Zidampidine, an aryl phosphate derivative of AZT: In vivo pharmacokinetics, metabolism, toxicity, and anti-viral efficacy against hemorrhagic fever caused by Lassa virus

Article abstract of DOI:10.1016/j.bmc.2005.02.031

The pharmacokinetics, metabolism, and toxicity of Zidampidine, an aryl phosphate derivative of AZT, 3′-azidothymidine-5′-[p-bromophenyl methoxyalaninyl phosphate] were investigated in CD-1 mice. Following iv injection, Zidampidine was rapidly converted to

Full text of DOI:10.1016/j.bmc.2005.02.031

Bioorganic & Medicinal Chemistry 13 (2005) 3279–3288
Zidampidine, an aryl phosphate derivative of AZT: in vivo
pharmacokinetics, metabolism, toxicity, and anti-viral
efficacy against hemorrhagic fever caused by Lassa virus
F. M. Uckun,a,* T. K. Venkatachalam,a D. Erbeck,a C. L. Chen,a
A. S. Petkevichb and A. Vassileva
aDrug Discovery Program, Parker Hughes Center for Clinical Immunology, 2699 Patton Road, St. Paul, MN 55113, USA
bResearch Institute for Epidemiology and Microbiology, 220050 MINSK, Belarus
Received 1 December 2004; accepted 14 February 2005
Abstract—The pharmacokinetics, metabolism, and toxicity of Zidampidine, an aryl phosphate derivative of AZT, 30-azidothymi-
dine-50-[p-bromophenyl methoxyalaninyl phosphate] were investigated in CD-1 mice. Following iv injection, Zidampidine was rapi-
dly converted to itsmetabolitesAla-AZT-MP and AZT. Zidampidine wasnot toxic to mice at dosesup to 250 mg/kg. We next
examined the therapeutic effect of Zidampidine in CBA mice challenged with intracerebral injections of the Josiah strain of Lassa
virus. Mice were treated either with vehicle or non-toxic doses of Zidampidine administered intraperitoneally 24 h prior, 1 h prior,
and 24, 48, 72, and 96 h after virus inoculation. The probability of survival following the Lassa challenge was significantly improved
for Zidampidine-treated mice (Kaplan Meier, Log-Rank p value < 0.0001). This pilot study provides the basis for future preclinical
evaluation of Zidampidine and its potential as a new agent for the treatment of viral hemorrhagic fevers caused by Lassa virus.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
We and othershave ysntheiszed a number of phenyl
phosphate derivatives of AZT (3-azido-30-deoxythymi-
18–35
Lassa virus is the causative agent of Lassa fever, an
acute viral disease found throughout West Africa, Eur-
ope, Asia, and the United States, which causes consider-
able morbidity and mortality.1–12 Severe multi-organ
involvement occursin 5–10% of infectionsand caes-
fatality ratesfor hops italized patientsrange from 15%
to 25%.13,14 The virus is transmitted by the respiratory
route and by direct contact with contaminated materi-
als. The potential use of hemorrhagic fever (HF) viruses
asagentsof biological warfare (BW) isa growing con-
cern. Because of the ability of Lassa virus to spread from
person to person, risk of its importation by international
travel, and renewed threatsabout the potential ues of
HF viruses for BW, Lassa fever has emerged as a world-
wide concern among public health officialsepsecially
among children.15–17
dine) aspotential anti-viral agents.
The purpose of
the present study was to examine the effects of the novel
AZT derivative Zidampidine (30-azidothymidine-50-
[p-bromophenyl methoxyalaninyl phosphate]) on the
survival outcome of mice challenged with fatal amounts
of Lassa virus.
2. Results and discussion
2.1. Pharmacokinetic profiles of Zidampidine in mice
following intravenous administration
Zidampidine wasimmediately converted to metabolite 1
and metabolite 2 after a single bolus intravenous injec-
tion. The plasma metabolite concentration–time curves
are presented in Figure 1A and B. The estimated phar-
macokinetic parameter valuesare presented in Table 1.
Following iv administration of Zidampidine (240 mg/
kg) to CD-1 mice, metabolite 1 (herein known as
Zidampidine-M1) wasformed almost immediately ( tmax
of 5.4 min) and showed rapid elimination (t1/2 of
*
Corresponding author. Tel.: +1 651 796 5450; fax: +1 651 796
5493; e-mail: fatih_uckun@ih.org
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.02.031

3280
F. M. Uckun et al. / Bioorg. Med. Chem. 13 (2005) 3279–3288
LC–MS conditions( Fig. 2B) showed a clean spectrum
with a mass value of 417. Based on this result we pro-
pose that Zidampidine-M1 was formed in the plasma
samples along with other unknown compounds.
Figure 1. Plasma concentration–time profiles of Zidampidine-M1 (A)
and Zidampidine-M2 (B) in CD-1 mice following iv bolusinjection of
Zidampidine at dose of 240/kg (five mice per time point).
22.3 min). The estimated AUC and Cmax valueswere
196.0 lM h and 124.2 lM, respectively.
When Zidampidine wasfurther metabolized to form
metabolite 2 (herein known asZidampidine-M2) the
time to reach the maximum plasma concentration was
14.7 min. The elimination half-life of Zidampidine-M2
following iv injection of Zidampidine was90.3 min.
The estimated AUC and Cmax valuesof metabolite 2
were 177.3 lM h and 73.0 lM, respectively.
Figure 2. Mass spectra of Zidampidine-M1 obtained from plasma
extraction (A) and synthetic Ala-AZT-MP: (A) corresponds to the
material extracted from the plasma after iv injection of Zidampidine
and (B) represents the authentic synthetic sample of Ala-AZT-MP.
The material obtained from the plasma shows complexity in the
pattern perhapsdue to the presence of other unknown material that
eluted with Zidampidine-M1. Our intent wasnot to identify all the
peaks from the plasma sample, but was instead to compare the peaks
in the spectrum corresponding to the authentic sample of Ala-AZT-
MP.
2.2. Identification of Zidampidine metabolites
The first Zidampidine metabolite (Zidampidine-M1)
wasextracted from the plasma of CD-1 mice following
iv injection of Zidampidine. The LC–MS spectrum
showed peaks corresponding to a mass value of 417
along with additional mass peaks (Fig. 2A). In contrast,
the synthetic compound, (30-azido-30-deoxythymidine-
50-methoxyalaninyl phosphate) under the same
Table 1. Estimated pharmacokinetic parameter values of Zidampidine metabolites in CD-1 mice
Measured
AUC (lM h)
Cmax (lM)
t1/2 (min)
tmax (min)
Ala-AZT-MP
AZT
124.2 (149.6 10.7)
177.3 (165.4 13.5)
196.0 (226.0 16.5)
73.0 (75.9 2.9)
22.3 (23.1 2.5)
90.3 (78.9 6.1)
5.4 (6.6 0.9)
14.7 (15.3 1.7)
Pharmacokinetic parameters in CD-1 mice following intravenous injection of 240 mg/kg are present as the average values estimated from composite
plasma concentration–time curves of pooled data. The mean SE values are indicated in the parentheses (N = 5 mice per time point). Abbreviations:
t1/2 isterminal elimination half-life; tmax isthe time required to reach the maximum plasma drug concentration following injection.

F. M. Uckun et al. / Bioorg. Med. Chem. 13 (2005) 3279–3288
3281
The second metabolite (Zidampidine-M2) was extracted
directly with ethyl acetate from the urine of Zidampidine-
treated mice. The NMR spectrum of Zidampidine-M2
wasfound to be identical to that of pure AZT. Therefore,
Zidampidine-M2 wasidentified asAZT.
HPLC profiles of the plasma samples, we observed forma-
tion of both Zidampidine-M1 (Ala-AZT-MP) aswell as
Zidampidine-M2 (AZT). If cleavage occurred via the
second pathway, then we would have observed forma-
tion of Zidampidine-M2 (AZT) only, rather than
Zidampidine-M1 and Zidampidine-M2 (Fig. 3). In the
case of the third proposed pathway, formation of only
Zidampidine-M2 would be visualized as Zidampidine-
M1 would lack the AZT moiety due to the reasons
explained earlier and hence we would not be able to ob-
serve the formation of Zidampidine-M1. From these re-
sults, we propose that formation of both the metabolites
demonstrates that the methyl ester side chain of
Zidampidine undergoes hydrolysis via the first proposed
pathway by first forming Ala-AZT-MP.
Zidampidine-M2 (AZT) may be produced from
Zidampidine-M1 (Ala-AZT-MP) through enzymatic
metabolism in vivo. Hence there are three pathways
through which AZT could be formed in vivo. In the first
pathway, Zidampidine enzymatically convertsinto
Zidampidine-M1 by hydrolysis of the methyl ester side
chain of the compound. The carboxylic acid cyclizes
intramolecularly by eliminating the bromophenoxy
group thusforming an untsable cyclic intermediate
(see Scheme 1). Thisintermediate isfurther hydrolyzed
by water to form Zidampidine-M1(Ala-AZT-MP), that
can further metabolize in the presence of other enzymes
to form Zidampidine-M2 (AZT).
2.3. In vivo toxicity of Zidampidine in mice
Six weeksold female CD-1 mice were adminitsered an
intraperitoneal bolusinjection of Zidampidine in
0.2 mL 10–20% DMSO/PBS solution, or 20% DMSO/
PBS alone (control mice). Groupsof 10 mice received
a single treatment at one of the three different dose levels
of Zidampidine (100 lg, 1 mg, and 5 mg). Ten control
mice were injected intraperitoneally with 20% DMSO/
PBS solution in accordance with the experimental
Alternatively, direct enzymatic cleavage of the nucleo-
side can break the nucleoside moiety thus forming
Zidampidine-M2 (AZT). In a third pathway, simulta-
neous attack on both the ester side chain in combination
with direct attack on the phosphorus center may form
both the metabolites. However, when we compared the
O
5
NH
O
6 N
O
5
H
O
O
P O
NH
NH
5’
Br
O
H 6
O
N
O
O
P O
NH
5’
Carboxy esterase
Br
O
COOMeN3
N3
COOH
Zidampidine
Unstable intermediate
O
O
5
Intramolecular cyclization and
elimination
5
NH
NH
6
N
O
6
N
O
H
O
H
O
O
5’
5’
O
P O
NH
Br
O
P O
NH
O
OH
O
N3
N3
O
Unstable cyclic intermediate
Hydrolysis
O
O
N
O
5
6
NH
NH
Other enzymes
O
H
O
N
O
H
5’
HO
P O
NH
H
O
O
N3
COOH
N3
Meabolite 1
[Alanine-AZT-monophosphate]
Metabolite 2
{AZT]
Scheme 1. Proposed hydrolysis pathway of Zidampidine in vivo.

3282
F. M. Uckun et al. / Bioorg. Med. Chem. 13 (2005) 3279–3288
Mice were then observed twice daily for 21 days for
morbidity and mortality. Of the 20 control mice, 2 died
on day 1 immediately after intracerebral injection due to
accidental brain injury and are not evaluable. All of the
remaining 18 vehicle-treated control mice developed
decreased mobility and scruffy fur as the clinical signs
of Lassa infection between days 6 and 10 (Table 3).
Fourteen of the 18 control mice developed seizures be-
tween days7 and 11. Thirteen mice experienced 4–10%
weight loss and died between days 8 and 11 (Table 3,
Fig. 4). Of the 10 mice treated with Zidampidine at
the 25 mg/kg dose level, 1 died accidentally immediately
after intracerebral Lassa virus inoculation. All of the
remaining nine mice developed decreased mobility and
scruffy fur as the clinical signs of Lassa infection be-
tween days 8 and 10. All nine mice survived the Lassa
challenge beyond the 21-day observation period and
did not experience any weight loss or seizures (Table
3, Fig. 4). The probability of survival following the
Lassa challenge was significantly improved for Zidampi-
dine-treated mice (Kaplan Meier, Log-Rank
p
value ꢀ 0.001). The probability of survival at 21 days
was28% (7–48%, 95% confidence limit)s for vehicle-
treated mice (median survival = 9 days), and 100% for
mice treated with Zidampidine at the 25 mg/kg dose
level (median survival >21 days).
These results provide unprecedented evidence that
Zidampidine exhibitsa potent prophylactic effect
against Lassa fever and therefore shows clinical poten-
tial asa new agent for the treatment of viral hemor-
rhagic fevers caused by Lassa virus.
Figure 3. Representative HPLC chromatograms of blank plasma (A)
and plasma sample at 10 min following iv injection 240 mg/kg
Zidampidine.
protocol. There were no immediate adverse clinical
effectson the test mice following Zidampidine administra-
tion. All mice were electively sacrificed healthy without
any weight loss on day 30 (Table 2). No significant
gross organ pathology was observed at the time of nec-
ropsy. Histopathologic examination of multiple tissues
revealed no toxic lesions.
3. Experimental
3.1. Chemicals
All the reagents used in this study were HPLC grade.
Deionized distilled water was prepared using Milli-Q
purification system (Millipore, Medford, MA). Metha-
nol, acetonitrile, acetic acidswere purchaesd from
Fisher Chemicals (Fair Lawn, NJ). AZT (30-azido-30-
deoxythymidine) wasfrom Toronto Research Chemicals
(Ontario, Canada).
2.4. In vivo anti-Lassa activity of Zidampidine
In order to evaluate the anti-Lassa activity of Zidampi-
dine, CBA mice were inoculated with intracerebral injec-
tions of the Josiah strain of Lassa36 at a 1000 PFU dose
level, which islethal to 70–100% of mice within 7–12
days.37 Mice were treated either with vehicle or
Zidampidine administered intraperitoneally 24 h prior,
1 h prior, and 24, 48, 72, and 96 h (total number of doses
administered to each mouse = 6) after virus inoculation.
3.2. Synthesis and characterization of Zidampidine
derivatives
Zidampidine [30-azidothymidine-50-[p-bromophenyl meth-
oxyalaninyl phosphate]] and Ala-AZT-MP [AZT-50-
Table 2. Life-table analysis of survival data and statistical analysis of weight change
Treatment group
# of mice
Mean weight change (g) SEM
Weight change p valuea
Zidampidine
1 mg
20% DMSO
100 lg
5 mg
20% DMSO
10
10
10
10
6.17 1.017
6.49 0.591
6.96 1.192
6.04 1.002
0.7886
0.6201
0.7280
0.9284
0.7035
0.5620
100 lg Zidampidine
1 mg Zidampidine
5 mg Zidampidine
0.7886
0.6201
0.9284
0.7280
0.7035
0.5620
a Weight p value determined by unpaired t-test analysis. A p value of <0.05 was considered statistically significant.

F. M. Uckun et al. / Bioorg. Med. Chem. 13 (2005) 3279–3288
Table 3. Anti-Lassa activity of Zidampidine in CBA mice
3283
Disease onset (days after inoculation with Lassa virus)
Decreased mobility
Scruffy fur
Convulsions
Weight loss (%)
Survival (days)
Group A—vehicle
Mouse #1a
Mouse #2a
Mouse #3
NA
NA
6.0
6.0
7.5
7.5
7.0
7.0
7.0
7.0
8.5
8.5
8.0
9.0
9.0
9.0
9.0
9.0
9.0
9.5
NA
NA
6.0
NA
NA
7.0
NA
NA
4.5
5.0
9.0
4.3
4.3
4.5
8.3
4.8
5.3
4.5
4.5
4.1
4.3
NO
NO
NO
NO
NO
61
61
8
Mouse #4
Mouse #5
6.5
7.0
8.0
8.0
9
9.5
Mouse #6
Mouse #7
7.5
7.0
8.0
8.0
9.0
9.0
Mouse #8
Mouse #9
7.0
7.0
8.5
8.5
9.5
9.5
Mouse #10
Mouse #11
Mouse #12
Mouse #13
Mouse #14
Mouse #15
Mouse #16
Mouse #17
Mouse #18
Mouse #19
Mouse #20
7.0
8.0
8.5
9.0
9.5
10.5
10.0
10.0
11.0
11.5
8.0
8.5
9.0
9.5
9.0
9.5
10.5
11.0
NO
NO
11.0
NO
NO
10.0
10.0
10.0
10.0
10.0
>21
>21
>21
>21
>21
Group B—Zidampidine 25 mg/kg
Mouse #1a
Mouse #2
NA
8.5
8.0
8.0
8.0
8.0
8.0
8.5
9.0
9.0
NA
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.0
8.5
NA
NO
NO
NO
NO
NO
NO
NO
NO
NO
NA
NO
NO
NO
NO
NO
NO
NO
NO
NO
61
>21
>21
>21
>21
>21
>21
>21
>21
>21
Mouse #3
Mouse #4
Mouse #5
Mouse #6
Mouse #7
Mouse #8
Mouse #9
Mouse #10
NA = not applicable; NO = not observed.
a Mouse died after traumatic intracerebral injection.
(methoxyalaninyl phosphase)] were synthesized as de-
scribed previously19,24–35 and asfollows.
wasevaporated under vacuum using a rotary evaporator
to yield crude phosphorodichloridate (1) asa pale yel-
low viscous oil. Vacuum distillation of the crude product
gave p-bromophenyl phosphorodichloridate (1) asa col-
orless, viscous oil. Bp 98–102 ꢁC/0.1 mm, UV: kmax: 242,
271 nm, IR: 3878, 3095, 2358, 1888, 1712, 1483, 1187,
3.2.1. Synthetic scheme for aryl phosphoramidates.
Scheme 2 illustrates the route followed for the synthesis
of aryl phosphoramidate derivatives of AZT.24–30
831 cmꢁ1 1H NMR (CDCl3): d 7.46–7.44 (m, 2H),
.
3.2.1.1. p-Bromophenyl phosphorodichloridate (1).
Phosphorus oxychloride (17.0 g, 111 mmol) in diethyl-
ether (100 mL) wasplaced in a three-neck RB flaks
equipped with an additional funnel. The contentswere
cooled to 0 ꢁC using an external ice bath. A solution
of p-bromophenol (17.3 g, 100 mmol) and triethylamine
(10.1 g, 100 mmol) in anhydrousdiethyl ether (250 mL)
wasadded dropwise while maintaining the temperature
of the mixture at 0 ꢁC throughout the addition (ca: 2–
3 h). After thisperiod, the ice bath wasremoved and
the mixture wasallowed to gradually attain room tem-
perature and wasstirred vigorously overnight. The addi-
tional funnel wasreplaced by a condenesr and the
reaction mixture washeated to reflux for an additional
2 h to ensure completion of the reaction. The reaction
mixture wascooled to room temperature and the precipi-
tated triethylammonium salt was filtered under vacuum
and the precipitate waswahsed with additional anhy-
drousether. The layerswere combined and the solvent
7.50–7.47 (m, 2H), 31P NMR (CDCl3): d 3.12.
3.2.1.2. p-Bromophenyl methoxyalaninyl phosphoro-
chloridate (2). p-Bromophenyl phosphorodichloridate
(1) (5.8 g, 20.02 mmol) and L-alaninemethylester hydro-
chloride (2.8 g, 20.08 mmol) were placed into a RB flask
under nitrogen atmosphere. Using a dry syringe, anhy-
drousmethylene chloride (160 mL) wasadded and the
mixture wascooled to ꢁ70 ꢁC using a dry ice/acetone
bath. A solution of triethylamine (4.0 g, 40.0 mmol) in
anhydrousmethylene chloride (120 mL) wasadded
dropwise with vigorous stirring over a period of 3 h.
After completion of the addition, the reaction mixture
wasallowed to gradually warm to room temperature
and stirred for 48–72 h until the reaction was complete
asevidenced from TLC. The contentswere concentrated
in vacuum, anhydrousether (60 mL) wasadded and the
precipitated triethylammonium hydrochloride salt was
filtered. The precipitate wasfurther wahsed with

3284
F. M. Uckun et al. / Bioorg. Med. Chem. 13 (2005) 3279–3288
Figure 4. Protective activity of Zidampidine in CBA mice challenged with Lassa virus. CBA mice were inoculated with intracerebral injections of the
Josiah strain of Lassa at a 1000 PFU dose level. Mice were treated either with vehicle or Zidampidine (25 mg/kg) administered intraperitoneally 24 h
prior, 1 h prior, and 24, 48, 72, and 96 h after virus inoculation. Mice were then observed twice daily for 21 days for morbidity and mortality. Results
are presented as the cumulative proportion of mice surviving after virus inoculation. See Table 3 for more detailed information of the treatment
outcome.
O
P
O
P
Cl
E N/Et O
3
2
Br
O
Br
Cl
OH
+
Cl
Br
Cl
Cl
1
Me
O
P
Cl
AZT/THF
Et N/CH Cl
3
2
2
O
CO Me
2
N
H
N-Methylimidazole
Me
NH
HCl
2
2
COOMe
O
H C
3
NH
O
H
O
N
O
Br
O
P O
NH
N
3
CO Me
Me
2
3
(Zidampidine)
Scheme 2. Synthetic scheme for Zidampidine.
additional ether (2 · 20 mL). The combined ether ex-
tractswere pooled together and rotary evaporated under
atmosphere. Using a dry syringe, anhydrous tetrahydro-
furan (15 mL) wasintroduced and the contentstsirred
to dissolve the phosphorochloridate. Using another
dry syringe, N-methyl imidazole (2.0 g, 24.5 mmol) was
added and the contentswere tsirred for additional
20 min. After thisperiod, a oslution of AZT (1.25 g,
4.1 mmol) in anhydroustetrahydrofuran (40 mL) was
added and the reaction mixture wastsirred vigoroulsy
at room temperature under a nitrogen atmosphere for
100–120 h. Throughout the reaction, the flask was cov-
ered with an aluminum foil to avoid light during the
reaction. After the reaction wascomplete asevidenced
by TLC, the mixture wasconcentrated to remove tetra-
hydrofuran under vacuum and the residue was dissolved
in chloroform (100 mL). The chloroform solution was
vacuum to yield (2) asa vicsousoil. UV:
273 nm, IR: 3212, 2989, 1747, 1483, 1270, 1209, 1147,
kmax: 241,
1
1010, 927, 831 cmꢁ1; HNMR (CDCl3): d 8.70 (br s,
1H), 7.16 (d, 2H, J = 9.0 Hz), 7.48 (d, 2H, J = 9.0 Hz),
3.79 (s, 1H), 3.77 (s, 1H), 1.51 (m, 3H), 1.40 (m, 3H);
MS (CI, m/e): 357.9 (M++2), 355.9 (M+), 322.0
(M++2ꢁCl), 320 (M+ꢁCl), 295.9 (M+ꢁCOOCH3),
186.0 (M++2ꢁBrꢁPhO), 184.0 (M+ꢁBrꢁPhO).
3.2.2. 30-Azidothymidine-50-[(p-bromophenyl)methoxy-
alaninyl phosphate] (3) (Zidampidine). p-Bromophenyl
methoxyalaninyl phosphorochloridate (2) (4.27 g,
12.0 mmol) wasplaced in a dry RB flask under nitrogen

F. M. Uckun et al. / Bioorg. Med. Chem. 13 (2005) 3279–3288
3285
washed consecutively with a 1 N aqueous HCl solution
3.5. Animal housing
(2 · 100 mL) saturated aqueous sodium bicarbonate
(2 · 75 mL), water (3 · 75 mL). The separated organic
layer was dried over anhydrous sodium sulfate, filtered,
and evaporated to dryness to afford (3) asa pale yellow
viscous oil. The crude product was further purified by
column chromatography using 2–5% MeOH/CHCl3 to
give analytically pure (3).
All mice were housed and cared for in accordance with
guidelinesetsablihsed by the Animal Welfare Act and
the National Institutes of Health Guide for the Care
and Use of Laboratory Animals, following approval of
the Parker Hughes Institute Animal Care and Use
Committee.
The structures of the resulting compounds were identi-
3.5.1. Toxicity studies in CD-1 mice. All CD-1 mice used
in this toxicity study were obtained from the specific
pathogen free (SPF) breeding facilitiesof CharlesRiver
Laboratories(Wilmington, MA) at 5 weeksof age. The
mice were housed in the animal housing facility of the
Parker Hughes Institute. All husbandry and experimen-
tal contact made with the mice maintained SPF condi-
tions. The mice were kept in microisolator cages (Lab
Products, Inc., Maywood, NJ) containing autoclaved
food, water, and bedding.
1
fied by H and 13C NMR (Varian Mercury 300 instru-
ment), mass spectra [MALDI-TOF spectrometer
(Model G2025A LD-TOF)], and IR [Nicolet Protege
460 spectrometer] and are shown in Figure 5.
3.3. Physico-chemical data of Zidampidine
White foamy solid; yield (83%); IR (neat): m 3205, 2109,
1
1745, 1691 cmꢁ1; H NMR (CDCl3): d 8.69 (br s, 1H),
7.45 (m, 2H), 7.33 (s, 1H), 7.14 (m, 2H), 6.18 (t, 1H)
6.13 (t, 1H), 4.44–3.77 (m, 4H), 3.73*, 3.72* (s, 3H),
2.18 (s, 3H), 1.39 (m, 3H), 1.41 (m, 3H); 13C NMR
(CDCl3): d 173.8, 163.6, 150.1, 149.2, 149.1, 135.4,
132.4, 121.8, 121.7, 117.8, 111.1, 85.0, 84.7, 81.9, 81.8,
82.2, 65.5, 60.1, 59.4, 52.4, 50.0, 49.9, 36.9, 20.6, 12.2;
GC/MS (CI) m/z 587.1 (C20H2479BrN6O8P+), 589.1
(C20H2481BrN6O8P+); HPLC tR 36.10, 37.51 min (aceto-
nitrile/water).
In thistoxicity tsudy, 40 weighed 6 weeksold female
CD-1 mice averaging 21.5 g were administered an intra-
peritoneal bolusinjection of Zidampidine in 0.2 mL 10–
20% DMSO/PBS solution, or 20% DMSO/PBS alone
(control mice). Groupsof 10 mice received a single treat-
ment of 1 of 3 different dose levels of Zidampidine
(100 lg, 1 mg, and 5 mg). Ten control mice were injected
intraperitoneally with 20% DMSO/PBS solution in
accordance with the experimental protocol. No sedation
or anesthesia was used throughout the treatment period.
Mice were monitored daily for mortality for determina-
tion of day 30 LD50 values. Mice surviving 30 days post-
treatment were sacrificed and the tissues were immedi-
ately collected and preserved in 10% neutral phosphate
buffered formalin. Mice were weighed at the time of sac-
rifice or death. At the time of necropsy the selected tis-
sues (bone, bone marrow, brain, spinal cord, cecum,
heart, kidney, large intestine, liver, lung, lymph node,
ovary, pancreas, skeletal muscle, skin, small intestine,
spleen, stomach, thymus, thyroid gland, urinary blad-
der, and uterus, as available) were immediately
collected from mice for histopathologic examination.
3.4. Statistical analysis
Statistical significance was determined using the Kap-
lan Meier Log-Rank test. Cumulative survival propor-
tionsobtained from usrvival curvesat 6, 9, 13, and
21-day time pointswere compared for all treatment.s
Z-Test was performed to calculate significant differ-
ences in the proportions that survived using the cumu-
lative proportion surviving and the standard error
values( z-score = (cumulative proportion 1 ꢁ cumula-
tive proportion 2)/square root(standard error 12 + stan-
dard error 22); p < 0.05 two tailed were deemed
significant).
Figure 5. Metabolic pathway of Zidampidine (30-azidothymidine-50-[p-bromophenyl methoxyalaninyl phosphate]). Zidampidine-M1 (Ala-AZT-MP =
30-azido-30-deoxythymidine-50-methoxyalaninyl phosphate) and Zidampidine-M2 (AZT = 30-azido-30-deoxythymidine).

3286
F. M. Uckun et al. / Bioorg. Med. Chem. 13 (2005) 3279–3288
For histopathologic studies, tissues were fixed in 10%
neutral buffered formalin, dehydrated, and embedded
in paraffin by routine methods. Glass slides with affixed
4–5 lm tissue sections were prepared and stained with
Hemotoxylin and Eosin (H&E).
3.7.2. Isolation of urinary metabolite (Zidampidine-M2).
Ten CD-1 mice were placed into a Nalgene metabolic
cage after being injected intravenously with 260 mg/kg
Zidampidine. Urine wascollected at 0–24 h. The urine
wascentrifuged and tsored at
analysis.
ꢁ20 ꢁC until further
3.6. HPLC determination of Zidampidine and its
metabolites
An aliquot of urine sample was extracted with ethyl ace-
tate and the ethyl acetate phase was transferred to a
clean tube. The ethyl acetate wasevaporated uisng a
steady stream of nitrogen and the residue was dissolved
in deuterated chloroform and used directly for NMR
analysis.
Zidampidine and itsmetaboliteswere determined by
adding 75 lL of methanol to 25 lL of plasma and vor-
texing for at least 30 s. Following centrifugation
(1500 rpm, 5 min), the supernatant was transferred into
a clean HPLC vial and 50 lL wasinjected for HPLC
analysis. The HPLC system was a Hewlett Packard
(Palo Alto, CA) series 1100 instrument consisting of a
quaternary pump, an auto sampler, an auto electronic
degasser, an automatic thermostatic column compart-
ment, diode array detector, and a computer having a
Chem Station software program for data analysis.
3.7.2.1. NMR analysis. 1H NMR spectra were run on
a Model Varian XL-400 NMR spectrometer operating
at 399.9 MHz, and employing a deuterium field-lock fre-
quency in the normal manner. The samples were dis-
solved in CDCl3 for NMR analysis.
3.8. Evaluation of pharmacokinetics of Zidampidine
The analytical column wasa Hewlett Packard,
250 · 4 mm Lichrospher 100, RP-18 (5 lM) column
and the guard column wasa 4 · 4 mm Lichrospher
100, RP-18 (5 lM). The mobile phase was degassed
automatically by the electronic degasser system. Before
the analysis, the column was equilibrated and a gradient
program was used for analysis of samples. The flow rate
wasmaintained at 1 mL/min and the column wasmain-
tained at room temperature. The linear gradient mobile
phase was 100% 10 mM ammonium phosphate buffer
(pH 3.7) at 0 min; 20% acetonitrile, 80% ammonium
phosphate buffer (pH 3.7) at 20 min; 32% acetonitrile,
68% water (0.1% HAC) at 30–60 min. The wavelength
of detection was set at 270 nm. Peak width, response
time, and slit were set at >0.03 min, 0.5 s, and 8 nm,
respectively.
Female CD-1 mice (7–9 weeksold) from CharlesRiver
Laboratories(Wilmington, MA) were housed in a con-
trolled environment (12-h light/12-h dark photoperiod,
22 1 ꢁC, 60 10% relative humidity), which isfully
accredited by the USDA (United StatesDepartment of
Agriculture). All rodentswere houesd in microios lator
caging systems (Lab Products, Inc., NJ) containing
autoclaved bedding. The mice were allowed free access
to autoclaved pellet food and tap water throughout
the experiments. Animal studies were approved by the
Parker Hughes Institute Animal Care and Use Commit-
tee, and all animal care proceduresconformed to the
Principlesof Laboratory Animal Care (NIH publication
#85-23, revised 1985).
Zidampidine was dissolved in DMSO and 50 lL of drug
solution was injected intravenously into each mouse.
Thisamount of DMSO waswell tolerated by mice when
administrated by rapid intravenously (iv) or extravascu-
lar injection.20,21 In pharmacokinetic studies, mice were
injected iv via the tail vein with a bolusdose of 240 mg/
kg Zidampidine. Blood samples (ꢂ200 lL) were ob-
tained from the ocular venousplexusby retro-orbital
veni-puncture at 0, 5, 10, 15, 30, 45 min and 1, 1.5, 2,
4, and 6 h following iv injection.
3.7. In vivo metabolism
3.7.1. Isolation and identification of plasma metabolite
(Zidampidine-M1). CD-1 mice were injected intrave-
nously with Zidampidine at dose of 240 mg/kg. After
10 min, all blood waswithdrawn from the ocular venous
plexusby retro-orbital veni-puncture, and the plams a
samples were collected. Acetonitrile was added to the
pooled plasma to precipitate the plasma protein. Subse-
quently, acetonitrile wasevaporated under a tseady
stream of nitrogen and the residue was further treated
with ethyl acetate (to remove other componentsfrom
the plasma samples). The above procedure was repeated
twice and the aqueouslayer wasesparated from the
ethyl acetate layer. The aqueouslayer, which contained
Zidampidine-M1, wasfurther lyophilized and the reis-
due was reconstituted in methanol. The reconstituted
All collected blood samples were heparinized and centri-
fuged at 7000g for 5 min in a microcentrifuge to obtain
plasma. Aliquots of plasma were used for extraction
within 2 h after withdrawal from the treated mice and
the extract wasusbjected for HPLC analyisasde-
scribed above.
solution was analyzed using
LC–MS instrument.
a
Hewlett Packard
3.9. Pharmacokinetic analysis
Pharmacokinetic modeling and pharmacokinetic
parameter calculationswere carried out using the phar-
macokinetic software, WinNonlin program, Profes-
sional Version 3.0. (Pharsight Inc., Mountain View,
CA).22,23 An appropriate pharmacokinetic model was
chosen on the basis of the lowest sum of weighted
The experimental conditions for mass spectrum analysis
were asfollows: fragmentor of 75, gasflow of 10 L/min,
nebulizer pressure of 25 psi, drying gas temperature of
350 ꢁC, peak width and gain were set at 0.03 s and 1.0,
respectively.

F. M. Uckun et al. / Bioorg. Med. Chem. 13 (2005) 3279–3288
3287
squared residuals, lowest Schwartz criterion (SC), low-
4. Conclusion
est AkaikeÕsinformation criterion (AIC) value, lowets
standard errors of the fitted parameters, and dispersion
of the residuals. The half-life was estimated by linear
regression analysis of the terminal phase of the plasma
concentration profile. The time (tmax) taken to achieve
peak concentration (Cmax) wascalculated using differen-
tial calculus.
In summary, treatment with Zidampidine at non-toxic
doses significantly improved the probability of survival
following Lassa challenge. Therefore, Zidampidine
shows clinical potential as a new agent for treatment
of viral hemorrhagic fevers caused by Lassa virus. The
elucidation of optimized prophylactic aswell aspots-
exposure treatment regimens will be the focus of our fu-
ture studies. It will also be important to determine if
Zidampidine is active against other viruses associated
with lethal viral hemorrhagic feversand/or encephelo-
myelitis, such as the Ebola viruses of the Filoviridae
family. Thisisthe firts report of pharmacokinetic and
metabolism profiles of Zidampidine and its metabolites
in mice. These pilot pharmacokinetic and metabolism
studiesaswell asthe preliminary toxicity studiesof this
aryl phosphate derivative of AZT combined with the
availability of the described quantitative HPLC method
for its detection in plasma provide the basis for future
preclinical evaluation of Zidampidine and itspotential
asan anti-viral agent.
3.10. Linearity and sensitivity of HPLC-based detection
method
Under the chromatographic separation conditions de-
scribed in Section 3.1, the retention times for Zidampi-
dine, Ala-AZT-MP and AZT were ꢂ56, 14.5, and
19.2 min, respectively. At the retention time of the
Zidampidine and the metabolites, no significant interfer-
ence peaks from blank plasma were observed (Fig. 3A
and B). The standard curve was linear over the concen-
tration–dose ranges tested. The linearity was statistically
confirmed using the Instat Program V3.0. The lowest
limit of detection of Zidampidine, Ala-AZT-MP, and
AZT was0.5, 1.0, 0.5 lM at a signal-to-noise ratio of
ꢂ4.
References and notes
3.11. Animal infection
1. Frame, J. D.; Baldwin, J. M., Jr.; Gocke, D. J., et al. Am.
J. Trop. Med. Hyg. 1970, 19, 670.
2. Frame, J. D. Bull. World Health Organ. 1975, 52, 593.
3. McCormick, J. B.; Webb, P. A.; Krebs, J. W., et al.
J. Infect. Dis. 1987, 155, 437.
4. Monath, T. P. Bull. World Health Organ. 1975, 52, 577.
5. Buckley, S. M.; Casals, J. Am. J. Trop. Med. Hyg. 1970,
19, 680.
6. Gunther, S.; Emmerich, P.; Laue, T., et al. Emerg. Infect.
Dis. 2000, 6, 466.
37
Animal infection wasperformed asdecsribed
in an
appropriate Animal BioSafety Level-3 Laboratory
(ABL-3) at BRIEM (Research Institute for Epidemiol-
ogy and Microbiology, MINSK, Belarus) with the
technician wearing appropriate facility clothing. The
culture wasthawed in a water bath at 37 ꢁC and then
diluted in normal saline to achieve the required concen-
tration. In thistsudy, all mice were challenged with
1000 PFU, which is100-timeshigher than the LD
dose. Each group of animals was placed in a separate
cage.
7. Hirabayashi, Y.; Oka, S.; Goto, H., et al. Nippon Rinsho
1989, 47, 71.
8. Mahdy, M. S.; Chiang, W.; McLaughlin, B., et al. Can.
Dis. Wkly. Rep. 1989, 15, 193.
50
9. Schmitz, H.; Kohler, B.; Laue, T., et al. Microbes Infect.
2002, 4, 43.
3.12. Lassa virus model
10. Shlaeffer, F.; Sikuler, E.; Keynan, A. Harefuah 1988, 114,
12.
11. Haas, W. H.; Breuer, T.; Pfaff, G., et al. Clin. Infect. Dis.
2003, 36, 1254.
12. Holmes, G. P.; McCormick, J. B.; Trock, S. C., et al. N.
Engl. J. Med. 1990, 323, 1120.
13. McCormick, J. B.; Walker, D. H.; King, I. J., et al. Am. J.
Trop. Med. Hyg. 1986, 35, 401.
14. Monson, M. H.; Frame, J. D.; Jahrling, P. B., et al. Trans.
R. Soc. Trop. Med. Hyg. 1984, 78, 549.
15. Fabiyi, A. Bull. Panam. Health Organ. 1976, 10, 335.
16. Webb, P. A.; McCormick, J. B.; King, I. J., et al. Trans.
R. Soc. Trop. Med. Hyg. 1986, 80, 577.
17. Monson, M. H.; Cole, A. K.; Frame, J. D., et al. Am. J.
Trop. Med. Hyg. 1987, 36, 408.
18. DÕCruz, O. J.; Waurzyniak, B.; Yiv, S. H., et al. Contra-
ception 2000, 61, 69.
CBA strain mice were intracerebrally infected with
1000 PFU of Lassa virus (Josiah strain) that resulted
in lethality of 80–100% of control (non-treated) animals
in 7–9 daysafter infection. Control animalswere given
physiologic salt solution as a placebo instead of the
compound. In general, for non-treated animals, clinical
signs of the disease manifested on the fifth and seventh
day by presenting: weight loss, immobility, disheveled
hair, convulsions, severe decubitus paralysis, and death.
All subjective measurement of decreased mobility and
scruffy fur were done in a blinded fashion as not to influ-
ence the results. The protective properties of the experi-
mental anti-viral agent were assessed by using the
following treatment-preventative regimen: mice were
treated either with vehicle or Zidampidine (25 mg/kg)
administered intraperitoneally 24 h prior, 1 h prior,
and 24, 48, 72, and 96 h after virusinoculation. Mice
were then observed for 21-days post infection. The pro-
tective effect of the experimental anti-viral drugswas
evaluated according to the rise of the survival rate and
prolongation of mean life of the experimental animals
19. Venkatachalam, T. K.; DÕCruz, O. J.; Uckun, F. M.
Antiviral Chem. Chemother. 2000, 11, 31.
20. DÕCruz, O. J.; Venkatachalam, T. K.; Uckun, F. M. Biol.
Reprod. 2000, 62, 37.
21. DÕCruz, O. J.; Zhu, Z.; Yiv, S. H., et al. Contraception
1999, 59, 319.
22. DÕCruz, O. J.; Shih, M. J.; Yiv, S. H., et al. Mol. Hum.
Reprod. 1999, 5, 421.
38
ascompared with the control animals.

3288
F. M. Uckun et al. / Bioorg. Med. Chem. 13 (2005) 3279–3288
23. Jan, S. T.; Zhu, Z.; Tai, H. L., et al. Antiviral Chem.
Chemother. 1999, 10, 47.
31. McGuigan, C.; Davies, M.; Pathirana, R. N.; Mahmood,
N.; Hay, A. J. J. Med. Chem. 1996, 39, 1748.
24. Balzarini, J.; Baba, M.; Herdewijn, P.; De Clercq, E.
Biochem. Pharmacol. 1988, 37, 2847.
25. Balzarini, J.; Herdewijin, P.; De Clercq, E. J. Biol. Chem.
1989, 264, 6127.
32. Curley, D.; McGuigan, C.; Devine, K. G.; OÕConnor, T. J.;
Jeffries, D. J.; Kinchington, D. Antiviral Res. 1990, 14, 345.
33. Devine, K. G.; McGuigan, C.; OÕConnor, T. J.; Nicholls,
S. R.; Kinchington, D. AIDS 1990, 4, 371.
26. McGuigan, C.; Devine, K. G.; OÕConnor, T. J.; Kinch-
ington, D. Antiviral Res. 1991, 15, 255.
27. McGuigan, C.; Pathirana, R. N.; Mahmood, N.; Devine,
K.; Hay, A. J. Antiviral Res. 1992, 17, 311.
28. McGuigan, C.; Devine, K. G.; OÕConnor, T. J.; Galpin, S.
A.; Jeffries, D. J.; Kinchington, D. Antiviral Chem.
Chemother. 1990, 1, 107.
34. McGuigan, C.; Slater, M. J.; Parry, N. R.; Perry, A.;
Harris, S. Bioorg. Med. Chem. Lett. 2000, 3, 645.
35. UCKUN, Fatih, M.; VIG, Rakesh, Aryl phosphate
derivativesof d4T having anti-HIV activity, European
Patent Number EP 1 090 018 B1.
36. Fidarov, F. M.; Surikova, L. E.; Erofeeva, N. I., et al.
Vop. Virusol. 1990, 35, 326.
29. McGuigan, C.; Davies, M.; Pathirana, R. N.; Mahmood,
N.; Hay, A. J. Antiviral Res. 1994, 24, 69.
37. IgnatÕev, G. M.; Kaliberov, S. A.; Pereboeva, L. A., et al.
Vop. Virusol. 1994, 39, 257.
30. McGuigan, C.; Pathirana, R. N.; Balzarini, J.; De Clercq,
E. J. Med. Chem. 1993, 36, 1048.
38. Uckun, F. M.; Petkevich, A. S.; Vassilev, A. O., et al.
BMC Infect. Dis. 2004, 4, 1.