Novel trypanocidal analogs of 5′-(methylthio)-adenosine

Article abstract of DOI:10.1128/AAC.00480-07

The purine nucleoside 5′-deoxy-5′-(hydroxyethylthio)-adenosine (HETA) is an analog of the polyamine pathway metabolite 5′-deoxy-5′- (methylthio)-adenosine (MTA). HETA is a lead structure for the ongoing development of selectively targeted trypanocidal agents. Thirteen novel HETA analogs were synthesized and examined for their in vitro trypanocidal activities against bloodstream forms of Trypanosoma brucei brucei LAB 110 EATRO and at least one drug-resistant Trypanosoma brucei rhodesiense clinical isolate. New compounds were also assessed in a cell-free assay for their activities as substrates of trypanosome MTA phosphorylase. The most potent analog in this group was 5′-deoxy-5′-(hydroxyethylthio)-tubercidin, whose in vitro cytotoxicity (50% inhibitory concentration [IC₅₀], 10 nM) is 45 times greater than that of HETA (IC₅₀, 450 nM) against pentamidine-resistant clinical isolate KETRI 269. Structure-activity analyses indicate that the enzymatic cleavage of HETA analogs by trypanosome MTA phosphorylase is not an absolute requirement for trypanocidal activity. This suggests that additional biochemical mechanisms are associated with the trypanocidal effects of HETA and its analogs. Copyright

Full text of DOI:10.1128/AAC.00480-07

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Jan. 2008, p. 211–219
0066-4804/08/$08.00ϩ0 doi:10.1128/AAC.00480-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.
Vol. 52, No. 1
Novel Trypanocidal Analogs of 5Ј-(Methylthio)-Adenosineᰔ†
Janice R. Sufrin,1,2* Arthur J. Spiess,1 Canio J. Marasco, Jr.,1‡
Donna Rattendi,3 and Cyrus J. Bacchi3,4
Department of Pharmacology and Therapeutics, Roswell Park Cancer Institute, Buffalo, New York 142631; Program of Molecular
Pharmacology and Cancer Therapeutics, Roswell Park Division, State University of New York at Buffalo, Buffalo,
New York 142632; and Haskins Laboratory3 and Department of Biological and Health Sciences,4
Pace University, New York, New York 10038-1502
Received 6 April 2007/Returned for modification 26 June 2007/Accepted 13 October 2007
The purine nucleoside 5؅-deoxy-5؅-(hydroxyethylthio)-adenosine (HETA) is an analog of the polyamine
pathway metabolite 5؅-deoxy-5؅-(methylthio)-adenosine (MTA). HETA is a lead structure for the ongoing
development of selectively targeted trypanocidal agents. Thirteen novel HETA analogs were synthesized and
examined for their in vitro trypanocidal activities against bloodstream forms of Trypanosoma brucei brucei LAB
110 EATRO and at least one drug-resistant Trypanosoma brucei rhodesiense clinical isolate. New compounds
were also assessed in a cell-free assay for their activities as substrates of trypanosome MTA phosphorylase. The
most potent analog in this group was 5؅-deoxy-5؅-(hydroxyethylthio)-tubercidin, whose in vitro cytotoxicity
(50% inhibitory concentration [IC50], 10 nM) is 45 times greater than that of HETA (IC50, 450 nM) against
pentamidine-resistant clinical isolate KETRI 269. Structure-activity analyses indicate that the enzymatic
cleavage of HETA analogs by trypanosome MTA phosphorylase is not an absolute requirement for trypanocidal
activity. This suggests that additional biochemical mechanisms are associated with the trypanocidal effects of
HETA and its analogs.
African sleeping sickness is endemic to vast areas of sub-
Saharan Africa. The World Health Organization estimates that
55 million people in 35 African countries are at risk for con-
tracting the disease, which is invariably fatal to untreated in-
dividuals (29). An alarming resurgence of African sleeping
sickness in Sudan and other parts of Central Africa has created
great concern among African governments and international
health care agencies. Pentamidine and the organoarsenicals
(e.g., melarsoprol and melarsen), which have been used to
treat African sleeping sickness for many years, are highly toxic
and have led to drug-resistant clinical strains that are refrac-
tory to chemotherapy (9, 29). The drug ␣-difluoromethylorni-
thine (ornidyl, eflornithine), which was approved by the FDA
in 1990 for treatment of the disease, is costly to prepare, cumber-
some to administer, and ineffective against the disease forms
prevalent in East Africa (29). At present, DB289, a prodrug of the
pentamidine analog 2,5-bis(4-amidinophenyl)furan, is the only
potential new treatment undergoing clinical trials for early-
stage human African sleeping sickness (41). New agents for the
treatment of African sleeping sickness are urgently needed (9,
14, 29). Ideally, these should be highly effective in eradicating
host-dwelling parasites, nontoxic to the host, inexpensive to
produce, and easy to administer to patients in rudimentary
medical care settings.
vided effective targets for the development of trypanocidal
agents (4). Difluoromethylornithine, the agent used to treat
late-stage African sleeping sickness, is an inhibitor of ornithine
decarboxylase (5, 6). Another polyamine pathway enzyme, 5Ј-
deoxy-5Ј-(methylthio)-adenosine phosphorylase (MTA-Pase),
has been studied by our laboratories as a potential trypanocidal
target. 5Ј-Deoxy-5Ј-(methylthio)-adenosine MTA, the nucleo-
side by-product of spermidine synthesis in trypanosomes, is
cleaved by MTA-Pase to generate methylthioribose-1-phos-
phate (MTRP) and adenine (Fig. 1). MTRP is recycled to
methionine, while adenine is redirected to nucleoside synthe-
sis. Trypanosomes lack de novo purine biosynthetic pathways
(11). In addition, they have continuously high demands for
S-adenosylmethionine (AdoMet), the methionine metabo-
lite that serves as a polyamine precursor (22). MTA-Pase
plays a critical role in these parasites by enabling the salvage
of adenine and methionine, both of which are essential for
trypanosome survival (11, 22). Moreover, MTA-Pase pre-
sents a selective chemotherapeutic target in African try-
panosomes, based on a significant difference in the substrate
activities of the MTA analog, 5Ј-(hydroxyethylthio)-adeno-
sine (HETA), toward mammalian and trypanosome forms of
MTA-Pase: the trypanososme form of MTA-Pase (Tryp-
MTA-Pase) rapidly metabolizes HETA, whereas the mam-
malian enzyme is comparably ineffective (8, 44). Although
the mechanism of HETA’s toxicity is not known, its rapid
breakdown, once it is inside trypanosomes, is an indication
that these effects are not caused by the inhibition of Tryp-
MTA-Pase. More likely, HETA’s toxicity is related to its
ability to interfere with methionine recycling. This is sup-
ported by the finding that HETA’s in vitro trypanocidal
effects can be reversed by the concomitant administration of
methionine or 2-keto-4-methylthio-butyrate, the immediate
precursor of methionine salvage from MTA (8). HETA’s
Polyamine biosynthetic pathways in trypanosomes have pro-
* Corresponding author. Mailing address: Grace Cancer Drug Cen-
ter, Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo,
NY 14263. Phone: (716) 845-4582. Fax: (716) 845-4445. E-mail: janice
.sufrin@roswellpark.org.
† Supplemental material for this article may be found at http://aac
.asm.org/.
‡ Present address: Department of Mathematics and Natural Sciences,
D’Youville College, 320 Porter Avenue, Buffalo, NY 14201.
ᰔ Published ahead of print on 22 October 2007.
211

212
SUFRIN ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
FIG. 1. Pathways of methionine salvage from MTA. (Adapted from reference 40 with permission.)
toxicity may also be associated, in part, with the generation
of a toxic methionine analog, as was observed in Klebsiella
pneumonia (32). Other key biochemical properties that con-
tribute to HETA’s trypanocidal potency include its rapid,
carrier-mediated accumulation inside trypanosomes via the
P2 purine transporter and a novel AdoMet transporter dis-
covered in these parasites (19, 20, 22). HETA was observed
to induce elevations in AdoMet and MTA levels in trypano-
somes (3) and to inhibit trypanosome protein methylation
reactions (21).
trypanosomes, which provided the initial basis for its potential
selectivity, was supported by results from in vivo studies dem-
onstrating its curative effects in mice infected with Trypano-
soma brucei brucei and T. brucei rhodesiense strains and the
absence of host toxicity (7, 8). These in vivo studies strongly
validated the designation of HETA as a lead compound for
further analog development. Our continuing efforts to identify
more potent structural analogs of our lead compound, de-
scribed in this study, have centered on the synthesis and in vitro
antitrypanosomal screening of novel purine- and ribose-mod-
ified analogs of HETA.
The preferential metabolism of HETA by MTA-Pase in
TABLE 1. Structures of exocyclic purine-modified MTA analogs
Compound no.a
Compound
X
Ia
Ib
Ic
Id
Ie
5؅-Deoxy-5؅-(hydroxyethylthio)-inosine (HETI)
5Ј-Deoxy-5Ј-(hydroxyethylthio)-adenosine (HETA)
5؅-Deoxy-5؅-(methylthio)-6-fluoropurine riboside (HET-FPR)
5؅-Deoxy-5؅-(hydroxyethylthio)-6-methylpurine riboside (HET-MPR)
5؅-Deoxy-5؅-(hydroxyethylthio)-nebularine (HETN)
OH
NH2
F
CH3
H
IIa
IIb
5Ј-Deoxy-5Ј-(hydroxyethylthio)-2-aminoadenosine (2-amino-HETA)
5Ј-Deoxy-5Ј-(hydroxyethylthio)-2-fluoroadenosine (2-fluoro-HETA)
NH2
F
a Novel HETA analogs are distinguished in boldface.

VOL. 52, 2008
NOVEL TRYPANOCIDAL ANALOGS OF 5Ј-METHYLTHIOADENOSINE
213
TABLE 2. Structures of heterocyclic ring-modified MTA analogs
Compound no.a
Compound
R
IIIa
IIIb
IV
V
VI
5؅-Deoxy-5؅-(hydroxyethylthio)-tubercidin (HETT)
5Ј-Deoxy-5Ј-(methylthio)-tubercidin (MTT)
5؅-Deoxy-5؅-(hydroxyethylthio)-8-aza-adenosine (8-aza-HETA)
5؅-Deoxy-5؅-(hydroxyethylthio)-ribavirin (HET-RV)
5Ј-Deoxy-5Ј-(hydroxyethylthio)-ethenoadenosine (etheno-HETA)
HOCH2CH2
CH3
a Novel HETA analogs are distinguished in boldface.
MATERIALS AND METHODS
5Ј-(aminopropylthio)-adenosine (APTA), 5Ј-(methylamino)-adenosine (MAA),
5Ј-(methylthio)-7-deaza-adenosine [5Ј-(methylthio)-tubercidin (MTT)], and 5Ј-(hy-
droxyethylthio)-immucillin A (HET-immucillin A), were synthesized by previously
published procedures (8, 13, 16, 26, 27, 30, 38, 39, 43–45).
MTA analogs. The structures of the MTA analogs that were evaluated for
trypanocidal activity in this study, as well as in previous studies (3, 8, 43), are depicted
in Tables 1 to 4. The structures of the 13 newly synthesized analogs are highlighted
by boldface type in all tables; the methods developed for their chemical synthesis
appear in the supplemental material (1, 10, 23, 25, 31, 33, 34, 35, 37, 47). Also
included in the supplemental material are detailed methods for the synthesis of
2-amino-HETA, 2-fluoro-HETA, and 2Ј-deoxy-HETA, analogs whose trypanocidal
effects were reported previously (3, 43). Confirmation of the structures of the newly
synthesized analogs was accomplished by 1H nuclear magnetic resonance (NMR)
spectroscopy. The purity of the compounds was assessed by thin-layer chromatog-
raphy and NMR spectroscopy. Other nucleosides shown in Tables 1 to 4, i.e.,
5Ј-(ethylthio)-adenosine (ETA), 5Ј-(propylthio)-adenosine (PTA), HETA, 2Ј,3Ј-
seco-HETA, 5Ј-(hydroxypropylthio)-adenosine (HPTA), 5Ј-(monofluoroethylthio)-
adenosine (MFETA), 5Ј-(monofluoropropylthio)-adenosine (MFPTA), 5Ј-(car-
boxymethylthio)-adenosine (CMTA), 5Ј-(aminoethylthio)-adenosine (AETA),
Trypanosome strains. Trypanosoma brucei brucei LAB 110 EATRO was ob-
tained from the late William Trager of the Rockefeller University (46). Clinical
isolates of T. brucei rhodesiense (KETRI 243 and KETRI 269) were obtained
from A. R. Njogu of the Kenya Trypanosomiasis Research Institute (KETRI).
KETRI 243 is resistant to pentamidine and melarsoprol, and KETRI 269 is
resistant to pentamidine (2). KETRI 243-As10-3 is a highly melarsen- and di-
amidine-resistant clone of KETRI 243 (2). The bloodstream forms of the strains
were adapted to grow under axenic conditions in Iscove’s modified Dulbecco
medium with hypoxanthine at 1 ␮M and 20% horse serum instead of synthetic
serum (HMI-18 medium) (24).
MTA phosphorylase enzyme assays. Cell-free enzyme preparations were ob-
tained from bloodstream trypanosomes by a previously published procedure (8).
TABLE 3. Structures of exocyclic (5Ј)-ribose-modified MTA analogs
Compound no.a
Compound
R
VIIa
VIIb
VIIc
VIId
VIIe
VIIf
VIIg
5Ј-Deoxy-5Ј-(methylthio)-adenosine (MTA)
CH3
5Ј-Deoxy-5Ј-(ethylthio)-adenosine (ETA)
CH3CH2
5Ј-Deoxy-5Ј-(monofluoroethylthio)-adenosine (MFETA)
5Ј-Deoxy-5Ј-(propylthio)-adenosine (PTA)
5Ј-Deoxy-5Ј-(hydroxypropylthio)-adenosine (HPTA)
5Ј-Deoxy-5Ј-(monofluoropropylthio)-adenosine (MFPTA)
5Ј-Deoxy-5Ј-(carboxymethylthio)-adenosine (CMTA)
FCH2CH2
CH3CH2CH2
HOCH2CH2CH2
FCH2CH2CH2
CH3OCCH2
\
O
VIIh
VIIj
VIII
5Ј-(Aminoethylthio)-5Ј-deoxy-adenosine (AETA)
5Ј-(Aminopropylthio)-5Ј-deoxy-adenosine (APTA)
5؅-Deoxy-5؅-(methylamino)-adenosine (MAA)
H2NCH2CH2
H2NCH2CH2CH2
CH3NH
a Novel HETA analogs are distinguished in boldface.

214
SUFRIN ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
TABLE 4. Structures of ribofuranose ring-modified MTA analogs
Compound no.a
Compound
X
Y
IXa
IXb
IXc
X
XIa
XIb
XII
XIII
2Ј-Deoxy-HETA
OH
H
H
H
OH
H
3؅-Deoxy-HETA
2؅,3؅-Dideoxy-HETA
2Ј,3Ј-seco-HETA
(S)-9-[2-Hydroxy-3-(hydroxyethylthiopropyl)]adenine [(S)-HHETP-adenine]
(R)-9-[2-Hydroxy-3-(hydroxyethylthiopropyl)]adenine [(R)-HHETP-adenine]
9-[(2-Hydroxyethylthioethoxy)methyl]adenine (HET-EMA)
5Ј-HET-immucillin A
a Novel HETA analogs are distinguished in boldface.
Briefly, bloodstream trypanosomes were harvested from rats. Cell lysates were
frozen-thawed in dry ice-methanol for three cycles and centrifuged at 5,000 ϫ g
for 10 min, and endogenous unbound phosphate was removed by dialysis for 1 h
(0.05 M Tris-HCl, 0.1 mM disodium EDTA, 1 mM 2-mercaptoethanol, pH 7.4).
Trypanosome extracts were frozen and stored at Ϫ70°C. MTA analogs were
assayed for their ability to act as substrates for Tryp-MTA-Pase with and without
added phosphate in the reaction mixture. Enzyme assays were based upon the
phosphorolysis of MTA to yield adenine. The conversion of adenine to 2,8-
dihydroxyadenine in the presence of commercial xanthine oxidase was measured
TABLE 5. Activities of MTA analogs as substrates for
Trypanosoma brucei brucei MTA-Pase
% of control activity
Compound group
and no.a
Substrate
With 50 mM
Without
Ϫ
Ϫ
PO4
PO4
Purine-modified
at 305 nm (change in E ϭ 15.5 ϫ 10 3
M
Ϫ1 cmϪ1). The reaction mixtures (1-ml
analogs
IIb
VIIa
Ib
2-Fluoro-HETA
MTA
148b
100c
96
61
56
42
29b
11
2
106b
100d
100
70
62
60
18b
4
final volume in quartz cuvettes) contained 0 or 50 mM potassium phosphate (pH
7.4), xanthine oxidase (1,000 U; type III; Sigma Chemical Co., St. Louis, MO),
and 100 ␮l of trypanosomal extract. MTA or its substrate analog (25 to 500 ␮M)
was added to start the reaction. Normally, 200 ␮M MTA or its analog was
saturating. The increase in absorbance over 1 h was measured in a recording
spectrophotometer (DU7; Beckman) with a temperature-controlled cuvette
holder (37°C). The rates were linear for the 1-h period (18).
Determination of in vitro antitrypanosomal activity. Drug studies were done
in duplicate in 24-well plates (1 ml per well) with final inhibitor concentra-
tions of 0.1, 1, 10, 25, and 100 ␮M. After 48 h, the number of parasites was
determined in a Z1 Coulter counter, and the approximate range of activity
was determined. The 50% inhibitory concentrations (IC50s) were then deter-
mined from additional studies with a narrower range of inhibitor concentra-
tions. Inhibitors with Ͻ50% inhibition at 100 ␮M were considered inactive
and their IC50 values were not further determined. Analogs were dissolved in
water or dimethyl sulfoxide. Dilutions were made with HMI-18 medium, such
that the dimethyl sulfoxide concentration never exceeded a noninhibitory
concentration of 0.3%.
HETA
V
VI
Ic
IIa
Ie
Id
Ia
HET-RV
Etheno-HETA
HET-FPR
2-Amino-HETA
HETN
HET-MPR
HETI
25
0
0
IIIb
IIIa
MTT
0
0
99
0
HETT
0
HETT ϩ MTAe
2
Ribose-modified
analogs
IXa
VIIg
VIIb
VIIc
VIIe
VIId
XII
2Ј-Deoxy-HETA
CMTA
100
84f
81
34f
ETA
76.2f
75.3f
72.5f
57.5f
56
26.9f
MFETA
29.3f
27.4f
26.0f
78
RESULTS
HPTA
PTA
MTA phosphorylase assays. MTA analogs, whose struc-
tures are depicted in Tables 1 to 4, were assayed for their
abilities to act as substrates for Tryp-MTA-Pase with and
without phosphate added to the reaction mixture. Inorganic
phosphate is an absolute requirement for MPA-Pase activ-
ity. Thus, the MTA substrate analogs display differential
activities in the presence and absence of phosphate: MTA
substrate activity is 20.86 nmol/mg protein/h with added
phosphate, and with no added phosphate, the baseline ac-
tivity is 7.67 nmol/mg protein/h, likely contributed by the
action of nonspecific nucleosidases (Table 5). As seen in
HET-EMA
MFPTA
VIIf
47.5f
46
44
39.4
17
4
31.3f
60
VIIj
APTA
X
2Ј, 3Ј-seco-HETA
3؅-Deoxy-HETA
2؅, 3؅-Dideoxy-HETA
MAA
0
IXb
37.8
18
IXc
VIII
0
a Novel HETA analogs are distinguished in boldface.
b Data are from reference 3.
c 20.86 nmol/mg protein/h.
d 7.67 nmol/mg protein/h.
e Equimolar (200 ␮M) concentrations of HETT and MTA.
f Data are from reference 8.

VOL. 52, 2008
NOVEL TRYPANOCIDAL ANALOGS OF 5Ј-METHYLTHIOADENOSINE
215
Table 5, the analogs demonstrated various abilities to act as
substrates. The substrate activity of 2-fluoro-HETA (com-
pound IIb) was significantly greater than that of MTA (com-
pound VIIa). The substrate activities of HETA (compound
Ib) and 2Ј-deoxy-HETA (compound IXa) were comparable
to the activity of MTA (compound VIIa). ETA (compound
VIIb), MFETA (compound VIIc), HPTA (compound VIIe),
and CMTA (compound VIIg) were also effective substrates,
with relative activities in the range of 72 to 84%. Five ana-
logs, 5Ј-deoxy-5Ј-(hydroxyethylthio)-inosine (HETI; com-
pound Ia), 5Ј-deoxy-5Ј-(hydroxyethylthio)-6-methylpurine
riboside (HET-MPR; compound Id), 5Ј-deoxy-5Ј-(hydroxy-
ethylthio)-tubercidin (HETT; compound IIIa), MTT (com-
pound IIIb), and MAA (compound VIII), were devoid of
substrate activity. The 7-deaza analog, HETT (compound
IIIa), which was also assayed for its inhibitory effects on
enzyme activity in the presence of MTA, was neither a
substrate nor an inhibitor of the trypanosome enzyme.
In vitro growth inhibitory activities of MTA analogs. The
IC50 values of the parent nucleoside analogs were determined
(Table 6). These compounds were grouped according to their
relative activities: the highly active analogs (those active at
concentrations of Յ1 ␮M), which consisted of HETA (com-
pound Ib), HETT (compound IIIa), and MFETA (compound
VIIc); active analogs (analogs active at concentrations of 2 to
10 ␮M), which consisted of 2-amino-HETA (compound IIa),
2-fluoro-HETA (compound IIb), and 8-aza-HETA (com-
pound IV); moderately active analogs (analogs active at con-
centrations of 10 to 50 ␮M), which consisted of HET-MPR
(compound Id), 5Ј-deoxy-5Ј-(hydroxyethylthio)-nebularine
(HETN; compound Ie), 3Ј-deoxy-HETA (compound IXb),
and 2Ј,3Ј-seco-HETA (compound X); active analogs (analogs
active at concentrations of 50 to 100 ␮M), which consisted of
MTT (compound IIIb); and inactive analogs (analogs active at
concentrations of Ͼ100 ␮M). Most analogs were either slightly
growth inhibitory or non-growth inhibitory for T. brucei brucei
LAB 110 EATRO. Unexpectedly, three analogs, 5Ј-deoxy-5Ј-
(hydroxyethylthio)-ribavirin (HET-RV; compound V), AETA
(compound VIIh), and 2Ј-deoxy-HETA (compound IXa),
were observed to stimulate the growth of a drug-resistant strain
(KETRI 243 or 243-As10-3).
ment of cytotoxic MTA analogs that could be selectively acti-
vated by the MTA-Pase of trypanosomes. We identified one
such MTA analog, HETA, and found, as anticipated, that
HETA has significantly greater cytotoxicity for trypanosomes
than for mammalian cells (8).
In seeking to enhance HETA’s trypanocidal activity, we syn-
thesized new analogs which were assayed for their susceptibil-
ity to cleavage by Tryp-MTA-Pase and for their in vitro
trypanocidal activity in T. brucei brucei LAB 110 EATRO and
at least one drug-resistant KETRI clinical isolate. The biolog-
ical results from the present study were compared with rele-
vant data from previous studies (3, 8, 43) in order to determine
the distinct effects of each structural modification on enzyme
activity and trypanocidal activity, respectively (Tables 5 and 6).
The most significant increase in substrate activity occurred
when an exocyclic fluoro substituent was added at the C-2
position of the purine moiety, as evidenced by the substrate
activity of 2-fluoro-HETA (compound IIb), which was 1.5
times greater than that of MTA and HETA. The greatest
enhancement in trypanocidal activity resulted from replace-
ment of the N-7 nitrogen of HETA by a C atom; thus, when
HETT (IC50 value, 10 nM) was evaluated with pentamidine-
resistant clinical isolate KETRI 269, it was 45 times more
potent than HETA (IC50 value, 450 nM).
MTA-Pases from mammalian and microbial sources are
known to have a broad tolerance for substrates with various
alkyl and aryl substituents as replacements for the terminal
5Ј-methyl group (3, 8, 12, 15, 18, 28, 44). Accordingly, a library
of novel S-alkyl and/or S-aryl analogs of MTA is likely to
contain a small subset of MTA analogs that are more suscep-
tible to cleavage by microbial MTA-Pases than by mammalian
MTA-Pases. Our initial success in identifying HETA as one
such MTA analog was further validated by a detailed compar-
ison of the mammalian and trypanosome MTA-Pase substrate
activities of a selected group of S-alkyl-modified MTA analogs.
The Tryp-MTA-Pase substrates had decreasing activity in the
following order: MTA (100%) ϭ HETA Ͼ ETA (76%) Х
MFETA Х HPTA Ͼ PTA (57%) Ͼ MFPTA (47%) Х APTA.
The mammalian-MTA-Pase substrates have decreasing activity
in the following order (44): MTA (100%) ϭ PTA Х ETA Ͼ
MFPTA (64%) ϭ MFETA Ͼ HETA (34%). The noteworthy
increase in the Tryp-MTA-Pase substrate activity of HETA
compared with that of ETA and the increase in the Tryp-
MTA-Pase substrate activity of HPTA compared with that of
PTA are a consequence of the addition of a terminal hydroxyl
group. By contrast, the addition of a hydroxyl group reduces
the substrate activity toward mammalian MTA-Pase by 60%
(i.e., for HETA compared with that of ETA).
DISCUSSION
The biochemical differences between the mammalian and
the microbial pathways of MTA metabolism as selective targets
for drug design have been studied extensively (3, 4, 7, 8, 18, 28,
36, 40, 42). Unlike mammalian cells, two distinct pathways of
MTA metabolism are known to exist among microbial species
(Fig. 1): one operates by the initial phosphorolytic cleavage of
MTA by MTA-Pase to produce adenine and MTRP, and the
other operates by the initial hydrolytic cleavage of MTA by
MTA nucleosidase to produce adenine and 5-methylthioribose
(MTR), which is phosphorylated by MTR kinase to produce
MTRP. The further metabolism of MTRP, an intermediate
common to both pathways, results in the recycling of MTA to
methionine. African trypanosomes appear to metabolize MTA
exclusively by MTA-Pase cleavage. Ghoda et al. (18) observed
differences in the substrate specificities of MTA-Pases from T.
brucei brucei and mammalian cells and proposed the develop-
Our initial in vitro studies of HETA and MFETA pointed to
a direct association between their trypanocidal effects and their
efficacies as substrates of Tryp-MTA-Pase and implied that
their cytotoxicity is mediated through interference with methi-
onine recycling (8). This is supported by the finding that
HETA’s in vitro trypanocidal effects can be reversed by the
concomitant administration of methionine or 2-keto-4-methyl-
thio-butyrate, the immediate precursor of methionine salvage
from MTA (8). However, the current studies, which encompass
a more extensive series of purine- and ribose-modified MTA
analogs, do not substantiate the link between substrate activi-
ties and IC50 values for many compounds. We have shown that

216
SUFRIN ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
TABLE 6. In vitro efficacies of MTA analogs and selected O-acetylated derivatives in trypanosome isolates grown as bloodstream formsa
IC50 (␮M)
Compound group and no.a
Compoundb
LAB 110 EATRO
KETRI 243
KETRI 243-As10-3
KETRI 269
ND
Purine-modified analogs
Ia
HETI
Ͼ100
Ͼ100
NDc
Ib
HETA
Tri-OAc-HETA
0.54d
0.32d
0.44d
0.31d
0.19
ND
0.45d
0.26d
Ic
HET-FPR
Tri-OAc-Ic
Ͼ100
Ͼ100
Ͼ100
Ͼ100
ND
ND
ND
ND
Id
HET-MPR
Tri-OAc-Id
24.5
9.3
29.5
22
29.0
22
19
20.5
Ie
HETN
Tri-OAc-Ie
34
3.65
66
2.9
54
6.4
72
6.7
IIa
IIb
IIIa
IIIb
2-Amino-HETA
3.9
1.75
43
3.45
ND
5.5
Tri-OAc-IIa
12.5
ND
2-Fluoro-HETA
Tri-OAc-IIb
1.9
2.75
1.4
2.2
ND
3.0
1.2
3.0
HETT
Tri-OAc-IIIa
0.042
0.145
0.015
0.155
0.09
0.10
0.010
0.150
MTT
Tri-OAc-IIIb
67
46
100
100
26.5
66
Ͼ100
Ͼ100
IV
8-aza-HETA
7.0
7.8
ND
6.4
V
HET-RV
Tri-OAc-HET-RV
9e
26.5
ϩf
39
ND
25
ND
ND
VI
Etheno-HETA
Tri-OAc-VI
Ͼ100
Ͼ100
ND
38.5
ND
ND
35
19
Ribose-modified analogs
VIIa
VIIb
VIIc
VIIc
VIId
VIIe
VIIf
VIIg
VIIh
VIIj
VIII
IXa
MTA
100d
138g
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Ͼ100
ND
ϩd,f
ND
ETA
ND
MFETA
Di-OAc-VIIIc
PTA
0.20d
0.83d
46g
0.32d
0.62d
ND
0.28d
0.60d
ND
HPTA
MFPTA
CMTA
AETA
APTA
MAA
170g
ND
ND
120g
ND
ND
130g
ND
ND
Ͼ100␮M
Ͼ100
Ͼ100
ϩf
ND
72.5
Ͼ100
15.75
ND
2Ј-Deoxy-HETA
Di-OAc-IXa
Ͼ100d
Ͼ100d
Ͼ100d
22d
30.6d
14.9d
IXb
3؅-Deoxy-HETA
Di-OAc-IXb
12.5
0.64
12.5
0.82
ND
ND
40.5
0.66
Continued on facing page

VOL. 52, 2008
NOVEL TRYPANOCIDAL ANALOGS OF 5Ј-METHYLTHIOADENOSINE
217
TABLE 6—Continued
IC50 (␮M)
Compound group and no.a
Compoundb
LAB 110 EATRO
100
KETRI 243
71
KETRI 243-As10-3
ND
KETRI 269
ND
IXc
2Ј, 3Ј-Dideoxy-HETA
X
2Ј, 3Ј-seco-HETA
Tri-OAc-X
14d
15d
ND
ND
25d
5.3d
5.2d
Ͼ100d
XIa
XIb
XII
(S)-HHETP-adenine
(R)-HHETP-adenine
HET-EMA
60
Ͼ100
64
Ͼ100
Ͼ100
100
ND
ND
ND
ND
Ͼ100
ND
Ͼ100
ND
XIII
5Ј-HET-immucillin A
Ͼ10
ND
a Novel HETA analogs are distinguished in boldface.
b OAc, O-acetyl.
c ND, not determined.
d Data are from reference 46.
e The datum is in percent.
f Growth stimulatory at analog concentration of 100 ␮M.
g Data are from reference 8.
HETA has many effects on the metabolism of African trypano-
somes. These include increases in AdoMet, S-adenosylhomo-
cysteine, and MTA concentrations to unphysiological levels;
decreases in spermidine concentrations; and a reduction of the
level of methylation of proteins (3). HETA’s trypanocidal ef-
fects are enhanced by its extremely rapid uptake and concen-
tration by parasites, with its concentration reaching Ͼ800 ␮M
in minutes (3). This suggests that additional factors, such as
differences in carrier-mediated transport and the involvement
of other biochemical targets, also contribute to the trypano-
cidal effects of HETA and its analogs. To overcome possible
barriers to the active transport of these novel HETA analogs,
we prepared, as potential prodrugs, their O-acetylated deriva-
tives, which are expected to readily diffuse across the outer
membrane. In fact, O acetylation of several analogs, HET-
MPR, HETN, HET-RV, etheno-HETA, 2Ј,3Ј-deoxy-HETA,
and 3Ј-deoxy-HETA, markedly enhanced their IC50 values
(Table 6).
throughout our studies as preliminary screens to select com-
pounds as possible candidates for in vivo evaluations. In seek-
ing compounds with IC50 values Յ1 ␮M, we identified HETA,
MFETA, and 2-fluoro-HETA in prior studies (3, 8); and in the
present studies, we identified HETT, tri-O-acetyl-HETT, and
di-O-acetyl-3Ј-deoxy-HETA. HETT, the most cytotoxic com-
pound that we have identified in our in vitro screening assays,
is neither a substrate nor an inhibitor of Tryp-MTA-Pase.
Table 5 indicates that while HETT is not an inhibitor of Tryp-
MTA-Pase, it inhibits the nonphosphorolytic cleavage of MTA
and, presumably, other nucleosides. Note the difference in
Ϫ
enzyme activity without PO4 in the presence of MTA alone
and in the presence of MTA and HETT in Table 5. Since
trypanosomes and related kinetoplastid protozoa have exten-
sive purine and purine nucleoside salvage pathways, this find-
ing suggests that some nucleoside analogs which are not cleav-
able by MTA-Pase may be converted to nucleotides by a
nonspecific nucleoside phosphotransferase, such as that which
happens with allopurinol in the T. brucei brucei group (17).
Enzyme and growth inhibition assays have been used
TABLE 7. Efficacies of HETA and O-acyl-HETA derivatives against T. brucei rhodesiense infections in mice
Efficacya
Isolate
HETA
Di-OAcb-HETA
Tri-OAc-HETA
Tri-O-propyl-HETA
LAB 110 EATRO
KETRI 243
Curative (5/5)
No effect
Curative (9/10)
No effect
Curative (5/5)
No effect
Curative (1/5)
No effect
No effect
Curative (3/5)
No effect
Curative (3/5)
No effect
Curative (3/5)
No effect
Curative (1/5)
Curative (3/5)
No effect
KETRI 269
No effect
No effect
No effect
KETRI 1992
No effect
No effect
Curative (3/5)
No effect
No effect
No effect
Curative (3/5)
No effect
Curative (2/5)
No effect
KETRI 2002
Curative (3/5)
Curative (5/5)
No effect
KETRI 2285
KETRI 2537
KETRI 2538
Curative (5/5)
No effect
Curative (4/5)
No effect
Curative (4/5)
Curative (4/5)
KETRI 2545
KETRI 2772
Curative (2/5)
No effect
ATCC 30027 (Wellcome CT)
ATCC 30119 (EATRO 105)
Curative (5/5)
a Data are from reference 7. Numbers in parentheses denote the number of cures/total number of mice tested at the reported optimal daily dose (50, 100, 150, or
200 mg/kg of body weight for seven days) of compound.
b OAc, O-acetyl.

218
SUFRIN ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
alogs of 5Ј-methylthioadenosine as trypanocides. Antimicrob. Agents Che-
mother. 35:1315–1320.
9. Balasegaram, M., S. Harris, F. Checchi, S. Ghorashian, C. Hamel, and U.
Karunakara. 2006. Melarsoprol versus eflornithine for treating late-stage
Gambian trypanosomiasis in the Republic of the Congo. Bull. W. H. O.
84:783–790.
Taken together, these observations indicate that the potent
trypanocidal effects of HETT and 3Ј-deoxy-HETA are elicited
by their interactions with a novel, unidentified molecular tar-
get(s). Since neither analog is a substrate of Tryp-MTA-Pase,
their potential for selective toxicity against trypanosomes may
be diminished. This concern needs to be clarified in future
studies. We now recognize that the trypanocidal effects of
HETA analogs are not always dependent upon their initial
phosphorolytic cleavage by Tryp-MTA-Pase. However, we
consider the limited ability of mammalian MTA-Pase to cleave
HETA to be a common characteristic of HETA analogs, which
protects these compounds from metabolism by their host and
enhances their bioavailability to bloodstream trypanosomes.
Our long-term goal is to develop new HETA analogs with
improved in vivo properties compared with those of HETA, a
compound which has already been demonstrated to have a
broad spectrum of curative effects (summarized in Table 7).
HETT is by far the most potent trypanocide that we have
identified in vitro; and further studies with this compound are
warranted to determine its biochemical properties, to identify
its molecular targets, and to compare its in vivo trypanocidal
effects with those of HETA.
10. Beachum, L. M. 1979. Convenient preparation of 5Ј-chloro-2Ј,5Ј-dideoxya-
denosine. J. Org. Chem. 44:3100–3101.
11. Berens, R. L., E. C. Krug, and J. J. Marr. 1995. Purine and pyrimidine
metabolism, p. 89–117. In J. J. Marr and M. Muller (ed.), Biochemistry and
molecular biology of parasites. Academic Press, Inc., New York, NY.
12. Cartini-Farina, M., A. Oliva, G. Romeo, G. Napolitano, M. De Rosa, A.
Giambacorta, and V. Zappia. 1979. 5Ј-Methylthioadenosine phosphorylase
from Caldariella acidophila. Eur. J. Biochem. 101:317–324.
13. Coward, J. K., N. C. Motola, and J. D. Moyer. 1977. Polyamine biosynthesis
in rat prostate. Substrate and inhibitor properties of 7-deaza analogues of
decarboxylated S-adenosylmethionine and 5Ј-methylthioadenosine. J. Med.
Chem. 20:500–505.
14. Dardonville, C. 2005. Recent advances in antitrypanosomal chemotherapy:
patent literature 2002–2004. Expert Opin. Ther. Patents 15:1–17.
15. Della Ragione, F., and A. E. Pegg. 1983. Effect of analogues of 5Ј-methyl-
thioadenosine on cellular metabolism. Biochem. J. 210:429–435.
16. Evans, G. B., R. H. Furneaux, V. L. Schramm, V. Singh, and P. C. Tyler.
2004. Targeting the polyamine pathways with transition-state analogue in-
hibitors of 5Ј-methylthioadenosine phosphorylase. J. Med. Chem. 47:3275–
3281.
17. Fish, W. R., J. J. Marr, R. L. Berens, D. L. Looker, D. J. Nelson, S. LaFon,
and A. E. Balber. 1985. Inosine analogs as chemotherapeutic agents for
African trypanosomes: metabolism in trypanosomes and efficacy in tissue
culture. Antimicrob. Agents Chemother. 27:33–36.
18. Ghoda, L. Y., T. M. Savarese, C. H. Northup, R. E. Parks, Jr., J. Garofalo,
L. Katz, B. B. Ellenbogen, and C. J. Bacchi. 1988. Substrate specificities of
5Ј-deoxy-5Ј-methylthioadenosine phosphorylase from Trypanosoma brucei
brucei and mammalian cells. Mol. Biochem. Parasitol. 27:109–118.
19. Goldberg, B., D. Rattendi, D. Lloyd, J. R. Sufrin, and C. J. Bacchi. 1998. The
effects of intermediates of methionine metabolism and nucleoside analogs
on S-adenosylmethionine transport by Trypanosoma brucei brucei and a drug-
restistant Trypanosoma brucei rhodesiense. Biochem. Pharmacol. 56:95–103.
20. Goldberg, B., D. Rattendi, D. Lloyd, J. R. Sufrin, and C. J. Bacchi. 2001. In
situ kinetic characterization of methylthioadenosine transport by the aden-
osine transporter (P2) of the African Trypanosoma brucei brucei and Trypano-
soma brucei rhodesiense. Biochem. Pharmacol. 61:449–457.
21. Goldberg, B., N. Yarlett, D. Rattendi, D. Lloyd, and C. J. Bacchi. 1997. Rapid
methylation of cell proteins and lipids in Trypanosoma brucei. J. Eukaryot.
Microbiol. 44:345–351.
22. Goldberg, B., N. Yarlett, J. R. Sufrin, D. Lloyd, and C. J. Bacchi. 1997. A
unique transporter of S-adenosylmethionine in African trypanosomes.
FASEB J. 11:256–260.
23. Hanna, N. B., K. G. Upadha, C. R. Petrie, R. K. Robins, and G. R. Revankar.
1986. Synthesis of certain 5Ј-substituted derivatives of ribavirin and tiazofu-
ran. Nucleosides Nucleotides 5:343–362.
24. Hirumi, H., and K. Hirumi. 1989. Continuous cultivation of Trypanosoma
brucei bloodstream forms in a medium containing a low concentration of
serum protein without feeder cell layers. J. Parasitol. 75:985–989.
25. Holy, A. 1975. Aliphatic analogues of nucleosides, nucleotides and olignucle-
otides. Coll. Czech. Chem. Commun. 40:187–214.
26. Kikugawa, K., K. K. Izuka, Y. Higuchi, H. Hirayama, and M. Ichino. 1972.
Platelet aggregation inhibitors. 2. Inhibition of platelet aggregation by 5Ј-,
2Ј-, 6Ј- and 8Ј-substituted adenosines. J. Med. Chem. 15:387–390.
27. Kuhn, R., and W. Jahn. 1965. Vom adenosine abgeleitete thioather und
S-oxide. Chemistry (Berlin) 98:1699–1704.
28. Kung, P.-P., L. R. Zehnder, J. J. Meng, S. W. Kupchinsky, D. J. Skalitzkt,
M. C. Johnson, K. A. Maegley, A. Ekker, L. A. Kuhn, P. W. Rose, and L. A.
Bloom. 2005. Design, synthesis and biological evaluation of novel human
5Ј-deoxy-5Јmethylthioadenosine phosphorylase (MTAP) substrates. Bioorg.
Med. Chem. Lett. 15:2829–2833.
ACKNOWLEDGMENTS
These studies were supported in part by grants 920082 (to J.R.S.)
and 95094 (to C.J.B.) from the United Nations Development Pro-
gramme/World Bank/World Health Organization Special Programme
for Research and Training in Tropical Diseases and Public Health
Service grant AI32975 (to J.R.S.). NMR and mass spectra were pro-
vided by the Roswell Park Cancer Institute NMR and Biopolymer
Core Facilities, which are supported in part by NCI core grant
CA16056 to the Roswell Park Cancer Institute. Support for the syn-
thesis of HET-immucillin A was from NIH and Science and Technol-
ogy grants to Industrial Research Ltd., Lower Hutt, New Zealand, and
the Albert Einstein College of Medicine.
We thank Pharmacia & Upjohn, Inc. (Kalamazoo, MI), for the
generous gift of tubercidin. HET-immucillin A was synthesized by
Peter C. Tyler of Industrial Research Ltd. and was provided by Vern
L. Schramm of the Albert Einstein College of Medicine.
REFERENCES
1. Anisuzzaman, A. K. M., and R. L. Whistler. 1978. Selective replacement of
primary hydroxyl groups in carbohydrates: preparation of some carbohydrate
derivatives containing halomethyl groups. Carbohydr. Res. 61:511–518.
2. Bacchi, C. J., J. Garofalo, M. A. Ciminelli, D. Rattendi, B. Goldberg, P. P.
McCann, and N. Yarlett. 1993. Resistance to DL-␣-difluoromethylornithine
by clinical isolates of Trypanosoma brucei rhodesiense: role of S-adenosyl-
methionine. Biochem. Pharmacol. 46:471–481.
3. Bacchi, C. J., B. Goldberg, D. Rattendi, T. E. Gorrell, A. J. Spiess, and J. R.
Sufrin. 1999. Metabolic effects of a methylthioadenosine phosphorylase sub-
strate analog on African trypanosomes. Biochem. Pharmacol. 57:89–96.
4. Bacchi, C. J., P. P. McCann, H. C. Nathan, S. H. Hutner, and A. Sjoerdsma.
1982. Antagonism of polyamine metabolism—a critical factor in chemother-
apy of African trypanosomiasis. Adv. Polyamine Res. 4:221–231.
5. Bacchi, C. J., H. C. Nathan, A. B. Clarkson, E. J. Bienen, A. J. Bitonti, P. P.
McCann, and A. Sjoerdsma. 1987. Effects of the ornithine decarboxylase
inhibitors DL-alpha-difluoromethyl-ornithine and alpha-monofluoromethyl-
dehydroornithine methyl ester alone and in combination with suramin
against Trypanosoma-brucei brucei central-nervous-system models. Am. J.
Trop. Med. Hyg. 36:46–52.
6. Bacchi, C. J., H. C. Nathan, N. Yarlett, B. Goldberg, P. P. McCann, A. J.
Bitonti, and A. Sjoerdsma. 1992. Cure of murine Trypanosoma brucei
rhodesiense infections with an S-adenosylmethionine decarboxylase inhibitor.
Antimicrob. Agents Chemother. 32:2736–2740.
7. Bacchi, C. J., K. Sanabria, A. J. Spiess, M. Vargas, C. J. Marasco, Jr., L. M.
Jiminez, B. Goldberg, and J. R. Sufrin. 1997. In vivo efficacies of 5Ј-meth-
ylthioadenosine analogs as trypanocides. Antimicrob. Agents Chemother.
41:2108–2112.
29. Legros, D., G. Ollivier, M. Gastellu-Etchegorry, C. Paquet, C. Burri, J.
Jannin, et. al. 2002. Treatment of human African trypanosomiasis—present
situation and needs for research and development. Lancet Infect. Dis. 2:437–
440.
30. Marasco, C. J., Jr., D. L. Kramer, J. Miller, C. W. Porter, C. J. Bacchi, D.
Rattendi, L. Kucera, N. Iyer, R. Bernacki, P. Pera, and J. R. Sufrin. 2002.
Synthesis and evaluation of analogues of 5Ј-([(Z)-4-amino-2-butenyl]meth-
ylamino)-5Ј-deoxyadenosine (MDL73811) as inhibitors of tumor cell growth,
trypanosomal growth and HIV-1 infectivity. J. Med. Chem. 45:5112–5122.
31. Marasco, C. J., Jr., P. J. Pera, A. J. Spiess, R. J. Bernacki, and J. R. Sufrin.
2005. Improved synthesis of ␤-D-6-methylpurine riboside and antitumor ef-
fects of the ␤-D- and ␣-D-anomers. Molecules 10:1015–1020.
32. Meyers, R. W., and R. H. Abeles. 1989. Conversion of 5-S-ethyl-5-thio-D-
ribose to ethionine in Klebsiella pneumoniae. Basis for the selective toxicity of
5-S-ethyl-5-thio-D-ribose. J. Biol. Chem. 264:10547–10551.
8. Bacchi, C. J., J. R. Sufrin, H. C. Nathan, A. J. Spiess, T. Hannan, J.
Garofalo, K. Alecia, L. Katz, and N. Yarlett. 1991. 5Ј-Alkyl-substituted an-

VOL. 52, 2008
NOVEL TRYPANOCIDAL ANALOGS OF 5Ј-METHYLTHIOADENOSINE
219
33. Montgomery, J. A., and K. Hewson. 1968. A convenient method for the
synthesis of 2-fluoroadenosine. J. Org. Chem. 33:432–434.
34. Montgomery, J. A., and K. Hewson. 1969. The preparation of 6-fluoropurines
by the modified Schiemann reaction. J. Org. Chem. 34:1396–1399.
35. Novotny, L., and J. Beranek. 1991. 5Ј-Chloro-5Ј-deoxyinosine. 5Ј-halogeno
purine nucleosides in the synthesis of anomalous nucleosides. Nucleic Acids
Chem. 4:189–193.
41. Sturk, L. M., J. L. Brock, C. R. Bagnell, J. E. Hall, and R. R. Tidwell. 2004.
Distribution and quantitation of the anti-trypanosomal diamidine 2,5-bis(4-
amidinophenyl)furan (DB75 and its N-methoxy prodrug DB289 in murine
brain tissue. Acta Trop. 9:131–143.
42. Sufrin, J. R., S. R. Meshnick, A. J. Spiess, J. Garofalo-Hannan, X.-Q. Pan,
and C. J. Bacchi. 1995. Methionine recycling and antimalarial drug design.
Antimicrob. Agents Chemother. 39:2511–2515.
36. Riscoe, M. K., A. J. Ferro, and J. H. Fitchen. 1989. Methionine recycling as
a target for antiprotozoal drug development. Parasitol. Today 5:330–333.
37. Robins, M. J., and P. W. Hatfield. 1982. Nucleic acid related compounds. 37.
Convenient and high-yield syntheses of N-[(2-hydroxyethoxy)methyl] hetero-
cycles as “acyclic nucleoside” analogues. Can. J. Chem. 60:547–553.
38. Robins, M. J., V. Neschadimenko, B.-O. Ro, C.-S. Yuan, R. T. Borchardt,
and S. F. Wnuk. 1998. Nucleic acid related compounds. 101. S-adenosyl-L-
homocysteine hydrolase does not hydrate (5Ј-fluoro)vinyl or (6Ј-halo)homovinyl
analogs derived from 3Ј-deoxyadenosine or 3Ј-(chloro or fluoro)-3Ј-deoxy-
adenosine. J. Org. Chem. 63:1205–1211.
39. Samejima, K., Y. Nakazawa, and I. Matsusaga. 1978. Improved synthesis of
decarboxylated S-adenosylmethionine and related sulfonium compounds.
Chem. Pharm. Bull. 26:1480–1485.
40. Sekowska, A., V. Denervaud, H. Ashida, K. Michoud, D. Haas, A. Yokota,
and A. Danchin. 2004. Bacterial variations on the methionine salvage path-
way. BMC Microbiol. 4:9. http://www.biomedcentral.com/1471-2180/4/9.
43. Sufrin, J. R., D. Rattendi, A. J. Spiess, S. Lane, C. J. Marasco, and C. J.
Bacchi. 1996. Antitrypanosomal activity of purine nucleosides can be en-
hanced by their conversion to O-acetylated derivatives. Antimicrob. Agents
Chemother. 40:2567–2572.
44. Sufrin, J. R., A. J. Spiess, D. L. Kramer, P. L. Libby, J. T. Miller, R. J.
Bernacki, R. T. Borchardt, Y. Lee, and C. W. Porter. 1991. Targeting 5Ј-
deoxy-5Ј-(methylthio)-adenosine phosphorylase by 5Ј-haloalkyl analogues of
5Ј-deoxy-5Ј-(methylthio)-adenosine. J. Med. Chem. 34:2600–2606.
45. Sufrin, J. R., A. J. Spiess, C. J. Marasco, S. L. Croft, D. Snowdon, V. Yardley,
and C. J. Bacchi. 1995. Purine 2Ј,3Ј-acyclonucleosides: improved synthesis
and antiparasitic activity. Bioorg. Med. Chem. Lett. 5:1961–1964.
46. Trager, W. 1978. Cultivation of parasites in vitro. Am. J. Trop. Med. Hyg.
27:216–222.
47. Vorbruggen, H., and B. Bennua. 1981. A new simplified nucleoside synthesis.
Chemistry (Berlin) 114:1279–1286.