Structure-activity relationship of 2,4,5-trioxoimidazolidines as inhibitors of thymidine phosphorylase

Article abstract of DOI:10.1016/j.ejmech.2011.01.035

Novel non-nucleobase-derived inhibitors of the angiogenic enzyme, thymidine phosphorylase, have been identified using molecular modelling, synthesis and biological evaluation. These inhibitors are 2,4,5-trioxoimidazolidines bearing N-(substituted)phenylalkyl groups, together with, in most cases, N′-(CH₂)_n-carboxylic acid, ester or amide side chains. The best compound from this series is 3-(2,4,5-trioxo-3-phenylethyl- imidazolodin-1-yl)propionamide, with an IC₅₀ of 40 μM against Escherichia coli TP. Molecular modelling suggests that this ligand, when complexed with closed-cleft human TP, would have the phenylalkyl group in the active site region normally occupied by a thymine-containing structure.

Full text of DOI:10.1016/j.ejmech.2011.01.035

European Journal of Medicinal Chemistry 46 (2011) 1165e1171
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Structureeactivity relationship of 2,4,5-trioxoimidazolidines as inhibitors
of thymidine phosphorylase
*
**
Mehdi Rajabi, David Mansell, Sally Freeman , Richard A. Bryce
School of Pharmacy and Pharmaceutical Sciences, University of Manchester, Oxford Road, Manchester M13 9PT, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Novel non-nucleobase-derived inhibitors of the angiogenic enzyme, thymidine phosphorylase, have been
identified using molecular modelling, synthesis and biological evaluation. These inhibitors are 2,4,5-triox-
oimidazolidines bearingN-(substituted)phenylalkyl groups, together with, in most cases, N0-(CH2)n-carboxylic
acid, ester or amide side chains. The best compound from this series is 3-(2,4,5-trioxo-3-phenylethyl-imida-
zolodin-1-yl)propionamide, with an IC50 of 40 mM against Escherichia coli TP. Molecular modelling suggests
that this ligand, when complexed with closed-cleft human TP, would have the phenylalkyl group in the active
site region normally occupied by a thymine-containing structure.
Received 3 August 2010
Received in revised form
14 October 2010
Accepted 22 January 2011
Available online 31 January 2011
Keywords:
Ó 2011 Elsevier Masson SAS. All rights reserved.
Thymidine phosphorylase
Trioxoimidazolidine
Parabanic acid
Hydantoin
Anti-angiogenic
Molecular modelling
1. Introduction
which 2-nitroimidazolyluracil prodrugs (5A) require bioactivation
under tumour conditions to form the active species, 20-amino-
Thymidine phosphorylase (TP, dThPase, E.C. 2.4.2.4), also known
as the platelet-derived endothelial cell growth factor (PD-ECGF),
catalyses the phosphorolysis of thymidine (1) to 2-deoxyribose-1-
phosphate (2) and thymine (3) (Fig. 1) [1]. TP is highly expressed in
manysolid humantumours, facilitatingtumour growthbypromoting
angiogenesis, metastasis and suppressing apoptosis [2e5].
imidazolyluracils (5B) (Fig. 2) [9e11], one of which demonstrated
activity similar to TPI (apparent IC50 0.023 M against human TP)
m
[10,11]. These zwitterionic tight-binding inhibitors are proposed to
mimic parts of the oxacarbenium ion-like transition state formed
during thymidine phosphoroylsis, see Fig. 1 [12]. Despite the
potency levels of these inhibitors, their highly ionic nature and poor
pharmacokinetic profiles remain as substantial limitations.
Recently, novel scaffolds for TP inhibitors have been targeted
with the aid of 3D-homology models of human pyrimidine nucle-
oside phosphorylase and Escherichia coli TP [12,13]. We have
previously employed structure-based virtual screening of the
National Cancer Institute (NCI) database against the open confor-
mation of the homology model of human TP [9] based on a crystal
structure of the E. coli enzyme [13]. This study identified hydantoin
7 (Fig. 3) as a TP inhibitor lead with micromolar activity against
human and E.coli TP [13], comparable to that of the known, non-
uracil-based TP inhibitor, 7-deazaxanthine (6). This novel scaffold
lacks the undesirable ionic sites of the existing tight-binding
nucleobase-derived inhibitors and therefore may possess improved
pharmacokinetic and cell-penetrating properties.
Furthermore, intracellular hydrolysis of 2 generates 2-deoxy-D-
ribose [6] which promotes angiogenesis, the chemotactic activity of
endothelial cells and also confers resistance to hypoxia-induced
apoptosis in some cancer cell lines [2,7]. Thus, TP is considered an
attractive therapeutic target for inhibition of tumour angiogenesis
and concomitant tumour growth and metastasis [5].
Almost all literature TP inhibitors are based on uracil or very
closely related ring systems. One of the most potent inhibitors, 5-
chloro-6-[1-(2-iminopyrrolidinyl)methyl]uracil hydrochloride (4,
TPI, Fig. 2), caused a reduction in the rate of tumour growth when
administered to mice carrying tumours that over-expressed human
TP [8]. Prodrugs of inhibitors of TP have also been synthesised in
Based on our knowledge of this hydantoin (2,4-dioxoimidazo-
lidine) lead, we here employ molecular modelling, synthesis and
biological assays to identify further active non-nucleoside lead
scaffolds as a basis for development of novel TP inhibitors.
* Corresponding author. Tel.: þ44 161 275 2366; fax: þ44 161 275 2396.
** Corresponding author. Tel.: þ44 161 275 8345; fax: þ44 161 275 2396.
E-mail addresses: Sally.Freeman@manchester.ac.uk (S. Freeman), Richard.Bryce@
manchester.ac.uk (R.A. Bryce).
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.01.035

1166
M. Rajabi et al. / European Journal of Medicinal Chemistry 46 (2011) 1165e1171
Fig. 1. Phosphorolysis of thymidine (1) to 2-deoxyribose-1-phosphate (2) and thymine (3).
2. Results and discussion
hydrolysis of esters 22e25 gave the 3-benzyl-2,4,5-trioxo-imida-
zolidin-1-yl-ethanoic acids 9 and 26e28 in good yields. The novel
amide analogues 29e32 were obtained by reaction of the imida-
zolidine-2,4,5-trione derivatives 18e21 with 3-bromopropiona-
mide (Scheme 1). Amides 29e32 were fully characterized by 1H and
13C NMR spectroscopy, elemental analysis and mass spectrometry.
In an initial modelling step, the NCI and Available Chemicals
Directory (ACD) databases were filtered to identify all imidazoli-
dine-containing compounds [14]. The identified compounds were
then screened in silico against an X-ray structure of closed-cleft
human TP [15], in which phosphate was modelled into the active
site; these compounds were ranked according to calculated dock-
ing scores [14]. Interestingly, the top-ranked NCI ligand from this
screen was the hydantoin, 4-hexyl-1-methyl-2,5-dioxo-4-imida-
zolidinecarbaldehyde semicarbazone (8). This ligand is very similar
to the previously identified hydantoin derivative 7 [13], the only
difference being the presence of an extra methylene group in the
aliphatic chain. The top-ranked ACD ligand from the screen was [3-
(2-methylbenzyl)-2,4,5-trioxo-imidazolidin-1-yl]ethanoic acid (9).
Its complex with TP was predicted to be more stable (by 36.4 kJ/
mol) than that of compound 8 [14].
2.2. Structureeactivity relationship
Compound 9 and 18e32 were tested for their TP inhibitory
activity using E. coli TP. A spectrophotometric assay was used to
measure the decrease in absorbance (at 265 nm) of the natural
substrate thymidine upon the addition of the compounds [10,20].
The experimental IC50 values are presented alongside their calcu-
lated docking scores in Table 1, together with the data for the
inhibitor TPI (4) as a positive control.
The data presented in Table 1 for the 1-arylalkyl-2,4,5-triox-
oimidazolidines (18e21) indicate that there is little difference in TP
inhibition activity between these compounds. This indicates that
the nature of substituent, X, on the phenyl ring has a rather weak
influence on the overall binding affinity of the ligands. Corre-
spondingly, these groups do not appear to interact directly with the
protein in the predicted binding poses. This is also evident from
their similar calculated energy scores. Indeed, compounds 18e21
all bind in very similar conformations with good interactions
formed between the imidazolidine-2,4,5-trione ring and Ser217
and Arg202 residues, with the phenyl ring located in the centre of
the active site.
2.1. Chemistry
Based on this predicted preference of imidazolidines for the
active site of TP, the chemical space surrounding the highest scoring
compound, 9, was explored: a series of 3-arylalkyl-2,4,5-triox-
oimidazolidine-1-ethanoic acids (parabanic acid derivatives),
including 9, and their corresponding esters and amides were
prepared (Scheme 1) and tested for their TP inhibitory activities
(Table 1). We note that some of these compounds have previously
been investigated for their therapeutic potential for reduction of
diabetic complications such as neuropathy, nephropathy, retinop-
athy, keratopathy, angiopathy and cataracts by selectively inhibit-
ing the enzyme aldose reductase [16e18].
The general synthetic route for the acids and their esters is
shown in Scheme 1 and is based on chemistry developed by Ishii
and coworkers [19]. The synthesis starts from benzylamine deriv-
atives 10e13, which are reacted with urea in the presence of acid to
yield N-(benzyl)urea derivatives 14e17. Treatment of the N-
(benzyl)urea derivatives 14e17 with oxalyl chloride gave the 1-
(benzyl)-imidazolidine-2,4,5-triones 18e21 in good yields. The
ethyl [3-benzyl-2,4,5-trioxo-imidazolidin-1-yl]ethanoates 22e25
were then prepared by reaction of 18e21 with ethyl bromoetha-
noate in the presence of potassium hydroxide. Acid-catalysed
Ligands 22e32 examine the effect of the presence of side chains
opposite the lipophilic aryl substituent (Table 1). The ester groups
in ligands 22e25 constitute hydrogen bond acceptors and the ethyl
groups have the ability to make lipophilic interactions. The pres-
ence of the ester functionality generally decreased TP inhibition
when compared with other analogues in Table 1. Replacing the
ester functional group with the carboxylic acid group led to
a
marked improvement in the calculated GoldScore value,
comparable to the value obtained for TPI (Table 1). The most stable
binding conformation predicted for compound 9, which was the
top-ranked ligand in the original screen of the ACD database and
displayed the highest calculated score (ꢀ72.2 kJ/mol) of all the
Fig. 3. Hydantoins 7 and 8 and trioxoimidazolidine 9 were TP inhibitors identified by
in silico screening of NCI and ACD databases.
Fig. 2. Examples of known TP inhibitors.

M. Rajabi et al. / European Journal of Medicinal Chemistry 46 (2011) 1165e1171
1167
group on the imidazole ring and the Thr154 residue at the centre of
the active site (O/HeN backbone distance of 1.9 Å, Fig. 5). This
orientationwas also observed in compound 29, but not in amides 30
and 31, which possess bulkier substituted phenyl rings. Therefore,
the top-ranked docking pose would suggest that the interaction
between the amide functionality in 29 and 32 with the Asp123 and
Thr145 drives the flip in binding orientation, whereas in 30 and 31
this orientation is energetically unfavourable due to the increased
steric bulk on the phenyl ring. Summarising, in the amide series,
substitution on the phenyl ring appears to have a more significant
influence on the predicted binding orientation of the ligand, when
compared with either the ester or carboxylic acid derivatives.
3. Conclusions
In silico screening of the NCI and ACD databases identified
several inhibitors of human TP [14]. From this, a series of 3-ary-
lalkyl-2,4,5-trioxoimidazolidine-1-ethanoic acids (parabanic acid
derivatives) and their corresponding esters and amides (Table 1)
were prepared and their biological activity evaluated against E.coli
TP. In this series, compounds 32, 31 and 28 showed the greatest
E.coli TP inhibition with IC50 values of 40, 66 and 85 mM, respec-
tively. These inhibitors are unrelated to the molecular framework of
the thymine substrate or any previously reported ligands, and
therefore constitute a new direction for design and synthesis of
novel TP inhibitors with potentially improved pharmacokinetic and
cell-penetrating properties.
4. Experimental
4.1. Molecular modelling
All docking/screening studies employed the X-ray crystal
structure of human TP in complex with the inhibitor TPI (4) and in
the absence of phosphate (resolution 2.1 Å, PDB code 1UOU) [15].
TPI was removed and positioning of the active site phosphate
molecule was performed via analogy with the crystal structure of
Bacillus stearothermophilus pyrimidine nucleoside phosphorylase
(PyNP), in which the phosphate group is present [22]. The location
of hydrogen atoms was modelled into the structure using SYBYL 6
(Tripos Inc.: St Louis, MO, USA 2003). The phosphate was modelled
in its dihydrogen form. We note that in silico docking of TPI
reproduced its crystallographically observed binding mode [15]:
the nucleobase interacted via hydrogen bonds with residues
including His116 (not shown), Ser217, Arg202 and Lys221. This is
the region of the active site that binds the nucleobase of the
thymidine substrate. TPI’s chloro-substituent projected into a lipo-
philic pocket formed by residues Leu148, Val208, Ile214 and
Val241; the cationic NH2 group of the aminopyrrolidinium ring
interacted with the carbonyl oxygen of Ser117 and an oxygen atom
of the phosphate co-ligand. Three-dimensional structures of
ligands 9 and 18e33 were assigned using ViewerLite (Accelrys
Software Inc.). Compounds 9 and 26e28 were modelled as their
carboxylate anions. Initial screening of the NCI and ACD databases
was performed using flexible ligand docking via the program
DOCK4 [23] using Gasteiger-Marsili partial charges for the protein
atoms [24] and ligand atomic charges consistent with the MMFF94s
force field [25]. Subsequent analysis of structureeactivity data was
performed using GOLD and the GoldScore scoring function [26].
Scheme 1. Synthesis of 2,4,5-trioxoimidazolidin-1-yl derivatives (9 and 18e32): (a)
Urea, HCl,
acid, HCl,
D
; (b) Oxalyl chloride, THF; (c) Ethyl bromoacetate, KOH, EtOH,
D; (d) Acetic
D
; (e) 3-Bromopropionamide, KOH, EtOH,
D.
ligands in this study, is shown in Fig. 4 and is representative of the
other carboxylic acids derivatives (26e28). The carboxylate func-
tionality appeared to form strong interactions with the Arg202 and
Ser217 residues (Fig. 4), the same region with which TPI interacts.
Consequently, these ligands and in particular compound 9, would
be predicted to possess the most potent TP inhibition activity.
However, the novel amide derivatives (29e32) generally displayed
better IC50 values compared to other ligands in the study, with the
IC50 values obtained for 31 and 32 being comparable to 7-deazax-
anthine [21]. Amongst the carboxylate and amide derivatives, the
ligands for which n ¼ 1 (eg. 29e31 for the amides) appear to be too
short to effectively span the active site, from Arg202 to Asp123. In
contrast, ligands 28 and 32, where n ¼ 2, appear to have the
dimensions to achieve this. The greater level of interaction for n ¼ 2
compounds is illustrated by the 2-fold difference in potency for
compounds 29 and 32; or for the carboxylate series, by the 3-fold
difference between compounds 26 and 28.
The superior experimental inhibitory effect exhibited by the
amide derivatives was not reflected in the predicted values obtained
for their binding energy. Interestingly however, the predicted top-
scoring binding orientation of compound 32, the most potent
inhibitor of all the compounds studied (Fig. 5), was inverted with
respect to the carboxylate-containing compounds (Fig. 4): in this
case, the phenyl ring was positioned upwards towards the hydro-
phobic pocket at the top of the active site due to interactions formed
between the amide group and (i) the Asp123 residue (ligand
NeH/O Asp123 backbone distance of 1.8 Å, Fig. 5), and (ii) the
phosphate hydroxyl group at the bottom of the active site. In
addition, a strong interaction was observed between the 5-oxo
4.2. Instrumentation and chemicals
NMR spectra were recorded on a Bruker Avance 300 MHz
spectrometer. 1H and 13C NMR spectra were reported as dH parts per
million (ppm) downfield from tetramethylsilane. The 13C NMR

1168
M. Rajabi et al. / European Journal of Medicinal Chemistry 46 (2011) 1165e1171
Table 1
Interaction energies and E. coli TP inhibition data for 1-arylalkyl-2,4,5-trioxoimidazolodine analogues (18e21), and 1-arylalkyl-3-substituted-2,4,5-trioxoimidazolodine esters
(22e25), carboxylic acids (9 and 26e28) and amides (29e32). Docking scores are calculated using GoldScore (kJ/mol).
Compound
X
n
m
R
Score
IC50 (mM)
TPI (4)
18
19
20
21
22
23
24
25
26
9
27
28
29
30
31
32
e
e
1
1
1
2
1
1
1
2
1
1
1
2
1
1
1
2
e
e
e
e
e
1
1
1
1
1
1
1
1
2
2
2
2
e
ꢀ70.1
ꢀ49.0
ꢀ51.4
ꢀ52.9
ꢀ48.6
ꢀ58.3
ꢀ56.3
ꢀ58.5
ꢀ59.2
ꢀ69.1
ꢀ72.2
ꢀ67.7
ꢀ69.3
ꢀ54.8
ꢀ60.1
ꢀ54.4
ꢀ49.9
0.036 ꢁ 0.01a
H
e
115 ꢁ 8
2-Me
3-Cl
H
e
128 ꢁ 10
117 ꢁ 14
103 ꢁ 4
e
e
H
OEt
OEt
OEt
OEt
OH
OH
OH
OH
NH2
NH2
NH2
NH2
342 ꢁ 12
303 ꢁ 17
292 ꢁ 43
357 ꢁ 25
264 ꢁ 30
236 ꢁ 26
v. weak
2-Me
3-Cl
H
H
2-Me
3-Cl
H
85 ꢁ 15
H
106 ꢁ 13
88 ꢁ 10
2-Me
3-Cl
H
66 ꢁ 9
40 ꢁ 6
a
represents an upper limit for IC50 [11].
spectra were assigned with the aid of a 13C DEPT NMR spectrum.
Mass spectra were determined by the School of Chemistry,
University of Manchester, using Micromass Trio 2000, chemical
ionisation (CI), using NH3 as an ionising gas. Electrospray (ES) mass
spectra were recorded on a Micromass platform spectrometer
using acetonitrileewater (1:1) as the mobile phase. Melting points
were determined using a Gallenkamp MPD.350.BM2.5 instrument
and remain uncorrected. Reactions were monitored using thin layer
chromatography (TLC) on silica gel plates (precoated F254, Merck
1.05554), spots were visualised using 254 nm UV light, KMnO4 or
Fig. 4. The top-scoring pose of 9 in the active site of crystal structure of human TP.
Fig. 5. The top-ranked pose of 32 in the active site of crystal structure of human TP.
Distances are shown in Å.
Distances are in Å.

M. Rajabi et al. / European Journal of Medicinal Chemistry 46 (2011) 1165e1171
1169
DNP stain. Flash column chromatography was preformed using
Prolabo silica gel 60 (35e75 m particle size), 220e440 mesh.
(AreCH2eN), 61.4 (OCH2) 127.4 (C-4), 127.5 (C-2 and C-6), 128.3 (C-
3 and C-5), 134.9 (C-1), 153.2 (C]O), 156.4 (C]O) and 156.6 (C]O,
urea), 166.4 (C]O, ester).
m
Solvents were distilled from the indicated drying agents, following
standard procedures. Chemicals were obtained from The Aldrich
Chemical Co., Dorset, UK. and Lancaster Synthesis LTD., Lancashire,
UK. Enzyme rate spectral scans, studies of inhibition curves and
absorbance readings at fixed wavelengths were conducted using
a Peltier-thermostatted cuvette holder in a Cary 4000 UVevisible
spectrophotometer equipped with Cary Enzyme Kinetics Software.
4.3.6. Ethyl [3-(2-methyl-benzyl)-2,4,5-trioxo-imidazolidin-1-yl]
ethanoate (23)
Trione 19 (0.22 g, 1 mmol) and ethyl bromoacetate (0.17 ml,
1.54 mmol) were added to a solution of potassium hydroxide
(80 mg, 1.2 mmol) in ethanol (6 ml). After heating at reflux for 9 h,
the mixture was cooled to 0 ꢂC and filtered. The solid was dissolved
in ethyl acetate (5 ml) and washed with water (15 ml) and brine
(10 ml). The organic phase was dried (anhydrous Na2SO4) and
passed through a short silica gel pad. After concentration, the crude
solid was recrystallised from ethyl acetateehexane to give 23
(0.24 g, 79%) as colourless crystals. M.p. 156e159 ꢂC; Lit. M.p.
158e160 ꢂC [19]. IR (KBr, cmꢀ1): 1720 cmꢀ1 (br, C]O). 1H NMR
4.3. Synthesis
4.3.1. 1-Benzyl-imidazolidine-2,4,5-trione (18)
Trione 18 was prepared as described for 19 using benzylurea
(14): colourless powder, yield: 88%. M.p. 168e171 ꢂC; Lit. M.p.
169e170 ꢂC [19]. IR (KBr) cmꢀ1: 3150 (NH), 1720 (C]O). 1H NMR
(DMSO-d6)
d
: 4.63 (2H, s, CH2), 7.28e7.46 (5H, m, AreH), 12.10 (1H,
: 41.8 (eCH2e), 127.9 (C-4), 128.5 (C-2
(DMSO-d6)
d
: 1.20 (3H, t, 3JHH ¼ 7.1 Hz, CH3), 2.34 (3H, s, AreCH3),
s, NH). 13C NMR (DMSO-d6)
d
4.15 (2H, q, 3JHH ¼ 7.1 Hz, OCH2), 4.43 (2H, s, NCH2CO2), 4.73 (2H, s,
and C-6), 129.3 (C-3 or C-5), 136 (C-1), 154.8 (C]O), 158.6 (C]O),
159.2 (C]O).
AreCH2), 7.13e7.26 (4H, m, AreH). 13C NMR (DMSO-d6)
d: 12.8
(AreCH3), 17.7 (CH2eCH3), 39.8 (AreCH2), 40.5 (NCH2CO), 60.6
(OCH2) 124.9 (C-5), 126.5 (C-4), 126.6 (C-6), 129.1 (C-3), 131.9 (C-2),
134.6 (C-1), 152.4 (C]O), 155.5 (C]O) and 155.9 (C]O, amide),
166.0 (C]O, ester).
4.3.2. 1-(2-Methylbenzyl)-imidazolidine-2,4,5-trione (19)
Oxalyl chloride (0.10 ml, 1.2 mol) was added dropwise to
a suspension of N-(2-methylbenzyl)urea (15) (0.164 g, 1 mmol) in
tetrahydrofuran (3 ml) at 0 ꢂC. The mixture was warmed to room
temperature and stirred vigorously for 5 h. After removal of the
precipitate by filtration, the filtrate was concentrated. The solid
residue was dissolved in ethyl acetate (10 ml) and washed with
water (2 ꢃ 10 ml) and then brine (10 ml). The organic layer was
dried (anhydrous Na2SO4) and passed through a short silica gel pad.
The solvent was evaporated. The solid was recrystallised from ethyl
acetateehexane to give 19 as colourless crystals (0.18 g, 82.5%). M.p.
192e194 ꢂC; Lit. M.p. 195e196 ꢂC [19]. IR (KBr) cmꢀ1: 3150 (NH),
4.3.7. Ethyl [3-(3-chloro-benzyl)-2,4,5-trioxo-imidazolidin-1-yl]
ethanoate (24)
The ester was prepared as described for 23 using trione 20
(0.24 g, 1.0 mmol): colourless powder, yield: 62%. M.p. 139e142 ꢂC;
Lit. M.p. 136e137 ꢂC [19]. IR (KBr, cmꢀ1): 1720 (br, C]O). 1H NMR
3
(DMSO-d6)
d
: 1.19 (3H, t, JHH ¼ 7.1 Hz, CH3), 4.15 (2H, q,
3JHH ¼ 7.1 Hz, OCH2), 4.15 (2H, s, NCH2CO2), 4.76 (2H, s, AreCH2),
7.30e7.45 (4H, m, AreH). 13C NMR (DMSO-d6)
d: 14.3 (CH2eCH3),
40.0 (NeCH2), 41.7 (AreCH2), 62.0 (OCH2) 126.6 (C-6), 127.7 (C-4),
128.0 (C-2), 130.8 (C-5), 133.6 (C-3), 138.0 (C-1), 153.8 (C]O), 157.0
(C]O) and 157.2 (C]O), 167.0 (C]O, ester).
1720 (C]O). 1H NMR (DMSO-d6)
CH2), 7.20e7.30 (4H, m, AreH), 12.10 (1H, s, NH). 13C NMR (DMSO-
d6) : 19.1 (AreCH3), 40.4 (eCH2e), 126.2 (C-5), 127.8 (C-4), 127.9
d: 2.33 (3H, s, CH3), 4.61 (2H, s,
d
(C-6), 130.4 (C-3), 133.7 (C-2), 135.8 (C-1), 154.9 (C]O), 158.7 (C]
O), 159.2 (C]O).
4.3.8. Ethyl (2,4,5-trioxo-3-phenethyl-imidazolidin-1-yl)ethanoate
(25)
The ester 25 was prepared as described for 23 using trione 21:
4.3.3. 1-(3-Chloro-benzyl)-imidazolidine-2,4,5-trione (20)
colourless powder, yield: 85%. M.p.106e110 ꢂC; Lit. M.p.106e108 ꢂC
Trione 20 was prepared as described for 19 using 3-chlor-
obenzylurea (16): colourless powder, yield: 84%. M.p. 164e166 ꢂC;
Lit. M.p. 162e163 ꢂC [19]. IR (KBr) cmꢀ1: 3210 (NH), 1720 (C]O). 1H
[19]. IR (KBr, cmꢀ1): 1720 (br, C]O). 1H NMR (DMSO-d6)
d: 1.20 (3H,
t, 3JHH ¼ 7.1 Hz, CH3), 2.89 (2H, t, 3JHH ¼ 7.3 Hz, PheCH2), 3.77 (2H, t,
3JHH ¼ 7.3 Hz, PheCH2eCH2), 4.17 (2H, q, 3JHH ¼ 7.1 Hz, OCH2), 4.52
NMR (DMSO-d6)
d
: 4.65 (s, 2H, CH2), 7.30e7.46 (m, 4H, AreH), 12.10
(2H, s, NeCH2), 7.20e7.32 (5H, m, Ph). 13C NMR (DMSO-d6)
d: 13.8
(s, 1H, NH). 13C NMR (DMSO-d6)
d: 46.4 (eCH2e), 125.3 (C-6), 127.5
(CH3), 33.0 (PheCH2), 39.6 (NeCH2), 40.2 (PheCH2eCH2), 61.5
(OCH2),126.5 (C-4),128.4 (C-2 and C-6),128.6 (C-3 and C-5),137.5 (C-
1), 153.2, 153.3 and 156.4 (C]O), 166.5 (C]O, ester).
(C-4), 128.2 (C-2), 129.4 (C-5), 134.5 (C-3), 135.8 (C-1), 156.7 (C]O),
158.9 (C]O), 159.4 (C]O).
4.3.4. 1-Phenethyl-imidazolidine-2,4,5-trione (21)
4.3.9. (3-Benzyl-2,4,5-trioxo-imidazolidin-1-yl)ethanoic acid (26)
Acid 26 was prepared as described for 9 using 22: colourless
powder, yield: 69%. M.p. 203e208 ꢂC, Lit. M.p. 207.5e209.5 ꢂC [19].
IR (KBr, cmꢀ1): 3200e2800 cmꢀ1 (br, CO2H), 1710 cmꢀ1 (br, C]O).
Trione 21 was prepared as described for 19 using phenethylurea
(17) (0.16 g, 1.0 mmol): colourless powder, yield: 96%. M.p.
189e192 ꢂC; Lit. M.p. 189e190 ꢂC [19]. IR (KBr) cmꢀ1: 3210 (NH),
1720 (C]O). 1H NMR (DMSO-d6)
d
: 2.86 (2H, t, JHH ¼ 7.6 Hz,
1H NMR (DMSO-d6)
d
: 4.22 (2H, s, NeCH2), 4.74 (2H, s, PheCH2),
3
3
PheCH2), 3.66 (2H, t, JHH ¼ 7.3 Hz, CH2eN) 7.20e7.33 (5H, m,
7.26e7.39 (5H, m, Ph), 13.2 (1H, br s, OH). 13C NMR (DMSO-d6)
d:
Phenyl), 12.10 (1H, s, NH). 13C NMR (DMSO-d6)
d
: 32.9 (PheCH2),
40.6 (PheCH2 or NeCH2), 41.5 (PheCH2 or NeCH2), 125.7 (C-4),
126.3 (C-2 and C-6), 127.3 (C-3 and C-5), 133.9 (C-1), 152.2 (C]O),
155.4 (C]O), 155.6 (C]O), 166.7 (COOH).
39.7 (eCH2eN), 126.2 (C-4), 128.1 (C-2 and C-6), 128.2 (C-2 and
C-5), 137.6 (C-1), 153.9 (C]O), 157.6 (C]O), 158.5 (C]O).
4.3.5. Ethyl (3-benzyl-2,4,5-trioxo-imidazolidin-1-yl)ethanoate (22)
The ester was prepared as described for 23 using trione 18:
colourless powder, yield: 89%. M.p.147e150 ꢂC, Lit. M.p.150e151 ꢂC
[19]. IR (KBr, cmꢀ1): 1720 cmꢀ1 (br, C]O). 1H NMR (DMSO-d6)
4.3.10. [3-(2-Methylbenzyl)-2,4,5-trioxo-imidazolidin-1-yl]
ethanoic acid (9)
A mixture of 23 (0.20 g, 0.66 mmol), acetic acid (0.6 ml) and
concentrated hydrochloric acid (0.3 ml) was heated at reflux for 3 h.
The reaction mixture was concentrated under reduced pressure to
give a residue, which was heated at reflux with acetic acid (0.6 ml)
and concentrated hydrochloric acid (0.3 ml) for a further 2 h. The
d
: 1.20 (3H, t, 3JHH ¼ 7.1 Hz, CH3), 4.15 (2H, q, 3JHH ¼ 7.1 Hz, OCH2),
4.42 (2H, s, NCH2CO2), 4.75 (2H, s, PheCH2), 7.30e7.39 (5H, m, Ph).
13C NMR (DMSO-d6)
d: 13.7 (CH2eCH3), 40.1 (NeCH2), 42.3

1170
M. Rajabi et al. / European Journal of Medicinal Chemistry 46 (2011) 1165e1171
3
solid obtained by concentration was dissolved in ethyl acetate,
washed with water, and extracted with 10% aqueous sodium
carbonate. The aqueous layer was washed with ethyl acetate and
acidified with concentrated hydrochloric acid. The precipitated
solid was extracted with ethyl acetate, washed with water and
brine, and dried (anhydrous Na2SO4). Concentration followed by
crystallisation from diethyl ether gave 9 as a cream solid (0.12 g,
65%). M.p. 193e196 ꢂC; Lit. M.p. 198e199 ꢂC [19]. IR (KBr, cmꢀ1):
3200e2800 cmꢀ1 (br, CO2H),1710 cmꢀ1 (br, C]O). 1H NMR (DMSO-
CH2CO), 3.70 (2H, t, JHH ¼ 7.6 Hz, NeCH2), 4.61 (2H, s, AreCH2),
6.90 (1H, br s, NH), 7.28e7.35 (4H, m, AreH), 7.39 (1H, br s, NH). 13C
NMR (DMSO-d6),
d: 19.1 (AreCH3), 33.2 (CH2CONH2), 35.5
(NCH2CH2), 40.2 (AreCH2), 126.2 (C-5), 127.8 (C-4), 127.9 (C-6),
130.4 (C-3), 133.7 (C-2), 135.8 (C-1), 154.9 (C]O), 155.2 (C]O),
158.7 (C]O), 171.6 (CONH2). MS (ESþ): 312 ([M þ Na]þ, 100%), 307
([M þ NH4]þ, 50%). Accurate mass 290.1138. C14H16N3O4 (M þ H)
requires 290.1135.
d6)
d
: 2.35 (3H, s, CH3), 4.31 (2H, s, NeCH2), 4.73 (2H, s, AreCH2),
4.3.15. 3-[3-(3-Chlorobenzyl)-2,4,5-trioxo-imidazolidin-1-yl]-
propionamide (31)
7.13e7.26 (4H, m, AreH), 13.5 (1H, br s, OH). 13C NMR (DMSO-d6)
d:
19.1 (CH3), 40.1 (AreCH2), 40.4 (NCH2CO2), 126.3 (C-5), 128.0 (C-4),
128.1 (C-6), 130.5 (C-3), 133.3 (C-2), 136.1(C-1), 154.0 (C]O), 157.0
(C]O), 157.4 (C]O), 168.4 (COOH).
Amide 31 was prepared as described for 30 using 20: colourless
powder, yield: 38%. M.p. 218e221 ꢂC. Anal. Calc. for C13H12ClN3O4:
50.42% C, 3.91% H, 13.57% N, 11.45% Cl; found: 50.19% C, 3.82% H,
13.11% N, 11.16% Cl. IR (KBr) cmꢀ1: 3200e3500 (NH2), 1720 (C]O),
4.3.11. [3-(3-chloro-benzyl)-2,4,5-trioxo-imidazolidin-1-yl]
ethanoic acid (27)
Acid 27 was prepared as described for 9 using 24: colourless
powder, yield: 87%. M.p. 210e214 ꢂC, Lit. M.p. 207e208 ꢂC [19]. IR
(KBr, cmꢀ1): 3200e2800 cmꢀ1 (br, CO2H), 1710 cmꢀ1 (br, C]O). 1H
1660 (C]O), 1630 (C]O). 1H NMR (DMSO-d6),
d: 2.48 (2H, t,
3JHH ¼ 7.6 Hz, CH2CO), 3.70 (2H, t, 3JHH ¼ 7.6 Hz, NeCH2), 4.61 (2H, s,
AreCH2), 6.90 (1H, br s, NH), 7.34e7.38 (4H, m, AreH), 7.46 (1H, br s,
NH).13C NMR (DMSO-d6),
d: 39.8 (CH2CONH2), 40.1 (NCH2CH2), 49.3
(AreCH2), 126.4 (C-6), 127.1 (C-4), 127.8 (C-2), 128.8 (C-5), 134.2 (C-
3), 138.9(C-1), 155.2 (C]O), 155.8 (C]O), 156.3 (C]O), 176.5
(CONH2). MS(ESþ): 332 (35Cl [M þ Na]þ,100%), 334 (37Cl [M þ Na]þ,
30%). MS(ES-): 308 (35Cl MeH, 100%), 310 (37Cl MeH, 30%).
NMR (DMSO-d6)
d: 4.30 (2H, s, NeCH2), 4.77 (2H, s, AreCH2),
7.22e7.44 (4H, m, AreH), 12.9 (1H, br s, OH). 13C NMR (DMSO-d6)
d:
40.7 (AreCH2 or NeCH2), 41.7 (AreCH2 or NeCH2), 126.6 (C-6),
127.7 (C-4), 128.1 (C-2), 130.8 (C-5), 133.6 (C-3), 138.0 (C-1), 153.8
(C]O), 157.1 (C]O), 157.3 (C]O), 168.3 (COOH).
4.3.16. 3-(2,4,5-Trioxo-3-phenethyl-imidazolidin-1-yl)-
propionamide (32)
4.3.12. (2,4,5-Trioxo-3-phenethyl-imidazolidin-1-yl)ethanoic acid
(28)
Amide 32 was prepared as described for 30 using 21: colourless
powder, yield: 96%. M.p. 146e149 ꢂC. Anal. Calc. for C14H15N3O4:
58.11% C, 5.23% H, N 14.53% N; found: 57.77% C, 5.21% H, 13.54% N. IR
Acid 28 was prepared as described for 9 using 25: colourless
powder, yield: 76%. M.p.153e155 ꢂC, Lit. M.p.154.5e155.5 ꢂC [19]. IR
(KBr) cmꢀ1: 3200e3500 (NH2), 1720 (C]O), 1660 (C]O), 1630 (C]
(KBr, cmꢀ1): 3200e2800 cmꢀ1 (br, CO2H), 1710 cmꢀ1 (br, C]O). 1H
O). 1H NMR (DMSO-d6)
d
: 2.48 (2H, t, JHH ¼ 7.6 Hz, CH2CO), 2.93
3
3
NMR (DMSO-d6)
d
:2.90 (2H, t, JHH ¼ 7.2 Hz, PheCH2), 3.78 (2H, t,
(2H, t, 3JHH ¼ 7.6 Hz, PheCH2), 3.70 (2H, t, 3JHH ¼ 7.6 Hz, NeCH2 or
CH2eN), 3.72 (2H, t, 3JHH ¼ 7.6 Hz, NeCH2 or CH2eN), 6.91 (1H, br s,
NH), 7.22e7.33 (5H, m, Ph), 7.44 (1H, br s, NH). 13C NMR (DMSO-d6)
3JHH ¼ 7.1 Hz, CH2eN), 4.30 (2H, s, CH2eCO), 7.24e7.39 (5H, m, Ph),
12.8 (1H, br s, OH). 13C NMR (DMSO-d6)
d: 32.4 (AreCH2), 43.2
(AreCH2eCH2), 46.8 (NCH2CO2), 124.3 (C-4), 129.3 (C-2 and C-6),
131.6 (C-3 and C-5), 135.6 (C-1), 158.1, 158.3 and 159.2 (C]O), 166.9
(COOH).
d: 36.4 (PheCH2), 36.9 (CH2CONH2), 42.3 (NCH2CH2), 49.3
(eCH2eN), 126.3 (C-4), 127.6 (C-2 and C-6), 128.3 (C-3 and C-5),
139.7 (C-1),158.2 (C]O),158.8 (C]O),159.4 (C]O),169.4 (CONH2).
MS (ESþ): 312 ([M þ Na]þ, 100%).
4.3.13. 3-(3-Benzyl-2,4,5-trioxo-imidazolidin-1-yl)-propionamide
(29)
4.4. TP inhibition assay
Amide 29 was prepared as described for 30 using 18: colourless
powder, yield: 72%. M.p. 201e203 ꢂC. Anal. Calc. for C13H13N3O4:
56.71% C, 4.76% H, N 15.27%; found 56.50% C, 4.11% H, 14.51% N. IR
The 1 ml assay mix contained 20 mM thymidine, 0.1 M potassium
phosphate buffer (pH 7.4), together with the TP inhibitor. The
reaction was initiated by addition of E. coli TP enzyme (0.22 units)
and the change in absorbance was monitored at 265 nm at 25 ꢂC
[11,20]. Enzyme kinetics data were analysed using Grafit version 3
software (Erithacus software).
(KBr) cmꢀ1: 3200e3500 (NH2), 1720 (C]O), 1660 (C]O), 1630 (C]
3
O). 1H NMR (DMSO-d6)
d
: 2.40 (2H, t, JHH ¼ 7.6 Hz, CH2CO), 3.68
3
(2H, t, JHH ¼ 7.6 Hz, NeCH2), 4.61 (2H, s, AreCH2), 6.90 (1H, br s,
NH), 7.30e7.35 (5H, m, Ph), 7.45 (1H, br s, NH). 13C NMR (DMSO-d6)
d: 34.5 (CH2CONH2), 38.8 (NCH2CH2), 48.2 (PheCH2), 126.7 (C-4),
127.3 (C-2/6), 127.8 (C-3/5), 139.6 (C-1), 156.2 (C]O), 156.9 (C]O),
Acknowledgements
157.7 (C]O), 173.2 (CONH2). MS (ESþ): 298 ([M þ Na]þ, 100%).
We thank the School of Pharmacy for partial funding (MR), COST
606 and EPSRC for studentship and PhD plus fellowship (DM). We
thank Waleed Zalloum for assistance in preparation of figures and
Philip N. Edwards (p.n.e@btinternet.com) for informative and
challenging discussions.
4.3.14. 3-(2-Methylbenzyl)-2,4,5-trioxoimidazolidin-1-yl-
propionamide (30)
A mixture of 19 (0.22 g,1 mmol) and 3-bromopropionamide [27]
(0.23 g, 1.54 mmol) was added to a solution of KOH (80 mg,
1.2 mmol) in ethanol (6 ml). After heating at reflux for 8 h, the
mixture was cooled to 0 ꢂC and filtered. The solid was dissolved in
ethyl acetate and washed with water (10 ml) and brine (10 ml). The
organic phase was dried over (anhydrous Na2SO4) and passed
through a short silica gel pad. After being concentrated, the crude
solid was recrystallised from ethyl acetateehexane to give (30)
(0.26 g, 90%) as colourless crystals. M.p. 199e201 ꢂC. Anal. Calc. for
C14H15N3O4: 58.13% C, 5.23% H; found: 58.27% C, 4.91% H. IR (KBr)
References
[1] T.A. Krenitsky, G.W. Koszalka, J.V. Tuttle, Purine nucleoside synthesis: an
efficient method employing nucleoside phosphorylases, Biochemistry 20
(1981) 3615e3621.
[2] M. Haraguchi, K. Miyadera, K. Uemura, T. Sumizawa, T. Furukawa, K. Yamada,
S.-I. Akiyama, Y. Yamada, Angiogenic activity of enzymes, Nature 368 (1994) 198.
[3] K. Miyadera, T. Sumizawa, M. Haraguchi, H. Yoshida, W. Konstanty, Y. Yamada,
S.-I. Akiyama, Role of thymidine phosphorylase activity in the angiogenic
effect of platelet-derived endothelial cell growth factor/thymidine phos-
phorylase, Cancer Res. 55 (1995) 1687e1690.
cmꢀ1: 3200e3500 (NH2), 1720 (C]O), 1660 (C]O), 1630 (C]O). 1H
3
NMR (DMSO-d6),
d
: 2.34 (3H, s, CH3), 2.42 (2H, t, JHH ¼ 7.6 Hz,

M. Rajabi et al. / European Journal of Medicinal Chemistry 46 (2011) 1165e1171
1171
[4] E.J. Yu, Y. Lee, S.Y. Rha, T.S. Kim, H.C. Chung, B.K. Oh, W.I. Yang, S.H. Noh,
H.-C. Jeung, Angiogenic factor thymidine phosphorylase increases cancer cell
invasion activity in patients with gastric adenocarcinoma, Mol. Cancer Res. 6
(2008) 1554e1566.
[5] A. Bronckaers, F. Gago, J. Balzarini, S. Liekens, The dual role of thymidine
phosphorylase in cancer development and chemotherapy, Med. Res. Rev. 29
(2009) 903e953.
[6] M. de Bruin, T. van Capel, K. Smid, K. van der Born, M. Fukushima, K. Hoekman,
H.M. Pinedo, G.J. Peters, Role of platelet derived endothelial cell growth factor/
thymidine phosphorylase in fluoropyrimidine sensitivity and potential role
of deoxyribose-1-phosphate, Nucleos. Nucleot. Nucl. Acids 23 (2004)
1485e1490.
[7] R. Ikeda, X.-F. Che, M. Ushiyama, T. Yamaguchi, H. Okumura, Y. Nakajima,
Y. Takeda, Y. Shibayama, T. Furukawa, M. Yamamoto, M. Haraguchi,
T. Sumizawa, K. Yamada, S.-I. Akiyama, 2-Deoxy-D-ribose inhibits hypoxia-
induced apoptosis by suppressing the phosphorylation of p38 MAPK, Bio-
chem. Biophys. Res. Commun. 342 (2006) 280e285.
coli thymidine phosphorylase by in silico screening, Bioorg. Med. Chem. Lett.
13 (2003) 3705e3709.
[14] M. Rajabi. Design, synthesis and biological evaluation of thymidine phos-
phorylase inhibitors as anti-cancer drugs, PhD Thesis (2007), The University of
Manchester.
[15] R.A. Norman, S.T. Barry, M. Bate, J. Breed, J.G. Colls, R.J. Ernill, R.W.A. Luke,
C.A. Minshull, M.S.B. McAlister, E.J. McCall, H.H.J. McMiken, D.S. Paterson,
D. Timms, J.A. Tucker, R.A. Pauptit, Crystal structure of human thymidine phos-
phorylase in complex with a small molecule inhibitor, Structure 12 (2004)
75e84.
[16] P.F. Kador, J.H. Kinoshita, N.E. Sharpless, Aldose reductase inhibitors:
a potential new class of agents for the pharmacological control of certain
diabetic complications, J. Med. Chem. 28 (1985) 841e849.
[17] P.F. Kador, W.G. Robison, J.H. Kinoshita, The pharmacology of aldose reductase
inhibitors, Annu. Rev. Pharmacol. Toxicol. 25 (1985) 691e714.
[18] P.F. Kador, The role of aldose reductase in the development of diabetic
complications, Med. Res. Rev. 8 (1988) 325e352.
[8] S. Matsushita, T. Nitanda, T. Furukawa, T. Sumizawa, A. Tani, K. Nishimoto,
S. Akiba, K. Miyadera, M. Fukushima, Y. Yamada, H. Yoshida, T. Kanzaki, S.-
I. Akiyama, The effect of a thymidine phosphorylase inhibitor on angiogenesis
and apoptosis in tumors, Cancer Res. 59 (1999) 1911e1916.
[9] C. Cole, P. Reigan, A. Gbaj, P.N. Edwards, K.T. Douglas, I.J. Stratford, S. Freeman,
M. Jaffar, Potential tumor-selective nitroimidazolylmethyluracil prodrug
derivatives: inhibitors of the angiogenic enzyme thymidine phosphorylase,
J. Med. Chem. 46 (2002) 207e209.
[19] A. Ishii, T. Kotani, Y. Nagaki, Y. Shibayama, Y. Toyomaki, N. Okukado, K. Ienaga,
K. Okamoto, Highly selective aldose reductase inhibitors. 1. 3-(Arylalkyl)-
2,4,5-trioxoimidazolidine-1-acetic acids, J. Med. Chem. 39 (1996) 1924e1927.
[20] C. Nakayama, Y. Wataya, R.B. Meyer Jr., D.V. Santi, M. Saneyoshi, T. Ueda,
Thymidine phosphorylase. Substrate specificity for 5-substituted 20-deoxy-
uridines, J. Med. Chem. 23 (1980) 962e964.
[21] J. Balzarini, A.E. Gamboa, R. Esnouf, S. Liekens, J. Neyts, E.D. Clercq,
M.-J. Camarasa, M.-J. Pérez-Pérez, 7-Deazaxanthine,
a novel prototype
[10] P. Reigan, A. Gbaj, E. Chinje, I.J. Stratford, K.T. Douglas, S. Freeman, Synthesis
and enzymatic evaluation of xanthine oxidase-activated prodrugs based on
inhibitors of thymidine phosphorylase, Bioorg. Med. Chem. Lett. 14 (2004)
5247e5250.
[11] P. Reigan, P.N. Edwards, A. Gbaj, C. Cole, S.T. Barry, K.M. Page, S.E. Ashton,
R.W.A. Luke, K.T. Douglas, I.J. Stratford, M. Jaffar, R.A. Bryce, S. Freeman,
Aminoimidazolylmethyluracil analogues as potent inhibitors of thymidine
phosphorylase and their bioreductive nitroimidazolyl prodrugs, J. Med. Chem.
48 (2005) 392e402.
inhibitor of thymidine phosphorylase, FEBS Lett 438 (1998) 91e95.
[22] M.J. Pugmire, S.E. Ealick, The crystal structure of pyrimidine nucleoside
phosphorylase in a closed conformation, Structure 6 (1998) 1467e1479.
[23] T.J.A. Ewing, S. Makino, A.G. Skillman, I.D. Kuntz, DOCK 4.0: search strategies
for automated molecular docking of flexible molecule databases, J. Comput.
Aid. Mol. Des 15 (2001) 411e428.
[24] J. Gasteiger, M. Marsili, Iterative partial equalization of orbital electronega-
tivity: a rapid access to atomic charges, Tetrahedron 36 (1980) 3219e3222.
[25] T.A. Halgren, Merck molecular force field. I. Basis, form, scope, parameteri-
zation, and performance of MMFF94, J. Comput. Chem. 17 (1996) 490e519.
[26] M.L. Verdonk, J.C. Cole, M.J. Hartshorn, C.W. Murray, R.D. Taylor, Improved
protein-ligand docking using GOLD, Proteins: Struct. Funct. Genet. 52 (2003)
609e623.
[12] M.P. Price, W. Guida, T. Jackson, J. Nydick, P. Gladstone, J. Juarez, F. Donate,
R. Ternansky, Design of novel N-(2,4-dioxo-1,2,3,4-tetrahydro-thieno[3,2-d]
pyrimidin-7-yl)-guanidines as thymidine phosphorylase inhibitors, and flex-
ible docking to
a homology model, Bioorg. Med. Chem. Lett. 13 (2003)
107e110.
[27] D.D. Tanner, C.P. Meintzer, Reexamination of the reaction of a “graded set” of
[13] V.A. McNally, A. Gbaj, K.T. Douglas, I.J. Stratford, M. Jaffar, S. Freeman,
radicals with N-bromosuccinimide: a kinetic argument concerning the
states of succinimidyl, J. Am. Chem. Soc. 107 (1985) 6584e6589.
p and
R.A. Bryce, Identification of a novel class of inhibitor of human and Escherichia
s