Design of novel potent inhibitors of human uridine phosphorylase-1: Synthesis, inhibition studies, thermodynamics, and in vitro influence on 5-fluorouracil cytotoxicity

Article abstract of DOI:10.1021/jm401389u

Uridine (Urd) is a promising biochemical modulator to reduce host toxicity caused by 5-fluorouracil (5-FU) without impairing its antitumor activity. Elevated doses of Urd are required to achieve a protective effect against 5-FU toxicity, but exogenous administration of Urd is not well-tolerated. Selective inhibitors of human uridine phosphorylase (hUP) have been proposed as a strategy to increase Urd levels. We describe synthesis and characterization of a new class of ligands that inhibit hUP type 1 (hUP1). The design of ligands was based on a possible S_N1 catalytic mechanism and as mimics of the carbocation in the transition state of hUP1. The kinetic and thermodynamic profiles showed that the ligands here presented are the most potent in vitro hUP1 inhibitors developed to date. In addition, a lead compound improved the antiproliferative effects of 5-FU on colon cancer cells, accompanied by a reduction of in vitro 5-FU cytotoxicity in aggressive SW-620 cancer cells.

Full text of DOI:10.1021/jm401389u

Article
pubs.acs.org/jmc
Design of Novel Potent Inhibitors of Human Uridine Phosphorylase-
1: Synthesis, Inhibition Studies, Thermodynamics, and in Vitro
Influence on 5‑Fluorouracil Cytotoxicity
Daiana Renck,†,‡ Pablo Machado,†,‡ Andre A. Souto,‡,§ Leonardo A. Rosado,†,‡ Thais Erig,∥,⊥
Maria M. Campos,‡,∥,⊥ Caroline B. Farias,# Rafael Roesler,#,▽ Luis F. S. M. Timmers,‡,○
Osmar N. de Souza,‡,○ Diogenes S. Santos,*,†,‡ and Luiz A. Basso*,†,‡
†Centro de Pesquisas em Biologia Molecular e Funcional (CPBMF), Pontifícia Universidade Catol
(PUCRS), 6681/92-A, TecnoPuc, Av. Ipiranga, 90619-900 Porto Alegre, Rio Grande do Sul, Brazil
́
ica do Rio Grande do Sul
‡Programa de Pos
́
-Graduaca
̧
o em Biologia Celular e Molecular, Pontifícia Universidade Catol
́
ica do Rio Grande do Sul (PUCRS),
̃
90619-900 Porto Alegre, Rio Grande do Sul, Brazil
§Faculdade de Química, Pontifícia Universidade Catol
Alegre, Rio Grande do Sul, Brazil
́
ica do Rio Grande do Sul (PUCRS), 6681/P12 Av. Ipiranga, 90619-900 Porto
∥Instituto de Toxicologia e Farmacologia, Pontifícia Universidade Catol
́
ica do Rio Grande do Sul (PUCRS), 6681/P12 Av. Ipiranga,
90619-900 Porto Alegre, Rio Grande do Sul, Brazil
⊥Faculdade de Odontologia, Pontifícia Universidade Catol
Porto Alegre, Rio Grande do Sul, Brazil
́
ica do Rio Grande do Sul (PUCRS), 6681/P6 Av. Ipiranga, 90619-900
#Laborator
Rio Grande do Sul, Brazil
▽Laborator
io de Neurofarmacologia e Biologia de Tumor Neuronal, Departamento de Farmaciologia, Instituto de Cien
da Saude, Universidade Federal do Rio Grande do Sul (UFRGS), 90040-060 Porto Alegre, Rio Grande do Sul, Brazil
́ ̂
io de Pesquisa em Cancer, CPE-HCPA, Universidade Federal do Rio Grande do Sul (UFRGS), 90040-060 Porto Alegre,
́
̂ ́
cias Basicas
́
○Laborator
Universidade Catol
́
io de Bioinformat
́ ́
ica, Modelagem e Simulaca̧ o de Biossistemas (LABIO), Faculdade de Informatica, Pontifícia
̃
ica do Rio Grande do Sul (PUCRS), 6681/P32 Av. Ipiranga, 90619-900 Porto Alegre, Rio Grande do Sul, Brazil
́
S
* Supporting Information
ABSTRACT: Uridine (Urd) is a promising biochemical
modulator to reduce host toxicity caused by 5-fluorouracil
(5-FU) without impairing its antitumor activity. Elevated doses
of Urd are required to achieve a protective effect against 5-FU
toxicity, but exogenous administration of Urd is not well-
tolerated. Selective inhibitors of human uridine phosphorylase
(hUP) have been proposed as a strategy to increase Urd levels.
We describe synthesis and characterization of a new class of
ligands that inhibit hUP type 1 (hUP1). The design of ligands
was based on a possible SN1 catalytic mechanism and as
mimics of the carbocation in the transition state of hUP1. The
kinetic and thermodynamic profiles showed that the ligands here presented are the most potent in vitro hUP1 inhibitors
developed to date. In addition, a lead compound improved the antiproliferative effects of 5-FU on colon cancer cells,
accompanied by a reduction of in vitro 5-FU cytotoxicity in aggressive SW-620 cancer cells.
clinical setting to treat solid tumors such as breast, colorectal,
and head and neck cancers.12,13 5-FU requires metabolic
activation and, as a uracil analogue, it is processed by the same
enzymes involved in pyrimidine salvage pathway.13,14 The
active metabolites act through two main pathways: by forming a
ternary complex with thymidylate synthase (TS) leading to
inhibition of DNA synthesis and interference with DNA repair
INTRODUCTION
■
Uridine (Urd) is a natural pyrimidine nucleoside that is
involved, directly or indirectly, in many cellular processes
including RNA and biomembrane synthesis,1 galactose
metabolism,2,3 modulation of reproduction,4 regulation of
body temperature,5 and peripheral and central nervous systems
activities.6,7 Furthermore, Urd has been described for decades
as a promising biochemical modulator of 5-fluorouracil (5-FU)
metabolism.8−10 5-FU was synthesized in 1957 by Charles
Heidelberger,11 and since then it has been broadly used in the
Received: September 9, 2013
Published: October 16, 2013
© 2013 American Chemical Society
8892
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
and by incorporating fluorouridine 5′-triphosphate (FUTP)
into RNA.13−16 It has been hypothesized that the therapeutic
activity of 5-FU is due to DNA damage, whereas the toxic side
effects are related to RNA damage by FUTP incorpora-
tion.17−19 In this context, there has been interest in the use of
Urd as a biochemical modulator to counteract the host toxicity
caused by chemotherapy with fluoropyrimidines, as 5-FU,
thereby enabling the increase in doses of these drugs. High
levels of Urd leads to an increase in UTP levels, which will
compete with FUTP for RNA incorporation.15,18,20 However,
sustainable high plasma concentration is required for Urd to
exert a protective effect against 5-FU toxicity. Unfortunately,
this approach is precluded by dose-limiting toxic side effects
and the rapidly restored levels when either oral or intravenous
infusions are administrated.15,21 Two pro-drugs of Urd were
synthesized to surmount this barrier, but the strategy failed due
to the low potency and unfavorable pharmacokinetic profile.22
The half-life of Urd in plasma is only 2 min, with more than
80% metabolized in liver by uridine phosphorylase (hUP, EC
2.4.2.3) to maintain its homeostasis concentration (1−5 μM) in
plasma and tissues.22,23 The hUP enzyme catalyzes the
reversible phosphorolysis of Urd to yield uracil and ribose-1-
phosphate (R1P) in the presence of inorganic phosphate (Pi).24
The enzyme is present in two isoforms: hUP125 that is widely
distributed in tissues and hUP226 that is expressed only in
kidney. The hUP1 is the most studied isoform, and we have
recently reported its mode of action.27 The kinetic mechanism
of recombinant hUP1 was determined as steady-state ordered
bi−bi with Pi substrate binding first followed by Urd binding. In
addition, measurements of pH rate profiles for hUP1 prompted
us to propose the amino acids that are likely to play a role in
substrate binding and catalysis.27 These proposals were also
based on reported structural data for hUP1.28 A number hUP
inhibitors have been developed, including pyrimidine acyclo-
nucleoside analogues. This class of compounds can be
represented by 5-benzyl/5-benzylbarbituric acid derivatives29,30
and phenylselenyl/phenylthio-substitute,29,31 the most potent
of which being 5-benzyl-1-(2′-hydroxyethoxymethyl)uracil
(BAU or 5-benzylacyclouridine) with an in vitro Ki value of
100 nM for Urd substrate.32,33 In the present study, we report
the design, synthesis, and determination of kinetic and
thermodynamic parameters for a new class of potent hUP1
inhibitors. The effect of these inhibitors on viability of normal
and tumor cells, and their effect on 5-FU toxicity are also
described.
It was thus deemed appropriate to test as an initial hit for
hUP1 enzyme inhibition the commercially available 6-hydroxy-
1H-pyridin-2-one (1) compound (Figure 1). This compound
Figure 1. Scaffold evolution from 6-hydroxy-1H-pyridin-2-one (1).
Inhibitory activity presented between brackets for 1 μM final
concentration of compounds.
(1) was able to inhibit the activity of hUP1 by 38% when tested
at a final concentration of 1 μM (Table 1). In the first attempt
Table 1. Observed in Vitro Inhibitory Activities over hUP1
Catalyzed Reaction
a
a
compd
inhibitory activity (%)
compd
inhibitory activity (%)
1
38
67
12
70
25
16
31
42
2
6j
60
16
9.5
80
82
78
81
25
49
51
2
2
7
3
8
6a
6b
6c
6d
6e
6f
6g
6h
6i
10a
10b
10c
10d
10e
10f
10g
11
b
ND
12
43
a
Assigned at compounds final 1 μM concentrations in assay mixture.
Nondetected inhibitory activity.
b
to improve the inhibitory activity of compound 1, we analyzed
the contacts made by 5-FU at the site of hUP1 according to
diffractometry studies. From these observations, one can note
that N-1 of 5-FU is positioned 4.1 Å and 3.8 Å from residue
Thr141, whereas the 5-fluoro moiety is directed toward the
hydrophobic cavity formed by Leu272, Leu273, and Ile281. On
the basis of these distances, the sum of the van der Walls radii
and geometry, N-1 does not appear to form hydrogen bonds
with hUP1. In contrast, the 5-fluoro substituent makes a polar
contact with Ser142.35 Our hypothesis was that 6-hydroxy-1H-
pyridin-2-one nitrile-containing compounds can correctly place
a hydrogen bond acceptor in proximity to Thr141 or participate
in interactions with the cavity occupied by the fluorine of 5-FU
because both the 3- and 5-positions are identical in this
aromatic structure. Moreover, the electron-withdrawing effect
caused by the nitrile group could facilitate the formation of
hydrogen bonds with residues of hUP1 because of the
reduction in the pKa values of compounds as the conjugated
RESULTS AND DISCUSSION
■
Design and Chemistry. First, in our strategy to obtain
novel compounds with hUP1 inhibitory activity, we maintained
the 2,6-dihydroxyl groups on the pyridine ring. This type of
substitution on an aromatic ring, such as pyridine, preserves the
substituents of the uracil base and serves as a classic bioisosteric
replacement for six-membered rings.34 In addition, the
importance of these chemical groups has been illustrated by
the analysis of the main inhibitors of UP (both human and not
human) that have been previously reported. The importance of
these groups is also supported by crystallographic studies,
which have shown that hydrogen bonds form between these
substituents at a molecular recognition level in the active site of
hUP1.28,35 Indeed, both BAU and 5-FU establish polar contacts
involving the uracil rings and the Gln217 and Arg219
residues.28,35
8893
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
base is stabilized. As the polar nitrile group places the nitrogen
atom at 2.6 Å from C-3 of the 1H-pyridin-2-one ring, according
to the data for the energy-minimized structure, there would be
room for interaction with either Thr141, or Ser142, Leu272,
and Leu273. This group is present in several drugs in clinical
use and undergoing clinical trials, and it typically makes
compounds more metabolically stable than their counterparts.36
The greater stability conferred by the nitrile group is
particularly important in 1H-pyridin-2-one 1 because this
compound can readily undergo oxidation by 2,6-dihydroxypyr-
idine-3-hydroxylase to yield 2,3,6-trihydroxypyridine.37 There-
fore, 1H-pyridin-2-one 2 was synthesized according to reported
methodologies38 and its hUP1 inhibitory activity was 67% at 1
μM (Figure 1 and Table 1). The replacement of the nitrile
group with a methyl substituent at the 3-position demonstrated
the importance of the nitrile group to the compound’s
inhibitory activity because 1H-pyridin-2-one 3 inhibited the
activity of hUP1 by only 12% (Figure 1 and Table 1). As the
CN unit is approximately 8 times smaller than a methyl
group (3),36 these results suggest that bulkier groups at this
position are not likely to be better hUP1 inhibitors. To evaluate
the effect of chemical substitutions at different positions, a
series of 6-hydroxy-1H-pyridin-2-one-3-carbonitriles were
synthesized via the cyclocondensation of β-ketoesters 4a−4i
with cyanoacetamide 5 in a basic reaction medium (Scheme 1).
ring of 6a, the inhibitory activity of commercially available
compounds 7 and 8 (Figure 1) were evaluated. The absence of
the nitrile group or its replacement with a methyl group led to a
dramatic decrease in the inhibitory activity. Alkyl, phenyl, and
benzyl substitutions at C4 (6b−6f) resulted in weaker
inhibitory activity (Table 1). Ethyl, phenyl, and benzyl
substitutions at C5 of 4-methyl 6-hydroxy-1H-pyridin-2-one
ring (6g−6i) were also detrimental to the inhibitory activity
(Table 1) of the parent compound (6a). On the basis of this
preliminary screening, compound 6a was employed in the next
round of chemical modifications. It appears worth mentioning
that the compounds that were synthesized may be obtained in
three tautomeric forms in both solid and liquid state.38,39,42 On
the basis of differences in stability, the most probable tautomer
is the 1H-pyridin-2-one form. Semiempirical calculations for 6a
using the AM1 method43 showed that this form is 1.96 and 3.7
kcal mol−1 more stable than 1H-pyridin-6-one and 2,6-
dihydroxypyridine forms, respectively. The stability of 1H-
pyridin-2-one form permits to infer that this tautomer is
obtained in almost 96% at 37 °C (Figure S1, Supporting
Information). This observation was supported by experimental
data as nitrile absorption was obtained in agreement with
presence of an ortho-electron withdrawing group.42 In addition,
no signal of others tautomers was observed in both NMR and
FTIR experiments.
Acyclic mimics of the ribooxacarbenium ion of the transition
state analogues of purine nucleoside phosphorylases have been
reported to inhibit the enzyme at low nanomolar concen-
trations.44−46 In addition to their low inhibition constant
values, these simplified analogues are readily accessible
chemically, and several of them do not contain asymmetric
carbon centers. We thus deemed appropriate to use this
structural template to develop potent hUP1 inhibitors.
Incidentally, formation of a ribooxacarbenium ion in the
transition state has been proposed for Escherichia coli UP47 and
hUP128 enzymes. A bimolecular ANDN mechanism with an
SN2-like transition state, in which the ribosyl moiety possesses
significant bond order to both nucleophile and leaving groups,
has been proposed for the arsenolysis of uridine catalyzed by
Trypanosoma cruzi UP.48 Therefore, the classic Mannich
reaction was performed using primary amino alcohols and
secondary cyclic and acyclic amines to access the desired
compounds. Compounds 10a−10g were obtained after heating
6a in the presence of the appropriate amino alcohol or
secondary amine and paraformaldehyde using methanol as
solvent (Scheme 2). The products were found in 44−80%
yields, and the reaction times ranged from 4 to 16 h (Scheme
2). As previously reported,44 the secondary amines led to
compounds with yields that were higher than those of the
amino alcohol derivatives. All the synthesized compounds had
spectroscopic data that were in agreement with the proposed
structures. The 6-hydroxy-1H-pyridin-2-one-3-carbonitriles
10a−10g were assayed for their ability to inhibit the catalytic
activity of hUP1, using a final-fixed concentration of 1 μM as
initial screening (Table 1). Despite the limited number of
compounds evaluated, some characteristics of the structure−
activity relationship are apparent. The presence of the hydroxyl
group on the alkyl chain was a feature of fairly active 6-hydroxy-
1H-pyridin-2-one-3-carbonitriles (10a−10c). A uracil derivative
11 (Figure 2) was synthesized according to a reported
protocol,49 and its inhibitory activity was measured.
Scheme 1
The alkyl and aryl chains attached to C-4 and C-5 on the 1H-
pyridin-2-one-3-carbonitrile ring were varied, and the synthesis
was performed according to reported methodologies,39−41
yielding compounds 6a−6i in 34−58% yields with reaction
times of 2−16 h. When the acid work up was not performed,
the compounds could be isolated in their salt forms.39 The
potassium salt 6j was isolated in 74% yield (Scheme 1). All the
synthesized compounds had spectroscopic data that were in
agreement with the proposed structures. The hUP1 enzyme
inhibition data for compounds 6a−6j (Table 1) show that only
6a displayed a better inhibitory activity as compared to the
parent compound 2 (Figure 1 and Table 1). To assess the role
of 3-nitrile and 4-methyl groups of 6-hydroxy-1H-pyridin-2-one
Inhibitors Selection and Time-Dependent Inhibitory
Activity Measurements. The initial screening of 23
8894
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
a
itriles (6b−6f) also showed decreased inhibitory activity. The
ramifications attached to 5-position, absent in 2 and 6a−6f,
resulted in different inhibitory effects, namely compounds 10a−
10d displayed improved inhibitory activity whereas 10e−10g
were less active (Table 1) as compared to the parent compound
(6a). It is noteworthy that the simplest amino alcohol led to the
most potent series of inhibitors (10a−10c). The subtlety of the
interactions with hUP1 can be assessed by the differing
inhibitory activities of compounds 10d (81%) and 10e (25%)
at 1 μM. The effects of constraining the side chain of 10a using
a morpholine derivative were tested based on the hypothesis
that this approach would reduce the entropic penalties, leading
to a more potent inhibitor. However, compound 10g (51%)
was less potent than 10a (80%) at 1 μM (Table 1).
Ramifications on C5 of 6-hydroxy-1H-pyridin-2-one-3-carbon-
itrile participate in hydrogen and hydrophobic interactions with
amino acids in hUP1 active site, as discussed below. The six
best compounds (2, 6a, 10a−10d) and BAU showed no time-
dependent inhibition of hUP1 enzyme activity (data not
shown). Accordingly, Ki measurements were carried out
following classic Michaelis−Menten experiments. DMSO was
investigated as a possible enzyme activity modulator, and no
interference was observed at the maximum percentage of this
solvent used to dissolve the compounds (1%).
Scheme 2
a
Reactants and conditions: i = OH(CH2O)nH, MeOH, 100 °C, 4−16
h.
Determination of Ki for Inhibitors of hUP1. The Ki
values were determined for six selected compounds (2, 6a,
10a−10d) and BAU. The Ki of BAU was determined for both
substrates and was compared with the Ki value for Urd in the
literature.32,33 The values determined corroborate the literature
data and allowed a directed comparison of the inhibitory ability
between the molecules here described and the best in vitro
hUP1 inhibitor reported in the literature (BAU). Lineweaver−
Burk plots (Figures S5−S11, Supporting Information)
suggested the mode of inhibition (competitive, noncompetitive
or uncompetitive) and data fitting to appropriate equations
gave values for the inhibition constants (Kis and/or Kii).50,51
The inhibition data are summarized in Table 2. The best
compounds in the series of synthesized analogues were 10a
with Ki values for Urd = 68 10 nM and Pi = 127 8 nM, and
10b with Ki values for Urd = 94 11 nM and Pi = 98 6 nM.
For both inhibitors, the types of inhibition were competitive
toward Urd and uncompetitive to Pi. A steady-state ordered
bi−bi kinetic mechanism has been proposed for hUP1, in which
Pi binds first followed by the binding of Urd, and uracil
dissociates first followed by R1P release.27 The inhibition
patterns for 10a and 10b indicate that the molecule binds to
hUP1−Pi binary complex and competes with Urd binding en
route to ternary complex formation, corroborating the ordered
bi−bi mechanism proposed for hUP1.27 Although one cannot
assume that because an inhibitor displays the characteristic
Figure 2. Structural modifications of compound 10a with apparent
requirements for potency. Percentage of inhibition of hUP1 activity
showed between brackets for compound concentrations of 1 μM.
compounds showed that six compounds (2, 6a, 10a, 10b, 10c,
and 10d) displayed reasonably good inhibition (>60%) of
hUP1 activity at final-fixed concentration of 1 μM (Table 1).
The nitrile group on 3-position with (6a) or without (2) a
methyl group on 4-position of the 6-hydroxy-1H-pyridin-2-one
ring appears to be required for inhibitory activity of this class of
compounds. The absence of the nitrile (1, 7) or its replacement
with a methyl group (3, 8) at the 3-position of 6-hydroxy-1H-
pyridin-2-one ring were detrimental to their inhibitory activity.
Compounds in which the C4-methyl group was replaced with
bulkier groups in the 6-hydroxy-1H-pyridin-2-one-3-carbon-
Table 2. Inhibitory Constants of Selected Molecules on hUP1 Activity
compd
Kis/Kii Urd (nM)
inhibition mode
Kis/Kii Pi (nM)
inhibition mode
2
2610 339/2546 330
375 145/635 91
68 10
noncompetitive
noncompetitive
competitive
competitive
competitive
competitive
competitive
565 113/666 59
332 32
noncompetitive
uncompetitive
uncompetitive
uncompetitive
uncompetitive
uncompetitive
noncompetitive
6a
10a
10b
10c
10d
127
98
8
94 11
6
109 10
182
107
7
7
99 13
a
BAU
130 21
547 44/317 61
a
5-Benzyl-1-(2′-hydroxyethoxymethyl)uracil.
8895
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
signature of a competitive inhibitor that it necessarily binds to
the enzyme active site,50,51 the uncompetitive patterns for 10a
and 10b toward Pi further suggest that these compounds bind
to the Urd binding site of hUP1. For the design and synthesis
of inhibitors, it is important to keep in mind the probable effect
of these compounds in vivo, synthesizing molecules with
different types of inhibition as their effectiveness (ability to
produce a beneficial effect) depends on both the inhibitor
binding affinity and the concentration of the natural ligands in
the organism. It has been described that hUP1 activity as well as
the enzymes in the salvage pathway are enhanced in tumor cells
in contrast to normal cells because the tumor cells are
constantly undergoing cell division thus requiring recycling of
bases and nucleosides for RNA and DNA synthesis.25,52 Owing
to the importance of Pi and Urd in many cell processes, their
balances are tightly controlled being vital for maintaining basic
cellular function.1,53−55 The Pi concentration is maintained at
approximately 4.4 mM,53 which is 2-fold larger than the Km
value for Pi (2.46 mM),27 and the Urd levels in plasma and
tissues are tightly maintained in the range 1−5 μM,22,23 values
lower than the Km for Urd (51 μM) for the hUP1-catalyzed
chemical reaction.27 Accordingly, it is tempting to suggest that
the competitive inhibition mode toward Urd and uncompetitive
mode toward Pi bodes well to in vivo activity of 10a−10d
compounds. As intracellular Pi concentration is larger than its
Km value, formation of hUP1−Pi is likely and ensuing binding
of 10a−10d compounds to this binary complex feasible. In
addition, these compounds would compete with an intracellular
concentration of Urd that is lower than its Km value, and
displacement of inhibitors from hUP1 active site by Urd would
thus be precluded.
Molecular docking experiments were performed to try to
determine the interactions that occur upon enzyme−inhibitor
complex formation. Interestingly, BAU showed a lower number
of interactions than 10a and 10b (Figure 3). The substituents
at the 5-position of 10a and 10b make hydrogen bonds with
Thr141 (chain A) and Arg138 (chain A), and the ligand 10b
also participates in a hydrophobic interaction with Arg94 (chain
B). The corresponding 1-hydroxyethoxymethyl group of BAU
makes hydrogen contacts with Thr141 (chain A) and His36
(chain B) and a hydrophobic contact with Ser142 (chain A).28
The benzyl ring of BAU makes hydrophobic interactions with
Phe213 (chain A), Ile281 (chain A), Met110 (chain A), and
Tyr35 (chain B).28 The nitrile groups of 10a and 10b make
hydrophobic interactions with Leu272 (chain A), Leu273
(chain A), Ser142 (chain A), and Gly143 (chain A). The nitrile
of 10b also makes a hydrophobic interaction with Phe213
(chain A). It has been shown that the fluorine moiety of 5-FU
forms a hydrogen bond with Ser142 and hydrophobic
interactions with Leu272, Leu273, and Ile281 residues.35 The
nitrile of 10a and 10b and the fluorine of 5-FU thus appear to
bind to the same site of hUP1. The compounds without the
ramification on 5-position (e.g., 2 and 6a) do not make
hydrogen bonds with Thr141 (chain A) and Arg138 (chain A)
(data not shown), which is consistent with the larger Ki values
for these compounds (Table 2).
Figure 3. Three-dimensional representations of hUP1−ligands
interactions. Docking simulations for hUP1−10a (A) and hUP1−
10b (B). Crystal structure (PDB ID: 3EUF) of hUP1−BAU complex
(C). The residues that are making contacts with 10a, 10b, and BAU
are drawn as ball and sticks and colored as follows. Hydrophobic
contacts: carbon (orange), nitrogen (blue), oxygen (red). Hydrogen
bonds: carbon (light-blue), nitrogen (blue), oxygen (red). The 5-
position of 10a and 10b and 1-position of BAU are colored in deep-
olive and drawn as sticks. Phosphates are drawn as ball and stick
representation and colored by atom (oxygen = black and phosphorus
= pink). hUP1 tertiary structure is represented as cartoon. Figures
prepared using PyMol (www.pymol.org).
results lend support to the specificity of compounds tested
toward hUP1 enzyme as has been reported for pyrimidine
acyclonucleosides inhibitors.29,30 As thymidine phosphorylase
may play a role in the therapeutic application of 5-FU because it
is known to catalyze the conversion of 5-FU to 5′-fluoro-2′-
deoxyuridine by the addition of 2-deoxyribose-1-phosphate, the
specificity of the compounds here described for hUP1 activity
appears to be a desirable feature to not interfere with antitumor
activity of 5-FU. The low Ki values of 10a and 10b (in the same
range as BAU), the mode of inhibition and specificity for hUP1
warranted further experimental efforts.
Enzyme activity measurements were carried out to assess
whether or not the compounds to which Ki and mode of
inhibition of hUP1 were determined (Table 2) could also
inhibit human thymidine phosphorylase enzyme (EC 2.4.2.4)
activity. The IC50 values (concentration of inhibitor that
reduces enzyme velocity by half) were >100 μM for
compounds 2, 6a, and 10a−10d (data not shown). These
8896
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
a
Table 3. Thermodynamics Parameters for Free Enzyme and for the hUP1−Pi and hUP1−R1P Complexes at 310.15 K (37 °C)
compd
hUP1 or complex
ΔH° (kcal/mol)
−9.6 0.3
−TΔS° (kcal/mol/K)
ΔG° (kcal/mol)
−7.8 1.5
Kd (nM)
2
hUP1
1.7 0.3
2900 554
hUP1−Pi
hUP1−R1P 50 μM
hUP1−R1P 150 μM
−15 0.1
−18 0.5
5.7 0.9
8.5 3.2
−9.8 1.6
−9.7 3.7
108 18
127 48
6a
hUP1
hUP1−Pi
−19 1.3
−31 1.2
−21 0.1
−22 0.1
12 1.4
22 5.4
11 1.5
11 1.3
−6.9 0.8
−8.3 2.0
−9.8 1.4
−11 1.3
12000 1395
1400 341
40 5.0
hUP1−R1P 50 μM
hUP1−R1P 150 μM
hUP1−R1P 200 μM
20 2.0
10a
10b
10c
10d
hUP1
−23 0.5
−16 0.2
−17 0.6
−30 1.5
14 1.5
6.4 1.4
9.5 1.7
23 3.3
−8.0 0.9
−9.5 2.1
−8.0 1.5
−7.6 1.1
2000 221
151 34
hUP1−Pi
hUP1−R1P 50 μM
hUP1−R1P 150 μM
2300 423
3800 553
hUP1
−24 0.6
−14 0.2
−18 0.4
−10 1.0
16 2.0
5.0 0.8
10 1.0
3.3 1.0
−7.4 1.0
−9.0 1.5
−7.3 0.7
−7.3 2.2
5300 674
502 83
hUP1−Pi
hUP1−R1P 50 μM
hUP1−R1P 150 μM
7000 661
7000 2115
hUP1
−22 2.2
−19 0.1
−22 0.5
−18 0.8
15 2.0
10 1.0
14 1.0
11 1.2
−6.7 0.9
−9.3 0.9
−7.5 0.5
−7.1 0.8
20000 2633
263 26
hUP1−Pi
hUP1−R1P 50 μM
hUP1−R1P 150 μM
4600 344
9400 1037
hUP1
−6.7 0.2
−12 0.2
−22 0.4
−23 0.1
0.8 0.1
2.6 0.6
14 0.7
16 2.1
−7.6 1.1
−9.1 2.1
−7.2 0.4
−7.0 0.9
4200 602
312 73
hUP1−Pi
hUP1−R1P 50 μM
hUP1−R1P 150 μM
7400 381
10000 1325
b
BAU
hUP1
−10 0.1
−13 0.3
−12 0.3
−20 0.7
1.5 0.2
4.6 0.8
4.3 0.9
12 1.6
−8.4 1.0
−8.2 1.4
−8.2 1.7
−7.3 1.0
1200 150
1400 239
1200 244
6600 898
hUP1−Pi
hUP1−R1P 50 μM
hUP1−R1P 150 μM
a
b
Cells in which no data are given are for interaction processes that yielded no reliable data. 5-Benzyl-1-(2′-hydroxyethoxymethyl)uracil.
Determination of Thermodynamic Parameters by
enthalpy and in entropy, resulting in no affinity gains (ΔG°).
Interestingly, the parent compound (6a) displayed a more
favorable enthalpy of binding to hUP1−Pi (−19 kcal mol−1)
than the derivatives 10a (−16 kcal mol−1), 10b (−14 kcal
mol−1), and 10d (−12 kcal mol−1), except 10c (−19 kcal
mol−1). However, the more pronounced entropic penalty for 6a
as compared to 10a−10d resulted in lower values for the
dissociation constant of the latter compounds and more
favorable ΔG° values (Table 3). These results are in agreement
with lower inhibition constant values for 10a−10d as compared
to 6a obtained from enzyme activity measurements (Table 2).
The lower dissociation constant values for 10a−10d binding to
hUP1−Pi as compared to binding to free enzyme (Table 3)
lends further support to the uncompetitive mode of inhibition
toward Pi (Table 2). Interestingly, the ITC data for compounds
2 and 6a show that there exists an increase in affinity (lower
dissociation constant values) for binding to hUP1−R1P binary
complex as compared to binding to free hUP1 for 2 and to
hUP1−Pi for 6a (Table 3). Compound 6a, in particular,
displays increasing affinity as a function of increasing R1P
concentration (Table 3), thereby suggesting formation of a
dead-end complex. On the other hand, compounds 10a−10d
showed decreasing affinity as a function of increasing R1P
Isothermal Titration Calorimetry (ITC). The interaction
phenomenon can be monitored by ITC measuring the
differential heat (released or absorbed) among the sample
and reference cells during a binding reaction, yielding values for
the following thermodynamic parameters: enthalpy change
(ΔH°), entropy change (ΔS°), Gibbs free energy change
(ΔG°), and association constant (Ka). The binding processes
for compounds 2, 6a, 10a−10d, and BAU to hUP1 were
assessed by ITC (Table 3). The experimental data were
analyzed in the differential mode, displaying sigmoidal plots
(Figures S12−S18, Supporting Information). The differential
mode treats each injection as an independent point and is
plotted as heat evolved per injection vs molar ratio (ratio of
total ligand concentration to hUP1 concentration).56 The
binding interactions between hUP1 and ligands were
characterized by exothermic reactions and favorable Gibbs
free energy changes, as negative values were observed for both
ΔH° and ΔG° (Table 3). Compounds 2, 6a, and 10a−10d
exhibited a rather frequent problem in biological systems on the
course of lead optimization, namely the enthalpy−entropy
compensation (ΔH°−ΔS°) behavior.57−59 This compensatory
system leads to a linear relationship between the change in
8897
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
Figure 4. The (A) HaCat, (B) HT-29, and (C) SW-620 cell lineages were cultured and treated with different concentrations of 5-FU followed by 72
h of incubation with 30 μM of the compounds 6a, 10a, or BAU. Control cells were exposed only to culture medium or culture medium with 1%
DMSO. Data are shown as mean standard error of mean of three independent experiments performed in quadruplicate. The statistical analysis was
performed by one-way ANOVA followed by Bonferroni’s post hoc test. * Significantly different from the control group (p < 0.05).
concentration (Table 3), suggesting overlapping binding sites.
In addition, the stoichiometry (n) of the reaction was different
for some inhibitors. The hUP1 enzyme is dimeric in solution,
with the two active sites in the interface of the monomers,28
consequently the n of the binding assays should be 1. The
compounds 2, 6a, and 10c showed n = 0.5, suggesting a
negative cooperativity in which the binding of one molecule of
inhibitor cause a conformational change that locks the protein
in a disable form, weakening the binding affinity of the other
active site of the enzyme.60 The thermodynamics parameters of
6a and 10a−10d show more favorable ΔH° values as compared
to BAU (Table 3), indicating a larger stability value for the
protein−ligand complex (less energy is required for complex
formation) due to the thermodynamic of noncovalent
interaction profile (hydrogen bonds and van der Waals
redistribution), suggesting a better complementarity between
the protein and ligands.58,59 On the other hand, the entropic
contributions of 6a and 10a−10c (except 10d) are less
favorable as compared to BAU (Table 3). These results suggest
that addition of hydrophobic groups could improve the
entropic contributions of these compounds, perhaps on 3-
position of the pyridine ring. According to previous reports,
which were confirmed by hUP1 crystallographic structure, there
is a hydrophobic pocket adjacent to the uracil binding site that
is formed by amino acids of the two subunits.28,32 There are
differences between the interaction of BAU and compounds
10a−10b with the hUP1 active site (Figure 3). These
differences include contacts with Arg94 of chain B for 10b
(Figure 3B) and a hydrogen bond with His36 of chain B and a
hydrophobic contact with Tyr35 of chain B for BAU (Figure
3C); the latter is consistent with the more favorable entropic
contribution for BAU (Table 3). However, there is a limit to
refine the hydrophobic nature of the analogues, such as
solubility of the compounds, which can reduce oral
bioavailability. For some compounds, it was not possible to
determine the thermodynamic parameters for all experimental
conditions tested due to low affinity of the molecule for some
forms of enzyme in solution (either in complex or free).
Cytotoxicity of hUP1 Inhibitors on Normal and Tumor
Cells. The possible in vitro cytotoxicity of the lead molecule
(6a) and the best in vitro inhibitor of hUP1 (10a) was
evaluated in normal and tumor cell lines using the 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)
assay. The in vitro incubation of the selected compounds (1−
100 μM) did not significantly affect cell viability of either
normal or tumor cells (Figure S19, Supporting Information).
Influence of hUP1 Inhibitors on 5-FU IC50. A
concentration−response curve for 5-FU was performed in
presence or absence of the compounds 6a and 10a to
determine the influence of these inhibitors on the cytotoxicity
of 5-FU in normal and tumor cell lines, using BAU as positive
control.61 The IC50 values for 5-FU in both the presence and
absence of 6a, 10a, and BAU are expressed as the
concentrations that corresponded to a reduction of cellular
growth by 50%. The 5-FU IC50 for HaCat and HT-29 were
similar to values previously described in the literature,61,62 with
mean values (accompanied by the 95% confidence interval) of,
respectively, 1.3 (6.9−0.27) and 6.9 (9.5−5.0) μM (Figure 4).
However, the IC50 value for the tumor SW-620 cell line [81
(107−61) μM] was significantly different from a previously
reported value (15.3
0.8 μM),62 probably because the
published assay was performed after only 24 h of exposure. The
analysis showed that HaCat cells were equally sensitive to 5-FU
with or without the compounds (Figure 4A), suggesting that
the inhibitors have no effect on cytotoxicity of 5-FU for normal
cells, as expected for BAU.61 Similarly, for HT-29 tumor cells,
the incubation of compound 6a or BAU (30 μM) did not cause
a significant alteration of IC50 values when compared to control
cells (Figure 4B). However, the coexposure of these cells to 5-
FU and compound 10a (30 μM) produced an increase of about
9-fold in IC50 values, suggesting a protective effect for HT-29
cells (Figure 4B). Interestingly, when the more aggressive cell
line SW-620 was exposed to increasing concentrations of 5-FU,
in the presence of the inhibitors, the cells were far more
sensitive to 5-FU, with a decrease in IC50 values of 12-fold for
6a and 6-fold for 10a (Figure 4C). 5-FU has multiple possible
mechanisms of interference with the proliferative activity of
cells.13−16 It is tempting to suggest that inhibition of hUP1 by
6a and 10a decreased the conversion rate of 5-FU into 5-
FUridine in the tumor cells, resulting in increased 5-FU
concentration that would thus inhibit the TS enzyme activity
and cause the DNA damage, responsible for the therapeutic
effect of the chemotherapy.17−19 However, determination of
intracellular concentrations of these metabolites in the presence
of 6a and 10a should be assessed to provide a solid basis for
these proposals. At any rate, both the low cytotoxicity and the
lower IC50 values for 5-FU (more so for 6a) appear to warrant
further efforts to be pursued.
CONCLUSIONS
■
The interest in the search for compounds that are able to
decrease 5-FU toxic effects on normal cells is not recent. The
initial studies were performed using exogenous Urd and its pro-
8898
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
methanol (2 × 10 mL). Finally, the product was dried under reduced
pressure for at least 24 h to give 0.481 g (54%) of 6c as white powder.
No further purification was needed. Melting point 247−249 °C.
HPLC 96.6% (tR = 16.72 min). 1H NMR (60 MHz, DMSO-d6): δ 0.93
(t, 3H, CH3), 1.54 (quint, 2H, CH2), 2.53 (t, 2H, CH2), 5.64 (s, 1H,
CH), 11.45 (br s, 2H). FT-IR (UATR): 2874, 2222, 1599, 1292, 840.
HRMS (ESI) calcd for C9H10N2O2 + H, 179.0815; found, 179.0814
(M + H)+.
General Procedure for Preparation of 6-Hydroxy-1H-pyridin-2-
one-3-carbonitriles 10. Selected Example for Compound 6-
Hydroxy-5-(((2-hydroxyethyl)amino)methyl)-4-methyl-1H-pyridin-
2-one-3-carbonitrile (10a). The mixture of 6-hydroxy-4-methyl-1H-
pyridin-2-one-3-carbonitrile (6a) (1.0 mmol, 0.150 g), ethanolamine
(9a) (1.0 mmol, 0.061 g), and paraformaldehyde (0.030 g) in
methanol (10 mL) was heated in a Fisher−Porter vessel at 100 °C for
16 h. After cooling to room temperature, the solvent was evaporated
under reduced pressure. The solid obtained was washed with water
(10 mL) and cold diethyl ether (10 mL). Finally, the product was
crystallized from methanol, using carbon black, and subsequently was
dried under reduced pressure to give 0.153 g (64%) of 10a as pale
drugs, but interest in these approaches diminished due to
pharmacokinetic problems. One promising approach is the use
of hUP inhibitors, which appears to be the best option to raise
and maintain the Urd levels by an endogenous process. There
are many classes of drug targets, and enzymes comprise an
important category. The pharmaceutical industry has a great
interest in the production of enzyme-targeted drugs due to the
catalysis advantage for the drug design beyond the binding
property, and until 2005 the FDA had approved 317 drugs that
had enzymes as targets.63 The hUP1 inhibitors have a potential
utility in cancer chemotherapy regimens as modulators of the
host toxicity of 5-FU. In this study, we synthesized a new
potent class of hUP1 inhibitors and determined the kinetic and
thermodynamic properties of these inhibitors. Data on cell
culture suggested that the use of hUP1 inhibitors allow the
reduction of 5-FU concentration, resulting in less damage to
the normal cells and fewer side effects for the patient. The data
here presented may provide key information to in vivo action
and could be used to design new molecules for cancer
chemotherapy.
white solid. Melting point 219−221 °C (dec). HPLC 95.8% (tR
=
1
11.73 min). H NMR (400 MHz, DMSO-d6): δ 2.13 (s, 3H, CH3),
2.97 (t, 2H, CH2), 3.65 (t, 2H, CH2), 3.86 (s, 2H, CH2), 5.05 (br s,
1H), 7.98 (br s, 2H), 10.00 (br s, 1H, NH). FT-IR (UATR, cm−1):
3189, 2937, 2775, 2202, 1572, 1338, 1302, 704. HRMS (ESI) calcd for
C10H13N3O3 + H, 224.1030; found, 224.1032 (M + H)+.
EXPERIMENTAL SECTION
■
Chemistry. Compounds 1, 3, 7, and 8, starting materials, and
solvents were used as obtained from commercial suppliers without
further purification. Although compounds 6a−b, 6d−6e, 6g, and 6i are
commercially available, they were synthesized by our research group.
Expression, Purification, and Enzyme Activity Assay.
Production and isolation of recombinant hUP1 were as described in
a previous study as well as hUP1 enzyme activity assay.27 In brief, pET-
23a(+)::hUP1 recombinant plasmid was transformed into Escherichia
coli Rosetta (DE3) host cells, cultured in Terrifc Broth medium and
grown for 36 h at 30 °C. The isolation of the recombinant hUP1 was
by a single-step HPLC protocol using a cation exchange column, and
the homogeneous protein was stored at −80 °C. Recombinant protein
concentration was determined by the Bradford’s method.63,64 The
enzyme assay was performed under initial rate conditions in which
hUP1 activity was linear function of both time and protein
concentration, monitoring a decrease change in absorbance at 280
nm upon conversion of Urd to uracil using a spectrophotometer (UV-
2550 UV/Visible, Shimadzu). The assay mixture contained 100 mM
Tris buffer pH 7.4, 100 nM of recombinant hUP1, Urd, and Pi
concentrations at their Km values (50 μM and 2 mM, respectively) in a
final volume of 500 μL. The inhibitors were added to the mixture in a
final concentration of 1 μM. All the compounds that inhibited more
than 60% of the initial enzyme activity were selected. Assays were
carried out at 37 °C for 60 s. For the compounds that were dissolved
in DMSO, a control reaction was performed to evaluated DMSO effect
at a maximum final concentration of 1%.
Time-Dependent Inhibitory Effect. Before embarking on
determination of inhibition dissociation constants, the presence of
time-dependent inhibitory activity was evaluated. For this analysis, 100
nM of recombinant hUP1 was preincubated with 1 μM of inhibitor
(final concentration), which was then added at different times (up to
30 min) to the reaction mixture described above. The change in initial
velocity with time was monitored and the percentage of inhibition was
calculated.
Determination of Ki for Selected Inhibitors of hUP1. The
determination of Ki values and the mode of inhibition (competitive,
noncompetitive, or uncompetitive) were performed for each selected
inhibitor and BAU toward both hUP1 substrates. The inhibition
studies were carried out at varying concentration of one substrate,
fixed concentration of other substrate (Km value), and four different
concentrations of inhibitor (0.1 to 5 μM). For Urd substrate analysis,
the experiment was repeated with Urd concentrations ranging from 20
to 600 μM and Pi at 2 mM, and for Pi substrate analysis, Pi
concentrations ranging from 0.1 to 12 mM with Urd fixed at 50
μM. The enzyme concentration was constant at 100 nM throughout
the assays. The Kis and Kii values were estimated from the slope and
intercepts, respectively, of Lineweaver−Burk plot,65 and the mode of
inhibition was determined from the straight line patterns.65 The data
́
All melting points were measured using a Microquimica MQAPF-302
apparatus. 1H NMR spectra were acquired on a Bruker DPX 400
(Federal University of Santa Maria, UFSM/Brazil) or Anasazi EFT-60
spectrometer (1H at 400.13 MHz or 60.13 MHz, respectively) at 25
°C, for DPX400, and 30 °C for EFT-60, in 5 mm sample tubes.
DMSO-d6 or D2O were used as solvent. For molecules solubilized in
DMSO-d6, TMS was used as internal standard. Chemical shifts were
expressed in ppm, and J values were given in Hz. High-resolution mass
spectra (HRMS) were obtained for all compounds on an LTQ
Orbitrap Discovery mass spectrometer (Thermo Fisher Scientific).
This hybrid system combines an LTQ XL linear ion trap mass
spectrometer and an Orbitrap mass analyzer. The experiments were
performed via direct infusion of the samples in phosphoric acid 0.1%
acetonitrile/water (1:1, v/v) in positive-ion mode using electrospray
ionization (ESI); flow rate of 10 μL min−1. Elemental composition
calculations were carried out using the specific tool included in the
Qual Browser module of Xcalibur (Thermo Fisher Scientific, release
2.0.7). Fourier transform infrared (FTIR) spectra were recorded using
a universal attenuated total reflectance (UATR) attachment on a
PerkinElmer Spectrum 100 spectrometer in the wavenumber range of
650−4000 cm−1 with a resolution of 4 cm−1. Purity of compounds was
̈
determined by HPLC using an Akta HPLC system (GE Healthcare
Life Sciences). HPLC analysis conditions: RP column 5 μm Nucleodur
C-18 (250 mm × 4.6 mm); flow rate 1.5 mL min−1; UV detection at
270 and 254 nm; 100% water (0.1% acetic acid) for 7 min, followed by
linear gradient from 100% water (0.1% acetic acid) to 90%
acetonitrile/methanol (1:1, v/v) in 16 min; the last partition was
maintained by 15 min and subsequently returned to 100% water (0.1%
acetic acid) in 5 min, remaining for an additional 6 min. All evaluated
compounds were >95% pure.
General Procedure for Preparation of 6-Hydroxy-1H-pyridin-2-
one-3-carbonitriles 6. Selected Example for Compound 6-Hydroxy-
4-propyl-1H-pyridin-2-one-3-carbonitrile (6c). To a solution of β-
ketoester 4c (0.791 g, 5 mmol) and cyanoacetamide 5 (0.420 g, 5
mmol) in methanol (8 mL) was added potassium hydroxide (0.421 g,
7.5 mmol) in methanol (8 mL). The mixture was stirred at 65 °C for 8
h. After cooling to room temperature, the amorphous solid formed was
filtered off and washed with cold methanol (2 × 15 mL). The solid
obtained was dissolved in warm water (20 mL), filtered, cooled in an
ice bath, and acidified with concentrated hydrochloric acid. The
resultant solid was then washed with cold water (2 × 10 mL) and
8899
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
were fitted to the following equations for competitive, noncompetitive,
or uncompetitive, respectively:
supplemented with 10% (v/v) fetal bovine serum (FBS, Invitrogen)
and antibiotics, and the human colon cancer cell lines HT-29 and SW-
620 were cultured on RPMI-1640 (Invitrogen) supplemented with
10% (v/v) FBS and antibiotics. The cells were incubated in plastic
culture dishes at 37 °C, a minimum relative humidity of 95%, and an
atmosphere of 5% CO2 in air. The medium was replaced with fresh
one every 48 h. When ∼80% the confluence was reached, cells were
treated with trypsin (2 mL) to separate them from the dishes, counted,
and the appropriate culture medium volume added before being
seeded. All the assays had as experimental control cultured cells with
medium only and medium with 1% of DMSO.
Cytotoxicity of hUP1 Inhibitors on Normal and Tumor Cells.
For cytotoxicity assays, 7 × 103 cells were seeded onto 96-well plates
and incubated for 48 or 24 h (HaCat or HT-29 and SW-620,
respectively) before drug treatments to promote their adhesion to the
plate. Thereafter, the medium cultured was removed, replaced with
varying concentrations of the inhibitors 6a and 10a (1−100 μM), and
exposed for 72 h. For the analysis of cell viability, the medium was
removed, the cells were washed with calcium magnesium-free buffer,
and 100 μL of 10% (v/v) MTT solution was added to the cells and
incubated for 3 h. The formazan crystals formed were dried at room
temperature for at least 24 h and then dissolved with 100 μL of
DMSO. The absorbance was measured at 495 nm (Spectra Max M2e,
Molecular Devices). The formazan concentration is directly propor-
tional to the number of live cells with active mitochondria. The
percentage of cell viability was calculated according to the following
equation:
v = VA/[Ka(1 + I/Kis) + A]
v = VA/[Ka(1 + I/Kis) + A(1 + Kii)]
or
v = VA/[A(1 + I/Kii) + Ka]
where V is the maximum velocity Vmax, A is the substrate
concentration, Ka is the Michaelis−Menten constant (Km), I is the
inhibitor concentration, Kis is the equilibrium dissociation constant for
enzyme−inhibitor complex, and Kii is the equilibrium constant
dissociation constant for enzyme−substrate−inhibitor complex.66 All
measurements were performed in duplicate.
Determination of Thermodynamics Parameters by Isother-
mal Titration Calorimetry (ITC). The binding interaction between
the enzyme and ligands was performed by ITC using ITC200
microcalorimeter (MicroCal, Inc., Northampton, MA). All assays were
performed under initial rate conditions at 310.15 K (37 °C), 56 μM of
recombinant hUP1 in Tris HCl 50 mM, pH 7.4, 10 mM of Pi (final
concentration), and R1P concentrations ranging from 50 to 200 μM
(final concentration). The reference cell (200 μL) was loaded with
MilliQ water for all assays. The compounds 2, 10a−10d, and BAU
were first dissolved in DMSO and then diluted in buffer (Tris HCl 50
mM pH 7.4), and the compound 6a was diluted only in buffer. The
same percentage of DMSO in the final stock was added to the sample
cell (200 μL) containing the reaction mixture. The formation of either
binary or ternary complex was carried out by one injection of 0.5 μL
(excluded of the analysis), followed by 26 injections of 1.5 μL of each
inhibitor to a solution containing either free enzyme, hUP1−Pi , or
hUP1−R1P binary complexes in the sample cell. Nonspecific
contributions arising from dilution effects of the Pi or R1P and
protein were corrected for by subtracting the small heat changes at the
end of titrations.56 The thermodynamics parameters ΔH° (enthalpy of
binding) and Keq (affinity constant at equilibrium) and stoichiometry
of binding (n) were obtained from fitting the data to one set of sites
binding model. The ΔS° (entropy of binding) and ΔG° (Gibbs free
energy of binding) were calculated by the following equations:
cell viability (%) = AbsT/AbsC × 100
where AbsT is the absorbance of treated cells and AbsC is the
absorbance of control (untreated) cells. The cell viability was reported
as the mean of three different experiments standard error mean and
each experiment was performed as quadruplicate. One-way ANOVA
was used as statistical analysis for cytotoxicity followed by Bonferroni’s
post hoc test. A p value <0.05 was considered statistically significant.
Values are reported as the mean standard error mean in Figure S19,
Supporting Information.
Influence of hUP1 Inhibitors on 5-FU IC50. For the 5-FU
concentration−response curve, 7 × 103 cells were seeded onto 96-well
plates and grown as described above. The cells were exposed for 72 h
to increasing concentrations of 5-FU: HaCat (0.25−20 μM), HT-29
(0.25−50 μM), and SW-620 (0.5−150 μM). The same concen-
tration−response curve was performed in the presence of 30 μM of
the compounds 6a, 10a, and also BAU as a positive control.62 Analysis
of cell sensitivity to the 5-FU, in the presence or absence of inhibitors,
was performed by MTT assay (described above) and expressed by
IC50, which was calculated from a semilogarithmic concentration−
response curve by nonlinear regression using Prism software
(GraphPad 5.0). Experiments were carried out in quadruplicate and
repeated three times. One-way ANOVA was used as statistical analysis
for IC50 analysis followed by Bonferroni’s post hoc test. A p value
<0.05 was considered statistically significant. Values are reported as the
mean accompanied by the 95% confidence interval.
ΔG° = ΔH° − TΔS°
ΔG° = −RT ln Keq
where R is the gas constant (1.98 cal K−1 mol−1 or 8.314 J K−1 mol−1),
T is the temperature in Kelvin (T = °C + 273.15), and ln Keq is the
natural logarithm of equilibrium constant.
The dissociation constant at equilibrium (Kd) was calculated by the
following equation:
Kd = 1/Keq
Molecular Docking of the Selected Compounds. Molecular
docking experiments were performed to analyze the interaction mode
of the selected compounds with the receptor hUP1. The receptor and
ligand structures were prepared using AutoDockTools1.5.2, while
docking simulations were performed with AutoDock4.2, allowing
flexibility to the ligands.67,68 The crystal structure of hUP1 associated
with BAU (PDB ID: 3EUF) was used as template. Because hUP1 is
biologically active as a dimer, this form was used to perform all
docking experiments. For all simulations, the 3D-grid dimension used
to define the hUP1 active site and to evaluate the scoring function was
60 × 60 × 60, with spacing of 0.375 Å. The Lamarckian Genetic
Algorithm (LGA) was employed as the docking algorithm with 50
runs and the remaining parameters set to their default values, except
for ga_num_evals, which was set to 2500000. All docking experiments
were performed with the most stable tautomeric form of the
compounds assessed by chemical synthesis analyses (NMR and FT-
IR) and semiempirical calculations.
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthesis and characterization of compounds (1H NMR, FTIR,
and high resolution mass spectrometry), tautomeric equili-
brium of 6a, Lineweaver−Burk plots of compounds 2, 6a, 10a−
10d, and BAU, isothermal titration calorimetry of compounds
2, 6a, 10a−10d, and BAU, and cell viability. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*For L.A.B.: phone/fax, +55-51-33203629; E-mail, luiz.basso@
pucrs.br.
■
Cell Culture. The human keratinocyte cell line HaCat was cultured
in Dulbecco’s Modified Eagle’s Medium (DMEM, Invitrogen),
*For D.S.S.: E-mail, diogenes@pucrs.br.
8900
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
(12) Pinedo, H. M.; Peters, G. J. Fluorouracil: Biochemistry and
Pharmacology. J. Clin. Oncol. 1988, 6, 1653−1664.
Author Contributions
The manuscript was written through contribution of all authors.
All authors have given approval to the final version of the
manuscript.
(13) Longley, D. B.; Harkin, D. P.; Johnston, P. G. 5-Fluorouracil:
Mechanisms of Action and Clinical Strategies. Nature 2003, 3, 330−
338.
Funding
(14) Rich, T. A.; Shepard, R. C.; Mosley, S. T. Four Decades of
Continuing Innovation with Fluorouracil: Current and Future
Approaches to Fluorouracil Chemoradiation Therapy. J. Clin. Oncol.
2004, 22, 2214−2232.
(15) van Groeningen, C. J.; Peters, G. J.; Pinedo, H. M. Modulation
of Fluorouracil Toxicity with Uridine. Semin. Oncol. 1992, 19, 148−
154.
(16) Grem, J. L. 5-Fluorouracil: Forty-Plus and Still Ticking. A
Review of Its Preclinical and Clinical Development. Invest. New Drugs
2000, 18, 299−313.
(17) Houghton, J. A.; Houghton, P. J.; Wooten, R. S. Mechanism of
Induction of Gastrointestinal Toxicity in the Mouse by 5-Fluorouracil,
5-Fluorouridine, and 5-Fluoro-2′-deoxyuridine. Cancer Res. 1979, 39,
2406−2413.
(18) al Safarjalani, O. N.; Rais, R.; Shi, J.; Schinazi, R. F.; Naguib, F.
N. M.; el Kouni, M. H. Modulation of 5-Fluorouracil Host-Toxicity
and Chemotherapeutic Efficacy Against Human Colon Tumors by 5-
(Phenylthio)acyclouridine, a Uridine Phosphorylase Inhibitor. Cancer
Chemother. Pharmacol. 2006, 58, 692−698.
(19) Wan, L.; Cao, D.; Zeng, J.; Yan, R.; Pizzorno, G. Modulation of
Uridine Phosphorylase Gene Expression by Tumor Necrosis Factor-α
Enhances the Antiproliferative Activity of the Capecitabine Inter-
mediate 5′-Deoxy-5-fluorouridine in Breast Cancer Cells. Mol.
Pharmacol. 2006, 69, 1389−1395.
(20) al Safarjalani, O. N.; Zhou, X.; Rais, R. H.; Shi, J.; Schinazi, R. F.;
Naguib, F. N. M.; el Kouni, M. H. 5-(Phenylthio)acyclouridine: a
Powerful Enhancer of Oral Uridine Bioavailability: Relevance to
Chemotherapy with 5-Fluorouracil and Other Uridine Rescue
Regimens. Cancer Chemother. Pharmacol. 2005, 55, 541−551.
(21) Pizzorno, G.; Yee, L.; Burtness, B. A.; Marsh, J. C.; Darnowski, J.
W.; Chu, M. Y. W.; Chu, S. H.; Chu, E.; Leffert, J. J.;
Handschumacher, R. E.; Calabresi, P. Phase I Clinical and
Pharmacological Studies of Benzylacyclouridine, a Uridine Phosphor-
ylase Inhibitor. Clin. Cancer Res. 1998, 4, 1165−1175.
(22) Ashour, O. M.; Naguib, F. N. M.; Goudgaon, N. M.; Schinazi, R.
F.; el Kouni, M. H. Effect of 5-(Phenylselenenyl)acyclouridine, an
Inhibitor of Uridine Phosphorylase, on Plasma Concentration of
Uridine Released from 2′,3′,5′-Tri-O-acetyluridine, a Prodrug of
Uridine: Relevance to Uridine Rescue in Chemotherapy. Cancer
Chemother. Pharmacol. 2000, 46, 235−240.
This work was supported by the National Institute of Science
and Technology on Tuberculosis, (DECIT/SCTIE/MS-MCT-
CNPq-FNDCT-CAPES) awarded to D.S.S. and L.A.B. L.A.B.
and D.S.S. also acknowledge financial support awarded by
FAPERGS-CNPq-PRONEX-2009. L.A.B., D.S.S., O.N.S.,
A.A.S., and R.R. are Research Career Awardees of the National
Research Council of Brazil (CNPq). R.R. acknowledges funds
awarded to the National Institute for Translational Medicine.
R.R. is also supported by CNPq research grants 303703/2009-1
and 484185/2012-8. D.R. and L.A.R. are recipients of a Ph.D.
scholarship awarded by CNPq.
Notes
The authors declare no competing financial interest.
ABBREVIATIONS USED
■
BAU, 5-benzylacyclouridine or 5-benzyl-1-(2′-
hydroxyethoxymethyl)uracil; FBS, fetal bovine serum; 5-FU,
5-fluorouracil; FUTP, fluorouridine 5′-triphosphate; MTT, 3-
(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; Pi,
inorganic phosphate; ITC, isothermal titration calorimetry;
Urd, uridine; UP, uridine phosphorylase; TS, thymidylate
synthase
REFERENCES
■
(1) Connolly, G. P.; Duley, J. A. Uridine and its Nucleotides:
Biological Actions, Therapeutic Potentials. Trends Pharmacol. Sci.
1999, 20, 218−225.
(2) Cardini, C. E.; Caputto, R.; Paladini, A. C.; Leloir, L. F. Uridine
Diphosphate Glucose: The Coenzyme of the Galactose−Glucose
Phosphate Isomerization. Nature 1950, 165, 191−192.
(3) Cardini, C. E.; Leloir, L. F.; Chiriboga, J. The Biosynthesis of
Sucrose. J. Biol. Chem. 1955, 214, 149−155.
(4) Persson, B. E.; Sjoman, M.; Niklasson, F.; Ronquist, G. Uridine,
̈
Xanthine and Urate Concentrations in Prostatic Fluid and Seminal
Plasma of Patients with Prostatitis. Eur. Urol. 1991, 19, 253−256.
(5) Peters, G. J.; van Groeningen, C. J.; Laurensse, E. J.; Lankelma, J.;
Leyva, A.; Pinedo, H. M. Uridine-Induced Hypothermia in Mice and
Rats in Relation to Plasma and Tissue Levels of Uridine and its
Metabolites. Cancer Chemother. Pharmacol. 1987, 20, 101−108.
(6) Connolly, G. P.; Harrison, P. J.; Stone, T. W. Action of Purine
and Pyrimidine Nucleotides on the Rat Superior Cervical Ganglion. Br.
J. Pharmacol. 1993, 110, 1297−1304.
(23) Gasser, T.; Moyer, J. D.; Handschumacher, R. E. Novel Single-
Pass Exchange of Circulating Uridine in Rat Liver. Science 1981, 213,
777−778.
(24) Cardine, C. E.; Paladini, A. C.; Caputto, R.; Leloir, L. F. Liver
Uridine Phosphorylase. Acta Physiol. Lat. Am. 1950, 1, 57−63.
(25) Liu, M.; Cao, D.; Russel, R.; Handschumacher, R. E.; Pizzorno,
G. Expression, Characterization, and Detection of Human Uridine
Phosphorylase and Identification of Variant Phosphorolytic Activity in
Selected Human Tumors. Cancer Res. 1998, 58, 5418−5424.
(26) Johansson, M. Identification of a Novel Human Uridine
Phosphorylase. Biochem. Biophys. Res. Commun. 2003, 307, 41−46.
(27) Renck, D.; Ducati, R. G.; Palma, M. S.; Santos, D. S.; Basso, L.
A. The Kinetic Mechanism of Human Uridine Phosphorylase 1:
Towards the Development of Enzyme Inhibitors for Cancer
Chemotherapy. Arch. Biochem. Biophys. 2010, 497, 35−42.
(28) Roosild, T. P.; Castronovo, S.; Fabbiani, M.; Pizzorno, G.
Implications of the Structure of Human Uridine Phosphorylase 1 on
the Development of Novel Inhibitors for Improving the Therapeutic
Window of Fluoropyrimidine Chemotherapy. BMC Struct. Biol. 2009,
16, 9−14.
(7) Choi, J. W.; Shin, C. Y.; Choi, M. S.; Yoon, S. Y.; Ryu, J. H.; Lee,
J.; Kim, W.; el Kouni, M. H.; Ko, K. H. Uridine Protects Cortical
Neurons from Glucose Deprivation-Induced Death: Possible Role of
Uridine Phosphorylase. J. Neurotrauma 2008, 25, 695−707.
(8) Leyva, A.; van Groeningen, C. J.; Kraal, I.; Gall, H.; Peters, G. J.;
Lankelma, J.; Pinedo, H. M. Phase I and Pharmacokinetic Studies of
High-Dose Uridine Intended for Rescue from 5-Fluorouracil Toxicity.
Cancer Res. 1984, 44, 5928−5933.
(9) Martin, D. S.; Stolfi, R. L.; Sawyer, R. C.; Spiegelman, S.; Young,
C. W. High-Dose 5-Fluorouracil with Delayed Uridine “Rescue” in
Mice. Cancer Res. 1982, 42, 3964−3970.
(10) al Safarjalani, O. N.; Rais, R.; Naguib, F. N. M.; el Kouni, M. H.
Potent Combination Therapy for Human Breast Tumors with High
Doses of 5-Fluorouracil: Remission and Lack of Host Toxicity. Cancer
Chemother. Pharmacol. 2012, 69, 1449−1455.
(11) Heidelberger, C.; Chaudhhuri, N. K.; Dannerberg, P.; Mooren,
P.; Griesbach, L.; Duschinsky, R.; Schnitzer, R. J.; Pleven, E.; Scheiner,
J. Fluorinated Pyrimidines, a New Class of Tumor-Inhibitory
Compounds. Nature 1957, 179, 663−666.
(29) Niedzwicki, J. G.; el Kouni, M. H.; Chu, S. H.; Cha, S.
Pyrimidine Acyclonucleosides, Inhibitors of Uridine Phosphorylase.
Biochem. Pharmacol. 1981, 15, 2097−2101.
8901
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902

Journal of Medicinal Chemistry
Article
(30) Naguib, F. N. M.; Levesque, D. L.; Wang, E.; Panzica, R. P.; el
Kouni, M. H. 5-Benzylbarbituric Acid Derivatives, Potent and Specific
Inhibitors of Uridine Phosphorylase. Biochem. Pharmacol. 1993, 46,
1273−1283.
(31) Goudgaon, N. M.; Naguib, F. N. M.; el Kouni, M. H.; Schinazi,
R. F. Phenylselenenyl- and Phenylthio-substituted Pyrimidines as
Inhibitors of Dihydrouracil Dehydrogenase and Uridine Phosphor-
ylase. J. Med. Chem. 1993, 39, 4250−4254.
(32) Cha, S. Development of Inhibitors of Pyrimidine Metabolism.
Yonsei Med. J. 1989, 30, 315−326.
(33) Temmink, O. H.; de Bruin, M.; Turksma, A. W.; Cricca, S.;
Laan, A. C.; Peters, G. J. Activity and Substrate Specificity of
Pyrimidine Phosphorylase and their Role in Fluoropyrimidine
Sensitivity in Colon Cancer Cell Lines. Int. J. Biochem. Cell Biol.
2007, 39, 565−575.
(34) Lima, L. M.; Barreiro, E. L. Bioisosterism: A Useful Strategy for
Molecular Modification and Drug Design. Curr. Med. Chem. 2005, 12,
23−49.
(35) Roosild, T. P.; Castronovo, S. Active Site Conformational
Dynamics in Human Uridine Phosphorylase 1. PLoS One 2010, 5, 1−
6.
(36) Fleming, F. F.; Yao, L.; Ravikumar, P. C.; Funk, L.; Shook, B. C.
Nitrile-Containing Pharmaceuticals: Efficacious Roles of the Nitrile
Pharmacophore. J. Med. Chem. 2010, 53, 7902−7917.
(37) Treiber, N.; Schulz, G. E. Structure of 2,6-Dihydroxypyridine 3-
Hydroxylase from a Nicotine-Degrading Pathway. J. Mol. Biol. 2008,
379, 94−104.
(38) Hirohashi, M.; Kido, M.; Yamamoto, Y.; Kojima, Y.; Jitsukawa,
K.; Fujii, S. Synthesis of 5-Fluorouracil Derivatives Containing an
Inhibitor of 5-Fluorouracil Degradation. Chem. Pharm. Bull. 1993, 41,
1498−1506.
(39) Mijin, D. Z.; Uscumlic, G. S.; Valentic, N. V. Synthesis and
Investigation of Solvent Effects on the Ultraviolet Absorption Spectra
of 5-Substituted-4-methyl-3-cyano-6-hydroxy-2-pyridones. J. Serb.
Chem. Soc 2001, 66, 507−516.
(48) Silva, R. G.; Vetticatt, M. J.; Merino, E. F.; Cassera, M. B.;
Schramm, V. L. Transition-State Analysis of Trypanosoma cruzi
Uridine Phosphorylase-Catalyzed Arsenolysis of Uridine. J. Am.
Chem. Soc. 2011, 133, 9923−9931.
(49) Delia, T. J.; Scovill, J. P.; Munslow, W. D.; Burckhalter, J. H.
Synthesis of 5-Substituted Aminomethyluracils via the Mannich
Reaction. J. Med. Chem. 1976, 19, 344−346.
(50) Copeland, R. A. Evaluation of Enzyme Inhibitors in Drug
Discovery: A Guide for Medicinal Chemists and Pharmacologists, 1st ed.;
John Wiley & Sons: Hoboken, NJ, 2005; pp 48−81.
(51) Schon, A.; Lam, S. Y.; Freire, E. Thermodynamics-Based Drug
̈
Design: Strategies for Inhibiting Protein−Protein Interactions. Future
Med. Chem. 2011, 3, 1129−1137.
(52) Pizzorno, G.; Cao, D.; Leffert, J. J.; Russell, R. L.; Zhang, D.;
Handschumacher, R. E. Homeostatic Control of Uridine and the Role
of Uridine Phosphorylase: A Biological and Clinical Update. Biochim.
Biophys. Acta 2002, 1587, 133−144.
(53) Khoshniat, S.; Bourgine, A.; Julien, M.; Weiss, P.; Guicheux, J.;
Beck, L. The Emergence of Phosphate as a Specific Signaling Molecule
in Bone and Other Cell Types in Mammals. Cell. Mol. Life Sci. 2011,
68, 205−218.
(54) Traut, T. W. Physiological Concentrations of Purines and
Pyrimidines. Mol. Cell. Biochem. 1994, 140, 1−22.
(55) Griffiths, L.; Stratford, I. J. Platelet-Derived Endothelial Cell
Growth Factor Thymidine Phosphorylase in Tumour Growth and
Response to Therapy. Br. J. Cancer 1997, 76, 689−693.
(56) Perozzo, R.; Folkers, G.; Scapozza, L. Thermodynamics of
Protein−Ligand Interactions: History, Presence, and Future Aspects. J.
Recept. Signal Transduction Res. 2004, 24, 1−52.
(57) Eftink, M. R.; Anusiem, A. C.; Biltonen, R. L. Enthalpy−Entropy
Compensation and Heat Capacity Changes for Protein−Ligand
Interactions: General Thermodynamic Models and Data for the
Binding of Nucleotides to Ribonuclease A. Biochemistry 1983, 22,
3884−3896.
(58) Ladbury, J. E.; Klebe, G.; Freire, E. Adding Calorimetric Data to
Decision Making in Lead Discovery: A Hot Tip. Nature Rev. Drug
Discovery 2010, 9, 23−27.
(59) Kawasaki, Y.; Freire, E. Finding a Better Path to Drug
Selectivity. Drug Discovery Today 2011, 16, 985−990.
(60) Copeland, R. A. Enzymes: A Practical Introduction to Structure,
Mechanism, and Data Analysis, 2nd ed.; John Wiley & Sons: Hoboken,
NJ, 2000; pp 367−383.
(61) Temmink, O. H.; de Bruin, M.; Turksma, A. W.; Cricca, S.;
Laan, A. C.; Peters, G. J. Activity and Substrate Specificity of
Pyrimidine Phosphorylases and Their Role in Fluoropyrimidine
Sensitivity in Colon Cancer Cell Lines. Int. J. Biochem. Cell Biol.
2007, 39, 565−575.
(62) Mans, D. R. A.; Grivicich, I.; Peters, G. J.; Schwartsmann, G.
Sequence-Dependent Growth Inhibition and DNA Damage For-
mation by the Irinotecan-5-fluorouracil Combination in Human Colon
Carcinoma Cell Lines. Eur. J. Cancer 1999, 35, 1851−1861.
(63) Robertson, J. G. Enzymes as a Special Class of Therapeutical
Target: Clinical Drugs and Modes of Action. Curr. Opin. Struct. Biol.
2007, 17, 674−679.
(64) Bradford, M. M. A Rapid and Sensitive Method for the
Quantitation of Microgram Quantities of Protein Utilizing the
Principle of Protein−Dye Binding. Anal. Biochem. 1976, 72, 248−254.
(65) Lineweaver, H.; Burk, D. The Determination of Enzyme
Dissociation Constant. J. Am. Chem. Soc. 1934, 56, 658−666.
(66) Copeland, R. A. Enzymes: A Practical Introduction to Structure,
Mechanism, and Data Analysis, 2nd ed.; John Wiley & Sons, Inc.:
Hoboken, NJ, 2000; pp 266−304.
(67) Morris, G. M.; Huey, R.; Lindstrom, W.; Sanner, M. F.; Belew,
R. K.; Goodsell, D. S.; Olson, A. J. AutoDock4 and AutoDockTools4:
Automated Docking with Selective Receptor Flexibility. J. Comput.
Chem. 2009, 30, 2785−2791.
(68) Goodsell, D. S.; Olson, A. J. Automated Docking of Substrates
to Proteins by Simulated Annealing. Proteins 1990, 8, 195−202.
(40) Bobbitt, J. M.; Scola, D. A. Synthesis of Isoquinoline Alkaloids.
II. The Synthesis and Reactions of 4-Methyl-3-pyridinecarboxaldehyde
and other 4-Methyl-3-substituted Pyridines. J. Org. Chem. 1960, 25,
560−564.
́
(41) Pevet, I.; Brule, C.; Tizot, A.; Gohier, A.; Cruzalegui, F.; Boutin,
J. A.; Goldstein, S. Synthesis and Pharmacological Evaluation of
Thieno[2,3-b]pyridine Derivatives as Novel c-Src Inhibitors. Bioog.
Med. Chem. 2011, 19, 2517−2528.
(42) Portnoy, S. Fluorinated Nitrogen Heterocycles via Cyclization. I.
Trifluoromethyl-2-pyridones from Fluorinated 1,3-Dicarbonyls and
Cyanoacetamide. J. Org. Chem. 1965, 30, 3377−3380.
(43) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P.
Development and Use of Quantum Mechanical Molecular Models. 76.
AM1: A New General Purpose Quantum Mechanical Molecular
Model. J. Am. Chem. Soc. 1985, 107, 3902−3909.
(44) Semeraro, T.; Lossani, A.; Botta, M.; Ghiron, C.; Alvarez, R.;
Manetti, F.; Mugnaini, C.; Valensin, S.; Focher, F.; Corelli, F.
Simplified Analogues of Immucillin-G Retain Potent Human Purine
Nucleoside Phosphorylase Inhibitory Activity. J. Med. Chem. 2006, 49,
6037−6045.
(45) Taylor, E. A.; Clinch, K.; Kelly, P. M.; Li, L.; Evans, G. B.; Tyler,
P. C.; Schramm, V. L. Acyclic Ribooxacarbenium Ion Mimics as
Transition State Analogues of Human and Malarial Purine Nucleoside
Phosphorylases. J. Am. Chem. Soc. 2007, 129, 6984−6985.
(46) Clinch, K.; Evans, G. B.; Frohlich, R. F. G.; Furneaux, R. H.;
̈
Kelly, P. M.; Legentil, L.; Murkin, A. S.; Li, L.; Schramm, V. L.; Tyler,
P. C.; Woolhouse, A. D. Third-Generation Immucillins: Syntheses and
Bioactivities of Acyclic Immucillin Inhibitors of Human Purine
Nucleoside Phosphorylase. J. Med. Chem. 2009, 52, 1126−1143.
(47) Caradoc-Davies, T. T.; Cutfield, S. M.; Lamont, I. L.; Cutfield, J.
F. Crystal Structure of Escherichia coli Uridine Phosphorylase in Two
Native and Three Complexed Forms Reveal Basis of Substrate
Specificity, Induced Conformational Changes and Influence of
Potassium. J. Mol. Biol. 2004, 337, 337−354.
8902
dx.doi.org/10.1021/jm401389u | J. Med. Chem. 2013, 56, 8892−8902