Sulfonyl-containing aldophosphamide analogues as novel anticancer prodrugs targeted against cyclophosphamide-resistant tumor cell lines

Article abstract of DOI:10.1021/jm0304764

A series of sulfonyl-group containing analogues of aldophosphamide (Aldo) were synthesized as potential anticancer prodrugs that liberate the cytotoxic phosphoramide mustards (PM, IPM, and tetrakis-PM) via β-elimination, a nonenzymatic activation mechanism. Kinetic studies demonstrated that all these compounds spontaneously liberate phosphoramide mustards with half-lives in the range of 0.08-15.2 h under model physiological conditions in 0.08 M phosphate buffer at pH 7.4 and 37 °C. Analogous to Aldo, the rates of β-elimination in all compounds was enhanced in reconstituted human plasma under same conditions. The compounds were more potent than the corresponding phosphoramide mustards against V-79 Chinese hamster lung fibroblasts in vitro (IC₅₀ = 1.8-69.1 μM). Several compounds showed excellent in vivo antitumor activity in CD2F1 mice against both P388/0 (Wild) and P388/CPA (CP-resistant) tumor cell lines.

Full text of DOI:10.1021/jm0304764

J . Med. Chem. 2004, 47, 3843-3852
3843
Su lfon yl-Con ta in in g Ald op h osp h a m id e An a logu es a s Novel An tica n cer P r od r u gs
Ta r geted a ga in st Cyclop h osp h a m id e-Resista n t Tu m or Cell Lin es
Monish J ain, J unying Fan, Nesrine Z. Baturay, and Chul-Hoon Kwon*
Department of Pharmaceutical Sciences, College of Pharmacy and Allied Health Professions, St. J ohn’s University,
J amaica, New York 11439
Received September 22, 2003
A series of sulfonyl-group containing analogues of aldophosphamide (Aldo) were synthesized
as potential anticancer prodrugs that liberate the cytotoxic phosphoramide mustards (PM, IPM,
and tetrakis-PM) via â-elimination, a nonenzymatic activation mechanism. Kinetic studies
demonstrated that all these compounds spontaneously liberate phosphoramide mustards with
half-lives in the range of 0.08-15.2 h under model physiological conditions in 0.08 M phosphate
buffer at pH 7.4 and 37 °C. Analogous to Aldo, the rates of â-elimination in all compounds was
enhanced in reconstituted human plasma under same conditions. The compounds were more
potent than the corresponding phosphoramide mustards against V-79 Chinese hamster lung
fibroblasts in vitro (IC50 ) 1.8-69.1 µM). Several compounds showed excellent in vivo antitumor
activity in CD2F1 mice against both P388/0 (Wild) and P388/CPA (CP-resistant) tumor cell
lines.
Sch em e 1
In tr od u ction
Cyclophosphamide (CP) is a widely used anticancer
alkylating agent that requires activation in vivo. The
pharmacology and chemistry of CP has been exten-
sively studied and reviewed.1-3 CP undergoes initial
hepatic mixed function oxidase-catalyzed activation to
4-hydroxycyclophosphamide (4-OH CP) that undergoes
ring-opening to aldophosphamide (Aldo), followed by
liberation of cytotoxic phosphoramide mustard (PM) and
acrolein by â-elimination. The cytotoxic activity of CP
is attributed to the reactive aziridinium ion derived from
PM, the ultimate alkylating species that cross-links
interstrand DNA (Scheme 1). Acrolein, a byproduct of
â-elimination, does not contribute significantly toward
anticancer activity of CP, although it is highly toxic to
cultured tumor cells. However, acrolein is responsible
for hemorrhagic cystitis, a side effect observed during
CP therapy.
Release of PM and acrolein from Aldo due to â-elim-
ination is subject to general base catalysis4 and may also
be catalyzed by intracellular 3′-5′ exonucleases.5,6 The
â-elimination of PM from 4-OHCP/Aldo is also acceler-
ated in plasma due to human serum albumin (HSA).7
Such influence of HSA on the pharmacokinetic profiles
of 4-OH CP/Aldo and PM metabolites might be of
importance in cancer chemotherapy of CP.
selectivity of CP is unclear and remains to be elucidated.
One of the hypotheses suggests that AlDH enzyme-
mediated conversion of Aldo to carboxyphosphamide is
less efficient in drug-sensitive tumor cells than normal
cells. As a result, Aldo liberates cytotoxic PM preferen-
tially in the tumor cells.8-11
Ifosfamide (IFA) (Chart 1), a structural isomer of CP,
has been developed as a useful anticancer agent with
greater activity than CP in certain experimental and
human tumors as well as lack of complete cross-
resistance with CP-resistant tumor cells.1,12-15 Analo-
gous to CP, IFA requires metabolic activation to its C-4
hydroxylated metabolite that equilibrates with its al-
dehyde tautomer (aldoifosfamide) and ultimately yields
cytotoxic isophosphoramide mustard (IPM). In contrast
to CP, IFA suffers substantial deactivation due to
One of the primary deactivation pathways of CP
results from oxidation of Aldo to inactive carboxy-
phosphamide due to enzyme aldehyde dehydrogenase
(AlDH).8,9 Elevation of either class 1 or class 3 AlDH
isozymes has been implicated in development of resist-
ance to 4-OH CP/Aldo but not PM in certain murine and
human tumor cell lines. The reason for high oncotoxic
* To whom correspondence should be addressed. Department of
Pharmaceutical Sciences, College of Pharmacy and Allied Health
Professions, St. J ohn’s University, 8000 Utopia Parkway, J amaica, NY
11439. Office: (718)-990-6161 x 5214, fax: (718)-990-6551, e-mail:
kwonc@stjohns.edu.
10.1021/jm0304764 CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/17/2004

3844 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
J ain et al.
Ch a r t 2. Structures of Substituted Sulfonylethyl
Ch a r t 1. Structures of Ifosfamide,
4-Hydroperoxycyclophosphamide, and
Chloroacetaldehyde
Phosphorodiamidates
do not depend on the action of hepatic mixed function
oxidases for their activation. Potential differences in
routes of metabolism between IFA and sulfonyl pro-
drugs might lead to decreased side-chain N-dealkylation
pathway in the latter, thereby minimizing generation
of potentially toxic chloroacetaldehyde. In contrast to
Aldo, the designed sulfonyl prodrugs are not likely to
be accepted as substrates by the enzyme aldehyde
dehydrogenase and therefore have the potential to be
active against CP-resistant tumors. Design of nonalde-
hyde prodrugs that would circumvent AlDH-mediated
deactivation and release phosphoramide mustards spon-
taneously without activation has been previously ex-
amined in phenylketoethyl phosphorodiamidate (phen-
ylketophosphamide)10,34 or perhydrooxazine analogues35
that liberate cytotoxic species directly, or via enamine
intermediates, respectively. In both cases, in vitro and
in vivo antitumor activity was achieved; however,
significant advantages over CP were not reported, with
toxicity observed for phenylketophosphamide.10
Sch em e 2
oxidative N-dealkylation metabolism on the chloroethyl
side chain to yield inactive metabolites, 2-dechloroethyl
ifosfamide and 3-dechloroethyl ifosfamide, along with
chloroacetaldehyde (Chart 1). It has been suggested that
the latter is the cause of the nephrotoxic16-18 and
neurotoxic19,20 side effects of IFA.
Despite the observation that extracellularly delivered
PM derivatives21,22 have demonstrated activity against
experimental tumors, it is well accepted that prodrugs
of PM (and IPM) will provide better chemotherapeutic
agents due to several reasons. Phosphoramide mustard
is chemically labile (t1/2 ) 18 min in 0.1 M HEPES buffer
at 37 °C)23 and has a narrow therapeutic index,24 and
it is shown that extracellularly generated PM would not
be readily transported to intracellular environment due
Resu lts a n d Discu ssion
Several substituted sulfonylethyl analogues were
examined as novel prodrugs of PM and its derivatives
(Chart 2). To optimize rate of â-elimination, the R
substituent was varied; electron-donating groups such
as methyl and p-tolyl would decrease the rate, whereas
electron-withdrawing phenyl and p-nitrophenyl groups
would increase the rate. The liberated vinyl sulfone may
produce toxicity of its own due to the chemical reactivity,
which is analogous to the case with acrolein; however,
the reactivity could be modulated by introduction of a
moderate to bulky substituent, e.g., R′ ) methyl or
phenyl group to the â-carbon.
to its anionic character at physiological pH (pKa
)
4.5).3,25 Aldophosphamide appears to be an appropriate
prodrug of PM that does not require metabolic activa-
tion. However, Aldo is difficult to synthesize and
isolate.26 Therefore, properly designed stable prodrugs
of PM and its derivatives can serve as useful anticancer
alkylating agents.
Since the first report of Kader and Stirling,27 who
described 2-(4-toluenesulfonyl)ethyloxycarbonyl (Tsc) as
an amine protecting group, a number of alkyl- or aryl-
substituted sulfonylethyloxycarbonyl groups have been
proposed for carboxyl as well as amino protection.28-30
The success of this method is based on the fact that
â-elimination is readily achieved under weakly alkaline
conditions, provided the leaving group is situated in a
position â to a sulfonyl group. The rate of that â-elim-
ination varies with different substituents in the sulfonyl
activating group31,32 and also changes in the leaving
group.33 As discussed earlier (Scheme 1), the further
activation of Aldo to PM follows the same â-elimination
mechanism except that the electron-withdrawing group
is an aldehyde and the leaving group is the cytotoxic
PM.
Ch em istr y. A model compound 2-(methylsulfonyl)-
ethyl 2,4-dinitrophenyl ether (2) was chosen for pre-
liminary studies to determine whether rate of â-elimi-
nation for substituted sulfonylethyl compounds is suitable
under physiological conditions. The compound was
designed on the basis that the â-elimination product,
2,4-dinitrophenol (2,4-DNP), has a visible chromophore,
and therefore its â-elimination from 2 can be easily
quantitated using a spectophotometric technique. Com-
pound 2 was synthesized by first reacting 2-(methyl-
thio)ethanol with 2,4-dinitrofluorobenzene and Et3N to
give 1, which in turn was oxidized by hydrogen peroxide/
ammonium molybdate in acetone (Scheme 3).
The 2-thioethanol starting materials for generation
of sulfonylethyl phosphorodiamidates were either avail-
able commercially or prepared according to published
procedures. Reactions of thiocresol or p-nitrothiophenol
with 2-chloroethanol gave 2-thioethanols 17 and 22,
respectively. Reaction of 1-mercapto-2-propanol with
methyl iodide yielded 9, and reaction of styrene oxide
with sodium methylthiolate provided 12.
On the basis of the above analogy, substituted sulfo-
nylethyl phosphorodiamidates (Scheme 2) might serve
as novel prodrugs of phosphoramide mustards (PM,
IPM, and tetrakis-PM) with several potential advan-
tages over CP and related analogues. These prodrugs

Aldophosphamide Analogues as Anticancer Prodrugs
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3845
Sch em e 3a
Sch em e 6a
a
Reagents: (a) CH3I, NaOH; (b) 2,4-dinitrofluorobenzene, Et3N;
(c) H2O2, ammonium molybdate.
Sch em e 4a
a
Reagents: (a) HCl‚HN(CH2CH2Cl)2, Et3N, CH2Cl2; (b) Et3N,
CH2Cl2 (c) t-BuOOH; (d) H2O2, ammonium molybdate; (e) Pd/C,
H2, 1 atm.
a
Reagents: (a) NaH, anhyd THF; (b) NH3; (c) H2O2, ammonium
molybdate.
Sch em e 5a
a
Reagents: (a) P(O)Cl3, Et3N, CH2Cl2, or benzene; (HCl‚H2NCH2-
CH2Cl, Et3N, oe benzene; (c) H2O2, ammonium molybdate.
F igu r e 1. â-Elimination profile of 2 in 0.08 M phosphate
buffer at pH 6.4 (1), 7.4 (b), and 8.4 (O) at 37 °C.
The prodrugs contained either N,N-bis(2-chloroethyl)
(5, 11, and 19) or N,N′-bis(2-chloroethyl) (6, 16, 21, and
24) phosphorodiamidates that upon â-elimination would
release the bis-alkylating mustard PM or IPM (Scheme
2), respectively. The release of potentially more potent
alkylating agent tetrakis-PM (Scheme 2) was sought by
design of compounds containing N,N,N′,N′-tetrakis(2-
chloroethyl) phosphorodiamidate (8, 14, and 26). Suc-
cessive reaction of appropriate 2-thioethanol with so-
dium hydride, N,N-bis(2-chloroethyl)phosphoramidic
dichloride, and ammonia gave N,N-bis(2-chloroethyl)-
phosphorodiamidate intermediates (4, 10, and 18)
(Scheme 4). Alternatively, the 2-thioethanol starting
materials were treated with phosphorus oxychloride,
followed by reaction of the mixture with 2 equiv of
2-chloroethylamine to yield N,N′-bis(2-chloroethyl) (15,
20, and 23) intermediates (Scheme 5). The obtained
thioethyl N,N-bis(2-chloroethyl)/N,N′-bis(2-chloroethyl)
intermediates were oxidized with excess H2O2 and
catalytic ammonium molybdate to give corresponding
sulfonylethyl phosphorodiamidates.
modified in case of analogue 6. Reaction of 2-(methyl-
sulfonyl)ethanol with phosphoryl chloride and 2-chloro-
ethylamine afforded 6 directly (Scheme 5).
Phosphorus trichloride was sequentially reacted with
2 equiv of bis(2-chloroethyl)amine and the appropriate
alcohol, and the resulting phosphorodiamidite was
oxidized with tert-butyl hydroperoxide to give interme-
diates thioethyl N,N,N′,N′-tetrakis(2-chloroethyl) phos-
phorodiamidates 7, 13, and 25 or benzyl N,N,N′,N′-
tetrakis(2-chloroethyl) phosphorodiamidate 29. The
obtained thioethyl intermediates were oxidized with
H2O2/ammonium molybdate to afford the corresponding
sulfonyl products 8, 14, and 26. Reduction of 29 yielded
tetrakis phosphoramide mustard 30 (Scheme 6).
Sta bility Stu d ies. Model sulfonyl compound 2 un-
derwent base-catalyzed â-elimination of 2,4-DNP with
half-life (t1/2) of 13.9 min at pH 8.4, 68 min at pH 7.4,
and 19.25 h at pH 6.4 in 0.08 M phosphate buffer at 37
°C (Figure 1). Additional studies with corresponding
sulfide 1 at similar pH values showed that the analogue
was stable for at least 24 h, thereby confirming the
importance of sulfonyl moiety as the activating group
in the â-elimination process. Having established that 2
Reaction of 2-(methylthio)ethanol with phosphoryl
chloride and 2-chloroethylamine did not yield the de-
sired thioethyl N,N′-bis(2-chloroethyl)phosphorodiami-
date intermediate; therefore, the synthetic strategy was

3846 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
J ain et al.
Ta ble 1. Half-Lives under Model Physiological Conditions and
in Vitro Cytotoxicity Profiles against V-79 Cellsa for Sulfonyl
Prodrugs and Phosphoramide Mustards
hindrance of â-methyl to the incoming nucleophile, may
explain slower base-catalyzed â-elimination in 11. The
underlying mechanism for enhancement of â-elimina-
tion due to the â-phenyl substituent could not be readily
explained. Presumably, the electron-inductive effects of
the â-phenyl, that enhance acidity of the R-proton and
accelerate elimination, outweigh the steric hindrance
effects it offers to the incoming nucleophile.
Among sulfonyl analogues, only the p-nitrophenyl-
sulfonyl compounds 24 and 26 showed comparable or
faster â-elimination rates than 4-OH CP/Aldo. Other
sulfonyl analogues showed greater stability than Aldo
under model physiological conditions. The data suggests
that sulfonyl analogues are chemically stable and may
serve as useful prodrugs of phosphoramide mustards.
The activating group confers the elimination reactivity
in the following order: p-nitrophenylsulfonyl g aldehyde
> phenylsulfonyl ≈ methylsulfonyl > p-tolylsulfonyl.
half-life (h)b
phosphate
bufferc
human
compound
plasmad
IC50 (µM)b
5
6
8
11
14
16
19
21
24
26
5.9 ( 0.6
6.4 ( 0.6
7.2 ( 1.2
15.2 ( 0.3
1.8 ( 0.2
6.4 ( 0.3
12.8 ( 2.8
12.1 ( 1.1
2.2(2.3 ( 0.3)e 56.5 ( 5.6
2.2 ( 0.1
1.6 ( 0.1
6.7 ( 0.7
1.0 ( 0.1
2.6 ( 0.2
4.8 ( 0.5
4.4 ( 0.4
68.0 ( 9.7
24.1 ( 3.3
69.1 ( 9.8
15.2 ( 2.4
13.9 ( 1.5
11.5 ( 1.1
9.5 ( 2.0
2.1 ( 0.5
1.8 ( 0.2
3.3 ( 0.8
77.0 ( 0.0
60.9 ( 5.8
42.6 ( 4.3
0.08 ( 0.01 < 0.01
0.69 ( 0.02
0.37 ( 0.16
4-OH CP/Aldof
27 (PM)
28 (IPM)
0.75g
0.25g
0.30h
1.0i
30 (Tetrakis-PM)
a
b
Drug exposure time: 3 h. Mean ( SD of three experiments.
It might be argued that sulfonylethyl phosphorodi-
amidates liberate phosphoramide mustards and as a
result disappear under model physiological conditions
due to mechanisms different than the suggested â-
elimination pathway such as hydrolysis of the phos-
phate ester linkage of the phosphorodiamidate group.
Thioethyl phosphorodiamidate analogues such as 4 and
18 lack a sulfonyl group and should be stable toward
â-elimination. These analogues were found to be stable
for at least 16 h in 0.08 M phosphate buffer (pH 7.4, 37
°C), suggesting that phosphate ester hydrolysis does not
contribute to the disappearance of sulfonylethyl phos-
phorodiamidates.
Aldophosphamide undergoes approximately 250-fold
faster decomposition in plasma than in phosphate buffer
alone, and human serum albumin (HSA) was identified
as the catalyst responsible for accelerated â-elimination
of PM from Aldo.7 By analogy, since, sulfonyl analogues
are expected to employ the same â-elimination mecha-
nism to generate phsophoramide mustards, they were
examined for similar enhanced decomposition behavior
in plasma. Interestingly, sulfonyl analogues underwent
accelerated decomposition in plasma (and HSA for 5)
than in phosphate buffer with the extent of acceleration
similar to that observed in Aldo.
c 0.08 M phosphate buffer. Human plasma reconstituted in 0.08
M phosphate buffer, pH 7.4, 37 °C. Half-life in human serum
d
e
albumin (Fraction V, 60 mg/mL) in 0.08 M phosphate buffer, pH
7.4, 37 °C. f 4-OOH CP, a precursor form, that readily converts to
4-OH CP/Aldo in aqueous solution was used for cytotoxicity testing.
Half-life values at pH 7.4, 37 °C in g0.1 M phosphate buffer, h0.1
M HEPES buffer, and i0.2 M phosphate buffer obtained from
literature.23
undergoes â-elimination under physiological pH, we
synthesized several substituted sulfonylethyl phos-
phorodiamidates as prodrugs (Chart 2), anticipating
their ability to yield cytotoxic phosphoramide mustards
as anticancer agents (Scheme 2). Rates of â-elimination
of phosphoramide mustards in sulfonyl prodrugs were
determined under model physiological conditions by
monitoring the rate of disappearance of prodrug upon
incubation in 0.08 M phosphate buffer, pH 7.4 and 37
°C, with the use quantitative TLC technique. Results
are presented in Table 1. The half-lives of 2-(methyl-
sulfonyl)ethyl phosphorodiamidates (5, 6, and 8) were
essentially similar (5.9-7.2 h) and showed greater
stability than model compound 2, apparently due to
change of the nature of the leaving group, i.e., 2,4-DNP
versus phosphorodiamidic acid. Changes within phos-
phorodiamidate moiety itself did not affect the elimina-
tion rate and were found similar for N,N-bis(2-chloro-
ethyl), N,N′-bis(2-chloroethyl), or N,N,N′,N′-tetrakis(2-
chloroethyl)phosphorodiamidic acid leaving groups. The
phenylsulfonyl compound 16 showed a half-life of 6.4 h
that was comparable to that of methylsulfonyl ana-
logues. The electron-donating p-tolylsulfonyl substituent
in analogues 19 and 21 reduces acidity of the R-proton,
thereby decreasing the â-elimination reaction rate (t1/2
) approximately 12 h) as compared to 16. Conversely,
the electron-withdrawing p-nitrophenylsulfonyl sub-
stituent in 24 and 26 enhanced the â-elimination rate
and showed a much shorter half-life (t1/2 ) 0.08 and 0.69
h, respectively) than 16. The â-methyl-substituted
compound 11 with a half-life of 15.2 h appeared more
stable than the corresponding non-â-substituted ana-
logues. The â-phenyl-substituted prodrug 14 showed a
short half-life of 1.8 h. The observation that â-methyl
substituent decreases and â-phenyl substituent en-
hances the â-elimination rate in sulfonylethyl com-
pounds is consistent with previously reported data.36
The electron-inductive effects of â-methyl that diminish
acidity of the R-proton, in conjunction with steric
The catalytic rate constants kcat under pH 7.4 and 37
°C for disappearance of 5 in phosphate buffer (0.08 M)
and human serum albumin (HSA, mol. wt. 66 000 Da;
actual concentration 48 mg/mL or 7.3 × 10-4 M) were
1.46 and 253 M-1 h-1, respectively, where, kcat ) kobs
/
concentration of phosphate buffer or concentration of
HSA. In case of phosphate, kobs (1.17 × 10-1 h-1) was
obtained from stability studies. The kobs (3.01 × 10-1
h-1) in HSA was corrected for contribution from phos-
phate and the result (3.01 × 10-1 h-1 - 1.17 × 10-1 h-1
) 1.84 × 10-1 h-1) used to calculate kcat in HSA. Based
on kcat values, the rate of decomposition of 5 due to HSA
is approximately 175-fold faster than in phosphate
buffer.
Analogous to Aldo, HSA appears to be the catalytic
component of human plasma that enhances â-elimina-
tion in 5 and other sulfonyl analogues. The significance
of albumin catalytic activity in CP metabolism and its
possible role in oncotoxic specificity of CP remains to
be explored. However, the sulfonyl analogues undergo-
ing similar acceleration of PM generation in albumin

Aldophosphamide Analogues as Anticancer Prodrugs
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3847
Ta ble 2. Antitumor Activity of Sulfonylethyl
Phosphorodiamidates against P388/0 and P388/CPA Leukemia
in Vivo
might share the pharmacokinetic properties of 4-OH CP/
Aldo and PM and also share their therapeutic profile.
In Vitr o Cytotoxicity. Sulfonyl analogues were
evaluated for in vitro cytotoxicity against Chinese
hamster lung fibroblast (V-79) cells in a clonogenic assay
using a 3 h drug exposure period. The drug concentra-
tion that produced a 50% reduction in clonogenic
survival (IC50) was determined from the slope of the log
survival versus concentration curves. The results are
presented in Table 1. Sulfonyl analogues are expected
to exert their cytotoxicity by the release of phosphor-
amide mustards; therefore, their ability to enter the cells
and the rate at which they liberate phosphoramide
species would influence their cytotoxic potential.
Methylsulfonyl analogues (5, 6, 8, 11, and 14) showed
appreciable cytotoxic activity with IC50 values ranging
between 15.2 µM and 68.0 µM. Increase in cytotoxic
potential was observable due to the presence of ad-
ditional nitrogen mustard in 8 or enhancement in
lipophilicity and â-elimination rates as in 14. In general,
irrespective of their â-elimination rates, aromatic ring-
substituted sulfonyl analogues showed greater cytotoxic
activity (IC50 ) 1.8-13.9 µM) than the corresponding
methylsulfonyl compounds. Presence of aromatic group
confers lipophilic character to the analogue and there-
fore may allow its easy transport across tumor cell
membrane. In comparison to 5, 6, and 8, the increased
cytotoxicity for 16, 19, and 21 may be attributed to their
greater lipophilicity, whereas nitrophenylsulfonyl com-
pounds 24 and 26 show substantial cytotoxicity, most
likely due to their lipophilic character as well as rapid
â-elimination rates. Nitrophenylsulfonyl analogues were
found even more potent than 4-hydroperoxy CP control
that spontaneously generates 4-OH CP/Aldo in aqueous
media. All sulfonyl analogues showed better cytotoxic
activity (IC50 ) 1.8-69.1 µM) than the corresponding
phsophoramide mustards (IC50 ) 42.6-77.0 µM), thereby
suggesting that neutral sulfonyl analogues might serve
as useful transport forms that liberate cytotoxic phos-
phoramide mustards intracellularly.
In Vivo An titu m or Activity. Antitumor activity of
sulfonyl analogues was assessed in vivo against P388/0
(Wild) and P388/CPA (CP-resistant) leukemia in mice
and efficacy measured by increase in life span. Ana-
logues were tested at three doses of 100, 200, and 400
mg/kg. Isophosphoramide mustard (IPM) at its maxi-
mum tolerable dose (LD10) of 100 mg/kg or CP at 200
mg/kg dose served as positive controls.33 The results are
presented in Table 2.
All the sulfonyl analogues tested for in vivo antitumor
activity against P388/0 (Wild) and P388/CPA (CP-
resistant) leukemia in mice were active, thereby sup-
porting the hypothesis that sulfonylethyl phosphorodi-
amidates can serve as useful prodrugs of phosphoramide
mustards. Analogues 5, 6, 19, 24, and 26 showed in vivo
antitumor activity against P388/0. At 200 mg/kg dose,
24 and 26 showed similar or better antitumor activity
against P388/0 than CP with 2/5 long-term survivors
for both analogues. At 400 mg/kg level, 5, 6 (1/5 long-
term survivor), and 19 showed equivalent or better
activity than IPM. Only 24 and 26 displayed toxicity at
400 mg/kg, whereas IPM was toxic at 200 mg/kg level.
% ILSa
P388/0 (wild):
dose (mg/kg)
P388/CPA (resistant):
dose (mg/kg)
compound 100
200
400
100
28
100 (1/5) 20
200
52
400
5
36 73
91
100
108
60
36
60
60
44
36
6
8
33 52
10 10
10 10
10 20
33 33
20 50
23 14
71 138 (2/5) toxic
100 120 (2/5) toxic
92 toxic
84
28
20
20
28
28
12
30
30
20
33
90
23
12
12
4
20
28
4
11
14
16
19
21
24
26
IPM
CP
60
92 (2/5) toxic
84 (1/5) 68 (3/5) toxic
128 (2/5)
toxic
a
Numbers in parentheses indicate long-term (33-day) survivors.
against P388/CPA in vivo with considerably superior
activity as compared to CP. Whereas CP did not show
activity against P388/CPA, p-nitrophenylsulfonyl ana-
logues 24 and 26 were most potent with 2/5 and 3/5
long-term survivors, respectively, at 200 mg/kg dose.
Similar to activity against P388/0, 5 and 6 also dem-
onstrated excellent antitumor activity against P388/
CPA. Compounds 8, 14, and 16 were more potent
against CP-resistant P388 than wild-type P388. P388/
CPA was not cross-resistant toward any of the tested
sulfonyl analogues, most likely due to nonsusceptibility
of sulfonyl analogues to AlDH-mediated deactivation.
Among the tested prodrugs, only 24 and 26 displayed
toxicity at the highest level, and better antitumor
activity is expected from other agents against both
P388/0 and P388/CPA at higher doses. In view of these
data, it is reasonable to suspect that AlDH-mediated
preferential deactivation of alodophosphamide in nor-
mal versus tumor cells may not be the sole contributing
factor that governs selective in vivo antitumor activity.
Detailed studies with sulfonyl prodrugs may provide
additional insights into the role of AlDH with respect
to mechanisms of tumor-selective toxicity of prodrugs
of phosphoramide mustards.
Con clu sion s
Several sulfonyl group-containing stable analogues of
aldophosphamide (Aldo) have been examined as novel
prodrugs of phosphoramide mustard (PM) derivatives.
These compounds do not require enzyme-mediated
activation but require only base-catalyzed â-elimination
to liberate cytotoxic species.
Analogues with electron-withdrawing substituents on
the sulfonyl group showed faster â-elimination rates as
compared to analogues containing electron-donating
substituents. Despite substituent dependent variation
in decomposition half-lives of sulfonyl compounds, the
elimination behavior was found strikingly similar to
Aldo such that acceleration in â-elimination rate was
observed in the presence of human plasma for all
analogues. The in vitro cytotoxicity potential of the
sulfonyl analogues seemed to rely on their lipophilicity
as well as the elimination rate characteristics. All
sulfonyl analogues demonstrated in vivo antitumor
activity. Furthermore, in vivo antitumor activity was
Consistent with the original hypothesis, sulfonylethyl
phosphorodiamidates were effective anticancer agents

3848 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
J ain et al.
added dropwise to a suspension of sodium hydride (0.74 g, 18
mmol; a 60% dispersion in mineral oil) in dry THF (60 mL) at
ice-bath temperature with stirring over a 30 min period. The
reaction mixture was stirred at ice-bath temperature for 3 h.
The resulting pale, cloudy suspension was added dropwise to
a solution of 3 (4.6 g, 18 mmol) in dry THF (50 mL) at ice-
bath temperature. The reaction mixture was stirred for 4 h
while warming to room temperature and evaporated in vacuo
to give a pale yellow viscous liquid. This liquid was dissolved
in CH2Cl2 (50 mL) and bubbled with ammonia for 45 min at
room temperature. The resulting milky liquid was evaporated
in vacuo and diluted with anhydrous diethyl ether (50 mL),
and the resulting precipitates were filtered. The filtrate was
evaporated in vacuo to yield a viscous, yellow oil. The crude
product was purified on a flash column with eluent EtOAc to
give 1.4 g (22% yield) of 4 as a yellow oil: TLC Rf 0.53 in
EtOAc; 1H NMR (CDCl3) δ 2.07 (s, 3H), 2.66 (t, J ) 9 Hz, 2H),
3.00-3.66 (m, 8H), 4.07 (q, J ) 10.4 Hz, 2H).
equal to or better than that of CP against wild-type
P388 cells. Consistent with the original hypothesis,
sulfonyl prodrugs were substantially superior to CP
against CP-resistant P388 cell line in vivo.
Exp er im en ta l Section
Melting points were determined on a Thomas-Hoover capil-
1
lary melting point apparatus and were uncorrected. H NMR
spectra were recorded on a Varian EM-360 (60 MHz) instru-
ment, and chemical shifts are reported as δ values (ppm)
downfield from tetramethylsilane as an internal standard.
NMR abbreviations used are as follows: s (singlet), d (doublet),
t (triplet), q (quartet), m (multiplet). Elemental analyses were
performed by Atlantic Microlab, Inc. Norcross, GA. Silica gel
GF plates (Analtech) were used for TLC (250 µm, 2.5 × 10
cm) and preparative TLC (1000 µm, 20 × 20 cm). Silica gel
(40 µm, Baker) was used for flash column chromatography.
NBP spray was used for the detection of potential alkylating
compounds as follows: the plates were sprayed with 5% 4-(p-
nitrobenzyl)pyridine (NBP) in acetone, heated at 100 °C for 5
min (20 min for the kinetic studies), and sprayed with 5%
methanolic KOH. Alkylating agents are indicated by the
appearance of a blue chromophore. Spectrophotometric analy-
sis was performed on a Milton Roy spectronic 301 spectropho-
tometer. Quantitative TLC analyses were done on an Analtech
Uniscan Video Densitometer. All organic reagents and solvents
were reagent grade and purchased from commercial vendors.
2-(Meth ylth io)eth yl 2,4-d in itr op h en yl eth er (1) was
prepared according to the published procedure37 with some
modifications. 2-(Methylthio)ethanol (1.2 g, 1.2 mL, 13 mmol)
and triethylamine (1.5 mL, 10.8 mmol) were added to 2,4-
dinitrofluorobenzene (1.0 g, 5.3 mmol). Benzene (25 mL) was
then added and the dark reaction mixture stirred overnight
at room temperature. The reaction mixture was then heated
under reflux with stirring for 6 h, cooled, and acidified with
15 mL of 1 N HCl. The organic phase was separated and dried
over anhydrous sodium sulfate and the filtrate evaporated to
give a yellow colored residue. The residue was dissolved in 25
mL of diethyl ether and washed with 2 N aqueous NaHCO3
(4 × 15 mL). The ether layer was evaporated in vacuo and
the residue recrystallized in methanol to afford 0.45 g (33%
yield) of 1 as a light yellow crystalline product: mp 53-55
°C; TLC Rf 0.57 in EtOAc/hexane (1:1); 1H NMR (CDCl3) δ 2.15
(s, 3H), 2.86 (t, J ) 6.6 Hz, 2H), 4.33 (t, J ) 9 Hz, 2H), 7.15
(d, J ) 14.4 Hz, 1H), 8.31 (d, J ) 16.2 Hz, 1H), 8.66 (s, 1H).
2-(Met h ylsu lfon yl)et h yl 2,4-d in it r op h en yl et h er (2)
was prepared according to the procedure of Hardy et al.28 with
some modifications. The corresponding sulfide 1 (0.39 g, 1.5
mmol) was dissolved in acetone (20 mL) and water (10 mL),
treated with aqueous hydrogen peroxide (30 wt %, 21 mL, 200
mmol), and mixed with 0.3 M ammonium molybdate (3 mL).
The dark red reaction mixture was stirred for 3 h at room
temperature after which the acetone was evaporated in vacuo
and the residue partitioned between CH2Cl2 (50 mL) and water
(50 mL). The aqueous phase was washed with CH2Cl2 (3 × 25
mL). The pooled CH2Cl2 extracts were washed with saturated
brine (2 × 75 mL) and dried over anhydrous sodium sulfate,
and the filtrate was evaporated under reduced pressure to
yield pale yellow, shiny plates. The product separated out as
white solids during an attempt to dissolve the crude product
in 25 mL of THF for purification by preparative TLC. The
solids were filtered and dried to afford 0.2 g of pure 2 (47%
yield): mp 150-155 °C; TLC Rf 0.50 in EtOAc/hexane (1:2);
1H NMR (CDCl3/DMSO 5:1) δ 3.12 (s, 3H), 3.62 (t, J ) 3.6 Hz,
2H), 4.76 (t, J ) 8.4 Hz, 2H), 7.64 (d, J ) 12 Hz, 1H), 8.31 (d,
J ) 16.2 Hz, 1H), 8.66 (s, 1H). Anal. (C9H10N2O7S) C, H, N, S.
N,N-Bis(2-ch lor oeth yl)p h osp h or a m id ic d ich lor id e (3)
was synthesized according to the published procedure38 in 84%
yield.
2-(Meth ylsu lfon yl)eth yl N,N-bis(2-ch lor oeth yl)p h os-
p h or od ia m id a te (5) was prepared from 4 on a 2.0 mmol (0.7
g) scale as described in 2 with the same quantities of aqueous
H2O2 (30 wt %, 21 mL, 200 mmol) and 0.3 M ammonium
molybdate (3 mL) used in the procedure. The product was
obtained as a viscous, yellow oil that was crystallized from
anhydrous diethyl ether to give 0.35 g (50% yield) of 5 as white
1
solids: mp 83-86 °C; TLC Rf 0.45 in EtOAc/EtOH (10:1); H
NMR (CDCl3) δ 3.0 (s, 3H), 3.11-3.90, (m, 10H), 4.39 (q, J )
6.6 Hz, 2H). Anal. (C7H17Cl2N2O4SP) C, H, Cl, N, S.
2-(Meth ylsu lfon yl)eth yl N,N-bis(2-ch lor oeth yl)p h os-
p h or od ia m id a te (6) was synthesized according to published
procedure39 with some modifications. A solution of 2-(methyl-
sulfonyl)ethanol (3.72 g, 30 mmol) and triethylamine (3.10 g,
30 mmol) in CH
2Cl2 (30 mL) was added dropwise to a solution
of phosphorus oxychloride (4.60 g, 30 mmol) in CH2Cl2 (50 mL)
at ice-bath temperature over a 20 min period. The resulting
mixture was stirred at room temperature for 1 h and then
cooled in an ice-bath. 2-Chloroethylamine monohydrogen
chloride (6.96 g, 60 mmol) was added, followed by a dropwise
addition of triethylamine (12.2 g, 120 mmol) solution in CH2Cl2
(40 mL) over a 40 min period. The reaction mixture was stirred
overnight at room temperature, and the resulting white solids
were filtered off. The filtrate was washed with 5% HCl (50
mL), 5% NaHCO3 (50 mL), and saturated brine (2 × 100 mL),
dried over anhydrous Na2SO4, and filtered. The filtrate was
evaporated in vacuo, and the crude residue was purified on a
flash column with eluent EtOAc/EtOH (10:1) to give a colorless
oil. The oil was crystallized with EtOAc/hexane to obtain 1.1
g (11% yield) of 6 as white solids: mp 58-60 °C; TLC Rf 0.43
1
in EtOAc/EtOH (10:1); H NMR (CDCl3) δ 3.0 (s, 3H), 3.07-
3.77 (m, 12H), 4.45 (q, J ) 8.8 Hz, 2H). Anal. (C7H17Cl2N2O4-
SP) C, H, Cl, N, S.
2-(Meth ylth io)eth yl N,N,N′,N′-tetr a k is(2-ch lor oeth yl)-
p h osp h or od ia m id a te (7) was prepared according to pub-
lished procedure40 with some modifications. Phosphorus tri-
chloride (5.0 mL, 2.0 M in CH2Cl2, 10.0 mmol) was diluted in
CH2Cl2 (100 mL) and cooled to 0 °C (ice-salt bath). Bis(2-
chloroethyl)amine hydrochloride (3.57 g, 20.0 mmol) was added
to the solution, followed by a dropwise addition of triethyl-
amine (8.3 mL, 60.0 mmol) in 45 mL of CH2Cl2 over a period
of 45 min. 2-(Methylthio)ethanol (0.83 g, 9.0 mmol) dissolved
in CH2Cl2 (15 mL) was added rapidly from a dropping funnel
and the mixture stirred at 0 °C for 45 min. The reaction
mixture was cooled to -24 °C (dry ice-CCl4 bath), and tert-
butyl hydroperoxide (7.6 mL, 3.0 M in 2,2,4-trimethylpentane,
22.0 mmol) was added. The mixture was then allowed to warm
to room temperature. Subsequently, EtOAc (150 mL) was
added to the pale yellow reaction mixture, and the resulting
white solids were filtered. The filtrate was evaporated in vacuo
to give a yellow oily residue. The residue was dissolved in 50
mL of CH2Cl2 and washed with cold 5% HCl (3 × 50 mL) and
cold 5% NaHCO3 (3 × 50 mL). The organic layer was dried
over anhydrous sodium sulfate and evaporated in vacuo to
yield the crude product as a yellow oil. The crude material
was purified on a flash column with eluent EtOAc/hexanes
2-(Meth ylth io)eth yl N,N-bis(2-ch lor oeth yl)p h osp h or o-
d ia m id a te (4) was synthesized by a modification of the
published procedure.38 A solution of 2-(methylthio)ethanol
(1.69 g, 1.7 mL, 18 mmol) in anhydrous THF (20 mL) was

Aldophosphamide Analogues as Anticancer Prodrugs
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3849
(1:1) to afford 1.0 g (23% crude yield) of slightly impure 7 as
a colorless, viscous oil: TLC Rf 0.43 in EtOAc/hexane (1:1);
1H NMR (CDCl3) δ 2.13 (s, 3H), 2.73 (t, J ) 6.8 Hz, 2H), 3.10-
3.90 (m, 16H), 4.12 (q, J ) 8.6 Hz, 2H).
ethanol (4.70 g, 30 mmol) and Et3N (3.10 g, 30 mmol) in
benzene (30 mL) was added dropwise to a stirred solution of
phosphorus oxychloride (4.60 g, 30 mmol) in benzene (50 mL)
at ice-bath temperature over 30 min. The resulting mixture
was stirred for 40 min at room temperature and then cooled
in ice-bath. 2-Chloroethylamine hydrochloride (6.96 g, 60
mmol) was added to the solution, followed by a dropwise
addition of Et3N (12.1 g, 120 mmol) in 30 mL of CH2Cl2 with
stirring over 30 min. The mixture was stirred overnight at
room temperature, and the resulting white solids were filtered
off. The filtrate was washed with 5% HCl (50 mL), 5% NaHCO3
(50 mL), and saturated brine (2 × 100 mL), dried over
anhydrous Na2SO4, and filtered. The filtrate was evaporated
in vacuo to yield crude product. The crude material was
purified on a flash column with eluent EtOAc/hexane (3:1) to
afford 2.0 g (19% yield) of 15 as a colorless oil: TLC Rf 0.45 in
2-(Met h ylsu lfon yl)et h yl N,N,N′,N′-t et r a k is(2-ch lor o-
eth yl)p h osp h or od ia m id a te (8) was prepared from 7 on a
2.3 mmol (0.98 g) scale, as described in 2 with the same
quantities of aqueous H2O2 (30 wt %, 21 mL, 200 mmol) and
0.3 M ammonium molybdate (3 mL) used in the procedure.
Crude product was obtained as a colorless, oily residue. This
was purified on a flash column with eluent EtOAc to remove
impurities and then EtOH to obtain 0.6 g (overall 12% yield)
of 8 as white solids: mp 98-101 °C; TLC Rf 0.41 in EtOAc;
1H NMR (CDCl3) δ 2.95 (s, 3H), 3.03-3.83, (m, 18H), 4.45 (q,
J ) 8.6 Hz, 2H). Anal. (C11H23Cl4N2O4SP) C, H, Cl, N, S.
2-Hyd r oxyp r op yl Met h yl Su lfid e (9). A mixture of
1-mercapto-2-propanol (2.00 g, 22 mmol) and 10 M NaOH (2.2
mL, 22 mmol) were added dropwise to a solution of methyl
iodide (5.00 g, 35 mmol) in 100 mL of MeOH with stirring over
a 25 min period. The reaction mixture was stirred at room
temperature for 2 h and then MeOH evaporated at atmo-
spheric pressure. The residue was poured in water (100 mL)
and extracted with diethyl ether (3 × 100 mL). The combined
ether extract was washed with saturated brine (2 × 50 mL),
dried over anhydrous MgSO4, and filtered. The filtrate was
evaporated in vacuo to give 1.0 g (41% yield) of 9: TLC Rf 0.42
in EtOAc/hexane (1:4), I2 as indicator; 1H NMR (CDCl3) δ 1.28
(q, J ) 5.4 Hz, 3H), 2.13 (s, 3H), 2.59 (d, J ) 13.2 Hz, 2H),
3.35-4.20 (m, 1H).
1
EtOAc/hexane (4:1); H NMR (CDCl3) δ 2.76-3.83 (m, 10H),
4.14 (q, J ) 9.6 Hz, 2H), 7.13-7.53 (s, 5H).
2-(P h en ylsu lfon yl)eth yl N,N′-bis(2-ch lor oeth yl)p h os-
p h or od ia m id a te (16) was prepared from 15 on a 2.6 mmol
(1.00 g) scale as described in the synthesis of 2 with the same
quantities of aqueous H2O2 (30 wt %, 21 mL, 200 mmol) and
0.3 M ammonium molybdate (3 mL) used in the procedure.
Product 16 was obtained as 1.0 g (94% yield) of white solids:
mp 64-65 °C; TLC Rf 0.52 in EtOAc; 1H NMR (CDCl3) δ 2.85-
3.78 (m, 10H), 4.15-4.58 (m, 2H), 6.88-8.15 (m, 5H). Anal.
(C12H19Cl2N2O4SP) C, H, Cl, N, S.
2-(p-Tolylth io)eth a n ol (17) was synthesized by the pro-
cedure of Amaral30 with some modifications. Aqueous 10 M
NaOH (5.6 mL, 56 mmol) was added dropwise over a 10 min
period to a stirring mixture of 2-chloroethanol (5.40 g, 67
mmol) and thiocresol (6.95 g, 56 mmol) in ethanol (100 mL)
kept at 40 °C. The reaction mixture was refluxed for 1 h and
then added to 2-chloroethanol (3.00 g, 37 mmol). Reflux was
continued for additional 1 h and then cooled and evaporated
in vacuo. The residue was partitioned between ethyl acetate
(100 mL) and water (100 mL). The aqueous layer was extracted
with ethyl acetate (3 × 100 mL). The combined EtOAc extract
was dried over anhydrous Na2SO4 and filtered and the filtrate
evaporated in vacuo to give 9.1 g (97% yield) of 17 as a colorless
oil: TLC Rf 0.35 in EtOAc/hexane (1:5); 1H NMR (CDCl3) δ
2.17 (s, 3H), 2.97 (t, J ) 12 Hz, 2H), 3.63 (t, J ) 9.9 Hz, 2H),
6.90-7.43 (m, 4H).
2-(p -Tolylth io)eth yl N,N-bis(2-ch lor oeth yl)p h osp h or o-
d ia m id a te (18) was prepared with the use of 2-(p-tolylthio)-
ethanol (17) on a 30 mmol (5.0 g) scale as in the procedure for
4 to give 10.1 g of a light yellow oil.. The crude oil was purified
on a flash column with eluent EtOAc/hexane (4:1) to give 6.4
g (58% yield) of 18 as a light yellow oil, which solidified upon
overnight storage in the freezer to a white, waxlike solid: TLC
Rf 0.18 in EtOAc/hexane (4:1); 1H NMR (CDCl3) δ 2.27 (s, 3H),
2.72 (t, J ) 11.1 Hz, 2H), 2.93-3.66 (m, 8H), 4.08 (q, J ) 8.8
Hz, 2H), 7.06-7.40 (m, 4H).
1-Meth yl-2-(m eth ylth io)eth yl N,N-bis(2-ch lor oeth yl)-
p h osp h or od ia m id a te (10) was prepared from 9 on a 30
mmol (3.18 g) scale as in the synthesis of 4 to give a yellow
liquid crude product. This was purified on a flash column with
eluent CH2Cl2/EtOH (20:1) to give 2.2 g (24% yield) of pure
10 as a yellow oil. The product was obtained as a mixture of
two diastereoisomers with very close Rf values: TLC Rf1 0.43
1
and Rf2 0.38 in CH2Cl2/EtOH (20:1); H NMR (CDCl3) δ 1.42
(d, J ) 4.8 Hz, 3H), 2.15 (s, 3H), 2.66 (d, J ) 16.2 Hz, 2H),
3.40-3.70 (m, 8H), 4.58-4.67 (m, 1H). Anal. (C8H19Cl2N2O2-
SP) C, H, Cl, N, S.
1-Meth yl-2-(m eth ylsu lfon yl)eth yl N,N-bis(2-ch lor oeth -
yl)p h osp h or od ia m id a te (11) was prepared from 10 on a 1
mmol (0.31 g) scale as described in the synthesis of 2 with the
same quantities of aqueous H2O2 (30 wt %, 21 mL, 200 mmol)
and 0.3 M ammonium molybdate (3 mL) used in the procedure.
Pure 11 was obtained as 0.3 g (yield 88%) of a light yellow oil
as a mixture of diastereoisomers: TLC Rf1 0.54 and Rf2 0.50
in CH2Cl2/EtOH (10:1); 1H NMR (CDCl3) δ 1.55 (d, J ) 8.4
Hz, 3H), 3.12 (s, 3H), 3.20-3.85 (m, 10H), 4.94-5.16 (m, 1H).
Anal. (C8H19Cl2N2O4SP), C, H, Cl, N, S.
2-Meth ylth io-1-p h en yleth a n ol (12) was synthesized ac-
cording to published procedure41 in 43% yield.
1-P h en yl-2-(m et h ylt h io)et h yl N,N,N′,N′,-t et r a k is(2-
ch lor oeth yl)p h osp h or od ia m id a te (13) was prepared with
the use of 12 on a 9 mmol (1.5 g) scale as described previously
in the synthesis of 7. The crude product was purified on a flash
column with EtOAc/hexane (3:7) eluent to give 2.5 g (65%
yield) of 13 as a yellow oil: TLC Rf 0.66 in EtOAc/hexane
(1:1); 1H NMR (CDCl3) δ 1.99 (s, 3H), 2.65-3.80 (m, 18H),
5.13-5.60 (m, 1H), 7.31 (s, 5H).
1-P h en yl-2-(m eth ylsu lfon yl)eth yl N,N,N′,N′,-tetr a k is-
(2-ch lor oeth yl)p h osp h or od ia m id a te (14) was prepared
from 13 on a 1.6 mmol (0.83 g) scale as described in 2 with
the same quantities of aqueous H2O2 (30 wt %, 21 mL, 200
mmol) and 0.3 M ammonium molybdate (3 mL) used in the
procedure. Crude product was obtained as a yellow oil. This
was purified on a flash column with EtOAc/hexane (1:1) as
eluent to obtain 0.6 g (63% yield) of 14 as a yellow oil: TLC Rf
0.31 in EtOAc/hexane (1:1); 1H NMR (CDCl3, 360 MHz) δ 2.80
(s, 3H), 2.90-3.90 (m, 18H), 5.68-6.20 (m, 1H), 7.45 (m, 5H).
Anal.(C17H27Cl4N2O4SP) C, H, Cl, N, S.
2-(p-Tolylsu lfon yl)eth yl N,N-bis(2-ch lor oeth yl)p h os-
p h or od ia m id a te (19) was synthesized from 18 on a 4.0 mmol
(1.5 g) scale as in 2 with the same quantities of aqueous H2O2
(30 wt %, 21 mL, 200 mmol) and 0.3 M ammonium molybdate
(3 mL) used in the procedure. The crude product was obtained
as white, waxlike solids. Upon recrystallization with EtOAc/
hexane, this afforded 1.4 g of pure 19 as white solids (87%
yield): mp 94-96 °C; TLC Rf 0.66 in EtOAc; 1H NMR (CDCl3)
δ 2.45 (s, 3H), 3.03 (t, J ) 8.1 Hz, 2H), 3.26-3.86 (m, 8H),
4.35 (q, J ) 10 Hz, 2H), 7.20-8.00 (m, 4H). Anal. (C13H21
Cl2N2O4SP) C, H, Cl, N, S.
-
2-(p-Tolylth io)eth yl N,N′-bis(2-ch lor oeth yl)p h osp h or o-
d ia m id a te (20) was prepared from 17 on a 20 mmol (3.36 g)
scale as described for the synthesis of 6 to give crude oil. The
crude product was purified on a flash column with eluent
EtOAc/hexane (3:1) to give 2.2 g (26% yield) of 20 as a light
yellow oil: TLC Rf 0.52 in EtOAc/hexane (3:1); 1H NMR
(CDCl3) δ 2.33 (s, 3H), 2.80-3.86 (m, 10H), 4.15 (q, J ) 10
Hz, 2H), 6.93-7.46 (m, 4H).
2-(P h en ylth io)eth yl N,N′-bis(2-ch lor oeth yl)ph osp h or o-
d ia m id a te (15) was synthesized according to the procedure
in 6 with some modifications. A mixture of 2-(phenylthio)-
2-(p-Tolylsu lfon yl)eth yl N,N′-bis(2-ch lor oeth yl)p h os-
p h or od ia m id a te (21) was synthesized from 20 on a 2.5 mmol

3850 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
J ain et al.
(1.00 g) scale as described for 2 with the same quantities of
aqueous H2O2 (30 wt %, 21 mL, 200 mmol) and 0.3 M
ammonium molybdate (3 mL) used in the procedure to give
1.0 g (99% yield) of 21 as white solids: mp 75.5-78.5 °C; TLC
solution was stirred at room temperature for 20 min and
evaporated in vacuo to yield 1.8 g of grayish semisolid. A
portion of crude 30 (0.15 g) was subjected to preparative TLC
and developed with EtOAc. The TLC region containing the
product was visualized by NBP-spray reagent, removed, and
extracted with EtOAc/EtOH (1:1) to afford 0.1 g (63% yield)
of 30 as a brown solid: mp 79-82 °C; 1H NMR (CDCl3/CD3OD,
1:1) δ 2.66-3.67 (m, 16H), 1.01-2.15 (m, 11H) (1H NMR of
crude 30).
1
Rf 0.27 in EtOAc; H NMR (CDCl3) δ 2.35 (s, 3H), 2.97-3.73
(m, 10H), 4.17-4.66 (m, 2H), 7.17-7.89 (m, 4H). Anal. (C13H21
Cl2N2O4SP) C, H, Cl, N, S.
-
2-(p-Nitr op h en ylth io)eth a n ol (22) was prepared accord-
ing to the published procedure30 in 73% yield.
4-Hyd r op er oxycyclop h osp h a m id e (31) was prepared
2-(p -Nit r op h en ylt h io)et h yl N,N′-b is(2-ch lor oet h yl)-
p h osp h or od ia m id a te (23) was prepared with the use of 22
on a 10 mmol (1.99 g) scale according to the method utilized
for synthesis of 6, except that benzene was used as solvent
instead of CH2Cl2. Crude product obtained as yellow oil was
purified on a flash column with eluent EtOAc/hexane (1:4) to
remove impurities and then EtOAc to give 1.2 g (30% yield) of
23 as yellow oil: TLC Rf 0.70 in EtOAc; H NMR (CDCl3) δ
2.93-3.76 (m, 10H), 4.16 (t, J ) 9.9 Hz, 2H), 7.30-8.33 (m,
4H).
according to published procedures43 in overall 14% yield.
Sta bility Stu d ies. Sta bility Test of 2-(Meth ylsu lfon yl)-
eth yl 2,4-d in itr op h en yl Eth er (2) in 0.08 M P h osp h a te
Bu ffer a t 37 °C a t p H Va lu es of 6.4, 7.4, a n d 8.4. Each
incubation mixture consisted of 5 mL of 0.08 M phosphate
buffer containing 0.2 mM of 2. Incubations were initiated by
the addition of 1 mL solution (1 mM) of 2 in THF to 4 mL of
preincubated 0.1 M phosphate buffer (pH 6.4, 7.4, or 8.4) and
terminated by immediately cooling the test tubes in crushed
ice. Incubations were carried out for varying time intervals in
a shaking water bath at 37 °C and absorbance values mea-
sured at 359 nm. A standard curve of 2,4-dinitrophenol at 359
nm was used to calculate the amount of 2,4-dinitrophenol
eliminated from 2 upon incubation. Results were expressed
as percent of total 2,4-dinitrophenol, where the total is
represented by 0.2 mM of 2,4-dinitrophenol. Subsequently, the
percent unchanged 2 was calculated against time of incuba-
tion. The slope (-kobs) of a linear plot of ln(% of 2 remaining)
versus time of incubation (t) gave the elimination rate constant
1
2-(p-Nitr op h en ylsu lfon yl)eth yl N,N′-bis(2-ch lor oeth -
yl)p h osp h or od ia m id a te (24) was prepared from 23 on a 1.5
mmol (0.6 g) scale as described for 2 with the same quantities
of aqueous H2O2 (30 wt %, 21 mL, 200 mmol) and 0.3 M
ammonium molybdate (3 mL) and reaction time of 1 h, to give
a yellow oil. Crystallization with EtOAc/hexane gave 0.5 g
(77% yield) of 24 as white solids: mp 86-88 °C; TLC; Rf 0.19
1
in EtOAc; H NMR (CDCl3) δ 2.83-3.77 (m, 10H), 4.43 (q, J
) 9.4 Hz, 2H), 8.03-8.60 (m, 4H). Anal. (C12H18Cl2N3SO6P)
C, H, Cl, N, S.
k
obs. Half-life (t1/2) of â-elimination was calculated by the use
2-(p-Nitr oph en ylth io)eth yl N,N,N′,N′-tetr akis(2-ch lor o-
eth yl)p h osp h or od ia m id a te (25) was prepared with the use
of 22 on a 10 mmol (1.8 g) scale as described in the synthesis
of 7 to give 5.0 g of yellow oil as the crude product. Initial
purification with a flash column with eluents EtOAc/hexanes
(1:2) and then EtOAc gave semipure product. Further purifica-
tion on a flash column with EtOAc/hexane (1:1) as the eluting
solvent gave 25 as a yellow oil (1.5 g, 29% yield): TLC Rf 0.47
in EtOAc/hexane (1:1). 1H NMR (CDCl3) δ 3.06-3.72, (m, 18H),
4.15 (q, J ) 10 Hz, 2H), 7.13-8.20 (m, 4H).
of equation: t1/2 ) 0.693/kobs
.
Sta bility Test of Su lfon yleth yl P h osp h or od ia m id a te
P r od r u gs u n d er Mod el P h ysiologica l Con d ition s. Sta bil-
ity in 0.08 M P h osp h a te Bu ffer p H 7.4 a n d 37 °C. Each
incubation mixture consisted of 0.5 mL of 0.08 M phosphate
buffer pH 7.4 containing a 3-9 mM concentration of all test
compounds, except 0.8 mM for 14 and 0.3 mM for 26.
Incubations were initiated with the addition of 0.1 mL of
solution of test compound to 0.4 mL of preincubated 0.1 M
phosphate buffer (pH 7.4) in 13 × 100 mm culture test tubes.
Drug solutions were prepared in water (5, 6, and 11), EtOH
(14 and 26), EtOH/H2O (2:1) (8 and 19), EtOH/H2O (1:1) (24),
EtOH/H2O (1:2) (21), or EtOH/H2O (1:3) (16). The test tubes
were incubated at 37 °C in a shaking water-bath for varying
time intervals and incubations terminated with the addition
of 0.5 mL of 2.0 M acetate buffer (pH 5.0) and 2.0 mL of EtOH.
An aliquot (16 µL for all compounds except 48 µL for 14 and
26) of the resulting mixture for all samples was then spotted
on a 10 × 20 mm silica gel TLC plate and eluted with
appropriate solvent [EtOAc/EtOH (10:1) for all analogues
except EtOAc for 8 and EtOAc/hexane (3:1) for 14 and 26].
The starting material from all the samples either appeared
as a dark spot under UV light (λ ) 254 nm), or the plates were
subjected to NBP assay treatment to visualize the starting
material (5, 6, 8, 11, and 14) as a purple colored spot with
varying intensity. The TLC profile of zero hour control samples
in all cases did not show appearance of degradation products.
The spots were then quantitatively analyzed using a Uniscan
Video Densitometer. Results were expressed as the percent
starting material remaining after each incubation period in
comparison to the zero time incubation control sample. The
slope (-kobs) of a linear plot of ln(% of test compound remain-
ing) versus time of incubation (t) gave the elimination rate
constant kobs. Half-life (t1/2) of â-elimination was calculated by
2-(p -Nit r op h en ylsu lfon yl)et h yl N,N,N′,N′-t et r a k is(2-
ch lor oeth yl)p h osp h or od ia m id a te (26) was prepared from
25 on a 2.0 mmol (1.0 g) scale in the procedure for 2 to give
crude product as a pale yellow viscous oil. The crude oil was
purified using flash column chromatography with EtOAc/
hexane (2:1) eluent to give 0.6 g of pale yellow oil. Crystal-
lization with EtOAc/hexane gave 0.6 g (49% yield) of 26 as a
white solid: mp 83-85 °C; TLC Rf 0.43 in EtOAc/hexane
1
(2:1); H NMR (CDCl3) δ 3.06-3.83, (m, 18H), 4.40 (q, J ) 8
Hz, 2H), 8.00-8.53 (m, 4H). Anal. (C16H24Cl4N3O6SP) C, H,
Cl, N, S.
N,N-Bis(2-ch lor oeth yl)p h osp h or od ia m id ic a cid , cyclo-
h exyla m m on iu m sa lt (P h osp h or a m id e m u sta r d ) (27) was
prepared as described42 in 79% yield.
N,N′-Bis(2-ch lor oeth yl)ph osph or odiam idic acid, cyclo-
h exyla m m on iu m sa lt (Isop h osp h or a m id e m u sta r d ) (28)
was prepared according to a published procedure22 in 30%
yield.
N ,N ,N ′,N ′-Te r t a k is(2-ch lor oe t h yl)p h osp h or od ia m i-
d a te p h en ylm eth yl ester (29) was prepared from benzyl
alcohol on a scale of 18 mmol (1.95 g) as described for 7. The
crude material was purified on a flash column with CH2Cl2/
acetone (100:3) eluent to give 5.1 g (67% yield) of 29 as a
colorless oil: TLC Rf 0.25 in CH2Cl2/acetone (100:3); 1H NMR
(CDCl3) δ 3.07-3.67 (m, 16H), 5.07 (d, J ) 12 Hz, 2H), 7.61
(s, 5H).
the use of equation: t1/2 ) 0.693/kobs
.
Sta bility Test in th e P r esen ce of Hu m a n P la sm a .
Lyophilized human plasma (Sigma Chemical Co., St. Louis,
MO) was used for the study and was reconstituted immediately
prior to the experiments. The protocol used for the study was
essentially the same as in phosphate buffer except the follow-
ing:
Each incubation mixture contained 0.4 mL of reconstituted
solution of human plasma in 0.1 M phosphate buffer pH 7.4,
instead of the buffer alone. After the termination of the
N,N,N′,N′-Tet r a k is(2-ch lor oet h yl)p h osp h or od ia m id -
ic a cid , cycloh exyla m m on iu m sa lt (Tetr a k is p h osp h or -
a m id e m u sta r d ) (30) was prepared by modification of a
previously reported procedure.42 A solution of 29 (2.0 g, 4.7
mmol) in absolute EtOH (150 mL) was added to palladium on
activated carbon (0.6 g, carbon content 10%), and the mixture
was hydrogenated at room temperature and 1 atm for 2.5 h.
The reaction mixture was filtered, and then cyclohexylamine
(0.5 mL) was added dropwise to the filtrate. The reaction

Aldophosphamide Analogues as Anticancer Prodrugs
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3851
(2) Ludeman, S. M. The chemistry of the metabolites of cyclophos-
phamide. Curr. Pharm. Des. 1999, 5, 627-643.
incubation, the test tubes were placed in crushed ice and
centrifuged at high speed (4750 rpm, approximately 3000g)
for 20 min to precipitate the proteins. Subsequently, an aliquot
of the resulting supernatant was spotted on 10 × 20 mm silica
gel TLC plates, developed, and analyzed as described in the
procedure for phosphate buffer.
(3) Colvin, O. M. An overview of cyclophosphamide development and
applications. Curr. Pharm. Des. 1999, 5, 555-560.
(4) Low, J . E.; Borch, R. F.; Sladek, N. E. Further studies on the
conversion of 4-hydroxyoxazaphosphorines to reactive mustards
and acrolein in inorganic buffers. Cancer Res. 1983, 43, 5815-
5820.
In Vitr o Cytotoxicity Eva lu a tion . V-79 Chinese hamster
lung fibroblasts (American Type Culture Collection, Rockville,
MD, ATCC #93) subcloned by this laboratory were used. The
procedures were similar to those reported previously.44 Cell
cultures were maintained at subconfluence in a 95% air, 5%
CO2 humidified atmosphere at 37 °C. The medium used for
routine subcultivation as well as for experimental determina-
tions was Minimum Essential Medium (Gibco, Grand Island,
NY) supplemented with 10% heat-inactivated fetal bovine
serum, Pen G sodium, streptomycin sulfate, amphotericin, and
essential and nonessential amino acids and vitamins.
Log-phase cultures fed 24 h prior to use were disassociated
from their growth substrate by scraping, and cells were
disaggregated by repeated pipet aspirations. Cells were counted
by trypan blue dye exclusion method using a Bright-line
Haemocytometer (Fisher Inc.). Polystyrene tissue culture
Petri-dishes (100 mm, Costar, Cambridge, MA) were seeded
with 500 cells in 10 mL of media and incubated for 12-15 h
in a 95% air, 5% CO2 humidified atmosphere. Following the
initial incubation, the medium was replaced with 9.75 mL of
fresh medium and the cells were treated with 0.25 mL of the
drug solution in appropriate concentration or the solvent as a
control. Cultures were then incubated undisturbed for 3 h, and
the drug exposure was terminated by removing the old
medium and rinsing with Hank’s Balanced Salt Solution
(HBSS, 1 × 10 mL). The Petri-dishes were then added with
fresh medium (10 mL) and then incubated under standard
conditions, undisturbed for 5-6 days to allow colony formation.
Colonies were then rinsed with HBSS (2 × 10 mL), fixed with
95% EtOH, and stained with crystal violet and counted.
Results were reported as the number of colonies surviving drug
treatment per number of colonies in the solvent-treated con-
trol. The IC50 values were determined by semilogarithmically
plotting the drug concentration versus cell viability as deter-
mined by the number of colonies surviving the treatments.
In Vivo An titu m or Activity. Preliminary antitumor activ-
ity of the sulfonyl analogues was evaluated against P388/0
(Wild) or P388/CPA (CP-resistant) leukemia cells in CD2F1
mice (24-26 g). To determine the ability of the test compounds
to increase the life span of mice, animals were implanted with
106 cells, ip, on day 0. The drugs were either dissolved in saline
or suspended in saline-Tween 80. The mice were injected ip
with drug, saline, or saline-Tween 80 on day 1. The survival
rate was compared with that of untreated controls over a 33-
day period. Isophosphoramide mustard (IPM) (100 mg/kg) or
CP (200 mg/kg) served as a positive control. Three doses (100,
200, and 400 mg/kg) were used for the sulfonyl analogues. Five
mice were used per dose per drug. The results were reported
as % ILS (percent increase in life span) as calculated by the
following equation:
(5) Bielicki, L.; Voelcker, G.; Hohorst, H. J . Activated cyclophos-
phamide: An enzyme-mechanism-based suicide inactivator of
DNA polymerase/3′-5′ exonuclease. J . Cancer Res. Clin. Oncol.
1984, 107, 195-198.
(6) Hohorst, H. J .; Bielicki, L.; Voelcker, G. The enzymatic basis of
cyclophosphamide specificity. Adv. Enzyme Regul. 1986, 25, 99-
122.
(7) Kwon, C. H.; Maddison, K.; LoCastro, L.; Borch, R. F. Accelerated
decomposition of 4-hydroxycyclophosphamide by human serum
albumin. Cancer Res. 1987, 47, 1505-1508.
(8) Sladek, N. E. Aldehyde-dehydrogenase mediated cellular relative
insensitivity to the oxazaphosphorines. Curr. Pharm. Des. 1999,
5, 607-625.
(9) Sladek, N. E.; Kollander, R.; Sreerama, L.; Kiang, D. T. Cellular
levels of aldehyde dehydrogenase (ALDH1A1 and ALDH3A1) as
predictors of therapeutic responses to cyclophosphamide-based
chemotherapy of breast cancer: a retrospective study Rational
individualization of oxazaphosphorine-based cancer chemothera-
peutic regimens. Cancer Chemother. Pharmacol. 2002, 49(4),
309-321.
(10) Ludeman, S. M.; Gamcsik, M. P. Mechanisms of resistance
against cyclophosphamide and ifosfamide: can they be overcom-
ing without sacrificing selectivity. Cancer Treat. Res. 2002, 112,
177-197.
(11) Bunting, K. D.; Townsend, A. J . De novo expression of trans-
fected human class 1 aldehyde dehydrogenase (ALDH) causes
resistance to oxazaphosphorine anti-cancer alkylating agents in
hamster V-79 cell lines. Elevated class 1 ALDH activity is closely
correlated with reduction in DNA interstarnd cross-linking and
lethality. J . Biol. Chem. 1996, 271, 11884-11890.
(12) Dittrich, C. Ifosfamide: actual data, new insights, and persisting
questions. Semin. Oncol. 2000, 27 (1 Suppl. 1), 1-2.
(13) Fleming, R. A. An overview of cyclophosphamide and ifosfamide
pharmacology. Pharmacotherapy 1997, 17 (5 Pt 2), 146S-154S.
(14) Kerbusch, T.; de Kraker, J .; Keizer, H. J .; van Putten, J . W.;
Groen, H. J .; J ansen, R. L.; Schellens, J . H.; Beijnen, J . H.
Clinical pharmacokinetics and pharmacodynamics of ifosfamide
and its metabolites. Clin. Pharmacokinet. 2001, 40 (1), 41-62.
(15) Boddy, A. V.; Yule, S. M. Metabolism and pharmacokinetics of
oxazaphosphorines. Clin. Pharmacokinet. 2000, 38 (4), 291-304.
(16) Woodland C.; Ito, S.; Granvil, C. P.; Wainer, I. W.; Klein, J .;
Koren, G. Evidence of renal metabolism of ifosfamide to nephro-
toxic metabolites. Life Sci. 2000, 68 (1), 109-117.
(17) Dubourg, L.; Michoudet, C.; Cochat, P.; Baverel, G. Human
kidney tubules detoxify chloroacetaldehyde, a presumed nephro-
toxic metabolite of ifosfamide. J . Am. Soc. Nephrol. 2001, 12 (8),
1615-1623.
(18) Springate, J . E. Ifosfamide metabolite chloroacetaldehyde causes
renal dysfunction in vivo. J . Appl. Toxicol. 1997, 17 (1), 75-79.
(19) Sood, C.; O’Brien, P. J . 2-Chloroacetaldehyde-induced cereberal
glutathione depletion and neuroxicity. Br. J . Cancer 1996, 27,
S287-S293.
(20) McVay, J . I.; Wood, A. M. Suspected ifosfamide-induced neuro-
toxicity. Pharmacotherapy 1999, 19(12), 1450-1455.
(21) Friedman, O. M.; Wodinsky, I.; Myles, A. Cyclophosphamide
(NSC-26271)-related phosphoramide mustards-Recent advances
and historical perspective. Cancer Treat. Rep. 1976, 60, 337-
346.
% ILS ) (treated median day of death/
control median day of death) × 100 - 100
Animals living beyond 33 days were reported as “long term
survivors” and were not used in calculations.
(22) Struck, R. F.; Dykes, D. J .; Corbett, T. H.; Suling, W. J .; Trader,
M. W. Isophosphoramide mustard, a metabolite of ifosfamide
with activity against murine tumors comparable to cyclophos-
phamide. Br. J . Cancer 1983, 47, 15-26.
(23) Kwon, C.-H.; Borch, R. F. Effects of N-substitution on the
activation mechanisms of 4-hydroxycyclophosphamide ana-
logues. J . Med. Chem. 1989, 32, 1491-1496.
(24) Brock, N. Comparative pharmacological study in vitro and in
vivo with cyclophosphamide (NSC-26271), cyclophosphamide
metabolites, and plain nitrogen mustard compounds. Cancer
Treat. Rep. 1976, 60, 301-307.
(25) Boal, J . H.; Ludeman, S. M.; Ho, C. K.; Engel, J .; Niemeyer, U.
Direct detection of the intracellular formation of carboxy-
phosphamide using nuclear magnetic resonance spectroscopy.
Arzneim-Forsch. 1994, 44 (1), 84-93.
(26) Struck, R. F. Aldophosphamide: Synthesis, characterization and
comparison with “Hohorst’s aldophosphamide”. Cancer Treat.
Rep. 1976, 60, 317-319.
(27) Kader, A. T.; Stiriling, C. J . M. Elimination-addition. Part III.
New procedures for the protection of amino groups. J . Chem.
Soc. 1964, 258-268.
Ack n ow led gm en t. The authors wish to express
sincere gratitude to Dr. R. F. Struck of Southern
Research Institute, Birmingham, AL, for kindly agree-
ing to conduct in vivo antitumor evaluations.
Su p p or tin g In for m a tion Ava ila ble: Elemental analyses
data. This material is available free of charge via the Internet
at http://pubs.acs.org.
Refer en ces
(1) Sladek, N. E. Metabolism of oxazaphosphorines. Pharm. Ther.
1988, 37, 301-355.

3852 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
J ain et al.
(28) Hardy, P. M.; Rydon, H. N.; Thompson, R. C. The use of
â-methylthioethyl esters for the protection of carboxyl groups
in peptide synthesis: removal through the â-methylsulphonyl-
ethyl ester. Tetrahedron Lett. 1968, 21, 2525-2526.
(29) Tesser, G. I.; Balvert-Geers, I. C. The methylsulfonylethyloxy-
carbonyl group, new and versatile amino protective function. J .
Pept. Protein Res. 1975, 7, 295-305.
(36) Barlow, K. N.; Marshall, D. R.; Stirling, C. J . M. Elimination
and Addition Reactions. Part 33. Formation of carbanions
derived from sulphones and nitriles bearing â-onium substitu-
ents. J . Chem. Soc., Perkin Trans. 2 1977, 1920-1928.
(37) Whalley, W. B. Organic Fluoro-compounds. Part II. The prepara-
tion of alkyl 2:4-dinitrophenyl ethers from 1-fluoro-2:4-dinitro-
benzene. J . Chem. Soc. 1950, 2241-2243.
(30) Amaral, M. J . S. A. The 2-(p-nitrophenylthio)ethyl group for
carboxy-group protection in peptide synthesis. J . Chem. Soc. (C)
1969, 2495-2497.
(38) Friedman, O. M.; Seligman, A. M. Preparation of N-Phospho-
rylated derivatives of bis-â-chloroethylamine. J . Am. Chem. Soc.
1954, 76, 655-658.
(31) Crosby, J .; Stirling, C. J . M. Elimination and addition reactions.
Part XIX. Elimination of phenoxide from â-substituted ethyl
phenyl ethers: the nature of activation in 1,2-elimination. J .
Chem. Soc. B 1970, 671-679.
(32) Crosby, J .; Stirling, C. J . M. Elimination and addition reactions.
Part XIX. Elimination of phenoxide from â-substituted ethyl
phenyl ethers: Evidence for E1cB mechanism. J . Chem. Soc. B
1970, 679-686.
(33) Marshall, D. R.; Thomas, P. J .; Stirling, C. J . M. Elimination
and addition reactions. Part 32. Discrimination between con-
certed and stepwise processes in activated elimination reactions.
J . Chem. Soc., Perkin Trans. 2 1977, 1913-1919.
(34) Ludeman, S. M.; Boyd, V. L.; Regan, J . B.; Gallo, K. A.; Zon, G.;
Ishii, K. Synthesis and antitumor activity of cyclophosphamide
analogues. 4. Preparation, kinetic studies, and anticancer
screening of “phenylketophosphamide” and similar compounds
related to the cyclophosphamide metabolite aldophosphamide.
J . Med. Chem. 1986, 29, 716-727.
(35) Borch, R. F.; Valente, R. R. Synthesis, activation and cytotoxicity
of aldophosphamide analogues. J . Med. Chem. 1991, 34, 3052-
3058.
(39) Misiura, K.; Okruszek, A.; Pankiewicz, K, Stec, W. J .; Czownicki,
Z.; Utracka, B. Stereospecific synthesis of chiral metabolites of
ifosfamide and their determination in the urine. J . Med. Chem.
1983, 26, 674-679.
(40) Mulcahy, R. T.; Gipp, J . J .; Schmidt, J . P.; J oswig, C.; Borch, R.
F. Nitrobenzyl phosphorodiamidates as potential hypoxia-selec-
tive alkylating agents. J . Med. Chem. 1994, 37, 1610-1615.
(41) Conte, V.; Di Furia, F.; Licini, G.; Modena, G.; Sbampato, G.;
Vallr, G. Enantioselective oxidation of â-hydroxythioethers.
Synthesis of optically active alcohols and epoxides. Tetra-
hedron: Asymmetry 1991, 2 (4), 257-276.
(42) Friedman, O. M.; Boger, E.; Grubliauskas, V.; Sommer, H.
Synthesis of N-phosphorylated derivatives of nitrogen mustards
with latent cytotoxicity. J . Med. Chem. 1963, 6, 50-58.
(43) Borch, R. F.; Millard, J . A. The mechanism of activation of
4-hydroxycyclophosphamide. J . Med. Chem. 1987, 30, 427-431.
(44) Kwon, C.-H.; Moon, K.-Y.; Baturay, N.; Shirota, F. Chemically
stable, lipophilic prodrugs of phosphoramide mustard as poten-
tial anticancer agents. J . Med. Chem. 1991, 34 (2), 588-592.
J M0304764