Molecularly imprinted polymer of 5-methyluridine for solid-phase extraction of pyrimidine nucleoside cancer markers in urine

Article abstract of DOI:10.1016/j.bmc.2008.08.063

Normal and modified urinary nucleosides represent potential biomarkers for cancer diagnosis. To selectively extract modified nucleosides, we developed a molecularly imprinted polymer (MIP) of 5-methyluridine as selective material for molecularly imprinted solid-phase extraction (MISPE). The MIPs were obtained from vinyl-phenylboronate ester derivative of the template, acrylamide and pentaerythritol triacrylate co-polymer, and were tested in batch and cartridge experiments with aqueous samples. Our results indicated that the imprinted polymer was selective for pyrimidine nucleosides with a K_d and a B_max of 46 μM and 18 μmol/g, respectively. Finally, a MISPE of the most common pyrimidine nucleoside cancer markers in urine sample was realized.

Full text of DOI:10.1016/j.bmc.2008.08.063

Bioorganic & Medicinal Chemistry 16 (2008) 8932–8939
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Molecularly imprinted polymer of 5-methyluridine for solid-phase extraction
of pyrimidine nucleoside cancer markers in urine
a
b
b
Damien Jégourel a, Raphaël Delépée a, , Florent Breton , Antoine Rolland , Richard Vidal ,
*
Luigi A. Agrofoglio a,
*
a Institut de Chimie Organique et Analytique (ICOA), CNRS UMR 6005, Université d’Orléans, BP 6759, Orléans 45067 Cedex 2, France
b ESTAPOR Merck Chimie, rue du Moulin de la Canne, 45800 Pithiviers, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 March 2008
Revised 15 July 2008
Accepted 26 August 2008
Available online 29 August 2008
Normal and modified urinary nucleosides represent potential biomarkers for cancer diagnosis. To selec-
tively extract modified nucleosides, we developed a molecularly imprinted polymer (MIP) of 5-methylu-
ridine as selective material for molecularly imprinted solid-phase extraction (MISPE). The MIPs were
obtained from vinyl-phenylboronate ester derivative of the template, acrylamide and pentaerythritol tri-
acrylate co-polymer, and were tested in batch and cartridge experiments with aqueous samples. Our
results indicated that the imprinted polymer was selective for pyrimidine nucleosides with a Kd and a
Keywords:
Bmax of 46
lM and 18 lmol/g, respectively. Finally, a MISPE of the most common pyrimidine nucleoside
5-Methyluridine
Cancer markers
Molecularly imprinting polymer
Pyrimidinic nucleosides
MISPE
cancer markers in urine sample was realized.
Ó 2008 Elsevier Ltd. All rights reserved.
Urine
1. Introduction
more excreted in patient with severe combined immunodefi-
ciency diseases, breast or ovary cancers, respectively.8,9
The hydrolytic action of ribonucleases and phosphatases liber-
ates normal and modified nucleosides during the ribonucleic acid
(RNA) turnover. This is made post-transcriptionally by specific
enzymes such as RNA-methyltransferases and RNA-synthetases1
and leads to structural modifications of the nucleosides such as
alkylation of the heterocycle, methylation of the sugar, heterocy-
cle isomerization, reduction, thiolation, and deamination. These
compounds cannot be used for the de novo RNA synthesis and
thus circulate freely in blood before elimination in urines by
the same excretion pathway than normal nucleosides.2 It has
been shown that urinary levels of natural and modified nucleo-
sides in patients with stage 2–4 cancer were significantly ele-
vated compared to human healthy as shown for leukemia,
lymphoma,3 lung cancer,4 oesophagus,5 breast,6 and renal cell
carcinoma.7 RNA and particularly tRNA is a primary source of
the modified nucleosides found in urine. A total of 93 natural
and modified nucleosides including cytidine (1), uridine (2), gua-
nosine (3), adenosine (4), 3-methylcytidine (5), 5-methylcytidine
(6), N4-acetylcytidine (7), or pseudouridine (8) have been identi-
fied (Fig. 1), the later been found to be between 3 and 6 times
Thus, the quantitation of normal and modified urinary nucleo-
sides should allow a premature cancer diagnosis and a monitoring
during its evolution. During the last decade, we and others have
used various techniques such as enzyme-linked immunoassay,10
capillary electrophoresis,11 cathodic stripping,12 HPLC–MS,13 or
GC–MS14 to analyze nucleosides in biological fluids. A common
limitation is the pre-treatment of the biological sample, which
can include an extraction, a hydrolysis or a derivation step and is
prohibitive in terms of time and cost when applied to a large num-
ber of samples as in diagnosis bio-analyses.
In the present study, we have used the molecularly imprinting
polymer (MIP) technology to set up a molecularly imprinted so-
lid-phase extraction (MISPE) for selective extraction from urine
of endogenous and modified pyrimidine nucleosides. The MIP
concept, historically introduced by Wulff15 and Mosbach16 are
speciality polymers generated via the interaction of functional
monomers with a target molecule (template) and a cross-linking
agent.17 With the non-covalent self-assembly approach, the
template, functional monomer, and cross-linking agents are equil-
ibrated to generate a pre-polymerisation cluster utilizing hydro-
gen bond interactions, electrostatic attraction, and associated
weak interactions. The mix is then polymerized to generate cavi-
ties and the template is subsequently removed via an effective
extraction.
* Corresponding authors. Tel.: +33 2 3841 7004; fax: +33 2 3849 4582 (R.D.).
E-mail addresses: raphael.delepee@univ-orleans.fr (R. Delépée), luigi.agrofoglio@
univ-orleans.fr (L.A. Agrofoglio).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.08.063

D. Jégourel et al. / Bioorg. Med. Chem. 16 (2008) 8932–8939
8933
NH2
N
O
O
NH2
N
N
N
N
N
NH
NH
O
O
HO
N
HO
N
HO
HO
N
N
NH2
O
O
O
O
HO OH
HO OH
Uridine (2)
HO OH
Guanosine (3)
HO OH
Adenosine (4)
Cytidine (1)
NH
N
NH2
NHAc
O
NH
O
N
N
HN
O
O
HO
N
O
O
HO
HO
N
HO
N
O
O
O
HO OH
HO OH
HO OH
5-methyluridine (6)
HO OH
N4-acetylcytidine (7)
3-methylcytidine (5)
Pseudouridine (8)
Figure 1. Studied natural and modified nucleosides.
The obtained cavities are supposed to have a complementary
shape and electronic affinity to the template. For the imprinting
of polar compounds, the classical approach consists in the deriva-
tion of polar compounds into less polar analogues, such as acety-
lated galactosides,18 acetylated nucleosides and nucleotides.19,20
and the heterocycle by hydrogen-bonding interactions. To com-
plex the nucleobase, several monomers have been reported such
as the acrylic acid,37,38 the acrylamide39,40 (reported to induce a
better affinity than methacrylic acid),41 the 2,6-bis(acrylamido)
pyridine.36,42 In our case, we have synthesized the 2,6-bis(meth-
acrylamido) pyridine monomer (11a) because of a lower ten-
dency of self-association compared to the 2,6-bis(acrylamido)
pyridine derivative (11b).35 It is also a less reactive polymeriz-
able monomer compared to the acrylamide analogue which
contains two vinyl groups. In order to mimic the natural base-
pairing interactions between adenosine and uridine, the 2,6-dia-
mino-9-(3/4-vinylbenzyl) purine monomer33 (13) was also
evaluated. 1H NMR titrations were conducted in order to moni-
tor hydrogen bond interactions between the functional co-mono-
mer (host) and compound 10a (guest) by increasing the amount
of host into a constant amount of guest. The complexation of
10a with MAP (11a), DAVBP (12), and AA (13), respectively, in-
duced a variation of the chemical shift of the imide proton of
10a as reported in Figure 3.
Other polar molecules like p-nitrophenyl-a-b-L
-furanoside21 have
been imprinted by a non-covalent approach as a cross-linked
hydrogel.22–24 For the unprotected sugar extraction, Wulff et al.
have reported a covalent approach using a boronic acid ester of
a-D
-mannopyranoside as template.25–28 The property of a boronic
acid to bind diols has been also used to prepare polymeric gels
for nucleotide adsorption using a boronic acid–hydroxyethylmeth-
acrylate co-polymer.29 MIP were developed as sensors on a quartz
crystal microbalance electrode30,31 or directly on an electrode32 for
the detection of adenosine monophosphate (AMP). Other pyrimi-
dine derivatives have been imprinted by Hall et al.33 using 2,6-
bis(acrylamido)pyridine as monomer,33–35 and a low complexation
was exhibited toward uridine.36
Thus, the preparation of highly cross-linked non-polar polymer
having a strong affinity for polar compounds (such as nucleosides)
is still challenging. Starting with the 5-methyluridine (9) as dum-
my template (i.e., a template structurally similar to the compounds
to be analyzed but different enough not to interfere with them), we
have synthesized the corresponding polymerizable 2,3-diol boro-
nate ester (10). We then turned our attention to complex the
pyrimidine moiety by specific hydrogen bonds. The efficiency of
the obtained polymer was evaluated as MISPE material for the
selective extraction of selected natural and modified pyrimidine
nucleosides in urine.
Based on the literature on host–guest systems,43–46 we assumed
to have a 1:1 stoichiometry for the association. As previously
published, the best complexing site for uracil moiety is the accep-
tor–donor–acceptor (ADA) triad obtained from the groups at C2,
N3, and C4 position.35,38,47 From the non-linear curve fitting of
isoterms,48 we can extract the apparent association constants
(Kapp) and the maximum complexation induced shifts (CISs), (see
Table 2).
As shown with the Kapp values, 10a has a better association with
acrylamide (13) than with DAVBP (12) or MAP (11a). The Kapp for
11a was at least twice lower than for 13. These data were not in
agreement with those reported for the association of N1-benzyl-
uracil33,35 with 2,6-bis(acrylamido)pyridine monomer (11b),
which was found to exhibit a better interaction than the acrylam-
ide (13) with the ADA system of uracil. The steric hindrance
brought by the two methyl groups of 11a but also found at the
C5-position of the heterocycle and at the 20,30-position of the sugar
could explained this result. The acrylamide (13), which has the big-
gest Kapp, can interact with the nucleobase by two complexes (at
C4–N3 or C2–N3), as reported for N1-methyluracil.41 The maxi-
mum complexation induced shifts (CIS) is related to the length of
the hydrogen bond and the dihedral angle of the bond.49 From
Table 2, it can be seen than the geometry of the complex with DAV-
BP was different than the two others since no self-association was
detected (results not shown).
2. Results and discussion
2.1. Chemistry
The synthesis of compounds 10, 11a, and 12 (Fig. 2) was real-
ized in our laboratory meanwhile acrylamide (13) and cross-link-
ers (15–17) were commercially available. These reagents were
used to synthesize six MIPs with different compositions as summa-
rized in Table 1.
2.2. Choice of monomers and NMR studies
To imprint the 5-methyluridine (9), we decided to covalently
bind the ribose moiety applying the boronic acid methodology

8934
D. Jégourel et al. / Bioorg. Med. Chem. 16 (2008) 8932–8939
O
NH2
O
O
NH
N
N
N
NH
NH
HO
N
O
O
O
B
NH2
N
HO
N
O
HO
N
O
O
B
O
O
O
O
R
R
N
N
N
O
O
HO
OH
H
H
Bu4N+
OH
2,6-diamino-9-(3/4-vinylbenzyl)purine
DAVBP
5-methyluridine
5-MeU
11a
11b
R=Me, 2,6-bis(methacrylamido)
pyridine MAP
12
R=Me, 2,6-bis(acrylamido)pyridine
AAP
9
O
5-methyluridine-2',3'-(4-vinylbenzyl)boronate
5-MeU-tri-B
5-Methyluridine-2',3'-(4-vinylbenzyl)hydroxyboronate
tetrabytylammonium
NH2
5-MeU-tetra-B
10a
10b
Acrylamide
AA
13
O
O
O
O
O
O
3
OH
O
3
O
Divinylbenzene
DVB
Pentaerythritol triacrylate
PETRA
Trimethylolpropane trimethacrylate
TRIM
Ethylene glycol dimethacrylate
EGDMA
14
17
16
15
Figure 2. Templates, functional monomers, and cross-linkers used for MIP optimization.
Table 1
Composition of the polymers synthesizeda
Polymer
A
B
C
D
E
F
Monomer (quantity)
Cross-linker (20 mmol)
Porogen volume (MeCN) (mL)
MAP (2 mmol)
EGDMA
3.6
DAVBP (2 mmol)
EGDMA
5.0
DAVBP (2 mmol)
DVB
3.3
AA (4 mmol)
EGDMA
3.2
AA (4 mmol)
TRIM
3.2
AA (4 mmol)
PETRA
3.2
a
Each polymers contain 1 mmol of template (10) and AIBN mass was 40 mg (for abbreviation see Fig. 2).
Table 2
6
Apparent association constants (Kapp) and maximum complexation induced shifts
dmax) of the imide proton of 10a
(D
5
4
3
2
1
0
Functional monomer
Kapp (Mꢀ1
)
Ddmax (ppm)
MAP (11a)
DAVBP (12)
AA (13)
116 36
271 15
312 29
1.75 0.27
5.87 0.13
2.68 0.09
rebinding studies (Table 3) were conducted following a protocol
described in Section 4.6.
From Table 3, the better rebinding was observed with the DAV-
BP-based MIP (polymer B) and not with the AA-based MIP (poly-
mer D) as it could be expected from NMR data. Since the
rebinding experiments were conducted in water and the 1H NMR
experiments in CDCl3, the solvent should strongly disrupt the
intramolecular hydrogen-bonding interactions. The major rebind-
ing effect in water was due to the covalent bonding of 20,30-diol
moiety of nucleosides with the boronic acid moiety, in agreement
with the literature.50 However, selectivity between uridine and
cytidine derivatives was observed, in agreement with 1H NMR
experiments. It is interesting to note that no selectivity for the het-
erocycle was observed when a commercially available phenylbo-
ronic acid material was used.
0
0.005
0.01
0.015
[M] (M)
0.02
0.025
Figure 3. 1H NMR complexation induced shifts (
Dd) of the imide proton of 10a
versus total concentration of 11a (open circles), 12 (filled squares), or 13 (filled
triangle) in CDCl3.
2.3. MIP synthesis
MIPs A–D were synthesized in the same conditions and varied
from each other only by their composition (see Table 1). The

D. Jégourel et al. / Bioorg. Med. Chem. 16 (2008) 8932–8939
8935
NIP
Table 3
Rebinding assays on the synthesized MIP and NIPa
Rebinding (%)
A
B
C
D
E
F
MIP
NIP
MIP
NIP
MIP
NIP
MIP
NIP
MIP
NIP
MIP
Uridine
Cytidine
Guanosine
64
67
85
5
7
6
81
86
96
10
11
26
21
35
51
8
6
20
55
38
71
7
3
22
38
19
40
6
5
10
71
57
88
3
2
6
a
Error on each point was less than 3%, n = 3, (see Table 1 for composition).
The polarity of the cross-linker was also investigated. The dif-
ference of rebinding rate between EGDMA-based MIP (polymer
B) and DVB-based MIP (polymer C) highlighted the importance
of the polarity of the material. Polymer C using DVB (14) as
cross-linker was less polar than polymer B based on EGDMA
(15) and exhibited a lower affinity (21%) for polar uridine com-
pared to polymer B (81%). The best specificity for uridine, cyti-
Under acidic and neutral pH, the trigonal 10a is major whereas tet-
rahedral 10b is mainly observed for pH > pKa (e.g., 8.8).55 It is only
under this last conformation (tetrahedral) that the diols can be
well complexed.56 The boronate ester bond is thus pH dependent
and is stable in aqueous solutions at pH above 9, so all further
experiments were achieved at pH 10.
The first unsuccessful assay of rebinding experiments was done
with a 0.1 M sodium hydroxide solution adjusted to pH 10 with
0.1 M HCl. Following the works by Smith et al.,57 tetra-butyl ammo-
nium hydroxide (TBAH) was added in the rebinding solutions (in
1000/1 molar ratio versus the rebound compound) and under this
condition, 74% of uridine (2) was rebound within 24 h. The N3 of
uridine has a pKa value of 9.5,58 so at pH 10 the heterocycle is neg-
atively charged. At that pH, uridine forms an ion pair with the qua-
ternary ammonium. The access of the nucleoside to the cavity of
the polymer could be thus facilitated by an ion pairing mechanism.
This was confirmed by the absence of complexation observed
when sodium hydroxide was only used. We then turned our atten-
tion to the length of the alkyl chain of the quaternary ammonium
ion (Fig. 5A). As expected, the longer the alkyl chain, the higher the
rebinding capacity of uridine. When the chains became longer than
butyl, problems of solubility in water and loss of rebinding rate
occurred. The last studied parameter to show the ion pairing
behavior of the uridine binding was the influence of the counter-
ion concentration (Fig. 5B). As expected, an increase of the concen-
tration of TBAH led to an increase of the rebinding of the MIP. With
concentrations higher than a molar ratio of 1000/1, a loss of selec-
tivity can be observed. Under these conditions, the rebinding was
almost immediate (data not shown).
dine, and guanosine, was obtained using
a AA-based MIP
(polymers D–F). It has been shown that trimer cross-linkers
(such as TRIM (16) and PETRA (17)) can increase the selectivity
of the rebinding step.51–54 TRIM- and PETRA-based MIP (poly-
mers E and F, respectively) raised selectivity of MIP, with very
good rebinding rate for the more polar PETRA cross-linker. Thus,
we have shown that for the MIP of nucleosides, the rebinding
and the selectivity were mainly governed by the polarity of
the monomer and the cross-linkers. For highly polar compounds,
the hydrogen-bonding capabilities were not the unique source of
selectivity. Finally, based on the targeted specificity, the polymer
F (AA-PETRA-based MIP) was chosen for all further experiments.
The obtained hard porous materials have a surface area of 321
and 337 m2 gꢀ1 for MIP and NIP respectively, with a cauliflower
form (Fig. 4). Their degree of swelling in volume in water was
estimated to be approximately 10%.
2.4. Influence of the nature and concentration of the counter-
ion
Since 5-methyluridine (9) was used as the dummy template, all
further experiments were carried out with uridine (2). In anhy-
drous aprotic solvents, a boronic acid derivative forms with 1,2-
diol a trigonal boronate ester (10a). In aqueous solution, trigonal
boronate is unstable and is either hydrolyzed to the starting com-
pounds or ionized to form an anionic tetrahedral boronate (10b).
2.5. Isotherm
The Scatchard representation was next explored to determine
the number and type of binding sites.34 The Scatchard plot of
MIP F displayed a strongly curved shape, reflecting that uridine
was adsorbed on either imprinting sites or non-imprinting sites,
(Fig. 6). In fact, the Scatchard plots indicate that MIP F shows high
and low concentration regimes that are nearly linear, intercon-
nected with strongly curved regions in the mid-concentration
range. As the Scatchard plot for NIP is curvilinear, it indicates that
a one site model should afford a good fit.
Bmax and Kd constants were determined according to Scat-
chard’s representation (Table 4). Bmax represents the number
of sites on the polymer and Kd the affinity for it. Kd values
for both sites highlight the interaction between the polymer
and uridine; a high affinity is correlated with a low Kd. Since
covalent rebinding is stronger than hydrogen-bonding interac-
tions, it is clear that the high affinity sites come from boronate
ester and by the way from selective sites. The second sites (low
affinity sites) probably come from the excess of acrylamide
introduced in the formulation of the MIP. As expected from
the excess of acrylamide, we can observe that there were twice
more non-selective sites than selective ones. Nevertheless, the
excess of acrylamide is necessary to favor the hydrogen-bond-
ing interactions.
Figure 4. SEM image of the obtained MIP.

8936
D. Jégourel et al. / Bioorg. Med. Chem. 16 (2008) 8932–8939
B
80
70
60
50
40
30
20
10
0
A
80
70
60
50
40
30
20
10
0
MIP
NIP
MIP
NIP
NaOH TMAH TPAH TBAH THAH TOAH
counter-ion
0
10
100
1000
10000
ratio template / tBuNOH (mol/mol)
Figure 5. Influence of the nature (A) and the concentration (B) of the counter-ion on the rebinding of uridine. TMAH, (Me)4NOH; TPAH, (Pr)4NOH; TBAH, (n-Bu)4NOH; THAH,
(n-Hex)4NOH; TOAH, (n-Oct)4NOH.
0.40
100
90
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
MIP NIP
80
70
60
50
40
30
20
10
0
Uridine
Cytosine Guanosine Adenosine
d4T
Figure 7. Rebinding rate of different nucleosides on MIP F and NIP F packed in a SPE
cartridge (100 mg).
0
5
10
15
20
25
30
35
B (µmol/g)
Figure 6. Scatchard representation of the binding of uridine on MIP F (filled
triangle) and NIP F (open triangle).
side lacking the 20,30-diol groups, were also evaluated. It appears
that purine nucleosides—guanosine (3) and adenosine (4)—were
more retained on the MIP than pyrimidine nucleosides—cytidine
(1) and uridine (2). These data could be explained by the differ-
ence of polarity of the selected nucleosides as well as by non-
specific interactions. Uridine (2) was more retained compared
to cytidine (1), which was in opposition with the logP values
at pH 10 (ꢀ2.88 for 2 and ꢀ2.05 for 1), respectively). These data
showed an imprinting effect for compounds with similar polari-
ties. Moreover, the contribution of the boronate ester bonds was
clearly demonstrated by the poor rebinding of d4T. Finally,
nearly no rebinding was observed on NIP.
Table 4
Constants from Scatchard plot for high and low affinity sites on MIP F
High affinity sites
Low affinity sites
mol/g) Kd
283
Bmax
(
l
mol/g)
Kd
(lmol/L)
Bmax
(l
(lmol/L)
18
46
44
With those optimized conditions for the MIP preparation, we
have evaluated this material in a cartridge for solid-phase extrac-
tion of nucleosides of interest (MIPSE).
2.7. MISPE of selected nucleosides
2.6. MIP selectivity
In order to optimize the MISPE conditions, a first experiment
was done with the selected nucleosides in water sample (Fig. 8).
The imprinting effect can be observed when non-specific
interactions between the template and the MIP are decreased.
We firstly evaluated the rebinding of some natural nucleosides,
for example, cytidine (1), uridine (2), guanosine (3), and adeno-
sine (4), on MIP and NIP packed in a SPE cartridge (100 mg)
(Fig. 7). The 30-deoxy-20,30-didehydrothymidine (d4T), a nucleo-
Typically, 10 lg of selected nucleosides in water (1 mL) were
mixed with 25 mM acetate buffer pH 10 and TBAH and then per-
colated on 100 mg of the 5-methyluridine–MIP cartridge. The
same experiment was carried out on NIP. A first TBAH washing
was done with buffer in order to elute the unbound polar com-
pounds. A second washing was done with acetonitrile to (a)

D. Jégourel et al. / Bioorg. Med. Chem. 16 (2008) 8932–8939
8937
A
B100
100
90
80
70
60
50
40
30
20
10
90
80
70
60
50
40
30
20
10
0
0
NIP
MIP
NIP
MIP
Figure 8. Specificity of MIP F in MISPE experiment. (A) Elution profiles obtained for 100 mg of 5-methyluridine MIP F and NIP and (B) recoveries for other nucleosides in the
same conditions. For conditions, see Section 4.
remove unbound non-polar compounds and (b) to increase the
selective interactions by favoring the hydrogen bounding at the
heterocycle moiety. Then, 1 mL of water/methanol/acetic acid
(50:45:5) was percolated to release the adsorbed nucleosides.
At each steps, the collected solutions were carefully analyzed
by LC–UV. Figure 8A represents the elution profile obtained for
uridine (2) on MIP F and NIP. The extraction recoveries were
80% for the MIP while only 3% for the NIP. This process was ap-
plied to other natural and modified nucleosides (Fig. 8B). The
extraction recoveries for all cytosine nucleosides (1, 5, and 6)
were quantitative excepted for N4-acetylcytidine (7), which at
pH 10, was hydrolyzed to cytidine (1). For all other nucleosides,
the extraction recoveries were ranging from 49% to 80%. This
data emphasized the selectivity of the prepared MIP for pyrimi-
dine nucleosides compared to purine ones.
by the prepared MIPSE. The pH value of urine being around 6, it
is necessary to adjust to pH 10 by a buffer solution of ammonium
acetate. TBAH was also added to the sample in order to load the ion
pair on the MISPE cartridge. Figure 9 reports the UV chromato-
grams corresponding to the injections of the standard mixture
(Fig. 9A), spiked-urine with nucleosides (Fig. 9C) and MISPE
extracted nucleosides from spiked-urine (Fig. 9B). As shown in
Figure 9C and B, the nucleosides were strongly extracted (thus
retained) by the prepared MISPE. In Fig. 9B, the increase of peak
1 corresponding to the cytidine (1) is due to the hydrolysis of
N4-acetylcytidine (2).
3. Conclusion
For the first time, this study described the synthesis of a
highly specific molecularly imprinted polymer of pyrimidine
nucleosides. The role of the polarity of polymer has been here
Finally, the extraction from urine of selected nucleosides,
including modified nucleosides as cancer markers was realized
500
450
400
350
300
250
200
150
100
50
(C)
(B)
1
2
4
6
3
5
7
(A)
0
0
5
10
15
20
25
Time (min)
30
35
40
45
50
Figure 9. Chromatograms corresponding to (A) standard solution (1 ppm), (B) elution profile after MISPE from urine-spiked, and (C) urine-spiked with 1 ppm of nucleosides.
Peak identification: 1, cytidine; 2, uridine; 3, guanosine; 4, adenosine; 5, 3-methylcytidine; 6, 5-methylcytidine; 7, N4-acetylcytidine. For conditions, see Section 4.

8938
D. Jégourel et al. / Bioorg. Med. Chem. 16 (2008) 8932–8939
clearly highlighted and the use of an ion pairing agent to in-
crease the rebinding has been depicted closely. Finally, the use
of this polymer as stationary phase for MISPE has been demon-
strated. The potential of MISPE for the direct extraction of cancer
marker pyrimidine nucleosides from urine has then largely
demonstrated.
4.3. Synthesis of templates
4.3.1. Synthesis of 20,30-(4-vinyl)phenylboronate-5-
methyluridine (10a)
4-Vinylphenylboronic acid (148 mg, 1 mmol) was dissolved in
anhydrous 1,4-dioxanne (20 mL). 5-Methyluridine (258 mg) was
added and the mixture was refluxed for 10 h. The solution was
then co-evaporated with toluene under reduced pressure. Yield:
97%; 1H NMR d (ppm) in CDCl3: 9.79 (s, 1H, NH), 7.78 (d,
J = 8.1 Hz, 2ArH), 7.44 (d, J = 8.1 Hz, 2H, ArH), 7.15–7.13 (m, 2H),
6.74 (q, J = 17.7, 10.9 Hz, 2H), 5.85 (d, J = 17.7 Hz, 1H), 5.53 (d,
J = 2.7 Hz, 1H), 5.43–5.28 (m, 3 H), 5.95 (s, 3H).
4. Experimental
4.1. Materials and reagents
Methacrylic acid (MAA), ethylene glycol dimethacrylate
(EGDMA), azo-bis-isobutyronitrile (AIBN), methylmethacrylate
(MMA), acrylamide (AA), divinylbenzene (DVB), and 5-methylu-
ridine were purchased from Acros (Noisy-Le-Grand, France).
4-Vinylphenylboronic acid was purchased from Aldrich (Saint-
Quentin-Fallavier, France). The acetonitrile was distillated on
calcium hydride. All chemicals and solvents were analytical or
HPLC grade and were used without further purification. The
ultra-pure water was provided by a UHQ system (Elga, High
Wycombe, UK).
4.3.2. Synthesis of 20,30-(4-vinyl)phenylboronate-5-
methyluridine tetra-butylammonium hydroxide (10b)
Compound 10a (370 mg, 1 mmol) was dissolved in 5 mL of
anhydrous methanol and 1 mL of 1 M tetra-butylammonium
hydroxide in methanol was added. The mixture was shaked for
10 min and the solvent was evaporated under reduced pressure
at 30 °C. Yield: 98%; 1H NMR
d (ppm) in CDCl3: 7.78 (d,
J = 8.05 Hz, 2H, ArH), 7.44 (d, J = 8.05 Hz, 2H, ArH), 7.15–7.13 (m,
2H), 6.74 (q, J = 17.7, 10.88 Hz, 2H), 5.85 (d, J = 17.7 Hz, 1H), 5.53
(d, J = 2.68, 1 H), 5.43–5.28 (m, 3H), 5.95 (s, 3H).
4.2. Instrumentation
4.4. Synthesis of monomers
4.2.1. NMR
All NMR spectra were realized on a Brücker Advance 2 450 MHz
instrument, (Wissenbourg, France) at room temperature. The tem-
plate (5 mg) was dissolved in deuterated chloroform (1 mL). If nec-
essary, monomer was added corresponding to 1 molar equivalent
for the 2,6-bismethacrylamido-pyridine 2 (MAP, 3.31 mg) and
2,6-diaminopurine-7-vinylbenzyl 3 (DAPVB, 3.62 mg) both func-
tional monomers reacting on the 2 carbonyl functions and NH from
the template or 2 equiv (1.92 mg) of AA because each monomer re-
act on 1 carbonyl. These data were analyzed with the Mestrec soft-
ware ver 4.9.6.0 (Mestrelab Research, Santiago de Compostela,
Spain).
4.4.1. Synthesis of 2,6-bis(methacrylamido)pyridine MAP (11a)
The synthesis of 11a follows an already reported procedure.38
2,6-Diamino-pyridine (109 mg, 1 mmol) was solubilized in dry
pyridine (4 mL) at 0 °C. Methacryloyl chloride (587 lL) was added
on a period of 15 min and the mixture was stirred for 30 min at this
temperature then for 3 h at room temperature. The mixture was
evaporated with toluene. The obtained brown solid was dissolved
in a solution of ammonium acetate saturated and the mixture
was extracted twice with methylene chloride. The organics phases
were dried on magnesium sulfate and then evaporated. The oil was
purified on silica column using methylene chloride/methanol
(99:1, v/v). Yield: 78%; 1H NMR d (ppm) in CDCl3: 8.01 (s, 2H,
NH), 7.92 (d, J = 7.85 Hz, 2H, ArH), 7.64 (t, J = 8.15 Hz, 1H), 5.81
(s, 2H), 5.47 (s, 2H), 2.01 (s, 6H, CH3).
4.2.2. Scanning electron microscopy
Imprinted polymers were deposited onto silicon slides, and
coated with 5 nm of gold using a thermal evaporator (Denton
DV-502A, Denton Vacuum, Moorestown, NJ, USA). Scanning
electron micrographs (SEM) were obtained at 4 kV with Hit-
4.4.2. Synthesis of 2,6-diamino-9-(3/4-vinylbenzyl) purine
DAVBP (12)
2,6-Amino-purine (150 mg, 1 mmol) was solubilized in DMF
(6 mL). Cesium carbonate (1.075 g, 3.3 mmol) was then added.
The mixture was heated at 90 °C, m and p-isomers of vinylbenzyl
achi S4500 equipped with
Japan).
a Field Emission Gun (Hitachi,
4.2.3. BET surface measurements
chloride (171 lL, 1.2 mmol) were then added. The reaction was
A 60 mg quantity of polymers was degassed overnight at 100 °C
to remove adsorbed gases and moisture. Surface areas analyses
were performed at 77 K by Brunauer–Emmett–Teller (BET) on an
ASAP 2020 surface area and porosity analyser (Micromeritics
Instrument Corporation, Creil, France).
kept at this temperature 10 h. The mixture was filtrated to remove
the salts and was evaporated with toluene. The oil was purified on
silica column by methylene chloride/methanol (95:5, v/v). Yield:
40%; 1H NMR d (ppm) in CDCl3: 7.47 (s, 1H, Hpurine), 7.37 (d,
J = 8.2 Hz, 2H, ArH), 7.21 (d, J = 8.2 Hz, 2 H, ArH), 6.69 (dd,
J = 10.9, 17.6 Hz, 1H, 1Hvinyl), 5.73 (d, J = 17.6 Hz, 1H, 1Hvinyl),
5.46 (s, 2H, NH2), 5.26 (d, J = 10.9 Hz, 1H, 1Hvinyl), 5.19 (s, 2H,
CH2), 4.77 (s, 2H, NH2).
4.2.4. HPLC
The instrumentation used to perform this work included a
VWR—Hitachi LaChrom Elite (Tokyo, Japan) HPLC system com-
posed by a quaternary pump equipped with a degasser, an auto-
4.5. Synthesis of MIP and NIP
sampler programmed to inject 20 lL, a column oven set at 25 °C
and a diode array detector acquiring spectra in the 200–300 nm
wavelength range. Analysis was performed in isocratic mode by
The template was introduced in a glass vial and dissolved in
acetonitrile. The functional monomer, the cross-linker, and the ini-
tiator (AIBN) were then added. The tube was degassed for 5 min by
nitrogen then sealed. The polymerization was photochemically ini-
tiated by a standard laboratory UV light source (BLX E-365 5 ꢁ 8W,
365 nm, Vilbert Lourmat, Marne la vallée, France) for 24 h and the
tubes were then heated for 2 h at 70 °C to complete the polymeri-
zation. During the first 2 h of polymerization, the tubes were
using a Hypercarb column (5
l
m, 100 ꢁ 4.6 mm, Thermo Electron,
Shandon, UK) and acetonitrile/25 mM ammonium acetate buffer
pH 5.5 mixture (15:85, v/v) as eluent. The flow rate was 1 mL/
min. These data were computed at a wavelength of 254 nm using
EZChromElite ver 3.1.7 software (Agilent Technologies, Pleasanton,
CA, USA).

D. Jégourel et al. / Bioorg. Med. Chem. 16 (2008) 8932–8939
8939
2. Liebich, H. M.; Müller-Hagedorn, S.; Klaus, F.; Meziane, K.; Kim, K. R.;
Frickenschmidt, A.; Kammerer, B. J. Chromatogr. A 2005, 1071, 271.
3. Rasmuson, T.; Björk, G. R. Acta Oncol. 1995, 34, 61.
4. McEntire, J. E.; Kuo, K. C.; Smith, M. E.; Stalling, D. L.; Richens, J. W. Jr.; Zumwalt,
R. W.; Gehrke, C. W.; Papermaster, B. W. Cancer Res. 1989, 49, 1057.
5. Masuda, M.; Nishihira, T.; Itoh, K.; Mizugaki, M.; Ishida, N.; Mori, S. Cancer
1993, 72, 3571.
6. Zheng, Y. F. Clin. Biochem. 2005, 38, 24.
7. Koshida, K. Urol. Res. 1985, 13, 213.
8. Schram, K. H. Mass Spectrom. Rev. 1998, 17, 131.
9. Yang, J.; Xua, G.; Zheng, Y.; Kong, H.; Pang, T.; Lv, S.; Yang, Q. J. Chromatogr. B
2004, 813, 59.
10. Itoh, K.; Aida, S.; Ishiwata, S.; Yamagushi, T.; Ishida, N.; Mizugaki, M. Clin. Chim.
Acta 1995, 234, 37.
11. Bézy, V.; Chaimbault, P.; Morin, P.; Unger, S. E.; Bernard, M. C.; Agrofoglio, L. A.
Electrophoresis 2006, 27, 2464.
12. Palecek, E. Anal. Chim. Acta 1985, 174, 103.
13. Cho, S. H.; Jung, B. H.; Lee, W. Y.; Kong, G.; Chung, B. C. Biomed. Chromatogr.
2006, 20, 1229.
14. Dudley, E.; Lemiere, F.; Van Dongen, W.; Langridge, J. I.; El-Sharkawi, S.; Games,
D. E.; Esmans, E. L.; Newton, R. P. Rapid Commun. Mass Spectrom. 2001, 15,
1701.
turned of 90° each 15 min. The obtained polymer was removed
from the vial, crushed and soxhlet extracted in methanol/acetic
acid 70:30 overnight. The particles were drilled and sized to
25
l
m.
For MIPs with DAPVB the tube were heated in acetonitrile be-
fore introduction of AIBN until the dissolution and directly poly-
merized on UV because of a poor insolubility of monomers in
the solvent. Porogen solvent mass was equivalent to cross-linker
and functional monomer together. The molar amount of AIBN
was corresponding to 1% molar amount of all compounds. For
each MIP, a corresponding non-imprinted polymer (NIP) was
synthesized in the same conditions but without vinyl-phenylbor-
onate ester template. By using a dummy template (imprinting 5-
methyluridine and rebinding uridine) no interference was possi-
ble in the analysis since the analyte and the template were
different.
15. Wulff, G.; Sarhan, A. Angew. Chem. 1972, 84, 364.
16. Arshady, R.; Mosbach, K. Macromol. Chem. Phys.—Macromol. Chem. 1981, 182,
687.
4.6. Batch rebinding studies
17. Alexander, C.; Andersson, H. S.; Andersson, L. I.; Ansell, R. J.; Kirsch, N.; Nicholls,
I. A.; O’Mahony, J.; Whitcombe, M. J. J. Mol. Recognit. 2006, 19, 106.
18. Nilsson, K. G. I.; Sakaguchib, K.; Gemeiner, P.; Mosbach, K. J. Chromatogr. A
1995, 707, 199.
A mass of 5 mg of polymers was placed in 1.5 mL polypropylene
microtubes and incubated in 0.5 mL of ammonium acetate buffer
solution 10 mM, pH 10, containing 10 mg/L of nucleosides and
19. Allender, C. J.; Brain, K. R.; Ballatore, C.; Cahard, D.; Siddiqui, A.; McGuigan, C.
Anal. Chim. Acta 2001, 435, 107.
54 lL/10 mL of tetra-butylammonium hydroxide 40% (w/v) in
water for 24 h with constant shaking. The polymer was then cen-
trifuged for 10 min at 18,000g (microfuge 22R centrifuge, Beckman
Coulter, Krefeld, Germany) and the concentration of nucleosides in
supernatant was measured by HPLC after filtration on disposable
0.20 lm cellulose acetate syringe filter supplied by Iwaki Glass
(Tokyo, Japan).
20. Spivak, D. A.; Shea, K. J. Macromolecules 1998, 31, 2160.
21. Mayes, A. G.; Andersson, L. I.; Mosbach, K. Anal. Chem. 1994, 222, 483.
22. Wizeman, W. J.; Kofinas, P. Biomaterials 2001, 22, 1485.
23. Parmpi, P.; Kofinas, P. Biomaterials 2004, 45, 1969.
24. Oral, E.; Peppas, N. A. Polymers 2004, 45, 6163.
25. Wulff, G.; Sarhan, A.; Zabrocki, K. Tetrahedron Lett. 1973, 14, 4329.
26. Wulff, G.; Vesper, R.; Grobe-Einsler, R.; Sharhan, A. Macromol. Chem. 1997, 178,
2799.
27. Wulff, G.; Vesper, W. J. Chromatogr. A 1978, 167, 171.
28. Wulff, G.; Lohmar, E. J. Chem. 1979, 18, 279.
4.7. Preparation of doped dog urine sample
29. Tuncel, A.; Ozdemir, A. J. Biomater. Sci. Polym. Ed. 2000, 11, 817.
30. Kanekiyo, Y.; Ono, Y.; Inoue, K.; Sano, M.; Shinkai, S. J. Chem. Soc. Perkins Trans.
2 1999, 557.
31. Kanekiyo, Y.; Ono, Y.; Inoue, K.; Sano, M.; Shinkai, S.; Reinhoudt, D. N. J. Chem.
Soc. Perkin Trans. 2 1999, 2719.
A volume of 1 mL of urine was introduced in 10 mL flask and
1 mL of buffer solution of ammonium acetate 25 mM, pH 10, con-
taining nucleosides at a concentration of 100 ppm was added to
32. Sallacan, N.; Zayats, N. M.; Bourenko, T.; Kharitonov, A. B.; Willner, I. Anal.
Chem. 2002, 74, 702.
33. Hall, A. J.; Manesiotis, P.; Mossing, J. T.; Sellergren, B. Mat. Res. Soc. Symp. Proc.
2002, 723, M1.3.1.
34. Molecularly imprinted polymers. Man made mimics of antibodies and their
applications in analytical chemistry, Techniques and instrumentation in analytical
chemistry; Sellergren, B., Ed.; Elsevier Science B.V.: Amsterdam, 2001; Vol. 23,.
35. Manesiotis, P.; Hall, A. J.; Sellergren, B. J. Org. Chem. 2005, 70, 2729.
36. Manesiotis, P.; Hall, A. J.; Courtois, J.; Irgum, K.; Sellergren, B. Angew. Chem. Int.
Ed. 2005, 44, 3902.
obtain a final concentration of 10 ppm. A volume of 54 lL of Bu4-
NOH (40% wt in water) was introduced in order to obtain a solution
of 80 mM and the volume was adjusted with a ammonium acetate
solution 25 mM at pH 10. The solution was directly extracted on
MISPE cartridges without others treatments.
4.8. Molecularly imprinted solid-phase extraction (MISPE)
37. Shea, K. J.; Spivak, D. A.; Sellergren, B. J. Am. Chem. Soc. 1993, 115, 3368.
38. Spivak, D.; Gilmore, M. A.; Shea, K. J. J. Am. Chem. Soc. 1997, 119, 4388.
39. Sallacan, N.; Zayats, M.; Bourenko, T.; Kharitonov, A. B.; Willner, I. Anal. Chem.
2002, 74, 702.
40. Galetich, I.; Kosevich, M.; Shelkovsky, V.; Stepanian, G.; Blagoi, Y. P.;
Adamowicz, L. J. Mol. Struct. 1999, 478, 155.
41. Bunte, G.; Hürttlen, J.; Pontius, H.; Hartlieb, K.; Krause, H. Anal. Chim. Acta 2007,
591, 49.
42. Yano, K.; Tanabe, K.; Takeuchi, T.; Matsui, J.; Ikebukuro, K.; Karube, I. Anal.
Chim. Acta 1998, 363, 111.
MISPE cartridges were prepared by adding 100 mg of dried and
sieved polymer in an empty solid-phase extraction cartridge be-
tween two glass frits. The cartridges were activated by methanol
then conditioned for 5 min with 2 mL of 10 mM ammonium ace-
tate, pH 11.5, buffer solution containing 800 mM of TBAH. Solution
of urine to analyze was loaded on the cartridge (1 mL). Then, the
cartridge was washed successively with 10 mM ammonium ace-
tate, pH 11.5, buffer solution containing 800 mM of TBAH, 1 mL
of acetonitrile and the analytes were eluted with 1 mL of 1 M HCl
in water/methanol/acetic acid 50:45:5, v/v/v.
43. Sherrington, D. C.; Taskinen, K. A. Chem. Soc. Rev. 2001, 30, 83.
44. Yu, L.; Schneider, H.-J. Eur. J. Org. Chem. 1999, 1619.
45. Beijer, F. H.; Kooijman, H.; Speck, A. L.; Sijbesma, R. P.; Meijer, E. W. Angew.
Chem. Int. Ed. 1998, 37, 75.
46. Beijer, F. H.; Sijbesma, R. P.; Vekemans, J. M.; Meijer, E. W. J. Org. Chem. 1996,
61, 6371.
47. Zhu, X.; Cao, Q.; Hou, N.; Wang, G.; Ding, Z. Anal. Chim. Acta 2006, 561, 171.
48. Connors, K. A. Binding constants. The measurement of molecular complex stability;
John Wiley & Sons: New York, 1987.
49. Abraham, R. J.; Mobli, M. Magn. Reson. Chem. 2007, 45, 865.
50. Li, F.; Zhao, X.; Wang, W.; Xu, G. Anal. Chim. Acta 2006, 580, 181.
51. Rajkumar, R.; Warsinke, A.; Möhwald, H.; Scheller, F. W.; Katterle, M. Biosens.
Bioelectron. 2007, 22, 3318.
52. Dong, X.; Sun, H.; Lü, X.; Wang, H.; Liu, S.; Wang, N. Analyst 2002, 127, 1427.
53. Cederfur, J.; Pei, Y.; Zihui, M.; Kempe, M. J. Comb. Chem. 2003, 5, 67.
54. Le Moullec, S.; Bégos, A.; Pichon, V.; Bellier, B. J. Chromatogr. A 2006, 1108, 7.
55. Branch, G. E. K.; Yabroff, D. L.; Bettman, B. J. J. Am. Chem. Soc. 1934, 56, 937.
56. Bosch, L. I.; Fyles, T. M.; James, T. D. Tetrahedron 2004, 60, 11175.
57. Smith, B. D.; Davis, J. P.; Draffin, S. P.; Duggan, P. J. Supramol. Chem. 2004, 16, 87.
58. Kirby, A. J.; Marriott, R. E. J. Chem. Perkin Trans. 2 2002, 422.
Acknowledgments
These studies were supported by an ‘ANR-05-BLAN-0368-03’.
F.B. is grateful to ANR for post-doctoral fellowship and D.J. to the
Région Centre—Merck Estapor for Ph.D. fellowship. Thanks are
due to Mr. Jean Battini from Micromeretics for the supply of BET
experiments.
References and notes
1. Kammerer, B.; Frickenschmidt, A.; Müller, C. E.; Laufer, S.; Gleiter, C. H.; Liebich,
H. Anal. Bioanal. Chem. 2005, 382, 1017.