Design and synthesis of a nitrogen mustard derivative stabilized by apo-neocarzinostatin

Article abstract of DOI:10.1021/jm040790d

Neocarzinostatin (NCS) is an antitumor antibiotic comprising a 1:1 protein-chromophore complex and exhibits cytotoxic action through DNA cleavage via H-abstraction. Cytotoxic activity resides with the chromophore 1 alone, while the protein (apoNCS) protec

Full text of DOI:10.1021/jm040790d

4710
J . Med. Chem. 2004, 47, 4710-4715
Design a n d Syn th esis of a Nitr ogen Mu sta r d Der iva tive Sta bilized by
Ap o-n eoca r zin osta tin
Michael D. Urbaniak,† J ohn P. Bingham,‡ J ohn A. Hartley,‡ Derek N. Woolfson,§ and Stephen Caddick*,|
Department of Chemistry, School of Life Sciences, University of Sussex, Falmer, Brighton BN1 9QJ , U.K.,
Cancer Research UK Drug-DNA Interactions Research Group, Department of Oncology, University College London,
London W1W 7BS, U.K., Department of Biochemistry, School of Life Sciences, University of Sussex,, Brighton BN1 9QJ , U.K.,
and Department of Chemistry, Christopher Ingold Laboratories, University College London, 20 Gordon Street,
London WC1H OAJ , U.K.
Received February 23, 2004
Neocarzinostatin (NCS) is an antitumor antibiotic comprising a 1:1 protein-chromophore
complex and exhibits cytotoxic action through DNA cleavage via H-abstraction. Cytotoxic
activity resides with the chromophore 1 alone, while the protein (apoNCS) protects and
transports labile 1. The naphthoate portion (2) of NCS chromophore (1) is important for binding
to apoNCS and DNA intercalation. In this paper we describe our attempts to use apoNCS to
improve the hydrolytic stability of novel bifunctional DNA alkylating agents. The nitrogen
mustards, melphalan and chlorambucil, were both conjugated to 2, and the biological activities
of these conjugates were assessed. Chlorambucil did not benefit from conjugation. The
melphalan conjugate (6) formed covalent DNA adducts at guanine bases and exhibited greater
in vitro cytotoxic activity than unmodified melphalan. Fluorescence and NMR spectroscopy
showed that 6 binds to apoNCS. Binding to apoNCS-protected 6 reduced the extent of hydrolysis
of the conjugate. This novel approach demonstrates for the first time that an enediyne apo-
protein can be used to improve the stability of substances that are of potential interest in
cancer chemotherapy.
In tr od u ction
The nitrogen mustards, chlorambucil and melphalan,
produce interstrand cross-links by DNA alkylation and
are widely used in chemotherapy. However, the ef-
ficiency of these agents is limited by deactivation
through reaction with nucleophiles such as water,
proteins, and thiols and further through low affinity for
DNA. Attempts to address the latter deficiency have met
with some success: conjugation to DNA targeting
moieties (binders and intercalators) produces agents
with increased affinity for DNA in vitro, though the
effectiveness of these species in vivo is varied.1 To our
knowledge, no attempts have been made to address the
lability of these compounds. Herein, we describe a novel
strategy for the improvement of the stability of this class
of chemotherapeutic agent.
Neocarzinostatin (NCS), an antitumor antibiotic iso-
lated from Streptomyces carzinostaticus, comprises a
highly reactive enediyne chromophore 1 (Figure 1)
noncovalently bound to an 11 kDa protein (apoNCS).2-4
NCS has found clinical application in J apan against
leukemia and cancers of the bladder, stomach, pancreas,
liver, and brain.5 The protein component provides
essential protection for the labile chromophore from
heat, UV light, and attack by nucleophiles and is
reported to transport 1 into cells.6 The protein alone is
not cytotoxic. The cytotoxic activity results from cy-
F igu r e 1.
cloaromatization of the enediyne ring of 1, producing a
diradical species that targets DNA.7 Minor groove
binding and intercalation of the naphthoate moiety 2
positions the diradical species to facilitate hydrogen
abstraction from the deoxyribose sugar backbone, lead-
ing to single and double stranded DNA cleavage.8,9
Naphthoate 2 is also an important element in the
binding of the chromophore to the apo-protein; we have
demonstrated that simple NCS analogues that contain
the naphthoate 2 bind within the apoNCS binding site.10
Some of the deficiencies of the nitrogen mustards
might be remedied by conjugation to 2. First, noncova-
lent binding of such a conjugate to apoNCS has the
potential to protect the reactive nitrogen mustard and
to improve transport in vivo; second, the introduction
of an intercalating motif may improve the affinity for
DNA. The possible use of the protein as a stabilizer and
transporter for reactive small molecules is significant,
and while there are potential difficulties in using
protein-based drug delivery systems, the clinical use of
NCS indicates that such approaches may hold promise
and warrant further investigation. We report the syn-
* Corresponding author. S.Caddick@ucl.ac.uk, Phone and Fax +44
207 6794694.
† Department of Chemistry, University of Sussex.
‡ Department of Oncology, University College London.
§ Department of Biochemistry, University of Sussex.
| Department of Chemistry, University College London.
10.1021/jm040790d CCC: $27.50 © 2004 American Chemical Society
Published on Web 08/17/2004

Nitrogen Mustard Stabilization by Apo-neocarzinostatin
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 19 4711
Sch em e 1. Synthesis of Compounds 4 and 6a
a
Reagents and conditions: (a) chlorambucil (1.0 equiv), ethylene
glycol (10 equiv), DCC (1.1 equiv), DMAP (0.1 equiv), CH2Cl2,
0 °C, 16 h, 96%; (b) 2 (1.0 equiv), DCC (1.1 equiv), CH2Cl2:THF
(1:1), 0 °C, 16 h, 82%. (c) DCC (1.1 equiv), ethylene glycol (10
equiv), CH2Cl2:THF (1:1), 0 °C, 16 h, 91%. (d) DCC (1.1 equiv),
N-Boc melphalan (1.0 equiv), CH2Cl2:THF (1:1), 0 °C, 16 h, 41%.
(e) TFA:CH2Cl2 (1:1), 0 °C, 45 min, 99%.
F igu r e 2. Taq polymerase stop assay. C: negative control;
lane 1: 4 (2.5 µM); lane 2: chlorambucil (2.5 µM); lane 3: 6
(2.5 µM); lane 4: melphalan (2.5 µM).
Ta ble 1. Cytotoxicity and Binding Affinity of 4, 6, Melphalan,
and Chlorambucil
thesis of naphthoate conjugates of chlorambucil and
melphalan and demonstrate that the inherent in vitro
cytotoxicity of the latter is improved. Furthermore, we
evaluate their binding to apoNCS and demonstrate that
such binding improves the stability toward hydrolysis.
compds
cytotoxicity,a IC50, µM
Kd (apoNCS)
c
melphalan
6
chlorambucil
4
30 ( 10
7 ( 4
-
47 ( 5 µM
c
95 ( 25
-
200 ( 50b
>1 mMd
Resu lts a n d Discu ssion
a
b
Average of three experiments. Estimated by extrapolation.
Ch em istr y. Conjugation of chlorambucil or mel-
phalan to the NCS naphthoate 2 was achieved through
sequential dicyclohexyl carbodiimide (DCC) coupling
reactions utilizing ethylene glycol as a short flexible
linker.11 Coupling of chlorambucil gave 3 and the
subsequent coupling of 2 produced 4 in high yield,
Scheme 1. Prior to coupling the amino group of mel-
phalan was protected as the N-tert-butyloxycarbonyl (N-
Boc) derivative by treatment with di-tert-butyl dicar-
bonate (Boc)2O.12 Convergent coupling of N-Boc mel-
phalan to the ester 5 was found to be higher yielding
than the linear approach. Deprotection of the coupled
product under acidic conditions gave the desired product
6 in high yield. After purification all compounds were
protected from light and stored at -20 °C to prevent
degradation.
c Insufficient change in fluorescence. Binding too weak for ac-
d
curate measurement.
observation is consistent with the work of Sartorius et
al. who concluded that a positively charged side chain
is required for intercalation of naphthalene deriva-
tives;15 which is present in 6 but not 4.
The in vitro cytotoxic activity of the compounds
against the human leukemia cell line K562 was mea-
sured using the MTT assay, Table 1.16 The ability of
the compounds to inhibit cell growth correlated well
with their DNA damaging properties. The potency of 4
was reduced compared to chlorambucil, while 6 had
increased potency compared to the parent compound
melphalan. The reduced potency of 4 may reflect the
increased hydrophobicity of the compound.
In Vitr o Eva lu a tion of Biologica l Activity. Ni-
trogen mustards form covalent adducts with DNA via
formation of a transient aziridinium species that attack
the nucleophilic N7 position of guanine nucleotides.1 The
ability of the naphthoate conjugates 4 and 6 to form
covalent DNA adducts was measured using the Taq
polymerase stop assay,13 Figure 2. Both conjugates
produced DNA adducts, with the level of damage 4 ≈
chlorambucil < melphalan < 6. The position of DNA
damage observed was similar for all compounds, strongly
suggesting that DNA alkylation at the N7 position of
guanine nucleotides occurs.13 No significant alkylation
of adenine nucleotides was detected, demonstrating
that, unlike some intercalators,14 the naphthoate did not
alter the intrinsic sequence selectivity of the nitrogen
mustards. Conjugation of 2 increased the level of DNA
alkylation exhibited by melphalan but did not alter that
of chlorambucil, indicating that the presence of 2 alone
is insufficient to produce effective intercalation. This
Bin d in g to a p oNCS. The ability of the naphthoate
conjugates to bind to apoNCS was examined using
fluorescence quenching titration experiments.10 The
binding of the chlorambucil derivative 4 was weak and
did not allow a dissociation binding constant (Kd) to be
determined accurately. However, the melphalan deriva-
tive 6 showed significant binding to apoNCS, Table 1.
The lack of binding for 4 was surprising given that we
have previously demonstrated that conjugation of 2 is
sufficient to sequester other small molecules within the
apoNCS binding site.10 The alkyl chain of 4 is one
carbon longer than that of 6, which may in part account
for the difference in binding through increased flexibility
of the linker region or unfavorable steric hindrance in
the binding site. In addition, 6 possesses an amino group
six bonds from the naphthoate carbonyl; curiously, the
corresponding carbonyl and methlyamino groups of 1
have the same separation. In 6, but not 4, this may
allow favorable interaction between the positively charged

4712 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 19
Urbaniak et al.
F igu r e 3. (A) Backbone NH Min∆δ analysis versus residue number; (B) Backbone NH Min∆δ mapped onto apoNCS.17 The
coloring corresponds to four groups: >0.1 ppm: ‘highly shifted’ as red, 0.1-0.075 ppm: ‘moderately shifted’ as orange, 0.075-
0.06 ppm: ‘slightly shifted: as yellow, <0.06 ppm: ‘unshifted’ as blue, and unassigned residues in gray The position of NCS
chromophore is shown to aid identification of the binding site.
amino group and the aromatic ring of Phe 78 analogous
to an interaction observed in the crystal structure of
holo-NCS.17 The nitrogen mustards alone, which were
anticipated not to bind apoNCS, showed negligible
change in fluorescence in the presence of apoNCS.
The location at which 6 bound to apoNCS was
examined by following the perturbation of chemical shift
of backbone amides upon complex formation using
1H,15N HSQC NMR spectroscopy.18,19 In the absence of
large change in the tertiary conformation of the protein,
residues that display perturbed chemical shifts upon
binding can be interpreted as being close to the position
at which binding occurs. The location of chemical shift
F igu r e 4. Hydrolysis of 6 in the presence and absence of
apoNCS. Representative time course experiment shown.
perturbations was visualized by using the minimum
change in chemical shift (Min∆δ) values of backbone
amide resonances to color a ribbon representation of
to protect it from solvent, producing measurable benefit
NCS, Figure 3. The interactions were clearly localized
on the longevity of the small molecule. The level of
to the apoNCS binding cleft, consistent with those
protection achieved will depend on the dissociation
naphthoate conjugates that we have previously shown
constant Kd of the complex; as Kd decreases, the amount
to occupy the binding site.10 Residues with high and
of time that the drug is in free solution, and hence
moderate shifts were found in the â-strands that form
unprotected, will decrease. However, the location, spe-
the base and two sides of the binding site, which are in
cific orientation and hence the burial of the labile
close contact with the naphthoate moiety in the natural
portion of the drug within the apoNCS binding site will
holo-complex.17 Residues of the loops surrounding the
also strongly affect the level of protection.
apoNCS binding site also showed slight to moderate
shifts, including the Phe 78 resonance, which showed a
We tested the ability of apoNCS to improve the in
vitro cytotoxicity of 4 and 6. Coincubation of 4 or 6 with
moderate shift. The perturbation of Phe 78 may reflect
apoNCS did not alter their IC50 against the cultured
the postulated charged-aromatic interaction, as the
human leukemia cell line K562 (data not shown). The
benzyl side chain is known to alter conformation upon
former result reflects the lack of interaction between 4
binding of the natural chromophore.20
and apoNCS. The latter result is consistent with the
Binding to apoNCS provides essential protection for
results of Goldberg and co-workers,21,22 who have ex-
its natural and labile chromophore 1 from inactivation
tensively studied the mechanism of action of NCS. They
by heat, UV light, and attack by nucleophiles.21 We
demonstrated that, despite the extreme lability of the
postulated that the labile nitrogen mustard conjugate
chromophore,21 the chromophore alone is as active as
6 might be afforded similar protection from hydrolysis
the holo-complex against cultured HeLa cells.22 How-
when bound to apoNCS. Reversed-phase high-perfor-
ever, the holo-complex is essential for in vivo activity.5
mance liquid chromatography (RP-HPLC) was used to
follow the hydrolysis of 6 in the presence and absence
Con clu sion s
of apoNCS, Figure 4. The half-life (t1/2) of 6 changed from
3.5 ( 0.5 h to 8.1 ( 1.5 h in the presence of 1 equiv of
apoNCS, a greater than 2-fold reduction in the extent
of hydrolysis. No effect on the extent of hydrolysis of 4
was observed in the presence or absence of apoNCS
(data not shown). These data demonstrate that 6 binds
apoNCS in an orientation and with a lifetime sufficient
Previously, we have shown that simple NCS ana-
logues that contain the naphthoate 2 bind within the
apoNCS binding site.10 Here, we have demonstrated
that drug-naphthoate conjugates may retain the ability
to bind within the apoNCS binding site, and that such
binding provides protection from hydrolysis. We have

Nitrogen Mustard Stabilization by Apo-neocarzinostatin
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 19 4713
Ch em istr y. 4-{4-[Bis-(2-ch lor o-eth yl)-a m in o]-p h en yl}-
bu tyr ic Acid 2-Hyd r oxy-eth yl Ester (3). Chlorambucil (100
mg, 0.33 mmol, 1.0 equiv) was added to a stirred solution of
ethylene glycol (184 µL, 3.3 mmol, 10 equiv), DMAP (20 mg,
0.165 mmol, 0.5 equiv), and DCC (75 mg, 0.36 mmol, 1.1 equiv)
in dry CH2Cl2 (25 mL) under N2 at 0 °C, and the resulting
solution was stirred for 16 h. The reaction mixture was diluted
with hexane (20 mL) and dried over Na2SO4 and the solvent
removed in vacuo. Purification by column chromatography
(40:1 CH2Cl2:MeOH) gave 3 as a viscous oil (0.11 g, 0.32 mmol,
96%).
2-Hydr oxy-7-m eth oxy-5-m eth yl-n aph th alen e-1-car box-
ylic Acid 2-(4-{4-[Bis-(2-ch lor o-eth yl)-a m in o]-p h en yl}-
bu tyr yloxy)-eth yl Ester (4). DCC (50 mg, 0.24 mmol, 1.1
equiv) was added to a stirred solution of 2 (51 mg, 0.22 mmol,
1.0 equiv) and 3 (75 mg, 0.22 mmol, 1.0 equiv) in dry CH2Cl2:
THF (1:1, 10 mL) under N2 at 0 °C and the resulting solution
was stirred for 16 h. The reaction mixture was washed with
brine (3 × 50 mL), extracted with CH2Cl2 (3 × 100 mL), dried
over MgSO4 and the solvent removed in vacuo. Purification
by column chromatography (3:1 PE:EtOAc) gave 4 as a viscous
oil (0.10 g, 0.18 mmol, 82%). RP-HPLC >95% unhydrolyzed.
2-Hydr oxy-7-m eth oxy-5-m eth yl-n aph th alen e-1-car box-
ylic Acid 2-Hyd r oxy-eth yl Ester (5). DCC (73 mg, 0.36
mmol, 1.1 equiv) was added to a stirred solution of 2 (75 mg,
0.32 mmol, 1.0 equiv) and ethylene glycol (0.2 mL, 3.6 mmol,
11 equiv) in dry CH2Cl2:THF (1:1, 15 mL) under N2 at 0 °C,
and the resulting solution was stirred for 16 h. The reaction
mixture was warmed to room temperature, and the solvent
was removed in vacuo. The residue was dissolved in ether (20
mL), passed though a pad of Celite, and dried over MgSO4 and
the solvent removed in vacuo. Purification by column chro-
matography (4:1 PE:EtOAc) gave 5 as white solid (79 mg, 0.29
mmol, 91%).
P r ep a r a tion of 2-Hyd r oxy-7-m eth oxy-5-m eth yl-n a p h -
th a len e-1-ca r boxylic Acid 2-(3-{4-[Bis-(2-ch lor o-eth yl)-
a m in o]-p h en yl}-2-ter t-bu toxyca r bon yla m in o-p r op ion yl-
oxy)-eth yl Ester (5′). DCC (62 mg, 0.29 mmol, 1.1 equiv) was
added to a stirred solution of 5 (75 mg, 0.27 mmol, 1.0 equiv)
and N-Boc melphalan (106 mg, 0.27 mmol, 1.0 equiv) in dry
CH2Cl2 (15 mL) under N2 at 0 °C, and the resulting solution
was stirred for 16 h. The reaction mixture was warmed to room
temperature, hexane (50 mL) added, and the suspension
filtered to remove excess urea. The resulting solution washed
with brine and dried over MgSO4 and the solvent removed in
vacuo. Purification by column chromatography (4:1 P:EtOAc)
gave 5′ as an oil (79 mg, 0.11 mmol, 41%).
P r ep a r a tion of 2-Hyd r oxy-7-m eth oxy-5-m eth yl-n a p h -
th a len e-1-ca r boxylic Acid 2-(2-Am in o-3-{4-[bis-(2-ch lor o-
eth yl)-am in o]-ph en yl}-pr opion yloxy)-eth yl Ester (6). TFA
(10 mL, excess) was added to a solution of 5′ (59 mg, 89 µmol)
in dry CH2Cl2 (10 mL) at 0 °C under N2 and the reaction stirred
for 1 h. The TFA was quenched with NaHCO3, extracted with
CH2Cl2, and dried over MgSO4 and the solvent removed in
vacuo. Purification by column chromatography (40:1 CH2Cl2:
MeOH) gave 6 as an oil (50 mg, 89 µmol, 99%). RP-HPLC
>95% unhydrolyzed.
Ta q P olym er a se Stop Assa y. Taq polymerase stop assays
were conducted in a manner analogous to that described
previously.13 Template DNA (pUC18 Hind III digest, 0.5 µg)
in FP buffer (50 mM KCl, 20 mM HEPES, 1 mM DTT, 1 mM
MgCl2, 0.5 mM EDTA) and 10% v/v DMSO was reacted with
test compound for 2.5 h at room temperature and then
precipitated twice with ethanol and vacuum-dried. The syn-
thetic oligonucleotide primer PUC1 (5′-CTC ACT CAA AGG
CGG TAA TAC-3′) was 5′-end labeled with [γ-32P] ATP using
T4 polynucleotide kinase and purified using a sephadex
column. Linear PCR (94 °C 3′, 30 cycles of {94 °C 1′, 60 °C
1′,72 °C 1′}, 72 °C 4′, 4 °C 5′) of the damaged template (0.5
µg) with labeled PUC1 primer (0.5 ng) was performed in a
thermal cycler using Taq DNA polymerase (10 units) in a final
volume of 100 µL of reaction buffer (200 µM each dNTP, 50
mM KCl, 10 mM Tris‚HCl pH 9.0, 1.5 mM MgCl2, 1% Triton
X-100). The PCR product was precipitated twice with ethanol
also demonstrated that reactive small molecule drugs
conjugated to 2 can retain their inherent cytotoxic
activity. Indeed, for the nitrogen mustard melphalan the
conjugation of 2, a known DNA intercalator, improved
the level of DNA damage and in vitro cytotoxicity.
However, for chlorambucil the inverse was observed,
supporting suggestions that the naphthoate 2 is a poor
intercalator in the absence of a positively charged side
chain.8,15
These data represent the first evidence that binding
to apoNCS may confer the similar benefits for small-
molecule conjugates as reported for binding of the
natural chromophore 1. However, as binding is only
moderate (micromolar) in comparison to the natural
ligand (nanomolar), an improvement in Kd for the drug-
naphthoate complexes can be expected to produce an
improvement in the beneficial effects obtained by such
binding. Protein engineering offers the potential to
modulate the affinity of binding of a given small
molecule to the protein by altering residues within the
apoNCS binding site. Indeed, Heyd et al. recently
selected proteins from a phage display library of apoNCS
binding site variants that showed improved binding to
an unnatural ligand (testosterone).23
In the natural holo-complex, apoNCS is believed to
provide a transport function for the natural chro-
mophore; both apo- and holo-NCS enter mammalian
cells and penetrate at least as far as the nuclear
membrane.6 Furthermore, conjugation of poly(styrene-
co-maleic acid anhydride) to apoNCS produces the
pharmaceutical SMANCS, which displays improved
tumor uptake and toxicological profile.24 Thus, other
small molecules bound to apoNCS, such as 6, may also
benefit from improved targeting in vivo.
Exp er im en ta l Section
Gen er a l. Glassware was oven dried at 160 °C. Solvents
quoted as dry were freshly distilled; all other chemicals used
were obtained from commercial sources and used without
further purification. Identical samples of 2-hydroxy-5-methyl-
7-methoxy-naphthoic acid 2 were obtained following the
procedures of Takahashi et al.25,26 or Rucker et al.27 N-Boc-
melphalan was synthesized as described previously.12 Neo-
carzinostatin apo-protein (apoNCS) was expressed from the
pCANTAB-(NCS) vector in E. coli (HB2151), with uniform
15N-labeling achieved using M9 minimal media containing
15NH4Cl as the sole nitrogen source.10
Nuclear magnetic resonance NMR spectra for small mol-
ecules were recorded on a Bruker DMX 300 MHz spectrometer
at ambient temperature and are reported as chemical shifts
in parts per million referenced against the undeuterated
solvent peak. 13C NMR and DEPT were run on the same
machine at 75 MHz and referenced against the solvent peak.
IR samples were recorded on a Perkin-Elmer FT-IR 298
spectrophotometer as thin films on NaCl plates. Accurate mass
values are within 5 ppm of the calculated value.
Reversed phase high performance liquid chromatography
(RP-HPLC) was conducted on a J asco system composed of two
PU-980 pumps, a UV-970M multiwavelength detector, and a
Hypersil HyPurity Elite C18 column (150 × 4.6 mm, 3 µm,
ThermoQuest). A solvent gradient was produced by varying
the proportion of two buffers; buffer A (5.0% v/v aqueous
MeCN, 0.1% v/v CF3CO2H) and buffer B (95% v/v MeCN, 5.0%
v/v H2O, 0.085% v/v CF3CO2H). Separations were performed
using a flow rate of 1.0 mL/min with a solvent gradient of 20%
B for 2.5 min, 20-80% B linear gradient in 30 min, 80-100%
B linear gradient in 2.5 min and 100-20% B linear gradient
in 2.5 min. The eluant was monitored simultaneously at
wavelengths of 210, 280, and 330 nm.

4714 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 19
Urbaniak et al.
and vacuum-dried, and fragments were separated on a 6%
polyacrylamide gel (7 M urea, Tris-boric acid-EDTA buffer)
for approximately 3 h at 3000 V, 55 °C. Gels were transferred
to filter paper and dried and autoradiographs obtained using
film.
Cytotoxicity Stu d ies. Human leukemia cell line K562 was
grown in RPMI 1640 supplemented with 10% heat-inactivated
foetal bovine serum and 2 mM L-glutamine without antibiotic
at 37 °C, 5% CO2, 95% air, 100% relative humidity. 200 µL of
K562 cells (5 × 104 cells/mL) were aliquoted into a 96-well
microtiter plate and incubated for 1 h with the test compound
(100 µM to 30 nM). After being washed with culture medium,
the cells were incubated for a further 4 days before the number
of viable cells was counted using the MTT assay.16
F lu or escen ce Sp ectr oscop y. Fluorescence spectra were
recorded on a Varian Cary Eclipse Fluorescence spectropho-
tometer with a Peltier temperature control device set to
maintain the sample temperature at 25 ( 0.5 °C. Fluorescence
quenching titration experiments were conducted using the
constant dilution method described by Kondo et al.,28 using
5 µM of naphthoate conjugates 4 or 6 with 0-150 µM of
apoNCS or 10 µM apoNCS with 0-300 µM of melphalan or
chlorambucil in 100 mM sodium acetate buffer pH 5.0 with
10% v/v MeOH. Dissociation constants (Kd) were calculated
as the average value of three separate titration experiments
analyzed using both a linear (eq 1) and a nonlinear method
(eq 2).28,29
Su p p or t in g In for m a t ion Ava ila b le: Spectral data for
compounds 2-5, 5′, 6. This material is available free of charge
via the Internet at http://pubs.acs.org.
Refer en ces
(1) Rajski, S. R.; Williams, R. M. DNA Cross-linking agents as
antitumor drugs. Chem. Rev. 1998, 98, 2723-2795.
(2) Ishida, N.; Miyuzaki, K.; Kumagai, K.; Rikimaru, M. Neocarzi-
nostatin, an antitumor antibioticum of high molecular weight.
J . Antibiot. 1965, 18, 68-76.
(3) Napier, M. A.; Holmquist, B.; Strydom, D. J .; Goldberg, I. H.
Neocarzinostatin: spectral characterisation and separation of
a nonprotein chromophore. Biochem. Biophys. Res. Com. 1979,
89, 635-639.
(4) Smith, A. L.; Nicolaou, K. C. The Enediyne Antibiotics. J . Med.
Chem. 1996, 39, 2103-2117.
(5) Maeda, H.; Edo, K.; Ishida, N., Eds. Neocarzinostatin: The Past,
Present and Future of an Anticancer Drug; Springer-Verlag:
Tokoyo, 1997; p 287.
(6) Maeda, H.; Aikawa, S.; Yamashita, A. Subcellular Fate of Protein
Antibiotic NCS in culture of a Lymphoid Cell line from Burkitt’s
Lymphoma. Cancer Res. 1975, 35, 554-559.
(7) Myers, A. G. Proposed structure of the neocarzinostatin chro-
mophore - methyl thioglycolate adduct; a mechanism for the
nucleophilic activation of neocarzinostatin. Tetrahedron Lett.
1987, 28, 4493-4496.
(8) Povirk, L. F.; Dattagupta, N.; Warf, B. C.; Goldberg, I. H.
Neocarzinostatin chromophore binds to deoxyribose nucleic acid
by intercalation. Biochemistry 1981, 20, 4007-4014.
(9) Gao, X.; Stassinopoulos, A.; Rice, J . S.; Goldberg, I. H. Structural
basis for the sequence-specific DNA strand cleavage by the
enediyne neocarzinostatin chromophore. Structure of the post-
activated chromophore-DNA complex. Biochemistry 1995, 34,
40-49.
Kd
1
∆F
1
1
)
)
+
(1)
(2)
R[ST] [LT] R[ST]
(10) Urbaniak, M. D.; Muskett, F. W.; Finucane, M. D.; Caddick, S.;
Woolfson, D. N. Solution structure of a novel chromoprotein
derived from apo-neocarzinostatin and a synthetic chromophore.
Biochemistry 2002, 41, 11731-11739.
[LT] - [ST]‚∆F
/
∆FMax
∆F ) ∆Fmax.
Kd + [LT] - [ST]‚∆F
/
∆FMax
(11) Valu, K. K.; Gourdie, T. A.; Boritzki, T. J .; Gravatt, G. L.;
Baguley, B. C.; Wilson, W. R.; Wakelin, L. P. G.; Woodgate, P.
D.; Denny, W. A. DNA directed alkylating agents. 3. Structure-
activity relationships for the acridine-linked analine mustards:
consequences of varying the length of the linker chain. J . Med.
Chem. 1990, 33, 3014-3019.
(12) Morris, A. D.; Atassi, G.; Guilbaud, N.; Cordi, A. A. The synthesis
of novel melphalan derivatives as potential antineoplastic
agents. Eur. J . Med. Chem. 1997, 32, 343-349.
(13) Ponti, M.; Forrow, S. M.; Souhami, R. L.; D′Incalci, M.; Hartley,
J . A. Measurement of the sequence specificity of covalent DNA
modification by antineoplastic agents using Taq DNA polym-
erase. Nucleic Acids Res. 1991, 19, 2929-2933.
(14) Prakash, A. S.; Denny, W. A.; Gourdie, T. A.; Valu, K. K.;
Woodgate, P. D.; Wakelin, L. P. G. DNA-directed alkylating
ligands as potential antitumor agents: sequence specificity of
alkylation by intercalating analine mustards. Biochemistry 1990,
29, 9799-9807.
(15) Sartorius, J .; Schneider, H.-J . Intercalation mechanisms with
double stranded DNA: binding modes and energy contributions
with benzene, naphthalene, quinoline and indole derivatives
including some antimalarials J . Chem. Soc., Perkins Trans. 2
1997, 2319-2327.
(16) Alley, M. C.; Scudiero, D. A.; Monks, A.; Hursey, M. L.;
Czerwinski, M. J .; Fine, D. L.; Abbott, B. J .; Mayo, J . G.;
Shoemaker, R. H.; Boyd, M. R. Feasibility of drug screening with
panels of human tumor cell lines using a microculture tetrazo-
lium assay. Cancer Res. 1988, 48, 589-601.
(17) Kim, K. H.; Kwon, B. M.; Myers, A. G.; Rees, D. C. Crystal
structure of neocarzinostatin, an antitumor protein - chro-
mophore complex. Science 1993, 262, 1024-1046.
(18) Falmer, B. T.; Constantine, K. L.; Goldfarb, V.; Friedrichs, M.
S.; Wittekind, M.; Yanchanus, J .; Robertson, J . G.; Mueller, L.
Localising the NADP+ binding site on the MurB enzyme by
NMR. Nat. Struct. Biol. 1996, 3, 995-997.
where [ST] and [LT] are the total concentration of substrate
and ligand respectively, R is a constant, ∆F is the change in
fluorescence, and ∆Fmax. is the change in fluorescence at
saturation.
P r otein Nu clea r Ma gn etic Reson a n ce Sp ectr oscop y.
Protein nuclear magnetic resonance (NMR) spectroscopy was
performed as described previously.10 The perturbation of
backbone amide chemical shifts was calculated relative to
apoNCS recorded in 10% v/v MeOH-d4 using the minimum
chemical shift procedure, eq 3.18,19
min∆δ ) min{(HN
∆
ppm)2 + (N∆ppm*RN)2}1/2
(3)
N
15
where HN
∆
ppm
and
∆
ppm
are the 1HHN and
N chemical shift
NH
changes and RN is a scaling factor to account for the difference
in spectral width of backbone 15N relative to 1H. A value of
RN ) 1/7 was used for all residues except glycine (RN ) 1/5).
Hyd r olysis Exp er im en ts. Hydrolysis of the nitrogen
mustard derivative 6 was conducted at 35 °C using 500 µM of
6 in 50 mM phosphate pH 5.0 with 10% v/v THF in the
presence or absence of 500 µM apoNCS. Aliquots were with-
drawn at appropriate times and analyzed immediately by RP-
HPLC. The percentage unhydrolyzed 6 was calculated as the
ratio of the unhydrolyzed peak to all species present, normal-
ized to give 100% at the start of the experiment.
Ack n ow led gm en t. We are grateful to Cancer Re-
search UK, EPSRC, BBSRC, AICR, AstraZeneca, Glaxo-
SmithKline, Novartis, and the Centre for Biomolecular
Design and Drug Development at the University of
Sussex for supporting this work. We thank the Medical
Research Council for NMR spectrometer time at the
National Institute for Medical Research, Mill Hill. We
also thank the EPSRC mass Spectrometry Service at
Swansea and Dr Ali Abdul-Sada for provision of mass
spectra.
(19) Williamson, R. A.; Carr, M. D.; Frenkiel, T. A.; Feeney, J .;
Freedman, R. B. Mapping the binding site for matrix metallo-
proteinase on the N-terminal domain of the tissue inhibitor of
metalloproteinase-2 by NMR chemical shift pertubation. Bio-
chemistry 1997, 36, 13882-13889.
(20) Imajo, S.; Ishiguro, M.; Tanaka, T.; Hirama, M.; Teplyakov, A.
On the conformation of Phe 78 of a chromoprotein antibiotic
neocarzinostatin. Bioorganic & Med. Chem. 1995, 3, 429-436.
(21) Kappen, L. S.; Goldberg, I. H. Stabilisation of neocarzinostatin
nonprotein chromophore activity by interaction with apoprotein
and with HeLa cells. Biochemistry 1980, 19, 4786-4790.

Nitrogen Mustard Stabilization by Apo-neocarzinostatin
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 19 4715
(22) Kappen, L. S.; Napier, M. A.; Goldberg, I. H. Role of chromophore
and apo-protein in neocarzinostatin action. Proc. Natl. Acad. Sci.
U.S.A. 1980, 77, 1970-1974.
(23) Heyd, B.; Pecorari, F.; Collinet, B.; Adjadj, E.; Desmadril, M.;
Minard, P. In vitro evolution of the binding specificity of
neocarzinostatin, an enediyne-binding chromoprotein. Biochem-
istry 2003, 42, 5674-5683.
(24) Maeda, H.; Ueda, M.; Morinaga, T.; Matsumoto, T. Conjugation
of poly(styrene-co-maleic acid) derivatives to the antitumour
protein NCS: pronounced improvements in pharmacological
properties. J . Med. Chem. 1985, 28, 455-461.
(25) Takahashi, K.; Suzuki, T.; Hirama, M. A simple and efficient
synthesis of the naphthoate moiety of neocarzinostatin chro-
mophore. Tetrahedron Lett. 1992, 33, 4603-4604.
(26) Takahashi, K.; Tanaka, T.; Suzuki, T.; Hirama, M. Synthesis
and Bbinding of simple NCS chromophore analogues to the
apoprotein. Tetrahedron 1994, 50, 1327-1340.
(27) Rucker, M.; Bruckner, R. Practical synthesis of the trisubstituted
naphthalene carboxylic acid from neocarzinostatin chromophore.
Synlett 1997, 1187-1189.
(28) Kondo, H.; Nakatani, M.; Hirami, K. Interactions of cyclodextrins
with fluorescence probes and its application to kinetic studies
of amylone. J . Biochem. 1976, 79, 393-405.
(29) Heyd, B.; Lerat, G.; Adjadj, E.; Minard, P.; Desmadril, M.
Reinvestigation of the protolytic acitivity of apoNCS. J . Bacteriol.
2000, 1812-1818.
J M040790D