Modulation of melphalan resistance in glioma cells with a peripheral benzodiazepine receptor ligand-melphalan conjugate

Article abstract of DOI:10.1021/jm960592p

Peripheral benzodiazepine receptors (PBRs) are located on the outer membrane of mitochondria, and their density is increased in brain tumors. Thus, they may serve as a unique intracellular and selective target for antineoplastic agents. A PBR ligand-melph

Full text of DOI:10.1021/jm960592p

1726
J . Med. Chem. 1997, 40, 1726-1730
Mod u la tion of Melp h a la n Resista n ce in Gliom a Cells w ith a P er ip h er a l
Ben zod ia zep in e Recep tor Liga n d -Melp h a la n Con ju ga te
Lidia Kupczyk-Subotkowska,† Teruna J . Siahaan,† Anthony S. Basile,‡ Henry S. Friedman,§ Patricia E. Higgins,∇
Di Song,∇ and J ames M. Gallo*,∇
Department of Pharmaceutical Chemistry, Simons Laboratories, The University of Kansas, Lawrence, Kansas 66047,
Laboratory of Neuroscience, NIDDK, National Institutes of Health, Bethesda, Maryland, Department of Pediatrics,
Duke University Medical Center, Durham, North Carolina 27710, and Department of Medical Oncology,
Fox Chase Cancer Center, 7701 Burholme Avenue, Philadelphia, Pennsylvania 19111
Received August 16, 1996X
Peripheral benzodiazepine receptors (PBRs) are located on the outer membrane of mitochondria,
and their density is increased in brain tumors. Thus, they may serve as a unique intracellular
and selective target for antineoplastic agents. A PBR ligand-melphalan conjugate (PBR-
MEL) was synthesized and evaluated for cytotoxicity and affinity for PBRs. PBR-MEL (9)
(i.e., 670 amu) was synthesized by coupling of two key intermediates: 4-[bis(2-chloroethyl)-
amino]-L-phenylalanine ethyl ester trifluoroacetate (6) and 1-(3′-carboxylpropyl)-7-chloro-1,3-
dihydro-5-phenyl-2H-1,4-benzodiazepin-2-one (8). On the basis of receptor-binding displacement
assays in rat brain and glioma cells, 9 had appreciable binding affinity and displaced a
prototypical PBR ligand, Ro 5-4864, with IC50 values between 289 and 390 nM. 9 displayed
differential cytotoxicity to a variety of rat and human brain tumor cell lines. In some of the
cell lines tested including rat and human melphalan-resistant cell lines, 9 demonstrated
appreciable cytotoxicity with IC50 values in the micromolar range, lower than that of melphalan
alone. The enhanced activity of 9 may reflect increased membrane permeability, increased
intracellular retention, or modulation of melphalan’s mechanisms of resistance. The combined
data support additional studies to determine how 9 may modulate melphalan resistance, its
mechanisms of action, and if target selectivity can be achieved in in vivo glioma models.
Sch em e 1a
Chemotherapy of malignant brain tumors remains an
enormous therapeutic challenge. Treatment regimens
are plagued by the inability to selectively deliver drugs
to tumor cells and the emergence of drug resistance. The
most recent advances in drug delivery to brain tumors
have utilized regional implants of drug-loaded polymers
that bypass the blood-brain barrier (BBB)1,2 and gene
therapy approaches.3 Moreover, in attempts to circum-
vent mechanisms of drug resistance, modulators of
alkylguanine transferase (i.e., O6-benzylguanine)4 or
glutathione (i.e., buthionine sulfoximine)5 have been
applied. Presently, these approaches are experimental,
and their ultimate place in the therapeutic arsenal will
require further analyses.
a
Reagents: (a) HClg, EtOH; (b) di-tert-butyl dicarbonate, TEA;
One way to enhance the selective delivery of antican-
cer drugs to tumors is to target receptors that are
overexpressed in tumors. The peripheral benzodiaz-
epine receptor (PBR) is such a novel therapeutic target,
as PBR expression is selectively increased in both
experimental and human brain tumors compared to
normal brain and peripheral tissues.6-8 In fact, PBR
expression by brain tumors correlates with the degree
of malignancy.7 The PBR is predominately expressed
on the outer surface of mitochondria,9-11 distinguishing
it as an intracellular target, rather than the more
common cell surface membrane targets. The identifica-
tion of the PBRs in brain tumors has also served as a
means to evaluate PBR ligands as diagnostic imaging
agents.7,8 Thus, the characteristics of PBRs in tumors
(c) 10% Pd/C, H2, MeOH; (d) ethylene oxide, CH3COOH/H2O; (e)
CCl4, Ph3P/CH2Cl2; (f) TFA/anisole.
suggest that PBR ligands can serve as receptor-medi-
ated drug carriers to selectively target anticancer drugs
to brain tumors.
The goals of the current study were to synthesize a
PBR ligand-melphalan conjugate able to bind to PBRs
in vitro and to further evaluate its cytotoxicity in
melphalan-sensitive and melphalan-resistant glioma
cell lines.
Resu lts a n d Discu ssion
Synthesis of PBR-MEL (9) was accomplished by
coupling 6 and 8 (Scheme 2). The synthesis of inter-
mediate 6 was started by esterification of 4-nitrophen-
ylalanine (1) with a mixture of ethanol and hydrogen
chloride gas (Scheme 1). Next, the amino group was
protected with N-tert-butyloxycarbonyl (t-Boc) group to
* To whom correspondence should be sent.
† The University of Kansas.
‡ NIDDK.
§ Duke University Medical Center.
∇ Fox Chase Cancer Center.
X Abstract published in Advance ACS Abstracts, May 1, 1997.
S0022-2623(96)00592-4 CCC: $14.00 © 1997 American Chemical Society

Modulation of Melphalan Resistance with PBR-MEL
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11 1727
Sch em e 2a
Ta ble 2. Cytotoxicity of Melphalan (MEL),
Benzodiazepine-Melphalan Ethyl Ester (PBR-MEL, 9), and
Peripheral Benzodiazepine Ligands in Glioma Cell Lines
IC50 (µM) a
cell line
(species source)
MEL PBR-MEL Ro 5-4864 nordiazepam
C6 (rat)
C6-MR (rat)
RG-2 (rat)
SF-126 (human)
SF-188 (human)
D341 (human)
D341 MR (human)
D283 (human)
D283 MR (human)
DAOY (human)
11.0
74.9
15.2
103.0
35.4
12.4
52.4
6.8
43.4
74.4
34.3
21.0
17.3
25.0
32.8
72.3
25.0
40.0
29.3
>200
>200
>100
>100
137.5
NDb
ND
ND
ND
ND
ND
>200
>100
>100
>200
>200
>200
>200
>200
>200
16.3
55.4
DAOY MR (human) 140.4
a
Data are expressed as mean of two separate experiments
except for SF-126, SF-188, DAOY, and DAOY MR in which three
b
separate studies were completed. ND ) not determined.
at the N-4 position were inserted.15 Using the model
PBR ligand Ro 5-4864, the binding affinity (i.e., Kd) and
receptor density (i.e., Bmax) in the rat glioma cells
compared favorably to values previously reported with
either Ro 5-4864 or PK-11195 in C6 and RG-2 tumors
in vivo.6-8 The in vitro binding affinities were about
1.5-2-fold greater, whereas the PBR densities were
about 20-fold greater than those reported in vivo. The
higher Bmax values obtained in vitro may in part be due
to the greater specificity of the assay since contamina-
tion from normal adjacent brain is not possible. The
higher density of PBRs in brain tumors compared to
normal brain is the basis of using PBR ligand-drug
conjugates to selectively target brain tumors.
a
Reagents: (a) NaH, DMF; (b) KOH, MeOH; (c) EDC, HOBt.
Ta ble 1. Receptor-Binding Characteristics of
Benzodiazepine-Melphalan Ethyl Ester (PBR-MEL, 9),
Nordiazepam, and Ro 5-4864 in Rat Brain and Glioma Cells
The prototypical isoquinoline ligand PK-11195 has
appreciable binding to both rat and human brain
PBRs16,17 and brain tumors.7,8 The Ro 5-4864 PBR
ligand and 9, both based on a benzodiazepine structure,
specifically bind only to rat PBRs. Thus, a pertinent
direction to follow for targeting human PBRs will be the
development of isoquinoline-based drug conjugates.
Cytotoxicity assays of MEL, 9, and the PBR ligands
nordiazepam and Ro 5-4864 were conducted in various
rat and human brain tumor cell lines using an SRB
assay18 for adherent cells or the MTS assay19 for
suspended cells (i.e., D341, D341 MR, D283, D283 MR)
(see Table 2). The SRB assay has been adopted for
routine use by the NCI.20 It can be seen that 9 has
greater activity in five cell lines with the greatest
difference being a 5-fold enhancement in cytotoxicity in
the human glioma SF-126 cell line. MEL was more
active in the other five cell lines. The two benzodiaz-
epines, nordiazepam and Ro 5-4864, did not demon-
strate appreciable cytotoxicity, and in only one cell line,
SF-188, was an IC50 value obtained.
Cytotoxicity assays (i.e., SRB assay) were also con-
ducted in RG-2 cells with MEL and 9 in the presence of
two PBR ligands, nordiazepam (a weak ligand) and Ro
5-4864 (a strong ligand), at concentrations from 70 nM
to 70 µM, to determine if cytotoxicity was attenuated
by competitive inhibition of PBR binding. For both
MEL and 9, there was no change in the IC50 values
obtained in the absence and presence of the PBR
ligands. In the presence of the PBR ligands, MEL’s IC50
values ranged from 12 to 15 µM (compared to 15.2 µM
without binding inhibitors), and for 9 the values ranged
from 32 to 38 µM (compared to 34.3 µM without binding
inhibitors).
IC50
Bmax (pmol/mg
of protein)
source
(nM)a
Kd (nM)b
1.3 ( 0.1
normal brain homogenate
327
520c
289
390
0.29 ( 11
C6
RG-2
4.0 ( 0.25
3.5 ( 0.64
22 ( 1.5
33 ( 3.1
a
Values indicate binding affinity of 9 unless otherwise indi-
cated. Kd and Bmax are binding constants for Ro 5-4864. c Value
b
for nordiazepam (desmethyldiazepam).
give compound 2. Reduction of the nitro group in
compound 2 was accomplished using 10% Pd/C with
ammonium formate to yield compound 3.12 The aro-
matic amino group was reacted with ethylene oxide in
acetic acid/water to give compound 4. Treatment of
compound 4 with triphenylphosphine, carbon tetrachlo-
ride, and dichloromethane gave compound 5.13 The
t-Boc protecting group in 5 was removed by trifluoro-
acetic acid to give intermediate 6. Intermediate 8 was
synthesized by treatment of 7-chloro-1,3-dihydro-5-
phenyl-2H-1,4-benzodiazepin-2-one14 with methyl 4-bro-
mobutyrate to generate compound 7 (Scheme 2). Basic
hydrolysis of compound 7 gave the desired intermediate
8. Intermediates 6 and 8 were coupled using EDC and
HOBt to yield the target molecule 9.
The receptor-binding assays (see Table 1) indicate
that 9 binds to PBRs in rat brain and rat glioma cells
(i.e., C6 and RG-2) at nanomolar concentrations that
should be achievable in vivo. The binding affinity of 9
in normal brain (i.e., IC50 ) 327 nM) was greater than
that obtained with nordiazepam (i.e., IC50 ) 520 in
normal brain), the benzodiazepine from which 9 is
derived. This greater affinity exhibited by 9 is consis-
tent with other reports where large alkyl substitutions

1728 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11
Kupczyk-Subotkowska et al.
Ta ble 3. Glutathione S-Transferase (GST) Activity and
Glutathione (GSH) Concentrations in Human Medulloblastoma
Cellsa
preliminary findings and characterize how 9 and other
analogues may modulate MEL’s mechanisms of resis-
tance.
cell line
D341
D341 MR
D283
D283 MR
DAOY
DAOY MR
GST (nmol/min/mg of protein)
GSH (µmol/g)
The 9 moiety is a unique drug conjugate, not only due
to its intracellular target but also because of its low
molecular weight (i.e., MW ) 670 DA). 9 is formed
through an amide linkage between a benzodiazepine
moiety and melphalan ethyl ester that is not susceptible
to acid or base hydrolysis in biological fluids. Other
amide linkages of melphalan conjugates have been
shown to be stable intracellularly,24 and thus, cleavage
of the amide bond of 9 will likely require intracellular
endopeptidases. This potentially enhanced stability
may have consequences for its pharmacological action,
since generation of free MEL may not be required to
initiate cytotoxic events.
The lipophilicity of 9, imparted by the benzodiazepine
moiety, may enable the conjugate to pass through cell
membranes by passive diffusion rather than by facili-
tated transport via the large neutral amino acid carrier
as for MEL.25 This could result in higher intracellular
concentrations of 9 compared to those obtained by MEL
alone. In addition, the presence of the ethyl ester group,
as opposed to the carboxylic acid group of MEL, could
also facilitate membrane uptake and cellular retention
following conversion to the free acid.
212.6 ( 21.7
327.9 ( 43.6
150.6 ( 35.8
417.3 ( 65.6
88.7 ( 7.1
17.5 ( 3.6
33.2 ( 8.5
18.4 ( 4.5
50.7 ( 16.1
11.4 ( 2.4
10.8 ( 1.5
84.0 ( 10.9
a
Data are expressed as mean of three to four experiments.
Since there is a species dependence in PBR ligand
binding,16,17 enhanced cytotoxicity of 9 may have been
anticipated in a rat glioma cell line, such as RG-2, in
which Ro 5-4867 and analogues have greater binding
affinity than in human glioma cell lines. However,
there was no direct relationship between in vitro cyto-
toxicity and binding to PBRs. Although this disjunction
may cause concern, it may, in fact, be an advantage to
separate receptor binding and cytotoxicity processes.
The ability of 9 to bind to PBR located on the outer
membrane of mitochondria is a mechanism that may
control the selective distribution of 9 to brain tumors
in vivo, as demonstrated with PBR ligands.6-8 Follow-
ing initial distribution to PBRs, 9 may exert its cytotoxic
action via multiple pathways upon dissociation from
PBRs. Either MEL, generated from the conjugate, or
the conjugate itself can interact with DNA forming
lethal adducts. If PBR binding and cytotoxicity were
correlated, it would suggest a single pathway of cyto-
toxicity via interaction of 9 with mitochondrial DNA,
contingent upon internalization of 9 into mitochondria.
Thus, the strategy of PBR-drug conjugates is to attain
target (brain tumor) selectivity via binding to PBRs
followed by subsequent cytotoxic actions, most likely
directed at genomic DNA. In vivo tissue distribution
studies will be needed to demonstrate the feasibility of
this strategy.
In summary, the novel design features of 9 may lead
to enhanced membrane transport, intracellular stability,
and retention. In addition, the interaction of 9 with
PBRs may serve as a selective target for in vivo delivery.
Development of this class of agents should be pursued.
Exp er im en ta l Section
Equ ip m en t. All reagents are commercially available.
Melting points were determined on a Fisher-J ohns melting
point apparatus and are uncorrected. NMR spectra were
determined in CDCl3 on a QE (300 MHz) spectrometer. High-
resolution mass spectra were obtained using a VG Analytical
ZAB double-focusing spectrometer. A Polytron (Brinkmann
Instruments, Westbury, NY) tissue homogenizer was used.
Glass fiber filters (#32) were obtained from Schleicher &
Schuell (Keene, NH). A Brandel (Brandel Instruments, Gaith-
ersburg, MD) M-24R filtering manifold was used for receptor-
It was of interest to contrast cytotoxicity of MEL and
9 in sensitive and MEL-resistant cell lines. Mecha-
nisms of resistance to MEL can vary depending on the
cell type and normally consist of membrane transport
defects, inactivation via glutathione, and enhanced
repair of DNA adducts.21-23 In two of the human-
derived MEL-resistant clones, D283 MR and D341 MR,
glutathione S-transferase (GST) activity, an enzyme
that catalyzes the formation of glutathione conjugates,22
and glutathione (GSH) concentrations were elevated
(see Table 3). MEL’s activity is reduced about 4-fold in
these lines compared to the parental lines, whereas 9’s
activity is either unchanged or enhanced. In the DAOY
and DAOY MR lines, MEL is approximately 3-fold less
toxic to the MR line, whereas 9 is equally toxic, with
IC50s less than 40 µM. Since 9 exhibits analogous
cytotoxicity to the MR lines as the parental lines, unlike
MEL, it is not susceptible to the mechanism of MEL
resistance in these cell lines. In contrasting the cyto-
toxicity of MEL and 9 in the rat C6 and C6 MR cell lines,
it is again shown that 9 is not nearly as influenced by
the mechanisms that control resistance to MEL. On the
basis of the combined data in both rat and human MR
cell lines, 9 may modulate the formation of inactive
drug-GSH conjugates and possibly bypass membrane
transport defects. It will be important to extend these
binding assays. Radioactivity was measured in
a liquid
scintillation spectrometer (model LS 5801, Beckman Instru-
ments, Fullerton, CA).
Syn t h esis.
N-(ter t-Bu t yloxyca r b on yl)-4-n it r o-L-
p h en yla la n in e Eth yl Ester (2). 4-Nitro-L-phenylalanine (1)
was prepared as described in the literature26 and converted
into its ethyl ester hydrochloride.27 To a suspension of 4-nitro-
L-phenylalanine ethyl ester hydrochloride (5.49 g, 0.02 mol)
in anhydrous dichloromethane (60 mL) were added triethyl-
amine (3.5 mL) followed by di-tert-butyl dicarbonate (4.36 g,
0.02 mol), and the reaction mixture was stirred at room
temperature for 4 h. TLC (silica gel, chloroform:petroleum
ether, 7:3) indicated complete reaction. Dichloromethane (200
mL) was added, and the reaction mixture was washed suc-
cessively with a 1 N solution of HCl, water, saturated NaHCO3
solution, water, and brine and then dried (MgSO4). Solvent
was evaporated, and product was crystallized from a diethyl
ether:hexane mixture to give 5.40 g (0.016 mol, 80%) of 2,
homogeneous by TLC (chloroform:petroleum ether, 7:3): mp
64-65 °C; 1H NMR (CDCl3, δ) 8.15 (d, J ) 8.1 Hz, 2H), 7.3 (d,
J ) 8.4 Hz, 2H), 5.0 (bd, 1H), 4.5-4.4 (m, 1H), 4.2 (q, J ) 7.0
Hz, 2H), 3.3-3.1 (m, 2H), 1.4 (s, 9H), 1.25 (t, 3H); FAB HRMS
[M + H] calcd for (C16H22N2O6) 339.1556, found 339.1574.
N-(ter t-Bu t yloxyca r b on yl)-4-a m in o-L-p h en yla la n in e
Eth yl Ester (3). To a solution of 2 (5.40 g, 0.016 mol) in
anhydrous methanol (100 mL) were added 10% Pd/C catalyst
(100 mg) followed by anhydrous ammonium formate (5.1 g,

Modulation of Melphalan Resistance with PBR-MEL
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11 1729
0.08 mol). The reaction mixture was stirred at room temper-
ature for 1 h and TLC (silica gel, diethyl ether) indicated
complete reduction. The reaction mixture was filtered through
a Celite 545 pad and washed with methanol, and combined
methanol solutions were evaporated. The solid residue was
triturated with water and the product extracted into diethyl
ether. The ether solution was dried (MgSO4) and evaporated
to give an oily product, homogeneous by TLC (diethyl ether)
and resulting in 4.89 g (0.016 mol, 100%) of 3: 1H NMR
(CDCl3, δ) 6.9 (d, J ) 8.1 Hz, 2H), 6.6 (d, J ) 8.4 Hz, 2H), 4.95
(bd, 1H), 4.5-4.4 (m, 1H), 4.15 (q, J ) 7.0 Hz, 2H), 2.95 (d, J
) 7.0 Hz, 2H), 1.4 (s, 9H), 1.2 (t, 3H); FAB HRMS [M+] calcd
for C16H24N2O4 308.1736, found 308.1740.
for 2 h. TLC (silica gel, ethyl acetate) indicated complete
reaction. Methanol was evaporated, water was added to the
residue, and the solution was acidified with 1 N HCl. Product
was extracted into ethyl acetate and dried over anhydrous
sodium sulfate. The solution was dried (Na2SO4) and solvent
evaporated. Crystallization from an ethyl ether:petroleum
ether mixture gave 0.91 g (0.0025 mol, 85%) of 8, homogeneous
1
by TLC (ethyl acetate): mp 160-163 °C; H NMR (CDCl3, δ)
7.6-7.2 (m, 8H), 4.8 (d, 1H), 4.35 (q, 1H), 3.75 (d, 1H), 3.75-
3.6 (m, 1H), 2.3-1.6 (m, 4H); FAB HRMS [M + H] calcd for
C19H18N2O3Cl 357.1006, found 357.1002.
1-[4-Bu t a n oyl-[p -b is(2-ch lor oet h yl)a m in o]-L-p h en yl-
a la n in e eth yl ester ]-7-ch lor o-1,3-d ih yd r o-5-p h en yl-2H-
1,4-ben zod ia zep in -2-on e (9). To a stirred and 0 °C cooled
solution of 8 (0.36 g, 0.001 mol) in dimethylformamide (20 mL)
were added 1-[3-(dimethylamino)propyl]-3-ethylcarbodiimide
hydrochloride (EDC) (0.20 g) and hydroxybenzotriazole (0.14
g). The reaction mixture was stirred for 30 min; then
compound 6 (0.45 g, 0.001 mol) followed by triethylamine (0.2
mL) were added. After overnight stirring at room tempera-
ture, the solvent was evaporated and the residue treated with
an ethyl acetate:water mixture. The organic layer was sepa-
rated, washed successively with citric acid solution, water,
saturated sodium bicarbonate solution, water, sodium chloride
solution, water, saturated NaHCO3 solution, water, and brine,
and then dried (Na2SO4). Solvent was evaporated and product
purified by column chromatography on silica gel (60-200
mesh) with ethyl acetate as eluent. Crystallization from an
ethyl acetate:petroleum ether mixture gave 0.21 g (0.00031
mol, 31%) of 9, homogeneous by TLC (ethyl acetate): mp 70-
72 °C; 1H NMR (CDCl3, δ) 7.65-7.35 (m, 8H), 6.9 (dd, 2H),
6.55 (dd, 2H), 6.05 (bd, 1H), 5.8 (bd, 1H), 4.85-4.6 (m, 2H),
4.5-4.1 (m, 3H), 3.8-3.55 (m, 9H), 3.05-2.85 (m, 2H), 2.1-
1.6 (m, 4H), 1.3 (t, 3H); FAB HRMS [M + H] calcd for
C34H38N4O4Cl3 671.1959, found 671.1959.
Recep tor -Bin d in g Assa y. Competitive binding assays
using Ro 5-4864, [3H]Ro 5-4864, and 9 were performed with
rat glioma cells and rat brain homogenates as previously
described.28 For whole brain homogenates, rats were sacrificed
(80% CO2, 20% O2 prior to decapitation) and the brains
removed and placed into ice-cold 0.32 M sucrose. The intact
brain (1.5-2 g) was homogenized in 50 vol of 50 mM Tris-HCl
(pH 7.4, 0-4 °C) and then centrifuged once at 20 000g for 20
min. The pellet was retained and resuspended in 100 vol of
50 mM Tris-HCl buffer. For RG-2 and C6 glioma cells, 6-8
× 106 cells were suspended in 25 mL of 50 mM Tris-citrate
buffer (pH 7.4) and homogenized for 10 s. The homogenate
was centrifuged at 20 000g for 20 min, and the resultant pellet
was resuspended in 16 mL of Tris-citrate buffer. Aliquots of
either cell suspension (125 µL, 4 µg of protein) or tissue
suspension (250 µL of brain, ∼200 µg of protein) were added
to each assay tube containing 25 µL (for cells) or 50 µL (for
tissues) of [3H]Ro 5-4864 (final concentration ) 1 nM, specific
activity ) 86.9 Ci/mmol; DuPont/New England Nuclear,
Boston, MA). Nonspecific binding was determined using
unlabeled PK-11195 (final concentration ) 10 µM). Sufficient
buffer was added to bring the final volume to either 250 µL
(for cells) or 500 µL (for tissues). All assays were performed
in duplicate. The assay was initiated by the addition of cells
or tissues and terminated after 60 min of incubation at 0-4
°C by rapid filtration over glass fiber filters using a filtering
manifold. The samples were washed with two 5 mL aliquots
of 50 mM Tris-HCl buffer (pH 7.4, 0-4 °C). The radioactivity
retained by the filters was measured using 4 mL of scintillation
cocktail (Cytoscint, ICN Biomedical, Aurora, OH). Nonlinear
regression analysis of the data was performed using Prism II
(GraphPad Software, San Diego, CA).
N-(ter t-Bu tyloxyca r bon yl)-4-[bis(2-h yd r oxyeth yl)a m i-
n o]-L-p h en yla la n in e Eth yl Ester (4). Solution of 3 (3.84
g, 0.0124 mol) in water:glacial acetic acid mixture (50 mL:34
mL) was cooled to 0 °C. Ethylene oxide (8.5 mL) was then
added and the reaction mixture left at room temperature for
24 h. Water and acetic acid were evaporated, and the product,
after neutralization with saturated Na2CO3 solution, was
extracted into ethyl acetate. The organic layer was washed
three times with water followed by brine and dried (MgSO4).
Ethyl acetate solution was evaporated and the product purified
by column chromatography on silica gel (70-230 mesh) with
ethyl acetate as eluent to give 4.27 g (0.0108 mol, 87%) of oily
4, homogeneous by TLC (ethyl acetate): 1H NMR (CDCl3, δ)
6.95 (d, J ) 8.1 Hz, 2H), 6.6 (d, J ) 8.4 Hz, 2H), 5.0 (bd, 1H),
4.55-4.45 (m, 1H), 4.15 (q, J ) 7.0 Hz, 2H), 3.8 (t, 6H), 3.5 (t,
4H), 3.0-2.95 (m, 2H), 1.4 (s, 9H), 1.2 (t, 3H); FAB HRMS [M
+ H] calcd for C20H33N2O6 397.23339, found 397.2357.
N-(ter t-Bu tyloxycar bon yl)-4-[bis(2-ch lor oeth yl)am in o]-
L-p h en yla la n in e Eth yl Ester (5). To the stirred and 0 °C
cooled solution of 4 (4.27 g, 0.011 mol) in anhydrous dichlo-
romethane (100 mL) was added anhydrous carbon tetrachlo-
ride (10 mL) following addition of 1 g portions of triphenylphos-
phine (8.20 g, 0.031 mol). After addition was completed, the
reaction mixture was stirred at room temperature overnight.
Solvent was evaporated and the product purified by column
chromatography on silica gel (70-230 mesh) with dichlo-
romethane followed by diethyl ether as eluent to give 3.10 g
(0.007 mol, 66%) of oily 5, homogeneous by TLC (diethyl
ether): 1H NMR (CDCl3, δ) 7.0 (d, J ) 8.1 Hz, 2H), 6.6 (d, J )
8.4 Hz, 2H), 4.95 (bd, 1H), 4.55-4.45 (m, 1H), 4.15 (q, J ) 7.0
Hz, 2H), 3.75-3.55 (m, 8H), 3.05-2.95 (m, 2H), 1.4 (s, 9H),
1.25 (t, 3H); FAB HRMS [M + H] calcd for C20H31N2O4Cl2
433.1661, found 433.1670.
4-[Bis(2-ch lor oeth yl)a m in o]-L-p h en yla la n in e Eth yl Es-
ter Tr iflu or oa ceta te (6). A solution of 5 (3.10 g, 0.007 mol)
in trifluoroacetic acid (50 mL) containing anisole (1 mL) was
stirred at room temperature for 1 h. TFA was evaporated and
diethyl ether was added to precipitate the product that
resulted in 2.60 g (0.0058 mol, 83%) of 6: mp 138-139 °C; 1H
NMR (CDCl3, δ) 7.1 (d, J ) 8.1 Hz, 2H), 6.6 (d, J ) 8.4 Hz,
2H), 4.25-4.05 (m, 3H), 3.8-3.55 (m, 8H), 3.15 (d, J ) 7.0
Hz, 2H), 1.25 (t, 3H); FAB HRMS [M + H] calcd for
C15H23N2O2Cl2 333.1137, found 333.1145.
1-(3′-Ca r b om e t h oxyp r op yl)-7-ch lor o-1,3-d ih yd r o-5-
p h en yl-2H-1,4-ben zod ia zep in -2-on e (7). 7-Chloro-1,3-di-
hydro-5-phenyl-2H-1,4-benzodiazepin-2-one (1.35 g, 0.005 mol)
was dissolved in anhydrous dimethylformamide (10 mL), and
sodium hydride (0.40 g) was added. The reaction mixture was
stirred at room temperature for 30 min, and then 4-bromobu-
tyric acid methyl ester (1.80 g, 0.010 mol) was added dropwise.
After overnight stirring at room temperature, solvent was
evaporated and the residue treated with an ethyl acetate:water
mixture. The organic layer was separated, washed with water
and brine, and dried (Na2SO4). Solvent was evaporated, and
the thick oily product was used for the next step to obtain 1.27
g (0.003 mol, 60%) of crude 7: 1H NMR (CDCl3, δ) 7.6-7.25
(m, 8H), 7.1 (d, 2H), 4.8 (d, 1H), 4.3 (q, 1H), 3.75 (d, 1H), 3.55
(s, 1H), 3.7-3.6 (m, 1H), 2.3-1.7 (m, 4H); FAB HRMS [M +
H] calcd for C20H20N2O3Cl 371.1162, found 371.1141.
Cytotoxicity Assa ys. The standard SRB (sulforhodamine
B) assay, used in the NCI screen, was used to determine
cytotoxicity for monolayer cell lines.18,19 For suspension cell
lines (i.e., D283, D283 MR, D341, and D341 MR), a com-
mercially available kit29 was used for the MTS (3-(4,5-
dimethylthiazol-2-yl)-5-[3-(carboxymethoxy)phenyl]-2-(4-sul-
fophenyl)-2H-tetrazolium, inner salt) assay. C6 glioma cells
were obtained from American Type Culture Collection (ATCC,
Rockville, MD); SF126 and SF188 were kindly provided by the
1-(3′-Ca r b oxylp r op yl)-7-ch lor o-1,3-d ih yd r o-5-p h en yl-
2H-1,4-ben zod ia zep in -2-on e (8). A solution of 7 (1.27 g,
0.003 mol) in methanol (50 mL) containing a 1 M solution of
potassium hydroxide (4 mL) was stirred at room temperature

1730 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11
Kupczyk-Subotkowska et al.
(7) Black, K. L.; Ikezaki, K.; Santori, E.; Becker, D. P.; Vinters, H.
V. Specific high-affinity binding of peripheral benzodiazepine
receptor ligands to brain tumors in rat and man. Cancer 1990,
65, 93-97.
(8) Ikezaki, K.; Black, K. L.; Toga, A. W.; Santori, E. M.; Becker, D.
P.; Smith, M. L. Imaging peripheral benzodiazepine receptors
in brain tumors in rats: in vitro binding characteristics. J .
Cerebral Blood Flow Metab. 1990, 10, 580-587.
(9) Marangos, P. J .; Patel, J .; Boulenger, J .-P.; Clark-Rosenberg,
R. Characterization of peripheral-type benzodiazepine binding
sites in brain using [3H]Ro 5-4864. Mol. Pharmacol. 1982, 22,
26-32.
(10) Basile, A. S.; Skolnick, P. Subcellular localization of peripheral-
type binding sites for benzodiazepine in rat brain. J . Neurochem.
1986, 46, 305-308.
(11) Anholt, R. R. H.; Pedersen, P.; De Souza, E.; Snyder, S. The
peripheral-type benzodiazepine receptor: localization to the
mitochondrial outer membrane. J . Biol. Chem. 1986, 261, 576-
583.
(12) Ram, S.; Ehrenkaufer, R. E. A General Procedure for Mild and
rapid Reduction of Aliphatic and aromatic Nitro Compounds
Using Ammonium Formate as a Catalytic Hydrogen Transfer
Agent. Tetrahedron Lett. 1984, 25, 3415-3418.
(13) Appel, R. Tertiary Phosphane/Tetrachloromethane, a Versatile
Reagent For Chlorination, Dehydration, and P-N Linkage.
Angew. Chem., Int. Ed. Engl. 1975, 14, 801-811.
(14) Sternback, L. H.; Fryer, R. I.; Metlesics, W.; Reeder, E.; Sach,
G.; Saucy, G.; Stempel, A. Quinazolines and 1,4-Benzodiazepines
VI. Halo-, Methyl-, and Methoxy-substituted 1,3-Dihydro-5-
phenyl-2H-1,4-benzodiazepin-2-ones. J . Org. Chem. 1962, 27,
3788-3796.
(15) Wang, J . K. T.; Taniguchi, T.; Spector, S. Structural require-
ments for the binding of benzodiazepines to their peripheral-
type sites. Mol. Pharmacol. 1984, 25, 349-351.
(16) Olson, J . M. M.; J unck, L.; Young, A. B.; Penney, J . B.; Mancini,
W. R. Isoquinoline and peripheral-type benzodiazepine binding
in gliomas: Implications for diagnostic imaging. Cancer Res.
1988, 48, 5837-5841.
(17) Basile, A. S.; Klein, D. C.; Skolnick, P. Characterization of
benzodiazepine receptors in the bovine pineal gland: Evidence
for the presence of an atypical binding site. Mol. Brain Res. 1986,
1, 127-135.
(18) Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J .;
Vistica, D.; Warren, J . T.; Bokesch, H.; Kenney, S.; Boyd, M. R.
New colorimetric cytotoxicity assay for anticancer-drug screen-
ing. J . Natl. Cancer Inst. 1990, 82, 1107-1112.
(19) Buttke, T. M.; McCubrey, J . A.; Owen, T. C. Use of an aqueous
soluble tetrazolium/formazan assay to measure viability and
proliferation of lymphokine-dependent cell lines. J . Immunol.
Methods 1993, 157, 233-340.
(20) Rubinstein, L. V.; Shoemaker, R. H.; Paull, K. D.; Simon, R. M.;
Tosini, K.; Skehan, P.; Scudiero, D. A.; Monks, A.; Boyd, M. R.
Comparison of in vitro anticancer-drug-screening data generated
with a tetrazolium assay versus a protein assay against a diverse
panel of human tumor cell lines. J . Natl. Cancer Inst. 1990, 82,
1113-1118.
(21) Rosenberg, M. C.; Colvin, O. M.; Griffith, O. W.; Bigner, S. H.;
Elion, G. B.; Horton, J . K.; Lilley, E.; Bigner, D. D.; Friedman,
H. S. Establishment of a melphalan-resistant rhabdomyosar-
coma xenograft with cross-resistance to vincristine and enhanced
sensitivity following buthionine sulfoximine-mediated glu-
tathione depletion. Cancer Res. 1989, 49, 6917-6922.
(22) Phillips, P. C. Antineoplastic drug resistance in brain tumors.
Neurol. Clin. 1991, 9, 383-404.
(23) Friedman, H. S.; Dolan, M. E.; Kaufmann, S. H.; Colvin, O. M.;
Griffith, O. W.; Moschel, R. C.; Schold, S. C.; Bigner, D. D.; Ali-
Osman, F. Elevated DNA polymerase a, DNA polymerase B, and
DNA topoisomerase II in a melphalan-resistant rhabdomyosa-
rcoma xenograft that is cross-resistant to nitrosourease and
topotecan. Cancer Res. 1994, 54, 3487-3493.
Brain Tumor Research Center, University of California, San
Francisco, CA. Development of a C6 MEL-resistant cell line
(i.e., C6 MR) was achieved by exposure of C6 cells to increasing
MEL concentrations, starting at 10 µM, at weekly intervals.
In each MEL treatment cycle, MEL would be exposed to cells
for 24 h, then removed, and split before addition of the next
higher MEL concentration. All other cell lines were supplied
by Dr. Darrell Bigner, Duke University Medical Center,
Durham, NC. For the three human melphalan-resistant cell
lines (i.e., D282 MR, D41 MR, and DAOY MR), resistance was
maintained by exposure to melphalan (between 70 and 78 µM)
for 24 h every third passage. After the 24 h exposure period,
melphalan and media were removed and then resuspended
in fresh medium prior to use in the cytotoxicity assays.
For monolayer cells (i.e., SRB assay), cell suspensions
prepared in their established culture medium were added to
96-well microtiter plates (200 µL/well at 10-20% confluency)
and incubated overnight at 37 °C in 95% CO2, 5% air.
Solutions of melphalan, 9, nordiazepam, and Ro 5-4864 were
added to the cells either as single agents or in combination
(i.e., 9 + Ro 5-4864) and incubated for 48 h at 37 °C.
Suspension cells (i.e., MTS assay) were prepared in their
established culture medium and added to 96-well microtiter
plates (100 µL/well at 10-20% confluency) followed by the
addition of drug solutions. Melphalan solutions were prepared
in normal saline, whereas all other compounds were prepared
in ethanol. The volume of ethanol added to each well did not
exceed 1% of the total volume. To avoid drug degradation, in
which melphalan is particularly susceptible, all drug solutions
were freshly prepared and rapidly added to the cells.
For the SRB assay, following the 48 h incubation period,
the medium was removed, and the cells were washed with
phosphate-buffered saline. Next, 10% trichloroacetic acid
(TCA) was added, and the plates were incubated for 60 min
at 4 °C. The TCA was removed, and the cells were washed
five times with distilled water followed by addition of 100 µL
of SRB solution to each well. The plates were then incubated
for 30 min at room temperature, the dye was removed, and
then the cells were washed four times with 1% acetic acid
followed by addition of 100 µL of 10 mM Tris buffer to each
well. Following a minimum incubation of 60 min, the absor-
bance at 560 nm of each well was measured in a microplate
reader. The absorbance values were converted to percent of
control cell growth, and the IC50 was calculated from an
algebraic equation.
For the MTS assay, following the 48 h incubation, 20 µL of
an MTS/PBS solution was added to each well. The background
absorbance was immediately determined, and then the plates
were incubated at 37 °C. The absorbance was continually
monitored every 30 min until a maximum was achieved.
GST activity and GSH concentrations in the medulloblas-
toma cell lines were determined as reported previously.21
Refer en ces
(1) Brem, H.; Mahaley, M. S., J r.; Vick, N. A.; Black, K. L.; Schold,
S. C., J r.; Burger, P. C.; Freidman, H.; Ciric, I. S.; Cozzens, T.
W.; Kenealy, J . N. Interstitial chemotherapy with drug polymer
implants for the treatment of recurrent gliomas. J . Neurosurg.
1991, 74, 441-446.
(2) Brem, H.; Piantadosi, S.; Burger, P. C.; Group, P. B. T. T.
Intraoperative chemotherapy using biodegradable polymers:
safety and effectiveness for recurrent glioma evaluated by a
prospective, multi-institutional placebo-controlled clinical trial.
Lancet 1995, 345, 1008-1012.
(3) Weller, M.; Malipiero, U.; Rensing-Ehl, A.; Barr, P. J .; Fontana,
A. Fas/APO-1 gene transfer for human malignant glioma. Cancer
Res. 1995, 55, 2936-2944.
(4) Felker, G. M.; Friedman, H. S.; Dolan, M. E.; Moschel, R. C.;
Schold, S. C., J r. Treatment of subcutaneous and intracranial
brain tumor xenografts with O6-benzylguanine and 1,3-bis(2-
chloroethyl)-1-nitrosourea. Cancer Chemother. Pharmacol. 1993,
32, 471-476.
(5) Skapek, S. X.; Colvin, O. M.; Griffith, O. W.; Groothius, D. R.;
Colapinto, E. V.; Lee, Y.; Hilton, J .; Elion, G. B.; Bigner, D. D.;
Friedman, H. S. Buthionine sulfoximine-mediated depletion of
glutathione in intracranial human glioma-derived xenografts.
Biochem. Pharmacol. 1988, 37, 4313-4317.
(6) Black, K. L.; Ikezaki, K.; Toga, A. W. Imaging of brain tumors
using peripheral benzodiazepine receptor ligands. J . Neurosurg.
1989, 71, 113-118.
(24) Kupczyk-Subotkowska, L.; Tamura, K.; Pal, D.; Sakaeda, T.;
Siahaan, T. J .; Stella, V. J .; Borchardt, R. T. Derivative of
melphalan designed to enhance drug accumulation in cancer
cells. J . Drug Targeting, accepted for publication.
(25) Greig, N. H.; Momma, S.; Sweeney, D. J .; Rapoport, S. I.
Facilitated transport of melphalan at the blood-brain barrier by
the large neutral amino acid carrier system. Cancer Res. 1987,
47, 1571-1576.
(26) Erlenmeyer, E.; Lipp, A. Synthese des Tyrosins. Ann. Chem.
1883, 219, 161-233.
(27) Bergel, F.; Stock, J . A. Cyto-active Amino-acid and Peptide
Derivatives. Part I. Substituted phenylalanines. J . Chem. Soc.
1954, 2409-2417.
(28) Basile, A. S.; Skolnick, P. Tissue specific regulation of “peripheral-
type” benzodiazepine receptor density after chemcial sympath-
ectomy. Life Sci. 1988, 42, 273-283.
(29) CellTiter 96 AQueous Assay; Promega: Madison, WI, 1994.
J M960592P