Pharmacokinetic profile of atenolol aspirinate

Article abstract of DOI:10.1002/ardp.200700070

We report microwave-assisted synthetic routes, the pharmacokinetic profile along with results from ulcerogenicity and mutagenicity studies of atenolol aspirinate, and an already described derivative, in which acetyl salicylic acid (aspirin) was connected to atenolol by an ester linkage. Atenolol aspirinate was stable towards aqueous hydrolysis but rapidly hydrolyzed in plasma (t_1/2 = 7.6 min). The results showed that the rapid and complete hydrolysis generates atenolol salicylate, which assumes a conformation stabilized by two intramolecular H-bonds, avoiding its further hydrolysis to salicylic acid and atenolol.

Full text of DOI:10.1002/ardp.200700070

Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
A. C. Montes-Gil et al.
445
Full Paper
Pharmacokinetic Profile of Atenolol Aspirinate
Ana C. Montes-Gil1, Marcos Zanfolin3, Cristina E. Okuyama1, Sergio Lilla4, Delma P. Alves3,
Vincenzo Santagada2, Elisa Perissutti2, Antonio Lavecchia2, Ferdinando Fiorino2,
Beatrice Severino2, Giuseppe Caliendo2, Fernanda B. M. Priviero1, Gustavo D. Mendes4,
Jose L. Donato1, Gilberto de Nucci1
1 Department of Pharmacology, Faculty of Medical Sciences, State University of Campinas, Campinas, S¼o
Paulo, Brazil
2 Department of Pharmaceutical Chemistry, University of Naples ”Federico II”, Naples, Italy
3 CEMIB -Multidisciplinary Center for Biological Investigation. State University of Campinas, Campinas, S¼o
Paulo, Brazil
4
Galeno Research Unit, Campinas, S¼o Paulo, Brazil
We report microwave-assisted synthetic routes, the pharmacokinetic profile along with results
from ulcerogenicity and mutagenicity studies of atenolol aspirinate, and an already described
derivative, in which acetyl salicylic acid (aspirinm) was connected to atenolol by an ester linkage.
Atenolol aspirinate was stable towards aqueous hydrolysis but rapidly hydrolyzed in plasma (tꢀ
= 7.6 min). The results showed that the rapid and complete hydrolysis generates atenolol salicy-
late, which assumes a conformation stabilized by two intramolecular H-bonds, avoiding its fur-
ther hydrolysis to salicylic acid and atenolol.
Keywords: Antihypertensive therapy / Atenolol / Microwave synthesis / Pharmacokinetics / Prodrug /
Received: April 4, 2007; accepted: May 30, 2007
DOI 10.1002/ardp.200700070
effective for both primary and secondary prevention of
MI, stroke and cardiovascular death [5, 6], and in the
acute management of MI, unstable angina, and embolic
stroke [7–9]. Acetyl salicylic acid imparts its primary
antithrombotic effects through the irreversible inhibi-
tion of platelet cyclooxygenase-1 (COX-1), thus prevent-
ing the formation of thromboxane A2 (TXA2) [10], potent
aggregating, and vasoconstrictor molecules [11, 12]. The
synthesis of isosorbide-based acetyl salicylic acid pro-
drugs with effective biological activity [13] has been pre-
viously reported.
Introduction
Atenolol antihypertensive therapy combined with acetyl
salicylic acid is effective in reducing cardiovascular
death, stroke, and myocardial infarction (MI) [1, 2]. Ate-
nolol is a selective b1-adrenoceptor antagonist that is
widely used to treat a wide range of cardiovascular disor-
ders, including mild to moderate hypertension, angina
pectoris and supraventricular arrhythmias, and arrhyth-
mia prophylaxis after acute MI [3, 4]. Clinical trials have
subsequently demonstrated that acetyl salicylic acid is
In this work, we report a very detailed evaluation of
the in-vitro and in-vivo activity, metabolic and pharmaco-
kinetic profile along with results from ulcerogenicity
and mutagenicity studies for atenolol aspirinate (ATA),
an already described derivative in which acetyl salicylic
acid (aspirinm) was joined to atenolol by an ester linkage
[14, 15]. This derivative was designed on the basis of the
prodrug strategy of coupling acetyl salicylic acid to fre-
quently co-prescribed cardiovascular agents. ATA could
act as a potential parallel mutual prodrug of both acetyl
salicylic acid and atenolol enhancing their bioavailabil-
Correspondence: Gilberto de Nucci, 415 Jesuino Marcondes Machado
Ave. 13092320 Campinas, SP, Brazil.
E-mail: denucci@dglnet.com.br
Fax: +55 19 3242-7439
Abbreviations: Atenolol aspirinate (ATA); atenolol salicylate (ATS); car-
bonyldimidazole (CDI); cyclooxygenase-1 (COX-1); 1-benzotriazoleoxy
(HOBt); benzotriazolyloxy tetramethyl uronium (HBTU); benzotriazolyl-
oxy-tris-dimethylaminophosphonium (BOP); dicyclohexylcarbodiimide
(DCC); N,N-dimethylformamide (DMF); isoproterenol (ISO); microwave-
assisted organic synthesis (MAOS); myocardial infarction (MI); salicylic
acid (SA); thromboxane A2 (TXA2); thromboxane B2 (TXB2)
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

446
A. C. Montes-Gil et al.
Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
ity and improving their therapeutic effects, as well as,
reducing (due to a potential lower dosage) the risk of gas-
trointestinal bleeding and toxicity associated with acetyl
salicylic acid. The large amount of ATA needed for these
detailed pharmacological studies, prompted us to opti-
mize and improve the synthetic route since the previ-
ously reported strategies (such as activation by DCC/
DMAP, acid chloride formation, etc.) only allowed low
yields (25–30%) [16]. Moreover, the ester linkage is
present in a very large number of pharmacologically
active compounds, therefore, the development of new
procedures for the easy, clean, and racemization-free syn-
thesis of ester is of topical interest in medicinal chemis-
try. On the basis of these considerations, ATA has been
prepared employing microwave-assisted organic synthe-
sis (MAOS) applied to two different esterfication routes.
Scheme 1. Synthetic route of 2-acetoxy-benzoic acid 2-(4-car-
bamoylmethyl-phenoxy)-1-(isopropylamino-methyl)-ethyl ester
hydrochloride (ATA, 3).
Results and discussion
Chemistry
dary amine group by forming a N-Boc derivative; b) activa-
tion of the carboxy group in acetyl salicylic acid and cou-
pling with N-Boc-protected atenolol; c) removal of the
protecting group from the atenolol secondary amine
group.
In general, the formation of esters from alcohols and car-
boxylic acids implies the activation of the carboxy group
by either the preventive conversion to a more reactive
acylating agent, such as chloride, mixed anhydride, acyl
azide, or active ester, or in-situ activation by using cou-
pling reagents, such as dicyclohexylcarbodiimide (DCC),
1-benzotriazoleoxy (HOBt), benzotriazolyloxy-trisdime-
thylaminophosphonium (BOP), carbonyldimidazole
(CDI), or benzotriazolyloxytetramethyluronium (HBTU).
However, these methods have some drawbacks such as
exothermic reaction, formation of byproducts, compli-
cated conditions, and low selectivity. Microwave-assisted
organic synthesis (MAOS) continues to affect synthetic
chemistry significantly by enabling rapid, reproducible,
and scalable chemistry development [16]. Numerous reac-
tions including condensations, cycloadditions, hetero-
cycle forming reactions, and metal-catalyzed cross-cou-
pling have been explored under microwave conditions,
some of which have been applied to medicinal chemistry
[16, 17]. MAOS can facilitate the discovery of new reac-
tions and reduce cycle time in optimization of reactions.
As it relates to our efforts to develop efficient and robust
methodologies and processes for high-throughput syn-
thesis of pharmacologically interesting libraries for drug
discovery, the exploration of microwave chemistry to
access new compounds has been of particular interest to
us [16]. Herein, we report, the pharmacological character-
ization and kinetic profile of ATA and two microwave-
assisted approaches for the synthesis of ATA from readily
available starting materials. In both methods, the process
comprises the steps of a) protection of the atenolol secon-
In the first method, we applied microwave irradiation
to the active pentafluorophenylester method (–OPfp,
Scheme 1, Method A) for the activation of the carboxylic
moiety, while in the second approach, microwave heat-
ing has been applied to a polymer-assisted solution-phase
technique using the polymer-supported reagent PS-carbo-
diimide as dehydration agent in presence of HOBt
(Scheme 1, Method B). In particular, in the first method,
the final compound ATA was obtained as follows: acetyl
salicylic acid 1 in anhydrous N,N-dimethylformamide
(DMF) in presence of DCC was treated with pentafluoro-
phenol to give the corresponding pentafluorophenyl
ester derivative 2. Intermediate 2 was dissolved in DMF,
N-Boc-atenolol was added, and the resulting solution was
heated by microwave irradiation for 30 min. After work
up, the intermediate N-Boc-atenolol-acetyl salicylic acid
aspirinate was isolated by silica gel column chromatogra-
phy (75% yield). Final compound ATA 3 was obtained by
removal of the Boc-group with 30% trifluoroacetic acid
(TFA) in dichloromethane (DCM) to give the correspond-
ing trifluoroacetate salt that was dissolved in methanol,
treated with 10% aqueous NaHCO3 and extracted with
dichloromethane. The solvent was evaporated and the
residue was crystallized with diethyl ether.
In the second method, condensation between N-Boc-
atenolol and acetyl salicylic acid 1 has been performed
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
Pharmacokinetic Profile of Atenolol Aspirinate
447
applying microwave heating to a polymer-assisted solu-
tion-phase technique using polymer-supported reagent
PS-carbodiimide as dehydration agent in the presence of
HOBt and using N,N-dimethylacetamide (DMA) as solvent.
The highly polar DMA ensured complete homogeneity
(excluding the polymer-supported carbodiimide resin)
and effective microwave heating of the reaction mixture.
In this case, 10 min of irradiation was found sufficient to
completely consume the starting acetyl salicylic acid 1
and the N-Boc-atenolol component from the reaction
mixture, affording a clean conversion to the desired ester
N-Boc-atenolol-aspirinate. Final compound
3
was
obtained according to the first method.
Pharmacological results
Antiplatelet effects
Aggregation induced by arachidonic acid was used
mainly to evaluate the effect of ATA on thromboxane-
dependent platelet activation, since this metabolic path-
way is inhibited by acetyl salicylic acid at very low con-
centrations [18]. ATA (30 lM), unlike acetyl salicylic acid,
had no inhibitory effect in all tested concentrations on
AA-induced platelet aggregation and thromboxane A2
(TXA2) synthesis in platelets (thromboxane B2 TXB2: basal
1.98 l 0.26 ng/mL, 845.81 l 35.77 and 1.87 l 0.75 ng/mL
for 200 lM of ATA and acetyl salicylic acid respectively,
p a 0.01). Ex-vivo studies showed that ATA significantly
inhibited serum-TXA2 production at the highest doses
administered, but was weaker than acetyl salicylic acid
in both mice (at 0.28 mmol/kg (vehicle, 71.04 l 17.8 ng/
mL; 34.2 l 11.2 and 4.3 l 1.7 ng/mL of TXB2 for ATA and
acetyl salicylic acid. respectively, p a 0.01)) and rats (at
0.17 mmol/kg (vehicle, 216.9 l 8.0 ng/mL; 70.3 l 7.3 and
8.7 l1.1 ng/mL of TXB2 for ATA and acetyl salicylic acid,
respectively, p a 0.05)).
x
Figure 1. Effect of ATA on N -nitro-L-arginine methyl ester (L-
NAME)-induced hypertension in rats. Mean Arterial Blood Pres-
sure (MABP; mm Hg) was measured by radiotelemetry twice a
week in Control (h), L-NAME (0), L-NAME + ATA (9), L-NAME +
atenolol (F) and L-NAME + acetyl salicylic acid (f) groups (n = 5
to 8). The arrow shows beginning of concomitant treatment with
the drugs, two weeks after L-NAME was initiated. Results are
expressed as mean l SEM. * p a 0.01 compared to control group
and # p a 0.01 compared to L-NAME group.
(9.93 l 0.16 and 163 l 21.75). As expected, atenolol addi-
tion caused a dose-dependent parallel shift to the right of
the ISO concentration-response curves reducing signifi-
cantly (p a 0.05) the pEC50 to ISO at concentrations of 1
and 10 lM (8.09 l 0.02 and 7.17 l 0.02, respectively) in
comparison to control animals (9.38 l 0.27).
Antihypertensive effect assessed by telemetry
In-vitro b1-adrenoceptor-blocking effects on chronotropic
responses in isolated rat right atria
Chronic administration of L-NAME (a non-selective NO-
synthase antagonist) caused a significant increase in
mean arterial blood pressure (Fig. 1). Atenolol treatment
initiated two weeks after starting L-NAME administra-
tion, caused a significant reduction in the blood pressure
(Fig. 1) and heart rate (302 l 6 and 266 l 4 bpm for L-
NAME and L-NAME + atenolol-treated animals, respec-
tively; values after four weeks of atenolol treatment).
Treatment with either acetyl salicylic acid or ATA did not
affect the hypertension or heart rate induced by chronic
administration of L-NAME (Fig. 1).
Addition of ATA (0.1–10 lM) to the bath had no effect on
the basal rate (300 l 11 and 315 l 5 bpm (beats per
minute));, before and after addition of ATA, respectively,
at the highest concentration of 10 lM). In contrast, addi-
tion of atenolol (0.1–10 lM) provoked a significant
reduction in the heart rate (270 l 13 and 210 l 10 bpm,
before and after addition of atenolol, respectively, at the
highest concentration of 10 lM). Concentration-response
curves to isoproterenol (ISO) in isolated rat right atria
were obtained in the absence or presence of atenolol and
ATA at concentrations of 0.1, 1 and 10 lM. The potency
(pEC50) and maximal responses of ISO were not signifi-
cantly unchanged by ATA 10 lM (9.61 l 0.17 and
135 l 5.0, respectively) in comparison to control animals
Hydrolysis experiments
ATA was rapidly hydrolyzed (tꢀ = 7.6 min, Kobs 9.16102
min– 1) in dilute plasma solution. Hydrolysis was sup-
pressed (tꢀ = 421.5 min, Kobs 0.166102 min– 1) when ATA
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

448
A. C. Montes-Gil et al.
Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
Table 1. Pseudo first-order rate constants and corresponding
half-lives for the hydrolysis of atenolol aspirinate in aqueous sol-
ution at 378C.
pH
2.5
7.4
9.4
kobs (h– 1610– 3
t1/2 (h)
)
12.27
56.47
27.85
24.88
108.0
6.41
was co-incubated with 3 lM of the esterase inhibitor neo-
stigmine, suggesting that ATA plasma hydrolysis was pre-
sumably due to the presence of plasma pseudocholines-
terase. Accurate mass measurement of the ATA hydroly-
sis product using a Q-TOF mass spectrometer showed
that it was formed by deacetylation of ATA and corre-
sponded to atenolol salicylate (ATS). The hydrolysis of
ATA in aqueous solution at 378C over the pH range 2.5–
9.4, followed pseudo-first order kinetics over several half-
life periods. Pseudo-first order plots for the decomposi-
tion of ATA were constructed from the logarithm of the
remaining ester versus time. The pseudo first-order con-
stants (kobs) obtained and the half-lives are shown in
Table 1. ATA exhibited high aqueous stability at all pH
values; at the low pH 2.5 the half-life was 56.5 h. ATA
aqueous stability is in agreement with the observation
that most esters of acetyl salicylic acid generally exhibit
higher aqueous stability than acetyl salicylic acid itself
[19] because esterfication masks the carboxylate, which
performs an autocatalytic role in acetyl salicylic acid
hydrolysis [20].
Figure 2. Kinetics of atenolol aspirinate (O) metabolization by
rat-liver microsomes, three unknown metabolites were formed
M1 (0), M2 (J), and M3 (g). Results are expressed as mean
l SEM for three experiments.
their nominal masses were m/z 403 and the accurate
masses of these compounds were determined to be m/z
403.1880 (M2) and m/z 403.1887 (M3). The mass shift for
each metabolite was 16 Da relative to atenolol salicylate,
suggesting the formation of monooxygenated metabo-
lites with elemental composition of C21H26N2O6 (theoreti-
cal monoisotopic mass 403.1875 Da; error –2.7 ppm for
M2 and –4.4 ppm for M3). A regenerating system of nico-
tinamide adenine dinucleotide phosphate (NADP+) was
present during the incubation; oxidation products were
therefore expected [21]. The MS/MS resulting spectra of
M2 showed an important fragment at m/z 283.16 which
would correspond to molecular mass of hydroxy ateno-
lol, oxidation would have taken place at the acetamide
group (atenolol subunit). Tandem MS spectra of M3
showed a signal only at m/z 267.17, demonstrating that
this metabolite was probably oxidized at the benzene
ring of the salicylic acid subunit, because the atenolol
subunit was kept intact as it is shown in Fig. 4. The oxida-
tion positions proposed for these metabolites were con-
sidered on the basis of mass fragments obtained after MS/
MS analysis, and the literature metabolic oxidation path-
ways available for salicylic acid [22] and atenolol [23].
In-vitro metabolism
The kinetics of ATA microsomal metabolization deter-
mined by peak area of the drug and its unknown metabo-
lites at different incubation times (Fig. 2) indicated that
ATA undergoes an immediate and complete biotransfor-
mation generating three metabolites: M1 was formed
within the first 5 min, then it was metabolized to the
other unknown minor metabolic species M2 and M3,
whose formation was time-dependent and parallel to the
kinetics of M1 consumption. Accurate mass measure-
ment of ATA metabolites using a Q-TOF mass spectrome-
ter showed for M1 an accurate mass of 387.1905 Da (ele-
mental composition of C21H26N2O5; error: 3.8 ppm with
respect to the theoretical monoisotopic mass of 387.1920
Da), this molecular formula matched with that of ATS
determined as the main hydrolysis product of ATA. The
resulting spectra obtained by MS/MS analysis of M1
(Fig. 3) showed the same intense signal seen in the
hydrolysis experiment that corresponded to the molecu-
lar mass of atenolol (m/z 267.1689). This confirmed that
ATS is the main metabolite formed after in-vitro hepatic
metabolism. Concerning M2 and M3 (as shown in Fig. 4)
Pharmacokinetics after intravenous administration
Under our experimental conditions, acetyl salicylic acid
was not detectable in any plasma sample taken from
treated dogs; thus, for salicylate quantification in acetyl
salicylic acid- or ATA-treated groups, the focus of the
determination was concentrated on salicylic acid (SA) lev-
els and kinetics, since acetyl salicylic acid and ATA are
rapidly hydrolyzed in the plasma as indicated by the
short half-lives of acetyl salicylic acid observed in
humans [24] and dogs [25] in concert with our ATA
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
Pharmacokinetic Profile of Atenolol Aspirinate
449
Figure 3. MS/MS product ion spectra and the predominant fragmentation pattern of M1. Atenolol salicylate was formed as a result of
atenolol aspirinate deacetylation.
Table 2. Pharmacokinetic parameters of atenolol salicylate, atenolol, and salicylic acid following intravenous administration of ateno-
lol aspirinate, acetyl salicylic acid, or atenolol.
Kinetic
Salicylic Acid
Atenolol
Acetyl salicylic
ATS
ATA group
Parameter
ATA group
Acetyl salicylic
ATA group
acid group
acid group
C0 (lg/L)
AUC0-24h (lgh/L)
AUC0-v (lgh/L)
132.6 l 60.1a)
110.9 l 82.9c)
159.4 l 85.6c)
0.8 l 0.3
5098.7 l 444.4
15182.6 l 7100
15366.8 l 5806
1.7 l 0.6
329.5 l 61.5b)
137.4 l 23.5b)
145.4 l 24.4b)
1.3 l 0.8b)
6272.0 l 4031
8724.0 l 1107
8847.0 l 1342
5.0 l 0.2
7583.1 l 2154.2
2849.0 l 405.2
3163.9 l 520.1
8.2 l 3.6
T1/2 (h)
Ke (L/h)
Vd (L/kg)
0.9 l 0.3
0.4 l 0.2
0.4 l 0.04
0.6 l 0.3
0.1 l 0.01
2.5 l 0.3
0.1 l 0.1
1.6 l 0.3
42.6 l 18.4a)
59.4 l 24.0b)
Atenolol aspirinate (ATA), Atenolol salicylate (ATS). Data are expressed as mean l SD (n = 3).
a)
p a 0.01.
b)
p a 0.05 when compared with the acetyl salicylic acid group.
p a 0.01 when compared with the atenolol group.
c)
plasma hydrolysis results. Similarly, concentrations of the ATA-treated dogs was 1.6% of that for the atenolol-
ATS instead ATA were found in the dog plasma samples. treated dogs. Plasma concentrations found for SA and
The mean plasma concentration time profiles and the atenolol in the ATA-treated group showed that both com-
pharmacokinetic parameters for SA, atenolol, and ATS pounds originated from cleavage of ATS molecule at the
after i.v. administration of ATA, acetyl salicylic acid, or benzoate ester linkage, generating concentration levels
atenolol are shown in Fig. 5 and Table 2. The SA AUC0 – 24h to a lesser extent than levels found after treatment with
for the ATA-treated dogs was 0.7% of that for the acetyl equimolar doses of the individual drugs.
salicylic acid-treated dogs and the atenolol AUC0 – 24h for
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

450
A. C. Montes-Gil et al.
Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
Figure 4. MS/MS product ion spectra and the predominant fragmentation patterns of the hydroxylated metabolites of atenolol salicy-
late, M2 (bottom panel) and M3 (top panel).
Gastric mucosal damage
ATA was significantly less ulcerogenic (0.66 mmol/kg)
than acetyl salicylic acid after a single equimolar admin-
istration to rats. Gastric damage characterized by
oedema associated with extensive gastric swelling,
necrosis, and hemorrhage was confirmed by light micro-
scopic examination of these tissues. Extensive gastric
damage was also noted visually and microscopically after
four weeks of oral treatment with 0.37 mmol/kg acetyl
salicylic acid; there was a remarkable and significant dif-
ference (p a 0.01) between the “lesion index” values for
acetyl salicylic acid and equivalent amounts of ATA
Figure 5. Mean plasma concentrations (l SEM) quantified for
(Fig. 6).
atenolol (0), salicylic acid (J), and atenolol salicylate (9) in dogs
(n = 3) after intravenous administration (bolus) of atenolol aspiri-
nate.
Conclusions
Our results showed that the coupling of atenolol and ace-
Mutagenicity assay (Ames Test)
tyl salicylic acid by means of an ester linkage did not pro-
duce good pharmacological results, neither in vitro nor in
vivo. The coupling of acetyl salicylic acid abolished the in-
vitro blocking activity of the atenolol moiety, as shown
by lacking the effect of ATA in the isolated atria prepara-
ATA failed to elicit a positive mutagenic response in the
Ames test after direct testing or in the presence of S9 mix
in Salmonella thyphimurium strains TA98, TA 100, TA102,
TA1535 and TA1537.
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
Pharmacokinetic Profile of Atenolol Aspirinate
451
the oral route did not exhibit any antihypertensive effect
in this particular model, indicating that no release of ate-
nolol in vivo was observed. The most likely explanation is
that ATA was metabolized in vivo to ATS, and this com-
plex made it difficult for plasma esterases to break the
linkage between the salicylate and the atenolol. To
inspect the different accessibility of the two ester
branches of ATA to the hydrolytic attack of esterases, a
conformational analysis was carried out by varying all
torsions in the random search option of the SYBYL pro-
gram (see Experimental, Section 4.12 for operational
details). Interestingly, among the lowest energy conform-
ers of ATA, a recurring feature was found: the presence of
an intramolecular H-bond between the protonated nitro-
gen atom of the isopropyl-amino-methyl group and the
ether oxygen of the phenyl-acetamide moiety so to form
a 6-membered ring (Fig. 7). Consequently, the linkage
between the salicylate and atenolol seems to be less acces-
sible to the attack of the plasma esterases, being pro-
tected by the bulky aminoalkyl chain. On the other hand,
the acetate group, lying in an exposed and unprotected
position, is more prone to the attack by the esterases. Fur-
thermore, when the hydrolysis of the acetate occurs, the
free phenolic group of ATS is in turn engaged in an addi-
tional intramolecular H-bond with the salicylate car-
bonyl oxygen (Fig. 8). This peculiar conformation
increases the thermodynamic stability of the molecule,
thus justifying the low metabolic reactivity of ATS. ATA
showed a safe gastric profile, as compared to acetyl sali-
cylic acid. Probably, the linkage also made it difficult for
acetyl salicylic acid to bind to the cyclooxygenase. This
was corroborated by the lack of inhibition on TXA2 syn-
thesis, either in human platelets in vitro, but also in the
ex-vivo evaluation made in mice and rats. Concerning the
synthetic procedure, we have developed two microwave-
enhanced, rapid and efficient methods for the synthesis
Figure 6. Gastric mucosal damage produced by oral administra-
tion of ATA or equivalent amounts of acetyl salicylic acid (0.37 or
0.66 mmol/kg, chronic or acute treatment, respectively). Mean
l SEM (n = 5) * p a 0.05 and ** p a 0.01 compared to vehicle and
ATA group.
tion. The L-NAME-induced hypertension is an established
chronic model for evaluating antihypertensive agents
[26] and atenolol has an important antihypertensive
effect [27]. However, chronic administration of ATA by
Figure 7. 3-D representation of the two lowest energy conformers of ATA (a and b). H-bonds are shown as dashed yellow lines. Non-
polar hydrogens are omitted for clarity.
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

452
A. C. Montes-Gil et al.
Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
Figure 8. Stereoview showing the lowest energy conformer of atenolol salicylate (ATS).
H-bonds are shown as dashed yellow lines. Non-polar hydrogens are omitted for clarity.
the filtrate was concentrated in vacuo to obtain 1.25 g of crude N-
Boc-atenolol. Further purification was performed by chromatog-
raphy on silica gel using diethyl ether as eluent to give the final
product N-Boc-atenolol as an almost colorless oil (yield: 1.07 g,
95%).
of ATA. One method uses readily available solid-phase
reagent that not only results in higher yield but also
greatly simplifies the purification process. Additionally,
the use of microwave technology allowed us to quickly
identify optimal reaction conditions with reaction times
reduced from hours to minutes.
2-Acetoxy-benzoic acid pentafluorophenyl ester 2
A solution of acetyl salicylic acid 1 (1 g, 5.5 mmol) in anhydrous
DMF (20 mL) was treated with pentafluorophenol (1.02 g,
5.5 mmol) in the presence of DCC (1.2 g, 6.0 mmol) at 408C for
20 min by microwave heating (300 W). After removal of the pre-
cipitated urea by filtration and DMF under reduced pressure,
the reaction mixture was dissolved in ethyl acetate, followed by
removal of more dicyclohexylurea. The solvent was removed
under reduced pressure and the residual oil was crystallized
from ethanol, to give the 2-acetoxy-benzoic acid pentafluoro-
We thank Fundaꢀ¼o de Amparo ꢁ Pesquisa do Estado de S¼o
Paulo (FAPESP) for the financial support.
Experimental
General procedures
1
phenyl ester derivative 2 (yield: 1.78 g, 95%; mp: 94–958C) H-
Some synthetic steps were performed using a microwave oven
ETHOS 1600 (Milestone, Sorisole, Italy) and a predefined micro-
wave program composed of appropriate temperature (fiber optic
temperature measurement), ramping, and holding steps. All
microwave reactions were carried out in standard Pyrex glass-
ware with a reflux condenser fitted through the roof of the
microwave cavity. All reactions were followed by thin-layer chro-
matography (TLC) carried out on precoated silica gel plates Kie-
selgel 60F254 (E. Merck, A.G., Darmstadt, Germany) with diethyl
ether/hexane (7 : 3, v/v) as solvent system. Visualization was
accomplished with UV light. Chemical reagents and solvents
were purchased from Aldrich Chemical Co and Carlo Erba
Reagents (Rodano, Italy) and were used without further purifica-
NMR (CDCl3): d H (selected peaks only) 2.31 (3H, s, OCOCH3), 7.20,
7.40, 7.68; and 8.21 (4H, d, t, t, d, acetyl salicylic acid, aromatic
proton resonances).
2-Acetoxy-benzoic acid 2-(4-carbamoylmethyl-phenoxy)-
1-(isopropylamino-methyl)-ethyl ester (ATA, 3)
Method A: The pentafluorophenyl ester derivative 2 (415 mg,
1.2 mmol) was dissolved in DMF (80 mL) and to this solution was
added N-Boc-atenolol (1.74 g, 4.8 mmol, 4 eq). After 30 min while
refluxing on microwave irradiation (300 W), TLC showed the for-
mation of the desired ester N-Boc-atenolol-aspirinate. Evapora-
tion of the solvent on the rotary evaporator left a gum, which
was dissolved in DCM and passed through a plug of silica, elut-
ing with ethyl acetate, to give N-Boc-atenolol-aspirinate as an oil
(yield: 475 mg, 75%).1H-NMR (CDCl3): d H selected peak only) 2.32
(3H, s, CH3COO). N-Boc-atenolol-aspirinate (324 mg, 0.6 mmol)
was added to a stirred solution of TFA (6 mL) in anhydrous DCM
(6 mL) at room temperature. After 1 h of stirring, the solvent was
removed by evaporation under reduced pressure, to leave the
alkylammonium trifluoroacetate salt derivative as a clear color-
less oil (yield: 300 mg, 92%). 1H-NMR (CDCl3): d H selected peaks
only) 1.33 (6H, m, CH(CH3)2), 2.29 (3H, s, CH3COO), 3.44 (2H, s,
PhCH2CONH2), 3.49 and 4.23 (5H, m, OCH2CH2NCH), 5.53, 5.60,
and 6.66 (3H, br, m, CH-OCO and CONH2), 6.85 and 7.08 (4H, 2d,
benzeneacetamide proton resonances), 7.09, 7.28, 7.57, and 7.99
(4H, d, t, t, d, aspirin aromatic proton resonances). 13C-NMR
(CDCl3): d C (selected peaks only) 19.0 (26C, CH(CH3)2, 21.4 (16C,
1
tion. H-NMR, FTIR, and MS spectra were obtained on a Bruker
AMX-500, Bruker IFS48 (Bruker Bioscience, Billerica, MA, USA),
and LCQ ThermoQuest instruments (ThermoQuest CO, San Jose,
CA, USA), respectively.
[3-(4-Carbamoylmethyl-phenoxy)-2-hydroxy-propyl]-
isopropyl-carbamic acid tert-butyl ester (N-Boc-atenolol)
A solution of di-tert-butyldicarbonate (0.75 g, 3.44 mmol) in t-
BuOH (3.5 mL) was added over 30 min to a solution of atenolol
(0.82 g, 3.1 mmol) in a mixture of t-BuOH (3.5 mL) and distilled
H2O (3.5 mL), and the mixture was stirred for 3 h at room tem-
perature. Then, the solution was poured into distilled water and
extracted with diethyl ether (4610 mL). The combined organic
layers were washed with saturated NaCl solution (3610 mL) and
then dried over anhydrous MgSO4. The mixture was filtered and
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
Pharmacokinetic Profile of Atenolol Aspirinate
453
CH3CO-O), 42.2 (1C, PhCH2CONH2), 55.0 and 70.1 (26C,
CHCH2NCH(CH3)2, 67.7 and 71.9 (2C, -NCH2-CHCH2OAr), 115 and
132 (26C, benzeneacetamide CH resonances), 124, 128, 134,
and 136 (46C, acetyl salicylic acid aromatic CH resonances),
122, 130, 152, and 157 (46C, quaternary aromatic carbons),
157, 164, and 172 (36C, COO, COO and CONH2). ATA N TFA (0.8 g)
was dissolved in 10 mL of methanol and treated with 10% aque-
ous NaHCO3 (150 mL). The aqueous solution was extracted with
DCM (3650 mL). The organic solution was dried on Na2SO4 and
evaporated yielding ATA that was crystallized from diethyl ether
(yield: 568 mg, 90%).
rated by centrifugation (1000 g610 min) and kept at –208C
until assayed for TXB2 by immunoenzymatic assay.
b1-Adrenoceptor antagonist effects; isolated right
atria and concentration response-curves
Right atria from male Wistar rats (200–300 g, n = 5) were iso-
lated, processed, and data were evaluated according to our
reported procedure [28]. Concentration–response curves for the
chronotropic actions of isoproterenol (ISO 10 pM-1 mM) were
obtained in the absence or in the presence of atenolol or ATA
(0.1, 1.0, and 10 lM).
2-Acetoxy-benzoic acid 2-(4-carbamoylmethyl-phenoxy)-
1-(isopropylamino-methyl)-ethyl ester (ATA, 3)
Antihypertensive effects
Male rats (Wistar-Uni specific pathogen free (SPF); 130–200 g)
provided by the Central Animal House (CEMIB-UNICAMP) were
maintained in individual cages under SPF, light and tempera-
ture-controlled conditions (12 h day / 12 h night, 258C). The anti-
hypertensive effects were assessed by telemetry, methodology
was developed according to our reported procedure [29]. The ani-
mals were anaesthetized, surgically manipulated, and equipped
with a radiotelemetry device into the descending aorta. Data
acquisition was performed during 60 s twice a week for six con-
secutively weeks (Data Science Inc., St. Paul, MN, USA). The ani-
mals were initially divided into control (n = 5, received tap water
alone) or L-NAME groups. Two weeks later, when the rats were
hypertensive, concomitant treatment was initiated with
0.37 mmol/kg of ATA (n = 8) or equimolar doses of atenolol (n =
7), or acetyl salicylic acid (n = 6); five rats received L-NAME alone.
L-NAME was dissolved in the drinking water and the daily dose
was 20 mg/rat/day. Atenolol, acetyl salicylic acid and ATA were
dissolved in 30% DMSO and administered by diary oral gavage
(vol. 0.8 mL) during four weeks, L-NAME and control animals
also received 30% DMSO.
Method B: PS-Carbodiimide resin (1 g, 1.2 mmol) was charged in
a two-necked flask. To the resin were added acetyl salicylic acid
(108 mg, 0.6 mmol) in DMA (5 mL), HOBt (81 mg, 0.6 mmol) in
DMA (5 mL) and N-Boc-atenolol (220 mg, 0.6 mmol) in DMA. The
mixture was heated by microwave irradiation at 1008C (300 W)
for 10 min. After cooling, the reaction mixture was diluted with
10 mL methanol and transferred to a pre-packed column of Si-
Carbonate (5 g, 0.7 mmol/g), which had been previously condi-
tioned with methanol and the eluent collected via gravity filtra-
tion. The column was washed with methanol (2615 mL) and the
solvent evaporated at reduced pressure to afford crude N-Boc-ate-
nolol-aspirinate which was purified on a silica gel column eluted
with ethyl acetate to give pure N-Boc-atenolol-aspirinate as an oil
(yield: 285 mg, 90%). Finally, ATA 3 was obtained as previously
described in the method A.
Platelet aggregation
Venous blood samples were collected from five human adult,
healthy volunteers, who had not taken any medication drug for
at least two weeks. Platelet-rich plasma (PRP) was prepared by
low-speed centrifugation of 10 mL citrated blood at 200
g610 min at room temperature. Increasing amounts of ATA or
acetyl salicylic acid (10 lM to 1 mM, dissolved in DMSO 10%)
were incubated with 400 lL aliquots of PRP at 370C for 5, 10, or
30 min before agonist arachidonic acid (AA) 1 mM addition.
Aggregation was monitored for 5 min in a double-channel
whole blood, Lumi-Aggregometer, (Chrono-log 560 CA; Chrono-
log Corp., Haverton, PA, USA). The antiaggregatory activity of
the drugs was evaluated as%-inhibition of platelet aggregation
compared to control samples. At the end of the incubation
period, PRP was collected in EDTA to stop aggregation; the tested
platelets were centrifuged at 2000 g64 min. The supernatant
was collected and stored at –208C. Concentrations of TXB2
(hydrolysis product of TXA2) were determined by immunoenzy-
matic assay using a commercial kit (Cayman Chemical Co., Ann
Arbor, MI, USA).
Hydrolysis kinetics in human plasma and aqueous
buffer solutions
The hydrolysis of ATA (initial concentration 0.2 mM) was studied
in 0.01 M phosphate buffer of pH 7.4 containing 10% human
plasma and aqueous buffer solutions of pH 2.5, 7.4, and 9.4 (ace-
tate, phosphate, and borate buffers, 0.01 M respectively) using
previously reported methodology [19]. Plasma hydrolysis was
also performed in the presence of neostigmine (3 lM) to confirm
the role of esterases in the hydrolysis of ATA. Analyses were per-
formed in triplicate, using RP–HPLC. Eluted peaks were col-
lected and introduced in a quadrupole time-of-flight mass spec-
trometer (QTOF-MS) source for accurate mass determination of
hydrolysis products under our previously reported chromato-
graphic and mass spectrometry conditions [30].
In-vitro hepatic metabolism
Thromboxane synthesis stimulated ex-vivo
Rat liver microsomes were prepared following a standard proce-
dure: tissues were homogenized in ice-cold 0.1 M potassium
phosphate, pH 7.5, containing 0.15 M KCl and 0.1 mM EDTA
(1 : 5 w/v). Crude homogenate was centrifuged at 12000 g for
20 min and the supernatant centrifuged at 105000 g for 60 min.
The microsomal pellets were resuspended in buffer phosphate
(0.2 M) containing 20% glycerol and frozen at –808C. Protein
contents (40.6 mg protein/mL) were estimated using the Lowry
[31] method and cytochrome P-450 content (1.3 nmol/mg pro-
tein) was assayed according to the method of Omura and Sato
Male Swiss mice weighing 25–30 g (n = 7) were treated orally
with acetyl salicylic acid 0.06, 0.17, and 0.28 mmol/kg or the
same doses of ATA dissolved in DMSO 10%. Additionally, male
Wistar rats weighing 150–200 g (n = 5), received 0.17 mmol/kg
of acetyl salicylic acid or the same doses of ATA. Three hours
after drug administration, the animals were anesthetized with
halothane and then blood samples were obtained by intracar-
diac puncture and transferred into glass tubes without anticoa-
gulant and allowed to clot at 378C for 60 min; serum was sepa-
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

454
A. C. Montes-Gil et al.
Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
[32]. ATA was incubated with microsomes under the following
conditions and final concentrations: ATA 50 lM, microsomal
protein 1 mg/mL, NADPH-generating system (glucose 6-phos-
phate 10 mM, MgCl2 10 mM, and glucose 6-phosphate dehydro-
genase 2 U, 0.5 mM) and Tris-HCl buffer (pH 7.4; 0.072 M) in a
total volume of 500 lL. After different incubation times at 378C,
reactions were terminated by adding ice-cold methanol and vor-
tex-mixing. Separated supernatant aliquots (50 lL) were submit-
ted to RP-HPLC analysis; chromatographic peaks obtained were
manually collected and introduced in a QTOF-MS source for
accurate mass measurement of ATA metabolites under our previ-
ously reported chromatographic and mass spectrometry condi-
tions [30].
Gastric mucosal damage
The ability to produce gastric damage was evaluated according
to reported procedures [33, 34]. Male Wistar rats (n = 5) weighing
180–200 g, were deprived of food but not of water 24 h before
the experiment. Three hours after oral administration of the
vehicle (DMSO 30%), acetyl salicylic acid 0.66 mmol/kg or an
equivalent amount of ATA, rats were sacrificed and their stom-
achs were removed, opened along the lesser curvature and
examined using magnifier lenses for the presence of macro-
scopically visible lesions. Each individual hemorrhagic lesion
was measured along its greatest length (a1mm = rating of 1; 1–
2 mm = rating of 2; A2 mm = rating according to their length in
mm). The overall total was designated as the “lesion index”. The
stomachs were fixed in buffered formalin (pH 7.4) and processed
by routine methods before mounting on glass slides and stain-
ing with hematoxylin and eosin. Coded slides (to avoid observer
bias) were evaluated using a light microscope. Additionally, tak-
ing advantage of chronic treatment followed by telemetry proto-
col, we evaluated gastric lesions after four weeks oral treatment
with 0.37 mmol/kg acetyl salicylic acid or equimolar doses of
ATA.
Pharmacokinetics after a single intravenous dose
ATA, atenolol, or acetyl salicylic acid at 11.2 lmol/kg of body
weight, were administered dissolved in 10% DMSO as a single
intravenous (i.v.) bolus into the medial saphenous vein from the
hind leg of male beagles weighing 10 to 15 kg (n = 3). Venous
blood samples (3 mL) were collected immediately before and at
0.02 to 24 h after drug administration in heparinized and
chilled tubes. Samples were then centrifuged at 2700 g for
10 min at 48C. Plasma was aspirated off, chilled with liquid N,
and stored at –208C until analysis. For plasma extraction of ate-
nolol, samples (100 lL) were spiked with the internal standard
nadolol; 0.025 lg/mL), and then 50 lL of ammonium acetate
(50%) were added to the sample and vortexed. Diethyl ether
(4 mL) – DCM (60 : 40 v/v) was added for liquid-liquid extraction,
centrifuged (3200 g for 1 min), frozen for 10 min, and after
phase separation, the collected organic phases were evaporated
to dryness under a nitrogen stream at 508C, taken up in 100 lL
of acetonitrile – water (10 : 90 v/v) – ammonium hydroxide
(2.5 mM), vortexed and transferred to an autosampler vial. For
acetyl salicylic acid and salicylic acid (SA) extraction, cinnamic
acid 0.025 lg/mL was used as an internal standard, formic acid
(3%) was added to acidify the sample before liquid-liquid extrac-
tion with ether – hexane (80 : 20 v/v) and used instead ammo-
nium hydroxide in the reconstitution solution.
Mutagenicity assay (Ames test)
Mutagenicity of ATA (31–500 lg/plate) dissolved in 30% DMSO
was assessed in the Ames–Salmonella assay with strains TA98,
TA100, TA102, TA1535, and TA1537 according to our previously
reported method [35].
Molecular modeling
Molecular modeling calculations and graphics manipulations
were performed on a Silicon Graphics Tezro workstation
equipped with four 700 MHz R16000 processors using the
SYBYL7.2 software package (Sybyl Molecular Modeling System
Tripos Inc., St. Louis, MO, USA). The core structures of ATA and
ATS were retrieved from Cambridge Structural Database (CSD)
[36] and modified according the standard bond lengths and
bond angles of the SYBYL fragment library with the TRIPOS force
field [37]. The ligands were modelled in the R configuration. Geo-
metric optimizations were carried out with the SYBYL/MAXI-
MIN2 minimizer by applying the BFGS (Broyden, Fletcher, Gold-
farb, and Shannon) algorithm [38] and setting a root-mean-
squares (rms) gradient of the forces acting on each atom of
0.05 kcal/molꢁ as the convergence criterion. A conformational
analysis was performed on ATA and ATS by varying all torsions
in the randomsearch option [39] of the SYBYL program. In this
Monte Carlo-like method, many conformations are generated by
randomly perturbing selected rotatable bonds and then the con-
formations are energy-minimized. Each new conformation is
compared against all others found so far, to see if it is unique.
Randomsearch was run under the following settings (reported
in parentheses): a) the maximum number of hits, which defines
the number of times each conformation must be found to stop
the search for new conformations (n = 6); b) the rms threshold,
which defines the maximum rms difference between two con-
formations before they are considered different (0.7 ꢁ); c) the
energy cut-off, which defines the maximum allowed energy for
a conformation to be kept and energy-minimized (E = energy of
the input conformer plus 50 kcal/mol); d) the maximum num-
ber of attempts to find a new conformation (1000 for both com-
pounds). Among the resulting conformations, only those associ-
ated with energy within 3 kcal/mol from the global minimum
LC-MS/MS analysis
The LC-MS/MS system comprised a liquid chromatograph Shi-
madzu model LC-10AVP (Shimadzu, Tokyo Japan); a degasser Agi-
lent model G1322-A (Agilent, Palo Alto, CA, USA) and an auto-
injector CTC analytics model HTS Pal (CTC Analytics, Zwingen,
Switzerland); coupled to an Applied Biosystems Sciex API-2000
triple-quadrupole mass spectrometer equipped with an ESI
source (Applied Biosystems, Foster City, CA, USA). The quadru-
poles were operated with unit resolution in the positive ion
mode (for ATA and atenolol) and negative ion mode (for acetyl
salicylic acid and SA). An Alltech, Prevail, C18 (5 lm, 15064.6
mm) column (Alltech, Deerfield, IL, USA)and mobile phase com-
posed of acetonitrile – water (70 : 30 v/v) – ammonium acetate
(10 mM) – acetic acid (16.7 mM) were used for atenolol analysis
and acetonitrile – water (70 : 30 v/v) – formic acid (10 mM) for
SA analysis. The run time was 5 min; quantification was per-
formed with Masslynx software version 3.5 (Micromass, UK).
Pharmacokinetic parameters were calculated using the software
WinNonLin Professional Network Edition 1.5 (Pharsight Corp.,
Mountain View, CA, USA)
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2007, 340, 445–455
Pharmacokinetic Profile of Atenolol Aspirinate
455
(11 for ATA and 12 for ATS) were further minimized with
MOPAC/AM1 [40] (default values; additional keywords MMOK
and PRECISE). Only low-energy conformations (energy less than
3 kcal/mol above the absolute minimum energy) were investi-
gated further.
[18] P. Minuz, C. Lechi, R. Tommasoli, S. Gaino, et al., Thromb.
Res. 1995, 80, 367–376.
[19] N. M. Nielsen, H. Bundgaard, J. Med. Chem. 1989, 32, 727–
734.
[20] T. St Pierre, W. P. Jencks, J. Am. Chem. Soc. 1968, 90, 3817–
3827.
Statistical analysis
[21] P. R. Tiller, A. P. Land, I. Jardine, D. M. Murphy, et al., J.
Chromatogr. A 1998, 794, 15–25.
Results are expressed as means l S.E.M or SD. Differences
between controls and treatments were analyzed by using the
Student's t test. A value p=0.05 was considered as statistically sig-
nificant.
[22] A. J. Hutt, J. Caldwell, R. L. Smith, Xenobiotica 1986, 16,
239–249.
[23] P. R. Reeves, D. J. Barnfield, S. Longshaw, D. A. McIntosh,
M. J. Winrow, Xenobiotica 1978, 8, 305–311.
[24] D. J. Siebert, F. Bochner, D. M. Imhoff, S. Watts, et al.,
Clin. Pharmacol. Ther. 1983, 33, 367–374.
References
[25] D. J. Morton, D. C. Knottenbelt, J. S. Afr. Vet. Assoc. 1989,
60, 191–194.
[1] E. Fossum, A. Moan, S. E. Kjeldsen, R. B. Devereux, et al., J.
Am. Coll. Cardiol. 2005, 46, 770–775.
[26] M. O. Ribeiro, E. Antunes, G. de Nucci, S. M. Lovisolo, R.
[2] V. Hinstridge, T. M. Speight, Drugs 1991, 42, 8–20.
Zatz, Hypertension 1992, 20, 298–303.
[3] A. N. Wadworth, D. Murdoch, R. N. Brogden, Drugs 1991,
42, 468–510.
[27] S. R. Pacca, A. P. de Azevedo, C. F. de Oliveira, I. M. de
Luca, et al., J. Cardiovasc. Pharmacol. 2002, 39, 201–207.
[4] K. E. Ellison, G. Gandhi, Drugs 2005, 65, 787–797.
[28] S. R. Riado, A. Zanesco, L. A. Barker, I. M. De Luca, et al.,
Hypertension 1999, 34, 802–807.
[5] Steering Committee of the Physicians' Health Study
[29] M. Zanfolin, R. Faro, E. G. Araujo, A. M. Guaraldo, et al., J.
Cardiovasc. Pharmacol. 2006, 47, 391–395.
Research Group. N. Engl. J. Med. 1989, 321, 129–135.
[6] Antiplatelet Trialists' Collaboration, BMJ 1994, 308, 81–
[30] B. Bicalho, G. C. Guzzo, S. Lilla, H. O. Dos Santos, et al.,
Curr. Drug Metab. 2005, 6, 519–529.
106.
[7] Antithrombotic Trialists' Collaboration, BMJ 2002, 324,
[31] O. H. Lowry, N. J. Rosebrough, A. L. Farr, R. I. Randall, J.
Biol. Chem. 1951, 193, 265–275.
71–86.
[8] S. Yusuf, F. Zhao, S. R. Mehta, S. Chrolavicius, et al., N.
Engl. J. Med. 2001, 345, 494–502.
[32] T. Omura, R. Sato, J. Biol. Chem. 1964, 239, 2370–2378.
[33] C. Cena, M. L. Lolli, L. Lazzarato, E. Guaita, et al., J. Med.
Chem. 2003, 46, 747–754.
[9] ISIS-2 (Second International Study of Infarct Survival)
Collaborative Group. Lancet 1988, 13, 349–360.
[34] J. L. Wallace, M. N. Muscara, W. McKnight, M. Dicay, et
al., Thromb. Res. 1999, 93, 43–50.
[10] M. Hamberg, J. Svensson, B. Samuelsson, Proc. Natl. Acad.
Sci. USA 1975, 72, 2994–2998.
[35] H. S. Cardoso, B. Bicalho, P. Genari, V. Santagada, et al.,
Eur. J. Med. Chem. 2006, 41, 408–416.
[11] G. de Gaetano, Haematologica 2001, 86, 349–356.
[12] P. J. Loll, D. Picot, R. M. Garavito, Nature Struct. Biol. 1995,
2, 637–643.
[36] F. H. Allen, S. Bellard, M. D. Brice, B. A. Cartwright, et al.,
Acta Crystallogr. 1979, B35, 2331–2339.
[13] J. F. Gilmer, L. M. Moriarty, M. N. Lally, J. M. Clancy, Eur.
J. Pharm. Sci. 2002, 16, 297–304.
[37] J. G. Vinter, A. Davis, M. R. Saunders, J. Comput-Aided Mol.
Des. 1987, 1, 31–51.
[14] W. Byrne, A. Rynne, PCT Int. Appl. WO9520568, 1995.
[15] W. Byrne, J. F. Gilmer, PCT Int. Appl. WO9705128, 1997.
[38] J. Head, M. C. Zerner, Chem. Phys. Lett. 1985, 122, 264–
270.
[16] V. Santagada, E. Perissutti, G. Caliendo, Curr. Med. Chem.
2002, 9, 1251–1238; V. Santagada, F. Frecentese, E. Peris-
sutti, L. Faretto, G. Caliendo, QSAR Comb. Sci. 2004, 23,
919–944.
[39] A. M. Treasurywala, E. P. Jaeger, M. L. Peterson, J. Comput.
Chem. 1996, 17, 1171–1182.
[40] MOPAC (version 6.0) is available from Quantum Chemis-
try Program Exchange, No. 455; M. J. S. Dewar, E. G. Zoe-
bisch, E. F. Healy, J. J. P. Stewart, J. Am. Chem. Soc. 1985,
107, 3902–3909.
[17] O. C. Kappe, D. Dallinger, Nat. Rev. Drug Discov. 2006, 5,
51–63.
i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com