A carbamate-based approach to primaquine prodrugs: Antimalarial activity, chemical stability and enzymatic activation

Article abstract of DOI:10.1016/j.bmc.2011.11.059

O-Alkyl and O-aryl carbamate derivatives of the antimalarial drug primaquine were synthesised as potential prodrugs that prevent oxidative deamination to the inactive metabolite carboxyprimaquine. Both O-alkyl and O-aryl carbamates undergo hydrolysis in alkaline and pH 7.4 phosphate buffers to the parent drug, with O-aryl carbamates being ca. 10⁶-10 ¹⁰ more reactive than their O-alkyl counterparts. In human plasma O-alkyl carbamates were stable, whereas in contrast their O-aryl counterparts rapidly released the corresponding phenol product, with primaquine being released only slowly over longer incubation periods. Activation of the O-aryl carbamates in human plasma appears to be catalysed by butyrylcholinesterase (BuChE), which leads to carbamoylation of the catalytic serine of the enzyme followed by subsequent slow enzyme reactivation and release of parent drug. Most of the O-aryl and O-alkyl carbamates are activated in rat liver homogenates with half-lives ranging from 9 to 15 h, while the 4-nitrophenyl carbamate was hydrolysed too rapidly to determine an accurate rate constant. Antimalarial activity was studied using a model consisting of Plasmodium berghei, Balb C mice and Anopheles stephensi mosquitoes. When compared to controls, ethyl and n-hexyl carbamates were able to significantly reduce the percentage of infected mosquitos as well as the mean number of oocysts per infected mosquito, thus indicating that O-alkyl carbamates of primaquine have the potential to be developed as transmission-blocking antimalarial agents.

Full text of DOI:10.1016/j.bmc.2011.11.059

Bioorganic & Medicinal Chemistry 20 (2012) 886–892
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
A carbamate-based approach to primaquine prodrugs: Antimalarial activity,
chemical stability and enzymatic activation
Graça Mata a, Virgílio E. do Rosário b, Jim Iley c, Luís Constantino a, Rui Moreira a,
⇑
a Research Institute for Medicines and Pharmaceutical Sciences (iMed.UL), Faculty of Pharmacy,University of Lisbon, Avenida das Forças Armadas, 1649-003 Lisboa, Portugal
b Centro de Malária e Outras Doenças Tropicais, Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa, Rua da Junqueira, 100, 1349-008 Lisboa, Portugal
c Department of Chemistry and Analytical Sciences, The Open University, Milton Keynes, MK7 6AA, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
O-Alkyl and O-aryl carbamate derivatives of the antimalarial drug primaquine were synthesised as poten-
tial prodrugs that prevent oxidative deamination to the inactive metabolite carboxyprimaquine. Both
O-alkyl and O-aryl carbamates undergo hydrolysis in alkaline and pH 7.4 phosphate buffers to the parent
drug, with O-aryl carbamates being ca. 106–1010 more reactive than their O-alkyl counterparts. In human
plasma O-alkyl carbamates were stable, whereas in contrast their O-aryl counterparts rapidly released
the corresponding phenol product, with primaquine being released only slowly over longer incubation
periods. Activation of the O-aryl carbamates in human plasma appears to be catalysed by butyrylcholin-
esterase (BuChE), which leads to carbamoylation of the catalytic serine of the enzyme followed by sub-
sequent slow enzyme reactivation and release of parent drug. Most of the O-aryl and O-alkyl carbamates
are activated in rat liver homogenates with half-lives ranging from 9 to 15 h, while the 4-nitrophenyl car-
bamate was hydrolysed too rapidly to determine an accurate rate constant. Antimalarial activity was
studied using a model consisting of Plasmodium berghei, Balb C mice and Anopheles stephensi mosquitoes.
When compared to controls, ethyl and n-hexyl carbamates were able to significantly reduce the percent-
age of infected mosquitos as well as the mean number of oocysts per infected mosquito, thus indicating
that O-alkyl carbamates of primaquine have the potential to be developed as transmission-blocking
antimalarial agents.
Received 6 October 2011
Revised 24 November 2011
Accepted 25 November 2011
Available online 3 December 2011
Keywords:
Malaria
Primaquine
Prodrugs
Carbamates
Chemical stability
Metabolic activation
Ó 2011 Elsevier Ltd. All rights reserved.
MeO
MeO
1. Introduction
N
N
Malaria remains the world’s top-priority tropical disease due to
its high death burden, as well as to its economic and social impacts
on the development of malaria-endemic countries.1 The emer-
gence and spread of multidrug-resistant Plasmodium falciparum,
the causative agent of the most lethal form of human malaria, is
still the major obstacle in achieving an effective control of the dis-
ease.2 Most antimalarials in clinical use or under development are
potent blood-schizontocides, that is they act rapidly against the
parasitic forms that invade erythrocytes and cause the usual symp-
toms to effect a cure from malaria within a reasonable time (ideally
3 days or less).3–5 However, the ultimate goal of global eradication
of malaria parasites from the human population requires the radi-
cal cure of all life cycle stages of all malaria species infecting
humans.6
HN
HN
CO2H
NH2
1, Primaquine
2, Carboxyprimaquine
MeO
MeO
N
O
N
O
H
N
HN
HN
N
NHMe
N
H
OR
H
O
3
4, R = aryl, alkyl
Currently, primaquine, 1, is the only available antimalarial that
displays a marked activity against gametocytes from all species of
parasite causing human malaria, including chloroquine-resistant
P. falciparum; it is thus capable of interrupting disease transmission
⇑
Corresponding author. Tel.: +351 217946477; fax: +351 217946470.
E-mail address: rmoreira@ff.ul.pt (R. Moreira).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.11.059

G. Mata et al. / Bioorg. Med. Chem. 20 (2012) 886–892
887
from the host to the mosquito vector.7 Primaquine is also the only
antimalarial effective against the latent exoerythrocytic forms
(hypnozoites) of Plasmodium vivax responsible for relapsing malar-
ia.7,8 In addition, primaquine also displays blood schizontocidal
activity, particularly against P. vivax and P. falciparum, but at doses
that can induce side effects such as methaemoglobinemia.
However, primaquine is rapidly metabolised in mammals to
carboxyprimaquine, 2, which is devoid of significant antimalarial
activity against the several forms of the parasite.7 After intrave-
nous administration of primaquine to rats, monkeys and humans
it was found that the plasma concentration of carboxyprimaquine
rapidly exceeded that of the parent drug after 15–30 min.9–11 The
oxidative deamination of the alkyl side chain of primaquine most
likely involves three enzymes in a two-step process: first, mono-
amine oxidase (MAO) or cytochrome P450 systems mediate the
oxidation of primaquine to the corresponding aldehyde; second,
the aldehyde intermediate is further oxidised to carboxyprimaqu-
ine by aldehyde dehydrogenase.7
The synthesis of prodrugs is a widely accepted approach used to
overcome metabolic deactivation and bioavailability problems
commonly found in amine drugs.12,13 In the case of primaquine,
any prodrug should be activated at a rate adequate to maintain
sustained levels of the parent drug while being unable to undergo
oxidative deamination prior to the release of primaquine. Since
both mechanisms of cytochrome P450- and MAO-catalyzed oxida-
tion of amines appear to involve an initial single electron transfer
to form a nitrogen-centred radical cation,14 increasing the ioniza-
tion potential through nitrogen acylation is expected to reduce sig-
nificantly the rate of oxidation. For example, acylation of the
terminal amino group of primaquine with dipeptides (e.g. 3)
blocked the formation of carboxyprimaquine in rat liver homoge-
nates, without affecting the antimalarial activity.15
In the present work we assess carbamates 4 as potential pro-
drugs for primaquine. The carbamate prodrug approach requires
the derivative to be enzymatically activated to a carbamic acid,
which rapidly decomposes to the parent amine.16 Carbamates
can be activated by esterases or cytochrome P450, and their chem-
ical and enzymatic reactivity can be modulated by appropriately
choosing the alcohol carrier.16,17 We have now synthesized a series
of O-alkyl and O-aryl carbamate derivatives 4 of primaquine com-
prising different alcohol and phenol pro-moieties, in order to
determine the impact of these leaving groups on (i) the chemical
stability, (ii) rates of activation in human plasma and by rat liver
homogenates and (iii) the potency of these derivatives. The target
compounds were evaluated in vivo for their gametocytocidal
activity.
was prepared according to the procedure reported by McChesney
et al.18
2.2. Hydrolysis in aqueous buffers
The rates of hydrolysis of primaquine carbamates, 4, were
investigated in aqueous buffers containing 20% (v/v) of acetonitrile.
The observed pseudo-first-order rate constants, kobs, were propor-
tional to the hydroxide ion concentration (Eq. 1), up to pH 12.
ꢀ
ꢀ
kobs ¼ kOH ½OH ꢁ
ð1Þ
In general, compounds 4 undergo alkaline hydrolysis to primaq-
uine and the corresponding alcohol or phenol with a wide range of
ꢀ
rates, as shown by the kOH values listed in Table 1. Primaquine and
the product phenols were readily observed by HPLC, their identi-
ties being confirmed by spiking with authentic samples. However,
compound 4h (R = CH2CO2Et) decomposes very rapidly to 4i
(R = CH2CO2H) via ester rather than carbamate hydrolysis.
It is possible to consider two different mechanisms that are con-
sistent with both the products of hydrolysis and the observed
kinetics. Carbamates 4 can undergo direct hydroxide attack at
the carbonyl carbon atom (path A, Fig. 1). Alternatively, com-
pounds 4 can undergo elimination by an E1cB mechanism, that is
involving the formation of an isocyanate 6, resulting from depar-
ture of the alcohol from the carbamate anion 5 (path B, Fig. 1).
The latter then reacts rapidly with hydroxide ion to yield the con-
jugate base of carbamic acid, 7. Eqs. (2) and (3) can be derived from
Fig. 2 for the addition–elimination (path A) and elimination–addi-
tion (path B) pathways, respectively:
k1Kw
kobs
kobs
¼
¼
ð2Þ
ð3Þ
Ka þ ½Hþꢁ
k1Ka
Ka þ ½Hþꢁ
where Kw is the auto-protolysis constant for water. For both equa-
tions, if [H+] ꢂ Ka then kobs varies linearly with 1/[H+], which is ex-
actly what is observed, as expressed by (Eq. 1).
Since both Eqs. (2) and (3) are kinetically equivalent they can-
not be used to distinguish between the two possible mechanisms.
However, the following strongly suggest that aryl carbamates, and
carbamates derived from alcohols with pKa <13, hydrolyse via the
elimination–addition pathway (path B, Fig. 1), while alkyl carba-
mates derived from alcohols with pKa P 13 hydrolyse via addi-
ꢀ
tion–elimination pathway (path A, Fig. 1). First, from the kOH
values for O-aryl carbamates 4d–g presented in Table 1, it is clear
that electron-withdrawing substituents in the phenol moiety en-
hance the reactivity. A plot of log kOH versus rꢀ values gives a
ꢀ
2. Results and discussion
2.1. Synthesis
Hammett
q value of 3.0 (Eq. 4), which is identical to that reported
for the alkaline hydrolysis of aryl N-phenylcarbamates.19 These
compounds are known to hydrolyse via a E1cB mechanism, with
formation of isocyanate intermediates.
Primaquine carbamates were synthesised either by reaction of
primaquine with the appropriate alkyl or aryl chloroformates or,
alternatively, with an equimolar amount of carbonyl diimidazole
and the appropriate alcohol (Scheme 1). Carboxyprimaquine, 2,
log kOH ¼ ð2:97 ꢃ 0:27Þrꢀ þ ð0:49 ꢃ 0:16Þ
ðn ¼ 4; r2 ¼ 0:98; SD ¼ 0:31Þ
ð4Þ
ꢀ
Second, O-aryl carbamates 4d–g are ca. 106–1010 more reactive
than their O-alkyl counterparts 4a,b, reflecting the better nucleofu-
gacity of phenols. The trifluoroethyl carbamate, 4c, also hydrolyses
ca. 104 times more rapidly than the other alkyl carbamates. A sim-
ilar result was observed for aryl N-phenylcarbamates when com-
pared to their alkyl counterparts, this difference of reactivity
being ascribed to a change of mechanism from E1cB to direct at-
tack of hydroxide ion at the carbamate carbonyl for those com-
pounds containing poorer leaving groups.18 Third, the Brønsted
MeO
MeO
(i) or (ii)
N
O
N
HN
HN
N
H
OR
NH2
4
1
ꢀ
plot of log kOH versus the pKa of the leaving alcohol (Fig. 2) shows
that only carbamates with good leaving groups (phenols and
Scheme 1. Reaction conditions: (i) ClCO2R, TEA; (ii) CDI, ROH.

888
G. Mata et al. / Bioorg. Med. Chem. 20 (2012) 886–892
Table 1
ꢀ
Second-order-rate constant for alkaline hydrolysis, kOH , half-lives, t1/2, for non-enzymatic hydrolysis at pH 7.4, first-order-rate constants and half-lives for activation in human
plasma and rat liver homogenates at 37 °C, kplasma and kliver, respectively, and lipophilicity data for primaquine carbamates 4
h
kOH (Mꢀ1 sꢀ1
)
ꢀ
Compound
R
pKa (ROH)
pH 7.4 buffer
Human plasma
kplasma (sꢀ1
t1/2 (min)
Rat liver homogenate
clogP
logD7.4
t1/2 (d)
)
kliver (hꢀ1
)
t1/2 (h)
4a
4b
4c
4d
4e
4f
Et
15.9a
16.2b
12.37a
10.20c
9.95c
9.38c
7.14c
12.9b
9.34 ꢄ 10ꢀ7
3.49 ꢄ 10ꢀ7
3.95 ꢄ 10ꢀ3
9.10 ꢄ 10ꢀ1
2.64
NRe
NR
NR
> 30
12
4
NR
NR
NR
NR
NR
NR
467
217
38.3
0.16
NDf
4.56 ꢄ 10ꢀ2
6.63 ꢄ 10ꢀ2
5.04 ꢄ 10ꢀ2
6.70 ꢄ 10ꢀ2
7.26 ꢄ 10ꢀ2
5.85 ꢄ 10ꢀ2
NDg
15.1
10.4
13.7
10.3
9.5
11.8
NDg
NDf
3.09
4.63
3.56
4.29
4.26
4.78
4.28
3.03
3.02
4.64
3.41
4.23
4.27
4.44
4.23
3.09
(CH2)5Me
CH2CF3
C6H4–4-MeO
C6H5
C6H4–4-Cl
C6H4–4-NO2
CH2CO2Et
2.45 ꢄ 10ꢀ5
5.34 ꢄ 10ꢀ5
3.06 ꢄ 10ꢀ4
7.12 ꢄ 10ꢀ2
NDf
7.17
4g
4h
2.40 ꢄ 104
2.1 min
2.5d
10d
NDf
a
b
c
Ref. 39.
Ref. 40.
Ref. 41.
Ester hydrolysis.
NR—no reaction detected.
ND—not determined.
d
e
f
g
h
Too fast to be determined accurately.
Ref. 39.
MeO
MeO
k2
N
O
N
O
HN
HN
C
N
Path B
N-
OR
Me
Me
5
6
Ka
fast OH-
MeO
MeO
k1[OH-]
N
O
N
O
HN
HN
O-
Path A
N
H
OR
N
H
Me
Me
4
7
- CO2
Figure 2. Brønsted plot of logkOHꢀ versus the pKa of the leaving alcohol, for the
alkaline hydrolysis of carbamates 4 (d, aryl carbamates; j, alkyl carbamates).
1
Figure 1. Mechanisms of hydrolysis for carbamates 4; path
addition–elimination pathway and path B referrers to the elimination–addition
pathway (E1cB mechanism).
A refers to the
4d–g in 50% human plasma followed strict first-order kinetics for
at least four half-lives, (r2 >0.95 in all cases), reflecting their en-
hanced chemical reactivity. From the rate data presented in Table
1, it is clear that human plasma enzymes markedly accelerate
the rate of hydrolysis of carbamates 4d–g. To assess whether the
rate-enhancement observed with human plasma, compared to
hydrolysis in pH 7.4 buffer, could be ascribed to the action of plas-
ma esterases, the effect of diisopropyl fluorophosphate (DIFP) on
the rate of hydrolysis of 4f was also studied. DIFP is a potent inhib-
itor of the serine hydrolase cholinesterase (E.C. 3.1.1.8, also called
pseudocholinesterase, plasma cholinesterase or butyrylcholinest-
erase, BuChE).20 Most significantly, a complete inhibition of the
hydrolysis of carbamate 4f was observed when the human plasma
was pre-incubated with 20 mM DIFP, indicating that plasma acti-
vation of aryl carbamates is BuChE-catalysed.
Carbamates 4d–g were hydrolysed in human plasma quantita-
tively to the corresponding phenol product, but none of the sub-
strates released primaquine over the time scale of the reactions
(ca. five half-lives). However, primaquine was detected when 4f
and 4g were incubated for longer periods (ca. 10 half-lives), corre-
sponding to 15% and 16% of prodrug conversion to the parent drug,
respectively (Fig. 3). This suggests that O-aryl carbamates 4d–g
trifluoroethylethanol) show high sensitivity to leaving group vari-
ation, consistent with a with considerable acyl-oxygen bond cleav-
age as expected for the E1cB mechanism.17,19
The reactivity of carbamates 4 was also studied in pH 7.4 phos-
phate buffer, at 37 °C. As revealed by the half-live values, t1/2, pre-
sented in Table 1, O-alkyl carbamates 4a–c were stable at pH 7.4
(no measurable decomposition for two weeks), while the O-aryl
carbamates 4d–f hydrolysed very slowly to primaquine and the
corresponding phenol product. In contrast, the 4-nitrophenol car-
bamate 4g was rapidly hydrolysed, with a half-life of only 2 min.
Overall, rates of hydrolysis at pH 7.4 largely reflect the nucleofu-
gacity of the phenol pro-moiety.
2.3. Activation in human plasma
The alkyl carbamates 4a–c proved to be stable in human plas-
ma, their concentration in the incubation mixtures remaining con-
stant for at least 3 h. In contrast, hydrolysis of O-aryl carbamates

G. Mata et al. / Bioorg. Med. Chem. 20 (2012) 886–892
889
primary and secondary amines via oxidative dealkylation, as
reported for ritonavir, amprenavir and loratidine.28–30 Importantly,
neither carboxyprimaquine or 8-amino-6-methoxyquinoline,
a
minor metabolite arising from the oxidative deamination at
C-10,31,32 were detected in any of the incubation mixtures over the
timescale of the hydrolysis reactions. The reason for the incomplete
release of primaquine from the corresponding carbamates in rat li-
ver homogenates is not yet fully understood, but one might specu-
late that for O-aryl carbamates 4d–g might form a stable carbamoyl
enzyme resulting from the reaction with liver carboxylesterase
(CE). It has been reported that rat liver contains CE concentrations
high enough to contribute significantly to the hydrolysis of carma-
bates,33,34 and thus this enzyme is likely to be involved in the
hydrolysis of derivatives 4d–g in rat liver homogenates.
2.5. In vivo studies: gametocytocidal activity
Figure 3. Reaction profile for the hydrolysis of carbamate 4f in human plasma (d,
4f; j, 4-chlorophenol; N, primaquine).
The potential of compounds 4 to prevent the transmission of
malaria was evaluated through their gametocytocidal activity
against Plasmodium berghei using a model consisting of Balb C
mice and Anopheles stephensi mosquitoes.35,36 Carbamates 4f–g
were excluded from this assay due to their chemical reactivity.
For each compound, the effect of the single dose of 15 mg/kg on
the appearance of oocysts in the midguts of mosquitoes was stud-
ied. The criteria to assess the activity were: (i) the percentage of
mosquitoes with oocysts and (ii) the mean number of oocysts
per infected mosquito. Sporogonic studies using laboratory-raised
vectors generates data with significant inherent variation which
might have several origins, such as a genetically heterogeneous
mosquito population or individual variation in vector feeding
duration (therefore affecting the number of ingested gameto-
cytes).37 In order to assure that this inherent variation would
not affect the interpretation of the experimental data, the largest
possible number of mosquitoes was used for each compound and
dose. As the result of the high number of mosquitoes necessary to
perform all experiments it was decided to use separate controls
for each compound (Table 2).
Inspection of the activity data presented in Table 2 shows that
O-alkyl carbamates 4a,b reduced the percentage of infected mos-
quitoes and the production of oocysts when compared to controls.
In contrast, carbamates 4c–e did not significantly affect the
percentage of mosquitoes with oocysts and the mean number of
oocysts per infected mosquito when compared to controls.
Compounds 4a,b were less active than primaquine, which almost
completely inhibited the production of oocysts at a dose of
15 mg/kg. These results suggest that the gametocytocidal activity
displayed by compounds 4a–e is not dependent on the release of
the parent drug and that O-alkyl carbamates 4a,b, the most potent
compounds in this series, are active per se.
react with the catalytic serine residue of BuChE to form a relatively
stable carbamoylated enzyme, which is subsequently slowly
hydrolysed to primaquine. This result is in line with reports show-
ing that aryl carbamates can rapidly carbamoylate esterases, while
the resulting modified enzyme hydrolyses at slower rates to regen-
erate the active enzyme and liberate the amine moiety.21–23
The general order of reactivity for the plasma-catalysed hydro-
lysis of 4d–g is identical to that observed in the alkaline hydrolysis
ꢀ
and a good correlation between log kOH and logkplasma is obtained,
as shown in (Eq. 5):
ꢀ
log kOH ¼ ð0:77 ꢃ 0:07Þ log kplasma ꢀ ð4:46 ꢃ 0:16Þ
ð5Þ
ðn ¼ 4; r2 ¼ 0:98; SD ¼ 0:24Þ
The logkplasma values also correlate with Hammett rꢀ values for
the substituents in the phenol moiety to give a value of 2.3
q
(r2 = 0.99). Overall, these results suggest that both alkaline and
plasma hydrolyses have similar structural requirements, that is
reactivity in human plasma appears to be dependent on the alcohol
leaving group ability rather than on lipophilicity of the prodrug,
and that plasma hydrolysis has a substantial contribution from
the E1cB process. A similar correlation between alkaline and plas-
ma hydrolyses has also been reported for the plasma activation of
acyloxymethyl-type prodrugs of sulfonamides.24 These results sug-
ꢀ
gest that the kOH values can be used as a crude indicator for the
metabolic susceptibility of a prodrug series towards serine-depen-
dent enzymes such as BuChE and carboxylesterases.25,26
2.4. Activation in rat liver homogenates
Based on previous studies reporting the nearly equal roles of
cytochrome P450, MAO and aldehyde dehydrogenase in the con-
version of primaquine into carboxyprimaquine,27 rat liver homog-
enates were specifically chosen to study the metabolism of
derivatives 4. Both aryl and alkyl carbamates were slowly hydroly-
sed in rat liver homogenates, with half-lives ranging from 9 to 15 h,
while 4-nitrophenyl carbamate, 4 g hydrolysed too rapidly to be
monitored by HPLC. No clear correlation emerged between the rate
data in Table 1 and the pKa of the alcohol/phenol leaving group,
logP or logD7.4 values for compounds 4a–g.
HPLC analysis of the incubation mixtures revealed that only O-
aryl carbamates released primaquine, corresponding to 12–15% of
prodrug conversion to the parent drug. Surprisingly, primaquine
was not detected when O-alkyl carbamates 4a–c were activated
in liver homogenates, contrasting with the observation that second-
ary and tertiary alkyl carbamates readily liberate the corresponding
Table 2
Effect of carbamates 4a–e and primaquine on the sporogonic development of P.
berghei in A. stephensi mosquitoes
Compound
% of infected
mosquitoes
Mean No. oocysts
per mosquito ( SE)a
Controls
4a
4b
4c
Controls
4d
4e
91
19
24
72
80
74
65
3.9
58.8 (8.9)
14.8 (6.8)
24.6 (5.4)
41.0 (5.5)b
46.1 (9.1)
39.7 (5.5)b
33.4 (8.9)
1.6 (1.3)
Primaquine
a
Counting of oocysts was carried out at day 10 post-feed.
P >0.05 versus control, by Student’s t-test.
b

890
G. Mata et al. / Bioorg. Med. Chem. 20 (2012) 886–892
FAB-MS, m/z (%) 332 (75, MH+), 201 (100), 175 (37). Anal. Calcd
for C18H25N3O3: C, 65.23; H, 7.60; N, 12.68. Found: C, 65.17; H,
7.91; N, 12.51.
3. Conclusions and relevance to prodrug design
The use of carbamates as prodrugs of basic primary amine drugs
is still open to discussion mainly due to the lack of in vitro/in vivo
prodrug activation data.16 The results presented herein show that
primaquine carbamates derived from phenols are readily hydroly-
sed both in alkaline solutions and human plasma, with chemical
and enzymatic reactivities being dependent on the electronic
effects on the alcohol moiety. Activation of O-aryl carbamates in
human plasma appears to be catalysed by BuChE, leading to car-
bamoylation of the catalytic serine and subsequent slow enzyme
reactivation and release of parent drug. Both O-aryl and O-alkyl
carbamates are slowly activated in rat liver homogenates, but only
O-aryl derivatives released primaquine at a measurable rate. When
assayed for gametocytocidal activity, ethyl, 4a, and n-hexyl, 4b,
derivatives were able to significantly reduce the percentage of
mosquitoes with oocysts and mean number of oocysts per infected
mosquito when compared to controls, thus indicating that carba-
mates of primaquine have the potential to be developed as
transmission-blocking antimalarial agents.
4.1.3. 8-[4-(1-Hexyloxycarbonyl)amino-1-methylbutylamino]-
6-methoxyquinoline (4b)
The procedure described for 4a was used with n-hexyl chloro-
formate (0.49 mL, 3 mmol) and primaquine diphosphate (0.911 g,
2 mmol) to afford 4b as a brown oil (42%). 1H NMR (CDCl3): d
0.86 (3H, t, J = 6.8), 1.14–1.21 (4H, m), 1.25 (3H, d, J = 6.4),
1.44–1.67 (8H, m), 3.12–3.21 (2H, m), 3.47–3.52 (1H, m), 3.77
(3H, s), 4.01 (2H, t, J = 7.0), 4.55 (1H, br s), 5.89 (1H, d, J = 8.4),
6.16 (1H, d, J = 2.6), 6.22 (1H, d, J = 2.6), 7.19 (1H, dd, J = 8.2, 4.2),
7.80 (1H, dd, J = 8.2, 1.6), 8.41 (1H, dd, J = 4.2, 1.6); 13C NMR
(CDCl3): d 14.4, 21.0, 22.9, 25.9, 27.2, 29.4, 31.9, 34.3, 41.4, 48.2,
55.6, 65.3, 92.1, 97.1, 122.2, 130.3, 135.2, 135.9, 144.7, 145.3,
157.2, 159.8; FAB-MS, m/z (%) 388 (47, MH+), 201 (100), 175 (64).
Anal. Calcd for C22H33N3O3: C, 68.19; H, 8.58; N, 10.84. Found: C,
68.25; H, 8.53; N, 10.86.
4.1.4. 8-[4-(2,2,2-Trifluoroethoxycarbonyl)amino-1-methylbu-
tylamino]-6-methoxyquinoline (4c)
4. Material and methods
4.1. Chemistry
Trifluoroethanol (0.07 mL, 1 mmol) in dried tetrahydrofuran
(5 mL) was slowly added to a solution of carbonyl diimidazole
(0.162 g, 1 mmol) in dried tetrahydrofuran (5 mL). After stirring
for 1 h at room temperature, a solution of primaquine free base
(0.263 g, 1 mmol; obtained by treating an aqueous solution of pri-
maquine diphosphate with an excess of Na2CO3 and then extract-
ing with dichloromethane, followed by drying and evaporation of
the solvent) in dried tetrahydrofuran (5 mL) was added to reaction
mixture. After overnight at room temperature the reaction mixture
was filtered and the solvent was evaporated under reduce pres-
sure. The residue was re-dissolved in dichloromethane (50 mL)
and the organic phase was washed (satd NaHCO3, water), dried
(anhydrous MgSO4) and evaporated to dryness. The residue was
further purified by dry column chromatography (silica gel; ether/
light petroleum, 7:3) to give 4c as brown oil in 43% yield. IR
4.1.1. General
Melting points were recorded using a Buchi 510 capillary melt-
ing-point apparatus and are uncorrected. 1H NMR spectra were re-
corded using a Bruker MSX-300 spectrometer. The 1H NMR data
are reported as follows: chemical shifts, expressed in ppm, using
internal TMS as reference; number of protons; multiplicity (s, sin-
glet; d, doublet; dd, double doublet; t, triplet; q, quartet; m, multi-
plet; br, broad); and coupling constants, J, quoted in Hertz. The IR
spectra were recorded using a Nicolet FTIR Impact 400 spectropho-
tometer either as thin films or KBr pellets. Mass spectra (FAB-MS)
were obtained on a VG Mass Lab 20–250 using glycerol as matrix.
Microanalyses were obtained from MEDAC (UK) or ITN laboratories
(Portugal). All kinetic measurements were made using either a Shi-
madzu UV-2100 spectrophotometer equipped with a Shimadzu
CPS-260 temperature controller or an HPLC system comprising a
Merck-Hitachi L-6000 pump, a Merck-Hitachi diode array detector
(cmꢀ1 mmax 3360, 1737; 1H NMR (CDCl3): d 1.23 (3H, d, J = 6.4),
)
1.58–1.68 (4H, m), 3.15–3.23 (2H, m), 3.51–3.60 (1H, m), 3.81
(3H, s), 4.36 (2H, q, J = 9.0), 4.89 (1H, br s), 5.92 (1H, d, J = 8.4),
6.20 (1H, d, J = 2.5), 6.27 (1H, d, J = 2.5), 7.23 (1H, dd, J = 8.2, 4.2),
7.85 (1H, dd, J = 8.2, 1.5), 8.45 (1H, dd, J = 4.2, 1.5); 13C NMR
(CDCl3): d 20.7, 26.6, 33.8, 41.4, 47.8, 55.3, 74.5, 91.8, 96.9, 121.9,
124.9, 130.0, 134.9, 135.6, 144.4, 144.9, 154.6, 159.5; FAB-MS,
m/z (%) 386 (48, MH+), 201 (100), 175 (60). Anal. Calcd for
and a 20 lL Rheodyne injector. Primaquine, (8-(4-amino-1-meth-
ylbutylamino)-6-methoxyquinoline), the alkyl and aryl chlorofor-
mates and carbonyl diimidazole (CDI) were purchased from Aldrich.
C
18H22F3N3O3: C, 56.10; H, 5.75; N, 10.90. Found: C, 56.36; H,
4.1.2. 8-(4-Ethoxycarbonylamino-1-methylbutylamino)-6-meth
oxyquinoline (4a)
5.59; N, 10.77.
Ethyl chloroformate (0.19 mL, 2 mmol) was added to a suspen-
sion containing primaquine diphosphate (0.911 g, 2 mmol) and tri-
ethylamine (3 mol equiv) in tetrahydrofuran (5 mL) at ꢀ10 °C. The
mixture was stirred at ꢀ10 °C for 30 min and then left at room
temperature for 90 min. The reaction mixture was filtered and
the filtrate evaporated. The residue was re-dissolved in dichloro-
methane and the organic layers were washed (satd NaHCO3,
water), dried (anhydrous MgSO4) and evaporated to dryness to
yield an oily residue that was subjected to column chromatogra-
phy on silica gel (ether/light petroleum, 7:3) to give 4a as a brown
oil in 53% yield. 1H NMR (CDCl3): d 1.15 (3H, t, J = 7.1), 1.23 (3H, d,
J = 6.4), 1.51–1.61 (4H, m), 3.11–3.20 (2H, m), 3.52–3.58 (1H, m),
3.81 (3H, s), 4.02 (2H, q, J = 7.1), 4.61 (1H, br s), 5.93 (1H, d,
J = 8.2), 6.20 (1H, d, J = 2.5), 6.26 (1H, d, J = 2.5), 7.23 (1H, dd,
J = 8.2, 4.2), 7.85 (1H, dd, J = 8.2, 1.6), 8.45 (1H, dd, J = 4.2, 1.6);
13C NMR (CDCl3): d 14.7, 20.9, 26.9, 34.0, 41.0, 47.9, 55.3, 60.7,
91.7, 96.8, 121.9, 130.0, 134.9, 135.4, 144.4, 145.0, 156.8, 159.5;
4.1.5. 8-[4-(4-Methoxyphenoxycarbonyl)amino-1-methylbu-
tylamino]-6-methoxyquinoline (4d)
The procedure described for 4a was used with 4-methoxy-
phenyl chloroformate (0.49 mL, 4 mmol) and primaquine diphos-
phate (1.82 g, 4 mmol) to afford 4d as a yellowish solid (49%).
Mp: 96–98 °C; IR (cmꢀ1 mmax 3330, 1703; 1H NMR (CDCl3): d
)
1.32 (3H, d, J = 6.4), 1.67–1.80 (4H, m), 3.57–3.69 (2H, m), 3.72–
3.79 (1H, m), 3.79 (3H, s), 3.89 (3H, s), 5.02 (1H, br s), 6.02 (1H,
d, J = 8.4), 6.29 (1H, d, J = 2.4), 6.34 (1H, d, J = 2.4), 6.85 (2H, d,
J = 9.0), 7.01 (2H, d, J = 9.0), 7.31 (1H, dd, J = 8.2, 4.2), 7.92 (1H,
dd, J = 8.2, 1.6), 8.53 (1H, dd, J = 4.2, 1.6); 13C NMR (CDCl3): d
20.8, 26.9, 33.8, 41.2, 47.8, 55.2, 55.6, 91.7, 97.8, 114.3, 121.8,
122.4, 129.9, 134.8, 135.3, 144.3, 144.7, 144.9, 154.9, 156.8,
159.4; FAB-MS, m/z (%) 410 (51, MH+), 286 (93), 201 (100), 175
(67). Anal. Calcd for C23H27N3O4: C, 67.45; H, 6.65; N, 10.26. Found:
C, 67.40; H, 6.53; N, 10.24.

G. Mata et al. / Bioorg. Med. Chem. 20 (2012) 886–892
891
4.1.6. 8-(4-Phenoxycarbonylamino-1-methylbutylamino)-6-
methoxyquinoline (4e)
treated with 1 N HCl and extracted with dichloromethane
(10 mL). The resulting organic phase was dried (anhydrous MgSO4)
and evaporated to dryness to afford 4i as a brown oil (36%). IR
The procedure described for 4a was used with phenyl chlorofor-
mate (0.50 mL, 4 mmol) and primaquine diphosphate (1.82 g,
(cmꢀ1 mmax 3480 (br), 3320, 1730, 1712; 1H NMR (CDCl3): d 1.23
)
4 mmol) to afford 4e as a light brown gum (42%). IR (cmꢀ1
)
mmax
(3H, d, J = 6.4), 1.50–1.84 (4H, m), 3.46–3.58 (2H, m), 3.60–3.81
(1H, m), 3.81 (3H, s, OCH3), 4.62 (2H, s), 4.82 (1H, br s), 5.92 (1H,
d, J = 8.4), 6.20 (1H, d, J = 2.4), 6.26 (1H, d, J = 2.5), 7.23 (1H, dd,
J = 8.2, 4.2), 7.84 (1H, dd, J = 8.2, 1.6), 8.45 (1H, dd, J = 4.2, 1.6),
11.51 (1H, br s); 13C NMR (CDCl3): d 20.5, 24.3, 33.5, 40.1, 47.6,
55.2, 67.7, 91.8, 96.8, 121.8, 129.8, 134.7, 135.3, 144.3, 144.8,
155.8, 159.4, 170.4. Anal. Calcd for C18H23N3O5: C, 59.82; H, 6.41;
N, 11.63. Found: C, 60.11; H, 6.33; N, 11.44.
3367, 1734; 1H NMR (CDCl3): d 1.33 (3H, d, J = 6.4), 1.67–1.80 (4H,
m), 3.29–3.34 (2H, m), 3.64–3.68 (1H, m), 3.89 (3H, s), 5.05 (1H, br
s), 6.02 (1H, d, J = 8.4), 6.30 (1H, d, J = 2.4), 6.34 (1H, d, J = 2.6), 7.11
(2H, d, J = 7.5), 7.15–7.37 (4H, m), 7.93 (1H, dd, J = 8.2, 1.6), 8.54
(1H, dd, J = 4.2, 1.6); 13C NMR (CDCl3): d 20.6, 26.6, 33.8, 41.2, 47.8,
55.2, 91.7, 96.7, 121.6, 121.8, 125.2, 129.2, 129.9, 134.8, 135.5,
144.3, 145.1, 151.0, 154.7, 159.5; FAB-MS, m/z (%) 380 (45, MH+),
201 (100), 175 (44). Anal. Calcd for C22H25N3O3: C, 69.64; H, 6.64;
N, 11.07. Found: C, 69.43; H, 6.67; N, 11.19.
4.1.11. Carboxyprimaquine (3)
Carboxyprimaquine, 3, was synthesized according to the proce-
dure reported by McChesney et al.18 Anal. Calcd for C15H18N2O3: C,
65.68; H, 6.61; N, 10.21. Found: C, 65.71; H, 6.55; N, 10.14.
4.1.7. 8-[4-(4-Chlorophenoxycarbonyl)amino-1-methylbutyla-
mino]-6-methoxyquinoline (4f)
The procedure described for 4a was used with 4-chlorophenyl
chloroformate (0.56 mL, 4 mmol) and primaquine diphosphate
(1.82 g, 4 mmol) to afford 4f as a yellowish solid (36%). Mp:
4.2. In vitro studies
106–108 °C; IR (cmꢀ1
)
mmax 3354, 1736; 1H NMR (CDCl3): d 1.34
4.2.1. Hydrolysis in aqueous solutions
(3H, d, J = 6.1), 1.68–1.80 (4H, m), 3.28–3.30 (2H, m), 3.64–3.69
(1H, m), 3.89 (3H, s), 5.05 (1H, br s), 6.02 (1H, d, J = 8.2), 6.29 (1H, d,
J = 2.5), 6.34 (1H, d, J = 2.4), 7.05 (2H, d, J = 8.7), 7.30 (2H, d, J = 8.7),
7.32 (1H, dd, J = 8.2, 4.5), 7.93 (1H, dd, J = 8.2; 1.6), 8.50 (1H, dd,
J = 4.5, 1.6); 13C NMR (CDCl3): d 20.6, 26.5, 33.8, 41.2, 47.9, 55.2,
91.7, 96.8, 121.9, 122.9, 129.2, 129.9, 130.0, 134.8, 135.5, 144.4,
145.2, 150.3, 154.0, 159.2; FAB-MS, m/z (%) 416 (11, MH++2), 414
(33, MH+), 201 (100), 175 (42). Anal. Calcd for C22H24ClN3O3: C,
63.84; H, 5.84; N, 10.15. Found: C, 64.01; H, 5.79; N, 9.99.
All kinetic experiments were carried out in aqueous borate and
NaOH buffers, containing 20% (v/v) acetonitrile to aid substrate sol-
ubility, at an ionic strength of 0.5 M (NaClO4). Kinetic studies on the
aryl carbamates were carried out by UV spectrophotometry at fixed
wavelength by recording the decrease of substrate absorbance in the
appropriate buffer solutions at 37 °C. In a typical run, reaction was
initiated by adding a 15 l
L aliquot of a 10ꢀ2 M stock solution of sub-
strate in acetonitrile to a cuvette containing 3 ml of the pre-equili-
brated buffer solution at 37 °C. The pseudo-first-order rate
constants, kobs, were obtained from the slopes of plots of ln(At ꢀ A
)
1
4.1.8. 8-[4-(4-Nitrophenoxycarbonyl)amino-1-methylbutyla-
mino]-6-methoxyquinoline (4g)
versus time, where At and A represent the absorbance at time t and
1
at infinity, respectively. For alkyl carbamates 4a–c and 4h, the ki-
The procedure described for 4a was used with 4-nitrophenyl
chloroformate (1.01 g, 5 mmol) and primaquine diphosphate
netic studies were carried out by HPLC following the loss of the sub-
strate and formation of primaquine. In this method a 15
10ꢀ2 M stock solution of substrate was added to a reaction flask con-
taining 3 ml of the buffer solution. At regular intervals, 50 L sam-
ples of the reaction mixture were neutralised with HCl solutions
and analysed using a Lichrosorb RP-8 column (5 m particles,
lL aliquot of
(2.28 g, 5 mmol) to afford 4g as a brown oil (10%). IR (cmꢀ1
) mmax
3332, 1705; 1H NMR d (CDCl3): 1.33 (3H, d, J = 6.2), 1.65–1.74
(4H, m), 3.31–3.35 (2H, m), 3.70 (1H, m), 3.89 (3H, s), 5.22 (1H,
br s), 6.01 (1H, d, J = 8.4), 6.30 (1H, d, J = 2.4 Hz), 6.35 (1H, d,
J = 2.4), 7.30 (1H, dd, J = 8.2, 4.0), 7.30 (2H, d, J = 8.9), 7.93 (1H,
dd, J = 8.2, 1.1), 8.20 (2H, d, J = 9.0), 8.53 (1H, dd, J = 4.0, 1.5); 13C
NMR (CDCl3): d 21.1, 26.8, 34.2, 41.8, 48.2, 55.6, 92.2, 97.3, 116.2,
122.3, 125.5, 126.6, 130.3, 135.3, 144.7, 144.8, 145.2, 153.1,
156.1, 159.8; FAB-MS, m/z (%) 425 (33, MH+), 201 (100), 175 (41).
Anal. Calcd for C22H24N4O5: C, 62.25; H, 5.70; N, 13.20. Found: C,
62.27; H, 5.69; N, 13.15.
l
l
4 ꢄ 250 mm; Merck) and an eluant consisting of acetonitrile/
water/KH2PO4 1 M (55:40:5) at a flow rate of 1 mL/min. The hydro-
lysis of 4a–f and 4h in pH 7.4 buffer were also studied by HPLC.
4.2.2. Hydrolysis in human plasma
Human plasma was obtained from the heparinised blood of
healthy donors, and was pooled and frozen at ꢀ70 °C before use.
Typically, a 35
was added to 350
pH 7.4 isotonic phosphate buffer, at 37 °C. At appropriate intervals,
75 L aliquots were added to acetonitrile (350 L) to quench the
l
L aliquot of 10ꢀ2 M stock solution of compound 4
4.1.9. 8-[4-(2-Ethoxycarbonylmethoxycarbonyl)amino-1-meth-
ylbutylamino]-6-methoxyquinoline (4h)
lL of the human plasma diluted to 50% (v/v) with
The procedure described for 4c was used with ethyl glycolate
(0.09 mL, 1 mmol) and primaquine free base (0.262 g, 1 mmol) to
l
l
reaction and precipitate plasma proteins. These samples were centri-
fuged (10,000 rpm for 10 min) and the clear supernatant analysed by
HPLC for the presence of both substrate and primaquine, using a
afford 4h as a brown oil (27%). IR (cmꢀ1
) mmax 3365, 1740, 1732;
1H NMR (CDCl3): d 1.26 (3H, d, J = 7.1), 1.29 (3H, t, J = 7.0),
1.62–1.76 (4H, m), 3.20–3.30 (2H, m), 3.57–3.65 (1H, m), 3.89
(3H, s), 4.19 (2H, q, J = 7.1), 4.65 (2H, s), 4.93 (1H, br s), 6.02 (1H,
d, J = 8.2), 6.28 (1H, d, J = 2.5), 6.34 (1H, d, J = 2.5), 7.30 (1H, dd,
J = 8.2, 4.2), 7.92 (1H, dd, J = 8.2, 1.6), 8.53 (1H, dd, J = 4.2, 1.6);
13C NMR (CDCl3): d 14.1, 20.6, 26.6, 33.8, 41.1, 47.8, 55.2, 61.1,
61.6, 91.7, 96.8, 121.8, 129.9, 134.8, 135.2, 144.2, 144.8, 155.5,
159.4, 168.7. Anal. Calcd for C20H27N3O5: C, 61.68; H, 6.99; N,
10.79. Found: C, 60.82; H, 6.73; N, 10.64.
Lichrosorb RP-8 column (5
lm particles, 4 ꢄ 250 mm; Merck) and
an eluant consisting of acetonitrile/buffer (40:60%) at a 1.0 mL/min
flow rate. The buffer contained 10 mM sodium hexanesulfonate,
2.5 mM sodium acetate and 2.5 mM phosphoric acid. For compounds
4d–g, the corresponding phenol products were also analysed. Quan-
titation of the substrate and corresponding products was achieved
from measurement of peak areas relative to those of corresponding
standards subjected to the same chromatographic conditions. The
first-order-rate constants, kplasma, were determined in triplicate.
4.1.10. ({4-[(6-Methoxyquinolin-8-yl)amino]pentyl}carbamoyl-
oxy) acetic acid (4i)
4.2.3. Hydrolysis in rat liver homogenates
A solution of 4h (72.2 mg, 0.2 mmol) in aqueous 0.1 N NaOH in
dioxane (3 mL) was stirred for 2 h. The reaction mixture was
The metabolism of primaquine carbamates 4 in rat liver homog-
enates15 was monitored using HPLC following either the loss of the

892
G. Mata et al. / Bioorg. Med. Chem. 20 (2012) 886–892
substrate and the formation of primaquine. Incubations were car-
ried out at 37 °C in pH 7.4 phosphate buffer saline (PBS) using
2 mg protein/mL of liver homogenates and a NADPH generating
system consisting of glucose-6-phosphate 6.25 mM, NADP
1.25 mM, MgCl2 6 mM and 2.5 U/mL of glucose-6-phosphate dehy-
drogenase. The substrates, dissolved in pH 7.4 PBS, were added to
give initial concentrations of 0.25–5.0 mM. At regular intervals,
References and notes
1. WHO, World Malaria Report 2011.
2. Rodrigues, T.; Moreira, R.; Lopes, F. Future Med. Chem. 2011, 3, 1.
3. Olliaro, P.; Wells, T. N. C. Clin. Pharmacol. Ther. 2009, 85, 584.
4. Wells, T. N. C.; Alonso, P. L.; Gutteridge, W. E. Nat. Rev. Drug Disc. 2009, 8,
879.
5. Gamo, F. J.; Sanz, L. M.; Vidal, J.; Cozar, C.; Alvarez, E.; Lavandera, J. L.;
Vanderwall, D. E.; Green, D. V. S.; Kumar, V.; Hasan, S.; Brown, J. R.; Peishoff, C.
E.; Cardon, L. R.; Garcia-Bustos, J. F. Nature 2010, 465, 305.
6. Alonso, P. L.; Djimde, A.; Kremsner, P.; Magill, A.; Milman, J.; Najera, J.; Plowe, C.
V.; Rabinovich, R.; Wells, T.; Yeung, S. PLoS Med. 2011, 8, e1000402.
doi:10.1371/journal.pmed.1000402.
7. Vale, N.; Moreira, R.; Gomes, P. Eur. J. Med. Chem. 2009, 44, 937. and references
cited herein.
8. Wells, T. N. C.; Burrows, J. N.; Baird, J. K. Trends Parasitol. 2010, 26, 145.
9. Clark, A. M.; Baker, J. K.; McChesney, J. D. J. Pharm. Sci. 1984, 73, 502.
10. Baker, J. K.; Bedford, J. A.; Clark, A. M.; McChesney, J. D. Pharm. Res. 1984, 1,
98.
25–100 lL samples of the liver incubates were withdrawn and
added to acetonitrile (1.2 mL), the resulting mixture was centri-
fuged at 10,000 rpm for 5 min. The supernatant was analysed using
the same HPLC described for studies in human plasma. The first-or-
der-rate constants, kliver, were determined in triplicate.
4.2.4. Apparent partition coefficients
The apparent partition coefficients between 1-octanol and pH
7.4 phosphate buffer, D7.4, were determined at 22 °C. Both the
aqueous and octanol phases were mutually saturated before the
experiment. The volumes of each phase were chosen so that solute
concentrations in the aqueous phase after distribution were readily
measured. The compounds were dissolved in 1-octanol and the
octanol–phosphate buffer mixtures were shaken for 30 min to
reach an equilibrium distribution. Each phase was analysed sepa-
rately by HPLC. Partition coefficients were calculated from the ratio
of the peak area in 1-octanol to the peak area in buffer and the re-
ported values are the average of three experiments. Partition coef-
ficients for compounds 4 were calculated using the ALOGPS 2.1
program (available at http://146.107.217.178/lab/alogps/).38
11. Mihaly, G. W.; Ward, S. A.; Edwards, G.; Leorme, M.; Breckenridge, A. M. Br. J.
Clin. Pharmacol. 1984, 17, 441.
12. Testa, B.; Krämer, S. D. Chem. Biodivers. 2009, 6, 591.
13. Guarino, V. R. In Prodrugs and Targeted Delivery: Towards Better ADME
Properties; Rautio, J., Ed.; Wiley-VCH: Weinheim, 2011; pp 31–78.
14. Silverman, R. B. The Organic Chemistry of Enzyme-Catalyzed Reactions; Academic
Press: San Diego, CA, 2000. Chapters 3 and 4.
15. Portela, M. J.; Moreira, R.; Valente, E.; Constantino, L.; Iley, J.; Pinto, J.; Rosa, R.;
Cravo, P.; do Rosário, V. E. Pharm. Res. 1999, 16, 949.
16. Vacondio, F.; Silva, S.; Mor, M.; Testa, B. Drug Metab. Rev. 2010, 42, 551.
17. Williams, A.; Douglas, K. T. Chem. Rev. 1975, 75, 627.
18. McChesney, J. D.; Sarangan, S. Pharm. Res. 1984, 1, 96.
19. Hegarty, A. F.; Frost, L. N. J. Chem. Soc., Perkin 2 1973, 1719.
20. Li, B.; Sedlacek, M.; Manoharan, I.; Boopathy, R.; Duysen, E. G.; Masson, P.;
Lockridge, O. Biochem. Pharmacol. 2005, 70, 1673.
21. Bar-On, P.; Millard, C. B.; Harel, M.; Dvir, H.; Enz, A.; Sussman, J. L.; Silman, I.
Biochemistry 2002, 41, 3555.
22. Wilson, I. B.; Harrison, M. A.; Ginsburg, S. J. Biol. Chem. 1961, 236, 1498.
23. Verheijen, J. C.; Wiig, K. A.; Du, S.; Connors, S. L.; Martin, A. N.; Ferreira, J. P.;
Slepnev, V. I.; Kochendorfer, U. Bioorg. Med. Chem. Lett. 2009, 19, 3243.
24. Lopes, F.; Moreira, R.; Iley, J. Bioorg. Med. Chem. 2000, 8, 707.
25. Sykes, N. O.; Macdonald, S. J. F.; Page, M. I. J. Med. Chem. 2002, 45, 2850.
26. Mulchande, J.; Guedes, R. C.; Tsang, W. Y.; Page, M. I.; Moreira, R.; Iley, J. J. Med.
Chem. 2008, 51, 1783.
27. Constantino, L.; Paixao, P.; Moreira, R.; Portela, M. J.; Do Rosario, V. E.; Iley, J.
Exp. Toxicol. Phatol. 1999, 51, 299.
28. Gangl, E.; Utkin, I.; Gerber, N.; Vouros, P. J. Chromatogr., A 2002, 974, 91.
29. Yumibe, N.; Huie, K.; Chen, K. J.; Snow, M.; Clement, R. P.; Cayen, M. N. Biochem.
Pharmacol. 1996, 51, 165.
30. Treluyer, J. M.; Bowers, G.; Cazali, N.; Sonnier, M.; Rey, E.; Pons, G.; Cresteil, T.
Drug Metab. Dispos. 2003, 31, 275.
31. Satoh, T.; Hosokawa, M. Annu. Rev. Pharmacol. Toxicol. 1998, 38, 257.
32. Yoon, K. J. P.; Morton, C. L.; Potter, P. M.; Danks, M. K.; Lee, R. E. Bioorg. Med.
Chem. 2003, 11, 3237.
33. Baker, J. K.; Yarber, R. H.; Nanayakkara, N. P. D.; McChesney, J. D.; Homo, F.;
Landau, I. Pharm. Res. 1990, 7, 91.
34. Abu-El-Hay, S.; Allahyari, R.; Chavez, E.; Frazer, I. M.; Strother, A. Xenobiotica
1988, 18, 1165.
35. Araújo, M. J.; Bom, J.; Capela, R.; Casimiro, C.; Chambel, P.; Gomes, P.; Iley, J.;
Lopes, F.; Morais, J.; Moreira, R.; Oliveira, E.; Rosário, V.; Vale, N. J. Med. Chem.
2005, 48, 888.
36. Vale, N.; Prudêncio, M.; Marques, C. A.; Collins, M. S.; Gut, J.; Nogueira, F.;
Matos, J.; Rosenthal, P. J.; Cushion, M. T.; do Rosário, V.; Mota, M. M.; Moreira,
R.; Gomes, P. J. Med. Chem. 2009, 52, 7800.
37. Vaughan, J. A.; Noden, B. H.; Beier, J. C. J. Parasitol. 1992, 78, 716.
38. Tetko, I. V.; Gasteiger, J.; Todeschini, R.; Mauri, A.; Livingstone, D.; Ertl, P.;
Palyulin, V. A.; Radchenko, E. V.; Zefirov, N. S.; Makarenko, A. S.; Tanchuk, V. Y.;
Prokopenko, V. V. J. Comput. Aided Mol. Des. 2005, 19, 453.
39. Ballinger, P.; Long, F. A. J. Am. Chem. Soc. 1960, 82, 795.
40. The pKa value for ethyl glycolate (the leaving group for 4h) and 1-hexanol (the
leaving group for 4b) were calculated using the equation pKa = 15.9–1.42r⁄
determined for the dissociation of RCH2OH, presented in Ref. 39, and using the
r⁄ values of 2.12 for R = CO2Et and ꢀ0.25 for R = (CH2)4CH3 (Zhang, J.;
Kleinoder, T.; Gasteiger, J. J. Chem. Inf. Model. 2006, 46, 2256).
41. Albert, A.; Serjeant, E. P. Ionization Constants of Acids and Bases; Methuen:
London, 1962.
4.3. In vivo studies
4.3.1. Sporogonic development of Plasmodium berghei
P. berghei ANKA 25R/10 strain, Balb C mice and A. stephensi mos-
quitoes were used throughout the course of this study. Mice were
infected by intraperitoneal inoculations of 107 erythrocytes parasi-
tised with P. berghei ANKA. After four days, when the presence of
parasites in the blood (namely gametocytes) was observed by
microscopic observation of Giemsa stained blood films, mice were
randomly separated into five different groups of six animals. Each
group was treated by intraperitoneal administration with one sin-
gle dose of each compound (1.86, 3.75, 7.50 and 15.0 mg/kg body
weight in inoculation volumes of 0.1–0.2 mL; controls consisted
of mice given a saline solution). Two hours after administration,
mice were anesthetised (diethyl ether) and placed on top of individ-
ual cages containing ca. 50 glucose-starved mosquitoes. These were
allowed to feed for 2 h. After the blood meal, unfed females were re-
moved from each cage. Engorged mosquitoes were maintained in
the insectarium at 21 1 °C and 60% relative humidity, receiving
a solution containing 10% glucose, 0.05% PABA and 0.001% gentam-
icine (all v/v). Ten days after the blood meal, 10 mosquitoes of each
cage were randomly collected and dissected for microscopic detec-
tion (100ꢄ and 400ꢄ) of oocysts in midguts. Dissection was per-
formed, using a stereomicroscope, in a drop of saline solution.
Acknowledgment
The authors are grateful to the Fundação para a Ciência e Tecn-
ologia (PEst-OE/SAU/UI4013/2011) for financial support of this
research.