Synthesis and analysis of urea and carbamate prodrugs as candidates for melanocyte-directed enzyme prodrug therapy (MDEPT).

Article abstract of DOI:10.1016/S0968-0896(02)00097-4

The suitability of 4-di(2-chloroethyl)aminoanilino-4-hydroxyphenethylaminomethanone 2 to act as a prodrug for melanocyte-directed enzyme prodrug therapy (MDEPT) is assessed. Thus its synthesis, ability to generate a cytotoxic agent upon exposure to tyrosinase, and stability within different sera are reported. A comparison is made to illustrate that the new urea prodrug 2 is a more suitable candidate for MDEPT than the corresponding carbamate prodrug 1.

Full text of DOI:10.1016/S0968-0896(02)00097-4

Bioorganic & Medicinal Chemistry10 (2002) 2625–2633
Synthesis and Analysis of Urea and Carbamate Prodrugs
as Candidates for Melanocyte-Directed Enzyme
Prodrug Therapy (MDEPT)
Allan M. Jordan,a Tariq H. Khan,b Hugh Malkina and Helen M. I. Osborna,*
aDepartment of Chemistry, University of Reading, Whiteknights, Reading RG6 6AD, UK
bDepartment of Medical Oncology, Imperial College of Science, Technology and Medicine,
Charing Cross Campus, London W6 8RP, UK
Received 25 January2002; accepted 12 March 2002
Abstract—The suitability of 4-di(2-chloroethyl)aminoanilino-4-hydroxyphenethylaminomethanone2 to act as a prodrug for melanocyte-
directed enzyme prodrug therapy (MDEPT) is assessed. Thus its synthesis, ability to generate a cytotoxic agent upon exposure to
tyrosinase, and stability within different sera are reported. A comparison is made to illustrate that the new urea prodrug 2 is a more
suitable candidate for MDEPT than the corresponding carbamate prodrug 1. # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
other cells, it is theoreticallypossible to consider loca-
lised drug deliverystrategies based on prodrugs for this
enzyme. Fortuitously, melanoma cells contain high
levels of the tyrosinase enzyme, and since this enzyme is
virtuallyabsent from other cells, it is a good target for
in-situ activation of prodrugs.4 A number of tyrosinase
dependent prodrug strategies have been investigated for
the treatment of melanoma. For example, non-toxic
phenol and catechol prodrugs have been oxidised by
tyrosinase to afford toxic quinones within the vicinity of
melanoma tumours.5 In addition, we have recently
reported a novel tyrosinase meditated drug delivery
system which utilises tyrosinase to mediate the release of
cytotoxic drugs, from prodrugs, via a cyclisation/delivery
system outlined in Scheme 1.6
Malignant melanoma continues to be a serious clinical
problem on account of its increasing incidence, the
relativelyyoung population that it affects, and the poor
prognosis of the disseminated disease.1 The high mor-
talityrate is a reflection of the failure of melanoma cells
to respond to current cytotoxic treatment in the form of
radiation and chemotherapy. Thus there is an urgent
need for improved methods of treatment for this parti-
cular cancer.
A number of deliverysystems have been developed for
the localised deliveryof prodrugs to certain cancers,
2
and these offer the advantage of minimal non-localised
drug toxicity. For example, the antibody-directed
enzyme prodrug therapy (ADEPT)3 utilises antibodies
to selectivelydeliver non-mammalian enzmyes to
tumour cells. Prodrugs, which are substrates for the
introduced enzyme, are then administered with the
knowledge that theywill onlyrelease the toxic drug at
the tumour site, where the antibody–enzyme conjugate
resides. If an enzyme is expressed in high levels within a
tumour cell, and if this enzyme is virtually absent from
Although taxol, daunomycin, gemcitabine and nitrogen
mustards have previouslybeen incorporated within our
prodrugs, previous results have illustrated that nitrogen
mustards are the preferred cytotoxic agents for our sys-
tem.6,7 This is a consequence of steric hinderance
affecting the abilityof prodrugs derived from larger
cytotoxic agents to act as substrates for tyrosinase.6
Together with the fact that nitrogen mustards are often
used in prodrug strategies such as ADEPT,8 and are
indeed employed in the clinic for the treatment of mel-
anoma,9 nitrogen mustards can be considered as sui-
table cytotoxic agents for incorporation within our
MDEPT programme.
*Corresponding author. Tel.: +44-118-987-5123; fax: +44-118-931-
6331; e-mail: h.m.i.osborn@reading.ac.uk
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00097-4

2626
A. M. Jordan et al. / Bioorg. Med. Chem. 10 (2002) 2625–2633
Scheme 1. Proposed drug release mechanism mediated by tyrosinase. Reagents: (i) Tyrosinase
Scheme 2. Synthesis of urea-linked aniline mustard 2. Reagents: (i) p-Nitrophenyl isocyanate 4, pyridine, 100%; (ii) 10% Pd on carbon (10 mol.%),
abs ethanol, H2, 1 atm, 100%; (iii) ethylene oxide, 54%; (iv) thionyl chloride, 52% (see Table 1); (v) chloroacetaldehyde, sodium cyanoborohydride,
HCl/MeOH, pH 6, 69%.
The success of a localised deliverystrategyrelies upon
generation of the cytotoxic unit from the prodrug purely
at the tumour site. This therefore necessitates stability
of the prodrug within serum. Previous work within our
group had illustrated that carbamate prodrugs of
representative structure 1 were good substrates for
tyrosinase, and their ability to release the cytotoxic
phenol mustard moietyupon exposure to tyrosinase had
been demonstrated.6 We now wish to report our recent
studies within this area, which have investigated the
relative stabilities of carbamate and urea prodrugs
within sera as well as the abilityof urea prodrug 2 to
release nitrogen mustard upon exposure to tyrosinase.
steps. Synthesis of the urea prodrug 2 was accomplished
using an isocyanate intermediate (Scheme 2). Thus for-
mation of the urea prodrug skeleton was easilyachieved
bycoupling tyramine 3 with p-nitrophenyl isocyanate 4
to afford urea 5. Subsequent nitro reduction gave amine
6, in quantitative yield over the two steps. Amine 6 was
then either reacted with ethylene oxide to afford the
corresponding bis-diol 7, which was subsequently
transformed to the bis-chloride 2 using several different
chlorinating reagents (Table 1), or taken straight
through to prodrug 2 bya reductive amination step,
using chloroacetaldehyde. This reductive amination has
previously been reported in the synthesis of mono-ethyl
mustards,10 but has not been reported for entryto bis-
chloroethyl mustards. Although both synthetic routes
successfullyprovided access to the urea prodrug 2,
reductive amination resulted in a superior overall yield
(69%), compared to the traditional ethylene oxide
approach (28%) and offered the added advantages of
being consistentlyreproducible whilst avoiding the use
of toxic ethylene oxide.
Results and Discussion
Synthesis of phenyl and aniline mustard prodrugs
Carbamate-linked prodrug 1 was synthesised as pre-
6
viouslyreported, with an overall yield of 62%, over 5

A. M. Jordan et al. / Bioorg. Med. Chem. 10 (2002) 2625–2633
Table 1. Chlorinating reagents used in step (iv) of Scheme 2
EntryChlorinating reagent (step iv) Yield of
2627
talysed standard conditions (entry 1). However, more
importantly, when the amount of aldehyde was
increased from 4 equiv (entry1) to 8 equiv (entry6) it
was observed that 50% of amine 6 was converted to the
corresponding tertiaryamine 2 in 3 h, compared to 48h
under the standard conditions. Furthermore, as expected,
when more concentrated conditions were employed
(entry6 vs entry7) further rate enhancement was
observed. Therefore, in this instance, it can be concluded
that instead of using an excess of amine to favour the slow
formation of the iminium intermediate, it was possible to
use an excess of the readilyavailable aldehyde to achieve
the same effect.
2
1
2
3
4
Thionyl chloride
Mesyl chloride
CH3N+CHCl2
CCl4, PPh3
52%
37%
<5%
<5%
In order to achieve appreciable yields in the reductive
amination the reaction was typically left to run over
several days. Under standard Borch reaction condi-
tions11 an excess of amine is used to favour the other-
wise slow formation of the iminium intermediate, thus
increasing the overall rate of the reaction. To overcome
the need for a large excess of amine, various modifi-
cations of the Borch conditions have been reported. For
instance, Mattson et al.12 reported the use of titanium (IV)
isopropoxide to catalyse rapid equimolar reductive
aminations between ketones/aldehydes and various
A convergent synthesis of urea prodrug 2 was also
attempted as outlined in Scheme 3. In this case, tyr-
amine 3 was coupled with p-nitrophenyl chloroformate
8 to give carbamate 9. This was subsequentlyreacted
with the aniline mustard 10, which was prepared bytwo
methods as outlined in Scheme 4.
13
primaryand secondaryamines and McCarthyet al.
employed titanium tetrachloride with diisopropylethyl-
amine or triethylamine in the synthesis of hindered
amines, with good effect. We therefore investigated a
range of conditions that were considered to promote
the rate of reductive amination between amine 6 and
chloroacetaldehyde, including: Lewis acids and base
catalysis, concentration and excess aldehyde (Table
2).
Synthesis of the aniline mustard 10 is outlined in
Scheme 4. Diol 12 could either be formed byreaction of
p-nitrophenyl aniline 11 with ethylene oxide, under
acidic conditions, or byan S N2Ar fusion between
p-fluoronitrobenzene 13 and diethanolamine. In either
case, diol 12 was then converted into the corresponding
bis-chloride 14, using mesyl chloride or thionyl chloride.
Iron mediated nitro reduction and treatment with
HCl(g) then gave 10 (HCl salt). The SN2Ar fusion route
was found to proceed with enhanced overall yield; 63%
versus 22% for the ethylene oxide route.
Table 2 shows that titanium (IV) isopropoxide (entry4
& 5) increases the reaction rate, compared to the unca-
Table 2. Affect of reaction conditions on the reductive amination of 6 in Scheme 2
EntryConditions
Reaction time (h) required for
Total%
Yield of 2
50% conversion of SM to product
1
2
3
4
5
6
7
4 eq. aldehyde, 10 mL MeOHꢀStandard conditions (SC)a
48
No product
No product
69
0
0
74
71
77
82
TiCl4 under SC
TiCl4+Et3N under SC
Ti(OiPr)4 under SC
12
12
3
Ti(OiPr)4+Et3N under SC
8 eq. aldehyde, 10 mL MeOH
8 eq. aldehyde, 5 mL MeOH
2
aSCꢀStandard conditions: 6 (1 mmol), aldehyde (4 mmol), MeOH (10 mL), NaBH3CN (2.2 mmol), 6N HClꢁpH 6.
Scheme 3. Convergent synthesis of urea prodrug 3. Reagents: (i) p-Nitrophenyl chloroformate 8, CH2Cl2, 97%; (ii) 10, dimethylformamide, tri-
ethylamine, 52%.

2628
A. M. Jordan et al. / Bioorg. Med. Chem. 10 (2002) 2625–2633
Scheme 4. Synthesis of aniline mustard hydrochloride 10. Reagents: (i) ethylene oxide, acetic acid, 30%; (ii) N,N-diethanolamine, 130 ꢂC, 84%; (iii)
MsCl, pyridine, 83% or thionyl chloride, pyridine/CH2Cl2, 86%; (iv) Fe, HCl, EtOH and then HCl(g), diethyl ether, 87%.
When the nitrogen mustard 10 was used in the route
outlined in Scheme 3 poor yields of urea prodrug 2 were
obtained, due to instabilityof the aniline mustard 10, in
basic solution and attack of the liberated p-nitrophenol
upon the newlyformed urea prodrug 2. Thus for further
studies, urea prodrug 2 was synthesised according to the
method outlined in Scheme 2.
Ability of urea prodrug 2 to act as a substrate for
tyrosinase
Upon establishing reliable synthetic entry to prodrug 2
we assessed its abilityto act as a substrate for tyrosinase
Graph 1. Oximetrydata for urea prodrug 2 (100 mL of a 10 mM solution).
6
and, as before, this was achieved via oximetrystudies.
If prodrug 2 is oxidised according to the pathwayillu-
strated in Scheme 1, molecular oxygen will be absorbed
from the surrounding solution. The resulting oxygen
depletion can be measured using an oxygen sensor, and
the rate of oxidation uptake provides an indication of
the rate of tyrosinase oxidation of prodrug 2. We have
alreadyillustrated that the carbamate prodrug 1 is a
good substrate for tyrosinase (Rmax=17.5 nanomol/min
compared with Rmax=17.5 nanomol/min for tyrosine
methyl ester, the methyl ester of the natural substrate for
tyrosinase).6 Oximetrydata illustrated that prodrug 2 was
also a good substrate for tyrosinase (Rmax=16.6 nano-
mol/min, which again compares favourablywith that for
tyrosine methyl ester) but in this case a longer lag phase
was evident at the begininning of the oxidation process
(Graph 1).
An initial lag phase is common during tyrosinase-
catalysed oxidation of monohydric phenols and is a
consequence of tyrosinase requiring activation by a
catechol before oxidation of the phenol can com-
mence.14 To diminish this lag period, further oximetry
studies were modified to allow addition of catalytic
Graph 2. Oximetrydata for urea prodrug 2 (100 mL of 10 mM solu-
tion) with N-methyl dopamine (5 mL of 10 mM solution) being added
at t=8 min.

A. M. Jordan et al. / Bioorg. Med. Chem. 10 (2002) 2625–2633
2629
quantities of a catechol, N-methyl dopamine. As expec-
ted, catechol mediated activation of tyrosinase then
allowed rapid oxidation of prodrug 2, as illustrated in
Graph 2. For comparison, oximetrydata for N-methyl
dopamine is presented in Graph 3.
Monitoring the ability of prodrug 2 to undergo
cyclisation/drug release upon exposure to tyrosinase
Having established that prodrug 2 was indeed a sub-
strate for tyrosinase, our attention next focussed on
determining whether tyrosinase mediated drug release
would occur from prodrug 2. Such a process had pre-
viouslybeen proven byHPLC for prodrug 1. Pleasingly,
HPLC analysis indicated that drug release also occurred
when prodrug 2 was exposed to mushroom tyrosinase at
25 ꢂC in H2O/DMSO (Trace 1).
Graph 3. Oximetrydata for N-methyl dopamine (100 mL of a 10 mM
solution).
Trace 1. This trace shows release of aniline mustard after 30 min exposure of urea prodrug 2 to mushroom tyrosinase at 25 ꢂC in H2O/DMSO (300
units of tyrosinase used per mL of solution).
Trace 2. Stabilityof carbamate prodrug 1 and urea prodrug 2 in human and bovine sera at 37 ꢂC, as analysed by HPLC.

2630
A. M. Jordan et al. / Bioorg. Med. Chem. 10 (2002) 2625–2633
tions and hydrolytic conditions. This was in contrast to
the results obtained for the carbamate prodrug 1. Graph
4 demonstrates that the bis-chloro moietyof the urea
mustard compound 2 was verystable to DO exchange
in a solution of D2O/d6-DMSO. In fact, little exchange
was observed over 48 h at room temperature and it was
onlyafter several days that appreciable exchange was
observed (40% mono-chloro mono-hydroxy compound
15 and 10% bis-hydroxy compound 16; see Scheme 5).
Presumablyexchange takes place via slow formation of
an aziridine intermediate which is rapidlyre-opened by
D2O, as shown in Scheme 5.
The enhanced stabilityand prolonged half life of the
urea prodrug, relative to the carbamate prodrug, within
sera, is a particularlyinteresting result. Since a number
of prodrugs employed within conventional ADEPT
strategies contain the carbamate functional group this
work highlights the potential of increasing the stability
and half lives of such prodrugs bythe introduction of
urea, rather than carbamate, linkages.
Graph 4. Investigation of halogen/DO exchange of compound 2 in
D2O/DMSO.
Stability of carbamate and urea prodrugs 1 and 2 in
bovine and human sera
Conclusions
Efforts were next focussed on determining the stability
of carbamate prodrug 1 and urea prodrug 2 in sera. To
avoid systemic toxicity within prodrug strategies it is
paramount that drug release from the prodrugs only
occurs at the targeted site, thus localising alkylation
within or around the malignant tumour. In order to
achieve this, and generate prodrugs with useful half
lives, in vivo stabilityof the urea/carbamate linker is
imperative. HPLC was used to assess the stabilityof the
prodrugs 1 and 2 in human and bovine sera at 37 ꢂC for
2.5 h (Trace 2). Stabilityof the mustard moietywas also
In conclusion, we have demonstrated that the urea pro-
drug 2 offers enhanced stabilityin sera compared with
our previous lead prodrug 1. Since prodrug 2 is a good
substrate for tyrosinase, and is able to release a cyto-
toxic moietyupon exposure to tyrosinase, it remains a
good candidate for the treatment of melanoma.
Experimental
1
investigated by H NMR in D2O (Graph 4).
All NMR spectra were recorded on a Bruker WM250,
Bruker AC250, Bruker Avance DPX 250, Bruker
AMX400 or Jeol AX400 spectrometer, using CHCl3 as
an internal standard unless stated otherwise (7.26 ppm
Pleasingly, it was evident that the urea prodrug 2 was
stable for prolonged times under physiological condi-
Scheme 5. Possible pathwayof halogen/DO exchange, via an aziridine intermediate.

A. M. Jordan et al. / Bioorg. Med. Chem. 10 (2002) 2625–2633
2631
for 1H NMR, 77.0 ppm for 13C NMR). 13C spectra were
recorded using Distortionless Enhancement byPolar-
isation Transfer. Mass spectra were recorded on a
Fisons VG Autospec. Infra red spectra were recorded
on a Perkin–Elmer Paragon 1000 FTIR spectrometer.
Melting points were determined using an Electrothemal
digital melting point apparatus, and are uncorrected.
LCMS was performed using a Waters 600 system with a
Micromass mass spectrometer. Stationaryphase for
LCMS was a Phenomenex Luna 5m, C18(2), 250Â4.6 mm.
Oximetrywas performed using a YSI model 5300 biologi-
cal oxygen monitor.
(C), 131.2, 4Â(CH), 131.7, 2Â(C), 157.3 (CO); (HRMS
(CI), found: [M+H]+ 396.1121. C19H23Cl2N3O2
[M+H]+, requires 395.1167); m/z (CI) (396 [M+H]+,
15%), 259 (45), 209 (100), 146 (20), 107 (40).
4-Hydroxyphenethylamino-4-nitroanilinomethanone (5).
Tyramine 3 (5 g, 36.5 mmol) was dissolved in an-
hydrous pyridine (30 mL) and cooled to 0 ꢂC. p-Nitro-
phenyl isocyanate 4 (5.98 g, 36.5 mmol) was added
slowlyand the mixture was stirred and allowed to warm
to room temperature. The reaction was monitored by
TLC (EtOAc) and upon completion (approx 3 h) the
mixture was concentrated in vacuo to afford urea 5 as a
yellow solid (10.98 g, 100%); mp 175–176 ꢂC; lmax (KBr
disc) 3350, 1680, 1615, 1560, 1512, 1370, 1285, 1110,
Unless stated otherwise, all chemicals and materials
were obtained from the Sigma–Aldrich Chemical Com-
pany, the B.D.H. Merk Chemical Company or Lan-
caster Chemicals and were used as received. Silica gel
for column chromatographywas obtained from Merck,
with a pore diameter of 6 nm. LCMS samples were pre-
pared byfiltering through Waters Sep-Pak cartridges
and run using a mobile phase of H2O (0.1% TFA) 90%:
acetonitrile 10% for 2 min to 100% acetonitrile over
10 min, at a flow rate of 1 mL/min. Mushroom tyrosinase
(3520 units/mg) was used at a concentration of 300 units/
mL in 0.1 M phosphate buffered saline (pH 7.4).
1
825 cmÀ1; d H NMR (400 MHz, CD3OD) 2.72 (2H, t,
J=6.3 Hz, PhCH2), 3.36 (2H, t, J=6.3 Hz, CH2), 6.71
(2H, d, J=7.9 Hz, Ar), 7.04 (2H, d, J=7.9 Hz, Ar), 7.54
(2H, d, J=7.9 Hz, Ar), 8.11 (2H, d, J=7.9 Hz, Ar); d
13C NMR (100 MHz, CD3OD) 36.6 (CH2), 42.9 (CH2),
116.7, 2Â(CH), 118.8, 2Â(CH), 126.3, 2Â(CH), 131.1,
2Â(CH), 131.5 (C), 143.3 (C), 148.2 (C), 157.4 (C and
CO); (HRMS (CI), found: [M+H]+ 302.1055.
C15H15N3O4 [M+H]+, requires 301.1063; m/z (CI)
(302 [M+H]+, 10%), 163 (20), 139 (100), 123 (25) 107
(65).
4 - Di(2 - chloroethyl)aminoanilino - 4 - hydroxyphenethyl -
aminomethanone (2)
4-Aminoanilino-4-hydroxyphenethylaminomethanone (6).
Compound 5 (5 g, 16.6mmol) was dissolved in ethanolabs
(250 mL) and 10% palladium on carbon was added
(1.5 g). The reaction mixture was stirred under a hydro-
gen atmosphere for 12 h, concentrated in vacuo and fil-
tered through Celite1 to afford 6 as an off white solid
(4.5 g, 100%); mp 188–190 ꢂC; nmax (KBr disc) 3350,
[Step e Scheme 3, thionyl chloride method]. Com-
pound 7 (1 g, 2.7 mmol) was suspended in a solution of
dichloromethane (20 mL) and pyridine (3 mL) and
cooled to 0 ꢂC. Thionyl chloride (0.42 mL, 5.8 mmol)
was added and the mixture was heated at reflux for 1 h,
allowed to cool and diluted with dichloromethane
(100 mL). The mixture was washed with water
(2Â50 mL) and the organics were dried (MgSO4) and
concentrated in vacuo to give an off white solid, which
was purified bycolumn chromatography(silica gel,
EtOAc) to yield mustard 2 as a white solid (550 mg,
52%).
1
1660, 1615, 1560, 1512, 1370, 1110, 825, 820 cmÀ1; d H
NMR (400 MHz, CD3OD) 2.56 (2H, t, J=7.0 Hz,
PhCH2), 3.22 (2H, t, J=7.0 Hz, CH2), 6.55–6.62 (4H,
m, Ar), 6.88–6.98 (4H, m, Ar); d 13C NMR (100 MHz,
CD3OD) 36.9 (CH2), 43.1 (CH2), 116.6, 2Â(CH), 117.5,
2Â(CH), 124.1, 2Â(CH), 131.2, 2Â(CH), 131.7 (C),
133.7 (C), 145.0 (C), 156.1 (C), 159.7 (CO); (HRMS
(CI), found: [M+H]+ 272.1386. C15H17N3O2
[M+H]+, requires 272.1399); m/z (CI) (272 [M+H]+,
40%), 165 (80), 137 (85), 108 (35).
[Step c Scheme 3, reductive amination method]. Com-
pound 6 (500 mg, 1.8 mmol) was dissolved in methanol
(5 mL). Chloroacetaldehyde (1.76 mL, 14.4 mmol) as a
45% aq solution and sodium cyanoborohydride
(260 mg, 4.25 mmol) were added slowly. The mixture
was acidified (pH 6) using acetic acid, stirred at room
temperature for 5 h, acidified with concd HCl (pH 2)
and concentrated in vacuo. The resulting oilyresidue
was dissolved in dichloromethane (150 mL) and washed
with a 10% sodium bicarbonate solution. The organic
layer was dried (MgSO4) and concentrated in vacuo to
give an off white solid. Purification bycolumn chroma-
4-Di(2-hydroxyethyl)aminoanilino-4-hydroxybenzylamino-
methanone (7). Compound 6 (2 g, 7.4 mmol) was dis-
solved in acetic acid (50 mL) and H2O (50 mL) and
cooled to 0 ꢂC. To this mixture 8 equiv of ethylene oxide
(59 mmol) was added slowly. The reaction mixture was
stirred and allowed to warm to room temperature. After
12 h a further 8 equiv of ethylene oxide (59 mmol) was
added slowlyand the mixture was stirred for a further
12 h. After this time, the solution was concentrated in
vacuo to give a black oil, which was purified byflash
chromatography(silica gel EtOAc, then MeOH). The
MeOH fraction was concentrated in vacuo and recrys-
talised (Acetone/EtOAc, 1.25:1) to give a brown solid,
further recrystalisation (MeOH) afforded diol 7 as an
off white solid (1.43 g, 54%); mp 186 ꢂC; nmax (KBr disc)
3350, 1680, 1615, 1560, 1400, 1285, 1110, 825 cmÀ1; d
1H NMR (400 MHz, CD3OD) 2.68 (2H, t, J=7.0 Hz,
PhCH2), 3.32 (2H, t, J=7.0 Hz, CH2), 3.47 (4H, t,
tography(silica gel, EtOAc) iyelded mustard
2 as a
white solid (583 mg, 82%); mp 95 ꢂC; nmax (KBr disc)
3350, 1680, 1670, 1615, 1560, 1512, 1370, 1110 cmÀ1; d
1H NMR (400 MHz, CD3OD) 2.69 (2H, t, J=7.0 Hz,
PhCH2), 3.34 (2H, t, J=7.0 Hz, CH2), 3.29–3.31 (8H,
m, 2ÂClCH2CH2N), 6.69–6.76 (4H, m, Ar), 7.04 (2H, d,
J=11.3 Hz, Ar), 7.14 (2H, d, J=11.3 Hz, Ar); d 13C NMR
(100MHz, CD3OD) 36.9, 2Â(CH2), 43.1, 2Â(CH2), 46.4
(CH2), 55.1, (CH2), 114.6, (C), 116.6, 4Â(CH), 124.2,

2632
A. M. Jordan et al. / Bioorg. Med. Chem. 10 (2002) 2625–2633
J=6.1 Hz, 2ÂCH2N), 3.68 (4H, t, J=6.1 Hz,
2ÂCH2OH), 6.69–6.74 (4H, m, Ar), 7.03–7.11 (4H, m,
Ar); d 13C NMR (100 MHz, CD3OD) 36.7 (CH2), 43.0
(CH2), 56.8, 2Â(CH2), 61.8, 2Â(CH2), 116.7, 2Â(CH),
120.8, 2Â(CH), 124.5, 2Â(CH), 131.2, 2Â(CH), 131.6
(C), 143.8 (C), 157.3 (C and CO), 158.0 (C); (HRMS
(CI), found: [M+H]+ 360.1845. C19H25N3O4
[M+H]+, requires 359.1845); m/z (CI) (360 [M+H]+,
100%), 315 (20), 223 (25), 133 (15), 100 (35).
[Step ii Scheme 4, fusion between 1-fluoro-4-nitro-ben-
zene and diethanolamine]. 1-Fluoro-4-nitro-benzene 13
(11.75 g,
83.3 mmol)
and
N,N-diethanolamine
(83.3 mmol) were mixed together and heated at 130 ꢂC
for 2 h. The mixture was allowed to cool to 60 ꢂC before
0.6% sodium hydroxide solution (890 mL) was added to
give a yellow precipitate. The precipitate was filtered
and dried (MgSO4) to afford diol 12 as a yellow solid
(16 g, 84%). Mp 104–105 ꢂC (lit.16 103–104 ꢂC); d H
1
NMR (400 MHz, (CD3OD) 3.66–3.81 (8H, m,
2ÂArNCH2CH2), 6.80 (2H, d, J=13.1 2ÂArH), 8.0
(2H, d, J=13.1 2ÂArH); d 13C NMR (100 MHz,
(CD3OD); 55.2 (2ÂCH2), 60.4 (2ÂCH2), 112.3 (2ÂCH),
127.3 (2ÂCH), 138.2 (C), 155.1 (C); (HRMS (CI),
found: [M+H]+ 227.1002. C10H14O2N2 [M+H]+,
requires 226.0954); m/z (CI) (227 [M+H]+, 100%), 195
(35).
4-Hydroxyphenethylamino-4-nitrophenoxymethanone (9).
3
Tyramine
(1 g, 7.3 mmol) and p-nitrophenyl-
chloroformate 8 (1.4 g, 7.3 mmol) were dissolved in
anhydrous dichloromethane and heated under reflux for
2 h. The reaction mixture was allowed to cool, con-
centrated in vacuo and purified bydryflash column
chromatography(silica, dichloromethane and then
ethylacetate) to afford carbamate 9 as a pale yellow
solid (2.2 g, 97%); mp 157–159 ꢂC; nmax (KBr disc) 3400,
1
1658, 1440, 1380 cmÀ1; d H NMR (400 MHz, CDCl3)
1N-(2-Chloroethyl)-1N-(3-chloropropyl)-4-nitroaniline
(14). Compound 12 (16 g, 71 mmol) was suspended in
dichloromethane (160 mL) and pyridine (10 mL) and
cooled to 0 ꢂC. Thionyl chloride (160 mmol) was added
slowlyand the mixture was stirred and heated at reflux
for 1 h. After cooling the mixture was diluted with di-
chloromethane (160 mL) and carefullywashed with
water (200 mL). The organics were dried (MgSO4) and
concentrated in vacuo to give mustard 14 as a dark yel-
low solid (16.1 g, 86%); mp 92–94 ꢂC (lit.17 94–95 ꢂC); d
1H NMR (400 MHz, (CD3OD) 3.58—3.68 (4H, m,
2ÂCH2Cl), 3.75–3.84 (4H, m, 2ÂArNCH2), 6.68 (2H,
dd, J=7.5 Hz, 2ÂArH), 7.93 (2H, dd, J=7.5 Hz,
2ÂArH); d 13C NMR (100 MHz, (CD3OD); 41.6
(2ÂCH2), 54.4 (2ÂCH2), 112.3 (2ÂCH), 127.4 (2ÂCH),
138.8 (C), 153.7 (C).
2.74 (2H, t, J=7.0 Hz, CH2Ar), 3.4 (2H, t, J=7.0 Hz,
CH2) 6.76 (2H, d, J=8.5 Hz, 2ÂArH), 7.06 (2H, d,
J=8.5 Hz, 2ÂArH), 7.29 (2H, d, J=9.2 Hz, 2ÂArH),
8.24 (2H, d, J=9.2 Hz, ArH); d 13C NMR (100 MHz,
CDCl3); 36.4 (CH2), 44.2 (CH2), 116.6 (2ÂCH), 123.7
(2ÂCH), 126.4 (2ÂCH), 131.2 (2ÂCH), 131.3 (C) 146.9
(C), 156.0 (C), 157.4 (C), 158.1 (C). m/z (CI) 163 (25%),
139 (10), 107 (100) 65 (15).
1N,1N-Di(2-chloroethyl)-1,4-benzenediamine hydrochloride
(10). Compound 14 (5 g, 19 mmol) was dissolved in eth-
anol (200 mL) and 10% palladium on carbon (500 mg)
was added. The mixture was stirred under an atmo-
sphere of hydrogen for 12 h, filtered through Celite1
and concentrated in vacuo to give a greysolid. The solid
was redissolved in ethanol (100 mL) and diethyl ether
(100 mL) and hydrogen chloride gas was passed through
it. The resulting precipitate was filtered off under a
nitrogen atmosphere to yield aniline mustard hydro-
chloride 10 as a white solid (4.4 g, 87%); mp 213–215 ꢂC
Acknowledgements
We gratefullyacknowledge Phairson Medical Ltd and
the BBSRC for their financial support of this work, and
Professor P. A. Rileyfor manyuseful discussions.
(lit.15 210–214 ꢂC); d H NMR (400 MHz, (CD3)2SO)
1
3.75–3.85 (8H, m, 2ÂArNCH2CH2), 6.76–6.81 (2H, m,
2ÂArH), 7.15–7.20 (2H, m, 2ÂArH); d 13C NMR
(100 MHz, (CD3)2SO); 44.3 (2ÂCH2), 58.3 (2ÂCH2),
113.9 (2ÂCH), 116.0 (2ÂCH), 134.5 (C), 136.2 (C);
(HRMS (CI), found: [M+H]+ 234.1367. C10H14Cl2N2
[M+H]+, requires 233.1370); m/z (CI) (233 [M+H]+,
63%), 183 (100), 120 (51), 66 (15).
References and Notes
1. (a) For example see: Grin-Jorgensen, C.; Rigel, D. S.;
Friedman, R. J. Cutaneous Melanoma, JB Lippincott 2nd ed.;
1992, p 27. (b) Armstrong, B. K.; Kricker, A. Cancer Surv.
1994, 19, 219. (c) Parker, S. L.; Tong, T.; Bolden, S. Cancer J.
Clin. 1997, 47, 5.
2. (a) For example see: Lloyd, K. O. Drug News Perspec.
2000, 13, 463. (b) Denny, W. A. Curr. Pharm. Des. 1996, 2,
281. (c) Syrigos, K. N.; Epenetos, A. A. Anticancer Res. 1999,
19, 605.
3. (a) For example see: Bagshawe, K. D.; Sharma, S. K.;
Springer, C. J.; Rogers, G. T. Annals. Oncol. 1994, 5, 879. (b)
NiculescuDuvaz, I.; Springer, C. J. Adv. Drug Del. Rev. 1997,
26, 151. (c) NiculescuDuvaz, I.; Springer, C. J. Curr. Med.
Chem. 1995, 2, 687. (d) de Groot, F. M. H.; Damen, E. W. P.;
Scheeren, H. W. Curr. Med. Chem. 2001, 8, 1093. (e) Tietze,
L. F.; Herzig, T.; Fecher, A.; Haunert, F.; Schuberth, I. Chem.
Biochem. 2001, 10, 758. (f) Wright, J. E.; Rosowsky, A.
Bioorg. Med. Chem. 2002, 10, 493.
2-(2-Hydroxyethyl-4-nitroanilino)-1-ethanol (12).
[Step i Scheme 4, N-alkylation of 4-nitroaniline]. 4-
Nitroaniline 11 (13.8 g, 100 mmol) was suspended in
glacial acetic acid (140 mL) and water (140 mL) and
cooled to 0 ꢂC. Ethylene oxide (240 mmol) was added
slowlyand the mixture was stirred and allowed to warm
to room temperature. The reaction mixture was stirred
(4 days) until no starting material remained by TLC
(ethyl acetate/methanol 9:1) and concentrated in vacuo
to give a black oil. The oil was dissolved in ethyl acetate
(1 L), washed with water (300 mL), dried (MgSO4) and
concentrated in vacuo to give a brown oil. Purification
bycolumn chromatography(hexane/ethyl acetate 1:1)
afforded diol 12 as a pale yellow solid (6.8 g, 30%).

A. M. Jordan et al. / Bioorg. Med. Chem. 10 (2002) 2625–2633
2633
4. (a) Simonova, M.; Wall, A.; Weissleder, R.; Bogdanov, A.
Cancer Res. 2000, 60, 6656. (b) Prota, G.; d’Ischia, M.; Mas-
cagna, D. Mel. Res. 1994, 4, 351. (c) Valmori, D.; Pittet, M. J.;
Rimoldi, D.; Lienard, D.; Dunbar, R.; Cerundolo, V.;
Lejeune, F.; Cerottini, J. C.; Romero, P. Cancer Res. 1999, 59,
2167.
8. NiculescuDuvaz, I.; Springer, C. J. Anti-Canc. Drug Des.
1995, 10, 361.
9. Kerr, D. E.; Li, Z. G.; Siemers, N. O.; Senter, P. D.;
Vrudhula, V. M. Bioconj. Chem. 1998, 9, 255.
10. Bagshawe, K. D.; Haase-Held, M.; Mann, J.; Springer,
C. J. Tetrahedron 1990, 46, 5377.
5. (a) For example see: Lant, N. J.; McKeown, P.; Kelland,
L. R.; Rogers, P. M.; Robins, D. J. Anti-Canc. Drug Des.
2000, 15, 295. (b) Riley, P. A.; Cooksey, C. J.; Johnson, C. I.;
Land, E. J.; Latter, A. M.; Ramsden, C. A. Eur. J. Cancer
1997, 33, 135. (c) Gili, A.; Thomas, P. D.; Ota, M.; Jimbow,
K. Mel. Res. 2000, 10, 9.
6. (a) Jordan, A. M.; Khan, T. M.; Osborn, H. M. I.; Photiou,
A.; Riley, P. A. Bioorg. Med. Chem. 1999, 7, 1775. (b) Jordan,
A. M.; Khan, T. H.; Malkin, H.; Osborn, H. M. I.; Photiou,
A.; Riley, P. A. Bioorg. Med. Chem. 2001, 9, 1549.
7. Osborn, H. M. I.; Malkin, H. unpublished results.
11. Borch, R. F. Org. Synth. 1972, 52, 124.
12. Mattson, R. J.; Pharm, K. M.; Leuck, D. J.; Cowen, K. A.
J. Org. Chem. 1990, 55, 2552.
13. Barney, C. L.; Huber, E. W.; McCarthy, J. R. Tetrahedron
Lett. 1990, 39, 5547.
14. For example see: Riley, P. A. J. Theoretical Biol. 2000, 203, 1.
15. Abdel-Rahman, M. A.; Elliot, H. W.; Binks, R.; Kung,
W.; Rapopart, H. J. Med. Chem. 1966, 9, 1.
16. Tong, L. K. J. J. Am. Chem. Soc. 1960, 82, 1988.
17. Palmer, B. D.; Wilson, W. R.; Pullen, S. M.; Denny, W. R.
J. Med. Chem. 1990, 33, 112.