Design, synthesis and biological evaluation of 10-CF3CO-DDACTHF analogues and derivatives as inhibitors of GAR Tfase and the de novo purine biosynthetic pathway

Article abstract of DOI:10.1016/S0968-0896(03)00458-9

The synthesis and evaluation of analogues and key derivatives of 10-CF ₃CO-DDACTHF as inhibitors of glycinamide ribonucleotide transformylase (GAR Tfase) and aminoimidazole carboxamide transformylase (AICAR Tfase) are reported. Polyglutamate analogues of 1 were evaluated as inhibitors of Escherichia coli and recombinant human (rh) GAR Tfase, and AICAR Tfase. Although the pentaglutamate 6 was found to be the most active inhibitor of the series tested against rhGAR Tfase (K_i=0.004 μM), little distinction between the mono-pentaglutamate derivatives was observed (K _i=0.02-0.004 μM), suggesting that the principal role of the required polyglutamation of 1 is intracellular retention. In contrast, 1 and its defined polyglutamates 3-6 were much less inactive when tested against rhAICAR Tfase (K_i=65-0.120 μM) and very selective (≥100-fold) for rh versus E. coli GAR Tfase. Additional key analogues of 1 were examined (7 and 8) and found to be much less active (1000-fold) highlighting the exceptional characteristics of 1.

Full text of DOI:10.1016/S0968-0896(03)00458-9

Bioorganic & Medicinal Chemistry 11 (2003) 4511–4521
Design, Synthesis and Biological Evaluation of 10-CF3CO-
DDACTHF Analogues and Derivatives as Inhibitors of GAR
Tfase and the De Novo Purine Biosynthetic Pathway
Joel Desharnais,a,c Inkyu Hwang,a,c Yan Zhang,b,c Ali Tavassoli,d Justin Baboval,d
d
a,c,
Stephen J.Benkovic, Ian A.Wilson b,c and Dale L.Boger
*
aDepartment of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA92037, USA
bDepartment of Molecular Biology, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA92037, USA
cThe Skaggs Institute for Chemical Biology, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA92037, USA
dDepartment of Chemistry, Pennsylvania State University, University Park, PA16802, USA
Received 22 April 2003; accepted 22 April 2003
Abstract—The synthesis and evaluation of analogues and key derivatives of 10-CF3CO-DDACTHF as inhibitors of glycinamide
ribonucleotide transformylase (GAR Tfase) and aminoimidazole carboxamide transformylase (AICAR Tfase) are reported.Poly-
glutamate analogues of 1 were evaluated as inhibitors of Escherichia coli and recombinant human (rh) GAR Tfase, and AICAR
Tfase.Although the pentaglutamate 6 was found to be the most active inhibitor of the series tested against rhGAR Tfase (Ki=0.004
mM), little distinction between the mono–pentaglutamate derivatives was observed (Ki=0.02–0.004 mM), suggesting that the prin-
cipal role of the required polyglutamation of 1 is intracellular retention.In contrast, 1 and its defined polyglutamates 3–6 were
much less inactive when tested against rhAICAR Tfase (Ki=65–0.120 mM) and very selective (ꢀ100-fold) for rh versus E. coli GAR
Tfase.Additional key analogues of 1 were examined (7 and 8) and found to be much less active (1000-fold) highlighting the
exceptional characteristics of 1.
# 2003 Elsevier Ltd.All rights reserved.
Glycinamide ribonucleotide transformylase (GAR Tfase)
and aminoimidazole carboxamide ribonucleotide trans-
formylase (AICAR Tfase) are folate-dependent enzymes
central to the de novo purine biosynthetic pathway.
GAR Tfase utilizes the cofactor (6R)-N10-formyltetra-
hydrofolate (Fig.1 ) to transfer a formyl group to the
primary amine of its substrate, glycinamide ribonucleo-
tide (GAR, Fig.1 ).This one carbon transfer incor-
porates the C-8 carbon of the purines and is the first of
two formyl transfer reactions.The second formyl trans-
fer reaction is catalyzed by the enzyme AICAR Tfase
which also employs (6R)-N10-formyltetrahydrofolate to
transfer a formyl group to the C-5 amine of its
substrate, aminoimidazole carboxamide ribonucleotide
(AICAR, Fig.1 ).1 The discovery that (6R)-5,10-dide-
azatetrahydrofolate (Lometrexol, (6R)-DDATHF, Fig.
2) achieves its potent anticancer activity by selective
GAR Tfase inhibition established GAR Tfase and the
purine de novo biosynthetic pathway as viable targets
for antineoplastic intervention.2À4 Herein, we report the
synthesis and evaluation of a series of analogues and key
derivatives of 1 incorporating the DDACTHF scaffold.
InhibitorDesign
In previous studies, we examined folate-based inhibitors
which incorporated electrophilic functional groups that
could potentially interact either with active site nucleo-
philes or the GAR/AICAR substrate amines.5À8 It was
envisioned that the properly positioned nontransferable
electrophilic carbonyl could potentially form an imine
or a tetrahedral adduct with the active site or substrate
nucleophiles or serve to stabilize gem diol formation of
the electrophilic carbonyl and promote active site bind-
ing by mimicking the tetrahedral intermediate of the
formyl transfer reactions.This latter effect was observed
with folate-based inhibitors bearing a nontransferable
formyl group and has provided potent, selective, and
efficacious GAR Tfase inhibitors.6,9
*Corresponding author.Te:l. +1-858-784-7522; fax: +1-858-784-
7550; e-mail: boger@scripps.edu
0968-0896/$ - see front matter # 2003 Elsevier Ltd.All rights reserved.
doi:10.1016/S0968-0896(03)00458-9

4512
J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
In the preceding articles, we examined folate-based
inhibitors bearing alternative electrophilic carbonyls.7,8
Embodied in these studies was the recognition that many
such folate-based inhibitors based on the DDACTHF
scaffold were not only effective enzyme inhibitors, but
also effectively transported into the cell by the reduced
folate carrier and polyglutamated by FPGS.6 In con-
junction with these studies, we also recently reported the
design, synthesis, and the biological evaluation of
10-trifluoroacetyl-DDACTHF (10-CF3CO-DDACTHF,
1).10 This analogue was shown to be a selective and
potent GAR Tfase inhibitor (Ki=0.015 mM against
rhGAR Tfase) and an effective cytotoxic agent (CCRF–
CEM IC50=16 nM).10 This GAR Tfase inhibitor was
shown not only to possess a potency similar to 10-for-
myl-DDACTHF (2, Ki=0.014 mM against rhGAR
Tfase and CCRF–CEM IC50=60 nM),6 but to exhibit
the necessary chemical stability for in vivo evaluation.
Thus, a facile oxidative decarbonylation of the formyl
group conveyed a chemical instability to 2 precluding
consideration for in vivo use, while the presence of the
trifluoromethyl ketone on the DDACTHF scaffold
with 1 provides a stable compound suitable for in vivo
evaluation.
A crystal structure of 1 with rhGAR Tfase was obtained
and the overall structure is represented in Figure 3.10
Interestingly, even though an inseparable and rapidly
equilibrating diastereomeric mixture was used in the
crystallization, only the R form of 1 was found in the
folate-binding site.The inhibitor 1 binds as the gem diol
mimicking the formyl transfer intermediate and making
extensive interactions with the catalytic residues at the
active site (Fig.4 ).10 The gem diol forms hydrogen
bonds with several residues present in the active site,
especially Asp144, His108, and Gly117.The terminal
carboxylate of Asp144 forms two hydrogen bonds with
each of the hydroxyls of the gem diol (2.5 and 2.7 A),
while one of the nitrogen atoms of the His108 imidazole
also forms two hydrogen bonds with the gem diol (3.1
and 3.2 A).In addition, the backbone carbonyl of
Gly117 forms a 3.0 A hydrogen bond with the gem diol.
These hydrogen bonding interactions were found to be
important for the activity since the alcohol variant of
10-CF3CO-DDACTHF was 60-fold less potent, and
DDACTHF, which lacks a C10 substituent and both
alcohols of the gem diol, was 115 times less active.10
Figure 1.
Herein, we report a series of analogues and derivatives
of 1 that were synthesized and evaluated to further
Figure 2.
Figure 3.

J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
4513
define the details of the inhibition of GAR Tfase (Fig.
5).The full series of polyglutamated derivatives of
enzyme active site catalytic residues including His108,
Asp144, and Gly117,10 an analogue with a C10 car-
boxylic acid could similarly bind by forming multiple
hydrogen bonds with the active site residues or even the
substrate amine.
1
(from di- to pentaglutamate, 3–6) were prepared since it
has been observed that inhibitors like 1 are converted to
polyglutamated homologues11À15 by FPGS that are not
only retained intracellularly,15 but that were often more
potent enzyme inhibitors than their monoglutamate
counterpart.12À14 Recent studies suggest that the poly-
glutamate metabolites of DDATHF are the principal
species causative of cell growth inhibition, and that
DDATHF itself may have minor cytotoxic activity.12
The pentaglutamate of DDATHF has been reported to
be about 1113 or 10012 times more potent against human
and mouse GAR Tfase, respectively, than DDATHF.
Chemistry
The di-, tri-, tetra- and pentaglumate analogues of 1 (3–
6) were prepared from the common intermediate 9,10 as
presented in Scheme 1.The carboxylic acid 9 was cou-
pled with the appropriate di-tert-butyl l-polyglutamate
hydrochloride 1016 to provide 11–14 (EDCI, NaHCO3,
DMF, 25 ꢁC, 24 h, 14–34%).Acid-catalyzed deprotec-
tion of 11–14 (1:4 v/v TFA/CHCl3, 24 h, 100%) pro-
vided the desired derivatives 3–6.For comparative
purposes, ketone 14 (pentaglutamate) was reduced with
NaBH4 (2.0 equiv, CH3OH, À20 ꢁC, 30 min) followed
by acid-catalyzed deprotection (4 N HCl–dioxane, 0–
25 ꢁC, 3 h, 50% from 14) to provide alcohol 15
(Scheme 2).
We also describe two key analogues of 1, one (7) that
contains an additional carbon in the chain linking the
benzoyl glutamate and the diaminopyrimidinone to
establish the importance of the length of this spacer.
The second (8) is an analogue where the trifluoromethyl
ketone was replaced by a carboxylic acid.Since the
hydrated form of 1 forms several hydrogen bonds with
Scheme 1.
Figure 4.
Figure 5.
Scheme 2.

4514
J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
The preparation of 7 follows closely that described for
1.10 NaH deprotonation of the known N,N-dimethyl-
hydrazone 1610 (DMF, 0 ꢁC, 15 min) and subsequent
treatment with excess 1,4-dibromobutane (10 equiv,
DMF, 25 ꢁC, 2.5 h) provided the monoalkylation pro-
duct 17.The preformed sodium salt of ethyl cyanoace-
tate (NaH, DMF, 0 ꢁC, 30 min) was alkylated with 17 to
give 18 (DMF, 25 ꢁC, 2 h, 28% from 16).Cyclization
with the free base of guanidine (CH3OH, reflux, 16 h,
68%) under basic conditions gave the desired pyr-
imidinone 19.Treatment of 19 with LiOH (2.1 equiv,
3:1 CH3OH–H2O, 25 ꢁC, 24 h) cleanly provided the
carboxylic acid 20 which was coupled with di-tert-butyl
l-glutamate hydrochloride (EDCI, NaHCO3, DMF,
25 ꢁC, 48 h, 36%) to provide 21.Deprotection of 21 was
accomplished by treatment with trifluoroacetic acid (1:4
v/v TFA/CHCl3, 25 ꢁC, 16 h, 100%) to provide 7
(Scheme 3).
hydroxy derivative pentaglutamate 15 was 34-fold more
potent than its monoglutamate 29 (Ki=20 mM) and
slightly more potent than the ketone monoglutamate 1
against E. coli GAR Tfase (0.58 mM for 15 vs 1.9 mM
for 1).The monoglutamate analogue 7 was only slightly
active against E. coli GAR Tfase (Ki=24 mM), whereas
8 was inactive in the assay (Ki>100 mM).The same
compounds were tested for inhibition of rhGAR Tfase
and all of them showed an increase in activity (Table 1).
Thus, the remarkable 100-fold selectivity for rhGAR
Tfase versus E. coli GAR Tfase observed with 110 and
earlier with the corresponding aldehyde 2 (10-formyl-
DDACTHF),6 is also observed with the full series of
polyglutamates.Importantly, it is this activity, not that
of the E. coli enzyme, that correlates with the CCRF–
CEM cytotoxic potency.Interestingly, each of the
polyglutamates exhibited a similar activity against
The synthesis of 8 was accomplished using a similar
procedure.Protection of 2217 as a tert-butyl ester
(tBuOH, DCC, DMAP, 25 ꢁC, 20 h, 44%), followed by
NaH deprotonation of 2318 (DMF, 0 ꢁC, 15 min) and
subsequent treatment with excess 1,3-dibromopropane
(10 equiv, DMF, 25 ꢁC, 2.5 h, 58%) provided the
monoalkylation product 24.The preformed sodium salt
of ethyl cyanoacetate (NaH, DMF, 0 ꢁC, 30 min) was
alkylated with 24 to give 25 (DMF, 25 ꢁC, 2 h, 73%).
Cyclization with the free base of guanidine (CH3OH,
reflux, 16 h, 51%) under basic conditions gave the
desired pyrimidinone 26.Treatment of 26 with LiOH
(3.0 equiv, 3:1 CH3OH–H2O, 25 ꢁC, 6 h) cleanly pro-
vided the carboxylic acid 27, which was coupled with di-
tert-butyl l-glutamate hydrochloride (EDCI, NaHCO3,
DMF, 25 ꢁC, 48 h, 29%) to provide 28.Deprotection of
28 was accomplished by treatment with trifluoroacetic
acid (1:4 v/v TFA/CHCl3, 25 ꢁC, 16 h, 100%) to provide
8 (Scheme 4).
GAR Tfase and AICAR Tfase Inhibition
Compounds 3–8 and 15 were tested for inhibition of
GAR Tfase and AICAR Tfase and the results are pre-
sented in Table 1 along with those derived from 1,
DDACTHF19 lacking a C10 substituent (Fig.2 ), and
the key analogues 29–30 previously disclosed (Fig.
6).7,10 The activity of the polyglutamated analogues of 1
against Escherichia coli GAR Tfase varied depending
upon the length of the glutamate chain.For the deriva-
tives with two (3), three (4) and four (5) glutamates, the
Ki’s were similar to the monoglutamate (5.6, 10 and 4.8
mM, respectively, vs 1.9 mM for 1), while the pentaglu-
tamate 6 was 7 times more active (Ki=0.27 mM).The
Figure 6.
Scheme 3.

J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
4515
rhGAR Tfase, where the di-, tri, tetra- and pentagluta-
mate had a Ki of 0.019, 0.01, 0.019 and 0.004 mM,
respectively, similar to that of the monoglutamate 1 in a
parallel assay (Ki=0.015 mM).Thus, the trends follow a
well-defined order of 5>3>1>2,4 glutamates,
although the distinctions are very small.The pentaglu-
tamate 6 was 2-fold more potent than the triglutamate
4, 3-fold more potent than the monoglutamate 1, and 4–
5-fold more potent than the di- and tetraglutamates.
These observations do not follow those made with
Lometrexol where a much more substantial increase in
activity was seen with the addition of each glutamate
side chain and where the pentaglutamate was estab-
lished to be 11- or 100-fold more potent than the
monoglutamate DDATHF.12,13 Identical to trends
observed in comparing 1 and 29 (monoglutamates,
Ki=15 and 900 nM),10 the pentaglutamate 15 of the
alcohol analogue was 10-fold less potent than the tri-
fluoromethyl ketone pentaglutamate 6 (40 nM vs 4 nM)
highlighting the importance of the active site gem diol
versus alcohol interaction.Most impressive in this ser-
ies, the trifluoromethyl ketone 1 (Ki=15 nM) was 60-
fold more potent than the corresponding alcohol 29
(Ki=900 nM) and roughly 100-fold more potent than
DDACTHF (Ki=1.7 mM) which lacks a C10 sub-
stituent, indicating (like the comparison of 6 and 15)
that each hydroxyl group of the bound gem diol
increases binding affinity roughly 10-fold.The impor-
tance of the benzoylglutamate subunit is defined by a
comparison with the simplified trifluoromethyl ketone
(30)7 which was inactive against both human and E. coli
GAR Tfase (Ki>100 mM).Similarly, 9 and its corre-
sponding methyl ester did not inhibit rhGAR Tfase
(Ki>100 mM) indicating that the potent inhibitory
activity of 1 requires the intact benzoylglutamate
including the glutamate subunit.10 Compounds 1 and 29
were both inactive against rhAICAR Tfase (Ki>100
mM), rhDHFR (Ki>100 mM), and rhTS (Ki>100 mM).
The latter observations are consistent with the demon-
strations below that 1 and 29 derive potent cytotoxic
activity through inhibition of the purine, not pyr-
imidine, biosynthesis and at a step preceding the action
of AICAR Tfase.Interestingly, the glutamate series
(n=1–5) did show a progressive increase in potency
against rhAICAR Tfase, but each was 10–160 fold less
potent for AICAR versus GAR Tfase.
Scheme 4.
Table 1. GAR and AICAR Tfase inhibition (Ki, mM)
Most surprising in the series of compounds examined
was 7.As observed with E. coli GAR Tfase (Ki=24
mM), 7 was only slightly active against rhGAR Tfase
(Ki=22 mM) and displayed no selectivity between the
two enzymes.This represents a remarkable 1000-fold
loss in rhGAR Tfase activity with 7 relative to 1 result-
ing from the one carbon increased linker length even
within a flexible linker.Clearly, the binding of 1 with
the enzyme represents achievement of a near optimal
interaction.
Compd
E. coli
GAR Tfasea
rhGAR
Tfaseb
rhAICAR
Tfasec
1
3
4
5
6
15
29
7
1.9
5.6
10
0.015
0.019
0.010
0.019
0.004
0.04
0.9
65
1.6
1.0
0.18
0.12
0.8
4.8
0.27
0.58
20
>100
nd
60
24
22
8
>100
4.6
0.1
5.3
1.7
0.06d
DDACTHF
Lometrexol
20
1
In the case of 8, modest activity against rhGAR Tfase
(Ki=5.3 mM) was observed, whereas it was inactive
against E. coli GAR Tfase (Ki >100 mM).Although the
activity of 8 is modest, like 1, it was >20-fold selective
for human versus E. coli GAR Tfase.
aE. coli GAR Tfase.
bRecombinant human GAR Tfase.
cRecombinant human AICAR Tfase.
dref.13

4516
J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
Table 2. In vitro cytotoxic activity
Table 4. In vitro cytotoxic activity
Compd
CCRF–CEM (IC50, mM)
Compd
CCRF–CEM/MTX (IC50, mM)
(+) T,
(+) Ha
(À) T,
(+) T,
(À) H
(À) T,
(À) H
(+) T,
(+) Ha
(À) T,
(+) T,
(À) H
(À) T,
(À) H
(+) H
(+) H
1
3
4
5
6
15
29
7
>100
>100
>100
>100
>100
nd
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
nd
>100
>100
>100
>100
>100
0.017
0.5
2.0
2.5
2.0
nd
1.4
50
40
0.016
0.5
1.5
3.0
4.0
nd
1.1
50
50
2.7
0.23
1
29
>100
>100
>100
>100
nd
nd
nd
nd
nd
nd
nd
nd
>100
>100
>100
>100
DDACTHF
Lometrexol
Compd
CCRF–CEM/FPGSÀ (IC50, mM)
(+) T,
(+) Ha
(À) T,
(+) T,
(À) H
(À) T,
(À) H
8
(+) H
DDACTHF
Lometrexol
3.6
0.52
1
29
>100
87
>100
>100
nd
nd
nd
nd
nd
nd
nd
nd
>100
25
>100
>100
aT, thymidine; H, hypoxanthine.
DDACTHF
Lometrexol
aT= thymidine (+10 mM); H, hypoxanthine (+100 mM).
Cytotoxic Activity
Compounds 3–8 and DDACTHF were examined for
cytotoxic activity both in the presence (+) and absence
(À) of added thymidine (pyrimidine) and hypoxanthine
(purine) against the CCRF–CEM cell line (Table 2).All
the analogues of 10-CF3CO-DDACTHF (1) were at
least 40-fold less active against the CCRF–CEM cell
line than 1.This was expected for the polyglutamated
analogues 3–6, and is due to ineffective cellular pene-
tration.Compound 1 exhibited very potent cytotoxic
activity (IC50=16 nM) against the CCRF–CEM cell
line when purines (hypoxanthine) are absent in the
media.10 Moreover, it is 14-fold more potent than
Lometrexol (IC50=230 nM) and both were inactive
(IC50>100 mM) in the presence of media purines.This
sensitivity to the presence of purines, but not pyr-
imidines (thymine), indicates that the cytotoxic activity
of 1 is derived from its inhibition of enzymes in the de
novo purine biosynthetic pathway.The corresponding
alcohol 29 and DDACTHF exhibited cytotoxic activity
(IC50=1.1 and 2.7 mM, respectively) which were also
sensitive to the presence of media purines.However, 29
and DDACTHF were ca.70 Â and 170Â less potent
than ketone 1 indicating the potentiation of biological
activity conveyed by the electrophilic carbonyl.
purine biosynthesis prior to the AICAR Tfase enzy-
matic step, consistent with the inhibition of GAR Tfase.
This selective sensitivity to GAR Tfase is the expected
behavior of the inhibitors 1 and 29 based on their inac-
tivity against AICAR Tfase in vitro.
The extent to which the cytotoxic activity of 1 and 29
was dependent on folate active transport across the cel-
lular membrane was established by assaying against a
resistant CCRF–CEM cell line (CCRF–CEM/MTX)
shown to have an impaired reduced folate active trans-
port system20 (Table 4).Like Lometrexol, 1 and 29 lost
activity against CCRF–CEM/MTX indicating a func-
tioning reduced folate carrier is required for functional
activity and implying they are effective substrates for
transport.Similarly, the importance of FPGS poly-
glutamation to the inhibitors cytotoxic activity was
established by examining them against a CCRF–CEM
cell line deficient in FPGS (CCRF–CEM/FPGSÀ).6
Like Lometrexol, 1 and 29 (to a lesser extent) lacked or
lost activity against this cell line (Table 4) indicating
polyglutamation is required for activity.Because of the
comparable rhGAR Tfase inhibitory activity of 1 with
its polyglutamates, this would seem to imply that the
polyglutamation requirement serves to promote intra-
cellular accumulation.
AICAR rescue experiments were performed using 1 and
29 in order to further elucidate the source of their cyto-
toxic activity (Table 3).In each case, the reversal of the
cytotoxicity with hypoxanthine (100 mM) or AICAR
monophosphate (100 mM) resulted in an ca.10 3–104
increase in the IC50 value.This indicates that the activ-
ity is being observed through selective inhibition of
As for 7 and 8, the addition of a carbon between the
diaminopyrimidinone and the trifluoromethyl ketone (7)
or the replacement of this ketone by a carboxylic acid
(8) had a great impact on the activity.Both exhibited
detectable purine sensitive cytotoxic activity that is
derived from their GAR Tfase inhibitory properties, but
at a level that is >1000-fold less potent than 1 and
>10-fold less potent than DDACTHF.
Table 3. Cytotoxic activity in the presence of AICAR
Compd
CCRF–CEM (IC50, mM)
(À) T, (À) H, (À) Aa
(À) T, (À) H, (+) A
Conclusions
1
29
0.016
1.1
0.23
>150
>150
>150
Several derivatives and key analogues of 10-CF3CO-
DDACTHF (1) were prepared and evaluated as inhibi-
tors of GAR Tfase and AICAR Tfase.Most prominent
among the observations was the unusually selective
Lometrexol
aT= thymidine; H, hypoxanthine, A=AICAR monophosphate
(+100 mM).

J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
4517
(ꢀ100-fold) and potent (Ki=4–20 nM) inhibition of
human (vs E. coli) GAR Tfase analogous to that
observed with 26 and the lack of activity of the analo-
gues against other representative folate dependent
reduced pressure.Chromatography (SiO , 9:1 CHCl3/
2
MeOH) provided 12 (21.2 mg, 34%) as a white solid: 1H
NMR (CD3OD, 400 MHz) d 7.90 (m, 2H), 7.40 (m, 2H),
4.47 (m, 1H), 4.26 (m, 3H), 3.61 (m, 1H), 2.31 (m, 23H),
1.42 (m, 30H); MALDIFTMS (DHB) m/z 1032.4885
(M+Na+, C48H70F3N7O13Na requires 1032.4876).
enzymes (rhTS, rhDHFR).The polyglutamates of
1
exhibited a well defined trend of n=5>3>1>2,4
against rhGAR Tfase although the distinctions in
potency were very modest (Ki=4, 10, 15, 19 and 19
nM).Consequently, the requirement for FPGS and
polyglutamation for observation of the potent, purine
sensitive cytotoxic activity of 110 is most likely the result
of enhanced intracellular accumulation (retention) and
not enhanced enzyme inhibitory potency.The ꢀ10-fold
loss in activity resulting from reduction of the ketone to
the alcohol 29 and the ꢀ100-fold loss in activity with its
removal (DDACTHF) highlights the importance of the
electrophilic carbonyl and each of the hydroxyls of the
enzyme-bound gem diol.10 Significantly, the key analo-
gues lacking the benzoylglutamate (30) or glutamate (9
or its methyl ester) subunits were inactive against
rhGAR Tfase and CCRF–CEM, whereas the mono-
glutamate analogue of 1 with a one carbon extension in
the flexible spacer (7) had a >1000-fold decrease in
activity against both rhGAR Tfase (Ki=22 mM) and
CCRF–CEM growth inhibition (IC50=50 mM) indicat-
ing how unique the active site interaction of 1 may be.
Similarly, the analogue 8 in which the electrophilic tri-
fluoromethyl ketone was replaced with a carboxylic acid
was >100-fold less active against rhGAR Tfase (Ki=5
mM) and >1000-fold less cytotoxic (IC50=40 mM) fur-
ther emphasizing the importance of the trifluoromethyl
ketone (gem diol) for activity of 1.
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(2,2,2-trifluoroacetyl)but-2-yl]benzoyl}-L-ꢀ-glutamyl-L-
ꢀ-glutamyl-L-ꢀ-glutamyl-L-ꢀ-glutamic acid octa-tert-
butyl ester( 13). A solution of 910 (28.6 mg, 0.072
mmol) and 10c16 (83.9 mg, 0.10 mmol, 1.4 equiv) in
DMF (0.5 mL) was treated with NaHCO3 (20.0 mg,
0.24 mmol) followed by EDCI (20.1 mg, 0.10 mmol)
and stirred at 25 ꢁC for 48 h.The reaction mixture was
diluted with EtOAc and washed with saturated aqueous
NaHCO3.The organic layer was dried (Na SO4) and
2
concentrated under reduced pressure.Chromatography
(SiO2, 9:1 CHCl3/MeOH) provided 13 (11.9 mg, 14%)
1
as a white solid: H NMR (CD3OD, 400 MHz) d 7.87
(m, 2H), 7.39 (m, 2H), 4.52 (m, 1H), 4.18 (m, 4H), 3.60
(m, 1H), 2.31 (m, 29H), 1.42 (m, 37H); MALDIFTMS
(DHB) m/z 1195.6106 (M+Na+, C66H100F3N9O19Na
requires 1195.6108).
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(2,2,2-trifluoroacetyl)but-2-yl]benzoyl}-L-ꢀ-glutamyl-L-
ꢀ-glutamyl-L-ꢀ-glutamyl-L-ꢀ-glutamyl-L-ꢀ-glutamic acid
deca-tert-butyl ester( 14). A solution of 910 (24.2 mg,
0.061 mmol) and 10d16 (66.6 mg, 0.067 mmol, 1.0 equiv)
in DMF (0.5 mL) was treated with NaHCO3 (16.8 mg,
0.2 mmol) followed by EDCI (36.6 mg, 0.19 mmol) and
stirred at 25 ꢁC for 48 h.The reaction mixture was
diluted with EtOAc and washed with saturated aqueous
NaHCO3.The organic layer was dried (Na SO4) and
2
Experimental
concentrated under reduced pressure.Chromatography
(SiO2, 9:1 CHCl3/MeOH) provided 14 (13.6 mg, 16%)
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(2,2,2-trifluoroacetyl)but-2-yl]benzoyl}-L-ꢀ-glutamyl-L-
ꢀ-glutamic acid tetra-tert-butyl ester( 11). A solution of
910 (23.6 mg, 0.059 mmol) and 10a16 (34.3 mg, 0.077
mmol, 1.3 equiv) in DMF (0.5 mL) was treated with
NaHCO3 (19.6 mg, 0.23 mmol) followed by EDCI (16.4
mg, 0.09 mmol) and stirred at 25 ꢁC for 48 h.The reac-
tion mixture was diluted with EtOAc and washed with
saturated aqueous NaHCO3.The organic layer was
dried (Na2SO4) and concentrated under reduced pres-
1
as a white solid: H NMR (CD3OD, 400 MHz) d 7.82
(m, 2H), 7.40 (m, 2H), 4.59 (m, 1H), 4.31 (m, 5H), 3.60
(m, 1H), 2.31 (m, 34H), 1.42 (m, 42H); MALDIFTMS
(DHB) m/z 1402.6959 (M+Na+, C66H100F3N9O19Na
requires 1402.6979).
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-1
-(2,2,2-trifluoroacetyl)but-2-yl]benzoyl}-L-ꢀ-glutamyl-L-
ꢀ-glutamic acid (3). A solution of 11 (11.1 mg, 0.01
mmol) in CHCl3 (1.0 mL), cooled to 0 ꢁC, was treated
with trifluoroacetic acid (0.3 mL). The solution was
stirred 25 ꢁC for 12 h before being concentrated under
reduced pressure.The solid residue was triturated with
ether and dried in vacuo to give 3–CF3CO2H (10.0 mg,
sure.Chromatography (SiO , 9:1 CHCl3/MeOH) pro-
2
vided 11 (13.1 mg, 27%) as a white solid: 1H NMR
(CD3OD, 400 MHz) d 7.72 (m, 2H), 7.39 (m, 2H), 4.47
(m, 1H), 4.27 (m, 2H), 3.86 (m, 1H), 2.31 (m, 17H), 1.42
(m, 23H); MALDIFTMS (DHB) m/z 825.4007
(M+H+, C39H56F3N6O10 requires 824.4004).
1
100%) as a tan solid: H NMR (CD3OD, 400 MHz) d
7.54 (m, 2H), 7.18 (m, 2H), 4.36 (s, 1H), 4.22 (m, 2H),
2.21 (m, 15H); MALDIFTMS (DHB) m/z 679.1972
(M+Na+, C27H31F3N6O10Na requires 679.1946).
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(2,2,2-trifluoroacetyl)but-2-yl]benzoyl}-L-ꢀ-glutamyl-L-
ꢀ-glutamyl-L-ꢀ-glutamic acid hexa-tert-butyl ester( 12).
A solution of 910 (24.4 mg, 0.061 mmol) and 10b16
(46.7 mg, 0.074 mmol, 1.2 equiv) in DMF (0.5 mL) was
treated with NaHCO3 (15.2 mg, 0.181 mmol) followed
by EDCI (14.7 mg, 0.077 mmol) and stirred at 25 ꢁC for
48 h.The reaction mixture was diluted with EtOAc and
washed with saturated aqueous NaHCO3.The organic
layer was dried (Na2SO4) and concentrated under
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(2,2,2-trifluoroacetyl)but-2-yl]benzoyl}-L-ꢀ-glutamyl-L-
ꢀ-glutamyl-L-ꢀ-glutamic acid (4). A solution of 12 (20.1
mg, 0.02 mmol) in CHCl3 (0.8 mL), cooled to 0 ꢁC, was
treated with trifluoroacetic acid (0.2 mL). The solution
was stirred 25 ꢁC for 12 h before being concentrated
under reduced pressure.The solid residue was triturated

4518
J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
with ether and dried in vacuo to give 4–CF3CO2H (17.9
mg, 100%) as a tan solid: 1H NMR (CD3OD, 400 MHz)
d 7.76 (m, 2H), 7.40 (m, 2H), 4.61 (s, 1H), 4.42 (m, 3H),
2.22 (m, 19H); MALDIFTMS (DHB) m/z 786.2576
(M+H+, C32H39F3N7O13 requires 786.2552).
equiv) was added quickly to the reaction mixture and
the cooling bath was removed.The reaction mixture
was stirred at 25 ꢁC for 2.5 h. The reaction was quen-
ched by the addition of saturated aqueous NH4Cl (15
mL).The reaction mixture was partitioned between
EtOAc (50 mL) and H2O (40 mL).The organic layer
was washed with H2O (3Â50 mL) and saturated aqu-
eous NaCl (1Â50 mL) followed by concentration under
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(2,2,2-trifluoroacetyl)but-2-yl]benzoyl}-L-ꢀ-glutamyl-L-
ꢀ-glutamyl-L-ꢀ-glutamyl-L-ꢀ-glutamic acid (5). A solu-
tion of 13 (10.1 mg, 0.008 mmol) in CHCl3 (0.8 mL),
cooled to 0 ꢁC, was treated with trifluoroacetic acid (0.2
mL).The solution was stirred 25 ꢁC for 12 h before
being concentrated under reduced pressure.The solid
residue was triturated with ether and dried in vacuo to
give 5–CF3CO2H (8.7 mg, 100%) as a tan solid: 1H NMR
(CD3OD, 400 MHz) d 7.88 (m, 2H), 7.37 (m, 2H), 4.61
(s, 1H), 4.39 (m, 4H), 2.26 (m, 22H); MALDIFTMS
(DHB) m/z 937.2782 (M+Na+, C37H45F3N8O16Na
requires 937.2798).
reduced pressure.Chromatography (SiO , 7:1 hexanes/
2
EtOAc) afforded 17 and this product was used without
further purification.
Methyl 4-[5-cyano-5-ethoxycarbonyl-1-(2,2,2-trifluoro-1-
dimethylhydrazonoethyl)pent-2-yl]benzoate (18).A sus-
pension of NaH (60% dispersion, 1.3 g, 32 mmol, 18
equiv) in anhydrous DMF (20 mL) at 0 ꢁC was treated
dropwise with ethyl cyanoacetate (3.9 mL, 37 mmol, 18
equiv).The reaction mixture was stirred at 0 ꢁC for 30
min, forming the sodium salt as a clear solution.This
anion was treated with a solution of 17 (crude, 0.76 g) in
anhydrous DMF (20 mL).The reaction mixture was
stirred at 25ÂꢁC for 2 h before being quenched by the
addition of saturated aqueous NH4Cl (5 mL).The
reaction mixture was diluted with EtOAc (50 mL) and
washed with H2O (3Â50 mL) and saturated aqueous
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(2,2,2-trifluoroacetyl)but-2-yl]benzoyl}-L-ꢀ-glutamyl-L-
ꢀ-glutamyl-L-ꢀ-glutamyl-L-ꢀ-glutamyl-L-ꢀ-glutamic acid
(6). A solution of 14 (13.6 mg, 0.001 mmol) in CHCl3
(0.8 mL), cooled to 0 ꢁC, was treated with trifluoroacetic
acid (0.2 mL). The solution was stirred 25 ꢁC for 12 h
before being concentrated under reduced pressure.The
solid residue was triturated with ether and dried in
vacuo to give 6–CF3CO2H (8.0 mg, 100%) as a tan
NaCl (50 mL).The organic layer was dried (Na SO4),
2
filtered, and concentrated under reduced pressure.The
excess ethyl cyanoacetate was distilled off and the resi-
dual product was purified by chromatography (SiO2, 7:1
hexanes/EtOAc) affording 18 (0.39 g, 28% from 16) as a
yellow oil: 1H NMR (CDCl3, 400 MHz) d 7.98 (d, J=8.5
Hz, 2H), 7.35 (d, J=8.2 Hz, 2H), 4.47 (t, J=7.7 Hz, 1H),
4.35 (m, 2H), 3.86 (s, 3H), 3.49 (m, 2H), 2.59 (s, 6H),
1.96 (m, 9H), 1.37 (m, 5H); MALDIFTMS (DHB) m/z
456.2101 (M+H+, C22H29F3N3O4 requires 456.2105).
1
solid: H NMR (CD3OD, 400 MHz) d 7.81 (m, 2H),
7.36 (m, 2H), 4.64 (s, 1H), 4.41 (m, 5H), 2.41 (m, 26H);
MALDIFTMS (DHB) m/z 1044.3406 (M+H+,
C42H53F3N9O19 requires 1044.3404).
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(2,2,2-trifluoro-1-hydroxymethyl)but-2-yl]benzoyl}-L-
ꢀ-glutamyl-L-ꢀ-glutamyl-L-ꢀ-glutamyl-L-ꢀ-glutamyl-L-
ꢀ-glutamic acid (15). A solution of 14 (7.9 mg, 0.0057
mmol) in anhydrous CH3OH (0.5 mL) at À20 ꢁC was
treated with NaBH4 (1.5 mg, 0.004 mmol, 0.7 equiv).
The reaction mixture was stirred at À20 ꢁC for 30 min
before it was quenched by the addition of H2O (1 mL).
The mixture was diluted with EtOAc (5 mL) and
washed with H2O (2Â1 mL).The organic layer was
dried (Na2SO4), filtered, and concentrated under reduced
pressure.The resulting product was treated with 4 N
HCl–dioxane (2 mL) at 0 ꢁC, and the solution was
allowed to warm and stir at 25 ꢁC for 3 h.The reaction
was purged with N2 and then concentrated under
Methyl 4-[5-(2,4-diamino-6-oxo-1,6-dihydropyrimidin-5-
yl)-1-(2,2,2-trifluoro-1-dimethylhydrazonoethyl)pent-2-
yl]benzoate (19). Sodium metal (0.027 g, 1.16 mmol, 2.0
equiv) was added to anhydrous CH3OH (4 mL) and the
reaction mixture was stirred at 25 ꢁC for 10 min to gen-
erate NaOCH3. Guanidine–HCl (0.055 g, 0.58 mmol,
1.0 equiv) was added and the reaction mixture was stir-
red at 25 ꢁC for 30 min.Separately, 18 (0.39 g, 0.58
mmol) was dissolved in anhydrous CH3OH (2 mL) and
this solution was added quickly to the stirring reaction
mixture.The resulting reaction mixture was stirred at
reflux for 16 h.The reaction mixture was applied
directly to a SiO2 plug.Impurities were removed by
washing with 3:1 hexanes/EtOAc.The product was
subsequently eluted by washing with 10:1 CHCl3/
reduced pressure.Et O (1 mL) was added and a pre-
2
cipitate formed.The precipitate was collected, triturated
with Et2O (3Â1 mL), and dried in vacuo to give 15–HCl
1
CH3OH to afford 19 (0.18 g, 68%) as a tan solid: H
1
(6.6 mg, 100% from 14) as a yellow solid: H NMR
NMR (CD3OD, 400 MHz) d 7.96 (d, J=8.4 Hz, 2H),
7.41 (d, J=8.4 Hz, 2H), 4.80 (t, J=7.7 Hz, 1H), 3.88 (s,
3H), 2.65 (s, 6H), 2.34 (m, 3H), 1.52 (m, 5H); MAL-
(CD3OD, 400 MHz) d 7.78 (m, 2H), 7.37 (m, 2H), 4.41
(m, 6H), 2.41 (m, 26H); MALDIFTMS (DHB) m/z
1046.3563 (M+H+, C42H55F3N9O19 requires 1046.3561).
DIFTMS
C21H27F3N6O3Na requires 491.1989).
(DHB)
m/z
491.1996
(M+Na+,
Methyl 4-[5-bromo-1-(2,2,2-trifluoro-1-dimethylhydrazo-
noethyl)pent-2-yl]benzoate (17).NaH (60% dispersion,
0.13 g, 3.21 mmol, 1.0 equiv) was added to a stirred
solution of 1610 (0.87 g, 3.03 mmol) in anhydrous DMF
(15 mL) at 0 ꢁC.The solution was stirred at 0 ꢁC for 15
min. 1,4-Dibromobutane (2.20 mL, 18.4 mmol, 6.0
4-[5-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-1-
(2,2,2-trifluoro-acetyl)pent-2-yl]benzoic acid (20).A
solution of 19 (0.18 g, 0.39 mmol) in 3:1 CH3OH–H2O
(8 mL) was treated with LiOH–1H2O (0.054 g, 1.3
mmol, 3.3 equiv) and the reaction mixture was stirred at

J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
4519
25 ꢁC for 24 h.The reaction mixture was diluted with
H2O (10 mL) and the aqueous layer was washed with
EtOAc (10 mL).The aqueous layer was acidified to
pH=4 by the addition of 1 N aqueous HCl.The reac-
tion mixture was concentrated under reduced pressure
and the residue was treated with MeCN (3Â10 mL) to
remove traces of H2O to provide 20 (0.16 g, 100%)
mmol, 1.6 equiv) was added to a stirred solution of 23
(0.10 g, 0.40 mmol) in anhydrous DMF (2.5 mL) at
0 ꢁC.The solution was stirred at 0 ꢁC for 15 min.1,3-
Dibromopropane (0.25 mL, 2.5 mmol, 6.0 equiv) was
added quickly to the reaction and the cooling bath was
removed.The reaction mixture was stirred at 25 ꢁC for
2.5 h. The reaction was quenched by the addition of
saturated aqueous NH4Cl (15 mL).The reaction mix-
ture was partitioned between EtOAc (50 mL) and H2O
(40 mL).The organic layer was washed with H 2O (3Â50
mL) and saturated aqueous NaCl (1Â50 mL) followed
by concentration under reduced pressure.Chromato-
graphy (SiO2, 9:1 hexanes/EtOAc) afforded 24 (0.086 g,
1
which was used without further purification: H NMR
(CD3OD, 400 MHz) d 7.95 (m, 2H), 7.32 (m, 2H), 4.26
(m, 1H), 2.15 (t, J=6.8 Hz, 2H), 1.38 (m, 6H); MAL-
DIFTMS (DHB) m/z 413.1453 (M+H+, C18H20F3N4O4
requires 413.1431).
1
N-{4-[5-(2,4-Diamino-6(1H)-pyrimidinon-5-yl)-1-(2,2,2-
trifluoroacetyl)pent-2-yl]benzoyl}-L-ꢀ-glutamic acid di-
tert-butyl ester( 21). A solution of 20 (0.24 g, 0.58
mmol) and di-tert-butyl l-glutamate hydrochloride
(0.26 g, 0.88 mmol, 1.5 equiv) in DMF (8 mL) was
treated with NaHCO3 (0.30 g, 3.62 mmol, 6.0 equiv)
followed by EDCI (0.36 g, 1.86 mmol, 3.2 equiv). The
reaction mixture was stirred at 25 ꢁC for 48 h.The
reaction mixture was partitioned between EtOAc (20
mL) and H2O (20 mL).The organic layer was washed
with saturated aqueous NaHCO3 (2Â20 mL) and satu-
rated aqueous NaCl (1Â20 mL) followed by concen-
58%) as a yellow oil: H NMR (CDCl3, 400 MHz) d
7.98 (d, J=7.9 Hz, 2H), 7.34 (d, J=7.9 Hz, 2H), 3.90 (s,
3H), 3.50 (t, J=7.1 Hz, 1H), 3.38 (t, J=6.2 Hz, 2H),
2.19 (m, 2H), 1.83 (m, 2H), 1.37 (m, 9H); MAL-
DIFTMS (DHB) m/z 371.0850 (M+H+, C17H24BrO4
requires 371.0852).
Methyl 4-[4-cyano-5-ethoxycarbonyl-1-(tert-butylbutoxy-
carbonylethyl)but-2-yl]benzoate (25).A suspension of
NaH (60% dispersion, 29 mg, 0.72 mmol, 3.0 equiv) in
anhydrous DMF (3 mL) at 0 ꢁC was treated dropwise
with ethyl cyanoacetate (75 mL, 0.70 mmol, 3.0 equiv).
The reaction mixture was stirred at 0 ꢁC for 30 min,
forming the sodium salt as a clear solution.This anion
was treated with a solution of 24 (86 mg, 0.23 mmol) in
anhydrous DMF (3 mL).The reaction mixture was
stirred at 25 ꢁC for 2 h before being quenched by the
addition of saturated aqueous NH4Cl (1 mL).The
reaction mixture was diluted with EtOAc (10 mL) and
washed with H2O (3Â10 mL) and saturated aqueous
tration under reduced pressure.Chromatography (SiO ,
2
10:1 CHCl3/CH3OH) afforded 21 (0.066 g, 17%) as a
1
yellow solid: H NMR (CD3OD, 400 MHz) d 7.71 (m,
2H), 7.37 (m, 2H), 4.50 (m, 1H), 3.83 (m, 1H), 2.42 (m,
8H), 1.43 (m, 18H); MALDIFTMS (DHB) m/z 676.2934
(M+Na+, C31H42F3N5O7Na requires 676.2928).
N-{4-[5-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(2,2,2-trifluoroacetyl)pent-2-yl]benzoyl}-L-ꢀ-glutamic
acid (7). A solution of 21 (53.4 mg, 0.08 mmol) in
CHCl3 (4 mL) cooled to 0 ꢁC was treated with tri-
fluoroacetic acid (1 mL).The reaction mixture was
allowed to warm and stirred at 25 ꢁC for 16 h.The
reaction was concentrated under reduced pressure.Et 2O
(1 mL) was added and a precipitate formed.The
precipitate was collected, triturated with Et2O (3Â1 mL)
and dried in vacuo to give 7–CF3CO2H (53.9 mg,
100%) as a white solid: 1H NMR (CD3OD, 400 MHz) d
7.88 (m, 2H), 7.36 (m, 2H), 4.64 (m, 1H), 4.27 (m, 1H),
2.33 (m, 12H); MALDIFTMS (DHB) m/z 542.1853
(M+H+, C23H27F3N5O7 requires 542.1857).
NaCl (10 mL).The organic layer was dried (Na SO4),
2
filtered, and concentrated under reduced pressure.The
excess ethyl cyanoacetate was distilled off and the resi-
dual product was purified by chromatography (SiO2, 4:1
hexanes/EtOAc) affording 25 (68 mg, 73%) as a yellow
1
oil: H NMR (CDCl3, 400 MHz) d 7.98 (d, J=8.3 Hz,
2H), 7.33 (d, J=8.3 Hz, 2H), 4.26 (q, J=7.7 Hz, 2H),
3.89 (s, 3H), 3.48 (m, 2H), 1.79 (m, 6H), 1.36 (s, 9H),
1.29 (t, J=7.7 Hz, 3H); MALDIFTMS (DHB) m/z
426.1874 (M+Na+, C22H29NO6Na requires 426.1887).
Methyl 4-[4-(2,4-diamino-6-oxo-1,6-dihydropyrimidin-5-
yl)-1-(tert-butylbutoxycarbonylethyl)but-2-yl]benzoate
(26). Sodium methoxide (22 mg, 0.41 mmol, 2.4 equiv)
was added to anhydrous CH3OH (1 mL) and the reac-
tion mixture was stirred at 25 ꢁC for 10 min.Guani-
dine–HCl (20 mg, 0.21 mmol, 1.2 equiv) was added
and the reaction mixture was stirred at 25 ꢁC for 30
min.Separately, 25 (68 mg, 0.17 mmol) was dissolved
in anhydrous CH3OH (1 mL) and this solution was
added quickly to the stirring reaction mixture.The
resulting reaction mixture was stirred at reflux for 16 h.
The reaction mixture was applied directly to a SiO2
plug.Impurities were removed by washing with 3:1
hexanes/EtOAc.The product was subsequently eluted
by washing with 10:1 CHCl3/CH3OH to afford 26 (36
Methyl 4-(tert-butylbutoxycarbonylethyl)benzoate (23)18.
Known acid 2217 (1.03 g, 5.31 mmol) was dissolved in
anhydrous CH2Cl2 (200 mL). tert-Butanol (1.87 g, 25.3
mmol) and N,N-dimethylaminopyridine (1.72 g, 8.34
mmol) were added slowly and the solution was stirred at
25 ꢁC for 10 min. DCC (0.40 g, 3.30 mmol) was added
at 0 ꢁC and the reaction mixture was stirred at 25 ꢁC for
20 h.The mixture was concentrated under reduced
pressure and the product was purified by chromato-
graphy (SiO2, 7:1 hexanes/EtOAc) afforded 23 (0.58 g,
1
44%) as a yellow oil: H NMR (CDCl3, 400 MHz) d
8.06 (d, J=8.0 Hz, 2H), 7.42 (d, J=8.0 Hz, 2H), 3.98 (s,
3H), 3.66 (s, 2H), 1.49 (s, 9H).
1
mg, 51%) as a tan solid: H NMR (CD3OD, 400 MHz)
d 7.96 (d, J=8.0 Hz, 2H), 7.42 (d, J=8.0 Hz, 2H), 3.90
(s, 3H), 3.63 (m, 1H), 2.33 (t, J=7.4 Hz, 1H), 2.10 (s,
1H), 1.70 (m, 2H), 1.41 (m, 11H); MALDIFTMS
Methyl 4-[4-bromo-1-(tert-butylbutoxycarbonylethyl)but-
2-yl]benzoate (24). NaH (60% dispersion, 0.025 g, 0.63

4520
J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
(DHB) m/z 439.1941 (M+Na+, C21H28N4O5Na
requires 439.1952).
98% purity when assessed by SDS-PAGE.The purified
protein was used in the inhibition assays, cytotoxic
assays and crystallization experiments.
4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-1-
(tert-butylbutoxycarbonylethyl)but-2-yl]benzoic acid (27).
A solution of 26 (10 mg, 0.02 mmol) in 3:1 THF–
CH3OH (0.4 mL) was treated with LiOH–1H2O (3.5
mg, 0.08 mmol, 3.0 equiv) dissolved in H2O (0.1 mL)
and the reaction mixture was stirred at 25 ꢁC for 6 h.
The aqueous layer was acidified by the addition of 1 N
aqueous HCl (0.1 mL). The reaction mixture was parti-
tioned between EtOAc (5 mL) and H2O (5 mL).The
organic layer was dried on Na2SO4 followed by con-
centration under reduced pressure to provide 27 as a tan
GAR and AICAR Tfase inhibition assay. The Ki values
for the folate analogues were measured as previously
described.10 For the GAR Tfase inhibition assay,
briefly, each compound was dissolved in DMSO and
then diluted in assay buffer.The concentration of
DMSO did not affect enzyme activity.Thus, all assays
were conducted by mixing 10 mM of fDDF, 20 mM of
inhibitor in total volumn of 1 mL buffer (0.1 M HEPES,
pH 7.5) at 26 ꢁC, and the reaction initiated by the addi-
tion of 76 nM E. coli or rh GAR Tfase.The assay
monitors the deformylation of fDDF (Áe=18.9 mMÀ1
cmÀ1 at 295 nm) resulting from the transfer of the
formyl group to GAR.If the inhibitor was found to be
active, a series of 1/ni versus 1/[GAR] at different, fixed
concentrations of I (4, 8, 12, 16, 20, 32 mM) were
generated in order to determine Ki using the Michaelis–
Menton equation for competitive inhibition.The
results for the inhibition assays are summarized in
Table 1.
1
solid (8.0 mg, 100%): H NMR (CD3OD, 400 MHz) d
8.04 (m, 2H), 7.48 (m, 2H), 3.67 (m, 1H), 2.08 (m, 4H),
1.46 (m, 11H); MALDIFTMS (DHB) m/z 425.1779
(M+Na+, C20H26N4O5Na requires 425.1795).
N-{4-[4-(2,4-Diamino-6(1H)-pyrimidinon-5-yl)-1-(tert-
butylbutoxycarbonylethyl)but-2-yl]benzoyl}-L-ꢀ-glutamic
acid di-tert-butyl ester( 28). A solution of 27 (4.0 mg,
0.011 mmol) and di-tert-butyl l-glutamate hydrochlor-
ide (7.4 mg, 0.025 mmol, 2 equiv) in DMF (0.1 mL) was
treated with NaHCO3 (3.0 mg, 0.036 mmol, 3.0 equiv)
followed by EDCI (4.4 mg, 0.023 mmol, 2.0 equiv). The
reaction mixture was stirred at 25 ꢁC for 48 h.The
reaction mixture was partitioned between EtOAc (1.0
mL) and H2O (1.0 mL). The organic layer was washed
with saturated aqueous NaHCO3 (2Â1.0 mL) and satu-
rated aqueous NaCl (1Â1.0 mL) followed by concen-
Cytotoxic assay. The cytotoxic activity of the com-
pounds was measured using CCRF–CEM human leu-
kemia cells as described previously.10
Acknowledgements
tration under reduced pressure.Chromatography (SiO ,
We gratefully acknowledge the financial support of the
National Institutes of Health (CA 63536, to D.L.B.,
I.A.W., and S.J.B.), the Skaggs Institute for Chemical
Biology, and a Skaggs predoctoral fellowship (Y.Z.).
2
10:1 CHCl3/CH3OH) afforded 28 (1.9 mg, 29%) as a
1
yellow solid: H NMR (CD3OD, 400 MHz) d 7.71 (m,
2H), 7.36 (m, 2H), 4.18 (m, 2H), 2.33 (m, 9H), 1.49 (m,
28H); MALDIFTMS (DHB) m/z 666.3464 (M+Na+,
C33H49N5O8Na requires 666.3473).
References and Notes
N-{4-[4-(2,4-Diamino-6-oxo-1,6-dihydropyrimidin-5-yl)-
1-(carboxyethyl)but-2-yl]benzoyl}-L-ꢀ-glutamic acid (8).
A solution of 28 (2.8 mg, 0.0044 mmol) in CHCl3 (0.3
mL) cooled to 0 ꢁC was treated with trifluoroacetic acid
(0.1 mL). The reaction mixture was allowed to warm
and stirred at 25 ꢁC for 16 h.The reaction was con-
1.Warren, L;.Buchanan, J.M.
J. Biol. Chem. 1957, 229, 613.
Buchanan, J.M;. Hartman, S.C. Adv. Enzymol. 1959, 21, 199.
Warren, M. S.; Mattia, K. M.; Marolewski, A. E.; Benkovic,
S.J. Pure Appl. Chem. 1996, 68, 2029.
2.Jackson, R.C;.Harkrader, R.J.In
Nucleosides and Cancer
Treatment; Tattersall, M. H. N., Fox, R. M., Eds.; Academic:
Sydney, 1981; p 18.Review: Christopherson, R.I;. Lyons,
S.D;. Wilson, P.K. Acc. Chem. Res. 2002, 35, 961.
3. Beardsley, G. P.; Moroson, B. A.; Taylor, E. C.; Moran,
R.G. J. Biol. Chem. 1989, 264, 328.
centrated under reduced pressure.Et O (0.5 mL) was
2
added and a precipitate formed.The precipitate was
collected, triturated with Et2O (3Â0.5 mL) and dried in
vacuo to give 8-CF3CO2H (2.8 mg, 100%) as a white
1
solid: H NMR (CD3OD, 400 MHz) d 7.80 (m, 2H),
4.For related analogues and studies, see: Caperelli, C.A. J.
Med. Chem. 1987, 30, 1254. Taylor, E. C.; Hamby, J. M.;
Shih, C.; Grindey, G. B.; Rinzel, S. M.; Beardsley, G. P.;
Moran, R.G. J. Med. Chem. 1989, 32, 1517.Taylor, E.C;.
Wong, G.S.K. J. Org. Chem. 1989, 54, 3618.Taylor, E.C;.
Harrington, P.M;. Warner, J.C. Heterocycles 1988, 27, 1925.
Taylor, E.C.; Harrington, P.M. J. Org. Chem. 1990, 55, 3222.
Moran, R. G.; Baldwin, S. W.; Taylor, E. C.; Shih, C. J. Biol.
Chem. 1989, 264, 21047. Kelley, J. L.; McLean, E. W.; Cohn,
N. K.; Edelstein, M. P.; Duch, D. S.; Smith, G. K.; Hanlon,
M.H;. Ferone, R. J. Med. Chem. 1990, 33, 561.Taylor, E.C;.
Gillespie, P.; Patel, M. J. Org. Chem. 1992, 57, 3218.Taylor,
E. C.; Schrader, T. H.; Walensky, L. D. Tetrahedron 1992, 48,
19. Bigham, E. C.; Hodson, S. J.; Mallory, W. R.; Wilson, D.;
Duch, D.S.; Smith, G.K.; Ferone, R. J. Med. Chem. 1992, 35,
1399. Taylor, E. C.; Kuhnt, D.; Shih, C.; Rinzel, S. M.; Grin-
7.42 (m, 2H), 4.62 (m, 1H), 3.66 (m, 1H), 2.33 (m, 8H);
MALDIFTMS (DHB) m/z 498.1590 (M+Na+,
C21H25N5O8Na requires 498.1595).
Recombinant human GAR Tfase protein preparation.
The recombinant human GAR Tfase construct includes
residues 808–1010 from human trifunctional enzyme
(purD-purM-purN).The gene was subcloned into
pet22b vector using NdeI/XhoI cloning site with hexa-
histine tag at the C-terminus.The plasmid was trans-
formed into the E. coli expression strain BL21 (DE3)
Gold.The protein was expressed and purified as descri-
bed previously.21 The yield of the protein is greater than
30 mg per liter LB broth after purification with at lease

J. Desharnais et al. / Bioorg. Med. Chem. 11 (2003) 4511–4521
dey, G. B.; Barredo, J.; Jannatipour, M.; Moran, R. G. J. J.H;. Benkovic, S.J.
4521
Bioorg. Med. Chem. 2000, 8, 1075.
Med. Chem. 1992, 35, 4450. Piper, J. R.; Johnson, C. A.;
Otter, G.M;. Sirotnak, F.M. J. Med. Chem. 1992, 35, 3002.
Shih, C.; Gossett, L. S.; Worzalla, J. F.; Rinzel, S. M.; Grin-
dey, G. B.; Harrington, P. M.; Taylor, E. C. J. Med. Chem.
1992, 35, 1109. Shih, C.; Grindey, G. B.; Taylor, E. C.; Har-
rington, P.M. Bioorg. Med. Chem. Lett. 1992, 2, 339.Shih, C.;
Hu, Y.; Gossett, L. S.; Habeck, L. L.; Mendelsohn, L. G.;
Grindey, G.B. Bioorg. Med. Chem. Lett. 1993, 3, 2657.Tay-
lor, E.C.In Chemistry and Biology of Pteridines and Folates;
Ayling, J. E., Nair, M. G., Baugh, C. M., Eds.; Plenum: New
York, 1993; p 387.Durucasu, I. Heterocycles 1993, 35, 1527.
Barnett, C. J.; Wilson, T. M.; Wendel, S. R.; Winningham,
M.J.; Deeter, J.B. J. Org. Chem. 1994, 59, 7038.Taylor, E.C.;
Yoon, C.-M.; Hamby, J. M. J. Org. Chem. 1994, 59, 7092.
Taylor, E. C.; Yoon, C.-M. J. Org. Chem. 1994, 59, 7096.
Pizzorno, G.; Moroson, B. A.; Cashmore, A. R.; Russello, O.;
Mayer, J. R.; Galivan, J.; Bunni, M. A.; Priest, D. G.;
Beardsley, G.P. Cancer Res. 1995, 55, 566.Piper, J.R;.
Ramamurthy, B.; Johnson, C. A.; Otter, G. M.; Sirotnak,
F.M. J. Med. Chem. 1996, 39, 614.Taylor, E.C;. Young,
W.B.; Spanka, C. J. Org. Chem. 1996, 61, 1261.Gossett, L.S.;
Habeck, L. L.; Gates, S. B.; Andis, S. L.; Worzalla, J. F.;
Schultz, R. M.; Mendelsohn, L. G.; Kohler, W.; Ratnam, M.;
Grindey, G.B;. Shih, C. Bioorg. Med. Chem. Lett. 1996, 6,
Boger, D. L.; Marsilje, T. H.; Castro, R. A.; Hedrick, M. P.;
Jin, Q.; Baker, S. J.; Shim, J. H.; Benkovic, S. J. Bioorg. Med.
Chem. Lett. 2000, 10, 1471.
6. Marsilje, T. H.; Labroli, M. A.; Hedrick, M. P.; Jin, Q.;
Desharnais, J.; Baker, S. J.; Gooljarsingh, L. T.; Ramcharan,
J.; Tavassoli, A.; Zheng, Y.; Wilson, I. A.; Beardsley, G. P.;
Benkovic, S.J;. Boger, D.L.
2739.
Bioorg. Med. Chem. 2002, 10,
7. Marsilje, T. H.; Hedrick, M. P.; Desharnais, J.; Tavassoli,
A.; Ramcharan, J.; Zhang, Y.; Wilson, I. A.; Benkovic, S. J.;
Boger, D.L. Bioorg. Med. Chem. Preceding article.
8. Marsilje, T. H.; Hedrick, M. P.; Desharnais, J.; Capps, K.;
Tavassoli, A.; Ramcharan, J.; Zhang, Y.; Wilson, I. A.; Ben-
kovic, S.J.; Boger, D.L. Bioorg. Med. Chem. Preceding article.
9. Greasley, S. E.; Yamashita, M. M.; Cai, H.; Benkovic, S. J.;
Boger, D.L;. Wilson, I.A. Biochemistry 1999, 38, 16783.For
a unrelated, but analogous enzyme-assemble multisubstrate
adduct formation within the active site of GAR Tfase, see:
Greasley, S. E.; Marsilje, T. H.; Cai, H.; Baker, S.; Benkovic,
S. J.; Boger, D. L.; Wilson, I. A. Biochemistry 2001, 40, 13538.
10. Zhang, Y.; Desharnais, J.; Marsilje, T. H.; Li, C.;
Hedrick, M. P.; Gooljarsingh, L. T.; Tavassoli, A.; Benkovic,
S. J.; Olson, A. J.; Boger, D. L.; Wilson, I. A. Biochemistry
2003, 42, 6043.
473.Taylor, E.C;. Dowling, J.E.
Bioorg. Med. Chem. Lett.
11. Hanlon, M. H.; Ferone, R.; Mullin, R. J.; Keith, B. R.
Cancer Res. 1990, 50, 3207.
12. Baldwin, S. W.; Tse, A.; Gossett, L. S.; Taylor, E. C.;
Rosowsky, A.; Shih, C.; Moran, R. G. Biochemistry 1991, 30,
1997.
13. Habeck, L. L.; Leitner, T. A.; Shackelford, K. A.; Gossett,
L. S.; Schultz, R. M.; Andis, S. L.; Shih, C.; Grindey, G. B.;
Mendelsohn, L.G. Cancer Res. 1994, 54, 1021.
14. Shih, C.; Chen, V. J.; Gossett, L. S.; Gates, S. B.; MacK-
ellar, W. C.; Habeck, L. L.; Sheckelford, K. A.; Mendelsohn,
L. G.; Soose, D. J.; Patel, V. F.; Andis, S. L.; Bewley, J. R.;
Rayl, E. A.; Moroson, B. A.; Beardsley, G. P.; Kohler, W.;
Ratnam, M.; Schultz, R. M. Cancer Res. 1997, 57, 1116.
Mendelsohn, L. G.; Shih, C.; Schultz, R. M.; Worzalla, J. F.
Invest. New Drugs 1996, 14, 287.
1997, 7, 453. Taylor, E. C.; Zhou, P.; Jennings, L. D.; Mao,
Z.; Hu, B.; Jun, J.-G. Tetrahedron Lett. 1997, 38, 521.Varney,
M. D.; Palmer, C. L.; Romines, W. H., III; Boritzki, T.;
Margosiak, S. A.; Almassy, R.; Janson, C. A.; Bartlett, C.;
Howland, E.J;. Ferre, R. J. Med. Chem. 1997, 40, 2502.Shih,
C.; Habeck, L. L.; Mendelsohn, L. G.; Chen, V. J.; Schultz,
R.M. Adv. Enz. Reg. 1998, 38, 135. Wall, M.; Shim, J. H.;
Benkovic, S.J. J. Med. Chem. 1999, 42, 3421.Taylor, E.C;.
Chaudhuri, R.P;. Watson, S.E. Tetrahedron 1999, 55, 1631.
Gossett, L. S.; Habeck, L. L.; Shackelford, K. A.; Mendel-
sohn, L. G.; Gates, S. B.; Worzalla, J. F.; Self, T. D.; Theo-
bald, K. S.; Andis, S. L.; Schultz, R. M.; Shih, C. Bioorg. Med.
Chem. Lett. 1999, 9, 75. Read, M. W.; Miller, M. L.; Ray, P. S.
Tetrahedron 1999, 55, 373.Taylor, E.C;. Wang, Y.
Hetero-
cycles 1998, 48, 1537. Borrell, J. L.; Teixido, J.; Martinez-Tei-
pel, B.; Matallana, J. L.; Copete, M. T.; Llimargas, A.;
Garcia, E. J. Med. Chem. 1998, 41, 3539.Roberts, J.D;. Shi-
bata, S.; Spicer, D. V.; McLeod, H. L.; Tombes, M. B.; Kyle,
B.; Carroll, M.; Sheedy, B.; Collier, M. A.; Pithavala, Y. K.;
15.McGuire, J.J;. Bertino, J.R.
39, 19.
Mol. Cell. Biochem. 1981,
16.Styles, V.L;.Kelley, J.L.
1809.
J. Heterocyclic Chem. 1990, 27,
17. Grell, W.; Hurnaus, R.; Griss, G.; Sauter, R.; Rupprecht,
E.; Mark, M.; Luger, P.; Nar, H.; Wittneben, H.; Mueller, P.
J. Med. Chem. 1998, 41, 5219.
Paradiso, L.J;. Clendeninn, N.J.
Cancer Chemother. Phar-
macol. 2000, 45, 423. Borrell, J. I.; Teixido, J.; Matallana, J. L.;
Martinez-Teipel, B.; Colominas, C.; Costa, M.; Balcells, M.;
Schuler, E.; Castillo, M. J. J. Med. Chem. 2001, 44, 2366.
5. Boger, D. L.; Haynes, N.-E.; Kitos, P. A.; Warren, M. S.;
Ramcharan, J.; Marolewski, A. E.; Benkovic, S. J. Bioorg.
Med. Chem. 1997, 5, 1817. Boger, D. L.; Haynes, N.-E.; War-
ren, M. S.; Gooljarsingh, L. T.; Ramcharan, J.; Kitos, P. A.;
Marolewski, A.E;. Benkovic, S.J. Bioorg. Med. Chem. 1997,
5, 1831. Boger, D. L.; Haynes, N.-E.; Warren, M. S.; Ram-
charan, J.; Kitos, P. A.; Benkovic, S. J. Bioorg. Med. Chem.
1997, 5, 1839. Boger, D. L.; Haynes, N.-E.; Warren, M. S.;
Ramcharan, J.; Marolewski, A. E.; Kitos, P. A.; Benkovic,
S.J. Bioorg. Med. Chem. 1997, 5, 1847.Boger, D.L;. Haynes,
N.-E.; Warren, M. S.; Ramcharan, J.; Kitos, P. A.; Benkovic,
S.J. Bioorg. Med. Chem. 1997, 5, 1853.Boger, D.L;.
Kochanny, M. J.; Cai, H.; Wyatt, D.; Kitos, P. A.; Warren,
M. S.; Ramcharan, L. T.; Gooljarsingh, L. T.; Benkovic, S. J.
Bioorg. Med. Chem. 1998, 6, 643. Boger, D. L.; Labroli, M. A.;
Marsilje, T. H.; Jin, Q.; Hedrick, M. P.; Baker, S. J.; Shim,
18.Agnelli, F;.Sulikowski, G.A.
8807.
19.Taylor, E.C;. Harrington, P.M.
1169. Bigham, E.; Duch, D.; Ferone, R.; Kelley, J.; Smith, G.
Chem. Biol. Pteridines 1989, 961.Sokoloski, J.A;. Beardsley,
Tetrahedron Lett. 1998, 39,
Heterocycles 1989, 28,
G.P;. Sartorelli, A.C.
Cancer Chemother. Pharmacol. 1991,
28, 39. Mullin, R. J.; Keith, B. R.; Bigham, E. C.; Duch, D. S.;
Ferone, R.; Heath, L. S.; Singer, S.; Waters, K. A.; Wilson,
H.R. Biochem. Pharmacol. 1992, 43, 1627.Shih, C;. Gossett,
L. S.; Worzalla, J. F.; Rinzel, S. M.; Grindey, G. B.; Harring-
ton, P.M;. Taylor, E.C. J. Med. Chem. 1992, 35, 1109.Big-
ham, E. C.; Hodson, S. J.; Mallory, W. R.; Wilson, D.; Duch,
D.S;. Smith, G.K;. Ferone, R. J. Med. Chem. 1992, 35, 1399.
20.Jansen, G;. Westerhof, G.R;. Kathmann, I;. Rademaker,
B.C.; Rijksen, G.; Schornagel, J.H. Cancer Res. 1989, 49, 2455.
21. Zhang, Y.; Desharnais, J.; Greasley, S. E.; Beardsley,
G. P.; Boger, D. L.; Wilson, I. A. Biochemistry 2002, 41,
14206.