Probing for a hydrophobic a binding register in prostate-specific membrane antigen with phenylalkylphosphonamidates

Article abstract of DOI:10.1016/j.bmc.2004.06.031

A series of phenylalkylphosphonamidate derivatives of glutamic acid were synthesized and evaluated for their inhibitory potencies against PSMA. The greatest inhibitory potency was observed for the inhibitors 1f (n = 5) and 1g (n = 6) suggesting the presen

Full text of DOI:10.1016/j.bmc.2004.06.031

Bioorganic & Medicinal Chemistry 12 (2004) 4969–4979
Probing for a hydrophobic a binding register in
prostate-specific membrane antigen with
phenylalkylphosphonamidates
Jack Maung,a Jeremy P. Mallari,a Teri A. Girtsman,a Lisa Y. Wu,a Jennifer A. Rowley,a
Nicholas M. Santiago,a Alan N. Brunelleb and Clifford E. Berkmana,*
aDepartment of Chemistry and Biochemistry, San Francisco State University, 1600 Holloway Ave., San Francisco, CA 94132, USA
bSCOLR, Inc., 3625 132nd Ave. SE, suite 300, Bellevue, WA 98006, USA
Received 29 April 2004; revised 22 June 2004; accepted 23 June 2004
Available online 27 July 2004
Abstract—To explore for the existence of an auxiliary hydrophobic binding register remote from the active site of PSMA a series of
phenylalkylphosphonamidate derivatives of glutamic acid were synthesized and evaluated for their inhibitory potencies against
PSMA. Both the phenyl- and benzylphosphonamidates (1a and 1b) exhibited only modest inhibitory potency against. The phenethyl
analog 1c was intermediate in inhibitory potency while inhibitors possessing a longer alkyl tether from the phenyl ring, resulted in
markedly improved Ki values. The greatest inhibitory potency was obtained for the inhibitors in which the phenyl ring was extended
furthest from the central phosphorus (1f, n=5 and 1g, n=6). The slightly serrated pattern that emerged as the alkyl tether increased
from three to six methylene units suggests that inhibitory potency is not simply correlated to increased hydrophobicity imparted by
the phenylalkyl chain, but rather that one or more hydrophobic binding registers may exist remote from the substrate recognition
architecture in the active site of PSMA.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
II (GCP II), is a 750-amino acid type II membrane gly-
coprotein4 and was discovered during the development
of the LNCaP cell line; one which retains most of the
known features of prostate cancer.5
Prostate cancer remains the most common male malig-
nancy and the second most common cause of cancer-
related mortality in most Western societies.1 Its inci-
dence is associated with age, family history, and life style
factors,2 and varies strikingly among ethnic, racial, and
national groups with noteworthy high rates of both inci-
dence and mortality among African Americans.3 As
basic research provides an ever-growing understanding
of this disease, the strategies for preventing, diagnosing,
and treating prostate cancer are being reshaped. One
notable discovery has been the identification of an
over-expressed membrane-bound cell surface protein
on prostate cancer cells, namely, prostate-specific mem-
brane antigen (PSMA). PSMA, also known as folate
hydrolase I (FOLH1) and glutamate carboxypeptidase
Although PSMA is primarily expressed in normal hu-
man prostate epithelium, the importance of this enzyme
lies with the fact that it is upregulated and strongly
expressed in prostate cancer cells, including those of
the metastatic disease state.6 Recent studies have
demonstrated PSMA expression in the endothelium of
tumor-associated neovasculature of multiple nonpro-
static solid malignancies,7 including metastatic renal car-
cinoma.8 It is not surprising that PSMA has attracted a
great deal of attention as a target for immunotherapy.9
In addition to its immunological importance, PSMA is
also reported to possess three, yet poorly understood,
enzymatic activities: the hydrolytic cleavage and libera-
tion of glutamic acid from both c-glutamyl derivatives
of folic acid10 and the neuropeptide NAAG11 (N-acetyl-
aspartylglutamate) as shown in Figure 1, and dipeptidyl
peptidase IV activity.12 PSMA is highly homologous
to NAALADase (N-acetylated alpha-linked L-amino
Keywords: PSMA; Glutamate carboxypeptidase II; Phosphonamidate;
HPLC; c-Diglutamate.
Corresponding author. Tel.: +1-415-338-6495; fax: +1-415-338-
2384; e-mail: cberkman@sfsu.edu
*
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.06.031

4970
J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
–
CO2
O
H
N
H
N
O
P
–
–
–
CO2
CO2
CO2
N
H
- O
O
–
hydrophobic
binding site?
CO2
2+
NAAG
Zn
–
O
CO2
OH
N
N
H
H
N
–
H
N
O
P
–
–
CO2
N
N
CO2
CO2
N
N
H
n
O
- O
H2N
Folic Acid
–
CO2
n = 0 - 6
1a-1g
Figure 1. Substrates of PSMA and a binding model for a hydrophobic probing inhibitor.
dipeptidase) which is specifically characterized by its
ability to hydrolyze the neuropeptide NAAG.13 How-
ever, in contrast to NAALADase, which has been exten-
sively studied due to its presumed regulatory role in
glutamate neurotransmission, questions of medical
interest remain to be answered for PSMA, especially
with regard to its role in prostate cancer. It is our
hypothesis that these enzymatic activities of PSMA
can be exploited for chemotherapeutic strategies such
as the selective inhibition by small molecule inhibitors.
was selected as the nucleophile for alkylation to provide
the phenylalkylphosphonate homologs 3. As such,
Michaelis–Becker alkylation of dibenzyl phosphite with
the readily available phenylalkyl bromides (Ph(CH2)n
Br, n=1, 3, or 5) provided the phenylalkylphosphonates
3b,d, and 3f in good yield. Not surprising, attempts at
alkylating dibenzyl phosphite with 2-phenylethylbro-
mide resulted in low yields with concomitant formation
of styrene. Employing the same alkyl halides as above,
the phenylalkylphosphonates 3c,e, and 3g were prepared
by alkylation of dibenzyl methylphosphonate 2. Care
was taken to minimize the extent of dimerization of
the methylphosphonate.16
There is no crystal structure of PSMA as of yet, and
thus the development of specific inhibitors must rely
upon rational strategies to identify putative binding sites
within or near the architecture of its active site.
Although there have been a number of studies aimed
at the acquisition of inhibitors of NAALADase,14 there
have been far fewer studies on the inhibition of PSMA
by rationally designed, synthetic, small-molecule inhibi-
tors.15 The focus of this work was to identify the exist-
ence and relative location of hydrophobic binding site
remote from the catalytic center of PSMA. To this
end, we chose to prepare a series of phenylalkylphos-
phonamidate derivatives of glutamic acid (1a–g) Figure
1. The glutamic acid residue was incorporated to pro-
vide a sufficient substrate-recognition element to these
inhibitors while the phosphonyl center would bind as a
tetrahedral-intermediate analog to the active-site zinc
(II) ions. A phenyl ring was selected as a probe for a
putative remote hydrophobic binding site and as such,
was tethered by an increasing number of (0–6) methyl-
ene units from the central phosphorus.
Monodealkylation of the phenylalkylphosphonates was
easily achieved with NaI17 and following an acidic work-
up provided the phosphonic acids 4b–g in good yield.
18
Formation of the phosphonochloridate with (COCl)2
immediately followed by reaction with glutamic acid
dibenzyl ester [H-Glu(OBn)OBn] afforded the phospho-
namidate precursors 5b–g. The phenylphosphonamidate
5a was prepared directly from phenylphosphonic dichlo-
ride in a two-step, one-pot procedure we established pre-
viously.19 Hydrogenolysis of the phosphonamidate
precursors 5 in the presence of 1.5Mequiv of potassium
carbonate provided the inhibitors 1a–g in quantitative
yield. We observed considerable degradation of the
inhibitors during the deprotection step if the reaction
mixture was not buffered with carbonate. Knowing that
P–N bond is labile to acidic conditions, we presumed
that as the acidic moieties were revealed during hydro-
genolysis, their acidity compromised the chemical integ-
rity of the the P–N bond.
2. Results and discussion
With the acquisition of the hydrophobic inhibitor
probes 1a–g, we were interested in comparing their rela-
tive inhibitory potencies against purified PSMA rather
than crude LNCaP cell lysates, which contains consider-
able amounts of PSMA0, a truncated, soluble cytosolic
form of PSMA.20 Since PSMA0 exhibits dual active-site
kinetics as opposed to the single active-site kinetics of
The series of phenylalkylphosphonamidate inhibitors
(1a–g) of PSMA were prepared as outlined in Scheme
1. Depending upon the length of the alkyl chain linking
the phenyl probe to the central phosphorus, either di-
benzyl methylphosphonate (2) or dibenzyl phosphite

J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
4971
O
1. KHMDS
H
P
OBn
OBn
2.
Br
n
n = 1, 3, 5
O
P
O
P
1. n-BuLi
2.
CH3
CH3
O
P
Cl
OBn
BnOH
DIPEA
1. NaI
2. H+
Cl
OBn
n
OBn
n = 1-6 OBn
Br
n
2
n = 1, 3, 5
3b-3g
O
O
P
CO2Bn
O
P
CO2– K+
1. (COCl)2
2. H-Glu(OBn)OBzl
DIPEA, 1H-tetrazole
P
H2, Pd-C
K2CO3
n
n
N
H
N
H
OH
OBn
n
BnO
K+ –O
n = 1-6
n = 0-6
CO2Bn
CO2– K+
n = 0 - 6
5a-5g
4b-4g
1a-1g
1. BnOH, DIPEA
O
1H-tetrazole
2. H-Glu(OBn)OBzl
P
Cl
Cl
Scheme 1. Synthesis of phenylalkylphosphonamidate derivatives of glutamic acid.
PSMA,20 it was thought that employing the full-length
membrane bound form of the enzyme would simplify
the competitive inhibition kinetics. After obtaining the
membrane-bound protein fraction of LNCaP cell
lysates, active full-length PSMA was purified by anti-
body-affinity chromatography.
the structure of a known substrate for PSMA and bear
a chromophore for convenient detection by HPLC.
Knowing that folylpoly-c-glutamates are substrates for
PSMA, we prepared a simplified analog 7, which bears
a phenylazobenzoyl chromophore (Fig. 2). Using re-
versed-phase chromatography, the putative product 6
was cleanly separated from our synthetic substrate 7
with baseline resolution (Fig. 3). In terms of detection
limits, we noted that this HPLC-based assay was suffi-
ciently sensitive to conveniently detect concentrations
of our synthetic substrate 7 and product 6 as low as
10nM. Kinetic parameters were subsequently deter-
mined for diglutamate 7 with (Fig. 4) and found to be
within the broad range reported for NAAG (2.6–
540nM).21 Specifically, Km and Vmax for 7 were
In order to monitor PSMA enzymatic activity, we chose
to develop a convenient HPLC-based assay rather than
use the conventional radioisotopic assay. The latter
method involves incubating PSMA with radiolabeled
NAAG, loading the enzymatic reaction mixture onto
an ion-exchange column, selectively eluting and count-
ing enzymatically liberated, radiolabeled glutamate.
Cost and waste issues notwithstanding, this method is
labor intensive and prone to various introductions of
error. Our hypothesis was that we could develop a more
convenient alternative assay for monitoring PSMA
activity. The constraints we enforced upon our design
of a novel substrate were that it must be based upon
280nM and 622nmolminꢀ1 mgproteinꢀ1
,
respec-
tively. Indeed, the determined kinetic parameters for 7
with PSMA demonstrated the success of our design
and development of a convenient alternative assay for
monitoring PSMA assay. It should also be noted that,
O
CO2H
O
CO2H
N
H
N
H
H
N
N
CO2H
CO2H
N
CO2H
H2N
CO2H
CO2H
N
+
N
PSMA
O
6 PABG
7 PABGγG
Figure 2. PSMA-mediated hydrolysis of PABGcG to PABG.

4972
J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
uv
14
12
10
8
14
12
10
8
PABGγG
6
6
PABG
4
4
2
2
0
0
0
2
4
6
8
Minutes
Figure 3. HPLC chromatogram of PABGcG and PABG. Lichrosphere C18 150·4.6mm 5lm; 40% ACN, 60% phosphate (pH2.0, 25mM) 1.0mL/
min; 325nm.
0.5
0.4
0.3
y = ((m0*m1)/(m0+m2))
Value
0.62247
0.28842
Error
0.024286
0.031249
NA
m1
m2
0.2
0.1
0
Chisq 0.0006253
0.99816
R
NA
0
0.2
0.4
0.6
0.8
1
[substate] µM
Figure 4. Kinetics of PSMA with PABGcG 7.
purified PSMA maintained consistent enzymatic activity
through numerous freeze-thaw cycles.
the conventional addition of millimolar concentrations
of CoCl2. Although CoCl2 is typically included in en-
zyme incubation systems to enhance measurable enzy-
matic activity, we found that the inclusion of 1.0mM
CoCl2 in our assay resulted in a 27% decrease in enzy-
matic activity. Dixon analyses (Fig. 5) were conducted
To ensure that the measured inhibition of PSMA by
phosphonamidate inhibitors 1a–g was physiologically
relevant, enzyme activity assays were performed without

J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
4973
Figure 5. Dixon analysis of phosphonamidates 1a–g.* (*Each point represents the average of triplicate determinations.)
for inhibitors 1a–g and the resulting Ki values are pre-
sented in Table 1. The initial substrate concentration
for these assays was generally 1lM while inhibitor con-
centrations were varied within the range of 0.25–2.0lM
depending upon the potency of the inhibitors.
phorus (1f and 1g). These results are not inconsistent
with the observations of Tenniswood and co-workers
that phosphorus-containing inhibitors bearing large,
yet remote fluorescent moieties were relatively potent
inhibitors of PSMA/PSMA0 from crude LNCaP cell
lysates.15a
Both the phenyl- and benzylphosphonamidates (1a and
1b) exhibited only modest inhibitory potency against
PSMA. It is likely that the phenyl ring in these com-
pounds imparts steric congestion or unproductive inter-
actions in the active site and precludes efficient binding
of the inhibitors. The phenethyl analog 1c was interme-
diate in inhibitory potency while inhibitors possessing a
longer alkyl tether from the phenyl ring, resulted in
markedly improved Ki values. The greatest inhibitory
potency was obtained for the inhibitors in which the
phenyl ring was extended furthest from the central phos-
The slightly serrated pattern that emerged as the alkyl
tether increased from three to six methylene units sug-
gests that inhibitory potency is not simply correlated
to increased hydrophobicity imparted by the phenyl-
alkyl chain nor does the pattern suggest that increasing
the tether length allows for reduced unfavorable interac-
tions by incrementally displacing the phenyl ring from
an active site hosting a number of electrostatic interac-
tions. Rather, the data suggests that one or more hydro-
phobic binding registers may exist remote from the
substrate recognition architecture in the active site of
PSMA as proposed by Wiest and co-workers.22
Table 1. Inhibition of PSMA by phenylalkylphosphonamidates
homologues
In order to better appreciate the significance of a poten-
tial hydrophobic binding site to inhibitory potency of
PSMA, the Ki of N-phosphoglutamate23 (8, Fig. 6)
was determined. Not unlike the potent NAALADase
inhibitor 2-PMPA, N-phosphoglutamate was identified
as a very potent inhibitor of PSMA with a Ki value of
0.068 ( 0.015)nM. Lacking the strong chelating interac-
tions of the central dibasic phosphonamidate group with
H
N
O
P
–
CO2
n
–O
–
CO2
Inhibitor
n
Ki (nM)a
1a
1b
1c
1d
1e
1f
0
1
2
3
4
5
6
603 (34)
459 (12)
159 (7.9)
98 (5.1)
107 (12)
70 (5.3)
77 (3.8)
O
P
CO2 Li
Li+ –O
N
H
Li+ –O
CO2– Li
8 N-phosphoglutamate
1g
a Standard deviation in parenthesis.
Figure 6.

4974
J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
the catalytic zinc ions in PSMA, the phenylalkylphos-
phonamidates 1d–g likely interact with a remote hydro-
phobic register site to elicit their enhanced inhibitory
potency over the shorter homologs 1a–c.
(CDCl3) d: 11.30, 13.20, 67.61, 67.67, 128.42, 128.93,
129.13, 136.90, 136.98. 31P NMR (CDCl3) d: 31.93.
4.1.2. General procedure for phenylalkylphosphonates
3b,d, and 3f. A 0.5M solution of potassium hexame-
thyldisilylzane (3.81mL, 1.91mmol) in toluene was
added dropwise via syringe to a stirring solution of dib-
enzyl phosphite (421lL, 1.91mmol) in THF (5mL) at
0ꢁC under an Ar(g) atmosphere. After 1h, the appropri-
ate phenylalkylbromide (2.29mmol) was added via syr-
inge and the reaction mixture was refluxed 2.5h. The
solvent was removed under reduced pressure and flash
chromatography afforded the phenylalkylphosphonates
as pale yellow oils.
3. Conclusion
In summary, we have developed a convenient and sensi-
tive HPLC-based assay for monitoring the enzymatic
activity of PSMA using a novel synthetic c-diglutamate
derivative bearing a N-acyl chromophore. A comparison
of the individual inhibitory potencies of phenylalkyl-
phosphonamidate derivatives of glutamate supports
the existence of a hydrophobic binding register remote
from the catalytic cleft of PSMA. We expect that further
elaboration of both the alkyl tether and the phenyl ring
will improve the inhibitory potency and selectivity of
these phosphonamidate derivatives of glutamic acid
and such studies are currently underway.
4.1.3. Dibenzyl benzylphosphonate (3b). Chromatogra-
phy conditions: hexanes/ethyl acetate 2:1, v:v,
1
Rf ꢁ0.19. Yield: 61.4%. H NMR (CDCl3) d: 3.18 (d,
J=21.7Hz, 2H), 4.91 (d, J=8.3Hz, 4H), 7.24–7.32 (m,
15H) 13C NMR (CDCl3) d: 33.84, 35.55, 35.67, 68.26,
66.35, 127.62, 128.58, 129.00, 129.20, 130.50, 130.58,
131.81, 131.92, 136.97, 137.05 31P NMR (CDCl3) d:
27.72.
4. Experimental
4.1. Synthesis
4.1.4. Dibenzyl 3-phenylpropylphosphonate (3d). Chro-
matography conditions: hexanes/ethyl acetate 3:1, v:v,
1
All solvents used in reactions and diisopropylethylamine
(DIPEA) were both anhydrous and obtained as such
from commercial sources. The HCl salt of glutamic acid
dibenzyl ester [H-Glu(OBn)OBn] was neutralized by
extraction with methylene chloride from a saturated
aqueous solution of sodium bicarbonate. All other rea-
gents were used as supplied unless otherwise stated. Liq-
uid flash chromatography (silica or C18) was carried out
Rf ꢁ0.11. Yield: 65.2%. H NMR (CDCl3) d: 1.67–1.93
(m, 4H), 2.62 (t, J=7.3Hz, 2H), 4.89–5.07 (m, 4H),
7.07–7.37 (m, 15H). 13C NMR (CDCl3) d: 24.61,
25.21, 27.08, 33.84, 36.88, 37.10, 67.71, 67.78, 126.70,
128.55, 129.01, 129.12, 129.22, 130.48, 130.58, 137.12,
137.20, 141.54. 31P NMR (CDCl3) d: 33.53.
4.1.5. Dibenzyl 5-phenylpentylphosphonate (3f). Chro-
matography conditions: hexanes/ethyl acetate 2:1 v:v,
1
using a Biotage 12i/40i system. H, 13C, and 31P NMR
1
spectra were recorded on a Bruker DRX 300MHz
NMR Spectrometer. H NMR chemical shifts are rela-
Rf ꢁ0.21. Yield: 58.1%. H NMR (CDCl3) d: 1.32–1.39
1
(m, 2H), 1.51–1.78 (m, 6H), 2.55 (t, J=7.5Hz, 2H),
4.90–5.08 (m, 4H), 7.11–7.34 (m, 15H). 13C NMR
(CDCl3) d: 22.85, 25.75, 27.62, 30.63, 30.87, 31.48,
36.26, 67.62, 67.76, 68.28, 126.37, 128.56, 128.95,
129.02, 129.24, 130.51, 130.61, 137.23, 137.29, 143.06.
31P NMR (CDCl3) d: 33.83.
tive to TMS( d=0.00ppm), CDCl3 (d=7.26ppm),
CD3OD (d=4.87 and 3.31ppm), or D2O (d=4.87). 13C
NMR chemical shifts are relative to CD3OD
(d=49.15ppm) or CDCl3 (d=77.23ppm). 31P NMR
chemical shifts in CDCl3, CD3OD, or D2O were exter-
nally referenced to 85% H3PO4 (d=0.00ppm) in CDCl3,
CD3OD, and D2O, respectively. Combustion analyses
were performed by Quantitative Technologies Inc.,
Whitehouse, NJ. High resolution mass spectra (FAB)
were performed by the University of Notre Dame Mass
Spectrometry Facility, Notre Dame, IN 46556-5670.
4.1.6. General procedure for dibenzyl phenylalkylphos-
phonates 3c,e and 3g. A 1.6M solution of n-butyllithium
(7.33mL, 11.7mmol) in hexanes was added dropwise via
syringe to a stirring solution of dibenzyl methylphos-
phonate (2.70g, 9.77mmol) in dry THF (27mL) at
ꢀ78ꢁC under a N2 atmosphere. After 15min, the respec-
tive phenylalkylbromide (11.7mmol) was added drop-
wise and the reaction stirred for 0.5h at ꢀ78ꢁC, then
additional 2h at room temperature. The aqueous layer
was extracted twice more with CH2Cl2, the organic lay-
ers were then combined, washed with brine, dried over
MgSO4, and the solvent was evaporated under reduced
pressure. The residue was dissolved in CH2Cl2, and
washed with a half-saturated aqueous solution of NaCl.
Flash chromatography afforded the dibenzyl phenyl-
alkylphosphonates as light yellow oils.
4.1.1. Dibenzyl methylphosphonate (2). Benzyl alcohol
(4.55mL, 44.0mmol) and DIPEA (7.87mL, 45.2mmol)
were sequentially added dropwise via syringe to a stir-
ring solution of methyl phosphonic dichloride (3.00g,
22.6mmol) and 1H-tetrazole (0.16g, 2.2mmol) in dry
benzene (30mL) at 0ꢁC under an Ar(g) atmosphere.
The reaction mixture was stirred for 0.5h at 0ꢁC, then
additional 3.5h at room temperature. After filtering
the salts, the solvent was removed under reduced pres-
sure, and the crude oil was subjected to flash chromato-
graphy using 2:1 hexanes/ethyl acetate to afford the
product as
a
light brown oil (Rf ꢁ0.08). Yield:
4.1.7. Dibenzyl 2-phenylethylphosphonate (3c). Chroma-
tography conditions: hexanes/ethyl acetate 1:1 v:v,
Rf ꢁ0.45. Yield: 47.2%. H NMR (CDCl3) d: 2.02–2.14
89%. 1H NMR (CDCl3) d: 1.46 (d, J=17.6Hz, 3H),
4.92–5.09 (m, 4H), 7.26–7.34 (m, 10H). 13C NMR
1

J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
4975
(m, 2H), 2.84–2.93 (m, 2H), 4.96–5.11 (m, 4H), 7.11–
7.41 (m, 15H). 13C NMR (CDCl3) d: 27.77, 29.12,
29.18, 29.61, 67.83, 67.90, 127.00, 128.61, 128.70,
129.07, 129.27, 129.23, 137.08, 137.14, 141.35, 141.58.
31P NMR (CDCl3) d: 32.18.
4.1.14. Benzyl 4-phenylbutylphosphonic acid (4e). Yield:
62.3%. H NMR (CDCl3) d: 1.67–1.84 (m, 6H), 2.57
1
(t, J=6.6Hz, 2H), 5.02 (d, J=7.8Hz, 2H), 7.11–7.338
(m, 10H). 13C NMR (CDCl3) d: 22.60, 22.65, 25.62,
27.51, 32.72, 32.94, 36.07, 67.07, 67.15, 26.48, 128.42,
128.94, 129.01, 129.21, 137.07, 137.30, 142.59. 31P
NMR (CDCl3) d: 35.88.
4.1.8. Dibenzyl 4-phenylbutylphosphonate (3e). Chroma-
tography conditions: hexanes/ethyl acetate 2:1 v:v,
1
Rf ꢁ0.20. Yield: 42.9%. H NMR (CDCl3) d: 1.58–1.83
4.1.15. Benzyl 5-phenylpentylphosphonic acid (4f). Yield:
1
(m, 6H), 2.52–2.57 (m, 2H), 4.91–5.08 (m, 4H), 7.08–
7.33 (m, 15H). 13C NMR (CDCl3) d: 22.59, 22.66,
25.67, 27.50, 32.66, 32.88, 35.95, 67.65, 67.71, 126.41,
128.52, 128.97, 129.19, 137.15, 137.21, 142.51. 31P
(CDCl3) d: 30.65.
86.0%. H NMR (CDCl3) d: 1.36–1.39 (m, 2H), 1.58–
1.71 (m, 6H), 2.56 (t, J=7.4Hz, 2H), 4.97 (d, J=
7.8Hz, 2H), 7.11–7.33 (m, 10H), 11.14 (s, 1H). 13C
NMR (CDCl3) d: 22.70, 22.76, 25.64, 27.55, 30.62, 30.85,
31.52, 33.75, 36.30, 66.95, 67.04, 67.71, 67.78, 126.37,
127.56, 128.41, 128.57, 128.95, 129.03, 129.19, 130.61,
130.69, 137.02, 137.12, 143.07. 31P NMR (CDCl3) d: 36.20.
4.1.9. Dibenzyl 6-phenylhexylphosphonate (3g). Chroma-
tography conditions: hexanes/ethyl acetate 2:1 v:v,
1
Rf ꢁ0.22. Yield: 38.5%. H NMR (CDCl3) d: 1.29–1.41
4.1.16. Benzyl 6-phenylhexylphosphonic acid (4g). Yield:
1
(m, 4H), 1.53–1.80 (m, 6H), 2.58 (t, J=7.5Hz, 2H),
4.94–5.10 (m, 4H), 7.11–7.36 (m, 15H). 13C (CDCl3) d:
22.89, 22.96, 25.81, 27.66, 29.34, 30.89, 31.12, 31.77,
36.51, 67.67, 67.76, 126.31, 128.56, 128.62, 128.92,
128.99, 129.05, 129.25, 137.27, 137.34, 143.28. 31P
(CDCl3) d: 33.94.
91.2%. H NMR (CDCl3) d: 1.26–1.40 (m, 4H), 1.56–
1.80 (m, 6H), 2.56 (t, J=7.6Hz, 2H), 4.98 (d,
J=8.1Hz, 2H), 7.13–7.35 (m, 11H). 13C NMR (CDCl3)
d 22.78, 25.715, 27.60, 29.38, 30.90, 31.14, 31.83, 36.51,
66.96, 67.75, 126.21, 126.31, 128.40, 128.98, 129.14,
137.03, 137.15, 143.25. 31P NMR (CDCl3) d: 36.32.
4.1.10. General procedure for benzyl phenylalkylphos-
phonic acids 4b–g. A stirring solution of NaI (0.729g,
4.86mmol) and dibenzyl phenylalkylphosphonate 3
(4.05mmol) in 2-butanone (15mL) was refluxed 2.5h
under an Ar(g) atmosphere. The solvent was removed
under reduced pressure and the resulting salt was col-
lected and washed with ice-cold acetone to remove start-
ing materials. Trace solvents were removed from the
filtered solid under reduced pressure. The mono-sodium
phosphonate salt was dissolved in a solution of 10% HCl
(30mL) and brine (30mL) then extracted thrice with
CH2Cl2 (40mL). The organic layers were combined,
dried over MgSO4, and concentrated in vacuo to
afford the benzyl phenylalkylphosphonic acids as white
solids.
4.1.17. N-[Benzyloxy(phenyl)phosphinyl]-L-glutamic acid
dibenzyl ester (5a). Benzyl alcohol (391lL, 3.78mmol)
and DIPEA (724lL, 4.16mmol) were sequentially
added dropwise via syringe to a stirring solution of phe-
nylphosphonic dichloride (579lL, 4.16mmol) and 1-H-
tetrazole (26.5mg, 0.038mmol) dissolved in dry benzene
(8.0mL) at 0ꢁC under an Ar(g) atmosphere. The reaction
mixture was stirred for 0.5h at 0ꢁC, then additional
2.5h at room temperature. The reaction mixture was
cooled to 0ꢁC and to it was sequentially added H-
Glu(OBn)OBn (1.237g, 3.78mmol) and DIPEA
(724lL, 4.16mmol) dropwise. The reaction mixture
was stirred for 0.5h at 0ꢁC and then 3.5h at room tem-
perature. The reaction mixture was filtered to remove
salts and the solvent was removed under reduced pres-
sure. Reversed-phase (C18) flash chromatography
(MeOH/H2O, 4:1, v:v) afforded the product as white sol-
ids (Rf ꢁ0.16). Yield: 39.5%. 1H NMR (CDCl3) d: 1.93–
1.98 (m, 1H), 2.05–2.09 (m, 1H), 2.34–2.43 (m, 2H),
3.50–3.55 (m, 1H), 3.91–4.09 (m, 1H), 5.03–5.08 (m,
6H), 7.25–7.47 (m, 18H), 7.75–7.79 (m, 2H). 13C NMR
(CDCl3) d: 30.11, 30.20, 30.27, 30.43, 30.51, 53.55,
53.80, 66.93, 67.00, 67.85, 128.33, 128.42, 128.82,
128.87, 128.91, 129.01, 129.09, 129.17, 129.21, 131.96,
132.09, 132.14, 132.27, 132.73, 132.77, 132.81, 135.73,
136.39, 136.42, 137.01, 172.97, 173.29, 173.37, 173.43.
31P NMR (CDCl3) d: 18.86, 19.52 (diastereomers). Anal.
Calcd for C32H32NO6P: C, 68.93; H, 5.78; N, 2.51.
Found: C, 68.82; H, 5.91; N, 2.45.
4.1.11. Benzyl benzylphosphonic acid (4b). Yield: 66.5%.
1H NMR (CDCl3) d: 3.05 (d, J=22.1Hz, 2H), 4.80 (d,
J=7.6Hz, 2H), 7.18–7.33 (m, 10H), 11.10 (s, 1H) 13C
NMR (CDCl3) d: 33.66, 33.52, 67.50, 127.57, 128.40,
128.88, 129.14, 130.59, 130.68, 131.89, 132.01, 136.90,
136.99. 31P NMR (CDCl3) d: 29.86.
4.1.12. Benzyl 2-phenylethylphosphonic acid (4c). Yield:
1
52.5%. H NMR (CD3OD) d: 1.86–1.98 (m, 2H), 2.69–
2.78 (m, 2H), 4.92 (d, J=8.4Hz, 2H), 7.043–7.285 (m,
10H), 11.15 (s, 1H). 13C NMR (CDCl3) d: 27.73,
29.04, 29.08, 29.59, 67.30, 67.38, 126.99, 128.43,
128.69, 129.00, 129.23, 136.92, 137.01, 141.36, 141.61.
31P NMR (CD3OD) d: 30.52.
4.1.18. General procedure for benzyl phenylalkylphospho-
namidates 5b–g. (COCl)2 (97.1lL, 1.13mmol) was
added dropwise via syringe to a stirring solution of phe-
nylalkylphosphonic acid 4 (0.942mmol) and DMF
(0.094mmol) in CH2Cl2 (6mL) at 0ꢁC under an Ar(g)
atmosphere. After 0.5h, the reaction mixture was
warmed to room temperature and stirred for an addi-
tional 1.5h. The solvent was removed under reduced
4.1.13. Benzyl 3-phenylpropylphosphonic acid (4d). Yield:
73.1%. H NMR (CDCl3) d: 1.70–1.81 (m, 2H), 1.89–
1
1.98 (m, 2H), 2.66 (t, J=7.2Hz, 2H), 5.00 (d, J=7.8Hz,
2H), 7.113–7.332 (m, 10H), 10.66 (s, 1H). 13C NMR
(CDCl3) d: 24.54, 24.59, 25.19, 27.09, 36.88, 37.11, 67.08,
67.16, 126.74, 128.31, 128.43, 128.97, 129.10, 129.17,
130.47, 136.96, 141.64. 31P NMR (CDCl3) d: 33.20.

4976
J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
pressure, the residue was dissolved in dry toluene (2mL),
and the solvent was again removed under reduced pres-
sure. The resulting crude phosphonochloridate was dis-
solved in benzene (2.54mL) and added dropwise to a
stirring solution of H-Glu(OBn)OBn (278mg,
0.848mmol), 1H-tetrazole (6.6mg, 0.094mmol), and
DIPEA (181lL, 1.04mmol) in benzene (2.54mL) at
0ꢁC under an Ar(g) atmosphere. After 0.5h, the reaction
mixture was warmed to room temperature and stirred
for additional 3.5h. The reaction mixture was filtered
and the solvent of the filtrate was removed under
reduced pressure. Reversed-phase (C18) flash chroma-
tography (MeOH/H2O, 4:1, v:v) afforded the products
as white solids.
30.51, 30.58, 30.71, 30.77, 32.87, 33.10, 36.10, 53.80,
53.93, 65.98, 66.02, 66.12, 66.22, 67.19, 68.03, 128.27,
128.83, 128.93, 129.00, 129.06, 129.22, 129.35, 135.88,
136.47, 142.67, 173.05, 173.77. 31P NMR (CDCl3) d:
34.62, 35.24 (diastereomers). Anal. Calcd for
C36H40NO6P: C, 70.46; H, 6.57; N, 2.28. Found: C,
68.15; H, 6.60; N, 2.36.
4.1.23. N-[Benzyloxy(5-phenylpentyl)phosphinyl]-L-gluta-
1
mic acid dibenzyl ester (5f). Rf ꢁ0.12. Yield: 49.1%. H
NMR (CDCl3) d: 1.33–1.36 (m, 2H), 1.52–1.69 (m,
6H), 1.85–1.94 (m, 1H), 2.05–2.14 (m, 1H), 2.39–2.43
(m, 2H), 2.56 (t, J=7.4Hz, 2H), 3.01–3.05 (m, 1H),
3.94–4.10 (m, 1H), 4.83–4.90 (m, 1H), 4.97–5.12 (m,
6H), 7.12–7.34 (m, 20H). 13C NMR (CDCl3) d: 22.97,
30.46, 30.50, 30.68, 30.73, 31.55, 36.30, 53.83, 65.92,
66.05, 67.16, 68.00, 126.34, 128.23, 128.43, 128.82,
128.94, 129.04, 129.34, 129.31, 136.40, 143.08, 173.04,
173.80. 31P NMR (CDCl3) d: 34.76, 35.40 (diastereo-
mers). Anal. Calcd for C37H42NO6P: C, 70.80; H,
6.74; N, 2.23. Found: C, 70.56; H, 6.60; N, 2.13.
4.1.19. N-[Benzyloxy(benzyl)phosphinyl]-L-glutamic acid
1
dibenzyl ester (5b). Rf ꢁ0.16. Yield: 33.1%. H NMR
(CD3OD) d: 1.76–1.95 (m, 2H), 2.12–2.48 (m, 4H),
3.11–3.22 (m, 2H), 3.90–3.96 (m, 1H), 5.06–5.16 (m,
4H), 7.17–7.35 (m, 20H). 13C NMR (CDCl3) d: 30.74,
30.79, 30.93, 31.02, 36.46, 36.55, 38.11, 38.27, 54.27,
54.55, 66.34, 66.67, 66.76, 66.96, 67.06, 67.49, 68.37,
127.93, 128.01, 128.51, 128.63, 128.80, 129.15, 129.21,
129.35, 129.39, 129.59, 129.63, 129.70, 129.78, 130.87,
130.95, 131.05, 132.97, 133.09, 136.28, 136.86, 137.66,
137.75, 173.43, 174.06. 31P NMR (CDCl3) d: 29.25,
29.54 (diastereomers). Anal. Calcd for C33H34NO6P:
C, 69.34; H, 6.00; N, 2.45. found: C, 69.14; H, 6.00;
N, 2.73.
4.1.24. N-[Benzyloxy(6-phenylhexyl)phosphinyl]-L-gluta-
1
mic acid dibenzyl ester (5g). Rf ꢁ0.10. Yield: 59.2%. H
NMR (CDCl3) d: 1.29–1.30 (m, 4H), 1.55–1.69 (m,
6H), 1.93–1.95 (m, 1H), 2.04–2.18 (m, 1H), 2.39–2.43
(m, 2H), 2.56 (t, J=7.6Hz, 2H), 3.02–3.06 (m, 1H),
3.95–4.05 (m, 1H), 4.83–4.87 (m, 1H), 4.98–5.12 (m,
6H), 7.14–7.31 (m, 20H). 13C NMR (CDCl3) d: 22.96,
23.01, 27.91, 29.41, 29.59, 30.52, 30.68, 30.75, 31.04,
31.26, 31.83, 36.54, 53.77, 53.89, 65.87, 66.16, 67.13,
68.01, 126.29, 128.23, 128.43, 128.81, 128.92, 129.04,
129.21, 129.32, 136.44, 143.30, 173.09, 173.80. 31P
NMR (CDCl3) d: 34.85, 35.49 (diastereomers). Anal.
Calcd for C38H44NO6P: C, 71.12; H, 6.91; N, 2.18.
Found: C, 71.08; H, 6.92; N, 2.16.
4.1.20. N-[Benzyloxy(2-phenylethyl)phosphinyl]-L-gluta-
1
mic acid dibenzyl ester (5c). Rf ꢁ15. Yield: 31.3%. H
NMR (CDCl3) d: 1.93–2.10 (m, 4H), 2.37–2.43 (m,
2H), 2.86–2.89 (m, 2H), 3.03–3.12 (m, 1H), 4.02–4.14
(m, 1H), 4.87–5.14 (m, 6H), 7.11–7.33 (m, 20H). 13C
NMR (CDCl3) d: 29.21, 29.26, 29.92, 30.40, 30.68,
30.72, 31.60, 53.78, 53.89, 126.97, 128.28, 128.50,
128.76, 128.93, 128.97, 129.05, 129.24, 129.30, 135.78,
136.38, 137.23, 137.33, 141.52, 141.73, 173.00, 173.07,
173.71, 173.78. 31P NMR (CDCl3) d: 33.23, 33.90 (dia-
stereomers). Anal. Calcd for C34H36NO6P: C, 69.73;
H, 6.20; N, 2.39. Found: C, 69.48; H, 6.21; N, 2.37.
4.1.25. General procedure for phenyl- and phenylalkyl-
phosphonamidates 1a–g. To a flask was added 5
(100mg), THF (1mL), 10% Pd/C (10mg), K2CO3
(1.55equiv), and water (1mL). The mixture was then
stirred vigorously under an N2(g) atmosphere for 1min.
The flask was then purged and charged with H2(g) (bal-
loon pressure) and stirred for 4h at room temperature.
The reaction mixture was filtered using a Whatman
0.2lm PTFE micropore filtration disk, the solvent was
removed under reduced pressure, and the products were
obtained as white solids in quantitative yield.
4.1.21. N-[Benzyloxy(3-phenylpropyl)phosphinyl]-L-glu-
tamic acid dibenzyl ester (5d). Rf ꢁ0.14. Yield: 54.5%.
1H NMR (CDCl3) d: 1.62–2.13 (m, 6H), 2.40–2.47 (m,
2H), 2.52–2.58 (m, 2H), 3.93–3.96 (m, 1H), 4.97–5.10
(m, 6H), 7.06–7.29 (m, 20H). 13C NMR (CD3OD) d:
24.67, 27.35, 27.41, 29.05, 29.16, 30.36, 30.45, 30.51,
30.69, 37.02, 37.23, 53.74, 53.87, 65.92, 66.00, 66.13,
66.20, 67.09, 67.96, 126.67, 128.25, 128.46, 128.81,
128.92, 129.02, 129.14, 129.21, 129.29, 135.85, 136.44,
137.34, 137.41, 141.68, 141.73, 173.00, 173.74. 31P
NMR (CDCl3) d: 33.66, 34.43 (diastereomers). Anal.
Calcd for C35H38NO6P: C, 70.10; H, 6.39; N, 2.34.
Found: C, 70.05; H, 6.34; N, 2.13.
4.1.26. N-[Hydroxy(phenyl)phosphinyl]-L-glutamic acid
1
tripotassium salt (1a). H NMR (D2O) d: 1.64–1.70 (m,
2H), 2.02–2.08 (m, 2H), 3.29–3.32 (m, 1H), 7.35–7.38
(m, 3H), 7.55–7.63 (m, 2H). 13C NMR (D2O) d: 32.98,
34.52, 57.26, 128.85, 129.01, 130.93, 131.28, 131.40,
136.71, 138.81, 161.62, 182.06, 183.83. 31P NMR
+
(D2O) d: 13.31. FAB-HRMS[M+K] calcd 439.8872,
found: 439.8872 for C11H11K4NO6P.
4.1.22. N-[Benzyloxy(4-phenylbutyl)phosphinyl]-L-gluta-
1
mic acid dibenzyl ester (5e). Rf ꢁ0.14. Yield: 47.6%. H
4.1.27. N-[Hydroxy(benzyl)phosphinyl]-L-glutamic acid
tripotassium salt (1b). 1H NMR (D2O) d: 1.79–1.85
(m, 2H), 2.15–2.23 (m, 2H), 2.99 (d, J=19.0Hz, 2H),
7.28–7.39 (m, 5H). 13C NMR (D2O) d: 27.22, 32.82,
33.75, 33.92, 36.31, 36.82, 37.65, 38.36, 56.81, 125.83,
NMR (CDCl3) d: 1.61–1.75 (m, 3H), 1.85–2.16 (m,
3H), 2.34–2.56 (m, 4H), 2.98–3.13 (m, 1H), 3.98–4.09
(m, 1H), 4.82–5.12 (m, 6H), 7.11–7.33 (m, 20H). 13C
NMR (CDCl3) d: 22.82, 27.80, 27.93, 29.66, 30.42,

J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
4977
126.02, 128.63, 129.81, 129.87, 136.13, 136.23, 181.83,
183.40. 31P NMR (D2O) d: 22.58. FAB-HRMS
[M+K]+ calcd 453.9029 found: 453.9029 for
C12H13K4NO6P.
washed, and vacuum filtered using the following sol-
vent wash cycle: 3·5mL DMF, 3·5mL CH2Cl2,
2·5mL CH3CN, 2·5mL diethyl ether. Step 2 (Chromo-
phore coupling): 4-(Phenylazo)benzoic acid (0.183g,
0.81mmol),
benzotriazole-1-yl-oxy-tris-pyrrolidino-
4.1.28. N-[Hydroxy(2-phenylethyl)phosphinyl]-L-glutamic
acid tripotassium salt (1c). H NMR (CD3OD) d: 1.75–
phosphonium hexafluorophosphate (PyBOP, 0.421g,
0.81mmol) and N-hydroxybenzotriazole (HOBt,
0.109g, 0.81mmol) were dissolved in DMF (5mL) and
added to the deprotected resin followed by the addition
of DIPEA (0.282mL, 1.62mmol) via syringe. The reac-
tion mixture was rocked 1h. The resin was vacuum fil-
tered then subjected to the solvent wash cycle described
above. A standard Kaiser test was used to confirm com-
plete loading. In the case that loading was incomplete,
this step was repeated. Step 3 (Cleavage of product from
resin): A 5mL solution of TFA/H2O (4:1, v:v) was added
to the resin and rocked 3h. Resin was vacuum filtered
then washed with CH2Cl2 (4·5mL). The filtrate was col-
lected and the solvent removed under reduced pressure to
provide an orange solid (0.138g). The product was puri-
fied by semi-prep HPLC (Alltech Semi-Prep Econosil
C18 10l, 250·22mm) using a mobile phase composed
of 65% A and 35% B (Solvent A=0.1% aqueous TFA
by volume; Solvent B=CH3CN) to give a light yellow so-
lid (0.032g, 13% yield). 1H NMR (D2O with 1equiv
K3PO4 for solubility) d: 1.98–2.18 (m, 2H), 2.22–2.32
(m, 2H), 4.25 (t, J=5.90Hz, 1H), 7.45–7.47 (m, 3H),
7.60–7.66 (m, 4H), 7.76 (d, J=8.07Hz, 2H). 13C NMR
(D2O with 1equiv K3PO4) d: 30.32, 36.03, 57.88,
124.21, 124.37, 130.14, 131.22, 133.96, 137.33, 153.53,
155.42, 170.42, 180.30, 183.89. Mp 188–189ꢁC.
1
1.92 (m, 4H), 2.21–2.28 (m, 2H), 2.74–2.82 (m, 2H),
3.56–3.69 (m, 1H), 7.24–7.40 (m, 5H). 13C NMR
(CDCl3) d: 30.16, 31.71, 32.92, 33.35, 34.08, 56.91,
125.09, 127.67, 127.86, 127.99, 143.71, 143.94, 181.04,
181.56. 31P NMR (CDCl3) d: 25.29. FAB-HRMS
[M+H]+ calcd 429.9627 found: 429.9626 for C13
H16K3NO6P.
4.1.29. N-[Hydroxy(3-phenylpropyl)phosphinyl]-L-gluta-
mic acid tripotassium salt (1d). 1H NMR (D2O) d:
1.48–1.59 (m, 2H), 1.79–1.88 (m, 4H), 2.69 (t,
J=7.2Hz, 2H), 3.41–3.48 (m, 1H), 7.25–7.42 (m, 5H).
13C NMR (D2O) d: 25.32, 28.20, 29.82, 32.70, 34.20,
36.56, 36.79, 56.80, 126.16, 128.83, 128.93, 142.97,
182.19, 183.31. 31P NMR (D2O) d: 29.56. FAB-HRMS
[M+K]+ calcd 481.9342 found: 481.9342 for
C14H17K4NO6P.
4.1.30. N-[Hydroxy(4-phenylbutyl)phosphinyl]-L-gluta-
mic acid tripotassium salt (1e). 1H NMR (D2O) d:
1.54–1.70 (m, 6H), 1.80–1.86 (m, 2H), 2.22 (t,
J=8.0Hz, 2H), 2.66 (t, J=7.2Hz, 2H), 3.46–3.49 (m,
1H), 7.25–7.41 (m, 5H). 13C NMR (D2O) d: 23.12,
28.42, 32.84, 32.69, 34.20, 34.99, 56.82, 126.00, 128.81,
143.68, 182.27, 183.34. 31P NMR (D2O) d: 28.33.
k
max =325nm, e=3.00·104 Mꢀ1 cmꢀ1
.
+
FAB-HRMS[M+H] calcd 457.9939 found: 457.9939
for C15H20K3NO6P.
4.1.34.
N-[4-(Phenylazo)benzyol]glutamyl-c-glutamic
acid (7). Step 1 (Deprotection of resin): Deprotection
of the resin was accomplished according to the proce-
dure described above for 6. Step 2 (Amino acid cou-
pling): A solution of Nꢀa-Fmoc-L-glutamic acid-c-t-
butyl ester (0.345g, 0.81mmol), PyBOP (0.421g,
0.81mmol) and HOBt (0.109g, 0.81mmol) in DMF
(5mL) was added to the deprotected resin. DIPEA
(0.282mL, 1.62mmol) was added via syringe and the
reaction mixture was rocked 1h. The resin was vacuum
filtered then subjected to the solvent wash cycle de-
scribed in Step 3 for 6. A standard Kaiser test was used
to confirm complete loading. In the case that loading
was incomplete, this step was repeated. Step 4 (Chromo-
phore coupling): The resin was first deprotected accord-
ing to Step 1. 4-(Phenylazo)benzoic acid (0.183g,
0.81mmol), PyBOP (0.421g, 0.81mmol) and HOBt
(0.109g, 0.81mmol) were dissolved in DMF (5mL)
and added to the deprotected resin followed by the addi-
tion of DIPEA (0.282mL, 1.62mmol) via syringe. The
reaction mixture was rocked 1h. The resin was vacuum
filtered then subjected to the solvent wash cycle de-
scribed in Step 3 for 6. A standard Kaiser test was used
to confirm complete loading. In the case that loading
was incomplete, this step was repeated. Step 5 (Cleavage
of product from resin): A 5mL solution of TFA/H2O
(4:1, v:v) was added to the resin and rocked 3h. Resin
was vacuum filtered then washed with CH2Cl2
(4·5mL). The filtrate was collected and the solvent re-
moved under reduced pressure to provide a reddish
4.1.31. N-[Hydroxy(5-phenylpentyl)phosphinyl]-L-gluta-
mic acid tripotassium salt (1f). 1H NMR (D2O) d:
1.351.61 (m, 8H), 1.81–1.83 (m, 2H), 2.21 (t,
J=7.8Hz, 2H), 2.62 (t, J=6.8 Hz, 2H), 3.43–3.46 (m,
1H), 7.24–7.35 (m, 5H). 13C NMR (D2O) d: 23.10,
28.49, 30.10, 30.22, 30.53, 32.77, 34.20, 35.10, 56.74,
56.85, 125.84, 126.06, 128.74, 128.84, 128.96, 143.82,
182.22, 183.34. 31P NMR (D2O) d: 28.57. FAB-HRMS
[MꢀK]ꢀ calcd 432.0381 found: 432.0390 for
C16H21K2NO6P.
4.1.32. N-[Hydroxy(6-phenylhexyl)phosphinyl]-L-gluta-
mic acid tripotassium salt (1g). 1H NMR (D2O) d:
1.34–1.61 (m, 10H), 1.79–1.81 (m, 2H), 2.22 (t,
J=7.8Hz, 2H), 2.63 (t, J=7.5Hz, 2H), 3.41–3.44 (m,
1H), 7.25–7.36 (m, 5H). 13C NMR (D2O) d: 23.35,
23.39, 28.36, 28.73, 30.35, 30.51, 30.74, 31.02, 32.89,
32.94, 34.37, 35.34, 56.96, 126.06, 128.88, 128.96,
144.07, 182.30, 183.40. 31P NMR (D2O) d: 28.71.
FAB-HRMS[M ꢀK]ꢀ calcd 446.0537 found: 446.0559
for C17H23K2NO6P.
4.1.33. N-[4-(Phenylazo)benzyol]-L-glutamic acid (6). Step
1. (Deprotection of resin): N-a-Fmoc-L-glutamic acid-
c-t-butyl ester preloaded on Wang resin (0.5g,
0.270mmol; 0.54mmol/g resin 100–200 mesh, Novabio-
chem San Diego, CA) was deprotected with 5mL piperi-
dine/DMF (20:80, v:v) for 30min, vacuum filtered,

4978
J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
brown solid (0.145g). The product was purified by semi-
prep HPLC (Alltech Semi-Prep Econosil C18 10l,
250·22 mm) using a mobile phase composed of 65%
A and 35% B (Solvent A=0.1% aqueous TFA by vol-
ume; Solvent B=CH3CN) to give an orange solid
(0.058g, 23% yield). 1H NMR (D2O with 1equiv
K3PO4 for solubility) d: 1.79–1.93 (m, 2H), 1.96–2.12
(m, 2H), 2.14–2.24 (m, 2H), 2.38–2.45 (m, 2H), 4.05–
4.09 (m, 1H), 4.27–4.31 (m, 1H), 7.49 (d, J=6.60Hz,
2H), 7.64–7.68 (m, 5H), 7.80 (d, J=8.40Hz, 2H). 13C
NMR (D2O) d: 28.76, 29.52, 33.37, 35.04, 56.24, 56.46,
123.36, 123.49, 129.28, 130.36, 133.07, 136.46, 152.75,
154.66, 169.61, 175.82, 178.88, 179.65, 182.94. Mp
the enzyme solution was quickly frozen in dry ice or
N2(l) then stored at ꢀ20ꢁC.
4.3. HPLC Quantification of PABGcG and PABG
PABGcG and its hydrolytic product (PABG) were sep-
arated and quantified using an analytical reversed-phase
HPLC column (Lichrosphere C18 5lm, 150·4.6mm;
Phenomenex, Torrence, CA) with a mobile phase con-
sisting of ACN/potassium phosphate [25mM, pH2.0
(adjusted with H3PO4)] at a respective volume ratio of
40:60. At a flow rate of 1.0mL/min, PABGcG and its
hydrolytic product (PABG) were detected at 325nm
with retention times of 4.8 and 6.9min, respectively.
185–186ꢁC. kmax =325nm, e=2.58·104 Mꢀ1 cmꢀ1
.
4.2. Purification of PSMA
4.4. Determination of Km and Vmax for PABGcG/with
purified PSMA
4.2.1. Cells and media. LNCaP cells were obtained from
ATCC (Manassus, VA). Cells were cultured in RPMI
1640 medium with 2mM L-glutamine adjusted to con-
tain 1.5g/L sodium bicarbonate, 4.5g/L glucose,
10mM HEPES, and 1.0mM sodium pyruvate, 90%; fe-
tal bovine serum, 10% at 37ꢁC in a 5% CO2 atmosphere.
Working solutions of the substrate (PABGcG) were
made in Tris buffer (50mM, pH7.5). A typical incuba-
tion mixture (final volume 250lL) was prepared by
the addition of 200lL Tris buffer (50mM, pH7.4) to
either 25lL of
a
solution of purified PSMA
(0.00214lg) or 25lL Tris buffer (50mM, pH7.4) as a
negative control. The enzymatic reaction was initiated
by the addition of 25lL PABGcG (1–20lM). The final
concentration of PABGcG ranged from 0.1 to 2.0lM.
The reaction was allowed to proceed for 15min with
constant shaking at 37ꢁC and was terminated by the
addition of 25lL methanolic TFA (2% trifluoroacetic
acid by volume in methanol) followed by vortexing
and centrifugation (10min at 7000g). An 85lL aliquot
of the resulting supernatant was subsequently quantified
by HPLC as described above. Under the assay condi-
tions described above, it was noted that the initial sub-
strate concentration was not substantially depleted
during the time course of the incubation (e.g., approxi-
mately 10% conversion to product was observed for in-
cubations with the lowest substrate concentration,
0.1lM).
4.2.2. Extraction of PSMA from Cell lysates. The media
from the cell cultures were poured off and LNCaP cells
were scraped from 200 75cc flasks, washed in PBSand
pelleted by centrifugation at 1000rpm. The supernatant
is discarded and the pellet was dounce-homogenized in
50mM Tris, 150mM NaCl pH7.4. The membrane-con-
taining fraction was isolated as a pellet by centrifugation
at 100,000g for 15min; the supernatant was discarded.
This process of homogenization and centrifugation
was repeated twice more. The pellet was then suspended
in 50mM Tris, 150mM NaCl and 1.0% Triton X-100 to
solubilize PSMA. After
100,000g the PSMA-containing supernatant was col-
lected and quickly frozen in dry ice or N2(l)
a final centrifugation at
.
4.2.3. PSMA purification. An antibody-affinity column
was prepared by adding 500lL Sepharose-3C6 Mab
affinity resin (Northwest Biotherapeutics; Seattle, WA)
into a Poly-Prep Chromatography Column (Bio-Rad
Laboratories, Hercules, CA). The column was sequen-
tially washed with 5mL each: 50mM Tris pH7.5 con-
taining 150mM NaCl; 100mM Glycine pH2.5
containing 150mM NaCl; and 50mM Tris pH7.5 con-
taining 150mM NaCl. Solubilized PSMA obtained from
cell lysates above was loaded on the column and sequen-
tially washed with 5mL each of the following buffers:
50mM Tris pH7.5 containing 150mM NaCl and 1.0%
Triton-X 100; 1.0M KCl containing 150mM NaCl
and 1.0% Triton-X 100; 50mM Tris pH7.5 containing
150mM NaCl and 1.0% Triton-X 100, 50mM Tris
pH9.0 containing 150mM NaCl and 1.0% Triton-X
100. Purified active enzyme was finally eluted with
5mL 100mM Glycine pH11.5 containing 150mM NaCl
directly into a Sephadex 10DG desalting column (Bio-
Rad Laboratories, Hercules, CA) equilibrated with
50mM Tris pH7.5 containing 150mM NaCl and 1.0%
Triton-X 100. The eluate was collected in 0.5mL aliqu-
ots, PSMA enzyme activity was determined by HPLC as
described below, and active aliquots were pooled and
4.5. PSMA inhibition assay
Working solutions of the substrate (PABGcG) and all
inhibitors were made in Tris buffer (50mM, pH7.5). A
typical incubation mixture (final volume 250lL) was
prepared by the addition of 175lL Tris buffer (50mM,
pH7.4) to either 25lL of an inhibitor solution or
25lL Tris buffer (50mM, pH7.4). 25lL of a solution
of purified PSMA (0.00855lg) was added and the enzy-
matic reaction was initiated by the addition of 25lL
PABGcG (10 or 20lM). In all cases, the final concentra-
tion of PABGcG was either 1 or 2lM while the final
inhibitor concentration varied: 0.5–2.0lM for 1a and
1b; 0.25–1.0lM for 1c–g; and 0.0005–0.002lM for 8.
The reaction was allowed to proceed for 15min with
constant shaking at 37ꢁC and was terminated by the
addition of 25lL methanolic TFA (2% trifluoroacetic
acid by volume in methanol) followed by vortexing
and centrifugation (10min at 7000g). An 85lL aliquot
of the resulting supernatant was subsequently quantified
by HPLC as described above. Under the assay condi-
tions described above, it was noted that the initial sub-

J. Maung et al. / Bioorg. Med. Chem. 12 (2004) 4969–4979
4979
Tong, W. D. W.; May, F.; Mukherjee, B.; Heston, W. D.
W. Clin. Cancer. Res. 1996, 2, 1445.
11. Carter, R. E.; Feldman, A. R.; Coyle, J. T. Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 749.
12. Pangalos, M. N.; Neefs, J. M.; Somers, M.; Verhasselt, P.;
Bekkers, M.; van der Helm, L.; Fraiponts, E.; Ashton, D.;
Gordon, R. D. J. Biol. Chem. 1999, 274, 8470.
strate concentration was not substantially depleted dur-
ing the time course of the incubation (e.g., approxi-
mately 10% conversion to product was observed for
incubations with the lowest substrate concentration
(1lM) in the absence of inhibitor.
13. Robinson, M. B.; Blakely, R. D.; Couto, R.; Coyle, J. T.
J. Biol. Chem. 1987, 262, 14498.
Acknowledgements
14. (a) Jackson, P. F.; Cole, D. C.; Slusher, B. S.; Stetz, S. L.;
Ross, L. E.; Donzanti, B. A.; Trainor, D. A. J. Med.
Chem. 1996, 39, 619; (b) Luthi-Carter, R.; Barczak, A. K.;
Speno, H.; Coyle, J. T. Brain Res. 1998, 795, 341; (c) Nan,
F.; Bzdega, T.; Pshenichkin, S.; Wroblewski, J. T.;
Wroblewska, B.; Neale, J. H.; Kozikowski, A. P.
J. Med. Chem. 2000, 43, 772.
15. (a) Tang, H.; Brown, M.; Ye, Y.; Huang, G.; Zhang, Y.;
Wang, Y.; Zhai, H.; Chen, X.; Shen, T. S.; Tenniswood,
M. Biochem. Biophys. Res. Commun. 2003, 307, 8; (b)
Rodriguez, C. E.; Lu, H.; Martinez, A. R.; Brunelle, A.;
Berkman, C. E. J. Enzym Inhib. 2001, 16, 359.
This work was supported in part by grants from the Na-
tional Institutes of Health, MBRSSCORE Program-
NIGMS(Grant No. 2S06-GM052588-09), the National
Cancer Institute, U56-Program (Grant No. CA 96217),
and the NIH-PREP Scholarship Program (Grant No. 1
R25 GM64078-03). The authors would also like to ex-
tend their gratitude to W. Tam and the NMR facility
at SFSU for their expert assistance.
16. Teulade, M.-P.; Savignac, P.; Aboujaoude, E. E.; Collin-
gnon, N. J. Organomet. Chem. 1986, 312, 283.
References and notes
17. Patel, D. V.; Gordon, E. M.; Schmidt, R. J.; Weller, H. N.;
Young, M. G.; Zahler, R.; Barbacid, M.; Carboni, J. M.;
Gullo-Brown, J. L.; Huniham, L.; Ricca, C.; Robinson, S.;
Seizinger, B. R.; Tuomari, A. V.; Manne, V. J. Med.
Chem. 1995, 38, 435; Delfino, J. M.; Stankovic, C. J.;
Schreiber, S. L.; Richards, F. M. Tetrahedron Lett. 1987,
28, 2323; Scheit, K. H. Tetrahedron Lett. 1967, 3243.
18. Malachowski, W. P.; Coward, J. K. J. Org. Chem. 1994,
59, 7616; Musiol, H.-J.; Grams, F.; Rudolph-Bohner, S.;
Moroder, L. J. Org. Chem. 1994, 59, 6144.
1. Brawer, M. K. World J. Surg. 2000, 24, 1200.
2. Cohen, S. P.; Jaskulsky, S. R. Geriatrics 2001, 56, 39.
3. Farkas, A.; Marcella, S.; Rhoads, G. G. Ethn. Dis. 2000,
10, 69.
4. Holmes, E. H.; Greene, T. G.; Tino, W. T.; Boynton, A.
L.; Aldape, H. C.; Misrock, S. L.; Murphy, G. P. Prostate
Suppl. 1996, 7, 25.
5. Horoszewicz, J. S.; Leong, S. S.; Kawinski, E.; Karr, J. P.;
Rosenthal, H.; Chu, T. M.; Mirand, E. A.; Murphy, G. P.
Cancer Res. 1983, 43, 1809.
19. Mlodnosky, K. L.; Holmes, H. M.; Lam, V. Q.; Berkman,
C. E. Tetrahedron Lett. 1997, 38, 8803.
6. Bacich, D. J.; Pinto, J. T.; Tong, W. P.; Heston, W. D.
Mamm. Genome 2001, 12, 117.
20. Tiffany, C. W.; Lapidus, R. G.; Merion, A.; Calvin, D. C.;
Slusher, B. S. Prostate 1999, 39, 28.
7. Chang, S. S.; OÕKeefe, D. S.; Bacich, D. J.; Reuter, V. E.;
Heston, W. D.; Gaudin, P. B. Clin. Cancer Res. 1999,
5(10), 2674.
8. Chang, S. S.; Reuter, V. E.; Heston, W. D.; Gaudin, P. B.
Urology 2001, 57, 801.
9. Tasch, J.; Gong, M.; Sadelain, M.; Heston, W. D.
Crit. Rev. Immunol. 2001, 21, 249; Salit, R. B.; Kast, W.
M.; Velders, M. P. Front. Biosci. 2002, 7, e204; Lu, J.;
Celis, E. Cancer Res. 2002, 62, 5807; Fracasso, G.;
Bellisola, G.; Cingarlini, S.; Castelletti, D.; Prayer-Galetti,
T.; Pagano, F.; Tridente, G.; Colombatti, M. Prostate
2002, 53, 9.
21. Several representative references and Km values for
NAAG with various sources of GCPII follow.
Km =2.6nM, Ref. 15a; Km =73nM, Ref. 14b; Km =87,
100, and 146, Luthi-Carter, R.; Barczak, A. K.; Speno, H.;
Coyle, J. T. J. Pharm. Exp. Ther. 1998, 286, 1020;
Km =130nM, Rojas, C.; Frazier, S. T.; Flanary, J.;
Slusher, B. S. Anal. Biochem. 2002, 310, 50; Km =200, Ref.
´
´
20; Km =427, Barinka, C.; Rinnova, M.; Sacha, P.; Rojas,
C.; Majer, P.; Slusher, B. S.; Konvalinka, J. J. Neurochem.
2002, 80, 477Km =540, Ref. 13.
22. Oliver, A. J.; Wiest, O.; Helquist, P.; Miller, M. J.;
Tenniswood, M. Bioorg. Med. Chem. 2003, 11, 4455.
23. Lu, H.; Ng, R.; Shieh, C. C.; Martinez, A. R.; Berkman,
C. E. Phosphorus, Sulfur, and Silicon 2003, 178, 17.
10. Heston, W. D. Urology 1997, 49, 104; Pinto, J. T.;
Suffoletto, B. P.; Berzin, T. M.; Qiao, C. H.; Lin, S. L.;