Analogues of N(α)-(4-amino-4-deoxypteroyl)-N(δ)-hemiphthaloyl-L- ornithine (PT523) modified in the side chain: Synthesis and biological evaluation

Article abstract of DOI:10.1021/jm9606453

Four heretofore undescribed side chain analogues of N(α)-(4-amino-4- deoxypteroyl)-N(δ)-hemiphthaloyl-L-ornithine (PT523, 4) were synthesized via straightforward methods of antifolate chemistry, and their properties were compared with those of PT523 and two related compounds with the aim of defining the contribution of the hemiphthaloyl-L-ornithine moiety to the exceptional in vitro antitumor activity of this novel non-polyglutamatable aminopterin analogue. The IC₅₀ values of N(α)-(4-amino-4-deoxypteroyl)- N(β)-hemiphthaloyl-L-2,3-diaminopropanoic acid (10) and N(α)-(4-amino-4- deoxypteroyl)-N(γ)-hemiphthaloyl-L-2,4-diaminobutanoic acid (9) against A549 human non-small-cell lung carcinoma cells in culture were 23 and 22 nM, whereas those of PT523 and N(α)-(4-amino-4-deoxypteroyl)-N(ε)- hemiphthaloyl-L-lysine (8) were 1.3 and 5.2 nM. A decrease in the in vitro activities of 8 and 9 relative to PT523 was also observed against the panel of cell lines used by the National Cancer Institute to screen new drugs. However the potency of 8 and 9 remained several times greater than that of the historical control methotrexate against many of the cell lines in the screening panel. In an in vivo tumor model, SCC-VII murine squamous cell carcinoma, 9 and methotrexate were well tolerated as 5-day continuous infusions at doses of 0.52 and 0.75 mg/kg/day, whereas the highest tolerated dose of PT523 on this schedule was 0.19 mg/kg/day, in agreement with its lower IC₅₀ in culture. To assess the importance of the hemiphthaloyl group in PT523, N(α)-(4-amino-4-deoxypteroyl)-N(δ)-isophthaloyl-L-ornithine (11), N(α)-(4-amino-4-deoxypteroyl)-N(δ)-terephthaloyl-L-ornithine (12), and N(α)-(4-amino-4-deoxypteroyl)-N(δ)-(4,5-dichlorohemiphthaloyl)-L-ornithine (13) were also synthesized. The IC₅₀ values of 11 and 12 against A549 cells were 45 and 3300 nM, as compared with 1.3 nM for PT523 and 23 nM for methotrexate. In a clonogenic assay against SCC25 human squamous cell carcinoma cells, the IC₅₀ values of 11 and 12 were 2.9 and 72 nM, as compared with 0.3 nM for PT523 and 27 nM for methotrexate. Thus, activity was decreased by moving the aromatic carboxyl group in PT523 to the meta position and was further diminished by moving it to the para position. The IC₅₀ of the halogenated analogue 13 against SCC25 human head and neck squamous carcinoma cells was 18 nM, suggesting lack of tolerance for this 4,5- disubstitution in the phthaloyl moiety. Our results suggest that the combination of a hemiphthaloyl group and three CH₂ groups in the side chain are critical determinants of the potent in vitro activity of PT523.

Full text of DOI:10.1021/jm9606453

286
J . Med. Chem. 1997, 40, 286-299
An a logu es of Nr-(4-Am in o-4-d eoxyp ter oyl)-Nδ-h em ip h th a loyl-L-or n ith in e (P T523)
Mod ified in th e Sid e Ch a in : Syn th esis a n d Biologica l Eva lu a tion 1
Andre Rosowsky,*,† Chitra M. Vaidya,† Henry Bader,† J oel E. Wright,† and Beverly A. Teicher‡
Dana-Farber Cancer Institute and the Departments of Biological Chemistry and Molecular Pharmacology,
Radiation Oncology, and Medicine, Harvard Medical School, Boston, Massachusetts 02115
Received September 12, 1996X
Four heretofore undescribed side chain analogues of NR-(4-amino-4-deoxypteroyl)-Nδ-hemi-
phthaloyl-L-ornithine (PT523, 4) were synthesized via straightforward methods of antifolate
chemistry, and their properties were compared with those of PT523 and two related compounds
with the aim of defining the contribution of the hemiphthaloyl-L-ornithine moiety to the
exceptional in vitro antitumor activity of this novel non-polyglutamatable aminopterin analogue.
The IC50 values of NR-(4-amino-4-deoxypteroyl)-Nâ-hemiphthaloyl-L-2,3-diaminopropanoic acid
(10) and NR-(4-amino-4-deoxypteroyl)-Nγ-hemiphthaloyl-L-2,4-diaminobutanoic acid (9) against
A549 human non-small-cell lung carcinoma cells in culture were 23 and 22 nM, whereas those
of PT523 and NR-(4-amino-4-deoxypteroyl)-Nꢀ-hemiphthaloyl-L-lysine (8) were 1.3 and 5.2 nM.
A decrease in the in vitro activities of 8 and 9 relative to PT523 was also observed against the
panel of cell lines used by the National Cancer Institute to screen new drugs. However the
potency of 8 and 9 remained several times greater than that of the historical control
methotrexate against many of the cell lines in the screening panel. In an in vivo tumor model,
SCC-VII murine squamous cell carcinoma, 9 and methotrexate were well tolerated as 5-day
continuous infusions at doses of 0.52 and 0.75 mg/kg/day, whereas the highest tolerated dose
of PT523 on this schedule was 0.19 mg/kg/day, in agreement with its lower IC50 in culture. To
assess the importance of the hemiphthaloyl group in PT523, NR-(4-amino-4-deoxypteroyl)-Nδ-
isophthaloyl-L-ornithine (11), NR-(4-amino-4-deoxypteroyl)-Nδ-terephthaloyl-L-ornithine (12),
and NR-(4-amino-4-deoxypteroyl)-Nδ-(4,5-dichlorohemiphthaloyl)-L-ornithine (13) were also
synthesized. The IC50 values of 11 and 12 against A549 cells were 45 and 3300 nM, as compared
with 1.3 nM for PT523 and 23 nM for methotrexate. In a clonogenic assay against SCC25
human squamous cell carcinoma cells, the IC50 values of 11 and 12 were 2.9 and 72 nM, as
compared with 0.3 nM for PT523 and 27 nM for methotrexate. Thus, activity was decreased
by moving the aromatic carboxyl group in PT523 to the meta position and was further
diminished by moving it to the para position. The IC50 of the halogenated analogue 13 against
SCC25 human head and neck squamous carcinoma cells was 18 nM, suggesting lack of tolerance
for this 4,5-disubstitution in the phthaloyl moiety. Our results suggest that the combination
of a hemiphthaloyl group and three CH2 groups in the side chain are critical determinants of
the potent in vitro activity of PT523.
The ability of the L-glutamic acid side chain in
classical dihydrofolate reductase (DHFR) inhibitors like
methotrexate (MTX, 1), aminopterin (AMT, 2), and
edatrexate (10-EDAM, 3) to form poly-L-glutamate
conjugates of the γ-carboxyl group is a critical determi-
nant of both their potency and therapeutic selectivity.1-3
The enzyme which catalyzes elongation of the side chain
in folates and classical antifolates, folyl γ-polyglutamate
synthetase (FPGS),4,5 is ubiquitous in normal cells and
is considered essential for cellular proliferation, since
mutant cells lacking this enzyme are auxotrophic for
thymidine and a purine.6 While the mono- and poly-
glutamates of MTX bind almost equally well to DHFR,7
polyglutamation causes a substantial increase in bind-
ing to two other enzymes of one-carbon metabolism,
thymidylate synthase (TS)8 and aminoimidazolecar-
boxamide ribotide formyltransferase (AICAR FTase).9
This enables MTX to act as a multienzyme-targeted
drug, blocking purine nucleotide as well as thymidylate
de novo synthesis. Because cells with low FPGS activity
tend to be less sensitive to classical antifolates than
those with higher FPGS activity unless they are con-
tinuously exposed to drug, the addition of one or more
glutamates by FPGS is an important determinant of the
potency of not only MTX10 but also 10-EDAM2,11 and
several of the newer TS12,13 and AICAR FTase inhibi-
tors14,15 currently in clinical trial.
In addition to the enhanced binding of classsical
antifolates to the enzymes of one-carbon metabolism,
polyglutamation plays a critical role in the cellular
retention of these drugs. Because of their polyanionic
nature, polyglutamates efflux much more slowly than
the parent monoglutamates from cells.16 Within a short
time after the end of treatment, typically as little as 1
h, the short-chained species are lost, leaving species
containing four to six glutamyl residues. While differ-
ences in the baseline activity of DHFR and other
enzymes of one-carbon metabolism may likewise play
a role in the MTX and 10-EDAM sensitivity of certain
cell lines,11b it is widely held that the therapeutic
selectivity of classical DHFR inhibitors (i.e., their cyto-
toxicity to tumor versus nontumor cells) depends mainly
on differential polyglutamation, with FPGS-competent
tumors being more sensitive than rapidly proliferating
† Department of Biological Chemistry and Molecular Pharmacology.
‡ Departments of Radiation Oncology and Medicine.
X Abstract published in Advance ACS Abstracts, J anuary 1, 1997.
S0022-2623(96)00645-0 CCC: $14.00 © 1997 American Chemical Society

Analogues of Substituted L-Ornithine
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3 287
resistance was due either to a defect in drug accumula-
tion or to an increase in DHFR activity.18a More
importantly, in a finding for which there was no
precedent in our laboratory, PT523 was more active
against the resistant cells than MTX was against the
wild-type parental cells. Biochemical studies demon-
strated that potent growth inhibition by PT523 was due
to a MTX-like effect on reduced folate pools, with
thymidine and a purine both needed in the medium to
block its effects.24 More recently, data have been
presented which strongly suggest that the enhanced
activity of PT523 compared with MTX reflects not only
stronger DHFR binding but also more efficient utiliza-
tion of the reduced folate carrier.25,26 In addition, it was
found that PT523 is synergistic with etoposide in vitro
and in vivo and thus has the potential to be useful in
multidrug regimens.27
The promising results obtained with PT523 have led
us to synthesize second-generation analogues in order
to identify the optimal structural features for biological
activity. Our initial steps in this direction included the
synthesis of the benzoyl ring halogenated analogue 7
and the L-lysine and L-2,4-diaminobutanoic acid ana-
logues 8 and 9.28 These three compounds showed
similar activity against isolated DHFR, suggesting that
binding was not very sensitive to halogenation of the
p-aminobenzoyl moiety or to small changes in the
number of CH2 groups in the side chain. On the other
hand, only 7 was similar to PT523 in its activity against
cultured tumor cells, suggesting (a) that the effect of
different chain lengths is primarily on cellular uptake
and (b) that a side chain with three CH2 groups, as in
L-ornithine, may be optimal. In the present paper we
report the synthesis of an additional group of heretofore
undescribed side chain analogues consisting of the L-2,3-
diaminopropanoic acid analogue 10, the isophthaloyl
host tissues like marrow and gut.2b However poly-
glutamation can be a two-edged sword in that some
tumors possess inherently low FPGS activity whereas
other tumors acquire this resistance phenotype during
treatment. If a patient does not respond to MTX
because of low FPGS activity in the tumor, the amount
of drug cannot be escalated because of dose-limiting host
toxicity.
With this hypothesis as the starting point, we became
interested some time ago in the therapeutic potential
of compounds that combine key pharmacodynamic
features of both the classical and nonclassical families
of DHFR inhibitors, specifically the ability to be actively
transported into cells via the reduced folate carrier
without being dependent on polyglutamation as a
prerequisite for full antifolate activity.17 Our basic
approach has been to keep the R-COOH group in place
and modify the γ-region of the side chain so as to retain
tight DHFR binding and the capacity for active trans-
port while preventing polyglutamation. Among the
numerous side chain analogues of this type that were
evaluated in the course of this work, an unusually
promising lead was NR-(4-amino-4-deoxypteroyl)-Nδ-
hemiphthaloyl-L-ornithine (PT523, 4).18,19 Two notable
properties of PT523, despite the presence of an ad-
ditional hydrophobic phenyl ring in the side chain, were
its good aqueous solubility and the ability to be actively
transported into cells. These properties are undoubt-
edly due to retention of two COOH groups in the side
chain, an advantage which sets PT523 apart from true
nonclassical DHFR inhibitors such as trimetrexate (5)
and piritrexim (6), which are not very water soluble
even as salts and are not substrates for the reduced
folate carrier utilized by MTX and other classical
antifolates. A further benefit of the two COOH groups
is that PT523 is an unlikely candidate for resistance
via the P-glycoprotein multidrug resistance phenotype.
This problem has been reported for both 5 and 6,20,21
but interestingly is not universal for all lipid-soluble
DHFR inhibitors.22,23
PT523 proved to be more active than MTX against
tumor cells whose level of acquired resistance was in
the clinically relevant 10- to 30-fold range and whose

288 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3
Rosowsky et al.
Sch em e 1
Sch em e 2
and terephthaloyl acid analogues 11 and 12, and the
4,5-dichlorophthaloyl analogue 13. Experiments di-
rected toward the synthesis of the homophthaloyl
analogues 14 and 15 and the 8-carboxynaphthoyl ana-
logue 16 are also described. Compound 10 was made
in order to complete the ω-hemiphthaloyl-R,ω-diami-
noalkanoic acid series.28 Its biological properties con-
firm that the ornithine moiety in PT523 has the optimal
chain length. Compounds 11 and 12 were prepared to
address the question of whether ortho positioning of the
COOH group is optimal. Compound 13 was made to
determine whether halogenation is as well tolerated in
the phthaloyl ring as it is in the p-aminobenzoyl group
(cf. 7).
the protected coupling product 22, which on treatment
with NaOH in DMSO (5 min at room temperature)18
underwent cleavage of the phthalimide ring and re-
moval of the N10-formyl and methyl ester groups to give
10 in ca. 39% overall yield from 21.
The synthesis of 11 (Scheme 2) began with the
previously unknown chelate bis[Nδ-[3-(methoxy-
carbonyl)benzoyl]-L-ornithinato]copper(II) (23), which
was prepared in 52% yield from the Cu(II) complex of
L-ornithine30 and 3-(methoxycarbonyl)benzoyl chloride
under Schotten-Baumann conditions in water. Re-
moval of the copper from 23 with EDTA gave Nδ-[3-
(methoxycarbonyl)benzoyl]-L-ornithine (24), which on
esterification (MeOH/SOCl2) was converted to 25 (65%
combined yield, two steps). Condensation with 21 via
the mixed anhydride method (i-BuOCOCl/Et3N) then
converted 25 to the protected intermediate 26 (33%
combined yield, two steps). Treatment of 26 with NaOH
in DMSO solution for 5 min at room temperature
simultaneously removed the N10-formyl group and
cleaved the two ester groups to afford 11 (81%).
The synthesis of the isomeric terephthaloyl derivative
12 from 21 and the previously unknown intermediate
methyl Nδ-[4-(methoxycarbonyl)benzoyl]-L-ornithinate
(27) is shown in Scheme 2. NR-(Benzyloxycarbonyl)-L-
ornithine was esterified (MeOH/SOCl2) to 28‚HCl (96%
yield), and a mixed anhydride reaction (i-BuOCOCl/
Et3N) was used to convert the latter into 29 (83% yield).
Hydrogenolysis then gave 27, which was isolated in 78%
yield as the HCl salt. Mixed anhydride coupling with
21, followed by removal of the protecting groups with
Ch em istr y
The synthesis of NR-(4-amino-4-deoxypteroyl)-Nâ-
hemiphthaloyl-L-2,3-diaminopropanoic acid (10) (Scheme
1) began with NR-(benzyloxycarbonyl)-L-2,3-diamino-
propanoic acid (17), which we had previously used in
connection with other work.29 Treatment of 17 with
phthalic anhydride in DMF solution in the presence of
a slight excess of Et3N at room temperature afforded
the phthalimide 18 (52% yield), which was converted
sequentially to the methyl ester 19 by reaction with
MeOH/SOCl2 and to the hydrochloride salt 20‚HCl by
catalytic hydrogenolysis over Pd-C in ethanolic HCl.
The overall yield for these last two steps was ca. 66%.
Condensation of 20‚HCl with 4-amino-4-deoxy-N10
-
formylpteroic acid (fAPA, 21) in DMF solution by the
mixed anhydride method (i-BuOCOCl/Et3N) yielded

Analogues of Substituted L-Ornithine
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3 289
Sch em e 3
NaOH in DMSO solution, afforded 30 and 12, respec-
tively. The overall yield of 12 in the final two steps was
38%.
In the attempted synthesis of the isomeric homoph-
thaloyl derivatives 14 and 15, reaction of homophthalic
anhydride with 28 (cf. Scheme 2 for structure) in
refluxing EtOH yielded 37 (Scheme 4), the product
expected from nucleophilic attack on the aliphatic
carbonyl group. Esterification of 37 with MeOH/SOCl2
gave 38 (95% yield), which on removal of the Cbz group
(H2/Pd-C) was converted to the amino diester 39 (44%
yield). The direction of anhydride ring opening was
confirmed by the infrared spectrum of 38, which con-
tained an aromatic CO2Me peak at 1710 cm-1 in
The synthesis of 13 (Scheme 3) began with com-
mercially available 4,5-dichlorophthalic acid, which we
converted into the hitherto unknown compound 4,5-
dichlorophthalimide (31, 44% yield) by heating the
dimethyl ester with urea and NaOMe as described for
the preparation of phthalimide.31 Treatment of 31 with
ethyl chloroformate in the presence of Et3N as described
for the reaction of phthalimide32 afforded the N-carb-
ethoxy derivative 32 (90%). Acylation of bis(L-ornithi-
nato)Cu(II) with 32 as in the preparation of 23 then
gave bis[L-2-amino-5-(4,5-dichlorophthalimido)pentana-
to]Cu(II) (33, 71% yield). Conversion of 33 to L-2-amino-
5-(4,5-dichlorophthalimido)pentanoic acid hydrochloride
(34‚HCl) was achieved in 62% yield by treatment with
6 N HCl, a procedure which we found to be more
convenient than the use of EDTA. Finally, in a se-
quence analogous to the one used to obtain 11 and 12,
esterification of 34 with MeOH/SOCl2 gave 35‚HCl (89%
yield), mixed anhydride coupling of 35 and 21 gave 36
(21%), and reaction of 36 with NaOH in DMSO gave
13 (67%).
addition to the aliphatic R-CO2Me peak at 1740 cm-1
,
and by its 1H NMR spectrum, which showed an aro-
matic CO2Me singlet at δ 3.76 in addition to the
aliphatic R-CO2Me singlet at δ 3.61. Mixed anhydride
condensation of 39 and 21 gave the protected coupling
product 40 (40%). To obtain the alternative products
of ornithine Nδ-acylation, homophthalic anhydride was
cleaved at the aliphatic carbonyl group with NaOMe,
and the resulting monoester (41, 80%) was coupled to
28 via a mixed anhydride reaction to obtain 42 (48%).
Compounds 42 and 38 had different melting points, and
1
their infrared and H NMR spectra were clearly distin-
guishable. The infrared spectrum of 42 contained only
a single aliphatic CO2Me peak at 1735 cm-1, and the
1H NMR spectrum contained, in addition to the usual
R-CO2Me singlet at δ 3.61, a second CO2Me singlet at δ
3.51 as compared with the aromatic CO2Me singlet of
38 at δ 3.76. Catalytic hydrogenolysis of the Cbz group
in 42 followed by esterification (MeOH/SOCl2) and a
mixed anhydride reaction with 21 then gave the ex-
pected products 43 (81%) and 44 (44%), respectively. No
rearrangements occurred during these steps, as evi-
The reaction of dimethyl 4,5-dichlorophthalate with
urea and NaOMe proved somewhat capricious, often
giving, instead of 31, a mixture of monomethyl 4,5-
dichlorophthalate and a compound we believe to be
N,N′-carbonyldi(4,5-dichlorophthalimide). Interestingly,
heating this compound separately with NaOMe in
refluxing MeOH converted it to 31. Similar results were
obtained when diethyl 4,5-dichlorophthalate was used
instead of the dimethyl ester. Alternative approaches
to the synthesis of 31 were also investigated without
success, including (a) heating 4,5-dichlorophthalic an-
hydride with (NH4)2CO3 at 260 °C, which gave a
sublimate consisting of a mixture of 31 and the starting
anhydride; (b) heating dimethyl 4,5-dichlorophthalate
in refluxing EtOH saturated with ammonia; and (c)
fusion of an intimate mixture of 4,5-dichlorophthaloyl
dichloride and NH4Cl at 130 °C. We concluded that the
urea method is workable but requires meticulous opti-
mization of reaction conditions.
1
denced by clear differences in the infrared and H NMR
spectra of 43 and 44 as compared with 39 and 40 (see
the Experimental Section).
While it had been our aim to convert 40 to 14 and 44
to 15 by treatment with NaOH in DMSO under our
usual conditions, we were surprised to find that the
identical product formed when either reaction mixture
was carefully adjusted to pH 4 and that this product
appeared from its 1H NMR spectrum to be the ring-

290 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3
Rosowsky et al.
Sch em e 4
closed structure 45. Critical to the interpretation of the
spectrum of 45 was that the signal assigned to the
δ-CH2 protons was a sharply resolved, highly sym-
metrical quintuplet at δ 3.86, whereas the δ-CH2
protons adjacent to the CONH group in the noncyclic
precursors 40 and 44 gave broader peaks at δ 3.02 and
3.18, respectively, a difference we ascribe to the fact that
45 is a cyclic imide and thus lacks the NH proton
reponsible for the line broadening seen for the δ-CH2
protons in amides 40 and 44. The spectrum of 45 also
contained two distinctive peaks at δ 4.11 and 7.61.
These peaks were absent in the spectra of 40 and 44,
and only the peak at δ 4.11 disappeared on addition of
D2O. We believe these peaks indicate the presence of
an enol structure (45A) in equilibrium with 45. This
cyclic product apparently forms so rapidly at pH 4 that
the open-chain compounds cannot be isolated as free
acids. HPLC analysis (C18 silica gel, 8% MeCN in 0.1
M NH4OAC, pH 7.5) of a solution obtained by dissolving
this product in base revealed three peaks, one of which
coeluted with authentic NR-(4-amino-4-deoxypteroyl)-L-
ornithine.29 The other two products, giving peaks of
roughly equal size with retention times of 19 and 21
min, were assumed to be 14 and 15. Because the
retention times on the analytical column were judged
to be too similar to allow efficient large-scale separation
on a preparative column, and because of the tendency
of 45/45A to partially break down to NR-(4-amino-4-
deoxypteroyl)-L-ornithine even after a few hours at pH
7.5 (presumably via 14 and 15), this approach was not
pursued further. The facile cleavage of the homoph-
thaloyl group stands in marked contrast with the
behavior of the phthaloyl group in PT523, which is
stable under these conditions.
formyl group and open the imide ring, the Nδ-8-carbox-
ynaphthoyl group was lost even more rapidly than the
corresponding group in the homophthaloyl series. Fac-
ile cleavage of the amide bond at the peri position was
presumably due to anchimeric assistance by the 8-car-
boxy group.
Biologica l Activity
The ability of 10 to bind to human DHFR and inhibit
the growth of human tumor cells in culture was com-
pared with that of PT523 and its previously synthesized
chain-lengthened and chain-shortened homologues, 8
and 9.28 Our interest in evaluating the analogue with
only a single CH2 group in the side chain was prompted
in part by the fact that the spatial separation between
the COOH groups in this molecule more closely ap-
proximates the distance between these groups in clas-
sical antifolates. As shown in Table 1, 10 was a DHFR
inhibitor. However, its titration curve was more cur-
vilinear than that of PT523 (results not shown), indicat-
ing that a greater concentration would be needed to
produce near-stoichiometric inhibition, a critical re-
quirement for complete arrest of de novo thymidylate
and purine nucleotide synthesis. The results in Table
1 suggest that the location of the aromatic COOH group
in PT523 may allow it to interact optimally with some
basic residue in DHFR and that altering the number of
CH2 groups diminishes this interaction. This basic
residue may lie in the distal part of the active site of
DHFR where it lies beyond the reach of the γ-COOH
group in the non-polyglutamated form of classical an-
tifolates but may be within striking distance of the
Work was also discontinued on the 8-carboxynaph-
thoyl derivative 16 because of a severe instability
problem. Condensation of 1,8-naphthalic anhydride
with NR-(benzyloxycarbonyl)-L-ornithine, followed by the
usual sequence of esterification (MeOH/SOCl2), removal
of the Cbz group (H2/Pd-C), and mixed anhydride
coupling with 21, led to the imide 49 via the intermedi-
ates 46-48. Unfortunately, when 49 was treated with
NaOH under a variety of conditions to remove the

Analogues of Substituted L-Ornithine
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3 291
Ta ble 1. Dihydrofolate Reductase Inhibition and in Vitro Cell Growth Inhibition by PT523 Analogues Modified in the
Hemiphthaloylornithine Moiety
in vitro antitumor activity: IC50 (nm)
rhDHFR activity
compd
x
R1
R2
R3
R4
IC50/[E] (nM/milliunit)a
A549 cellsb
SCC25 cellsc
8
9
10
11
12
4
2
1
3
3
3
3
COOH
COOH
COOH
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
2.9 (0.69)
2.5 (0.80)
3.5 (0.57)
2.2 ( 0.4 (0.91)
2.8 ( 0.1 (0.71)
2.4 (0.83)
2.0 ( 0.3 (1.0)
2.2 ( 0.6 (0.91)
5.2 ( 2.2 (0.25)
22 ( 1 (0.059)
23 ( 1 (0.057)
45 ( 5 (0.029)
3300 ( 580 (0.00039)
ND
ND
ND
ND
COOH
H
Cl
H
2.9 ( 0.7 (0.10)
72 ( 16 (0.0042)
18 ( 3 (0.017)
0.3 ( 0.1 (1.0)
27 ( 3 (0.011)
H
COOH
Cl
H
13
COOH
COOH
PT523 (4)
MTX (1)
1.3 ( 0.2 (1.0)
23 ( 5 (0.057)
a
Human recombinant DHFR was expressed in Escherichia coli and purified as described earlier.25 Activity was measured at 22 °C by
measuring the change in UV absorbance at 340 nm in an assay solution containing 60 µM NADPH, 50 µM dihydrofolate, and 3-6 milliunits
of enzyme in 50 mM Tris HCl, pH 7.0. The reaction was initiated by addition of dihydrofolate after preincubation of the other components
for 2 min. Results are averaged from a minimum of three experiments on different days for 8-10 and 13, which were assayed once. IC50
values are expressed in normalized units of nM/milliunit of DHFR activity to compensate for the fact that precisely the same amount of
b
enzyme was not added to every titration. Values in parentheses are normalized relative to PT523 (1.0). The A459 human non-small
lung carcinoma cells (growth assay) were plated at a density of 2000-3000/well in RPMI medium containing 10% fetal bovine serum.
After 24 h at 37 °C in a 5% CO2 humidified atmosphere to allow the cells to adhere to the dishes, they were treated for another 72 h with
a range of drug concentrations (0.1-1000 nM), and then stained with Sulforhodamine B as described.35 Cell growth, expressed as a
percent of untreated controls, was plotted as a function of the drug concentration on a log scale. c The SCC25 human head and neck
squamous carcinoma cells (clonogenicity assay) were grown in tissue culture flasks in Dulbecco’s minimal essential medium (MEM)
containing 10% fetal bovine serum and hydrocortisone (400 µg/L). Cells in exponential growth cells were preincubated without drug for
48 h, treated with a range of drug concentrations (0.01-5000 nM) for another 72 h, and replated in fresh medium. After 14 days, the
colonies were stained with crystal violet and counted. Only colonies containing g 5 cells were counted. The surviving fraction was
plotted as a percentage of untreated controls, and the IC50 was estimated from a log-log plot. Each experiment was performed in replicate
plates and was repeated at least three times on different days. Numbers in parentheses are normalized potencies (PT523 ) 1.0). ND )
not determined.
R-COOH groups in the second glutamyl residue of a
polyglutamate. A residue that could conceivably fulfill
this role, most likely via hydrogen bonding through an
intervening molecule of water, is Arg-28.33
of 11 and PT523 against DHFR suggests that the meta
and ortho COOH groups are able to interact with the
same basic amino acid residue in the enzyme complex.
However it is also possible that they can interact equally
with two different residues. X-ray crystallographic or
high-field NMR data for the complexes of these isomeric
compounds with DHFR may allow these possibilities to
be distinguished.
In a growth inhibition assay against human A549
non-small-cell lung carcinoma cells, the IC50 of 10 (one
CH2 group) for 72 h of continuous exposure was found
to be 23 ( 1 nM, a value essentially identical to that of
9 (two CH2 groups) or MTX, but 4-fold higher than that
of 8 (four CH2 groups) and 18-fold higher than that of
PT523 (three CH2 groups). It thus appears three CH2
groups in the side chain are optimal for activity against
intact cells as well as for binding to the enzyme. The
finding that the IC50 differences were greater for growth
inhibition than for DHFR inhibition suggested that the
length of the side chain affects not only DHFR binding
but also cellular accumulation. Since compounds 8-10,
unlike MTX, cannot form polyglutamates, their effect
on cell growth is not dependent on the level of FPGS
activity inside the cell.
The effect of moving the COOH group in PT523 from
the ortho to the meta and para position, as in com-
pounds 11 and 12, was likewise examined (Table 1). The
difference in IC50 values for DHFR inhibition between
PT523 and 11 was not statistically significant (Student’s
t-test). In contrast, the IC50 of 12 was 40% higher than
that of PT523, suggesting that there is at least some
loss of binding when the aromatic COOH group is moved
from the ortho to the para position and supporting the
view that this group plays an active role in enzyme
binding. As with 10, however, the titration curves for
11 and 12 both showed greater curvilinearity relative
to PT523, indicating decreased capacity for near-sto-
ichiometric inhibition. The similarity in the IC50 values
While the differences in IC50 values betwen PT523
and the positional isomers 11 and 12 were relatively
minor in the DHFR assay, the ability of these com-
pounds to inhibit cell growth differed substantially.
Thus, while the IC50 of PT523 against A549 cells after
72 h of continuous drug treatment was 1.3 ( 0.2 nM,
the IC50 of the meta isomer 11 was 45 ( 5 nM (35-fold
decreased activity) and that of the para isomer 12 was
3300 ( 580 nM (2500-fold decreased activity). We were
concerned about the possibility that 11 and 12 might
be less stable than PT523 under tissue culture condi-
tions (37 °C, pH 7.5), but this proved not to be the case,
since the three compounds were completely stable by
HPLC after being dissolved in growth medium and kept
in the incubator for 48 h. The lower activity of 11 and
12 relative to PT523 can be taken to mean that the
aromatic COOH is more than just a passive water-
solublizing group on the Nδ-aroyl ring and that its
position can dramatically influence biological activity.
It is worth noting that the ortho COOH group in PT523
can form an internal H-bond to the CdO or NH group
in the adjacent amide bond, which is obviously not
possible in the meta and para isomers.
The cytotoxicity of 11 and 12 was evaluated also in a
clonogenic assay, using human SCC25 squamous cell
carcinoma cells. In this assay the IC50 of PT523

292 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3
Rosowsky et al.
Ta ble 2. Comparative Activity of MTX with PT523 and Its
Nearest Chain Length Homologues 8 and 9 against Selected
Human Carcinoma Cell Lines in Culture
was 0.3 ( 0.1 nM, whereas that of 11 and 12 was 2.9
nM (10-fold decreased activity) and 72 nM (240-fold
decreased activity). Thus, moving the aromatic COOH
group to the meta and para position was detrimental
to activity in two cell lines representing different
histopathologic types and different tissues of origin.
Although the differences among the three isomers were
less pronounced against the SCC25 cells than against
the A549 cells, the same qualitative trend in IC50 values
was obtained (i.e., ortho < meta , para). Whether the
greater differences among these compounds against
A549 cells than against SCC25 cells reflect the proper-
ties of these cell lines or the assay method (i.e., cell
growth versus clonogenicity) was not determined.
IC50 (nM)a
tumor typeb
NSCL
cell line
MTX
PT523c
8
9
NCI-H460
NCI-H23
NCI-H522
EKVX
HCT-116
HCT-15
HT29
SW-620
KM12
SF-539
28
43
229
>103
30
30
32
33
42
0.5
4.4
23
4.2
9.2
>103
36
0.6
14
32
>103
21
23
colon
0.3
1.2
1.2
1.9
3.2
3.7
2.5
>104
0.4
1.8
6.0
2.0
4.6
2.4
2.5
>102
0.6
5.5
2.4
3.6
2.7
>102
9.5
5.0
5.2
5.7
3.4
7.2
9.8
6.3
4.9 >103
CNS
35
26
57
8.4 >102
>103
SF-268
52
SNB-75
LOX IMVI
UACC-62
SK-MEL-5
>104
26
The last compound tested in these experiments was
the dichlorophthaloyl derivative 13. In the clonogenic
assay against SCC25 cells, its IC50 was found to be 18
nM, a 60-fold decrease in activity relative to that of
PT523. It thus appears that 4,5-dichloro substitution
on the hemiphthaloyl moiety has an undesirable influ-
ence on DHFR binding and/or transport. It may be
noted that the van der Waals radius of a chlorine atom
is 1.8 Å versus 1.2 Å for hydrogen and that the predicted
effect of two Cl atoms meta and para to the COOH group
would be to increase acidity by approximately 0.6 pK
unit. Although it is unknown how much each of these
factors individually influence DHFR binding and cel-
lular uptake, our results indicate that the net effect of
this particular structural modification is a marked
decrease in antitumor activity.
melanoma
0.9
4.4
28
87
>103
12
85
91
MALME-3M >103
>103
>103
SK-MEL-28
OVCAR-8
OVCAR-5
OVCAR-3
786-0
UO-31
ACHN
MCF7
MCF7-ADR
PC-3
>103
31
>103
>103
ovarian
renal
5.9
8.1
>103
398
33
>103
>103
86
>103
9.2
8.0
45
191
40
36
78
2.0
23
7.8
5.2
breast
5.0 >103
9.1 >103
prostate
5.0
9.1
6.7
34
DU-145
a
Assays were performed according to the standard NCI proto-
col,33 in which cells are plated in 96-well plates, treated continu-
ously with drug for 48 h, and stained for total protein with
Sulforhodamine B. All the data were generously provided by the
Division of Cancer Treatment, National Cancer Institute (DCT/
NCI). The results for PT523 against all the cell lines except the
breast and prostate carcinomas are from two experiments (ID#
9009RC34 and 9510SR81), whereas those for the breast and
prostate carcinomas are from a single experiment (ID# 9510SR81).
The results for 8 and 9 against all the tumors are likewise from a
single experiment (ID# 9407RM60). Although assays were carried
out with all the tumors in the NCI solid tumor panel, data are
included in the table only for those cell lines whose IC50 values
(GI50 values according to NCI terminology) agreed closely and
could be averaged. The IC50 values of MTX are the historical
averages of >50 assays per cell line and were kindly provided by
Dr. J . W. H. Watthey (Starks C. P., Rockville, MD). b NSCL ) non-
small-cell lung. c In a separate experiment (ID# 9510LC81) in
which cells that were relatively insensitive to PT523 during the
standard 48-h exposure were treated for 144 h, the following IC50
values were obtained: NCI-H522, 3.7 nM; EKVX, 1.6 nM; SNB-
75, 2.9 nM; OVCAR-3, 2.8 nM; PC-3, 0.72 nM.
Because the in vitro activity of 10 was substantially
lower than that of the other congeners in the Nω-
hemiphthaloyl-R,ω-diaminoalkanoic acid series,28 this
compound was not studied further. However, we did
compare the activity of PT523 and its two nearest chain
length homologues, 8 and 9, against the panel of human
tumor cell lines currently used by the National Cancer
Institute to screen new agents for activity.35 In addi-
tion, we decided to compare the activity of PT523 and
its chain-shortened analogue 8 in an in vivo tumor
growth delay assay using SCC VII squamous cell
carcinoma. The latter experiment was done to test the
possibility that 8 might have more favorable in vivo
properties even though it is less active in vitro. This
could arise, for example, if 8 was cleared more slowly,
resulting in a higher plasma AUC. The results of these
experiments are summarized in Tables 2 and 3, respec-
tively.
a potential advantage over these drugs in being imper-
vious to P-glycoprotein-mediated drug resistance. The
second interesting result is that, for several of the cell
lines against which the IC50 (48 h) of PT523 was >10
nM (i.e., NCI-H522, EKVX, SNB-75, OVCAR-3, and PC-
3), nanomolar sensitivity was achieved by extending the
treatment time to 144 h (cf. Table 2, footnote c). For
cells whose IC50 (48 h) was in the 1-10 nM range,
sensitivity was increased to the subnanomolar range in
several cases (data not shown). Since PT523 cannot
form polyglutamates, differences in FPGS activity among
the various cells cannot account for their different
sensitivity to this drug. The finding that sensitivity of
some of the resistant cell lines is increased dramatically
by increasing the duration of exposure suggests that,
other things being equal, activity may depend on the
time needed for the intracellular drug to reach a critical
concentration in excess of the DHFR level. Therefore,
in order for in vivo treatment with PT523 to be effective,
Compounds 8 and 9 showed reasonably good activity
against a number of NCI solid tumor panel cell lines
(Table 2) but were consistently less active than PT523,
which had mean IC50 values in the low nanomolar (1.0-
10 nM) or subnanomolar (0.1-1.0 nM) range in 19 (70%)
of the 27 cell lines shown. By comparison, 8 and 9 had
IC50 values in the 0.1-10 nM range against 16/27 (59%)
and 9/27 (33%) tumors, respectively, whereas the IC50
of MTX was outside this concentration range against
all 27 cell lines and was >100 nM in a number of
instances. Two additional results in Table 2 are of
interest. The first is that MCF-7/ADR cells, which are
pleiotropically drug-resistant by virtue of an increase
in P-glycoprotein expresssion, were not cross-resistant
to PT523. Thus, even though PT523 shares with
lipophilic antifolates like trimetrexate (5) and piritrexim
(6) a lack of dependence on FPGS for its activity, it has

Analogues of Substituted L-Ornithine
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3 293
Ta ble 3. Effect of Na-(4-Amino-4-deoxypteroyl)-Nγ-hemiphthaloyl-L-2,4-diaminobutanoic acid (9), PT523, and MTX on the Growth of
Subcutaneously Implanted SCC-VII Squamous Cell Carcinoma in Mice
tumor size (mm)b
day 0
range
day 7
range
time to reach 10 mm (days)
compd
dosea
no. of mice
mean
mean
range
mean
TGD (days)c
0.9% NaCl
9
PT523
MTX
23d
5
1-2.5
1-1.5
1-3
1.3
1.1
2.1
1.6
4.5-15
1-9
3.5-9.5
6-8
9.8
5.1
6.8
7.3
4-21
9-27
7.2
9.8
12
0.52
0.19
0.75
2.6 (35)
5 (67)
5 (67)
8e,f
5f
8-25
1-2
10-14
12
a
Dose in mg/kg/day infused over 5 days via Alzet osmotic minipumps (Model 2001, volume 200 µL, flow rate 1.0 µL/h). Drugs were
injected in water adjusted to pH 7.5-8.0. SCC-VII cells (2 × 106, >90% dye exclusion) were implanted subcutaneously in the hind limb
b
of C3H male mice (21-30 g). Pumps were implanted subcutaneously in the scapular region when the tumor became palpable (day 0) and
were removed on day 5. Animals were euthanized in the approved manner when the tumor reached 18-20 mm, or sooner if they appeared
moribund. Tumor size was defined as the mean of two orthogonal measurements using calipers. Tumors were generally measured every
day, but on a few occasions were measured every second or third day. For calculation purposes, any tumor of <1 mm on day 0 or 1 was
scored as being 1 mm in size. c Tumor growth delay (TGD) was calculated by subtracting the average time required to reach 10 mm in
the control group from the average time to reach this size in the treatment group. Numbers in parentheses represent the percent increase
in the time to reach the endpoint. Results of four separate experiments each using five to seven mice. e Results of two experiments each
d
using four mice. f One mouse died on day 7 in one of the PT523 experiments, and one died on day 10 in the MTX experiment.
a nanomolar concentration of the drug may have to be
maintained in plasma for as long as 144 h.
determinants of the potent in vitro biological activity
of PT523. Modifications of other regions of the molecule,
such as the B ring and the bridge, are being studied in
our laboratory as part of a larger program of structure-
activity optimization.
PT523 and 9 were also tested in vivo against a
subcutaneously implanted murine tumor, SCC VII
squamous cell carcinoma.37 In view of our earlier in
vitro evidence suggesting that cell killing by PT523
depends markedly on the duration of exposure,27 the
drugs were administered over 5 days from subcutane-
ously implanted Alzet osmotic minipumps. For com-
parison purposes MTX was included in the experiment
even though we recognized that a 5-day infusion would
not necessarily be optimal for a polyglutamated drug.
The therapeutic endpoint was taken to be the number
of days for the tumor to grow from 1-2 to 10 mm. The
SCC VII tumor was chosen because it is relatively
unresponsive to MTX and because we wanted to show
that, when PT523 and MTX are administered by
continuous infusion, a lower dose of PT523 can retard
tumor growth to the same degree even though poly-
glutamation is possible only with MTX. Compound 9
was chosen over 8 because there was a greater differ-
ence between its activity and that of PT523 in vitro, and
we therefore felt that there was a better likelihood of
seeing a difference in vivo. As indicated in Table 3, an
identical tumor growth delay (TGD) of 4.8 days was
achieved with PT523 as with MTX, but the dose of
PT523 giving this effect was only 0.19 mg/kg/day
whereas that of MTX was 0.75 mg/kg/day. This 4-fold
difference was consistent with the previously reported
6.5-fold potency difference between PT523 and MTX
against SCC VII cells treated continuously for 72 h.28
Both higher and lower doses of the two drugs were also
tested, but the lower doses were less effective whereas
the higher doses were toxic (data not shown). Com-
pound 9 was well tolerated at 0.52 mg/kg/day, but gave
a somewhat lower TGD of 2.6 days. The most pertinent
aspect of these results was that 9 was obviously better
tolerated than PT523, in qualitative agreement with
their respective in vitro activities (cf. Tables 1 and 2).
Since we previously found 9 to be 10-20-fold less potent
than PT523 in culture depending on the cell line
tested,28 it is conceivable that a dose greater than 0.52
mg/kg/day would increase the TGD beyond 2.6 days
without producing toxicity.
Exp er im en ta l Section
Infrared spectra were obtained on a Perkin-Elmer Model
781 double-beam instrument. With the exception of the
homophthaloyl derivatives, for which full spectra are reported
in order to validate the structural integrity of the two series,
peaks below 1400 cm-1 (as well as weak peaks and shoulders)
are omitted. Ultraviolet absorption spectra were obtained on
a Varian Model 210 spectrophotometer. 1H NMR spectra were
obtained at 60 MHz on a Varian EM360L instrument with
Me4Si as the reference or at 500 MHz on a Varian Model
ML500 instrument. TLC analyses were on fluorescent Baker
250F silica gel plates, and spots were visualized under 254-
nm UV illumination or with the aid of iodine or ninhydrin.
Column chromatography was on Baker silica gel (regular
grade, 60-200 or 70-230 mesh; flash grade, 40-µm particle
size) or Whatman DE-52 DEAE-cellulose (preswollen). Sol-
vents for moisture sensitive reactions were dried over Linde
4A molecular sieves. Analytical HPLC separations were done
on C18 silica gel radial compression cartridges (Waters, Milford,
MA; analytical: 5-µm particle size, 5 × 100 mm; preparative:
15-µm particle size, 25 × 100 mm). Solid reaction products
were generally dried in vacuo over P2O5 at 50-70 °C, in some
instances after preliminary drying in a benchtop lyophilization
1
apparatus. Elemental analysis and H NMR spectra showed
that fractional molar amounts of organic solvents were re-
tained in some of the analytical samples even after careful
drying. Melting points (not corrected) were measured on a
Fisher-J ohns hot-stage microscope or in open Pyrex capillary
tubes in a Mel-Temp apparatus (Cambridge Laboratory De-
vices, Cambridge, MA). Synthetic starting materials and other
chemicals were purchased from Aldrich, Milwaukee, WI, and
from Fluka, Ronkonkoma, NY. Microanalyses were performed
by Robertson Laboratory, Madison, NJ , and were within
(0.4% of theoretical values unless otherwise specified.
2-L-[N-(Ben zyloxyca r bon yl)a m in o]-3-p h th a lim id op r o-
p a n oic Acid (18). Solid NR-(benzyloxycarbonyl)-L-2,3-diami-
nopropanoic acid (17) (477 mg, 2 mmol) was stirred for 24 h
in a solution of phthalic anhydride (326 mg, 2.2 mmol) and
Et3N (304 mg, 418 µL, 3 mmol) in dry DMF (25 mL) at room
temperature. The solution was evaporated to dryness under
reduced pressure, and the residue was redissolved in CHCl3.
The solution was washed with 1 N HCl (10 mL), rinsed with
H2O, dried (MgSO4), and evaporated. The oily residue was
chromatographed on a silica gel column (30 g, 20 × 2 cm) with
28:12:1 CHCl3-MeOH-28% NH4OH as the eluent. Fractions
were analyzed by TLC with the same eluent, and fractions
containing a single spot with Rf 0.53 were pooled and evapo-
In summary, the results in this paper demonstrate
that the combination of an unsubstituted hemiphthaloyl
group and three CH2 groups in the side chain are critical

294 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3
Rosowsky et al.
rated to a semisolid: crude yield 406 mg (52%); mp 81-83 °C
was removed in vacuo, and the residue was recrystallized from
petroleum ether (bp 30-60 °C) to obtain 3-(methoxycarbonyl)-
benzoyl chloride (4.5 g, 81%) as a low-melting solid which
liquified at room temperature (lit.36 mp 41-43 °C). The low-
melting solid was used directly in the next step.
(CHCl3-Et2O); IR (KBr) ν 2500-3500 br, 1775, 1710, 1660-
1
1680, 1600, 1520, 1470, 1440 cm-1; H NMR (CDCl3) δ 4.01
(m, 2H, â-CH2), 4.40 (m, 1H, R-CH), 4.75 (s, 2H, OCH2Ph),
6.0-7.7 (m, 9H, aromatic protons). Anal. (C19H16N2O6‚
0.75NH3‚0.75 H2O) C, H, N.
To a solution of L-ornithine hydrochloride (3.82 g, 22.7 mmol)
in H2O (30 mL) containing NaOH (1.8 g, 45.4 mmol) was added
a solution of Cu(OAc)2‚H2O (2.26 g, 11.3 mmol) in H2O (30 mL),
followed sequentially by solid NaHCO3 (1.9 g, 22.7 mmol) and
3-(methoxycarbonyl)benzoyl chloride (4.5 g, 22.7 mmol). The
reaction mixture was stirred for 1 h and filtered. The solid
was washed with hot MeOH and dried, first in a lyophilization
apparatus and then overnight over P2O5 to obtain a blue
powder (3.83 g, 52%): mp 220 °C with darkening; IR (KBr) ν
Both the solid and the material recovered from the mother
liquor were of sufficient purity to be used in the next step.
Met h yl 2-L-[N-(Ben zyloxyca r b on yl)a m in o]-3-p h t h a l-
im id op r op a n oa te (19). SOCl2 (0.37 mL) was added dropwise
to a stirred solution of 18 (137 mg, 0.347 mmol) in MeOH (3
mL) over a period of 3 min while the temperature was kept at
-30 to -15 °C. After 18 h at room temperature, the solvent
was evaporated under reduced pressure to obtain a solid (142
mg, 100%). Recrystallization from i-PrOH afforded colorless
needles: mp 139-140 °C; Rf 0.86 (silica gel, 95:5 CHCl3-
MeOH); IR (KBr) ν 3420, 3300, 1770, 1750, 1720, 1680, 1530,
1440, 1400 cm-1. Anal. (C20H18N2O6) C, H, N.
Meth yl 2-L-Am in o-3-p h th a lim id op r op a n oa te Hyd r o-
ch lor id e (20‚HCl). A solution of 19 (320 mg, 0.837 mmol) in
a mixture of 95% EtOH (70 mL) and 1 N HCl (1.5 mL) was
shaken with H2 and 10% Pd-C for 6 h in a Parr apparatus at
50 lb/in.2 initial pressure. The catalyst was filtered, the
solution was evaporated to dryness, the residue was taken up
in a mixture of CHCl3 and EtOH, and the solution was re-
evaporated. The solid residue was stirred in absolute EtOH,
a small amount of insoluble material was removed, and the
filtrate was concentrated to 10 mL and cooled to 0 °C.
Filtration yielded a solid (61 mg): mp 219-220 °C; Rf 0.39
(silica gel, 95:5 CHCl3-MeOH), ninhydrin positive; IR (KBr)
ν 3400, 3000, 2900, 1770, 1750, 1710, 1590, 1560, 1500, 1460,
1430, 1410 cm-1. Anal. (C12H13ClN2O4) C, H, Cl, N.
3260, 2940, 1720, 1660, 1530, 1430, 1400 cm-1
(C28H34N4O10Cu) C, H, N, Cu.
.
Anal.
Nδ-[3-(Meth oxycar bon yl)ben zoyl]-L-or n ith in e (24). The
Cu complex 23 (3.28 g, 5.06 mmol) was treated with 0.12 M
EDTA (100 mL, pH 8). A white solid precipitated immediately.
The solution was left in the refrigerator overnight, and the
precipitate was filtered, washed with ice-cold H2O (2 × 10 mL),
and dried in vacuo at 70 °C to obtain a white solid (1.93 g,
65%) of sufficient purity for direct use in the next step: mp
225 °C dec; IR (KBr) ν 3250, 2940, 1720, 1630, 1580, 1540,
1
1520, 1430, 1400 cm-1; H NMR (TFA) δ 2.1 (m, 4H, â- and
γ-CH2), 3.6 (m, 2H, δ-CH2), 3.9 (s, 3H, CO2CH3), 4.4 (m, 1H,
R-CH), 7.6-8.4 (m, 4H, isophthaloyl protons). Anal. (C14H18
N2O5) C, H, N.
-
Meth yl Nr-(4-Am in o-4-d eoxy-N10-for m ylp ter oyl)-Nδ-[3-
(m eth oxyca r bon yl)ben zoyl]-L-or n ith in a te (26). A stirred
suspension of 24 (1.93 g, 6.6 mmol) in absolute MeOH (100
mL) at -10 °C was treated dropwise with SOCl2 (7.5 mL) while
keeping the temperature below 0 °C. When addition was
complete, the reaction mixture was left to stir at toom
temperature for 18 h. The solvent was evaporated under
reduced pressure and the residue taken up in a minimum
volume of CHCl3. Excess Et2O was added, and the precipitated
solid was filtered and dried to obtain ester 25 as a white
powder (2.25 g, 100%) of sufficient purity for direct use in the
next step: mp 70 °C; IR (KBr) ν 3400, 2920, 1740, 1720, 1630,
Evaporation of the mother liquor and crystallization of the
residue from i-PrOH/MeOH gave another 95 mg of material
indistinguishable from the analytical sample; total yield 156
mg (66%).
Meth yl 2-L-[(4-Am in o-4-deoxy-N10-for m ylpter oyl)am in o]-
3-p h th a lim id op r op a n oa te (22). A suspension of 20‚HCl
(183 mg, 0.5 mmol) in dry DMF (10 mL) was treated sequen-
tially with Et3N (152 µL, 111 mg, 1.1 mmol) and i-BuOCOCl
(65.3 µL, 68 mg, 0.5 mmol) and stirred for 20 min at room
temperature. Solid 21 (142 mg, 0.5 mmol) was then added,
and after another 20 min of stirring the reaction was quenched
by adding 2 drops of AcOH. The solution was concentrated to
dryness by rotary evaporation, the residue was dissolved in
2:5:5 MeOH-MeCN-CHCl3, and the solution was applied onto
a flash silica gel column (10 g, 22 × 1 cm), which was eluted
with the same solvent mixture. Fractions giving a blue-
fluorescent TLC spot (Rf 0.33, silica gel, 2:5:5 MeOH-MeCN-
CHCl3) were pooled and evaporated, and the residue was
redissolved in CHCl3. The solution was washed twice with
H2O, dried (MgSO4), and evaporated; yield 107 mg (36%). The
product was dissolved in CHCl3-MeOH, and the solution was
added to Et2O. The mixture was cooled, and the precipitate
was filtered and dried to obtain a yellow powder (86 mg): mp
158-160 °C; IR (KBr) ν 3450, 3350, 3200, 1770, 1740, 1720,
1530, 1430 cm-1 1H NMR (CD3OD) δ 1.9 (m, 4H, â- and
;
γ-CH2), 3.5 (m, 2H, δ-CH2), 3.85 (s, 3H, R-CO2CH3), 3.95 (s,
3H, aromatic CO2CH3), 4.2 (m, 1H, R-CH), 8.0 (m, 4H,
isophthaloyl protons).
To a stirred suspension of 25 (183 mg, 0.5 mmol) in dry DMF
(10 mL) were added sequentially Et3N (140 µL, 1.01 mg, 1
mmol) and i-BuOCOCl (65 µL, 69 mg, 0.5 mmol). After 20
min, 21 (190 mg, 0.5 mmol) was added and the reaction
mixture was stirred for 40 min. The solvent was evaporated
under reduced pressure, and the residue was chomatographed
on a silica gel column (15 g, 17 × 2.0 cm), which was packed
and eluted first with 98:2 and then with 95:5 CHCl3-MeOH.
Fractions giving a single TLC spot (Rf 0.18, silica gel, 9:1
CHCl3-MeOH) were pooled and evaporated, and the residue
was taken up in a small volume of 95:5 CHCl3-MeOH. This
solution was then added dropwise to excess Et2O, and the
yellow precipitate was collected and dried (105 mg, 33%): mp
145 °C; IR (KBr) ν 3350, 2950, 1725, 1650, 1600, 1550, 1500,
1710, 1660, 1600, 1560, 1530, 1500, 1440 cm-1
.
Anal.
(C27H23N9O6‚2H2O) C, H, N.
1
Evaporation of the mother liquor yielded a second crop
which was indistinguishable from the analytical sample.
Nr-(4-Am in o-4-d eoxyp ter oyl)-Nâ-h em ip h th a loyl-L-2,3-
d ia m in op r op a n oic Acid (10). A solution of 22 (90 mg, 0.149
mmol) in DMSO (1.0 mL) was treated dropwise with 2.5 N
NaOH (0.4 mL) over a period of 30 s. After 5 min, the solution
was diluted with H2O (5 mL) and adjusted to pH 5.0 with 10%
AcOH to obtain a flocculent precipitate. After 1 h at 0 °C, the
solid was filtered, washed with a large volume of H2O, dried
in a lyophilization apparatus, and finally dried in vacuo over
P2O5 to obtain a yellow powder (70 mg, 80%): mp >225 °C
dec; Rf 0.15 (silica gel, 5:4:1 CHCl3-MeOH-28% NH4OH); IR
1450 cm-1; H NMR (DMSO-d6) δ 1.6 (m, 2H, â-CH2), 1.8 (m,
2H, γ-CH2), 3.3 (m, 2H, δ-CH2), 3.60 (s, 3H, R-CO2CH3), 3.86
(s, 3H, aromatic CO2CH3), 4.42 (m, 1H, R-CH), 5.21 (s, 2H,
9-CH2), 7.56 (m, 2H, 3′- and 5′-H), 7.61 (m, 1H, isophthaloyl
CONH), 7.85 (m, 2H, 2′- and 6′-H), 7.58 (m, 1H, proton meta
to isophthaloyl CO2CH3), 8.05 (s, 2H, proton para to iso-
phthaloyl CONH and proton ortho to isophthaloyl CONH and
para to isophthaloyl CO2CH3), 8.40 (s, 1H, proton ortho to
isophthaloyl COOH and CONH), 8.66 (s, 1H, 7-H), 8.71 (m,
1H, benzoyl CONH), 8.78 (s, 1H, CHO). Anal. (C30H31N9O7‚
H2O) C, H, N.
Nr-(4-Am in o-4-d eoxyp ter oyl)-Nδ-(isop h th a loyl)-L-or n i-
th in e (11). NaOH (2.5 N, 0.4 mL) was added to a stirred
solution of 26 (95 mg, 0.15 mmol) in DMSO (1 mL) at room
temperature, and after 5 min the reaction was quenched by
diluting it with H2O (10 mL) and adjusting the pH to 4.2 with
1 N HCl. The yellow precipitate was filtered immediately,
washed with H2O (70 mL), and dried (78 mg, 81%): mp 220
(KBr) ν 3340 br, 1635, 1600, 1510, 1450 cm-1
(C25H23N9O6‚2.7H2O) C, H, N.
.
Anal.
Bis[Nδ-[3-(m eth oxyca r bon yl)ben zoyl]-L-or n ith in a to]-
cop p er (II) (23). SOCl2 (15 mL) and dry DMF (2 drops) were
added to monomethyl isophthalate (5 g, 17.7 mmol), and the
reaction mixture was refluxed at 80 °C for 6 h. The SOCl2

Analogues of Substituted L-Ornithine
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3 295
°C dec; TLC Rf 0.19 (1:4:5 28% NH4OH-MeOH-CHCl3);
HPLC 33.5 min (C18 silica gel; 0.1 M NH4OAc, pH 5.5, with
2.0-35% MeCN linear gradient over 90 min; 1 mL/min); IR
0.66 g (78%); mp 230 °C dec; IR (KBr) ν 3315, 2960, 1750, 1725,
1650, 1500, 1440 cm-1; 1H NMR (CD3OD) δ 1.8 (m, 2H, â-CH2),
2.0 (m, 2H, γ-CH2), 3.5 (m, 2H, δ-CH2), 3.84 (s, 3H, R-CO2-
CH3), 3.93 (s, 3H, aromatic CO2CH3), 4.12 (m, 1H, R-CH), 7.9
(dd, 4H, terephthaloyl protons). Anal. (C15H20N2O5‚HCl.
0.1Et2O) C, H, Cl, N.
(KBr) ν 3400, 2900, 1640, 1600, 1500, 1440 cm-1 1H NMR
;
(DMSO-d6) δ 1.6 (m, 2H, â-CH2), 1.8 (m, 2H, γ-CH2), 3.4 (m,
2H, δ-CH2), 4.3 (m, 1H, R-CH), 4.5 (m, 2H, 9-CH2), 6.6 (br s,
2H, NH2), 6.7 (m, 2H, 3′- and 5′-H), 6.80 (m, 1H, 10-H), 7.7
(m, 2H, 2′- and 6′-H), 7.56 (m, 1H, proton meta to isophthaloyl
COOH), 8.05 (m, 1H, proton para to isophthaloyl COOH), 8.15
(m, 1H, proton ortho to isophthaloyl COOH and para to
isophthaloyl CONH), 8.40 (s, 1H, proton ortho to isophthaloyl
COOH and CONH), 8.68 (s, 1H, 7-H), 8.05 (m, 1H, benzoyl
Meth yl Nr-(4-Am in o-4-d eoxy-N10-for m ylp ter oyl)-Nδ-[4-
(m eth oxyca r bon yl)ben zoyl]-L-or n ith in a te (30). i-BuO-
COCl (65 µL, 68 mg, 0.5 mmol) and Et3N (140 µL, 101 mg, 1
mmol) were added to a stirred suspension of 21 (183 mg, 0.5
mmol) in dry DMF (10 mL). After 20 min, 28‚HCl (190 mg,
0.5 mmol) was added, followed 20 min later by i-BuOCOCl
(32 µL, 34 mg, 0.25 mmol) and Et3N (35 µL, 25 mg, 0.25 mmol).
After another 20 min, additional portions of 27‚HCl (95 mg,
0.25 mmol) and Et3N (35 µL, 25 mg, 0.25 mmol) were added,
and the mixture was stirred for another 20 min. The solvent
was evaporated under reduced pressure, and the residue was
dissolved in CHCl3 (100 mL). The solution was washed with
distilled H2O (2 × 50 mL), dried (MgSO4), and applied onto a
column of silica gel (20 g, 27 × 2.0 cm) which was packed with
98:2 CHCl3-MeOH and eluted with the same mixture, fol-
lowed by 95:5 and finally 9:1 CHCl3-MeOH. Fractions giving
a single TLC spot (Rf 0.18, silica gel, 9:1 CHCl3-MeOH) were
pooled and evaporated. The residue was redissolved in a small
amount of 4:1 CHCl3-MeOH and the solution added dropwise
to an excess of Et2O. The yellow precipitate was collected and
dried: yield 120 mg (38%); mp 130 °C dec; IR (KBr) ν 3330,
2950, 1725, 1640, 1610, 1565, 1540, 1505, 1450 cm-1; 1H NMR
(DMSO-d6) δ 1.6 (m, 2H, â-CH2), 1.8 (m, 2H, γ-CH2), 3.25 (m,
2H, δ-CH2), 3.59 (s, 3H, R-CO2CH3), 3.85 (s, 3H, aromatic CO2-
CH3), 4.4 (m, 1H, R-CH), 5.19 (s, 2H, 9-CH2), 7.57 (m, 2H, 3′-
and 5′-H), 7.9-8.0 (m, 4H, terephthaloyl protons), 8.60 (m, 1H,
terephthaloyl CONH), 8.63 (m, 2H, 2′- and 6′-H), 8.63 (s, 1H,
7-H), 8.70 (m, 1H, benzoyl CONH), 8.77 (s, 1H, CHO). Anal.
(C30H31N9O7‚H2O).
CONH), 8.65 (s, 1H, isophthaloyl CONH).
Anal.
(C27H27N9O6‚1.1H2O) C, H, N.
Met h yl
Nr-(Ben zyloxyca r b on yl)-Nδ-[4-(m et h oxy-
ca r bon yl)ben zoyl]-L-or n ith in a te (29). A stirred suspension
of NR-(benzyloxycarbonyl)-L-ornithine (3 g, 11.3 mmol) in
absolute MeOH (100 mL) was treated dropwise with SOCl2
(10 mL) while the temperature was kept below 0 °C. When
addition was complete the cooling bath was removed and the
reaction mixture left to stir for 18 h. The solvent was
evaporated under reduced pressure, and the residue was
recrystallized from a mixture of MeOH and EtOAc to obtain
methyl NR-(benzyloxycarbonyl)-L-ornithinate hydrochloride
(28‚HCl) as white solid (3.24 g, 96%) which was of sufficient
purity for direct use in the next step: mp 139-140 °C; IR (KBr)
ν 3340, 2960, 2890, 1735, 1685, 1530, 1470 cm-1; 1H NMR (CD3-
OD) δ 1.85 (m, 4H, â- and γ-CH2), 3.00 (m, 2H, δ-CH2), 3.7 (s,
3H, CO2CH3), 4.3 (m, 1H, R-CH), 5.1 (s, 2H, benzylic CH2),
7.3 (s, 5H, aryl protons).
A solution of monomethyl terephthalate (0.6 g, 3.3 mmol)
in dry DMF (50 mL) was treated with i-BuOCOCl (428 µL,
452 mg, 3.3 mmol) and Et3N (459 µL, 330 mg, 3.3 mmol). After
20 min, 28‚HCl (1.04 g, 3.3 mmol) was added, followed by a
second portion of Et3N (459 µL, 330 mg, 3.3 mmol). After 40
min of continued stirring, additional portions of i-BuOCOCl
(214 µL, 226 mg, 1.65 mmol) and Et3N (225 µL, 164 mg 1.62
mmol) were added, and after another 40-min period the solvent
was evaporated under reduced pressure. The residue was
taken up in CHCl3 (120 mL), and the solution was washed
with distilled H2O (2 × 70 mL), dried (MgSO4), and evaporated
under reduced pressure. The residue was chromatographed
on a silica gel column (60 g, 27 × 3 cm) which was packed and
eluted with 98:2 CHCl3-MeOH. Fractions containing a major
TLC spot with Rf 0.56 (silica gel, 95:5 CHCl3-MeOH), along
with several lesser spots, were pooled and chromatographed
on a silica gel column (40 g, 27 × 2.5 cm) which was packed
and eluted with CHCl3 followed by 99.5:0.5 CHCl3-MeOH.
Fractions giving a single TLC spot (Rf 0.56, 95:5 CHCl3-
MeOH) were pooled and evaporated to a white solid (1.21 g,
83%): mp 130 °C; IR (KBr) ν 3330, 2960, 1750, 1725, 1695,
1635, 1540, 1440 cm-1; 1H NMR (CDCl3) δ 1.8 (m, 4H, â- and
γ-CH2), 3.5 (m, 2H, δ-CH2), 3.75 (s, 3H, R-CO2CH3), 3.95 (s,
3H, aromatic CO2CH3), 4.4 (m 1H, R-CH), 5.1 (s, 2H, benzylic
CH2), 7.3 (s, 5H, phenyl protons), 7.8-8.2 (dd, 4H, terephtha-
loyl protons). Anal. (C23H26N2O7) C, H, N.
Met h yl Nδ-[4-(Met h oxyca r b on yl)b en zoyl]-L-or n it h i-
n a te Hyd r och lor id e (27‚HCl). A solution of 29 (1.21 g, 2.75
mmol) in a mixture of MeOH (30 mL) and glacial AcOH (9
mL) was shaken under H2 (2 atm) in a Parr apparatus in the
presence of 10% Pd-C (0.17 g) for 14 h. The reaction mixture
was filtered through Celite, and the filter pad was washed with
MeOH. The combined filtrate and wash were evaporated
under reduced pressure, and the residue was treated with CH2-
Cl2 (30 mL) and H2O (60 mL). The mixture was stirred
vigorously while solid NaHCO3 was added in small portions
until the pH was 8. After CH2Cl2 (30 mL) and H2O (50 mL)
were added, the layers were separated. The aqueous layer was
extracted once more with CH2Cl2 (30 mL), and the combined
organic layers were dried (MgSO4) and evaporated. The
residue was dissolved in MeOH (10 mL), and 6 N HCl (2 mL)
was added with constant stirring. Excess MeOH and HCl were
removed by rotary evaporation, and the solid residue was
redissolved in a minimum amount of MeOH. The solution was
added to an excess of Et2O with continuous stirring, and the
precipitate was collected, washed with Et2O, and dried: yield
Nr-(4-Am in o-4-d eoxyp ter oyl)-Nδ-ter ep h th a loyl-L-or n i-
th in e (12). A solution of 30 (72 mg, 0.11 mmol) in DMSO (1
mL) was treated with 2.5 N NaOH (0.4 mL). The mixture was
stirred for 5 min, diluted with H2O (10 mL), adjusted to pH
4.2 with 1 N HCl, and filtered immediately. The yellow solid
was washed with H2O (50 mL) and dried: yield 63 mg (100%);
mp 210 °C dec; TLC Rf 0.45 (silica gel, 5:4:1 CHCl3-MeOH-
28% NH4OH); HPLC 33.5 min (C18 silica gel; 0.1 M NH4OAc,
pH 5.5, with 2.0-35% MeCN linear gradient over 90 min; 1
mL/min); IR (KBr) ν 3400, 2910, 1640, 1610, 1550, 1510, 1450,
1
1400 cm-1; H NMR (DMSO-d6) δ 1.6 (m, 2H, â-CH2), 1.8 (m,
2H, γ-CH2), 3.6 (m, 2H, δ-CH2), 4.33 (m, 1H, R-CH), 4.47 (s,
2H, 9-CH2), 6.58 (br s, 2H, NH2), 6.72 (m, 2H, 3′- and 5′-H),
6.80 (m, 1H, 10-H), 7.70 (m, 2H, 2′- and 6′-H), 7.89-7.97 (m,
4H, terephthaloyl protons), 8.14 (m, 1H, benzoyl CONH), 8.60
(m, 1H, terephthaloyl CONH), 8.67 (s, 1H, 7-H). Anal.
(C27H27N9O6‚2H2O) C, H, N.
4,5-Dich lor op h th a lim id e (31). A stirred solution of 4,5-
dichlorophthalic acid (2.0 g, 8.5 mmol) in absolute MeOH (100
mL) was cooled to -10 °C, and SOCl2 (19 mL) was added
dropwise while the temperature was kept below 0 °C. After
18 h of stirring, the solvent was evaporated under reduced
pressure and MeOH was added several times and re-
evaporated to obtain the dimethyl ester, which was used
directly in the next step without purification (2.2 g, 100%):
1H NMR (CD3OD) δ 4.0 (s, 6H, OMe), 7.9 (s, 2H, aromatic).
Urea (510 mg, 8.5 mmol) was added to a solution of NaOMe
prepared by allowing metallic Na (200 mg, 8.5 mmol) to
dissolve in absolute MeOH (50 mL). To this solution was
added a solution of dimethyl 4,5-dichlorophthalate in a mini-
mum volume of MeOH. The reaction mixture was refluxed
for 5 h, and the solvent was evaporated under reduced
pressure. The solid residue was triturated with H2O, and the
insoluble portion was filtered off and resuspended in H2O to
which NH4OH was added dropwise to bring the pH to 10. The
material which remained undissolved was again filtered off,
and the filtrate was acidified to pH 2 with 1 N HCl. The white
solid which precipitated was collected, washed with H2O, and
dried to obtain 31 (0.8 g, 44%): IR (KBr) ν 3320, 3280, 3220,
3080, 3020, 1735, 1725, 1710, 1600 cm-1; 1H NMR (CD3OD) δ

296 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3
Rosowsky et al.
8.1 (s, aromatic protons). Anal. (C8H3Cl2NO2) C, H, N; Cl:
calcd, 32.82; found, 32.34.
1H, benzoyl CONH), 8.67 (d, J ) 2 Hz, 2H, aromatic protons
on phthalimide moiety), 8.70 (s, 1H, C7-H), 8.80 (CHdO).
A stirred solution of 36 (30 mg, 0.045 mmol) in DMSO 0.25
mL) was treated with 2 N NaOH (0.15 mL), and after 5 min
the mixture was diluted with H2O (3.5 mL). The pH was
adjusted to 4.2 with 10% AcOH, and the precipitate was
collected, washed with H2O (50 mL), and dried to obtain a
yellow powder (19 mg, 67%); mp 215 °C, dec >170 °C; TLC Rf
0.71 (silica gel, 5:4:1 CHCl3-MeOH-28% NH4OH); HPLC 14
min (C18 silica gel, 15% MeCN in 0.1 M NH4OAc, pH 7.5, 1
4,5-Dich lor o-N-ca r beth oxyp h th a lim id e (32). A stirred
solution of 31 (665 mg, 3.08 mmol) and Et3N (477 µL, 3.42
mmol) in dry DMF (12 mL) was cooled to 5-10 °C and treated
with EtOCOCl (327 µL, 3.42 mmol). The reaction mixture was
stired for 1.5 h and poured with stirring into H2O (200 mL).
The precipitate was collected, washed with H2O (100 mL), and
dried to obtain 32 as a white solid (0.8 g, 90%): mp 138 °C
(EtOH); IR (KBr) ν 3400, 3330, 3080, 3020, 2980, 2925, 1805,
1760, 1720, 1600, 1470, 1440 cm-1; 1H NMR (DMSO-d6) δ 1.4
(t, 3H, CH2CH3), 4.5 (q, 2H, CH2CH3), 8.3 (s, 2H, aromatic
protons). Anal. (C11H7Cl2NO4) C, H, N; Cl: calcd, 24.61;
found, 24.19.
mL/min); IR (KBr) ν 3400, 2900, 1640, 1600, 1510, 1450 cm-1
;
1H NMR (DMSO-d6) δ 1.54-1.61 (m, 2H, â-CH2), 1.73-1.86
(m, 2H, γ-CH2), 3.19 (m, 2H, δ-CH2), 4.32 (m, 1H, R-CH), 4.48
(s, 2H, 9-CH2), 6.71 (m, 2H, 3′- and 5′-H), 6.81 (m, 1H, N10-H),
7.66 (s, 1H, aromatic proton ortho to CONH), 7.71 (m, 2H, 2′-
and 6′-H), 7.89 (s, 1H, aromatic proton ortho to COOH), 8.13
(s, 1H, R-CONH), 8.50 (m, 1H, phthaloyl CONH), 8.68 (s, 1H,
C7-H). Anal. (C27H25Cl2N9O6‚0.5AcOH‚H2O) C, N, Cl (H:
calcd 4.23, found 3.72).
Bis[L-2-a m in o-5-(4,5-d ich lor op h th a lim id o)p en ta n a to]-
Cu (II) (33). A solution of L-ornithine hydrochloride (454 mg,
269 mmol) in H2O (10 mL) containing NaOH (215 mg, 5.38
mmol) was treated with Cu(OAc)2‚H2O (269 mg, 1.34 mmol)
in H2O (15 mL). To the mixture were then added NaHCO3
(226 mg, 2.69 mmol) and 32 (775 mg, 2.69 mmol). The reaction
mixture was stirred for 2 h and filtered, and the solid was
washed successively with H2O, EtOH, CHCl3, and Et2O and
then dried to obtain 33 as a blue powder which was used
directly for the next reaction (690 mg, 71%): mp 205 °C with
darkening; IR (KBr) ν 3450, 3250, 2920, 1770, 1710, 1620,
Meth yl Nr-(Ben zyloxyca r bon yl)-Nδ-[(2-ca r boxyp h en -
yl)a cetyl]-L-or n ith in e (37). To a solution of 28 (1 g, 3.15
mmol) in absolute EtOH (100 mL) were added i-Pr2NEt (1.1
mL, 6.3 mmol) and homophthalic anhydride (0.510 g, 3.15
mmol), and the mixture was stirred under reflux for 6 h. The
solvent was evaporated under reduced pressure, and the
residue was dissolved in CH2Cl2 (100 mL). The solution was
washed with 0.1 N HCl (30 mL) and distilled H2O (2 × 30 mL),
dried over MgSO4, and evaporated. The residue was chro-
matographed on a silica gel column (60 g, 27 × 3 cm), which
was packed and eluted with 98:2 CH2Cl2-MeOH. Fractions
showing a single TLC spot (Rf 0.30, silica gel, 95:5 CH2Cl2-
MeOH) were pooled and evaporated to obtain a white solid
(0.75 g, 54%): IR (KBr) ν 3300, 3060, 2950, 1740, 1690, 1650,
1600, 1540, 1450, 1430, 1350, 1280, 1130, 1075, 1010, 720, 700
1430, 1400 cm-1
.
L-2-Am in o-5-(4,5-d ich lor op h th a lim id o)p en ta n oic Acid ,
Hyd r och lor id e Sa lt (34‚HCl). The finely pulverized Cu(II)
complex 31 (690 mg, 0.096 mmol) was treated with 6 N HCl
for 1 h. The solid was filtered, washed with 6 N HCl, air-
dried overnight, and redissolved in hot MeOH (7 mL). The
solution was added to EtOAc (75 mL), and the mixture left in
the refrigerator for 2 h. The white precipitate was collected,
washed with EtOAc, and dried: yield 435 mg (62%); mp 230
°C dec; IR (KBr) ν 3550, 3350, 2900, 1770, 1730, 1700, 1600,
1500, 1400 cm-1; 1H NMR (DMSO-d6) δ 1.65-1.80 (m, 4H, â-
and γ-CH2), 3.58 (m, 2H, δ-CH2), 4.0 (m, 1H, R- CH), 8.19 (s,
aromatic proton), 8.39 (m, 3H, NH2 and aromatic proton). Anal.
(C13H12Cl2N2O4‚HCl‚0.5MeOH‚0.5H2O) C, H, Cl, N.
1
cm-1; H NMR (CDCl3) δ 1.8 (m, 4H, â- and γ-CH2), 3.22 (m,
2H, δ-CH2), 3.61 (s, 3H, R-CO2CH3), 3.85 (s, 2H, homophthaloyl
CH2CONH), 4.05 (m, 1H, R-CH), 5.10 (s, 2H, PhCH2O), 7.2-
7.4 (m, 7H, C6H5CH2O and 2 aromatic protons meta to COOH),
7.5 (m, 1H, aromatic proton para to COOH), 8.0 (m, 1H,
aromatic proton ortho to COOH). Anal. (C23H26N2O7) C, H,
N.
Met h yl
L-2-Am in o-5-(4,5-d ich lor op h t h a lim id o)-
p en ta n oa te, Hyd r och lor id e Sa lt (35‚HCl). A stirred solu-
tion of 34‚HCl (0.42 g, 1.14 mmol) in MeOH (50 mL) at -10
°C was treated dropwise with SOCl2 (3.2 mL) while the
temperature was kept below 0 °C. After addition was complete
the mixture was left at room temperature for 18 h. The solvent
was evaporated under reduced pressure, the residue was
redissolved in a minimum of MeOH, the solution was added
to EtOAc (100 mL), and the mixture was kept at 0 °C until a
solid precipitated. The white solid was collected, washed with
EtOAc, and dried: yield 387 mg (89%); mp 198-199 °C dec;
IR (KBr) ν 3450, 3000, 1770, 1750, 1710, 1580, 1500, 1450,
1440, 1400 cm-1; 1H NMR (DMSO-d6) δ 1.64-1.77 (m, 4H, â-
and γ-CH2), 3.59 (m, 2H, δ-CH2), 3.70 (s, 3H, CO2CH3), 4.05
(m, 1H, R-CH), 8.20 (s, 1H, aromatic proton), 8.50 (m, 3H, NH2
and aromatic proton). Anal. (C14H14Cl2N2O4‚HCl.0.1MeOH)
C, H, Cl, N.
Meth yl Nr-(Ben zyloxyca r bon yl) Nδ-[[2-(m eth oxyca r -
bon yl)p h en yl]a cetyl]-L-or n ith in e (38). Thionyl chloride
(5.2 mL) was added dropwise to a stirred suspension of 37 (0.75
g, 1.67 mmol) in MeOH (100 mL) while the temperature was
kept below 0 °C. The reaction mixture was left to stir at room
temperature for 18 h and was then evaporated to dryness
under reduced pressure. Crystallization from MeOH and
EtOAc and drying afforded a white solid (0.73 g, 95%): mp
85-86 °C; IR (KBr) ν 3300, 3060, 2950, 1740 (aliphatic ester),
1710 (aromatic ester), 1690, 1650, 1600, 1540, 1450, 1430,
1
1350, 1280, 1130, 1075, 1010, 720, 700 cm-1; H NMR (CD3-
OD) δ 1.8 (m, 4H, â- and γ-CH2), 3.09 (m, 2H, δ-CH2), 3.61 (s,
3H, CH2CO2CH3), 3.76 (s, 3H, aromatic CO2CH3), 3.81 (s, 2H,
CH2CONH), 4.08 (m, 1H, R-CH), 5.00 (s, 2H, PhCH2OCO),
7.20-7.34 (m, 7H, C6H5CH2O and 2 aromatic protons meta to
CO2Me), 7.43 (m, 1H, aromatic proton para to CO2Me), 7.48
(m, 1H, aromatic proton ortho to CO2Me). Anal. (C24H28N2O7)
C, H, N.
Nr-(4-Am in o-4-d eoxyp ter oyl)-Nδ-(4,5-d ich lor oh em ip h -
th a loyl)-L-or n ith in e (13). To a stirred suspension of 21 (172
mg, 0.47 mmol) in dry DMF (20 mL) were added sequentially
Et3N (130 µL, 0.94 mmol) and i-BuOCOCl (61 µL, 0.47 mmol).
After 30 min, 35‚HCl (180 mg, 0.47 mmol) was added, and
the mixture was stirred for 1 h. The solvent was evaporated
under reduced pressure, and the residue was chromatographed
on a silica gel column (20 g, 17 × 2 cm) with 98:2, 96:4, and
95:5 CHCl3-MeOH as the eluents. Fractions giving a single
TLC spot with Rf 0.49 (silica gel, 9:1 CHCl3-MeOH) were
pooled and evaporated, the residue was redissolved in a small
volume of 95:5 CHCl3-MeOH, and the solution was added
dropwise to excess Et2O. The precipitate was collected by
centrifugation and dried to obtain 36 as a yellow powder which
was of sufficient purity for use in the next step (65 mg, 21%):
mp 185 °C dec; IR (KBr) ν 3400, 2900, 1770, 1710, 1640, 1600,
1500, 1470, 1440, 1400 cm-1; 1H NMR (DMSO-d6) δ 1.65-1.80
(m, 4H, â- and γ-CH2), 3.56 (m, 5H, δ-CH2 and R-CO2CH3),
4.40 (m, 1H, R-CH), 5.22 (s, 2H, 9-CH2), 7.0 (br s, 2H, NH2),
7.55 (m, 2H, 3′- and 5′-H), 7.82 (m, 2H, 2′- and 6′-H), 8.14 (s,
Meth yl Nδ-[[2-(m eth oxyca r bon yl)p h en yl]a cetyl]-L-or -
n ith in e Hyd r och lor id e (39‚HCl). A solution of 38 (0.4 g,
1.47 mmol) in a mixture of EtOH (30 mL), glacial AcOH (6.5
mL), and 6 N HCl (0.33 mL) was shaken under 30 psi of H2 in
a Parr apparatus in the presence of 10% Pd-C (0.13 g) for 14
h. The catalyst was removed by filtration through Celite, and
the filter pad was washed with MeOH. The combined filtrate
and wash were evaporated under reduced pressure. Rerys-
tallization from MeOH and i-PrOH gave an off-white solid
(0.31 g, 44%): mp 132 °C; IR (KBr) ν 3340, 2960, 1745, 1730,
1650, 1580, 1540, 1500, 1450, 1435, 1400, 1380, 1350, 1300,
1280, 1250, 1220, 1140, 1085, 950, 800, 750, 735, 690, 630
1
cm-1; H NMR (DMSO-d6) δ 1.6-1.8 (m, 4H, â- and γ-CH2),
3.02 (m, 1H, δ-CH2), 3.72 (s, 3H, R-CO2CH3), 3.75 (s, 3H,
aromatic CO2CH3), 3.78 (s, 2H, CH2CONH), 4.01 (m, 1H,
R-CH), 7.29 (m, 1H, aromatic proton ortho to CO2Me and meta
to CH2CONH), 7.35 (m, 1H, aromatic proton meta to CO2Me

Analogues of Substituted L-Ornithine
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3 297
and para to CH2CONH), 7.48 (m, 1H, aromatic proton para to
CO2Me), 7.79 (m, 1H, aromatic proton ortho to CO2Me), 8.0
(m, 1H, CONH), 8.48 (broad s, NH2). Anal. (C16H22ClN2O5‚
HCl‚0.1MeOH) C, H, Cl, N.
3105, 2960, 1740, 1640, 1600, 1540, 1300, 1260, 1220, 1170,
1
1100, 1010, 840, 720, 690 cm-1; H NMR (DMSO-d6) δ 1.6-
1.8 (m, 4H, â- and γ-CH2), 3.17 (m, 2H, δ-CH2), 3.54 (s, 3H,
CH2CO2CH3), 3.73 (s, 3H, R-CO2CH3), 3.83 (s, 2H, CH2CO2-
Me), 4.05 (m, 1H, R-CH), 7.28 (m, 2H, aromatic proton meta
to CH2NHCO and ortho to CH2CO2Me), 7.32 (m, 1H, aromatic
proton meta to CH2NHCO and para to CH2CO2Me), 7.39 (m,
1H, aromatic proton para to CH2CONH), 7.44 (m 1H, aromatic
proton ortho to CH2CONH). Anal. (C22H22N2O5‚HCl) C, H,
Cl, N.
Meth yl Nr-(4-Am in o-4-d eoxy-N10-for m ylp ter oyl)-Nδ-[2-
[(m eth oxycar bon yl)m eth yl]ben zoyl]-L-or n ith in e (44). This
compound was made from 43 (170 mg, 0.66 mmol) and 21 (173
mg, 0.46 mmol) as in the synthesis of 40 to obtain a yellow
powder (131 mg, 44%): mp 145 °C, darkening above 125 °C;
TLC Rf 0.26 (silica gel, 9:1 CHCl3-MeOH; IR (KBr) ν 3320,
2975, 1740, 1640, 1600, 1550, 1350, 1300, 1220, 1020, 970, 850,
770 cm-1; 1H NMR (DMSO-d6, 500 MHz) δ 1.6-1.8 (m, 4H, â-
and γ-CH2), 3.28 (m, 2H, δ-CH2), 3.48 (s, 3H, CH2CO2CH3),
3.61 (s, 3H, R-CO2CH3), 3.82 (s, 2H, CH2CO2Me), 4.41 (m, 1H,
R-CH), 5.24 (s, 2H, 9-CH2), 7.27 (m, 1H, aromatic proton meta
to CH2CONH and ortho to CH2CO2Me), 7.32 (m, 1H, aromatic
proton meta to CH2CONH and para to CH2CO2Me), 7.38 (m,
1H, aromatic proton para to CH2CONH), 7.41 (m, 1H, aromatic
proton ortho to CH2CONH), 7.54 (m, 2H, 3′- and 5′-H), 7.87
(m, 2H, 2′- and 6′-H), 8.32 (m, 1H, CONH), 8.72 (m, 1H,
CONH), 8.79 (s, 1H, CHdO), 8.70 (s, 1H, C7-H). Anal.
(C31H33N9O7‚H2O) C, H, N.
Meth yl Nr-(4-Am in o-4-d eoxy-N10-for m ylp ter oyl)-Nδ-[[2-
(m eth oxyca r bon yl)p h en yl]a cetyl]-L-or n ith in e (40). To a
stirred suspension of 21 (87 mg, 0.23 mmol) in dry DMF (10
mL) at room temperature were added sequentially Et3N (65
µL, 0.46 mmol) and i-BuOCOCl (30 µL, 0.23 mmol). After 30
min, 39 (85 mg, 0.23 mmol) was added, and the mixture was
left to stir for 1 h. The solvent was evaporated under reduced
pressure, and the residue was dissolved in 9:1 CHCl3-MeOH
and chromatographed on a silica gel column (10 g, 17 × 2 cm)
with 98:2, 95:5, and 92:8 CHCl3-MeOH as the eluents.
Fractions showing a single TLC spot (Rf 0.26, silica gel, 9:1
CHCl3-MeOH) were pooled and evaporated. The residue was
taken up in a small volume of 9:2 CHCl3-MeOH, the solution
was added dropwise to excess Et2O, and the precipitate was
collected and dried to obtain a yellow powder (60 mg, 40%):
mp >130 °C dec; IR (KBr) ν 3320, 2975, 1735, 1715, 1640,
1600, 1550, 1500, 1450, 1350, 1270, 1220, 1130, 1020, 970
1
cm-1; H NMR (DMSO-d6) δ 1.6-1.8 (m, 4H, â- and γ-CH2),
3.02 (m, 2H, δ-CH2), 3.60 (s, 3H, R-CO2CH3), 3.70 (s, 3H,
aromatic CO2CH3), 3.76 (s, 2H, CH2CONH), 4.38 (m, 1H,
R-CH), 5.24 (s, 2H, 9-CH2), 7.28 (m, 1H, aromatic proton ortho
to CH2CONH and meta to CO2Me), 7.33 (m, 1H, aromatic
proton para to CH2CONH and meta to CO2Me), 7.46 (m, 1H,
aromatic proton para to CO2Me), 7.75 (m, 1H, aromatic proton
ortho to CO2Me), 7.58 (m, 2H, 3′- and 5′-H), 7.86 (m, 2H, 2′-
and 6′-H), 8.70 (m, 2H, C7-H and CONH), 8.78 (s, 1H, CHdO).
Anal. (C31H33N9O7‚H2O) C, H, N.
L-2-[N r-(4-Am in o-4-d e o xyp t e r oy l)a m in o ]-5-(h o m o -
p h th a lim id o)p en ta n oic Acid (45). P r oced u r e A. A solu-
tion of 40 (40 mg, 0.06 mmol) in DMSO (0.5 mL) was treated
for 5 min with 2.5 N NaOH (0.25 mL), diluted with H2O (5
mL), and acidified to pH 4.5 with 10% AcOH. The precipitate
was filtered, washed with H2O (50 mL), and dried to a yellow
powder (30 mg, 86%): mp 185 °C dec; IR (KBr) ν 3350, 2950,
1715, 1660, 1610, 1510, 1540, 1400, 1260, 1335, 1260, 1200,
Meth yl 2-Ca r boxyp h en yla ceta te (41). Sodium metal
(0.142 g, 0.16 mmol) cut into small pieces was allowed to
dissolve in absolute MeOH (50 mL), and homophthalic anhy-
dride (1 g, 0.16 mmol) was added. The reaction mixture was
stirred overnight at room temperature and then neutralized
with 6 N HCl (1.02 mL, 6.16 mmol). The solvent was
evaporated under reduced pressure, and the residue was taken
up in EtOAc. An insoluble material was removed by filtration,
and the filtrate was concentrated to dryness. The product was
chromatographed twice on silica gel with 98:2 CH2Cl2-MeOH
as the eluent. Fractions showing a single TLC spot (Rf 0.36,
silica gel, 9:1 CH2Cl2-MeOH) were combined and evaporated
to obtain a white solid (0.24 g, 80%): mp 87-88 °C; IR (KBr) ν
3420, 3000, 2940, 1725, 1680, 1600, 1580, 1490, 1445, 1400
cm-1; 1H NMR (DMSO-d6) δ 3.55 (s, 3H, CO2CH3), 3.98 (s, 2H,
CH2CO2Me), 7.32 (m, 1H, 6-H), 7.39 (m, 1H, 4-H), 7.39 (m,
1H, 5-H), 7.90 (m, 1H, 3-H). Anal. (C10H10O4) C, H.
1
1150, 850, 770, 750, 700, 650 cm-1; H NMR (DMSO-d6, 500
MHz) δ 1.6-1.8 (m, 4H, â- and γ-CH2), 3.86 (m, 2H, δ-CH2),
4.11 (m, 1H enolic OH of 45A, disappearing on addition of
D2O), 4.30 (m, 1H, R-CH), 4.47 (s, 2H, 9-CH2), 6.60 (br s, 2H,
NH2), 6.70 (m, 2H, 3′- and 5′-H), 6.80 (m, 1H, N10-H), 7.36 (m,
1H, 5′′-H), 7.44 (m, 1H, 7′′-H), 7.61 (m, 1H, 4′′-H of 45A), 7.65
(m, 2H, 2′- and 6′-H), 8.01 (m, 1H, 6′′-H), 8.11 (m, 1H, 8′′-H),
.
8.68 (s, 1H, C7-H). Anal. (C28H29N9O6 0.6H2O) C, H, N.
P r oced u r e B. A sample of 44 (40 mg, 0.06 mmol) was
treated with NaOH in DMSO exactly as in the preceding
experiment to obtain a yellow powder (29 mg, 82%); mp 185
°C dec. The IR and 1H NMR spectra of this material were
Meth yl Nr-(Ben zyloxyca r bon yl)-Nδ-[2-[(m eth oxyca r -
bon yl)m eth yl]ben zoyl]-L-or n ith in e (42). i-BuOCOCl (516
µL, 3.98 mmol) and Et3N (1.11 mL, 7.96 mmol) were added to
a solution of 41 (0.77 g, 3.98 mmol) in dry DMF (50 mL). After
30 min, 28‚HCl (1.2 g, 3.98 mmol) was added, and the reaction
mixture was stirred for 1 h. The solvent was evaporated under
reduced pressure, and the residue was dissolved in CH2Cl2 (100
mL). The solution was washed with H2O (2 × 50 mL), dried
(MgSO4), and evaporated. The product was purified by flash
chromatography on silica gel, using 70:30 CH2Cl2-EtOAc as
the eluent. Fractions containing a major TLC spot (Rf 0.32,
silica gel, 75:25 CH2Cl2-EtOAc) along with a slower-moving
impurity were rechromatographed on another flash column
with 75:25 CH2Cl2-EtOAc as the eluent, and TLC-homoge-
neous fractions were pooled and evaporated to obtain a white
solid (0.86 g, 48%): mp 100-101 °C; IR (KBr) ν 3380, 2950,
1735, 1690, 1530, 1465, 1430, 1290, 1255, 1220, 1130, 1060,
identical with those of the product obtained from 40.
A
microsample of the product via either route, dissolved in dilute
NaOH and analyzed immediately by HPLC (C18 silica gel, 8%
MeCN in 0.1 M NH4OAc, 1 mL/min), showed a small peak at
11 min, coeluting with authentic NR-(4-amino-4-deoxypteroyl)-
L-ornithine, and two larger peaks of ca. equal area at 19 and
21 min, which are presumed to be 14 and 15.
L-2-[Nr-(Ben zyloxyca r b on yl)a m in o]-5-(1,8-n a p h t h a l-
im id o)p en ta n oic Acid (46). N,N-Diisopropylethylamine
(544 mL, 3.12 mmol) and 1,8-naphthalic anhydride (0.5 g, 208
mmol) were added to a suspension of NR-(benzyloxycarbonyl)-
L-ornithine (0.83 g, 3.12 mmol) in dry DMF, and the reaction
mixture was heated for 5 h at an internal temperature of 100
°C (all the solid dissolved after 30 min). The solvent was then
evaporated at reduced pressure, and the residue was dissolved
in CH2Cl2 (100 mL). The solution was washed with 1 N HCl
(20 mL), rinsed with H2O (50 mL), dried (MgSO4), and
evaporated. The residue was applied onto a silica gel column
(50 g, 27 × 3 cm), packed with CHCl3, and eluted sequentially
with CHCl3 and 97:3 CHCl3-MeOH. Fractions containing a
major spot at Rf 0.43 (silica gel, 9:1 CHCl3-MeOH) were pooled
and evaporated to a solid (0.54 g, 39%). The analytical sample
was obtained by recrystallization from CH2Cl2 and cyclohex-
ane: mp 184-185 °C; IR (KBr) ν 3400 br, 1950, 1695, 1655,
1625, 1590, 1535, 1440 cm-1; 1H NMR (DMSO-d6) δ 1.62-1.77
(m, 4H, â- and γ-CH2), 3.94 (m, 1H, R-CH), 4.08 (m, 2H, δ-CH2),
4.98 (s, 2H, C6H5CH2O), 7.31 (s, 5H, C6H5CH2), 7.56 (m, 1H,
1010, 700 cm-1 1H NMR (DMSO-d6) δ 1.8 (m, 4H, â- and
;
γ-CH2), 3.14 (m, 2H, δ-CH2), 3.54 (s, 3H, homophthaloyl CO2-
CH3), 3.61 (s, 3H, R-CO2CH3), 3.82 (s, 2H, CH2CO2Me), 4.03
(m, 1H, R-CH), 5.01 (s, 2H, PhCH2OCO), 7.20-7.42 (m, 9H,
aromatic protons). Anal. (C24H22N2O7), C, H, N.
Met h yl Nδ-[2-[(Met h oxyca r b on yl)m et h yl]b en zoyl]-L-
or n ith in e Hyd r och lor id e (43‚HCl). Compound 42 (0.82 g,
1.8 mmol) was hydrogenated in a Parr apparatus as described
for 38‚HCl except that, instead of being recrystallized, the
product was precipitated from a mixture of i-PrOH and Et2O
to obtain a white solid (0.52 g, 81%): mp 132 °C; IR (KBr) ν

298 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3
Rosowsky et al.
CONH), 7.86 (m, 2H, naphthyl 3- and 6-H), 8.46 (m, 4H,
naphthyl 2-, 4-, 5-, and 7-H). Anal. (C25H22N2O6) C, H, N.
Meth yl L-2-[Nr-(Ben zyloxycar bon yl)am in o]-5-(1,8-n aph -
th a lim id o)p en ta n oa te (47). A stirred suspension of 46 (200
mg, 0.45 mmol) in MeOH (12 mL) was cooled to -10 °C and
treated dropwise with SOCl2 (0.6 mL) while the temperature
was kept below 0 °C. When addition was complete, the
reaction mixture was left to stir at room temperature for 18 h
and evaporated to dryness under reduced pressure. Recrys-
tallization from CH2Cl2 and cyclohexane gave a solid (150 mg,
75%): mp 132-133 °C; IR (KBr) ν 3400, 2900, 1745, 1700,
Refer en ces
(1) Fabre, I.; Fabre, G.; Goldman, I. D. Polyglutamylation, an
important element in methotrexate cytotoxicity and selectivity
in tumor versus murine granulocytic progenitor cells in vitro.
Cancer Res. 1984, 44, 3190-3195.
(2) (a) Samuels, L. L.; Moccio, D. M.; Sirotnak, F. M. Similar
differential for total polyglutamylation and cytotoxicity among
various folate analogues in human and murine tumor cells in
vitro. Cancer Res. 1985, 45, 1488-1495. (b) Rumberger, B. G.;
Barrrueco, J . R.; Sirotnak, F. M. Differing specificities for
4-aminofolate analogues of folylpolyglutamate synthetase from
tumors and proliferative intestinal epithelium of the mouse and
significance for selective antitumor action. Cancer Res. 1990, 50,
4639-4643.
(3) Barredo, J .; Moran, R. G. Determinants of antifolate cytotoxic-
ity: folylpolyglutamate synthetase activity during cellular pro-
liferation and development. Mol. Pharmacol. 1992, 42, 687-694.
(4) McGuire, J . J .; Bertino, J . R. Enzymatic synthesis and function
of folylpolyglutamates. Mol. Cell. Biochem. 1981, 38, 19-48.
(5) For papers on the primary sequence, structural organization,
and polymorphism of the human FPGS gene, see: (a) Garrow,
T. A.; Admon, A.; Shane, B. Expression cloning of a human cDNA
encoding folylpoly(γ-glutamate) synthetase and determination
of its primary structure. Proc. Natl. Acad. Sci. U.S.A. 1992, 89,
9151-9155. (b) Freemantle, S. J .; Taylor, S. M.; Krystal, G.;
Moran, R. G. Upstream organization of and multiple transcripts
from the human folylpoly-γ-glutamate synthetase gene. J . Biol.
Chem. 1995, 270, 9574-9584. (c) Taylor, S. M.; Freemantle, S.
J .; Moran R. G. Structural organization of the human folyl-γ-
polyglutamate synthetase gene: Evidence for a single genomic
locus. Cancer Res. 1995, 55, 6030-6034.
(6) Taylor, R. T.; Hanna, M. L. Folate dependent enzymes in
cultured Chinese hamster cells: folylpolyglutamate synthetase
and its absence in mutants auxotrophic for glycine + adenosine
+ thymidine. Arch. Biochem. Biophys. 1977, 181, 331-344.
(7) Kumar, P.; Kisliuk, R. L.; Gaumont, Y.; Freisheim, J . H.; Nair,
M. G. Inhibition of human dihydrofolate reductase by antifolyl
polyglutamates. Biochem. Pharmacol. 1989, 38, 541-543.
(8) Allegra, C. J .; Chabner, B. A.; Drake, J . C.; Lutz, R.; Rodbard,
D.; J olivet, J . Enhanced inhibition of thymidylate synthase by
methotrexate polyglutamates. J . Biol. Chem. 1985, 260, 9720-
9726.
1
1660, 1625, 1590, 1535, 1440 cm-1; H NMR (CDCl3) δ 1.81-
1.98 (m, 4H, â- and γ-CH2), 3.70 (s, 3H, OMe), 4.21 (m, 2H,
δ-CH2), 4.44 (m, 2H, R-CH), 5.08 (s, 2H C6H5CH2O), 7.25 (m,
5H, C6H5CH2), 7.75 (m, 2H, naphthyl 3- and 6-H), 8.22 (m,
2H, naphthyl 4- and 5-H), 8.60 (m, 2H, naphthyl 2- and 7-H).
Anal. (C26H24N2O6‚0.1CH2Cl2) C, H, N.
Met h yl L-2-[N-(4-Am in o-4-d eoxy-N10-for m ylp t er oyl)-
a m in o]-5-(1,8-n a p h th a lim id o)p en ta n oa te (49). A solution
of 47 (142 mg, 0.3 mmol) in MeOH (30 mL) and 6 N HCl (0.1
mL) was shaken with 10% Pd-C (20 mg) under H2 (30 psi) in
a Parr apparatus for 18 h. The reaction mixture was filtered
through Celite, and the filter pad was washed with MeOH.
The solvent was evaporated under reduced pressure, the
residue was redissolved in MeOH (5 mL), and the solution was
added to a large volume of Et2O with continuous stirring. The
precipitate was collected, washed with Et2O, and dried to
obtain 48‚HCl (80 mg, 71%): mp 140 °C, decomposing slowly
above 85 °C with gas evolution; IR (KBr) ν 3450, 2950, 1745,
1695, 1655, 1625, 1590, 1510, 1460, 1440 cm-1 1H NMR
;
(DMSO-d6) δ 1.70-1.81 (m, 4H, â- and γ-CH2), 3.70 (s, 3H,
OMe), 4.06 (m, 3H, R-CH and δ-CH2), 7.87 (m, 2H, naphthal-
imide 3- and 6-H), 8.37 (br s, 2H, NH2), 8.44 (m, 4H,
naphthalimide 2-, 4-, 5-, and 7-H). This material was used
directly in the next reaction.
i-BuOCOCl (24 µL, 0.19 mmol) and Et3N (26 µL, 0.19 mmol)
were added to a stirred suspension of 21 (70 mg, 0.19 mmol)
in dry DMF (7 mL). After 30 min, the solid obtained in the
preceding step (70 mg, 0.19 mmol) was added and stirring was
continued for 1 h. The solvent was evaporated to dryness
under reduced pressure, the residue was disolved in a mini-
mum volume of 9:1 CHCl3-MeOH, and the solution was
loaded onto a column of silica gel (10 g, 27 × 2 cm), which
was eluted first with 95:5 and then 9:1 CHCl3-MeOH.
Fractions giving a single TLC spot with Rf 0.43 (silica gel, 9:1
CHCl3-MeOH) were pooled and evaporated, the residue was
redissolved in a minimum volume of 9:1 CHCl3-MeOH, and
the solution was added dropwise with stirring to a large
volume of Et2O. The yellow precipitate was filtered and
dried: yield 25 mg (20%); mp 168-170 °C; IR (KBr) ν 3400,
(9) Allegra, C. J .; Drake, J . C.; J olivet, J .; Chabner, B. A. Inhibition
of phosphoribosyl aminoimidazolecarboxamide transformylase
by methotrexate and dihydrofolic acid polyglutamates. Proc.
Natl. Acad. Sci. U.S.A. 1985, 82, 4881-4885.
(10) (a) Pizzorno, G.; Chang, Y. M.; McGuire, J . J .; Bertino, J . R.
Inherent resistance of human squamous carcinoma cell lines to
methotrexate as a result of decreased polyglutamylation of this
drug. Cancer Res. 1989, 49, 5275-5280. (b) McCloskey, D. E.;
McGuire, J . J .; Russell, C. A.; Rowan, B. G.; Bertino, J . R.
Decreased folylpolyglutamate synthetase activity as a mecha-
nism of methotrexate resistance in CCRF-CEM human leukemia
sublines. J . Biol. Chem. 1991, 266, 6181-6187. (c) Li, W.-W.;
Lin, J . T.; Schweitzer, B. I.; Tong, W. P.; Niedziecki, D.; Bertino,
J . R. Intrinsic resistance to methotrexate in human soft tissue
sarcoma cell lines. Cancer Res. 1992, 52, 3908-3913.
(11) (a) Rumberger, B. R.; Schmid, F. A.; Otter, G. M.; Sirotnak, F.
M. Preferential selection during therapy in vivo by edatrexate
compared to methotrexate of resistant L1210 cell variants with
decresed folylpolyglutamate synthetase activity. Cancer Com-
mun. 1990, 2, 305-310. (b) Braakhuis, B. J . M.; J ansen, G.;
Noordhuis, P.; Kegel, A.; Peters, G. J . Importance of pharma-
codynamics in the in vitro antiproliferative activity of the
antifolates methotrexate and 10-ethyl-10-deazaaminopterin
against human head and neck squamous cell carcinoma. Bio-
chem. Pharmacol. 1993, 46, 2155-2161.
(12) (a) J ackman, A. L.; Taylor, G. A.; Gibson, W.; Kimbell, R.; Brown,
M.; Calvert, A. H.; J udson, I. R.; Hughes, L. R. ICI D1694, a
quinazoline antifolate thymidylate synthase inhibitor that is a
potent inhibitor of L1210 tumor cell growth in vitro and in vivo.
A new agent for clinical study. Cancer Res. 1991, 51, 5579-5586.
(b) Lu, K.; Yin, M.-B.; McGuire, J . J .; Bonmasssar, E.; Rustum,
Y. M. Mechanisms of resistance to N-[5-[N-(3,4-dihydro-2-
methyl-4-oxoquinazolin-6-ylmethyl)-N-methylamino]-2-thenoyl-
L-glutamic acid (ZD1694), a folate-based thymidylate synthase
inhibitor, in the HCT-8 human ileocecal adenocarinoma cell line.
Biochem. Pharmacol. 1995, 50, 391-398.
(13) (a) Duch, D. S.; Banks, S.; Dev, I. K.; Dickerson, S. H.; Ferone,
R.; Heath, L. S.; Humphreys, J .; Knick, V.; Pendergast, W.;
Singer, S.; Smith, G. K.; Waters, K.; Wilson, H. R. Biochemical
and cellular pharmacology of 1843U89, a novel benzoquinazoline
inhibitor of thymidylate synthase. Cancer Res. 1993, 53, 810-
818. (b) Dev, I.; Dallas, W. S.; Ferone, R.; Hanlon, M.; McKee,
D. D.; Yates, B. B. Mode of binding of folate analogs to
thymidylate synthase. J . Biol. Chem. 1994, 269, 1873-1882.
(14) (a) Matherly, L. H.; Angeles, S. M.; McGuire, J . J . Determinants
of the disparate antitumor activities of (6R)-5,10-dideaza-5,6,7,8-
tetrahydrofolate and methotrexate toward human lymphoblastic
2950, 1735, 1690, 1650, 1620, 1590, 1555, 1500, 1450 cm-1
;
1H NMR (DMSO-d6, 500 MHz) δ 1.69-1.80 (m, 4H, â- and
γ-CH2), 3.58 (s, 3H, R-CO2Me), 4.02-4.11 (m, 2H, δ-CH2), 4.41
(m, 1H, R-CH), 5.17 (s, 2H, 9-CH2), 6.6 (br s, 2H, NH2), 7.52
-7.55 (m, 2H, 3′- and 5′-H), 7.80-7.82 (m, 2H, 2′- and 6′-H),
7.82-7.84 (m, 2H, 3′′-H and 6′′-H), 8.42-8.46 (m, 4H, 2′′-, 4′′-,
5′′-, and 7′′-H), 8.62, s, 1H, C7-H), 8.66-8.67 (m, 1H, CONH),
.
8.76 (s, 1H, NCHdO). Anal. (C33H29N9O6 1.2H2O) C, H, N.
Ack n ow led gm en t. This work was supported by
Grants CA25394, CA19589, and CA0516 from the
National Cancer Institute, DHHS. PT523 is the subject
of a licensing agreement between the Dana-Farber
Cancer Institute (A.R. as consultant) and Sparta Phar-
maceuticals, Inc., Horsham, PA. The authors gratefully
acknowledge the skilled technical assistance of J orge
Pardo in the DHFR inhibition assays, Ana Maria Pardo
and Anatoly Tretyakov in the in vitro cell growth assays
against A549 cells and the in vivo antitumor assays
against murine SCC-VII squamous cell carcinoma, and
Ying-Nan Huang in the clonogenic assays against
SCC25 cells. We also want to honor the memory of the
late Dorothy Trites, who assisted with the HPLC
analyses.

Analogues of Substituted L-Ornithine
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 3 299
leukemia cells, characterized by severely impaired antifolate
membrane transport. Biochem. Pharmacol. 1993, 46, 2185-
2195. (b) Pizzorno, G.; Moroson, B. A.; Cashmore, A. R.; Russello,
O.; Mayer, J . R.; Galivan, J .; Bunni, M. A.; Priest, D. G.;
Beardsley, G. P. Multifactorial resistance to 5,10-dideazatet-
rahydrofolic acid in cell lines derived from human lymphoblastic
leukemia CCRF-CEM. Cancer Res. 1995, 55, 566-573.
(15) Itoh, F.; Russello, O.; Akimoto, H.; Beardley, G. P. Novel pyrrolo-
[2,3-d]pyrimidine antifolate TNP-351: Cytotoxic effect on meth-
otrexate-resistant CCEF-CEM cells and inhibition of trans-
formylases of de novo purine biosynthesis. Cancer Chemother.
Pharmacol. 1994, 34, 273-279.
(16) Schlemmer, S. R.; Sirotnak, F. M. Retentiveness of methotrexate
polyglutamates in cultured L1210 cells. Evidence against a role
for mediated plasma membrane transport outward. Biochem.
Pharmacol. 1993, 45, 1261-1266.
(17) (a) Rosowsky, A.; Forsch, R.; Uren, J .; Wick, M.; Kumar, A. A.;
Freisheim, J . H. Methotrexate analogues. 20. Replacement of
glutamate by long-chain amino diacids: effects of dihydrofolate
redutase inhibition, cytotoxicity, and in vivo antitumor activity.
J . Med. Chem. 1983, 26, 1719-1724. (b) Rosowsky, A.; Bader,
H.; Kohler, W.; Freisheim, J . H.; Moran, R. G. Methotrexate
analogues. 34. Replacement of the glutamate moiety in meth-
otrexate and aminopterin by long-chain 2-aminoalkanedioic
acids. J . Med. Chem. 1988, 31, 1338-1344.
(18) (a) Rosowsky, A.; Bader, H.; Cucchi, C. A.; Moran, R. G.; Kohler,
W.; Freisheim, J . H. Methotrexate analogues. 33. Nδ-Acyl-NR-
(4-amino-4-deoxypteroyl)-L-ornithine derivatives: synthesis and
in vitro antitumor activity. J . Med. Chem. 1988, 31, 1332-1337.
(b) Rosowsky, A.; Bader, H.; Forsch, R. A. Synthesis of the
folylpolyglutamate synthetase inhibitor NR-pteroyl-L-ornithine
and its Nδ-benzoyl and Nδ-hemiphthalyl derivatives, and an
improved synthesis of NR-(4-amino-4-deoxypteroyl)-Nδ-hemiph-
thaloyl-L-ornithine. Pteridines 1989, 1, 91-98.
(19) For other early examples of MTX and AMT analogues modified
in the γ-region that were synthesized in this context, see: (a)
Rosowsky, A.; Freisheim, J . H.; Bader, H.; Forsch, R. A.; Susten,
S. S.; Cucchi, C. A.; Frei, E., III. Methotrexate analogues. 25.
Chemical and biological studies on the γ-tert-butyl esters of
methotrexate and aminopterin. J . Med. Chem. 1985, 28, 660-
667. (b) Rosowsky, A.; Bader, H.; Radike-Smith, M.; Cucchi, C.
A.; Wick, M. M.; Freisheim, J . H. Methotrexate analogues. 28.
Synthesis and biological evaluation of new γ-monoamides of
aminopterin and methotrexate. J . Med. Chem. 1986, 29, 1703-
1709.
(20) Klohs, W. D.; Steinkampf, R. W.; Besserer, J . A.; Fry, D. W. Cross
resistance of pleiotropically drug resistant P388 leukemia cells
to the lipophilic antifolates trimetrexate and BW 301U. Cancer
Lett. 1986, 31, 253-260.
(21) Assaraf, Y. G.; Molina, A.; Schimke, R. T. Cross-resistance to
the lipid-soluble antifolate trimetrexate in human carcinoma
cells with the multidrug resistance phenotype. J . Natl. Cancer
Inst. 1989, 81, 290-294.
(25) Chen, G.; Wright, J . E.; Rosowsky, A. Dihydrofolate reductase
binding and cellular uptake of nonpolyglutamatable anti-
folates: Correlates of cytotoxicity toward methotrexate sensitive
and resistant human head and neck squamous carcinoma cells.
Mol. Pharmacol. 1995, 48, 758-765.
(26) 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 activ-
ity. Mol. Pharmacol. 1995, 48, 459-471.
(27) Holden, S. A.; Teicher, B. A.; Robinson, M. F.; Northey, D.;
Rosowsky, A. Antifolates can potentiate topoisomerase II inhibi-
tors in vitro and in vivo. Cancer Chemother. Pharmacol. 1995,
36, 165-171.
(28) Rosowsky, A.; Bader, H.; Wright, J . E.; Keyomarsi, K.; Matherly,
L. H. Synthesis and biological activity of Nω-hemiphthaloyl-R,ω-
diaminoalkanoic acid analogues of aminopterin and 3′,5′-dichlo-
roaminopterin. J . Med. Chem. 1994, 37, 2167-2174.
(29) Rosowsky, A.; Freisheim, J . H.; Moran, R. G.; Solan, V. C.; Bader,
H.; Wright, J . E.; Radike-Smith. M Methotrexate analogues. 26.
Inhibition of dihydrofolate reductase and folylpolyglutamate
synthetase activity and in vitro tumor cell growth by methotr-
exate and aminopterin analogues containing a basic amino acid
side-chain. J . Med. Chem. 1986, 29, 655-660.
(30) Rosowsky, A.; Wright, J . E. Nω-Alkoxycarbonylation of R,ω-
diaminoacids with 2-(trimethylsilyl)ethyl 4′-nitrophenylcarbon-
ate. J . Org. Chem. 1983, 48, 1539-1541.
(31) Roeder, G. On the condensation of urea with acid esters. Ber.
1913, 46, 2560-2564.
(32) Nefkens, G. H. L.; Tesser, G. I.; Nivard, R. J . F. A simple
preparation of phthaloyl amino acids via a mild phthaloylation.
Rec. Trav. Chim. Pays-Bas 1960, 79, 658-698.
(33) We are grateful to Dr. Vivian Cody, Hauptmann-Woodward
Medical Research Institute, Buffalo, NY, for her help in recog-
nizing these potential PT523 interaction sites from a crystal-
lographic analysis of the ternary complex of human DHFR with
NADPH and MTX-γ-tetrazole.34
(34) Cody, V; Luft, J . R.; Ciszak, E.; Kalman, T. I.; Freisheim, J . H.
Crystal structure determination at 2.3 Å of recombinant human
dihydrofolate reductase ternary complex with NADPH and
methotrexate-γ-tetrazole. Anti-Cancer Drug Des. 1992, 7, 483-
491.
(35) Monks, A.; Scudiero, D.; Skehan, P.; Shoemaker, R.; Paull, K.;
Vistica, D.; Hose, C.; Langley, J .; Cronise, P.; Vaigro-Wolff, A.;
Gray-Goodrich, M.; Campbell, H.; Mayo, J .; Boyd, M. Feasibility
of a high-flux anticancer drug screen using a diverse panel of
cultured human tumor cell lines. J . Natl. Cancer Inst. 1991, 83,
757-766.
(36) Miyajima, K.; Takemoto, M.; Achiwa, K. Prostanoids and related
compounds. V. Synthesis of 6-aza-5-oxo-2,3,4-trinor-1,5-inter-m-
phenylene prostacyclin derivatives. Chem. Pharm. Bull. 1991,
39, 3175-3179.
(22) Baccanari, D. P.; Tansik, R. L. Multiple drug resistance and the
relationship to brain penetration by lipophilic dihydrofolate
reductase inhibitors. Proc. Am. Assoc. Cancer Res. 1990, 31, 339.
(23) Kuyper, L. F.; Baccanari, D. P.; J ones, M. L.; Hunter, R. N.;
Tansik, R. L.; J oyner, J . S.; Boytos, C. M.; Rudolph, S. K.; Knick,
V.; Wilson, H. R.; Cadell, J . M.; Friedman, H. S.; Comley, J . C.
W.; Stables, J . N. High-affinity inhibitors of dihydrofolate
reductases: antimicrobial and anticancer activities of 7,8-dialkyl-
1,3-diaminopyrrolo[3,2-f]quinazolines with small molecular size.
J . Med. Chem. 1996, 39, 892-903.
(37) We are indebted to Dr. J . M. Brown, Department of Radiology,
Stanford University School of Medicine, Palo Alto, CA, for
providing us with the SCC VII tumor used in these experiments.
The SCC VII cell line was originally isolated from a spontaneous
tumor of the abdominal wall of a C3H mouse by Dr. Herman
Suit, Massachusetts General Hospital, Boston, and has been
used primarily to study tumor growth delay by chemotherapeutic
agents and radiation; cf. Weinberg, M. J .; Rauth, A. M. 5-Fluo-
rouracil infusions and fractionated doses of radiation: studies
with a murine squamous cell carcinoma. Int. J . Radiation Oncol.
Biol. Phys. 1987, 13, 1691-1699.
(24) Rhee, M. S.; Galivan, J .; Wright, J . E.; Rosowsky, A. Biochemical
studies on PT523, a potent nonpolyglutamatable antifolate, in
cultured cells. Mol. Pharmacol. 1994, 45, 783-791.
J M9606453