New photoactivated protecting groups. 6. p-Hydroxyphenacyl: A phototrigger for chemical and biochemical probes

Article abstract of DOI:10.1021/ja9635589

p-Hydroxyphenacyl, a new photoactive, aqueous soluble protecting group is proposed as a second generation α-keto 'cage' reagent, a phototrigger for the efficient, rapid release of bioactive phosphates, e.g., inorganic phosphate (P(i)) and ATP (Givens, R.S.; Park, C.-H. Tetrahedron Lett. 1996, 37, 6259-6262). p-Hydroxyphenacyl esters 6c and 7 trigger the release of P(i) and ATP when irradiated at wavelengths between 300-350 nm also yielding p-hydroxyphenylacetic acid (8) from the rearrangement of the intermediate α-keto carbocation or its equivalent. In contrast, unsubstituted and m-substituted phenacyl esters yield only photoreduction and radical coupling products and none of the rearrangement product. Quantum efficiencies of 0.38 ± 0.04 were measured for the disappearance of the p-hydroxyphenacyl phosphate esters 6c and 7; the appearance efficiencies for 8 and ATP were 0.30 ± 0.03. Rates of release of ~10⁷ s^-1 or better are observed for these esters with only minor variations in efficiencies and rate constants between these two examples of the p-hydroxyphenacyl phototrigger. Just as was found for the desyl 'cage' series reported earlier (Givens, R.S.; Athey, P.S.; Kueper, L.W., III; Matuszewski, B.; Xue, J.-y.; Fister, T.J. Am. Chem. Soc. 1993, 115, 6001-6010), the p-hydroxyphenacyl derivatives react via their triplet states. Amino substituents, i.e., p-amino-, p-acetamido-, and p-(carbomethoxyamino)phenacyl phosphates 6f-h, were also investigated, but these analogues proved to be inferior as phototriggers when compared with p-hydroxyphenacyl.

Full text of DOI:10.1021/ja9635589

J. Am. Chem. Soc. 1997, 119, 2453-2463
2453
New Photoactivated Protecting Groups. 6. p-Hydroxyphenacyl:
A Phototrigger for Chemical and Biochemical Probes1,2
Chan-Ho Park and Richard S. Givens*
Contribution from the Department of Chemistry, UniVersity of Kansas, Lawrence, KS 66045
ReceiVed October 11, 1996X
Abstract: p-Hydroxyphenacyl, a new photoactive, aqueous soluble protecting group is proposed as a second generation
R-keto “cage” reagent, a phototrigger for the efficient, rapid release of bioactive phosphates, e.g., inorganic phosphate
(Pi) and ATP (Givens, R. S.; Park, C.-H. Tetrahedron Lett. 1996, 37, 6259-6262). p-Hydroxyphenacyl esters 6c
and 7 trigger the release of Pi and ATP when irradiated at wavelengths between 300-350 nm also yielding
p-hydroxyphenylacetic acid (8) from the rearrangement of the intermediate R-keto carbocation or its equivalent. In
contrast, unsubstituted and m-substituted phenacyl esters yield only photoreduction and radical coupling products
and none of the rearrangement product. Quantum efficiencies of 0.38 ( 0.04 were measured for the disappearance
of the p-hydroxyphenacyl phosphate esters 6c and 7; the appearance efficiencies for 8 and ATP were 0.30 ( 0.03.
Rates of release of ∼107 s-1 or better are observed for these esters with only minor variations in efficiencies and rate
constants between these two examples of the p-hydroxyphenacyl phototrigger. Just as was found for the desyl
“cage” series reported earlier (Givens, R. S.; Athey, P. S.; Kueper, L. W., III; Matuszewski, B.; Xue, J.-y.; Fister,
T. J. Am. Chem. Soc. 1993, 115, 6001-6010), the p-hydroxyphenacyl derivatives react via their triplet states. Amino
substituents, i.e., p-amino-, p-acetamido-, and p-(carbomethoxyamino)phenacyl phosphates 6f-h, were also
investigated, but these analogues proved to be inferior as phototriggers when compared with p-hydroxyphenacyl.
Photoactivated release of biochemical substrates has received
considerable attention since Engels and Schlaeger’s3 first
reported that the o-nitrobenzyl (2-nitrobenzyl or 2NB) ester of
cAMP released free cAMP upon photolysis, thus providing the
potential for controlling both the temporal and spatial release
of the nucleotide in viscous or structured environments.
Subsequently, Kaplan, Forbush, and Hoffman4 reported the
efficient photorelease of ATP and inorganic phosphate (Pi) from
the corresponding 2NB and the o-nitrophenethyl (1-(2-nitro-
phenyl)ethyl or 2NPE) esters (eq 1). These authors assigned
the general term “cage” to the 2-nitrobenzyl group to emphasize
its application as a photoactivated protecting group. Previous
studies5 had clearly demonstrated that the 2-nitrobenzyl group
could serve the function of a photoprotecting group in synthesis
(1)
of phosphates, amines, alcohols, carbamates, and carboxylic
acids. However, the reports of Engels and Schlaeger3 and
Kaplan et al.4 were among the first to demonstrate the
photorelease of a nucleotide from a “caged” substrate which
then could elicit a biochemical response, thus opening the way
for future exploitation of photoprotecting groups for biological
investigations. The release of biochemical substrates from
2-nitrobenzyl esters of nucleotides, oligopeptides, proteins, and
amino acids has since rapidly gained wide use among biochem-
ists and physiologists.6-14
X Abstract published in AdVance ACS Abstracts, February 15, 1997.
(1) For a recent study on R-keto “caged” amino acids, see: Gee, K. R.;
Kueper, L. W., III; Barnes, J.; Dudley, G.; Givens, R. S. J. Org. Chem.
1996, 61, 1228-1233.
(6) See reviews: (a) Lester, H. A.; Nerbonne, J. M. Annu. ReV. Biophys.
Bioeng. 1982, 151-175. (b) Nerbonne, J. M. Opt. Methods Cell Physiol.
1986, 40 (Chapter 24), 418-445. (c) Gurney, A. M.; Lester, H. A. Physiol.
ReV. 1987, 67, 583-617. (d) Homsher, E.; Millar, N. C. Annu. ReV. Physiol.
1990, 52, 875 -896. (e) Somlyo, A. P.; Somlyo, A. V. Annu. ReV. Physiol.
1990, 52, 857-874. (f) Corrie, J. E. T.; Katayama, Y.; Reid, G. P.; Anson,
M.; Trentham, D. R. Philos. Trans. R. Soc. London, Ser. A. 1992, 340,
233-243. (g) Adams, S. R.; Tsien, R. Y. Annu. ReV. Physiol. 1993, 5,
755-784.
(7) (a) McCray, J. A.; Trentham, D. R. Annu. ReV. Biophys. Chem. 1989,
18, 239-70. (b) Corrie, J. E. T.; Reid, G. P.; Trentham, D. R., Mazid, M.
A.; Hursthouse, M. B. J. Chem. Soc., Perkin Trans 1 1992, 1015-1019.
(8) Corrie, J. E. T.; Trentham, D. R. Caged nucleotides and neurotrans-
mitters. In Biological Applications of Photochemical Switches; Morrison,
H., Ed.; J. Wiley and Sons: New York, 1994; pp 243-305.
(2) For previous studies on R-keto phosphates, see: (a) Givens, R. S.,
Kueper, L. W., III Chem. ReV. 1993, 93, 55-66. (b) Givens, R. S.; Athey,
P. S.; Kueper, L. W., III; Matuszewski, B.; Xue, J.-y. J. Am. Chem. Soc.
1992, 114, 8708-8709. (c) Givens, R. S.; Athey, P. S.; Kueper, L. W., III;
Matuszewski, B.; Xue, J.-y.; Fister, T. J. Am. Chem. Soc. 1993, 115, 6001-
6010. (d) Givens, R. S.; Matuszewski, B. J. Am. Chem. Soc. 1984, 106,
6860-6861.
(3) Engels, J.; Schlaeger, E.-J. J. Med. Chem. 1977, 20, 907-911.
(4) Kaplan, J. H.; Forbush, G., III; Hoffman, J. F. Biochemistry 1978,
17, 1929-1935.
(5) Pillai, V. N. R. Synthesis 1980, 1-26.
S0002-7863(96)03558-5 CCC: $14.00 © 1997 American Chemical Society

2454 J. Am. Chem. Soc., Vol. 119, No. 10, 1997
Park and GiVens
The major advantages of a photoactivated process for the
release of nucleotides, as summarized in many reviews,6-8 are
control of the spatial and temporal distribution of a substrate
and the ability to effect significant increases in the concentration
of a released substrate. An additional factor of significance is
the convenience of a “noninvasive” photochemical activation
process.
The substantial number of studies that apply some variation
of the 2-nitrobenzyl strategy6-8 for the investigation of bio-
chemical mechanisms have established the 2NB cage as the
archetypal chromophore. The reactions are generally unaffected
by modest changes in the temperature and media, occur with
high quantum efficiencies (∼0.1-0.6), and are activated by UV-
vis wavelengths greater than 300 nm. While 2NB and its
derivatives would appear to be ideal cages, the photoactivated
release of a substrate from 2-nitrobenzyl cages typically is
limited to the millisecond to second time regime (kr ) 1-103
s-1; see eq 1).
Nevertheless, the effectiveness of the 2NB strategy to probe
biochemical mechanisms has been abundantly demonstrated3,4,6-13
over the past decade, e.g., the mechanism of release of inorganic
phosphate in skeletal muscle,9 the action of actomyosin in
muscle contraction induced by ATP,10 the role of cAMP in the
relaxation of distal muscle,11 and the action of Ca-ATPase in
the sarcoplasmic reticulum in active calcium transport during
ATP hydrolysis.12 In a recent example, Corrie and Trentham
et al.13 reported the use of rapid scan time-resolved Fourier
transform infrared (TR-FTIR) spectroscopy to measure the rate
constant of ATP release at 1251 cm-1 and the appearance of
the free phosphate group of ATP at 1119 cm-1. The decay of
the signal for the aci-nitro anion intermediate and for the
formation of ATP were fitted to a single exponential to give a
mean rate constant of 218 ( 33 s-1 at pH 7.0 and 22 °C.
Among the more recent applications of 2-nitrobenzyl caged
nucleotides have been the time-resolved X-ray crystallographic
investigations of a substrate residing in the active sites of an
enzyme.14
relatively slow, rate-limiting hydrolysis of the aci-nitro inter-
mediate formed by the initial rapid hydrogen atom abstraction
by the excited 2-nitrobenzyl moiety (eq 1).
The relatively modest microscopic rate constant (k ≈ 1-103
s-1) limits its application to the study of rapid biochemical
events and, for example, in at least one case has prevented a
complete kinetic analysis of an enzymatic hydrolysis reaction.
Niggli and Lederer17 found that enzymatic-catalyzed formation
of ADP from ATP was more rapid than the rate of photoinitiated
release of ATP from its caged precursor in a study of myocardial
muscle contraction. Additional disadvantages encountered with
the 2NB and 2NPE cage release strategy for nucleotides have
included: (1) the high UV absorptivity of the aci-nitro and
arylnitroso intermediates, (2) the reactivity of the nitroso
functional group with primary amines and other nucleophiles,
(3) the instability of the caged derivative which occasionally
gives rise to premature hydrolysis and biochemical reaction prior
to photorelease, and (4) the competing pathways for energy
migration and electron transfer prior to C-O bond lysis.
Our initial studies on phosphate photochemistry2d provided
the first example of a new R-keto phototrigger as an alternative
to 2-nitrobenzyl. Desyl phosphates2 (eq 2) and other R-keto
phosphates derivatives18 have subsequently been shown to
release the phosphate ligand efficiently with rate constants of
105 s-1 or higher as a consequence of the direct cleavage of the
covalent bond to the caged substrate from the excited state of
the phototrigger.c,18,19 Furthermore, the byproducts of the R-keto
cage ligands appeared to be biochemically benign.
There have been other investigations which have focused from
modifying 2-nitrobenzyl cages to optimizing the conditions for
photorelease of substrate.15 Substituted 2-nitrobenzyl cages have
been designed to improve the photochemical efficiencies and
to increase the rates of release of several caged neurotransmit-
ters.16 However, few reports have appeared that describe the
discovery, development, and application of entirely new pho-
toactive cages. This is surprising, since there are several
disadvantages which limit the application of the nitrobenzyl
chromophore. An especially significant shortcoming is the
(2)
Others, particularly the group of Corrie and Trentham,18 have
also reported the development of caged phosphates, e.g., 3′,5′-
dimethoxydesyl ATP 12, as alternatives to the 2NP and 2NPE.
Their work parallels our own investigations of desyl phosphate
and desyl cAMP, further substantiating the potential of desyl
derivatives. Subsequent studies by Baldwin,19 Pirrung,20 and
Iwamura21 have also tested the efficacy of R-keto phosphates
as phototriggers. These studies demonstrated, however, that
desyl derivatives are limited by low solubility in aqueous buffer
and by hydrolytic instability. These limitations have prompted
us to further explore other candidates for phototriggers, and
(9) Walker, J. W.; Lu, Z.; Moss, R. L. J. Biol. Chem. 1992, 267, 2459-
2466.
(10) Ostap, E. M.; Thomas, D. D. Biophys. J. 1991, 59, 1235-1241.
(11) Willenbucher, R. F.; Xie, Y. N.; Eysselein, V. E.; Snape, W. J., Jr.
Am. J. Physiol. (Gastroint. LiVer Physiol. 14) 1992, 262, G159-G164.
(12) Lewis, S. M.; Thomas, D. D. Biochemistry 1991, 30, 8331-8339.
(13) Barth, A.; Hauser, K.; Mantele, W.; Corrie, J. E. T.; Trentham, D.
R. J. Am. Chem. Soc. 1995, 117, 10311-10316.
(14) (a) Schlichtling, I.; Almo, S. C.; Rapp, G.; Wilson, K.; Petratos,
K.; Lentfer, A.; Wittnghoffer, A.; Kabsch, W.; Pai, E. F.; Petsko, G. A.;
Goody, R. S. Nature 1990, 345, 309-315. (b) Hajdu, J.; Andersson, I. Annu.
ReV. Biophys. Biomol. Struct. 1993, 22, 467-498.
(15) (a) Walker, J. W.; Reid, G. P.; McCray, J. A.; Trentham, D. R. J.
Am. Chem. Soc. 1988, 110, 7170-7177. (b) Walker, J. W.; Reid, G. P.;
Trentham, D. R. Methods Enzymol. 1989, 172, 288-301. (c) Thirlwell,
H.; Corrie, J. E. T.; Reid, G. P.; Trentham, D. R.; Ferenczi, M. A. Biophys.
J. 1994, 67, 2436-2447.
(16) (a) Milburn, T.; Matsubara, N.; Billington, A. P.; Udgaonkar, J. B.;
Walker, J. W.; Carpenter, B. K.; Webb, W. W.; Marque, J.; Denk, W.;
McCray, J. A.; Hess, G. P. Biochemistry 1989, 28, 49-55. (b) 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. (c) Gee, K. R.; Wieboldt,
R.; Hess, G. P. J. Am. Chem. Soc. 1994, 116, 8366-8367.
(17) Niggli, E.; Lederer, W. J. Biophys. J. 1991, 59, 1123-1135.
(18) (a) Corrie, J. E. T.; Trentham, D. R. J. Chem. Soc., Perkin Trans.
1 1992 2409-2417. (b) Corrie, J. E. T.; Reid, G. P.; Trentham, D. R.;
Mazid, M. A.; Hursthouse, M. B. J. Chem. Soc., Perkin Trans 1 1992,
1015-1019.
(19) Baldwin, J. E.; McConnaughie, A. W.; Moloney, M. G.; Pratt, A.
J.; Shim, S. B. Tetrahedron 1990, 46, 6879-6884.
(20) Pirrung, M. C.; Shuey, S. W. J. Org. Chem. 1994, 59, 3890-3897.
(21) Furuta, T.; Torigai, H.; Sugimoto, M.; Iwamura, M. J. Org. Chem.
1995, 60, 3953-3956.

New PhotoactiVated Protecting Groups. 6
J. Am. Chem. Soc., Vol. 119, No. 10, 1997 2455
accordingly, we report here our recent studies on the synthesis,
exploratory photochemistry, and quantitative mechanistic studies
for substrate release from p-substituted phenacyl phototriggers.22
The photochemical release of phosphates and carboxylic acids
from the p-methoxyphenacyl ester precursors had been reported
previously by Sheehan,23a,b Epstein,23c and Baldwin.19 However,
for each of these examples, the fate of the phenacyl moiety was
found to be the photoreduced form, i.e., the corresponding
acetophenone. These results were puzzling for two reasons:
(1) Anderson and Reese24 had reported as early as 1962 that o-
and p-methoxyphenacyl as well as p-hydroxyphenacyl chloride
gave a substantial amount of the rearranged substituted phenyl
acetate ester when irradiated in methanol. (2) The results of
our2 study and that of Corrie and Trentham15,18 on desyl
phosphates would suggest a carbocation reaction intermediate
which cyclizes in the desyl series to a furan, but in the
p-hydroxyphenacyl analogues to a spiro diene dione proceeding
through a carbonyl migration to the phenylacetic acid which
was reported by Anderson and Reese.24 For these reasons, we
have reinvestigated and extended the study of some of these
reactions and report here their application to the release of ATP
(eq 3).22
Scheme 1
next converted to the diethyl phosphates (3a-d) with tetra-
methylammonium diethyl phosphate in 76-89% yields by direct
displacement of bromide,27 taking advantage of the enhanced
nucleofugacity of the leaving bromide when located R to a
carbonyl. Similarly, 4-hydroxyphenacyl bromide (2c) was
converted to 4-hydroxyphenacyl dibenzyl phosphate (3c, 85%)
by reaction with tetramethylammonium dibenzyl phosphate in
benzene under anhydrous conditions. Strictly anhydrous condi-
tions were important for the displacement reactions with
tetramethylammonium phosphates and the R-keto bromide
because the presence of even small amounts of H2O resulted in
much lower yields.
(3)
Protection of ketone 3e as its ketal 4e was necessary to
remove both benzyl protecting groups in the hydrogenolysis step
and was accomplished with ethylene glycol in the presence of
a catalytic amount of p-toluenesulfonic acid (p-TsOH) in
benzene (Scheme 2).28 Interestingly, without the protection of
ketone 3e, only one of the two benzyl groups was removed even
after extended hydrogenation. Hydrogenolysis29 of the benzyl
groups on 4e with H2 on Pd/C followed by the treatment with
1% HCl gave 4-hydroxyphenacyl dihydrogen phosphate, which
was further purified on an anion-exchange DEAE Sephadex
column, producing 4-hydroxyphenacyl diammonium phosphate
(6c) in a yield of 96%.
Results
A. Synthesis of p-Substituted Phenacyl Phosphates. In
order to explore the effect of aryl substituents on the photo-
chemistry, a series of phenacyl phosphates 3a-c and 6c-h were
synthesized by the procedures shown in Scheme 1. p-Meth-
oxyphenacyl phosphate 3d, which had been reported previously,
is included here for comparison.
The synthesis of the p-hydroxyphenacyl-caged ATP (7,
Scheme 2) required the coupling18,30 of the monophosphate 6c
with activated ADP under rigorously anhydrous reaction condi-
tions. Thus, the monophosphate 6c and ADP were dried by
sequential cycles of dissolution and evaporation in scrupulously
dried solvents,31 i.e., pyridine or DMF, and then transformed
Unsubstituted 1a and 3-methoxyacetophenone (1b) were
converted to the R-bromoacetophenones (2a, 63% and 2b, 76%,
respectively) by acid-catalyzed bromination.25 However, R-bro-
mo-4-hydroxyacetophenone (2c) was synthesized by bromina-
tion of 4-hydroxyacetophenone (1c, 71%) with cupric bromide26
to avoid dibromination. The R-bromoacetophenones 2a-d were
(22) This work has been communicated: Givens, R. S.; Park, C.-H.
Tetrahedron Lett. 1996, 37, 6259-6262.
(27) (a) Kluba, M.; Zwierzak, A. Ann. Soc. Chim. Polonorum 1974, 48,
1603. (b) Zwierzak, A.; Kluba, M. Tetrahedron 1971, 22, 3163-3170. (c)
Serebryakov, E. P.; Suslova, L. M.; Kucherov, V. F. Tetrahedron 1978,
34, 345-351.
(23) (a) Sheehan, J. C.; Wilson, R. M.; Oxford, A. W. J. Am. Chem.
Soc. 1971, 93, 7222-7228 and references therein. ( b) Sheehan, J. C.;
Umezawa, J. J. Org. Chem. 1973, 38, 3771-3774. (c) Epstein, W. W.;
Garrossian, M. J. Chem. Soc., Chem. Commun. 1987, 532-533.
(24) Anderson, J. C.; Reese, C. B., Tetrahedron Lett. 1962, 1-4.
(25) (a) Cowper, R. M.; Davidson, L. H. Org. Synth. 1939, 19, 24-26.
(b) Rather, J. B.; Reid, E. E. J. Am. Chem. Soc. 1919, 41, 75-83.
(26) (a) Buu-Hoi, N. P.; Lavit, D. J. Chem. Soc. 1955, 18-20. (b)
Durden, D. S.; Juorio, A. V.; Davis, B. A. Anal. Chem. 1980, 52, 1815-
1820.
(28) (a) Amos, A. A.; Ziegler, P. Can. J. Chem. 1959, 37, 345. (b) Ireland,
R. E.; Walba, D. M. Organic Syntheses; Wiley: New York, 1988; Collect.
Vol. 6, p 567.
(29) Heathcock, C. H.; Ratcliffe, R. J. Am. Chem. Soc. 1971, 93, 1746-
1757.
(30) Hoard, D. E.; Ott, D. G. J. Am. Chem. Soc. 1965, 87, 1785-1788.
(31) Burfield, D. R.; Smithers, R. H. J. Org. Chem. 1978, 43, 3966-
3969.

2456 J. Am. Chem. Soc., Vol. 119, No. 10, 1997
Park and GiVens
Scheme 2
The syntheses of 4-aminophenacyl diammonium phosphate
(6f), 4-acetamidophenacyl diammonium phosphate (6g), and
4-(carbomethoxyamino)phenacyl diammonium phosphate (6h)
were accomplished by the same general strategy employed for
4-hydroxyphenacyl diammonium phosphate (Scheme 3). 4-Ami-
noacetophenone (1e) was converted to 4-((carbobenzyloxy)-
amino)acetophenone (1h, 89%) before the bromination step to
prevent formation of the undesired dibrominated products. This
strategy also permitted the removal of the carbobenzyloxy group
along with benzyl phosphate protecting groups under the same
hydrogenolysis conditions as the phosphate group deprotection.
Bromination of 1f-h either by bromine25 or by cupric bromide
26 gave the corresponding R-bromo derivatives 2f-h (89-98%)
followed by conversion27 to the phenacyl dibenzyl phosphates
3f-h (73-89%). Reaction of 3f-h with ethylene glycol and
catalytic amount of p-TsOH gave the ketals 4f-h28 in yields
of 77-96%.
Hydrogenolysis29 removed the two benzyl groups from the
phosphate ligand along with the one in the carbobenzyloxy
group of 4f. Treatment with small amounts of 1% HCl easily
removed the ketal group. Interestingly, the removal of the ketal
from 4h was much slower, requiring 48 h, in contrast to the 12
h needed for 4f and 4g. Conversion of the phenacyl dihydrogen
phosphates to the diammonium phosphates was performed on
the weakly anionic DEAE cellulose column by elution with
ammonium bicarbonate. The ammonium salts were recrystal-
lized from H2O/MeOH solution to give the 4-amino- (6f, 95%),
4-acetamido- (6g, 96%), and 4-(carbomethoxyamino)phenacyl
diammonium phosphate (6h, 96%), respectively.
The stability of 4-hydroxyphenacyl ATP (7) was examined
in buffered and unbuffered aqueous media. Solutions of 7 in
Tris buffer (50 mM, pH 7.3), three different Ringer’s solutions
(pH 6.5), D2O, and H2O were stored in the dark at 25 °C and
were periodically monitored by HPLC. The samples showed
no loss of 7 over a 24 h period. In fact, for the D2O solution
of p-hydroxyphenacyl ATP, the NMR sample of the phototrigger
was stable for over 100 days.
to the tri-n-octylammonium salt. ADP was activated by first
treating it with pyridine and tri-n-butylamine and then converting
the salt to imidazolyl ADP by reaction with carbonyl diimida-
zole. The triphosphate was obtained from reaction of the two
substrates for 72 h in HMPA (hexamethylphosphoramide) at
room temperature with 20 min of sonication every 12 h, due to
the poor solubility of the reactants. The product was purified
by DEAE cellulose chromatography, eluting with a step gradient
of aqueous ammonium bicarbonate. Purification by DEAE
Sephadex chromatography with 10% methanol/ammonium
acetate gave product of slightly higher purity but was a more
tedious process. The yield of 4-hydroxyphenacyl ATP (7) from
6c and ADP was 42%.
B. Photochemical Studies. The photochemistry of this
limited collection of phenacyl phosphates is shown in Scheme
4 and eqs 4 and 5. Irradiation of all four diethyl esters 3a-d
in MeOH as well as 3c,d in t-BuOH at 300 nm produced the
corresponding reduction products, acetophenones 1a-d, as
primary products (Scheme 4). m-Methoxyphenacyl phosphate
(3b) also formed radical coupling products 9b and 10b; whereas,
the p-hydroxy (3c) and p-methoxy esters (3d) gave rearrange-
ment products 8c,d as their methyl or tert-butyl esters, respec-
tively. Phosphates 3c,d reacted with higher efficiencies and
Scheme 3

New PhotoactiVated Protecting Groups. 6
J. Am. Chem. Soc., Vol. 119, No. 10, 1997 2457
Scheme 4
4-aminoacetophenone (1e), were also detected by HPLC of the
photolysis mixture. Likewise, the reduction product 1g from
4-acetamidophenacyl diammonium phosphate (6g) was obtained
along with the corresponding dimers. In contrast, 4-(car-
bomethoxyamino)phenacyl diammonium phosphate (6h) did not
produce the expected reduction products, instead giving a
complex mixture of unidentified products (by HPLC). Quantum
efficiencies for the disappearance of 6g,h and the reduction
products are shown in Table 3. 31P NMR spectroscopy
confirmed that all three aminoacetophenones triggered the
release of inorganic phosphate (Pi) upon photolysis. However,
the rearranged substituted phenyl acetates were not detected.
Interestingly, it appears that these phototriggers release phos-
phate through a photoreduction process. We have not pursued
the mechanism of this photorelease process.
A determination of the change in the pH (∆pH) for a solution
of each of the three phenacyl diammonium phosphates 6c, 6f,
and 6g in H2O was measured as a function of photolysis time
(Table 4). The ∆pH for solutions of 4-hydroxyphenacyl
diammonium phosphate (6c) and 4-acetamidophenacyl diam-
monium phosphate (6g) were 0.7 and 1.1, respectively. How-
ever, the ∆pH of solutions of 4-aminophenacyl diammonium
phosphate (6f) was practically 0, a result of the low quantum
efficiency for the release of inorganic phosphate (Pi) from this
phototrigger.
gave much better mass balances than the other two phenacyl
diethyl phosphate derivatives.
Discussion
(4)
(5)
From among the limited selection of substituted phenacyl
phosphates examined in our study, the most versatile photo-
trigger appears to be p-hydroxyphenacyl. The meta-substituted
and unsubstituted phosphates yield complex mixtures of prod-
ucts in MeOH including acetophenone products from photore-
duction and dimeric products resulting from radical coupling
processes. In contrast, both p-methoxy- and p-hydroxyphenacyl
phosphate rearranged to the p-methoxyphenyl- and p-hydrox-
yphenylacetates when photolyzed in hydroxylic solvents, MeOH
or H2O, respectively.
In contrast to the complex photochemistry of 3a-d, p-
hydroxyphenacyl phosphate (6c) and the caged ATP 7 were
each converted at 300 nm exclusively to p-hydroxyphenylacetic
acid and the released inorganic phosphate (Pi) or ATP,
respectively. Essentially identical results were obtained whether
the irradiations were conducted in Tris buffer or in the
unbuffered aqueous media (e.g., H2O or D2O, eq 4). No
evidence of reduction or coupling products was obtained under
these conditions.
Quantum efficiencies for the disappearance of p-hydroxy-
phenacyl derivatives and diethyl p-methoxyphenacyl
phosphate2b,c,22 are given in Table 1. Appearance efficiencies
were also measured for the rearrangement, for some of the
reduction products, and, in one case, for the substrate appear-
ance, i.e., ATP.
This predominance of the photorearrangement is even more
remarkable since apparently only these two oxygen-containing
electron donors are capable of promoting the phenacyl to
phenylacetate rearrangement. The three p-amino derivatives
6f-h gave only photoreduction products of the trigger.
For the p-hydroxyphenacyl derivatives which possess both a
requisite para electron-donating group and a good nucleofuge,
the rearrangements are efficient and virtually free of byproducts,
as shown in Table 1. The disappearance efficiencies for the
caged phosphates 6c and 7 are 0.37 ( 0.04 and product yields
are essentially quantitative by NMR. The measured appearance
quantum efficiencies were 0.30((0.03) for the formation of the
two photoproducts from 7.
The results for the phosphate series are in excellent agreement
with the results reported earlier by Anderson and Reese.24 In
that study, only strong electron-donating substituents, i.e., OH
or OCH3, located either ortho or para to the carbonyl underwent
aryl group migration to the substituted phenylacetic acids.
Additional examples of aryl migrations bearing strongly electron-
donating para substituents are found in ground state chemistry
as well, for example, the well-established neighboring phenyl
group participation32 observed in the solvolysis of â-anisylethyl
tosylates, brosylates, and bromides. Our results with the
p-hydroxyphenacyl derivatives are congruent with the earlier
suggestion that arylmethyl phosphates yield a photoproduct
distribution consistent with a carbocation intermediates2 and give
a Hammett substituent correlation with a negative slope of 0.9
Sodium 2-naphthalenesulfonate successfully quenched the
photorelease of ATP from p-hydroxyphenacyl ATP (7) and gave
good Stern-Volmer (SV) kinetics (Table 2). The rate constant
for the decay of the triplet, which is equated with the release
rate for ATP, was 5.5 × 108 s-1; whereas, the rate constant for
disappearance of 7 was 6.9 × 108 s-1, obtained from the SV
slopes assuming a rate constant for diffusion in Tris buffer of
6.6 × 109 M-1 s-1
.
Photolyses of 4-amino- (6f), 4-acetamido- (6g), and 4-(car-
bomethoxyamino)phenacyl diammonium phosphate (6h) are
shown in eq 5. In contrast to the 4-hydroxy derivatives,
photoreactions for 4-aminophenacyl diammonium phosphate (6f)
in buffered media at pH 7.3 had a low quantum efficiency for
disappearance (<5%). Small amounts of the reduction product,
(32) Cram, D. J. J. Am. Chem. Soc. 1952, 74, 2129-2137 and references
therein.

2458 J. Am. Chem. Soc., Vol. 119, No. 10, 1997
Park and GiVens
Table 1. Quantum Efficiencies for the Photorearrangements of 4-Hydroxy- (3c, 6c, and 7) and 4-Methoxyphenacyl (3d)2b,c,19,23c Phosphatesa,b
aryl substituent
p-MeOb (3d)
(3d)
R1
Et
Et
Et
Et
R2
Et
Et
Et
Et
solvent
Φdis
Φred
Φrearrange
Φphosphate
ref
CH3OH
CD3OD
CH3OD
CH3OH
aq CH3CN
Tris
0.42
0.07
0.013
0.053
0.05
0.00
0.00
0.20
0.14
0.11
0.33
0.12
0.31
2b,c, 23c
2b
2b
e
e
e
(3d)
p-OHc (3c)
p-OHc,d (6c)
(7)
0.77
0.38
0.37
NH4+
ATP
NH4+
NH4+
0.30
a Error limits for the quantum efficiencies are estimated to be (10%. b Irradiated at 300 nm. c Irradiated at 350 nm. d 10%CH3CN/90% Tris
buffer. e This work.
Table 2. Stern-Volmer Quenching Data for Photolysis of 7 in
Scheme 5
Tris Buffer at 350 nma
sodium
naphthalenesulfonate
[mM]
Φ
dis (7)
Φ
rearr (8)
ΦATP
0
50
0.37 (0.04)
0.31
0.31 (0.03)
0.25
0.30 (0.03)
0.25
100
150
0.27
0.24
0.20
0.16
0.23
0.19
slope (Ksv), M-1
3.56
6.38
3.61
τ, ns
0.54 (0.09)
6.90 (1.15)
0.97 (0.10)
3.20 (0.33)
0.55 (0.06)
5.50 (1.00)
108 k, s-1
a Error shown in parenthesis.
Table 3. Quantum Efficiencies for p-Substituted Aminophenacyl
Diammonium Phosphates 6f-h in Tris Buffer at 300 nma
p-substituent
concn (mM)
Φdis
Φred
others
4-NH2 (6f)
4-CH3CONH (6g)
4-CH3OCONH (6h)
19
6.5
4.2
<0.05 <0.05
0.38
0.34
0.11 dimers
2 unknowns
a Error limits for the quantum efficiencies are estimated to be 10%.
a radical pair that undergoes subsequent electron transfer to an
ion pair, as outlined in Scheme 5 for release of phosphate form
6c.34 Pincock34a,b has proposed this mechanistic pathway for
the photochemical release of carboxylates from benzyl and
naphthyl esters via the excited singlet state during irradiation
of the esters in methanol. Recently, Peters34c provided additional
evidence for this pathway for the photosolvolysis of benzhydryl
chlorides in polar media such as acetonitrile. Peters suggested
that as much as 30% of the contact ion pairs generated from
the singlet state of p-methoxybenzhydryl chloride originates
from a geminate radical pair via an electron transfer (ET) process
that competes with diffusion of the radical pair from the cage.
The effect of substituents on the partitioning of the ET-diffusion
competition is particularly significant in the benzhydryl series.
The ion pair formation must be negligible for unsubstituted and
p-methylbenzhydryl chlorides but dramatically increases to 20%
or higher with methoxy substitution on one or both of the aryl
rings.
For our studies, either water or buffer (highly polar, ionic
media) is the solvent, the substituent is a p-hydroxy group (a
better electron donor), and the radical pair is initially a triplet.
While this combination of factors has not been present in any
of the photochemical substrates to date that have been studied
mechanistically, a sequence of steps parallel to those reported
by Peters34c may very well obtain here. This proposed mech-
anism is shown in Scheme 5.
Table 4. pH Changes for the Nonbuffered Photolysis Solution of
Phenacyl Diammonium Phosphates 6c, 6f, and 6g in H2Oa
pH
p-substituent
4-OH (6c)
4-NH2 (6f)
concn (mM)
before
after
∆pH
7.5
7.5
7.5
7.4
6.8
4.7
6.7
6.7
3.6
0.7
0.1
1.1
4-CH3CONH (6g)
a Irradiated at 300 nm for 10 min.
versus Hammett σ reflecting positive charge character in the
product-forming step.33
Quenching studies of 7 with sodium 2-naphthalenesulfonate
(Table 2) confirmed the triplet as the reactive excited state for
the photorelease and the rearrangement reactions. The linear
dependence of the Stern-Volmer quenching by 2-naphthale-
nesulfonic acid to greater than 90% of the phosphate photore-
lease accords with an exclusively triplet state pathway for the
reaction and parallels our earlier results for the desyl phosphates2b,c
and desyl amino acids.1 The rate constants calculated for release
of phosphates derived from the SV quenching were 6.9 × 108
s-1 for the disappearance of 7 and 5.5 × 108 s-1 for the
appearance of ATP and the rearranged p-hydroxyphenylacetic
acid (8). These rate constants compare very favorably with the
value of 2.9 × 108 s-1 for desyl-caged cAMP2b,c and 1.2 × 107
s-1 obtained for caged γ-O-desyl glutamate.1
Structural features of this new R-keto phototrigger and its
efficient photorearrangement to p-hydroxyphenylacetic acid
invite some additional comments. First, this new caging group
lacks a chiral center, allaying any concerns about generating a
Other than the determination of the multiplicity, the overall
mechanistic picture for the photoreaction remains to be estab-
lished. A reasonable suggestion for the triplet is homolysis to
(34) (a) Hilborn, J. W.; MacKnight, E.; Pincock, J. A.; Wedge, P. J. J.
Am. Chem. Soc. 1994, 116, 3337-3346. (b) Pincock, J. A. Acc. Chem.
Res. 1997, 30, 43-49. (c) Lipson, M.; Deniz, A. A.; Peters, K. S. J. Am.
Chem. Soc. 1996, 118, 2992-2997.
(33) (a) Givens, R. S.; Matuszewski, B.; Athey, P. S.; Stoner, R. M. J.
Am. Chem. Soc. 1990, 112, 6016-6023. (b) Givens, R. S.; Singh, R.
Tetrahedron Lett. 1991, 7013-7017.

New PhotoactiVated Protecting Groups. 6
J. Am. Chem. Soc., Vol. 119, No. 10, 1997 2459
Table 5. Photoefficiencies of 4-Hydroxyphenacyl Adenosine
5′-Triphosphate (7) in Buffered Solution
cAMP. In our series, we have not encountered the equivalent
hydrolyses of p-hydroxyphenacyl-caged phosphates 6c or 7 in
Ringer’s solutions I, II, and III or Tris or in H2O or D2O at
temperatures between 25 and 40 °C over a 24 h period to yield
Pi and ATP, which would have been accompanied by R,p-
dihydroxyacetophenone (11). No 11 was detected in any of
these control experiments. Thus, caged phosphates 6c and 7
are stable to most common solvents, yet trigger the release of
the phosphate ligand upon 300-350 nm radiation.
phototriggers
conditionsa
tris Bufferb
pH Φ4(dis) ΦATP(app) Φ8(app)
7
6c
7.3
0.37
0.38
0.30
c
0.31
0.12
45% Tris/55% 6.5
CH3CN
a Irradiated at 300 nm at room temperature. b Stability was demon-
strated for this buffer, H2O, D2O and for Ringer’s solutions I, II, and
III for greater than 24 h. For D2O, the phototrigger was stable for
over 100 h by HPLC. c Not determined
Conclusions
mixture of diastereomers when employed with a chiral substrate.
This feature will significantly simplify the purification and
characterization of the synthetic caged substrates for phenacyl
cages. Second, unlike 2-NB, 2-NPE, and even the desyl cages,
the chromophore of the rearranged photoproduct, p-hydrox-
yphenylacetic acid, does not compete for the incident radiation
in the 300-400 nm region, since it is blue-shifted relative to
the phenacyl cage. This eliminates product interference of the
incident light absorption by the phototrigger which would reduce
the yields and prevent complete conversion in the release of
the substrate. Third, the attachment of the substrate to a primary
methylene, R to a carbonyl, should stabilize the derivative to
any “SN1-like” ground state release of the substrate. Direct
nucleophilic substitution, however, remains a potential debilitat-
ing reaction.
The p-hydroxyphenacyl cage fulfills several of the criteria
advanced by Lester,6c especially the requirements of a high
efficiency, a benign photobyproduct, and good stability of the
caged derivative in ionic media. Most importantly, the p-
hydroxyphenacyl phototrigger has no chiral center but does
possess a much improved aqueous buffer solubility when
compared with that of the desyl analogues. The photorelease
is a primary process, occurring with rate constants of 107-108
s-1, several orders of magnitude faster that the 2NB and 2NPE
derivatives and approximately the same as the desyl analogues.
These very desirable features of the new generation of photot-
riggers provide a promising future for their application to
mechanistic biochemistry, physiology, and related fields.
Experimental Section
As a test of the solvolytic stability, p-hydroxyphenacyl ATP
was subjected to various buffers and to aqueous media employed
in typical biological studies. Six different media were explored
in this study: H2O, D2O, Tris, and Ringers solutions I, II, and
III. Solutions of the two ester in the series of solvents and
buffers were sampled periodically over a 24 h period and the
aliquots examined by HPLC. As shown in Table 5, the
p-hydroxyphenacyl ATP phototrigger was stable for at least 24
h under all conditions examined. In fact, 7 remained unchanged
in D2O for over 100 h. Additional studies with other p-
hydroxyphenacyl carboxylic esters have also been stable to
solvolytic conditions.35 In addition, HPLC analysis confirmed
that neither of these two esters were measurably degraded or
hydrolyzed to R,p-dihydroxyacetophenone (11) when stored in
the dark.
General Method. All nonaqueous reactions were performed under
an atmosphere of argon with a slight positive pressure and continuous
magnetic stirring. All reagents were used as received without further
purification unless otherwise stated. All solvents used for chromato-
graphic purposes were of reagent grade or better or were distilled prior
to used. Acetophenone was distilled under vacuum. Triethylamine
was dried by refluxing over potassium hydroxide and distilling under
argon, collecting the middle fraction, and stored in the dark. Pyridine
and N,N-dimethylformamide were dried over phosphorus pentoxide and
distilled under reduced pressure, collecting the middle fraction, and
stored over sodium hydroxide and molecular sieves, respectively.
Benzene was distilled from sodium/benzophenone ketyl prior to use.
Technical grade diethyl ether, methanol, methylene chloride, and
tetrahydrofuran were distilled from calcium hydride. Flash column
chromatography was performed with ethyl acetate/methylene chloride
to purify the phosphate triesters. All NMR spectra are reported in ppm
(δ) with either tetramethylsilane (1H and 13C) or 85% phosphoric acid
(31P) as the internal standards. Melting points are uncorrected. Mass
spectra were provided by the Mass Spectrometry Laboratory at the
University of Kansas.
O
OH
HO
2-Βromoacetophenone (2a) was prepared by the method of Cowper
11
and Davidson.25
2-Bromo-3′-methoxyacetophenone (2b)25 was prepared by the same
method as above to yield 11.6 g (76%) of 2-bromo-3′-methoxyac-
In contrast, Corrie and Trentham et al.7 found that 3′,5′-
dimethoxybenzoin-caged ATP (12) undergoes slow hydrolysis
to ADP and 3′,5′-dimethoxybenzoin phosphate both in phos-
phate and carbonate buffers. The instability of 12 was attributed
by the authors to anchimeric assistance of the carbonyl of the
hydrated benzoyl group, as shown in Scheme 6.7,8 That this
was not observed for the p-hydroxyphenacyl series may be due
in part to the greater electron donating ability of the hydroxy
group to the carbonyl, rendering it less electrophilic and thus
less reactive toward hydration. It should be noted that Iwamura
et al.21,36 reported that desyl cAMP hydrolyzed spontaneously
in the absence of light in Ringer’s solution containing 1%
DMSO to yield benzoin and cAMP. This hydrolysis must be
occurring by some other mechanism, e.g., SN2 displacement of
1
etophenone (2b): mp 55-57 °C; H NMR (CDCl3) δ 3.82 (s, 3H),
4.51 (s, 2H), 7.13-7.60 (m, 4H); IR (CHCl3) 3100, 2954, 1684, 1598,
1583, 1037 cm-1; UV-vis (CH3CN) λmax (ꢀ) 312 (2881), 253 (8413),
218 (21 900).
2-Bromo-4′-hydroxyacetophenone (2c) was prepared by the method
of Buu-Hoi and Lavit26 to give 2.53 g (11.8 mmol, 53.9%) of 2-bromo-
1
4′-hydroxyacetophenone (2c): mp 129-131 °C; H NMR (acetone-
d6) δ 4.64 (s, 2H), 6.97 (d, J ) 8.7 Hz, 2H), 7.97 (d, J ) 8.7 Hz, 2H);
13C NMR (acetone-d6) δ 38.38, 122.22, 132.83, 138.29, 169.31, 196.35.
Phenacyl diethyl phosphate (3a), 3-methoxyphenacyl diethyl
phosphate (3b), and 4-hydroxyphenacyl diethyl phosphate (3c) were
prepared by the general method of Zwiezak and Kluba.27,37
4-Hydroxyphenacyl Dibenzyl Phosphate (3e).27 A mixture of
tetramethylammonium dibenzyl phosphate (1.60 g, 4.70 mmol), 4-hy-
droxyphenacyl bromide (2c, 1.00 g, 4.70 mmol), and 20 mL of benzene
was placed in a 50 mL round-bottom flask and refluxed with efficient
(35) C.-H. Park, W. Bartlett, A. Jung, J. Weber, and R. Givens, work in
progress.
(36) For additional details, see: ref 21. Futura, T.; Torigai, H.; Iwamura,
M. Chem Lett. 1993, 1179-1182. Futura, T.; Torigai, H.; Osawa, T.;
Iwamura, M. J. Chem. Soc., Perkin Trans. 1 1993, 3139-3142.
(37) See Supporting Information for details including experimental
procedures and spectral data.

2460 J. Am. Chem. Soc., Vol. 119, No. 10, 1997
Park and GiVens
Scheme 6
stirring for 2 h. The solution was then cooled to room temperature
and extracted with H2O/EtOAc solution. After the solvent was
removed, a pale yellow solid product was purified by silica gel flash
column chromatography (CH2Cl2/EtOAc 80:20) to yield 1.63 g (85%)
of 4-hydroxyphenacyl dibenzyl phosphate (3e): mp 116.5-118 °C;
1H NMR (DMSO-d6) δ 5.13 (d, J ) 7.8 Hz, 4H), 5.36 (d, J ) 10.7
Hz, 2H), 6.87 (d, J ) 7.6 Hz, 2H), 7.38 (m, 10H), 7.82 (d, J ) 7.6 Hz,
2H), 10.51 (s, 1H); 13C NMR (DMSO-d6) δ 68.65 (d, J ) 22.3 Hz),
68.73 (d, J ) 25.8 Hz), 115.39, 125.14, 127.81, 128.30, 128.41, 130.34,
136.01 (d, J ) 29.8 Hz), 162.65, 190.80 (d, J ) 16.5 Hz); 31P NMR
(DMSO-d6) δ 0.56; IR (CHCl3) 3212, 1684, 1600, 1456, 1373, 1281,
1115, 984 cm-1; exact mass calcd for C22H21O6P 412.1154, found
412.1163. Anal. Calcd for C22H21O6P: C, 64.08; H, 5.13. Found:
C, 64.46; H, 5.18.
of ammonium acetate. The products were eluted with stepping gradient
of ammonium acetate solution (200 mL each of 0.0-0.2 M solution
with 0.05 M increments). Diammonium salt 6c was eluted at 0.2 M
ammonium acetate. Excess ammonium acetate and water were
lyophilized from the solution to yield a solid which was recrystallized
from H2O/MeOH to give 1.05 g (96.2%) of 4-hydroxyphenacyl
diammonium phosphate (6c): mp 177-192 °C (dec); 1H NMR (D2O)
δ 5.08 (d, J ) 5.6 Hz, 2H), 6.92 (d, J ) 8.8 Hz, 2H), 7.89 (d, J ) 8.8
Hz, 2H); 13C NMR (D2O) δ 69.37 (d, J ) 14.2 Hz), 118.46, 128.59,
133.42, 165.03, 200.50 (d, J ) 33.3 Hz); 31P NMR (D2O) δ 1.18; IR
(KBr) 3350, 3088, 2950, 1679, 1594, 1570, 1440, 1268, 1090 cm-1
;
UV-vis (CH3CN/H2O) λmax (ꢀ) 220 (8400), 282 (14 000). Anal. Calcd
for C8H15N2O6P: C, 36.10; H, 5.68; N, 10.52. Found: C, 35.82; H,
5.38; N, 10.33.
4-Hydroxyphenacyl Dibenzyl Phosphate (Ethylene Ketal) (4e).28
To the solution of 4-hydroxyphenacyl dibenzyl phosphate (3e, 3.45 g,
8.4 mmol) in 100 mL of benzene, a catalytic amount of p-TsOH (79.9
mg, 0.42 mmol) was added followed by the addition of excess amount
of ethylene glycol (8.86 g, 145.1 mmol). The solution was refluxed
for 12 h with a Dean-Stark apparatus to remove the water that was
generated. When reaction was complete, NaHCO3 (0.71 g, 8.4 mmol)
was added to neutralize the mixture. After benzene was removed in
Vacuo, ethylene glycol was removed by extraction with H2O/EtOAc.
Further purification was done by recrystallization from Et2O to give
3.19 g (82.1%) of 4-hydroxyphenacyl dibenzyl phosphate (ethylene
ketal) (4e): mp 97-98 °C; 1H NMR (DMSO-d6) δ 3.80 (m, 2H), 4.00
(m, 2H), 4.03 (d, J ) 6.7 Hz, 2H), 4.95 (d, J ) 7.7 Hz, 4H), 6.74 (d,
J ) 8.6 Hz, 2H), 7.27 (d, J ) 8.6 Hz, 2H), 7.33-7.38 (m, 10 H), 9.53
(s, 1H); 13C NMR (DMSO-d6) δ 64.91, 68.34 (d, J ) 22.7 Hz), 69.03
(d, J ) 24.9 Hz), 107.06 (d, J )33.4 Hz), 114.71, 127.21, 127.71,
128.28, 128.40, 129.11, 135.96 (d, J ) 28.3 Hz), 157.60; 31P NMR
(DMSO-d6) δ -0.02; IR (CHCl3) 3280, 1610, 1512, 1455, 1273, 1168,
1002 cm-1; exact mass calcd for C24H25O7P 456.1416, found 456.1400.
Anal. Calcd for C24H25O7P: C, 63.16; H, 5.52. Found: C, 63.00; H,
5.49.
Exploratory Hydrogenation of 4-Hydroxyphenacyl Dibenzyl
Phosphate (3e).29 To 4-hydroxyphenacyl dibenzyl phosphate (3e, 0.35
g, 0.84 mmol) dissolved in 10 mL of MeOH was added 35 mg of 10%
Pd/C. The same hydrogenation procedure (see below) was used as
that employed for 4-hydroxyphenacyl dibenzyl phosphate (ethylene
ketal) (4e). Only one benzyl group was removed even with extended
hydrogenation to give 4-hydroxyphenacyl benzyl phosphate (5c).
4-Hydroxyphenacyl Diammonium Phosphate (6c).29 To 4-hy-
droxyphenacyl dibenzyl phosphate (ethylene ketal) (4e, 1.86 g, 4.1
mmol) dissolved in 30 mL of MeOH was added 186 mg of 10% Pd/C.
The solution was hydrogenated under 10 psi of H2 with 45 min of
stirring followed by the addition of 1% HCl (1 mL). After filtration,
the filtrate was concentrated in Vacuo to yield a viscous liquid which
was loaded on 5 g of DEAE Sephadex column pretreated with 10 mM
4-Hydroxyphenacyl Adenosine 5′-Triphosphate, Triammonium
Salt (7).30 Dowex 50W resin (30 g) in a sintered glass funnel was
treated with 5% hydrochloric acid (60 mL × 2) and washed with water
until the filtrate was neutral. The resin was then treated with a 20%
aqueous pyridine solution and washed with water until the filtrate was
neutral. A solution of adenosine 5′-diphosphate (ADP, potassium salt,
1.105 g, 2.33 mmol) in 20 mL of water was stirred with the resin for
5 min, the resin was filtered and washed with water (20 mL × 4). Into
the filtrate was added tributylamine (0.87 g, 4.7 mmol) followed by
30 min of stirring. The solution was lyophilized to yield 1.95 g of a
white powder which was further dried by sequential cycles of
dissolution and evaporation of dry pyridine (10 mL × 2) and dry DMF
(10 mL × 2). The dried residue was redissolved in 10 mL of dry
DMF, carbonyl diimidazole (1.65 g, 10 mmol) was added, and the
resulting solution was stirred under argon for 24 h at ambient
temperature. The reaction was quenched by adding methanol (0.3 mL,
8 mmol); the resulting solution was stirred for 1 h, and the solvent
was removed in Vacuo. Meanwhile, a solution of 4-hydroxyphenacyl
diammonium phosphate (6e, 1.17 g, 4.44 mmol) in 40 mL of water
was stirred for 30 min with the activated Dowex 50W resin (30 g,
pyridinium form) as described above. The resin was filtered and
washed with water (20 mL × 3). Tri-n-octylamine (1.65 g, 4.44 mmol)
was added to the filtrate, and the mixture was lyophilized to yield 3.21
g of a white powder which was dissolved in dry pyridine (15 mL × 2)
followed by evaporation and then dissolved in dry DMF (15 mL × 2)
and evaporated. The gummy residue was redissolved in 10 mL of dry
DMF and added to the above ADP imidazole solution. The DMF was
evaporated in Vacuo at room temperature and hexamethylphosphora-
mide (HMPA, 30 mL) added to the pale yellow residue. The solution
was stirred under argon for 3 days during which time the solution was
sonicated for two 20 min periods each day. The reaction was quenched
by adding 120 mL of water and washed with chloroform (100 mL ×
4) and hexane (100 mL). The aqueous layer was lyophilized and
purified by DEAE-cellulose (HCO3- form, 500 mL dry volume) column
eluted with a stepping gradient of ammonium bicarbonate solution with

New PhotoactiVated Protecting Groups. 6
J. Am. Chem. Soc., Vol. 119, No. 10, 1997 2461
the concentrations from 0.0 to 0.3 M in 0.05 M increments. Fractions
with only one component were combined and lyophilized to yield 0.55
g (42.3%) of 4-hydroxyphenacyl adenosine 5′-triphosphate, triammo-
nium salt (7): mp 137 °C (dec); 1H NMR (D2O) δ 4.21 (m, 1H), 4.27
(s, 2H), 4.37 (t, J ) 4.3 Hz, 1H), 4.44 (t, J ) 21.5 Hz, 1H), 5.10 (m,
2H), 5.77 (d, J ) 5.1 Hz, 1H), 6.52 (d, J ) 8.5, 2H), 7.46 (d, J ) 8.8
Hz, 2H), 8.02 (s, 1H), 8.29 (s, 1H); 13C NMR (D2O) δ 68.12 (d, J )
20.7 Hz), 70.87 (d, J ) 19.4 Hz), 72.92, 77.77, 86.51 (d, J ) 37.1
Hz), 90.20, 117.93, 121.14, 128.20, 133.00, 142.91, 151.02, 152.90,
156.33, 164.12, 197.95 (d, J ) 33.2 Hz); 31P NMR (D2O) δ -25.42 (t,
J ) 48.3 Hz), -14.09 (d, J ) 46.1 Hz), -13.74 (d, J ) 49.4 Hz); IR
(KBr) 3320, 3150, 2991, 2821, 1710, 1679, 1594, 1570, 1453, 1372,
1248, 1112, 1041 cm-1; UV-vis (CH3CN/H2O) λmax (ꢀ) 286 (14 600);
exact mass calcd for C18H22O15P3 (free acid) 641.0482, found 641.0460.
4-((Carbobenzyloxy)amino)acetophenone (1f). To a solution of
4-aminoacetophenone (1e, 5 g, 37 mmol) and NaOH (1.49 g) in 50
mL of dioxane/H2O (70:30) was added dropwise benzyl chloroformate
(6.3 g) for 30 min at 0 °C followed by 2 h of vigorous stirring at room
temperature. After the reaction was complete, the solvent was removed
in Vacuo. The resulting solid was extracted with EtOAc/H2O. The
organic layer was concentrated and purified by silica gel column
chromatography (CH2Cl2/hexane 50:50). Further purification was
accomplished by recrystallization from CH2Cl2/hexane giving 8.84 g
(89%) of yellowish crystalline 4-((carbobenzyloxy)amino)acetophenone
h using a Dean-Stark apparatus to remove the water. After the reaction
was complete, the mixture was cooled to ambient temperature and
neutralized by addition of NaHCO3 (0.54 g) followed by removal of
the solvent in Vacuo. Extraction of the crude product with H2O/EtOAc
and purification by silica gel column chromatography (EtOAc/hexane
50:50) gave 3.62 g (96%) of 4-(((carbobenzyloxy)amino))phenacyl
dibenzyl phosphate (ethylene ketal) (4f): 1H NMR (acetone-d6) δ 2.89
(m, 2H), 4.10 (m, 2H), 4.12 (d, J ) 6.8 Hz, 2H), 4.99 (d, J ) 7.8 Hz,
4H), 5.18 (s, 2H), 7.34-7.39 (m, 15H), 7.44 (d, J ) 8.7 Hz, 2H), 7.59
(d, J ) 8.7 Hz, 2H), 8.85 (s, 1H); 13C NMR (acetone-d6) 64.38, 72.28,
75.32 (d, J ) 20.2 Hz), 75.79 (d, J ) 22.6 Hz), 125.35 (d, J ) 30.0
Hz), 134.79, 134.95, 135.01, 135.09, 135.28, 135.38, 136.04, 137.52,
137.79 (d, J ) 16.2 Hz), 143.41, 152.15, 152.38, 160.38; 31P NMR
(acetone-d6) δ -2.63; IR (CHCl3) 3428, 3050, 2931, 2881, 1687, 1598,
1525, 1502, 1435, 1410, 1212, 1182 cm-1
.
4-Aminophenacyl Diammonium Phosphate (6f).29 To 4-((car-
bobenzyloxy)amino) phenacyl dibenzyl phosphate (ethylene ketal) (4f,
3.62 g, 6.14 mmol) dissolved in 30 mL of MeOH was added 362 mg
of 10% Pd/C. The solution was hydrogenated at 10 psi of H2 for 25
min with stirring followed by the addition of 1% HCl (1 mL). After
filtration, the filtrate was evaporated in Vacuo. The resulting viscous
liquid was loaded on 5 g of a DEAE Sephadex (A-50-120, Sigma)
column pretreated with 10 M of ammonium acetate. The diammonium
salt of was eluted at 100 M ammonium acetate. Excess ammonium
acetate was lyophilized, and the resulting solid was recrystallized from
H2O/MeOH to give 1.55 g (95%) of 4-aminophenacyl diammonium
1
(1f): mp 123.5-124.5 °C; H NMR (CDCl3) δ 2.56 (s, 3H), 5.22 (s,
2H), 7.39-7.41 (m, 5H), 7.49 (d, J ) 8.7 Hz, 2H), 7.93 (d, J ) 8.7
Hz, 2H); 13C NMR (DMSO-d6) δ 26.28, 66.27, 117.22, 128.08, 128.12,
128.40, 129.49, 130.99, 136.21, 143.55, 153.11, 196.43; IR (CHCl3)
3426, 3050, 2935, 1739, 1676, 1600, 1524, 1500, 1409, 1359, 1313,
1272, 1178 cm-1; FABMS m/z (rel intensity) 185 (72), 270 (M + 1,
100); exact mass calcd for C16H15NO3 (M + H) 270.1130, found
270.1113.
1
phosphate (6f): mp 160 °C (dec); H NMR (D2O) δ 5.08 (d, J ) 5.6
Hz, 2H), 6.92 (d, J ) 8.8 Hz, 2H), 7.89 (d, J ) 8.8 Hz, 2H); 13C NMR
(D2O) δ 69.26 (d, J ) 14.2 Hz), 117.16, 126.30, 133.25, 156.14, 199.80
(d, J ) 31.6 Hz); 31P NMR (D2O) δ -1.71; IR (mineral oil) 3239,
3205, 3043, 2923, 1659, 1603, 1567, 1419, 1321, 1111 cm-1; UV-vis
(CH3CN/H2O) λmax (ꢀ) 230 (5960), 316 (19 100); FABMS (free acid)
m/z (rel intensity) 110 (18), 185 (100), 202 (32), 232 (M + 1, 22);
exact mass calcd for C8H11NO5P (free acid, M + H) 232.0375, found
232.0350.
4-((Carbobenzyloxy)amino)phenacyl Bromide (2f).25 To a solu-
tion of 4-((carbobenzyloxy)amino)acetophenone (1f, 0.92 g, 3.7 mmol)
in 30 mL of THF containing a catalytic amount of AlCl3 (10 mg) was
added 0.55 g of Br2 dropwise for 30 min at 0 °C. After completion of
the addition, the solution turned to pale yellow within 10 min. The
solvent was removed in Vacuo followed by the extraction of the resulting
solid with H2O/EtOAc. The organic layer was concentrated and purified
by silica gel column chromatography (EtOAc/hexane 30:70). Further
purification was accomplished by recrystallization from EtOAc/hexane
to give 1.1 g (93%) of 4-((carbobenzyloxy)amino)phenacyl bromide
4-Acetamidoacetophenone (1g). To a solution of 4-aminoac-
etophenone (1e, 10.0 g, 74.0 mmol) in 50 mL of dioxane/H2O (50:50)
was added acetic anhydride (6.3 g, 111.0 mg) over a 30 min period at
0 °C followed by 2 h of vigorous stirring at room temperature. The
solvent was removed in Vacuo, and the resulting solid was filtered and
rinsed with dioxane. The crude product was dried and recrystallized
from EtOAc or EtOAc/benzene to give 12.5 g (95%) of 4-acetami-
1
1
(2f): mp 165 °C (dec); H NMR (acetone-d6) δ 4.70 (s, 2H), 5.22 (s,
doacetophenone (1g): mp 165.5-166.5 °C; H NMR (DMSO-d6) δ
2H), 7.38-7.46 (m, 5H), 7.75 (d, J ) 8.8 Hz, 2H), 8.04 (d, J ) 8.8
Hz, 2H), 9.23 (s, 1H); 13C NMR (DMSO-d6) δ 33.68, 66.13, 117.31,
127.92, 128.13, 128.20, 128.43, 130.23, 136.16, 144.22, 153.08, 190.16;
IR (mineral oil) 3355, 3190, 3100, 2929, 2854, 1732, 1682, 1592, 1533,
2.06 (s, 3H), 2.52 (s, 3H), 7.71 (d, J ) 8.8 Hz, 2H), 7.91 (d, J ) 8.7
Hz, 2H), 10.29 (s, 1H); 13C NMR (DMSO-d6) δ 24.10, 26.30, 118.04,
129.38, 131.40, 143.56, 168.85, 196.35; IR (CHCl3) 3431, 3048, 2939,
1695, 1684, 1599, 1515, 1403, 1362, 1312, 1270, 1178 cm-1; FABMS
m/z (rel intensity) 120.1 (11), 136.1 (27), 178.1 (M + 1, 100); exact
mass calcd for C10H12NO2 (M + H) 178.0868, found 178.0874.
4-Acetamidophenacyl Bromide (2g),26 4-Acetamidophenacyl Diben-
zyl Phosphate (3g).27 4-Acetamidophenacyl Dibenzyl Phosphate
(Ethylene Ketal) (4g),28 and 4-Acetamidophenacyl Diammonium
Phosphate (6g).29 These were prepared according to the same
procedure used for 2f, 3f, 4f, and 6f, respectively.37
4-(Carbomethoxyamino)acetophenone (1h). To a solution of
4-aminoacetophenone (1e, 5.0 g, 37.0 mmol) and NaOH (1.5 g, 37.0
mmol) in 50 mL of dioxane/H2O (50:50) was added dropwise methyl
chloroformate (3.5 g, 37.0 mmol) for 30 min at 0 °C followed by 2 h
of vigorous stirring at room temperature. After the reaction was
complete, the solvent was removed in Vacuo. The resulting solid was
extracted with EtOAc/H2O. The organic layer was concentrated and
purified by silica gel column chromatography (EtOAc/CH2Cl2/hexane
30:50:20). Further purification was accomplished by recrystallization
from CH2Cl2/hexane to give 7.1 g (98%) of white crystals of
4-(carbomethoxyamino)acetophenone (1h): mp 160-162 °C; 1H NMR
(DMSO-d6) δ 2.51 (s, 3H), 3.70 (s, 3H), 7.59 (d, J ) 8.7 Hz, 2H),
7.90 (d, J ) 8.8 Hz, 2H), 10.07 (s, 1H); 13C NMR (DMSO-d6) δ 26.28,
51.83, 117.15, 129.49, 130.93, 143.63, 153.72, 196.33; IR (CHCl3)
3428, 3050, 2948, 1739, 1675, 1603, 1525, 1408, 1359, 1314, 1272,
1179 cm-1; FABMS m/z (rel intensity) 185.1 (100), 194.0 (M + 1,
45); exact mass calcd for C10H12NO3 (M + H) 194.0817, found
194.0800.
1460, 1415, 1284, 1219, 1181, 1045, 972 cm-1
.
4-((Carbobenzyloxy)amino)phenacyl Dibenzyl Phosphate (3f).27
To a solution of tetramethylammonium dibenzyl phosphate (0.6 g, 1.7
mmol) in hot benzene was added 0.5 g (1.44 mmol) of 4-((carboben-
zyloxy)amino)phenacyl bromide (2f). After 2 h at reflux, the resulting
solution was cooled to ambient temperature and the solvent was
removed in Vacuo. The resulting solid was purified by silica gel column
chromatography (EtOAc/CH2Cl2 20:80) to give 0.70 g (89%) of
4-((carbobenzyloxy)amino)phenacyl dibenzyl phosphate (3f): mp 112-
1
113 °C; H NMR (acetone-d6) δ 5.17 (s, 2H), 5.21 (d, J ) 5.2 Hz,
4H), 5.38 (d, J ) 11.1 Hz, 2H), 7.36-7.47 (m, 15H), 7.45 (d, J ) 8.9
Hz, 2H), 7.95 (d, J ) 8.9 Hz, 2H), 9.23 (s, 1H); 13C NMR (acetone-
d6) δ 73.29, 75.56 (d, J ) 22.2 Hz), 75.83 (d, J ) 24.6 Hz), 124.35 (d,
J ) 30.0 Hz), 134.79, 134.95, 135.01, 135.09, 135.28, 135.38, 136.04,
137.52, 143.41, 151.15, 151.25, 159.97, 197.79 (d, J ) 16.2 Hz); 31P
NMR (acetone-d6) δ -2.07; IR (CHCl3) 3430, 3048, 2928, 2879, 1737,
1698, 1598, 1525, 1500, 1438, 1410, 1209, 1178 cm-1; UV-vis (CH3-
CN) λmax (ꢀ) 289 (22 500); FABMS m/z (rel intensity) 105 (23), 117
(84), 185 (100), 277 (30), 456 (17), 546 (M + 1, 83); exact mass calcd
for C30H29NO7P (M + H) 546.1682, found 546.1703.
4-((Carbobenzyloxy)amino)phenacyl Dibenzyl Phosphate (Eth-
ylene Ketal) (4f).28 To a solution of 4-((carbobenzyloxy)amino)-
phenacyl dibenzyl phosphate (3f, 3.5 g, 6.4 mmol) in 50 mL of benzene
was added a catalytic amount of p-TsOH (60.9 mg) and an excess of
ethylene glycol (6.75 g). The solution was refluxed and stirred for 24

2462 J. Am. Chem. Soc., Vol. 119, No. 10, 1997
Park and GiVens
4-(Carbomethoxyamino)phenacyl Bromide (2h),26 4-(Car-
bomethoxyamino)phenacyl Dibenzyl Phosphate (3h),27 4-(Car-
bomethoxyamino)phenacyl Dibenzyl Phosphate (Ethylene Ketal)
(4h),28 and 4-(Carbomethoxyamino)phenacyl Diammonium Phos-
phate (6h).29 These compounds were prepared according to the same
procedure used for 2f, 3f, 4f, and 6f, respectively.37
were introduced 302 mg (0.90 mmol) of 3-methoxyphenacyl diethyl
phosphate (3b) and 0.2 g (1.88 mmol) of anisole as an internal standard.
The solution was deaerated and irradiated at 300 nm for 105 min for
MeOH and 600 min for tert-butyl alcohol solutions. Aliquots were
taken at 30 min intervals and prepared for HPLC analysis. For both
solvents, the photoproducts were isolated by silica gel column chro-
matography and identified as 3-methoxyacetophenone, 1,4-bis-[3-
methoxyphenyl]-butane-1,4-dione (9b) [1H NMR (CDCl3) δ 3.45 (s,
4H), 3.86 (s, 6H), 7.69-8.20 (m, 8H); IR (CHCl3) 3050, 2897, 1684,
1600, 1029 cm-1; mass spectrum m/e (rel intensity) 299 (28.9, M +
1), 298 (59.1, M+), 163 (32.6), 135 (100.0), 107 (73.1), 92 (54.1), 77
(59.9)], and 2,3-bis-[4-methoxyphenyl]-butan-2,3-diol (10b) [1H NMR
(CDCl3) δ 1.61 (s, 6H), 2.60 (s, 2H), 3.85 (s, 6H), 7.11-7.68 (m, 8H);
IR (CHCl3) 3300, 3050, 1606, 909 cm-1]. Further characterization of
products was not attempted.
Photolysis of 4-Hydroxyphenacyl Diethyl Phosphate (3c). Into
five Pyrex tubes containing 10 mL of MeOH or tert-butyl alcohol were
introduced 100 mg (0.46 mmol) of 4-hydroxyphenacyl diethyl phos-
phate (3c) and 0.2 g (1.88 mmol) of anisole as an internal standard.
The solution was deaerated and irradiated at 300 nm for 30 min.
Aliquots were taken at 5 min intervals and prepared for HPLC analysis.
Methyl 4-hydroxyphenylacetate2c (8c, R ) Me, in MeOH) [1H NMR
(CDCl3) δ 3.56 (s, 2H), 3.70 (s, 3H), 6.74 (d, J ) 8.5 Hz, 2H), 7.11
(d, J ) 8.5 Hz, 2H); IR (CHCl3) 3350, 3050, 1732, 1607, 1600, 1514,
1442, 1302, 1156 cm-1; mass spectrum m/e (rel intensity) 167 (5.4, M
+ 1), 166 (8.8, M+), 107 (22.8)] and tert-butyl 4-hydroxyphenylacetate
(8c, R ) t-Bu, in t-BuOH) [1H NMR (CDCl3) δ 1.43 (s, 9H), 3.45 (s,
2H), 6.75 (d, J ) 8.4 Hz, 2H), 7.11 (d, J ) 8.4, 2H), 7.26 (s, 1H); IR
(CHCl3) 3350, 3050, 2947, 1715, 1610, 1541, 1453, 1367, 1145, 958,
871, 837 cm-1; mass spectrum m/e (rel intensity) 208 (81.3, M+), 153
(13.2), 121 (18.0), 107 (99.3), 57 (100.0)] were isolated and identified
as photoproducts by the injection on HPLC with authentic samples.
Further characterization of products was not attempted. The quantum
efficiencies for the disappearance of 4-hydroxyphenacyl diethyl phos-
2,4′-Dihydroxyacetophenone (11).38 To the solution of 2-bromo-
4′-hydroxyacetophenone (2c, 0.5 g, 2.3 mmol) and formic acid (0.13
g, 2.8 mmol) in 30 mL of benzene was added DBU (0.43 g, 2.8 mmol)
at 0 °C. The reaction was monitored by TLC. After the reaction was
complete, the solvent was removed in Vacuo, and the residue was
purified by silica gel column chromatography (90% CH2Cl2/10%
1
EtOAc). The intermediate 11 was identified by H NMR [(DMSO-
d6) δ 5.49 (s, 2H), 6.88 (d, J ) 8.7 Hz, 2H), 7.86 (d, J ) 8.7 Hz, 2H),
8.40 (s, 1H), 10.53 (s, 1H)]. To a solution of the intermediate 11
dissolved in 20 mL of methanol was added sodium hydroxide (0.2 g,
4.0 mmol) at room temperature. The solution was stirred for 1 h and
neutralized by hydrochloric acid. After the solvent was removed in
Vacuo, the mixture was extracted with H2O/EtOAc. The crude product
was recrystallized from EtOAc/hexane to give 0.35 g (99%) of 2,4′-
dihydroxyacetophenone (11): mp 165-167 °C; 1H NMR (CD3CN) δ
3.4 (br, 1H), 4.72 (s, 2H), 6.87 (d, J ) 8.7 Hz, 2H), 7.81 (d, J ) 8.7
Hz, 3H); 13C NMR (CD3CN) δ 64.55, 115.09, 129.92, 161.72, 196.81;
IR (mineral oil) 3362, 3124, 2928, 1677, 1596, 1537, 1411, 1280, 1228,
1207, 1169, 1076 cm-1; HRMS (EI, CH3OH) 153* (M+ + 1), 152
(M+), 136 (M+ - 16 [O]), 121 (M+ - CH3O), 93 (M+ - C2H3O), 65.
General Procedure for Photolysis. All photochemical starting
materials were synthesized and used as described below. Acetonitrile
(HPLC grade) was used without further purification. All water was
distilled and passed through a Nanopure deionizing system. Phosphate
buffer for analytical HPLC was prepared using 85% phosphoric acid,
tetrabutylammonium phosphate (TBAP, 0.5 mM), and potassium
hydroxide to afford a solution of 0.04 M buffer at pH 6. The HPLC
system employed consisted of two pumps, a controller, an auto injector
fitted with a 10 µL loop, an UV-vis spectrophotometric detector set at
254 or 280 nm, a recorder and a C18 5 µm 250 mm × 4.6 mm column.
All analytical HPLC analyses employed either a solvent gradient or an
isocratic elution with a flow rate of 1.0-1.5 mL/min. Photolyses were
performed in a Southern New England photoreactor fitted with a merry-
go-round apparatus using 16 × 300 or 4 × 350 nm lamps. The light
output for the determination of quantum yields was measured using
the potassium ferrioxalate method.39
Samples for irradiation were placed in 20 × 180 mm Pyrex tubes
and either 5 or 10 mL of an aqueous or alcoholic solution containing
the phosphate ester was introduced into the tubes along with the
appropriate internal standard. The concentration of the phosphate was
adjusted to assure complete absorption of the incident radiation, i.e.,
greater than 3 absorbance units at the excitation wavelength. The tube
was sealed with a septum, deaerated with argon for at least 20 min at
0 °C, and photolyzed. Aliquots (100 µL) were removed periodically,
stored in the dark, and analyzed by HPLC.40
Photolysis of Phenacyl Diethyl Phosphate (3a). Into three Pyrex
tubes containing 10 mL of MeOH or tert-butyl alcohol were introduced
217 mg (0.86 mmol) of phenacyl diethyl phosphate (3a) and 0.2 g
(1.88 mmol) of anisole as an internal standard. The solution was
deaerated and irradiated at 300 nm for 3 h for MeOH and 48 h for
tert-butyl alcohol solutions. Aliquots were taken at 30 min intervals
and prepared for HPLC analysis. For both solvents, the photoproducts
were isolated by silica gel column chromatography and identified as
acetophenone and 1,4-diphenylbutane-1,4-dione (9a) [1H NMR (CDCl3)
δ 3.48 (s, 4H), 7.56-8.05 (m, 10H); IR (CHCl3) 3050, 1682, 1600,
1050 cm-1; mass spectrum m/e (rel intensity) 239 (18.1, M + 1), 238
(78.7, M+), 149 (20.1), 133 (38.7), 105 (100.0), 77 (53.5), 51 (22.0)],
and acetophenone. Further characterization of the products was not
attempted.
phate (3c), ΦDis, the appearance of 4-hydroxyacetophenone (1c), Φred
,
and methyl 4-hydroxyphenylacetate (8c, R ) CH3), Φrearr, are shown
in Table 1.
Photolysis of 4-Hydroxyphenacyl Diammonium Phosphate (6c).
Into five Pyrex tubes containing 10 mL of 10% CH3CN in buffer (pH
7.3) were introduced 100 mg (0.38 mmol) of 4-hydroxyphenacyl
diammonium phosphate (6c) and 0.2 g (1.88 mmol) of anisole as an
internal standard. The solution was deaerated and irradiated at 300
nm for 30 min. Aliquots were taken at 5 min intervals and prepared
for HPLC analysis. The only photoproduct observed, 4-hydroxyphe-
nylacetic acid (8), was identified by co-injection on HPLC with
authentic sample. The quantum efficiencies for the disappearance of
4-hydroxyphenacyl diammonium phosphate (6c), ΦDis, and the appear-
ance of 4-hydroxyphenylacetic acid (8), Φrearr, are shown in Table 1.
Photolysis of 4-Hydroxyphenacyl Adenosine 5′-Triphosphate,
Triammonium Salt (7). Into three Pyrex tubes containing 5 mL of
Tris buffer (0.05 M, pH 7.3) were introduced 20 mg (28.9 µmol) of
4-hydroxyphenacyl adenosine 5′-triphosphate, triammonium salt (7) and
15.3 mg (125 µmol) of benzoic acid as an internal standard. The
solution was deaerated and irradiated at 300 nm for 12 min. Aliquots
were taken at 3 min intervals and prepared for HPLC analysis.
4-Hydroxyphenylacetic acid (8) and ATP were identified as photo-
products by co-injection on HPLC with authentic samples. The
quantum efficiencies for the disappearance of triammonium salt (7),
Φ
Dis, and the appearance of 4-hydroxyphenylacetic acid (8), Φrearr, and
ATP, ΦATP, are shown in Table 1.
Photolysis of 4-hydroxyphenacyl adenosine 5′-triphosphate, triam-
monium salt (7) in lactated Ringer’s solution Into a Pyrex tube
containing 5 mL of Ringer’s solution (each 100 mL contains 5 g of
dextrose hydrous, 600 mg of sodium chloride, 310 mg of sodium lactate,
30 mg of potassium chloride, and 20 mg of calcium chloride, pH 6.5)
was introduced 20 mg (28.9 µmol) of 4-hydroxyphenacyl adenosine
5′-triphosphate, triammonium salt (7). The solution was deaerated and
irradiated at 300 nm for 12 min. The resulting solution was prepared
for HPLC analysis. 4-Hydroxyphenylacetic acid (8c, R ) H) and ATP
were identified as photoproducts by co-injection on HPLC with
authentic samples. Quantitative measurements were made during the
quenching studies given below.
Photolysis of 3-Methoxyphenacyl Diethyl Phosphate (3b). Into
three Pyrex tubes containing 10 mL of MeOH or tert-butyl alcohol
(38) (a) Fodor, G.; Kovacs, O.; Mercher, T. Acta Chim. Sci. Hung. 1951,
1, 395-402. (b) Patzlaff, M.; Barz, W. Z. Naturforsch. 1978, 33c, 675-
684.
(39) Hatchard, C. G.; Parker, C. A. Proc. R. Soc. London A 1956, 235,
518-522.
(40) Hoffman, N. E.; Liao, J. C. Anal. Chem. 1977, 49, 2231-2234.

New PhotoactiVated Protecting Groups. 6
J. Am. Chem. Soc., Vol. 119, No. 10, 1997 2463
Competitive Quenching Study of 7 with Sodium 2-Naphthale-
nesulfonate. To a 10 mL volumetric flask were added 15 mg (21.7
µmol) of 4-hydroxyphenacyl adenosine 5′-triphosphate, triammonium
salt (7), 61 mg (0.5 mmol) of benzoic acid, and Tris buffer (1 M, pH
7.3) to the fill line. To each of four Pyrex tubes was added 1 mL of
the above solution. To three of these tubes was added 11.5 mg (50
µmol), 23.0 mg (100 µmol), or 34.5 mg (150 µmol) of sodium
2-naphthalene sulfonate. The tubes were deaerated for 20 min with
argon at room temperature and photolyzed using 4 × RPR 350 nm
lamps; the reactions were monitored by HPLC. A least-squares analysis
was performed on the data obtained. The results are given in Table 2.
Photolysis of 4-Aminophenacyl Diammonium Phosphate (6f).
Into each of four Pyrex tubes containing 10 mL of 10% CH3CN/Tris
buffer (50 mM, pH 7.3) was introduced 50 mg (0.19 mmol) of
4-aminophenacyl diammonium phosphate (6f) and 0.2 g (1.88 mmol)
of anisole as an internal standard. One of the four phototubes was
wrapped with aluminum foil and used as a thermal control for reference.
The solutions were deaerated and irradiated at 300 nm for 30 min.
Aliquots were taken at 5 min intervals. All aliquots were diluted for
HPLC analysis. Only one product, 4-aminoacetophenone (1e), was
identified by co-injection on HPLC with authentic sample. The
quantum efficiencies for the disappearance of 4-aminophenacyl diam-
monium phosphate (6f) and the appearance 4-aminoacetophenone (1e)
are shown in Table 3. The reference tube did not show any product.
Photolysis of 4-acetamidophenacyl diammonium phosphate (6g)
and 4-carbomethoxyaminophenacyl diammonium phosphate (6h)
were performed out using the same procedures employed for 6f.37
Stability Test of 4-Hydroxyphenacyl Adenosine 5′-Triphosphate,
Triammonium Salt (7) in Various Media. Into each of six 10 mL
of test tubes, respectively, containing 5 mL of H2O, D2O, Tris Buffer
(50 mM, pH 7.3), Ringer’s solution I (each 100 mL contained 600 mg
of sodium chloride, 30 mg of potassium chloride, and 20 mg of calcium
chloride, pH 6.5), Ringer’s solution II (each 100 mL contained I and
310 mg of sodium lactate, pH 6.5), and Ringer’s solution III (each
100 mL contained (II) and 5 g of dextrose hydrous, pH 6.5) was
introduced 5 mg (7.2 µmol) of 4-hydroxyphenacyl adenosine 5′-
triphosphate, triammonium salt (7). Solutions were maintained at
ambient temperature for 24 h. Aliquots were removed periodically to
determine the stability of 7 in the six different media by HPLC. In all
cases, 7 was stable at least 24 h or longer.
pH Test of Three Phenacyl Diammonium Phosphates. Into each
of three Pyrex tubes containing 5 mL of H2O were introduced 10 mg
(37.6 µmol) of 4-hydroxyphenacyl diammonium phosphate (6c), 10
mg (37.6 µmol) of 4-aminophenacyl diammonium phosphate (6f), and
11.5 mg (37.6 µmol) of 4-acetamidophenacyl diammonium phosphate
(6g). The pH of each solution was determined to be 7.4, 6.8, and 4.7,
respectively. The solutions were deaerated and irradiated at 300 nm
for 10 min. The resulting photolysis solutions gave pH measurements
of 6.7, 6.7, and 3.6, respectively (Table 4).
31P Experiments of p-Substituted Phenacyl Diammonium Phos-
phates for the Release of Inorganic Phosphate (Pi). Each (2 mg) of
three phosphates, 4-aminophenacyl diammonium phosphate (6f), 4-ac-
etamidophenacyl diammonium phosphate (6g), and 4-(carbomethoxyami-
no)phenacyl diammonium phosphate (6h) was introduced to three NMR
tubes, respectively. All of the tubes were filled with 1 mL of D2O,
and the solutions were deaerated and irradiated for 10 min. 31P NMR
analyses were conducted before and after the irradiation. Since the
31P reference peak (measured by 85% phosphoric acid) was not stable,
the analyzing peaks being evaluated often changed Pi chemical shifts
slightly. However, all three samples showed that the phosphoric acid
was released upon irradiation. All of the released Pi peaks were
observed between 0.2 and 0.0 ppm in the NMR spectra.
Acknowledgment. This material is based on work supported
by the National Science Foundation under grant no. NSF/OSR-
9255223. Additional support from PRF Grant No. 17541-AC4
is also gratefully acknowledged. We thank Dr. Andreas Jung
for obtaining the spectral data for 11.
Supporting Information Available: Experimental and
spectral details for 1f-h, 2f-h, 3a-c, f-h, 4g-h, and 6g-h
and the photochemical experimental details for 6g and h (14
pages). See any current masthead page for ordering and Internet
access instructions.
JA9635589