.1 ± 0.1
74
Introduction of one ‘blocked’ or methylated phenolic group (5) resulted in a three-fold increase
in binding affinity (K = 10.4 μM) relative to cYY 1 (K = 30 μM). Introduction of the second O-
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
D
D
Me group (6) returned the binding affinity to a value similar to 1 (K = 29 μM). Intriguingly, these
D
results are the opposite of those observed where the phenols are removed rather than replaced: cYF
3
lacking one phenol has a decreased affinity for CYP121 (K = 80.8_μM), whereas cFF 4 lacking
D
both phenols has a considerably increased binding affinity (K = 2.4 μM) (Table 1).
D
Replacement of the phenols with halogens (I or F, 7 and 8) resulted in an increase in binding
affinity. The fluorine-containing analogue 7 displayed a K = 6.3 μM, between that of the natural
D
substrate cYY 1 and the cFF analogue 3, though it displayed a sigmoidal rather than hyperbolic
binding curve (Fig. S6). The iodine-containing analogue displayed an increased binding affinity
(
K = 0.7 μM).
D
Introduction of halogen substituents at position-3 resulted in increased binding affinity for
CYP121. Binding affinity increased with increasing size of the halogen; F<Cl<Br<I (Table 1, 11–
4). As the iodinated analogues 8 and 11 exhibited sub M affinities, we investigated the effect
1
of additional iodine substitution. Analogue 9 possessing 3-iodo substituents on both aromatic rings
displayed similar binding affinity to 11 (K = 0.45  M cf. 0.28  M), suggesting the second iodine
D
did not significantly affect affinity. Analogue 10 possessing a 3,5-diiodinated aromatic ring
displayed reduced affinity relative to 11 (K = 9.1  M cf. 0.28  M), though still improved
D
compared to cYY 1.
The 3-methylated analogue 15 displayed an increased binding affinity (K = 1.1 μM) relative to
D
cYY 1, though not as great as the equivalent 3-iodo analogue 11. Introduction of a second methyl
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 56
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
group in the form of the 3,5-dimethyl analogue reduced affinity (K = 15.1 μM), showing an effect
D
similar to the 3,5-diiodo analogue 10. The 2-methylated analogue 17 displayed slightly decreased
binding affinity (K = 40 μM) relative to cYY 1. The 2,6-dimethyl analogue, however, exhibited
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
no binding to CYP121 (K >>100 μM), with no change observed in the UV-Vis spectrum even at
D
1
20 μM (Figure S17A).
EPR studies
The binding interactions of 1 and 3–18 were analyzed using EPR spectroscopy. EPR
spectroscopy is extensively used in studies of CYP heme environment, ligand binding mode and
oxidation state, and can provide an in-depth insight into the ligand–porphyrin interactions affecting
the entire heme environment, particularly in the absence of further structural data. The EPR
spectrum for substrate-free CYP121 shows a typical rhombic trio of g-tensor elements, observed
for low spin heme iron.7
The EPR spectrum of CYP121 bound to the natural substrate 1 is shown in Figure 4B. The native
CYP121 enzyme is almost completely low-spin, with negligible high-spin content. The g-values,
together with an estimate of the amount of high spin complex observed, for substrate-free CYP121
and CYP121 bound to 1, 3–18 are shown in Table 2.
Table 2: Summary of EPR analysis for CYP121 and in the presence of 1, 3–18a
g values
LS)
g values
(HS)
g values
(LS)
g values
(HS)
Compd
~% HS Compd
~% HS
(
–
1
2.49, 2.25, 1.88
–
0
10
11
2.45, 2.25, 1.90
2.46, 2.24, ND
8.04, 3.54, 1.68
7.98, 3.56, 1.68
38
86
2.46, 2.25, 1.90 8.05, 3.44, ND
2.7
ACS Paragon Plus Environment
12

Page 13 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
4
5
6
7
8
9
2.45, 2.25, 1.90 8.05, 3.44, ND
2.49, 2.25, 1.88
1.4
0
12
13
14
15
16
17
18
2.46, 2.25, 1.90
2.46, 2.25, 1.90
2.46, 2.26, 1.89
2.46, 2.25, 1.90
2.46, 2.25, 1.90
2.46, 2.25,1.90
2.47, 2.25, 1.89
8.06, 3.46, 1.66
7.98, 3.49, 1.67
7.96, 3.51, ND
8.01, 3.49, 1.66
8.05, 3.47,1.66
8.01, 3.35, ND
7.94, 3.48, ND
68
21
–
2.45, 2.25, 1.90 8.00, 3.50, 1.66
2.47, 2.25, 1.89 7.96, 3.53, 1.68
64
1.2
11
56
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
2.49, 2.25, 1.88
2.49, 2.25, 1.88 7.83, 3.64, ND
7.96, 3.53, 1.68
7.8, ND, ND
1.5
7.5
100
62
0.6
1.9
–
a
EPR spectra available in the SI (Figures S1–S17).
Figure 4: EPR spectra of CYP121 and with ligands 1, 3, 5, 9 and 15 bound. (* = HS signal)
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 56
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
For cYF 3 ~1.4% high spin signal was observed (Figure 4C), while for cFF analogue 4 (Figure
S3), no high spin signal was observed, despite improved binding affinity compared to cYF 3. For
the 4-F analogue 7 and the 4-I analogue 8, the amount of high spin signal observed was comparable
to the natural substrate 1. Intriguingly, the EPR spectra for the mono- and bis-O-methyl analogues
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
5
and 6 showed much greater shift towards the high spin state (~64% and 56%, respectively,
Figures 4D and S5). The 3-iodo analogues 9 and 11 induced an almost complete shift towards the
high spin signal (Figures 4E and S10), indicative of tight binding with displacement of the H O
2
ligand.
X-ray crystallography studies
X-ray crystal structures were obtained for CYP121 bound to 5, 11 and 14–18. The 3-methyl
analogue with the substituted aromatic ring 15 (Figure 5A) binds within the enzyme active site.
The 3-methyl substituent is bound efficiently in the CYP121 binding pocket of the enzyme active
site distal to the heme, with no steric clashes or enzyme conformational changes observed. The
III
binding of the ligand does not displace the water molecule associated with the Fe ion as the sixth
ligand. The analogue 15 shows a direct hydrogen bonding interaction between the carbonyl group
of the DKP ring and the side chain of the Asn85 residue. Other hydrogen bonding interactions to
the ligand are formed through water molecules between the phenolic groups of the ligand and
Thr77, Gln385 and Arg386 residues of the enzyme. There are also additional hydrophobic
interactions between the 3-methyl substituent and the Phe168 and Val78 residues.
ACS Paragon Plus Environment
14

Page 15 of 56
Journal of Medicinal Chemistry
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
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 5: X-ray crystal structure of the 3-methyl analogue 15 and 3,5-dimethyl analogue 16
bound to CYP121 (PDB IDs 6RQ0, 6RQ5). The heme group is in red and the protein side chains
in green. The red spheres denote water molecules. The dashed lines represent probable hydrogen
bonds. (A); CYP with ligand 15 (purple). (B) and (C); two binding modes of CYP121 with ligand
1
6 (yellow).
The 3,5-dimethyl analogue 16 binds in two conformations: one conformation is similar to the 3-
methyl analogue 15 (Figure 5B), while the second conformation is nearly perfectly overlaid with
the first, but with the 3,5-dimethyl substituted aromatic ring orientated towards the heme (Figure
5
C). The 3,5-dimethyl analogue 16 shows the same hydrogen bonding interactions with the
residues Asn85, Thr77, Gln385 and Arg386 as observed for the analogue 15. Further, similar van
der Waals interactions between the 3-methyl substituent and Phe168/Val78 in the distal pocket are
observed. However, the extra methyl substituent in each orientation buttresses against active site
residues. In the first orientation, the 5-Me group is buttressed against the I helix, close to Val228,
while in the second orientation, the 5-Me group forces the Gln385 side-chain to move away to
accommodate the hydrophobic methyl group.
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 56
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 2,6-dimethyl analogue 18 (Figure 6A) binds within the enzyme active site in a flipped
conformation compared to the 3- and 3,5-substituted analogues, 15 and 16, with the DKP ring
III
interacting with the water molecule bound to the heme Fe ion. The substituted aromatic ring 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
1
8 binds in the distal pocket, orientated away from the heme moiety, while the unsubstituted
phenolic group is situated in the space occupied by the DKP ring of the natural substrate 1. Thus,
the phenolic group on the non-methylated aromatic ring shows a hydrogen bonding interaction
with the Asn85 residue and the DKP ring shows a hydrogen bonding interaction with the Arg386
residue – the reverse of the interactions observed for the 3-methylated analogues. The water-
mediated hydrogen-bonding interaction from the carbonyl groups on the DKP ring to Thr77 and
Gln385 is still observed for this flipped binding conformation of analogue 18, and the phenolic
group on the aromatic ring bound in the distal pocket retains similar hydrogen bonding interactions
to those observed for the other analogues. Arg386 makes a direct hydrogen bond with the carbonyl
group on the DKP ring which is pointing towards the heme group. An additional hydrogen bonding
interaction is observed from the nitrogen of the DKP ring to a water molecule. This flipped
conformation resembles the binding mode of the related TxtC–thaxtomin D enzyme–substrate
3
6
complex reported by Challis. TxtC catalyzes the double hydroxylation of the DKP substrate,
thaxtomin, via two distinct binding modes.
ACS Paragon Plus Environment
16

Page 17 of 56
Journal of Medicinal Chemistry
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
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 6: X-ray crystal structure of the 2-methyl analogue 17 and 2,6-dimethyl analogue 18 bound
to CYP121 (PDB IDs 6RQ1, 6RQ3). The heme group is in red and the protein side chains are in
green. The red spheres denote the water molecules. The dashed lines represent probable hydrogen
bonds. (A); CYP with ligand 18 (cyan). (B) and (C); two binding modes of CYP121 with ligand
1
7 (wheat).
The 2-methyl substituted analogue 17 exhibits two binding conformations (Figure 6B and C).
The first binding conformation is similar to the binding mode for 2,6-dimethylated analogue 18.
The second binding mode for the 2-methylated analogue 17 is similar to that of the 3,5-dimethyl
analogue 16 with the substituted aromatic ring in the pocket near the heme prosthetic group.
The 3-F analogue 14 binds in a similar manner to natural substrate 1, with the substituted
aromatic ring binding in the distal pocket (Figure 7). Two bound conformations are observed: one
binding mode is similar to the 3-methyl substituted analogue 15 with the fluoro substituent facing
toward the Thr77 residue. In the second binding mode, the substituted aromatic ring is flipped so
that the fluoro substituent is facing towards the Gln385 residue. The second binding mode is only
possible because the small size of the fluorine atom allows it to be accommodated within the active
site in the rotated conformation. All hydrogen bonding interactions, through a network of water
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 56
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
molecules, are similar for both binding conformations of analogue 14, and are the same as those
observed for the 3-methyl analogue 15.
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 7: X-ray crystal structures of the 3-fluoro analogue 14 bound to CYP121 (PDB ID 6RQ6).
The heme group is in red and the protein side chains are in green. The red spheres denote water
molecules. The dashed lines represent probable hydrogen bonds. (A) and (B); two binding modes
of CYP121 with ligand 14 (magenta).
Figure 8: X-ray crystal structures of the 3-chloro analogue 13 and 3-bromo analogue 12 bound
to CYP121 (PDB IDs 6RQD, 6RQB). The heme group is in red and the protein side chains are in
green. The red spheres denote the water molecules. The dashed lines represent probable hydrogen
ACS Paragon Plus Environment
18

Page 19 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
bonds. (A); CYP with ligand 13 (olive). (B) and (C); two binding modes of CYP121 with ligand
1
2 (blue).
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
The 3-Cl and 3-Br analogues, 13 and 12, bind in a similar manner to natural substrate 1, but in
contrast to the fluoro analogue 14 the halogen-substituted aromatic ring binds in the heme pocket
(Figure 8). The binding of the ligands does not displace the water molecule associated with the
III
Fe ion as the sixth ligand, and potential interactions between the halogen and the water molecule
are observed in each case. Two bound conformations are observed for the 3-bromo analogue 12:
one with the bromo substituent in the heme binding pocket and the other with the bromo substituent
facing towards the Thr77 residue in the pocket distal to the heme.
Figure 9: X-ray crystal structures of the 4-O-methyl analogue 5 and 3-iodo analogue 11 bound
to CYP121 (PDB IDs 6RQ9, 6RQ8). The heme group is in red and the protein side chains are in
green. The red spheres denote the water molecules. The dashed lines represent probable hydrogen
bonds. (A); CYP with ligand 5 (blue). (B); CYP121 with ligand 11 (light blue).
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 56
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
For the 4-O-methyl analogue 5 the substituted aromatic ring binds in the proximal pocket close
III
to the heme, with the O-methyl group situated very close to the heme Fe ion (Figure 9A). The
sixth (distal) H O ligand is completely displaced, which correlates with the large high-spin signal
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
observed in both the EPR and UV-visible spectra. The aromatic side chain rotates down slightly
III
towards the heme Fe ion so that the OMe group is directly above the Fe ion. It may be speculated
III
that there are specific CH –Fe interactions between the O-methyl hydrogens and the Fe ion that
3
contribute to the high binding affinity.37 The OMe moiety is also participating in a direct
hydrogen-bonding interaction with Arg386, bringing the ligand closer to the heme group.
The 3-iodo analogue 11 also binds with the substituted aromatic ring in the proximal pocket,
with the sixth H O ligand completely displaced (Figure 9B). The iodine atom is situated very close
2
III
to the heme Fe ion occupying the space otherwise occupied by the H O ligand, which again
2
correlates with the large high-spin signal observed in both the EPR and UV-visible spectra. The
phenolic group on the substituted aromatic ring is within hydrogen-bonding distance of Arg386
which indicates tight binding and also correlates with the high binding affinity observed for this
compound.
Turnover experiments
Turnover experiments were undertaken for compounds 2–18, but they failed to show any
significant level of CYP-catalyzed turnover for any compound except natural substrate cYY 2.
Antimicrobial assays
Most of the synthesized substrate analogues have a greater binding affinity for CYP121 than the
natural substrate cYY 1, with the iodinated analogues 8, 9 and 11 exhibiting up to a 100-fold
ACS Paragon Plus Environment
20

Page 21 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
greater binding affinity. It was therefore envisaged that these analogues might be able to function
as competitive inhibitors of CYP121 in vitro, thereby exerting anti-mycobacterial activity.
Various analogues were therefore tested for their MIC values in a resazurin-based Mtb cell
viability assay. Mtb H37Rv at an OD of 0.001 and the bacterial suspension was incubated with the
compounds at different dilutions for 7 days. Resazurin was then added to the wells and incubated
for a further 24 hours, and then the Mtb survival was calculated as a percentage of negative
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
3
8
controls. Unfortunately, all compounds exhibited little or no antimicrobial activity with MIC
values greater than 100 M (see SI). The dense, waxy cell wall of Mtb presents a major obstacle
to the penetration of molecules into the bacterial cell, which is a possible cause of the lack of
activity of the assayed compounds. Further modifications are underway to develop compounds
with better mycobacterial cell penetration.
DISCUSSION
The binding affinities of substrate analogues containing modifications at the phenolic group (4-
position) of the aromatic rings were scrutinized for structure-activity relationships of these
modifications. Firstly, it is apparent that the phenolic groups of substrate cYY 1 are not essential
for binding. Compounds with one or both phenols either removed, methylated or replaced with
iodine display similar or increased binding affinity. Removal of one phenol to give cYF results in
a decreased binding affinity, as may be expected with the loss of binding interactions (van der
Waals and hydrogen bonding) with this OH group. However, there is no apparent explanation for
the increased binding affinity of cFF which lacks both phenolic groups: no increase in binding
interactions are evident, and no high spin enzyme adduct is formed upon binding. Blocking of one
phenol with a methyl ether (compound 5) results in 3-fold improved binding and a large shift to
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 56
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
III
high spin Fe upon binding. X-ray crystallographic analysis shows that this is due to the OMe
th
group occupying the space of the 6 ligand water molecule, with possible OMe–Fe interactions
accounting for the improved binding. The bis-OMe compound 6 also induced a large shift to high
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
III
spin Fe , though it has slightly weaker binding than 5. Though X-ray data were not available for
CYP121 bound to 6, presumably the proximal pocket interacts with 6 in much the same way as 5
III
(
with favourable interactions of the CH group with the heme Fe ), with the additional OMe group
3
in the distal pocket resulting in the loss of a H-bond donor, thereby accounting for the slight loss
of affinity.
The 4-F and 4-I compounds 7 and 8 both display increased binding affinity compared with cYY
1. The fluoro analogue 7 displays an anomalous sigmoidal UV-Vis trace, which suggests allosteric
binding interactions. However, as the EPR spectrum shows little shift to high spin state and no X-
ray data could be obtained, little interpretation of this behavior can be postulated. The iodo
analogue 8 results in a moderate amount of the high spin state upon binding, and high binding
affinity. Though no X-ray data are available, the data for the 3-I compound 11 (vide infra) suggest
III
that the 4-I compound 8 binds with the iodine atom close to the heme Fe , such that ‘halogen
th
binding’ is invoked, and that this compound possibly displaces the 6 ligand water molecule.
The aryl-methylated compounds 15–18 provide us with information regarding the steric effects
of substrate binding to CYP121. Substituents at the 2-position of the aromatic ring are not well
tolerated. Though introduction of a single 2-Me substituent on the ring results in only a slight
decrease in binding affinity, the X-ray data demonstrate that, when binding in the standard
orientation and in order to avoid unfavorable interactions with the heme, the 2-Me group is forced
to be positioned close to the DKP carbonyl oxygen. Hence, a flipped binding conformation, where
the DKP ring and the tyrosyl side chain essentially swap positions, becomes a competitive binding
ACS Paragon Plus Environment
22

Page 23 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
orientation. In the 2,6-diMe compound 18, binding in the standard orientation becomes highly
unfavorable, as one of the Me substituents would have a severe steric clash with the heme group.
Accordingly, this compound binds only in the flipped conformation, though only very weakly (KD
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
>
100 M).
The 3-Me substituted analogue 15 displays excellent affinity for CYP121, and induces a
III
moderate amount of high spin Fe upon binding. In contrast the 3,5-dimethyl compound 16
exhibits moderate binding affinity, yet induces a high amount of the high spin state. The X-ray
data enable an explanation to be forthcoming, which highlights the intriguing variations that can
arise when a symmetrical substrate (or pseudo-symmetrical substrate analogue) binds to an
asymmetric active site. The 3-Me analogue 15 binds with the Me-substituted aromatic ring in the
distal binding pocket. The increase in binding affinity is presumably a result of increased
hydrophobic binding interactions of the additional Me group with Phe168 and Val78. The standard
tyrosyl group binds adjacent to the heme in the proximal pocket, such that only a moderate amount
III
of high spin Fe is induced. In contrast the 3,5-diMe analogue 16 binds in two conformations,
with the substituted aromatic ring equally disposed to binding in the distal pocket or the proximal
pocket. Additional hydrophobic interactions may be gained upon binding in the distal pocket,
while in the proximal pocket one of the Me groups binds close to the heme Fe. Though the X-ray
th
structure shows the 6 ligand water is still in place, steric interactions with the 3-Me substituent
may favor its displacement, thereby increasing the conversion to the high-spin Fe state. The second
methyl group (5-Me) in each conformation gives rise to slightly unfavorable interactions such that
the overall binding affinity is weaker than the 3-Me analogue 15.
The 3-halo analogues 9 and 11–14 all bind with increased affinity relative to 1, with the smallest
improvement being for the fluoro compound 14 and then a stepwise increase in affinity with
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 56
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
increasing size to the iodo compound 11 (i.e. K F<Cl<Br<I). The difference observed for the
D
halogenated analogues may be due to steric and/or electronic effects. Analysis of the X-ray data
for the 3-F analogue 14 shows the fluorinated aromatic ring binds in the pocket away from the
heme group, and in both conformations – i.e. with the C–F bond pointing towards helix I or away
from it. Presumably the small size of the fluorine atom results in insignificant steric clashes with
nearby active site residues, with minor increases in van der Waals contacts resulting in slightly
improved binding affinity, relative to 1. No significant electronic influence is apparent, as the
electron withdrawing fluorine group should have a greater impact on hydrogen bonding from the
adjacent phenol than the other halogens. The 3-Cl and 3-Br analogues, 13 and 12, bind with the
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
III th
halogen in the heme pocket: though they do not displace the heme Fe 6 ligand water molecule,
the halogens interact with the water molecule in a manner similar to that reported for the nitrile
3
9
moiety of bicalutamide to CYP46A1. The 3-I analogue 11 binds with the substituted ring situated
III
close to the heme prosthetic group. The iodine atom is situated very close to the heme Fe and
th
intriguingly this analogue displaces the 6 ligand water molecule upon binding. This observation
is in good correlation with both the UV-Vis and EPR spectroscopic data, which show the enzyme
undergoes an almost complete shift to the high-spin state upon binding 11. It is interesting to
compare the 3-I, 3-Br and 3-Me analogues, which all exhibit tight binding to CYP121 (1.1, 0.5,
0
.3 M, respectively). The size of an iodine substituent is slightly larger than a Me group (2.15 vs
2.00 Å), with a bromine substituent (1.95 Å) slightly smaller than a methyl group. Could van der
Waals interactions thus account for the difference in affinity? The X-ray data show that this is not
the case, with 11, 12 and 15 binding in different orientations. The smallest substituent, the 3-Br
group of 12, binds in two conformations; either away from the heme or in the pocket close to the
heme but without displacing the water ligand. The slightly larger substituent, the 3-Me group of
ACS Paragon Plus Environment
24

Page 25 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
5, only binds in the pocket away from the heme, whereas the largest group, the 3-I of 11, binds
in the pocket close to the heme and does displace the water ligand. Thus, we presume that the
iodine, due to its large size and polarizability, binds such that favorable halogen-bonding
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
40
interactions with the heme group give rise to sub M level binding affinity. The 3-Br and 3-Cl
analogues 12 and 13, bind in a similar fashion, with decreased size and polarizability (c.f. 3-I)
resulting in gradually poorer affinity and a lack of displacement of the water molecule. The non-
polarizable 3-F substituent binds with the substituted aromatic ring in the pocket distal to the heme,
where small improvements in van der Waals contacts give a marginal improvement in binding
affinity.
CONCLUSION
In conclusion, we have synthesized and analyzed a family of analogues of the substrate of the
Mtb CYP121 enzyme to unveil structural information about this enzyme and its ligands. CYP121
in the presence of these analogues was analyzed by UV-Vis and EPR spectroscopy, and by X-ray
crystallography, yielding information on the binding affinity of the substrate analogues, enzyme
III
Fe spin state and the active site structure. The introduction of a single iodine atom onto the
substrate of CYP121 results in sub-M binding affinity and a complete shift to the high spin state
III
of the heme Fe . We speculate that the high binding affinity of the 3-iodo analogue 11 is a result
III
of a specific interaction between the iodine atom and either the Fe ion or the heme group, such
40
as ‘halogen bonding’ or ‘-hole bonding’. The introduction of halogens that are able to interact
with heme groups is thus a reasonable approach to the development of next-generation, tight
binding inhibitors of the CYP121 enzyme, in the search for novel anti-tubercular compounds.
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 56
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
EXPERIMENTAL SECTION
General experimental methods
Commercially available reagents and solvents were used without further purification unless
otherwise stated. Acetone was dried over calcium sulfate, potassium carbonate was dried in a hot
air oven (130 °C) overnight. Molecular sieves were activated with a conventional microwave and
allowed to cool under vacuum. DMF was dried over activated sieves (4 Å) for 16 h before use.
Progress of the reactions was monitored by analytical thin-layer chromatography using Merck pre-
coated silica gel 60 aluminium plates with F254 indicator. Visualization of spots was accomplished
by using UV light (254 nm), DNPH (2,4-dinitrophenyl hydrazine), p-anisaldehyde or
phosphomolybdic acid stain. Purification of crude mixtures was performed by flash
chromatography on silica gel 40–63 μa with eluting solvents reported as % v/v mixtures. Yields
refer to chromatographically and spectroscopically pure compounds, unless otherwise indicated.
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
1
13
All compounds were determined to be of >95% purity by H NMR analysis. H NMR and C
NMR spectra were recorded in CDCl , CD OD or d -dimethyl sulfoxide (DMSO-d ) solution, on
3
3
6
6
1
13
1
13
a Varian Unity Inova 400 (400 MHz H, 100 MHz C) or 500 (500 MHz H, 125 MHz C)
spectrometer. Chemical shift values (δ) are reported in ppm relative to CHCl (δ = 7.26 ppm),
3
1
DMSO-d (δ = 2.50 ppm) and CD OD (δ = 3.31 ppm) for H NMR and relative to the central
6
3
CDCl resonance (δ = 77.16 ppm), central DMSO-d resonance (δ = 39.52 ppm) and central MeOD
3
6
1
3
resonance (δ = 49.00 ppm) for C NMR. The proton spectra are reported as follows δ (multiplicity,
coupling constant J in Hz, number of nuclei). Multiplicities are indicated by s (singlet), d (doublet),
dd (doublet of doublet), dt (doublet of triplet), dq (doublet of quartet), t (triplet), q (quartet), m
(multiplet) and br s (broad singlet), etc. Mass spectra were recorded using a NanoLC/OrbiTrap
FusionTM LumosTM spectrometer.
ACS Paragon Plus Environment
26

Page 27 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
General procedure A: Synthesis of DKPs
The Boc dipeptide methyl ester was dissolved in formic acid (5 mL/mmol) and stirred at room
temperature. When the Boc-deprotection was complete (by TLC analysis, 2–3 hours) formic acid
was evaporated under reduced pressure. A mixture of sec-butanol and toluene (1:4, 0.01 M) was
added and the solution was heated at 90 °C for 4 h. The solvent was evaporated under reduced
pressure to yield an off-white solid. Trituration of the solid with ethyl acetate afforded the pure
DKP.
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
General procedure B: Synthesis of N-acetyl-(-dehydro)-DKPs
To a solution of N,N-diacetyl-cyclo(glycine-L-tyrosine(OBn)) (394 mg, 1 mmol) in dry DMF (3
mL) under nitrogen atmosphere was added Cs CO (326 mg, 1 mmol), the benzaldehyde derivative
2
3
(2 mmol) and molecular sieves 3Å (100 mg). The reaction mixture was stirred at room temperature
until the reaction was complete. The reaction mixture was diluted with water (300 mL) and then
extracted with ethyl acetate (3 × 100 mL). The aqueous layer was discarded and the ethyl acetate
layer was washed with brine (3 × 50 mL), dried over MgSO , filtered and evaporated under reduced
4
pressure to afford the crude product. The crude product was purified by flash chromatography,
eluting with a gradient of 20–30% ethyl acetate in hexane to obtain the olefin.
General procedure C: Synthesis of N-acetyl DKPs
To a solution of olefin (1 equiv.) in ethyl acetate/methanol (1:1, 0.01 M) was added 10% Pd/C
(
50% by weight) and the reaction mixture was stirred at room temperature under a hydrogen
atmosphere until the reaction was complete. The reaction mixture was filtered through a silica plug
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 56
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
followed by removal of the solvent under reduced pressure to obtain the crude product. The crude
product was purified by flash chromatography using 5% MeOH in DCM to obtain the N-acetyl
DKP.
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
General procedure D: N-Acetyl deprotection
The N-acetyl DKP (1 equiv.) was stirred in 10% aqueous NH (0.2 M) for 15 minutes followed
3
by removal of aqueous NH by lyophilization. The crude product was purified by flash
3
chromatography in 10–20% MeOH in DCM.
cyclo(L-Tyrosine-L-tyrosine) 1
The title compound was prepared from Boc-L-tyrosine-L-tyrosine-OMe (688 mg, 1.5 mmol)
according to general procedure A and was obtained as a white solid (450.0 mg, 92%). M.p. 288–
290 °C (lit. m.p. 270–275 °C).24 1H-NMR (400 MHz, DMSO-d ): δ 9.20 (s, 2H), 7.76 (d, J = 2.0
6
Hz, 2H), 6.84 (d, J = 8.4 Hz, 4H), 6.67 (d, J = 8.4 Hz, 4H), 3.85–3.83 (m, 2H), 2.53 (dd, J = 4.6
1
3
Hz, 2H, partially obscured by DMSO peak), 2.10 (dd, J = 13.7, 6.5 Hz, 2H). C-NMR (125 MHz,
DMSO-d ): δ 166.2, 156.0, 130.7, 126.5, 115.0, 55.7, 38.8. NMR data consistent with that
6
2
4
+
+
reported. HRMS (ESI) calcd. for C H N O [(M+H) ] 327.1345, found 327.1346.
1
8
19
2
4
cyclo(L-Phenylalanine-L-tyrosine) 3
The title compound was prepared from Boc-L-phenylalanine-L-tyrosine-OMe (664 mg, 1.5
mmol) according to general procedure A and was obtained as a white solid (399.0 mg, 86%). M.p.
2
90–293 °C (lit. m.p. 291–293 °C).41 1H-NMR (400 MHz, DMSO-d ): δ 9.22 (s, 1H), 7.85 (br s,
6
2H), 7.29–7.18 (m, 3H), 7.03 (d, J = 7.1 Hz, 2H), 6.83 (d, J = 8.4 Hz, 2H), 6.67 (d, J = 8.3 Hz,
ACS Paragon Plus Environment
28

Page 29 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
H), 3.94 (m, 1H), 3.89 (m, 1H), 2.58 (dd, J = 13.6, 4.8 Hz, 1H), 2.22–2.15 (m, 2H), one dd
1
3
obscured by the DMSO peak. C-NMR (125 MHz, DMSO-d ): δ 166.3, 166.2, 156.1, 136.6,
6
1
3
130.8, 129.7, 128.2, 126.5, 126.4, 115.0, 55.7, 55.4, 38.5, (one C peak obscured by DMSO).
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
NMR data consistent with that reported.42 HRMS (ESI) calcd. for C H N O + [(M+H)+]
1
8
19
2
3
3
1
11.1390, found 311.1390.
cyclo(L-Phenylalanine-L-phenylalanine) 4
The title compound was prepared from Boc-L-phenylalanine-L-phenylalanine-OMe (640 mg,
.5 mmol) according to general procedure A and was obtained as a white solid (396 mg, 90%).
M.p. 296–298 °C (lit. m.p. 297–300 °C).43 1H-NMR (400 MHz, DMSO-d ): δ 7.92 (br s, 2H),
6
7
2
.30–7.19 (m, 6H), 7.04 (d, J = 7.1 Hz, 4H), 3.95–3.94 (m, 2H), 2.57 (dd, J = 13.6, 4.8 Hz, 2H),
1
3
.23 (dd, J = 13.6, 6.2 Hz, 2H). C-NMR (100 MHz, DMSO-d ): δ 166.2, 136.6, 129.8, 128.2,
6
126.5, 55.4, 39.4. NMR data consistent with that reported.44
cyclo(L-Tyrosine(OMe)-L-tyrosine) 5
The title compound was prepared from the mixture of regoisomers 21b (270 mg, 0.57 mmol)
according to general procedure A and was obtained as a white solid (131.0 mg, 68%). M.p. 252–
1
2
54 °C. H-NMR (400 MHz, DMSO-d ): δ 9.22 (s, 1H), 7.83 (br s, 1H), 7.80 (br s, 1H), 6.95 (d,
6
J = 8.6 Hz, 2H), 6.86–6.83 (m, 4H), 6.68 (d, J = 8.5 Hz, 2H), 3.88 (m, 2H), 3.70 (s, 3H), 2.56–
1
3
2.50 (m, 2H), 2.21–2.08 (m, 2H). C-NMR (100 MHz, DMSO-d ): δ 166.26, 166.24, 158.0, 156.1,
6
1
30.8, 130.8, 128.5, 126.5, 115.0, 113.7, 55.7, 55.6, 55.0, 39.5, 38.6. HRMS (ESI) calcd. for
+
+
C H N O [(M+H) ] 341.1496, found 341.1496.
1
9
21
2
4
ACS Paragon Plus Environment
29