Photoconversion of β-Lapachone to α-Lapachone via a Protonation-Assisted Singlet Excited State Pathway in Aqueous Solution: A Time-Resolved Spectroscopic Study

Article abstract of DOI:10.1021/acs.joc.5b00086

The photophysical and photochemical reactions of β-lapachone were studied using femtosecond transient absorption, nanosecond transient absorption, and nanosecond time-resolved resonance Raman spectroscopy techniques and density functional theory calculations. In acetonitrile, β-lapachone underwent an efficient intersystem crossing to form the triplet state of β-lapachone. However, in water-rich solutions, the singlet state of β-lapachone was predominantly quenched by the photoinduced protonation of the carbonyl group at the β position (O₉). After protonation, a series of fast reaction steps occurred to eventually generate the triplet state α-lapachone intermediate. This triplet state of α-lapachone then underwent intersystem crossing to produce the ground singlet state of α-lapachone as the final product. 1,2-Naphthoquinone is examined in acetonitrile and water solutions in order to elucidate the important roles that water and the pyran ring play during the photoconversion from β-lapachone to α-lapachone. β-Lapachone can also be converted to α-lapachone in the ground state when a strong acid is added to an aqueous solution. Our investigation indicates that β-lapachone can be converted to α-lapachone by photoconversion in aqueous solutions by a protonation-assisted singlet excited state reaction or by an acid-assisted ground state reaction.

Full text of DOI:10.1021/acs.joc.5b00086

Article
pubs.acs.org/joc
Photoconversion of β‑Lapachone to α‑Lapachone via a Protonation-
Assisted Singlet Excited State Pathway in Aqueous Solution: A Time-
Resolved Spectroscopic Study
Lili Du,†,§ Ming-De Li,†,§ Yanfeng Zhang,† Jiadan Xue,‡ Xiting Zhang,† Ruixue Zhu,†
Shun Cheung Cheng,† Xuechen Li,† and David Lee Phillips*,†
†Department of Chemistry, The University of Hong Kong, Hong Kong S.A.R., China
‡Department of Chemistry, Zhejiang Sci-Tech University, Hangzhou 310018, China
S
* Supporting Information
ABSTRACT: The photophysical and photochemical reactions
of β-lapachone were studied using femtosecond transient
absorption, nanosecond transient absorption, and nanosecond
time-resolved resonance Raman spectroscopy techniques and
density functional theory calculations. In acetonitrile, β-
lapachone underwent an efficient intersystem crossing to form
the triplet state of β-lapachone. However, in water-rich
solutions, the singlet state of β-lapachone was predominantly
quenched by the photoinduced protonation of the carbonyl
group at the β position (O9). After protonation, a series of fast
reaction steps occurred to eventually generate the triplet state α-
lapachone intermediate. This triplet state of α-lapachone then
underwent intersystem crossing to produce the ground singlet
state of α-lapachone as the final product. 1,2-Naphthoquinone is examined in acetonitrile and water solutions in order to
elucidate the important roles that water and the pyran ring play during the photoconversion from β-lapachone to α-lapachone. β-
Lapachone can also be converted to α-lapachone in the ground state when a strong acid is added to an aqueous solution. Our
investigation indicates that β-lapachone can be converted to α-lapachone by photoconversion in aqueous solutions by a
protonation-assisted singlet excited state reaction or by an acid-assisted ground state reaction.
was obtained under visible-light-induced photoreduction.21,22
The probable mechanism for this photo-oxidation of the
alcohols involves a H atom abstraction quenching of the excited
state followed by an electron transfer/proton transfer sequence
INTRODUCTION
■
β-Lapachone (β-LA) is an ortho-naphthoquinone obtained
from the heartwood of Tabebuia avellanedae, which is mainly
found in Brazil.1,2 β-LA has been extensively studied due to its
anticancer properties.3−14 β-LA has been found to inhibit the
in which ground state β-LA is reduced. Lower limiting
growth of sarcoma tumors in mice3 and to be cytotoxic to
efficiencies for the photoreduction of β-LA by the alcohols
leukemia cells4 and human cancers such as breast, colon,
are attributed to inefficiencies of the net H atom transfer in the
quenching step.21 In addition, both triplet states of β-LA and β-
prostate, and lung.5−14 The mechanism of the observed
LA-3-sulfonic acid were efficiently quenched by amino acids
cytotoxicity in cancer cells appears to be due to NAD(P)H
and nucleobases or nucleosides induced by an initial electron
quinine oxidoreductase-1 performing a two-electron reduction
transfer and followed by a fast proton transfer.19,23 The
quinone cation radicals were also detected in the study of β-
lapachone derivatives.24,25
of β-LA to produce its hydroquinone,12,15−17 which may then
generate reactive oxygen species to induce DNA damage that
can lead to cell death.18 Therefore, there has been some
research effort conducted on the photoreactions of β-LA.19−22
Netto-Ferreira and co-workers19,20 observed that β-LA is able
to react with hydrogen donors, such as 2-propanol, 1,4-
cyclohexadiene, 4-methoxyphenol, or indole, to form a
corresponding ketyl radical. The low values observed for the
hydrogen abstraction rate constant for β-LA when using 2-
propanol and 1,4-cyclohexadiene as quenchers are independent
of the solvent polarity, and this is thought to be due to the π,π*
character of its excited triplet.20 In the presence of alcohols, a
semireduced radical (QH•), which is toxic to a normal cell,18
Although a number of studies have examined β-LA in organic
solvents, the phototoxicity mechanism of β-LA toward cancers
cells remains poorly understood because the organic solution
conditions are very different from those of the biological
environment. Therefore, it will be useful to investigate the
photophysics and photochemistry of β-LA in aqueous solutions
that are more relevant to the actual solvent environment in
Received: January 13, 2015
Published: July 2, 2015
© 2015 American Chemical Society
7340
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
biological systems. To the best of our knowledge, we report
here the first time-resolved spectroscopic study of the
photoinduced reaction(s) of β-LA in aqueous solutions. Proton
transfer reactions and protonation are fundamental classes of
reactions in chemical transformations and play an important
role as fundamental processes throughout nature.26 Photo-
induced intermolecular proton transfer reactions are also
ubiquitous in numerous chemical and biological processes.
Results from some previous studies of selected aromatic
carbonyl compounds in aqueous solutions found that photo-
induced protonation of a ketone may often open some novel
photochemical reactions dependent on the water concentration
and acidity of the solutions.27−29 Since β-LA contains a couple
of ketone groups, it may exhibit similar behavior in aqueous
solutions.
We are pleased to report in this paper femtosecond and
nanosecond transient absorption (fs-TA and ns-TA) and
nanosecond time-resolved resonance Raman (ns-TR3) spectro-
scopic methods to investigate the early events taking place for
β-LA after photoexcitation in aqueous solutions. As far as we
know, this is the first time that fs-TA, ns-TA, and ns-TR3
experiments have been performed on β-LA in different
acetonitrile (MeCN)/H2O mixture solutions in order to
investigate the role of water on the excited states, intermediates,
and dynamics of reactive intermediates that are involved in the
photochemistry of β-LA. To help determine the geometries and
vibrational spectra of the intermediate species and assignments
of the experimental vibrational bands, density functional theory
(DFT) calculations were done using the B3LYP methods with
a 6-311G** basis set for all of the species examined here. The
structure and properties of the triplet state of β-LA (β-3LA) are
also briefly discussed. In order to distinguish the important role
of the pyran ring in the photochemistry, the photoreaction of
the parent compound 1,2-naphthoquinone (NQ) that does not
contain a pyran ring (see Scheme 1) was also studied by time-
resolved spectroscopy methods in both MeCN and aqueous
solutions and compared to the results found for β-LA here. Our
time-resolved spectroscopy results indicate that a photoinduced
protonation of the carbonyl group at the β position takes place
for the singlet excited state of β-LA with the assistance of water
in aqueous solutions. This protonation initiates a series of
processes that lead to the conversion of some β-LA to α-LA in
water-rich solutions and acidic aqueous solutions. This
photoinduced protonation reaction provides insight into the
role of water in the cytotoxicity and phototoxicity of β-LA in
biological systems and its potential use in photodynamic
therapy applications.
RESULTS AND DISCUSSION
■
Transient Absorption and Ns-TR3 Spectroscopic
Study of NQ and β-LA in MeCN. Fs-TA spectra were
obtained for the parent compound NQ in MeCN and are
shown in Figure 1. The spectra at early (Figure 1a) and later
delay times (Figure 1b) are given separately to more easily
discern the spectral changes that take place on different time
scales. After irradiation by 267 nm light, the ground state of NQ
is transferred to a higher excited singlet state Sn, and the very
early time TA spectra are assigned to the relaxation from the Sn
to S1 state.30 The initial species that appears at 370 (sharp
band) and 480 nm (broad band) at 2 ps can be assigned to S1
to Sn absorption from the singlet excited state of NQ. The time
constant for internal conversion from Sn to S1 is within the
instrument response time and will not be discussed here. In
Figure 1b, the bands at 370 and 480 nm are noticeably shifted
with time toward a longer wavelength, ultimately yielding a
sharp band at 375 nm and a broad absorption band centered at
593 nm that extends from below 460 nm to above 680 nm. An
isosbestic point at 535 nm indicates a dynamical conversion
between two distinct states, and the spectrum recorded after 30
ps appears identical to that assigned previously to the triplet
state of NQ (3NQ).19,31 From the inserted graph in Figure 1b,
the kinetics of the 480 nm decay can be fitted simultaneously
by a single-exponential function with a 7 ps time constant,
which is very close to the ISC time constant of benzophenone
recorded in MeCN.32,33 This indicates that the spectral
transformation observed here is due to the ISC conversion
from the lowest singlet (S1) the triplet (T1) of NQ.
Scheme 1. Structures of β-LA and NQ
From Figure 2, it can be seen that the temporal evolution of
the early time spectra for β-LA obtained after 267 nm excitation
in MeCN is very similar to that of NQ. To clearly discern the
spectral changes on different time scales, the spectra of the early
Figure 1. Fs-TA spectra of NQ in MeCN solution obtained after 266 nm excitation from 0.2 to 2.0 ps (a) and 2.0 to 31.5 ps (b). Temporal
dependence of the transient absorption spectra for NQ recorded at early picosecond times at 480 nm is inserted at the top of the right panel. The
solid lines indicate the best-fit kinetics to the experimental data points.
7341
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
Figure 2. (a−d) Fs-TA and ns-TA spectra of β-LA in neat MeCN solvent acquired after 267 nm irradiation. Kinetics of the characteristic fs-TA
absorption bands observed at 530 nm are inserted at the top of (c). Kinetics of the characteristic ns-TA absorption bands observed with nitrogen
saturated (left) and air saturated (right) at 373 nm and excited by 267 nm in MeCN solvent are inserted at the top of (d). The solid lines indicate the
kinetics fitting to the experimental data points.
(before 3.5 ps), 3.5−8 ps, and later (8−160 ps) times are given
separately in Figure 2. It can be seen that both the 370 and 530
nm bands grow in rapidly within 3.5 ps (see Figure 2a). Later
(Figure 2b), the 374 nm band increases in intensity slightly,
while the 530 nm band does not change much and the shoulder
at 494 nm appears. Meanwhile, both of the two bands blue shift
slightly and become narrower. In the next 160 ps (Figure 2c),
with the decrease in intensity of the 530 and 370 nm bands,
another broad absorption band (around 610 nm) appears
within 100 ps and an isosbestic point at ∼570 nm is seen, which
indicates a dynamical conversion between two distinct states.
The two transient absorption bands at 373 and 610 nm as well
as the three small bands around 400−500 nm appear to be
stable until 3 ns. Based on the fs-TA results of NQ and the
transient absorption features of the triplet of NQ in MeCN, the
transient absorption bands after 160 ps are likely to be due to
the β-3LA intermediate. From the inserted graph shown in
Figure 2c, the kinetics of the 530 nm decay can be simulated
satisfactorily by a triexponential function with the time
constants of 524 fs (τ1), 5 ps (τ2), and 29 ps (τ3), weighted
by different proportionality factors. The spectrum at very early
time (from 0.5 to 3.5 ps in Figure 2a) is assigned to an internal
conversion from the Sn to S1 (n,π*), whose lifetime is 524 fs.
Then, the S1 state is thermalized by vibrational cooling in 5 ps,
as evidenced by the narrowing and blue shifting of the bands.
With the reduction of the S1 signal, the subsequent changes in
the spectra are mainly attributed to the S1 (n,π*) to T1 (π,π*)
ISC process (τ3 = 29 ps) that leads to the increase of the
transient absorption at 610 nm, which corresponds to the T1 to
Tn transition with maxima at 373 and 610 nm in neat MeCN
solvent. Spectra at later delay times resemble those obtained
from nanosecond laser flash photolysis spectra of β-LA in
MeCN and the time evolution analysis of the absorbance at 373
nm of β-LA in MeCN under open air and nitrogen-purged
conditions (these results are given in Figure 2d). The observed
quenching by oxygen provides further evidence that the species
observed in the later fs-TA and ns-TA is a triplet species in
3
nature. The β-LA intermediate was reported to have a sharp
band 380 nm and a broad absorption band at 650 nm in
MeCN,19,20 which is analogous to the observation in our fs-TA
and ns-TA experiments.
Time-dependent DFT (TD-DFT) calculations were carried
out on the singlet and triplet states of NQ and β-LA in order to
predict their UV−vis spectra, and these results are shown in
Table 1S of the Supporting Information. The frontier orbitals
of NQ and β-LA obtained from these calculations are shown in
Figure 1S of the Supporting Information. As seen in Table 1S,
the experimental data are consistent with the calculation-
predicted spectra of the excited states. The first singlet states of
both molecules are calculated to have two strong absorption
bands in the visible region, which are also observed in the
experimental results: (1) 349, 331 nm transitions correspond-
ing to a transition from the benzyl ring to the dione ring
(LUMO) for both NQ and β-LA, which explains the almost
identical wavelength (370 nm) for these two compounds in the
experimental spectra; and (2) 441, 471 nm transitions due to
π,π* transitions for NQ and β-LA with little difference, while
the pyran ring of β-LA contributes to the transition on the
higher wavelength where the band is 50 nm red-shifted for β-
LA (530 nm) compared to the analogous band for NQ (480
nm) in the experimental spectra.
In an attempt to characterize the vibrational structure and
properties of the intermediates involved for β-LA in MeCN
after photoexcitation, ns-TR3 experiments were done using 266
nm as the pump laser wavelength and 368.9 nm as the probe
laser wavelength for β-LA in MeCN. Figure 2S displays the ns-
7342
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
TR3 spectra of β-LA in MeCN, and mainly one species was
observed. Fs-TA and ns-TA experiments suggest that the
intermediate detected in the nanosecond time range is the
β-3LA intermediate. DFT calculations were done to predict the
Raman spectrum of β-3LA, and the calculated result was
compared with the experimental TR3 spectrum obtained at 10
ns in Figure 3. Examination of Figure 3 reveals that the
toward the hydrogen abstraction reaction with hydrogen donor
reagents. Two types of T1 states, dominated respectively by
n,π* and π,π* character, have been identified to be responsible
for the differences in the T1 photophysical and photochemical
behavior.34−36,38 A T1 state with mainly n,π* nature exhibits a
high efficiency for a hydrogen abstraction reaction, whereas a
triplet state with a π,π* nature is barely reactive for that type of
reaction. Previous TR3 and TRIR studies found that the
frequency of the CO stretch mode appears in the 1400−
1600 cm−1 range for a π,π* predominant character triplet state,
while it is in the 1200−1400 cm−1 region for a typical n,π*
nature triplet state.37−40 As illustrated above, the CO stretch
frequencies for the β-3LA state are at 1581 and 1498 cm−1, and
this indicates that the β-3LA intermediate possesses mainly a
π,π* electronic configuration. This may account for the low
efficiency of the hydrogen abstraction for β-LA when using 2-
propanol and 1,4-cyclohexadiene as quenchers.20,21
Transient Absorption Study of β-LA and NQ in MeCN
and H2O Mixed Solutions. Protonation Accounts for the
Decay of the Singlet Excited State of β-LA. Fs-TA spectra of
NQ excited by 267 nm in 1:1 MeCN/H2O mixed solution were
obtained and are shown in Figure 3S (Supporting Information).
It is obvious that the temporal evolution of the early time
spectra is almost the same as that in neat MeCN, with bands at
370 and 480 nm appearing as the singlet of NQ (1NQ) that
then undergoes the ISC process (7 ps) to generate its triplet
excited state (3NQ) with absorption bands at 375 and 593 nm.
Figures 4S and 5S (Supporting Information) and Figure 4a
display the fs-TA results for β-LA in 9:1 MeCN/H2O, 4:1
MeCN/H2O, and 1:1 MeCN/H2O mixed solutions excited at
267 nm. Similar to the fs-TA results obtained in neat MeCN
solution, bands at 370 and 530 nm grow in rapidly within 3 ps
in the MeCN/H2O mixed solutions (see Figure 4S in the
Supporting Information), which are due to an internal
conversion process from Sn to S1 of β-LA. In Figure 5Sa
(Supporting Information), with decreasing intensity of the
bands at 370 and 530 nm, the band at 610 increases slightly
without an obvious isosbestic point in 9:1 MeCN/H2O mixed
solvent. However, no obvious increase in intensity at 610 nm
was detected in the 4:1 MeCN/H2O mixed solvent in Figure
5Sb (Supporting Information). This clearly indicates that the
generation of β-3LA was inhibited with the presence of H2O.
The transient absorption at 610 nm almost does not decay any
more from 3.7 to 126.7 ps (see Figure 5Sa, Supporting
Information); this demonstrates that some β-1LA still under-
goes the ISC to transform into the β-3LA while a small portion
of β-1LA undergoes an unknown reaction with water molecules.
Figure 3. Comparison the experimental resonance Raman spectrum
(10 ns time delay) of β-LA obtained in pure MeCN (pump 266 nm,
probe 368.9 nm) with the DFT-calculated normal Raman spectrum of
β-3LA. The star symbols indicate the regions affected by solvent
subtraction artifacts.
calculated Raman spectrum shows reasonable agreement with
the experimental TR3 spectrum for the β-3LA species in terms
of the vibrational frequency pattern of the two spectra. The
slight differences between the relative intensity patterns of the
experimental spectrum and the calculated spectrum can be
attributed to the experimental spectrum being resonantly
enhanced while the calculated spectrum is a normal Raman
spectrum. Most of the Raman bands seen in the TR3 spectra of
β-3LA are due to the vibrations associated with the ring C−C
stretch and C−H bend motions. Six major Raman bands at
1581, 1548, 1498, 1455, 1424, and 1344 cm−1 were observed.
The 1581, 1548, and 1344 cm−1 Raman bands are mainly due
to the ring C−C stretch, CO stretch, and H−C−H scissor
vibrational modes. The 1498 cm−1 Raman band is associated
with a CO stretch and a C−H in-plane rocking of the ring
vibrational modes. The 1440 cm−1 Raman band is due to a
combination of the C−O stretching and C−H wagging modes.
The 1424 cm−1 Raman feature is attributed by the C−H
scissoring mode.
For aromatic carbonyl compounds, the electronic config-
uration of the T1 state determines the T1 state’s reactivity
Figure 4. Fs-TA spectra of β-LA in (a) 1:1 MeCN/H2O and (b) 1:1 MeCN/H2O (pH 0) mixed solutions obtained after 267 nm excitation.
7343
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
When the concentration of water increases to 50% (1:1
MeCN/H2O mixed solvent), no increase in intensity at 610 nm
can be observed with the significant decrease of intensity of
both bands at 370 and 530 nm (see Figure 4a). To confirm the
participation of water molecules, Figure 5 presents the singlet
Supporting Information), Figure 4b, and Figure 5 (◊) display
the fs-TA spectra and kinetics for β-LA in 1:1 MeCN/H2O
(pH = 0) mixed solutions excited at 267 nm. Examination of
Figure 4b indicates that the bands at 370 and 530 nm in pH 0
acidic solution decrease in intensity much faster than those in
the 50% water solution. As shown in Figure 5, the time
constant for the singlet decay dynamics in the pH 0 solution is
about 11 ps, which is much shorter than that obtained in the
50% water neutral mixed solutions. This provides a direct piece
of evidence to support that the β-1LA may be protonated in
aqueous solutions.
Competition between Intersystem Crossing into β-3LA in
Lower Water Concentration Solutions and Protonation
Conversion into α-3LA in Water-Rich Solutions. Ns-TA
experiments were also performed for β-LA in different aqueous
solutions. The ns-TA spectra and temporal evolution at 373 nm
in the (a) MeCN, (b) 95:5 MeCN/H2O, (c) 9:1 MeCN/H2O,
(d) 8:2 MeCN/H2O, (e) 7:3 MeCN/H2O, and (f) 1:1 MeCN/
H2O mixed solutions are presented in Figures 7S−9S in the
Supporting Information. The time constants for the transient
species’ decay dynamics obtained from the ns-TA spectra are
summarized in Table 1 for β-LA in the MeCN and mixed
Figure 5. Temporal dependences of the transient absorption intensity
○
for β-LA in 9:1 MeCN/H2O (Δ), 4:1 MeCN/H2O ( ), 1:1 MeCN/
◊
□
H2O ( ), and 1:1 MeCN/H2O (pH 0) ( ) mixed solutions at 530
nm are shown. Solid lines indicate fitting using a single-exponential
function to fit the experimental data.
Table 1. Time Constants of the Decay Dynamics at 373 nm
in Different Solutions
decay dynamics at 530 nm of β-LA in the four solvents of
varying water concentration. The time constants for the singlet
decay dynamics are 38, 32, and 20 ps in 9:1 MeCN/H2O, 4:1
MeCN/H2O, and 1:1 MeCN/H2O mixed solutions, respec-
tively. These results show that the rate of the unknown reaction
of β-LA with water in the different aqueous solutions increases
noticeably with the increasing concentration of water because
the β-1LA decays much faster than that obtained in the lower
water concentration solutions. This indicates that the β-1LA
intermediate may be involved in a reaction with H2O. The high
reactivity of the ortho-diketones gives rise to the high potential
in the low excited state reduction potential.41 Solvent
polarization also induces a small barrier to proton transfer for
the zero-point motion.42 Therefore, protonation can likely
occur from β-1LA in aqueous solution, as since Hooker43 has
already pointed out that conversion from β-LA into α-LA can
take place in the presence of hydrogen chloride on the ground
state potential surface. It appears that the protonation of the
carbonyl group (O9) may also be the driving force to promote
the transformation from β-LA into α-LA on an excited state
potential surface. Since the ISC process of NQ is not affected
by the increase of the concentration of water molecules, this
implies that the pyran ring of β-LA determines the reaction
between β-1LA and water molecules. Given that the
protonation of the carbonyl group may drive β-LA into α-LA
in the acidic aqueous solution on the ground state,43 the excited
state of β-LA can also probably be easily protonated compared
to the ground state of β-LA. Fs-TA results indicate that the
lifetime of β-1LA will gradually decrease with the increase of the
concentration of water. This reveals that β-1LA may be
protonated even in the neutral high water concentration
solutions, and this is consistent with observations that
benzophenone is a very weak base in the ground state and its
protonation requires a very strong acid; however, the acidity of
the conjugate acid, pKa(S0) = 4.7, is reduced by 4 orders of
magnitude in the excited triplet state, pKa(S1) = 0.4.27,44 This
would suggest that the protonation of the excited state of β-LA
would be much easier than for its ground state. Figure 6S (see
MeCN/ MeCN/ MeCN/ MeCN/ MeCN/
solvent
t/τ
N2
MeCN
H2O
H2O
H2O
H2O
H2O
95:5
9:1
8:2
7:3
1:1
purged
τ1
6.8 μs
2.6 μs
5.8 μs
2.1 μs
4.5 μs
2.0 μs
3.8 μs
3.3 μs
2.7 μs
τ2
open air
τ
264 ns
327 ns
378 ns
427 ns
463 ns
695 ns
aqueous solvents with varying water content. The time
constants obtained in the open air condition are dramatically
shorter than that obtained in a nitrogen-purged condition. This
indicates that the transient species observed in the ns-TA
experiments have triplet character. The different lifetimes in
different solvents under open air are caused by the different
concentration of oxygen dissolved in the solutions; the lifetime
is 264 ns in the neat MeCN, while the lifetime is 695 ns in 50%
water concentration solution; this is because the oxygen has a
higher dissolve ability in MeCN than in water solution. It is
interesting to note that, for solvents with water content larger
than 30% and in the neat MeCN solvents, only one time
constant can be obtained (see Table 1), which is in accordance
with the results achieved in the fs-TA spectra recorded in 1:1
MeCN/H2O mixed solution, where the β-1LA is quenched by
the protonation and no β-3LA is generated, while in neat
MeCN solvent, the β-1LA completely transforms into the β-3LA
by ISC. Therefore, the 6.8 μs is the lifetime of β-3LA, while the
3.3 and 2.7 μs accounts for the lifetime of a new species that has
a triplet character (generated by the protonation of β-1LA). In
less than 30% water solution, there are two time constants that
can be obtained by a biexponential fitting of the dynamics of
373 nm. According to the preceding interpretation, the short
time constants (2.6, 2.1, and 2.0 μs obtained in 5, 10, and 20%
water solution, respectively) are associated with the new species
produced by the protonation of the singlet excited state of β-
LA; in contrast, the longer time constants (5.8, 4.5, and 3.8 μs)
are the lifetime of the β-3LA species.
7344
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
According to the discussion above, the transient species with
the longer lifetime is β-3LA, whose formation decreases as the
concentration of water increases. The dependence of the
transient species’ decay dynamics on water concentration
reflects the effect of the local solvent environment on the triplet
quenching process. The different lifetimes between the
nitrogen-purged and open air conditions indicate that the
species obtained in the ns-TA in 1:1 MeCN/H2O mixed
solution may be triplet in nature. Figure 6 shows the ns-TA
(c) 9:1 MeCN/H2O, (d) 8:2 MeCN/H2O, and (e) 1:1
MeCN/H2O mixed solutions using a 368.9 nm probe
wavelength and a 266 nm pump wavelength. In the MeCN
solvent, the main resonance Raman bands of β-LA are at 641,
818, 953, 1019, 1042, 1144, 1166, 1204, 1266, 1344, 1424,
1455, 1498, 1548, and 1581 cm−1. Based on the study of β-LA
in MeCN, the transient species detected at 10 ns could be
assigned to β-3LA. In the 50% water solution, the main
resonance Raman bands appear at 1150, 1171, 1469, 1502,
1553, and 1581 cm−1. According to the ns-TA study of β-LA,
the transient species detected in the 50% aqueous solution are
likely to be the α-3LA species. Thus, DFT calculations were
conducted to predict the normal Raman spectra of α-3LA, and
Figure 8 displays the comparison between the experimental
Figure 6. Ns-TA spectra of β-LA in 1:1 MeCN/H2O mixed solution
obtained after 266 nm excitation. Inset: Transient absorption decay of
○
β-LA in 1:1 MeCN/H2O mixed solution purged with N2 ( ) and
□
saturated air ( ). The solid lines indicate the kinetics fitting to the
experimental data points.
Figure 8. Comparison the experimental resonance Raman spectra (10
ns time delay) of β-LA obtained in 1:1 MeCN/H2O (pump 266 nm,
probe 368.9 nm) with the DFT-calculated normal Raman spectrum of
α-3LA.
spectra acquired after photolysis of β-LA in 1:1 MeCN/H2O
mixed solution under open air. Only one species is obtained
through the ns-TA experiment, which displays absorption
maxima at 304, 374, and 440 nm. The transient spectra here are
almost the same as the transient absorption of the triplet state
of α-LA (α-3LA) obtained in MeCN by Netto-Ferreira45 with a
small shift in the wavelength. Therefore, the character of the
transient species observed in 50% water solution can probably
be assigned to the α-3LA intermediate.
TR3 spectrum obtained at 10 ns and the B3LYP/6-311G**
DFT-calculated normal Raman spectrum of α-3LA. Inspection
of Figure 8 shows that there is excellent agreement between the
vibrational frequency patterns of the calculated normal Raman
spectrum and the ns-TR3 spectrum. The dotted lines indicate
the correspondence of the vibrational features in the calculated
and experimental spectra in Figure 8. This further indicates that
the transient species observed in the nanosecond time scale can
be assigned to α-3LA. To better understand the ns-TR3 spectra
variation in the different solutions, DFT calculations have been
employed to optimize the structures and predict normal Raman
spectra of β-3LA and α-3LA. The optimized structures of β-3LA
and α-3LA are displayed in Figure 10S. The bond lengths of
C11−C12, C14−C15, C4−O5, C1−O10, and C2−O9 are 1.417,
1.435, 1.329, 1.249, and 1.259 Å for β-3LA and 1.407, 1.419,
1.254, 1.254, and 1.315 Å for α-3LA, respectively. The DFT
calculations indicate that the Raman bands of 1498 and 1548
cm−1 are mainly contributed from the C−C stretch of C11−C12
and C14−C15 and CO stretch of C1−O10 and C2−O9 for
β-3LA. The bond lengths of C11−C12, C14−C15 (1.407 and
1.419 Å), and C4−O5 (1.254 Å) for α-3LA are shorter than
those of C11−C12, C14−C15 (1.417 and 1.435 Å), and C1−O10
(1.259 Å) for β-3LA. This suggests that Raman bands
contributed by the C−C stretch of C11−C12 and C14−C15
and CO stretch of C1−O10 and C2−O9 will shift up to higher
wavenumber when β-3LA converts into α-3LA. This is
consistent with the ns-TR3 experimental Raman spectra
changes. Figure 7 shows that the Raman bands at 1498 and
1548 cm−1 shifted to 1502 and 1553 cm−1, and there is also a
remarkable change in their relative intensity as there is an
Confirmation of the Assignment of α-3LA by ns-TR3
Spectra. To gain structural information for the reactive
intermediates, ns-TR3 experiments were performed for β-LA
in different aqueous solutions. Figure 7 displays the ns-TR3
spectra at 10 ns of β-LA in (a) MeCN, (b) 95:5 MeCN/H2O,
Figure 7. Ns-TR3 spectra of β-LA obtained with 266 nm excitation
and a 368.9 nm probe wavelength in (a) MeCN, (b) 95:5 MeCN/
H2O, (c) 9:1 MeCN/H2O, (d) 8:2 MeCN/H2O, and (e) 1:1 MeCN/
H2O mixed solutions at 10 ns time delay. The sharp features labeled by
(*) are due to solvent subtraction artifacts and/or stay light.
7345
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
Scheme 2. Proposed Protonation Photoinduced Conversion Mechanism from β-LA into α-LA in Aqueous Solutions
increase of the water concentration. Furthermore, the 1424 and
1455 cm−1 Raman bands were observed in MeCN solvent,
while they almost disappear in the 50% water solution. The
Raman bands at 1424 and 1455 cm−1 are associated with the
C−O stretch of C4−O5 (1.329 Å) for β-3LA. When the increase
in the water concentration reaches 50%, some α-3LA species are
generated and the cleavage of the C6−O5 bond gives the C4−
O5 double bond character. This causes the Raman bands
associated with the single-bond C−O stretch of C4−O5 (1.329
Å) for β-3LA to disappear. In summary, the ns-TR3 spectra
changes in the different solutions are in accordance with the
features of the optimized structures and DFT-calculated normal
Raman spectra for β-3LA and α-3LA. This demonstrates that the
transient species detected in the 50% water solution are
convincingly assigned to the α-3LA species.
Discussion of the Photochemical Reaction Mecha-
nism of β-LA in Aqueous Solutions. Based on the data
presented above, the proposed photochemical reaction path-
ways of β-LA in aqueous solutions are shown in Scheme 2.
After photoexcitation in MeCN, both the first singlet excited
state of β-LA and NQ undergo an ISC process to transform
into the β-3LA and 3NQ species with time constant of 7 and 29
ps, respectively. In the 50% water solution, the photochemical
reaction pathway of NQ is the same as that observed in the
MeCN solvent. However, β-LA exhibits a different photo-
chemical reaction in the aqueous solutions compared with the
one observed in MeCN solvent. In the 50% water solution, the
carbonyl group at the β position of first excited state of β-LA
tends to be quenched by protonation. The fs-TA results of β-
LA in the pH 0 acidic solution further indicate that the
protonation may accelerate the quenching of the β-1LA species.
Therefore, the excited state of β-LA is first protonated in the
aqueous solution or acidic solution, and then the resonant
species of the protonation of the excited state of β-LA will
promptly transform into the excited state of α-LA by the
cleavage of the O5−C6 bond and the cyclization reaction to
form the singlet excited state of α-LA. The singlet excited state
of α-LA is a very short-lived intermediate, which was quickly
consumed through an intersystem crossing and formed the
triplet state of α-LA. This apparently occurs very fast, and the
singlet excited state of α-LA was not resolved under the
resolution of the present fs-TA experiments. Finally, only the
triplet state of α-LA was detected by ns-TA and ns-TR3
experiments.
Given that Hooker43 and Ettlinger46 reported that the
ground state of β-LA can be thermodynamically converted into
α-LA in the presence of hydrogen chloride. To explore the
formation of the ground singlet state of α-LA, DFT calculations
were carried out on the cyclization reactions from the
protonated β-LA to ground singlet state of α-LA. Figure 10S
shows the optimized structures of the reactant complex (RCn),
transition state (TSn), and product complex (PCn) (for which n
represents the sequence of the reaction steps) obtained from
B3LYP/6-311G** calculations. Selected bond lengths are given
in the optimized structures shown in Figure 11S. These DFT
computations found that formation of the ground singlet state
of α-LA needed a pathway through two steps. The first step
involves the carbocation attacking the oxygen (O9) of the
alcohol group, and the second step involves a deprotonation to
produce the singlet state of α-LA. The energy barriers of the
two steps are 2.94 and 1.87 kcal/mol, respectively (see Figure
9). The very small energy barriers for these two steps
Figure 9. Reactive energy profiles obtained from (U)B3LYP/6-
311G** calculations are shown from the carbocation 2 transient to the
singlet state of α-LA (see text for more details).
demonstrate that the cyclization reactions easily proceed in
acidic aqueous solutions. Using hydrogen chloride as an acidic
catalyst for β-LA to prepare α-LA, Ettlinger reported 70%
conversion efficiency, and the final product was only
characterized by its melting point and an elemental analysis.48
In order to confirm the conversion from β-LA into α-LA in the
acidic solution, we repeated the experimental procedure and
obtained a 90% conversion efficiency from β-LA to α-LA. The
final product was characterized by NMR, mass spectra, and
7346
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
Scheme 3. Alternatively Proposed Mechanism from β-LA into α-LA in Aqueous Solutions
UV−vis absorption spectra, which are displayed in the
Supporting Information, and the NMR and mass spectra of
the final product agree well with the literature ones reported for
the spectroscopic and spectrometric properties of α-LA.45,47
Since protonation of the ground state of β-LA in the acidic
solution efficiently undergoes the decyclization and cyclization
to produce α-LA, it is believed that excited state of β-LA
(β-1LA) will also able to convert into α-LA in the high water
concentration solutions or acidic aqueous solution. As
mentioned above, the time-resolved spectra results demon-
strated that the β-1LA species undergo ISC to form the β-3LA
species in MeCN in low water concentration solutions, while
the β-1LA mostly convert into the α-3LA in high water
concentration solutions or acidic solution. This implies that
water plays a very important role in the conversion process, and
that the protonation of the β-1LA species enables the
transformation from β-1LA into the α-3LA intermediate
observed in the ns-TA and ns-TR3 spectra that eventually
forms the α-LA final product. This implies that the water
environment may significantly change the photochemical
pathways of β-LA. Our previous studies on Ketoprofen (KP)
also found that an increase in the water concentration may
shuttle a proton from the carboxyl group to the carbonyl group
via the water hydrogen bonded network that could lead to KP
undergoing a photodecarboxylation reaction in neutral high
water concentration solutions.29,48,49 This makes KP in aqueous
solutions exhibit a different photochemical pathway compared
with that obtained in a neat MeCN or low water concentration
solutions.
β-LA may also possibly be converted into α-LA by another
alternative pathway. The protonation of singlet excited state β-
LA (1β-LAH+) could be susceptible to conjugate β-addition of
water at C6, which then leads to formation of a hemiacetal (see
Scheme 3). Ketonization of the C6 carbon with the departure of
the alcohol side chain produces a benzo-1,2,4-trione with a 2-
methyyl-2-hydroxybutanyl side chain. The hydroxyl group then
bonds at the C2 carbonyl followed by dehydration to form α-
LA. In order to assess the susceptibility of the conjugate β-
addition of water at C6, the reaction energy profile of β-addition
of water at C6 was scanned for the protonated β-LAH+ species.
The computational result indicates that the energy surface of
the β-addition of water at C6 for the protonated β-LAH+
species is an endothermic reaction throughout the scan (see
Figure 12S); hence, it can be concluded that the β-addition of
water at C6 is not very likely to occur in a thermodynamic
sense. However, kinetically, we cannot exclude the possibility of
the conjugate β-addition of water at C6 because the protonated
β-LAH+ species might occur to some degree. However, the
DFT calculation results reveal that the bond lengths of O5−C4
and O5−C6 are 1.346 and 1.472 Å, respectively, for the ground
state of β-LA, while the bond lengths of O5−C4 and O5−C6 are
1.293 and 1.526 Å, respectively, after the protonation of ground
state of β-LA (see Figure 13S). This reveals that the protonated
ground state of β-LA tends to undergo the cleavage of the O5−
C6 bond more easily. Winter and co-workers reported that
some unstabilized carbocations have been found to have
structurally nearby, low-energy conical intersections that are
often favored for photochemical heterolysis, whereas some
stabilized carbocations have been found to have high-energy,
unfavorable conical intersections, which may be favored for
thermal heterolysis.50 For the formation of the carbocation
intermediates in Scheme 2, the carbocation intermediate
formed by the cleavage of the O5−C6 bond could be considered
unstabilized carbocations, so the cleavage of the O5−C6 bond
may be a photo-heterolysis control process rather than a
thermal heterolysis process.50 Therefore, the protonation of
excited state β-LA may directly help induce the cleavage of the
O5−C6 bond and is hence proposed as being the predominant
pathway (as shown in Scheme 2).
It is worth noting that the conversion from β-LA into α-LA
only takes place in aqueous solutions. Previous studies of the
photophysics and photochemistry of β-LA were conducted in
organic solvents by nanosecond laser flash photolysis19−22 and
found that β-3LA will be quenched by hydrogen donors and L-
tryptophan, L-tryptophan, methyl ester, L-tyrosine, L-tyrosine
methyl ester, and L-cysteine to form the corresponding radical
pair resulting from an initial electron transfer from the amino
acids or their esters to the excited quinone that was followed by
a fast proton transfer. However, in aqueous solutions, a
different reaction occurs to convert β-LA into α-LA. It has been
noted that β-LA can act as an anticancer drug that can inhibit
the growth of sarcoma tumors in mice3 and exhibits cytotoxicity
to leukemia cells4 and to human cancers such as breast, colon,
prostate, and lung.5−14 All of these studies were performed in
physiological and biological conditions, and the previous
studies on β-LA in the organic solvents do not appear to
explain the phototoxiciy mechanism of β-LA to the cancer cells
because of the different microenvironments. In fact, the
7347
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
aqueous solutions are more like biological conditions. Some
previous studies proposed that the mechanism of cytotoxicity in
cancer cells is due to NAD(P)H quinine oxidoreductase-1
performing a two-electron reduction of β-LA to generate its
hydroquinone.12,15−17 However, we found that the protonation
of O9 will drive β-LA to convert into α-LA after photoexcitation
in aqueous solution. Hooker43 pointed out that β-LA will
transform into α-LA in the presence of hydrogen chloride
acidic solution in the ground electronic state, and our similar
experiments with ground state β-LA also demonstrated that
adding acid can convert it into α-LA, with about 90%
conversion efficiency in acidic solution. Measurement by
microelectrodes found that the microenvironment is often
more acidic in tumors than in normal tissues.51 This indicates
that it is probable that β-LA tends to convert into α-LA in the
cancer cells’ microenvironment. Therefore, the cytotoxicity in
cancer cells for β-LA is probably caused, at least in part, by
some of its conversion into α-LA. α-LA has been demonstrated
to be a strong candidate for the development of drugs to treat
resistant cell lines with hyperactivity against multi-drug-resistant
tumors with lower expression of topoisomerase II.52,53 It can
induce DNA cleavage mediated by topoisomerase II, which is
an essential enzyme that interconverts different topological
forms of DNA and is the target for some anticancer drugs.54,55
In brief, our time-resolved spectroscopic investigation of the
excited state reaction in aqueous solutions and examination of
the ground state reaction in acidic aqueous solution of β-LA
indicate that under these two conditions a noticeable amount of
β-LA can be converted to α-LA. The cytotoxicity and
phototoxicity of β-LA in biological systems will likely depend
on the pH of the microenvironment and the presence or
absence of photoexcitation of β-LA.
the ground singlet state of the α-lapachone final product. We
also observed that some β-lapachone can be converted to α-
lapachone on the ground electronic state when a strong acid is
added to an aqueous solution of β-lapachone. Our present
study indicates that some β-lapachone can be converted to α-
lapachone by photoconversion in aqueous solutions by a water-
assisted excited state reaction or by an acid-assisted ground
electronic state reaction. We briefly discussed the implications
of these new results in aqueous solutions for the phototoxicity
and cyctotoxicity of β-lapachone in biological systems.
EXPERIMENTAL AND SECTION
■
The sample of β-LA was synthesized following a literature procedure,
and its spectroscopic and spectrometric properties are in full accord
with the structure proposed; the characterization spectra and data are
provided in the Supporting Information. NQ and spectroscopic grade
MeCN were purchased and used to prepare the sample solutions.
Unless specified, all of the mixed solvent ratios are of volume ratio.
Perchloric acid (HClO4) was used to adjust the pH value of the
aqueous solutions.
Femtosecond Transient Absorption Experiment. The fs-TA
experiments were done by employing an experimental setup and
methods detailed previously,56 and only a brief description is provided
here. Fs-TA measurements were done using a femtosecond
regenerative amplified Ti:sapphire laser system, in which the amplifier
was seeded with the 120 fs laser pulses from an oscillator laser system.
The laser probe pulse was produced by utilizing ∼5% of the amplified
800 nm laser pulses to generate a white-light continuum (350−800
nm) in a CaF2 crystal, and then this probe beam was split into two
parts before traversing the sample. One probe laser beam goes through
the sample, while the other probe laser beam goes to the reference
spectrometer in order to monitor the fluctuations in the probe beam
intensity. For the present experiments, the sample solution was excited
by a 267 nm pump beam (the third harmonic of the fundamental 800
nm from the regenerative amplifier). The 40 mL solutions were
studied in a flowing 2 mm path-length cuvette with an absorbance of 1
at 267 nm throughout the data acquisition.
Nanosecond Transient Absorption Experiment. The ns-TA
measurements were carried out with a LP920 laser flash spectrometer
provided by Edinburgh Instruments Ltd. The probe light source is a
450 W ozone-free Xe arc lamp with 10 Hz to a single shot operation
versatile sample chamber with an integral controller, high-speed pump
and probe port shutters, sample holder, and filter holders, which
produces a continuous spectrum between 280 and 800 nm.
Measurements of the ns-TA spectra were performed according to
the following procedure. The fresh sample solutions were excited by a
Q-switched Nd:YAG laser (fourth harmonic line at λ = 266 nm). The
probe light from a pulsed xenon arc lamp was passed through various
optical elements, samples, and a monochromator before being
detected by a fast photomultiplier tube and recorded with a TDS
3012C digital signal analyzer. In the kinetics mode, a photomultiplier
detector or InGaAs PIN detector is used and the transient signal
acquired using a fast, high-resolution oscilloscope. In the spectral
mode, an array detector is fitted to the spectrograph exit port to
measure a full range of wavelengths simultaneously. Unless specified
otherwise, the ns-TA experiments were performed in air-saturated
solutions and the sample solutions were made up to have an
absorbance of 1 at 266 nm.
CONCLUSION
■
The present work reported a fs-TA, ns-TA, and ns-TR3
spectroscopic investigation of the photophysical and photo-
chemical reactions of β-LA in MeCN and aqueous solutions.
For comparison, the photoreactions of the parent compound
NQ were also studied by fs-TA in MeCN and aqueous
solutions in order to help better understand the role of water
and the pyran ring in the photochemistry of β-LA in aqueous
solutions. ISC processes were observed for NQ in both MeCN
and 1:1 MeCN/H2O mixed solutions with time constants of
about 7 ps. The ISC process of β-LA takes place in MeCN to
produce the β-3LA species within 29 ps. In contrast, a
photoinduced protonation of the carbonyl group at the β
position takes place for the singlet excited state of β-LA with
the assistance of water in aqueous solutions. The protonation is
the initial step to lead to the conversion of some β-LA to α-LA
in water-rich solutions. After protonation of the carbonyl group
at the β position, the photochemical heterolysis control
cleavage of the oxygen−carbon bond (O5−C6) on the pyran
ring appears to allow the formation of an unstabilized
carbocation intermediate, which can then attack the oxygen
of the alcohol group (O9) and lead to a cyclization reaction to
form the singlet excited state of α-LA that then undergoes ISC
to generate the triplet state of α-LA that was observed in the ns-
TA and ns-TR3 experiments. The details of the ultrafast events
and possible role of a nearby conical intersection that lead to
the formation of the triplet state of α-LA observed in the ns-TA
and ns-TR3 experiments remain somewhat ambiguous, and
further work is planned to address this in the future. The
photochemically produced triplet state of α-LA then formed
Nanosecond Time-Resolved Resonance Raman Experi-
ments. The ns-TR3 experiments were done using an experimental
apparatus and methods discussed in detail previously,57,58 and only a
short description will be given here. The pump laser pulse with a
wavelength of 266 nm generated from the fourth harmonic of a
Nd:YAG nanosecond pulsed laser, a 309.1 nm probe laser pulse
produced from the first anti-Stokes hydrogen Raman-shifted laser line
from the third (355 nm) harmonic, and a 368.9 nm probe laser pulse
produced from the second anti-Stokes hydrogen Raman-shifted laser
line from the second (532 nm) harmonic were employed in the TR3
7348
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
experiments. The two Nd:YAG lasers were synchronized electronically
by a pulse delay generator to control the time delay of pump and
probe lasers, and the time delay between the laser pulses was
monitored by a fast photodiode and a 500 MHz oscilloscope. The time
resolution for the TR3 experiments was approximately 10 ns. The
pump and probe laser beams were lightly focused onto the sampling
system, and the Raman light was collected using reflective optics into a
spectrometer, whose grating dispersed the light onto a liquid-nitrogen-
cooled CCD detector. The Raman signal was acquired for 10−30 s by
the CCD before reading out in the interfaced PC computer, and 10−
30 scans of the signal were accumulated to produce a resonance
Raman spectrum. The TR3 spectra presented here were obtained by
the subtraction of a resonance Raman spectrum with negative time
delay of −100 ns (probe-before-pump spectrum) from the resonance
Raman spectrum with a positive time delay (pump−probe spectrum).
The TR3 spectra in this work were calibrated by the known MeCN
solvent’s Raman bands with an estimated accuracy of (5 cm−1).
Samples of the β-LA solutions were prepared to have a UV absorption
of ∼1 at 266 nm in a 1 mm path-length cuvette and then were used in
the TR3 experiments.
Density Functional Theory Calculations. DFT calculations
were performed employing the (U)B3LYP method with a 6-311G**
basis set. The Raman spectra were found by computing the Raman
intensities from transition polarizabilities computed by numerical
differentiation, with an assumed zero excitation frequency. A
Lorentzian function with a 15 cm−1 bandwidth for the vibrational
frequencies and a frequency scaling factor of 0.974 was used in the
comparison of the calculated results with the experimental spectra.
TD-DFT was used to calculate the excitation energies and oscillator
strengths, and the simulation of UV−vis spectra of selected
intermediates and excited state was obtained from (U)B3LYP DFT
calculations employing a 6-311G** basis set in PCM solvent mode.
No imaginary frequency modes were observed at the stationary states
of the optimized structures. All of the reactions have been explored by
optimizing the structures of the reactant (RC), transition states (TS),
and product complexes (PC). Transition states were located using the
Berny algorithm. Frequency calculations at the same level of theory
have also been performed to identify all of the stationary points as
minima for transition states (one imaginary frequency). Intrinsic
reaction coordinates were calculated for the transition states to
confirm that the relevant structures connect the two relevant minima.
All of the calculations were done using the Gaussian 03 program
suite59 operated on the high-performance computing cluster
(HPCPOWER2) at the University of Hong Kong.
ACKNOWLEDGMENTS
■
This work was supported by grants from the Research Grants
Council of Hong Kong (HKU 7035/13P) to D.L.P. Partial
support from the Grants Committee Areas of Excellence
Scheme (AoE/P-03/08) and the Special Equipment Grant
(SEG HKU/07) are also gratefully acknowledged.
REFERENCES
■
(1) Docampo, R.; Lopes, J. N.; Cruz, F. S.; de Souza, W. Exp.
Parasitol. 1977, 42, 142−149.
(2) Docampo, R.; Cruz, F. S.; Boveris, A.; Muniz, R. P. A.; Esquivel,
D. M. S. Arch. Biochem. Biophys. 1978, 186, 292−297.
(3) Docampo, R.; Cruz, F. S.; Boveris, A.; Muniz, R. P. A.; Esquivel,
D. M. S. Biochem. Pharmacol. 1979, 28, 723−728.
(4) Schaffnersabba, K.; Schmidtruppin, K. H.; Wehrli, W.; Schuerch,
A. R.; Wasley, J. W. F. J. Med. Chem. 1984, 27, 990−994.
(5) Luo, X. Q.; Patel, M.; Li, L. S.; Deberardinis, R.; Boothman, D. A.
Cancer Res. 2013, 73, 4614.
(6) Luo, X. Q.; Li, L. S.; Huang, X. M.; Cao, L. F.; Moore, Z.;
Deberardinis, R.; Brekken, R.; Gerson, S.; Lu, L. L.; Boothman, D. A.
Cancer Res. 2013, 73, 3344.
(7) Park, M. T.; Song, M. J.; Lee, H.; Oh, E. T.; Choi, B. H.; Jeong, S.
Y.; Choi, E. K.; Park, H. J. PLoS. One 2011, 6, e25976.
(8) Li, L. S.; Bey, E. A.; Dong, Y.; Meng, J. R.; Patra, B.; Yan, J. S.;
Xie, X. J.; Brekken, R. A.; Barnett, C. C.; Bornmann, W. G.; Gao, J. M.;
Boothman, D. A. Clin. Cancer Res. 2011, 17, 275−285.
(9) Imanishi, M.; Yoshida, S.; Mizuno, K. J. Pharmacol. Sci. 2011, 115,
234p−234p.
(10) Hori, T.; Kondo, T.; Lee, H.; Song, C. W.; Park, H. J. Int. J.
Hyperthermia 2011, 27, 53−62.
(11) Planchon, S. M.; Pink, J. J.; Tagliarino, C.; Bornmann, W. G.;
Varnes, M. E.; Boothman, D. A. Exp. Cell Res. 2001, 267, 95−106.
(12) Pink, J. J.; Planchon, S. M.; Tagliarino, C.; Varnes, M. E.; Siegel,
D.; Boothman, D. A. J. Biol. Chem. 2000, 275, 5416−5424.
(13) Wuerzberger, S. M.; Pink, J. J.; Planchon, S. M.; Byers, K. L.;
Bornmann, W. G.; Boothman, D. A. Cancer Res. 1998, 58, 1876−1885.
(14) Planchon, S. M.; Wuerzberger, S.; Frydman, B.; Witiak, D. T.;
Hutson, P.; Church, D. R.; Wilding, G.; Boothman, D. A. Cancer Res.
1995, 55, 3706−3711.
(15) Lamberti, M. J.; Rumie Vittar, N. B.; de Carvalho da Silva, F.;
Ferreira, V. F.; Rivarola, V. A. Phytomedicine 2013, 20, 1007−1012.
(16) Byun, S. J.; Son, Y.; Cho, B. H.; Chung, H. T.; Pae, H. O. J. Clin.
Biochem. Nutr. 2013, 52, 106−111.
(17) Park, E. J.; Choi, K. S.; Kwon, T. K. Chem.-Biol. Interact. 2011,
189, 37−44.
(18) Bentle, M. S.; Bey, E. A.; Dong, Y.; Reinicke, K. E.; Boothman,
D. A. J. Mol. Histol. 2006, 37, 203−218.
ASSOCIATED CONTENT
■
S
* Supporting Information
(19) Netto-Ferreira, J. C.; Lhiaubet-Vallet, V.; de Oliveira Bernardes,
B.; Ferreira, A. B. B.; Miranda, M. A. Photochem. Photobiol. 2009, 85,
153−159.
Additional fs-TA and ns-TA spectra and data obtained after
photoexcitation of NQ and β-LA in MeCN and aqueous
solutions of varying water concentration and pH values.
Optimized geometries, Cartesian coordinates, total energies,
and vibrational zero-point energies for the optimized geometry
from the (U)B3LYP/6-311G** calculations for the compounds
and intermediates considered in this paper. The Supporting
Information is available free of charge on the ACS Publications
website at DOI: 10.1021/acs.joc.5b00086.
(20) Netto-Ferreira, J. C.; Bernardes, B.; Ferreira, A. B. B.; Miranda,
M. A. Photoch. Photobio. Sci. 2008, 7, 467−473.
(21) Ci, X. H.; Silveira da Silva, R.; Nicodem, D.; Whitten, D. G. J.
Am. Chem. Soc. 1989, 111, 1337−1343.
(22) Ci, X. H.; Dasilva, R. S.; Goodman, J. L.; Nicodem, D. E.;
Whitten, D. G. J. Am. Chem. Soc. 1988, 110, 8548−8550.
(23) Netto-Ferreira, J. C.; Lhiaubet-Vallet, V.; da Silva, A. R.; da Silva,
A. M.; Ferreira, A. B. B.; Miranda, M. A. J. Braz. Chem. Soc. 2010, 21,
966−972.
AUTHOR INFORMATION
(24) Netto-Ferreira, J. C.; Lhiaubet-Vallet, V.; Bernardes, B.;
Buarque, A. B.; Ferreiraa, A. B. B.; Miranda, M. A. Photoch. Photobio.
Sci. 2014, 13, 1655−1660.
(25) Armendariz-Vidales, G.; Hernandez-Munoz, L. S.; Gonzalez, F.
́ ́ ́
̃
J.; de Souza, A. A.; de Abreu, F. C.; Jardim, G. A. M.; da Silva, E. N.,
Jr.; Goulart, M. O. F.; Frontana, C. J. Org. Chem. 2014, 79, 5201−
5208.
(26) Bell, R. P. The Proton in Chemistry; Chapman and Hall: London,
■
Corresponding Author
*E-mail: phillips@hku.hk.
Author Contributions
§L.D. and M.-D.L. contributed equally to this work.
Notes
The authors declare no competing financial interest.
1973.
7349
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350

The Journal of Organic Chemistry
Article
(27) Klan
́
, P.; Wirz, J. Photochemistry of Organic Compounds; John
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J.
B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.
E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J.
J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;
Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. L.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision
C.02; Gaussian, Inc.: Pittsburgh, PA, 2004.
Wiley & Sons, Ltd.: New York, 2009; pp 183−226.
(28) Li, M. D.; Huang, J.; Liu, M.; Li, S.; Ma, J.; Phillips, D. L. J. Phys.
Chem. B 2015, 119, 2241−2252.
(29) Li, M. D.; Yeung, C. S.; Guan, X. G.; Ma, J.; Li, W.; Ma, C.;
Phillips, D. L. Chem. - Eur. J. 2011, 17, 10935−10950.
(30) Hubig, S. M.; Bockman, T. M.; Kochi, J. K. J. Am. Chem. Soc.
1997, 119, 2926−2935.
(31) Becker, R. S.; Natarajan, L. V. J. Phys. Chem. 1993, 97, 344−349.
(32) Li, M. D.; Li, W.; Ma, J.; Su, T.; Liu, M.; Du, Y.; Phillips, D. L. J.
Phys. Chem. A 2011, 115, 14168−14174.
(33) Aloise, S.; Ruckebusch, C.; Blanchet, L.; Rehault, J.; Buntinx, G.;
Huvenne, J. P. J. Phys. Chem. A 2008, 112, 224−231.
(34) Wagner, P. J.; Truman, R. J.; Puchalski, A. E.; Wake, R. J. Am.
Chem. Soc. 1986, 108, 7727−7738.
(35) Wagner, P. J.; Siebert, E. J. J. Am. Chem. Soc. 1981, 103, 7329−
7335.
(36) Wagner, P. J.; Puchalski, A. E. J. Am. Chem. Soc. 1980, 102,
6177−6178.
(37) Toscano, J. P., Ed. Structure and Reactivity of Organic
Intermediates as Revealed by Time-Resolved Infrared Spectroscopy;
Wiley: New York, 2001; Vol. 26.
(38) Schwartz, B. J.; Peteanu, L. A.; Harris, C. B. J. Phys. Chem. 1992,
96, 3591−3598.
(39) Webb, S. P.; Philips, L. A.; Yeh, S. W.; Tolbert, L. M.; Clark, J.
H. J. Phys. Chem. 1986, 90, 5154−5164.
(40) Webb, S. P.; Yeh, S. W.; Philips, L. A.; Tolbert, M. A.; Clark, J.
H. J. Am. Chem. Soc. 1984, 106, 7286−7288.
(41) de Lucas, N. C.; Correa, R. J.; Albuquerque, A. C. C.; Firme, C.
L.; Garden, S. J.; Bertoti, A. R.; Netto-Ferreira, J. C. J. Phys. Chem. A
2007, 111, 1117−1122.
(42) Marx, D.; Tuckerman, M. E.; Hutter, J.; Parrinello, M. Nature
1999, 397, 601−604.
(43) Hooker, S. C. J. Chem. Soc., Trans. 1892, 61, 611−650.
(44) Ramseier, M.; Senn, P.; Wirz, J. J. Phys. Chem. A 2003, 107,
3305−3315.
(45) Netto-Ferreira, J. C.; Lhiaubet-Vallet, V.; Bernardes, B. O.;
Ferreira, A. B. B.; Miranda, M. A. Phys. Chem. Chem. Phys. 2008, 10,
6645−6652.
(46) Ettlinger, M. G. J. Am. Chem. Soc. 1950, 72, 3090−3095.
(47) Ribeiro, F. W.; Pinto, M. C. F. R.; Pinto, A. V. J. Braz. Chem. Soc.
1990, 1, 55−57.
(48) Li, M. D.; Du, Y.; Chuang, Y. P.; Xue, J.; Phillips, D. L. Phys.
Chem. Chem. Phys. 2010, 12, 4800−4808.
(49) Li, M. D.; Su, T.; Ma, J.; Liu, M.; Liu, H.; Li, X.; Phillips, D. L.
Chem. - Eur. J. 2013, 19, 11241−11250.
(50) Buck, A. T.; Beck, C. L.; Winter, A. H. J. Am. Chem. Soc. 2014,
136, 8933−8940.
(51) Griffiths, J. R. Br. J. Cancer 1991, 64, 425−427.
(52) Krishnan, P.; Bastow, K. F. Biochem. Pharmacol. 2000, 60,
1367−1379.
(53) Krishnan, P.; Bastow, K. F. Cancer Chemother. Pharmacol. 2001,
47, 187−198.
(54) Frydman, B.; Marton, L. J.; Sun, J. S.; Neder, K.; Witiak, D. T.;
Liu, A. A.; Wang, H. M.; Mao, Y.; Wu, H. Y.; Sanders, M. M.; Liu, L. F.
Cancer Res. 1997, 57, 620−627.
(55) Fortune, J. M.; Velea, L.; Graves, D. E.; Utsugi, T.; Yamada, Y.;
Osheroff, N. Biochemistry 1999, 38, 15580−15586.
(56) Li, M. D.; Ma, J.; Su, T.; Liu, M.; Yu, L.; Phillips, D. L. J. Phys.
Chem. B 2012, 116, 5882−5887.
(57) Du, L.; Zhu, R.; Xue, J.; Du, Y.; Phillips, D. L. J. Raman Spectrosc.
2015, 46, 117−125.
(58) Xue, J.; Du, L.; Zhu, R.; Huang, J.; Phillips, D. L. J. Org. Chem.
2014, 79, 3610−3614.
(59) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
7350
DOI: 10.1021/acs.joc.5b00086
J. Org. Chem. 2015, 80, 7340−7350