1H NMR study of the complexation of aromatic drugs with dimethylxanthine derivatives

Article abstract of DOI:10.1016/j.molstruc.2011.11.045

With an aim of searching efficient interceptors of aromatic drugs, the self- and hetero-association of dimethylxanthine derivatives with different structures, selected according to Strategy 1 (variation of the position of methyl groups) and Strategy 2 (variation of the length of (CH₂) _nCOOH group), with aromatic drug molecules: Ethidium Bromide, Proflavine and Daunomycin, were studied using ¹H NMR spectroscopy. It was found that the association proceeds in a form of stacking-type complexation and its energetics is relatively independent on the structure of the dimethylxanthines. However, on average, the dimethylxanthines possess higher hetero-association constant and, hence, higher interceptor ability as compared to the trimethylxanthine, Caffeine, used during the past two decades as a typical interceptor molecule.

Full text of DOI:10.1016/j.molstruc.2011.11.045

Journal of Molecular Structure 1010 (2012) 139–145
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
1H NMR study of the complexation of aromatic drugs
with dimethylxanthine derivatives
A.A. Hernandez Santiago a, M. Gonzalez Flores a, S.A. Rosas Castilla a, A.M. Cervantes Tavera a,
R. Gutierrez Perez a, V.V. Khomich b, D.V. Ovchinnikov c, H.G. Parkes c, M.P. Evstigneev b,
⇑
a Faculty of Chemical Sciences, Autonomous University of Puebla, Puebla, Mexico
b Department of Physics, Sevastopol National Technical University, Sevastopol 99053, Ukraine
c Institute of Cancer Research, University of London, Sutton, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 13 October 2011
Received in revised form 25 November 2011
Accepted 27 November 2011
Available online 6 December 2011
With an aim of searching efficient interceptors of aromatic drugs, the self- and hetero-association of dim-
ethylxanthine derivatives with different structures, selected according to Strategy 1 (variation of the
position of methyl groups) and Strategy 2 (variation of the length of A(CH2)nACOOH group), with aro-
matic drug molecules: Ethidium Bromide, Proflavine and Daunomycin, were studied using 1H NMR spec-
troscopy. It was found that the association proceeds in a form of stacking-type complexation and its
energetics is relatively independent on the structure of the dimethylxanthines. However, on average,
the dimethylxanthines possess higher hetero-association constant and, hence, higher interceptor ability
as compared to the trimethylxanthine, Caffeine, used during the past two decades as a typical interceptor
molecule.
Keywords:
Aromatic molecules
Dimethylxanthine
Hetero-association
NMR
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
means of directed search of xanthine molecules which specifically
target certain aromatic drugs and feature high xanthine–drug het-
Xanthine-type molecules, in particular, Caffeine and its deriva-
tives, have long been known to interfere with biological action of
DNA-targeting aromatic drugs, leading in some cases to lowering
of their toxicity in vitro [1–5]. This effect may potentially have a
very important application in clinical practice, e.g. for rapid regula-
tion of the toxicity level after overdosing [6]. It appears as being
generally accepted now that the mechanism of such synergistic
action in drug-xanthine systems can be explained in terms of
two basic molecular processes acting simultaneously, viz. the inter-
ceptor (hetero-association of the xanthine and the drug) and pro-
tector (removal of the drug from DNA due to xanthine–DNA
binding) [for review see Ref. [6]]. Although there is still some de-
bate on the interrelation of these two processes, in both cases on
addition of xanthine molecules to drug-containing biological sys-
tem, the concentration of drug–DNA complexes decreases result-
ing in alteration of the net biological effect of the drug. The
binding density of the aromatic drug on DNA has been shown to
correlate with xanthine–drug hetero-association constant [7,8],
which suggests the principal possibility of amplifying the intercep-
tor action and subsequent decrease in the net drug’s toxicity by
ero-association constant.
Within the whole group of xanthine derivatives, the dimethylx-
anthines, containing two methyl groups in positions 1, 3 or 7 of the
xanthine chromophore (Fig. 1), act as physiological derivatives
(metabolites) of Caffeine [9], and for some of them the interceptor
action towards aromatic drugs has been reported [3,5,10]. With an
aim of seeking the relation between the structure of dimethylxan-
thine molecules and the hetero-association constant with aromatic
drugs under conditions close to physiological (pH ꢀ 7), in the pres-
ent work we employed two strategies:
(i) Strategy 1, viz. the variation of the position of two methyl
groups in xanthine chromophore, giving Theophylline
(THP, 1,3-dimethylxanthine), Theobromine (THB, 3,7-dim-
ethylxanthine) and Paraxanthine (PARA, 1,7-dimethylxan-
thine) (Fig. 1). The rationale behind this strategy is an
assumption that the alteration of the position of the methyl
groups may modulate the hydrophobic contribution, known
as one of the principal determinants of the hetero-associa-
tion of aromatic molecules [11], and
(ii) Strategy 2, viz. the introduction of A(CH2)nACOOH group to
position 7 of the xanthine chromophore, giving Theophyl
line-7-acetic acid (AA, R = CH2ACOOH), Theophylline-7-pro-
pionic acid (PA, R = (CH2)2ACOOH) and Theophylline-7-
butyric acid (BA, R = (CH2)3ACOOH) (Fig. 1). The rationale
⇑
Corresponding author. Address: Department of Physics, Sevastopol National
Technical University, Universitetskaya Str., 33, Sevastopol 99053, Ukraine. Fax:
+380 692 243 590.
E-mail address: max_evstigneev@mail.ru (M.P. Evstigneev).
0022-2860/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2011.11.045

140
A.A. Hernandez Santiago et al. / Journal of Molecular Structure 1010 (2012) 139–145
Caffeine
Theophylline
Theobromine
Paraxanthine
Theophylline-7-Acetic acid (AA, R=CH2-COOH)
-Propionic acid (PA, R=(CH2)2-COOH)
-Butyric acid (BA, R=(CH2)3-COOH)
Daunomycin
Proflavine
Ethidium Bromide
Fig. 1. Structures of the dimethylxanthines and aromatic drugs.
behind this strategy is an assumption that A(CH2)nACOOH
Antibiotic Daunomycin (DAU), mutagens, Ethidium Bromide
(EB) and Proflavine (PF) (Fig. 1), which were used in the past to tes-
tify the interceptor action of the xanthines in vitro [for review see
Ref. [6]], were taken as typical representatives of aromatic drugs.
DAU, EB and PF ligands are known as being charged under neutral
pH (see the charge localisation in Fig. 1 [6]), whereas the A(CH2)n
ACOOH group in THP derivatives is ionised and bears negative
charge. The latter provides alternative to the formation of the
moiety, which is flexible and contains hydrogen donor/
acceptor groups, may form intermolecular hydrogen bond
to aromatic drug molecule within the hetero-complex. The
intermolecular H-bonds are known to provide additional
stabilization of the hetero-complexes of aromatic molecules
in solution which is monitored by increased magnitude of
the hetero-association constant [12,13].

A.A. Hernandez Santiago et al. / Journal of Molecular Structure 1010 (2012) 139–145
141
H-bond mechanism of possible additional stabilization of the het-
ero-complexes within the Strategy 2 due to specific electrostatic
attraction.
corresponded to the MM3 force field. Prior to energy minimisa-
tions the initial structures of the dimethylxanthine dimers and
1:1 dimethylxanthine–drug hetero-complexes were built up from
analysis of the induced chemical shifts of aromatic protons of the
interacting molecules.
2. Experimental
3. Results and discussion
2.1. Materials
3.1. The self-association of dimethylxanthines
Daunomycin from ‘Fluka’, Proflavine, Ethidium Bromide, Theo-
bromine, Paraxanthine, Theophylline-7-acetic acid from ‘Sigma’
(Fig. 1) were used without further purification. The samples were
lyophilised from D2O solutions and re-dissolved in 0.1 M phos-
phate buffer in 99.95% D2O at pD 7.1, containing 10ꢁ4 M EDTA.
The typical concentrations of aromatic molecules, used in NMR
experiment, fall in millimolar concentration range which requires a
consideration of indefinite self-association of dimethylxanthines
and the drug molecules, as well as the indefinite hetero-association
between the xanthines and the drugs. Hence, prior to the analysis
of the hetero-association, it is necessary to get the full set of equi-
librium parameters, describing the self-association, to be used fur-
ther as an input data in the hetero-association analysis [10,14]. The
self-association study of the dimethylxanthines was performed in
the present work in similar solution conditions (pD 7.1, 0.1 M
Na-phosphate buffer). The same conditions were also used before
in the study of the self-association of DAU, PF, EB and Theophylline
[15], so a comparative analysis of these systems is meaningful.
Experimental concentration dependences of chemical shifts of
aromatic protons of the dimethylxanthines (data not shown) were
fitted by means of indefinite self-association model based on com-
mon assumption that aromatic molecules aggregate with the equi-
librium constant, K, independent on the number of molecules in
aggregates (EK-model [16]):
2.2. Preparation of Theophylline-7-propionic acid (PA) and
Theophylline-7-butyric Acid (BA)
Forty eight milligrams (1 mmol) of sodium hydride (as 50% dis-
persion) were added to 180 mg (1 mmol) of Theophylline, dis-
solved in 5 ml of anhydrous dimethylformamide at room
temperature, while stirring. After the evolution of hydrogen has
ceased, 215 mg (1.1 mmol) of 4-bromobutyric acid ethyl ester
were added under stirring and the mixture was then heated to
90° C for 4 h. Volatiles were removed in vacuo, and the residue
was taken up in ethyl acetate, filtered and washed twice with
water. After removing of solvent in vacuo and crystallization of
the residue (iPrOH/petrol ether), 271 mg (92%) of ester were
obtained. The purified ester (236 mg, 0.8 mmol) was dissolved in
3 ml of methanol/1 M NaOH solution and the solution was gently
warmed for 2 h. After TLC-monitoring had shown no more starting
material, the mixture was cooled down to 0 °C and acidified with
diluted HCl. Precipitated white solid was filtered off, washed with
water and crystallized from aqueous methanol to give 184 mg
(86%) of pure Theophylline-7-butyric acid. The structure was con-
firmed by MS(ESI) and NMR data. Similarly, Theophylline-7-propi-
onic acid was prepared.
!
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2Kx0 þ 1 ꢁ ð4Kx0 þ 1Þ
d ¼ dm þ 2ðdd ꢁ dmÞ
;
ð1Þ
2Kx0
where dm, dd are proton chemical shifts in monomer and dimer
forms (or on the edge of aggregates) of a dimethylxanthine in solu-
tion, respectively; x0 is the total concentration.
The adjustable parameters in this model are K/dm/dd. Their val-
ues are presented in Table 1. The self-association parameters for
the trimethylxanthine, Caffeine (CAF), are also presented here for
the purposes of comparison. In numerical analysis of the self-,
and hetero-association of AA/PA/BA (see below), we used only
the aromatic protons directly attached to the xanthine chromo-
phore (H8, 1Me, 3Me) and ignored the ethyl protons of the
A(CH2)nACOOH group. This is because the latter are strongly af-
fected by local shielding effects, which may not feature an additive
property of magnetic shielding due to aromatic ring-current effect,
put into basis of the models of self- and hetero-association used in
the present work.
2.3. NMR measurements
500 MHz 1H NMR spectra were recorded on Varian spectrome-
ter with the residual water peak saturated during relaxation. Signal
assignments of the non-exchangeable protons of the drugs were
obtained using both two-dimensional homonuclear TOCSY and
ROESY experiments. Chemical shift measurements of the non-
exchangeable protons of the aromatic molecules were made as a
function of concentration of the drug, maintaining the concentra-
tion of dimethylxanthine fixed (y0 = 2 mM) – in the hetero-associ-
ation analysis, and as
a function of the dimethylxanthines
Due to relatively low solubility of THB (62 mM) the concentra-
tion dependence of d for this molecule within the available concen-
tration range (0.05–2 mM) used in NMR experiment, was too weak
to perform reliable self-association analysis (chemical shift
changes were within 0.003–0.006 ppm). In such case the magni-
tudes of dm parameter for THB protons were obtained by extrapo-
lation of the concentration curves down to zero concentration.
It is seen from Table 1 that the magnitudes of equilibrium self-
association constants for the dimethylxanthines, PARA/THP/AA/PA/
BA, are similar within the error limits which means that the posi-
tion of the methyl groups in the xanthine chromophore and the
distance of the ACOOH group from the chromophore do not affect
the energetics of the self-association. As discussed above, the value
of K for THB was not possible to determine from experimental data
available, however, the insensitivity of the equilibrium constant to
the position of the methyl groups allows to take the K value for
THB as intermediate between the K for PARA and THP (see Table
1), which is required for further hetero-association analysis. It
concentration – in the self-association analysis. All measurements
were made at T = 298 K.
All sets of NMR measurements were made in the fast-exchange
condition on the NMR timescale. Chemical shifts were measured
relative to TMA (tetramethylammonium bromide) as an internal
reference and recalculated with respect to DSS (sodium 2,2-di-
methyl-2-silapentane-5-sulphonate), i.e. DSS = TMA + 3.178 (ppm).
2.4. Structure calculations
Calculations of the spatial structures of the 1:1 dimethylxan-
thine–drug complexes in an aqueous environment have been made
by molecular mechanics methods using X-PLOR software with
the Charmm22 force field, as previously [10,12,13]. Modelling of
the aqueous environment was performed by water molecules in
the form of TIP3P, placed in rectangular box (1100 molecules).
Parameters of non-valent interactions between atoms

142
A.A. Hernandez Santiago et al. / Journal of Molecular Structure 1010 (2012) 139–145
Table 1
Calculated parameters of the self-association of dimethylxanthines (0.1 M Na-phosphate buffer in D2O, pD 7.1, T = 298 K).
a
Xanthine
dm (ppm)
dd (ppm)
K (Mꢁ1
)
H8
7Me
3Me
1Me
H8
7Me
3Me
1Me
Strategy 1 (variation of the position of methyl groups)
THB
7.90
7.83
8.02
7.89
3.94
3.94
–
3.49
–
3.58
3.54
–
–
7.2b
7.1 0.5
7.4 0.4
11.8 0.3
PARA
3.32
3.37
3.35
7.80
7.81
7.83
3.89
–
3.82
–
3.28
3.33
3.26
3.12
3.16
THP [15]
CAF [14]
3.95
Strategy 2 (variation of the length of A(CH2)nACOOH group)
AA (n = 1)
PA (n = 2)
BA (n = 3)
7.92
7.93
7.96
–
–
–
3.56
3.53
3.54
3.35
3.34
3.35
7.87
7.90
7.79
–
–
–
3.32
3.32
3.37
3.19
3.16
3.21
7.0 3.0
7.8 2.0
6.4 1.0
a
The errors for the equilibrium constant were calculated from the values of K, computed over individual protons.
The value of K for THB was taken as intermediate between the self-association constants for PARA and THP (see discussion in the text).
b
should, however, be noted that the trimethylxanthine, Caffeine,
aggregates with the equilibrium constant, K, being higher than that
for the dimethyxanthines. This result may be explained by the
hydrophobic contribution from the additional methyl group in
the structure of CAF as compared to the dimethylxanthines.
Analysis of the computed magnitudes of the monomer (dm) and
dimer (dd) chemical shifts in Table 1 enables to draw certain con-
clusions on the influence of the structure of the dimethylxanthine
molecule on the magnetic shielding of aromatic protons. The val-
ues of dm for 7Me group are nearly identical for THB and PARA,
hence, the position of the second methyl group (in 1 or 3 position)
does not affect the chemical shift of this group. On the other hand,
H8 proton of THB and PARA appears to be sensitive to the presence
of the 3Me or 1Me groups (i.e. dm(H8) experiences 0.07 ppm shift in
these dimethylxanthines) which is expected because of much low-
er electron density in the vicinity of H8 as compared to the methyl
protons. At the same time the dm chemical shifts of 1Me group for
THP and CAF are also close to each other suggesting that the exis-
tence of 7Me group does not affect the monomer chemical shift of
the 3Me group. Hence, the variation of the position of two methyl
groups in the dimethylxanthines (i.e. the Strategy 1) does not influ-
ence the magnetic shielding of the methyl groups, but affects the
shielding of H8 proton. In general, similar conclusions can be made
a consequence of a stacking-type aggregation when the xanthine
chromophores in a complex are parallel to each other creating
favourable conditions for the magnetic shielding of all aromatic
protons. It thus may be stated that the mode of the self-association
of all of the investigated dimethylxanthine derivatives is likely to
be the stacking-type aggregation.
3.2. The hetero-association of dimethylxanthines with aromatic drugs
The typical concentration dependences of the proton chemical
shifts for the drugs and the dimethylxanthines studied in the
mixed solutions are given in Fig. 2. It is seen that on variation of
the drugs’ concentration the signals from aromatic protons of the
dimethylxanthines move upfield. The latter indicates that the
interaction between the drug and dimethylxanthine molecules oc-
curs in solution and most likely proceeds in a form of the stacking-
type association, also reported before for other drug–xanthine sys-
tems [14,15,17] and stated above for the self-association of the
dimethylxanthines. Such concentration dependences were fitted
employing the same model as used before in analysis of various
drug–xanthine systems, which takes into account indefinite self-
and hetero-association of X (drug) and Y (dimethylxanthine) mol-
ecules [14,15]:
ð2Þ
with respect to the introduction of acetic, propionic and butyric
groups into the structure of Theophylline (i.e. the Strategy 2, see
Table 1), viz. the changes of dm (3Me/1Me) are relatively small
and the changes of dm(H8) are relatively high as compared with
the same protons of THP.
An important conclusion which comes from inspection of the
data in Table 1 is that for the whole set of aromatic protons, with-
out exceptions, the magnetic shielding effect is observed as a result
of the self-association i.e. dm > dd. This effect has previously been
reported for other xanthine derivatives [14,15] and interpreted as
where dh and Kh are the proton chemical shift in hetero-complex
and the equilibrium hetero-association constant, respectively, act-
ing as search parameters; x1, y1 are the concentrations of
monomers.
The magnitudes of dm, dd and equilibrium self-association con-
stants, KX, KY, for PARA/THB/AA/PA/BA/DAU/EB/PF molecules are
known from previous self-association studies (see Table 1 for the
dimethylxanthines and Ref. [15] for DAU/EB/PF). Monomeric con-
centrations, x1 and y1, in Eq. (2) may be obtained from the solution
of the system of mass balance Eq. (3) in each experimental point:

A.A. Hernandez Santiago et al. / Journal of Molecular Structure 1010 (2012) 139–145
143
δ, ppm
δ, ppm
H8 (THB)
8
H8 (PARA)
H2 (DAU)
H2 (DAU)
H1 (DAU)
H3 (DAU)
H1 (DAU)
H3 (DAU)
6
H1' (DAU)
H1' (DAU)
4Me (DAU)
4
4
2
4Me (DAU)
7Me (PARA)
3Me (PARA)
H10e (DAU)
H10a (DAU)
H8e (DAU)
H8e (DAU)
7Me (THB)
3Me (THB)
H10e (DAU)
H10a (DAU)
H8e (DAU)
2
H8e (DAU)
Сdau, mM
Cdau, mM
0,0
0,5
1,0
1,5
2,0
2,5
3,0
0,5
1,0
1,5
2,0
2,5
3,0
(a)
(b)
9
δ, ppm
9
8
7
δ, ppm
H1 (EB)
H10 (EB)
H8 (PARA)
H1 (EB)
8
7
H10 (EB)
H8 (THB)
H9 (EB)
H4 (EB)
H2 (EB)
H9 (EB)
H4 (EB)
H2 (EB)
H7 (EB)
6
4
H7 (EB)
6
4
7Me (PARA)
7Me (THB)
3Me (THB)
Сeb, mM
1Me (PARA)
Сeb, mM
3
3
0
2
4
6
8
10
12
0
2
4
6
8
10
12
(d)
(c)
δ, ppm
δ, ppm
8
8
H9 (PF)
H8 (PARA)
H1/8 (PF)
H9 (PF)
H1/8 (PF)
H2/7 (PF)
H4/5 (PF)
H2/7 (PF)
6
4
H4/5 (PF)
6
4
H8 (THB)
7Me (PARA)
1Me (PARA)
7Me (THB)
3Me (THB)
Сpf, mM
Сpf, mM
2
2
0
2
4
6
8
10 12 14 16
0
2
4
6
8
10 12 14
(e)
(f)
Fig. 2. Experimental dependences of the chemical shifts of aromatic protons of PARA and THB on EB/PF/DAU concentration in the mixed solutions: (a) THB–DAU, (b) PARA–
DAU, (c) THB–EB, (d) PARA–EB, (e) THB–PF, (f) PARA–PF.
8
h
i
i
Kh2
2
y12
x1
y1
1ꢁKY y1
x1y1
Analysis of the computed values of equilibrium hetero-associa-
tion constants in Table 2 suggests that there is a relatively
insignificant change in Kh on variation of the structure of dimeth-
ylxanthine molecules in either the Strategy 1 and Strategy 2. It
was previously shown that the existence of apparent additional
stabilization of the hetero-complexes of aromatic molecules via
>
<
x0 ¼
y0 ¼
1 þ Kh
þ
þ
þ K2
ð1ꢁKX x1Þ2
ð1ꢁKY
y
1Þ2
h
ð1ꢁKY y1Þð1ꢁKX x1Þ
ð3Þ
h
Kh2
2
x12
>
:
y1
x1
1ꢁKX x1
x1y1
1 þ Kh
þ K2
h
ð1ꢁKY
y
1Þ2
ð1ꢁKX x1Þ2
ð1ꢁKX x1Þð1ꢁKY y1Þ
The results of computation of the hetero-association parame-
ters for the systems studied are given in Tables 2 and 3.

144
A.A. Hernandez Santiago et al. / Journal of Molecular Structure 1010 (2012) 139–145
Table 2
is likely a consequence of the presence of additional methyl group
in the structure of CAF molecule providing additional steric hin-
drance in the hetero-complexes. It follows that the interceptor
ability of the dimethylxanthines towards aromatic drugs is higher
than that of CAF to the same drugs and is relatively unspecific to
the drug itself.
Calculated values of equilibrium constants (Mꢁ1
) of the hetero-association of
dimethylxanthines with aromatic drugs (0.1 M Na-phosphate buffer in D2O, pD 7.1,
T = 298 K).
System
DAU
PF
EB
Strategy 1 (variation of the position of methyl groups)
THB
PARA
THP[15]
CAF[14]
160 40
110 20
190 30
250 30
155 15
180 20
160 17
170 20
The calculated magnitudes of the induced proton chemical
102
102
62
9
6
4
shifts,
3. It is seen that the aromatic protons of THB and PARA are sys-
tematically shielded ( dh > 0) in the complexes with all drugs
D
dh = dm ꢁ dh, in 1:1 hetero-complexes are given in Table
72
4
D
Strategy 2 (variation of the length of A(CH2)nACOOH group)
studied, which is in accord with the previously reported results
on the hetero-association of THP and CAF with the same drugs
[14,15]. It also confirms the conclusion, made above on the basis
of analysis of experimental concentration dependences, on the
stacking-type hetero-association of the dimethylxanthines with
aromatic drugs. Similar shielding is also observed for PF and EB
aromatic protons, but some of the DAU protons appear to be
AA
PA
BA
180 30
160 10
150 10
250 30
120 15
200 20
125 10
120 10
170 20
Note: the errors were calculated from the values of Kh, computed over individual
protons.
intermolecular H-bond, charge-transfer interactions or specific
slightly deshielded (Ddh < 0, see Table 3). The deshielding of the
electrostatic attraction is exerted through high magnitude of Kh
DAU protons in the hetero-complex with Theophylline was previ-
ously interpreted as the result of considerable difference in
dimensions of the antibiotic and THP molecules [15], and posi-
tioning of the xanthine molecule above the two central rings of
DAU. This leads to the situation when peripherial aromatic pro-
tons H2, H3, 4Me, H10 of the antibiotic fall out of the shielding
cone of the THP chromophore and results in the net deshielding
(or nearly zero shielding) effect. It is possible to conclude that
similar deshielding effect is the case for the drug-dimethylxan-
thine systems studied in the present work. Relatively high shield-
ing of the H1 proton as compared to the neighbouring H2/H3
protons in the structure of DAU (see Fig. 1) is an artefact and is
most likely due to local magnetic shielding from neighbouring
polar C@O group of DAU, which was also noted before in THP-
drug systems [15].
as compared to the self-association of X and Y molecules [12,13].
Kh
A formal criterion fh ¼
was suggested in order to estimate
relative importance of KtXhþeKhhþKeYtero-association reactions in overall
dynamic equilibrium in solution, and the threshold, fh P 35%,
was established enabling to separate the systems which are likely
stabilized by some additional factor such as H-bonds or charge-
transfer [18]. The magnitude of the fh factor takes the maximum
value, fh = 35%, only for EB-BA system and the lower values for all
other systems. This result suggests that the studied systems most
likely do not feature any additional stabilization of 1:1 hetero-
complexes in solution and the hetero-association is governed by
a balance of van der Waals, electrostatic and hydrophobic forces,
as observed before for the majority of other hetero-associations
of aromatic molecules [11]. The existence of intermolecular
charge-transfer, H-bonding or specific electrostatic attraction,
although being possible in principle, does not contribute signifi-
cantly to the hetero-association of the dimethylxanthines with
aromatic drugs. The highest, on average, value of Kh is observed
for PF-dimethylxanthine systems (Table 2) which is explained by
the absence of bulky side chains in the structure of PF molecule
as compared to EB and DAU (see Fig. 1), thus lowering the steric
hindrance of the interacting molecules in the hetero-complexes.
This conclusion is also supported by the fact that for all the drugs
studied the Kh values for the trimethylxanthine, CAF, are lower
than the Kh for any of the dimethylxanthines (see Table 2). This
Based on the calculated magnitudes of
Ddh in Table 3 it is pos-
sible to build initial structures of 1:1 hetero-complexes of the
investigated systems for further energy minimisation by means
of molecular modelling methods (see Supplementary material).
As seen from the Table 3 the shielding profile for the drugs’ protons
is qualitatively similar for all dimethylxanthines studied, which
suggests qualitatively similar structures of their hetero-complexes,
also confirmed by the calculated structures. Relatively high shiel-
dings of the dimethylxanthines’ protons (mainly >0.5 ppm) means
that they are located in close vicinity of the aromatic chromoph-
ores of the drugs.
Table 3
Induced chemical shifts of the protons of dimethylxanthines and aromatic drugs on hetero-association (0.1 M Na-phosphate buffer in D2O, pD 7.1, T = 298 K).
System
DAU protons
H2
Xanthine protons
H1
0.06
H3
4Me
H1’
H10e
H10a
H8e
H8
7Me
3Me
1Me
DAU + THB
DAU + PARA
DAU + THP [15]
ꢁ0.02
ꢁ0.07
0.03
ꢁ0.02
ꢁ0.08
0.03
ꢁ0.01
ꢁ0.04
0.02
ꢁ0.07
ꢁ0.68
ꢁ0.06
0.01
ꢁ0.31
0.01
ꢁ0.02
ꢁ0.18
ꢁ0.03
0.03
0.01
0.004
0.36
0.71
0.46
0.30
0.75
–
0.37
–
0.55
–
0.72
0.47
ꢁ0.01
0.25
PF protons
Xanthine protons
H8
H9
H1/8
H2/7
H4/5
7Me
3Me
1Me
PF + THB
PF + PARA
PF + THP [15]
0.34
0.31
0.53
0.20
0.17
0.31
0.10
0.06
0.16
0.40
0.35
0.61
0.62
0.86
0.54
0.62
0.89
–
0.66
–
0.63
–
0.85
0.57
EB protons
H1
Xanthine protons
H10
H4
H2
H7
H9
H8
7Me
3Me
1Me
EB + THB
EB + PARA
EB + THP [15]
0.15
0.35
0.54
0.45
0.39
0.56
0.25
0.22
0.32
0.16
0.16
0.23
0.14
0.13
0.20
0.14
0.16
–
0.38
0.54
0.53
0.53
0.68
–
0.59
–
0.66
–
0.85
0.77

A.A. Hernandez Santiago et al. / Journal of Molecular Structure 1010 (2012) 139–145
145
Analysis of the shielding profile of aromatic protons of the drugs
Acknowledgements
and dimethylxanthines studied (data not shown) for the THP-7-
AA/BA/PA systems (i.e. the Strategy 2) did not reveal any clear
This work was supported, in part, by BUAP PROMEP/103.5/07/
2208 grant. Prof. David Gadian (ICH) is thanked for use of the
NMR spectrometer.
specificity in terms of some pattern in the dependence of
Ddh on
the length of A(CH2)nACOOH group. The latter is quite expected
as the shielding of aromatic protons should be little affected by
the length of the side chains attached to the chromophore. Analysis
of the calculated spatial structures of the drug-AA/BA/PA hetero-
complexes revealed no apparent hydrogen bonding between the
ACOOH group of the dimethylxanthines and the drugs (see Supple-
mentary material) which is in accord with the same conclusion
made above on the basis of the calculated values of equilibrium
hetero-association constant and the fh factor.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molstruc.2011.11.045.
References
[1] F. Traganos, J. Kapuscinski, Z. Darzynkiewicz, Cancer Res. 51 (1991) 3682.
[2] J. Piosik, M. Zdunek, J. Kapuscinski, Biochem. Pharmacol. 63 (2002) 635.
[3] J. Piosik, K. Ulanowska, A. Gwizdek-Wisniewska, A. Czyz, J. Kapuscinski, G.
Wegrzyn, Mutat. Res. 530 (2003) 47.
4. Conclusions
[4] A. Osowski, M. Pietrzak, Z. Wieczorek, J. Wieczorek, J. Toxicol. Environ. Health
Part A 73 (2010) 1141.
[5] A. Woziwodska, A. Gwizdek-Wisniewska, J. Piosik, Bioorg. Chem. 39 (2011) 10.
[6] M.P. Evstigneev, DNA-Binding Aromatic Drug Molecules, Physico-Chemical
Interactions and their Biological Roles, Lambert Academic Publishing,
Germany, 2010.
[7] M.P. Evstigneev, A.O. Lantushenko, V.P. Evstigneev, Y.V. Mykhina, D.B. Davies,
Biophys. Chem. 132 (2008) 148.
[8] M.B. Lyles, I.L. Cameron, Cell Biol. Int. 26 (2002) 145.
[9] A. Lelo, J.O. Miners, R. Robson, D. Birkett, Clin. Pharmacol. Theor. 39 (1986) 54.
[10] A.A. Hernandez Santiago, D.D. Andrejuk, A.M. Cervantes Tavera, D.B. Davies,
M.P. Evstigneev, J. Biol. Phys. 35 (2009) 115.
[11] V.V. Kostjukov, N.M. Khomytova, A.A. Hernandez Santiago, A.M. Cervantes
Tavera, J. Salas Alvarado, M.P. Evstigneev, J. Chem. Thermodyn. 43 (2011) 1424.
[12] D.B. Davies, D.A. Veselkov, V.V. Kodintsev, M.P. Evstigneev, A.N. Veselkov, Mol.
Phys. 98 (2000) 1961.
[13] M.P. Evstigneev, Yu.V. Mukhina, D.B. Davies, Mol. Phys. 104 (2006) 647.
[14] M.P. Evstigneev, V.V. Khomich, D.B. Davies, Eur. Biophys. J. 36 (2006) 1.
[15] D.D. Andrejuk, A.A. Hernandez Santiago, V.V. Khomich, V.K. Voronov, D.B.
Davies, M.P. Evstigneev, J. Mol. Struct. 889 (2008) 229.
[16] D.B. Davies, L.N. Djimant, A.N. Veselkov, J. Chem. Soc. Faraday Trans. 92 (1996)
383.
The results of the present work suggest that the self- and het-
ero-association of dimethylxanthine derivatives with different
structures, selected according to Strategy 1 (variation of the po-
sition of the methyl groups) and Strategy 2 (variation of the
length of A(CH2)nACOOH group), with aromatic drug molecules
proceeds in a form of a stacking-type complexation. The energet-
ics of the self- and hetero-association was found to be relatively
independent on the structure of the dimethylxanthines, which
means that the introduction of a potential H-bond donor/accep-
tor A(CH2)nACOOH group into the structure of the xanthine chro
mophore does not lead to additional stabilization of the hetero-
complexes with aromatic drugs. However, it was also found that,
on average, the dimethylxanthines possess higher hetero-associ-
ation constant and, hence, higher interceptor ability as compared
to the trimethylxanthine, Caffeine, used during the past two
decades as a typical interceptor molecule. These results provide
further step towards search of more efficient interceptors of aro-
matic drugs.
[17] G. Gattuso, G. Manfredi, S. Sammartano, Fluid Phase Equilib. 308 (2011) 47.
[18] M.P. Evstigneev, D.B. Davies, A.N. Veselkov, Chem. Phys. 321 (2006) 25.