Photorelease of phosphates: Mild methods for protecting phosphate derivatives

Article abstract of DOI:10.3762/bjoc.10.212

We have developed a new photoremovable protecting group for caging phosphates in the near UV. Diethyl 2-(4-hydroxy-1-naphthyl)- 2-oxoethyl phosphate (14a) quantitatively releases diethyl phosphate upon irradiation in aq MeOH or aq MeCN at 350 nm, with quantum efficiencies ranging from 0.021 to 0.067 depending on the solvent composition. The deprotection reactions originate from the triplet excited state, are robust under ambient conditions and can be carried on to 100% conversion. Similar results were found with diethyl 2-(4-methoxy-1-naphthyl)-2-oxoethyl phosphate (14b), although it was significantly less efficient compared with 14a. A key step in the deprotection reaction in aq MeOH is considered to be a Favorskii rearrangement of the naphthyl ketone motif of 14a,b to naphthylacetate esters 25 and 26. Disruption of the ketone-naphthyl ring conjugation significantly shifts the photoproduct absorption away from the effective incident wavelength for decaging of 14, driving the reaction to completion. The Favorskii rearrangement does not occur in aqueous acetonitrile although diethyl phosphate is released. Other substitution patterns on the naphthyl or quinolin-5-yl core, such as the 2,6-naphthyl 10 or 8-benzyloxyquinolin-5-yl 24 platforms, also do not rearrange by aryl migration upon photolysis and, therefore, do not proceed to completion. The 2,6-naphthyl ketone platform instead remains intact whereas the quinolin-5-yl ketone fragments to a much more complex, highly absorbing reaction mixture that competes for the incident light.

Full text of DOI:10.3762/bjoc.10.212

Photorelease of phosphates: Mild methods for protecting
phosphate derivatives
Sanjeewa N. Senadheera§, Abraham L. Yousef¶ and Richard S. Givens*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2014, 10, 2038–2054.
Department of Chemistry, University of Kansas, Lawrence, Kansas
66045, U.S.A
doi:10.3762/bjoc.10.212
Received: 22 May 2014
Accepted: 07 August 2014
Published: 29 August 2014
Email:
Richard S. Givens* - givensr@ku.edu
* Corresponding author
This article is part of the Thematic Series "Organophosphorus chemistry".
Guest Editor: P. R. Hanson
§ Present address: Department of Medicinal Chemistry, University of
Kansas, Lawrence, Kansas 66045, U.S.A.
¶ Present address: Department of Chemistry, Sweet Briar College,
Sweet Briar, VA 24595, U.S.A.
© 2014 Senadheera et al; licensee Beilstein-Institut.
License and terms: see end of document.
Keywords:
caged phosphates; hydroxynaphthylacetyl; organophosphorus;
photo-Favorskii rearrangement; photoremovable protecting groups
Abstract
We have developed a new photoremovable protecting group for caging phosphates in the near UV. Diethyl 2-(4-hydroxy-1-naph-
thyl)-2-oxoethyl phosphate (14a) quantitatively releases diethyl phosphate upon irradiation in aq MeOH or aq MeCN at 350 nm,
with quantum efficiencies ranging from 0.021 to 0.067 depending on the solvent composition. The deprotection reactions originate
from the triplet excited state, are robust under ambient conditions and can be carried on to 100% conversion. Similar results were
found with diethyl 2-(4-methoxy-1-naphthyl)-2-oxoethyl phosphate (14b), although it was significantly less efficient compared
with 14a. A key step in the deprotection reaction in aq MeOH is considered to be a Favorskii rearrangement of the naphthyl ketone
motif of 14a,b to naphthylacetate esters 25 and 26. Disruption of the ketone-naphthyl ring conjugation significantly shifts the
photoproduct absorption away from the effective incident wavelength for decaging of 14, driving the reaction to completion. The
Favorskii rearrangement does not occur in aqueous acetonitrile although diethyl phosphate is released. Other substitution patterns
on the naphthyl or quinolin-5-yl core, such as the 2,6-naphthyl 10 or 8-benzyloxyquinolin-5-yl 24 platforms, also do not rearrange
by aryl migration upon photolysis and, therefore, do not proceed to completion. The 2,6-naphthyl ketone platform instead remains
intact whereas the quinolin-5-yl ketone fragments to a much more complex, highly absorbing reaction mixture that competes for the
incident light.
Introduction
Phosphates have long held an important formative position in in the photosolvolysis of aryl phosphates that showed unexpect-
the development of organic photochemistry beginning with the edly pronounced meta activation of substituted aryl phosphates.
seminal report by Havinga [1] of the unusual substituent effects A strong ‘meta effect’ resulting in enhanced reaction efficiency
2038

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
by electron withdrawing, meta substituents is contrary to their photoremovable protecting groups (PPGs) in which the at-
ground state effects on solvolysis reactions which, instead, tached chromophore masks the biological activity of a substrate.
display enhanced reactivity for para substituents. This Irradiation releases the substrate, allowing it to return to its
contrasting substituent reactivity is a consequence of two normal bioactivity.
different potential energy surface (PES) contours that control
the reactivity, an observation first rationalized by Zimmerman The desirable properties of any new PPG candidate are 1)
through a comparison of the change in PES electron distribu- absorption at wavelengths near or above 400 nm, 2) enhanced
tion in the ground (HOMO) and excited states (LUMO) using chemical and photochemical quantum yields and 3) improved
Hückel molecular orbital theory [2,3], subsequently attributed rate of release, ideally in the range of ps to ns time constants; all
by him to the positioning of conical intersections between the three are useful properties for applications in organic chemistry
HOMO–LUMO PE surfaces [4,5]. He termed this phenomenon and biochemistry, but they are particularly important for mecha-
the “meta effect”, and generalized it for photochemical solvol- nistic studies in biochemistry and biophysics [6,8-11]. Most
ysis reactions. Several additional photosolvolysis studies PPG candidates require sufficiently high energy excited states
substantiated the generality of the meta effect for phosphate for heterolysis of a carbon–oxygen or carbon–heteroatom bond
leaving groups in heterolytic photoreactions.
that binds the chromophore to a substrate. This limits the useful
absorption range to ca. 350 to 400 nm for heterolysis by a pri-
Over the past four decades, additional examples of phosphate mary photochemical pathway.
esters attached to reactive chromophores other than phenyl and
benzyl [6-11] have been investigated for their propensity to Phosphates hold a central historic position in caged photochem-
undergo heterolytic photosolvolysis reactions, most notably istry through their cross-disciplinary significance in both
o-nitrobenzyl (o-NB) [12], benzyl phenyl ketone (benzoin) biology and chemistry. Nucleotides (especially ATP, cAMP,
[13,14], coumarin-4-ylmethyl [15], and, more recently, and GTP) were among the first to be covalently bound to chro-
p-hydroxyphenacyl (pHP) [15] phosphate esters (Figure 1). mophores in a cage or PPG format that were demonstrably
While these chromophores exhibit a range of photophysical released upon photolysis. Benzoin caged cAMP and o-nitro-
properties, all share a conjugated aromatic structural motif that phenethyl caged ATP, seminal examples of caged phosphates,
facilitates UV–vis absorption and serves as a traceless reagents, are often the two classic caged biochemical substrates cited
orthogonal to common ground state deprotection processes and [6-13,15-17]. Subsequent interest in and application of PPGs
essentially independent of pH effects. Thus, the chromophores has exploded largely because they provide researchers in inter-
are particularly useful at neutral pH and no added reagents are disciplinary fields with a tool for initiating biological [18] and
required for protecting group release.
chemical processes [6,8,10,11] by remote control with light as
the activator. Spatial, temporal and concentration parameters for
Many photosolvolysis studies have targeted key biological releasing substrates are controlled by adjusting the focus, time
substrates [6,8-12] for mechanistic and phenomenological resolution, and intensity of the light source in conjunction with
studies. The culmination of these studies has resulted in the other variables such as the photochemical reactivity and molec-
development of an entirely new branch of photochemistry with ular reaction pathway. Since photolysis reactions are generally
application in both organic chemistry and biochemistry. The considered to be kinetically much faster than ground state
reagents are generally described as “caged” compounds or processes (most photorelease rates have sub μsec time
Figure 1: Common photoremovable protecting groups (PPGs) for phosphates depicted as diethyl phosphate (DEP) esters.
2039

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
constants), the use of time resolved lasers to activate PPGs broader absorption band closer to the visible region, making the
offers much greater temporal resolution and more precise PPG more accessible for photodeprotection. Thus, elaborating
spatial control of the reaction variables in chemical processes, the two motifs by optimally appending a hydroxy and carbonyl
increasingly important features for biophysics and biochem- positioned for excited state interaction, in accord with the pHP
istry.
motif 4a, invited serious examination.
Results
Our studies on photosolvolysis reactions have uncovered
several new chromophores with superior PPG properties Synthesis of phosphate esters
amenable to photodeprotection. Earlier work defined the advan- A series of diethyl phosphate (DEP) esters caged with extended
tages and limitations of p-hydroxyphenacyl (pHP, 4), a PPG 2-(6-hydroxynaphthalen-2-yl)-2-oxoethyl (2,6-HNA, 10), 2-(4-
that is now finding application in biochemistry and chemistry hydroxynaphthalen-1-yl)-2-oxoethyl (1,4-HNA, 14a) and its
[15]. We now report new pHP analogs with a fused aromatic or methoxy ether (1,4-MNA, 14b), and 8-(benzyloxy)quinolin-5-
heteroaromatic extension of pHP. Our intent was to impose the yl)-2-oxoethyl (5,8-BQA, 24), each modeled after the
pHP motif on the naphthyl and indolin-5-yl platforms (main- p-hydroxyphenacyl (pHP, 4) chromophore were synthesized as
taining the critical substituent pattern) in order to foster aryl illustrated in Scheme 1 and Scheme 2.
ketone migration by a photo-Favorskii rearrangement concomi-
tant with ligand release. The combination of these features Synthesis of the extended 2,6-HNA phosphate ester 10 was
shifts and extends the chromophore, exhibiting a more intense, accomplished by first demethylating 2-acetyl-6-methoxynaph-
Scheme 1: Synthesis of 2,6-HNA DEP (10), 1,4-HNA DEP (14a), and 1,4-MNA DEP (14b) DEP esters. Reagents and conditions: a) PhSH, K2CO3,
N-methyl-2-pyrrolidinone, 194 °C, 45 min; b) TBSCl, Et3N, DBU, CH2Cl2, rt, 28 h; c) CuBr2, 1:1 CHCl3/EtOAc; d) KHSO4, MeOH (aq), CH2Cl2, rt,
43 h; e) tetramethylammonium diethyl phosphate, DMF, 55 °C, 2–3 h; f) Ac2O, AlCl3, CH2Cl2, 0 °C to rt over 3 h then stirred at rt.
2040

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
thalene (5) with thiophenol and 0.1 mol percent potassium quantitatively protecting the hydroxy group as its benzyl ether
carbonate in N-methyl-2-pyrrolidinone at 194 °C [19] to [24] followed by installation of ketal protection with iodoben-
generate the hydroxynaphthalene 6 (72%) followed by zene diacetate in alkaline methanol [25] to give 22 (90%,
protecting the hydroxy group with TBSCl to give 7 (89%, Scheme 2). Deprotection of 22 with 50% acetic acid gave the
Scheme 1A). α-Bromination with copper(II) bromide [20] gave α-hydroxy ketone 23 (98%) which was esterified with diethyl
8 (96%, direct bromination of 6 gave poor yields). The TBS phosphoryl chloride in pyridine [26] to afford 2-(8-
group was removed in 30% aqueous methanol containing (benzyloxy)quinolin-5-yl)-2-oxoethyl diethyl phosphate (5,8-
10–20% methylene chloride and several molar equivalents of BQA DEP, 24, 30%).
potassium hydrogen sulfate to provide hydroxy bromo ketone 9
(yield, 75%) [21], which was reacted with tetramethylammo- The UV–vis spectra of 1,4-substituted esters 14a,b (λmax range
nium diethyl phosphate under anhydrous conditions to afford from 319 to 325 nm) are substantially red-shifted relative to the
diethyl (2-(6-hydroxynaphthalen-2-yl)-2-oxoethyl) phosphate pHP ester 4a (λmax at 287 nm) and exhibit molar extinction
(2,6-HNA DEP, 10, 62%) [22].
coefficients (ε) of ca. 104 M−1cm−1 in aq MeCN (Figure 2). For
the 2,6-HNA DEP (10) and 2,6-HNA GABA, the λmax appears
Diethyl (2-(4-methoxynaphthalen-1-yl)-2-oxoethyl) phosphate at 325 nm whereas the λmax for the 5,8-BQA phosphate 24
(1,4-MNA DEP, 14b) was obtained, as shown in Scheme 1B, occurs at 321 nm in 10% aq MeCN. For the 2,6-HNA series, a
beginning with Friedel–Crafts acylation of 1-methoxynaphtha- strong fluorescence emission is observed at 470 nm as shown
lene (11) yielding 1-acetyl-4-methoxynaphthalene (12, 73%). here for the more aqueous soluble GABA ester derivative. The
α-Bromination with copper(II) bromide gave 13 (44%) which 2,6-HNA GABA had better aqueous solubility for fluorescence
was converted to the phosphate ester using tetramethylammo- studies. (See Supporting Information File 1 for synthetic and
nium diethyl phosphate in dimethoxyethane (DME) at room spectral details).
temperature to give diethyl (2-(4-methoxynaphthalen-1-yl)-2-
oxoethyl) phosphate (1,4-MNA DEP, 14b, 92%).
Photolysis of phosphate esters
Irradiation of 14a at 350 nm in Pyrex vessels under ambient
The hydroxy analog (1,4-HNA DEP, 14a, Scheme 1C), was conditions in 1% aq CD3OD resulted in the release of diethyl
synthesized from methoxynaphthone 12 by demethylation with phosphate as confirmed by NMR (Scheme 3, Table 1). The
thiophenol giving acetylnaphthol 15 in moderate yield (42%). protecting group underwent a photo-Favorskii rearrangement
Protection with TBSCl gave 16 (86%) followed by α-bro- (Scheme 4), yielding predominantly methyl 4-hydroxy-1-naph-
mination to 17 (89%). TBS deprotection gave 2-bromo-1-(4- thylacetate (25) with a trace amount of reduction product 15
hydroxy-1-naphthyl)ethanone, (18, 59%) which was converted (Scheme 3). In anhydrous methanol, the ratio of 25/15 was
to diethyl (2-(4-hydroxynaphthalen-1-yl)-2-oxoethyl) phos- approximately 3:1. When the photolysis was performed in 10%
phate (1,4-HNA DEP, 14a, 31%) with tetramethylammonium aq MeCN, 15 was the only assigned structure from the 1H NMR
diethyl phosphate.
spectrum of a complex product mixture.
Benzyl protected diethyl 2-(8-hydroxyquinolin-5-yl)-2-oxoethyl Photolysis of the methoxy analog 14b under similar conditions
phosphate (24) was synthesized from 8-hydroxyquinoline (19) either in CD3OD or 1% aq MeCN resulted in the release of DEP
by first performing acylation [23] with acetyl chloride and then and formation of 12 as confirmed by reversed-phase HPLC
Scheme 2: Synthesis of diethyl 8-(benzyloxy)quinolin-5-yl)-2-oxoethyl phosphate (5,8-BQA DEP, 24). Reagents and conditions: (a) CH3COCl, AlCl3,
nitrobenzene, 14 h, 40%; (b) PhCH2Br, K2CO3, DMF, rt, 24 h, 100%; (c) PhI(OAc)2, MeOH, KOH, 0 °C to rt, 24 h, 90%; (d) 50% CH3COOH, rt, 12 h,
98%; (e) P(O)Cl(OEt)2, pyridine, −5 °C to rt, 24 h, 30%.
2041

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
Figure 2: A. UV–vis spectrum of 14a (1,4-HNA DEP) in 1% aq MeCN. B. Fluorescence emission/excitation spectra of 2,6-HNA GABA (0.042 mM) in
pH 7.3 TRIS buffer containing 1% MeCN.
Scheme 3: Photolysis of 1,4-HNA and 1,4-MNA diethyl phosphates 14a and 14b in aq MeOH.
alkylation of 1-methoxynaphthalene (11) with methyl bromo-
Table 1: Photolysis efficiencies and product yields for 1,4-HNA DEP
(14a) and 1,4-MNA DEP (14b).
acetate and ferric oxide [27].
ester
% conversiona
deprotection
% yieldb
Φdisc
Quantum yields for disappearance and percentage conversions
for 14a are higher than that of 14b as indicated in Table 1. After
one hour of photolysis at 350 nm in CD3OD, 14a reached 98%
conversion whereas 14b managed to attain only 40% conver-
sion. The disappearance quantum efficiency (Φdis) for 14a in
1% aq CD3OD was 0.028, more than three times that for 14b
under the same conditions. However, the deprotection yields for
both esters were excellent, resulting in quantitative release of
DEP based on conversion.
14a OH
98
40
100
90
0.028
14b OMe
0.0076
aOne hour photolysis in CD3OD at 350 nm, determined by 1H NMR
using DMF as an internal standard, bdiethyl phosphate, corrected for
conversion, cdetermined by RP-HPLC; photolysis in 1% aqueous
MeOH.
(RP-HPLC) and NMR (Scheme 3, Table 1) but the conversion
was not complete and the reaction was lethargic due to competi- The quantitative conversion and apparent good efficiency of the
tive absorption by the unrearranged product chromophore 12. In 1,4-derivative 14a prompted further investigation. The quantum
1% aq MeOH, however, the major product from the protecting yield for disappearance of 14a in 1% aq MeCN (Φdis = 0.021) is
group was methyl 4-methoxy-1-naphthylacetate (26) which was comparable to that observed in aq MeOH (Φdis = 0.028, Table 1
characterized by the signals at δ 4.02 and 3.68 ppm and and Table 2) but the product mixture was more complex (vide
confirmed by FAB–MS (m/z 230) and comparison with an inde- infra). Interestingly, quantum yields for the disappearance for
pendently synthesized sample obtained from Friedel–Crafts 14a are enhanced three-fold (Φdis = 0.067) when the photolysis
2042

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
Scheme 4: The photo-Favorskii rearrangement of 14a.
solution was purged free of oxygen (Table 2). The quantum Irradiation of 2,6-HNA DEP (10) at 350 nm in 1% aq MeCN
yield of 14a for the triplet sensitized photolysis in the presence under conditions comparable to those employed with 14a,b also
of benzophenone (BP), a well-established triplet sensitizer, at released phosphate. The disappearance quantum yield for 10
254 nm under O2-free conditions was Φdis = 0.022. These (Φdis = 0.031) was nearly the same as that obtained for 14a but
combined results demonstrate that the reactive excited state is a the conversion after a 10 min irradiation was only 67%. Unlike
triplet that is partially quenched by O2 under ambient condi- 14a,b, however, the chromophore did not undergo the photo-
tions.
Favorskii rearrangement, forming only the reduction product
2-acetyl-6-hydroxynaphthalene (6) in addition to other unidenti-
fied photoproducts (Scheme 5). These results paralleled our
observations for 14a,b in aq MeCN.
Table 2: Photolysis of 1,4-HNA diethyl phosphate (14a) in the pres-
ence and absence of O2 and benzophenone in 1% aq MeCN.
Φdisa
conditions
Finally, photolysis of 5,8-BQA diethyl phosphate (24) in 10%
aq MeCN at 300 or at 350 nm under degassed conditions
released DEP at only 30% conversion, even after 18 hours as
confirmed by both 1H and 31P NMR analyses (Scheme 6). The
quantum yields for disappearance and product appearance
(Φdis = Φapp = 0.00024) were exceptionally low (Table 3),
considerably less than any of the naphthyl (10 and 14) or
0.067b
0.022
0.021
degassed with Ar
BP sensitizedc
saturated with Air
aDetermined with RP-HPLC; bEstimated value (see experimental
section); cBenzophenone (11 mM) as the sensitizer and 14a (0.59 mM)
in 1% aq MeCN at 254 nm and purged with Ar.
Scheme 5: Photolysis of 2,6-HNA DEP (10) in 1% aq MeCN.
Scheme 6: Photolysis of 5,8-BQA diethyl phosphate (24).
2043

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
Table 3: Quantum yields and conversions of 24 with and without Ar purging.a
% conversionb
ambient O2
% conversionb
Ar purged
Φdis (× 103)c
ambient O2
Φapp (× 103)d
ambient O2
ester
24
32
30
0.24
0.24
aSolvent was 10% H2O in MeCN. Photolyses were conducted with 300 or 350 nm lamps. bDetermined using 1H NMR in 10% D2O in CD3CN;
cDetermined with 1H NMR for the disappearance of the starting material at 350 nm with air; dDetermined with 1H NMR for the appearance of diethyl
phosphate at 350 nm with air.
phenacyl (4a) analogs (see Table 4, Discussion section). A or rearrange to more strongly absorbing chromophores which
complex product mixture of byproducts of the indolinyl compete for the effective radiation wavelengths, often resulting
chromophore was obtained which was not investigated further. in incomplete conversions that compromise their synthetic
utility. These chromophores and especially their photoproducts
Discussion
are also highly fluorescent, which frequently prove to be disad-
In our design of new PPG chromophores, we have attempted to vantageous.
circumvent or avoid the inherent deficiencies of the classical
o-NB, benzoin, and coumaryl analogs (see Figure 1). A The recent candidates are based on a pHP motif. Phosphate
noisome limitation of o-NB PPGs that mitigate their general use esters of 10, 14, 24, and acetate 27 [19] each possess a hydroxy
in biological and mechanistic studies are their inherently slow donor coupled with an acetyl function bearing the leaving group
reaction rates. o-NB rate constants are decidedly among the (Figure 3). One of these, the 5-hydroxy-1-naphthylacetyl motif
slowest for a photochemical heterolysis (e.g., usually not larger as its acetate ester 27, has already been reported to be photo-
than μsec time constants) due to mechanistic constraints chemically inert in 1:1 aq MeCN by Wan [19]. Accordingly, we
imposed by the rate determining step of a ground state hemiac- did not pursue the study of 27 (vide infra). Extending the
etal hydrolysis, generated photochemically, that releases the aromatic portion of the chromophore with either an added
substrate. Furthermore, the o-NB PPG chromophore, itself, is benzo or pyridyl group using the pHP model red-shifted the
converted to an o-nitroso aldehyde (or ketone) that reacts with chromophore absorption for the remaining three, 10, 14, and 24
endogenous nucleophiles such as amines, thiols, etc. either (e.g., the maximum of 4a at 280 nm shifted to 325 nm for 14a
present on the substrate, on target proteins or in the media. tailing nearly to 400 nm, see Figure 2).
Poignant evidence of these limitations surfaced during mecha-
nistic investigations of GAP protein GTPases and enzymatic Similar extensions of the benzyl PPG expanding it to naphthyl-
studies of ATPase [28-34].
methyl [35,36] (28 and 29, Figure 4) or modifications by
attaching substituents to phenacyl (e.g., 4b,c) or benzyl (30b–d)
We [15,28,29,35] and others [6-11] have successfully offered moieties, and naphthylacetyl [37] reportedly lowered the singlet
alternatives to the o-NB PPG series, initially with benzoin phos- and triplet state energies only modestly [10,11] and were unsuit-
phate 2 [13,14] and coumarin-4-ylmethyl phosphate 3. able because these chromophores remained intact and in compe-
However, the chromophores for these candidates remain intact tition for incident radiation.
Figure 3: Naphthyl and quinolin-5-yl caged phosphate esters 10, 14, 24 and 27 (acetate ester).
2044

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
Figure 4: Previously studied caged diethyl phosphate PPGs possessing aromatic (benzyl, phenacyl, and naphthylmethyl) phosphates.
Photolysis of 1,4-HNA DEP (14a) in 1% aq MeOH released In contrast to the results in 1% aq MeOH, photolysis of 14a,b in
DEP with modest efficiency, Φ = 0.028. Similarly, the methoxy anhydrous media or changing the co-solvent to MeCN resulted
analog, 1,4-MNA DEP (14b), released DEP, but with a much in a decrease in conversion (40–60%) and more complex reac-
lower efficiency (Φ = 0.0078). Both quantitatively produced tion mixtures. The only products identified were DEP (quanti-
DEP and, pleasingly, underwent photo-Favorskii rearrange- tative) and reduction products 12 or 15, signaling a change in
ments converting the aryl ketones to methyl naphthylacetates, mechanism from heterolysis to homolysis. The low conver-
blue-shifting the byproduct chromophore absorptions.
sions are primarily due to competition for incident radiation by
the unrearranged byproduct chromophore 12 or 15.
Both 14a and 14b now become the first new examples of photo-
Favorskii rearrangements, here on a naphthyl platform. The role of H2O and hydroxylic solvents observed in this series
Previous examples are limited to phenacyl frameworks is in complete accord with solvent effects on the parent pHP
[6-9,15,38,39]. It is interesting to note, however, that replace- photoreactions [6,13,15,40-46]. The photo-Favorskii reaction
ment of the OH with OMe in the 1,4-HNA series results in both (Scheme 7) is favored by aqueous-based solvents that serve
lower efficiency and decreased yield, a further analogy to that both as a proton donor and an acceptor to the conjugate base
found with p-methoxy vs p-hydroxyphenacyl DEPs 4a,b generated from the chromophore’s triplet state heterolytic
(Table 4) [6,13,15,40] and with other leaving groups.
cleavage of the leaving group. The dual behavior of H2O is
Scheme 7: Photo-Favorskii mechanism based on pHP DEP 4a photochemistry as applied to 1,4-HNA DEP (14a).
2045

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
manifest in accepting a proton from the naphthol OH while (ET = 59.5 kcal/mol) are higher energy contributors than the
simultaneously solvating the developing anionic charge on the same two components of the 1,4 and 1,5 patterns, 1-naphthol
departing nucleofuge. The resulting biradical 31, which must be (ET = 58.6 kcal/mol) and 1-acetylnaphthalene (ET =
a triplet by Wigner’s spin rule [47] is formed from the chromo- 56.4 kcal/mol) [10,11]. Of the two motifs, the 1,4 arrangement
phore while on the excited state PES. Intersystem crossing (isc) has the lower energy triplet. Yet the 2,6 pattern with the higher
to a diradicaloid ground state intermediate 32 (vide infra) is fol- triplet energy is the less reactive and fails to participate in the
lowed by cyclization to form the ‘Favorskii intermediate’ 33 photo-Favorskii rearrangement.
[6,8,15,26,38,39,48].
Wan’s work clearly demonstrates, however, that the excited
In contrast to the 1,4-naphthyl design, neither 10 nor 24 under- state acidity of 1-naphthols is much greater than 2-naphthols
went Favorskii rearrangements. While both gave stoichiometric and that excited-state intramolecular proton transfer (ESIPT) for
release of DEP, the chromophores either remain intact (for 10) 1-naphthols occurs at both the 5- and 8-positions by H–D
as also reported for other 2-naphthylacetyl analogs [49] or are exchange [19]. Furthermore, dehydration of 5-(1-hydroxy-
consumed by oxidation and fragmentation as in the case of 24. ethyl)-1-naphthol (34) is very efficient, leading to quinone
Thus, the conversion and chemical yields were low and the methide 35 which is trapped by solvent MeOH (Scheme 8). The
photolysates were complex mixtures of byproducts. In the case corresponding 4-(1-hydroxyethyl)-1-naphthol, however, is unre-
of 24, although the reaction efficiency was insensitive to O2 active. Comparisons of the two photoreactions, photodehydra-
(implying singlet reactivity, not quenchable by ambient O2), the tion and photo-Favorskii [48,50], and Wan’s results [19] for
product mixture included several oxidation byproducts of the H–D exchange reactions demonstrate the importance of the
chromophore. Thus, photoreactions of 24 with ambient O2 were relative location of the two functional groups and the role of
a complication not encountered with 10 or 14a,b.
ESIPT for both reactions. The hydroxy group, as Wan has
elegantly determined, is the source of the acidic proton (espe-
The failure to deconjugate the carbonyl of either chromophore cially for singlet state reactions). The carbonyl appendage,
by rearrangement once again impinged on the overall conver- however, must be responsible for the enhanced heterolysis effi-
sion to released substrate. The formation of competing chro- ciency [28,29,38] and may influence the formation of the triplet
mophoric byproducts limits the conversion and yields excited state through intersystem crossing (isc) [26] which is
[6,10,11,15]. For example, changing the leaving group for 10, the origin of the heterolysis and rearrangement processes for
i.e., photolysis of 6-hydroxy-2-naphthylacetyl GABA ester pHP.
(Figure 2B) inefficiently released 4-aminobutyrate (GABA) in
low yield and is absent of any chromophore rearrangement.
Wan’s alternative suggestion that a disruption of both aromatic
rings' π-conjugation would be less favorable than only one
It is surprising that 2,6-HNA DEP (10) and Wan’s 1,5-HNA ring’s disruption has merit. For this to be a determining factor,
acetate 27 [19] do not undergo a photo-Favorskii rearrange- the disruption would have to occur prior to or during the prod-
ment. In the case of 27, Wan suggested that the lack of reactiv- uct determining formation of the ‘Favorskii’ intermediate 33
ity may have arisen from factors such as a lower excited state [38,48,50]. The putative key intermediates for 2,6-HNA and
energy of the naphthyl ketone platform or the disruption of both 1,5-HNA photo-Favorskii rearrangements (Scheme 9) illustrate
aromatic rings for the 1,5 substituted analog 27 whereas only the stage at which disruption in the π-network takes place. As in
one ring is involved for the 1,4 substituted derivative [19].
the case for the 1,4-HNA rearrangement, neither triplet birad-
ical 337 nor 340 experience a change in connectivity within the
Both of the contributing components of the 2,6 pattern, aromatic nucleus. The formation of the triplet biradical is irre-
2-naphthol (ET = 60.2 kcal/mol) and 2-acetylnaphthalene versible so that once generated, it must proceed on to a final
Scheme 8: Photodehydration and substitution of 5-(1-hydroxyethyl)-1-naphthol 34 [19].
2046

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
Scheme 9: Putative rearrangement intermediates for 1,5- and 2,6- HNA chromophores.
product. Since no rearrangement products are formed for either methyl ketone, probably results from homolysis. The lower
the 2,6-HNA or 1,5-HNA, it is tempting to conclude that the excited state acidity and the poor intersystem crossing in the
decay to a ground state biradicaloid or zwitterion is the product 2,6-HNA platform and strong fluorescence of the 2,6-HNA
determining step. In both cases, the aromaticity of both rings is chromophore (Figure 2) also diminishes formation of the triplet
lost upon formation of the Favorskii cyclopropanone intermedi- and, thus, its participation in a photo-Favorskii rearrangement.
ates 38 and 41.
Finally, an indication of the photostability of the 2,6-HNA
derivatives had earlier precedence in the lack of product forma-
However, this rationale also falls short because the reactions of tion for 2-hydroxy-6-trifluoromethylnaphthylene photo-dehy-
the similar biradicaloid or zwitterion are known to undergo drofluorination reported by Seiler and Wirz [51]. For the 1,5-
nucleophilic substitution and not reduction to the methyl ketone HNA acetate, the less reactive (higher pKa) of the leaving OAc
products 6, 12 or 15 [38,39]. Substitution by solvent would trap group of 27 vs DEP, fluorescence decay and reversible ESIPT
the intermediate biradicaloid (or zwitterion) to form, in the case processes are major factors disfavoring photorelease.
of 2,6-HNA, the α-methoxy ketone 42, which was not detected
among the photoproducts.
A summary of the most frequently encountered examples of
caged phosphates is given in Table 4. The data are reported for
The controlling feature leading to the change in mechanism for DEP leaving groups, when available, since DEP has proven to
these two examples is more reasonably imbedded in the route to be a commonly employed test leaving group for PPG compar-
the reactive triplet excited state configuration [10,11]. For pHP, isons. It is generally found that the DEP model is reliable when
the reactive excited state is a π,π* triplet that is favored only in extended to more complex phosphate leaving groups including
hydroxylic solvents, especially H2O [15,45,46,48]. In non- nucleotides such as ATP and GTP and tyrosyl phosphates and
hydroxylic solvents, the (n,π*)3 dominates which leads to the thiophosphates [52-55]. The key comparisons for practical
classical α- and β-homolytic cleavage and H-abstraction reac- applications of photochemical deprotection are the maximum
tions. For 2,6-HNA and 5,8-BQA, the reduction product, a conversion which controls the chemical yield, the quantum
2047

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
yield (or photochemical efficiency) and the complexity of the Table 4 reveals that most chromophores either form complex
photochemical products and byproducts in the photolyzate mix- mixtures of photoproducts complicating the isolation of pure,
ture. Solvents and excitation wavelengths are also given in unprotected phosphates or they remain essentially unchanged
Table 4 as a guide on selecting a PPG using these reaction para- (intact) and continue to absorb incident radiation, thus compro-
meters.
mising the conversion to products and lengthening exposure of
Table 4: A comparison of photorelease from cages for diethyl phosphates. Chemical yields, quantum yields, efficiencies and complexity based on
solvents and chromophores.
PPG chromophore
derivatives
solventa
λexc nm
Φdis and
(conversion
yield/%)
S1 or T1 chromo-
phore fateb
Ref
o-nitrobenzyl
1a X=1
1b X=2
H2O
H2O
254, 308 N/Ac (28d)
308, 355 N/A (90d)
mixture
N/A
[56]
[56]
S1
benzyl
30a (R, X,
Y = H)
MeOH,
n-BuOH,
t-BuOH
254
0.39 (50–60)
S1
intact
[35,57]
S(−)-30b (X,
Y = H, R = Me) t-BuOH
MeOH
254
254
254
300
0.21–0.9e (77/28
rac)
S1
S1
S1
T1
S1
nr
intact
[35,58]
[58]
30c (R, Y = H, n-BuOH,
X = OMe)
0.42 (47)
intact
t-BuOH
30d (R, X = H, t-BuOH
Y = OMe)
0.18 (28)
intact
[58]
phenacyl
4a (R = H,
X = OH)
MeOH
0.77 (87–96)
Favorskii
mixture
intact
[6,8,15,26,
28,29]
4b (R = H,
X = OMe)
H2O, MeOH 300, 355 0.42f (27d/90f)
[28,36,57]
4c (R = Me,
MeOH
313
0.71 (94)
[59]
X = H)
benzoin
2a (X = H)
2a (X = H)
benzene
350
350
0.28 (72–79)
T1
T1
benzofuran [35,56]
H2O/MeCN
0.37
N/Ad,
[56,58,60]
(50d, 100)
benzofuran
2b (X = OMe)
H2O
308, 350 N/A (55, 100)
benzofuran [8,56,61]
coumarin-4-ylmethyl
3a
(X = OH)
MeOH
350
0.038 (<50)
S1
S1
S1
intact
intact
intact
[62,63]
[58,60]
[62,64]
3b
(X = OMe)
H2O/CH3CN, 333–350 0.04–0.08
MeOH
3c
H2O
389
0.0041 (N/A)
(X = NMe2)
naphthylmethyl
28 α
29 β
MeOH
MeOH
300
300
0.08 (N/A)
0.3 (N/A)
S1
S1
intact
intact
[35,58]
[35,58]
2048

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
Table 4: A comparison of photorelease from cages for diethyl phosphates. Chemical yields, quantum yields, efficiencies and complexity based on
solvents and chromophores. (continued)
naphthylacetyl
10 (2,6)
(X = OH, R,
Y = H)
1% aq
MeCN
350
350
350
300
0.031 (67)
0.028 (98)
0.0076 (40)
nr (nrg)
T1
T1
T1
intact
this work
this work
this work
[19]
14a (1,4)
(X, Y = H,
R = OH)
1% aq
MeOH
Favorskii
Favorskii
intact
14b (1,4)
(X ,Y = H,
R = OMe)
1% aq
MeOH
27 (1,5)
MeCN/H2O
(X, R = H,
Y = OH)
hydroxyindolinyl
24
MeCN/H2O
350
24 × 10−5 (30–44)
S1
mixture
this work
aThe ratio of H2O/MeCN was 1:1 unless stated otherwise. bWhen the chromophore is unchanged upon photolysis, fate of the chromophore is noted
as “intact”. When the chromophore rearranges by the photo-Favorskii migration or closure to a benzofuran group, the fate is noted as Favorskii or
benzofuran, respectively. When complex photoreactions occur with the chromophore, it is noted as “mixture”. cN/A = not available. dInorganic phos-
phate released in H2O ([56]), eWhen internal return is included. fIn MeOH, gnr = no reaction (leaving group was acetate).
the sample to radiation. Three chromophores, benzoin, pHP, lower quantum efficiency of 1,4-MNA in MeOH is fully in
and now 1,4-hydroxynaphthylacetyl substrates cleanly undergo accord with a similar methylation of the hydroxy group of pHP
rearrangements that shift the absorption chromophore. Benzoins DEP, emphasizing the role of the phenolic OH on the reaction
form benzofurans that, in fact, exacerbate the problem of light and rearrangement [6,15,46,48]. In anhydrous, non hydroxylic
competition and fluorescence. Only pHP and 1,4-substituted media, the mechanism of the reaction changes from heterolysis
naphthylacetyl 14 generate a less competitive chromophore and reverts to the more traditional photochemically induced
allowing maximum conversion and quantitative chemical yield. homolysis mechanism, leading to more complicated product
mixtures and poor conversions.
While pHP is highly efficient as a PPG (Φ’s approach 1.0) [15]
the lower efficiency of 1,4-HNA (0.028) is competitive with The heterolysis mechanism of the photo-Favorskii rearrange-
many currently available PPGs. A measure often applied in bio- ment and release of DEP is initiated at the triplet state of 14 and
logical decaging studies is the product of absorptivity and effi- proceeds according to the more extensively characterized
ciency (εmax × Φ) which is 232 for photolysis of 1,4-HNA DEP rearrangement for pHP DEP. Due to a more favorable excita-
at 325 nm, competitive with PPGs listed in Table 4. By com- tion wavelength of 350 nm for 1,4-HNA (vs 280 nm for pHP), it
parison, although its absorptivity is very low at 325 nm, pHP is anticipated that the 4-hydroxy-1-acylnaphthyl PPG will find
DEP has a much larger εmax at 280 nm that gives a value of more favorable applications as a PPG not only for phosphates
6000 for excitation at 280 nm, an excitation wavelength that is but many other acidic functional groups. Additional studies on
biologically harmful [6,8,9].
2PE activation of 1,4-HNA DEP at 650 nm are also in progress.
Conclusion
2,6-HNA DEP and 5,8-BQA DEP also release DEP but do not
We have designed and tested several new candidates for near- undergo the beneficial aryl ketone deconjugative rearrange-
UV absorbing photoprotecting groups. 1,4-HNA DEP (14a) and ment, instead follow a homolytic pathway. This results in lower
1,4-MNA DEP (14b) upon irradiation at 350 nm in 1% aq yields, poorer efficiencies, and incomplete conversion for 2,6-
MeOH quantitatively release diethyl phosphate concomitant HNA, and intractable mixtures of chromophoric byproducts and
with a photo-Favorskii rearrangement with quantum yields of secondary oxidation reactions with O2, for 5,8-BQA. Neither
0.028 for 1,4-HNA DEP and 0.0076 for 1,4-MNA DEP. The chromophore is suitable for application as a PPG.
2049

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
Experimental
Synthesis
2. Photoproduct analysis in aqueous methanol. The photol-
ysis mixture of 14b from the previous experiment was concen-
The details on the synthesis of all new compounds can be found trated under reduced pressure and the residue washed with
in the Supporting Information File 1. The synthetic procedures water and extracted with ethyl acetate. The organic extract was
and spectral data for all new compounds are reported in concentrated under vacuum and the residue analyzed with
Supporting Information File 1.
1H NMR (CD3OD) and FAB–MS. The data for the major prod-
uct matched that of an authentic sample of methyl (4-methoxy-
1-naphthyl)acetate (26).
Photochemistry
General methods
Photolyses were performed in a Rayonet RPR-100 photochem- 3. Photoproduct analysis in aqueous acetonitrile. Diethyl
ical reactor fitted with a merry-go-round apparatus and (2-(4-methoxynaphthalen-1-yl)-2-oxoethyl) phosphate (14b,
16 × 350 nm (RPR 3500 Å) or 16 × 300 nm (RPR 3000 Å) 6 mg, 0.017 mmol) was dissolved in acetonitrile (5 mL)
lamps. The Rayonet reactor was turned on to warm up the containing water (50 µL) in a Pyrex tube and irradiated at
lamps for 15 minutes prior to the irradiation. Samples were irra- 350 nm with 9–3500 Å bulbs. Sample aliquots were taken after
diated in NMR tubes or Pyrex test tubes at 40–45 °C in the 0, 45, and 90 min of photolysis and analyzed by HPLC.
presence of air unless indicated otherwise. The tubes were The detection wavelength was 319 nm. The phosphate ester
placed in a merry-go-round apparatus and the time of irradi- had a retention time of 5.6 min. A new peak emerged at
ation was recorded. Light output for the determination of 6.3 min that grew in intensity throughout the photolysis.
quantum efficiencies was measured by using the potassium Co-injection with an authentic sample of 1-acetyl-4-methoxy-
ferrioxalate method [65] under identical conditions used for the naphthalene (12) confirmed the ketone as a photoproduct from
photolysis of the esters. Photolysis samples were analyzed for the reaction.
product formation and ester disappearance using 1H NMR and
RP-HPLC methods. Dimethylformamide (DMF, 2 μL, 4. Diethyl (2-(4-methoxynaphthalen-1-yl)-2-oxoethyl) phos-
2.6 × 10−2 mmol) was used as an internal standard in 1H NMR phate (14b): Dark reaction control experiment. An NMR
analyses. For HPLC analyses the solvent was 70% CH3CN and tube was charged with 14b (ca. 5 mg), which was dissolved in
30% H2O unless otherwise noted. The detection wavelength methanol-d4 (<1 mL). The sample was heated to 46 °C in a
varied depending on the ester analyzed as noted and the flow warm water bath in the dark for 1 h. 1H NMR scans were taken
rate was 1.0 mL/min. Sample injections were performed in trip- before and after heating; no significant difference was observed
licate.
in the NMR spectra. Also, 14a was shown to be stable for 30 h
in 10% aq MeCN at rt in the dark.
Photolysis of diethyl (2-(4-methoxynaphthalen-1-yl)-
2-oxoethyl) phosphate (14b)
Photolysis of diethyl (2-(4-hydroxynaphthalen-1-yl)-
An NMR tube was charged with 14b (16 mg, 0.045 mmol) 2-oxoethyl) phosphate (14a)
along with CD3OD (0.95 mL) and DMF (2 µL) as an internal An NMR tube was charged with 14a (17 mg, 0.05 mmol) along
standard. The sample was irradiated at 350 nm and 1H NMR with CD3OD (0.95 mL) and DMF (2 µL). The sample was irra-
scans were collected after 0, 30, and 60 min of photolysis. The diated at 350 nm and the percent conversion and deprotection %
depletion of 14b and the appearance of released diethyl phos- yield were determined as described above for 14b. Results are
phate were quantified from the integrations of the signals at shown in Table 1.
δ 5.4 and 4.1 ppm, respectively. Results are shown in Table 1.
1. Quantum efficiency determination in aqueous methanol.
1. Quantum efficiency determination. A Pyrex tube was The same general procedure and HPLC conditions were
charged with 14b (6.2 mg, 0.018 mmol) in methanol (5 mL) and employed as described for 14b. Amounts used: 14a (7.6 mg,
water (50 µL) and the contents were mixed thoroughly. The 0.022 mmol), methanol (5 mL), water (50 µL). The detection
sample was irradiated at 350 nm and sample aliquots (150 µL) wavelength was 323 nm. The phosphate ester had a retention
were taken after 0, 15, 30, and 45 min. The aliquots were time of 3.1 min. Results are shown in Table 1.
analyzed by HPLC with a mobile phase of 90% acetonitrile and
10% water and a detection wavelength of 319 nm. The deple- 2. Quantum efficiency determination in aqueous aceto-
tion of 14b was quantified from the peak areas of the signal at nitrile. A Pyrex tube was charged with 14a (4.7 mg,
3.5 min corresponding to the phosphate ester. The quantum effi- 0.014 mmol), acetonitrile (5 mL) and water (50 µL). The
ciency was determined as described above. Results are shown in sample was irradiated at 350 nm; aliquots (200 µL) were taken
Table 1.
after 0, 2, 5, 7, and 10 min of photolysis and analyzed by HPLC
2050

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
on a C18 analytical column. The mobile phase contained 70% ical shifts observed for methyl (4-methoxy-1-naphthyl)acetate
acetonitrile and 30% water, and the detection wavelength was (26).
323 nm. The phosphate ester had a retention time of 3.5 min.
The depletion of 14a was accompanied by the appearance of 6. Photoproduct analysis in aqueous acetonitrile. An NMR
four new peaks at 1.7, 2.3, 2.6, and 3.9 min, which were not tube containing 14a (ca. 5 mg) was charged with CD3CN
identified. Results are shown in Table 2.
(1 mL) and D2O (0.1 mL). The solution was mixed thoroughly
and irradiated at 350 nm for 30 min. The 1H NMR spectrum
3. Quantum efficiency determination in aqueous aceto- contained a singlet at δ 2.66 ppm that was assigned to the
nitrile (degassed). A Pyrex tube was charged with 14a (5 mg, methyl protons of 1-acetyl-4-hydroxynaphthalene (15). The
0.015 mmol) along with acetonitrile (5 mL) and water (50 µL). presence of 15 was confirmed upon spiking with an authentic
The tube was fitted with a rubber septum and sparged with sample.
argon for 15 min. Sample aliquots (0.2 mL) were taken after 0,
2, 5, 7, and 10 min of photolysis using a needle syringe to avoid Photolysis of diethyl (2-(6-hydroxynaphthalen-2-yl)-
the introduction of air into the sample during irradiation. The 2-oxoethyl) phosphate (10)
aliquots were analyzed as described above. The percent conver- Quantum efficiency determination. A Pyrex tube was charged
sion was 44% after 2 min of photolysis, thus the first two data with 10 (2.2 mg, 0.0065 mmol) along with acetonitrile (5 mL)
points from a plot of mmol 14a vs time were used to estimate and water (50 µL). The solution was irradiated at 350 nm and
the slope of the regression line for depletion of 14a which led to sample aliquots (200 µL) were taken after 0, 2, 5, 7, and 10 min
an approximate value of 0.067 for the disappearance quantum of photolysis. The aliquots were analyzed by HPLC. The detec-
efficiency (Table 2).
tion wavelength was 315 nm. The phosphate ester had a reten-
tion time of 3.4 min and its depletion was accompanied by the
4. Triplet sensitization experiment. A quartz tube was charged appearance of signals at 1.8 and 4.0 min. The quantum effi-
with 14a (2.0 mg, 0.0059 mmol) and benzophenone (20 mg, ciency was determined as described above and displayed in
0.11 mmol), and the contents were dissolved in acetonitrile Scheme 5 (Φdis= 0.031).
(10 mL) to which was added water (100 µL) to make a 1%
aqueous solution. The solution was degassed with argon and Photolysis of 4-[2-(6-hydroxy-2-naphthyl)-2-
irradiated with 6-RPR 2540 Å bulbs. Sample aliquots (0.2 mL) oxoethoxy]-4-oxobutan-1-aminium trifluoroacetate
were taken after 0, 2, 5, 7, and 10 min of photolysis and (2,6-HNA GABA)
analyzed by HPLC on a C18 analytical column with a mobile Photoproduct analysis. A Pyrex tube was charged with 2,6-
phase containing 70% acetonitrile and 30% water. The detec- HNA GABA (19.5 mg, 0.068 mmol) along with acetonitrile
tion wavelength was 323 nm. The benzophenone had a reten- (1.0 mL) and 50 mM TRIS buffer (9.0 mL, pH 7.3). The solu-
tion time of 6.5 min and remained constant throughout the tion was photolyzed at 350 nm for 1 h, and sample aliquots of
photolysis. The phosphate ester had a retention time of 3.6 min 50 µL were taken after 0, 30, and 60 min of photolysis. The
and its depletion was accompanied by the appearance of three aliquots were diluted with 150 µL of TRIS buffer and injected
new peaks at 1.6, 1.9, and 4.4 min, which were not assigned. into an Econosphere C18 analytical column. The mobile phase
The quantum efficiency was determined as described above. contained 90% CH3CN with 0.1% TFA and 10% H2O. The
Results are shown in Table 2.
flow rate was 1.0 mL/min and the detection wavelength was
244 nm. The depletion of 2,6-HNA GABA (retention time ca.
5. Photoproduct analysis in aqueous methanol. A Pyrex tube 6 min) was accompanied by an increase in a peak with a reten-
was charged with 14a (5 mg, 0.015 mmol), methanol (5 mL) tion time of ca. 3 min corresponding to the photoproduct(s). At
and water (50 µL). The sample was irradiated at 350 nm for the end of the photolysis, the solution contained an orange-
15 min. After photolysis the solution changed from colorless to colored precipitate. The organic soluble components were
light yellow in appearance. The solvent was removed under extracted with ethyl acetate, and the organic extract was washed
reduced pressure and the remaining residue was dissolved in with water, dried over magnesium sulfate, and the solvent
methylene chloride and washed with water. The organic extract removed under reduced pressure to afford ca. 6 mg of a yellow
was collected and the solvent concentrated to afford ca. 3 mg of residue. The residue was dissolved in acetonitrile and injected
an orange-colored residue. The residue was analyzed with into the HPLC column, resulting in a peak with a retention time
1H NMR in methanol-d4. Two signals were observed in the of ca. 3 min. Co-injection of an authentic sample of 2-acetyl-6-
spectrum at δ 4.00 and 3.68 ppm, in a ratio of ca. 1:1.5, which hydroxynaphthalene (6) also produced the same peak. The
were assigned to the methylene and methyl protons of methyl residue was analyzed with 1H NMR (CDCl3) and found to
(4-hydroxy-1-naphthyl)acetate (25), based on the similar chem- contain the characteristic methyl singlet at δ 2.7 ppm corres-
2051

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
ponding to 6, in addition to other unassignable peaks further spectrum. No significant change was observed when the sample
downfield in the spectrum. TLC analysis with 2:1 hexane/ethyl was re-examined 48 h later.
acetate further suggested the presence of 6 as a photoproduct
from the reaction.
Supporting Information
Fluorescence measurements. A solution of 2,6-HNA GABA
Supporting Information File 1
(0.042 mM in pH 7.3 TRIS buffer containing 1% acetonitrile)
was placed in a quartz cell in a Carey Eclipse fluorescence spec-
trometer. Measurements were made under ambient conditions.
Synthetic procedures and spectral data for all new
compounds.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-10-212-S1.pdf]
Photolysis of 2-(8-(benzyloxy)quinolin-5-yl)-2-
oxoethyl diethyl phosphate (24)
An NMR tube was charged with 24 (5.0 mg, 0.01 mmol)
dissolved in CD3CN (900 μL), D2O (100 μL), and DMF (2 μL,
2.6 × 10−2 mmol). The sample was irradiated without degassing
with 16 × 300 nm or 16 × 350 nm lamps and 1H NMR spectra
(16 scans) were collected after 0, 30, 60, 90, and 120 min of
photolysis. The depletion of 24 and the appearance of released
diethyl phosphate were quantified from the integrations
of the methylene signals at δ 5.30 ppm and 3.89 ppm, respect-
ively.
Acknowledgements
Support of this work was provided by NIH grant R01 GM72910
(RSG) and by the University of Kansas.
References
1. Havinga, E.; De Jongh, R. O. Bull. Soc. Chim. Belg. 1962, 71,
803–810. doi:10.1002/bscb.19620711131
2. Zimmerman, H. E.; Sandel, V. R. J. Am. Chem. Soc. 1963, 85,
915–922. doi:10.1021/ja00890a019
3. Zimmerman, H. E.; Somasekhara, S. J. Am. Chem. Soc. 1963, 85,
922–927. doi:10.1021/ja00890a020
1. Photolysis of 24 in the absence of oxygen. Using the same
general procedure, the photolysis was repeated under degassed
conditions. In these experiments, photolysis samples were
purged with argon for 30 min before photolysis. The results are
shown in Table 3.
4. Zimmerman, H. E. J. Am. Chem. Soc. 1995, 117, 8988–8991.
doi:10.1021/ja00140a014
5. Buck, A. T.; Beck, C. L.; Winter, A. H. J. Am. Chem. Soc. 2014, 136,
8933–8940. doi:10.1021/ja501777r
6. Klán, P.; Šolomek, T.; Bochet, C. G.; Blanc, A.; Givens, R.; Rubina, M.;
Popik, V.; Kostikov, V.; Wirz, J. Chem. Rev. 2013, 113, 119–191.
doi:10.1021/cr300177k
7. Bochet, C. G.; Blanc, A. In CRC Handbook of Organic Photochemistry
and Photobiology, 3rd ed.; Griesbeck, A. G.; Oelgemöller, M.;
Ghetti, F., Eds.; CRC Press: Boca Raton, Florida, 2012; pp 73–93.
8. Goeldner, M.; Givens, R. S., Eds. Dynamic Studies in Biology;
Wiley-VCH: Weinheim, Germany, 2005. doi:10.1002/3527605592
9. Mayer, G.; Heckel, A. Angew. Chem., Int. Ed. 2006, 45, 4900–4921.
doi:10.1002/anie.200600387
2. Photoproduct analysis by 1H NMR. An NMR tube was
charged with 24 (10 mg, 0.02 mmol) which was dissolved in
CD3CN (900 μL) and D2O (100 μL). The contents were mixed
thoroughly and the sample irradiated with air or under Ar
purged conditions with 16 × 300 or 350 nm lamps. 1H NMR
spectra were collected after 60 min of photolysis and used to
calculate the percent conversion. The depletion of 24 and the
appearance of released diethyl phosphoric acid were deter-
mined from the NMR signals for the methylene protons of the
phosphate ester and methylene hydrogens of diethyl phosphoric
acid at δ 5.30 ppm and 3.88 ppm, respectively. The photolysis
sample was spiked with an authentic sample of diethyl phos-
phoric acid synthesized from diethyl chlorophosphate [66]. A
dramatic increase in the intensities of the new methylene and
methyl signals at 3.88 and 1.19 ppm, respectively, confirmed
the presence of diethyl phosphate.
10.Turro, N. J.; Ramamurthy, V.; Scaiano, J. C. Modern Molecular
Photochemistry of Organic Molecules, 1st ed.; University Science
Books: Sausalito, California, 2010.
11.Klán, P.; Wirz, J. Photochemistry of Organic Compounds: From
Concepts to Practice, 1st ed.; John Wiley & Sons: Chichester, 2009.
doi:10.1002/9781444300017
12.Mewes, J.-M.; Dreuw, A. Phys. Chem. Chem. Phys. 2013, 15,
6691–6698. doi:10.1039/c3cp44338h
13.Givens, R. S.; Kueper, L. W., III. Chem. Rev. 1993, 93, 55–66.
doi:10.1021/cr00017a004
14.Pincock, J. A. In CRC Handbook of Organic Photochemistry and
Photobiology, 2nd ed.; Horspool, W.; Lenci, F., Eds.; CRC Press: Boca
Raton, Florida, 2010; Chapter 66.
15.Givens, R. S.; Rubina, M.; Wirz, J. Photochem. Photobiol. Sci. 2012,
11, 472–488. doi:10.1039/c2pp05399c
3. Control experiments with 24; stability in the absence of
irradiation. An NMR tube was charged with 24 (10.0 mg,
0.02 mmol) along with 900 μL of CD3CN and 100 μL of D2O.
The contents were mixed thoroughly and warmed in a water
bath at 40 °C in the dark for 2 h. 1H NMR analysis before and
after the heating showed no significant change in the NMR
16.Engels, J.; Schlaeger, E.-J. J. Med. Chem. 1977, 20, 907–911.
doi:10.1021/jm00217a008
17.Kaplan, J. H.; Forbush, B., III; Hoffman, J. F. Biochemistry 1978, 17,
1929–1935. doi:10.1021/bi00603a020
2052

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
18.Givens, R. S.; Weber, J. F. W.; Jung, A. H.; Park, C.-H. In Methods in
Enzymology; Marriott, G., Ed.; Academic Press: New York, 1998; Vol.
291, pp 1–29.
43.Ma, C.; Kwok, W. M.; Chan, W. S.; Du, Y.; Kan, J. T. W.; Toy, P. H.;
Phillips, D. L. J. Am. Chem. Soc. 2006, 128, 2558–2570.
doi:10.1021/ja0532032
19.Lukeman, M.; Veale, D.; Wan, P.; Munasinghe, V. R. N.; Corrie, J. E. T.
Can. J. Chem. 2004, 82, 240–253. doi:10.1139/v03-184
20.Bauer, D. P.; Macomber, R. S. J. Org. Chem. 1975, 40, 1990–1992.
doi:10.1021/jo00901a027
44.Zuo, P.; Ma, C.; Kwok, W. M.; Chan, W. S.; Phillips, D. L.
J. Org. Chem. 2005, 70, 8661–8675. doi:10.1021/jo050761q
45.Chen, X.; Ma, C.; Kwok, W. M.; Guan, X.; Du, Y.; Phillips, D. L.
J. Phys. Chem. A 2006, 110, 12406–12413. doi:10.1021/jp064490e
46.Phillips, D. L. In Hydrogen bonding and transfer in the excited state;
Han, K.-L.; Zhao, G.-J., Eds.; John Wiley & Sons: Chichester, 2011;
pp 287–312.
21.Arumugam, P.; Karthikeyan, G.; Perumal, P. T. Chromatographia 2004,
33, 1146–1147.
22.Zwierzak, A.; Kluba, M. Tetrahedron 1971, 27, 3163–3170.
doi:10.1016/S0040-4020(01)98109-8
47.Calvert, J. G.; Pitts, J. N., Jr. Photochemistry; John Wiley & Sons: New
York, 1966; p 88.
23.Matsumura, K. J. Am. Chem. Soc. 1930, 52, 4433–4436.
doi:10.1021/ja01374a037
48.Givens, R. S.; Heger, D.; Hellrung, B.; Kamdzhilov, Y.; Mac, M.;
Conrad, P. G., II; Cope, E.; Lee, J. I.; Mata-Segreda, J. F.;
Schowen, R. L.; Wirz, J. J. Am. Chem. Soc. 2008, 130, 3307–3309.
doi:10.1021/ja7109579
24.Iwakuma, T.; Tsunashima, A.; Ikezawa, K.; Takaiti, O. Novel carbostyril
derivative and process for preparing same. Eur. Pat. Appl. EP0147719
A2, July 10, 1985.
25.Recuero, V.; de Gonzalo, G.; Brieva, R.; Gotor, V. Eur. J. Org. Chem.
2006, 18, 4224–4230. doi:10.1002/ejoc.200600385
26.Ma, C.; Kwok, W. M.; Chan, W. S.; Zuo, P.; Kan, J. T. W.; Toy, P. H.;
Phillips, D. L. J. Am. Chem. Soc. 2005, 127, 1463–1472.
doi:10.1021/ja0458524
49.Suzuki, T.; Kaneko, Y.; Ikegami, M.; Arai, T. Bull. Chem. Soc. Jpn.
2004, 77, 801–806. doi:10.1246/bcsj.77.801
50.Rajesh, C. S.; Givens, R. S.; Wirz, J. J. Am. Chem. Soc. 2000, 122,
611–618. doi:10.1021/ja993070i
51.Seiler, P.; Wirz, J. Tetrahedron Lett. 1971, 12, 1683–1686.
doi:10.1016/S0040-4039(01)87434-7
27.Gielen, H.; Alonso-Alija, C.; Hendrix, M.; Niewöhner, U.; Schauss, D.
Tetrahedron Lett. 2002, 43, 419–421.
52.Zou, K.; Miller, W. T.; Givens, R. S.; Bayley, H. Angew. Chem., Int. Ed.
2001, 40, 3049–3051.
doi:10.1016/S0040-4039(01)02162-1
28.Givens, R. S.; Park, C.-H. Tetrahedron Lett. 1996, 37, 6259–6262.
doi:10.1016/0040-4039(96)01390-1
doi:10.1002/1521-3773(20010817)40:16<3049::AID-ANIE3049>3.0.CO
;2-N
29.Park, C.-H.; Givens, R. S. J. Am. Chem. Soc. 1997, 119, 2453–2463.
doi:10.1021/ja9635589
53.Kawakami, T.; Cheng, H.; Hashiro, S.; Nomura, Y.; Tsukiji, S.;
Furuta, T.; Nagamune, T. ChemBioChem 2008, 9, 1583–1586.
doi:10.1002/cbic.200800116
30.Kötting, C.; Güldenhaupt, J.; Gerwert, K. Chem. Phys. 2012, 396,
72–83. doi:10.1016/j.chemphys.2011.08.007
54.Aemissegger, A.; Carrigan, C. N.; Imperiali, B. Tetrahedron 2007, 63,
6185–6190. doi:10.1016/j.tet.2007.03.023
31.Kötting, C.; Gerwert, K. Chem. Phys. 2004, 307, 227–232.
doi:10.1016/j.chemphys.2004.06.051
55.Rothman, D. M.; Shults, M. D.; Imperiali, B. Trends Cell Biol. 2005, 15,
502–510. doi:10.1016/j.tcb.2005.07.003
32.Kötting, C.; Kallenbach, A.; Suveyzdis, Y.; Wittinghofer, A.; Gerwert, K.
Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 6260–6265.
doi:10.1073/pnas.0712095105
56.Baldwin, J. E.; McConnaughie, A. W.; Moloney, M. G.; Pratt, A. J.;
Shin, S. B. Tetrahedron 1990, 46, 6879–6884.
33.Kötting, C.; Kallenbach, A.; Suveyzdis, Y.; Eichholz, C.; Gerwert, K.
ChemBioChem 2007, 8, 781–787. doi:10.1002/cbic.200600552
34.Du, X.; Frei, H.; Kim, S.-H. Biopolymers 2001, 62, 147–149.
doi:10.1002/bip.1007
doi:10.1016/S0040-4020(01)87875-3
57.Givens, R. S.; Athey, P. S.; Kueper, L. W., III; Matuszewski, B.;
Xue, J. Y. J. Am. Chem. Soc. 1992, 114, 8708–8710.
doi:10.1021/ja00048a059
35.Givens, R. S.; Matuszewski, B. J. Am. Chem. Soc. 1984, 106,
6860–6861. doi:10.1021/ja00334a075
58.Givens, R. S.; Matuszewski, B.; Athey, P. S.; Stoner, M. R.
J. Am. Chem. Soc. 1990, 112, 6016–6021. doi:10.1021/ja00172a016
59.Klán, P.; Pelliccioli, A. P.; Pospišil, T.; Wirz, J.
36.Givens, R. S.; Athey, P. S.; Matuszewski, B.; Kueper, L. W., III; Xue, J.;
Fister, T. J. Am. Chem. Soc. 1993, 115, 6001–6012.
doi:10.1021/ja00067a015
Photochem. Photobiol. Sci. 2002, 1, 920–923. doi:10.1039/B208171G
60.Schade, B.; Hagen, V.; Schmidt, R.; Herbrich, R.; Krause, E.;
Eckardt, T.; Bendig, J. J. Org. Chem. 1999, 64, 9109–9117.
doi:10.1021/jo9910233
37.Khade, P. K.; Singh, A. K. Tetrahedron Lett. 2007, 48, 6920–6923.
doi:10.1016/j.tetlet.2007.07.155
38.Šolomek, T.; Heger, D.; Ngoy, B. P.; Givens, R. S.; Klán, P.
J. Am. Chem. Soc. 2013, 135, 15209–15215. doi:10.1021/ja407588p
39.Kammath, V. B.; Šolomek, T.; Ngoy, B. P.; Heger, D.; Klán, P.;
Rubina, M.; Givens, R. S. J. Org. Chem. 2013, 78, 1718–1729.
doi:10.1021/jo300850a
61.Corrie, J. E. T.; Katayama, Y.; Reid, G. P.; Anson, M.; Trentham, D. R.;
Sweet, R. M.; Moffat, K. Philos. Trans. R. Soc., A 1992, 340, 233–244.
doi:10.1098/rsta.1992.0063
62.Schmidt, R.; Geissler, D.; Hagen, V.; Bendig, J. J. Phys. Chem. A
2005, 109, 5000–5004. doi:10.1021/jp050581k
40.Epstein, W. W.; Garrossian, M. J. Chem. Soc., Chem. Commun. 1987,
532–533. doi:10.1039/c39870000532
63.Schmidt, R.; Geissler, D.; Hagen, V.; Bendig, J. J. Phys. Chem. A
2007, 111, 5768–5774. doi:10.1021/jp071521c
41.Ma, C.; Zuo, P.; Kwok, W. M.; Chan, W. S.; Kan, J. T. W.; Toy, P. H.;
Phillips, D. L. J. Org. Chem. 2004, 69, 6641–6657.
doi:10.1021/jo049331a
64.Pinheiro, A. V.; Parola, A. J.; Baptista, P. V.; Lima, J. C.
J. Phys. Chem. A 2010, 114, 12795–12803. doi:10.1021/jp103045u
65.Hatchard, C. G.; Parker, C. A. Proc. R. Soc. London, Ser. A 1956, 235,
518–536. doi:10.1098/rspa.1956.0102
42.Chan, W. S.; Ma, C.; Kwok, W. M.; Phillips, D. L. J. Phys. Chem. A
2005, 109, 3454–3469. doi:10.1021/jp044546+
66.Schole, J.; Schole, C.; Eikemeyer, J.; Krebs, H. C. Tetrahedron 1994,
50, 1125–1128. doi:10.1016/S0040-4020(01)80823-1
2053

Beilstein J. Org. Chem. 2014, 10, 2038–2054.
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.10.212
2054