Tropical Medicine and International Health
VOLUME I N O . 2 PP
2 5 5 - 2 6 3 APRIL 1996
(S)-9-(3-hydroxy-2-phosphonylmethoxypropyl)adenine
[(S)-H PMPA]: a purine analogue with trypanocidal
activity in vitro and in vivo
Ronald Kaminsky', Cecile Schmid', Yvonne Grether', Antonin Holy', Erik De Clercq3, Lieve Naesens3
and Reto Brun'
Swiss Tropical Institute, Socinstr. 57, 4002 Basel, Switzerland
of Organic Chemtstry and Biochemistry, Academy of Sciences, Praha, Czech Republic
Rega Institute for Medical Research, Katholieke Universiteit Leuven, Belgium
' Institute
Summary
The unique features of purine salvage systems of pathogenic haemoflagellates render them
selectively susceptible to the cytotoxic effects of purine analogues. A series of acyclic
nucleoside phosphonates were evaluated for activity against pathogenic haemoflagellates in
vitro. One of the phosphonylmethoxyalkylpurines,namely
(S)-9-(3-hydroxy-2-phosphony1methoxypropyl)adenine[(S)-HPMPA],was active in vitro
against bloodstream forms of Trypanosoma brucei rhodesiense, T. b. gambiense,
multidrug-resistant T. b. brucei, T. congolense and T. evansi, but not against intracellular
T. crwi or Leishmania donovani. Cytotoxic effects against mammalian cells were observed
a t 4900-27 300-fold higher concentrations than those necessary to inhibit T. 6. rhodesiense.
(S)-HPMPAwas able to eliminate T. 6. rhodesiense and multidrug-resistant T. b. brucei in
an acute rodent model with two administrations of 10mg/kg each.
keywords African trypanosomes, chemotherapy, purines, (S)-HPMPA, in vitro assays,
cytotoxicity
correspondence Dr R. Kaminsky, Swiss Tropical Institute, Socinstr. 5 7, 4002 Basel,
Switzerland
Introduction
New compounds effective against the protozoan
parasites of the genus Trypanosoma are urgently
needed. Treatment of the lethal human African
sleeping sickness is unsatisfactory because one of the
only two drugs available (melarsoprol) is associated
with severe side-effects (Pepin et al. 1989; Kuzoe
1993) while the other drug for late stage disease,
DL-a-difluoromethylornithine (DFMO), is effective
only against Trypanosoma brucei gambiense infections but not for T. b. rhodesiense caused disease
(Bales et a[. 1989; Bacchi et al. 1990; Iten et al.
1995). Additionally, efficacy of the few drugs avail-
255
0 1996 Blackwell Science Lrd
able for animal pathogenic trypanosomes is threatened by the occurrence of drug-resistant trypanosomes (Ainanshe et al. 1992; Codjia et al. 1993).
African trypanosomes differ substantially in many
ways from their hosts regarding metabolic pathways.
These biochemical peculiarities represent potential
targets for chemotherapeutic attack. In contrast to
man and domestic animals, African trypanosomes
rely on preformed purine bases and nucleosides
present in the circulatory system of their host since
they are unable to synthesize them de novo (Fish
et al. 1982; Davies et al. 1983; Gottlieb 1985).
Bases, such as hypoxanthine and xanthine, are produced continuously in man by catabolism of purine
Tropical Medicine and International Health
VOLUME I N O 2 PP
255-263 APRIL 1996
R Karninsky et ol. (S)-HPMPA a trypanocidal purine analogue
nucleotides. Thus there is considerable opportunity
for bloodstream form trypanosomes to satisfy their
purine requirements by salvage. In addition, trypanosoma1 nucleoside hydrolase activity resulting in
cleavage of purine ribonucleosides into the respective
free bases has been demonstrated (Davies et al.
1983). Phosphoribosyltransferase activities towards
all four purine bases in Trypanosoma brucei brucei
(Hassan & Coombs 1986) show that it is these
enzymes that enable trypanosomes to salvage purine
bases and channel them as nucleotide components
into metabolic processes. All purine bases and
ribonucleosides are interconverted.
The unique features of the purine salvage systems
of trypanosomes render them selectively susceptible
to the cytotoxic effects of several purine analogues,
such as formycin B, ydeazainosine (Fish et al.
1985;Kaminsky & Zweygarth 1989), or methylthioadenosine analogues (Bacchi et al. 1991). We have
shown that some phosphonylmethoxyalkylpurines
express antitrypanosomal activity (Kaminsky et al.
rqgqb). These acyclic nucleoside phosphonates are
unique in that the phosphorus atom is attached to
the alkyl side-chain of the purine base via an 0-C-P
bond which is very resistant to enzymatic degradation when compared with the usual C-0-P bond.
Phosphonylmethoxyalkylpurines were initially synthesized as antivirals (De Clercq et al. 1986; 1989;
Rosenberg & Holji 1989). Nevertheless, some phosphonylmethoxyalkylpurines were able to eliminate
trypanosomes (Kaminsky et al. 1994b) and Plasmodium (de Vries et al. 1991). Among the acyclic
nucleoside phosphonates tested, (S)-9-(3-hydroxy-2phosphonylmethoxypropy1)adenine[(S)-HPMPA]
showed the highest activity against Trypanosoma
brucei brucei in in vitro assays (Kaminsky et al.
1994b). The aim of this study was to evaluate the
antitrypanosomal activity of (S)-HPMPA against
human and animal pathogenic trypanosome species
in vitro and in rodent models.
Hospital, Ifakara, Tanzania, from a human patient.
After several passages in rodents and a cyclical
passage in Glossina morsitans morsitans, a cloned
population was adapted to axenical growth in vitro.
Trypanosoma congolense STIB 910 is a cloned
derivative of stock STIB 249 which was isolated in
1971 from a lion in Tanzania (Geigy & Kauffmann
1973). Trypanosoma 6. brucei STIB 920 is a derivative of STIB 348 which was isolated in 1971 in
the Serengeti National Park (Tanzania) from a
hartebeest (Alcelaphus buselaphus cokii). Following
several passages in rodents and a cyclical passage in
G. m. morsitans, a cloned population was adapted
to axenical growth. Trypanosoma 6. gambiense STIB
930 is a derivative of TH-1/78E(o3 I ) which was isolated in 1978 from a patient from the Ivory Coast
(Mehlitz et al. 1981). Trypanosoma 6. brucei STIB
950 is a derivative of CP (Chemotryp-Project) 2469
which was isolated in 1985 from a bovine in
Hakaka, Soakow District, Somalia. The culture
adapted population shows a multidrug-resistant
phenotype (Kaminsky et al. 1989). Trypanosoma
evarrsi STIB 806 was isolated in 1983 from a water
buffalo in China (Lun et al. 1992). T . 6. brucei
GVR-3s is a derivative of LUMP 22 which was
isolated from a wildebeest in the Serengeti in 1966.
Leishmania donovani, MHOM/ET/67/L8 2, and
Trypanosoma cruzi MHOM/Br/oo/Y-strain, were
propagated in mouse peritoneal macrophages and in
the human foetal lung fibroblast cell line WI-3 8
(ATCC C C L ~ S )respectively.
,
HT-29 human adenocarcinoma cells were isolated
from a primary tumour in 1964 and were obtained
from the American Type Culture Collection (ATCC,
reference number HTB 38). HT-29 cells grow as a
monolayer and express an epithelial-like morphology. Bovine aorta endothelial (BAE) cells were
isolated in 1987 from material obtained from a
slaughterhouse in Basel, Switzerland.
Cultivation of parasites and cell lines
Materials and methods
Parasites and cell lines
Trypanosoma brucei rhodesiense STIB (Swiss
Tropical Institute Basel) 900 is a derivative of STIB
704 which was isolated in 1982 a t St Francis
256
0 1996 Blackwell Science Ltd
T. 6. rhodesiense, T. b. gambiense, T. 6. brucei, and
T . evansi were propagated in vitro in a liquid
medium consisting of Minimum Essential Medium
(MEM, Gibco-RBL No. 072-1100 powder) with
Earle’s salts with I 8/1 glucose, 10 ml/l MEM nonessential amino acids ( IOO x ), 2 . 2 gl/l NaHCO, and
Tropical Medicine and International Health
VOLUME I NO. 2 PP
255-263
APRIL
1996
R Kaminsky et ol. (S)-HPMPA a trypanocidal purine analogue
10 mM HEPES. The medium was further supplemented with (final concentration) 2 mM sodium
pyruvate, 0.2 mM 2-mercaptoethanol, 0.01 6 mM
thymidine (Baltz et al. 1985), 0.1mM hypoxanthine
and 15% heat inactivated horse serum (prepared
from horse blood obtained from a local slaughterhouse). The medium for T. b. gambiense was supplemented with heat inactivated 10% human serum
(STI human serum pool) and 5 % foetal bovine
serum (Biological Industries, Israel). Trypanosoma
congolense was propagated according to Kaminsky
et al. (1994a) in Iscove’s medium (GIBCO No. 074ozzoo,Life Technologies, Basel, Switzerland) supplemented with 3.024 g/l sodium pyruvate, 0.05 mM
bathocuproine disulphonic acid, I .5 mM L-cysteine,
0.5 mM hypoxanthine, z mM L-glutamine, 0.12mM
2-mercaptoethanol, I mM sodium pyruvate, 0.16 mM
thymidine and 20% heat inactivated goat serum
(c.c.PR0 GmbH, Karlsruhe, Germany). All cultures
were kept in ~4-multiwellplates (Costar, Cambridge,
MA, USA) at 37’C (or 34OC for T. congolense) in a
humidified atmosphere in 5% CO,. Cultures were
subpassaged to a density of 1o2-1oS
trypanosomes/ml every second or third day.
Trypanosomes in logarithmic growth phase were
used for determination of drug sensitivities.
The medium for cultivation of T. cruzi consisted
of M E M (Gibco-RBL No. 072-1100 powder) supplemented with I % M E M non-essential amino acids
( IOO x ) and 10% heat inactivated foetal bovine
serum. Monolayers of the human foetal lung
fibroblast cell line WI-38 (ATCC CCL75) were
subsequently infected with trypomastigote forms of
T. cruzi.
All mammalian cells were propagated in a liquid
medium consisting of M E M supplemented with 10%
heat inactivated foetal bovine serum (Biological
Industries, Israel). Stock cultures of mammalian cells
were maintained in T-z5 flasks (Falcon, Becton
Dickinson, Basel, Switzerland) in a humidified atmosphere at 3 7 T in 5 % CO,..Cells were subpassaged to
the appropriate split ratio (1:4-1:6) once a week.
Drugs
Melarsoprol (Arsobal) was a gift of Specia (Paris,
France). (S)-9-(j-hydroxy-z-phosphonylmethoxypropy1)adenine [(S)-HPMPA] and the
257
0 1996 Blackwell Science Ltd
corresponding guanine [(S)-HPMPG] and 2,6diaminopurine [(S)-HPMPDAP] counterparts, as well
as the (R)-enantiomer [(R)-HPMPA] were synthesized according to Holj. (1993). (S)-cHPMPA was
prepared according to Rosenberg and Holj. (1987).
The structures are illustrated in Figure I. For animal
experiments, (S)-HPMPAwas dissolved in distilled
water by neutralization with sodium hydroxide to
give a stock solution of I mg/ml.
In vitro assays
Trypanosomes or cells were seeded at appropriate
densities such that after the 7z-hour incubation
period trypanosomes in suspension cultures were
at the end of the logarithmic growth phase and
T. congolense and mammalian cells growing in
adherent cultures were forming an almost confluent
cell layer. Cells were propagated in the appropriate
medium for 72 hours in the presence of various drug
concentrations. All assays were performed in 96-well
plates (Costar, Cambridge, MA, USA). Trypanosomes or cells were incubated in a total volume of
IOO pl/well at 37’C except for T. congolense cultures
which were incubated at 34’C. First, the minimum
inhibitory concentration (MIC, defined as the concentration at which no trypanosomes or cells with
normal motility or morphology were detected) and
the maximum tolerated concentration (MTC, defined
as the concentration at which trypanosomes or cells
behaved the same as control cultures regarding
growth, motility and morphology) were determined
microscopically by viewing cultures with an inverted
microscope ( x zoo magnification).
Secondly, a fluorometric assay was performed
according to Obexer et al. (1995). Briefly, the
(6)fluorochromes ~’,7’-bis-carboxyethyl-~
carboxyfluoresceinpentaacetoxymethylester
(BCECF-AM) or Calcein-AM were added to the
cultures subsequently to drug exposure. In viable
cells, the non-fluorescent BCECF-AM or Calcein-AM
(used for mammalian cells only) was cleaved by
unspecific esterases to fluorescent products. After
incubation for I hour fluorescence was quantified
with a fluorescence plate reader (CytoFluor 2300,
Millipore Corp., Bedford, USA) employing an excitation wavelength of 485 f 20 nm and an emission
wavelength of 530 f zs nm.
Tropical Medicine and International Health
VOLUME 1 N O 2 PP
255-263
APRIL
1996
R Kaminsky et a/. (S)-HPMPA a trypanocidal purine analogue
(S)-HPMPA
(R)-HPMPA
(S)-H PMPDAP
0
(S)-HPMPG
(S)-cH PMPA
Figure I Chemical structures of HPMD-analogues.
Fluorescence of samples after drug exposure was
expressed as percentage of control cultures. The
percentage values of fluorescence were plotted
against the corresponding drug concentration on a
semi-logarithmic scale. EC,, values (defined as the
concentration which inhibited fluorescence intensity
by 5 0 % ) were quantified by linear interpolation
according to Huber and Koella (1993).
In vitro activity of (S)-HPMPA against T. cruzi
was determined using a 5-day assay by Brun et al.
(unpublished). Briefly, monolayers of the human foetal lung fibroblast cell line WI-38 were infected with
trypomastigote forms of T. cruzi. WI-38 cells were
seeded at a density of 10s cells/ml in I ml samples
into 24-well culture plates (Costar, USA). After 48
hours, the medium was removed a n d the cell layer
was infected with 1 0 5 trypomastigotes T. cruzi. The
infection was allowed to develop for 48 hours after
which the medium was replaced with fresh medium
containing the appropriate drug concentration. After
an additional 48 hours, the medium was replaced
with fresh medium containing the drug. Propagation
of amastigotes and appearance of trypomastigotes
258
0 1996 Blackwell Science Ltd
under drug pressure were determined microscopically
after a total drug exposure period of 5 days.
Susceptibility of L. doizovani to (5)-HPMPA in
vitro was tested following the procedure described by
Neal and Croft (1984). Briefly, Swiss ICR mouse
peritoneal macrophages were infected with L.
donovuni amastigotes, from hamster spleen at a ratio
of 5-10 parasites per macrophage. Test cultures were
maintained in Lab-tek 8-chamber slides (Nunc,
Napperville, USA) at s%CO, and 3 7 T in a humidified incubator. Subsequently to infection of macrophages for 24 hours, activity of (S)-HPMPAon intracellular L. donovani was evaluated by exposing
infected macrophages to concentrations up to 30 pd
ml for 5 days. Finally, cells were fixed and Giemsa
stained. The percentage of L. donovani-infected
macrophages was determined microscopically.
In vivo drug susceptibility test
Female Swiss ICR mice, weighing 25-3 5 g, were
used for the in vivo drug tests. Mice were inoculated intraperitoneally (i.p.) each with I x 1 0 5
Tropical Medicine and International Health
R Kaminsky et 01.
1996
(S)-HPMPA a trypanocidal purine analogue
T. 6. rhodesiense, T. 6. brucei STIB 950, and T. congolense, and treatment was administered 24 hours
after inoculation. When multiple treatment was
administered, the first treatment was given 24 hours
after inoculation. (S)-HPMPAwas administered i.p.
or orally at the appropriate concentration. The tail
blood of mice was examined for the presence of
trypanosomes 2-3 times a week for a total of 60
days using the wet blood film technique. Mice were
considered cured when no trypanosomes were
detected during the observation period. A similar
procedure was used to evaluate (S)-HPMPA against
T. 6. gambiense except that Mastomys natalensis
were pretreated with 20 mg/kg cyclophosphamide
24 hours prior to inoculation with T. 6. gambiense.
Parasitaemia in Mastomys natalensis was followed
by examination of tail blood using the haematocrit
centrifuge technique (Woo 1970).
To evaluate activity of (S)-HPMPA against late
stage central nervous system (CNS) infections, the
rodent late stage model according to Jennings and
Gray (1983) was used. Briefly, mice were inoculated
i.p. with T. 6. 6rucei GVR-35 and the infection was
allowed to develop for 21 days before treatment was
initiated. Infected mice were treated with (S)HPMPA and the parasitaemia was followed by
examination of mouse tail blood. Diminazene
aceturate (Berenil) was used as a control; after the
initial clearing of the peripheral blood from trypanosornes, those parasites which escaped drug treatment
while located in the CNS re-invaded the blood system and caused relapses. In this late stage model,
animals were considered cured if they survived and
were aparasitaemic for at least 180 days after the
end of the treatment.
Results
The in vitro antitrypanosomal activity of various
compounds against T. 6. rhodesiense STIB 900 is
summarized in Table I. (S)LHPMPA was the most
active compound which eliminated T. 6. rhodesiense
completely when trypanosomes were exposed for 72
hours to 0.22 pg/ml. The results of experiments on
the activity of (S)-HPMPA against various trypanosome species and against mammalian cells is summarized in Table 2. Human pathogenic T. b. rhodesiense and T. 6. gambiense were eliminated with 0.22
259
VOLUME T NO. 2 PP 2 j 5-263 APRIL
@ 1996 Blackwell Science Ltd
Table I In vitro activity of HPMP-analogues against T.
rhodesiense (STIB 9 0 0 )
(S)-HPMPA
(S)-cHPMPA
(R)-HPMPA
(SJ-HPMPG
(SJ-HPMPDAP
0.22
0.010
3.21
0.138
6.66
4.38
0.25
> 20
20
6.66
6.
0.028
0.248
7
6.55
0.72
Trypanosomes were exposed for 7 2 hours to the compound. MIC values (minimum inhibitory concentrations)
and M T C values (maximum tolerated concentrations) were
determined microscopically. EC,, values were determined
using a fluorescence assay. All values represent average
values of at least two experiments each performed in
triplicate. (For details cf. Materials and methods.)
Table 2 In vitro activity of (S)-HPMPA against various
haemoflagellate species and mammalian cells
Species/cells
T. 6. rhodesiense (STIB 9 0 0 )
T. 6. gambiense (STIB 930)
T. b. 6rucei (STIB 9 2 0 )
T. 6. brticei (mdr) (STIB 950)
T. congotense (STIB 910)
T. evunsi (STIB 806)
T. crtizi (Y/Br strain)
L. donovuni (L82)
Human carcinoma cells (HT-29)
Bovine aorta endothelial cells (BAE)
MIC
(pg/ml)
0.22
0.61
0.3
Selectivity
index
I 3 600
4900
T 0000
0.44
6800
0.23
o.ri
>50
13000
27300
>30
<TOO
3000
n.a.
n.a.
I800
<60
Parasites and mammalian cells were exposed to
(S)-HPMPA for 72 hours. MIC values (minimum inhibitory
concentrations) were determined microscopically. All MIC
values represent average values of a minimum of two
experiments each performed in triplicate. The selectivity
index was determined by dividing the MIC of HT-29 cells
by the MIC of the trypanosome species or by L. donovani,
respectively. (For details cf. Materials and methods.)
n.a., Not applicable.
and 0.61 pg/ml, respectively. Within or below this
concentration range all other animal pathogenic
trypanosome stocks, including the multidrugresistant STIB 9 5 0 , were eliminated. The MIC values
for mammalian cells were 8000-13 ooo higher than
the MIC value for T. 6. rhodesiense STIB 900. The
Tropical Medicine and International Health
VOLUME 1 N O 2 PP Z j
5-263 APRIL 1996
R Kaminsky et ol. (S)-HPMPA a trypanocidal purine analogue
Trypanosoma
species
Dosage'
(mglkg)
Route
0
(STIB 900)
1x5
2
10(,d)
i.p.
i.p.
1.p.
i.p.
i.p.
4 x
TOO
oral
10
i.p.
i.p.
i.p.
j(zd)
5(7d)
T. b. brucei mdr
0
(STIB 9 5 0 )
I
x I0
x
10(,d)
2
4 x 5
T. congolense
(STIB 9 I 0 )
x
10
1°(7d)
4 x 5
Late stage model
T. b. brucei
(GVR 35)
diminazene
4
5(4d)
2
IO(7d)
016
015
315
415
515
6-13
20-4 3
21, 59; >60
27, 44, 5 0 ; >60
24; >60
> 60
214
45; >60
012
6, 7
5 0 ; >60
> 60
> 60
115
415
515
415
012
> 60
i.p.
i.p.
i.p.
015
015
n.a.
n.a.
n.a.
1.p.
1.p.
i.p.
014
> 60
014
n.a.
n.a.
0
I
11s
014
in vitro selectivity index (MIC of HT-zq cells
divided by MIC of parasites) was in the range
4900-27 ooo for the extracellular trypanosomes.
The MIC values for the intracellular parasites
T. cruzi and for L. donovani were higher than 50
and 30 pg/ml, respectively, which were the highest
concentrations tested.
The results of experiments on the antitrypanosoma1 activity of (S)-HPMPA in rodents are summarized in Table 3. Two administrations (i.p.) of
10 mg/kg (bodyweight) 7 days apart were sufficient
to cure all mice infected with T. .6. rhodesiense or T.
6. 6rucei. Using the same dosage, it was not possible
to cure all mice infected with T. congolense or
Mastomys infected with T. 6. gambiense. Oral
administration of 4 x IOO mg/kg cured z of 4 mice
infected with T.6. rhodesiense STIB 900, while
those mice which were not cured had an extended
survival time when compared with untreated mice.
260
0 1996 Blackwell Science Ltd
Survival
(days)
~-
-.
T. 6. rhodesiense
I
Cured1
treated
Table 3 Activity of (S)-HPMPA
against various trypanosome species
in rodents a t various dosages.
Treatment commenced 24 hours after
infection except for T. 6. brucei
GVR-3 j for which treatment started
21 days post infection
' Interval between administrations
for multiple treatment is indicated in
parenthesis.
Significant weight loss of all
Mastomys natalensis during
treatment.
i.p., Intraperitoneally.
n.a., Not applicable.
When mice infected with T. 6. 6rucei GVR-35
were treated i.p. on days 21-24 with 4 x 5 mglkg or
on days 21 and 28 with 2 x 10mg/kg (S)-HPMPA
relapses occurred, after initial clearing of the blood,
4-16 days following treatment. None of the 8
infected mice was cured using the two dosage
schedules.
Discussion
The results obtained in the series of HPMP analogues (Table I ) confirm earlier findings on the
antiparasitic activity of phosphonylmethoxyalkylpurines against T. 6. 6rucei (Kaminsky eta[. 1994b).
(S)-HPMPA was the most active compound against
T. b. rhodesiense STIB 900, while the other basemodified HPMPA analogues expressed only minor
activities. Such a pattern is in accordance with
data on the activity of these compounds against
Tropical Medicine and International Health
VOLUME I NO. 2 PP
255-263
APRIL
1996
R Kaminsky et ol. (S)-HPMPA: a trypanocidal purine analogue
Plasmodium sp. (de Vries et al. 1991).However, in
view of the fact that (S)-HPMPA and its phosphates
are very potent inhibitors of purine nucleotide
phosphorylase, thus severely disturbing the purine
cell pool (Sediva et al. 1991),we have no explanation of why this nucleotide analogue did not express
antiprotozoal activity.
The antitrypanosomal activity of (S)-HPMPA in
vitro was similar against all extracellular trypanosomes with the highest activity against T. evansi.
The multidrug-resistant T. b. brucei clone STIB 9 5 0
was also susceptible to (S)-HPMPA which suggests a
different mode of action of (S)-HPMPA compared to
diminazene, isometamidium and quinapyramine, the
most commonly used trypanocides against animal
pathogenic trypanosomes. The (R)-enantiomer [ ( R ) HPMPA] is devoid of antitrypanosomal activity.
This enantiospecificity might be explained by an
enantiospecific activation of the (S)-HPMPA only by
the nucleotide kinases (e.g. AMP/dAMP kinase
(Merta et al. 1992;Balzarini et al. 1993)).The cyclic
prodrug of (S)-HPMPA, (S)-cHPMPA, showed an
approximately tenfold lower antitrypanosomal
activity compared to the parent drug. It is consistent
with the hypothesis that the increased activity of the
prodrug (e.g. Baba et al. 1987) is due to its better
transport over the cellular membrane compared to
the open form, a quality that does not apply to the
effect upon the extracellular trypanosomes.
The intracellular haemoflagellates T. cruzi and L.
donovani were not affected in vitro at the highest
concentration tested. This is in contrast to the
reported inhibition of intraerythrocytic P. falciparum
and P. berghei by (S)-HPMPA (de Vries et al. 1991).
Possibly, the slow endocytosis-mediated uptake of
(S)-HPMPA by mammalian cells (Palu et al. 1991)is
concennot sufficient to accumulate trypanocidal
._
trations in the host cells. Furthermore, (S)-HPMPA
needs to be diphosphorylated inside the cell to
become active (Votruba et al. 1987; Merta et al.
1992).It remains to be established whether in our
case the observed lack of activity against intracellular parasites was due to the slow uptake of the
drug into the host cell, to the low efficacy of the
activation process, to a hindered uptake of diphosphorylated (S)-HPMPA by the intracellular haemoflagellates or, finally, to the lack of its inhibitory
activity at the target.
26 I
0 1996 Blackwell Science Ltd
It has been established that mammalian cells are
hardly affected by (S)-HPMPA (De Clercq et al.
1989).This could be confirmed by our investigations
where (S)-HPMPA expressed a high selective toxicity
against extracellular trypanosomes resulting in an in
vitro selectivity index of 4900-27 300. This selectivity index is in the range of presently used veterinary
trypanocides like diminazene (16000-33 000) and is
higher than the selectivity index of melarsoprol (800I ~ O O )the
, drug of choice for late stage sleeping sickness (Kaminsky & Schmid, unpublished results). The
high in vitro selectivity index prompted infectiontreatment experiments in rodent models. Acute infections of mice caused by T. b. rhodesiense STIB 900
and T.b. brucei mdr (STIB 950)were cleared with z
administrations of 10 mg/kg, but chronic infections
caused by T. b. gambiense and T. congolense could
not be cured by this dosage regimen.
It was not possible to achieve cure of infected mice
with central nervous system (CNS) involvement using
the described dosage regimen. However, the fact that
(S)-HPMPA reduced mortality in mice infected intracerebrally with herpesviruses (De Clercq et al.
1989) suggests that the drug is able to cross the
blood-brain barrier. This ability is a prerequisite for
any new drug aimed against human trypanosomiasis
because T. b. rhodesiense and T. b. gambiense
are able to invade the CNS. The availability of
pharmacokinetic data would show to what extent
(S)-HPMPA or properly designed prodrugs are able
to cross the blood-brain barrier, and would help to
improve the application regimen, resulting in a better
bioavailability of the drug in the CNS. Such studies
are planned in a vervet monkey model.
Acknowledgements
This work was financially supported by a grant of
the Swiss Government for the European COST
Project 8 I 5 ACRIVAL (Antiparasitic Chemotherapy)
and by a grant of T D W H O . We gratefully
acknowledge technical assistance of Ms R. Nuesch
and M r R. Trebo for work with rodent models.
References
Ainanshe OA, Jennings FW & Holmes PH ( 1 9 9 ~Isolation
)
of drug-resistant strains of Trypanosoma congolense
Tropical Medicine and International Health
R Karninsky et a/.
255-263
APRIL
1996
(S)-HPMPA a trypanocidal purine analogue
from the lower Shabelle region of Southern Somalia.
Tropical Animal Health Production 24, 65-73.
Baba M, Mori S, Shigeta S & De Clercq E (1987) Selective
inhibitory effect of (S)-9-(j-hydroxy-z-phosphonylmethoxypropy1)adenine and 2'-nor-cyclic GMP on adenovirus
replication in uitro. Antimicrobial Agents and
Chemotherapy 3 I , 33 7-339.
Bacchi CJ, Nathan HC, Livingston T et al. (1990)
Differential susceptibility to DL-a-difluoromethylornithine
in clinical isolates of Trypanosoma brucei rhodesiense.
Antimicrobial Agents and Chemotherapy 34, 1183-1188.
Bacchi CJ, Sufrin JR, Nathan H C et al. (1991)
5'-alkyl-substituted analogs of 5'-methylthioadenosine as
trypanocides. Antimicrobial Agents and Chemotherapy
3.5, 1315-1320.
Bales JD, Harrison SM, Mbwabi DL & Schechter PJ
(1989) Treatment of arsenical refractory Rhodesian
sleeping sickness in Kenya. Annals of Tropical Medicine
and Parasitology 83 Suppl. I , 111-114.
Baltz T, Baltz D, Giroud Ch & Crockett J (1985)
Cultivation in a semi-defined medium of animal infective
forms of Trypanosoma brucei, T. equiperdum, T. evansi,
T. rhodesiense and T. gambiense. EMBO]ournal4,
1273-1 277.
Balzarini J, Holy A, Jindrich J et al. (1993) Differential
anti-retrovirus effects of the (S)-and (R)-enantiomers of
acyclic nucleoside phosphonates. Potent and selective in
uitro and in vivo anti-retrovirus activity of the
(R)-9-(~-phosphonomethoxypropyl)
derivatives of
heterocyclic bases. Antimicrobial Agents and
Chemotherapy 37, 332-338.
Codjia V, Mulatu W, Majiwa P A 0 et al. (1993)
Epidemiology of bovine trypanosomiasis in the Ghibe
valley, southwest Ethiopia. 3 Occurrence of populations
of Trypanosoma congolense resistant to diminazene,
isometamidium and homidium. Acta Tropica 5 3 ,
151-163.
Davies MJ, Ross AM & Gutteridge WE (1983) The
enzymes of purine salvage in Trypanosoma cruzi,
Trypanosoma brucei and Leishmania mexicana.
Parasitology 87, 211-217.
De Clercq E, Holy A, Rosenberg I, Sakuma T, Balzarini J
& Maudgal PC (1986) A novel-selective broad-spectrum
anti-DNA virus agent. Nature 323, 464-467.
De Clercq E, Holj. A & Rosenberg I (1989) Efficacy of
phosphonylmethoxyalkyl derivatives of adenine in
experimental Herpes simplex virus and Vaccinia virus
infections in viuo. Antimicrobial Agents and
Chemotherapy 33, 185-191.
de Vries E, Scam JG, Franssen FFJ et al. (1991) Inhibition
of the growth of Plasmodium falciparum and
262
VOLUME I N O 2 PP
0 I996 Blackwell Science Ltd
Plasmodium berghei by the DNA polymerase inhibitor
HPMPA. Molecular and Biochemical Parasitology 47,
43-50,
Fish WR, Looker DL, Marr JJ & Berens RI, (1982)Purine
metabolism in the bloodstream forms of Trypanosomu
gambiense and Trypanosoma rhodesiense. Biochimicu et
Biophysica Acta 719, 223-231.
Fish WR, Marr JJ, Berens RL et al. (1985) Inosine analogs
as chemotherapeutic agents for African trypanosomes:
metabolism in trypanosomes and efficacy in tissue
culture. Antimicrobial Agents and Chemotherapy 27(1),
33-36.
Geigy R & Kauffmann M (1973) Sleeping sickness survey
in the Serengeti area, Tansania. Acta Tropica 30, 14-48.
Gottlieb M (1985) Enzyme regulation in a trypanosomatid:
effect of purine starvation on levels of 3'-nucleotidase
activity. Science 227, 72-74.
Hassan H F & Coombs G H (1986) A comparative study of
purine- and pyrimidine-metabolising enzymes of a range
of trypanosomatids. Comparative Biochemistry and
Physiology 84B (z),217-233.
Holj. A (1993) Syntheses of enantiomeric
N-(3-hydroxy-2-phosphonomethoxypropyl)
derivatives of
purine and pyrimidine bases. Collection of Czechoslovak
Chemical Communications 5 8, 649-674.
Huber W & Koella J C (1993) A comparison of three
methods of estimation EC,, in studies of drug resistance
of malaria parasites. Acta Tropica 55, 257-261.
Iten M, Matovu E, Brun R & Kaminsky R (1995) Innate
lack of susceptibility of Ugandan Trypanosoma brucei
rhodesiense to DL-a-difluoromethylornithine (DFMO).
Tropical Medicine and Parasitology 46, 190-194.
Jennings FW & Gray AR (1983) Relapsed parasitemia
following chemotherapy of chronic Trypanosoma brucei
infections in mice and its relationship to cerebral
trypanosomes. Contributions to Microbiology and
Immunology 7, 147-154.
Kaminsky R, Chuma F & Wasike RPN (1994a) Time-dose
response of Trypanosoma congolense bloodstream forms
to diminazene and isometamidium. Veterinary
Parasitology 52, 235-242.
Kaminsky R, Chuma R & Zweygarth E (1989)
Trypanosoma brucei brucei: expression of drug resistance
in uitro. Experimental Parasitology 69, 28 1-289.
Kaminsky R & Zweygarth E (1989) Feeder layer-free in
vitro assay for screening antitrypanosomal compounds
against Trypanosoma brucei brucei and T. b. evansi.
Antimicrobial Agents and Chemotherapy 33, 881-885.
Kaminsky R, Zweygarth E & De Clercq E (199413)
Antitrypanosomal activity of
phosphonylmethoxyalkylpurines. ]ournal of Parasitology
80, 1026-1030.
Tropical Medicine and International Health
VOLUME I NO. 2 PP
255-263
APRIL
1996
R Kaminsky et a/. (S)-HPMPA: a trypanocidal purine analogue
Kuzoe FAS (1993)Current situation of African
trypanosomiasis. Acta Tropica 54, 153-162.
Lun Z-R, Allingham R, Brun R & Lanham SM (1992)The
isoenzyme characteristics of Trypanosoma evansi and T.
equiperdum from domestic stocks in China. Annuls of
Tropical Medicine and Parasitology 86, 33 3-340.
Mehlitz D, Brinkmann U & Heller R ( 1 9 8 1 )
Epidemiological studies o n the animal reservoir of
gambiense sleeping sickness. Part I Review of literature
and description of the study areas. Tropenmedizin und
Parasitenkunde 32, 129-133.
Merta A, Votruba I, Jindrich J et al. (1992)
Phosphorylation of y - ( 2-phosphonomethoxyethy1)adenine
and y-(S)-(3-hydroxy-z-phosphonomethoxypropyl)
adenine by AMP (dAMP) kinase from LIZIO cells.
Biochemical Pharmacology 44, 2067-2077.
Neal RA & Croft SL (1984)An in vitro system for
determining the activity of compounds against the
intracellular amastigote form of Leishmania donovani.
Journal of Antimicrobial Agents and Chemotherapy 14,
463-4 75 .
Obexer W, Schmid C & Brun R (1995) A novel in vitro
screening assay for trypanocidal activity using the
fluorescent dye BCECF-AM. Tropical Medicine and
Parasitology 46, 45-48.
Palu G, Stefanelli S, Rassu M, Parolin C, Balzarini J & De
Clercq E (1991) Cellular uptake of
phosphonylmethoxyalkylpurine derivatives. Antiviral
Research 16, 115-119.
263 0 1996 Blackwell Science Ltd
Pepin JF, Milord F, Guern C, Mpia B, Ethier L & Mansin
D (1989) Trial of prednisolone for prevention of
melarsoprol-induced encephalopathy in gambiense
sleeping sickness. Lancet i, 1246-1250.
Rosenberg I & Holy A (1987) Synthesis of potential
prodrugs and metabolites of
y - ( S ) - (3-hydroxy-2-phosphonylmethoxypropyl)adenine.
Collection of Czechoslovak Chemical Communications
5 2 , 2792-2800.
Sedivi K, Ananiev AV, Votruba I, Holj. A & Rosenberg I
(1991) Inhibition of purine nucleoside phosphorylase by
phosphonylmethyl analogues of nucleotides.
International Journal of Purine and Pyrimidine Research
2, 35-40,
Votruba I, Bernaerts R, Sakuma T et al. (1987)
Intracellular phosphorylation of broad-spectrum
anti-DNA virus agent
( S ) - y-( j-hydroxy-2-phosphonyImethoxypropy1)adenine
and inhibition of viral DNA synthesis. Molecular
Pharmacology 32, 524-529.
Woo PTK (1970)The haematocrit centrifuge technique for
the diagnosis of African trypanosomiasis. Acta Tropica
27, 384-386.