Lipoamino acid conjugates of methotrexate with antitumor activity

Article abstract of DOI:10.1021/js970194p

The synthesis, characterization, and in vitro antitumor activity against a wild and a transport-resistant CCRF-CEM cell line is described for a series of α,γ-bisamide lipoamino acid and oligomer conjugates of methotrexate. The influence of the lipophilicity of the conjugates on the cytotoxicity and the dihydrofolate reductase inhibition was investigated. All compounds were more active than their fatty acid conjugate analogues. Compound 1e with a 12- carbon atom aliphatic side chain showed the highest in vitro activity.

Full text of DOI:10.1021/js970194p

Lipoamino Acid Conjugates of Methotrexate with Antitumor Activity
†,‡
§
§
|
,†
ROSARIO PIGNATELLO , GERRIT JANSEN , IETJE KATHMANN , GIOVANNI PUGLISI , AND ISTVAN TOTH*
Contribution from The Department of Pharmaceutical and Biological Chemistry, The School of Pharmacy,
University of London, London, WC1N 1AX, UK, The Department of Oncology, University Hospital Vrije, Amsterdam,
1077 MB, NL, and Dip. di Scienze Farmaceutiche, Universita` di Catania, I-95125 Catania, Italy.
Received May 12, 1997. Accepted for publication December 8, 1997.
CEM line, which lacks the physiological transport system
Abstract 0 The synthesis, characterization, and in vitro antitumor
activity against a wild and a transport-resistant CCRF-CEM cell line
is described for a series of R,γ-bisamide lipoamino acid and oligomer
conjugates of methotrexate. The influence of the lipophilicity of the
conjugates on the cytotoxicity and the dihydrofolate reductase inhibition
was investigated. All compounds were more active than their fatty
acid conjugate analogues. Compound 1e with a 12-carbon atom
aliphatic side chain showed the highest in vitro activity.
for the folates.15
To investigate this area further, it was reasoned that
the bifunctional nature of various lipoamino acids (LAAs)16,17
could be exploited to allow the incorporation of different
levels of lipidic substitution onto the MTX molecule,
thereby increasing the propensity for passive cellular
uptake of the chemotherapeutic conjugate. We would
therefore like to report the synthesis and biological evalu-
ation of a series of novel MTX R,γ-bis LAA conjugates 1b-
r .
Introduction
Materials and Methods
Methotrexate (4-amino-4-deoxy-N10-methylpteroyl-L-
glutamic acid, MTX, 1a ), a classical inhibitor of dihydro-
folate reductase (DHFR), is widely used alone or as a basic
component of combination regimes for treatment of human
neoplastic diseases, like choriocarcinoma, lymphocytic
leukemia, and diffuse lymphomas and sarcomas, and for
the therapy of severe cases of rheumatoid arthritis and
psoriasis.1 However, as well as its marrow-related toxicity,
the clinical efficacy of MTX is often limited by inherent or
acquired cell resistance, due to different mechanisms. In
particular, impairment of the active transport system,
associated with both reduced folates and the majority of
MTX uptake, has been recognized as the basis of a
“transport resistance” in some cases.2,3
It was reasoned that an overall reduction in the polarity
of the MTX molecule may enhance passive cellular uptake
into tumor cells and therefore evade the transport-
resistance barrier. Modification of the carboxyl groups
present in the glutamic acid moiety of MTX, defined as a
region of “bulk tolerance”, appeared an ideal target, since
the high-affinity binding of MTX to target enzymes derives
from the diamino pyrimidine portion of the pteridine ring.4
Unsurprisingly, this has been the focus of many studies,
mainly exploiting mono- and diesters of MTX5-7 and
aromatic or short chain aliphatic mono- and bis-amides.8-13
In our previous work,14 a series of linear aliphatic MTX
R,γ-bis(amides) were prepared and tested in vitro to
investigate the effects of different substitutions of the free
carboxyl groups present in MTX on the affinity toward
DHFR and its growth inhibitory activity on different tumor
cell lines. Preliminary results demonstrated that bis-
(tetradecyl)amide had the highest in vitro antitumor
potency. Interestingly, despite these derivatives being less
active than MTX against wild CCRF-CEM cells, they
retained almost the same activity against an MTX-resistant
MTX (purity >98%, HPLC) was kindly gifted by Cyan-
amid (Catania, Italy). Lipoamino acids and their esters
were racemic and prepared as described by Gibbons et al.17
1H NMR spectra were recorded in DMSO-d6, CDCl3, or a
2:1 CDCl3-MeOD-d4 mixture, with either a Bruker AM250
or AM500 instrument operating at 250 or 500 MHz,
respectively; chemical shifts are reported in ppm downfield
from internal TMS. Fast-atom bombardment (FAB) mass
spectra were run on a VG analytical ZAB-SE instrument,
using a 20 kV Cs+ bombardment and 2 µL matrix (m-
nitrobenzoic acid or thioglycerol-glycerol-TFA); a NaI or
KI methanol solution was added to produce salted species
when no protonated molecular ions were observed. Electron-
impact (EI) mass spectra were carried out on a VG
analytical ZAB-SE instrument, using electrons at 70 eV.
Elemental analysis was performed on a Carlo Erba 1106
analyzer; experimental values are within (0.3% of theo-
retical ones (Table 1). TLC was carried out on Merck
F254+366 silica gel aluminum-backed plates. Purification
was achieved by gravity column chromatography on either
Merck Kieselgel G60 or dry G60 silica gel (230-400 mesh),
with the eluent systems indicated. Fractions identified by
MS mass matching were further checked for purity using
RP-HPLC. Since we used racemic lipoamino acids for
conjugation, the MTX lipidic conjugates were diastereo-
meric mixtures (melting points not reported).
Met h od AsMethotrexate R,γ-bis(methyl 2-amino-
decanoate) (1b)sMethotrexate 1a (90 mg, 0.2 mmol),
triethylamine (50 mg, 0.5 mmol), 1-hydroxybenzotriazole
hydrate (65 mg, 0.42 mmol), and 1-ethyl-3-(3-dimethyl-
aminopropyl)carbodiimide hydrochloride (EDAC; 100 mg,
0.52 mmol) were dissolved in 4 mL of dry dimethyl-
formamide at 0 °C in an ice bath; after 30 min, a solution
of methyl R-aminodecanoate hydrochloride16 (100 mg, 0.42
mmol) in 2 mL of dry dichloromethane was added and the
mixture was stirred at 0 °C for 2 h and then at room
temperature for 24 h. Solvents were removed under high
vacuum, and the residue was dissolved in 30 mL of
dichloromethane and washed with water (30 mL), 5% acetic
acid solution (20 mL), 10% sodium bicarbonate solution (30
mL) and brine (2 × 30 mL). The organic layer was then
* Corresponding author: Tel: +44 171 753 5873. Fax: +44 171 278
1939. E-mail: itoth@cua.ulsop.ac.uk.
† University of London.
‡ On sabbatical leave from the Department of Pharmaceutical
Sciences, University of Catania, Italy.
§ University Hospital Vrije.
| Universita di Catania.
© 1998, American Chemical Society and
American Pharmaceutical Association
S0022-3549(97)00194-9 CCC: $15.00
Published on Web 01/23/1998
Journal of Pharmaceutical Sciences / 367
Vol. 87, No. 3, March 1998

Table 1sAnalytical Data of MTX Lipoamino Acid Conjugates
% yield
TLC
f
HPLC tR
compd
method
A
appearance
R
(min)
formula
elemental analysis
1b
75
0.51,
0.52a
0.53,
0.57a
0.54a
6.63
9.15
C H N O
calcd C 60.12, H 7.93, N 16.69 (‚H O)
found C 60.15, H 8.00, N 16.82
calcd C 62.76, H 8.92, N 14.16 (‚1.5H O)
found C 62.83, H 8.80, N 14.14
calcd C 61.11, H 8.36, N 15.49 (‚1.5H O)
found C 61.01, H 8.49, N 15.69
calcd C 63.13, H 8.69, N 14.72 (‚H O)
found C 62.93, H 8.78, N 14.52
calcd C 67.79, H 9.63, N 12.16
found C 67.59, H 9.60, N 11.93
42 64 10
7
2
yellow
66
yellow
65
yellow
68
yellow
71
yellow
80
yellow
71
yellow
80
yellow
87
orange-yellow
68
dark yellow
82
dark yellow
74
dark yellow
77
dark yellow
90
dark yellow
75
dark yellow
88
orange-yellow
(821.044)
1c
1d
1e
1f
A
A
A
A
A
A
A
B
B
B
B
B
B
B
B
C H102N O
62
12
9
2
(1159.59)
7.57
C H N O
46 72 10
7
2
(877.15)
0.51,
0.53a
0.53a
9.75
C H N O
50 80 10
7
2
(933.26)
11.88
10.77
11.77
11.63
6.08
C H134N O
78 12 9
(1384.02)
1g
1h
1i
0.52,
0.54a
0.56,
0.59a
0.52,
0.53a
0.18b
C H N O
calcd C 62.70, H 9.06, N 13.54 (‚2.5H O)
54 88 10
7
2
(989.42)
found C 62.78, H 9.24, N 13.71
calcd C 69.04, H 10.10, N 11.23
found C 69.30, H 10.27, N 11.09
C H150N O
86
12 9
(1496.23)
C H N O
calcd C 64.95, H 9.30, N 13.06 (‚2H O)
found C 64.75, H 9.40, N 13.14
calcd C 57.33, H 7.82, N 16.71 (‚2.5H O)
found C 57.48, H 7.80, N 16.55
calcd C 62.69, H 8.77, N 14.62 (‚H O)
found C 62.66, H 8.80, N 14.59
calcd C 60.95, H 8.14, N 16.15 (‚H O)
found C 60.87, H 8.26, N 16.08
calcd C 63.69, H 8.46, N 15.47
found C 63.88, H 8.65, N 15.34
calcd C 63.92, H 9.74, N 11.77 (‚4H O)
found C 63.76, H 9.94, N 11.77
58 96 10
(1045.53)
C H N O
7
2
1j
40 60 10
(792.99)
7
2
1k
1l
0.18b
0.21b
0.26b
0.27b
0.28b
0.28b
0.29b
8.77
C H N O
60 98 12
(1131.53)
C H N O
9
2
7.02
44 68 10
(849.10)
7
2
1m
1n
1o
1p
1r
8.65
C H N O
48 76 10
7
(905.25)
11.08
9.57
C H130N O
76
12
9
2
(1355.94)
C H N O
calcd C 59.40, H 9.01, N 13.32 (‚5H O)
52 84 10
7
2
(961.36)
found C 59.58, H 8.86, N 13.52
calcd C 68.72, H 10.02, N 11.45
found C 68.90, H 10.23, N 11.12
11.25
10.74
C H146N O
84 12 9
(1468.22)
C H N O
calcd C 63.84, H 9.18, N 13.29 (‚2H O)
found C 64.03, H 9.22, N 13.06
56 92 10
7
2
(1017.47)
TLC systems: aCH Cl2-MeOH, 8:2. b CH Cl2−MeOH−AcOH 8.5:1:0.5.
2
2
γ
dried (MgSO4) and the solvent removed in vacuo. The
crude product was purified by column chromatography on
silica gel [eluent system, CH2Cl2 (400 mL), CH2Cl2-MeOH,
95:5 (1000 mL)]: yield 123 mg (75%) yellow solid; 1H NMR
(CDCl3) δ 8.56 (1H, s, H-7), 8.10 (1H, d, CONH-Glu), 7.63,
6.68 (4H, 2d, arom), 6.55 (1H, d, CONH), 4.68 (2H, s,
9-CH2), 4.44 (2H, bs, 2 R-CH), 3.69-3.59 (6H, m, 2
COOCH3), 3.11 (3H, s, NCH3), 2.50-2.05 (2H, m, CH2),
1.16, 1.15 (28H, 2s, 14 CH2), 0.79 (6H, t, 2 CH3); FAB-MS
(m/z, %) 844, 843 [M + Na]+ (22, 55), 822 [M + 1]+ (30),
821 [M]+ (50), 669 (10), 659 (15), 621,620 (10, 21), 461 (14),
446 (13), 439 (12), 308 (100); EI-MS (m/z, %) 662 (29), 648,
647 (37, 93), 530 (38), 515 (32), 441 (15), 316 (19), 220 (25),
191 (25), 136 (34), 109 (100).
COOCH3), 3.11 (3H, m, NCH3), 2.60-1.70 (4H, m, âCH2 CH2-
CO), 1.23-1.18 (44H, 7s, 22 CH2), 0.85 (6H, t, 3 CH3); FAB-
MS (m/z, %) 957, 956 [M+ + Na]+ (10, 22), 934 [M + 1]+
(47), 676 (10), 338 (12), 330 (9), 308 (100), 290 (18), 280
(15), 258 (15).
Methotrexate R,γ-Bis[methyl 2-(2-aminotetradecanoyl)-
aminotetradecanoate] (1f)s1H NMR (CDCl3) δ 8.65 (1H, br,
H-7), 8.05 (1H, br, NH2), 7.70, 6.77 (4H, br, arom), 4.75
(2H, s, 9-CH2), 4.40 (4H, br m, 4 R-CH), 3.71 (6H, m, 2
COOCH3), 1.27-1.20 (88H, 6s, 44 CH2), 0.87 (12H, t, 4
CH3); FAB-MS (m/z, %) 1407 [M + Na]+ (64), 1384 [M +
1]+ (21), 1108 (12), 1022 (37), 912 (72), 795 (33), 753 (42),
694 (48), 662 (73), 647 (87), 530 (100), 512 (38).
Methotrexate R,γ-Bis(methyl 2-aminohexadecanoate)
(1g)s1H NMR (CDCl3) δ) 7.70, 6.70 (4H, br, arom), 4.75
(2H, bs, 9-CH2), 4.45 (2H, m, 2 R-CH), 3.72 (6H,d, 2
COOCH3), 3.15 (3H, s, NCH3), 1.25 (52H, s, 26 CH2), 0.87
(6H, t, CH3); FAB-MS (m/z, %) 1012 [M+ + Na]+ (63), 991
[M + 1]+ (100), 838 (13), 684 (17), 410 (87), 331 (55).
Methotrexate R,γ-Bis[methyl 2-(2-aminohexadecanoyl)-
aminohexadecanoate] (1h )s1H NMR (CDCl3) δ 8.63 (1H,
bs, H-7), 7.70, 6.75 (4H, m, arom), 7.30 (2H, bs, NH2), 4.76
(2H, bs, 9-CH2), 4.60, 4.40 (4H, m, 4 R-CH), 3.70 (6H, m, 2
COOCH3), 3.20 (3H, m, NCH3), 1.24 (104H, s, 52 CH2), 0.87
(12H, t, 4 CH3); FAB-MS (m/z, %) 1519 [M+ + Na]+ (17),
1497 [M + 1]+ (27), 694 (23), 665 (42), 639 (48), 612 (28),
539 (100), 519 (85), 460 (99), 435 (57), 369 (43), 350 (49).
Methotrexate R,γ-Bis(methyl 2-aminooctadecanoate)
(1i)s1H NMR (CDCl3) δ 8.65 (1H, s, H-7), 7.70, 6.75 (4H,
m, arom), 5.29 (2H, br, CH), 4.75 (2H, s, 9-CH2), 4.50 (2H,
m, 2 R-CH), 3.70 (6H, m, 2 COOCH3), 3.18 (3H, s, NCH3),
1.27-1.21 (60H, 3s, 30 CH2), 0.87 (6H, t, 2 CH3); FAB-MS
(m/z, %) 1068 [M + Na]+ (93), 1046 [M + 1]+ (77), 884 (26),
732 (100), 558 (50), 488 (29), 459 (35), 426 (58).
Methotrexate R,γ-Bis[methyl 2-(2-aminodecanoyl)-ami-
nodecanoate] (1c)s1H NMR (CDCl3) δ 8.60 (1H, bs, H-7),
7.65, 6.75 (4H, 2 d, arom), 6.55 (1H, s, CONH-Glu), 4.50
(4H, m, 4 R-CH), 3.70 (6H, bs, 2 COOCH3), 3.18 (3H, s,
NCH3), 2.40-1.60 (4H, m, CH2γCH2CO), 1.24 (56H, s, 28
â
CH2), 0.85 (12H, t, 4 CH3); FAB-MS (m/z, %) 1183, 1182
[M + Na]+ (30, 42), 1160 [M + 1]+ (15), 1006 (11), 789 (12),
445 (16), 437 (18), 279 (19).
Methotrexate R,γ-Bis(methyl 2-aminododecanoate) (1d)s1H
NMR (CDCl3) δ 8.65 (1H, d, H-7), 7.71, 6.75 (4H, 2 m,
arom), 5.26 (1H, bs, CH), 4.76 (2H, s, 9-CH2), 4.60, 4.50
(2H, 2m, 2 R-CH), 3.73 (6H, m, 2 COOCH3), 3.18 (3H, d,
NCH3), 2.45-2.00 (4H, br m, âCH2γCH2CO), 1.27-1.21
(36H, 4s, 18 CH2), 0.88 (6H, t, 2 CH3); FAB-MS (m/z, %)
900 [M+ + Na]+ (100), 878 [M + 1]+ (82), 725 (20), 716 (24),
648 (27), 583 (76), 473 (36), 409 (30), 402 (39).
Methotrexate R,γ-Bis(methyl 2-aminotetradecanoate)
(1e)s1H NMR (CDCl3) δ 8.55 (1H, bs, H-7), 8.00 (1H, br,
NH), 7.70, 6.65 (4H, br, arom), 7.40 (2H, br, NH2), 4.75 (2H,
s, 9-CH2), 4.60, 4.50 (2H, m, 2 R-CH), 3.48 (6H, m,
368 / Journal of Pharmaceutical Sciences
Vol. 87, No. 3, March 1998

Meth od BsMethotrexate R,γ-bis(2-aminodecanoic acid)
(1j)sCompound 3a (75 mg) was dissolved in MeOH (10
mL), a 1 M MeOH/water (4:1) solution of NaOH (10 mL)
was added, and the mixture was stirred at 40 °C. The
progress of the reaction was followed by TLC. Upon
completion (usually, 10-20 min), the organic solvent was
removed in vacuo and the mixture was diluted with water
and acidified with a saturated citric acid solution to pH 6.
The product was extracted with dichloromethane (2 × 40
mL) and the organic layer was dried (MgSO4) and evapo-
rated off in vacuo, to give an yellow-orange product (63 mg,
87% yield): 1H NMR (DMSO-d6) δ 8.55 (1H, s, H-7), 7.99
(1H, br s, CONH-Glu), 7.72-7.70, 6.82-6.78 (4H, 2d,
arom), 6.57 (1H, s, CO-NH), 4.77 (2H, s, 9-CH2), 4.40, 4.10
(2H, br, 2 R-CH), 3.19 (3H, s, NCH3), 2.00-1.40 (4H, br m,
âCH2γCH2CO), 1.35 (4H, s, 2 â-CH2), 1.17 (24H, 3s, 12 CH2),
0.82 (6H, t, 2 CH3); FAB-MS (m/z, %) 837, 815 [M + 2Na,
M + Na]+ (10, 35), 794 [M + 1]+ (54), 641 (18), 606 (12),
482 (68), 460 (100), 443 (29), 430 (72), 410 (56); EI-MS (m/
z, %) 662 (20), 648, 647 (26, 62), 631 (3), 530 (63), 515 (49),
441 (83), 337 (29), 320 (27), 276 (22), 219 (48), 136 (58),
121 (67), 109 (100).
br, arom), 7.52 (2H, bs, NH2), 4.46 (2H, bs, 9-CH2), 4.30-
3.99 (4H, m, 4 R-CH), 3.50 (3H, s, NCH3), 1.90-1.50 (4H,
br m, CH2γCH2CO), 1.24 (104H, s, 52 CH2), 0.86 (12H, t,
â
4 CH3); FAB-MS (m/z, %) 1491, 1490 [M + Na]+ (41, 75),
1468 [M]+ (100), 1446 (16), 1365 (19), 1234 (22), 1197 (36),
1147 (33), 1090 (31), 1037 (41), 1002 (34).
Methotrexate R,γ-bis(2-aminooctadecanoic acid) (1r )s1H
NMR (DMSO-d6): δ ) 8.55 (1H, s, H-7), 8.01 (1H, br, CO-
NH-Glu), 7.69, 6.79 (4H, 2d, arom), 6.55 (1H, bs, CO-NH),
4.75 (2H, s, 9-CH2), 4.40, 4.10 (2H, m 2 R-CH), 3.18 (3H, s,
â
NCH3), 1.9-1.6 (4H, m, CH2-γCH2-CO), 1.20 (60H, s, 30
CH2), 0.82 (6H, t, 2 CH3); FAB-MS (m/z, %) 1084, 1062,
1040 [M + 3Na, +2Na, +Na]+ (68, 79, 75), 1017, 1016 [M,
M - 1]+ (20, 22), 994 (23), 907 (35), 886 (38), 780 (63), 757
(64), 654 (62), 629 (70), 605 (100).
HP LC An a lysissThe purity of compounds was assessed
by reversed phase HPLC, using a Waters instrument
equipped with a 717 plus automatic sampler, a 616 pump,
a 600S flow controller, and a 486 UV detector, using a
Vydac C4 column connected to a Beckman Ultrasphere C8
precolumn, at a constant 1.2 mL/min flow rate. Wave-
length and sensitivity were set at 310 nm and 1.0 AUFS,
respectively. The mobile phase consisted of 0.1% TFA (v/
v) (A) and 90% v/v acetonitrile-0.1% TFA (B); solvents
were filtered through a 23 µM membrane filter and
degassed by helium flow before use. The solvent gradient
ranged from 50% to 100% solvent B in 5 min and then was
kept at 100% B for 5 min and finally decreased to 50%
acetonitrile in 3 min. Samples were dissolved in 30% (v/
v) aqueous acetic acid and filtered through on a 0.8-0.22
µM polyamide filter (Acrodisc PF, Gelman Sciences, UK)
before injection. Retention times of conjugates usually
ranged between 8.5 and 11 min; the peaks of artificial
mixtures of conjugates and MTX always displayed a good
separation, with baseline resolution of each peak, under
the above conditions (MTX tR ) 7.36).
In Vitr o Cytotoxicity Assa ysLymphoblastoid CCRF-
CEM (CEM/S and CEM/R) cells were grown at 37 °C, in a
5% CO2 atmosphere, in RPMI 1640, supplemented with
10% fetal calf serum containing 2 mM glutamine and
antibiotics (penicillin and streptomycin). Cells concentra-
tion at initial inoculum was 1 × 105 cell/mL. One milliliter
of cells was put in the individual well of a 24-well plate
containing 10 µL of drug solution in anhydrous DMSO, at
the required drug concentration. Control cultures were
included containing 10 µL of DMSO without drugs. The
final DMSO dilution (1:100) appeared to be nontoxic to cell
growth. Six representative conjugates from the synthe-
sized pool were chosen for the cellular experiments. Com-
pounds 1j, 1m , 1r , and 1n represented acid conjugates with
increasing lipophilicity, and compounds 1e and 1f repre-
sented methyl ester conjugates with increasing lipophilicity
(a medium length monomer and dimer lipoamino acid were
chosen). Stock solutions of 5 mM 1e, 1j, and 1m and 2.5
mM 1f and 1r were made in DMSO, and compound 1n was
examined in 10 µM solution. After 72 h incubation with
the compounds, cell counts and viability were determined
with a hemocytometer by tripan blue exclusion assay.
DHF R In h ibition Assa ysDHFR activity was measured
by following NADPH oxidation at 340 nm as described
previously.14
Methotrexate R,γ-bis[2-(2-aminodecanoyl)-aminodecanoic
acid] (1k)s1H NMR (CDCl3-MeOD) δ 7.80, 6.85 (4H, br,
arom), 4.89 (2H, s, 9-CH2), 4.50 (4H, m, 4 R-CH), 3.31 (3H,
â
s, NCH3), 2.00-1.75 (4H, br m, CH2γCH2CO), 1.37 (56H,
s, 28 CH2), 0.97 (12H, t, 4 CH3); FAB-MS (m/z, %) 1197,
1176, 1154 [M + 3Na, + 2Na, + Na]+ (12, 33, 100), 1132
[M + 1]+ (37), 984 (16), 978 (12), 951 (11), 663 (20), 646
(15), 530 (12).
Methotrexate R,γ-bis(2-aminododecanoic acid) (1l)s1H
NMR (CDCl3-MeOD) δ 8.60 (1H, s, H-7), 8.05 (1H, br,
CONH-Glu), 7.77-6.80 (4H, 2d, arom), 6.65 (2H, br, NH2),
4.83 (2H, s, 9-CH2), 4.15 (2H, br, 2 R-CH), 3.40 (3H, s,
NCH3), 2.10-1.50 (4H, m, âCH2γCH2CO), 1.25 (36H, 2s, 18
CH2), 0.90 (6H, t, 2 CH3); FAB-MS (m/z, %) 916, 894, 872
[M + 3Na, +2Na, +Na]+ (3, 11, 33), 849 [M + 1]+ (18), 613
(11), 483, 482 (29, 65), 460 (100), 443 (25), 413 (27).
Methotrexate R,γ-bis(2-aminotetradecanoic acid) (1m)s1H
NMR (DMSO-d6) δ 8.55 (1H, s, H-7), 8.00 (1H, m, CO-
NH-Glu), 7.69, 6.79 (4H, 2d, arom), 7.40 (2H, bs, NH2), 6.53
(1H, s, CO-NH), 4.75 (2H, s, 9-CH2), 4.35, 4.11 (2H, m, 2
R-CH), 3.17 (3H, s, NCH3), 2.2-1.5 (4H, m, âCH2γCH2CO),
1.18 (44H, s, 22 CH2), 0.81 (6H, t, 2 CH3); FAB-MS (m/z,
%) 971, 949, 928 [M + 3Na, +2Na, +Na]+ (13, 37, 100),
906 [M + 1]+ (46), 881 (13), 753 (31), 663 (59), 649 (52),
590 (43), 530 (50), 491 (47), 469 (70), 460 (66), 441 (85),
421 (62), 412 (52).
Methotrexate R,γ-bis[2-(2-aminotetradecanoyl)-amino-
tetradecanoic acid] (1n )s1H NMR (CDCl3-MeOD) δ 8.55
(1H, s, H-7), 8.20 (1H, br, CO-NH-Glu), 7.75, 6.75 (4H,
2d, arom), 4.72 (2H, s, 9-CH2), 4.40 (4H, m, 4 R-CH), 3.25
â
(3H, s, NCH3), 1.75-1.60 (4H, br m, CH2γCH2CO), 1.24
(88H, m, 44 CH2), 0.86 (12H, t, 4 CH3); FAB-MS (m/z, %)
1379 [M + Na]+ (43), 1356 [M + 1]+ (24), 1204 (21), 1004
(21), 928 (15), 663 (23), 635 (43), 618 (87), 591 (100), 505
(44).
Methotrexate R,γ-bis(2-aminohexadecanoic acid) (1o)s1H
NMR (CDCl3-MeOD) δ 8.55 (1H, s, H-7), 8.15 (1H, m, CO-
NH-Glu), 7.70, 6.75 (4H, 2d, arom), 7.50 (2H, bs, NH2), 6.55
(1H, bs, CO-NH), 4.67 (2H, s, 9-CH2), 4.25 (2H, m, 2
R-CH), 3.31 (3H, s, NCH3), 1.8-1.6 (4H, m, âCH2γCH2CO),
1.21-1.13 (52H, 4s, 26 CH2), 0.84 (6H, t, CH3); FAB-MS
(m/z, %) 1050, 1028, 1006, 983 [M + 4Na, M + 3Na, +2Na,
+Na]+ (13, 100, 42, 7), 960 [M - 1]+ (24), 851 (23), 754
(32), 685 (77), 647 (47), 553 (59), 530 (64), 496 (70), 441
(86), 418 (83), 414 (74).
Results and Discussion
Ch em ist r ysDiastereomeric methotrexate conjugates
1b-i were obtained by coupling the corresponding racemic
methyl R-aminoalkanoates to 1a . The reactions were
carried out in the presence of 1-hydroxybenzotriazole,18
using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hy-
drochloride assisted coupling method, for 24 h at room
Methotrexate R,γ-bis[2-(2-aminohexadecanoyl)-amino-
hexadecanoic acid (1p)s1H NMR (CDCl3-MeOD) δ 8.45
(1H, bs, H-7), 8.05 (1H, br, CO-NH-Glu), 7.70, 6.87 (4H,
Journal of Pharmaceutical Sciences / 369
Vol. 87, No. 3, March 1998

Table 2sIC50 Values (µM) of MTX Conjugates Against CEM Cells
compd
CEM/S line
CEM/R line
selectivity ratioa
1a (MTX)
1e
1f
0.013 ± 0.003
3.7 ± 0.8
24.9 ± 2.6
>50
16.4 ± 1.8
>10
22.4 ± 3.6
1.7 ± 0.2
2.2 ± 1.0
23.0 ± 4.7
>50
13.5 ± 1.9
>10
−99.23
68.2
8.3
1j
1m
1n
1r
21.5
26.6
17.7 ± 4.1
a Expressed as [(IC CEM/S)/(IC CEM/R) − 1] × 100.
50
50
Table 3sInhibitory Activity of MTX and Its Lipoamino Acid
Conjugates against Bovine Liver DHFR
% enzyme inhibition
compd
10-5
M
10-6
M
10-7
M
10-8
M
10-9
M
1a (MTX)
1e
1f
100
100
100
95
100
90
100
94
88
59
64
70
70
100
75
40
32
54
62
54
92
51
31
23
39
43
37
89
29
12
11
32
40
21
1j
1m
1n
1r
87
a Results indicate percentage of enzyme inhibition against control.
Figure 1sGrowth-inhibitory activity of lipoamino acid conjugates of MTX 1e,
1f, 1j, 1m, 1n, and 1r against CEM/s (A) and CEM/R cells (B).
(with a seven-carbon alkyl chain), which appeared to be
inactive. It should be noted that the maximum achievable
concentration for compound 1n was 10 µM, due to its poor
solubility in the culture medium.
The above results correlate well with the reported
observations that substitution on the MTX R-carboxyl
group could hinder the binding of the compound to
DHFR.8,11 It is however noteworthy than many R-substi-
tuted MTX derivatives showed interesting features as
possible prodrugs in enzyme-assisted therapeutic schemes
(e.g., ADEPT approach).19,20
temperature. Although the resulting yields were always
higher than 65%, a small amount of starting compound 1a
and some monosubstituted side products were always
found in the reaction mixture, but they were easily removed
by chromatography.
The activity of compounds 1e, 1f, 1j, 1m , 1n , 1r , and 1a
were also tested against a CEM cell subline (CEM/R),
which is defective in the physiological “reduced folate”
carrier system (RFC). MTX (1a ) is 100 times less potent
against CEM/R than against the parental CEM cells. The
growth inhibitory activity of conjugates 1e, 1f, 1j, 1m , 1n ,
and 1r was maintained and in some cases even increased
(Figure 1B, Table 2). Compound 1e was the most active
against CEM/R cells, with a very similar IC50 value to that
of MTX. Compounds with the 12-carbon long alkyl chains
(1e, 1m ) showed the highest activity. The more lipophilic
methyl ester 1e displayed higher activity than the corre-
sponding compound with free acids, 1m . Interestingly,
compound 1j, with octyl side chains, was practically inac-
tive, and compounds with higher lipophilicity, 1n (4 × 12-
carbon atom chains) and 1r (2 × 16-carbon atom chains)
also showed reduced activity.
It is worth mentioning that incubations of MTX bis-
conjugates at 37 °C under different pH conditions [buffers
at pH 4.5 (acetate), 7.4 (PBS), and 8.0 (phosphate)] did not
show any decomposition or hydrolysis to MTX, up to 48 h
incubation.
In conclusion, the lipidic conjugates 1e, 1f, 1j, 1m , 1n ,
and 1r were able to penetrate tumor cells passively because
of their increased lipophilic character and possessed an in
vitro cytotoxic effect on MTX-transport-resistant CCRF-
CEM cells.
The lipid-modified compounds were always insufficently
soluble in water. To increase the water solubility and to
introduce a functional group for further potential deriva-
tization, the methyl esters were hydrolyzed by alkaline
treatment, resulting in compounds 1j-r .
In Vitr o CytotoxicitysTo assess the influence of the
conjugation and increased lipophilicity, representatives of
MTX conjugates (1e, 1f, 1j, 1m , 1n , and 1r ) were assayed
for their in vitro growth inhibitory activity against two lines
of lymphoblastoid CCRF-CEM cells, using both the wild
(sensitive) line (CEM/S) and a MTX-resistant mutant
(CEM/R). The activites were directly compared with that
of MTX 1a .
All the tested compounds possessed a reduced level of
activity against CEM/S cells compared with 1a (Figure 1A,
Table 1). Among the MTX conjugates, compound 1e
(tetradecanoyl derivative) showed the highest activity, with
an IC50 value of 3.7 µM. The other derivatives of MTX had
IC50 values between 15 and 25 µM, except compound 1j
In Vit r o DHF R In h ib it or y Act ivit ysThe effects of
MTX on cell proliferation is due to its inhibitory activity
on DHFR. Therefore, compounds 1a , 1e, 1f, 1j, 1m , 1n ,
and 1r were examined for their ability to inhibit purified
bovine liver DHFR. The “inhibition index”, defined as [IC50
370 / Journal of Pharmaceutical Sciences
Vol. 87, No. 3, March 1998

11. Rosowsky, A.; Bader, H.; Radike-Smith, M.; Cucchi, C. A.;
Wick, M. M.; Freisheim, J . H. Methotreaxate analogues 28.
Synthesis and biological evaluation of new γ-monoamides of
aminopterin and methotrexate. J . Med. Chem. 1986, 29, 9,
1703-1709.
12. Piper, J . R.; Montgomery, J . A.; Sirotnak, F. M.; Chello, P.
L. Syntheses of R- and γ-substituted amides, peptides and
esters of methotrexate and their evaluation as inhibitors of
folate metabolism. J . Med. Chem. 1982, 25, 5, 182-187.
13. Antonjuk,.J .; Boadle, D. K.; Cheung, H. T. A.; Friedler, M.
L.; Gregory, P. M.; Tattersall, M. H. N. R-Monoamides of
methotrexate as potential prodrugs. Arzneim.-Forsch./ Drug
Res. 1989, 39, 12-15.
MTX/IC50 bis(amide)] × 100, was employed to compare the
activities (Table 3). Interestingly, all compounds still
retained DHFR inhibitory activity even at 10-9 M concen-
tration, although these measured activities were still lower
than that of the parent MTX (1a ). It is worth noting that
conjugation of MTX with lipoamino acids resulted in
compounds posessing higher DHFR inhibitory activity than
the analogous bis(amides) synthesized via condensation of
MTX with alkylamines.14
14. Pignatello, R.; Sorrenti, V.; Spampinato, G.; Pecora, T.;
Fresta, M.; Di Giacomo, C.; Panico, A.; Vanella, A.; Puglisi,
G. Synthesis and preliminary in vitro sreening of lipophilic
R,γ-bis (amides) as potential prodrugs of methotrexate. Anti-
Cancer Drug Design 1996, 14, 253-264.
15. Westerhof, G. R.; Schornagel, J . H.; Kathmann, I.; J ackman,
A. L.; Rosowsky, A.; Forsch, R. A.; Hynes, J . B.; Boyle, F. T.;
Peters, G. J .; Pinedo, H. M.; J ansen, G. Carrier and receptor-
mediated transport of folate antagonists targeting folate-
dependent enzymes; correlates of molecular-structure and
biological activity Mol. Pharmacol. 1995, 48, 459-471.
16. Gibbons, W. A.; Hughes, R. A.; Charalambous, M.; Christodou-
lou, M.; Szeto, A.; Aulabaugh, A. E.; Mascagni, P.; Toth I.
Synthesis resolution and structural elucidation of fatty amino
acids and their homo- and hetero-oligomers. Liebigs Ann.
Chem. 1990, 1175-1183.
17. Toth, I. Review: A novel chemical approach to drug deliv-
ery: Lipidic amino acid conjugates J . Drug Targeting 1994,
2, 217-239.
18. Ko¨nig, W.; Geiger, R. New method for the synthesis of
peptides: Activation of carboxyl group with dicyclohexyl-
carbodiimide by using 1-hydroxybenzotriazoles as additives.
Chem. Ber. 1970, 103, 788-798.
19. J ungheim, L. N.; Shepherd, T. A. design of antitumor
prodrugs: substrates for antibody targeted enzymes. Chem.
Rev. (Washington, D.C.) 1994, 94, 1553-1566.
20. Huennekens, F. M. Tumor targeting: activation of prodrugs
by enzyme-monoclonal antibody conjugates. TIB TECH 1994,
12, 234-239.
References and Notes
1. Bannwarth, B.; Labat, L.; Moride, Y.; Schaeverbeke, T.
Methotrexate in Rheumatoid Arthritis. Drugs 1994, 47, 25-
50.
2. Niethammer, D.; J ackson, R. C. Changes in the molecular
properties associated with the development of resistance
against methotrexate in human lymphoblastoid cells. Eur.
J . Cancer 1975, 11, 845-854.
3. Goldman, I. D.; Matherly, L. H. The cellular pharmacology
of methotrexate. Pharm. Ther. 1985, 28, 77-102.
4. Matthews, D. A.; Alden, R. A.; Bolin, J .T; Freer, S. T.;
Hamlin, R.; Xuong, N.; Kraut, J .; Poe, M.; Williams, M.;
Hoogsteen, K. Dihydrofolate reductase: X-ray structure on
the binary complex with methotrexate. Science 1977, 197,
452-455.
5. Rosowsky, A.; Yu, C.-S. Methotrexate analogues 18. En-
hancement of antitumor effect of methotrexate and 3′,5′-
dichloromethatraxate by the use of lipid-soluble diesters. J .
Med. Chem. 1983, 26, 6, 1448-1452.
6. Rahman, L. K. A.; Chhabra, S. R., The chemistry of metho-
trexate and its analogues. Med. Res. Rev. 1988, 8, 95-156.
7. Fort, J . J .; Mitra, A. K. Effects of epidermal/dermal separa-
tion methods and ester chain configuration on the biocon-
version of a homologous series of methotrexate dialkyl esters
in dermal and epidermal homogenates of hailess mouse skin.
Int. J . Pharm. 1994, 102, 241-247.
8. Rosowsky, A.; Ensminger, W. D.; Lazarus, H.; Yu, C.-S.
Methotrexate analogues 8. Synthesis and biological evalua-
tion of bisamide derivatives as potential prodrugs. J . Med.
Chem. 1977, 20, 925-930.
9. Rosowsky, A.; Forsch, R.; Uren, J .; Wick, M. Methotrexate
analogues 14. Synthesis of new γ-substituted derivatives as
dihydrofolate reductase inhibitors an potential anticancer
agents. J . Med. Chem. 1981, 24, 4, 1450-1455.
Acknowledgments
This work was supported by Wellcome Trust Grant 046965/Z/
96. The authors wish to thank Dr. Barry Kellam for the fruitful
discussions, Mrs. Lily Randall, Mr. Mike Cocksedge, and Mr. Mire
Zloh for collecting the microanalysis, MS, and NMR data.
10. Rosowsky, A.; Yu, C.-S.; Uren, J .; Lazarus, H.; Wick, M.
Methotrexate analogues 13. Chemical and pharmacological
studies on amide, hydrazide and hydroxamic acid derivatives
of glutamate side chain. J . Med. Chem. 1981, 24, 559-567.
J S970194P
Journal of Pharmaceutical Sciences / 371
Vol. 87, No. 3, March 1998