Desyl esters of amino acid neurotransmitters. Phototriggers for biologically active neurotransmitters

Article abstract of DOI:10.1021/jo951635x

The application of the desyl or 2-oxo-1,2-diphenylethyl moiety as a photolabile ligand for the release of phosphates such as cAMP and inorganic phosphate (P(i))^1,2 is extended to include selected excitatory amino acids. The synthesis and photochemical studies of N- and O-desyl-caged versions of the endogenous amino acid neurotransmitters glutamate and γ-aminobutyric acid (GABA) are reported. Photolysis at 350 nm of solutions of γ-O-desyl glutamate (11) and O-desyl GABA (14) in 1:1 H₂O-acetonitrile cleanly produced free glutamate and GABA, respectively, with rate constants of ca. 10⁷ s^-1; 2-phenylbenzo[b]furan (2) was the only photobyproduct detected by HPLC. Photolysis quantum efficiencies for the disappearance of O-desyl amino acid esters were in the range of 0.29-0.31, and the appearance efficiencies of furan 2 (and the corresponding amino acid) were 0.14. The photolysis of 14 was efficiently quenched with sodium 2-naphthalenesulfonate, yielding a triplet lifetime of ca. 10 ns. Photolysis of 11 in mammalian brain tissue slices resulted in glutamate receptor activation, as indicated by whole cell electrophysiological measurements. Photolysis of the other desyl amino acids resulted in decomposition and produced several products but did not lead to the formation of furan 2.

Full text of DOI:10.1021/jo951635x

1228
J . Org. Chem. 1996, 61, 1228-1233
Desyl Ester s of Am in o Acid Neu r otr a n sm itter s. P h ototr igger s for
Biologica lly Active Neu r otr a n sm itter s†
Kyle R. Gee,*,‡ L. William Kueper, III,§ J effrey Barnes,§ Gregory Dudley,§ and
Richard S. Givens*,§
Molecular Probes, Inc., 4849 Pitchford Avenue, Eugene, Oregon 97402, and Department of Chemistry,
The University of Kansas, Lawrence, Kansas 66045
Received September 6, 1995X
The application of the desyl or 2-oxo-1,2-diphenylethyl moiety as a photolabile ligand for the release
of phosphates such as cAMP and inorganic phosphate (Pi)1,2 is extended to include selected excitatory
amino acids. The synthesis and photochemical studies of N- and O-desyl-caged versions of the
endogenous amino acid neurotransmitters glutamate and γ-aminobutyric acid (GABA) are reported.
Photolysis at 350 nm of solutions of γ-O-desyl glutamate (11) and O-desyl GABA (14) in 1:1 H2O-
acetonitrile cleanly produced free glutamate and GABA, respectively, with rate constants of ca.
107 s-1; 2-phenylbenzo[b]furan (2) was the only photobyproduct detected by HPLC. Photolysis
quantum efficiencies for the disappearance of O-desyl amino acid esters were in the range of 0.29-
0.31, and the appearance efficiencies of furan 2 (and the corresponding amino acid) were 0.14.
The photolysis of 14 was efficiently quenched with sodium 2-naphthalenesulfonate, yielding a triplet
lifetime of ca. 10 ns. Photolysis of 11 in mammalian brain tissue slices resulted in glutamate receptor
activation, as indicated by whole cell electrophysiological measurements. Photolysis of the other
desyl amino acids resulted in decomposition and produced several products but did not lead to the
formation of furan 2.
“Cages” or phototriggers for the rapid release of bio-
the R-keto cage derivative, of which the desyl or 2-oxo-
1,2-diphenylethyl group is a leading candidate for a
phototrigger.6
active substrates to elicit biological responses are becom-
ing increasingly important research tools for introduction
of reagents with precise spatial and temporal control.2,3
The most commonly used “cage” for nucleotides, neuro-
transmitters, and a variety of other important biochemi-
cal effectors is the o-nitrobenzyl functional group.4 Criti-
cal features of an effective cage group include high
efficiency for release of the substrate, rapid rate of
release, and production of a biologically benign byproduct
from the cage itself. Efforts to optimize photolysis
quantum efficiencies and rates by substitution at the
benzylic position or on the aromatic nucleus of the
o-nitrobenzyl group have usually resulted in unpredict-
able and undesirable changes in the photochemistry and
the bioavailability of the substrate.3a Moreover, the
release rates tend to be highly sensitive to the nature of
the leaving group, and the byproduct from the cage, e.g.,
o-nitrosobenzaldehyde, is not biologically benign.5 For
these reasons, attention has been drawn to the design
and study of new, potentially faster and less damaging
cage groups. An emerging class that shows promise is
Whereas the first reports of the photochemistry of desyl
derivatives by Sheehan et al.,7 were not encouraging
because of the many and varied photoproducts obtained
often in low yields, the more recent studies by our group,8
Corrie and Trentham,2,6 Baldwin et al.,9 Pirrung and
Shuey,10 Futura et al.,11 and Cameron et al.12 have since
independently demonstrated the utility of the desyl cage
for the release of phosphates, nucleotides, and carbam-
ates. Of particular importance is the demonstration that
cAMP, as its caged desyl ester (1), was released by
irradiation at 350 nm with a rate constant of 7.2 × 108
s-1, and a quantum efficiency of 0.34 (eq 1).1 This
represented a remarkable increase by 3 orders of mag-
nitude in the rate constant for a phototrigger while
maintaining a reasonably high efficiency relative to the
best nitrobenzyl analogues.3b The desyl ligand was
expelled as 2-phenylbenzo[b]furan (2) concomitant with
release of the cAMP and is a biochemically benign
byproduct, in contrast to the release of the more reactive
nitroso ketones and aldehydes from o-nitrobenzyl
analogues.3a,b
† Dedicated to Professor Glen A. Russell on the occasion of his 70th
birthday.
‡ Molecular Probes, Inc.
(6) (a) Corrie, J . E. T.; Trentham, D. R. J . Chem. Soc., Perkin Trans.
1 1992, 2409-2417. (b) Thirlwell, H.; Corrie, J . E. T.; Reid, G. P.;
Trentham, D. R.; Ferenczi, M. A. Biophys. J . 1994, 67, 2436.
(7) (a) Sheehan, J . C.; Wilson, R. M. J . Am. Chem. Soc. 1964, 86,
5277-5281. (b) Sheehan, J . C.; Wilson, R. M.; Oxford, A. W. J . Am.
Chem. Soc. 1971, 93, 7222-7228. (c) Sheehan, J . C.; Umezawa, J . J .
Org. Chem. 1973, 38, 3771-3774.
§ The University of Kansas.
X Abstract published in Advance ACS Abstracts, February 1, 1996.
(1) (a) Givens, R. S.; Athey, P. S.; Kueper, L. W., III; Matuszewski,
B.; Xue, J .-y. J . Am. Chem. Soc. 1992, 114, 8708. (b) Givens, R. S.;
Athey, P. S.; Kueper, L. W., III; Matuszewski, B.; Xue, J .-y.; Fister, T.
J . Am. Chem. Soc. 1993, 115, 6001.
(2) Corrie, J . E. T.; Reid, D. R.; Trentham, M. B.; Mazid, M. A. J .
Chem. Soc., Perkin Trans. 1 1992, 1015.
(8) Givens, R. S.; Matuszewski, B. J . Am. Chem. Soc. 1984, 106,
6860.
(3) (a) Adams, S. R.; Tsien, R. Y. Ann. Rev. Physiol. 1993, 55, 755-
783. (b) Corrie, J . E. T.; Trentham, D. R. In Bioorganic Photochemistry;
Morrison, H., Ed.; Wiley: New York, 1993; Vol. 2, pp 243-305. (c) Hess,
G. P. Biochemistry 1993, 32, 989-1000. (d) Givens, R. S.; Kueper, L.
W., III. Chem. Rev. 1993, 93, 55-66.
(9) Baldwin, J . E.; McConnaughie, A. W.; Moloney, M. G.; Pratt, A.
J .; Shim, S. B. Tetrahedron 1990, 46, 6879-6884.
(10) Pirrung, M. C.; Shuey, S. W. J . Org. Chem. 1994, 59, 3890-
3897.
(11) (a) Futura, T.; Torigai, H.; Sugimoto, M; Iwamura, M. J . Org.
Chem. 1995, 60, 3953-3956. (b) Futura, T.; Torigai, H.; Osawa, T.;
Iwamura, M. Chem. Lett. 1993, 1179-1182.
(12) Cameron, J . F.; Willson, C. G.; Fre´chet, J . M. J . J . Chem. Soc.,
Chem. Commun. 1995, 923-924.
(4) Patchornik, A.; Amit, B.; Woodward, R. B. J . Am. Chem. Soc.
1970, 92, 6333-6335.
(5) Kaplan, J . H.; Forbush, G., III; Hoffman, J . F. Biochemistry 1978,
17, 1929.
0022-3263/96/1961-1228$12.00/0 © 1996 American Chemical Society

Desyl Esters of Amino Acid Neurotransmitters
J . Org. Chem., Vol. 61, No. 4, 1996 1229
Sch em e 1. Gen er a l Syn th esis of N- a n d O-Desyl
Am in o Acid s
We report here the results of our investigations on a
new class of phototriggers for the excitatory amino acid,
glutamate, and for GABA, and contrast their photo-
chemistry with the corresponding o-nitrobenzyl amino
acids13,14 which have proven to be very valuable for
studies of receptor activation kinetics14,15 and synaptic
connection mapping.15
The desyl triggers for amino acids have not heretofore
been extensively or thoroughly evaluated. Corrie and
Trentham3b reported the synthesis of 3′,5′-dimethoxy-
desyl derivatives of GABA and alanine and noted that
the esters were hydrolytically unstable, prompting the
authors to abandon pursuit of the photolysis studies since
the probes would not be useful in aqueous solution. In
contrast, Sheehan’s report7c of successful release of
t-BOC-protected amino acids from the R-phenacyl esters,
including the N-protected, O-desyl glycinates,7b encour-
aged us to investigate the feasibility of rapid photorelease
of GABA and glutamate from the N- or O-desyl-protected
precursors in more detail.14 Another report extending
the studies on the release of amines though the corre-
sponding carbamates from p-methoxy-R-substituted phen-
acyl cage further illustrated the generality of this ap-
proach.16 Two caveats to the use of R-keto phototriggers
for release of amino acids in biological media, however,
were the assertions by Sheehan et al.7b,c that these
reactions proceed via a free radical mechanism and that
even small amounts of water in the reaction media
produce lower yields.
N-tert-butoxycarbonyl amino acids. Reaction of the
unprotected nucleophilic site with desyl bromide, followed
by treatment with trifluoroacetic acid (TFA) to remove
the tert-butyl and t-BOC groups, provided desyl-caged
9-11 and 14 in reasonable yields and high purity
(Scheme 1). Phototriggers 9-11 were isolated as mix-
tures of diastereomers. It was anticipated that the esters
11 and 14 would be more stable to hydrolysis because of
the greater separation between the hydrolyzable ester
function and the electron-withdrawing ammonium group
in 9.13a
P h otoch em istr y. Photolysis of the desyl-caged amino
acids gave mixed results. The photochemistry of 11 and
14 is shown in Scheme 2. Table 1 gives the quantum
efficiencies measured for both the photolytic generation
of the furan 2 and the disappearance of the starting
desyl-caged amino acid. Photolysis of O-desyl GABA (14)
was very efficient (ΦDis ) 0.31) and produced free GABA
and 2. No other products were observed by HPLC.
Quantum efficiency measurements on the formation of
2 gave a photolysis efficiency of 0.13. Since the comple-
mentary products furan and GABA were the only two
products detected (vide infra), it was assumed that the
appearance efficiencies for both 2 and GABA would be
Resu lts
Syn th esis. O-Desyl amino acid esters and N-desyl-
glutamate were synthesized starting with the com-
mercially available, partially protected O-tert-butyl and
(13) (a) Wieboldt, R.; Gee, K. R.; Nui, L.; Ramesh, D.; Carpenter, B.
K.; Hess, G. P. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 8752-8756. (b)
Gee, K. R.; Wieboldt, R.; Hess, G. P. J . Am. Chem. Soc. 1994, 116,
8366-8367.
(14) A preliminary account of these results was presented at the
208th National Meeting of the American Chemical Society, Organic
Division, Washington, DC, August 22-26, 1994, Abstract #351.
(15) (a) Dalva, M. B.; Katz, L. C. Science 1994, 265, 255-258. (b)
Callaway, E. M.; Katz, L. C. Proc. Natl. Acad. Sci. U.S.A. 1993, 90,
7661-7665. (c) Katz, L. C.; Dalva, M. C. J . Neurosci. Methods 1994,
54, 205-218.
(16) Church, G; Ferland, J .-M.; Gauthier, J . Tetrahedron Lett. 1989,
30, 1901.

1230 J . Org. Chem., Vol. 61, No. 4, 1996
Gee et al.
Sch em e 2. P h otoch em istr y of O-Desyl-Ca ged
In the case of the N-desylglutamate (10), however, no
furan was produced upon photolysis. Similar results
have been observed by Sheehan7 in the photochemistry
of other desyl caged glutamates.
GABA a n d Glu ta m a te
In a preliminary study,17 γ-O-desyl glutamate (11) was
evaluated for its efficacy as a “caged” glutamate in the
photogeneration of glutamate at mammalian neurons.
Samples of hippocampal pyramidal neurons from ferrets
were bathed in 1mM γ-O-desyl glutamate in ASCF
bathing solution and then photolyzed with an 8 ms pulsed
50 mW UV argon laser. In a typical experiment, a 400
µm thick sample received 17 mW of radiation at 351-
364 nm. Central neuron response was monitored by the
whole cell patch clamp technique to determine the
photostimulated action potentials and the inward ion
currents. The responses of the glutamate receptors to
released glutamate were determined by measuring the
resulting inward ion currents at neurons synaptically
connected to those being activated and were not opti-
mized.15 Figure 1 shows a series of traces obtained upon
photolysis of 11.
Discu ssion
Our photochemical studies of the desyl-caged amino
acid neurotransmitters (Scheme 2) indicates that the
desyl esters of γ-O-desyl glutamate (11) and GABA (14)
may be useful as caged amino acid precursors for triggers
of biochemical processes. The photolysis quantum ef-
ficiencies were competitive with those reported for the
o-nitrobenzyl cages, and the rate of release of the free
amino acid is considerably higher, fast enough to be
useful for probing the initial stages of even the most rapid
biological processes, i.e., receptor activation and de-
sensitization.13
Ta ble 1. P h otolysis Qu a n tu m Efficien cies for
Disa p p ea r a n ce of 9-11 a n d 14 a n d th e Ap p ea r a n ce of 2
phototrigger
Φfuran
Φdisappear
R-O-desyl glutamate (9)
N-desylglutamate (10)
γ-O-desyl glutamate (11)
O-desyl GABA (14)
a
a
b
b
0.14 ( 0.03
0.14 ( 0.02
0.29 ( 0.02
0.31 ( 0.02
The photolytic behaviors of 11 and 14 reflect a recur-
ring pattern of reactivity in which the desyl moiety is
converted to 2-phenylbenzo[b]furan (2).1,3d,6,7a,b,8 The
process is highly efficient, varying only slightly from a
value of ca. 0.3. Bond cleavage occurs entirely through
the desyl group’s triplet state as established by linear
Stern-Volmer quenching of the reaction by 2-naphtha-
lenesulfonate and accords with our earlier observations
of the photochemical behavior of several other desyl
derivatives.1,3d,8 While the mechanistic details are not
complete at this stage of our studies, reasonable path-
ways would involve an electron transfer from the ketone
triplet either before or following homolysis of the carboxyl
group from the desyl protecting ligand to yield the
carboxylate-desyl R-keto carbocation ion pair as depicted
in Scheme 3.8,18,19 Cyclization to the resulting oxacylco-
pentadienyl cation followed by loss of a proton yields 2.
This scheme is in accord with the mechanistic picture
previously developed from investigations of photolysis of
benzylic phosphates.3d,19 The recent studies by Pincock
a
Benzil formed upon irradiation; no furan was detected from
b
10 and only traces were observed from 9. Not determined.
the same as had been the case in our previous studies
for the desyl series of phosphates.1 In preparative runs,
the liberated GABA was isolated and compared with an
authentic sample by analytical and preparative TLC.
In a separate series of experiments, it was shown that
photolysis of 14 could be readily quenched by sodium
2-naphthalenesulfonate. The triplet lifetime of ca. 10
ns was derived from the Stern-Volmer slope from the
quenching study, based on a rate constant for diffusion1
of 1.0 × 1010 M-1 s-1 estimated for rate of diffusion in
the binary solvent mixture of 1:1 aqueous buffer:dioxane
employed. From these data, a rate constant for release
of GABA was determined to be 1.4 (( 0.7) × 107 s-1
.
γ-O-Desyl glutamate (11) showed behavior similar to
14 upon photolysis, yielding a quantum efficiency for
disappearance of 0.29, a rate constant of kr ) 1.2 (( 0.4)
× 107 s-1, and an appearance efficiency for the furan of
0.14, based again on linear Stern-Volmer quenching by
2-naphthalenesulfonate.
(17) The absorbance of the desyl group is 150 M-1 cm-1 while that
of the R-carboxy-2-nitrobenzyl group is 180 M-1 cm-1 at 360 nm.13
Similar experiments had been performed in which photolysis of caged
R-(4,5-dimethoxy-2-nitrobenzyl) glutamate (R-DMNB glutamate) gen-
erated free glutamate. The magnitude of the response for free 11 was
approximately equal to that observed in experiments using R-CNB and
much more efficacious than the R-DMNB caged glutamate.15 A limiting
factor in these determinations remains the low absorptivity of 11 at
the concentrations employed for these experiments, however.19
(18) (a) Decosta, D. P.; Pincock, J . A. J . Am. Chem. Soc. 1993, 115,
2180. (b) Hilborn, J . W.; MacKnight, E.; Pincock, J . A.; Wedge, P J . J .
Am. Chem. Soc. 1994, 116, 3337. (c) Kim, J . M.; Pincock, J . A. Can. J .
Chem. 1995, 73, 885-895.
For the other two desyl-caged amino acids (R-O-desyl
and N-desylglutamates (9 and 10)), the photochemistry
diverges from pathways that lead to glutamate. The
major byproduct detected by HPLC was benzil, possibly
a product of the coupling of two benzoyl radicals formed
in a type 1 photocleavage of the desyl derivatives. This
was also a major pathway in one of the Sheehan desyl
series.7 Some furan 2 was also formed in the photolysis
of 9, indicative of photolysis to uncage the amino acid.
(19) Givens, R. S.; Matuszewski, B.; Athey, P. S.; Stoner, R. J . Am.
Chem. Soc. 1990, 112, 6016 and references therein.

Desyl Esters of Amino Acid Neurotransmitters
J . Org. Chem., Vol. 61, No. 4, 1996 1231
Sch em e 3. A Ra tion a l Mech a n ism for th e
P h otor elea se of “Ca ged ” Am in o Acid s fr om Desyl
Ester s
ate formation of benzil (15), either thermolytically or
photolytically, in solutions of 9 and 10 at present lacks
sufficient experimental explanation. The diketone is
formed by the coupling of two benzoyl radicals, which
could be generated from a photochemical type 1 cleavage,
the most expedient explanation for the photochemical
pathway.7 No products resulting from the accompanying
benzyl radicals were observed by HPLC upon photolysis
of 9 or 10. Conformational effects may also play a
significant role in determining whether photosolvolysis
or type I chemistry dominates in each individual case.
Work is continuing on this aspect of the photochemistry.
Regardless of the mechanism, it is clear that the
competing route to benzil renders the latter two probes
unsuitable for the photorelease of free amino acid due to
the competing absorption of the benzil, which is also a
known triplet quencher. A further drawback is that 9 is
hydrolytically unstable, rapidly degrading to free
glutamate and benzoin in neutral aqueous solution in the
absence of light.
The work described herein provides two examples of a
new class of desyl-caged phototriggers which release
biologically active amino acids very rapidly upon il-
lumination at 350 nm. The concomitant formation of
biologically inert 2-phenylbenzo[b]furan (2) is another
salutary feature of this reaction when contrasted with
the o-nitrobenzyl series. The results for N-desylglutamate
(10) and R-O-desyl glutamate (9) point to a lack of
generality of the desyl phototrigger series for amino acid
photorelease, however, which cautions against developing
strategies for other new desyl probes as phototriggers and
encourages further pursuit of other cage chromophores.
This work is in progress.
F igu r e 1. Scanning laser photostimulation of hippocampal
pyramidal neurons with γ-desyl glutamate (11).17 Arrows mark
the beginning of the 8 ms uncaging laser flash. (A) In current
clamp mode, uncaging desyl glutamate (1mM in ASCF bathing
solution) in ten consecutive trials at the same location evokes
an action potential from direct stimulation of the cell’s
glutamate receptors. The latency to peak is consistent at
approximately 35 ms. (B) Stimulation of another location
results in a subthreshold response. Three trials are shown.
As in A, the latency to peak is also approximately 35 ms.
Between cells latency to peak varied somewhat, ranging from
20 to 45 ms. (C) In voltage clamp mode, uncaging of the desyl
glutamate near to the record neuron results in a large inward
current, which causes the cells to escape clamp and fire an
action potential. The response to uncaged glutamate begins
as soon as the laser flash begins. Uncaging of γ-desyl glutamate
results in large currents which can generate action potentials,
but it also appears to have considerable intrinsic activity in
its caged form. Addition of this compound results in a large
(200-300%) increase of the holding current, and it was difficult
to record for neurons for a long period of time in the presence
of the caged compound.
et al.,18 on the photosolvolysis of benzyl and naphthyl-
methyl carboxylic esters provide convincing evidence in
favor of the pathway illustrated in Scheme 3 for photoly-
sis reactions carried out in hydroxylic solvents. While
the comparison of the triplet reactions of desyl esters with
the results on benzylic esters reported by Pincock must
account for the differences in the multiplicity and the
relative oxidation potentials of the benzylic and R-keto-
benzyl radicals, the striking similarity of the solvent
effect on these two very similar reactions invites a
parallel mechanistic interpretation. The almost immedi-
Exp er im en ta l Section
Syn th esis. Gen er a l. All reactions were carried out under
a dry argon atmosphere under subdued light. All reagents
and solvents were used as received. Flash chromatography
was performed according to the method of Still et al.20 All
melting points are uncorrected.
r-O-Desyl Glu t a m a t e (9). A solution of N-t-BOC-L-
glutamic acid, γ-tert-butyl ester (3, 239 mg, 0.788 mmol), desyl
(20) Still, W. C.; Kahn, M.; Mitra, A. J . Org. Chem. 1978, 43, 2923-
2925.

1232 J . Org. Chem., Vol. 61, No. 4, 1996
Gee et al.
bromide (204 mg, 0.741 mmol), and DBU (0.12 mL, 0.80 mmol)
in benzene (17 mL) was heated at reflux for 1 h. The reaction
solution was washed with water (1 × 25 mL) and brine (1 ×
20 mL), dried (sodium sulfate), and concentrated to a pale
brown oil. This oil was purified by flash chromatography
(ethyl acetate/chloroform 0% to 10%) to give R-O-desyl γ-tert-
butyl N-tert-butylglutamate (6) in a yield of 0.37 g (95%) as a
colorless solid mixture of diastereomers: mp 87-90 °C; 1H
NMR (CDCl3) 7.90 (m, 2H), 7.51 (m, 1H), 7.4 (m, 7H), 6.85
(two s, 1H), 5.13 (d, J ) 8.1 Hz, 1H), 4.44 (m, 1H), 2.5-2.3
(m, 2H), 2.2-1.9 (m, 2H), 1.45 (s, 9H), 1.42 (s, 9H). Anal. Calcd
for C28H35NO7: C, 67.59; H, 7.09; N, 2.81. Found: C, 67.23;
H, 7.05; N, 2.64.
131.61, 131.71, 131.81, 131.89, 132.02, 132.43, 132.92, 135.45,
136.35, 137.82, 176.52, 176.68; mass spectrum m/ e (relative
intensity) 342 (M+ + 1, 95), 230 (45), 212 (95), 195 (64); exact
mass calcd 341.1341, found 341.1346; UV λmax (ꢀ) 252 (9000),
300 (1000), 350 (200). Anal. Calcd for C19H19NO5‚0.5H2O: C,
65.13; H, 5.75; N, 4.00. Found: C, 64.97; H, 5.69; N, 3.88.
O-Desyl γ-Am in obu t yr ic Acid , Tr iflu or oa cet a t e Sa lt
(14). As for the synthesis of 6, N-t-BOC-γ-aminobutyric acid
(12, 500 mg, 2.46 mmol), desyl bromide (677 mg, 2.46 mmol),
and DBU (396 mg, 2.60 mmol) were heated at reflux in
benzene (25 mL) to give O-desyl N-t-BOC-γ-aminobutyrate (13)
in a yield of 0.94 g (96%) as a clear colorless oil: 1H NMR
(CDCl3) 7.92 (d, J ) 7.8 Hz, 2H), 7.4 (m, 8H), 6.86 (s, 1H), 4.7
(br s, 1H), 3.20 (d, J ) 5.9 Hz, 2H), 2.53 (m, 2H), 1.87 (dt, J )
6.9, 2.8 Hz, 2H), 1.4 (br s, 9H).
A solution of 6 (0.17 g, 0.34 mmol) and TFA (2.0 mL, 26
mmol) in dichloromethane (4 mL) was stirred at room tem-
perature for 3 h. The volatiles were removed in vacuo, and
the residue was dissolved in water (15 mL). The resulting
solution was frozen and lyophilized to give 9 as a mixture of
diastereomers, in a yield of 151 mg (97%) as a cream-colored
A solution of 13 (0.93 g, 2.3 mmol) and TFA (8.0 mL, 100
mmol) in chloroform (25 mL) was stirred overnight at room
temperature. The volatiles were removed in vacuo, and
benzene (1 × 15 mL) was evaporated from the residue, leaving
a colorless oil. This oil was purified by chromatography on
Sephadex LH-20, using water as eluant. The pure product
fractions were combined and lyophilized to give O-desyl
γ-aminobutyrate (O-desyl GABA, 14) as 0.78 g (81%) of a
1
hygroscopic powder: mp 89-92 °C; H NMR (D2O) 7.92 (t, J
) 8.7 Hz, 1H), 7.49 (m, 3H), 7.31 (m, 5H), 7.23 (two s, 1H),
4.40 (two t, J ) 7.2 H, 1H), 2.81 (t, J ) 7.3 Hz, 1H), 2.52 (m,
1H), 2.38 (q, J ) 7.1 Hz, 1H), 2.27 (d, J ) 7.0 Hz 1H); mass
spectrum m/ e (relative intensity) 342 (M+ - CF3C)2, 100), 195
(52), 167 (40); exact mass calcd 342.1345, found 342.1340.
Anal. Calcd for C21H20NO7F3‚H2O: C, 53.28; H, 4.68; N, 2.96.
Found: C, 53.28; H, 4.69; N, 2.77. Attempted purification via
reverse phase chromatography on Sephadex LH-20, with water
as the mobile phase, resulted in extensive hydrolysis to free
glutamic acid and benzoin, as determined by isolation and
coelution with authentic samples on TLC.
1
colorless hygroscopic powder: mp 82-84 °C; H NMR (D2O)
1.82 (m, 2H), 2.38 (m, 2H), 2.84 (t, 2H, J ) 7.92 Hz), 6.88 (s,
1H), 6.99 (m, 5H), 7.12 (t, 1H), 7.25 (d, 2H, J ) 7.23), 7.69 (d,
2H, 7.23); 13C NMR (D2O) 24.6, 32.9, 41.1, 80.6, 131.1 d, 131.2
d, 131.3 d, 131.7 d, 132.0, 135.3, 136.1, 136.7, 175.9, 198.8.
Anal. Calcd for C20H20NO5F3: C, 58.39; H, 4.90; N, 3.40.
Found: C, 58.13; H, 4.75; N, 3.39.
P h otoch em istr y. Gen er a l. Dioxane and acetonitrile
were used without further purification. All water was distilled
and passed through a Nanopure deionizing system. Phosphate
buffer for analytical HPLC was prepared using potassium
hydrogen phosphate to afford a solution of 0.05 M KH2PO4 with
an adjusted pH of 2.5. The HPLC system consisted of two
HPLC pumps, a controller, an injector equipped with a 20 µL
loop, a Econsosil C18 FU 25.0 × 4.6 cm column, a spectropho-
tometric absorbance detector set at 280 nm, and a recording
integrator.
P h otolysis of γ-O-Desyl Glu ta m a te (11). In a 50 mL
volumetric flask, 500 mg of 11 and 38 mg of p-anisic acid were
dissolved in 50 mL of 50:50 acetonitrile:distilled water. Ali-
quots (5 mL) were placed in Pyrex phototubes. The solutions
were degassed for 20 min at 0 °C. The solution was then
photolyzed in a Rayonet reactor with four RPR 350 nm lamps
for 32 min. Aliquots (0.05 mL) were removed at the following
time intervals: 0, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 5.0, 8.0, 12.0, 17.0,
22.0, 27.0, 32.0 min. A 50 µL sample of each aliquot was
diluted to 450 µL and then assayed by HPLC using the
following conditions: 10 µL loop, mobile phase; 30% aceto-
nitrile:70% 50 mM NH4OAc pH 4.5 for 2 min to 75% aceto-
nitrile at 20 min, held at 75% acetonitrile from 20-25 min
and then to 30% acetonitrile for 30 min. The product was
identified by coinjection with a known sample of 2-phenyl-
benzo[b]furan (2) by HPLC. Relative response factors for 11
and the product 2-phenylbenzo[b]furan were determined from
standard solutions of known concentrations. p-Anisic acid was
used as an internal standard. Another unidentified product
formed as noted by HPLC. The yield of the minor product was
estimated to be less than 10% of the furan. Light output was
determined by the potassium ferrioxalate method.21
P h otolysis of O-Desyl γ-Am m on iobu tyr a te, Tr iflu or o-
a ceta te Sa lt (14). In a 50 mL volumetric flask, 452 mg of 14
and 38 mg of p-anisic acid were dissolved in 50 mL of 50:50
acetonitrile:distilled water. Aliquots (5 mL) were taken and
placed in Pyrex phototubes. The solutions were degassed for
20 min at 0 °C and were then photolyzed in a Rayonet reactor
with four RPR 350 nm lamps for 32 min. Aliquots (0.05 mL)
were removed at the following intervals: 0, 0.5, 1.0, 1.5, 2.0,
2.5, 3.0, 5.0, 8.0, 12.0, 17.0, 22.0, 27.0, 32.0 min. A 50 µL
sample of each aliquot was diluted to 450 µL with acetonitrile.
These samples were then assayed by HPLC using the following
conditions: 10 µL loop, mobile phase 55% acetonitrile: 45%
of a solution of 50 mM NH4OAc pH 4.5 for 8 min to 75%
acetonitrile at 15 min, held at 75% acetonitrile from 15 to 20
N-Desylglu ta m ic Acid , Tr iflu or oa ceta te Sa lt (10).
A
solution of di-tert-butyl L-glutamate hydrochloride (4, 2.03 g,
6.86 mmol), desyl bromide (1.89 g, 6.86 mmol), and diisopro-
pylethylamine (1.81 g, 14.0 mmol) in acetonitrile (30 mL) was
stirred overnight at room temperature. Concentration gave
a pale brown solid, which was purified by flash chromatogra-
phy (ethyl acetate/chloroform 0% to 5%) to give a mixture of
diastereomers of di-tert-butyl N-desylglutamate (7) in a yield
of 1.77 g (57%) as a clear pale brown oil: 1H NMR (CDCl3)
8.10 (d, J ) 7.5 Hz, 1H), 7.98 (d, J ) 7.2 Hz, 1H), 7.8 (m, 2H),
7.45 (m, 6H), 4.05 (m, 1H), 2.56 (m, 1H), 2.3 (m, 2H), 2.2 (m,
2H), 1.50 (s, 9H), 1.37 (s, 9H).
A solution of 7 (1.75 g, 3.86 mmol) and TFA (20 mL) in
dichloromethane (80 mL) was stirred at room temperature
overnight. The volatiles were removed in vacuo, and benzene
(2 × 20 mL) was evaporated to yield a residue as a pale brown
solid foam. Half of this crude product was purified on
Sephadex LH-20, using water as eluant, to give a mixture of
diastereomers of N-desylglutamate (10) in a yield of 0.44 g
(50%) as a hygroscopic colorless powder after lyophilization:
1
mp 110-114 °C dec; H NMR (D2O) 7.98 (d, J ) 7.8 Hz, 2H),
7.65 (t, J ) 7.4 Hz, 1H), 7.51 (m, 7H), 6.29 (s, 1H), 3.94 (t, J
) 6.4 Hz, 1H), 2.8-2.1 (m, 4H). Anal. Calcd for C21H20NO7-
F3‚H2O: C, 53.28; H, 4.68; N, 2.96. Found: C, 53.42; H, 4.66;
N, 3.07.
γ-O-Desyl Glu ta m a te (11). As in the procedure to form
6, a solution of desyl bromide (605 mg, 2.2 mmol), N-t-BOC-
glutamic acid, R-tert-butyl ester (672 mg, 2.2 mmol), and DBU
(0.35 mL, 2.3 mmol) was heated at reflux in benzene (25 mL)
and stirred overnight to give γ-O-desyl tert-butyl N-t-BOC-
glutamate (8) in a yield of 1.06 g (96%) as a colorless oil: 1H
NMR (CDCl3) 7.92 (d, J ) 8.1 Hz, 2H), 7.51 (t, J ) 7.4 Hz,
1H), 7.47-7.32 (m, 7H), 6.86 (s, 1H), 5.12 (dd, J ) 13.6, 7.3
Hz, 1H), 4.5 (br s, 1H), 2.6 (m, 2H), 2.22 (m, 1H), 1.97 (m,
1H), 1.45 (s, 9H), 1.42 (two s, 9H).
As in the procedure for synthesis of 9, γ-O-desyl tert-butyl
N-t-BOC-glutamate (8) (1.06 g, 2.13 mmol) and TFA (2.0 mL)
were stirred at room temperature in dichloromethane (5 mL)
and gave, after LH-20 purification, γ-O-desyl glutamate (11)
in a yield of 0.51 g (52%) as a hygroscopic colorless powder;
the trifluoroacetic acid was removed from the product salt
1
during lyophilization: mp 118-120 °C dec; H NMR (CDCl3)
2.18 (m, 2H), 2.71 (m, 2H), 3.76 (q, 1H), 7.13 (s, 1H), 7.39-
7.49 (m, 7H), 7.62 (t, 1H, J ) 7.33 Hz), 7.98 (d, 2H, J ) 7.35
Hz); 13C NMR (D2O) 28.39, 32.654, 56.786, 79.04, 130.95,

Desyl Esters of Amino Acid Neurotransmitters
J . Org. Chem., Vol. 61, No. 4, 1996 1233
min and then to 55% acetonitrile at 25 min. Light output was
determined by the potassium ferrioxalate method.21 The
product was identified by coinjection with known samples of
2-phenylbenzo[b]furan by HPLC. Relative response factors for
2 and 14 were determined from standard solutions of known
concentrations. p-Anisic acid was used as an internal stand-
ard. The results are given in Table 1.
placed in Pyrex phototubes. To two of the phototubes was
added enough sodium 2-naphthalenesulfonate to make 0.02
and 0.058 M solutions of the quencher. The solutions were
degassed for 20 min at 0 °C and then photolyzed in a Rayonet
reactor with four RPR 350 nm lamps for 17 min. Aliquots of
0.05 mL were removed at the following times: 0, 12, and 17
min. A 50 µL sample of each aliquot was diluted to 450 µL
which was assayed by HPLC using the following conditions:
10 µL loop, mobile phase 30% acetonitrile:70% of a solution of
50 mM NH4OAc at pH 4.5 for 2 min to 75% acetonitrile at 20
min, held at 75% acetonitrile from 20 to 25 min and then to
30% acetonitrile for 30 min. A rate of disappearance of 1.2 (
0.4 × 107 s-1 was determined from the slope of the Stern-
Volmer concentration dependence, assuming a diffusion rate
for triplet quenching of 1.1 × 1010 M-1 s-1 in 1:1 CH3CN:H2O.1
Relative response factors for 14 and 2-phenylbenzo[b]furan
were determined from standard solutions of known concentra-
tions. p-Anisic acid was used as the internal standard.
Qu en ch in g Stu d ies of 11. In a 50 mL volumetric flask,
500 mg of 11 and 38 mg of p-anisic acid were dissolved in 50
mL of 1:1 acetonitrile:distilled water. Aliquots (3 × 5 mL) were
placed in Pyrex phototubes. To two of the phototubes was
added enough sodium 2-naphthalenesulfonate to make 0.021
and 0.061 M solutions of the quencher. The solutions were
degassed for 20 min at 0 °C and then photolyzed in a Rayonet
reactor with four RPR 350 nm lamps for 17 min. Aliquots of
0.05 mL were removed at the following times: 0, 12, and 17
min. A 50 µL sample of each aliquot was diluted to 450 µL
which was assayed by HPLC using the following conditions:
10 µL loop, mobile phase 30% acetonitrile:70% of a solution of
50 mM NH4OAc at pH 4.5 for 2 min to 75% acetonitrile at 20
min, held at 75% acetonitrile from 20 to 25 min and then to
30% acetonitrile for 30 min. A rate of disappearance of 1.4 (
0.7 × 107 s-1 was determined from the slope of the Stern-
Volmer concentration dependence of the efficiencies, assuming
a diffusion-controlled rate constant for triplet quencher of 1.1
× 1010 M-1 s-1 in 1:1 CH3CN/H2O.1 Relative response factors
for 11 and 2-phenylbenzo[b]furan were determined from
standard solutions of known concentrations. p-Anisic acid was
used as the internal standard.
Ack n ow led gm en t. This material is based upon
work supported by the National Science Foundation
under Grant No. NSF/OSR-9255223. Additional sup-
port from PRF Grant No. 17541-AC4 was also provided.
We thank Matthew Dalva and Professor Lawrence Katz
in the Department of Neurobiology, Duke University
Medical Center, for performing the photolysis experi-
ments of 11 in brain tissue slices.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
7, 8, and 13 (3 pages). This material is contained in libraries
on microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
Qu en ch in g Stu d ies of 14. In a 50 mL volumetric flask,
452 mg of 14 and 38 mg of p-anisic acid were dissolved in 50
mL of 1:1 acetonitrile:distilled water. Aliquots (3 × 5 mL) were
(21) Hatchard, C. G.; Parker, C. A. Proc. R. Soc. London, Ser. 4 1956,
235, 518-526.
J O951635X