Identification of Adenosine Deaminase Inhibitors by Metal-binding Pharmacophore Screening

Article abstract of DOI:10.1002/cmdc.202000271

Adenosine deaminase (ADA) is a human mononuclear Zn²⁺ metalloenzyme that converts adenosine to inosine. ADA is a validated drug target for cancer, but there has been little recent work on the development of new therapeutics against this enzyme. The lack of new advancements can be partially attributed to an absence of suitable assays for high-throughput screening (HTS) against ADA. To facilitate more rapid drug discovery efforts for this target, an in vitro assay was developed that utilizes the enzymatic conversion of a visibly emitting adenosine analogue to the corresponding fluorescent inosine analogue by ADA, which can be monitored via fluorescence intensity changes. Utilizing this assay, a library of ～350 small molecules containing metal-binding pharmacophores (MBPs) was screened in an HTS format to identify new inhibitor scaffolds against ADA. This approach yielded a new metal-binding scaffold with a K_i value of 26±1 μM.

Full text of DOI:10.1002/cmdc.202000271

Accepted Article
Title: Identification of Adenosine Deaminase Inhibitors by Metal-binding
Pharmacophore Screening
Authors: Rebecca N Adamek, Paul Ludford, Stephanie M Duggan,
Yitzhak Tor, and Seth Mason Cohen
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemMedChem 10.1002/cmdc.202000271
Link to VoR: https://doi.org/10.1002/cmdc.202000271
01/2020

10.1002/cmdc.202000271
ChemMedChem
FULL PAPER
Identification of Adenosine Deaminase Inhibitors by Metal-
binding Pharmacophore Screening
Rebecca N. Adamek,[a] † Paul Ludford,[a]† Stephanie M. Duggan,[a] Yitzhak Tor,*[a] and Seth M. Cohen*[a]
[a]
Rebecca N. Adamek, Paul Ludford, Stephanie M. Duggan, Prof. Yitzhak Tor, Prof. Seth M. Cohen
Department of Chemistry and Biochemistry
University of California, San Diego
La Jolla, CA 92093, United States
†
These authors contributed equally to this work
Abstract: Adenosine deaminase (ADA) is a human mononuclear
Zn2+ metalloenzyme that converts adenosine to inosine. ADA is a
validated drug target for cancer, but there has been little recent work
on the development of new therapeutics against this enzyme. The
lack of new advancements can be partially attributed to an absence
of suitable assays for high-throughput screening (HTS) against ADA.
To facilitate more rapid drug discovery efforts for this target, an in
vitro assay was developed that utilizes the enzymatic conversion of
a
visibly emitting adenosine analogue to the corresponding
fluorescent inosine analogue by ADA, which can be monitored via
fluorescence intensity changes. Utilizing this assay, a library of
~350 small molecules containing metal-binding pharmacophores
(MBPs) was screened in an HTS format to identify new inhibitor
scaffolds against ADA. This approach yielded a new metal-binding
scaffold with a Ki value of 26±1 µM.
Introduction
Adenosine deaminase (ADA EC 3.5.4.4), also known as
adenosine dehydrolyase, is found across nearly all domains of
life, and plays
a
vital role in purine metabolism.
This
mononuclear Zn2+-dependent metalloenzyme functions to
deaminate adenosine and 2’-deoxyadenosine to yield inosine
and 2’-deoxyinosine, respectively.[1]
ADA activity has been implicated in certain types of
cancers, including lymphoma and leukemia.[2] As ADA has been
associated with other lymphoproliferative disorders, inhibitors of
ADA have been sought after as potential therapeutic agents.[3]
However, in consideration that ADA is also necessary for
immune system activity, reversible ADA inhibitors are preferable
to avoid potential on-target toxicity associated with irreversible
inhibitors. Currently there are only two FDA-approved ADA
inhibitors, pentostatin and cladribine (Figure 1), which are both
4]
clinically used in the treatment of hairy cell leukemia.[1a,
Notably, the potent ADA inhibitor erythro-9-(2-hydroxy-3-
nonyl)adenine (EHNA), is often used as a research tool, but
failed to enter clinical trials due to poor metabolism-related
5]
pharmacokinetic properties.[4b,
Therefore, the discovery of
viable new ADA inhibitors remains a significant challenge.[6]
Figure 1. Top: Mechanistic depiction of ADA-mediated deamination of
adenosine as well as tzA deamination to yield the active fluorophore tzI.
Middle: Structure and Ki values of ADA inhibitors Pentostatin, Cladribine, and
EHNA. Bottom: Crystal structure of EHNA (green) bound to the ADA active
site (PDB ID: 2Z7G). Key residues, catalytic Zn2+ (orange sphere), and Zn2+
bound water (red sphere) are highlighted.
-
1
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000271
ChemMedChem
FULL PAPER
Current bioaffinity methods to analyze ADA activity include
and SI; Figure S1), and was determined to have an IC50 value of
chromatographic separations,[6c] which are labor and time
consuming, and absorption spectroscopy, the latter of which
suffers from optical interference by inhibitors that display
absorption in the wavelengths between 200–300 nm, where
adenosine and its derivatives absorb.[4a, 6a, 6b, 6d] While newer
methods have been disclosed,[7] to the best of our knowledge,
their application in high-throughput screening (HTS) has not
been realized. Herein, we demonstrate the use of a new HTS
assay for ADA based on a recently described, fluorescent
reporter of ADA activity. The method relies on isothiazolo
adenosine (tzA), a visibly emitting adenosine surrogate as a
mimic of the native substrate.[8] This isothiazolo-based purine
analog (KM = 19 µM) undergoes deamination as effectively as
adenosine (KM = 28 µM) yielding tzI, the corresponding distinctly
6±2 nM, consistent with literature values.[1a,
All MBP
7d, 7e]
fragments were screened in triplicate with a positive (EHNA 10
µM) and negative (buffer only) control. From this screen, 36
MBPs were identified to have >50% ADA inhibition at 200 µM,
and an additional 20 compounds had >75% ADA inhibition this
concentration (Figure 2). Of these, compound 1 had the highest
overall activity in the initial screen with 89% inhibition of ADA
activity relative to the negative control (Figure 3). Scaffolds 2
and 3 also emerged as potential lead fragments, with activities
of ~83% inhibition at 200 µM against ADA (Figure 3).
emissive inosine isostere.[7e]
These easily detectable
fluorescence changes serve as the foundation for a real-time
and sensitive assay for the discovery of ADA inhibitors, which
can be executed using commercial plate readers,[7e] thereby
facilitating more rapid drug discovery efforts.
The deamination mechanism of ADA has been thoroughly
investigated (Figure 1).[1a] The catalytic Zn2+ is coordinated by
His15, His17, His214, and Asp295 with a water/hydroxide
completing the distorted trigonal bipyramidal coordination
geometry.[9] This Zn2+-activated nucleophilic water attacks the
adenosine at the C6 position, and the resulting rehybridization,
involving the departure of ammonia affords inosine (Figure 1).
Notably, active site amino acid side-chain residues and water
molecules mediate the necessary proton transfer reactions to
complete the deamination mechanism. While the majority of
ADA inhibitors are adenosine surrogates or transition state
analogs, the mechanism articulated above suggests that binding
to the catalytic Zn2+ could effectively inactivate this
metalloenzyme.[10] Therefore, it was decided to employ a metal-
binding pharmacophore (MBP) strategy against ADA in an effort
to discover new inhibitory motifs.
Figure 2. Results of the MBP library screen against ADA, represented as a
bar graph of percent inhibition (relative to no inhibitor) of ADA in the presence
of 200 µM MBP fragment. The green dashed line denotes compounds with
greater than 50% inhibition, and the red dashed line denotes compounds with
greater than 75% inhibition.
The use of MBPs to target the metal ion co-factor in
To further validate the fluorescent assay transition to a
plate reader format, an IC50 curve for 2 was generated and gave
a value of 30±1 µM (Figure S2). Quite nicely, from this curve,
the percent inhibition of ADA at 200 µM 2 was 84%, which was
consistent with the value previously obtained in the initial MBP
library screen; serving as a secondary control to demonstrate
conversion of the assay to a HTS, plate-based format.
metalloenzymes has been demonstrated to be
a viable
approach for inhibitor development.[10-11] One of the primary
benefits of the MBP approach is the intrinsic strength of the non-
covalent, reversible coordinate covalent bond, allowing for a
high ligand efficiency from the MBP alone. This MBP approach
is highly compatible with fragment-based drug discovery (FBDD).
Herein, we utilize the previously described tzA to tzI biophysical
assay[7e] in an HTS format, and use this tool to evaluate a library
of ~350 MBPs in an effort to discover new, metal-binding
inhibitors of ADA. Subsequent lead elaboration, as guided by
structure-activity relationship (SAR) analysis, resulted in the
identification of structural motifs that contributed to fragment
activity, resulting in the discovery of a new imidazoline scaffold
for ADA inhibitors.
Results
Figure 3. MBP compounds tested against ADA in this study.
A library of ~350 MBPs was screened at an initial
concentration of 200 µM in an HTS format, based on the tzA to tzI
reaction[7e] in a 96-well plate format. Fluorescence changes
were monitored following excitation at 322 nm and an emission
wavelength of 410 nm. As a positive control, EHNA was
screened in a 10-point inhibition curve (see Experimental part
Oxazoline Derivative Synthesis and Inhibitory Activity
The phenol derivatives 2-imidazolyl phenol and 2-oxazolyl
phenol (3 and 4, respectively) inhibited ADA-mediated
deamination by 83% and 61%, respectively at 200 µM.
Complete titrations of ADA activity with 3 and 4 yielded IC50
2
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000271
ChemMedChem
FULL PAPER
values of 93±17 and 260±14 µM, respectively (Figures S3 and
S4, respectively). Although these oxazoline- and imidazoline-
based compounds are known as components of certain
siderophores,[12] to the best of our knowledge, they have not
been explored as metalloenzyme inhibitors. To further explore
this structural motif, a total of 11 derivatives of 4 were prepared
Table 1. IC50 and Ki Values of prepared oxazoline derivatives. Standard
deviations listed are calculated from three independent experiments.
Compound
IC50 (µM)
Ki (µM)
0.005 ± 0.002a
0.019 ± 0.004b
~0.005c
0.0012 ± 0.0002b
77 ± 14
0.006 ± 0.002a
0.023 ± 0.004b
EHNA
Pentostatin
0.0014 ± 0.0001b
93 ± 17
260 ± 14
> 700
to develop
a structure-activity relationship (SAR) of the
oxazoline scaffold against ADA (Figure 3). Table 1 lists the IC50
values for these derivatives.
3
4
215 ± 12
n.d.
The synthesis of these oxazoline derivatives was
straightforward (Scheme 1). Cyclization to form the oxazoline
ring was achieved by heating pre-functionalized 2-
hydroxybenzonitriles with ethanolamine and catalytic amounts of
ZnCl2 at 110 ºC in toluene for 16 h. Subsequent work-up gave 3
– 5 and 10, 11, and 13 in good yield. For 6–9, the desired
stereochemistry was achieved by substituting the ethanolamine
respectively with (R)-1-aminopropan-2-ol, (S)-1-aminopropan-2-
ol, (S)-2-aminopropan-1-ol, or (R)-2-aminopropan-1-ol to attain
cyclization with the desired pre-installed stereochemistry.
Compounds 3, 5, and 13 were also prepared using the same
5
6
194 ± 4
188 ± 10
455 ± 8
> 700
161 ± 3
7
156 ± 8
8
377 ± 7
9
n.d.
10
11
12
13
14
15
16
17
18
> 700
n.d.
618 ± 9
372 ± 24
> 700
512 ± 7
308 ± 20
n.d.
Insoluble
273 ± 19
102 ± 17
31 ± 1
Insoluble
226 ± 15
85 ± 14
cyclization
method,
but
with
ethane-1,2-diamine,
2-
aminoethane-1-thiol, or 3-aminopropan-1-ol, respectively,
instead of ethanolamine. As an alternative method for oxazoline
cyclization, when the requisite benzonitrile was unavailable, a
pre-functionalized salicylic acid was used as the basis to
generate an ethanolamine-amide moiety. This ethanolamine-
amide was then cyclized in the presence of thionyl chloride to
form the desired oxazoline in 12 and 14–16.
26 ± 1
Insoluble
Insoluble
a Values were determined using the HTS assay format reported
here (Figure S1)
b Values determined using a non-HTS assay format8e
c Value reported in Cristalli et al.[1a]
Discussion
The use of MBPs to discover new, metal-binding
metalloenzyme inhibitors has been a fruitful strategy, and has
been employed against a variety of metalloenzyme targets,
including glyoxalase-1, New Delhi metallo-β-lactamase-1,
influenza PAN endonuclease, proteasomal deubiquitinase Rpn11,
MMP-12, and hCAII.[10, 13] These MBPs produce metalloenzyme
inhibition by coordinating the metal ions within the enzyme
active site, thereby blocking catalysis.[10, 13] Here we combine
this MBP fragment library with a recently developed sensitive,
real-time, fluorescence-based assay, in an effort to identify new
motifs that can serve as scaffolds for new ADA inhibitors.
The key to this screening campaign was the use of an
emissive adenosine surrogate (tzA, Figure 1), which is effectively
deaminated by ADA to the corresponding emissive inosine
derivative.[8] tzA was previously found to have a similar KM and
kcat to native adenosine.[7e] Its conversion to tzI can be observed
by fluorescence upon excitation in the near UV wavelength
range, a range typically outside the absorption spectra of the
Scheme 1. General method for oxazoline synthesis by: a) cyclization
reaction between ethanolamine or aminopropanol with an aromatic nitrile or b)
by cyclizing amide-coupled ethanolamine. Reagents and conditions: (a)
potassium acetate, hydroxylamine hydrochloride, formic acid, 100 ºC, 24 h, 57
% yield; (b) ZnCl2, ethanolamine, toluene, 130 ºC, 24 h, 75 % yield; (c)
ethanolamine, triethylamine, T3P, tetrahydrofuran, ethyl acetate, 70 ºC, 16 h,
24 % yield; (d) EDC, HOBt, triethylamine, ethanolamine, DMF, 60 ºC, 16 h, 20
% yield.
Representative data and IC50 value determinations are
found in Figures S2-S14. Of the prepared compounds, 16 had
the greatest activity with an IC50 value of 102±17 µM (Figure
S12). Compounds 9, 18, and 13 all displayed <10% inhibition at
100 µM (Figure S14). Both 15 and 18 were found to be
insoluble at concentrations >250 µM with 2.5 v/v% DMSO
present in the assay buffer, preventing determination of accurate
percent inhibition. Compound 5 showed no inhibition at a
concentration of 100 µM. Based on this SAR, an additional
inhibitor, 17, was synthesized in a fragment merge strategy to
combine the molecular features that appear to enhance potency.
Combining the 2-naphthol in 16 with the imidazoline 3 resulted
in 17 with an IC50 value of 31±1 µM, the best among the
molecules prepared in this study (Figure S13).
native nucleosides and simple pharmacophores.
Taking
advantage of the emissive properties of tzA and its similar
reaction time to adenosine, we were able to translate this
enzymatic assay to a more useful HTS format using standard
plate readers. To validate the HTS format and methods used,
EHNA, a known ADA inhibitor, was assessed and found to have
an IC50 = 6 nM, consistent with previously reported values
(Figure S1).[1a, 7d, 7e] One of the more potent fragments identified
in the screen, 3 was found to have an IC50 = 93 µM, providing an
excellent starting point for additional study (Figure S3).
To preliminarily establish SAR, two of the fragments, 3 and
4, were used as core scaffolds for inhibitor exploration. Although
3
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000271
ChemMedChem
FULL PAPER
3 was more active against ADA than 4, the oxazoline scaffold
was selected for preliminary SAR study, as it was more
amenable towards derivatization. It was decided to perform a
cursory SAR exploration of 4 with the intent of applying
discovered trends towards the more active scaffold 3. As such,
eleven compounds were synthesized, which included
modifications probing for interactions with hydrophobic or
hydrophilic pockets (Figure 3). For these prepared compounds,
an initial screen at 100 µM was performed to provide preliminary
insight into the most active modifications. Compounds that
showed <10% inhibition at 100 µM were not further analyzed.
The remaining compounds were then screened at various
concentrations to derive IC50 values, as described in the
experimental section.
The SAR study indicated a methyl group at the 4 position
of the dihydro-oxazole ring in 6, and 7 slightly increased potency.
However, incorporation of a methyl substituent at the 3 position
of the oxazoline ring in 8 and 9 resulted in a loss of activity.
Expansion of the oxazoline ring to the corresponding
dihydrooxazine derivative 13 also caused a significant decrease
in potency, possibly due to a disruption of the binding angle
necessary for metal coordination. Additionally, incorporation of
a methyl group to the aromatic ring in 10, 11, and 12 also
decreased activity. Conversely, replacement of the phenol ring
with an α-naphthol at the 2-position in 16 resulted in a greater
than two-fold increase in potency compared to parent compound
4. Whereas replacement with a β-naphthol attached at the 1-
position in 15 had little to no effect on activity, potentially
indicating a selectivity for preferred location of lipophilic bulk.
Finally, altering the heteroatoms had a significant impact on
inhibitor potency in that replacing the oxazoline ring in 4 with the
for medicinal chemistry that have not yet been explored by prior
ADA inhibitor classes.
Experimental Section
Unless otherwise noted, reagents were purchased from
commercially available sources, and were used with no additional
purification. Solvents were purchased from Sigma-Aldrich and Fisher
Scientific, and dried by standard techniques.
Silica gel column
chromatography was performed using either a CombiFlash Rf or Rf+
Teledyne ISCO system, using hexane, ethyl acetate, CH2Cl2 or MeOH as
eluents. Separations were monitored via UV absorbance, reading at 250
and 280 nm, as well as by a Teledyne ISCO RF+ PurIon ESI-MS detector
with 1 Da resolution. 1H and 13C NMR spectra were obtained using
either a Varian 400 MHz or a Varian 500 MHz spectrometer at the
Department of Chemistry and Biochemistry at UC San Diego. NMR data
is reported in parts per million relative to the residual non-deuterated
solvent signals. When available, coupling constants (J) are reported in
hertz (Hz). Standard resolution mass spectrometry was performed at
either the UC San Diego Molecular Mass Spectrometry Facility or on the
previously described Teledyne ISCO RF+ PurIon ESI-MS detector.
Analytical HPLC and high-resolution mass spectrometry using an ESI-
TOF probe were performed at the U.C. San Diego Molecular Mass
Spectrometry Facility.
Synthesis
tzA was synthesized based on previously published procedures.[8]
Compound 18 was purchased from Matrix Scientific. Compound 1 was
prepared as previously reported[13f]
. Lead compound 17 was prepared
following the exact protocol as reported in Sapegin, A. et. al.[14] and
matched all literature characterization. Experimental 1H and 13C NMR for
all prepared compounds are reported in the Supplementary Information
(Figures S15-S30). The purity of 3–17 was determined to be at least
97% by analytical HPLC (Figures S31-S45).
imidazoline in 3 caused an increase in inhibition activity.
A
possible explanation for this increase in potency could be that
the imidazoline ring serves as a better substrate mimetic of the
native purine than oxazoline.
General Method for Oxazoline Derivative Synthesis
The results described above suggested that substitution of
an oxazoline with an imidazoline ring and replacing the phenol
with the fused aromatic derivatives such as naphthols would
result in a fragment with enhanced potency. As a result,
fragment 17 was synthesized and found to have an IC50 value of
31±1 µM, the most potent inhibitor among the molecules studied
here (Figure S13).
General Method for Oxazoline Cyclization: To a solution of 2-
nitrilephenol (1 eq.) in toluene (5 mL), ZnCl2 (0.02 eq.) and aminoethanol
(1.5 eq.) were sequentially added. The mixture was purged with nitrogen
and heated to reflux at 130 ºC for 20 h. After cooling to RT, the reaction
was concentrated under reduced pressure and the residue was taken up
in water and ethyl acetate. The aqueous phase was extracted with ethyl
acetate (5 × 5 mL), the combined organic phase was dried over
magnesium sulfate, and the filtrate was concentrated under reduced
pressure. The resulting pink-brown residue was purified by column
chromatography with a gradient of 0% to 20% ethyl acetate in hexanes,
Conclusion
unless otherwise noted.
Desired fractions were combined and
concentrated under reduced pressure to obtain the desired oxazoline
derivative. Compounds that were isolated as oils were cooled to –20 ºC
for 1 h to solidify, and remained solid thereafter.
A rapid and simple HTS format for ADA inhibitors has been
established using tzA, an emissive and isofunctional adenosine
surrogate. A library of ~350 MBPs were screened using this
fluorescence-based assay in a 96-well plate format. Among
them, 36 compounds inhibiting ADA by more than 50 % at 200
µM were discovered. To generate refined SAR, 11 derivatives
(S)-2-(5-Methyl-4,5-dihydrooxazol-2-yl)phenol (6): From the general
method for oxazoline cyclization, from 2-hydroxybenzonitrile (0.500 g,
4.20 mmol), ZnCl2 (11.4 mg, 83.9 µmol), and (S)-1-aminopropan-2-ol
(473 mg, 6.30 mmol) in toluene (5 mL) at 130 ºC for 20 h, the desired
product 6 (0.270 g, 1.52 mmol, 36%) was obtained as a clear oil that
solidified upon cooling to -20 ºC to a white crystalline solid. 1H NMR (400
MHz, DMSO-d6): δ 12.28 (br s, 1H), 7.60 (d, J = 8.0, 1H), 7.41 (t, J = 7.2,
1H), 6.96 (d, J = 8.4, 1H), 6.90 (t, J = 7.6, 1H), 4.97 – 4.87 (m, 1H), 4.17
(ddd, J1 = 14.4, J2 = 9.6, J3 = 2.8, 1H), 3.61 (ddd, J1 = 14.4, J2 = 7.6, J3 =
2.8, 1H), 2.49 (d, J = 1.6, 3H); 13C NMR (500 MHz, DMSO-d6): δ 165.2,
159.6, 134.0, 128.2, 119.3, 116.9, 110.7, 76.4, 59.6, 21.0; HRMS (ESI-
TOF): m/z calcd for C10H12NO2+H+: 178.0863 [M+H]+; found: 178.0864.
were synthesized and analyzed.
This focused collection,
shedding light on the molecular features that contribute to
inhibitory activity, has guided the synthesis of a merged
molecule with the highest potency among these MBP fragments.
Future work in this study will entail further use of this fluorescent
HTS assay to guide rational elaboration of 17 to attain more
potent ADA inhibitors. Importantly, we are unaware of a metal-
centric approach for the development of ADA inhibitors. It is
possible that MBP-based inhibitors could open up new avenues
4
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000271
ChemMedChem
FULL PAPER
(R)-2-(5-Methyl-4,5-dihydrooxazol-2-yl)phenol (7): From the general
method for oxazoline cyclization, from 2-hydroxybenzonitrile (0.500 g,
4.20 mmol), ZnCl2 (11.4 mg, 83.9 µmol), and (R)-1-aminopropan-2-ol
(473 mg, 6.30 mmol) in toluene (5 mL) at 130 ºC for 20 h, the desired
product 7 (0.440 g, 2.48 mmol, 59%) was obtained as a clear oil that
solidified upon cooling to -20 ºC to a white crystalline solid. 1H NMR (400
MHz, DMSO-d6): δ 12.28 (br s, 1H), 7.59 (d, J = 8.0, 1H), 7.41 (t, J = 7.6,
1H), 6.96 (d, J = 8.0, 1H), 6.90 (t, J = 8.0, 1H), 4.97 – 4.88 (m, 1H), 4.16
(dd, J1 = 14.0, J2 = 9.6, 1H), 3.61 (dd, J1 = 14.4, J2 = 7.6, 1H), 1.37 (d, J
= 6.4, 3H); 13C NMR (500 MHz, DMSO-d6): δ 165.2, 159.6, 134.0, 128.2,
119.3, 116.9, 110.7, 76.4, 59.6, 21.0; HRMS (ESI-TOF): m/z calcd for
C10H12NO2+H+: 178.0863 [M+H]+; found: 178.0864.
In the final two wells of the row, phosphate buffer and EHNA were added,
respectively, in place of 2. Two more rows were prepared in the same
manner and a fourth row was prepared with phosphate buffer in place of
ADA. After two minutes of incubation, tzA (50 µL of 7.8 µM) was added
and mixed thoroughly. Wells were then excited at 322 nm and emission
monitored at 410 nm. Intensity of emission was measured every 37
seconds for just over 30 minutes.
IC50 Analysis of 4 and Derivatives
Concentrated stock solutions were prepared for tzA (1 mM) and potential
inhibitors (100 mM) in DMSO. Bovine spleen ADA was obtained from
Sigma Aldrich (EC Number 232-817-5). The commercial solution was
diluted to 1.15 U mL-1 by dissolving an aliquot (1 µL) in freshly prepared
phosphate buffer (999 µL, 50 mM, pH 7.4). The enzyme stock solution
was freshly prepared and kept on ice prior to use. Fresh dilutions of
each compound tested were prepared (1, 10 mM) in DMSO from the 100
mM stock solution. For 17, an additional 0.1 mM solution in DMSO was
prepared. Emission change was monitored on a Horiba Fluoromax-4
equipped with a cuvette holder with a stirring system setting both the
excitation and the emission slits at 3 nm, the resolution at 1 nm and the
integration time 0.1 s. The Horiba Fluoromax-4 was equipped with a
thermostat controlled ethylene glycol-water bath fitted to specially
designed cuvette holder and the temperature was kept at 25.0 ± 0.1 ºC.
Reactions were monitored in a 3 mL quartz cuvette while stirring. 1.7 mL
of phosphate buffer was added to the cuvette. A volume equal to the
sum of the tzA, ADA, and compound volumes was removed from the
cuvette. If the final concentration of inhibitor was 10, 100, or 1000 µM,
17 µL of 1, 10, or 100 mM compound solution was added, respectively. If
the final concentration of inhibitor was 1 µM, 17 µL of 1 mM 4 derivative
solution was added. If the final concentration of inhibitor was 25 or 250
µM, 4.25 µL of 10 or 100 mM 4 derivative solution was added,
respectively. For a final concentration of 0.1 µM 17, 1.7 µL of 0.1 mM
solution was added. 6.6 µL of tzA was added to the cuvette and the
solution was stirred for 1 minute. Initial emission at 410 nm upon
General Method for High Throughput Screen
For all assay screening, bovine spleen ADA was used, as it has 91%
sequence similarity overall with human ADA, and is highly conserved at
the active site.[15] Therefore, the values obtained with bovine spleen ADA
should be comparable with human ADA. Concentrated stock solutions
were prepared for tzA (8.4 mM) and potential inhibitors (50 mM) in DMSO.
Bovine spleen ADA was obtained from Sigma Aldrich (EC Number 232-
817-5). The commercial suspension (1150 U mL-1 in 3.2 M (NH4)2SO4,
0.01 M potassium phosphate, pH 6.0) was diluted to 1.15 U mL-1 by
dissolving an aliquot (1 µL) in freshly prepared phosphate buffer (999 µL,
50 mM, pH 7.4). A freshly prepared enzyme stock solution was kept on
ice prior to use. Fresh dilutions of tzA (7.8 µM) and enzyme (13.8 mU mL-
1) in previously prepared phosphate buffer were made just before the
reaction and kept at room temperature. All high throughput assays were
run on a BioTek Synergy H4 microplate reader. Reactions were run in
Costar black 96-well plates that were filled and then inserted into the
plate reader. Temperature was maintained at 37.0 ± 0.1 ºC. Reagents
and enzymes were added in the following order: 20 µL of enzyme first, 30
µL of inhibitor second, and 50 µL of tzA last. Wells were excited at 322
nm and emission monitored at 410 nm. Intensity of emission was
measured every 37 seconds for just over 30 minutes.
excitation at 322 nm was measured.
ADA was added and
measurements continued every 10 seconds for 20 minutes.
High Throughput Screen of EHNA
Note: For fragments 14, 15, and 16 at concentrations of 25 and 250 µM,
42.5 µL of 1 and 10 mM were added, respectively.
Fresh dilutions of EHNA (0.05, 0.5, 5, 12.5, 50, 125, 500, 1250, 5000,
50000 nM) in phosphate buffer were prepared. ADA (20 µL of 13.8 mU
mL-1 solution) and EHNA (30 µL) were added to successive wells in a
row and thoroughly mixed. Phosphate buffer was used as a negative
control. Two more rows were prepared in the same manner and a fourth
row was prepared with phosphate buffer in place of ADA. After two
minutes of incubation, tzA (50 µL of 7.8 µM) was added and mixed
thoroughly. Wells were then excited at 322 nm and emission monitored
at 410 nm. Intensity of emission was measured every 37 seconds for
just over 30 minutes.
Data Analysis
The signal intensity of each time point of tzA conversion to tzI was
converted to a percent change relative to the initial intensity. The
corresponding percent change values of the run without ADA were then
subtracted from the runs with inhibitor. The final point of each curve was
then divided by the final point of the run without inhibitor to yield the
percent activity for that run. The percent activity of each inhibitor
concentration was then subtracted from one to yield the percent inhibition.
For IC50 value determination, percent inhibition points were plotted
against the respective inhibitor concentration on a logarithmic scale x-
axis. The resulting plot was fit with a Hill curve and the IC50 value
calculated from the constants. IC50 values were converted to KI values
using the Cheng-Prusoff equation.
High Throughput Screen of MBP Library
Fresh dilutions of each inhibitor (1 mM) in phosphate buffer were
prepared. 20 µL of 13.8 mU mL-1 ADA and 30 µL of each inhibitor were
added to successive wells in a row and thoroughly mixed. In the final two
wells of the row, phosphate buffer and EHNA were added, respectively,
in place of inhibitor. Two more rows were prepared in the same manner
and a fourth row was prepared with phosphate buffer in place of ADA.
After two minutes of incubation, tzA (50 µL of 7.8 µM) was added and
mixed thoroughly. Wells were then excited at 322 nm and emission
monitored at 410 nm. Intensity of emission was measured every 37
seconds for just over 30 minutes.
Acknowledgements
The authors would like to thank Dr. Yongxuan Su for mass
spectrometry sample analysis at the U.C. San Diego Chemistry
and Biochemistry Mass Spectrometry Facility. We also thank
the remaining members of the Cohen and Tor Laboratory groups,
both former and current, for their support and advice; in
particular Dr. Cy Credille for preparing compound 2. This work
IC50 Analysis of 2
Fresh dilutions of 2 (0.005, 0.05, 0.125, 0.5, 5, 12.5, 50, 125, 500, 1000
µM) in phosphate buffer were prepared. 20 µL of 13.8 mU mL-1 ADA and
30 µL of 2 were added to successive wells in a row and thoroughly mixed.
5
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000271
ChemMedChem
FULL PAPER
2019, 20, 718-726; f) K. Hu, Y. Huang, S. Wang, S. Zhao,
J. Pharm. Biomed. Anal. 2014, 95, 164-168.
A. R. Rovira, A. Fin, Y. Tor, J. Am. Chem. Soc. 2015, 137,
14602-14605.
was supported by National Institute of Health grant R01
GM098435, R01 AI149444, R01 GM069773, and the Chemical
Biology Interfaces at U.C. San Diego training grant
T32GM112584-01. S.M.C. is a cofounder of and has an equity
interest in Cleave Therapeutics and Forge Therapeutics,
companies that may potentially benefit from the research results.
S.M.C. also serves on the Scientific Advisory Board for these
companies. The terms of this arrangement have been reviewed
and approved by the University of California, San Diego in
accordance with its conflict of interest policies.
[8]
[9]
a) D. K. Wilson, F. A. Quiocho, Biochemistry 1993, 32,
1689-1694; b) D. K. Wilson, F. B. Rudolph, F. A. Quiocho,
Science 1991, 252, 1278-1284.
A. Y. Chen, R. N. Adamek, B. L. Dick, C. V. Credille, C. N.
Morrison, S. M. Cohen, Chem. Rev. 2019, 119, 1323-1455.
Y. Yang, X. Q. Hu, Q. S. Li, X. X. Zhang, B. F. Ruan, J. Xu,
C. Liao, Curr Top Med Chem 2016, 16, 384-396.
a) R. J. Bergeron, Chem. Rev. 1984, 84, 587-602; b) H. R.
Hoveyda, V. Karunaratne, S. J. Rettig, C. Orvig, Inorg.
Chem. 1992, 31, 5408-5416.
[10]
[11]
[12]
[13]
a) C. Perez, A. M. Barkley-Levenson, B. L. Dick, P. F.
Glatt, Y. Martinez, D. Siegel, J. D. Momper, A. A. Palmer,
S. M. Cohen, J. Med. Chem. 2019, 62, 1609-1625; b) J. Li,
Y. Zhang, B. Da Silva Sil Dos Santos, F. Wang, Y. Ma, C.
Perez, Y. Yang, J. Peng, S. M. Cohen, T. F. Chou, S. T.
Hilton, R. J. Deshaies, Cell. Chem. Biol. 2018, 25, 1350-
1358 e1359; c) C. V. Credille, B. L. Dick, C. N. Morrison, R.
W. Stokes, R. N. Adamek, N. C. Wu, I. A. Wilson, S. M.
Cohen, J. Med. Chem. 2018, 61, 10206-10217; d) A. Y.
Chen, P. W. Thomas, A. C. Stewart, A. Bergstrom, Z.
Cheng, C. Miller, C. R. Bethel, S. H. Marshall, C. V.
Credille, C. L. Riley, R. C. Page, R. A. Bonomo, M. W.
Crowder, D. L. Tierney, W. Fast, S. M. Cohen, J. Med.
Chem. 2017, 60, 7267-7283; e) A. Y. Chen, P. W. Thomas,
Z. Cheng, N. Y. Xu, D. L. Tierney, M. W. Crowder, W. Fast,
S. M. Cohen, ChemMedChem 2019, 14, 1271-1282; f) R.
N. Adamek, C. V. Credille, B. L. Dick, S. M. Cohen, J. Biol.
Inorg. Chem. 2018, 23, 1129-1138.
Keywords: Fluorescent Probes
• Fragment Based Drug
Discovery • High Throughput Screening • Medicinal Chemistry •
Metalloenzymes
References
[1]
a) G. Cristalli, S. Costanzi, C. Lambertucci, G. Lupidi, S.
Vittori, R. Volpini, E. Camaioni, Med. Res. Rev. 2001, 21,
105-128; b) V. Tardif, R. Muir, R. Cubas, M. Chakhtoura,
P. Wilkinson, T. Metcalf, R. Herro, E. K. Haddad, Nat.
Commun. 2019, 10, 823.
[2]
a) Y. Nakajima, T. Kanno, T. Nagaya, K. Kuribayashi, T.
Nakano, A. Gotoh, T. Nishizaki, Cell. Physiol. Biochem.
2015, 35, 51-60; b) B. Kutryb-Zajac, P. Koszalka, P.
Mierzejewska, A. Bulinska, M. A. Zabielska, K. Brodzik, A.
Skrzypkowska, L. Zelazek, I. Pelikant-Malecka, E. M.
Slominska, R. T. Smolenski, J. Cell. Mol. Med. 2018, 22,
5939-5954.
[14]
[15]
A. Sapegin, A. Osipyan, M. Krasavin, Org. Biomol. Chem.
2017, 15, 2906-2909.
T. Kinoshita, N. Nishio, A. Sato, M. Murata, Acta.
Crystallogr. D Biol. Crystallogr. 1999, 55, 2031-2032.
[3]
a) A. V. Sauer, I. Brigida, N. Carriglio, A. Aiuti, Front.
Immunol. 2012, 3, 265; b) A. Mediero, B. N. Cronstein,
Trends Endocrinol. Metab. 2013, 24, 290-300; c) B.
Kutryb-Zajac, L. Mateuszuk, P. Zukowska, A. Jasztal, M. A.
Zabielska, M. Toczek, P. Jablonska, A. Zakrzewska, B.
Sitek, J. Rogowski, R. Lango, E. M. Slominska, S.
Chlopicki, R. T. Smolenski, Cardiovasc. Res. 2016, 112,
590-605; d) A. Cortes, E. Gracia, E. Moreno, J. Mallol, C.
Lluis, E. I. Canela, V. Casado, Med. Res. Rev. 2015, 35,
85-125; e) L. Antonioli, B. Csoka, M. Fornai, R. Colucci, E.
Kokai, C. Blandizzi, G. Hasko, Drug Discovery Today
2014, 19, 1051-1068; f) L. Antonioli, R. Colucci, C. La
Motta, M. Tuccori, O. Awwad, F. Da Settimo, C. Blandizzi,
M. Fornai, Curr. Drug Targets 2012, 13, 842-862.
[4]
[5]
[6]
a) H. J. Schaeffer, C. F. Schwender, J. Med. Chem. 1974,
17, 6-8; b) I. Gillerman, B. Fischer, J. Med. Chem. 2011,
54, 107-121.
a) W. R. McConnell, S. M. El Dareer, D. L. Hill, Drug
Metab Dispos 1980, 8, 5-7; b) C. U. Lambe, D. J. Nelson,
Biochem Pharmacol 1982, 31, 535-539.
a) S. C. Zimmermann, J. M. Sadler, P. I. O'Daniel, N. T.
Kim, K. L. Seley-Radtke, Nucleosides, Nucleotides Nucleic
Acids 2013, 32, 137-154; b) T. Tite, N. Lougiakis, V.
Myrianthopoulos, P. Marakos, E. Mikros, N. Pouli, R.
Tenta, E. Fragopoulou, T. Nomikos, Tetrahedron 2010, 66,
9620–9628; c) G. Li, I. Nakagome, S. Hirono, T. Itoh, R.
Fujiwara, Pharmacol. Res. Perspect. 2015, 3, e00121; d)
F. Da Settimo, G. Primofiore, C. La Motta, S. Taliani, F.
Simorini, A. M. Marini, L. Mugnaini, A. Lavecchia, E.
Novellino, D. Tuscano, C. Martini, J. Med. Chem. 2005, 48,
5162-5174.
[7]
a) L. Zhang, J. Zhao, J. Jiang, R. Yu, Chem. Commun.
(Cambridge, U. K. 2012, 48, 10996-10998; b) Y. Xu, B. J.
Venton, Phys. Chem. Chem. Phys. 2010, 12, 10027-
10032; c) X. J. Xing, X. G. Liu, H. Yue, Q. Y. Luo, H. W.
Tang, D. W. Pang, Biosens. Bioelectron. 2012, 37, 61-67;
d) R. W. Sinkeldam, L. S. McCoy, D. Shin, Y. Tor, Angew.
Chem. Int. Ed. Engl. 2013, 52, 14026-14030; e) P. T.
Ludford, 3rd, A. R. Rovira, A. Fin, Y. Tor, ChemBioChem
6
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000271
ChemMedChem
FULL PAPER
Entry for the Table of Contents
Metal-binding inhibitors of the metalloenzyme Adenosine Deaminase (ADA) were discovered by high-throughput screening. A
fragment merge strategy of combining active scaffolds lead to the discovery of an imidazoline-based inhibitor. This work
demonstrates the potential of metal-binding inhibitors to provide novel routes for ADA inhibition for therapeutic intervention.
7
This article is protected by copyright. All rights reserved.