Selective inhibition of Helicobacter pylori methionine aminopeptidase by azaindole hydroxamic acid derivatives: Design, synthesis, in vitro biochemical and structural studies

Article abstract of DOI:10.1016/j.bioorg.2021.105185

Methionine aminopeptidases (MetAPs) are an important class of enzymes that work co-translationally for the removal of initiator methionine. Chemical inhibition or gene knockdown is lethal to the microbes suggesting that they can be used as antibiotic targ

Full text of DOI:10.1016/j.bioorg.2021.105185

Bioorganic Chemistry 115 (2021) 105185
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Selective inhibition of Helicobacter pylori methionine aminopeptidase by
azaindole hydroxamic acid derivatives: Design, synthesis, in vitro
biochemical and structural studies
a
,
b
,
1
c
,
1
c
Sandeepchowdary Bala
, Kalisha vali Yellamanda , Anilkumar Kadari , Venkata.S.
a
a,
b
c
c,*
U. Ravinuthala , Bhavita Kattula , Om V. Singh , Rambabu Gundla
,
a,b,*
Anthony Addlagatta
a
Division of Applied Biology, CSIR-Indian Institute of Chemical Technology, Hyderabad 500 007, Telangana, India
Academy of Scientific and Innovative Research (AcSIR), Ghaziabad, Uttar Pradesh 201 002, India
b
c
Department of Chemistry, School of Science, GITAM Deemed to be University, Hyderabad 502 102, Telangana, India
A R T I C L E I N F O
A B S T R A C T
Keywords:
Methionine aminopeptidases (MetAPs) are an important class of enzymes that work co-translationally for the
removal of initiator methionine. Chemical inhibition or gene knockdown is lethal to the microbes suggesting that
they can be used as antibiotic targets. However, sequence and structural similarity between the microbial and
host MetAPs has been a challenge in the identification of selective inhibitors. In this study, we have analyzed
several thousands of MetAP sequences and established a pattern of variation in the S1 pocket of the enzyme.
Based on this knowledge, we have designed a library of 17 azaindole based hydroxamic acid derivatives which
selectively inhibited the MetAP from H. pylori compared to the human counterpart. Structural studies provided
the molecular basis for the selectivity.
Methionine aminopeptidase
Selective inhibition
Azaindole
Hydroxamic acid
Helicobacter pylori
1
. Introduction
Ribosome assisted protein synthesis in almost all living cells begins
that include ovalicin inhibit Type 2 MetAP by a million-fold better than
Type 1 enzyme, suggesting that there are differences between the two
active sites [9,10]. All prokaryotes contain only Type 1 MetAPs.
Depending upon N-terminal extensions or insertions within the Type 1
map genes, these have been sub-classified as Type 1a, 1b, 1c, 1d, and 1n
(Fig. S1) [11]. All MetAPs are metalloenzymes and the active site ac-
cepts at least two metal ions from first row transition elements. Because
of their criticality, MetAPs are identified as good drug targets in anti-
microbial and anticancer therapies.
with the amino acid methionine in eukaryotes while it is formyl
methionine in prokaryotes [1,2]. Methionine aminopeptidase (MetAP)
class of enzymes cleave the initiator methionine in 60–70% of proteins
in all living cells [3,4]. In prokaryotes, the peptide deformylase (PDF)
removes the N-terminal formyl group before the methionine hydrolysis
[
5]. Both these hydrolytic processes on the amino-terminus at the
ribosome exit tunnel are essential for the survival of all living cells [6].
Most prokaryotes contain a single gene of map while eukaryotes have
redundant genes. Chemical inhibition or gene knockout in E. coli and
other microbes was detrimental [7,8]. Eukaryotic cytosolic MetAP
proteins are classified as Type 1b and Type 2 while the mitochondrial
MetAP is classified as Type 1d. Deletion of a single map gene in eu-
karyotes results in slow growth phenotypic cells but detrimental when
both were deleted [8]. Natural product fumagillin family compounds
Since prokaryotic MetAPs are good antibacterial targets, several
studies have been reported to identify inhibitors [12,13–15]. However,
the biggest challenge is in selective inhibition of microbial enzymes
because of the similarity of their active site with human enzyme [12,13].
For more than a decade, others and we have studied active sites of a
large number of MetAPs from microbes, specifically from point of view
of inhibitor selectivity [12,13,16–24]. Based on our earlier studies, we
have identified that the S1 pocket could be separated into two parts: a
Abbreviations: AMC, 7-Amido-4-methylcoumarin; HATU, (1-[Bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate.
*
Corresponding authors.
E-mail addresses: rgundla@gmail.com (R. Gundla), anthony@csiriict.in (A. Addlagatta).
Authors equally contributed to work
1
https://doi.org/10.1016/j.bioorg.2021.105185
Received 1 March 2021; Received in revised form 4 June 2021; Accepted 14 July 2021
Available online 21 July 2021
0
045-2068/© 2021 Elsevier Inc. All rights reserved.

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
conserved and a variable lobes. Based on these fine differences, we have
identified selective inhibitors against MetAP from pathogenic microbes
OD280 reached zero. Continued the column wash with buffer B (50 mM
HEPES buffer pH 8.0, 150 mM NaCl and 10 mM imidazole) until the
absorption at OD280 reached zero. Pure protein was eluted by gradient
elution of imidazole (10 mM–200 mM) in buffer B. Pooled fractions were
dialyzed into storage buffer (25 mM HEPES pH 8.0 and 150 mM NaCl),
[
12,13].
Almost half of the world’s population is infected with H. pylori [25].
It is a spiral shaped gram-negative bacterium that grows in the digestive
tract. H. pylori invade the stomach lining causing ulcer formation which
in turn has a high risk of stomach cancer. Eradication of H. pylori has
been difficult due to its drug resistance towards antibiotics (amoxicillin,
clarithromycin, and other drugs metronidazole) [26]. Triple therapy
such as levofloxacin or rifabutin in combination with amoxicillin and
esomeprazole has been used to overcome the drug resistance curing
nearly 90% of the cases [26]. There is a necessity for new drugs that can
selectively and effectively target the H. pylori enzymes. Based on our
earlier understanding, we identified H. pylori MetAP1a (HpMetAP1a) as
a starting point to discover specific inhibitors against this organism. We
have designed and synthesized a library of 17 compounds that contain
hydroxamic acid moiety as a metal binding scaffold. Using biochemical
and structural biology tools we have determined molecular insights into
selective inhibition of enzyme from H. pylori compared to the human
counterpart.
◦
concentrated to 10 mg/ml, and stored at ꢀ 80 C until further use.
2
.4. Effect of different metal ions on enzyme activity
HsMetAP1b was shown to be active in the presence of Co [12,27].
2+
Metal selectivity was tested for HpMetAP1a (5 µM of the enzyme) using
2+
2+
,
various concentrations (1–1000 µM) of divalent metal ions (Ni , Co
2+,
2+
Mn
and Zn ) in 25 mM HEPES pH 8.0 and 150 mM NaCl and
◦
incubated for 20 min at 37 C. After incubation for 30 min, 50
μ
M of
methionine 7-amino-4-methyl coumarin (Met-AMC) was added to the
reaction and the release of free AMC was monitored continuously using
a microtiter plate reader (TECAN, Austria) with excitation and emission
wavelengths of 380 nm and 460 nm, respectively.
2
.5. Determining the optimum pH
2
. Materials and methods
The effect of pH on the activity of HpMetAP1a was determined using
acetate buffer (pH 5.0, 5.5), sodium phosphate buffer (pH 6.0 and 6.5),
HEPES (pH 7.0, 7.5 and 8.0), Tris (pH 8.5) and sodium carbonate (pH
2
.1. Materials
9
.0) at 25 mM concentration with 1 µM of the enzyme, two equivalents
Phusion high fidelity DNA polymerase was obtained from NEB (USA)
of NiCl
AMC).
2
, and 50 M of methionine 7-amino-4-methyl coumarin (Met-
μ
while Taq DNA polymerase was obtained from LAADH Biotech Pvt. Ltd,
Hyderabad, India. All oligonucleotides were procured from BioArtis Life
Sciences Pvt. Ltd, Hyderabad. All buffers, fine chemicals, solvents, and
substrates used for synthesis and enzyme assays were obtained from
Sigma-Aldrich (USA), SRL (India), GenPro Biotech (India), Alfa Aesar
2
.6. Determining the substrate specificity
AMC derivatives of various amino acids) (Asp, Asn, Ala, Glu, Gln,
(
India), and TCI (India).
Arg, His, Ile, Leu, Lys, Met, Ser, Phe, and Thr) were used to test the
activity of the HpMetAP1a. All substrates used for this study have free
amino terminus.). Reactions were carried out with a final volume of 100
µl having 25 mM HEPES (pH 8.0), 150 mM NaCl, 5 µM of HpMetAP1a,
2
.2. Cloning
map gene (UniProt ID: P56102) was amplified from the genomic DNA
of H. pylori (ATCC 26695) using polymerase chain reaction and cloned
into the pET15b vector using the NdeI and BamHI restriction sites. The
Δ91 human MetAP1b (HsMetAP1b) which is devoid of the N-terminal
zinc fingers is used in the present study. Details about its characteriza-
tion and biochemistry are published elsewhere [12,27]. A list of oligos
used in this study is given in the Supplementary Information (Table S1).
2
00 µM of substrate concentration, and two equivalents of NiCl
2
, in a 96-
◦
well black flat-bottom microtiter plate (Corning Inc., USA) at 37 C.
Kinetic studies were carried out using different concentrations of sub-
strates and the data were analyzed using SigmaPlot 13.0.
2
.7. Enzyme kinetics
2
.3. Protein expression and purification
The K
NiCl
M
and kcat of HpMetAP1a towards Met-AMC in the presence of
2
were determined (Table S2). At a fixed concentration of 5 µM of
Purification of HsMetAP1b was performed as described previously
HpMetAP1a, the substrate concentration was varied between 25 µM and
(
Fig. S2) [12]. A similar method was followed for the HpMetAP1a.
◦
1
200 µM and the reaction was performed at 37 C in triplicates. K
M
and
Briefly, the pET15b plasmid was transformed into BL21 (DE3) E. coli
strain for protein expression. A single colony was inoculated in LB broth
having 50 µM of ampicillin. The overnight culture was added to four
V
max were determined from slopes of various concentrations of the
substrate by applying a non-linear curve fit analysis. The turnover
number and kcat values were determined manually by applying the
formula kcat = Vmax/[E], where Vmax is the maximum velocity and [E] is
the total enzyme concentration. Data were fitted against the Michaelis-
◦
liters of LB media and incubated at 37 C by shaking at 250 rpm. Protein
expression was induced by the addition of IPTG to a final concentration
◦
of 1.0 mM and incubated at 25 C by shaking at 150 rpm for 16 h. The
Menten equation: v = Vmax × [S]/(K + [S]), using SigmaPlot.
M
bacterial cell pellet was collected by centrifugation of the culture at
◦
6
,000g for 20 min at 4 C. The cell pellet was resuspended in 50 mL of
buffer A(50 mM HEPES buffer pH 8.0 150 mM NaCl, 0.1% Triton X-100,
0% glycerol, and 10 mM imidazole), in the presence of protease
cocktail inhibitor (5 mg), 50 mg of lysozyme, 10 mg of DNase and 2.5
mM of MgCl . After lysozyme treatment, the cell suspension was sub-
2.8. Inhibitor studies
1
10 mM stocks of compounds were prepared in dimethyl sulfoxide
(DMSO) solution. Initially, all compounds were tested for inhibition of
the enzyme at 10 µM against MetAP enzymes from H. pylori, E. coli,
V. coralliilyticus, S. pneumoniae, and human MetAP1b (Table S6). Ki
values for the compounds within the range of tight binding inhibition
were determined (using the Morrison method) [28] by fitting the data
using non-linear regression in the formula listed below.
2
jected to sonication (sonicator with model number PKS-250F (PCI An-
alytics Pvt. Ltd.)) with a frequency of 20 kHz and power 250 W and with
◦
a probe diameter of ¼ inch (70% amplitude at 4 C, 10 min (2 s on and 4
s off)) to achieve complete lysis. Cell debris was removed by centrifu-
gation at 35,000g for 40 min. The lysate was passed through a nickel
NTA affinity column pre-equilibrated with buffer A. After passaging the
lysate, the column was washed with buffer A until the absorption at
2

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
√
̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
(
(
(
) ) )
(
(
(
) ) )
2
1 + [
S]
ꢀ
[E] + [I] +
K
1 +
[S]
ꢀ 4[E][I]
[E] + [I] +
K
i
i
KM
KM
v
i
=
1 ꢀ
v
0
2[E]
2
.9. Crystallization and X-ray data collection
Crystallization of HpMetAP1a was set up with 12 different crystal-
Cryoprotectant solution was prepared with 25% glycerol in well solu-
tion. Crystals were soaked with cobalt and compound 7d or compound
7e5. Single Crystal was frozen in the stream of liquid nitrogen and
shipped to Elettra Synchrotron, Trieste, Italy. X-ray diffraction data were
collected on the XRD2 beamline. Diffraction data were processed and
scaled using XDS [29]. Since all structures were homologous to the wild-
type HsMetAP1b structure, they were directly refined using the 2B3H
PDB [30] coordinates using the REFMAC5 [31] in the CCP4 [32] crys-
tallographic suite after removing all water molecules. Structure
lization screens (Hampton research and Molecular Dimensions), which
corresponds to about 1,200 reagents. No condition yielded diffraction
quality crystals. Crystallization of HsMetAP1b was set up using the
following method: 10 mg/ml protein in storage buffer was mixed in 1:1
ratio with well solution (0.1 M BISTRIS pH 6.2, 19% PEG 3350, and 5%
◦
glycerol) and incubated at 25 C. Rod-like crystals appeared in 24–48 h.
3

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
Table 1
Showing active site residue variations within Type
aminopeptidase.
1
methionine
S.
Left lobe
Right lobe
No
1
2
Site-1 (P-192): P(40.76), F(19.23), Y(16.38), N(5), T
F(73.5),Y(26.5)
(
(
4.61), L(3.53), Q(3.46), S(2.76), C(2.23), E(1.53), A
0.153), G(0.153), R(0.07)
Site-2 (Y-195): Y(72.46), P(5.53), T(5.153), C(3.61),
G(3.07), N(2.07), F(1.92), V(1.07), S(0.53), L(0.61),
D(0.69), E(0.384), Q(0.15), A(0.15), I(0.15), H
C(97),N(2.53), L
(0.30),T(0.15)
(
0.076), M(0.07)
Site-3 (F-309): F(59.30), M(17), L(12.84), Y(5.38), I
2.30), V(1.30), P(0.07)
3
W(100)
(
Fig. 2. Structural alignment showing S1 pocket residues of HsMetAP1b in
sticks (green) in complex with methionine (violet) (PDBID: 4U6J) and HpMe-
tAP1a (pink) homology model. (For interpretation of the references to colour in
this figure legend, the reader is referred to the web version of this article.)
and Gasteiger charges, using AutoDock Tools [40] of the MGLTools
package. A cube with a side of 14 Å was used for generating 20 docking
poses. Visualized the docking poses and generated representative figures
using open-source PyMOL [34].
2
.12. Chemistry
The chemistry used for the formation of azaindole esters 5a-5h and
Fig. 1. Active site residue variations among Type 1 methionine aminopepti-
dases S1 pocket residues are shown in sticks (green) (PDB ID: 4U6J) are shown.
Metal-binding residues and gatekeeper residues are shown in grey. (For inter-
pretation of the references to colour in this figure legend, the reader is referred
to the web version of this article.)
5
7
a1-5j10, azaindole acids 6a-6h and 6a1-6j10, and final products 7a-
h and 7a1-7j10 are depicted in Schemes 1 and 2. Bromo azaindole was
converted to the corresponding cyano with zinc cyanide in the presence
of catalytic amounts of Pd(PPh provided nitrile 2 and 2a, which was
3 4
)
converted to acid 3 and 3a followed by esterification to provide ester 4
and 4a. Azaindole esters were alkylated using benzyl halides and po-
tassium carbonate to provide esters 5a-5h and 5a1-5j10. Saponification
modeling against observed structure factors was performed in COOT
[
33]. All graphics generated for this manuscript were prepared using
2
of esters with LiOH⋅H O provided acids 6a-6h and 6a1-6j10.
Pymol [34].
Hydroxamic acids 7a-7h and 7a1-7j10 were prepared by treatment of
hydroxylamine hydrochloride by using HATU and triethylamine to get
corresponding N-alkylated hydroxamic acids.
2
.10. Homology modeling
Scheme for the synthesis of azaindole hydroxamic acid
derivatives:
SWISS-MODEL [35] is used to generate 3D homology models of
proteins which is a web-based bioinformatics tool. The homology model
of the HpMetAP1a was built using HsMetAP1b (PDB ID: 4U6J) as a
template (32.7% sequence identity) (Fig. S3). The model was validated
using the RAMPAGE [36] (Fig. S4) for the assessment of the Ram-
achandran plot.
Target molecules:
Nuclear magnetic resonance spectra (1H NMR: 400 MHz and 13C
3 6
NMR: 100 MHz) were recorded in CDCl and DMSO‑d using an Avance
III 400 MHz spectrometer (Bruker). Chemical shifts are reported in parts
per million (ppm) and the coupling constants (J) are expressed in Hertz
(
Hz). Splitting patterns are designated as follows: s, singlet; d, doublet; t,
2
.11. Docking of azaindole hydroxamic acid derivatives on human Type 2
triplet; q, quadruplet; m, multiplet; brs, broad singlet; dd, double of
doublets. Analytical thin-layer chromatography (TLC) was carried out
on silica gel F-254 plates (Merck). The solvents used in MS measures
were acetone, acetonitrile (Chromasolv grade), purchased from Sigma-
Aldrich, and milli-Q water 18 MX, obtained from Millipore’s
Simplicity system. The LC mass spectra were obtained using a 1200 L
triple quadrupole system (Varian, Palo Alto, CA) equipped with an
electron spray source (ESI) operating in both positive and negative ions.
MetAP
Docking studies were carried out on the crystal structure of MetAP2
PDB ID: 2ADU) (Fig. 6a and 6b) [37]. Protein was prepared for docking
(
by removing the ligand and water molecules. The coordinates for ligand
molecules were generated using the PRODRG server [38]. The input files
for AutoDock Vina [39] were generated after adding polar hydrogens
4

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
2
+
Fig. 3. a. Metal specificity of HpMetAP1a towards Met-AMC. The best activity was observed with Ni . b. The best activity of the HpMetAP1a activity was shown at
pH 8.0 towards Met-AMC. c. Substrate Specificity of HpMetAP1a towards different amino acid-AMCs in percentage activity. HpMetAP1a cleaves only methionine-
2
AMC in the presence of NiCl at pH 8.0.
2
2
.12.1. Chemistry: Experimental protocols
.12.1.1. General procedure for the synthesis of 7d4 and 7g7. Solvents
flash column chromatography using 20% ethyl acetate in pet ether as a
mobile phase. Collected solvent concentrated to get 8 g (yield: 92%) of
1H-pyrrolo[2,3-b]pyridine-4-carbonitrile as a pale orange solid. TLC: R
f
:
and all the reagents were purchased from Sigma-Aldrich (India), Alfa
0.4 (40% EtOAc in pet ether). LCMS (ESI positive) m/z = 143.9. LC
1
1
Aesar (India), and TCI (India). Nuclear magnetic resonance spectra ( H
6
Purity: 97.3%. HNMR (400 MHz, DMSO‑d ): δ 12.36 (brs, 1H), 8.4 (d, J
NMR: 400 MHz and 13C NMR: 100 MHz) were recorded in CDCl
and
using an Avance III 400 MHz spectrometer (Bruker). Chemical
shifts are reported in parts per million (ppm) and the coupling constants
J) are expressed in Hertz (Hz). Splitting patterns are designated as
= 9.2 Hz, 1H), 7.83 (d, J = 1.2 Hz, 1H), 7.55 (d, J = 8.4 Hz, 1H), 6.65 (d,
3
DMSO‑d
6
J = 1.6 Hz, 1H).
(
2.12.2.2. 1H-pyrrolo[2,3-b]pyridine-4-carboxylic acid (3). To a solution
of 1H-pyrrolo[2,3-b]pyridine-4-carbonitrile (2) (8.4 g, 58.74 mmol, 1
eq) in 70 mL of ethanol was added 40 mL of an aqueous 20% KOH so-
follows: s, singlet; d, doublet; t, triplet; q, quadruplet; m, multiplet; brs,
broad singlet; dd, double of doublets. Analytical thin-layer chromatog-
raphy (TLC) was carried out on silica gel F-254 plates (Merck). The
solvents used in MS measures were acetone, acetonitrile (Chromasolv
grade), purchased from Sigma-Aldrich, and Milli-Q water 18 MX, ob-
tained from Millipore’s Simplicity system. The LC mass spectra were
obtained using a 1200 L triple quadrupole system (Varian, Palo Alto, CA)
equipped with an electron spray source (ESI) operating in both positive
and negative ions.
◦
lution dropwise at RT. The reaction mixture was heated to 100 C and
stirred for 16 h. TLC showed completion of the reaction. The reaction
mixture was cooled to RT and concentrated to remove ethanol under
vacuum. The aqueous layer was diluted with 50 mL of water and washed
with ethyl acetate (2 × 50 mL). The aqueous layer was acidified to pH 6
by using 1 M HCl. The precipitated solid was collected, washed with
water, pet ether, and dried under a high vacuum to get 7.85 g (yield:
8
2.6%) of 1H-pyrrolo[2,3-b]pyridine-4-carboxylic acid as a pale yellow
solid. TLC: R
: 0.1(EtOAc). LCMS (ESI positive) m/z = 162.8. LC Purity:
9.1%. HNMR (400 MHz, DMSO‑d ): δ 13.3 (brs, 1H), 11.9 (brs, 1H),
2
.12.2. Preparation of target compounds
f
1
9
6
2
.12.2.1. Synthesis of 1H-pyrrolo[2,3-b]pyridine-4-carbonitrile (2). To a
8.34 (d, J = 4.8 Hz, 1H), 7.65 (t, J = 2.8 Hz, 1H), 7.56 (d, J = 4.8 Hz, 1H)
solution of 4-bromo-1H-pyrrolo[2,3-b]pyridine (12 g, 60.91 mmol, 1 eq)
in 80 mL of DMF in a sealed tube was added zinc cyanide (14.3 g, 121.83
mmol, 2 eq) at RT and the reaction mixture was purged with Argon for
and 6.87 (m, 1H).
2.12.2.3. Methyl 1H-pyrrolo[2,3-b]pyridine-4-carboxylate (4). To an ice-
cold solution of 1H-pyrrolo[2,3-b]pyridine-4-carboxylic acid (7.8 g,
48.14 mmol, 1 eq) in 150 mL of methanol was added thionyl chloride
(14 mL, 192.6 mmol, 4 eq) dropwise. The reaction mixture was heated to
reflux and stirred for 18 h. crude LCMS showed completion of the re-
action. The reaction mixture cooled to RT and concentrated. The crude
compound was dissolved in water (250 mL), pH adjusted to 12 by using
1 N NaOH solution. The precipitated solid was collected, washed with
water (2 × 30 mL), pet ether (2 × 50 mL), and dried under a high
1
5 min. Pd(Tetrakis) (3.52 g, 3.045 mmol, 0.05 eq) was added, and
purging continued for 15 min more. The seal tube was closed and the
◦
reaction mixture was heated to 100 C and stirred for 18 h. TLC showed
completion of the reaction. The reaction mixture cooled to RT, quenched
in 500 mL of ice-cold water with stirring, and extracted with ethyl ac-
etate (3 × 100 mL). The combined organic layer was washed with water
(
2 4
50 mL), brine (50 mL), dried over Na SO , filtered, and concentrated to
get the crude compound. The crude compound was purified by using
5

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
Table 2
Ki values of hydroxamic acid derivatives of azaindoles against HsMetAP1b and HpMetAP1a.
Compound code
Structure
HpMetAP1a-Ki (µM)
0.11 ± 0.010
HsMetAP1b-Ki (µM)
57.33 ± 3.30
Inhibition favoring the HpMetAP1a
7
d4
521.1
7
g7
0.14 ± 0.050
57.58 ± 14.17
411.2
7
c
0.17 ± 0.096
34.42 ± 3.6
202.4
7
g
0.84 ± 0.35
135.9 ± 16.3
161.7
7
e5
0.62 ± 0.096
36.61 ± 0.44
59.0
7
i9
0.23 ± 0.079
10.79 ± 0.78
46.9
7
f
0.82 ± 0.154
36.59 ± 4.9
43.5
7
c3
0.59 ± 0.106
13.76 ± 1.514
23.2
7
d
0.53 ± 0.073
11.05 ± 1.92
20.84
(
continued on next page)
6

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
Table 2 (continued)
Compound code
Structure
HpMetAP1a-Ki (µM)
HsMetAP1b-Ki (µM)
Inhibition favoring the HpMetAP1a
7
f6
0.28 ± 0.007
5.427 ± 0.88
19.3
7
a1
0.44 ± 0.054
8.18 ± 2.71
18.5
7
j10
0.78 ± 0.072
10.07 ± 1.39
12.9
7
a
5.21 ± 0.72
60.37 ± 12.82
11.5
7
h
1.49 ± 0.370
8.656 ± 0.81
5.8
7
b2
5.37 ± 0.696
22.41 ± 1.24
4.1
7
e
0.07 ± 0.024
ND
–
–
7
b
1.17 ± 0.23
ND
(
continued on next page)
7

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
Table 2 (continued)
Compound code
Structure
HpMetAP1a-Ki (µM)
HsMetAP1b-Ki (µM)
Inhibition favoring the HpMetAP1a
ND – Not determined.
purified using flash column chromatography using 10% ethyl acetate in
pet ether as mobile phase. Collected solvent concentrated to get 6.5 g
Table 3
Data collection and refinement statistics of HsMetAP1b in complex with 7d and
(yield: 89.1%) as pale orange solid. TLC: R
ether). MS (ESI + ) C
: 0.4 (40% EtOAc in pet
f
7
e5.
+
8
H
5
N
3
for m/z 143.9) [M + H] ; LC purity 94.15%,
1
HsMetAP1b-7d
HsMetAP1b-7e5
(ret. time, 4.16 min); H NMR (400 MHz, DMSO‑d
.58 (s, 1H), 8.50–8.49 (d, J = 0.8 Hz, 1H), 7.70–7.69 (m, 1H),
6.609–6.601 (s, 1H).
6
) δ 12.30 (brs, 1H),
8
Cell parameters
Space group
P2
1
P2
1
Unit cell (Å) (a, b, c)
47.55, 77.22, 48.65
91.549
47.57, 77.25, 48.62
91.54
◦
◦
α
= γ = 90 , β ( )
2
.12.3.2. The general procedure of benzylation (5d4 and 5g7). To a so-
lution of Methyl 1H-pyrrolo[2,3-b], pyridine-5-carboxylate 4a in
anhydrous DMF was added K CO at room temperature and the reaction
a
Data collection
b
Resolution range (Å)
47.53–1.35
40.5–1.29
(1.337–1.291)
580,633
87,090 (8351)
98.96 (95.44)
17.9
(
1.401–1.353)
2
3
Total reflections
519,285
75,652 (7550)
98.90 (99.16)
20.1
mixture was stirred for 30 min. To this was added dropwise benzyl
bromide/ chloride and the reaction mixture was stirred for 16 h and
monitored by TLC. The reaction mixture was quenched in ice-cold water
and extracted with ethyl acetate (3 × 30 mL). The combined organic
b
Unique reflections
b
Completeness (%)
Mean I/sigma (I)
Multiplicity
6.8
6.6
R-merge (%)
4.2
4.7
2 4
layer was washed with water, brine, dried over anhydrous Na SO ,
R-meas (%)
4.6
5.1
filtered, and concentrated to get the crude compound. The crude com-
pound was purified by flash column chromatography (silica gel 230–400
mesh) using 0–20% ethyl acetate in pet ether as a mobile phase.
Refinement statistics
R-work (%)
18.99
20.20
2646
2464
304
18.40
21.20
2631
2436
304
R-free (%)
No. of non-H atoms
Macromolecule
Protein residues
2
.12.3.3. Methyl
5d4). Yield: 600 mg (95.6%) as pale yellow solid.; H NMR (400 MHz,
CDCl
1-benzyl-1H-pyrrolo[2,3-b]pyridine-5-carboxylate
1
(
Wilson B-factor
15.40
28.00
98.90
1.00
15.70
25.00
98.34
1.32
2
3
) δ 9.02–9.01 (s, J = 4 Hz, 1H), 8.59 (s, 1H), 7.31–7.20 (m, 6H),
Average B-factor (Å )
Ramachandran favored (%)
Ramachandran allowed (%)
Ramachandran outliers (%)
6.58–6.57 (d, J = 4 Hz, 2H), 5.52 (s, 2H), 3.96 (s, 3H). MS (ESI + )
+
C
16
H
14
N
2
O
2
for m/z 267) [M + H] ; LC purity 95.11%, (ret. time, 3.51
0.37
0.33
min).
c
R.M.S.D (bonds) (Å)
0.018
2.26
0.025
2.29
c
◦
R.M.S.D (angles) ( )
Protein Data Bank code
6LZC
6LZB
2.12.3.4. Methyl
dine-5-carboxylate (5g7). Yield: 411 mg (43.3%) as off white solid.
NMR (400 MHz, CDCl
) δ 9.01–9.00 (s, J = 4 Hz, 1H), 8.61–8.60 (s, 1H),
1-(4-(trifluoromethyl)benzyl)-1H-pyrrolo[2,3-b]pyri-
1
H
a
X-ray source: XRD2 beamline, Elettra Synchrotron Trieste, Italy.
Statistics for the highest-resolution shell are shown in parentheses.
RMSD, root mean square deviation.
b
c
3
7.56 (d, J = 4 Hz, 1H), 7.54 (s, 1H), 7.30–7.28 (d, J = 8 Hz, 2H),
7
3
.24–7.23 (d, J = 4 Hz, 1H), 6.62–6.62 (d, J = 4 Hz, 1H), 5.58 (s, 2H),
+
.96 (s, 3H). MS (ESI + ) C17
H
13
F
3
2
N O
2
for m/z 335.1) [M + H] ; LC
vacuum to get 7.7 g (yield: 90.8%) of methyl 1H-pyrrolo[2,3-b]pyridine-
-carboxylate as pale yellow solid. TLC: R : 0.4 (EtOAc). LCMS (ESI
): δ
purity 95.99%, (ret. time, 3.75 min).
4
f
1
positive) m/z = 177. LC Purity: 99.8%. HNMR (400 MHz, DMSO‑d
6
2
.12.3.5. General procedure of hydrolysis (6d4 and 6g7). To an ice-cold
1
0.1 (brs, 1H), 8.44 (d, J = 4.8 Hz, 1H), 7.74 (d, J = 4.8 Hz, 1H), 7.52 (t,
solution of SM in THF: MeOH: water was added LiOH⋅H2O portion-wise.
The reaction mixture was stirred for 16 h at RT. TLC showed completion
of the reaction. The reaction mixture was concentrated to remove vol-
atiles and diluted with water. The aqueous layer was washed with ethyl
acetate and the aqueous layer was acidified to pH-2 using 1 N aqueous
HCl. The precipitated solid was collected by filtration, washed with
water, and dried under a high vacuum. The solid was triturated with
ethyl acetate, decanted, and dried under a high vacuum to get the
desired compound as a white solid.
1
H), 7.07 (t, 1H) and 4.03 (s, 3H).
2
.12.3. Preparation of target compounds
2
.12.3.1. 1H-pyrrolo[2,3-b]pyridine-5-carbonitrile (2a). To a solution of
-bromo-1H-pyrrolo[2,3-b]pyridine 1a (10 g, 51.02 mmol) in DMF in a
5
sealed tube was added zinc cyanide (11.9 g, 102.04 mmol) at room
temperature and the reaction mixture was purged with Argon for 15
min. Pd(Tetrakis) (2.9 g, 2.55 mmol) was added, and purging continued
for 15 min more. The seal tube was closed and the reaction mixture was
◦
2.12.3.6. 1-benzyl-1H-pyrrolo[2,3-b]pyridine-5-carboxylic acid (6d4).
heated to 100 C and stirred for 18 h and monitored by TLC. The reaction
1
Yield: 360 mg (86.3%) as white solid. H NMR (400 MHz, CDCl
3
) δ 9.11
mixture cooled to room temperature, quenched in ice-cold water with
stirring, and extracted with ethyl acetate. The combined organic layer
(
s, 1H), 8.67 (s, 1H), 7.33–7.32 (d, J = 4 Hz, 1H), 7.28–7.22 (m, 5H),
6
.62–6.61 (d, J = 4 Hz, 1H), 5.54 (s, 2H). MS (ESI + ) C15
12 2 2
H N O for m/
was washed with water, brine, dried over anhydrous Na
2 4
SO , filtered,
+
z 253.1) [M + H] ; LC purity 99.92%, (ret. time, 2.68 min).
and concentrated to get the crude compound. The crude compound was
8

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
Fig. 4. a) HsMetAP1b S1 pocket residues are shown in sticks (green) in complex with compound 7d is shown in pink b) Hydroxamic acid moiety coordinates with the
metal cofactors. c) HsMetAP1b S1 pocket residues are shown in sticks (green) in complex with compound 7e5 shown in yellow. d) Hydroxamic acid moiety co-
ordinates with the metal cofactors. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
2
.12.3.7. 1-(4-(trifluoromethyl)benzyl)-1H-pyrrolo[2,3-b]pyridine-5-car-
chromatography (silica gel 230–400 mesh) using 40–50% ethyl acetate
in pet ether as a mobile phase. Collected solvent concentrated to get the
desired compound as white solid.
boxylic acid (6g7). Yield: 380 mg (79.3%) as white solid. MS (ESI + )
+
C
16
H
11
F
N
3 2
O
2
for m/z 321.1) [M + H] ; LC purity 99.91%, (ret. time,
5
.32 min).
2
.12.3.9. 1-benzyl-N-hydroxy-1H-pyrrolo[2,3-b]pyridine-5-carboxamide
1
2
.12.3.8. The general procedure of amide coupling (7d4 and 7g7). To a
3
(7d4). Yield: 63 mg (17%) as white solid. H NMR (300 MHz, CDCl ) δ
solution of 6a1 in anhydrous THF was added triethylamine, hydroxyl-
11.21 (s, 1H), 9.03 (s, 1H), 8.64 (s, 1H), 8.35–8.34 (d, J = 1.8 Hz, 1H),
◦
amine hydrochloride at 0 C and the reaction mixture was stirred for 30
7.72–7.71 (d, J = 3.3 Hz, 1H), 7.32–7.21 (m, 5H), 6.62–6.61 (d, J = 3.6
min. To this was added HATU and the reaction mixture was stirred for
Hz, 1H), 5.50 (s, 2H). 13C NMR (75 MHz, DMSO-D
): δ 148.0, 141.8,
6
1
6 h and monitored by TLC. The reaction mixture filtered through a
138.1, 130.71, 128.5, 127.7, 127.4, 127.2, 121.0, 119.1, 100.6, and
+
celite pad and filtrate evaporated under reduced pressure to get the
crude compound. The crude compound was purified by flash column
47.3. MS (ESI + ) C15
H
13
3
N O
2
for m/z 268.1) [M + H] ; LC purity
99.08%, (ret. time, 3.93 min).
9

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
Fig. 5. a) Structural alignment of HsMetAP1b (green) in complex with compound 7d (pink) and HpMetAP1a homology model (brown) b) Structural alignment of
HsMetAP1b (green) in complex with compound 7e5 (yellow) and HpMetAP1a homology model (brown). (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)
Fig. 6. a) Molecular modeling of 7d4 (yellow) in the active site of HsMetAP2 (green). b) Molecular modeling of 7c (pink) in the active site of HsMetAP2 (green).
Metal binding residues are shown in gray. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
2
.12.3.10. N-hydroxy-1-(4-(trifluoromethyl)benzyl)-1H-pyrrolo[2,3-b]
pyridine-5-carboxamide (7g7). Yield: 80 mg (25.4%) as white solid. 1H
NMR (300 MHz, DMSO-D
) δ 11.22 (s, 1H), 9.04 (s, 1H), 8.64–8.63 (d, J
human counterpart. We believe that this understanding will allow sci-
entists to study only those microbial enzymes that have appreciable
differences from the human enzyme. In this study, we have analyzed 13
sets of 25 hundred sequence alignments that are based on homology
driven secondary structure prediction (HSSP) [51]. For each protein of
known 3D structure from the Protein Data Bank (PDB) [30], the data-
base has a multiple sequence alignment of all available homologues and
a sequence profile characteristic of the family. Each HSSP file contains a
maximum of 2,500 homolog proteins with a lower limit of 30% sequence
identity. However, there could be redundancy of protein sequences
between each alignment, which we did not attempt to eliminate (Table 1
and Fig. 1).
6
=
1.5 Hz, 1H), 8.37–8.36 (d, J = 1.5 Hz, 1H), 7.77–7.76 (m, 1H),
7
5
.69–7.66 (m, 2H), 7.40–7.38 (m, 2H), 6.66–6.65 (d, J = 3.6 Hz, 1H),
+
.62 (s, 2H). MS (ESI + ) C16
H
12
F
N
3 3
O
2
for m/z 336.1) [M + H] ; LC
purity 98.88%, (ret. time, 4.65 min).
. Results and discussion
Since Type 1 MetAPs are recognized as drug targets against microbial
3
infections, they have been studied extensively [12,16,41–43]. Several
classes of inhibitors have been reported against these enzymes that
include pyridyl pyrimidines, phosphonic acids, bengamides, hydroxa-
mic acids, etc. [12,13,44–50]. One of the limitations of most of the re-
ported inhibitors is the lack of selectivity. All earlier studies have been
dedicated to understanding a single enzyme at a time based on the
disease of interest and the inhibitors have been designed specifically for
that enzyme. To the best of our knowledge, there is no report till date on
understanding the variability of the active site Type 1 MetAP from their
Five residues (one histidine, two glutamates, and two aspartates) co-
ordinate with the active site metal ions, while two other histidine resi-
dues are placed at either side of the entrance of the active site. All these
seven residues are conserved among all MetAPs reported till date
(Fig. 1). For the convenience of description, we divide the S1 pocket
where the substrate methionine side chain binds into two halves, the left
lobe and the right lobe that contain three residues each. Three residues
in the right lobe of the S1 pocket of human MetAP1b are F198, C203,
1
0

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
Fig. 7. a) Structural alignment of HsMetAP1b (green) in complex with compound 7d (pink) and HsMetAP2 (PDB ID:2ADU) (brown) b) Structural alignment of
HsMetAP1b (green) in complex with compound 7e5 (yellow) and HsMetAP2 (PDB ID:2ADU) (brown). Metal binding residues are shown in gray. (For interpretation
of the references to colour in this figure legend, the reader is referred to the web version of this article.)
and W353. Based on the 13 alignment files, tryptophan is conserved
00% among all Type 1 MetAPs while the F198 position is either
proline and a phenylalanine were switched between the proteins at site 1
and site 2. The bulky aromatic Y195 in the human enzyme is replaced by
a relatively smaller amino acid (L61) in the pylori enzyme making the S1
pocket wider in the later. Based on these fine differences, we believe that
the inhibitors designed will show differential inhibition against the
H. pylori and human MetAPs.
1
phenylalanine (73.5%) or a tyrosine (26.5%). Similarly, the C203 po-
sition is occupied by cysteine at 97%, asparagine at 2.5%, while the
other 0.5% is occupied by leucine, threonine, and valine. Compared to
the right lobe of the S1 pocket, the left lobe is more variable. In the
human MetAP1b, these residues are P192, Y195 and F309 labeled as Site
1
, Site 2, and Site 3 in Fig. 1. Site 1 is most variable in the left lobe of the
S1 pocket represented by P(40.7%), F(19.2%), Y(16.3%), N(5%), T
4.6%), L(3.5%), Q(3.4%), S(2.7%), C(2.2%), E(1.5%) and few others in
less than 1% each. Site 2 ranks second in variation substituted by Y
3
.2. Design and synthesis of hydroxamic acid-based azaindoles
(
Hydroxamic acid is a well-known metal chelating moiety used in the
design of inhibitors against several metalloproteases [22,45,46,53,54].
Since both the aromatic residues as in human MetAP1b at Site 2 and Site
(
(
(
72.4%), P (5.5%), T (5.1%), C (3.6%), G (3%), N (2%), F (1.9%) and V
1%). Site 3 is substituted by F (59.3%), M (17%), L (12.8%), Y (5.3%), I
2.3%) and V (1.3%) (Table 1 and Fig. 1).
3
are truncated to smaller amino acids (leucine and proline, respec-
tively) in HpMetAP1a in the left lobe, an extra space is created. Using
this understanding, we designed azaindole derivatives with two substi-
tution points. We analyzed that the aromatic groups at Site 2 and Site 3
in the human enzyme would resist the binding of these molecules. To
optimize the binding, we varied the hydroxamic acid on the 4th (7a-7h)
and 5th (7a1-7j10) positions of the azaindole ring. In addition, an ar-
omatic ring with different substitutes has been added in all compounds
at position 1. Substitutions on the aromatic group in compounds 7a-7g
are repeated in compounds 7a1-7g7 and in the same order. Together, we
synthesized 17 molecules.
Recently, we have developed species-specific inhibitors based on the
variation of Site 3 [12]. F309 in the HsMetAP1b is represented by a
methionine in S. pneumoniae MetAP1a’ (SpMetAP1a’). This single amino
acid difference was sufficient to discover selective inhibitors SpMe-
tAP1a’ [12]. Similarly, ovalicin inhibits Type 2 MetAPs by a million-fold
compared to the Type 1 MetAPs from E. coli and human which have
phenylalanine at Site 2. Nanomolar inhibition is noticed in Type 1 en-
zymes like SpMetAP1a’, where phenylalanine is replaced by methionine
[
27].
Column 2 describes the variation of residues at Site-1, Site2, and Site-
of the left lobe of the MetAP active site and column 3 describes the
3
3
.3. Cloning, expression, purification, and characterization of
conservation of residues at the right lobe of the MetAP active site.
HpMetAP1a and HsMetAP1b
3
.1. Sequence and structural comparison of HpMetAP1a with human
Human MetAP1b used in this study was prepared as reported earlier
[12,13,27,55]. HpMetAP1a was cloned into pET15b vector. Both pro-
teins were expressed and purified to homogeneity (Fig. S2). Both the
proteins were inactive after purification. Enzymes were activated after
the addition of transition metal ions. MetAPs are metalloproteases with
variable specificity to first-row transition metal ions. Most of the MetAPs
MetAP1b:
The protein sequence of HpMetAP1a is 32.7% identical to human
MetAP1b (Fig. S3). It is important to note that the human MetAP1b used
in this study is devoid of N-terminal zinc fingers as reported earlier [12].
prefer Co2 better than others, including the human Type 1 enzyme.
+
3
D model of the HpMetAP1a was developed and validated by RAMPAGE
[
36]. Structural alignment between the human (PDB ID: 4U6J) and
Surprisingly, HpMetAP1a displayed activity in the presence of NiCl and
2
2
+
pylori MetAPs is performed using Clustal Omega [52]. As expected, the
left side of the S1 pocket is more varied in the HpMetAP1a in comparison
with the human enzyme. P192, Y195, and F309 in Site 1, Site 2, and Site
negligible activity with other metal ions (Fig. 3a). Using Ni as a co-
factor, the pylori enzyme had optimum activity at pH 8.0 (Fig. 3b).
Among all the tested AMC attached amino acids, only Met-AMC was
hydrolyzed (Fig. 3c). All further enzyme kinetic and inhibition studies
3
, respectively in human MetAP are replaced by F58, L61, and P175 in
for HpMetAP1a were carried out with Ni2 as co-factor at pH 8.0 and
+
pylori enzyme, while all three residues in the right lobe are absolutely
conserved between the two (Fig. 2). We note that the position of a
Met-AMC as substrate. K and kcat for HpMetAP1a were determined as
M
1
1

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
M and 298.50 sec 1, respectively (Table S2).
ꢀ
3.6. Comparison of binding of inhibitors with H. Pylori MetAP1a
9
4.17 ± 6.8
μ
3
.4. Selective inhibition of Helicobacter pylori MetAP1a compared to its
To understand the structural basis for the selective inhibition of
H. pylori MetAP1a, we have developed the 3D model of the full-length
protein and compared it with the HsMetAP1b-7d and HsMetAP1b-7e5
structures. H. pylori MetAP1a and the human MetAP1b share 32.7%
identity with all-metal binding and gatekeeper histidines conserved. Of
the six residues that make up the S1 pocket, all three residues in the right
lobe are conserved (F198, C203, and W353 in humans and F64, C69, and
W223 in the H. pylori protein). However as seen in other proteins
described in Table 1, all residues in the left lobe are different in
HpMetAP1a (F58, L61, and P175) compared to the human counterpart
(P192, Y195, and F309). It is important to note that in 2,500 sequences
aligned with the human enzyme in the HSSP database as discussed
above, there was no sequence with leucine at Site 2 and proline in Site 3.
In the HsMetAP1b-7d and HsMetAP1b-7e5, methylene carbon of the
benzyl group in inhibitors makes a very short contact with the side chain
of the Y195 (3.4 Å). In the H. pylori MetAP1a, leucine being small and
branched could rotate away with ease without affecting the neighboring
residues thereby accommodating the inhibitor with better ease
compared to that in the human enzyme thus providing the structural
basis for selectivity (Fig. 5a and 5b). In this process, tryptophan (W223)
may not have to flip away from the active site which could be costly in
terms of energy.
human counterpart
The library of 17 compounds synthesized in this study were screened
for inhibition and determined their inhibition constants (Ki) against
pylori and human Type 1 MetAPs (Table 2). As we hypothesized from
the design, all 17 molecules inhibited the pylori enzyme with high
selectivity compared to the human enzyme. Of the eight 4-azaindole
substituted compounds, five showed inhibition better than 1 µM
against HpMetAP1a, while eight out of nine 5-azaindole substituted
compounds showed lower activity than 1 µM. The best compound in the
4
-substituted series (7c) showed 200-fold selective inhibition of pylori
enzyme compared to human MetAP1b. Among the 5-substituted series,
d4 showed 520 folds and 7g7 showed 400-fold selectivity against
7
HpMetAP1a. Among the substituents of the aromatic rings at the 1st
position on the azaindole, compounds with trifluoromethyl groups (7g
and 7g7) are better inhibitors with more than 100-fold selectivity.
Surprisingly, the best molecule with more than 500-fold selectivity has
no substitution on the aromatic ring (7d4). Compound 7i9 with two
fluorine atoms at position 2 and 4 though inhibit the HpMetAP1a at a
low concentration similar to 7j10, it also inhibits the human enzyme at
low concentration (Ki = 10.07 µM). Compound 7d inhibits the HpMe-
tAP1a at 0.53 µM and human MetAP1b at 11.05 µM. On the other hand,
homologous compound, 7d4 inhibits at 0.11 µM and 57.33 µM, pylori,
and human MetAPs respectively. Together, the structure activity rela-
tionship analysis suggests that azaindoles with hydroxamic acid at the
3.7. In-silico analysis for the binding of azaindole hydroxamic acid
derivatives on human Type 2 MetAP
4
th position and aromatic rings with halogen substitution at position 1
We have carried our docking studies to understand whether azain-
dole derivatives non-specifically target human Type 2 MetAP (MetAP2)
are better and selective inhibitors against HpMetAP1a (Table 2).
(
Fig. 6 and Fig. 7). The major interaction of metal chelating inhibitors is
3
.5. Crystal structure of inhibitors in complex with human enzyme
their ability to interact strongly with metal ions in the active site. The
active site of MetAP2 is having two histidine residues (H339, H231)
which might be helping the enzyme in specificity. These two histidine
residues are restricting the metal ion access for the bulky azaindole
molecules used in this study.
Several attempts to crystallize the HpMetAP1a were not successful.
To get a clue on the mode of binding, we have determined the crystal
structure of two inhibitors in complex with human enzyme using the
reported conditions (Table 3) [12]. Several inhibitors were soaked but
crystals in complex with compounds 7d (4-azaindole derivative) and
4. Conclusions
7
e5 (5-azaindole derivative) yielded the diffraction quality crystals. The
overall structure of inhibitor complexes is similar to the human holo-
structure (PDB ID: 2B3H) with less than 0.15 Å RMSD (all main chain
atoms). Except for W353 all other amino acids in the active site are
We have identified MetAP1a from Helicobacter pylori that have var-
iations within the S1 pocket, identified as left lobe compared to human
MetAP1b. Using these differences in the active site we have designed
and synthesized a library of 17 compounds that contain hydroxamic acid
moiety as a metal binding scaffold. All 17 compounds inhibited the
pylori enzyme with selectivity as high as 500 compared to the human
enzyme. The crystal structure of the human enzyme with two com-
pounds provided the molecular basis for selective inhibition.
◦
unaltered. Tryptophan side chain flips away by 180 to make space for
the aromatic ring of the inhibitor. In compound 7d, the hydroxamic acid
at 4th position of the azaindole moiety binds to the bimetalo center
placing the hydroxyl group between the two metal ions (Fig. 4a and 4b).
In the holo structure, a water molecule acts as a bridge between the two
metal ions that are believed to act as a nucleophile in hydrolyzing the
peptide substrates. The ketone in the hydroxamic acid replaces another
conserved water molecule and makes interactions with one of the metal
ions in the metal center (2.3 Å) while making two other hydrogen bonds
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
(
T231 (3.4 Å) and a water molecule (2.9 Å)). The azaindole moiety
extends into the hydrophobic S1 pocket formed by Y195, F198, H212,
F309, and W253 (Fig. 4a and 4b). The benzyl group at N-1 pushes the
W353 away by flipping it compared to the native position. In the new
Acknowledgments
position, W353 makes edge- interaction (4.2 Å) with the inhibitor on
π
one side and partially solvent exposed on the other.
AA (EMR/2015/000461 and CRG/2019/006013) and RG (ECR/
2016/000288) thank Science and Engineering Research Board (SERB),
India for funding. RG acknowledges the lab facility at GITAM University.
SCB and BK acknowledge fellowship from DBT (Department of
Biotechnology), India. We acknowledge funding from DST (Department
of Science and Technology), India for the transport of our crystals and
data collection at Elettra Synchrotron. We thank Dr. Babu Manjashetty,
Beamline Scientist, Elettra Synchrotron for collecting data on our behalf.
The communication number issued by CSIR-IICT for this article is IICT/
Pubs./2021/021.
The overall conformation of the inhibitor in the HsMetAP1b-7e5 is
similar to that of 7d in HsMetAP1b-7d except for the orientation of the
carbonyl group in the hydroxamic moiety at the metal center (Fig. 4c
and Fig. 4d). In this structure, the carbonyl group flips out and makes
strong contact with H310 (2.6 Å). The rest of the molecule adopts a
similar conformation including the flipping of the W353 side chain.
1
2

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
Author contributions
[16] Travis R. Helgren, Phumvadee Wangtrakuldee, Bart L. Staker, Timothy J. Hagen,
Advances in Bacterial Methionine Aminopeptidase Inhibition, Curr. Top. Med.
Chem. 16 (4) (2015) 397–414.
AA and RG conceptualized and supervised the study and reviewed
the manuscript. SCB performed all the experiments, analyzed the data,
and drafted the manuscript. KV, AK, and OVS had synthesized the
compounds and drafted the chemistry part. RVS and BK helped during
crystallization trials of proteins.
[
17] V. Pillalamarri, T. Arya, N. Haque, S.C. Bala, A.K. Marapaka, A. Addlagatta,
Discovery of natural product ovalicin sensitive type 1 methionine aminopeptidases:
molecular and structural basis, Biochem. J. 476 (6) (2019) 991–1003.
18] Sandeep Chowdary Bala, Neshatul Haque, Vijaykumar Pillalamarri,
Ravikumar Reddi, Rajnandani Kashyap, Anil Kumar Marapaka,
[
Anthony Addlagatta, Discovery of a new class of type 1 methionine
aminopeptidases that have relaxed substrate specificity, Int. J. Biol. Macromol. 129
(
2019) 523–529.
Enzyme
[
19] R. Reddi, T. Arya, C. Kishor, R. Gumpena, R.J. Ganji, S. Bhukya, A. Addlagatta,
Selective targeting of the conserved active site cysteine of Mycobacterium
tuberculosis methionine aminopeptidase with electrophilic reagents, FEBS J. 281
Methionine aminopeptidase (EC: 3.4.11.18)
Databases
(
18) (2014) 4240–4248.
[
[
20] C. Kishor, R. Gumpena, R. Reddi, A. Addlagatta, Structural studies of Enterococcus
faecalis methionine aminopeptidase and design of microbe specific 2,2[prime or
minute]-bipyridine based inhibitors, MedChemComm 3 (11) (2012) 1406–1412.
21] Saleem Yousuf Bhat, Sonal Bhandari, Pavitra Suresh Thacker,
Structural data are available in PDB under accession numbers 6LZC,
Mohammed Arifuddin, Insaf Ahmed Qureshi, Development of quinoline-based
hybrid as inhibitor of methionine aminopeptidase 1 from Leishmania donovani,
Chem. Biol. Drug Des. 97 (2) (2021) 315–324.
6
LZB.
[
[
[
22] Natalie B. Vinh, Nyssa Drinkwater, Tess R. Malcolm, Michael Kassiou,
Leonardo Lucantoni, Peter M. Grin, Georgina S. Butler, Sandra Duffy, Christopher
M. Overall, Vicky M. Avery, Peter J. Scammells, Sheena McGowan, Hydroxamic
acid inhibitors provide cross-species inhibition of Plasmodium M1 and M17
aminopeptidases, J. Med. Chem. 62 (2) (2019) 622–640.
Funding sources and disclosure of conflicts of interest
Funding agencies have no role in study design, in the collection,
analysis, and interpretation of data; in the writing of the report; and in
the decision to submit the article for publication.
23] Travis R. Helgren, Elif S. Seven, Congling Chen, Thomas E. Edwards, Bart L. Staker,
Jan Abendroth, Peter J. Myler, James R. Horn, Timothy J. Hagen, The
identification of inhibitory compounds of Rickettsia prowazekii methionine
aminopeptidase for antibacterial applications, Bioorg. Med. Chem. Lett. 28 (8)
Appendix A. Supplementary data
(
2018) 1376–1380.
24] Komagal Kannan Sivaraman, Alessandro Paiardini, Marcin Sie n´ czyk,
Chiara Ruggeri, Christine A. Oellig, John P. Dalton, Peter J. Scammells,
Marcin Drag, Sheena McGowan, Synthesis and structure-activity relationships of
phosphonic arginine mimetics as inhibitors of the M1 and M17 aminopeptidases
from Plasmodium falciparum, J. Med. Chem. 56 (12) (2013) 5213–5217.
25] B.A. Salih, Helicobacter pylori infection in developing countries: the burden for
how long? Saudi J. Gastroenterol. 15 (3) (2009) 201–207.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bioorg.2021.105185.
[
[
[
References
26] M. Safavi, R. Sabourian, A. Foroumadi, Treatment of Helicobacter pylori infection:
Current and future insights, World J. Clin. Cases 4 (1) (2016) 5–19.
[
[
1] M. Kozak, Initiation of translation in prokaryotes and eukaryotes, Gene 234 (2)
(
1999) 187–208.
27] V. Pillalamarri, T. Arya, Discovery of natural product ovalicin sensitive type 1
methionine aminopeptidases: molecular and structural basis, Biochem. J. 476 (6)
2] C. Giglione, T. Pierre, M. Fau-Meinnel, T. Meinnel, Peptide deformylase as a target
for new generation, broad spectrum antimicrobial agents, Mol. Microbiol. 36 (6)
(
2019) 991–1003.
(
2000) 1197–1205.
[
[
[
28] R.A. Copeland, Enzymes, A Practical Introduction to Structure, Mechanism, and
Data Analysis, Wiley, 2004.
[
[
[
3] R.A. Bradshaw, W.W. Brickey, K.W. Walker, N-terminal processing: the methionine
aminopeptidase and N alpha-acetyl transferase families, Trends Biochem. Sci. 23
29] Karine M. Sparta, Michael Krug, Udo Heinemann, Uwe Mueller, Manfred S. Weiss,
XDSAPP2.0, J. Appl. Cryst 49 (3) (2016) 1085–1092.
(
7) (1998) 263–267.
4] J.A. Vetro, B. Dummitt, Y.-H. Chang, Methionine Aminopeptidase, in: N.
M. Hooper, U. Lendeckel (Eds.), Aminopeptidases in Biology and Disease, Springer,
US, Boston, MA, 2004, pp. 17–44.
30] H.M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T.N. Bhat, H. Weissig, I.
N. Shindyalov, P.E. Bourne, The Protein Data Bank, Nucleic Acids Res. 28 (1)
(
2000) 235–242.
5] R. Bingel-Erlenmeyer, R. Kohler, G. Kramer, A. Sandikci, S. Antoli ´c , T. Maier,
C. Schaffitzel, B. Wiedmann, B. Bukau, N. Ban, A peptide deformylase-ribosome
complex reveals mechanism of nascent chain processing, Nature 452 (7183) (2008)
[
[
31] Garib N. Murshudov, Pavol Skub a´ k, Andrey A. Lebedev, Navraj S. Pannu, Roberto
A. Steiner, Robert A. Nicholls, Martyn D. Winn, Fei Long, Alexei A. Vagin,
REFMAC5 for the refinement of macromolecular crystal structures, Acta
Crystallogr. D Biol. Crystallogr. 67 (4) (2011) 355–367.
1
08–111.
[
[
[
[
6] J. Solbiati, A. Chapman-Smith, J.L. Miller, C.G. Miller Jr., J.E. Cronan Jr.,
Processing of the N termini of nascent polypeptide chains requires deformylation
prior to methionine removal, J. Mol. Biol. 290 (3) (1999) 607–614.
32] M.D. Winn, C.C. Ballard, K.D. Cowtan, E.J. Dodson, P. Emsley, P.R. Evans, R.
M. Keegan, E.B. Krissinel, A.G. Leslie, A. McCoy, S.J. McNicholas, G.N. Murshudov,
N.S. Pannu, E.A. Potterton, H.R. Powell, R.J. Read, A. Vagin, K.S. Wilson, Overview
of the CCP4 suite and current developments, Acta Crystallogr. D Biol. Crystallogr.
7] S.G. Bernier, N. Taghizadeh, C.D. Thompson, W.F. Westlin, G. Hannig, Methionine
aminopeptidases type I and type II are essential to control cell proliferation, J. Cell.
Biochem. 95 (6) (2005) 1191–1203.
6
7 (Pt 4) (2011) 235–242.
[
33] P. Emsley, B. Lohkamp, W.G. Scott, K. Cowtan, Features and development of Coot,
Acta Crystallogr. D Biol. Crystallogr. 66 (Pt 4) (2010) 486–501.
8] X. Li, Y.H. Chang, Amino-terminal protein processing in Saccharomyces cerevisiae
is an essential function that requires two distinct methionine aminopeptidases,
Proc. Natl. Acad. Sci. U. S. A. 92 (26) (1995) 12357–12361.
[
[
34] L.L.C. Schrodinger, The PyMOL Molecular Graphics System, Version 1 (2015) 8.
35] A. Waterhouse, M. Bertoni, S. Bienert, G. Studer, G. Tauriello, R. Gumienny, F.
T. Heer, T.A.P. de Beer, C. Rempfer, L. Bordoli, R. Lepore, T. Schwede, SWISS-
MODEL: homology modelling of protein structures and complexes, Nucleic Acids
Res. 46 (W1) (2018) W296–W303.
9] J.J. Alvarado, A. Nemkal, J.M. Sauder, M. Russell, D.E. Akiyoshi, W. Shi, S.C. Almo,
L.M. Weiss, Structure of a microsporidian methionine aminopeptidase type 2
complexed with fumagillin and TNP-470, Mol. Biochem. Parasitol. 168 (2) (2009)
1
58–167.
[
[
36] S.C. Lovell, I.W. Davis, W.B. Arendall 3rd, P.I. de Bakker, J.M. Word, M.G. Prisant,
J.S. Richardson, D.C. Richardson, Structure validation by Calpha geometry: phi, psi
and Cbeta deviation, Proteins 50 (3) (2003) 437–450.
[
10] C.M. Brdlik, C.M. Crews, A single amino acid residue defines the difference in
ovalicin sensitivity between type I and II methionine aminopeptidases, J. Biol.
Chem. 279 (10) (2004) 9475–9480.
37] L.S. Kallander, Q. Lu, W. Chen, T. Tomaszek, G. Yang, D. Tew, T.D. Meek, G.
A. Hofmann, C.K. Schulz-Pritchard, W.W. Smith, C.A. Janson, M.D. Ryan, G.-
F. Zhang, K.O. Johanson, R.B. Kirkpatrick, T.F. Ho, P.W. Fisher, M.R. Mattern, R.
K. Johnson, M.J. Hansbury, J.D. Winkler, K.W. Ward, D.F. Veber, S.K. Thompson,
[
[
11] C. Giglione, S. Fieulaine, T. Meinnel, N-terminal protein modifications: Bringing
back into play the ribosome, Biochimie 114 (1638–6183) (2015) 134–146.
12] T. Arya, R. Reddi, C. Kishor, R.J. Ganji, S. Bhukya, R. Gumpena, S. McGowan,
M. Drag, A. Addlagatta, Identification of the molecular basis of inhibitor selectivity
between the human and streptococcal type I methionine aminopeptidases, J. Med.
Chem. 58 (5) (2015) 2350–2357.
4
-Aryl-1,2,3-triazole: A Novel Template for a Reversible Methionine
Aminopeptidase 2 Inhibitor, Optimized To Inhibit Angiogenesis in Vivo, J. Med.
Chem. 48 (18) (2005) 5644–5647.
[
13] C. Kishor, T. Arya, R. Reddi, X. Chen, V. Saddanapu, A.K. Marapaka, R. Gumpena,
D. Ma, J.O. Liu, A. Addlagatta, Identification, biochemical and structural
evaluation of species-specific inhibitors against type I methionine
aminopeptidases, J. Med. Chem. 56 (13) (2013) 5295–5305.
[
[
[
38] A.W. Schüttelkopf, D.M. van Aalten, PRODRG: a tool for high-throughput
crystallography of protein-ligand complexes, Acta Crystallogr. D Biol. Crystallogr.
6
0 (Pt 8) (2004) 1355–1363.
39] O. Trott, A.J. Olson, AutoDock Vina: improving the speed and accuracy of docking
with a new scoring function, efficient optimization, and multithreading, J. Comput.
Chem. 31 (2) (2010) 455–461.
[
[
14] M.K. Haldar, M.D. Scott, N. Sule, D.K. Srivastava, S. Mallik, Synthesis of
barbiturate-based methionine aminopeptidase-1 inhibitors, Bioorg. Med. Chem.
Lett. 18 (7) (2008) 2373–2376.
40] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.
J. Olson, AutoDock4 and AutoDockTools4: Automated docking with selective
receptor flexibility, J. Comput. Chem. 30 (16) (2009) 2785–2791.
15] T.R. Helgren, P. Wangtrakuldee, B. Staker, T.J. Hagen, Advances in Bacterial
Methionine Aminopeptidase Inhibition, Curr. Top Med. Chem. 16 (4) (2016)
3
97–414.
1
3

S. Bala et al.
Bioorganic Chemistry 115 (2021) 105185
[
41] C. Oefner, A. Douangamath, A. D’Arcy, S. Hafeli, D. Mareque, A. Mac Sweeney,
J. Padilla, S. Pierau, H. Schulz, M. Thormann, S. Wadman, G.E. Dale, The 1.15A
crystal structure of the Staphylococcus aureus methionyl-aminopeptidase and
complexes with triazole based inhibitors, J. Mol. Biol. 332 (1) (2003) 13–21.
42] Sergio C. Chai, Wen-Long Wang, De-Rong Ding, Qi-Zhuang Ye, Growth inhibition
of Escherichia coli and methicillin-resistant Staphylococcus aureus by targeting
cellular methionine aminopeptidase, Eur. J. Med. Chem. 46 (8) (2011) 3537–3540.
43] V. Pillalamarri, C.G. Reddy, S.C. Bala, A. Jangam, V.V. Kutty, A. Addlagatta,
Methionine aminopeptidases with short sequence inserts within the catalytic
domain are differentially inhibited: Structural and biochemical studies of three
proteins from Vibrio spp, Eur. J. Med. Chem. 209 (2021) 112883.
aminopeptidases by bengamide derivatives, ChemMedChem 6 (6) (2011)
1041–1048.
[49] Pengtao Zhang, Xinye Yang, Feiran Zhang, Sandra B. Gabelli, Renxiao Wang,
Yihua Zhang, Shridhar Bhat, Xiaochun Chen, Manuel Furlani, L. Mario Amzel, Jun
O. Liu, Dawei Ma, Pyridinylpyrimidines selectively inhibit human methionine
aminopeptidase-1, Bioorg. Med. Chem. 21 (9) (2013) 2600–2617.
[50] D.H. Quan, G. Nagalingam, I. Luck, N. Proschogo, V. Pillalamarri, A. Addlagatta,
E. Martinez, V. Sintchenko, P.A.-O. Rutledge, J.A. Triccas, Bengamides display
potent activity against drug-resistant Mycobacterium tuberculosis, Sci. Rep. 9 (1)
(2019) 14396.
[
[
[51] R. Schneider, A. de Daruvar, C. Sander, The HSSP database of protein structure-
sequence alignments, Nucleic Acids Res. 25 (1) (1997) 226–230.
[52] F. Sievers, A. Wilm, D. Dineen, T.J. Gibson, K. Karplus, W. Li, R. Lopez,
H. McWilliam, M. Remmert, J. Soding, J.D. Thompson, D.G. Higgins, Fast, scalable
generation of high-quality protein multiple sequence alignments using Clustal
Omega, Mol. Syst. Biol. 7 (2011) 539.
[53] R.J. Ganji, R. Reddi, R. Gumpena, A.K. Marapaka, T. Arya, P. Sankoju, S. Bhukya,
A. Addlagatta, Structural basis for the inhibition of M1 family aminopeptidases by
the natural product actinonin: Crystal structure in complex with E. coli
aminopeptidase N, Protein Sci. 24 (5) (2015) 823–831.
[
44] Mir Mohammad Masood, Vijay K. Pillalamarri, Mohammad Irfan, Babita Aneja,
Mohamad Aman Jairajpuri, Md. Zafaryab, M. Moshahid A. Rizvi, Umesh Yadava,
Anthony Addlagatta, Mohammad Abid, Diketo acids and their amino acid/
dipeptidic analogues as promising scaffolds for the development of bacterial
methionine aminopeptidase inhibitors, RSC Adv. 5 (43) (2015) 34173–34183.
45] X. Hu, J. Zhu, S. Srivathsan, D. Pei, Peptidyl hydroxamic acids as methionine
aminopeptidase inhibitors, Bioorg. Med. Chem. Lett. 14 (1) (2004) 77–79.
46] F. Huguet, A. Melet, R. Alves de Sousa, A. Lieutaud, J. Chevalier, L. Maigre,
P. Deschamps, A. Tomas, N. Leulliot, J.-M. Pages, I. Artaud, Hydroxamic acids as
potent inhibitors of Fe(II) and Mn(II) E. coli methionine aminopeptidase: biological
activities and X-ray structures of oxazole hydroxamate-EcMetAP-Mn complexes,
ChemMedChem 7 (6) (2012) 1020–1030.
[
[
[54] N. Drinkwater, N.B. Vinh, S.N. Mistry, R.S. Bamert, C. Ruggeri, J.P. Holleran,
S. Loganathan, A. Paiardini, S.A. Charman, A.K. Powell, V.M. Avery, S. McGowan,
P.J. Scammells, Potent dual inhibitors of Plasmodium falciparum M1 and M17
aminopeptidases through optimization of S1 pocket interactions, Eur. J. Med.
Chem. 110 (2016) 43–64.
[
[
47] J.P. Lu, X.H. Yuan, Q.Z. Ye, Structural analysis of inhibition of Mycobacterium
tuberculosis methionine aminopeptidase by bengamide derivatives, Eur. J. Med.
Chem. 47 (1) (2012) 479–484.
[55] Anthony Addlagatta, Xiaoyi Hu, Jun O. Liu, Brian W. Matthews, Structural basis for
the functional differences between type I and type II human methionine
aminopeptidases, Biochemistry 44 (45) (2005) 14741–14749.
48] Jing-Ping Lu, Xiu-Hua Yuan, Hai Yuan, Wen-Long Wang, Baojie Wan, Scott
G. Franzblau, Qi-Zhuang Ye, Inhibition of Mycobacterium tuberculosis methionine
1
4