New photoactivated protecting groups. 7. p-Hydroxyphenacyl: A phototrigger for excitatory amino acids and peptides

Full text of DOI:10.1021/ja971331n

J. Am. Chem. Soc. 1997, 119, 8369-8370
8369
from 1a or b gave, in addition to major yields of the cor-
responding acetophenones (2), rearranged p-methoxyphenylac-
etate and p-hydroxyphenylacetate (3a,b), respectively (eq 1).
Earlier, we confirmed that rearrangement was also a major
New Photoactivated Protecting Groups. 7.
p-Hydroxyphenacyl: A Phototrigger for Excitatory
Amino Acids and Peptides1
Richard S. Givens,* Andreas Jung, Chan-Ho Park,
Jo¨rg Weber, and Wenzel Bartlett
Department of Chemistry, UniVersity of Kansas
Lawrence, Kansas 66045
ReceiVed April 28, 1997
–
We report here our results for p-hydroxyphenacyl as the
phototrigger for the excitatory amino acids L-glutamate and
GABA and for a model peptide, the dipeptide, ala-ala. Our
initial studies in this arena began with the photorelease of
cAMP,2,3 L-glutamate,4 and GABA4 from their benzoin (desyl)
esters which demonstrated that release occurred much faster with
rate constants of 107-108 s-1 than are observed from the
archetypical o-nitrobenzyl analogues6-8 which typically release
pathway for p-methoxyphenacyl phosphates1,3,5 in alcohol
solvents3 and more recently the exclusive pathway for p-
hydroxyphenacyl phosphate (Pi) and ATP1,5 in aqueous media.
These results warranted further elaboration of this R-keto
phototrigger for excitatory amino acids, e.g., L-glutamate and
γ-aminobutyric acid (GABA) that are frequently employed
by others for studies of neurotransmission,8a for brain neuronal
mapping,8b and the temporal and spatial studies on neuronal
stimulation.8c-e
substrates at rates of 1-100 s-1
. The photorelease of the
substrate from desyl ester occurs with a high efficiency (∼0.30)
and is free of complicating side reactions. However, these
reactions are plagued by low aqueous solubility and the
incorporation of an additional chiral center with these esters
and a UV absorbing photoproduct. Our search for a phototrigger
that was free of these limitations resulted in the design of
p-hydroxyphenacyl moiety, and we subsequently demonstrated
that it functioned as a phototrigger for ATP.1,5
Scheme 1
Since protecting groups for amino acids, peptides, and
proteins have been extensively studied and developed, they have
been applied in biochemistry and physiology6 and in synthesis.7
The o-nitrobenzyl protecting group has been thoroughly ex-
ploited for the biological and physiological applications and
remains the principal component employed in the field of
“caged” compounds.8 However, researchers have begun focus-
sing on designing cages with higher efficiencies and better
absorption properties.9 With the increased interest in the
mechanistic and kinetic details of the molecular events sur-
rounding substrate-activated biochemical processes, attention is
now shifting toward designing more rapid photoreleasing cages
or phototriggers.1-5,9
Previous studies by Anderson and Reese,10 Sheehan et al.,11
and Epstein and Garrossian12 had clearly demonstrated that the
photochemistry of the p-methoxyphenacyl derivatives (1a,c,d)
was effective in releasing a variety of nucleofugal groups (e.g.,
RCO2-, Cl-, and (RO)2 PO2-) in organic solvents. Anderson
and Reese further demonstrated that the release of chloride10
Possible precursors for the synthesis of these “caged” amino
acid esters are the bromoketone 1e or the hydroxyketone 1f
which are readily available from p-hydroxyacetophenone.13 We
chose to synthesize the caged amino acids and dipeptide through
an SN2 displacement of the ketobromide 1e with the N-protected
amino acid catalyzed by DBU followed by deprotection with
TFA.14,15 The resulting esters 4a-c displayed excellent stability
in water, D2O, and Ringers solutions, showing no hydrolysis
after 24 h at room temperature. In TRIS, 4c hydrolyzed slowly
with a half-life of 214 min, whereas the GABA and glutamate
esters were stable.
Irradiation of buffered solutions of O-p-hydroxyphenacyl
GABA (4a), γ-O-p-hydroxyphenacyl L-glutamate (4b), and the
O-p-hydroxyphenacyl ala-ala (4c) resulted in the release of the
amino acid or dipeptide accompanied by the quantitative
rearrangement of the phenacyl moiety to p-hydroxyphenylacetic
(1) For paper 6, see: Park, C-H.; Givens, R. S. J. Am. Chem. Soc. 1997,
119, 2453-2463.
(2) Givens, R. S.; Athey, P. S.; Kueper, III, L. W.; Matuszewski, B.;
Xue, J-y. J. Am. Chem. Soc. 1992, 114, 8708-8709.
(3) Givens, R. S.; Athey, P. S.; Kueper, III, L. W.; Matuszewski, B.;
Xue, J.-y.; Fister, T. J. Am. Chem. Soc. 1993, 115, 6001-6010.
(4) Gee, K. R.; Kueper, III, L. W.; Barnes, J.; Givens, R. S. J. Org.
Chem. 1996, 61, 1228-1233.
(5) Givens, R. S.; Park, C.-H. Tetrahedron Lett. 1996, 37, 6259-6262.
(6) (a) Adams, S. R.; Tsien, R. Y. Am. ReV. Physiol. 1993, 55, 755-
783. (b) Hess, G. P. Biochemistry 1993, 32, 989-1000.
(7) Pillai, V. N. R. Synthesis 1980, 1-26.
(8) (a) Katz, L. C.; Dalva, M. B. J. Neurosci. Methods 1994, 54, 205-
218. (b) Dalva, M. B.; Katz, L. C. Science 1994, 265, 255-258. (c) Wiebolt,
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. (d) Wiebolt, R.; Ramesh, D.;
Carpenter, B. K.; Hess, G. P. Biochemistry 1994, 33, 1526-1533. (e)
Callaway, E. M.; Katz, L. C. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 7661-
7665.
(13) The bromide (1e) was obtained in 47% yield by treating p-
hydroxyacetophenone (2b) with cuprous bromide in refluxing ethyl acetate:
CHCl3. The bromide was converted to the alcohol (1f) in 49% yield by
treatment with formic acid and DBU in CH2Cl2 followed by hydrolysis of
the formate ester with basic (NaOH) methanol (mp 165-167 °C).
(14) The protected N-BOC GABA, N-BOC R-O-t-Bu L-glutamate, and
N-BOC ala-ala carboxylic acids were reacted with p-hydroxyphenacyl
bromide (1e) in benzene with DBU followed by treatment with TFA to
give the p-hydroxyphenacyl protected amino acids 4a and 4b or dipeptide
4c in yields of 70-85%. Details will be provided in our full paper.
(15) The esters of R-amino acids glycine, tyrosine, and L-glutamate were
also synthesized but were found to be hydrolytically unstable, most likely
due to the inductive influence of the protonated R-amino group at neutral
pH.4 The dipeptide 4c was synthesized for the intended purpose of
demonstrating the hydrolytic stability of a less basic peptide nitrogen.
(9) (a) Corrie, J. E. T.; Trentham, D. R. J. Chem. Soc., Perkin Trans. 1
1992, 2409-2417. (b) Furuta, T.; Torigai, H.; Sugimoto, M.; Iwamura, M.
J. Org. Chem. 1995, 60, 3953-3956. (c) Pirrung, M. C.; Shuey, S. W. J.
Org. Chem. 1994, 59, 3890-3897. (d) Niu, L.; Wieboldt, R.; Ramesh, D.;
Carpenter, B. K.; Hess, G. P. Biochemistry 1996, 35, 8136-8142.
(10) Anderson, J. C.; Reese, C. B. Tetrahedron Lett. 1962, 1, 1-4.
(11) (a) Sheehan, J. C.; Wilson, R. M. J. Am. Chem. Soc. 1964, 86, 5277.
(b) Sheehan, J. C.; Wilson, R. M.; Oxford, A. W. J. Am. Chem. Soc. 1971,
93, 7722. (c) Sheehan, J. C.; Umezawa, K. J. Org. Chem. 1973, 38, 3371.
(12) Epstein, W. W.; Garrossian, M. J. Chem. Soc., Chem. Commun.
1987, 532-533.
S0002-7863(97)01331-0 CCC: $14.00 © 1997 American Chemical Society

8370 J. Am. Chem. Soc., Vol. 119, No. 35, 1997
Communications to the Editor
of rat cortex through thin fused-silica fiber optics (Figure 1)
which demonstrate the degree of temporal and spatial control
provided by the release of L-glutamate from phototrigger 4b.
As suggested for the photorelease of phosphates, amino acid
release probably passes through a spirodienedione 5 or its
equivalent (Scheme 1). The reaction proceeds from the phena-
cyl triplet18 in an overall efficiency for disappearance of 4a-c
of 0.35, 0.12, and 0.27, respectively, determined by HPLC and
NMR. The formation efficiencies of p-hydroxyphenylacetic
acid were 0.08 and 0.24 for 4b and 4c, respectively. The yields
of the amino acids were quantitative within the detection limits
of NMR and HPLC and matched the appearance efficiencies
for 6, i.e., 0.08 (glu) and 0.25 (ala-ala). A rate constant of 7 ×
107 s-1 was determined for the release of the dipeptide 4c from
Stern-Volmer quenching with sodium 2-naphthalenesulfonate18
and agrees well with the observed kr values obtained for the
corresponding phosphates.3,19
The rate constants for release of a carboxylate moiety are
evidence of an excited state fragmentation process, possibly from
a π* f σ* electron transfer to the ester bond. Saveant et al20
recently correlated the rates of fragmentation of a series of
substituted phenacyl bromides, esters, and ethers from their
radical anions with standard free energies of reaction and found
that the p-methoxyphenacyl radical anion was among the most
reactive in the series. They attributed the enhanced reactivity
to substituent assisted electron transfer from the phenacyl radical
anion to the σ* orbital of the carbon-oxygen bond of the ester,
inducing rapid fragmentation of the nucleofuge from the
phenacyl moiety. In our studies, the transfer of the π* electron
of the photoexcited phenacyl moiety to the σ* orbital of the
amino acid ester bond could be envisaged as initiating the rapid
fragmentation of the phototrigger.
A second plausible mechanistic pathway involving initial
homolytic fragmentation of the ester CO2-C bond from the
excited triplet of the phenacyl chromophore followed by rapid
electron transfer21 has been suggested for the singlet state
photochemistry of benzyl acetates,21 pivalates,21 and phos-
phates,22 in polar, hydroxylic solvents and has been extended
to desyl amino acids8 and to p-hydroxyphenacyl and desyl
phosphates.1,7,9 However, additional studies are required to
resolve this dual mechanistic dilemma. These are in progress
as well as further development of new phototriggers.
Acknowledgment. We thank the NSF (NSF/05R-9255223) and PRF
(17541-ACS) for partial support of this research. W.B. also thanks the
NSF-REU program (CHE-9322061) and A.J. and J.W. thank University
of Kansas College of Liberal Arts and Sciences for support. Dedicated
to Professor Dietrich Do¨pp on the occassion of his 60th birthday.
Figure 1. Effect of irradiated rat cortical brain slices perfused with
p-hydroxyphenacyl L-glutamate (4b). (a, top) Inward currents of 100
to 1500 pA were evoked in whole cell voltage-clamped at -70 mV
(1) in the absence of 4b (trace 1; 10 ms exposure time) and in the
presence of with 50 µM concentration of 4b (traces 2-5; at shutter
openings of 3, 5, 7, and 10 ms exposure times, respectively. The cells
generated three action potentials when illuminated with 10 ms flashes.
(b, middle) Voltages obtained with current-clamp whole-cell patch
recordings of a CA1 pyramidal neuron in a hippocampal slice
superfused with 100 µM 4b in the presence of 2 µM TTX at the
indicated shutter openings. Above 7 ms shutter opening, a regenerative
Ca2+ spike generates a significantly greater response. (c, bottom)
Response monitored by fluorescence of FURA-2 labeled cortical slice,
neonatal rat cells to intracellular Ca2+ increases following a 10 ms flash
in the presence of 200 µM concentration of 4b. Images in upper row
represent video frames taken 5 s before, 0 s, and 20 s after flash
photolysis monitored at 385 nm. The lower graph shows the mean
fluorescence changes occurring near the tip of the optic fiber (frame)
expressed as ∆F/F (%).
JA971331N
(16) The formation of 1-(2,4′-dihydroxy)phenylethanone (1f, <3%) was
observed during the photolysis of 4c in aqueous solution and in Ringers
which was attributed to a slow background hydrolysis which occurred at
the slightly higher temperature (∼35 °C) of the photoreaction chamber
during the photolysis. The hydrolysis product also formed at a comparable
rate for controls with samples in the buffered media in the chamber but
protected from the incident light.
(17) We thank Professor L. Katz and Dr. K. Kandler, Howard Hughes
Medical Institute at the Duke University Medical School, for carrying out
these experiments. Details on the experimental procedures will appear in
their publications of the applications of the glutamate phototrigger.
(18) For example, Stern-Volmer quenching of 4c by 2-naphthalene-
sulfonic acid established the triplet state as the photoactive precursor to
the release of the dipeptide. The KSV measure in D2O was 30 ( 3 M-1
which yields a rate constant of 7.0 ( 1.0 × 107 s-1 for release of ala-ala.
(19) Wirz, J. A.; Hellrung, B.; Givens, R. S. Unpublished results. The
decay of the triplet for the diethyl phosphate model, which is equal to the
rate of formation of the benzo[b]furan, is complete in 50 ns. The lifetime
of the triplet is 12.5 ns and is readily quenchable by naphthalene.
(20) (a) Andrieux, C. P.; Save´ant, J.-M.; Tallec, A.; Tardivel, R.; Tardy,
C. J. Am. Chem. Soc. 1996, 118, 9788-9789. (b) For additional confirmation
of Save´ant’s results, see: Anderson, M. L.; Long, W.; Wayner, D. D. M.
J. Am. Chem. Soc. 1997, 119, 6590-6595.
acid (6, Scheme 1).9,16 As evidence of their efficacy as
phototriggers for the release of excitatory amino acid substrates,
Katz and Kandler17 performed patch-clamp and calcium-
sensitive dye studies by Hg flash irradiation of perfused slices
(21) Hilborn, J. W.; MacKnight, E.; Pincock, J. A.; Wedge, P. J. J. Am.
Chem. Soc. 1994, 116, 3337-3346.
(22) Givens, R. S.; Matuszewski, B.; Athey, P. S.; Stoner, M. R. J. Am.
Chem. Soc. 1990, 112, 6016-6021.