Real-Time Monitoring of Human Guanine Deaminase Activity by an Emissive Guanine Analog

Article abstract of DOI:10.1021/acschembio.1c00232

Guanine deaminase (GDA) deaminates guanine to xanthine. Despite its significance, the study of human GDA remains limited compared to other metabolic deaminases. As a result, its substrate and inhibitor repertoire are limited, and effective real-time activity, inhibitory, and discovery assays are missing. Herein, we explore two emissive heterocyclic cores, based on thieno[3,4-d]pyrimidine (thN) and isothiazole[4,3-d]pyrimidine (tzN), as surrogate GDA substrates. We demonstrate that, unlike the thieno analog, thGN, the isothiazolo guanine surrogate, tzGN, does undergo effective enzymatic deamination by GDA and yields the spectroscopically distinct xanthine analog, tzXN. Further, we showcase the potential of this fluorescent nucleobase surrogate to provide a visible spectral window for a real-time study of GDA and its inhibition.

Full text of DOI:10.1021/acschembio.1c00232

pubs.acs.org/acschemicalbiology
Articles
Real-Time Monitoring of Human Guanine Deaminase Activity by an
Emissive Guanine Analog
Marcela S. Bucardo, You Wu, Paul T. Ludford, III, Yao Li, Andrea Fin, and Yitzhak Tor*
Cite This: ACS Chem. Biol. 2021, 16, 1208−1214
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Guanine deaminase (GDA) deaminates guanine to
xanthine. Despite its significance, the study of human GDA remains
limited compared to other metabolic deaminases. As a result, its
substrate and inhibitor repertoire are limited, and effective real-time
activity, inhibitory, and discovery assays are missing. Herein, we
explore two emissive heterocyclic cores, based on thieno[3,4-
d]pyrimidine (thN) and isothiazole[4,3-d]pyrimidine (tzN), as
surrogate GDA substrates. We demonstrate that, unlike the thieno
analog, thGN, the isothiazolo guanine surrogate, tzGN, does undergo
effective enzymatic deamination by GDA and yields the
spectroscopically distinct xanthine analog, tzXN. Further, we
showcase the potential of this fluorescent nucleobase surrogate to
provide a visible spectral window for a real-time study of GDA and its inhibition.
inhibitors, innovative methods for monitoring their activity
are needed. Although fluorescence spectroscopy might a priori
appear most attractive due to its high sensitivity, it cannot be
applied in such contexts due to the lack of any useful emissive
features of the native purine nucleobases.
INTRODUCTION
■
Nucleobase and nucleoside deaminases play key roles in
cellular pathways. Being responsible for the cellular levels of
purine- and pyrimidine-based heterocycles, their over (or
under) expression is frequently related to pathological
conditions.1−4 Additionally, differences between human and
bacterial enzymes provide intriguing opportunities for new
therapeutic approaches, exploiting their distinct fidelity and
inherent activities.1
Guanine deaminase (GDA), a hydrolytic zinc-based enzyme,
converts guanine to xanthine.5,6 Recent observations have
shown it impacts neuronal morphology7,8 and implicated GDA
in traumatic brain injury,9 memory dysfunction, and
psychiatric diseases.10 The study of human GDA, from the
aminohydrolase superfamily, has remained somewhat sparse
compared to other deaminases, including adenosine deaminase
(ADA)11,12 and cytidine deaminase (CDA),13 which convert
adenosine to inosine and cytidine to uridine, respectively. As a
result, real-time activity and inhibitory assays for GDA,
especially the human isoform, have yet to be advanced, and
its substrate scope remains somewhat elusive.14
Diverse methods have been utilized for assessing the
enzymatic activity and inhibition of deaminases. Most
protocols capable of directly measuring GDA activity rely on
the distinct chromatographic behavior or the UV absorption
signature of the substrate/product pair.15 The former
techniques are accurate but can be time-consuming, while
the latter tools frequently suffer from spectral interference as
many inhibitors share similar chromophoric properties with
the substrate or product. To facilitate the biochemical study of
deaminases, define their substrate scope, and identify
Over the past two decades our laboratory has designed,
synthesized, and implemented minimally perturbing and
responsive fluorescent nucleoside analogs.15−29 The guiding
criterion for their design has been the diminution of structural
and functional perturbations, which are inevitable consequen-
ces of modifying any native building block. Nucleosides that
fulfill such critical constraints are thus coined isomorphic. If
they function identically to their native counterparts, we define
them as being isofunctional. We have advanced two emissive
RNA alphabets, based on thieno[3,4-d]pyrimidine (1st
generation; thN)30 and isothiazole[4,3-d]pyrimidine (2nd
generation; tzN)31 heterocyclic cores (Figure 1a). The emissive
adenosine analogs have been successfully used to advance real-
time activity and inhibition assays of adenosine deaminase,
exploiting the distinct emission profiles of the substrates (thA
and tzA), and their corresponding deamination products (thI
and tzI, respectively), with the latter (tzA) being essentially
isofunctional (Figure 1b).32,33 The red-shifted absorption and
Received: March 29, 2021
Accepted: June 16, 2021
Published: June 30, 2021
https://doi.org/10.1021/acschembio.1c00232
© 2021 American Chemical Society
ACS Chem. Biol. 2021, 16, 1208−1214
1208

ACS Chemical Biology
pubs.acs.org/acschemicalbiology
Articles
Figure 1. a) Heterocycle evolution pertaining to this study starting
from the native purine core to thN to tzN. b) Adenosine deaminase
conversion of A to I and isofunctional nucleoside analog tzA to tzI. c)
Guanine deaminase conversion of G to X, isomorphic nucleobase
analog thGN to thXN, and isofunctional nucleobase analog tzGN to tzXN.
Figure 2. a) Structures of thGN, thXN, tzGN, and tzXN nucleobases. b)
Absorption (dashed lines) and emission (solid lines) of thGN (green),
thXN (red), tzGN (blue), and tzXN (orange) in water. c) Stokes shift
correlations versus solvent polarity [ET(30) of water/dioxane
mixtures] for thGN (green), thXN (red), tzGN (blue), and tzXN
(orange). d) Change in optical density versus pH for thGN (green),
thXN (red), tzGN (blue), and tzXN (orange).
visible-range emission of such fluorescent substrate surrogates
opens a spectral window not accessible by other means, which
facilitates the real-time monitoring of these reactions even in
the presence of potentially interfering chromophores.34
Applying such approaches to nucleobase-processing en-
zymes such as human GDA represents, however, a minimally
explored territory, as highly emissive nucleobase analogs have
not been broadly tested and scrutinized as substrates.6,13 The
lack of the rather large and contact forming D-ribose residue
presents the challenge of altering guanine without disturbing
contacts at the active site of such a potentially fastidious
nucleobase-processing enzyme.35 We therefore set out to
investigate two emissive nucleobase analogs, thieno[3,4-
d]pyrimidine (thGN) and isothiazole[4,3-d]pyrimidine (tzGN)
(Figure 1c), and demonstrate that while the former is not a
viable substrate, tzGN does undergo facile GDA-mediated
deamination to yield the fluorescently distinct xanthine analog,
tzXN. The spectral differences displayed by the substrate and
product are exploited for a real-time monitoring of the
For the isothiazole-based nucleobases tzGN and tzXN, methyl 4-
aminothiazole 3-carboxylate hydrochloride, a key precursor,
was prepared from methyl thioglycolate and N-tosylated
Oxyma, followed by decarboxylation and esterification
(Scheme S2).37 Conversion of the isothiazole-based precursor
to tzGN and tzXN followed the same cyclization procedures
described above for the thiophene derivatives (Scheme S3).
Crystal structures have been determined for all nucleobases
(Figure 3a−d, Tables S1−S4).
tz
enzymatic reaction. Moreover, the potential of GN as a tool
for GDA screenings is showcased by measuring GDA
inhibition by a previously reported inhibitor and identifying
two new inhibitors. Taken together, these findings highlight
the necessity of the N7 for substrate recognition by GDA and
provide information regarding its substrate and inhibitor scope.
RESULTS AND DISCUSSION
■
The four nucleobases, containing either a thienopyrimidine-
core, thGN and thXN, or an isothiazolepyrimidine-core, tzGN and
tzXN, were prepared according to established procedures
(Figure 2a, Schemes S1−S3).30,31,35,36 Briefly, treatment of
commercially available methyl 4-aminothiophene 3-carboxylate
hydrochloride with chloroformamidine hydrochloride in
DMSO2 at 125 °C yields thGN in one step. The corresponding
xanthine analog, thXN, was synthesized in two steps by treating
the same starting material with KOCN to yield the
corresponding urea, which was then cyclized under basic
conditions with sodium methoxide in methanol (Scheme S1).
Figure 3. X-ray crystal structures of nucleobases. a) thGN, b) thXN, c)
tzGN, and d) tzXN. MOE docking of e) xanthine as reference (ΔΔG 0),
f) tzXN (ΔΔG 0.28), and g) thXN (ΔΔG 1.89) in GDA active site
(PDB ID 2UZ9, see Methods).
1209
https://doi.org/10.1021/acschembio.1c00232
ACS Chem. Biol. 2021, 16, 1208−1214

ACS Chemical Biology
pubs.acs.org/acschemicalbiology
Articles
Table 1. Photophysical Data for Nucleobase Analogs
a
b
a
λabs,max
ε
λem,max
ϕ
thGN
thXN
tzGN
tzXN
315
308
320
315
3.0
3.2
5.4
6.1
439
420
446
394
0.40
0.46
0.07
0.02
a
b
λabs,max and λem,max are in nm. ε is in 103 M−1 cm−1. Measured in
triplicate in water.
To assess the likelihood of spectroscopically distinguishing
between the substrates and their corresponding products, the
fundamental spectroscopic properties of the nucleobases were
determined (Tables 1 and S5). Absorption spectra in water
displayed red-shifted maxima of the nucleobase analogs
compared to their native counterparts, showing maxima at
315, 308, 320, and 315 nm for thGN, thXN, tzGN, and tzXN,
respectively. Excitation at their absorption maxima gave rise to
visible emission, which peaked at 439, 420, 446, and 394 nm
for thGN, thXN, tzGN, and tzXN, respectively (Figure 2b). As
expected, the emission quantum yields of the thiophene
analogs, thGN and thXN (0.40 and 0.46, respectively) were
Figure 4. a) Conversion of guanine to xanthine and conversion of
isofunctional tzGN to tzXN by GDA. b) HPLC relative peak area versus
time for consumption of guanine to xanthine at 260 nm (blue) and
consumption of tzGN to tzXN at 320 nm (red). Inset: HPLC relative
peak area versus time for the consumption of guanine (blue) and
formation of xanthine (green) at 260 nm and the consumption of
tzGN (red) and formation of tzXN (orange) at 320 nm from 0 to 100 s.
c) Absorption changes upon enzymatic deamination of guanine to
xanthine at 270 nm (blue) and tzGN to tzXN at 355 nm (red). Inset:
Enzymatic deamination of tzGN to tzXN monitored by absorbance (at
355 nm, red), fluorescence (450 nm, excitation at 328 nm, green),
and HPLC relative peak area over time (at 320 nm, blue).
tz
tz
higher than their isothiazole counterparts, GN and XN (0.07
and 0.02, respectively.)
Additionally, the sensitivity of photophysical parameters to
environmental polarity was evaluated, and spectroscopically
derived pKa values were determined (Table S5). The
absorption and emission parameters of each nucleobase were
measured in dioxane, water, and mixtures thereof. By linearly
correlating the calculated Stokes shift against the solvent
polarity of each sample, the chromophore’s responsiveness was
defined. The four nucleobases showed differing levels of
sensitivity to polarity, with the thiophene derivatives showing
substantial impact on Stokes shift compared to the isothiazole
derivatives (Figure 2c, Figure S1). The nucleobases also
showed responsiveness to pH changes. pKa values were
extrapolated by plotting the change in optical density versus
deamination of guanine is comparable to previously published
figures for rabbit liver GDA (KM of 12.5 μM),40 and the k2/KM
is comparable to k1/[GDA] (G k1/[GDA] = 2.6 μM−1 s−1)
from the pseudo-first-order kinetic curves (Table 2).
th
pH (Figure 2d, Figure S2). The guanine derivatives, GN and
tzGN, show two pKa values (pKa = 4.41, 10.19 and pKa = 3.28,
8.96, respectively), while the xanthine derivatives, thXN and
tzXN, show one pKa value (pKa = 9.92 and pKa = 8.8,
respectively), as expected.31,38,39
Rewardingly, spectroscopic analyses agree with the chroma-
tographic analysis (Figure 4c). When monitoring the
enzymatic deamination of tzGN spectroscopically, an increase
in absorbance at 355 nm and an emission decrease at 450 nm
when exciting at 328 nm, the isosbestic point, were observed.
To compare the susceptibility of thGN and tzGN to GDA-
mediated deamination, enzymatic reactions were first analyzed
by HPLC and then monitored in real-time by absorption and
emission spectroscopy. HPLC analysis confirmed the complete
transformation of native guanine to xanthine by GDA within
500 s (Figure 4a and S3a). The reaction of tzGN with GDA
showed conversion to the corresponding product, tzXN
(Figures 4a and S3b). However, the reaction of thGN showed
no conversion of substrate across all time points studied
(Figure S3c). By plotting the area under the curve at different
reaction times (Figure 4b), comparable reaction half-life values
for guanine deamination (G t1/2 = 27 s) and tzGN deamination
(tzGN t1/2 = 21 s) were calculated assuming pseudo-first-order
reaction conditions (Table 2). Kinetic parameters were also
calculated by fitting HPLC-monitored deamination reactions
to a set of ordinary differential equations consistent with
Michaelis−Menten kinetics (see Methods). Similar values were
obtained for guanine (KM = 12 9 μM; k2 = 36 27 s−1; k2/
KM = 3.0 0.3 μM−1 s−1) and its isothiazolo analog tzGN (KM
= 8 3 μM; k2 = 32 7 s−1; k2/KM = 4.2 1.0 μM−1 s−1; see
Table 2). We note that the KM value obtained for the
tz
The reaction half-life for GN was calculated using the same
assumptions mentioned previously (tzGN abs t1/2 = 27 s; t1/2
em = 29 s) and was found to be comparable to the native
substrate (G abs t1/2 = 31 s), as shown in Table 2. The
tz
transformation of both native guanine to xanthine and GN to
tzXN by GDA is also illustrated by the absorption maxima shift
(Figure S4). Thienopyrimidine thGN showed no change in
absorption or emission in the presence of GDA over 500 s.
Reactions monitored by HPLC as well as real-time
absorption and emission spectroscopy suggest that tzGN, the
isothiazolepyrimidine substrate, is equally susceptible to GDA-
mediated deamination as native guanine. Intriguingly, thGN, the
thienopyrimidine guanine analog, is unreactive and does not
undergo GDA-mediated deamination. These findings showcase
the functionality and increased isomorphicity of the
isothiazolepyrimidine-based nucleobase alphabet, which rein-
stated the N7 moiety, compared to the thienopyrimidine-based
nucleobases, even without the large and contact forming D-
ribose residue the nucleoside counterparts possess. Further
support is obtained by Molecular Operating Environment
1210
https://doi.org/10.1021/acschembio.1c00232
ACS Chem. Biol. 2021, 16, 1208−1214

ACS Chemical Biology
pubs.acs.org/acschemicalbiology
Articles
Table 2. GDA Reaction Rates
HPLC
absorbance
emission
a
b
c
d
d
d
a
b
c
a
f
b
c
λmntr
k1
26
NR
32
t1/2
KM
12
NR
k2
k2/KM
λmntr
k1
22
NR
25
t1/2
λmntr
k1
t1/2
G to X
260
320
260
2
1
27
NR
21
9
36 27
NR
3.0 0.3
NR
270
355
355
1
1
31
NR
27
e
thGN to thXN
tzGN to tzXN
450
NR
NR
29
8
3
32
7
4.2 1.0
450
24
1
a
b
λmntr reported in nm and represents wavelength monitored. Pseudo-first-order reaction kinetics slope of the exponential approximation in 10−3
c
d
s−1. t1/2 is reaction half-life calculated assuming pseudo-first-order kinetics. KM, k2, and k2/KM are reported in μM, s−1, and 106 M−1 s−1,
e
f
respectively. λexc at 360 nm for thGN reactions. λexc at 328 nm for tzGN reactions. Experiments done in triplicate.
(MOE) molecular docking experiments (Figures 3e−g, Figures
S5 and S6) using the published crystallographic data for
xanthine-bound human GDA (PDB ID 2UZ9). The enzyme
recognition preferences for tzXN over thXN are illustrated by the
markedly higher positive ΔΔG values obtained for the latter vs
the former (ΔΔG 1.89 and 0.28, respectively). A similar trend
was observed for the substrates (Figure S6). Docking also
shows a plausible structure deformation induced by Arg235,
which is projected toward the N7 position at the native
substrate. This basic site, present in native G/X and the
isothiazolo analogs, tzGN/tzXN, likely hydrogen bonds to
Arg235. Clashing of this arginine side chain with the CH
group of thGN at the same position likely renders the
Figure 5. a) GDA inhibitor structures of AICA, ATCA, and ICA. b)
thiophenopyrimidine-based substrate unrecognizable by GDA.
As stated, the presence of the nitrogen at a position
equivalent to N7 in the purine skeleton endows tzGN, the
isothiazole purine surrogate, with substrate recognition
features that appear lacking in thGN. While comprehensive
investigations of the GDA substrate scope have not been
pursued, previous substrate and inhibitor analyses of amino-
hydrolase isoforms suggest that the O6, N3, and N7 points of
contact are important for substrate recognition.6,17,40,41
Intriguingly, while tzGN retains these contact points, it is
lacking the NH group found in the purine’s 9 position.
Previously reported inhibitors, such as valciclovir and
derivatives of azepinomycin, containing substituents at the
N9 position suggest GDA can tolerate diverse groups at that
position, which may explain the high tolerance and the native
Overlay of % inhibition at 100 s and sigmoidal Hill plot for AICA
(black), ATCA (red), and ICA (blue).
tz
tz
sion of GN to XN can successfully be tracked in real-time as
the reaction substrate and product display distinct photo-
physical properties. The strategic replacement of the GDA
substrate with emissive counterparts from both previously
published emissive alphabets demonstrate the significance of
N7 for GDA substrate recognition. Moreover, this discovery
provides a biochemical tool to study the activity of human
GDA and offers a spectral window for fabricating real-time
high throughput discovery assays of GDA. We illustrate this
through a small pilot study of inhibitor effects on GDA activity,
which includes one previously reported inhibitor and two new
inhibitors.
14,42
tz
deamination rate displayed by GN.
The change in emission signal intensity upon GDA-
mediated deamination of tzGN to tzXN facilitates real-time
monitoring of the enzymatic reaction and its inhibition. To
demonstrate its potential, three inhibitors were tested, 5-
aminoimidazole-4-carboxamide (AICA), 4-aminoisothiazole-3-
carboxamide (ATCA), and 4-imidazolecarboxylic acid (ICA,
Figure 5a). AICA is an established competitive inhibitor of
GDA.40 We synthesized ATCA, the isothiazole counterpart
(Scheme S4). Reaction conditions used for previous experi-
ments were sustained with the exception of inhibitor addition
at various concentrations. Percent inhibition plots were fitted
with sigmoidal Hill curves to obtain IC50 values of the tested
inhibitors. AICA (IC50 = 100 μM) showed comparable
inhibition to ATCA (IC50 = 80 μM), while ICA is
comparatively a less potent inhibitor (IC50 = 2 mM, Figure
5b).
METHODS
■
Expression and Purification of Human GDA. Wild-type human
GDA gene, cloned into pET-28B vector (Genescript), was expressed
with an N-terminal hexahistidine tag. The construct was sequence-
verified (GENEWIZ). The plasmid (4 μg) was dissolved in water to a
concentration of 28.8 ng/μL. E. coli BL21 (DE3) cells were
transformed with the GDA containing plasmid. Overnight starter
cultures were grown in Lubria Broth (LB) medium at 37 °C with 50
μg/mL kanamycin. The overnight culture was used to inoculate 1 L of
LB medium with 50 μg/mL kanamycin. Cells were grown while
shaking at 37 °C until an OD600 of 0.6 was reached, and the culture
was down tempered to 18 °C over a period of 1 h. To induce target
protein production, 0.5 mM IPTG was added, and the medium was
left to shake at 18 °C overnight. Cells were harvested after overnight
growth by centrifugation at 3500 rpm for 35 min at 4 °C.
Cells were resuspended in lysis buffer [20 mM HEPES, 500 mM
NaCl, 10% glycerol, 0.5 mM TCEP (pH 7.5), supplemented with one
tablet of Complete EDTA-free protease inhibitor (Roche Applied
Science)] and lysed by sonication. Cell lysate was centrifuged at
10,000 rpm for 1 h at 4 °C before the supernatant was decanted and
filtered through a 0.45 μm syringe filter. GDA was purified by
immobilized metal affinity chromatography. An IMAC Ni-charged
column [2 mL HisPur Ni-NTA Resin (Thermo Scientific)] was
equilibrated with lysis buffer to which the lysate was then added. The
column was washed with 50 mL of wash buffer supplemented with 20
CONCLUSION
■
This study set out to assess the utility of two emissive
nucleobase analogs, thGN and tzGN, as substrates for the
enzymatic reaction of GDA. We demonstrate that isothiazole-
[4,3-d]pyrimidine, tzGN, behaves as an isomorphic and
isofunctional emissive surrogate for guanine. The interconver-
1211
https://doi.org/10.1021/acschembio.1c00232
ACS Chem. Biol. 2021, 16, 1208−1214

ACS Chemical Biology
pubs.acs.org/acschemicalbiology
Articles
mM imidazole. The bound protein was eluted from the column with
elution buffer (wash buffer supplemented with 200 mM imidazole).
Fractions were analyzed by SDS-PAGE, and those containing the
target protein were pooled, subsequently concentrated, and buffer
exchanged into storage buffer [20 mM Tris-HCl buffer (pH 8.0), 10%
glycerol, 1 mM DTT] using a centrifugal filter device with a 10 kDa
molecular weight cutoff. After protein expression and purification, the
final protein concentration was 5.3 mg mL−1 (0.1 mM) in a total
volume of 1.7 mL. Protein aliquots were snap-frozen and stored at
−80 °C until further use. Before being used in enzymatic reactions,
the GDA protein stock was diluted 1:100 in sodium phosphate buffer
[50 mM (pH 7.4)].
Synthetic Procedures. All synthetic procedures and character-
izations of compounds are reported in the Supporting Information.
Photophysical Properties: General Methods. Absorption
spectra were measured on a Shimadzu UV-2450 spectrometer with
0.5 mm resolution and setting the slit at 1 nm. Emission spectra were
measured on a Horiba Fluoromax-4 equipped with a cuvette holder
and stirring system. Emission measurements were taken with
resolution at 1 nm and setting both the excitation and emission
slits at 3 nm.
with sodium hydroxide. Samples were prepared in a 10.00 mm four-
sided quartz cuvette from Helma. Reactions had a total reaction
volume of 3 mL with nucleobase and enzyme concentrations of 3 μM
and 10 nM in sodium phosphate buffer [50 mM (pH 7.4)]. All
measurements were taken at 37 °C, and GDA was introduced after a 3
min temperature equilibration period.
Real-Time Monitoring of GDA Reactions via Absorption
and Emission. GDA-mediated conversion of guanine and its analogs
(thGN and tzGN) was monitored by absorption and emission
spectroscopy. Absorbance measurements of the enzymatic conversion
of guanine (and its analogs) were performed on a Shimadzu UV-2459
spectrometer taking a point every 10 s for 600 s after the addition of
GDA with a slit setting of 1 nm. The conversion was monitored at
270 nm for guanine to xanthine and at 355 nm for the analogs (thGN
and tzGN to tzXN). Emission measurements of the conversion of
guanine (and analogs) were performed on a Horiba Fluoromax-4 with
a cuvette holder and a built-in stirring system with excitation and
emission slits set to 3 nm and taking points every 10 s for 600 s after
the addition of GDA.
Emission was monitored at 450 nm with excitation at 360 and 328
nm, respectively, for thGN and tzGN. The thGN reaction showed no
change in absorption or emission intensity in the presence of GDA
over 600 s. Each experiment was done in triplicate. There is a 6 s lag
time after GDA is added to the cuvette after the time 0 measurement.
Steady State Absorption Measurements in the Presence of
GDA. Steady state absorption spectra over time were performed on a
Shimadzu UV-2450 spectrophotometer setting the slit at 5 nm, using
a resolution of 0.5 nm, taking a measurement every 20 s. All spectra
were corrected for the blank. Smoothing of data was done on spectra
for plotting.
HPLC Analysis of Enzymatic Conversion of Native G to X
and of tzGN to tzXN. GDA-mediated conversion was monitored by
chromatography. HPLC was carried out with an Agilent 1200 system
with a Polaris 5 C18-A 250 × 4.9 mm column. Solutions of 0.1%
formic acid (Honeywell Fluka) were prepared by dissolving 1 mL of
formic acid in 1 L total volume of acetonitrile (J. T. Baker) or water.
Solutions were filtered using Millipore type GNWP 0.2 μm filters
before use. Each injection (75 μL) for guanine or tzGN experiments
was subjected to a linear gradient of 0.5% to 10% acetonitrile in water
with 0.1% formic acid for 20 min, followed by a flush and
equilibration for 10 min. thGN injections were subjected to a linear
gradient of 0.5% to 10% acetonitrile in water with 0.1% formic acid for
30 min, followed by a flush and equilibration for 10 min. Each run was
monitored at 260, 280, and 320 nm with a calibrated reference at 650
nm and slit set at 1 nm.
All measurements were carried out in a 10 mm four-sided quartz
cuvette purchased from Helma. All spectra were corrected for the
blank. Both instruments were equipped with a thermostat-controlled
ethylene glycol-water bath, and all measurements were taken at 37 °C.
Measurements were recorded after a 3 min temperature equilibration
period.
Concentrated stock solutions for xanthine, thGN, thXN, tzGN, and
tzXN were prepared in DMSO, and a stock solution of guanine was
prepared in water basified to pH 12 with sodium hydroxide. Samples
for experiments were prepared with stock nucleobase diluted to a total
sample volume of 3 mL in deionized water, mixed with a pipet for 10
s, and placed in the cuvette holder. All samples contain 0.3 v/v %
DMSO, except guanine samples.
Quantum Yield Measurements. All sample concentrations were
adjusted to optical density lower than 0.1 at the excitation wavelength
(λex). The fluorescence quantum yield (ϕ) of each nucleobase was
evaluated based on 2-aminopurine (0.68 in water, λex 320 nm) as an
external standard by using the following equation
ODSTD
ISTD OD
I
n2
ϕ = ϕSTD
nS2TD
(1)
where ϕSTD is the fluorescence quantum yield of the standard, I and
ISTD are the integrated area of the emission band of the sample and
the standard, respectively, OD and ODSTD are the optical density at
the excitation wavelength for the sample and standard, respectively,
and n and nSTD are the solvent refractive index of the sample and the
standard solutions, respectively.
Sensitivity to pH. Sodium phosphate buffers with a final
concentration of 50 mM were prepared and adjusted to the desired
experimental pH values using HCl or NaOH prior to spectral
measurements. Changes in optical density, at 310 and 350 nm, were
plotted versus pH. The pKa values were determined by interpolation
of the fitting curves.
Sensitivity to Polarity. Experiments evaluating the effect of
polarity were performed in water, dioxane, and mixtures of 20, 40, 60,
and 80 v/v % water in dioxane. The sample ET(30) values were
determined by dissolving Reichardt’s dye in an aliquot of the same
solvent used to dilute the nucleobase DMSO sample. The observed
wavelength absorption maximum (λmabasx) was then converted to the
ET(30) values (Table S6) using the following equation
A concentrated stock solution of guanine, thGN, and tzGN was
diluted in phosphate buffer. The solution was warmed to 37 °C for 3
min with stirring before addition of a GDA stock solution. After the
addition of GDA, the enzymatic conversion was quenched in aliquots
(after 20, 40, 60, 80, 100, 200, and 500 s) by adding formic acid (0.55
M) and placing the aliquots on ice. Each 100 μL aliquot was filtered
and analyzed by HPLC.
HPLC traces were corrected for the blank, and the relative areas
were plotted as a function of time. Trend lines represent a loss of
substrate and product apparition over time for a pseudo-first-order
kinetic reaction.
A set of ordinary differential equations (ODEs) consistent with
Michaelis−Menten kinetics (eqs 3−6) was solved using the Runge−
Kutta method with a method with a variable time step in MatLab
(function ode45). Initial concentrations used for enzyme and
substrate were 10 nM and 3 μM, respectively. The product and
enzyme substrate complex were assumed to have initial concen-
trations of 0 μM. The resulting fitted curves for each species were
optimized by iteratively testing k values that maximized R.2 This
yielded k1, k−1, and k2 values from which KM and k2/KM values were
derived.
28591
ET(30) =
max
λ
(2)
abs
Enzymatic Deamination: General Methods. Reaction con-
ditions were the same for all GDA reactions monitored by
spectroscopy or chromatography. Concentrated stock solutions for
xanthine, thGN, thXN, tzGN, and tzXN were prepared in DMSO, and the
stock solution of guanine was prepared in water basified to pH 12
d[E]
dt
= −k1[E][S] + k−1[ES] + k2[ES]
(3)
1212
https://doi.org/10.1021/acschembio.1c00232
ACS Chem. Biol. 2021, 16, 1208−1214

ACS Chemical Biology
pubs.acs.org/acschemicalbiology
Articles
d[S]
Paul T. Ludford, III − Department of Chemistry and
= −k1[E][S] + k−1[ES]
dt
(4)
(5)
(6)
Biochemistry, University of California, San Diego, La Jolla,
California 92093, United States
d[ES]
dt
Yao Li − Department of Chemistry and Biochemistry,
University of California, San Diego, La Jolla, California
92093, United States
Andrea Fin − Department of Chemistry and Biochemistry,
University of California, San Diego, La Jolla, California
92093, United States; Present Address: Dipartimento di
Scienza e Tecnologia del Farmaco, University of Turin,
Via P. Giuria, 9, 10125 Turin, Italy.; orcid.org/0000-
0002-7567-4646
= k1[E][S] − k−1[ES] − k2[ES]
d[P]
dt
= k2[ES]
Inhibition Studies. Fluorescence real-time monitoring of the
enzymatic conversion of tzGN to tzXN with GDA in the presence of
inhibitors was followed by emission spectroscopy by monitoring the
intensity signal at 450 nm with excitation at 328 nm. 5-Amino-
imidazole-4-carboxamide (AICA) was tested as an inhibitor in the
enzymatic reaction for tzGN to tzXN using the same reaction conditions
but supplemented with inhibitor concentrations of 0.1, 1, 10, 100, and
1000 μM. 4-Aminoisothiazole-3-carboxamide (ATCA) was also tested
as an inhibitor in the reaction under the same conditions but
supplemented with inhibitor concentrations of 0.1, 1, 10, 100, 250,
and 500 μM. 4-Imidazolecarboxylic acid (ICA) was also tested as a
potential inhibitor under the same reaction conditions with 10, 100,
1000, 2000, and 3000 μM concentrations. IC50 values were
determined from plots of percent inhibition against respective
inhibitor concentration on a logarithmic axis, fitted with sigmoidal
Hill curves. All measurements were done in triplicate.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acschembio.1c00232
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
Docking of Guanine, Xanthine, and Analogs in GDA.
Docking simulations were performed with the Molecular Operating
Environment (MOE) 2020 software suite. The crystal structure of
human guanine deaminase (guaD) in complex with zinc and its
product xanthine was obtained from the Protein Data Bank (PDB ID
2UZ9). The crystal structure was chosen because it contains the
native product xanthine. The chosen ligands were placed by the
Triangle Matcher method and ranked with the London dG values
scoring function. A total of 30 poses were refined using the rigid
receptor method. Then five poses were rescored using a GBVI/WSA
dG scoring function. The pose with the ligand most closely aligned
with the xanthine ligand was used for comparison, and the ΔΔG value
was calculated for tzGN and thGN with native guanine as the reference
and for tzXN and thXN with native xanthine as the reference. The
equation below was used where ΔGref is the GBV1/WSA dG scoring
of the reference native ligand and ΔGlig is the GBVI/WSA dG scoring
of the ligand analogs (eq 7).
ACKNOWLEDGMENTS
■
We thank the National Institutes of Health (through grant no.
GM 069773) for generous support, the Chemistry &
Biochemistry MS facility, and the UCSD X-ray Crystallography
Facility (especially M. Gembicki and C. Moore). We also
thank the Burkhart Lab at UCSD for advice on protein
expression and purification and R. Anand for helpful
discussions and insight.
REFERENCES
■
(1) Gaded, V., and Anand, R. (2018) Nucleobase Deaminases: A
Potential Enzyme System for New Therapies. RSC Adv. 8 (42),
23567−23577.
(2) Wang, J., Bing, T., Zhang, N., Shen, L., He, J., Liu, X., Wang, L.,
and Shangguan, D. (2019) The Mechanism of the Selective
Antiproliferation Effect of Guanine-Based Biomolecules and Its
Compensation. ACS Chem. Biol. 14 (6), 1164−1173.
ΔΔG = ΔGlig − ΔGref
(7)
(3) Slater, P. G., Cammarata, G. M., Monahan, C., Bowers, J. T.,
Yan, O., Lee, S., and Lowery, L. A. (2019) Characterization of
Xenopus Laevis Guanine Deaminase Reveals New Insights for Its
Expression and Function in the Embryonic Kidney. Dev. Dyn. 248 (4),
296−305.
(4) Jung, J. M., Noh, T. K., Jo, S. Y., Kim, S. Y., Song, Y., Kim, Y.-H.,
and Chang, S. E. (2020) Guanine Deaminase in Human Epidermal
Keratinocytes Contributes to Skin Pigmentation. Molecules 25 (11),
2637.
(5) Paletzki, R. F. (2002) Cloning and Characterization of Guanine
Deaminase from Mouse and Rat Brain. Neuroscience 109 (1), 15−26.
(6) Shek, R., Hilaire, T., Sim, J., and French, J. B. (2019) Structural
Determinants for Substrate Selectivity in Guanine Deaminase
Enzymes of the Amidohydrolase Superfamily. Biochemistry 58 (30),
3280−3292.
(7) Akum, B. F., Chen, M., Gunderson, S. I., Riefler, G. M., Scerri-
Hansen, M. M., and Firestein, B. L. (2004) Cypin Regulates Dendrite
Patterning in Hippocampal Neurons by Promoting Microtubule
Assembly. Nat. Neurosci. 7 (2), 145−152.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acschembio.1c00232.
■
sı
Synthetic schemes, experimental methods, photophysical
data, spectra, and X-ray crystallographic data for thGN,
tz
tz
thXN, GN, and XN (PDF)
AUTHOR INFORMATION
Corresponding Author
Yitzhak Tor − Department of Chemistry and Biochemistry,
University of California, San Diego, La Jolla, California
92093, United States; Email: ytor@ucsd.edu
■
Authors
(8) Rodríguez, A. R., O’Neill, K. M., Swiatkowski, P., Patel, M. V.,
and Firestein, B. L. (2018) Overexpression of Cypin Alters Dendrite
Morphology, Single Neuron Activity, and Network Properties via
Distinct Mechanisms. J. Neural Eng. 15 (1), No. 016020.
(9) Swiatkowski, P., Sewell, E., Sweet, E. S., Dickson, S., Swanson, R.
A., McEwan, S. A., Cuccolo, N., McDonnell, M. E., Patel, M. V.,
Varghese, N., Morrison, B., Reitz, A. B., Meaney, D. F., and Firestein,
Marcela S. Bucardo − Department of Chemistry and
Biochemistry, University of California, San Diego, La Jolla,
California 92093, United States
You Wu − Department of Chemistry and Biochemistry,
University of California, San Diego, La Jolla, California
92093, United States
1213
https://doi.org/10.1021/acschembio.1c00232
ACS Chem. Biol. 2021, 16, 1208−1214

ACS Chemical Biology
pubs.acs.org/acschemicalbiology
Articles
B. L. (2018) Cypin: A Novel Target for Traumatic Brain Injury.
(31) Rovira, A. R., Fin, A., and Tor, Y. (2015) Chemical
Mutagenesis of an Emissive RNA Alphabet. J. Am. Chem. Soc. 137
(46), 14602−14605.
Neurobiol. Dis. 119, 13−25.
̀
(10) Zuccarini, M., Giuliani, P., Frinchi, M., Mudo, G., Serio, R. M.,
(32) Sinkeldam, R. W., McCoy, L. S., Shin, D., and Tor, Y. (2013)
Enzymatic Interconversion of Isomorphic Fluorescent Nucleosides:
Adenosine Deaminase Transforms an Adenosine Analogue into an
Inosine Analogue. Angew. Chem., Int. Ed. 52 (52), 14026−14030.
(33) Ludford, P. T., Rovira, A. R., Fin, A., and Tor, Y. (2019)
Fluorescing Isofunctional Ribonucleosides: Assessing Adenosine
Deaminase Activity and Inhibition. ChemBioChem 20 (5), 718−726.
(34) Adamek, R. N., Ludford, P., Duggan, S. M., Tor, Y., and Cohen,
S. M. (2020) Identification of Adenosine Deaminase Inhibitors by
Metal-Binding Pharmacophore Screening. ChemMedChem 15 (22),
2151−2156.
(35) Srivatsan, S. G., Weizman, H., and Tor, Y. (2008) A Highly
Fluorescent Nucleoside Analog Based on Thieno[3,4-d]Pyrimidine
Senses Mismatched Pairing. Org. Biomol. Chem. 6 (8), 1334−1338.
(36) Rovira, A. R., Fin, A., and Tor, Y. (2017) Expanding a
Fluorescent RNA Alphabet: Synthesis, Photophysics and Utility of
Isothiazole-Derived Purine Nucleoside Surrogates. Chem. Sci. 8 (4),
2983−2993.
(37) Gewald, K., and Bellmann, P. (1979) Synthese und Reaktionen
von 4-Aminoisothiazolen. Liebigs Ann. Chem. 1979 (10), 1534−1546.
(38) Rogstad, K. N., Jang, Y. H., Sowers, L. C., and Goddard, W. A.
(2003) First Principles Calculations of the PKa Values and Tautomers
of Isoguanine and Xanthine. Chem. Res. Toxicol. 16 (11), 1455−1462.
(39) Didier, P., Kuchlyan, J., Martinez-Fernandez, L., Gosset, P.,
Léonard, J., Tor, Y., Improta, R., and Mély, Y. (2020) Deciphering the
PH-Dependence of Ground- and Excited-State Equilibria of
Thienoguanine. Phys. Chem. Chem. Phys. 22 (14), 7381−7391.
(40) Lewis, A. S., and Glantz, M. D. (1974) Rabbit Liver Guanine
Deaminase. Chemical, Physical, and Kinetic Properties. J. Biol. Chem.
249 (12), 3862−3866.
Belluardo, N., Buccella, S., Carluccio, M., Condorelli, D. F., Caciagli,
F., Ciccarelli, R., and Di Iorio, P. (2018) Uncovering the Signaling
Pathway behind Extracellular Guanine-Induced Activation of NO
System: New Perspectives in Memory-Related Disorders. Front.
Pharmacol. 9, 110.
(11) Wilson, D. K., Rudolph, F. B., and Quiocho, F. A. (1991)
Atomic Structure of Adenosine Deaminase Complexed with a
Transition-State Analog: Understanding Catalysis and Immunodefi-
ciency Mutations. Science 252 (5010), 1278−1284.
(12) Kinoshita, T., Nakanishi, I., Terasaka, T., Kuno, M., Seki, N.,
Warizaya, M., Matsumura, H., Inoue, T., Takano, K., Adachi, H.,
Mori, Y., and Fujii, T. (2005) Structural Basis of Compound
Recognition by Adenosine Deaminase. Biochemistry 44 (31), 10562−
10569.
(13) Bitra, A., Biswas, A., and Anand, R. (2013) Structural Basis of
the Substrate Specificity of Cytidine Deaminase Superfamily Guanine
Deaminase. Biochemistry 52 (45), 8106−8114.
(14) Egeblad, L., Welin, M., Flodin, S., Gräslund, S., Wang, L.,
Balzarini, J., Eriksson, S., and Nordlund, P. (2012) Pan-Pathway
Based Interaction Profiling of FDA-Approved Nucleoside and
Nucleobase Analogs with Enzymes of the Human Nucleotide
Metabolism. PLoS One 7 (5), e37724.
(15) Wojczewski, C., Stolze, K., and Engels, J. W. (1999)
Fluorescent Oligonucleotides - Versatile Tools as Probes and Primers
for DNA and RNA Analysis. Synlett 1999 (10), 1667−1678.
(16) Hawkins, M. (2001) Fluorescent Pteridine Nucleoside Analogs:
A Window on DNA Interactions. Cell Biochem. Biophys. 34, 257−281.
(17) Rist, M. J., and Marino, J. P. (2002) Fluorescent Nucleotide
Base Analogs as Probes of Nucleic Acid Structure, Dynamics and
Interactions. Curr. Org. Chem. 6 (9), 775−793.
(18) Millar, D. P. (1996) Fluorescence Studies of DNA and RNA
Structure and Dynamics. Curr. Opin. Struct. Biol. 6 (3), 322−326.
(19) Okamoto, A., Saito, Y., and Saito, I. (2005) Design of Base-
Discriminating Fluorescent Nucleosides. J. Photochem. Photobiol., C 6,
108−122.
(20) Ranasinghe, R. T., and Brown, T. (2005) Fluorescence Based
Strategies for Genetic Analysis. Chem. Commun. 44, 5487−5502.
(21) Silverman, A. P., and Kool, E. T. (2006) Detecting RNA and
DNA with Templated Chemical Reactions. Chem. Rev. 106 (9),
3775−3789.
(41) Bergstrom, J. D., and Bieber, A. L. (1979) Characterization of
Purified Guanine Aminohydrolase. Arch. Biochem. Biophys. 194 (1),
107−116.
(42) Chakraborty, S., Shah, N. H., Fishbein, J. C., and Hosmane, R.
S. (2011) A Novel Transition State Analogue Inhibitor of Guanase
Based on Azepinomycin Ring Structure: Synthesis and Biochemical
Assessment of Enzyme Inhibition. Bioorg. Med. Chem. Lett. 21 (2),
756−759.
(22) Wilson, J. N., and Kool, E. T. (2006) Fluorescent DNA Base
Replacements: Reporters and Sensors for Biological Systems. Org.
Biomol. Chem. 4 (23), 4265−4274.
(23) Dodd, D., and Hudson, R. (2009) Intrinsically Fluorescent
Base-Discriminating Nucleoside Analogs. Mini-Rev. Org. Chem. 6,
378−391.
(24) Wilhelmsson, L. M. (2010) Fluorescent Nucleic Acid Base
Analogues. Q. Rev. Biophys. 43 (2), 159−183.
(25) Wierzchowski, J., Antosiewicz, J. M., and Shugar, D. (2014) 8-
Azapurines as Isosteric Purine Fluorescent Probes for Nucleic Acid
and Enzymatic Research. Mol. BioSyst. 10 (11), 2756−2774.
(26) Jones, A. C., and Neely, R. K. (2015) 2-Aminopurine as a
Fluorescent Probe of DNA Conformation and the DNA-Enzyme
Interface. Q. Rev. Biophys. 48 (2), 244−279.
(27) Wilhelmsson, M. Fluorescent Analogs of Biomolecular Building
Blocks: Design and Applications; 2016; DOI: 10.1002/9781119179320.
(28) Xu, W., Chan, K. M., and Kool, E. T. (2017) Fluorescent
Nucleobases as Tools for Studying DNA and RNA. Nat. Chem. 9
(11), 1043−1055.
(29) Saito, Y., and Hudson, R. H. E. (2018) Base-Modified
Fluorescent Purine Nucleosides and Nucleotides for Use in
Oligonucleotide Probes. J. Photochem. Photobiol., C 36, 48−73.
(30) Shin, D., Sinkeldam, R. W., and Tor, Y. (2011) Emissive RNA
Alphabet. J. Am. Chem. Soc. 133 (38), 14912−14915.
1214
https://doi.org/10.1021/acschembio.1c00232
ACS Chem. Biol. 2021, 16, 1208−1214