Preactivated oxazaphosphorines designed for isophosphoramide mustard delivery as bulk form or nanoassemblies: Synthesis and proof of concept

Article abstract of DOI:10.1021/jm501224x

Oxazaphosphorines are alkylating agents used in routine clinical practices for treatment of cancer for many years. They are antitumor prodrugs that require cytochrome P450 bioactivation leading to 4-hydroxy derivatives. In the case of ifosfamide (IFO), the bioactivation produces two toxic metabolites: acrolein, a urotoxic compound, concomitantly generated with the isophosphoramide mustard; and chloroacetaldehyde, a neurotoxic and nephrotoxic compound, arising from the oxidation of the side chains. To improve the therapeutic index of IFO, we have designed preactivated IFO derivatives with the covalent binding of several O- and S-alkyl moieties including polyisoprenoid groups at the C-4 position of the oxazaphosphorine ring to avoid cytochrome bioactivation favoring the release of the active entity and limiting the chloroacetaldehyde release. Thanks to the grafted terpene moieties, some of these new conjugates demonstrated spontaneous self-assembling properties into nanoassemblies when dispersed in water. The cytotoxic activities on a panel of human tumor cell lines of these novel oxazaphosphorines, in bulk form or as nanoassemblies, and the release of 4-hydroxy-IFO from these preactivated IFO analogues in plasma are reported.

Full text of DOI:10.1021/jm501224x

Subscriber access provided by NEW MEXICO STATE UNIV
Article
Pre-activated oxazaphosphorines designed for isophosphoramide mustard
delivery as bulk form or nano-assemblies: synthesis and proof of concept.
Charles Skarbek, Lea L. Lesueur, hubert Chapuis, Alain Deroussent, Catherine Pioche-
Durieu, Aurore Daville, Joachim Caron, Michael Rivard, Thierry Martens, Jean-Remi
Bertrand, Eric Le Cam, Gilles Vassal, Patrick Couvreur, Didier Desmaêle, and Angelo Paci
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm501224x • Publication Date (Web): 11 Dec 2014
Downloaded from http://pubs.acs.org on December 21, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Recherche Scientifique (CNRS), Laboratoire de Vectorologie et
Thérapeutiques Anticancéreuses UMR 8203; Gustave Roussy Cancer
Campus Grand Paris,
Couvreur, Patrick; Université Paris-Sud, Institut Galien UMR 8612
Desmaêle, Didier; Université Paris-Sud, Institut Galien UMR 8612
Paci, Angelo; Université Paris-Sud, Laboratoire de Vectorologie et
Thérapeutiques Anticancéreuses UMR 8203; Centre National de la
Recherche Scientifique (CNRS), Laboratoire de Vectorologie et
Thérapeutiques Anticancéreuses UMR 8203; Gustave Roussy Cancer
Campus Grand Paris, ; Gustave Roussy Cancer Campus Grand Paris,
Service Interdépartemental de Pharmacologie et d'Analyse du Médicament
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Preꢀactivated oxazaphosphorines designed for
isophosphoramide mustard delivery as bulk form
or nanoꢀassemblies: synthesis and proof of
concept.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
,2,3
1,2,3
4
1,2,3
Charles Skarbek , Lea L. Lesueur , Hubert Chapuis , Alain Deroussent ,Catherine
5
1,2,3
4
6
6
Pioche–Durieu , Aurore Daville , Joachim Caron , Michael Rivard , Thierry Martens , Jean-
1,2,3
5
1,2,3
4
4
Rémi Bertrand , Eric Le Cam , Gilles Vassal , Patrick Couvreur , Didier Desmaele and
1
,2,3,7*
Angelo Paci
.
*
Corresponding Author.
(1) Université ParisꢀSud, Laboratoire de Vectorologie et Thérapeutiques Anticancéreuses,
UMR 8203, Villejuif, Franceꢀ94805;
2) Centre National de la Recherche Scientifique (CNRS), Laboratoire de Vectorologie et
(
Thérapeutiques Anticancéreuses, UMR 8203, Villejuif, Franceꢀ94805;
(3) Gustave Roussy Cancer Campus Grand Paris, Laboratoire de Vectorologie et
Thérapeutiques Anticancéreuses, UMR 8203, Villejuif, Franceꢀ94805;
(
4) Université ParisꢀSud, Institut Galien, UMR 8612, ChâtenayꢀMalabry, Franceꢀ92296;
(5) CNRS UMR8126, Université Paris Sud 11, Institut Gustave Roussy, Villejuif, Franceꢀ
4805.
6) Université Paris Est Créteil, Institut de Chimie et des Matériaux ParisꢀEst (ICMPE),
UMR 7182, Thiais, Franceꢀ94320;
9
(
1
ACS Paragon Plus Environment

Page 3 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
7) Gustave Roussy Cancer Campus Grand Paris, Service Interdépartemental de
Pharmacologie et d'Analyse du Médicament (SIPAM), Villejuif, Franceꢀ94805.
KEYWORDS: Cancer, Cytotoxicity, Drug delivery; Ifosfamide, Nanomedicine, Nanoꢀ
assemblies, Oxazaphosphorines;.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT: Oxazaphosphorines are alkylating agents used in routine clinical practices for
treatment of cancer for many years. They are antitumor prodrugs that require cytochrome P450
bioꢀactivation leading to 4ꢀhydroxy derivatives. In the case of ifosfamide (IFO), the
bioꢀactivation produces two toxic metabolites; acrolein, an urotoxic compound, concomitantly
generated with the isophosphoramide mustard and, chloroacetaldehyde, a neurotoxic and
nephrotoxic compound, arising from the oxidation of the side chains. To improve the therapeutic
index of IFO, we have designed preꢀactivated IFO derivatives with the covalent binding of
several O- and S-alkyl moieties including polyꢀisoprenoid groups at the Cꢀ4 position of the
oxazaphosphorine ring to avoid cytochrome bioꢀactivation favoring the release of the active
entity and limiting the chloroacetaldehyde release. Thanks to the grafted terpene moieties, some
of these new conjugates demonstrated spontaneous selfꢀassembling properties into nanoꢀ
assemblies when dispersed in water. The cytotoxic activities on a panel of human tumor cell
lines of these novel oxazaphosphorines, as bulk form or nanoꢀassemblies, and the release of 4ꢀ
hydroxyꢀIFO from these preꢀactivated IFO analogues in plasma are reported.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table of contents graphic
3
ACS Paragon Plus Environment

Page 5 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
. Introduction.
Oxazaphosphorines belong to alkylating agents that have been widely used in routine clinical
practices to treat several types of cancer from soft tissue tumor to lymphoma. Ifosfamide (IFO,
1), cyclophosphamide and trofosfamide are oxazaphosphorines which contain two or three
chloroethyl groups bounded to the nitrogen atoms (Scheme 1). As prodrugs, these compounds
require a metabolic activation in position Cꢀ4 of the ring. This is fulfilled by specific liver
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
cytochrome P450 (CYP) . In the case of IFO, this leads to a hydroxylated intermediate, the 4ꢀ
hydroxyꢀifosfamide (4ꢀHOꢀIFO, 2), which after a ring opening mechanism can release the active
isophosphoramide mustard (3) displaying cytotoxicity by DNA crossꢀlinks. This main activation
pathway carried out by CYP3A4 leads concomitantly to the formation of acrolein (4) through
tautomeric equilibrium and retroꢀMichael process. It is noteworthy that urotoxicity due to 4 and
characterized by hemorrhagic cystitis, could be neutralized by coꢀadministration of sodium
2
1
mercaptoethanesulfonate (mesna). A competitive CYP bioꢀactivation of IFO by CYP2B6
toxigenic pathway) occurring on the lateral chains (in positions Cꢀ7 and Cꢀ9) leads to the
(
production of metabolites responsible for limiting sideꢀeffects. This secondary metabolism
pathway, revealed during high dose protocols, oxidizes the chloroethyl side chains and leads to
2
3
2
3
the release of Nꢀdeschlorethyl metabolites (N ꢀDCEꢀIFO, N ꢀDCEꢀIFO, N , N ꢀDCEꢀIFO) along
with chloroacetaldehyde (5) responsible for neuroꢀ and nephrotoxicity which remain not
countered until today.
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 1: Oxazaphosphorines and metabolism of Ifosfamide.
Cl
Cl
Cl
Cl
Cl
7
9
H
O
P
O
P
O
P
N
N
N
3
NH
N
N
Cl
Cl
4
5
2
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
6
O
O
Cyclophosphamide
Ifosfamide (IFO, 1)
Trofosfamide
Activating
pathway
CYP3A4
Toxigenic
pathway
CYP2B6
Cl
Cl
Cl
Cl
O
P
O
H
O
P
H
O
P
HO
N
NH
N
NH2
N
NH
N
NH2
P
O
O
O
O
2
3
2
3
Hydroxyꢀifosfamide (HOꢀIFO, 2)
N ꢀDCEꢀIFO
N ꢀDCEꢀIFO
N ,N ꢀDCEꢀIFO
ring opening
+
Cl
O
Chloroacetaldehyde (5)
neuroꢀ and nephrotoxic
Cl
Cl
O
+
CHO
HN
NH
P
Acrolein (4)
urotoxic
OH
Isophosphoramide mustard (3)
To circumvent the toxicity related to the release of chloroacetaldehyde, pharmacomodulation of
IFO have been investigated. Synthesis of analogues by engraftment of methyl groups on the side
3
4
chains (to avoid chloroacetaldehyde release) or replacement of the chlorine atoms by bromine
5
was described. Agents that do not require in vivo biotransformation for activation including
6
,7
8,9
isophosphoramide mustard and glufosfamide have also been proposed, some of them
10
reaching phase I clinical trial such as mafosfamide or phase III for palifosfamide. Another
strategy consists in the preꢀactivation of the oxazaphosphorine ring by chemical oxidation
regioselectively centered on the Cꢀ4 position in order to bypass the cytochrome bioꢀactivation.
1
1
12,13
Many derivatives have already been prepared (4ꢀOH, 4ꢀOOH,
4ꢀOR, 4ꢀSPh, 4ꢀSCOR, 4ꢀ
1
4
SCSR, 4ꢀONHCONH ). Although some of them showed improved in vitro activity, they were
2
5
ACS Paragon Plus Environment

Page 7 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
found either too unstable for further development or with no advantages over the use of IFO even
so the concept of activation of the position Cꢀ4 was not shown to be a drawback and needs to be
pursued.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In this context, we assume that IFO analogues bearing long chain 4ꢀalkoxy substituent might be
valuable candidates since steric hindrance around the N-3 nitrogen atom should slow down the
hydrolysis process to 4ꢀHOꢀIFO and design more stable derivatives. However, the introduction
of simple alkoxy derivatives would dramatically reduce the water solubility. To address the
problem, we therefore chose to introduce various polyꢀisoprenoid chains at the Cꢀ4 position.
Indeed, we have previously shown that the covalent linkage of squalene, a natural triterpene
constituted of six isoprene units, to a drug belonging to various therapeutic classes, produced a
1
5
bioconjugate that selfꢀassembled into nanoꢀassemblies (NAs) in aqueous media . In most cases,
the resulting NAs displayed an improved pharmaceutical profile compared to the parent
1
6,17,18,19
compound
. The “squalenoylation” method, initially developed with highly hydrophilic
nucleoside analogues has been further extended to more hydrophobic drugs such as penicillin
20
21
G
or paclitaxel. Furthermore, we recently showed that other polyꢀisoprenoid compounds
derived from geraniol (2 isoprene units) and farnesol (3 isoprene units), were able to provide
2
2
selfꢀassembling properties to the conjugates with nucleoside analogues. Based on these
findings, we hypothesize that the covalent binding of a polyꢀisoprenoid chain of a suitable size to
the Cꢀ4 carbon of IFO would deliver conjugates able to selfꢀassemble into supramolecular
objects in aqueous media. In order to modulate the release of 4ꢀHOꢀIFO, we decided to use
simple ether linker but also aminoethylsulfanyl linker. Thus, we report herein the synthesis of a
set of lipophilic derivatives of preꢀactivated IFO bearing, either a geranyl, a farnesyl, or a
squalene derived substituent at Cꢀ4 position (9cꢀf). To investigate the influence of the structural
features within the lipophilic side chain, IFO analogs bearing a simple alkyl group (9a) and
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tetrahydrogeranyl group (9b) were also synthesized. The nanoꢀassembly formulation of the preꢀ
activated IFO derivatives and the in vitro cytotoxic activities on a panel of cancer lines of the
NAs are also reported including in vitro investigations concerning the release of the key
metabolite, i.e. 4ꢀHOꢀIFO in mice plasma.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
. Results
2
.1. Chemistry.
The synthesis of preꢀactivated conjugates 9aꢀf is depicted in Scheme 2. According to our
previous finding, anodic oxidation of IFO in methanol afforded the expected
11
4ꢀmethoxyꢀifosfamide (MeOꢀIFO, 6) as a mixture of isomers with an optimized yield of 96%.
The substitution of the methoxy group by the relevant alkyl chain took advantage of the
formation of the iminium salt 7 upon Lewis acid treatment of 6. A short screening of the Lewis
acids (TMSOTf, Ti(OEt) , TiCl BF .Et O) showed that BF .Et O was the most efficient reagent.
4
4,
3
2
3
2
Thus, treatment of 6 at ꢀ78 °C by one equivalent of BF .Et O followed by addition of a slight
3
2
excess of nucleophiles 8aꢀf gave compounds 9aꢀf in 25ꢀ53% yield. All preꢀactivated IFO
derivatives were stable enough to be purified by chromatography on silica gel. However a
substantial loss of material was observed with the less lipophilic derivatives 9aꢀc that accounted
1
13
31
for their reduced yields. Conjugates 9aꢀf were characterized by H, C, P NMR, MS and
HRMS. All compounds were shown to be obtained as diastereomeric mixtures due to the
presence of the phosphorous stereogenic center together with the Cꢀ4 asymmetric center.
7
ACS Paragon Plus Environment

Page 9 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 2: Synthesis of the preꢀactivated conjugates 9aꢀf.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Alcohols 8aꢀd were commercially available, whereas squalene derivatives 8e,f were obtained by
synthesis from squalene and depicted in Scheme 3. The synthesis of 1,1’,2ꢀtrisnorsqualenol (8e)
was carried out from squalene via 1,1’,2ꢀtrisnorsqualenic aldehyde (10) according to previously
23,24
reported methods.
On the other hand, oxidation of aldehyde 10 gave the known 1,1’,2ꢀ
2
3
trisnorsqualenic acid (11). Acid activation by NHS using DCC as coupling agent, followed by
simple cysteamine treatment afforded 9f in 72% overall yield.
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 3: Synthesis of the squalenyl appendages 8e,f .
a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Ref 22,23
8
e
H
O
1
0
Squalene
b
X
O
11: X = OH
c
1
2: X = NHS
f: X = NHCH CH SH
d
8
2
2
a
Reagents and conditions : (a) NaBH , EtOH, 20 °C, 0.5 h; (b) CrO , H SO , acetone 0°C; (c)
4
3
2
4
NHS, DCC, THF, 20 °C, 24 h; (d) HSCH CH NH .HCl, Et N, CH Cl , 20 °C, 24h.
2
2
2
3
2
2
2.2. Preparation of supramolecular nano-assemblies.
NAs from prodrugs 9cꢀf were prepared at a concentration of 2 mg/mL in a single step by
nanoprecipitation, occurring spontaneously by addition of an acetonic solution of the
bioconjugate (8 mg/ml) into MilliQ® water under vigourous stirring. Formation of NAs occurred
immediately without the use of any surfactant. The organic solvent was then evaporated at room
temperature (9cꢀf) under vacuum using a Rotavapor®. Then, NAs colloidal dispersions were
stored at 4°C. These NAs that exhibited the characteristic Tyndall effect of colloidal systems had
a polydispersity index lower than 0.12 as measured by quasiꢀelastic light scattering. The mean
diameters of the obtained NAs are depicted in Table 1. Interestingly, prodrug 9c with a short
twoꢀisoprene unit appendage gave satisfactory NAs. This finding is in agreement with our
9
ACS Paragon Plus Environment

Page 11 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
previous results about polyꢀisoprenoyl conjugates of gemcitabine establishing that twoꢀisoprene
2
1
units are sufficient to provide conjugates with selfꢀassembling properties. No significant
changes neither in the size nor in the polydispersity index of the NAs 9cꢀf were detected over a
three day storage period at 4°C. The colloidal stability of 9cꢀf could be correlated with the fairly
negative Zeta potentials observed (ξ = −30 mV for 9c, ξ = −20 mV for 9d, ξ = −21 mV for 9e
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
and ξ = −34 mV for 9f). Furthermore, these highly hydrolysable compounds were found
1
chemically unchanged by H NMR and by MS after 24 h for 9c and up to three days for 9dꢀf.
These results seem to indicate that, within the NAs, the polar head group was protected from the
external aqueous medium by the hydrophobic isoprenoid core, and that longer the polyꢀisoprene
chain is more stable the NAs are.
As expected, compounds 9a and 9b did not provide any autoꢀassembling properties, clearly
demonstrating that the triꢀsubstituted double bonds of the isoprenoid linker group played a
crucial role to provide NAs. NAs of conjugates 9cꢀf (2 mg/mL) were characterized using
Dynamic Light Scattering (DLS) to determine their size, polydispersity index and zeta potential
which are summarized in Table 1.
Table 1: Physicoꢀchemical characteristics of the NAs of prodrugs 9cꢀf at 2 mg/mL.
Mean
diameter
Zeta
potential
(mV)
Polydispersity
Index (PdI)
Drug
Loading (%)
Compound
(nm)
9
9
c
126
195
147
167
0.10
0.16
0.12
0.06
ꢀ30
ꢀ20
ꢀ21
ꢀ34
63.1
54.2
40.4
36.3
d
9e
9
f
10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
As no surfactant or any excipient was needed to obtain NAs from these materials, the drug
loading could be directly deduced from respective molecular mass of the drugs and isoprenoyl
moieties. This led to conclude that obtained NAs were characterized by high drug loadings
compared to the currently available nanoꢀcarriers.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2.3. Transmission Electron Microscopy.
Transmission Electron Microscopy (TEM) allowed to characterize individual particles, their
shape and their size. However we know that this characterization depends on their spreading
onto the grids. We first studied nanoꢀassemblies with conventional negative staining method and
we observed spherical shape (150 ꢀ 200 nm) as shown in Figure 1), but unusually a few
assemblies were present on the grids. We therefore explored others methods by varying the dye
or the staining method. Hence, large field of particles could be observed with various sizes (90 –
300 nm) which are consistent with measurements by the nanosizer method.
Figure 1: GeranyloxyꢀIFO NAs suspended in water are observed by transmission electron
microscopy using negative staining method by phosphotungstic acid solution. Magnification : A)
50,000; B) 85,000; C) and D) 140,000. scale bar : 100 nm.
11
ACS Paragon Plus Environment

Page 13 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 2: FarnesyloxyꢀIFO NAs suspended in water observed by transmission electron
microscopy using positive staining method (A, B) or by negative staining method (C, D).
Magnification: A) 50,000; B) 85,000; C) and D) 140,000 ꢀ scale bar: 100 nm.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.4. In vitro cytotoxic evaluation
In vitro cytotoxicity of compounds 6, 9aꢀf has been tested on Rhabdomyosarcoma (RMSꢀ1) and
Ewing sarcoma (A673) cell lines by MTS assay, using IFO as a control. In the absence of
cytochrome, IFO showed no activity, whereas all the preꢀactivated compounds, either as bulk
form or as NAs, demonstrated cytotoxic activities (Table 2) with IC ranging from 0.6 µM to
5
0
3
1.4 µM. This inhibition is shown in Figure 2 which depicts the action of different SQꢀIFO
concentrations on the cell proliferation of both cell lines after a 72h treatment. At a low
concentration (1 µM) there is no inhibition of cell growth, whereas the concentrations of 20 and
8
0 µM show an inhibition cell of 50 and 80%, respectively.
Figure 3: Illustration of SQꢀIFO (9e) cytotoxicity on RMSꢀ1 and A673 cell lines after 72h
incubation.
12
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
[SQꢀIFO] non toxic dose
1µM)
[SQꢀIFO] around IC50
(20µM)
[SQꢀIFO] toxic dose
(80µM)
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A673
RMSꢀ1
All these preꢀactivated compounds showed a cytotoxic activity after 72h incubation for the tested
cell lines, except 9b, which needed to be incubated as long as 96h to reach its IC . Surprisingly,
5
0
for the two cell lines, the NAs formulation of compound 9c led to a significant improved activity
by comparing the IC obtained as bulk form (Table 2).
5
0
Table 2: Cytotoxic activities of IFO derivatives on Ewing sarcoma (A673) and
Rhabdomyosarcoma (RMSꢀ1) cell lines (n=3).
Compound
In vitro Activity (IC50 µM)
A673
RMSꢀ1
3.8 ± 1.4
5.3 ± 1.7
11.3 ± 5.8
9.3 ± 4.2
0.6 ± 0.2
12.3 ± 2.0
11.4 ± 5.0
6
1.5 ± 0.5
17.7 ± 2.3
24.5 ± 2.1
25.3 ± 1.5
14.7 ± 5.0
18.7 ± 3.6
22.0 ± 5.
23.2 ± 1.3
9a
9b**
9c
9
c NAs
d NAs
9e NAs
f NAs
9
*
9
31.4 ± 4.7
(
*) value obtained with the RD rhabdomyosarcoma cell line; (**) 96 h incubation
13
ACS Paragon Plus Environment

Page 15 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The NAs squalenoyl prodrugs 9e and 9f were previously evaluated on a panel of cancer cell lines
(Table 3). As usually observed with other squalene conjugates, owing to their prodrug nature, 9e
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
and 9f displayed variable activities on most cell lines.
Table 3: Cytotoxic activities of polyꢀisoprenoyl prodrugs NAs (9e,f) on various cancer cell lines
(n=4).
Cell line
In vitro Activity (IC µM)
50
9e NAs
9f NAs
RD
SKꢀNꢀMC
UWꢀ479
SAOSꢀ2
MCFꢀ7 MDR
MCFꢀ7
37.0 ± 1.0
31.5 ± 1.1
31.4 ± 4.7
23.1 ± 1.1
30.5 ± 1.1
31.1 ± 4.0
119.0 ± 27.1
45.6 ± 23.2
38.6 ± 22.3
22.0 ± 8.6
35.4 ± 21.6
23.0 ± 4.3
15.5 ± 8.4
65.2 ± 1.1
139.1 ± 0.9
157.0 ± 52.2
160.9 ± 78.2
158.1 ± 83.8
>200
A549
M109
KB 3.1
128.5 ± 52.7
134.5 ± 41.9
132.1 ± 67.9
B16F10
MiaPaCAꢀ2
Regarding cytotoxic activities of SQꢀIFO (9e) and SQꢀthioꢀIFO (9f), both compounds were
active on rhabdomyosarcoma (RD) and neuroblastoma (SKꢀNꢀMC). SQꢀIFO (9e) activity on the
other cell lines was poor as IC were above 50.0 µM ( Table 3). However, it can be underline
5
0
that SQꢀthioꢀIFO (9f) showed interesting activities on glioma (UWꢀ479), osteosarcoma (SAOSꢀ
), lung cancer (M109) melanoma (B16F10) and pancreatic cancer (MiaPaCaꢀ2) cell lines with
similar activities ranging between 15.5 to 23.0 µM (Table 3). Moreover, SQꢀthioꢀIFO (9f)
2
14
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
showed intermediate cytotoxic activities on the other cell lines but reverse the multidrug
resistance on the breast cell line (MCFꢀ7 MDR) as shown in Table 3.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
.5. In vitro release of 4-HO-IFO from X-IFO prodrugs.
The release kinetic study of the 4ꢀHOꢀIFO (2) from the preꢀactivated compounds (6, 9bꢀf) was
performed over 24 h in mice plasma incubated at 37°C. All preꢀactivated IFO compounds (6, 9bꢀ
f) were studied using stock solutions in ACN. The concentrations of 4ꢀHOꢀIFO and of the preꢀ
activated compounds were assessed in mice plasma using a quantitative HPLCꢀMS/MS assay
and allowed to display the decrease profile of the preꢀactivated compounds (6, 9bꢀf) and the
appearance profile of the 4ꢀHOꢀIFO metabolite concentrations, both expressed in ng/mL, taking
into account that the 4ꢀHOꢀIFO compound was analyzed as the semicarbazone derivative. The
kinetics curves show that 4ꢀHOꢀIFO was indeed produced from all these preꢀactivated
derivatives, except for 9f. The Figure 4 shows the results obtained with the bulk form of the preꢀ
activated compounds. On one hand, it was observed that concentrations of 9b (13,17ꢀ
dimethyloctyloxyꢀIFO), 9c (geranyloxyꢀIFO), 9d (farnesyloxyꢀIFO) and 9e (SQꢀIFO) decreased
especially during the first two hours and then remained stable, as shown in Figure 4A. On
another hand, the concentration of compound 6 (MeOꢀIFO) decreased quickly during the first
th
two hours and became very low after the 6 hour. Moreover, for compound 9f (SQꢀthioꢀIFO),
the concentration decreased gradually to reach a low value after 23h of incubation. The Figure
4B shows the appearance curves of released 4ꢀHOꢀIFO from each XꢀIFO compound. We
observed that the compound 6 provided increasing 4ꢀHOꢀIFO concentrations after 2h with a
maximum value of 1250 ng/mL, and then slowly decreased until 23h. After 0.5h, compounds 9b
and 9d produced maximum 4ꢀHOꢀIFO concentrations of 200 and 700 ng/mL, respectively.
15
ACS Paragon Plus Environment

Page 17 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compounds 9c and 9e displayed 4ꢀHOꢀIFO concentrations of 350 and 400 ng/mL after 4 and 6h,
respectively. By contrast, concentration lower than 30 ng/mL (at the bottom of Figure 4B) were
detected with compound 9f in which the alkoxy group is replaced by an alkylsulfanyl residue.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 4: Release study of 4ꢀHOꢀIFO from XꢀIFO as bulk form in mice plasma incubated at
37°C for 24h. (♦ 6 (MeOꢀIFO), ▼ 9b (13, 17ꢀdimethyloctyloxyꢀIFO), ▲ 9c (geranyloxyꢀIFO), ×
9d (farnesyloxyꢀIFO), ● 9e (SQꢀIFO), ○ 9f (SQꢀthioꢀIFO and ꢀ 1 (IFO)): Timeꢀconcentration
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
profiles of XꢀIFO A) and of 4ꢀHOꢀIFO B).
A)
3000
2500
2000
1500
1000
500
0
0
2
4
6
8
10
12
14
16
18
20
22
24
Time (h)
B)
2500
2250
2000
1750
1500
1250
1000
750
500
250
0
0
2
4
6
8
10
12
14
16
18
20
22
24
Time (h)
17
ACS Paragon Plus Environment

Page 19 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 5A) shows the timeꢀconcentration profile obtained for the compounds 9cꢀf prepared as
NAs. Concentrations of compounds 9cꢀe first decreased for 4.5 h and then remained stable for
24h. Regarding compound 9f, its concentration decreased gradually and reached a minimum
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
value after 24h. Figure 5B) shows the formation of released 4ꢀHOꢀIFO from the NAs of the preꢀ
activated compounds. It was observed that both compounds 9c and 9d produced after 13h
incubation a maximum 4ꢀHOꢀIFO concentration of 1300 and 900 ng/mL respectively. On the
other hand, compounds 9e and 9f could release low concentrations of 4ꢀHOꢀIFO with
concentration of 350 ng/mL after 2h and 100 ng/mL after 4.5h, respectively.
18
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 5: Release study of 4ꢀHOꢀIFO from XꢀIFO as NAs in mice plasma incubated at 37°C for
24h. (▲ 9c (geranyloxyꢀIFO), × 9d (farnesyloxyꢀIFI), ● 9e (SQꢀIFO) and ○ 9f (SQꢀthioꢀIFO):
Timeꢀconcentration profiles of XꢀIFO A) and of 4ꢀHOꢀIFO B).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A)
2500
2000
1500
1000
500
0
0
2
4
6
8
10
12
14
16
18
20
22
24
time (h)
B)
2500
2250
2000
1750
1500
1250
1000
750
500
250
0
0
4
8
12
16
20
24
Time (h)
19
ACS Paragon Plus Environment

Page 21 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
. Discussion and Conclusion.
The pharmacomodulation strategy described herein of IFO has led to the synthesis of seven new
preꢀactivated IFO derivatives (MeOꢀIFO 6, pentyloxyꢀIFO 9a, 13,17ꢀdimethyloctyloxyꢀIFO 9b,
geranyloxyꢀIFO 9c, farnesyloxyꢀIFO 9d, SQꢀIFO 9e and SQꢀthioꢀIFO 9f). The engraftment of
geraniol, farnesol, squalenol or squalenecysteamide to IFO allowed obtaining four compounds,
which were able to selfꢀassemble into NAs.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
1
As described previously, the in vitro cytotoxicity of MeOꢀIFO (6) in absence of cytochrome
was similar to IFO with CYP bioꢀactivation. Concerning the other preꢀactivated IFO derivatives,
the in vitro cytotoxic studies demonstrated their activity in the same range, with IC between 0.6
5
0
and 31.4 µM, as expected. Indeed, the aim of this strategy was not to improve their intrinsic
cytotoxic activity as the released entity (4ꢀHOꢀIFO) is the same for all these preꢀactivated
compounds. The aim was to control the delivery of the active agent through the modulation of its
release depending on the properties of the engrafted moieties.
It was shown that 72hꢀincubation were needed for the determination of the IC , except for
5
0
1
3,17ꢀdimethyloctyloxyꢀIFO (9b), which seemed to act more slowly, as IC was reached after
50
96h. This may be tentatively explained by the increased steric hindrance due to the branched
methyl group in the 3’ position of the alkyl appendage. Such configuration could significantly
modify the release kinetics of the key entity (4ꢀHOꢀIFO). Concerning the in vitro kinetic study of
4ꢀHOꢀIFO release, we demonstrated that all the preꢀactivated IFO derivatives produced the key
entity leading to the isophosphoramide mustard, as they were designed for this purpose.
Nevertheless, SQꢀthioꢀIFO (9f) was not hydrolyzed into 4ꢀHOꢀIFO, likely due to the different
type of bound linkage between the squalenecysteamide and IFO: a sulfide rather than an ether
bound is present in the case of the other compounds (6, 9aꢀe). The relative hydrolytic stability of
20
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
9
f in respect to the O-alkyl derivatives 9aꢀe could be attributed to the unfavorable protonation of
the soft sulfur center, S-ethers being weaker bases than Oꢀether.
3
1
13
All studied prodrugs were obtained as diasteromeric mixtures as evidenced by P and C NMR.
Although each isomer could potentially release the parent drug at different rates, we chose to
evaluate the mixture as racemic without separation for two reasons: one of the reasons was that
these acidꢀsensitive prodrugs do not crystallize and must be separated by chromatography
leading to a substantial loss of material. However, the main reason was that the mechanism
leading to the formation of the 4ꢀOHꢀIFO metabolite does not involve any enzymatic process but
is a simple acid catalyzed hemiaminal ether hydrolysis via protonation, leading to the release of
the 4ꢀOꢀalkyl sideꢀchain moiety with formation of an iminium ion and water addition. Thus, the
Cꢀ4 asymmetric center disappears in this process and the subsequent water addition on the
iminium ion would be expected to give 4ꢀOHꢀIFO. Moreover, the produg stereochemistry
corresponds to the one observed in the metabolites resulting of CYP3A4 mediated bioactivation
of IFO (see Scheme 1) because 4ꢀOHꢀIFO is in equilibrium with the ringꢀopened aldoꢀ
isophosphoramide which leads to epimerization of the Cꢀ4 center in the more stable
stereochemistry. The hypothesis that the Cꢀ4 stereogenic center of these prodrugs displays only a
small influence on the cytotoxicity has been reinforced by the observation that both 4ꢀMeOꢀIFO
isomers were found to have very similar cytotoxicity on carcinoma KB cell line (20 and 22 µM
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
respectively). This result could probably be related to the axial conformation of the methoxy
group in both diastereomers differing only by the conformation of the phosphoryl (P=O) bond.
Similarly, we can postulate that the O-alkyl groups engrafted in the Cꢀ4 position of these
prodrugs are in the same axial position due to stereoelectronic control in the addition step. Thus,
the same unique relationship with the nitrogen lone pair in both isomers should lead to roughly
21
ACS Paragon Plus Environment

Page 23 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
similar rate of iminium formation when hydrolyzed. Finally, we have chosen to examine these
prodrugs in their racemic form since IFO used in the clinic is racemic and because it was
demonstrated that there is no difference in efficacy through the release of 4ꢀHOꢀIFO between
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
25
racemic IFO and its enantiomers. In addition, when in vivo pharmacokinetic (PK) studies were
performed, no statistically differences of therapeutic concentrations and PK parameters were
2
6
shown between them.
Taking into account the results of both studies, all the designed preꢀactivated compounds were
active without any CYP bioꢀactivation, contrary to IFO. This observation confirms the proof of
concept concerning the design of these new preꢀactivated isoprenic IFO derivatives, as these
compounds generated directly the key entity leading to the release of the isophosphoramide
12
13
mustard, as it was previously described by Hohorst and Takamizawa in the mid 70s. This will
allow further investigations on the administered dose reduction in order to limit the toxic side
effects occurring in high dose protocols. Moreover, the NAs formulation of the drug could lead
15
to nanomedecines designed for a controlled release of the key entity.
Among the designed compounds, four seemed to stand out by their interesting results. As our
main goal was to administer a safer drug, the high activity and fast 4ꢀHOꢀIFO release profile of
MeOꢀIFO (6) were an impediment for controlling optimal pharmacokinetics development of this
compound. Due to the lack of stability, the compound MeOꢀIFO released 4ꢀHOꢀIFO very
quickly (Figure 4B), the use of this compound in chemotherapy may be considered as equivalent
to the direct administration of 4ꢀHOꢀIFO, implying that 6 may never reach its target and could
lead to harm the patient. However, compounds 9cꢀf have interesting properties as they selfꢀ
assemble into NAs with important drug loading. They also show a gradual 4ꢀHOꢀIFO release
22
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
kinetic profile which could be useful for a future active targeting strategy using nanoꢀassembly
functionalization.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The next step of this pharmacomodulation strategy will be first to investigate and to explain the
reason of the increased activity of 9c as NAs, which could be imaged by cell internalization
investigations using labelled NAs. Then, the in vivo efficacy using human tumor xenografted
mice and the toxicological impact of these new preꢀactivated oxazaphosphorine prodrugs (bulk
and NAs) will be further investigated.
23
ACS Paragon Plus Environment

Page 25 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
. Experimental section.
.1. Chemistry
.1.1. General
The structure of the synthesized XꢀIFO products was characterized by H, C and P NMR
4
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
13
31
1
13
31
spectra, which were recorded on Bruker Avance 400 (400, 100 and 160MHz, for H, C and P,
respectively) spectrometers Bruker Daltonik, Bremen, Germany). Recognition of methyl,
(
1
3
methylene, methine, and quaternary carbon nuclei in C NMR spectra rests on the Jꢀmodulated
spinꢀecho sequence and DEPT135 experiments. Positive electrospray mass spectra (+ESIꢀMS)
®
were performed using a QuattroLC mass spectrometer (Waters, Manchester, UK) and high
resolution mass spectra (HRꢀMS) were achieved with ±3 ppm accuracy with an Agilent 6520
series QꢀTof mass spectrometer (Agilent Technologies, Santa Clara, USA) operating in positive
electrospray ionization. The size of the obtained NAs was measured using a Malvern particle
®
size analyzer (Zetasizer ). Analytical grade solvents were provided by Carlo Erba (ValꢀdeꢀReuil,
France). Ifosfamide (Holoxan®) was supplied from Baxter (Baxter Healthcare Ltd, Auckland,
New Zealand) with 99.5% purity. Borontrifluoride diethyl etherate, squalene, geraniol, pentanol,
farnesol transꢀtrans and 3,7ꢀdimethyloctanol were purchased to Alfa Aesar (Schiltigheim,
France). Dichloromethane was distilled over calcium hydride, under nitrogen. All reactions
involving airꢀ or waterꢀsensitive compounds were performed under positive nitrogen pressure.
Analytical thinꢀlayer chromatography was performed on silica gel 60F254 (Merck) and the
preparative chromatography was performed using column filled with Silica Gel 60A (Merck,
Darmstadt, Germany).
4.1.2. HPLC-MS/MS assay of X-IFO purity
24
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The six synthesized and purified XꢀIFO products 9aꢀf (pentyloxyꢀIFO, geranyloxyꢀIFO, 13,17ꢀ
dimethyloctyloxyꢀIFO, farnesyloxyꢀIFO, SQꢀIFO, SQꢀthioꢀIFO) could contain the residual
product MeOꢀIFO (6) and showed a batch purity of 99%. Purity of each XꢀIFO batch was
calculated using a quantitative HPLCꢀMS/MS assay of the concentration of MeOꢀIFO impurity.
Based on the percentage ratio of MeOꢀIFO and XꢀIFO concentrations, purity (%) of XꢀIFO was
expressed as: P = 100 x ((1ꢀ c MeOꢀIFO/c XꢀIFO)).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Briefly, HPLCꢀMS/MS analyses of reference MeOꢀIFO standards (10ꢀ400 ng/mL) and XꢀIFO
samples (diluted at 10 µg/mL) were performed using a 1100 series HPLC system (Agilent
®
Technologies, Waldbronn, Germany) fitted with an Uptisphere C18 5µm column, (2.1 × 100
mm). Isocratic elution was achieved with a flow rate of 0.25 mL/min using the mobile phase
(acetonitrile/ammonium formate 5mM, 95:5, v/v). The total run time was 16.0 min. Analytes
®
were detected using a QuattroLC triple quadrupole mass spectrometer (Waters, Manchester,
UK), operating in the positive electrospray ionization (capillary voltage, 3.5kV and cone voltage,
2
5V) using tandem mass spectrometry with their respective MRM transitions (as example, m/z
91 → 259 for MeOꢀIFO (6) using argon as collision gas and collision energy of 20 eV. The
2
unknown concentration of MeOꢀIFO in XꢀIFO samples was quantified using calibration curve of
®
MeOꢀIFO standards, thanks to Masslynx software. In the same way, the concentration of the
IFO impurity was quantified in the initial MeOꢀIFO product using HPLCꢀMS/MS (MRM) using
the transition m/z 261 → 154 for IFO.
4
.2. Pre-activated drug synthesis
4.2.1 Synthesis of MeO-IFO (6)
A solution of ifosfamide (1) (200 mg, 0.766 mmol,) and sodium tetrafluoroborate (NaBF ) (170
4
mg, 1.54 mmol) in methanol (10 mL) was introduced into an undivided cell equipped with two
25
ACS Paragon Plus Environment

Page 27 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
graphite rod electrodes at room temperature. The anodic oxidation was followed by checking the
full consumption of the starting material with thinꢀlayer chromatography (CH Cl /(CH ) CO
2
2
3 2
(
50:50, v/v). After which 3.1 F/mol of electricity was passed into the solution at an intensity of
0 mA, sodium carbonate (100 mg, 1 mmol) was added and the solvent was removed under
reduced pressure. The residue was then filtered after dissolution in CH Cl . After concentration
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
2
2
in vacuo, we obtained an oil residue corresponding to 4ꢀmethyloxyꢀifosfamide (4ꢀMeOꢀIFO)
1
(
215 mg, 96% yield). This product was then used as raw material without further purification. H
NMR ((CD ) CO, 400 MHz): δ 1.25 (m, 1 H, H ), 2.72 (m, 1 H, H ), 3.42 (m, 4 H, H and H ),
3
2
5α
5β
7
9
3
.52 (m, 4 H, H and H ), 3.84 (m, 3 H, H and H ), 4.26 (m, 1 H, H ), 4.62 (td, 1 H, J=3 and
10 8 6α 11 6β
1
3
21 Hz, H ). C NMR ((CD ) CO, 100 MHz): δ 29.50 (C ), 43.29 (C ), 44.23 (C ), 45.73 (C ),
4
3 2
5
7
8
10
3
1
4
9.67 (C ) 54.79 (C ) 62.36 (C ), 90.75 (C ). P NMR ((CD ) CO, 160 MHz): δ 7.73 (s) and
9 11 6 4 3 2
8.40 (s).
+
+
MS (+ESI): m/z 291.1 (100%) [M+H] ; 293.1 (70%) [M+2+H] .
+
HRMS (+ESI): m/z 291.0426 ([M+H] calcd for C H N O P Cl : 290.0354.
8
17
2
3
1
2
4.2.2. Synthesis of pentyloxy-IFO (9a)
To a stirring solution of 4ꢀMeOꢀIFO (6) (200 mg, 0.69 mmol) in anhydrous CH Cl (10 mL) was
2
2
added dropwise BF .OEt (86 µL, 0.69 mmol) under inert atmosphere at ꢀ78 °C. After 45
3
2
minutes, a solution of pentanol (148µL, 1.37 mmol) in CH Cl (0.5 mL) was added dropwise.
2
2
The mixture was stirred for an additional 1h at ꢀ78 °C and 10 mL of a saturated Na CO solution
2
3
were added. After extraction with CH Cl (3×20 mL), the organic phases were dried over MgSO
4
2
2
and concentrated using reduced pressure. The obtained residue was purified by column
chromatography (silica gel, CH Cl /(CH ) CO/Et N (95:5:0.5, v/v/v)) to yield 9a as a yellow oil
2
2
3 2
3
1
(
90 mg, yield 33%). H NMR ((CD ) CO, 400 MHz): δ 0.92 (t, 3H, J = 7.1 Hz, O(CH ) CH ),
3
2
2 4
3
26
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
.30ꢀ1.45 (m, 4H, CH CH CH ), 1.62 (m, 2H, OCH CH ), 2.02ꢀ2.07 (m, 3H, H and NH), 3.16ꢀ
3
2
2
2
2
5
3
.30 (m, 2H, OCH CH ), 3.47ꢀ3.55 (m, 4H, NCH CH Cl and NHCH CH Cl), 3.60ꢀ3.65 (m, 2H,
2
2
2
2
2
2
NCH CH Cl), 3.70ꢀ3.75 (m, 2H, NHCH CH Cl), 4.05ꢀ4.20 (m, 1H, H ), 4.35ꢀ4.50 (m, 1H, H ),
2
2
2
2
6
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
13
4.70ꢀ4.80 (dd, 1H, J = 20.5 Hz and = 2.9 Hz, H ). C NMR ((CD ) CO, 100 MHz): δ 14.3 (CH ,
4 3 2 3
O(CH ) CH ) 23.2 (CH , O(CH ) CH CH ), 29.3 (CH , O(CH ) CH CH CH ), 30.4 (CH ,
2
4
3 ,
2
2 3
2
3
2
2 2
2
2
3
2
OCH CH (CH ) CH ), 30.9 (CH , C ), 43.6 (CH , NCH CH Cl), 44.0 (CH , NHCH CH Cl),
2
2
2 2
3
2
5
2
2
2
2
2
2
46.6 (CH , NHCH CH Cl), 50.2 (CH , NCH CH Cl), 63.2 (CH , C ), 68.6 (CH ,
2 2 2 2 2 2 2 6 2
31
OCH (CH ) CH ), 89.8 (CH, C ). P NMR ((CD ) CO, 160 MHz): δ 7.60 (s) and 8.46 (s).
2
2 3
3
4
3 2
+
+
MS (+ESI): m/z 347.1 (100%) [M+H] ; 349.1 (70%) [M+2+H] .
+
HRMS (+ESI): m/z 347.1068 ([M+H] calcd for C H N O P Cl : 347.1058).
1
2
25
2
3
1
2
4.2.3. Synthesis of 13,17-dimethyloctyloxy-IFO (9b)
To a stirring solution of 4ꢀMeOꢀIFO (6) (200 mg, 0.69 mmol) in anhydrous CH Cl (10 mL) was
2
2
added dropwise BF .OEt (86 µL, 0.69 mmol) under inert atmosphere at ꢀ78°C. After 45
3
2
minutes, 3, 7ꢀdimethyloctanol (262 µL, 1.37 mmol) dissolved in a small volume of CH Cl (0.5
2
2
mL) was added dropwise, the mixture was continuously stirred for an additional 1 h. The
reaction was then stopped by addition at low temperature of 10 mL of a saturated Na CO3
2
solution. After extraction with CH Cl , the organic phases were dried over MgSO and
2
2
4
concentrated under reduced pressure. The obtained residue was subject to column
chromatography purification (silica gel, CH Cl /(CH ) CO/Et N (95:5:0.5, v/v/v). The different
2
2
3 2
3
fractions containing the compound of interest were pooled and concentrated under reduced
pressure to yield a yellow to brown oily residue corresponding to the compound of interest 9c
1
(90 mg, yield 28%).
H
NMR ((CD ) CO, 400 MHz):
δ
0.74 (m, 9H,
3
2
O(CH ) CH(CH )(CH ) CH(CH ) ), 1.05ꢀ1.20 (m, 4H, (CH ) CH(CH ) ), 1.60ꢀ1.70 (m,
2 2
3
2 2
3 2
2 2
3 2
27
ACS Paragon Plus Environment

Page 29 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
H,NCH CH Cl), 2.01ꢀ2.07 (m, 2H, H ), 2.91 (s,1H, NH), 3.15ꢀ3.30 (m, 4H, NCH CH Cl and
2
2
6
2
2
NHCH CH Cl), 3.45ꢀ3.60 (m, 4H, NCH CH Cl and NHCH CH Cl), 3.64 (m, 2H, H ),4.05ꢀ4.19
2
2
2
2
2
2
11
1
3
(m, 1H, H ), 4.35ꢀ4.50 (m, 1H, H ), 4.70 (m, 1H, H ). C NMR ((CD ) CO, 100 MHz): δ 20.0
5 5 4 3 2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
CH , CH(CH ) ), 23.0 (CH , CH(CH ) ), 25.4 (CH , CH CH(CH )), 28.7 (CH ,
3
3 2
3
3 2
3
2
3
3
CH (CH ) CH(CH ) ), 30.5 (CH2, C5), 40.0 (CH2, NHCH CH Cl), 43.6 (CH2,
2
2 2
3 2
2
2
CH CH CH(CH ) ), 44.0 (CH , NCH CH Cl), 46.7 (CH NCH CH Cl), 49.7 (CH
2,
2
2
3 2
2
2
2
2,
2
2
OCH CH CH(CH )), 50.3 (CH2, NHCH CH Cl), 63.2 (CH2, C6), 66.7 (CH2,
2
2
3
2
2
31
OCH CH CH(CH )), 89.8 (CH, C ). P NMR ((CD ) CO, 160 MHz): δ 8.47 (s) and 7.55 (s).
2
2
3
4
3 2
+
+
MS (+ESI): m/z 417.2 (100%) [M+H] ; 419.2 (70%) [M+2+H] .
+
HRMS (+ESI): m/z 417.1828 ([M+H] Calcd for C H N O P Cl : 417.1840)
1
7
35
2
3
1
2
4.2.4. Synthesis of geranyloxy-IFO (9c)
To a stirring solution of 4ꢀMeOꢀIFO (6) (200 mg, 0.68 mmol) in anhydrous CH Cl (10 mL) was
2
2
added dropwise BF .OEt (86 µL, 0.68 mmol) at ꢀ78°C. After 45 minutes, a solution of geraniol
3
2
(237 µL, 1.37 mmol) in CH Cl (0.5 mL) was added dropwise, the mixture was stirred for an
2 2
additional 1 h at –78 °C and quenched by addition of 10 mL of a saturated Na CO solution.
2
3
After extraction with CH Cl (3×20 mL), the combined organic phases were dried over MgSO
4
2
2
and concentrated under reduced pressure. The obtained residue was purified by column
chromatography (silica gel, CH Cl /(CH ) CO/Et N (95:5:0.5, v/v/v)) to yield 9c as a yellow oil
2
2
3 2
3
1
(80 mg, yield 25%). H NMR ((CD ) CO, 400 MHz): δ 1.62 (s, 3H, HC=C(CH )), 1.68 (s, 3H,
3
2
3
HC=C(CH ) ), 1.72 (s, 3H, HC=C(CH ) ), 2.01ꢀ2.09 (m, 6H, (CH ) CH=C(CH ) and
3
2
3 2
2 2
3 2
CH CH=(CH )), 2.09ꢀ2.16 (m, 2H, H ), 2.85 (m, 1H, NH), 3.15ꢀ3.28 (m, 4H, NCH CH Cl and
2
3
5
2
2
NHCH CH Cl), 3.62ꢀ3.67 (m, 2H, NCH CH Cl), 3.70ꢀ3.75 (m, 2H, NHCH CH Cl), 4.08ꢀ4.12
2
2
2
2
2
2
(m, 2H, OCH CH=CH(CH )), 4.12ꢀ4.19 (m, 1H, H ), 4.39ꢀ4.49 (m, 1H, H ), 4.75 (dt, 1H, J =
2
3
6
6
28
ACS Paragon Plus Environment