Enantiospecificity of glutamate carboxypeptidase II inhibition

Article abstract of DOI:10.1021/jm049258g

Two representative glutamate carboxypeptidase II (GCP II) inhibitors, 2-(hydroxypentafluorophenylmethyl-phosphinoylmethyl)pentanedioic acid 2 and 2-(3-mercaptopropyl)pentanedioic acid 3, were synthesized in high optical purities (>97%ee). The two enantiom

Full text of DOI:10.1021/jm049258g

J. Med. Chem. 2005, 48, 2319-2324
2319
Enantiospecificity of Glutamate Carboxypeptidase II Inhibition
Takashi Tsukamoto,* Pavel Majer, Dilrukshi Vitharana, Chiyou Ni, Bunda Hin, Xi-Chun M. Lu, Ajit G. Thomas,
Krystyna M. Wozniak, David C. Calvin, Ying Wu, Barbara S. Slusher, David Scarpetti, and
George W. Bonneville
Guilford Pharmaceuticals Inc., 6611 Tributary Street, Baltimore, Maryland 21224
Received September 13, 2004
Two representative glutamate carboxypeptidase II (GCP II) inhibitors, 2-(hydroxypentafluo-
rophenylmethyl-phosphinoylmethyl)pentanedioic acid 2 and 2-(3-mercaptopropyl)pentanedioic
acid 3, were synthesized in high optical purities (>97%ee). The two enantiomers of 2 were
prepared from previously reported chiral intermediates, (R)- and (S)-2-(hydroxyphosphinoyl-
methyl)pentanedioic acid benzyl esters 8. The synthesis of (R)- and (S)-3 involves the hydrolysis
of (R)- and (S)-3-(2-oxo-tetrahydro-thiopyran-3-yl)propionic acids, (R)- and (S)-11, the corre-
sponding optically pure thiolactones delivered by chiral chromatographic separation of the
racemic 11. GCP II inhibitory assay revealed that (S)-2 is 40-fold more potent than (R)-2. In
contrast, both enantiomers of 3 inhibited GCP II with nearly equal potency. The efficacy
observed in subsequent animal studies with these enantiomers correlated well with the
inhibitory potency in a GCP II assay.
Introduction
Chart 1. Representative GCP II Inhibitors
In the brain, glutamate carboxypeptidase II (GCP II,
EC 3.4.17.21) hydrolyzes N-acetylaspartylglutamate
1
(
NAAG) into N-acetylaspartate and glutamate. Because
of this function, the enzyme has previously been referred
to as N-acetylated alpha-linked acidic dipeptidase
(
NAALADase). As NAAG is believed to be one of the
major sources of glutamate in the nervous system, there
has been a substantial interest in the discovery of potent
and selective GCP II inhibitors for the treatment of
neurological disorders associated with glutamate exci-
totoxicity such as stroke, spinal cord injury, amyotrophic
lateral sclerosis (ALS), peripheral neuropathies, chronic
pain, schizophrenia, and epilepsy.2
sents another class of inhibitors and exhibits the
inhibitory effects through the interaction of its sulfhy-
dryl group with the active site zinc atom(s). More
importantly, compound 3 is the first reported orally
bioavailable GCP II inhibitor and exhibits antinocice-
ptive effects in a rat chronic constrictive injury (CCI7)
model of neuropathic pain by daily oral administration,
thereby potentially extending the therapeutic utility of
GCP II inhibitors to multiple chronic disorders associ-
ated with glutamate excitotoxicity. Consequently, we
found that the treatment of G93A FALS (familial
amyotrophic lateral sclerosis) transgenic mice with 3 by
oral administration results in statistically significant
prolongation in median survival.9
Both of these representative GCP II inhibitors were
previously provided as racemates for the pharmacologi-
cal evaluation. Recently, studies on the two enantiomers
of 2-PMPA 1 have revealed that potent inhibition of
GCP II is specific to (S)-2-PMPA, which has an absolute
configuration corresponding to L-glutamate.10 The re-
sults suggest that only single enantiomers of 2 and 3
potently inhibit GCP II, and therefore it is essential to
identify the active enantiomers of 2 and 3 prior to
further preclinical characterization in order to eliminate
the potential for undesired pharmacological effects by
the inactive enantiomer. In this paper, we report the
A prototype GCP II inhibitor, 2-(phosphonomethyl)-
pentanedioic acid (2-PMPA 1; Chart 1),3 has been
extensively utilized in models studying the mechanism
and physiological role of GCP II as well as the potential
4
,5
therapeutic effects of GCP II inhibition. While these
studies have demonstrated the potential therapeutic
benefit of GCP II inhibition, its poor pharmacokinetic
profile has prompted us and other groups to explore
other small molecule GCP II inhibitors. Using 1 as a
template, we have extended our SAR (structure-activ-
ity relationship) studies to other zinc binding groups and
have identified two structurally distinct GCP II inhibi-
tors, 2-(hydroxy-pentafluorophenylmethyl-phosphinoyl-
6
methyl)pentanedioic acid 2 and 2-(3-mercaptopropyl)-
7
pentanedioic acid 3. The phosphinate-based inhibitor
2, which inhibits the metallopeptidase GCP II by
coordinating with the active site zinc atom(s), is highly
water-soluble and stable in media commonly used for
intravenous administration. Indeed, compound 2 shows
neuroprotective effects in the rat middle cerebral artery
occlusion (MCAO) model by intravenous administra-
6
,8
tion. 2-(3-Mercaptopropyl)pentanedioic acid 3 repre-
*
To whom correspondence should be addressed. Phone: (410) 631-
6
762. Fax: (410) 631-6797. E-mail: tsukamotot@guilfordpharm.com.
1
0.1021/jm049258g CCC: $30.25 © 2005 American Chemical Society
Published on Web 02/19/2005

2
320 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7
Tsukamoto et al.
Scheme 1. Synthesis of (()-2a
Scheme 3. Synthesis of (R)- and (S)-11a
a
Reagents and conditions: (a) BTSP, CH2Cl2. rt, 21%; (b)
BnOH, EDC, CH2Cl2, rt, 28%; (c) dibenzyl 2-methyleneglutarate,
NaH, THF, rt, 39%; (d) H2, Pd/C, H2O-THF, rt, 85%.
a
Reagents and conditions: (a) (i) triethylsilane, TFA, CH2Cl2,
5 °C; (ii) 10-camphorsulfonic acid, toluene, reflux, 38% for two
4
Scheme 2. Synthesis of (R)- and (S)-2a
steps; (b) chiral separation, 43% for (R)-11, 42% for (S)-11.
be easily attached to the phosphorus atom of 8. As
illustrated in Scheme 2, this coupling reaction was
successfully accomplished by using a procedure similar
to the one reported by Reiter’s group.11 Thus, treatment
of (R)-8 with chlorotrimethylsilane and triethylamine
in the presence of benzyl bromide 4 afforded (R)-9 in
quantitative crude yield. Complete elimination of oxygen
from the reaction media is essential for preventing the
phosphonate byproduct formation. A subsequent base-
mediated hydrolysis of (R)-9 gave (R)-2 in 42% yield.
The other enantiomer (S)-2 was successfully obtained
in a similar manner using (S)-8 as a starting material
in 61% yield. Both final products (R)- and (S)-2 were
shown to be of high optical purity (>98%ee) by chiral
HPLC analysis. In addition to its capacity to produce
enantiomers, the new synthetic route offers several
advantages over the conventional route such as reduced
number of steps and higher overall yield. The shortened
route is particularly attractive as it allows us to
synthesize a wide variety of optically pure phosphinate-
based GCP II inhibitors in two steps by substituting
other electrophilic reagents for pentafluorobenzyl bro-
mide 4.
a
Reagents and conditions: (a) 4, TMSCl, triethylamine, CH2Cl2,
3
5 °C, >100% crude yield for (R)-9, 97% crude yield for (S)-9, (b)
aq NaOH, rt, 42% for (R)-2, 61% for (S)-2.
synthesis of both enantiomers of the phosphinate-based
compound 2 and the thiol-based compound 3. Biological
evaluation of these compounds in a GCP II assay and
animal models of neurological disorders will be also
described.
Our initial attempt to prepare two enantiomers of 3
involved direct separation of racemate 3 by chiral
chromatography. After unsatisfactory attempts to sepa-
rate the two enantiomers in practical quantities, how-
ever, we shifted our focus to the corresponding thiolac-
tone 11 expecting that its cyclic system would increase
the degree of enantioseparation by chiral chromatogra-
phy. The thiolactone 11 was readily prepared from
S-trityl derivative 10 (Scheme 3). Chiral separation of
the racemic thiolactone 11 was successfully achieved on
a CHIRALPAK AD column using supercritical fluid
chromatography with carbon dioxide/methanol (85/15,
v/v) as a mobile phase to give (R)-11 in 99.9%ee and
(S)-11 in 98.1%ee, respectively. The stereochemistry at
C-2 of thiolactone 11 was established by X-ray structure
analysis of (R)-11 as shown in Figure 1.
Acid-mediated hydrolysis of thiolactone (R)-11 to (R)-3
(step a in Scheme 4) resulted in formation of completely
racemized final product 3. Under the basic conditions
(step b in Scheme 4), compound (R)-3 was consistently
obtained in slightly lower optical purity (∼90%ee)
compared to that of the starting material (R)-11. Since
we observed that the level of racemization is indepen-
dent of reaction time, we believe that racemization
Chemistry
The original synthesis of racemic 2 is outlined in
6
Scheme 1. Pentafluorobenzyl bromide 4 was treated
with bis(trimethylsilyl)phosphonite (BTSP) generated
in situ to afford phosphinic acid 5. This material was
converted into benzyl ester 6. Michael addition of 6 to
dibenzyl 2-methyleneglutarate gave the tribenzyl ester
derivative 7, which was converted into 2 by catalytic
hydrogenation. The penultimate step (synthesis of 7) of
this synthetic route not only creates a chiral center of
our interest (R-carbon of glutarate moiety) but also turns
the phosphorus atom into the second chiral center, thus
providing 7 as a mixture of four diastereomers. This
stereochemical complexity makes this synthetic route
unfeasible for the synthesis of the enantiomers of 2 and
prompted us to explore alternative synthetic routes.
We have previously reported the preparation of both
enantiomers of hydroxyphosphinylmethyl-pentanedioic
dibenzyl ester 8 with defined absolute configurations
1
0
(
Scheme 2). These compounds were successfully uti-
lized for the synthesis of (R)- and (S)-1 and also should
serve as ideal chiral intermediates for the synthesis of
2
provided that a pentafluorophenylmethyl group can

Enantiospecificity of Glutamate Carboxypeptidase II
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7 2321
Table 1. Inhibition of GCP II and Neuroprotective Effects in a
Tissue Culture Model of Ischemia
compd
IC50 (nM)
EC50 (nM)
a
(
(
(
(
(
(
(
(
(
RS)-1
0.3
30
1.2
a
R)-1
70
a
S)-1
0.1
0.2
b
RS)-2
82 ( 14
9.5 (11; 8.0)
410 (584, 227)
4.1 ( 1.5
13 ( 6
b
R)-2
1400 (1190; 1530)
34 ( 18
b
S)-2
b
RS)-3
90 ( 26
b
c
R)-3
85 ( 33
nd
nd
b
c
S)-3
67 ( 29
a
Values have been previously reported (ref 10) and are included
b
herein for reference purposes. Values are the mean ( SD for
experiments performed at least in triplicate. When only two
separate experiments were performed, values are expressed as the
mean of duplicate determinations followed by each measurement
c
separated by a semicolon in parentheses. Not determined.
an IC50 value of 34 nM. Significant difference in IC50
values between the two compounds demonstrates that
the potent inhibition of GCP II by 2 is attributable to
its S-enantiomer, which has an absolute configuration
corresponding to L-glutamate. The same trend was
previously observed with the two enantiomers of 2-P-
MPA (R)-1 and (S)-1 with IC50 values of 30 nM and 0.1
nM, respectively.10
Figure 1. Crystal structure of (R)-(-)-11.
Scheme 4. Synthesis of (R)-3 and (S)-3a
Surprisingly, no significant difference in GCP II
inhibitory potency was observed between (R)-3 and (S)-3
with IC50 values of 85 nM and 67 nM, respectively. The
lack of enantiospecificity observed with the thiol-based
inhibitor 3 is unexpected, considering the evident enan-
tiospecificity demonstrated by the two phosphorus-based
inhibitors 1 and 2. To assess the possibility of racem-
ization under the GCP II assay conditions, we incubated
(S)-3 with the assay components for 15 min. Chiral
HPLC analysis of the resulting assay mixture showed
no sign of racemization.
a
Reagents and conditions: (a) 6 M HCl, 100 °C, 0%ee; (b)
The loss of enantiospecificity cannot be solely at-
tributed to the change of the zinc-binding group from a
phosphinate to a thiol group. For example, a thiol-based
drug, (2S)-1-(3-mercapto-2-methylpropionyl)-L-proline
(captopril), is 100-fold more potent than the correspond-
ing (2R) analogue in inhibiting angiotensin-converting
NaOH, dioxane-H2O, rt, ∼90%ee; (c) LiOH, H2O2, rt; (d) TCEP,
H2O, rt, 97.1%ee for (R)-3, 97.5%ee for (S)-3.
occurs prior to cleavage of thiolactone 11, which pos-
sesses a relatively acidic R-proton compared to 3. To
prevent the partial racemization, we applied Evans’s
procedure to hydrolyze thioesters. Treatment of (R)-
1 with lithium hydroxide and hydrogen peroxide gave
1
2
15
enzyme (ACE). Kozikowski’s group has proposed the
1
fixed pharmacophore model consisting of two carboxylic
groups of the glutarate moiety and a zinc-binding group
on the basis of their predictive GCP II structures in
complex with 2-PMPA and one other phosphorus-based
inhibitor.16 Both of these compounds contain the phos-
phorus group linked to the chiral carbon through a
single methylene unit, and the model suggests that the
configuration of the chiral center has significant influ-
ence on the orientation of phosphorus group. Thus GCP
II inhibition by these phosphorus-based compounds
exhibits a higher enantiospecificity as demonstrated by
the compounds 1 and 2. We presume that the lack of
enantiospecificity observed with 3 is at least partially
associated with the flexible alkyl chain linker consisting
of three methylene units between the thiol group and
the chiral carbon. Regardless of its absolute configura-
tion, the flexible linker could allow the thiol group to
gain the same degree of interaction with an active site
zinc atom while its glutarate group optimally interacts
with glutamate recognition site of the enzyme, thus
resulting in equally potent inhibition by the two enan-
tiomers of 3.
the disulfide derivative (R,R)-12. The formation of
disulfide bond can be attributed to hydrogen peroxide-
mediated oxidation of free thiol. The disulfide (R,R)-12
was converted into (R)-3 using tris(carboxyethyl)phos-
phine (TCEP). Chiral HPLC analysis of the product
revealed that the two-step procedure provides the
product with minimal racemization, yielding (R)-3 in
9
7.1%ee. In a similar manner using (S)-11 as a starting
material, the other enantiomer (S)-3 was obtained in
7.5%ee.
9
Biological Results and Discussion
In Vitro GCP II Assay. Inhibitory potencies of these
enantiomers were evaluated using N-acetyl-L-aspartyl-
3
13
[
H]-L-glutamate as a substrate and a purified recom-
1
4
binant GCP II. The results are summarized in Table
1
10
along with previously reported values for (RS)-, (R)-,
and (S)-1 for comparison purposes.
In a GCP II assay, (R)-2 was a relatively weak
inhibitor with an IC50 value of 1400 nM while (S)-2
displayed nearly 40-fold greater inhibitory potency with

2
322 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7
Tsukamoto et al.
Figure 2. Reduction of ischemic injury by (R)- and (S)-2 in
the rat MCAO model. (R)-2 was dosed at 30 mg/kg bolus iv
(
1 h postoclusion) then at 20 mg/kg/h iv for 22 h. (S)-2 was
dosed at 3 mg/kg bolus iv (1 h postoclusion) then at 2 mg/kg/h
iv for 22 h. *37% protection (p ) 0.009), **33% protection (p
)
0.016), ***59% protection (p ) 0.005).
Cell Culture Model of Cerebral Ischemia. Since
compound 2 was intended for the treatment of ischemic
stroke, the two enantiomers of 2 were also tested in an
established tissue culture model of cerebral ischemia.17
In this model, rat cortical cultures were exposed to
potassium cyanide and 2-deoxyglucose to simulate is-
chemia. Cellular injury was quantified by measurement
of cytosolic lactate dehydrogenase released into medium.
As shown in Table 1, (RS)-2 and (S)-2 exhibited strong
neuroprotection against ischemic injury with median
effective concentration (EC50) values of 9.5 nM and 4.1
nM, respectively. (R)-2 exhibited much weaker potency
in this model with an EC50 value of 410 nM. The good
correlation between GCP II inhibitory and neuro-
protective potencies of (R)-2 (less potent in both assays)
and (S)-2 (more potent in both assays) provides further
experimental evidence that the neuroprotective effect
of these compounds is due to their ability to inhibit GCP
II.
Neuroprotective Effects of (R)- and (S)-2 in the
Rat Middle Cerebral Artery Occlusion (MCAO)
Model. The two enantiomers (R)- and (S)-2 were
evaluated for neuroprotection against ischemic injury
in a transient middle cerebral artery occlusion model
in the anesthetized rat (Figure 2). The less potent
enantiomer (R)-2 was given in bolus (30 mg/kg iv) at 1
h after occlusion followed by continuous infusion (20 mg/
kg/h iv) for 22 h. The more potent enantiomer (S)-2 was
given in a similar manner but at 10-fold lower dose (3
mg/kg bolus iv then 2 mg/kg/h iv). As shown in Figure
Figure 3. Antinociceptive effects of (R)- and (S)-3 (10 mg/kg/
day, po) in the rat chronic constriction injury (CCI) model of
neuropathic pain. Both (R)-3 (A) and (S)-3 (B) significantly
attenuated CCI-induced hyperalgesic state relative to the
vehicle-treated control (*p < 0.05).
by one measurement time point, both (R)-3 and (S)-3
significantly attenuated CCI-induced hyperalgesic state
relative to the vehicle-treated control. The fact that both
(R)- and (S)-3 as well as (()-3 exhibit a similar in vivo
efficacy profile suggests that the two enantiomers not
only inhibit GCP II with equal potency but also share a
similar pharmacokinetic profile when administered
orally.
Conclusions
The MCAO model of stroke in rats has demonstrated
a marked difference in efficacy between the two enan-
tiomers of 2, with (S)-2 providing superior neuropro-
tection. Furthermore, the similarly potent enantiomers
(R)- and (S)-3 showed comparable efficacy in the
neuropathic pain model. A good correlation between
GCP II potency and in vivo efficacy of these compounds
supports GCP II inhibition as the underlying mecha-
nisms of action in these animal models. As far as the
phosphinate-based inhibitor 2 is concerned, our studies
clearly suggest that further drug development activity
should be conducted exclusively with the more potent
enantiomer (S)-2. As for the thiol-based inhibitor 3,
additional comparative assessment (e.g., toxicology) of
the two enantiomers will be required prior to selection
of a single enantiomer for clinical development.
2
, treatment with (R)-2 produced no effect on ischemic
core infarction volumes. On the other hand, significant
reduction in the volume of ischemic injury was observed
in both cortical and subcortical regions of animals
treated with more potent enantiomer (S)-2.
Antinociceptive Effects of (R)- and (S)-3 in the
Rat Chronic Constriction Injury (CCI) Model. The
two enantiomers (R)- and (S)-3 were evaluated for their
antinociceptive effects using the rat chronic constriction
18
injury (CCI) model of neuropathic pain by oral admin-
istration. Since the two enantiomers are equally potent
GCP II inhibitors, both of them were tested at 10 mg/
kg/day by oral administration, the minimally effective
dose for the racemate (()-3 from the previously con-
Experimental Section
1H, 13C, and 31P NMR spectra were recorded on a Bruker
DRX-400 instrument at 400, 100, and 162 MHz, respectively.
Optical purity analyses were performed on a Jasco HPLC
system equipped with a chiral chromatography column. El-
emental analyses were obtained from Atlantic Microlabs,
7
ducted studies. Figure 3 summarizes the results of the
studies with (R)-3 (Figure 3A) and (S)-3 (Figure 3B).
Although onset of antinociceptive effect was different

Enantiospecificity of Glutamate Carboxypeptidase II
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7 2323
Norcross, GA, and optical rotations were obtained from Rob-
ertson Microlit Laboratories, Inc., Madison, NJ.
containing 10-camphorsulfonic acid (14 g, 60 mmol) and
refluxed for 6 h. The generated water was removed using a
Dean-Stark trap. The solvent was then removed under
reduced pressure and the residue was purified by silica gel
chromatography (Hexanes/EtOAc, 4:1). Fractions containing
the product were collected, concentrated, and recrystallized
(R)-2-(Hydroxy-pentafluorophenylmethyl-phosphinoyl-
methyl)pentanedioic Acid Dibenzyl Ester ((R)-9). A solu-
tion of (R)-8 (417.4 g, 1.069 mol) and 4 (161.5 mL, 1.069 mol)
in dichloromethane (3.13 L) was degassed and purged with
argon three times. Trimethylchlorosilane (407 mL, 3.207 mol)
and triethylamine (447 mL, 3.207 mol) were added to the
solution while maintaining the temperature below 20 °C. The
mixture was allowed to warm to room temperature and was
heated at 35 °C for 45 h. The reaction was cooled in an ice
bath and quenched with water (167 mL). Water (1.503 L) was
added and stirred gently over 20 min. The organic layer was
separated and concentrated in vacuo to give 644.8 g of (R)-9
from EtOAc/hexanes to give 23.3 g of 11 as a white solid (38%
1
yield): mp 85-86 °C; H NMR (CD
3
OD) δ 1.62-1.75 (m, 2H),
2.00-2.15 (m, 4H), 2.30-2.45 (m, 2H), 2.60-2.70 (m, 1H), 3.13
1
3
(dt, J ) 13.1, 6.0 Hz, 1H), 3.21 (dt, J ) 13.1, 6.5 Hz, 1H);
NMR (CD OD) δ 23.3, 27.4, 29.3, 31.2, 32.3, 50.0, 177.0, 206.4.
Anal. (C S) C, H, S.
C
3
8 12 3
H O
Separation of Enantiomers of 11 by Chiral Supercriti-
cal Fluid Chromatography. Racemic 11 (2.70 g, 14.3 mmol)
was passed through a CHIRALPAK AD column (21 mm id ×
as a light yellow oil (>100% crude yield). The crude material
1
was used for the next step without further purification:
H
2
50 mm length) using carbon dioxide/methanol (85/15, v/v) as
NMR (CD
3
OD) δ 1.8-2.1 (m, 3H), 2.2-2.5 (m, 3H), 2.8-3.0
eluent at a flow rate of 5 mL/min at 25 °C. Fractions containing
the first eluting enantiomer (detected by UV at 220 nm) was
concentrated to afford 1.15 g of (R)-11 as a white solid (43%
(
(
m, 1H), 3.25 (d, J ) 15.4 Hz, 2H), 5.0-5.2 (m, 4H), 7.2-7.4
1
3
3
m, 10H); C NMR (CD OD) δ 25.8 (d, J ) 85.9 Hz), 29.9 (d,
J ) 11.5 Hz), 32.1, 32.2 (d, J ) 95.1 Hz), 40.0 (d, J ) 3.8 Hz),
1
13
yield): H and C NMR spectral data were identical to those
6
7.3, 67.9, 129.0-130.0 (multiple aromatic peaks), 137.3, 137.6,
37.0-148.0 (multiple fluoroaromatic peaks), 174.0, 175.5 (d,
25
of (()-11; mp 78-79 °C; [R]
CHIRALPAK AD-RH 4.6 mm × 150 mm with hexanes/EtOH/
TFA, 90/10/0.1). Anal. (C S) C, H, S. Concentration of
D 3
) -28.8 (c 1.0, CH CN); 99.9%ee
1
(
3
1
3
J ) 6.9 Hz); P NMR (CD OD) δ 43.0 (m).
8 12 3
H O
(
R)-2-(Hydroxy-pentafluorophenylmethyl-phosphinoyl-
methyl)pentanedioic Acid ((R)-2). The above crude product
R)-9 (640 g, 1.122 mol) was dissolved in an aqueous solution
fractions containing the second eluting enantiomer gave 1.13
1
13
g of (S)-11 as a white solid (42% yield): H and C NMR
(
spectral data were identical to those of (()-11; mp 78-79 °C;
of sodium hydroxide (256 g, 6.400 mol) in water (3.2 L) and
cooled with an ice bath. The mixture was allowed to warm to
room temperature slowly overnight and stirred for 24 h. The
reaction mixture was washed with tert-butyl methyl ether (1
L × 3) and the aqueous phase was acidified to pH 1 with conc
HCl (340 mL). The acidified aqueous phase was extracted with
tert-butyl methyl ether (3 L × 3) after saturating with salt.
The combined extracts were concentrated in vacuo and the
resulting oil was azeotroped with dichloromethane (2 L). The
resulting pale yellow solid was taken up in THF (2 L) and a
colorless solid was filtered off. The resulting solid was taken
up in 6 M HCl (1.5 L) and extracted with tert-butyl methyl
ether (1 L × 3). The extracts were combined with the filtrate
and concentrated in vacuo. The residue was triturated with
25
[
R]
D
) +28.8 (c 1.0, CH
3
CN); 98.1% (CHIRALPAK AD-RH
4
.6 mm × 150 mm with hexanes/EtOH/TFA, 90/10/0.1). Anal.
(
C
8
H
12
O
3
S) C, H, S.
R)-2-(3-Mercaptopropyl)pentanedioic Acid (R)-3. To
O-THF (v/v )
O (36 mg, 1.25 mmol) and
O, 0.3 mL, 2.6 mmol) at 0 °C. The mixture
was stirred at 0 °C for 3 h and quenched by the addition of
saturated aq Na (2.0 mL). The mixture was acidified to
pH 1 by 10% aq KHSO and extracted with diethyl ether (10.0
mL × 3). The combined extracts were washed with brine, dried
over MgSO , and concentrated to give (R,R)-12 as a colorless
oil. To a solution of (R,R)-12 in 0.5 N NaHCO (3.0 mL) was
(
a solution of (R)-11 (100 mg, 0.5 mmol) in H
2
3
H
/5, 0.8 mL) were added LiOH‚H
2
2
O
2
(30% in H
2
2 2 3
S O
4
4
3
added tris(carboxyethyl)phosphine hydrochloride (144 mg, 0.5
mmol), and the mixture was stirred at room temperature for
10% tert-butyl methyl ether in dichloromethane (2.0 L) over-
night. The resulting solid was filtered and washed with
dichloromethane (200 mL × 2). The solid was triturated one
more time with 10% tert-butyl methyl ether in dichlo-
romethane (2.0 L) for 2 days. The resulting solid was filtered
and washed with dichloromethane (1.6 L). The solid was dried
2
min. The reaction mixture was acidified to pH 1 with 10%
KHSO
and extracted with EtOAc (10.0 mL × 3). The
combined extracts were washed with brine, dried over MgSO
and concentrated to give 70 mg of (R)-3 as a colorless oil (70%
4
4
,
1
in vacuo at 22 °C to give 182 g of (R)-2 as a white solid (42%
3
overall yield): H NMR (CDCl ) δ 1.35 (t, J ) 7.9 Hz, 1H),
1
yield for the two steps): mp 152-154 °C; H NMR (D
2
O) δ
1.55-1.80 (m, 4H), 1.80-2.05 (m, 2H), 2.44 (t, J ) 7.2 Hz, 2H),
2.40-2.50 (m, 1H), 2.53 (q, J ) 7.1 Hz, 2H), 10.7-11.20 (br,
1
.7-1.9 (m, 3H), 2.09 (ddd, J ) 9.1, 13.1, 15.4 Hz, 1H), 2.31
dt, J ) 16.9, 7.3 Hz, 1H), 2.36 (dt, J ) 16.9, 7.3 Hz, 1H), 2.6-
.8 (m, 1H), 3.15 (d, J ) 15.2 Hz); 13C NMR (D
O) δ 22.8 (d, J
86.6 Hz), 26.5 (d, J ) 12.3 Hz), 28.8 (d, J ) 92.8 Hz), 29.2,
6.8 (d, J ) 3.8 Hz), 134.0-145.0 (multiple fluoroaromatic
1
3
(
2H); C NMR (CDCl
3
) δ 24.3, 26.6, 30.5, 31.4, 31.7, 44.1, 179.4,
) +1.70 (c 1.0, CH CN); 97.1%ee (CHIRALCEL
OJ-H 4.6 mm × 250 mm with hexanes/i-PrOH/TFA, 90/10/
0.1). Anal. (C S‚0.25H O) C, H, S.
S)-2-(3-Mercaptopropyl)pentanedioic Acid (S)-3. The
25
2
2
181.7; [R]
D
3
)
3
8
H
14
O
4
2
3
1
peaks), 175.6, 176.6 (d, J ) 6.1 Hz); P NMR (D
2
O) δ 44.9
O); 98.3%ee (CHIRALPAK AD-H
.6 mm × 250 mm with hexanes/i-PrOH/TFA, 90/10/0.1). Anal.
P) C, H.
S)-2-(Hydroxy-pentafluorophenylmethyl-phosphinoyl-
(
2
5
(
D 2
m); [R] ) +1.85 (c 1.0, H
(
(
S)-enantiomer (S)-3 was synthesized as described above for
4
1
13
R)-3 except starting with (S)-11 (70% overall yield): H,
C
(
13
C H
12
F
5
O
6
25
NMR spectral data were identical to those of (R)-3; [R]
D
)
(
-2.26 (c 1.0, CH CN); 97.5%ee (CHIRALCEL OJ-H 4.6 mm ×
3
methyl)pentanedioic Acid ((S)-2). The (S)-enantiomer (S)-2
was synthesized as described above for (R)-2 except starting
2
0
8 14 4
50 mm with hexanes/i-PrOH/TFA, 90/10/0.1). Anal. (C H O S‚
.25H
2
O) C, H, S.
1
with (S)-8 (59% yield for the two steps): mp 155-157 °C; H,
Biological and Pharmacological Studies. The GCP II
13
31
C, and P NMR spectral data were identical to those of (R)-
13
4,17
assay and cell culture model of cerebral ischemia
were
2
5
2
; [R]
.6 mm × 250 mm with hexanes/i-PrOH/TFA, 90/10/0.1). Anal.
P) C, H.
-(2-Oxo-tetrahydro-thiopyran-3-yl)propionic Acid (11).
D 2
) -1.64 (c 1.0, H O); 98.6%ee (CHIRALPAK AD-H
carried out as outlined previously. The rat middle cerebral
4
8
artery occlusion model was performed as previously described.
(
C
13
H
12
F
5
O
6
The chronic constrictive injury models were performed follow-
3
18
ing the procedure reported by Bennett’s group.
To a solution of 10 (150 g, 334 mmol) in dichloromethane (500
mL) was dropwise added trifluoroacetic acid (110 mL) over 30
min. The resulting brown solution was stirred at room tem-
perature for additional 30 min. A solution of triethylsilane (45
mL, 0.33 mol) in dichloromethane (50 mL) was added to the
mixture and stirred at 45 °C for 1h. The volatiles were removed
in vacuo and the residue was triturated with hexanes (500
mL × 2). The oily residue was dissolved in toluene (500 mL)
Acknowledgment. Synthesis of both enantiomers
of 2 and chiral separation of 11 were codeveloped with
Oxford Asymmetry International and Chiral Technol-
ogy, respectively. X-ray structure analysis of (R)-(-)-
11 was performed by Rigaku MSC.

2
324 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7
Supporting Information Available: X-ray crystallo-
Tsukamoto et al.
(9) Ghadge, G. D.; Slusher, B. S.; Bodner, A.; Canto, M. D.; Wozniak,
K.; Thomas, A. G.; Rojas, C.; Tsukamoto, T.; Majer, P.; Miller,
R. J.; Monti, A. L.; Roos, R. P. Glutamate carboxypeptidase II
inhibition protects motor neurons from death in familial amyo-
trophic lateral sclerosis models. Proc. Natl. Acad. Sci. U.S.A.
graphic analysis of (R)-11. This material is available free of
charge via the Internet at http://pubs.acs.org.
References
2
003, 100, 9554-9559.
(
1) Carter, R. E.; Coyle, J. T. Glutamate Carboxypeptidase II. In
Handbook of Proteolytic Enzymes, Barrett, A. J., Rawlings, N.
D., Woessner, J. F., Eds. Academic: New York, 1998; pp 1434-
(10) Vitharana, D.; France, J. E.; Scarpetti, D.; Bonneville, G. W.;
Majer, P.; Tsukamoto, T. Synthesis and biological evaluation of
(
R)- and (S)-2-(phosphonomethyl)pentanedioic acids as inhibitors
of glutamate carboxypeptidase II. Tetrahedron: Asymm. 2002,
3, 1609-1614.
1437.
(2) Whelan, J. NAALADase inhibitors: A novel approach to glutamate
1
regulation. Drug Discovery Today 2000, 5, 171-172.
(
11) Reiter, L. A.; Jones, B. P. Amide-assisted hydrolysis of â-car-
boxamido-substituted phosphinic acid esters. J. Org. Chem.
(
3) Jackson, P. F.; Cole, D. C.; Slusher, B. S.; Stetz, S. L.; Ross, L.
E.; Donzanti, B. A.; Trainor, D. A. Design, synthesis, and
biological activity of a potent inhibitor of the neuropeptidase
N-acetylated alpha-linked acidic dipeptidase. J. Med. Chem.
1
997, 62, 2808-2812.
(12) Evans, D. A.; Wu, L. D.; Wiener, J. J. M.; Johnson, J. S.; Ripin,
D. H. B.; Tedrow, J. S. A general method for the synthesis of
enantiomerically pure â-substituted, â-amino acids through
R-substituted succinic acid derivatives. J. Org. Chem. 1999, 64,
1
996, 39, 619-622.
(
4) Slusher, B. S.; Vornov, J. J.; Thomas, A. G.; Hurn, P. D.;
Harukuni, I.; Bhardwaj, A.; Traystman, R. J.; Robinson, M. B.;
Britton, P.; Lu, X. C.; Tortella, F. C.; Wozniak, K. M.; Yudkoff,
M.; Potter, B. M.; Jackson, P. F. Selective inhibition of NAALA-
Dase, which converts NAAG to glutamate, reduces ischemic
brain injury. Nat. Med. 1999, 5, 1396-1402.
6
411-6417.
(
13) Rojas, C.; Frazier, S. T.; Flanary, J.; Slusher, B. S. Kinetics and
inhibition of glutamate carboxypeptidase II using a microplate
assay. Anal. Biochem. 2002, 310, 50-54.
14) Barinka, C.; Rinnova, M.; Sacha, P.; Rojas, C.; Majer, P.; Slusher,
B. S.; Konvalinka, J. Substrate specificity, inhibition and enzy-
mological analysis of recombinant human glutamate carbox-
ypeptidase II. J. Neurochem. 2002, 80, 477-487.
(15) Cushman, D. W.; Cheung, H. S.; Sabo, E. F.; Ondetti, M. A.
Design of potent competitive inhibitors of angiotensin-converting
enzyme. Carboxyalkanoyl and mercaptoalkanoyl amino acids.
Biochemistry 1977, 16, 5484-5491.
16) Rong, S. B.; Zhang, J.; Neale, J. H.; Wroblewski, J. T.; Wang,
S.; Kozikowski, A. P. Molecular modeling of the interactions of
glutamate carboxypeptidase II with its potent NAAG-based
inhibitors. J. Med. Chem. 2002, 45, 4140-4152.
(
5) Chen, S. R.; Wozniak, K. M.; Slusher, B. S.; Pan, H. L. Effect of
(
2
-(phosphono-methyl)-pentanedioic acid on allodynia and affer-
ent ectopic discharges in a rat model of neuropathic pain. J.
Pharmacol. Exp. Ther. 2002, 300, 662-667.
(
6) Jackson, P. F.; Tays, K. L.; Maclin, K. M.; Ko, Y. S.; Li, W.;
Vitharana, D.; Tsukamoto, T.; Stoermer, D.; Lu, X. C.; Wozniak,
K.; Slusher, B. S. Design and pharmacological activity of
phosphinic acid based NAALADase inhibitors. J. Med. Chem.
2
001, 44, 4170-4175.
(
7) Majer, P.; Jackson, P. F.; Delahanty, G.; Grella, B. S.; Ko, Y. S.;
Li, W.; Liu, Q.; Maclin, K. M.; Polakova, J.; Shaffer, K. A.;
Stoermer, D.; Vitharana, D.; Wang, E. Y.; Zakrzewski, A.; Rojas,
C.; Slusher, B. S.; Wozniak, K. M.; Burak, E.; Limsakun, T.;
Tsukamoto, T. Synthesis and biological evaluation of thiol-based
inhibitors of glutamate carboxypeptidase II: discovery of an
orally active GCP II inhibitor. J. Med. Chem. 2003, 46, 1989-
(
(
17) Vornov, J. J. Toxic NMDA-receptor activation occurs during
recovery in a tissue culture model of ischemia. J. Neurochem.
1
995, 65, 1681-1691.
1996.
(
8) Williams, A. J.; Lu, X. M.; Slusher, B.; Tortella, F. C. Electro-
encephalogram analysis and neuroprotective profile of the
N-acetylated-alpha-linked acidic dipeptidase inhibitor, GPI5232,
in normal and brain-injured rats. J. Pharmacol. Exp. Ther. 2001,
(18) Bennett, G. J.; Xie, Y. K. A peripheral mononeuropathy in rat
that produces disorders of pain sensation like those seen in man.
Pain 1988, 33, 87-107.
299, 48-57.
JM049258G