Selenium Blue-α and -β: Turning on the fluorescence of a pyrenyl fluorophore via oxidative cleavage of the Se-C bond by reactive oxygen species

Article abstract of DOI:10.1016/j.tetlet.2012.05.022

Rapid oxidation of nonfluorescent pyrenyl-CH₂SeAr (Ar = o-nitrophenyl) by hypochlorite yielded pyrenyl-CH₂Cl and pyrenyl-CH₂OH and turns on blue fluorescence, while slow oxidation of pyrenyl-CH₂SeAr with excess

Full text of DOI:10.1016/j.tetlet.2012.05.022

Tetrahedron Letters 53 (2012) 3843–3846
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Selenium Blue-
a
and -b: turning on the fluorescence of a pyrenyl fluorophore
via oxidative cleavage of the Se–C bond by reactive oxygen species
⇑
⇑
Wei Chen, Wan Ping Bay, Ming Wah Wong , Dejian Huang
Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
Rapid oxidation of nonfluorescent pyrenyl-CH2SeAr (Ar = o-nitrophenyl) by hypochlorite yielded pyrenyl-
CH2Cl and pyrenyl-CH2OH and turns on blue fluorescence, while slow oxidation of pyrenyl-CH2SeAr with
excess H2O2 leads to pyrenyl-CHO which emits a bluish-green fluorescence. The homolog, pyrenyl-
CH2CH2SeAr0 (Ar0 = o-nitrophenyl) reacts slower with H2O2 and ClOꢀ giving the same product, vinyl
pyrene.
Received 1 December 2011
Revised 18 April 2012
Accepted 3 May 2012
Available online 10 May 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Pyrene
Reactive oxygen species
Fluorescence
Organoselenium
Molecular probes
Ultrasensitive and selective detection and imaging of reactive
oxygen species (ROS) has drawn significant attention because of
the need to study the purported dual roles of ROS as causative
agents for oxidative stress,1 cell signal transduction molecules,2,3
and immunity.4 In the search for new types of turn-on fluorescent
probes for ROS, we were inspired by the fact that Se-dependent
peroxidases, such as glutathione peroxidase (GPx), utilize Se in
the active center by transferring oxygen from peroxides to give sel-
enol, which is also proposed as the reaction intermediate of GPx
mimics including ebselen and its analogs.5 Recently, a redox-
responsive organoselenium probe, benzylselenide–tricarbocyanine
was used in the detection of peroxynitrite.6 Selenium could be a
good fluorescent quencher due to its reducing properties and
possible heavy atom effect.7 We prepared new organoselenium
compounds 1 and 2 containing pyrenyl fluorophores through
one-step substitution reactions from the corresponding halide
and phenyl selenocyanate in the presence of sodium borohydride
and found that they demonstrated interesting reactivity with dif-
ferent ROS (Scheme 1).
but it is nonessential in fluorescence quenching. Reaction of 1a in
deuterated chloroform with one equivalent of hydrogen peroxide
under ambient conditions yielded firstly the oxygenated product
3 as revealed by the 1H NMR spectrum, which had a characteristic
AB pattern for the two magnetically inequivalent methylene pro-
tons at 5.05 ppm due to the chiral Se center. As the reaction pro-
gressed in the presence of excess H2O2, the concentration of 3
decreased and new products formed, as was judged from the com-
plex proton NMR signals in the aromatic region. A characteristic
singlet appeared at 10.8 ppm indicating the presence of an alde-
hyde proton. Indeed, 1-pyrenecarboxaldehyde (4) was detected
O
(1 : 2 molar ratio)
Se
+
OH
PY
OH
Cl
PY
PY-CHO
4
+
)
7
6
(PY =
NO2
5
ClO
CDCl
H2O2
We coined nicknames for compounds 1a, 1b: ‘Selenium Blue-a’
water, ClO-
PY
H2O2, 5 h
PY
Se
O
3
and 2: ‘Selenium Blue-b’ based on their unique fluorescent proper-
ties upon oxidation by hypochlorite and hydrogen peroxide. The
phenyl analog, 1b, does not fluoresce; hence selenium is responsi-
ble for fluorescence quenching. The nitro group in 1a may also help
PY
OH
Se
NO2
10 min
R
6
1, Selenium Blue-α
1a,
1b,
R = NO2,
R = H
ClO-
water, rt.
PY
PY
PY
Se+
Se
NO2
NO2
⇑
Corresponding authors. Tel.: +65 65164320; fax: +65 67791691 (M.W.W.); tel.:
O-
8
+65 65168821; fax: +65 67757895 (D.H.).
E-mail addresses: chmwmw@nus.edu.sg (M.W. Wong), chmhdj@nus.edu.sg
(D. Huang).
2, Selenium Blue-β
Scheme 1.
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.05.022

3844
W. Chen et al. / Tetrahedron Letters 53 (2012) 3843–3846
800
700
600
500
400
300
200
100
0
360
410
460
510
560
800
700
600
500
400
300
200
100
0
Figure 1. Top: Contrast reaction products and fluorescence signals between 1a and H2O2/ClOꢀ. Bottom: Selectivity of 1a (right, kex = 337 nm and kem = 392 nm, blue: 5 s, red:
10 min) and (left, kex = 365 nm and kem = 392 nm, blue: 1 h, red: 3 h), rt in water, with kex = 337 nm and emission at 392 nm. TBHP = t-butyl hydroperoxide,
ONOOꢀ = peroxynitrite. [Probe] = 50 nM, [ROS] = 1 M. Image conditions: [Probe] = 50 nM. Reaction time: 1a + H2O2, 5 h, 1a + ClOꢀ, 10 min.
2
l
by comparing the 1H NMR spectrum with that of an authentic sam-
ple of 4. The isolated yield of 4 was 85% on preparative scale. The
identity of 4 was further confirmed from the EI-MS spectrum,
(TBHP) and singlet oxygen exhibited only weak fluorescence
enhancement.
We propose that oxidative C–Se bond cleavage of 1a involves a
two-step process. Firstly, the Se in 1a is oxidized to a selenoxide,
and the resulting ArSe(O) group becomes a better leaving group.
In the presence of a nucleophile, ligand substitution occurs to
break the C–Se bond. Nucleophilic substitution by H2O2 would lead
to unobserved PY-CH2OOH, which decomposes to water and PY-
CHO. When excess tetrabutylammonium chloride (TBAC) was
added to the reaction mixture, there was no PYCHO detected and
only PY-CH2Cl formed. However, when ClOꢀ was the oxidant in
chloroform, we did not observe the formation of 3. The reaction
products were found to be highly dependent on the solvents used.
PYCH2Cl and PYCH2OH were detected when chloroform was the
solvent. In a nucleophilic solvent such as methanol, PYCH2OCH3
was the sole product, and PYCH2OH was the only product if water
was used as the solvent. The high reactivity of ClOꢀ made organic
amine buffers (such as Tris and HEPES) unsuitable for sensing ClOꢀ
(Fig. S9, Supplementary data) as the amine group competes the
reaction with hypochlorite. In the nonprotic solvent chloroform,
in situ generated Clꢀ from ClOꢀ is a much stronger nucleophile
and thus gave rise to PYCH2Cl, which was also found to be reactive
toward water (or HOꢀ) and methanol to give PYCH2OH and PY-
CH2OCH3. Organoselenium-containing fluorescent probes are rare.
An aryl-Se-derived rhodamine probe was reported and used for
sensing thiols through substitution of aryl-Se by RS, which turns
on the fluorescence; remarkably, the probe was not responsive to
reactive oxygen species.12
which showed
a molecular ion peak at m/z = 230.0731 for
C17H10O (calcd 230.0732). The Se-containing species was identified
from the mass spectrum as 2-nitrophenylseleninic acid (5) (m/
z = 234, ESI-MS anion mode), with the correct isotope distribution
for [C6H4NO4Se]ꢀ. The fluorescence spectrum of a dilute reaction
solution in water demonstrated a strong emission peak (excited
at 337 nm) at 470 nm and the color of this solution was greenish
blue (Fig. 1).8 It is remarkable that selenium oxide 3 showed only
weak fluorescence. This is in sharp contrast to the diphenyl-1-
pyrenylphosphine (DPPP), which has weak fluorescence.9 Upon
oxidation by lipid peroxidase to the phosphine oxide (DPPP@O),
the pyrene fluorescence was turned on. This property was applied
in sensing lipid peroxides.10
When testing the selectivity of 1a toward different reactive oxy-
gen species, we were surprised to find that it reacted with equimo-
lar hypochlorite within minutes at room temperature in water to
give highly fluorescent species 6 and 7 in a 1:2 ratio as revealed
by the 1H NMR spectrum. The identities of 6 and 7 were confirmed
by comparing their 1H NMR spectra with authentic compounds
prepared independently. Consistently, the products 6 and 7 gave
strong blue fluorescence as shown in Figure 1. The fluorescence
spectra of the reaction product of 1a and hypochlorite showed
two emission maxima, one at 392 nm and the other at 380 nm,
but no 1-pyrenecarboxaldehyde fluorescence peak was present.
Organoselenium compound 1a is a highly selective and highly sen-
sitive probe for rapid detection of ClOꢀ in buffer solutions or in
water. Very weakly fluorescent Selenium Blue-b, 2, also reacted
with H2O2 and ClOꢀ but gave rise to the same product, vinyl pyrene
(8), probably via an unobserved intermediate selenoxide that
To elucidate further the mechanism of the reactions of ClOꢀ
with 1a and 2, density functional theory (DFT) calculations were
carried out. The calculated relative free energies (DG298) of the var-
ious reaction pathways are summarized in schematic energy dia-
grams, as shown in Figures 2 and 3. For both Selenium Blue
derivatives (1a and 2), the formation of a selenoxide intermediate
is the rate-determining step of the overall reaction with ClOꢀ. This
oxidation step is predicted to be strongly exothermic, ꢀ195 and
eliminates ArSe–OH readily.11 Compound
2 shows a slower
reaction rate in comparison with 1a. We also treated 2 with differ-
ent ROS and found that, after longer reaction times, hypochlorite
still showed the best response while t-butyl hydroperoxide

W. Chen et al. / Tetrahedron Letters 53 (2012) 3843–3846
3845
PYCH2OH + ArSeO− + Cl−
[-344]
O
6
PY
+ OH−
HO
C
Se
Ar
HH
TS2 [-110]
O
+ ClO−
Cl
+ Cl−
[0]
Se
PY
PYCH2Cl + ArSeO−
Se
PY
Ar
CH2
1a
Ar
CH2
7
[-162]
PY
C
3
O
Se
+ Cl−
[-195]
O
Cl
+ ClOH
(0)
Se
PY
Ar
HH
TS3 [-68]
Ar
CH2
TS1 [49]
Cl
Se
+ OH−
PY
PYCH2OH + ArSeCl
PY = 1-pyrenyl
Ar = -nitrophenyl
Ar
CH2
o
6
9
(-150)
(81)
Figure 2. Schematic potential energy diagram showing the various reaction pathways for the reactions of Selenium Blue-
PCM-B3LYP/6-31G⁄ relative free energies ( G298) are given in kJ molꢀ1. Energies related to ClOꢀ and HOCl reactions are in square brackets and parentheses, respectively.
a
(1a) with ClOꢀ and HOCl. PCM-B3LYP/6-311+G⁄⁄//
D
+ ArSeOH + Cl−
[-319]
Py CH CH2
8
O
Se
H
H
− [0]
Cl
Se
CH2
+ ClO
CH2
Py
O
Ar
C
Py
H2
TS6
Ar
CH2
Se
Ar
CH2
2
[-77]
Py
CH2
O
+ ClOH
(0)
10
Se
CH2
+ Cl−
+ Cl−
[-175]
Ar
Py
Cl
Se
TS4
[84]
CH2
CH2PY
+ OH−
O
PyCH2CH2Cl
+ ArSeO−
[-144]
Ar
Py
CH2
Cl
C
Se
(97)
Ar
H H
Py = 1-pyrenyl
Ar = -nitrophenyl
[-49]
TS5
PyCH2CH2OH + ArSeCl
(-128)
o
Figure 3. Schematic potential energy diagram showing the various reaction pathways for the reactions of Selenium Blue-b (2) with ClOꢀ and HOCl. PCM-B310LYP/6-
311+G⁄⁄//PCM-B3LYP/6-31G⁄ relative free energies (
D
G298) are given in kJ molꢀ1. Energies related to ClOꢀ and HOCl reactions are in square brackets and parentheses,
respectively.
ꢀ175 kJ molꢀ1 for 1a and 2, respectively. In agreement with the
experimental observation, 1a is calculated to have a significantly
lower activation barrier of 49 kJ mol-1, via transition state TS1 (cf.
84 kJ molꢀ1 in 2, via TS4).
mic (ꢀ319 kJ molꢀ1). The calculated activation barrier of
98 kJ molꢀ1, via a coplanar transition state, is smaller than that of
the Clꢀ displacement reaction (Fig. 3). Hence, the formation of vi-
nyl pyrene is calculated to be kinetically and thermodynamically
more favorable, and pleasingly, in agreement with the experimen-
tal findings. These DFT calculations have shed light on the reaction
mechanism and confirmed the remarkable observed reactivity and
product preference of the two organoselenium probes.
In summary, new types of turn-on fluorescent probes taking
advantage of the reactivity of organoselenium compounds were
synthesized and can be used for selectively sensing hypochlorite
and H2O2. Se probes with longer wavelength excitation and emis-
sion would be of greater practical interest for bioimaging of reac-
tive oxygen species.
Selenoxide intermediate 3 can undergo nucleophilic substitu-
tion by HOꢀ or Clꢀ to form substitution products 6 or 7, respec-
tively. These SN2 displacement reactions are exothermic and
have a moderate energy barrier (Fig. 2). Not surprisingly, HOꢀ sub-
stitution, via TS2, is calculated to be more favorable than Clꢀ sub-
stitution, via TS3, both thermodynamically and kinetically. TS2 has
a significantly lower activation barrier than TS3 (by 42 kJ molꢀ1).
Hence, we predict that HOꢀ substitution, leading to pyren-1-yl-
methanol 6, is dominant in aqueous medium. This is in excellent
accord with experimental observations. In CDCl3, both 6 and 7
were observed. In this case, 6 is probably formed through reaction
of 1a with HOCl, which yields intermediate 9 and OHꢀ (Fig. 2).
Reaction of 9 with OHꢀ will finally lead to 6. Based on the energies
of the intermediates and products, the formation of 6 is less favor-
able than the formation of 7. In the case of 2, selenoxide can readily
undergo a well-established Grieco–Sharpless elimination to form
vinyl pyrene.11 This syn elimination reaction is strongly exother-
Acknowledgments
The authors are grateful for the financial support of the Science
and Engineering Research Council (SERC) of the Agency for Science,
Technology and Research (A⁄Star) of Singapore. Grant Number:
072-101-0015.

3846
W. Chen et al. / Tetrahedron Letters 53 (2012) 3843–3846
3. Zinkevich, N. S.; Gutterman, D. D. Am. J. Physiol. Heart C 2011, 301, H647–H653.
Supplementary data
4. Gabrilovich, D. I.; Nagaraj, S. Nat. Rev. Immunol. 2009, 9, 162–174.
5. Mugesh, G.; du Mont, W. W.; Sies, H. Chem. Rev. 2001, 101, 2125–2180.
6. Xu, K.; Chen, H.; Tian, J.; Ding, B.; Xie, Y.; Qiang, M.; Tang, B. Chem. Commun.
2011, 47, 9468–9470.
7. De, A. K.; De, S. K.; Mallik, A. K.; Ganguly, T. Spectrochim. Acta, Part A 2001, 57A,
1427–1441.
8. Ellison, E. H.; Moodley, D.; Hime, J. J. Phys. Chem. B 2006, 110, 4772–4781.
9. Akasaka, K.; Suzuki, T.; Ohrui, H.; Meguro, H. Anal. Lett. 1987, 20, 797–807.
10. Okimoto, Y.; Warabi, W.; Wada, Y.; Niki, E.; Kodama, T.; Noguchi, N. Free
Radical Biol. Med. 2003, 35, 576–585.
Supplementary data (synthetic methods, conditions for the ana-
lytical experiments for quantifying and imaging ROS and the coor-
dinates of all optimized equilibrium and transition state structures
(together with total energies and imaginary frequencies)) associ-
ated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.tetlet.2012.05.022.
11. (a) Grieco, P. A.; Gilman, S.; Nishizawa, M. J. Org. Chem. 1976, 41, 1485–1486;
(b) Sharpless, K. B.; Young, M. W. J. Org. Chem. 1975, 40, 947–949.
12. (a) Tang, B.; Xing, Y.; Li, P.; Zhang, N.; Yu, F.; Yang, G. J. Am. Chem. Soc. 2007,
129, 11666–11667; (b) Tang, B.; Yin, L.; Wang, X.; Chen, Z.; Tong, L.; Xu, K.
Chem. Commun. 2009, 45, 5293–5295.
References and notes
1. Dickinson, B. C.; Chang, C. J. Nat. Chem. Biol. 2011, 7, 504–511.
2. Finkel, T. J. Cell Biol. 2011, 194, 7–15.