A new class of conformationally rigid analogues of 4-amino-5- halopentanoic acids, potent inactivators of γ-aminobutyric acid aminotransferase

Article abstract of DOI:10.1021/jm9904755

Recently, we found (Qiu, J.; Pingsterhaus, J. M.; Silverman, R. B. J. Med. Chem. 1999, 42, 4725-4728) that conformationally rigid analogues of the GABA aminotransferase (GABA-AT) inactivator vigabatrin were not inactivators of GABA-AT. To determine if thi

Full text of DOI:10.1021/jm9904755

706
J . Med. Chem. 2000, 43, 706-720
A New Cla ss of Con for m a tion a lly Rigid An a logu es of 4-Am in o-5-h a lop en ta n oic
Acid s, P oten t In a ctiva tor s of γ-Am in obu tyr ic Acid Am in otr a n sfer a se
J ian Qiu and Richard B. Silverman*
Department of Chemistry and Department of Biochemistry, Molecular Biology, and Cell Biology,
Northwestern University, Evanston, Illinois 60208-3113
Received September 17, 1999
Recently, we found (Qiu, J .; Pingsterhaus, J . M.; Silverman, R. B. J . Med. Chem. 1999, 42,
4725-4728) that conformationally rigid analogues of the GABA aminotransferase (GABA-AT)
inactivator vigabatrin were not inactivators of GABA-AT. To determine if this is a general
phenomenon of GABA-AT inactivators, several mono- and di-halogen-substituted conforma-
tionally rigid analogues (7-15) of other GABA-AT inactivators, 4-amino-5-halopentanoic
acids, were synthesized as potential inactivators of GABA-AT. Four of them, (+)-7, (-)-9,
(+)-10, and (+)-15, were inactivators, although not as potent as the corresponding open-
chain analogues. The maximal inactivation rate constants, kinact, for the fluoro- and bromo-
substituted analogues were comparable, indicating that cleavage of the C-X bond is not
rate determining. Consistent with that observation is the finding that [3-2H]-10 exhibits a
deuterium isotope effect on inactivation of 3.3, suggesting that C-H bond cleavage is the
rate-determining step. The rate of inactivation of GABA-AT by the fluorinated analogue 7
is 1/15 that of inactivation by the corresponding open-chain analogue, 4-amino-5-fluoro-
pentanoic acid (3a ). Whereas inactivation by 3a releases only one fluoride ion, inactivation
by 7 releases 148 fluoride ions, accounting for the less efficient inactivation rate. Inactiva-
tion leads to covalent attachment of 2 equiv of inactivator after gel filtration; upon urea
denaturation, 1 equiv of radioactivity remains bound to the enzyme. This suggests that, un-
like the open-chain anlogue, the conformationally rigid analogue becomes, at least partially,
attached to an active-site residue. It appears that the conformational constraint has a larger
effect on inactivators that inactivate by a Michael addition mechanism than by an enamine
mechanism.
Sch em e 1
γ-Aminobutyric acid aminotransferase (GABA-AT)
is a pyridoxal 5′-phosphate (PLP)-dependent enzyme
that catalyzes the transformation of the inhibitory
neurotransmitter γ-aminobutyric acid (1, GABA) to
succinic semialdehyde and to the excitatory neuro-
transmitter L-glutamic acid (Scheme 1).1 Inhibitors
of GABA-AT exhibit anticonvulsant activity by in-
creasing the concentration of 1 in the brain.2 Vigaba-
trin (2) is a time-dependent, mechanism-based, inacti-
vator of GABA-AT currently in medical use for the
treatment of epilepsy3 and is under investigation as a
treatment for drug addiction.4 Mechanism-based inac-
tivators are unreactive prior to their turnover by the
target enzyme, so they tend to cause fewer side reactions
than affinity labeling agents.5 One of the general
strategies for the design of mechanism-based inactiva-
tors of PLP-dependent enzymes is to incorp-
orate a good leaving group in the â-position of an
amino acid that acts as a substrate for the targeted
enzyme.5 For example, â-haloalanines are valuable
mechanism-based inactivators of aminotransferases,6
racemases,7 and decarboxylases.8 4-Amino-5-halopen-
tanic acids (3) were rationally designed as mechanism-
based inactivators of GABA-AT by introduction of a good
leaving group â to the amino group.9 The fluoro ana-
logue of this series (3a ) was shown to be a potent
inactivator of GABA-AT with a highly efficient partition
ratio (release of product per inactivation event) of 0.10
Its inactivation mechanism was shown to proceed via
* Address correspondence to this author at the Department of
Chemistry. Phone: (847) 491-5653. Fax: (847) 491-7713. E-mail:
Agman@chem.nwu.edu.
10.1021/jm9904755 CCC: $19.00 © 2000 American Chemical Society
Published on Web 02/04/2000

Inactivators of γ-Aminobutyric Acid Aminotransferase
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 707
Sch em e 2
Sch em e 3
an enamine pathway in which an active enamine
intermediate (4), generated by GABA-AT, modifies the
enzyme-bound PLP (5, Scheme 2).
Ch em istr y
All analogues (7-15) were synthesized from a com-
mon starting material, 2-azabicyclo[2.2.1]hept-5-en-3-
one (16), which is commercially available in both
enantiomeric forms from Aldrich. It is necessary to
protect the lactam of 16 to prevent formation of complex
mixtures during the syntheses. Enantiopure (+)-7 was
synthesized from enantiopure (-)-16 as shown in Scheme
3. Treatment of (-)-16 with benzylbromide in the
presence of powdered potassium hydroxide gave the
The anticonvulsant drug vigabatrin (2) inactivates
GABA-AT by two mechanisms, mostly by a Michael
addition pathway, but a small amount by an enamine
pathway.11 Recently, (1R,4S)-4-aminocyclopent-2-ene-
1-carboxylic acid (6), a conformationally rigid analogue
of vigabatrin, was shown to be a potent inhibitor and
substrate of GABA-AT, but it was not a time-dependent
inactivator.12 It was surprising that the corresponding
conformationally rigid analogue (6) exhibited no time-
dependent inhibition. To determine if conversion of
acyclic inactivators into cyclic ones generally interferes
with mechanism-based inactivation of GABA-AT, we
constructed a variety of conformationally rigid ana-
logues of 4-amino-5-halopentanoic acids (3), compounds
known to inactivate GABA-AT exclusively by an enam-
ine pathway (pathway a, Scheme 2). Several of the
analogues described here are time-dependent, mecha-
nism-based inactivators, as in the case of the corre-
sponding open-chain (GABA) analogues.
protected lactam 17,13 which was treated with a solution
of 1,3-dibromo-5,5-dimethylhydantoin in glacial acetic
acid at room temperature to afford 18. The absolute
stereochemistry of 17 is switched in going to 18, as

708 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Qiu and Silverman
Sch em e 4
Sch em e 5
previously reported,14 but this is the required stereo-
chemistry for the final product (+)-7. Basic hydrolysis
of 18 with potassium carbonate and fluorination of 19
with (diethylamino)sulfur trifluoride (DAST)15 gave
desired product 20; dehydrobromination of 20 with
tributyltin hydride in the presence of a catalytic am-
mount of azoisobutyronitrile (AIBN) in dry benzene
furnished the monofluoro lactam 21 in an excellent
yield.14 Birch reduction was applied to deprotect the
benzyl group in 21 and afford lactam 22. Without any
purification, the resultant lactam 22 was hydrolyzed in
2 N HCl/acetic acid at 65 °C to give the final product
(+)-7.
F igu r e 1. Achimeric assistance by the carbamate group of
31 to give retention of configuration in 32.
Sch em e 6
The difluoro analogue (()-8 was synthesized from (()-
23 (Scheme 4), which was prepared in four steps.14 The
oxidation of 23 with a catalytic amount of tetrapropy-
lammonium perruthenate (TPAP) in the presence of the
co-oxidant, 4-methylmorpholine N-oxide (NMO), af-
forded ketone 24.16 Difluorination of 24 with DAST, in
benzene at room temperature, gave the difluorolactam
25, which was deprotected with potassium persulfate17
in the presence of potassium hydrogen phosphate at 80
°C to afford the difluorolactam 26. Unsuccessful at-
tempts at deprotection included oxidation of the 4-meth-
oxybenzyl group in 25 with ceric ammonium nitrate
(CAN), which predominantly afforded the oxidation
product instead of the required deprotected product 26
(less than 10% yield), and neat, anhydrous trifluoro-
acetic acid18 and trifluoroacetic acid-anisole in boiling
1,2-dichloroethane.19 Acidic hydrolysis of 26 gave the
final racemic product (8).
lyzed in a suspension of potassium carbonate, methanol,
and water to 30. Treatment of 30 with diazomethane
afforded 31 in high yield. Bromination of 31 with carbon
tetrabromide and triphenyl phosphine gave 32 with
retention of the stereochemistry at C-4, as evidenced
by the observation of a NOE between H-1 and H-3 and
between H-4 and one of the protons on C-2. The lack of
a NOE between H-4 and H-1 strongly supports a trans
relationship for protons H-4 and H-1. Retention of the
stereochemistry in going from 31 to 32 was probably
facilitated by participation of the neighbor group (t-
Boc)20 in the bromination step (Figure 1). Acid hydroly-
sis gave (-)-9.
Compound (-)-9 was synthesized starting from (-)-
16 (Scheme 5). The enantiopure compound 28 was
obtained in four steps (Scheme 5).14 Protection of 2814
gave the t-Boc protected lactam 29, which was hydro-
The synthesis of dibromo analogue 10 was straight-
forward starting from (-)-16 (Scheme 6). Bromination

Inactivators of γ-Aminobutyric Acid Aminotransferase
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 709
Sch em e 7
Sch em e 8
Sch em e 9
of (-)-16 had been shown to give dibromo compound 33
by way of an elegant “molecular somersault”, which was
hydrolyzed to (+)-10 (Scheme 6).21 The synthesis of the
2-bromo analogue 11 was achieved from racemic 4-meth-
oxybenzyl protected lactam 3414 (Scheme 7). Treatment
of (()-34 with bromine at room temperature gave
dibromolactam 35 in high yield. Elimination of hydro-
bromide in 35, by treatment with 1,5-diazobicyclo[4.3.0]-
non-5-ene (DBN) in refluxing toluene and xylene, did
not give satisfactory results. In refluxing toluene the
reaction did not go to completion and gave rise to a
complex mixture of 35 and 36. In refluxing xylene the
reaction temperature was too high, thereby decomposing
the starting material. However, the elimination of
hydrobromide from 35 to 36 was successful using neat
DBN at 120 °C. Compound 36 was converted by
hydrogenation, deprotection (CAN), and acidic hydroly-
sis to (()-11, which was purified by ion-exchange
chromatography.
Enantiomerically pure analogue 12 was made by
hydrolysis of enantiopure lactam 39 (Scheme 8), which
was prepared in five steps following a known proce-
dure.14 Analogue 13 was obtained as a racemic mixture
starting from racemic 16 (Scheme 9). Deprotection of
the 4-methoxybenzyl group of 4014 gave lactam 41; acid
hydrolysis of 41 gave (()-13. The synthesis of analogue
14 was straightforward (Scheme 10). Deprotection of
(()-42 with K2S2O8 gave 43, which was hydrolyzed to
racemic 14.
Sch em e 10
at 0 °C. None of these conditions gave one exclusive
product. It was expected that there would be only one
rearranged product, as was the case after acidic opening
of norbornene oxide.22 The mixture of 45 and 46,
resulting from the opening of 44 with 48% HBr, could
not be separated by either flash column chromatography
on silica gel or recrystallization from ethyl acetate/
hexane. However, after the hydroxyl groups of 45 and
46 were masked by trimethylsilylation, the resulting
products (47 and 48) could be easily separated by flash
column chromatography on silica gel. The structures of
45 and 46 were identified by a chemical derivatization
method (Scheme 12) and by NOE experiments. After
separation by column chromatography, 47 and 48 were
each treated with tetrabutylammonium fluoride (TBAF)
to afford pure 45 and 46, respectively. Oxidation of 45
and 46 with TPAP afforded ketones 51 and 52. NMR
spectral analyses of these products readily identified the
protons attached to the carbons bearing the hydroxyl
groups in 45 and 46. Furthermore, NOE experiments
with 45 showed that there were no interactions between
H-7 and H-5, and between H-6 and H-4, but there are
The synthesis of (+)-15 is shown in Scheme 11.
Epoxidation of (-)-34 with m-chloroperbenzoic acid in
boiling CHCl3 gave exclusively the exo-epoxide21 44.
Opening of the epoxide was attempted under different
conditions, e.g., HBr(g) in dry dichloromethane at 0 °C,
HBr(g) in CH3CN, and aqueous HBr (48%) in CH3CN

710 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Qiu and Silverman
Sch em e 11
Sch em e 12
interactions between H-6 and H-5 and between H-6 and
H-1. These NOE experiments support the proposed
structure for the rearranged product 45 depicted in
Scheme 11. The structure of 46 was identified indirectly
by determining the structure of 52 as depicted in
Scheme 12. NOE experiments performed on 52 il-
lustrated interactions between H-4 and H-5, H-5 and
H-7, and no interactions between H-5 and H-1. The
opening of the epoxide 44 with HBr gave two products:
the required rearranged product 45, and the directly
opened product 46. The optimized conditions for ring
opening were found to be aqueous 48% HBr in CH3CN
at 0 °C, giving the highest ratio of 45 to 46 (3:1) in a
75% overall yield. Fluorination of 45 with DAST gave a
single product (49), via an SN2 mechanism (Scheme 11).
The final product ((+)-15) was obtained after deprotec-
tion of 49 and acid hydrolysis of 50.
Two isotopically labelled analogues also were synthe-
sized for mechanistic studies. The deuterated dibromo
analogue ((()-61; Scheme 13) was synthesized from (()-
53. TPAP oxidation of the alcohol gave ketone 54.
Introduction of the deuterium from the favored endo
side with NaBD4 gave 55, which was converted to the
dibromide 56. Elimination of HBr followed by tributyltin
hydride-induced debromination gave 58. Bromination,
deprotection, and hydrolysis produced the deuterated
analogue 61.
To incorporate a radiolabel into 10 (Scheme 14),
enantiopure 62 was treated with tributyltin chloride and
tritiated sodium borohydride (all of the reactions in this
scheme were first performed in deuterated forms, then
with tritium). Tritiated 63 (63-T) was dibrominated and
deprotected and the lactam hydrolyzed to give [5-3H]-
(+)-66 (i.e., tritiated (+)-10).
Resu lts a n d Discu ssion
Kinetic constants for the inhibition and/or inactivation
of GABA-AT by each of the conformationally rigid
analogues are given in Table 1. Analogues (+)-7, (-)-9,
(+)-10, and (+)-15 are time-dependent inactivators of
GABA-AT. After incubation of GABA-AT with each of
these compounds until less than 10% of the enzyme
activity remained, gel filtration to remove excess and
noncovalently bound inactivators did not restore the
enzyme activity. This shows that all four of these
analogues irreversibly inactivate GABA-AT. Inactiva-
tion by (+)-10 was prevented by addition of GABA to
the incubation solution, indicating that inactivation is

Inactivators of γ-Aminobutyric Acid Aminotransferase
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 711
Sch em e 13
Sch em e 14
Ta ble 1. Kinetic Constants for Inhibition and Inactivation of
GABA-AT by Conformationally Rigid Analogues
derivative, (()-4-amino-5-fluoropentanoic acid (3a ; KI
) 0.40 mM),9b is comparable to that for (+)-7, but the
kinact value of 3a (0.50 min-1) is 17 times greater. The
smaller kinact value for 7 could be a function of its
constrained structure, which could either misorient the
proton abstracted so that it does not have maximal
orbital overlap with the PLP p-orbitals, thereby slowing
down the reaction rate, or misorient the reaction of the
activated species with the enzyme. However, it was
found that for (+)-7 148 ( 8 fluoride ions are released
per enzyme dimer inactivated (i.e., the partition ratio
is 147 ( 8), whereas for 3a the partition ratio is 0 (only
one fluoride ion is released during inactivation).10 The
lower inactivation rate, therefore, is, more likely, a
function of the lack of efficiency of inactivation with 7
relative to 3a as a result of the constraint of the
molecule. With both 3a and 7, no transamination occurs
during inactivation, only fluoride ion release. The fact
that 7 is an inactivator at all, however, is quite surpris-
kinact
kinact/KI
compd
KI (mM)
(min-1) (min-1 mM-1
)
Ki (mM)
(+)-7 0.43
0.03
0.070
0.012 ( 0.002
0.19 ( 0.03
3.6 ( 0.4
4.2 ( 0.3
12 ( 2
(D,L)-8 no inactivation
(-)-9 16
0.02
0.18
0.0037
0.0013
(+)-10 49
(D,L)-11 no inactivation
(+)-12 no inactivation
(D,L)-13 no inactivation
(D,L)-14 no inactivation
(+)-15 3.14
4.7 ( 0.6
13 ( 2
5.9 ( 0.6
1.6 ( 0.2
0.023
0.0073
an active-site phenomenon. If the PLP is converted to
PMP with GABA in the absence of R-ketoglutarate to
convert it back to PLP, then (+)-10 does not inactivate
the enzyme. Therefore, PLP is required for inactivation.
Of the four inactivators, (+)-7 is the most potent based
on the values of kinact/KI. The KI value for the open-chain

712 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Qiu and Silverman
Sch em e 15
about 22% of the total of one fluoride ion (2.4 ( 0.1 mM
fluoride ion) was released in a period of 16 h. Elimina-
tion of HF from 8, apparently, does not lead to inactiva-
tion. Presumably, the product of elimination is not
susceptible to Michael addition by an active-site nu-
cleophile (Scheme 2, pathway b) nor can it undergo the
enamine mechanism (Scheme 2, pathway a).
F igu r e 2. Deuterium isotope effects on the rate of inactivation
of GABA-AT by (()-10 and (()-61. From a plot of the log of
the remaining activity vs the time of incubation, values for
the t1/2 of inactivation at varying concentrations of (()-10 (4)
and (()-61 (b) were obtained.
Above, it was demonstrated that 10 inactivates
GABA-AT, and the rate-determining step is C-H bond
cleavage. In the case of 10, abstraction of the C-3 proton
gives a stabilized carbanion that can lead to elimination
of one or both of the two bromines in the molecule. It
was previously shown that 4-amino-3-halobutanoic acids
(67, Scheme 15) are good substrates for GABA-AT but
do not cause inactivation; GABA-AT-dependent elimi-
nation of the halide ion and transfer of the PLP back
onto the active-site lysine residue gives enamine 68,
which presumably is not oriented properly for addition
to the enzyme-bound PLP (Scheme 15).24 This suggests
that, in the case of 10, elimination of the C-2 bromide
should not be responsible for inactivation but elimina-
tion of the C-4 bromide should, because the C-4 bromide
mimics the structure of the 4-amino-5-halopentanoic
acid inactivators. Analogues (-)-9 and (()-11 support
that hypothesis. Only (-)-9, the 4-amino-5-halopen-
tanoic acid mimic, is an inactivator, not 11, the 4-amino-
3-halobutanoic acid mimic (Table 1). Unfortunately, 11
was too unstable to allow for isolation of the expected
elimination product.
Because elimination of both the C-2 and C-4 bromines
can occur, we wondered if competition between these
pathways could be influenced by a change in the leaving
group ability of the substituent. Consequently, (+)-12
was synthesized. Because the better leaving group is
at C-2, preferential HBr elimination should occur,
leading to the enamine pathway, such as in Scheme 15,
which does not cause inactivation. In fact (Table 1), (+)-
12 is not an inactivator of GABA-AT. However, when
these two halogens are reversed (the compound in which
the fluorine has R stereochemistry is much more dif-
ficult to synthesize, so the fluorine was incorporated
with S stereochemistry), as in the case of (+)-15, the
compound is an inactivator; elimination of HBr from
enzyme-bound 15 gives an intermediate equivalent to
that from the inactivator (+)-10. Similarly, 13 does not
inactivate GABA-AT. When the bromine and hydroxyl
groups were reversed (14), which should give a good
inactivator, only slow inactivation (10% loss of enzyme
activity in 4 h at 19.4 mM concentration) was observed.
ing because the conformationally rigid analogue of
4-amino-5-hexenoic acid (vigabatrin, 2), namely 6, is not
an inactivator. This may be a function of the difference
in inactivation mechanisms between vigabatrin (a
Michael addition mechanism; see Scheme 2, pathway b
for a Michael addition mechanism with 3a ) and 3a (an
enamine mechanism; Scheme 2, pathway a) and sug-
gests that the leaving group orientation (such as the
fluorine in 7) is not as important as the double bond
orientation (in 6) for inactivation.
The similarity of the maximal inactivation rate con-
stants (kinact) for (-)-9 and (+)-7, which differ only by
the halogen atom, suggests that the cleavage of the
carbon-halogen bond is not the rate-determining step;
if it were, then the kinact for 9 (X ) Br) would have been
greater than that for 7 (X ) F). Further support for this
conclusion comes from a comparison of the kinetic
constants for (()-10 with (()-61 (i.e., [3-2H]-(()-10; see
Scheme 13). As calculated from the results in Figure 2,
Hkinact/Dkinact ) 3.3 and (Hkinact/HKI)/(Dkinact/DKI) ) 8.1.
Therefore, abstraction of the proton adjacent to the
amino group is the rate-determining step, as is the case
with 3a .9c The isotope effect result also confirms that
10 is a mechanism-based inactivator; inactivation is
coupled to an enzyme-catalyzed process, namely, proton
removal.
Because of the effectiveness of (+)-7 as an inactivator,
and the fact that 4-amino-5,5-difluoropentanoic acid (the
difluoro analogue of 3a ) also is an inactivator of GABA-
AT,23 the corresponding conformationally rigid analogue
of 4-amino-5,5-difluoropentanoic acid, (()-8, was syn-
thesized. Surprisingly, 8 exhibits no time-dependent
inhibition, although it is a competitive inhibitor with a
Ki value of 190 µM (Table 1). Incubation of GABA-AT
with 8 in the presence of [14C]-R-ketoglutarate, even for
16 h, produces no [14C]-L-glutamate, indicating that 8
does not act as a substrate, reducing PLP to PMP.
However, a fluoride ion electrode was used to show that
fluoride ion is slowly released from 8 by GABA-AT;

Inactivators of γ-Aminobutyric Acid Aminotransferase
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 713
hydride. Sodium borohydride [3H] (100 mCi, 15 Ci/mmol) was
purchased from the American Radiolabeled Chemicals, Inc.
Possibly, there is hydrogen bonding to the hydroxyl
group which interferes with the compound from assum-
ing the appropriate orientation for inactivation.
Gen er a l P r oced u r e for th e Con ver sion of a Hyd r oxyl
Gr ou p in to Flu or in e. (Diethylamino)sulfur trifluoride (DAST)
(1.2 equiv) was added to a stirred solution of an alcohol (1.0
equiv) in anhydrous CH2Cl2 (15 mL) at -80 °C. After the
resulting solution was stirred for 1.5 h at -80 °C, it was
allowed to stir for 9 h at room temperature, then it was diluted
with dichloromethane (100 mL). The resulting organic layer
was washed with saturated NaHCO3 (20 mL) and brine (20
mL), dried over anhydrous Na2SO4, and concentrated under
reduced pressure. The residue was purified by flash column
chromatography on silica gel, eluting with ethyl acetate/
hexane, to afford the fluorobromo product.
Inactivation by, at least, (+)-10 is a covalent process.
Following inactivation of GABA-AT by 66-T (i.e., [5-3H]-
10), dialysis, and gel filtration, 2.2 ( 0.4 equiv of tritium
was found to remain associated with the protein. After
urea denaturation of the inactivated enzyme, 1.06 (
0.08 equiv remained bound per dimer. This suggests
that, at least, part of the inactivation occurs by protein
modification, which is different from what was observed
with 3a . Possibly, the structural constraints now permit
the Michael addition pathway to be a viable option as
well as the enamine pathway, which is the case with
vigabatrin (2).
Gen er a l P r oced u r e for th e 4-Meth oxyben zyl Dep r o-
tection by Cer iu m (IV) Am m on iu m Nitr a te (CAN). A
solution of cerium(IV) ammonium nitrate (CAN) (3.0 equiv)
in water (5 mL) was added to a stirred solution of the
4-methoxybenzyl protected lactam (1.0 equiv) in acetonitrile
(20 mL). After being stirred for 5 h, the reaction mixture was
diluted with ethyl acetate (200 mL). The combined organic
solvents were washed with water (4 × 5 mL) and brine (25
mL) and dried over anhydrous Na2SO4. The solvent was
concentrated under reduced pressure to afford crude depro-
tected lactam. The resultant lactam was normally not purified
and used directly for the next step.
All of the compounds are competitive reversible
inhibitors of GABA-AT (Table 1), ranging from a Ki
value of 12 µM for (+)-7 to 13 mM for (()-13. It appears
that bulkiness at C-2 is not very well tolerated.
In conclusion, unlike the five-membered conforma-
tionally rigid analogue of vigabatrin (6), which does not
inactivate GABA-AT, five-membered conformationally
rigid analogues of 4-amino-5-halopentanoic acids exhibit
inactivation properties similar to, although not as potent
as, the corresponding open-chain analogues. Because
the inactivation mechanisms require strict orbital over-
lap requirements, constraint of these orbitals can have
a detrimental effect on inactivation rate constants; in
some cases, the constraint prohibits inactivation. From
the studies with conformationally rigid analogues of
vigabatrin and 4-amino-5-halopentanoic acids, it ap-
pears that conformational constraint affects inactivation
by a Michael addition mechanism more so than inacti-
vation by an enamine pathway.
Gen er a l P r oced u r e for th e Acid ic Hyd r olysis of th e
La cta m to th e Cor r esp on d in g Am in o Acid . A lactam was
added to a stirred solution of acetic acid (10 mL) and 4 N HCl
(10 mL). The resulting solution was stirred at 70 °C for 5 h
and concentrated in vacuo to afford a solid mixture. The solid
was purified by ion-exchange chromatography (AG 50W-X8),
eluting with either 1 M aqueous pyridine or a gradient from
0.4 N to 2.0 N HCl, giving the final amino acid hydrochloride
product.
(1R,4R,6S,7R)-6-Acetoxy-7-br om o-2-ben zyl-2-azabicyclo-
[2.2.1]h ep ta n -3-on e (18). 1,3-Dibromo-5,5-dimethylhydantoin
(1.10 g, 3.78 mmol) was added to a stirred solution of (1R,4S)
2-benzyl-2-azabicyclo[2.2.0]heptan-3-one14 (17; 1.37 g, 6.9 mmol)
in acetic acid (15 mL). After being stirred overnight, the
resulting solution was diluted with dichloromethane (200 mL).
The organic layer was washed with water (3 × 20 mL), 10%
NaHSO3 (30 mL), saturated NaHCO3 (3 × 20 mL), and brine
(30 mL). The organic layer was dried over anhydrous Na2SO4
and concentrated under reduced pressure. The residue was
purified by flash column chromatography on silica gel, eluting
with ethyl acetate/hexane (1:3) to afford the desired product
(18; 1.91 g, 85%) as a clear oil: 1H NMR (300 MHz, CDCl3) δ
7.28-7.41 (5H, m, C6H5), 4.79 (1H, m, H6), 4.77 (1H, d, J 14.7
Hz, CH2Ph), 4.20 (1H, m, H7), 4.01 (1H, d, J 14.7 Hz, CH2Ph),
3.87 (1H, m, H1), 2.96 (1H, m, H4), 2.37 (2H, m, H5), 2.07 (3H,
s, CH3CO) ppm; 13C NMR (300 MHz, CDCl3) δ 173.5, 171.2,
136.1, 129.5, 128.9, 128.7, 73.1, 64.5, 51.0, 48.8, 45.1, 30.5, 21.5;
m/z (EI) 339, 337, 216, 186, 91. HRMS (EI) C15H1679BrO3N
calcd M 337.0314, found M 337.0310; C15H1681BrO3N calcd M
339.0294, found M 339.0305.
Exp er im en ta l Section
Gen er a l Meth od s. Optical spectra and GABA-AT assays
were recorded on a Perkin-Elmer Lambda 10 UV/vis spectro-
photometer. 1H NMR spectra were recorded on a Varian
Gemini 300 MHz NMR spectrometer. Chemical shifts are
reported as δ values in parts per million downfield from Me4-
Si (δ 0.0) as the internal standard in CDCl3. For samples run
in D2O, the HOD resonance was arbitrarily set at 4.60 ppm.
CClF3 was selected as an external standard with δ 0.0 ppm
for 19F NMR. Melting points were determined on a Fisher-
J ohns melting point apparatus and are uncorrected. Combus-
tion analyses were performed by Oneida Research Laborato-
ries, NY. Radioactivity was measured by liquid scintillation
counting using a Packard Tri-Carb 2100TR counter and
Packard Ultima Gold scintillation cocktail. HPLC analyses and
preparations were performed using a Beckman System Gold
system with a 125 solvent delivery module and a 166 UV
detector. High-resolution mass spectra and accurate mass
(1R ,4R ,6S ,7R )-7-B r o m o -6-h y d r o x y -2-b e n zy l-2-a za -
bicyclo[2.2.1]h ep ta n -3-on e (19). K2CO3 (0.46 g, 3.4 mmol)
was added to a stirred solution of (1R,4R,6S,7R)-18 (0.27 g,
0.84 mmol) in methanol (5 mL) and water (1 mL). The
resulting mixture was stirred for 2 h. After methanol was
removed under reduced pressure, the resulting aqueous layer
was extracted with CHCl3 (60 mL). The organic layer was
washed with brine (15 mL), dried over anhydrous Na2SO4, and
evaporated under reduced pressure to afford 19 (0.22 g, 99%)
as a solid: 1H NMR (300 MHz, CDCl3) δ 7.30 (5H, m, C6H5),
4.68 (1H, d, J 14.7 Hz, CH2Ph), 4.27 (1H, m, H6), 4.20 (1H, m,
H7), 4,01 (1H, d, J 14.7 Hz, CH2Ph), 3.96 (1H, m, H7), 3.87
(1H, m, H1), 2.96 (1H, m, H4), 2.37 (2H, m, H5) ppm; 13C NMR
(300 MHz, CDCl3) 172.8, 136.1, 129.3, 128.4, 73.1, 65.9, 50.8,
50.4, 44.7, 33.8 ppm; m/z (EI) 295, 297, 216, 186, 173, 91.
HRMS (EI) C13H1479BrO2N calcd M 295.0208, found M 295.0209;
C13H1481BrO2N calcd M 297.0189, found M 297.0231.
spectra were obtained on
a VG70-250SE high-resolution
spectrometer and a Micromass Quatro II LC/MS spectrometer.
An Orion Research model 702A pH meter with a general
combination electrode was used for pH measurements. Fluo-
ride ion concentration measurements were obtained using an
Orion Research model 702A pH meter with an Orion Research
model 96-09 combination fluoride electrode. Flash column
chromatography was carried out with Merck silica gel 60 (230-
400 mesh ASTM). TLC was run with EM Science silica gel 60
F254 precoated glass plates.
R ea gen t s. All reagents were purchased from Aldrich
Chemical Co. and were used without further purification
except anhydrous ether and tetrahydrofuran (THF), which
were distilled from sodium metal under nitrogen and anhy-
drous dichloromethane which was distilled from calcium

714 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Qiu and Silverman
(1S,4R,6S,7R)-2-Ben zyl-7-br om o-6-flu or o-2-a za bicyclo-
[2.2.1]h ep ta n -3-on e (20). Following the general fluorination
procedure described above, (1R,4R,6S,7R)-19 (0.15 g, 0.51
mmol) was treated with DAST (0.11 g, 0.66 mmol) to afford
the fluorobromo product 20 (0.15 g, 88%) as a solid: 1H NMR
(300 MHz, CDCl3) δ 7.30 (5H, m, C6H5), 4.79 (1/2H, m, H6),
4.61 (1/2H, m, H6), 4.63 (1H, d, J 14.8 Hz, CH2Ph), 4.27 (1H,
m, H7), 4.10 (1H, d, J 14.8 Hz, CH2Ph), 3.94 (1H, m, H1), 3.87
The organic solution was washed with brine (25 mL), dried
over anhydrous MgSO4, and concentrated under reduced
pressure. The residue was purified by flash column chroma-
tography on silica gel to afford 25 (0.3 g, 85%) as a solid: 1H
NMR (300 MHz, CDCl3) δ 7.15 (2H, m, ArH), 6.86 (2H, m,
ArH), 4.97 (1H, d, J 14.7 Hz, ArCH2), 3.87 (1H, d, J 15.3 Hz,
ArCH2), 3.78 (3H, s, OCH3), 3.59 (1H, m, H1), 2.83 (1H, m,
H4), 2.40-1.80 (4H, m, H5 and H7) ppm; 13C NMR (300 MHz,
CDCl3) δ 171.1, 154.5, 128.8, 125.3, 124.8, 123.5, 121.9, 100.5,
56.9 (quartet), 50.5, 40.6 (d), 38.9 (t), 34.1 (t), 32.0 (t) ppm;
19F NMR -92.2 (1/2F, m), -93.0 (1/2F, m), -111.5 (1/2F, m),
-112.3 (1/2F, m) ppm; m/z (EI) 229, 187, 135, 91, 77, 45. HRMS
(EI) calcd for C14H15F2NO3 M 267.1071, found M 267.1076.
(1H, m, H1), 2.99 (1H, m, H4), 2.30-2.60 (2H, m, H5) ppm; 19
F
NMR (300 MHz, CDCl3) -173 (1F, ddd, J 57, 29, 12 Hz) ppm;
m/z (EI) 296, 298, 218, 172, 91. HRMS (EI) C13H1379BrONF
calcd M 297.0164, found M 297.0172; C13H1481BrONF M calcd
299.0145, found M 299.0142.
(1S,4R,6S)-2-Ben zyl-6-flu or o-2-azabicyclo[2.2.1]h eptan -
3-on e (21). Tributyltin hydride (1.35 g, 4.6 mmol) and 2,2′-
azobisisobutyronitrile (AIBN) (20 mg) were added to a solution
of (1S,4R,6S,7R)-20 (1.06 g, 3.87 mmol) in dry benzene (25
mL). The resultant solution was heated at reflux and stirred
for 12 h. After the reaction was concentrated under reduced
pressure, the residue was purified by flash column chroma-
tography on silica gel, eluting with ethyl acetate/hexane (1:
3), to afford the desired debrominated product 21 (0.71 g, 94%)
as a colorless oil: 1H NMR (300 MHz, CDCl3) δ 7.30 (5H, m,
C6H5), 4.74 (1/2H, m, H6), 4.57 (1/2H, m, H6), 4.52 (1H, d, J
15.1 Hz, CH2Ph), 4.14 (1H, d, J 15.1 Hz, CH2Ph), 3.80 (1H, m,
6,6-Diflu or o-2-a za bicyclo[2.2.1]h ep ta n -3-on e (26). Po-
tassium persulfate (3.5 g, 12.9 mmol) and potassium hydrogen
phosphate trihydrate (1.7 g, 7.4 mmol) were added to a stirred
solution of 25 (0.6 g, 2.2 mmol) in CH3CN/H2O (40 mL, 1:1).
The resultant suspended solution was heated and stirred at
82 °C for 1.5 h. After removal of CH3CN under reduced
pressure, the aqueous residue was extracted with ethyl acetate
(3 × 30 mL). The combined organic layers were washed with
brine (25 mL), dried over anhydrous Na2SO4, and concentrated
under reduced pressure. The residue was purified by flash
column chromatography on silica gel to afford 26 (0.21 g, 64%)
as a solid: 1H NMR (300 MHz, CDCl3) δ 3.54 (1H, m, H1), 2.83
(1H, m, H4), 2.40-1.80 (4H, m, H5 and H7) ppm; 19F NMR
-95.3 (1/2F, m), -96.1 (1/2F, m), -111.8 (1/2F, m), -112.6
(1/2F, m) ppm; m/z (EI) 147, 128, 84. HRMS (EI) calcd for
C6H7F2NO M 147.1256, found M 147.1259.
H1), 2.83 (1H, m, H4), 2.17-1.70 (4H, m, H7 and H5) ppm; 19
F
NMR (300 MHz, CDCl3) -173 (1F, ddd, J 50, 32, 18 Hz) ppm;
m/z (EI) 235, 219, 173, 91, 65. HRMS (EI) C13H14ONF calcd
M 235.1008, found M 235.1008.
(1R,3S,4S)-3-Am in o-4-flu or ocyclop en ta n e-1-ca r boxyl-
ic Acid ((+)-7). Freshly cut pieces of sodium metal (0.45 g)
were added to a stirred solution of liquid NH3 (8 mL) and tert-
butyl alcohol (2 mL) at -78 °C to afford a deep blue solution.
Then a solution of (1S,4R,6S)-21 (0.71 g, 3.0 mmol) in THF (8
mL) was added portionwise to the stirred sodium-liquid
ammonium solution at -78 °C. After the resulting solution
was stirred for 10 min at -78 °C, the temperature was allowed
to rise to -30 °C and stirring continued for 4 min. The solution
was cooled back to -78 °C, and acetic acid (2 mL) was added
slowly. After the solution was allowed to warm to room
temperature, the resultant slurry was filtered and washed
with ethyl acetate (100 mL). The organic solution was con-
centrated under reduced pressure to afford a crude solid
residue (22, 0.80 g). Without any purification, the residue 22
was hydrolyzed following the general acidic hydrolysis proce-
dure, with a basic elution, to give the product (7, 0.46 g, 86%)
as a colorless solid: mp 225 °C (decomp.); [R]23.5 ) + 8.85° (c,
1.92, H2O); 1H NMR (300 MHz, D2O) 5.25 (0.5H, m, H4), 5.08
(0.5H, m, H4), 3.76 (1H, dtd, J 21.7, 7.8, 3.6 Hz, H3), 2.94 (1H,
quintet, J , 8.0 Hz, H1), 1.69-2.45 (4H, m, H2 and H5) ppm;
19F NMR (300 MHz, D2O) -173.5 (sextet, J 54, 27 Hz) ppm;
m/z (EI) 148, 127, 102, 83, 56. HRMS (EI) calcd for C6H11O2-
NF M 148.0774, found M 148.0775.
6-Oxo-2-(4′-m eth oxyben zyl)-2-a za bicyclo[2.2.1]h ep ta n -
3-on e (24). 4-Methylmorpholine N-oxide (0.77 g, 6.6 mmol),
tetrapropylammonium perruthenate (TPAP) (5 mg), and 4 Å
sieves were added to a stirred solution of (()-2314 (0.67 g, 2.7
mmol) in anhydrous CH2Cl2 (10 mL). After being stirred for
14 h, the mixture was concentrated under reduced pressure.
The resultant slurry was loaded onto a flash silica gel column
directly, eluting with ethyl acetate/hexane (2:3), to afford 24
(0.5 g, 75%) as a colorless solid; 1H NMR (300 MHz, CDCl3) δ
7.20 (2H, m, ArH), 6.85 (2H, m, ArH), 4.68 (1H, d, J 14.7 Hz,
ArCH2), 3.86 (1H, d, J 14.7 Hz, ArCH2), 3.78 (3H, s, OCH3),
3.50 (1H, m, H1), 3.01 (1H, m, H4), 2.33-2.15 (4H, m, H5 and
H7) ppm; m/z (EI) 246, 230, 184, 104, 91. HRMS (EI) calcd for
3â-Am in o-4,4-d iflu or ocyclop en ta n e-1â-ca r boxylic Acid
(8). Following the general acid hydrolysis procedure, compound
(()-26 (0.2 g, 1.35 mmol) was treated with HCl/acetic acid,
followed by ion-exchange chromatography with basic elution,
to afford 8 (0.20 g, 86%) as a colorless solid: mp 207-209 °C;
1H NMR (300 MHz, CDCl3) δ 3.99 (1H, m, H3), 3.20 (1H, m,
H1), 2.40-2.70 (3H, m, H2 and H5), 2.06 (1H, m, H2) ppm; 19
F
NMR -97.0 (1/2F, m), -97.8 (1/2F, m), -102.15 (1/2F, m),
-103.0 (1/2F, m) ppm; m/z (EI) 145, 120, 100, 83, 56, 36. HRMS
(EI) calcd for (C6H10O2NF2 + H+) (M + H) 166.0691, found M
+ H 166.0687.
(1S,4R,6S)-6-Acet oxy-2-(ter t-b u t oxyca r b on yl)-2-a za -
bicyclo[2.2.1]h ep ta n -3-on e (29). Di-tert-butyl dicarbonate
(1.79 g, 8.0 mmol), DMAP (200 mg), and triethylamine (4 mL)
were added to a stirred suspended solution of (1S,4R,6S)-6-
acetoxy-2-azabicyclo[2.2.1]heptan-3-one 2814 (0.62 g, 4.0 mmol)
in dichloromethane (15 mL). After being stirred for 5 h, the
resultant mixture was diluted with ethyl acetate (100 mL) and
washed with 5% HCl (25 mL) and brine (25 mL). The organic
layer was dried over MgSO4 and concentrated under reduced
pressure. The residue was purified by flash column chroma-
tography on silica gel, eluting with ethyl acetate/hexane (1/
3), to yield 29 (0.74 g, 71%) as a colorless solid: 1H NMR
(CDCl3) δ 4.92 (1H, m, H6), 4.43 (1H, m, H1), 2.77 (1H, m, H4),
2.30-1.60 (4H, m, H4 and H7), 2.02 (3H, s, CH3CO2), 1.47 (9H,
s, -O-t-C(CH3)3) ppm; m/z (EI) 269, 227, 196, 171, 109, 83, 58.
HRMS (EI) calcd for C13H19NO5 M 269.1263, found M 269.1261.
Meth yl (1R,3S,4S)-3-ter t-bu toxyca r bon yla m id o-4-h y-
d r oxycyclop en ta n e-1-ca r boxyla te (31). K2CO3 (1.18 g, 8.4
mmol) was added to a stirred solution of t-Boc protected lactam
29 (0.72 g, 2.8 mmol) in methanol (15 mL) and water (5 mL).
After being stirred for 4 h, the reaction mixture was concen-
trated under reduced pressure to remove the methanol. The
resultant aqueous solution was carefully acidified with 1 N
HCl to a pH of 2.0, monitoring with a pH meter. Then, the
acidified aqueous solution was extracted with ethyl acetate (4
× 50 mL). The organic solution was washed with brine (25
mL), dried over anhydrous Na2SO4, and concentrated under
reduced pressure. The resultant residue 30 (0.64 g, 94%) was
dissolved in methanol (10 mL) and ether (10 mL). Excess of
freshly made diazomethane in ether was added portionwise
to the stirred methanol-ether solution until the yellow color
did not disappear. After being stirred for another 30 min, the
solution was concentrated under reduced pressure to afford
31 (0.64 g, 95%) as a colorless solid.
C
14H15NO3 M 245.1052, found M 245.1058.
2-(4′-Meth oxyben zyl)-6,6-d iflu or o-2-a za bicyclo[2.2.1]-
h ep ta n -3-on e (25). DAST (0.72 g, 4.4 mmol) was added to a
stirred solution of 24 (0.32 g, 1.30 mmol) in anhydrous benzene
under N2. The solution was stirred for 4 days until no starting
material was observed by TLC. The resultant solution was
poured into a saturated NaHCO3 (30 mL) solution, and the
aqueous solution was extracted with ethyl acetate (3 × 30 mL).

Inactivators of γ-Aminobutyric Acid Aminotransferase
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 715
For (1R,3S,4S)-30: 1H NMR (300 MHz, DMSO) δ 6.72 (1H,
br, N-H), 4.78 (d, J 4.2 Hz, OH), 3.77 (1H, m, H4), 3.51 (1H,
m, H3), 2.78 (1H, m, H1), 1.4-2.2 (4H, m, H2 and H5), 1.35
(9H, s, O-t-Bu) ppm; m/ z (EI) 171 (M-OC4H9-H), 154, 126,
102, 83, 58. HRMS (ES), calcd for C11H19NO5H M + 1 246.1341,
found 246.1330.
For methyl (1R,3S,4S)-31: 1H NMR (400 MHz, CDCl3) δ
5.12 (1H, br, N-H), 4.09 (1H, m, H4), 3.75 (1H, m, H3), 3.66
(3H, s, OCH3), 3.02 (1H, m, H1), 1.6-2.4 (4H, m, H2 and H5),
1.44 (9H, s, O-t-Bu) ppm; m/z (EI) 185 (M-OC4H9-H), 140,
116, 83, 58. HRMS (ES) calcd for C12H21NO5H M + H 260.1497,
found M + H 260.1479.
as a colorless solid, which was used directly in the next step:
1H NMR (300 MHz, CDCl3) δ 7.19 (2H, m, ArH), 6.91 (2H, m,
ArH), 4.60 (1H, d, J 14.6 Hz, ArCH2), 4.24 (1H, m, H7), 4.00
(1H, d, J 14.6 Hz, ArCH2), 3.91 (1H, m, H6), 3.84 (3H, s,
-OCH3), 3.79 (1H, m, H1), 2.98 (1H, m, H4), 2.52-2.71 (2H,
m, H5) ppm; m/z (EI) 388 (M + H), 308, 198, 162, 121, 77, 65.
HRMS (EI) calcd for C14H15NO279Br2 calcd M 386.9470, found
M 386.9468; C14H15NO279Br81Br calcd M 388.9450, found M
388.9443; C14H15NO281Br2 calcd M 390.9420, found M 390.9418.
7-a n ti-Br om o-2-(4′-m eth oxyben zyl)-2-a za bicyclo[2.2.1]-
h ep t-5-en -3-on e (36). Dibromolactam 35 (0.80 g, 2.1 mmol)
was mixed with neat 1,5-diazobicyclo[4.3.0]non-5-ene (DBN)
(2 g), stirred, and heated to 120 °C for 5 h. The resultant dark
oil was diluted with dichloromethane (100 mL), then the
organic layer was washed with 1 N HCl (2 × 25 mL) and brine
(25 mL) and dried over anhydrous Na2SO4. The organic layer
was concentrated under reduced pressure. The residue was
purified by flash column chromatography on silica gel to afford
36 (0.48 g, 89%) as a colorless solid; 1H NMR (300 MHz, CDCl3)
δ 7.06 (2H, m, ArH), 6.86 (2H, m, ArH), 6.43 (2H, m, H5 and
H6), 4.63 (1H, m, H7), 4.29 (1H, d, J 14.4 Hz, ArCH2), 4.12
(1H, d, J 14.6 Hz, ArCH2), 4.08 (1H, m, H1), 3.80 (3H, s,
-OCH3); 3.49 (1H, m, H4) ppm; m/z (EI) 307, 228, 163, 121,
76, 65. HRMS (EI) calcd for C14H1479BrNO2 M 307.0208, found
M 307.0204; calcd for C14H1481BrNO2 M 390.0201, found M
390.0186.
Meth yl (1R,3S,4S)-4-Br om o-3-ter t-bu toxyca r bon yla m i-
d ocyclop en ta n e-1-ca r boxyla te (32). Carbon tetrabromide
(0.67 g, 2.0 mmol), triphenylphosphine (0.53 g, 2.0 mmol), and
tetra-n-butylammonium bromide (20 mg) were added to a
stirred solution of methyl (1R,3S,4S)-31 (0.27 g, 1.0 mmol) in
anhydrous THF (15 mL). After being stirred for 16 h, the
reaction mixture was concentrated under reduced pressure.
The residue was purified by flash column chromatography on
silica gel to afford 32 (0.28 g, 85%) as a colorless solid: 1H
NMR (400 MHz, CDCl3) δ 4.93 (1H, d, br, J 8.0 Hz, N-H),
4.51 (1H, m, H4), 3.97 (1H, m, H3), 3.71 (3H, s, OCH3), 2.88
(1H, m, H1), 2.62 (2H, m, H2), 2.05-2.30 (2H, m, H5), 1.44 (9H,
s, O-t-Bu) ppm; 13C NMR (400 MHz, CDCl3) 175.99, 156.20,
1.20.43, 80.93, 58.14, 56.01, 53.30, 39.88, 38.10, 32.97, 29.39
ppm; m/z (EI) 321, 248, 216, 190, 185, 142, 126, 82, 58. HRMS
(EI) calcd for C12H2079BrNO4 M 321.0576, found M 321.0571,
calcd for C12H2081BrNO4 M 323.0555, found M 323.0559.
(1R,3S,4S)-3-Am in o-4-br om ocyclop en ta n e-1-ca r boxy-
lic Acid ((-)-9). Methyl (1R,3S,4S)-32 (0.24 g, 0.72 mmol) was
hydrolyzed according to the general acid hydrolysis procedure,
and the product was purified by recrystallization from ethanol/
ether, to give (-)-9 (0.15 g, 74%) as a colorless solid: mp 172-
7-a n ti-Br om o-2-(4′-m eth oxyben zyl)-2-a za bicyclo[2.2.1]-
h ep ta n -3-on e (37). Palladium on charcoal (10%, 20 mg) was
added to a solution of 36 (0.20 g) in ethyl acetate (8 mL). The
resultant mixture was stirred under a hydrogen atmosphere
for 2 h. After careful filtration of the catalyst through a bed of
Celite 450 and evaporation of the solvent under reduced
pressure, 37 (0.20 g, 98%) was obtained as a colorless solid:
1H NMR (300 MHz, CDCl3) δ 7.15 (2H, m, ArH), 6.86 (2H, m,
ArH), 6.43 (2H, m, H5 and H6), 4.75 (1H, d, J 14.4 Hz, ArCH2),
4.63 (1H, m, H7), 4.01 (1H, d, J 14.6 Hz, ArCH2), 3.80 (3H, s,
-OCH3), 3.65 (1H, m, H1), 2.89 (1H, m, H4), 2.20-2.10 (2H,
m, H5 or H6), 1.4-1.75 (2H, m, H5 or H6) ppm; m/z (EI) 309,
1
174 °C; [R]23.5 ) - 12.4 ° (c, 0.88, H2O); H NMR (300 MHz,
DMSO) δ 4.62 (1H, m, H4), 3.55 (1H, m, H3), 2.96 (1H, m, H3),
2.68 (1H, m, H1), 2.43 (2H, m, H2), 2.20-2.30 (2H, m, H5)
ppm; m/z (EI) 164, 162, 121, 101, 82, 56. HRMS (ES) calcd for
C6H979BrNO2H M + H 207.9974, found M + H 208.0005; calcd
for C6H981BrNO2H M + H 209.9948, found M + H 209.9939.
(1R,2S,3R,4S)-3-Am in o-2,4-dibr om ocyclopen tan e-1-car -
boxylic Acid ((+)-10). Bromine (0.80 g, 5.0 mmol) in CH2Cl2
(2 mL) was added to a stirred solution of (1R)-(-)-azabicyclo-
[2.2.1]hept-5-en-3-one (16, 0.50 g, 4.58 mmol) in CH2Cl2 (10
mL) at 0 °C. After being stirred for 2 h, the reaction mixture
was concentrated under reduced pressure to (1R,4R,6S,7R)-
6,7-dibromo-2-azabicyclo[2.2.1]heptan-3-one (33; 1.20 g, 98%).
Without any purification, the lactam 33 was hydrolyzed
following the general acid hydrolysis procedure, followed by
ion-exchange chromatography with a basic elution, giving a
light yellow solid. The solid was dissolved in warm water and
briefly treated with charcoal. After filtration of the charcoal
and lyophilization, the product was obtained as a colorless solid
((+)-10, 0.38 g, 29%).
229, 164, 122, 78, 65. HRMS (EI) calcd for C14H1679BrNO2
309.0365, found M 309.0370, calcd for C14H1681BrNO2
311.0344, found M 311.0350.
M
M
3-r-Am in o-2â-br om ocyclop en ta n e-1r-ca r boxylic Acid
(11). Following the general deprotection procedure with CAN,
compound (()-37 (0.14 g, 0.45 mmol) was treated with CAN
(0.74 g, 1.35 mmol) to afford crude 38 (0.14 g). Without further
purification, 38 was hydrolyzed following the general acid
hydrolysis procedure and purified by ion-exchange chroma-
tography with acid elution, to yield (()-11 (0.034 g, 34%) as a
light-yellow solid.
For 38: 1H NMR (300 MHz, CDCl3) δ 4.24 (1H, m, H7), 3.80
(3H, s, -OCH3), 3.93 (1H, m, H1), 2.87 (1H, m, H4), 2.23-2.40
(2H, m, H5 or H6), 1.20-1.60 (2H, m, H5 or H6) ppm; m/z (EI)
189, 135, 110, 82, 69, 45. HRMS (EI) calcd for C6H879BrNO M
188.9790, found M 188.9798; calcd for C6H881BrNO M 190.9769,
found M 190.9768.
For (()-11: mp 130 °C (decomp); 1H NMR (300 MHz, D2O)
δ 4.19 (1H, t, J 9.0 Hz, H2), 3.73 (1H, dd, J 9.0, 8.7, H3), 3.02
(1H, dd, J 9.3, 8.7 Hz, H1), 2.02-2.30 (2H, m, H4 or H5),
1.60-1.90 (2H, m, H4 or H5) ppm. HRMS calcd for
C6H1079BrNO2H M + H 207.9938, found M + H 207.9974; calcd
for C6H1081BrNO2H M + H 209.9953, found M + H 210.0036.
(1S,4R,6S,7R)-7-Br om o-6-flu or o-2-a za b icyclo[2.2.1]-
h ep ta n -3-on e (39). Compound 39-P MB14 (0.68 g, 2.08 mmol)
was treated with CAN (3.4 g, 6.24 mmol), following the general
CAN deprotection procedure, to afford crude 39. Further
purification by flash column chromatography on silica gel,
eluting with ethyl acetate/hexane (1:1), gave 39 (0.36 g, 84%)
as a colorless solid: mp 160 °C (decomp.); 1H NMR (300 MHz,
CDCl3) δ 6.21 (1H, s, NH), 4.97 (1/2H, d, J 6 Hz, H6), 4.79
(1/2H, d, J 6 Hz, H6), 4.12 (1H, s, H1), 2.86 (1H, s, H4), 2.24-
2.55 (2H, m, H5 and H6) ppm; 19F NMR (300 MHz, CDCl3)
-171.4 (ddd, J 56.7, 29.4, 12.0 Hz) ppm; m/z (EI) 207, 127,
108, 82, 59. HRMS (EI) calcd for C6H779BrFNO M 206.9696,
For the dibromolactam 33: 1H NMR (300 MHz, D2O) δ 8.48
(1H, s, NH), 4.57 (1H, d, J 1.38 Hz, H6), 4.08 (2H, m, H1 and
H7), 2.76 (1H, m, H4), 2.41 (2H, m, H5) ppm.
For (+)-10: mp 134 °C (decomp.); [R] 23.5 ) + 48.6° (c, 1.36,
H2O); 1H NMR (300 Hz, D2O) δ 4.27 (2H, m, H2 and H4), 4.10
(1H, m, H3), 3.27 (1H, quart. J 9.0 Hz, H1), 2.55 (2H, m, H5)
ppm; m/z (ES) M+ 286.0 (50%), 288.0 (100%), 290 (50%). HRMS
calcd for C6H9NO279Br2H M + H 285.9079, found M + H
285.9048; calcd for C6H9NO279Br81BrH M + H 287.9054, found
M + H 287.9109; calcd for C6H9NO281Br2H M + H 289.9038,
found M + H 289.9097.
6-exo-7-a n t i -D ib r o m o -2-(4′-m e t h o x y b e n zy l)-2-a za -
bicyclo[2.2.1]h ep ta n -3-on e (35). Bromine (0.45 g, 2.8 mmol)
was added to a stirred solution of 2-(4′-methoxybenzyl)-2-
azabicyclo[2.2.1]hept-5-en-3-one ((()-34)14 (0.62 g, 2.5 mmol)
in dichloromethane (10 mL) at 0 °C. After being stirred for 1
h, the reaction mixture was diluted with dichloromethane (50
mL) and washed with saturated Na2SO3 solution (20 mL) and
brine (20 mL). The organic solution was dried over MgSO4 and
concentrated under reduced pressure to give 35 (0.95 g, 95%)

716 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Qiu and Silverman
found M 206. 9701; calcd for C6H781BrFNO M 208.9675, found
M 208. 9681.
(1.28 g, 92%) as a light brown oil: 1H NMR (300 MHz, CDCl3)
δ 7.16 (2H, m, ArH), 6.84 (2H, m, ArH), 4.42 (1H, d, J 14.6
Hz, ArCH2), 4.20 (1H, d, J 14.6 Hz, ArCH2), 3.77 (3H, s, OCH3),
3.74 (1H, m, H1), 3.50 (1H, m, H6), 3.20 (1H, m, H5), 2.96 (1H,
m, H4), 1.53-1.77 (2H, m, H7) ppm; m/z (EI) 245, 217, 163,
121, 77, 66. HRMS (EI) calcd for C14H15NO3 M 245.1052, found
M 245.1052.
Mixtu r e of(1S,4S,6S,7R)-6-Br om o-7-h ydr oxy-2-(4′eth oxy-
ben zyl)-2-a za bicyclo[2.2.1]h ep ta n -3-on e (45) a n d (1R,-
4R,5S,6S)-5-Br om o-6-h yd r oxy-2-(4′-m et h oxyb en zyl)-2-
a za bicyclo[2.2.1]h ep ta n -3-on e (46). HBr (48%, 1.5 mL) was
added to a stirred solution of 44 (1.2 g, 3.67 mmol) in
acetonitrile (30 mL) at 0 °C. After being stirred for 6 h, the
resultant reaction mixture was diluted with ethyl acetate (120
mL) and washed with saturated sodium bicarbonate (30 mL)
and brine (25 mL). The organic layer was dried over anhydrous
Na2SO4 and concentrated under reduced pressure. The result-
ant residue was recrystallized from hexane and ethyl acetate
to afford a mixture (1.2 g, 75%) of 45 and 46 as a colorless
solid.
(1S ,4S ,6S ,7R )-6-Br om o-2-(4′-m e t h oxyb e n zyl)-7-t r i-
m eth ylsiloxy-2-a za bicyclo[2.2.1]h ep ta n -3-on e (47) a n d
(1R,4R,5S,6S)-5-Br om o-2-(4′-m eth oxyben zyl)-6-tr im eth -
ylsiloxy-2-a za bicyclo[2.2.1]h ep ta n -3-on e (48). 2,6-Lutidine
(1.10 mL, 14.6 mmol), trimethylsilyl triflate (2.0 mL, 14.6
mmol) and DMAP (0.2 g) were added to a stirred solution of
the mixture of 45 and 46 (1.8 g, 5.5 mmol) in anhydrous CH2-
Cl2 (20 mL). After being stirred overnight, the resultant
reaction mixture was diluted with CH2Cl2 (100 mL) and briefly
washed with a 5% HCl solution (2 × 20 mL) and brine (25
mL). The resultant mixture was stirred and heated at reflux
overnight. After being cooled, the dark mixture was concen-
trated under reduced pressure. The resultant residue was
purified by flash column chromatography on silica gel, eluting
with ethyl acetate/hexane (1/4), giving 47 (1.2 g, 55%) and 48
(0.4 g, 18%), both as colorless solids.
For 47: 1H NMR (300 MHz, CDCl3) δ 7.19 (2H, m, ArH),
6.90 (2H, m, ArH), 4.52 (1H, d, J 14.6 Hz, ArCH2), 4.10 (1H,
d, J 14.6 Hz, ArCH2), 4.07 (1H, m, H7), 3.83 (3H, s, OCH3),
3.71 (1H, m, H6), 3.58 (1H, m, H1), 2.71 (1H, m, H4), 2.39-
2.46 (2H, m, H5), 0.141 (9H, s, Si(CH3)3) ppm; m/z (EI) 397,
320, 227, 162, 121, 73. HRMS (EI) calcd for C17H2479BrNO3Si
M 397.0708, found M 397.0710; calcd for C17H2481BrNO3Si M
399.0689, found M 399.0688.
(1S,2R,3R,4S)-3-Am in o-2-br om o-4-flu or ocyclop en ta n e-
1-ca r boxylic Acid ((+)12). Lactam 39 (0.36 g, 1.4 mmol) was
hydrolyzed to afford crude 12 by the general acid hydrolysis
procedure. Further recrystallization from ethanol/ether gave
pure (+)-12 (0.22 g, 48%) as the HCl salt: mp 160 °C (decomp.);
[R]23.5 ) + 36.5° (c, 1.70, H2O); 1H NMR(300 MHz, D2O) δ 5.30
(1/2H, m, H4), 5.17 (1/2H, m, H4), 4.32 (1H, t, J 10 Hz, H2),
4.13 (1H, ddd, J 23.1, 10, 4.4 Hz, H3), 3.54 (1H, q, J 10.4 Hz,
H1), 2.30-2.55 (2H, m, H5) ppm; 13C NMR (D2O) 175.83, 94.53,
92.70, 65.31, 65.03, 50.38, 45.78, 45.73, 34.36 ppm; 19F NMR
-175. 5 (1F, m) ppm; m/z (EI) 182 (M-HCO2), 180 (M-HCO2),
146, 126, 100, 74. HRMS (ES) calcd for C6H1079BrFNO2H M +
H 225.9879, found M + H 225.9894; calcd for C6H1081BrFNO2H
M + H 227.9859, found M + H 227.9865.
6-exo-Acetoxy-7-a n ti-br om o-2-a za bicyclo[2.2.1]h ep ta n -
3-on e (41). 6-exo-Acetoxy-7-anti-bromo-2-(4′-methoxybenzyl)-
2-azabicyclo[2.2.1]heptan-3-one (40)14 (1.33 g, 3.6 mmol) was
treated with CAN (5.9 g, 10.8 mmol) to afford crude 41 (general
deprotection by CAN). Purification by flash column chroma-
tography on silica gel, eluting with ethyl acetate/hexane (1:
1), afforded 41 (0.54 g, 61%) as a colorless solid: 1H NMR (300
MHz, CDCl3) δ 6.30 (1H, s, NH), 4.81 (1H, m, H6), 4.30 (1H,
m, H7), 4.06 (1H, m, H1), 2.86 (1H, m, H4), 2.32 (2H, m, H5),
2.08 (3H, s, CH3CO2) ppm. HRMS (ES) calcd for C8H979BrNO3
M 245.9766, found M 245.9781; calcd for C8H981BrNO3 M
247.9657, found M 247.9691.
3r-Am in o-2â-b r om o-4â-h yd r oxycyclop en t a n e-1r-ca r -
boxylic Acid ((()-13). The deprotected lactam (()-41 (0.46
g, 1.86 mmol) was hydrolyzed by the general acid hydrolysis
procedure, then purified by ion-exchange chromatography with
acid elution, to give (()-13 (0.32 g, 78%) as the HCl salt: mp
39-42 °C; 1H NMR (300 MHz, D2O) δ 4.07-4.21 (2H, H2 and
H4), 3.55 (1H, dd, J 7.2, 9.0 Hz, H3), 3.30 (1H, q, J 9.4 Hz, H1),
2.20-1.96 (2H, m, H5) ppm; 13C NMR (D2O) 175.7, 71.2, 65.4,
49.1, 45.1, 34.6 ppm. HRMS (ES) calcd for C6H11O3N79Br (M
+ H) 223.9917, found (M + H) 223.9938; calcd for C6H11O3N81
Br (M + H) 225.9897, found (M + H) 225.9957.
-
6-exo-Br om o-7-a n ti-h ydr oxy-2-azabicyclo[2.2.1]h eptan -
3-on e (43). Following the same procedure for the preparation
of 26 from 25, 6-exo-bromo-2-(4′-methoxybenzyl)-7-anti-trim-
ethylsiloxy-2-azabicyclo[2.2.1]heptan-3-one (42, 0.76 g, 1.15
mmol) was treated with K2S2O8 (1.87 g, 6.93 mmol) and
potassium hydrogen phosphate trihydrate (0.89 g, 3.91 mmol)
to afford 43 (0.21 g, 53%) as a colorless solid: 1H NMR (300
MHz, CDCl3) δ 5.83 (1H, s, NH), 4.32 (1H, m, H7), 4.00 (1H,
m, H6), 3.92 (1H, m, H1), 2.80 (1H, m, H4), 2.50-2.52 (2H, m,
H5) ppm. HRMS (ES) calcd for C6H979BrNO2 (M + H) 205.9811,
found (M + H) 205.9801; calcd for C6H981BrNO2 (M + H)
207.9791, found (M + H) 207.9818.
For 48: 1H NMR (300 MHz, CDCl3) δ 7.25 (2H, m, ArH),
6.91 (2H, m, ArH), 4.53 (1H, d, J 14.6 Hz, ArCH2), 4.25 (1H,
d, J 14.6 Hz, ArCH2), 3.89 (1H, m, H5), 3.82 (3H, s, OCH3),
3.82 (1H, m, H6), 3.36 (1H, m, H1), 2.98 (1H, m, H4), 2.04 (2H,
m, H7), 0.141 (9H, s, Si(CH3)3) ppm; m/z (EI) 397, 320, 246,
180, 121, 78, 66. HRMS (EI) calcd for C17H2479BrNO3Si M
397.0708, found M 397.0712; calcd for C17H2481BrNO3Si M
399.0689, found M 399.0694.
3r-Am in o-4â-b r om o-2â-h yd r oxycyclop en t a n e-1r-ca r -
boxylic Acid ((()-14). Lactam (()-43 (0.2 g, 0.97 mmol) was
hydrolyzed by the general acid hydrolysis procedure and
purified by ion-exchange with acid elution to afford (()-14 (0.16
(1S,4S,6S,7R)-6-Br om o-7-h ydr oxy-2-(4′-m eth oxyben zyl)-
2-a za bicyclo[2.2.1]h ep ta n -3-on e (45). Tetra-n-butylammo-
nium fluoride (1 M) in THF (1.5 mL) was added to a stirred
solution of 47 (0.39 g, 0.98 mmol) in THF (10 mL). After being
stirred for 30 min, the solvent was concentrated under reduced
pressure. The resultant residue was diluted with ethyl acetate
(100 mL), and the solution was washed with 0.5 N HCl (25
mL), saturated KHCO3 (25 mL), and brine (25 mL). The
organic layer was dried over anhydrous Na2SO4 and concen-
trated under reduced pressure. The resultant residue was
purified by recrystallization from hexane/ethyl acetate to give
45 (0.27 g, 84%) as colorless flakes: 1H NMR (300 MHz, CDCl3)
δ 7.18 (2H, m, ArH), 6.90 (2H, m, ArH), 4.54 (1H, d, J 14.6
Hz, ArCH2), 4.19 (1H, m, H7), 4.06 (1H, d, J 14.6 Hz, ArCH2),
3.84 (3H, s, OCH3), 3.76 (1H, m, H6), 3.72 (1H, m, H1), 2.85
(1H, m, H4), 2.48-2.51 (2H, m, H5) ppm; m/z (EI) 325, 327,
1
g, 76%) in the HCl form: mp 48-50 °C; H NMR (300 MHz,
D2O) δ 4.19-4.08 (2H, H2 and H4), 3.59 (1H, t, J 9.0 Hz, H3),
2.97 (1H, m, H1), 2.53 (1H, m, H5a or H5b), 2.35 (1H, m, H5a or
H
5b) ppm; 13C NMR (D2O) 175.7, 71.2, 65.4, 49.1, 45.1, 34.6
ppm. HRMS (ES) calcd for C6H11O3N79Br (M + H) 223.9917,
found (M + H) 223.9938; calcd for C6H11O3N81Br (M + H)
225.9897, found (M + H) 225.9957.
(1R,4S,5R,6S)-5,6-E p oxy-2-(4′-m et h oxyb en zyl)-2-a za -
bicyclo[2.2.1]h ep ta n -3-on e (44). An excess of mCPBA (1.9
g, 57%) was added to a stirred solution of (1R,4S)-2-(4′-
methoxybenzyl)-2-azabicyclo[2.2.1]hept-5-en-3-one (-)-34 (1.30
g, 5.7 mmol) in chloroform (20 mL) and was heated at reflux
for 8 h. After being cooled, the resultant mixture was diluted
with chloroform (80 mL) and washed with saturated Na2SO3
solution (30 mL), saturated Na2CO3 solution (2 × 30 mL), and
brine (25 mL). The organic layer was dried over anhydrous
Na2SO4 and concentrated under reduced pressure. The result-
ant residue was purified by flash column chromatography on
silica gel, eluting with ethyl acetate/hexane (1/2), to afford 44
246, 162, 121, 83. HRMS (EI) calcd for C14H1679BrNO3
325.0314, found M 325.0318; calcd for C14H1681BrNO3
327.0294, found M 327.0288.
M
M
(1S,4S,6S,7S)-6-Br om o-7-flu or o-2-(4′-m eth oxyben zyl)-
2-a za bicyclo[2.2.1]h ep ta n -3-on e (49). Following the general
fluorination procedure, 45 (0.6 g, 1.8 mmol) was treated with

Inactivators of γ-Aminobutyric Acid Aminotransferase
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 717
DAST (0.35 g, 2.2 mmol) affording 49 (0.37, 63%) as a colorless
solid: 1H NMR (300 MHz, CDCl3) δ 7.15 (2H, m, ArH), 6.88
(2H, m, ArH), 4.92 (1/2H, m, H7), 4.74 (1/2H, m, H7), 4.49 (1H,
dd, J 2.4, 14.7 Hz, ArCH2), 4.10 (1H, d, J 14.7 Hz, ArCH2),
3.86 (1H, m, H6), 3.81 (3H, s, OCH3), 3.79 (1H, m, H1), 2.99
(1H, m, H4), 2.37-2.55 (2H, m, H5) ppm; m/z (EI) 327, 248,
aqueous portion was extracted with ethyl acetate (3 × 30 mL).
The combined organic layers were washed with saturated
NaHCO3 and brine, dried over anhydrous MgSO4, and con-
centrated under reduced pressure. The residue was purified
by column chromatography on silica gel, eluting with ethyl
acetate/hexane (2:3), giving 55 (0.94 g, 92%) as a colorless
solid: 1H NMR (300 MHz, CDCl3) δ 7.18 (2H, m, ArH), 6.91
(2H, m, ArH), 4.60 (1H, d, J 14.8 Hz, ArCH2), 4.27 (1H, d, J
1.3 Hz, H7), 3.93 (3H, s, OCH3), 3.91 (1H, d, J 14.6 Hz, ArCH2),
3.73 (1H, m, H1), 2.99 (1H, m, H4), 2.45 (2H, m, H5), 2.23 (2H,
m, H5) ppm; m/z (EI) 327, 325, 201, 199, 121, 77. HRMS (EI)
calcd for C14H1579BrNO3D M 326.0376, found M 326.0368; calcd
for C14H1581BrNO3 M 328.0357, found M 328.0390.
6-exo-7-a n ti-Dibr om o-6-d eu ter iu m -2-(4′-m eth oxyben -
zyl)-2-a za bicyclo[2.2.1]h ep ta n -3-on e (56). Triphenylphos-
phine (1.41 g, 5.4 mmol), carbon tetrabromide (1.80 g, 5.4
mmol), and terta-n-butylammonium bromide (100 mg) were
added to a stirred solution of 55 (0.90 g, 2.7 mmol) in
anhydrous THF (20 mL). The resultant solution was heated
at reflux for 9 h. The organic solvents were concentrated under
reduced pressure, and the residue was purified by column
chromatography on silica gel, eluting with ethyl acetate/hexane
(1:4), yielding 56 (0.95 g, 90%) as a colorless solid: 1H NMR
(300 MHz, CDCl3) δ 7.19 (2H, m, ArH), 6.91 (2H, m, ArH),
4.62 (1H, d, J 14.8 Hz, ArCH2), 4.25 (1H, d, J 1.3 Hz, H7),
4.00 (1H, d, J 14.8 Hz, ArCH2), 3.89 (1H, m, H1), 3.84 (3H, s,
-OCH3), 2.99 (1H, m, H4), 2.52-2.71 (2H, m, H5) ppm; m/z
(EI) 390, 311, 309, 121. HRMS (EI) calcd for C14H14NO279Br2D
M 387.9532, found M 387.9549; calcd for C14H14NO279Br81BrD
M 389.9512, found M 389.9515; calcd for C14H15NO281Br2D M
391.9493, found M 391.9466.
7-a n t i-Br om o-6-d e u t e r iu m -2-(4′-m e t h oxyb e n zyl)-2-
a za bicyclo[2.2.1]h ep t-5-en -3-on e (57). Following the prepa-
ration of 36 from 35, dibromolactam 56 (0.80 g, 2.1 mmol) was
treated with neat DBN (2 g) to afford 57 (0.48 g, 89%) as a
colorless solid: 1H NMR (300 MHz, CDCl3) δ 7.11 (2H, m,
ArH), 6.87 (2H, m, ArH), 6.47 (1H, m, H6), 4.66 (1H, m, H7),
4.32 (1H, d, J 14.5 Hz, ArCH2), 4.16 (1H, d, J 14.5 Hz, ArCH2),
4.12 (1H, m, H1), 3.82 (3H, s, -OCH3), 3.55 (1H, m, H4) ppm;
m/z (EI) 311, 309, 241, 243, 229, 199, 163, 121. HRMS (EI)
calcd for C14H1379BrNO2D M 308.0271, found M 308.0279; calcd
for C14H1381BrNO2D M 310.0252, found M 310.0299.
6-Deu ter iu m -2-(4′-m eth oxyben zyl)-2-a za bicyclo[2.2.1]-
h ep t-5-en -3-on e (58). As described in the preparation of 21
from 20, compound 57 (0.46 g, 1.5 mmol) was allowed to react
with tributyltin hydride (0.65 g, 2.23 mmol) in the presence
of AIBN (20 mg), giving 58 (0.22 g, 68%) as a colorless oil: 1H
NMR (300 MHz, CDCl3) δ 7.11 (2H, m, ArH), 6.87 (2H, m,
ArH), 6.47 (1H, dd, J 1.38, 3.2 Hz, H6), 4.3 (1H, d, J 14.8 Hz,
ArCH2), 4.00 (1H, d, J 14.8 Hz, ArCH2), 4.05 (1H, m, H1), 3.82
(3H, s, -OCH3); 3.40 (1H, m, H4), 2.30 (1H, m, H7), 2.09 (1H,
m, H7) ppm; m/z (EI) 230, 163, 121, 69. HRMS (EI) calcd for
162, 136, 121, 78. HRMS (EI) calcd for C14H1579BrFNO2
327.02706, found M 327.02702; calcd for C14H1581BrFNO2
329.0250, found M 329.0227.
M
M
(1R,2S,3S,4S)-3-Am in o-4-br om o-2-flu or ocyclop en ta n e-
1-ca r boxylic Acid ((+)-15). In the general deprotection
procedure by CAN, the protected lactam 49 (0.37 g, 1.1 mmol)
was treated with CAN (1.8 g, 3.4 mmol), giving 50 as a solid.
Without further purification, the solid was hydrolyzed by the
general acid hydrolysis procedure with a basic elution from
an ion-exchange column, giving (+)-15 (0.08 g, 35%) as a
colorless solid. The product was further purified by recrystal-
lization from methanol/ether.
For 50: 1H NMR (300 MHz, D2O) δ 5.17 (1/2H, m, H7), 4.97
(1/2H, m, H7), 4.21 (1H, m, H6), 4.14 (1H, m, H1), 2.85 (1H, m,
H4), 2.30-2.60 (2H, m, H5); 19F NMR (300 MHz, D2O) -182.38
(dd, J 59.5, 37.5 Hz) ppm; m/z (EI) 207, 127, 108, 82, 59. HRMS
(EI) calcd for C6H779BrFNO M 206.9695, found M 206.9696;
calcd for C6H781BrFNO M 208.9675, found M 208.9679.
For (+)-15: mp 126 °C (decomp.); [R]23.5 ) + 45.7° (c, 1.36,
H2O); 1H NMR (300 MHz, D2O) δ 5.16 (1/2H, t, J 6.0 Hz, H2),
4.99 (1/2H, t, J 6.0 Hz, H2), 4.12 (1H, quart., J 8.4 Hz, H4),
3.90 (1H, m, H3), 3.06 (1H, m, H1), 2.38-2.53 (2H, m, H5);
19F NMR (300 MHz, D2O) -179.85 (dt, J 55.8, 21.6 Hz)
ppm; m/z (EI) 199, 201, 162, 121, 74, 56. HRMS (ES) calcd for
C6H979BrFNO2H M + H 225.9879, found M + H 225.9889;
calcd for C6H981BrFNO2H M + H 227.9859, found M + H
227.9845.
6-exo-Br om o-2-(4′-m eth oxyben zyl)-7-oxo-2-a za bicyclo-
[2.2.1]h ep ta n -3-on e (51). Following the same oxidation pro-
cedure from 23 to 24, compound 45 (20 mg) was treated with
excess of NMO in the presence of TPAP to afford crude 51.
Purification by flash column chromatography yielded pure (()-
51 (14 mg, 70%) as a solid: 1H NMR (300 MHz, CDCl3) δ 7.22
(2H, m, ArH), 6.92 (2H, m, ArH), 4.64 (1H, d, J 14.7 Hz,
ArCH2), 4.40 (1H, d, J 14.7 Hz, ArCH2), 3.84 (3H, s, OCH3),
3.74 (1H, m, H6), 3.72 (1H, m, H1), 3.04 (1H, dd, J 4.1, 1.2 Hz,
H4), 2.62 (1H, dd, J 14.3, 9.1 Hz, H5-endo or H5-exo), 2.49 (1H,
m, H5-endo or H5-exo) ppm.
5-en do-Br om o-2-(4′-m eth oxyben zyl)-6-oxo-2-azabicyclo-
[2.2.1]h ep ta n -3-on e (52). Following the same procedure as
for the preparation of 45 (20 mg), compound 48 was treated
with excess TBAF to afford crude 46. Without any purification,
compound 46 was oxidized to form 52 (10 mg, 50%) by the
same method as for the preparation of 51: 1H NMR (400 MHz,
CDCl3) δ 7.17 (2H, m, ArH), 6.86 (2H, m, ArH), 4.70 (1H, d, J
14.8 Hz, ArCH2), 4.29 (1H, d, J 4.0 Hz, H5), 3.89 (1H, d, J
14.8 Hz, ArCH2), 3.78 (3H, s, OCH3), 3.68 (1H, s, H1), 3.25 (1H,
m, H4), 2.50 (1H, m, H7-syn), 2.02 (1H, m, H7-anti) ppm.
7-a n ti-Br om o-2-(4′-m eth oxyben zyl)-6-oxo-2-a za bicyclo-
[2.2.1]h ep ta n -3-on e (54). Following the same oxidation pro-
cedure used to convert 23 to 24, 7-anti-bromo-6-exo-hydroxyl-
2-(4′-methoxybenzyl)-2-azabicyclo[2.2.1]heptan-3-one (53)11 (2.3
g, 7.0 mmol) was treated with NMO (1.64 g, 14 mmol) in
the presence of a catalytic amount of TPAP (5 mg) to afford
54 (1.8 g, 78%) as a colorless solid: 1H NMR (300 MHz, CDCl3)
δ 7.16 (2H, m, ArH), 6.88 (2H, m, ArH), 4.72 (1H, d, J 14.6
Hz, ArCH2), 4.37 (1H, m, H7), 3.90 (1H, d, J 14.6 Hz, ArCH2),
3.81 (3H, s, OCH3), 3.68 (1H, m, H1), 3.17 (1H, m, H4), 2.72
(2H, d, J 17.8, 4.1 Hz, H5), 2.23 (2H, d, J 17.8, 2.5 Hz, H5)
ppm; m/z (EI) 323, 325, 244, 216, 121. HRMS (EI) calcd for
C
14H14NO2D M 230.1166, found M 230.1166.
6-exo-7-a n ti-Dibr om o-1-d eu ter iu m -2-(4′-m eth oxyben -
zyl)-2-a za bicyclo[2.2.1]h ep ta n -3-on e (59). Bromine (0.15 g,
0.94 mmol) was added to a stirred solution of 58 (0.18 g, 0.78
mmol) in dichloromethane (10 mL) at 0 °C. After being stirred
for 1 h, the reaction mixture was diluted with dichloromethane
(50 mL) and washed with a saturated Na2SO3 solution (20 mL)
and brine (20 mL). The organic solution was dried over MgSO4
and concentrated under reduced pressure to afford 59 (0.29 g,
95%) as a colorless solid: 1H NMR (300 MHz, CDCl3) δ 7.19
(2H, m, ArH), 6.91 (2H, m, ArH), 4.60 (1H, d, J 14.8 Hz,
ArCH2), 4.24 (1H, t, J 1.5 Hz, H7), 4.00 (1H, d, J 14.8 Hz,
ArCH2), 3.84 (3H, s, -OCH3), 3.78 (1H, m, H6), 2.99 (1H, m,
H4), 2.68 (1H, m, H5), 2.52 (1H, m, H5) ppm. HRMS (ES) calcd
for C14H14NO279Br2D + H 388.9600, found M + H 388.9648;
calcd for C14H14NO279Br81BrD + H 390.9590, found M + H
390.9620; calcd for C14H14NO281Br2D + H 392.9570, found M
+ H 392.9598.
C
C
14H1479BrNO3 M 323.0157, found M 323.0153; calcd for
14H1481BrNO3 M 325.0138, found M 323.0134.
7-a n ti-Br om o-6-d eu t er iu m -6-h yd r oxy-2-(4′-m et h oxy-
ben zyl)-2-a za bicyclo[2.2.1]h ep ta n -3-on e (55). NaBD4 (0.13
g, 3.1 mmol) was added to a stirred solution of 54 (1.02 g, 3.1
mmol) in EtOD (10 mL). After being stirred for 1 h, the
reaction mixture was concentrated under reduced pressure.
The residue was treated with 0.5 N HCl (10 mL), and the
3r-Am in o-2â,4â-d ibr om o-3â-d eu t er iu m cyclop en t a n e-
1r-ca r boxylic Acid (61). Following the general deprotection
procedure by CAN, compound 59 (0.22 g, 0.45 mmol) was
treated with CAN (0.92 g, 1.7 mmol) to afford crude 60.

718 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Qiu and Silverman
Without further purification, 60 was hydrolyzed following the
general acid hydrolysis procedure, then purified by ion-
exchange chromatography with 1 N aqueous pyridine, giving
61 (0.09 g, 60%) as a light-yellow solid: mp 134 °C (decomp.);
1H NMR (300 MHz, D2O) δ 4.14-4.22 (2H, H4 and H2), 3.16
(1H, quart., J 9.0, H1), 2.41-2.48 (2H, m, H5) ppm. HRMS
calcd for C6H879Br2NO2D + H M + H 286.9520, found M + H
286.9518; calcd for C6H881Br79BrNO2D + H M + H 288.9946,
found M + H 288.9956; calcd for C6H881Br2NO2D + H M + H
290.9848, found M + H 290.9865.
7-Deu ter iu m -2-(4′-m eth oxyben zyl)-2-a za bicyclo[2.2.1]-
h ep t-5-en -3-on e (63-D). Tributyltin chloride (0.058 g, 0.18
mmol) in anhydrous ethanol (1 mL) was added to a stirred
suspension of NaBD4 (0.026 g, 0.60 mmol) in ethanol (2 mL)
at 0 °C. A precipitate formed. Then 7-anti-bromo-2-(4′-meth-
oxybenzyl)-2-azabicyclo[2.2.1]hept-5-en-3-one (62, 0.28 g, 0.90
mmol), which was prepared following the same procedure as
the preparation of 36, in anhydrous ethanol (3 mL) was added
to the above solution. The resultant mixture was stirred and
heated at reflux for 15 h and then was diluted with ethyl
acetate (100 mL). The organic solution was washed with
saturated NaHCO3 and brine and dried over anhydrous Na2-
SO4. After the organic solvents were concentrated under
reduced pressure, the residue was purified by column chro-
matography on silica gel, eluting with ethyl acetate/hexane
(1:2), to afford 63-D (48 mg, 40%): 1H NMR (300 MHz, CDCl3)
δ 7.10 (2H, m, ArH), 6.84 (2H, m, ArH), 6.52 (2H, m, H5 and
H6), 4.34 (1H, d, J 14.7 Hz, CH2Ar), 4.02 (1H, m, H1), 4.01
(1H, d, J 14.7 Hz, CH2Ar), 3.80 (3H, s, OCH3), 3.37 (1H, m,
H4), 2.26 (1/2 H, s, H7-syn or H7-anti), 2.04 (1/2H, s, H7-syn or
in 30 min at a flow rate of 1.0 mL/min. The column effluent
was monitored at 214 nm. The collected fraction was 1 mL/
min. All fractions were also checked by liquid scintillation
counting. Compound 66-T has a rention time of 22 min.
Finally, 66-T was mixed with (+)-10 to make a 100 mM
solution with a specific radioactivity of 4.2 mCi/mmol for the
enzyme inactivation studies.
En zym e a n d Assa ys. GABA aminotransferase was isolated
from pig brains by a modified procedure.25 Succinic semialde-
hyde dehydrogenase (SSDH) was isolated from GABAse, a
commercially available mixture of SSDH and GABA ami-
notransferase, by inactivation of the GABA aminotransferase
with gabaculine as described previously.26 GABA activity
assays were carried out using a modification of the coupled
assay developed by Scott and J akoby.27 The assay solution
contains 11 mM GABA, 5.3 mM R-ketoglutarate, 1.1 mM
NADP+, and 5 mM â-mercaptoethanol in 100 mM potassium
pyrophosphate, pH 8.5, and excess SSDH. Using this assay,
the change in absorbance at 340 nm indicates production of
NADPH which is directly proportional to the activity of GABA
aminotransferase.
Tim e-Dep en d en t In a ctiva tion of GABA Am in otr a n s-
fer a se by 7-15 a n d 61. GABA aminotransferase (15 µL, 1.80
mg/mL) was added to solutions of 7-15 (110 µL in final, with
varied concentrations of compounds), in 100 mM potassium
pyrophosphate buffer, pH 8.5, containing 5 mM R-ketoglut-
arate and 1 mM â-mercaptoethanol at 25 °C. For 8, 11, 12,
13, and 14, concentrations of 15 mM and incubation times of
more than 1 h were used with no time-dependent inhibition
observed. Concentrations of (+)-7, (-)-9, (+)-10, and (+)-15
used were as follows: (+)-7, 22, 51, 64, 88, and 130 µM; (-)-9,
4.6, 9.2, 14, and 19 mM; (+)-10, 2.8, 5.5, 8.3, 11, 17, and 25
mM; (+)-15, 3.0, 6.1, 9.1, 14, and 18 mM. For the measurement
of the deuterium isotope effect, the concentrations of (()-10
and (()-61 were as follows: 4.2, 8.4, 17, 25, and 34 mM for
(()-10 and 8.4, 17, 21, 25, and 34 mM for (()-61. At timed
intervals, aliquots (15 µL) were withdrawn and added to the
assay solution (585 µL) with excess SSDH. Rates were
measured spectrophotometrically at 340 nm, and the logarithm
of the remaining activity was plotted against time for each
concentration of inhibitor. A secondary plot of t1/2 obtained from
the first plot versus 1/[inactivator] was constructed to deter-
mine KI and kinact values28 for those four inactivators.
H
7-anti)) ppm.
6-exo-7-a n ti-Dibr om o-5-d eu ter iu m -2-(4′-m eth oxyben -
zyl)-2-a za bicyclo[2.2.1]h ep t-5-en -3-on e (64-D). Following
the same procedure as for the preparation of 59, compound
63-D (48 mg, 0.2 mmol) was treated with excess bromine to
yield 64-D (75 mg, 95%): 1H NMR (300 MHz, CDCl3) δ 7.19
(2H, m, ArH), 6.91 (2H, m, ArH), 4.60 (1H, d, J 14.8 Hz,
ArCH2), 4.24 (1H, m, H7), 4.00 (1H, d, J 14.8 Hz, ArCH2), 3.
96 (1H, m, H1), 3.84 (3H, s, -OCH3), 3.78 (1H, m, H6), 2.99
(1H, m, H4), 2.68 (1/2H, m, H5-exo or H5-endo), 2.52 (1/2H, m,
H5-exo or H5-endo) ppm.
3r-Am in o-2â,4â-d ibr om o-5r/â-d eu ter iu m cyclop en ta n e-
1r-ca r boxylic Acid (66-D). Following the same procedure as
for 61, 64-D (75 mg, 0.19 mmol) was deprotected with cerium-
(IV) ammonium nitrate (0.32 g, 0.60 mmol), giving 65-D (65
mg). Acid hydrolysis of 65-D and ion-exchange chromatogra-
phy, as with 10, gave 66-D (29 mg, 53%): 1H NMR (300 Hz,
D2O) δ 4.27 (2H, m, H2 and H4), 4.10 (1H, m, H3), 3.27 (1H, m,
H1), 2.55 (1H, m, H5) ppm.
Su bstr a te P r otection a n d Requ ir em en t for P LP for
In a ctiva tion . Substrate protection was carried out as de-
scribed for inactivation of GABA-AT except 10 (12 mM) was
incubated with GABA aminotransferase in the presence of
GABA (0.6 mM and 5.1 mM, respectively) and in the absence
of GABA.
(1R,2S,3R,4S)-3-Am in o-2,4-d ibr om o-5r/â-tr itiu m cyclo-
p en ta n e-1-ca r boxylic Acid ((+)-66-T). Following the same
procedure used for the preparation of 63-D, 100 mCi of [3H]-
NaBH4 (15 Ci/mmol) in anhydrous ethanol (1 mL) was added
to a stirred solution of tributyltin chloride (150 mg) in
anhydrous ethanol (1 mL). After being stirred for 25 min,
NaBH4 (25 mg) in anhydrous ethanol (1 mL) was added to the
above solution, then (1S,4R)-62 (198 mg) in anhydrous ethanol
(2 mL) was added. The resultant solution was stirred and
heated at reflux for 20 h. Workup and column purification
afforded the tritium-labeled product 63-T (0.36 mCi) with a
radioactivity yield of 1.4%.
To determine the importance of PLP for inactivation, GABA
aminotransferase (30 µL, 1.80 mg/mL) was preincubated with
1.2 mM GABA in a 100 mM potassium pyrophosphate buffer
(160 µL, pH 8.5) containing 1 mM â-mercaptoethanol at 25
°C, but no R-ketoglutarate, for 1 h. The incubation solution
was divided equally into two parts. One part was treated with
(+)-10 (20 mM) in the absence of R-ketoglutarate, and the
other part was treated with (+)-10 (20 mM) in the presence of
R-ketoglutarate (5 mM). The remaining enzyme activities were
monitored following the procedure described in the time-
dependent inactivation experiment.
Com p etitive In h ibition of GABA Am in otr a n sfer a se by
7-15. The activity of GABA aminotransferase (20 µL, 0.15 mg/
mL) at 25 °C in 600 µL of 120 mM potassium pyrophosphate
buffer (pH 8.5) containing excess succinic semialdehyde de-
hydrogenase, 5.3 mM R-ketoglutarate, and 1 mM NADP+ was
determined upon the introduction of varying concentrations
of 7-15 (two different concentrations for each GABA concen-
tration) and at different GABA concentrations. The percentage
of inhibition was obtained by comparison to an untreated
enzyme control. The Ki values were calculated based on the
following equation: % Inhibition ) 100[I]/{[I] + Ki(1 + [GABA]/
Km)}.29 The range of GABA concentrations is from 2.7 to 6.0
mM. The concentrations for these analogues are as follows:
0.04, 0.053, and 0.080 mM for (+)-4; 0.68 and 1.4 mM for 5;
Following the same procedure described for the preparation
of 64-D, compound 63-T was treated with excess bromine (60
mg) to afford the crude dibromo product 64-T (40 mg).
The crude 64-T was treated with cerium(IV) ammonium
nitrate (310 mg), which gave 65-T (0.32 mCi). Hydrolysis of
65-T with 4 N HCl (5 mL) and acetic acid (5 mL) at 75 °C
afforded 66-T, which was purified with AG 50W-X8 and
washing with water and then with 1 M aqueous pyridine
solution to give 66-T (0.22 mM; specific radioactivity 40 mCi/
mmol). The total activity yield was 0.9%. Compound 66-T was
further purified on a Hypresil C-18 Elite analytical column
(5.4 × 250 mm), initially eluting with water for 10 min, then
eluting with methanol solution in a gradient from 0% to 100%

Inactivators of γ-Aminobutyric Acid Aminotransferase
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 719
4.1, 6.8, 11, and 14 mM for (+)-6; 2.5, 5.0, 6.2, and 12 mM for
(-)-7; 6.2, 9.3, and 15 mM for 8; 3.8, 4.9, 5.4, and 11.4 mM for
(+)-9; 4.0, 12, 24, and 28 mM for 10; 1.78, 5.35, 8.9, and 11
mM for 11; 0.56, 1.12, and 1.7 mM for (+)-12.
presoaked with 8 M urea). A 40 µL aliquot of the inactivated
enzyme sample was counted by liquid scintillation. The protein
concentrations of enzyme sample (20 µL in 8 M urea) and BSA
standard (20 µL in 8 M urea) with 700 µL of Coomassie Plus
Protein Assay Reagent (Pierce) were measured by a UV
spectrometer at 595 nm.
Relea se of F lu or id e Ion s d u r in g In a ctiva tion GABA-
AT by (+)-7. Five different amounts of GABA-AT (28, 38, 47,
56, and 75 µg) were incubated for 16 h with (+)-7 (16 mM) in
100 mM potassium pyrophosphate buffer, pH 8.5, containing
5.0 mM â-mercaptoethanol and 3.0 mM R-ketoglutarate in a
total volume of 200 µL at 23 °C. No enzyme activity was
observed. A control containing no enzyme also was run. After
each sample (200 µL) was mixed with 5.00 mL of distilled
water and 5.00 mL of total ionic strength adjusting buffer
(TISAB II, Orion), the concentration of fluoride ions in each
sample (200 µL) was measured with a fluoride ion electrode.
Su p p or tin g In for m a tion Ava ila ble: Elemental analyses,
time-dependent inactivation plots for (+)-7, (-)-9, (+)-10, and
(+)-15, substrate protection of (+)-10, and effect of R-ketoglu-
tarate on inactivation of (+)-10. This information is available
free of charge via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. The authors are grateful to the
National Institutes of Health (NS 15703) for financial
support of this research.
Relea se of F lu or id e Ion s d u r in g In a ctiva tion GABA-
AT by (()-8. Analogue (()-8 (11.6 mM) was incubated for 16
h with GABA-AT (40 µL, 0.94 mg/mL) in 100 mM potassium
pyrophosphate buffer, pH 8.5, containing 5.0 mM â-mercap-
toethanol and 3.0 mM R-ketoglutarate in a total volume of 200
µL. The resultant incubation solution was mixed with 5.00 mL
of distilled water and 5.00 mL of total ionic strength adjusting
buffer (TISAB II, Orion) for the fluoride ion concentration
measurement.
Tr a n sa m in a tion Even ts d u r in g In cu ba tion w ith (+)-7
a n d (()-8. Analogue (+)-7 or (()-8 (16 mM) was incubated
for 16 h with GABA-AT (30 µL, 0.94 mg/mL) in 100 mM
potassium pyrophosphate buffer, pH 8.5, containing 5.0 mM
â-mercaptoethanol in a total volume of 200 µL in the presence
of 5.0 mM [14C]-R-ketoglutarate at 23 °C. No [14C]-L-glutamate
was detected.
Refer en ces
(1) Baxter, C. F.; Roberts, E. J . The γ-Aminobutyric Acid-R-
ketoglutaric acid Transaminase of Bee Brain. J . Biol. Chem.
1958, 233, 1135-1139.
(2) (a) Leach, M, J .; Walker, J . M. G. Effect of Ethanolamine-O-
sulphate on Regional GABA Metabolism in the Mouse Brain.
Biochem. Pharmacol. 1977, 26, 1569-1572. (b) Phillips, N. I.;
Fowler, L. J . The Effects of Sodium Valproate on γ-Aminobu-
tyrate Metabolism and Behaviour in Naive and Ethanolamine-
O-sulphate Pretreated Rats and Mice. Biochem. Pharmacol.
1982, 31, 2257-2261.
(3) Gale, K. GABA in Epilepsy: the Pharmacological Basis. Epi-
lepsia 1989, 30 (Suppl 3), S1-S11.
(4) Kushner, S. A.; Dewey, S. L.; Kornetsky, C. The Irreversible
gamma-Aminobutyric Acid (GABA) Transaminase Inhibitor
gamma-Vinyl-GABA Blocks Cocaine Self-Administration in Rats.
J . Pharmacol. Exp. Ther. 1999, 290, 797-802.
(5) Nanavati, S. M.; Silverman, R. B. Design of Potential Anticol-
vulsant Agents: Mechanistic Classification of GABA Ami-
notransferase Inactivators. J . Med. Chem. 1989, 32, 2413-2421.
(6) Morino, Y.; Kojima, H.; Tanase, S. Affinity Labeling of Alanine
Aminotransferase by 3-Chloro-L-alanine, J . Biol. Chem. 1979,
254, 279-285.
(7) Wang, E.; Walsh, C. Suicide Substrates for the Alanine Race-
mase of Escherichia Coli B. Biochemistry 1978, 17, 1313-1321.
(8) Relyea, N. M.; Tate, S. S.; Meister, A. Affinity Labeling of the
Active Center of L-Aspartate-â-decarboxylase with â-chloro-L-
alanine. J . Biol. Chem. 1974, 249, 1519-1524.
(9) (a) Silverman, R. B.; Levy, M. A. Synthesis of (S)-5-Substituted
4-Aminopentanoic Acids: A New Class of γ-Aminobutyric Acid
Transaminase Inactivators. J . Org. Chem. 1980, 45, 815-818.
(b) Silverman, R. B.; Levy, M. A. Irreversible Inactivation of Pig
Brain γ-Aminobutyric Acid-R-ketoglutarate Transaminase by
4-Amino-5-halopentanoic Acids. Biochem. Biophys. Res. Com-
mun. 1980, 95, 250-255. (c) Silverman, R. B.; Levy, M. A.
Mechanism of Inactivation of γ-Aminobutyric Acid-R-Ketoglu-
taric Acid Aminotransferase by 4-Amino-5-halopentanoic Acids.
Biochemistry 1981, 20, 1197-1203.
(10) Silverman, R. B.; Invergo, B. J . Mechanism of Inactivation of
γ-Aminobutyrate Aminotransferase by 4-Amino-5-fluoropen-
tanoic Acid. First Example of an Enamine Mechanism for
γ-Amino Acid with a Partition Ratio of 0. Biochemistry 1986,
25, 6817-6820.
(11) Nanavati, S. M.; Silverman, R. B. Mechanisms of Inactivation
of γ-Aminobutyric Acid Aminotransferase by the Antiepilepsy
Drug γ-Vinyl GABA (Vigabatrin). J . Am. Chem. Soc. 1991, 113,
9341-9349.
(12) Qiu, J .; Pingsterhaus, J .; Silverman, R. B. Inhibition and
Substrate Activity of Conformationally-Rigid Vigabatrin Ana-
logues with γ-Aminobutyric Acid Aminotransferase J . Med.
Chem. 1999, 42, 4725-4728.
(13) Faith, W. C.; Booth, C. A.; Foxman, B. M.; Snider, B. B. An
Approach to the Synthesis of Neplanocin. J . Org. Chem. 1985,
50, 1983-1985.
Deter m in a tion of th e Nu m ber of Equ iva len ts of 66-T
Bou n d to GABA Am in otr a n sfer a se a fter In a ctiva tion . In
a
typical experiment 66-T (16 mM, 4.2 mCi/mmol) was
incubated with GABA-AT (50 µL, 1.80 mg/mL) in the presence
of 2 mM R-ketoglutarate and 5 mM â-mercaptoethanol in 100
mM potassium pyrophosphate buffer (pH 8.5) for 16 h. More
than 60% of the enzyme was inactivated. The incubation
solution was loaded into a Centricon concentrator (10 000 MW
cutoff 3000) and was diluted to 1.5 mL with 50 mM potassium
pyrophosphate buffer (pH 8.5). Then the resultant solution was
concentrated to about 0.1 mL. The concentrated solution was
diluted to 1 mL with the same buffer and then was concen-
trated to 0.1 mL. After the removal of excess inactivator by
the method of Penefsky30 (Sephadex G-50, presoaked with 50
mM potassium pyrophosphate buffer, pH 8.5), a 30 µL aliquot
of the inactivated enzyme sample was counted by liquid
scintillation. The protein concentrations of the inactivated
enzyme sample (20 µL) and BSA standard (20 µL) were
determined in 50 mM potassium pyrophosphate buffer, pH 8.5.
A control experiment also was carried out with the PMP form
of GABA-AT, which was obtained by pretreatment of GABA-
AT with GABA (5 mM) in the absence of R-ketoglutarate. The
number of equivalents of inactivator bound to the enzyme was
calculated as the ratio of millimoles of inactivator in the
sample (determined by scintillation counting and the specific
activity of the inactivator) to the millimoles of enzyme in the
sample (determined by protein assay). This calculation also
was used for the control experiment. Finally, the number of
equivalents of inactivator bound to the enzyme was obtained
after subtraction of the control experiment from the experi-
ment.
Equ iva len ts of 66-T Bou n d to GABA Am in otr a n sfer a se
a fter In a ctiva tion a n d Den a tu r a tion w ith Ur ea . Tritium-
labeled dibromo analogue 66-T (10 mM, 4.2 mCi/mmol) was
incubated for 16 h with GABA aminotransferase (80 µL, 0.94
mg/mL) in 100 mM potassium pyrophosphate buffer, pH 8.5,
containing 5.0 mM â-mercaptoethanol and 3.0 mM R-ketoglu-
tarate in a total volume of 200 µL; >50% of the enzyme was
inactivated. The resultant incubation solution was loaded into
a Slide-A-Lyzer mini dialysis unit (Pierce, 10 000 molecular
weight cutoff) and dialyzed against 8 M urea (200 mL, changed
each 8 h) for 4 days. The resultant protein sample was further
purified by the Penefsky spin method30 (Sephadex G-50,
(14) Palmer, C. F.; Parry, K, P.; Roberts, S. M.; Sik, V. Rearrange-
ment of 2-Azabicyclo[2.2.1]hept-5-en-3-ones: Synthesis of cis-
3-Aminocyclopentane Carboxylic Acid Derivatives. J . Chem. Soc.,
Perkin Trans. 1 1992, 1021-1028.
(15) Middleton, W. New Fluorinating Reagents. Dialkylaminosulfur
Fluorides. J . Org. Chem. 1975, 40, 574-578.
(16) Ley, S. V.; Norman, J .; Griffith, W. P.; Marsden, S. P. Tetra-
propylammonium Perruthenate, Pr4N+RuO-4, TPAP: A Cata-
lytic Oxidant for Organic Synthesis. Synthesis 1994, 639-666.
(17) Shiozaki, M.; Ishida, N.; Hiraoka, T.; Marayuma, H. Ste-
reospecfic Synthesis of Chiral Precursor of Thienamycin from
L-Threonine Tetrahedron 1984, 40, 1795-1802.
(18) Smith, A. B., III; Rano, T. A.; Chida, N.; Sulikowski, C. A.; Wood,
J . L. Total Synthesis of the Cytotoxic macrocyclo (+)-Hitachi-
mycin. J . Am. Chem. Soc. 1992, 114, 8008-8022.

720 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Qiu and Silverman
(19) Couture, A.; Deniau, E.; Grandclaudon, P.; Lebrun, S. A New
Approach to the Synthesis of Aristolactams. Total Synthesis of
Cepharanone A and B. Synlett 1997, 1475-1477.
(25) Koo, Y. K.; Nandi, D.; Silverman, R. B. Multiple Active Enzyme
Species of γ-Aminobutyric Acid Aminotransferase are not
Isozymes. Arch. Biochem. Biophys., submitted.
(26) Hopkins, M. H.; Bichler, K. A.; Su, T.; Charmberlain, C. L.;
Silverman, R. B. Inactivation of γ-Aminobutyric Acid Ami-
notransferase by Various Amine Buffers. J Enzymol. Inhibition
1992, 6, 195-199.
(27) Scott, E. M.; J akoby, W. B. Soluble γ-Aminobutyric-Glutamic
Transaminase from Pseudomonas Fluorescens W. B. J . Biol.
Chem. 1959, 234, 932-936.
(28) (a) Silverman, R. B. Mechanism-Based Enzyme Inactivation:
Chemistry and Enzymology, Vol. 1; CRC Press: Boca Raton, FL,
1988; p 12. (b) Silverman, R. B. Methods Enzymol. 1995, 249,
240-283.
(29) Segal, I. H. Enzyme Kinetics; J ohn Wiley & Sons: New York;
1975; p 106.
(20) Katagiri, N.; Matsuhashi, Y.; Kokufuda, H.; Takebayashi, M.;
Kaneko, C. A Highly Efficient Synthesis of the Antiviral Agent
(+)-Cycloaradine Involving the Regioselective Cleavage of Ep-
oxide by Neighboring Participation. Tetrahedron Lett. 1997, 38,
1961-1964.
(21) Faith, W. C.; Booth, C. A.; Foxman, B. M.; Snider, B. B. An
Approach to the Synthesis of Neplanocin A. J . Org. Chem. 1985,
50, 1983-1985.
(22) (a) Winstein, S.; Stafford, J . Syn-7-Norbornenyl Toluene-
sulfonate. J . Am. Chem. Soc. 1957, 79, 505-507. (b) Walborsky,
H. M.; Loncrini, D. F. Reaction of Exo-norbornylene Oxide with
Hydrogen Bromide. J . Org. Chem. 1957, 22, 1117-1119.
(23) Bey, P.; J ung, M. J .; Gerhart, F.; Schirlin, D.; Van Dorsselaer,
V.; Casara, P. ω-Fluoromethyl Analogues of ω-Amino Acids as
Irreversible Inhibitors of 4-Aminobutyrate: 2-Oxoglutarate Ami-
notransferase. J . Neurochem. 1981, 37, 1341-1344.
(30) Penefsky, H. Reversible Binding of Pi by Beef Heart Mitochon-
drial Adenosine Triphosphatase. J . Biol. Chem. 1977, 252,
2891-2899.
(24) Silverman, R. B.; Levy, M. A. Substituted 4-Aminobutanoic
Acids: Substrates for γ-Aminobutyric Acid R-Ketoglutaric Acid
Aminotransferase J . Biol. Chem. 1981, 256, 11565-11568.
J M9904755