Hybrids of a 9-anthracenyl moiety and fluorescein as chemodosimeters for the detection of singlet oxygen in live cells

Article abstract of DOI:10.1039/c9ob02070e

Singlet oxygen (¹O₂) plays an important role in human innate immune response, plant physiology and anticancer photodynamic therapy (PDT). Therefore, its monitoring by convenient and sensitive methods (e.g. by detecting a fluorescence signal) by using non-toxic reagents would be advantageous. Known fluorogenic ¹O₂-chemodosimeters can potentially consume reducing agents in cells leading to the generation of toxic side products that limit their applications. In this paper we report on a series of 9-anthracenyl-fluorescein hybrids, which do not require any reducing agents for their reaction with ¹O₂. The selected compound 8d at a very low concentration of 100 nM is able to detect ¹O₂ in live human promyelocytic leukemia HL-60 cells with over 35-fold fluorescence signal enhancement within only 20 min assay time. This chemodosimeter is not toxic to HL-60 cells at concentrations ≤1 μM (higher concentrations were not tested) even at long incubation times ≤48 h.

Full text of DOI:10.1039/c9ob02070e

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Hybrids of a 9-anthracenyl moiety and fluorescein
as chemodosimeters for the detection of singlet
oxygen in live cells†
Cite this: Org. Biomol. Chem., 2019,
17, 9883
Serghei Chercheja,a Steffen Daum,‡a Hong-Gui Xu,a Frank Beierleinb and
a
Andriy Mokhir
*
Singlet oxygen (1O2) plays an important role in human innate immune response, plant physiology and
anticancer photodynamic therapy (PDT). Therefore, its monitoring by convenient and sensitive methods
(e.g. by detecting a fluorescence signal) by using non-toxic reagents would be advantageous. Known
fluorogenic 1O2-chemodosimeters can potentially consume reducing agents in cells leading to the gene-
ration of toxic side products that limit their applications. In this paper we report on a series of 9-anthracenyl-
fluorescein hybrids, which do not require any reducing agents for their reaction with 1O2. The selected
1
Received 23rd September 2019,
Accepted 1st November 2019
compound 8d at a very low concentration of 100 nM is able to detect O2 in live human promyelocytic
leukemia HL-60 cells with over 35-fold fluorescence signal enhancement within only 20 min assay time.
This chemodosimeter is not toxic to HL-60 cells at concentrations ≤1 μM (higher concentrations were
not tested) even at long incubation times ≤48 h.
DOI: 10.1039/c9ob02070e
rsc.li/obc
Excellent chemiluminescent chemodosimeters suitable
for the detection of O2 in live cells are known: e.g. commer-
Introduction
1
Singlet oxygen (1O2) is an excited state of triplet oxygen (3O2). cially available trans-1-(2′-methoxyvinyl)pyrene6 and recently
This compound can be produced either via photosensitizer- reported SOCL-CPP.7 Though a lot of effort was invested in the
mediated1 or thermal chemical reaction e.g. between ClO− and search for suitable fluorogenic chemodosimeters, the best
H2O2.2 Though 1O2 is highly reactive, it is also formed in known systems are still not ideal for practical applications. For
nature. In particular, the photosensitized synthesis takes place example, Singlet Oxygen Sensor Green (SOSG, Thermo Fischer,
in plants (a natural pigment chlorophyll acts as a photosensiti- Scheme 1A) was the first commercially available fluorogenic
zer3) and human skin,1 whereas the thermal one – in neutro- reagent for 1O2 detection. Supposedly, it is a hybrid system
phils thereby mediating the killing of bacteria.4 This is one of containing a fluorescein dye covalently linked to a 9-methyl-
the key chemical reactions responsible for the normal function anthracen-10-yl fragment (compound 1, where Ri = Riv = Rv =
1
of the innate immune system in humans. Furthermore, O2 is H, Rii
= = 2′-substituted fluorescein residue,
CH3, Riii
a cytotoxic mediator that is photochemically generated during Scheme 1). SOSG suffers from a relatively high residual fluo-
the photodynamic therapy (PDT) of cancer diseases.5 rescence in the absence of 1O2 (see e.g. Fig. 1A) that causes the
Therefore, the detection of 1O2 directly in live cells is impor- undesirable background signal thereby reducing sensitivity in
tant for understanding its biological function and has impli- the detection of 1O2. Furthermore, SOSG can itself catalyse 1O2
cations in medicine, e.g. in controlling of PDT efficacy as well formation and induce its own decomposition.8 Finally, for the
as diagnosis of neutrophil pathologies.
related compound HOCD-IR-780 (Scheme 1A) it is known that
the endoperoxide intermediate initially formed in the reaction
1
with O2 is converted to the di-ol derivative (Scheme 1A).9 The
aDepartment of Chemistry and Pharmacy, Organic Chemistry II,
Friedrich-Alexander-University of Erlangen-Nürnberg (FAU), 91058 Erlangen, Germany.
E-mail: Andriy.Mokhir@fau.de
bComputer-Chemistry-Center and Interdisciplinary Center for Molecular Materials,
Department of Chemistry and Pharmacy, Friedrich-Alexander University
Erlangen-Nürnberg (FAU), 91052 Erlangen, Germany
latter redox process consumes an endogenous reducing agent
(Red) that might lead to the generation of toxic radicals and
disturbance of redox balance in live cells. For other previously
reported fluorogenic chemodosimeters, e.g. Aarhus sensor
green10 (Scheme 1A), Si-DMA (now commercially available, e.g.
†Electronic supplementary information (ESI) available: Complementary syn-
thetic schemes and characterization of new probes. See DOI: 10.1039/
c9ob02070e
from Merck),11 compound
2
developed by our group12
(Scheme 1B), and 3 13 (Scheme 1B) as well as SOSG, similar
‡Present addresses: Merck KGaA, Schaffhausen, Switzerland.
redox processes are expected to occur in the presence of 1O2
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9883–9891 | 9883

View Article Online
Paper
Organic & Biomolecular Chemistry
leading to the formation of toxic species as outlined in
Scheme 1A and B: toxic products/moieties are labelled with
red colour.
The goal of this work was to develop chemodosimeters with
1
low background fluorescence in the absence of O2 and strong
1O2-mediated fluorescence increase. We wanted to achieve that
the endoperoxide intermediate formed in the latter reaction is
quickly converted to stable products without external reducing
agents. That would solve the problem of undesired redox pro-
cesses induced by the known 1O2-chemodosimeters in live
cells.
Results and discussion
1
A structure of O2-chemodosimeter 4 reported in this paper is
shown in Scheme 1C. In this molecule a fluorescein dye is
masked by 3′-O-alkyl and 6′-O-aryl groups. The 3′-O-alkyl con-
tains a carboxylic group to provide for the solubility of the
probe in neutral aqueous solution. Another group is an 1O2
−
reactive 9-anthracenyl fragment. Chemodosimeter 4 will react
with 1O2 with the formation of endoperoxide 4_O2, which is
expected to decompose cleanly to 9,10-anthraquinone AQ and
strongly fluorescent 5. Electrons from a C(10)–H bond will be
used in this reaction step as intramolecular reductants. Thus,
no external electrons will be required to form the stable final
products. These considerations are based on our previous
experience with 9-anthracenyl derivatives of RNAs and
DNAs.14–21
We prepared compound 4 as outlined in Scheme 2. In par-
ticular, fluorescein was first 3′-O-alkylated with 2-bromo-2-
methylpropionate to obtain derivative 6 (yield 15%). Further
alkylation of the second phenolic oxygen with 9-bromoanthra-
cene in the presence of a catalyst Pd(OAc)2/t-BuXPhos (1/1)
(leading to 7a with the yield of 12%), followed by hydrolysis of
the methyl ester by LiOH furnished non-fluorescent com-
Scheme 1 (A, B) Known probes (1–3) for the detection of singlet
oxygen (1O2) and mechanisms of their reactions with 1O2; toxic species/
reactive chemical fragments released in these reactions are red
coloured; sign “?” indicates unknown species, which should have been
formed to account for the atom balance in the chemical equation of the
shown reaction. (C) A structure of new 1O2-probe 4 and the mechanism
of its reaction with 1O2.
Fig. 1 (A) Fluorogenic response of newly prepared probes 4 and 8b–e
(each 2 µM) and known 1O2 probe SOSG (Singlet Oxygen Sensor Green;
Thermo Fischer, 2 µM) in the presence of photochemically generated
1O2: F0 – initial fluorescence intensity (λex = 495 nm, λem = 515 nm); F60
– fluorescence intensity after 60 min of irradiation with red light in the
presence of photosensitizer In(pyropheophorbide-a)chloride (0.5 µM) in
phosphate/NaCl buffer (pH 7.4); (B) fluorogenic response (F60/F0) of
newly prepared probes 4 and 8b–e (each 2 µM) in the presence of
different reactive oxygen species. Other experimental conditions are
described in the “Experimental” section of the paper.
Scheme 2 Synthesis of 1O2 probes 4 and 8b–8e: (a) t-BuOK, methyl
2-bromo-2-methylpropionate; (b) R-Hal, where Hal = Br or Cl, further
details are described in “Synthetic procedures” section; (c) LiOH, THF,
water.
9884 | Org. Biomol. Chem., 2019, 17, 9883–9891
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
pound 4 with the overall yield of 1.4% over three steps. The
low yield is caused by a low conversion of the starting
materials in the first and second steps. The final step (hydro-
lysis) occurs quantitatively according to the chromatographic
analysis, whereas the isolated yield was 78%.
Next, we studied the reaction of 4 with 1O2 by using electro-
spray mass spectrometry (ESI-MS). Compound 4 (final concen-
tration: 30 μM) was dissolved in CH3CN and H2O2 was added:
9/1 (v/v), final concentration of H2O2 – 30 mM. After incu-
bation for 1 h at 22 °C the ESI mass spectra of the resulting
solution in both negative and positive modes were acquired.
We detected in the spectra peaks corresponding to 9,10-
anthraquinone (AQ, calcd for C14H9O2 [M + H]+ 209.0602,
found m/z 209.0608, ESI+) and compound 5 (calcd for C24H7O7
[M − H]− 417.3935, found 417.3940, ESI−). These results (for-
Fig. 2 Relative HOMO energies of simple anthracene derivatives. These
compounds are models of probes 4 and 8a–e. 9-Amino-10-phenoxyan-
thracene was used as a reference compound (relative HOMO energy 0).
This compound was found to be easily oxidized even in the presence of
triplet oxygen.
1
mation of AQ in the reaction of probe 4 with O2) were further
corroborated with 1H NMR spectroscopy (Fig. S25, ESI†).
Moreover, in the fluorescence spectrum of a solution of 4
treated with H2O2 we observed the characteristic of compound
5 emission band centred at λem = 515 nm (λex = 488 nm). These pound. We observed that the reference compound was mostly
1
data confirm the mechanism of the reaction of O2 with com- electron rich, followed by the group of derivatives with similar
pound 4 outlined in Scheme 1C.
To evaluate the performance of 4, we compared the 1O2- a substantially more electron deficient model of 8b.
mediated increase of fluorescence of this compound with that
To validate these theoretical data, we prepared the corres-
HOMO energies (models of 8b, 8d, 4, 8c) and concluding with
of commercially available chemodosimeter SOSG (Fig. 1A). The ponding derivatives 8b–8e as outlined in Scheme 2 and
fluorescence intensity (λex = 495 nm, λem = 515 nm) of a solu- Schemes S1, S2 in the ESI.† In particular, we found that
tion of 4 (2 μM) in phosphate/NaCl buffer (pH 7.4) was 9-bromo-10-nitroanthracene is reactive enough to directly
increased from 2.4 arbitrary units (a.u., F0) to 26.3 a.u. (F60) alkylate intermediate 6 in the presence of slight excess Cs2CO3
1
upon reaction with O2, which was generated in the presence leading to 7b with the moderate isolated yield of 57%. After
of the photosensitizer In(pyropheophorbide-a)chloride11 the hydrolysis of 7b by LiOH compound 8b (replacement of
(0.5 µM) upon irradiation with red light. This corresponds to –H(10) for –NO2) was obtained with the yield of 51%. To reduce
an 11.2-fold fluorescence increase (F60/F0). In contrast, under a nitro-group in 7b we used SnCl2·2H2O. However, the product
the same conditions F60/F0 for SOSG was found to be only 4.2- (the reference compound, where the nitro-group is converted
fold that is due to the rather high background fluorescence of to an amino-group) was found to be extremely unstable under
the latter chemodosimeter (F60 = 194.2 a.u). Importantly, we aerobic conditions. This did not allow the isolation of this
observed that the fluorogenic response of compound 4 is product in pure form. Therefore, we conducted the reduction
specific towards 1O2 as compared to other reactive oxygen of 7b in the presence of acetic acid anhydride (Ac2O) to trap
species including H2O2 (F60/F0 = 1.7), O2 (F60/F0 = 1.1), HO• the unstable intermediate as the N-acylated product, 7c. Under
•−
(F60/F0 = 1.4) and control conditions (hν = red light irradiation the experimental conditions used in the synthesis of 4, as
in the absence of the photosensitizer, F60/F0 = 1.6, Fig. 1B).
described above, the ester group of 7c was hydrolysed leading
Furthermore, we explored the role of substituents in the to compound 8c (replacement of –H(10) for –NHAc).
10th position of the anthracene moiety (replacement of –H(10) Compounds 7d and 7e were obtained by the alkylation of inter-
for –NO2, –NHAc and –NH2) and the type of linker between mediate 6 with the corresponding 9-(chlormethoxy)anthracene
this moiety and a fluorophore (comparing the direct linkage to and 2,6-di-tert-butyl-9-(chlormethoxy)anthracene. The latter
the fluorophore with that via a methylene bridge) in the reac- two reagents were generated in situ from 9-(methyl-
tion of 4 with 1O2.
thiomethoxy)anthracene derivatives and sulfuryl chloride as
Since 1O2 is an electron deficient molecule, we expected described in the Experimental section. The following hydro-
that the energies of HOMOs of the anthracene-containing che- lysis furnished the desired products 8d and 8e.
modosimeters will correlate with their reactivity in the
We found that, in agreement with the DFT calculations,
addition of 1O2 (the first step in the reaction outlined in derivative 8b carrying the strongest electron-acceptor moiety
Scheme 1C). Therefore, we determined the energies of HOMOs –NO2 reacts less efficiently with 1O2 compared to other pre-
of derivatives modelling 4 and its analogues 8b–8e, where the pared compounds (Fig. 1B). Interestingly, the reactivities of 8e,
fluorescein fragment was replaced with a phenyl group, by 8d, 4, and 8c are substantially different from each other (8e >
using the B3LYP hybrid density functional and the 6-31G** 8d > 4 > 8c), whereas according to the DFT calculation the
basis set (Fig. 2, experimental description is given in the ESI†). addition of 1O2 to all of these compounds would go on with
9-Amino-10-phenyloxyanthracene was used as a reference com- similar efficacy (Fig. 3). This might indicate that both reaction
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9883–9891 | 9885

View Article Online
Paper
Organic & Biomolecular Chemistry
Commercially available chemicals were used without
further purification. Anhydrous reagents: o-xylene, CH2Cl2,
N,N-dimethylformamide (DMF), dioxane, trimethylamine
(TEA), dimethyl sulfoxide (DMSO), CHCl3 as well as fluo-
rescein, potassium t-butoxide (t-BuOK), methyl 2-bromo-
2-methylpropionate, 9-nitroanthracene, 9-bromoanthracene,
CH3SCH2Cl, sodium hydride (60% dispersion in mineral oil),
bromine, SnCl2·2H2O, Pd(OAc)2, 2-di-tert-butylphosphino-
2′,4′,6′-triisopropylbiphenyl (t-BuXPhos), Cs2CO3, SO2Cl2,
anthracene, tert-butyl chloride (t-BuCl), AlCl3, (NH4)2[Ce(NO3)6],
glacial acetic acid (AcOH), Sn dust, xanthine oxidase (XO) and
xanthine were purchased from Sigma-Aldrich. CH3NO2 was
purchased from Alfa Aesar and Singlet Oxygen Sensor Green
(SOSG) from Thermo Fisher Scientific. Thin layer chromato-
Fig. 3 (A) Monitoring the formation of 1O2 in HL-60 cells loaded with In
(pyropheophorbide-a)chloride (InPpa, 200 nM) upon irradiation with red
(650 nm) light. 8d (100 nM) was used as a probe (black bars). Cells
loaded only with InPpa (striped bars) or only probe 8d (grey bars) were
used as controls; (B) effect of probe 8d on the viability of HL-60 cells
(incubation for 48 h). Similar effects were obtained for probe 4 (data not
shown).
graphy (TLC) was performed on
a Merck DF-Alufoilien
(60F254, 0.2 mm pre-coated plates). Compounds on TLC plates
were visualized by exposure to UV light. NMR spectra were
acquired on a 300 MHz Bruker Avance spectrometer. 1H and
13C NMR spectra were referenced to the residual protonated
solvent (1H) or the solvent itself (13C). All chemical shifts are
reported in parts per million (ppm). For CDCl3, the shifts are
referenced to 7.27 ppm for 1H NMR spectroscopy and
77.0 ppm for 13C NMR spectroscopy. For acetone-d6, the shifts
are referenced to 2.05 ppm for 1H NMR spectroscopy and
29.9 ppm for 13C NMR spectroscopy. Abbreviations used in
the description of resonances are: s (singlet), d (doublet),
t (triplet), q (quartet), quin (quintet), br (broad), m (multiplet),
dd (doublet of doublets), and ddd (doublet of doublet of doub-
lets). Coupling constants (J) are quoted to the nearest 0.1 Hz.
ESI and APPI measurements were recorded on a micrOTOF II
focus mass spectrometer (Bruker Daltonik GmbH, Germany)
in a positive and negative mode. Fluorescence spectra were
acquired on an Agilent Cary Eclipse fluorescence spectrophoto-
meter by using fluorescence cuvettes (Hellma GmbH,
Germany) with a sample volume of 1 mL. The fluorescence of
live cells was quantified using a Guava easyCyte™ 6-2L flow
cytometer (Merck Millipore). Channel FL1 was used to quantify
steps (addition of 1O2 as well as AQ formation, Scheme 1C)
contribute to the overall reaction rate. It could also be noted
that linking of the anthracene moiety to the dye via a methyl-
ene bridge (8d, 8e) is more favorable for the chemodosimeter
reactivity than its direct linking (compare with the data for 4
and 8c, Fig. 1B).
1
Based on the high reactivity with O2 (Fig. 1B) and its mod-
erate synthetic accessibility we selected chemodosimeters 4
and 8d for further experiments. The goal was to confirm that
the effects observed in cell free settings are transferrable to the
cells. We selected human promyelocytic leukemia cells HL-60
for these experiments based on their accessibility in our lab-
oratory and the possibility of convenient and quantitative
monitoring of the intracellular fluorescence of these non-
adherent cells by using flow cytometry (Fig. 3A). We observed
that 8d (100 nM) remains intact (non-fluorescent) in live
HL-60 cells for the duration of at least 20 min. The same result
was obtained for the cells loaded with both 8d (100 nM) and
photosensitizer In(pyropheophorbide-a)chloride (200 nM).
However, when the latter cells were irradiated with red light,
1
the conditions that are known to lead to the generation of O2
the activation of probes 4 and 8d (λex = 488 nm, λem
=
in cells,14,15 the fluorescence of 8d was dramatically increased
(38.6-fold in 20 min) in a time dependent fashion (black bars
in Fig. 3A).
515–540 nm). Channel FL2 was used to quantify dead cells,
which were loaded with propidium iodide (PI, λex = 488 nm,
λem = 570–606 nm). The data were processed using the
inCyte™ software package (Merck Millipore).
It is important that 8d is not toxic to HL-60 cells at con-
centrations of up to 1 μM (higher concentrations were not
tested) even at incubation times up to 48 h (Fig. 3B). Probe 4
exhibited similar effects in the flow cytometry assay (data not
shown).
Finally, we confirmed that probe 4 can also be used in com-
bination with fluorescence microscopy for the detection of 1O2
in adherent cancer cells: prostate cancer DU-145 cells we used
as an example (Fig. S26, ESI†).
Synthetic procedures
Compound 6. A suspension of fluorescein (26.6 g, 80 mmol)
in DMF (500 mL) was prepared and t-BuOK (9 g, 80 mmol) was
added. After the solution became clear, 2-bromo-2-methyl-
propionate (5.6 mL, 40 mmol) was added in one portion. The
reaction mixture was stirred at 22 °C for 48 h. After that, the
reaction mixture was diluted with water and extracted with
EtOAc (3 × 150 mL). The combined organic extracts were
washed with brine (50 mL), dried (MgSO4), and concentrated
in vacuo. Purification of the residue by flash column chromato-
graphy (silica gel, elution gradient: from 30 : 1 CH2Cl2/MeOH
Experimental
General information
Unless otherwise specified, all reactions were carried out (v/v) to 20 : 1 CH2Cl2/MeOH, both solvent mixtures contained
under an atmosphere of nitrogen using oven-dried glassware. 0.5% TEA) gave compound 6 as a brown syrup (5.0 g,
9886 | Org. Biomol. Chem., 2019, 17, 9883–9891
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
11.6 mmol, 14.5%). Rf = 0.29 (silica gel, 20 : 1 : 1 CH2Cl2/ 7.70–7.86 (m, 2H), 7.50–7.60 (m, 4H), 7.29–7.37 (m, 1H),
MeOH/TEA, v/v/v); 1H NMR (300 MHz, CDCl3) δ (ppm) = 6.59–6.87 (m, 5H), 6.55 (d, J = 2.5 Hz, 1H), 1.59 (s, 6H);
7.96–8.05 (m, 1H), 7.56–7.75 (m, 2H), 7.12–7.19 (m, 1H), 13C NMR (100.5 MHz, acetone-d6) δ (ppm) = 175.1, 169.5,
6.80–6.73 (m, 1H), 6.70–6.61 (m, 2H), 6.60–6.54 (m, 2H), 6.51 162.7, 158.8, 153.7, 153.5, 152.8, 145.6, 136.5, 133.5, 131.2,
(dd, J = 9.0, 2.4 Hz, 1H), 3.79 (s, 3H), 1.64 (s, 6H); 13C NMR 131.0, 129.9, 129.7, 127.9, 127.7, 127.0, 125.7, 125.7, 125.5,
(75.5 MHz, CDCl3) δ (ppm) = 174.6, 170.1, 158.7, 157.3, 152.6, 125.2, 122.7, 116.3, 114.4, 113.5, 113.3, 106.3, 103.7, 82.9, 80.2,
152.5, 152.2, 135.1, 129.8, 129.2, 128.9, 126.7, 125.1, 124.2, 25.6; ESI-HRMS: calcd for C38H25O7 [M − H]−: 593.1606, found
115.0, 112.7, 112.4, 110.6, 105.7, 103.1, 79.5, 52.8, 25.4, 25.2; 593.1604. C, H, N analysis: calcd for C38H26O7: C 76.8%,
electrospray ionisation (ESI) – high-resolution mass spectro- H 4.4%; found C 77.0%, H 4.2%.
metry (HRMS): calcd for C25H21O7 [M + H]+: 433.1282, found
433.1278.
9-Bromo-10-nitroanthracene. A slightly modified literature
procedure was used for the synthesis of this compound.22 To a
For the synthesis of intermediates 7a, 7b, 7d and 7e the stirred solution of 9-nitroanthracene (4.90 g, 22.0 mmol) in
corresponding halogenated anthracene derivatives R_a-Br nitromethane (100 mL), bromine (1.12 mL, 22.0 mmol) was
(9-bromoanthracene), R_b-Br (9-bromo-10-nitroanthracene), added dropwise at 22 °C. The reaction mixture was heated at
R_d-Cl (9-(chloromethoxy)anthracene) and R_e-Cl (9-(chloro- 50 °C for 2 h. The reaction mixture was quenched slowly with
methoxy)-2,6-di-(tert-butyl)anthracene) were used as reactants saturated NaHCO3 solution and the aqueous layer was
(Scheme 2). Compound R_a-Br is commercially available. separated and extracted with CH2Cl2 (2 × 50 mL). The com-
Compound R_b-Br was prepared according to a procedure bined organic layers were dried (MgSO4), filtered, and con-
reported elsewhere.22 Compounds R_d-Cl and R_e-Cl were gen- centrated in vacuo. Purification of the residue by flash
erated in situ from synthetically accessible derivatives R_d- column chromatography (silica gel, 1 : 10 CH2Cl2/petroleum
SCH3 and R_e-SCH3 by using sulfuryl chloride as outlined in ether) gave the title compound as a yellow solid (1.28 g,
Schemes S1 and S2 in the ESI.†
4.2 mmol, 19%). Rf = 0.2 (silica gel, 1 : 10 CH2Cl2/petroleum
Compound 7a. A suspension of Pd(OAc)2 (1.25 g, 5.6 mmol), ether); 1H NMR (300 MHz, CDCl3) δ (ppm) = 8.53–8.68 (m,
t-BuXPhos (2.36 g, 5.6 mmol), and Cs2CO3 (9.04 g, 27.8 mmol) 2H), 7.84–7.97 (m, 2H), 7.63–7.73 (m, 4H); 13C NMR
in o-xylene (15 mL) was stirred at 22 °C for 30 min. (75.5 MHz, CDCl3) δ (ppm) = 129.9, 129.0, 128.2, 128.0, 126.4,
9-Bromoanthracene (1.43 g, 5.5 mmol) and compound 6 122.7, 121.7. Atmospheric pressure photo-ionisation (APPI)-
(2.40 g, 5.6 mmol) in dioxane (15 mL) were added and the HRMS: calcd for C14H879BrNO2 [M]+: 300.97384, found
resulting mixture was heated at 110 °C for 18 h. The reaction 300.97339.
mixture was quenched with water and the aqueous layer was
Compound 7b. A mixture of compound 6 (4.64 g,
separated and extracted with CH2Cl2 (3 × 50 mL). The com- 10.7 mmol), 9-bromo-10-nitroanthracene (4.21 g, 14.0 mmol,
bined organic layers were dried (MgSO4), filtered, and concen- 1.3 eq.) and Cs2CO3 (5.25 g, 16.1 mmol, 1.5 eq.) in DMF
trated in vacuo. Purification of the residue by flash column (40 mL) was stirred at 75 °C for 24 h. The reaction mixture was
chromatography (alumina (grade III), eluent – CH2Cl2) gave quenched with water and extracted with CH2Cl2 (3 × 50 mL).
the title compound 7a as a yellow solid (400 mg, 0.66 mmol, The combined organic layers were dried (MgSO4), filtered, and
12%). Rf = 0.72 (alumina, eluent – CH2Cl2); 1H NMR (300 MHz, concentrated in vacuo. Purification of the residue by flash
CDCl3) δ (ppm) = 8.43 (s, 1H), 7.98–8.12 (m, 5H), 7.58–7.72 column chromatography (silica gel, eluent – CH2Cl2) gave the
(m, 2H), 7.41–7.55 (m, 4H), 7.20 (d, J = 7.3 Hz, 1H), 6.67–6.77 title compound 7b as a yellow solid (4.02 g, 6.1 mmol, 57%).
(m, 2H), 6.59–6.67 (m, 1H), 6.45–6.55 (m, 3H), 3.73 (s, 3H), Rf = 0.47 (silica gel, eluent – CH2Cl2); 1H NMR (300 MHz,
1.59 (s, 6H); 13C NMR (75.5 MHz, CDCl3) δ (ppm) = 174.2, CDCl3) δ (ppm) = 8.12 (d, J = 8.85 Hz, 2H), 7.98–8.07 (m, 3H),
169.2, 161.6, 157.2, 152.7, 152.6, 151.9, 144.7, 135.0, 132.3, 7.59–7.72 (m, 4H), 7.53 (ddd, J = 0.94, 6.64, 8.81 Hz, 2H),
129.8, 129.5, 128.8, 128.4, 126.8, 126.2, 125.8, 125.0, 124.4, 7.16–7.25 (m, 1H), 6.69–6.81 (m, 2H), 6.61–6.69 (m, 1H),
124.3, 124.1, 122.0, 115.0, 112.5, 112.3, 112.2, 105.5, 103.0, 6.43–6.58 (m, 3H), 3.74 (s, 3H), 1.59 (s, 6H); 13C NMR
82.9, 52.7, 25.4, 25.1; ESI-HRMS: calcd for C39H29O7 [M + H]+: (75.5 MHz, CDCl3) δ (ppm) = 174.1, 169.1, 160.9, 157.2, 152.6,
609.1908, found 609.1912.
152.5, 151.8, 147.3, 142.0, 135.0, 129.8, 129.2, 128.8, 127.0,
Probe 4. To a solution of intermediate 7a (110 mg, 126.7, 125.1, 124.0, 123.9, 123.6, 122.4, 121.7, 115.1, 113.3,
0.18 mmol) in THF (0.58 mL) was added a 1.9 M aqueous solu- 112.1, 112.0, 105.4, 103.1, 82.6, 79.4, 52.6, 25.3, 25.1;
tion of LiOH (0.29 mL). The mixture was degassed (3 × freeze/ ESI-HRMS: calcd for C39H28NO9 [M + H]+: 654.1759, found
pump/thaw cycles) and stirred at 22 °C overnight. Next, the 654.1761.
reaction mixture was quenched with HCl (pH = 2) and the
Probe 8b. Intermediate 7b (392 mg) was hydrolysed accord-
resulting mixture was extracted with CH2Cl2 (3 × 1 mL). The ing to the protocol used for the preparation of probe 4. Probe
organic layers were combined, dried (MgSO4), and concen- 8b was obtained as a brown solid (196 mg, 0.31 mmol, 51%).
1
trated in vacuo. The residue was recrystallized from ethyl Rf = 0.18 (alumina, 20 : 1 MeOH/TEA, v/v); H NMR (300 MHz,
acetate/n-hexane (5 mL, 2/3, v/v) to give probe 4 as white solid CDCl3) δ (ppm) = 8.11 (d, J = 8.7 Hz, 2H), 7.95–8.07 (m, 3H),
(84 mg, 0.14 mmol, 78%). Rf = 0.39 (alumina, 20 : 1 MeOH/ 7.59–7.74 (m, 4H), 7.49–7.54 (m, 2H), 7.19 (d, J = 7.2 Hz, 1H),
TEA, v/v); 1H NMR (300 MHz, acetone-d6) δ (ppm) = 8.61 (s, 6.70–6.76 (m, 2H), 6.62–6.68 (m, 2H), 6.55–6.60 (m, 1H), 6.53
1H), 8.15–8.24 (m, 2H), 8.03–8.11 (m, 2H), 7.95–8.02 (m, 1H), (d, J = 2.4 Hz, 1H), 1.60 (s, 6H); 13C NMR (75.5 MHz, CDCl3)
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9883–9891 | 9887

View Article Online
Paper
Organic & Biomolecular Chemistry
δ (ppm) = 178.1, 169.2, 161.0, 156.7, 152.6, 152.5, 151.8, 147.3, was extracted with CH2Cl2 (3 × 50 mL). The organic layers
142.0, 135.1, 129.9, 129.9, 129.3, 128.9, 127.0, 126.7, 125.1, were combined, dried (MgSO4), and concentrated in vacuo.
124.0, 123.9, 123.6, 122.4, 121.8, 116.0, 113.3, 112.8, 112.0, Purification of the residue by flash column chromatography
106.7, 103.1, 82.6, 79.4, 25.3, 25.0; ESI-HRMS: calcd for (silica gel, from 1 : 9 to 2 : 8 CH2Cl2/i-hexane, v/v) gave
C38H24NO9 [M − H]−: 638.1457, found 638.1471; C, H, N ana- the title compound as a white solid (367 mg, 1.44 mmol,
lysis: calcd for C38H25NO9: C 71.3%, H 3.9%, N 2.2%; found 7%). Rf = 0.42 (silica gel, 2 : 8 CH2Cl2/i-hexane); 1H NMR
C 71.0%, H 4.1%, N 2.0%.
(300 MHz, CDCl3) δ (ppm) = 8.35 (d, J = 8.4 Hz, 2H); 8.25 (s,
Probe 8c. SnCl2·2H2O (10.0 g, 44.4 mmol) was suspended 1H); 7.93–8.04 (m, 2H); 7.42–7.56 (m, 4H), 5.39 (s, 2H), 2.38
in acetic anhydride (30 mL) and heated to 80 °C for 30 min. (s, 3H); 13C NMR (75.5 MHz, CDCl3) δ (ppm) = 150.0, 132.3,
Next, intermediate 7b (2.9 g, 4.44 mmol) was added and the 128.3, 125.5, 125.3, 124.8, 122.8, 122.5, 80.2, 16.0;
resulting mixture was heated at 80 °C for 24 h and concen- ESI-HRMS: calcd for C16H14OSNa [M + Na]+: 277.0663, found
trated in vacuo. The resulting solid was suspended in CH2Cl2 277.0670.
(200 mL), filtered through a plug of alumina (grade III) and
Probe 8d. Sulfuryl chloride (381 μL, 637 mg, 4.7 mmol) was
concentrated in vacuo to give a brown residue containing dissolved in 50 mL of anhydrous CH2Cl2. Next the solution
intermediate 8d. We did not manage to purify this intermedi- was ice cooled and 9-(methylthiomethoxy)anthracene (1.20 g,
ate. Therefore, the mixture was directly subjected to hydro- 4.7 mmol) was added. The resulting yellow reaction solution
lysis. In particular, the brown residue was suspended in THF was stirred for 1 h at 22 °C and then the solvent was removed
(4 mL) and 1 M aqueous solution of LiOH (3 mL) was added. in vacuo. The black residue was dissolved in anhydrous DMF
The resulting mixture was degassed (3 × freeze/pump/thaw (2 mL) and added rapidly to a solution obtained by stirring
cycles) and stirred at 22 °C overnight. Next, the reaction compound 6 (2.04 g, 4.7 mmol) and t-BuOK (530 mg, 1.0 eq.)
mixture was quenched with HCl (pH = 2) and the resulting dissolved in anhydrous DMF (20 mL) for 1 h. The resulting
mixture was extracted with CH2Cl2 (3 × 20 mL). The organic solution was then stirred for 24 h at 22 °C and slowly
layers were combined, dried (MgSO4), and concentrated quenched with water (50 mL). Next 20 mL of CH2Cl2 was
in vacuo. The residue was dissolved in CH2Cl2 (5 mL) and added, the layers were separated, and the aqueous layer was
loaded into a small column filled with alumina (grade III). further extracted with CH2Cl2 (3 × 50 mL). The organic layers
For removing impurities the column was first washed with were combined, dried (MgSO4), and concentrated in vacuo. To
CH2Cl2 (100 mL) and then with methanol (100 mL). The the resulting residue, THF (4 mL) was added followed by the
resulting fractions were disposed. Finally the column was addition of 1 M aqueous solution of LiOH (3 mL). The mixture
washed with 20 : 1 MeOH/TEA (v/v) and the respective frac- was degassed (3 × freeze/pump/thaw cycles) and stirred at
tions were combined and concentrated in vacuo. The result- 22 °C overnight. Next, the reaction mixture was quenched with
ing white solid was dissolved in CH2Cl2 (100 mL) and washed HCl (pH = 2) and the resulting mixture was extracted with
with 0.1 M aq. HCl (15 mL) and water (3 × 20 mL). The CH2Cl2 (3 × 20 mL). The organic layers were combined, dried
CH2Cl2 layer was dried (MgSO4), and concentrated in vacuo to (MgSO4), and concentrated in vacuo. The residue was dissolved
give probe 8c as a brown solid (410 mg, 0.63 mmol, 14% over in CH2Cl2 (5 mL) and loaded into a small column filled with
2 steps). Rf = 0.52 (alumina, 20 : 1 MeOH/TEA, v/v); 1H NMR alumina grade III. For removing impurities, the column was
(300 MHz, acetone-d6) δ (ppm) = 9.56 (s, 1H), 8.23–8.26 (m, first washed with CH2Cl2 (100 mL) and then with methanol
2H), 8.02–8.08 (m, 2H), 7.95–8.02 (m, 1H), 7.67–7.82 (m, 2H), (100 mL). The resulting fractions were disposed. Finally, the
7.46–7.60 (m, 4H), 7.26 (d, J = 7.35 Hz, 1H), 6.78–6.88 (m, column was washed with 20 : 1 MeOH/TEA (v/v) and the
2H), 6.57–6.73 (m, 3H), 6.48 (m, 1H), 2.42 (s, 3H), 1.55 (s, respective fractions were combined and concentrated in vacuo.
6H); 13C NMR (75.5 MHz, acetone-d6) δ (ppm) = 175.1, 170.8, The resulting white solid was dissolved in CH2Cl2 (100 mL)
170.7, 169.3, 162.5, 158.7, 153.5, 153.3, 152.6, 144.8, 136.3, and washed with 0.1 M aq HCl (15 mL) and water (3 × 20 mL).
131.0, 130.9, 130.3, 129.6, 129.2, 129.0, 127.7, 127.4, 127.2, The CH2Cl2 layer was dried (MgSO4), and concentrated in
125.6, 125.5, 125.3, 125.0, 122.8, 116.2, 114.3, 113.2, 106.1, vacuo to give the title compound as a brown solid (63 mg,
103.6, 82.9, 80.2, 25.7, 25.6, 23.3; ESI-HRMS: calcd for 0.10 mmol, 2% over 2 steps). Rf = 0.12 (alumina, 20 : 1 MeOH/
1
C
40H28NO8 [M − H]−: 650.1820, found 650.1827; C, H, N ana- TEA, v/v); H NMR (300 MHz, CDCl3) δ (ppm) = 8.29 (s, 1H);
lysis: calcd for C40H29NO8: C 73.7%, H 4.5%, N 2.2%; found 8.12–8.21 (m, 2H), 7.96–8.09 (m, 3H), 7.60–7.75 (m, 2H),
C 73.4%, H 4.5%, N 2.3%. 7.38–7.52 (m, 4H), 7.20–7.26 (m, 2H), 6.86–6.91 (m, 2H),
9-(Methylthiomethoxy)anthracene (compound R_d-SCH3). 6.77–6.83 (m, 1H), 6.70–6.77 (m, 1H), 6.65 (dd, J = 2.4, 9.0 Hz,
To a solution of anthrone (4.00 g, 20.6 mmol, 1 equiv.) in 1H), 5.93 (s, 2H), 1.68 (s, 6H); 13C NMR (75.5 MHz, CDCl3)
anhydrous DMSO (100 mL) sodium hydride (60% dispersion δ (ppm) = 177.6, 169.4, 159.1, 156.7, 152.9, 152.4, 152.0, 149.6,
in mineral oil) (907 mg, 22.7 mmol, 1.1 eq.) was added in 135.1, 132.1, 129.8, 129.4, 128.9, 128.2, 126.7, 125.7, 125.5,
small portions. After stirring for 30 min at 22 °C, 125.1, 124.6, 124.0, 123.4, 122.3, 115.8, 113.0, 112.9, 112.8,
CH3SCH2Cl (1.72 mL, 1.99 g, 20.6 mmol) was added drop- 106.9, 103.6, 96.4, 82.9, 79.6, 25.3, 25.1; ESI-HRMS: calcd for
wise resulting in a deep red solution, which was then stirred C39H27O8 [M − H]−: 623.1711, found 623.1720; C, H, N ana-
for 24 h at 22 °C. The reaction mixture was then slowly lysis: calcd for C39H28O8: C 75.0%, H 4.5%; found C 75.4%,
quenched with water (50 mL). The resulting aqueous mixture H 4.4%.
9888 | Org. Biomol. Chem., 2019, 17, 9883–9891
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
1
2,6-Di-tert-butyl-9-(methylthiomethoxy)anthracene. To
a
v/v); H NMR (300 MHz, acetone-d6) δ (ppm) = 8.36–8.45 (m,
solution of 2,6-di-tert-butylanthrone23 (11.7 g, 38.1 mmol) in 2H), 8.29 (d, J = 9.2 Hz, 1H), 8.15 (d, J = 1.7 Hz, 1H), 8.01–8.06
anhydrous DMSO (300 mL) sodium hydride (60% dispersion (m, 1H), 7.84–7.93 (m, 1H), 7.72–7.84 (m, 3H), 7.27–7.41 (m,
in mineral oil) (1.68 g, 41.9 mmol, 1.1 eq.) was added in small 2H), 7.00 (dd, J = 2.5, 8.8 Hz, 1H), 6.90 (d, J = 8.67 Hz, 1H),
portions. After stirring for 30 min at 22 °C, CH3SCH2Cl 6.76–6.86 (m, 2H), 6.67–6.76 (m, 1H), 6.10–6.23 (m, 2H), 1.65
(3.19 mL, 3.68 g, 38.1 mmol) was added dropwise to obtain a (s, 6H), 1.48 (s, 9H), 1.22 (s, 9H); 13C NMR (75.5 MHz, acetone-
deep red solution, which was stirred for 24 h at 22 °C. The red d6) δ (ppm) = 169.5, 160.1, 158.8, 153.9, 153.6, 152.9, 150.8,
reaction solution was then slowly quenched with water 150.0, 149.3, 136.3, 131.1, 130.7, 2 × 129.9, 128.9, 128.0, 127.8,
(150 mL), the layers were separated, and the aqueous layer was 126.7, 126.5, 125.7, 2 × 125.1, 124.5, 124.4, 123.6, 119.8, 118.4,
further extracted with CH2Cl2 (3 × 75 mL). The organic layers 116.5, 114.6, 113.9, 113.7, 106.5, 104.3, 97.4, 83.0, 80.4, 36.1,
were combined, dried (MgSO4), and concentrated in vacuo. 35.7, 3 × 31.2, 3 × 31.1, 25.8, 25.7. The signal for carboxyl
Purification of the residue by flash column chromatography carbon was not detected in the spectrum; ESI-HRMS: calcd for
(silica gel, from 1 : 9 to 2 : 8 CH2Cl2/i-hexane, v/v) gave the title C47H42ClO8 [M − H]−: 769.2574, found 769.2591; C, H, N ana-
compound as a white solid (730 mg, 2.0 mmol, 5%). Rf = 0.26 lysis: calcd for C47H43ClO8: C 73.2%, H 5.6%; found C 73.0%,
(silica gel, 1 : 4 CH2Cl2/i-hexane, v/v); 1H NMR (300 MHz, H 5.3%.
CDCl3) δ (ppm) = 8.25–8.34 (m, 2H); 8.20 (s, 1H); 7.87–7.99 (m,
2H); 7.54–7.67 (m, 2H), 5.42 (s, 2H); 2.47 (s, 3H); 1.50 (s, 9H),
Monitoring the fluorogenic response of 1O2 probes in cell free
settings in the presence of different reactive oxygen species
1.49 (s, 9H); 13C NMR (75.5 MHz, CDCl3) δ (ppm) = 149.8;
(2 × 147.4), 132.2, 131.1, 128.0, 124.9, 124.8, 124.4, 123.3,
(typical procedures)
122.6, 122.2, 122.1, 116.7, 79.9, 35.2, 34.9, 31.0, 30.9, 16.0;
ESI-HRMS: calcd for C24H30NaOS [M + Na]+: 389 1910, found
389.1903.
“Singlet oxygen” test. The solutions were pipetted into a
cuvette in the following order: water (867 μL), 1 M phosphate
Compound 7e. To an ice cold solution of 2,6-di-tert-butyl-9- buffer (100 μL, pH 7.4), 5 M NaCl (30 μL), 1.6 mM probe 8d in
(methylthiomethoxy)anthracene (655 mg, 1.8 mmol) in an- DMF (1.3 μL) and 487 µM In3+ (pyropheophorbide-a)chloride
hydrous CH2Cl2 (25 mL) was added sulfuryl chloride (145 μL, (InPPa) in DMF (1 μL). Final concentrations were as follows:
243 mg, 1.8 mmol). The resulting yellow solution was stirred 100 mM phosphate buffer; 150 mM NaCl; 2 µM 8d; 0.5 µM
for 1 h at 22 °C and then the solvent was removed in vacuo. InPpa; 0.3% DMF (v/v). The cuvette was closed by using a
The black residue was dissolved in anhydrous DMF (3 mL) and Teflon® cap and the fluorescence of the solution was
added rapidly to a solution obtained from stirring for 1 h com- measured (λex = 495 nm, λem = 515 nm; F0). The cuvette was
pound 6 (778 mg, 1.8 mmol) and t-BuOK (202 mg, 1.8 mmol) irradiated for 60 min with LED Array 672 (λmax = 650 nm) in a
in anhydrous DMF (7 mL). The resulting solution was then paper cardboard box covered with aluminum foil to induce
stirred for 24 h at 22 °C and slowly quenched with water 1O2 production catalysed by InPpa. Next, the cuvette was
(50 mL). Next, 20 mL of CH2Cl2 was added, the layers were sep- removed from the cardboard box and the fluorescence of the
arated, and the aqueous layer was further extracted with solution was measured (λex = 495 nm, λem = 515 nm; F60). The
CH2Cl2 (3 × 50 mL). The organic layers were combined, dried same experiment was repeated in the absence of the photo-
(MgSO4), and concentrated in vacuo. Purification of the residue catalyst (InPpa) to evaluate the possibility of direct probe acti-
by flash column chromatography (alumina, from 9 : 1 to 8 : 2 vation by red light.
i-hexane/ethyl acetate, v/v) gave the title compound as a brown
“Hydrogen peroxide” test. The solution of probe 8d (2 µM)
oil (143 mg, 0.2 mmol, 11%). Rf = 0.38 (8 : 2 i-hexane/ethyl in phosphate buffer (100 mM) containing NaCl (150 mM) and
acetate, v/v); 1H NMR (300 MHz, CDCl3) δ (ppm) = 8.38–8.45 DMF (0.3%, v/v) was prepared as described above (“singlet
(m, 2H), 8.14 (d, J = 9.23 Hz, 1H), 8.01–8.10 (m, 2H), 7.59–7.78 oxygen” test). H2O2 (final concentration: 10 mM) was added
(m, 4H), 7.31 (d, J = 2.45 Hz, 1H), 7.19–7.26 (m, 1H), 6.86–6.95 and the fluorescence of the resulting solution was measured
(m, 1H), 6.79–6.82 (m, 1H), 6.67–6.76 (m, 2H), 6.51–6.61 (m, (λex = 495 nm, λem = 515 nm) immediately after H2O2 addition
1H), 5.95 (s, 2H), 3.81 (s, 3H), 1.66 (s, 6H), 1.49 (s, 9H), 1.24 (s, (F0) and after 60 min incubation at 22 °C (F60).
9H); 13C NMR (100.6 MHz, CDCl3) δ (ppm) = 174.3, 169.3,
“Superoxide anion radicals” test. This experiment was con-
159.1, 157.3, 152.9, 152.6, 152.1, 149.2, 148.6, 148.2, 135.0, ducted similar to the “hydrogen peroxide” test, except that a
133.4, 131.4, 131.2, 129.8, 129.5, 128.9, 127.9, 127.3, 126.8, mixture of xanthine (33 mM) and xanthine oxidase (4 units per
126.5, 125.6, 125.3, 125.1, 124.5, 124.5, 124.0, 123.9, 123.4, L) was added in place of H2O2. The preparation of a stock solu-
122.1, 119.4, 117.0, 115.0, 113.1, 112.8, 112.4, 105.8, 103.5, tion of xanthine oxidase (100 units per L): xanthine oxidase
96.4, 82.8, 79.5, 52.7, 35.3, 35.0, 30.9, 30.6, 25.4, 25.3; (lyophilised powder, 3.57 mg, 5 units) was dissolved in 0.1 M
APPI-HRMS: calcd for C48H45ClO8 [M]+: 784.2803, found sodium phosphate buffer (1000 μL) at pH 7.4 which contained
784.28162.
1% DMF (v/v).
Probe 8e. Probe 8e was obtained by the hydrolysis of inter-
“Hydroxyl radicals” test. This experiment was conducted
mediate 7e (143 mg, 0.18 mmol) analogously to the prepa- similar to the “hydrogen peroxide” test, except that FeSO4
ration of probe 4. The title compound was obtained as a brown (0.6 mM) was also added to convert H2O2 to HO•. The same
solid (21 mg, 0.03 mmol, 15%). Rf = 0.13 (20 : 1 MeOH/TEA, experiment was repeated in the absence of H2O2 to evaluate
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9883–9891 | 9889

View Article Online
Paper
Organic & Biomolecular Chemistry
the direct effect of Fe2+ ions on the activation of the probes. cytic leukemia HL-60 cells at concentrations below 1 μM
Furthermore, Fe2+ ions in the latter experiment were replaced (higher concentrations were not tested) even at long incu-
1
with the same amounts of other bioavailable metal ions bation times (≤48 h) and retain its ability to detect O2 in live
including K+, Mg2+, Ca2+, Cu2+ and Ni2+ to evaluate their HL-60 cells.
1
possible effects on the O2 probes. We did not observe fluoro-
genic response of the new chemodosimeters in any of these
cases.
Conflicts of interest
Cellular assays
There are no conflicts to declare.
The human promyelocytic leukemia cell line (HL-60) was cul-
tured in Roswell Park Memorial Institute (RPMI) 1640 medium
supplemented with 10% FCS and 5 μg mL−1 penicillin/strepto-
mycin (all media and supplements from Gibco Invitrogen
Corp., Karlsruhe, Germany).
Acknowledgements
This work was supported by German Research Council (DFG),
grant number MO1418/8-1.
1
Monitoring the formation of O2 in live HL-60 cells by flow
cytometry. The cells, grown as described above, were centri-
fuged, and the medium was replaced with RPMI 1640 medium
(5% FCS, 1% L-glutamine, 1% penicillin/streptomycin) to
obtain suspensions containing 106 cells per mL. Stock solu-
tions (1 µL) of probe 8d (5 and 50 µM in DMF) and InPpa
(100 µM) were added to the cellular suspensions (500 µL).
After incubation for 20 min, washing of the cells with the
medium and their resuspension in fresh medium, the cells
were irradiated with red light (LED Array 672, λmax = 650 nm;
time = 0, 2, 5, 10, 15 and 20 min) to induce the production of
1O2. Intracellular activation of the probe was monitored by
using flow cytometry: λex = 488 nm, λem = 515–540 nm (FL1
channel).
Notes and references
1 P. Di Mascio, G. R. Martinez, S. Miyamoto, G. E. Ronsein,
M. H. G. Medeiros and J. Cadet, Chem. Rev., 2019, 119,
2043.
2 C. Kiryu, M. Makiuchi, J. Miyazaki, T. Fujinaga and
K. Kakinuma, FEBS Lett., 1999, 443, 154.
3 C. Triantaphylides and M. Havaux, Trends Plant Sci., 2009,
14, 219.
4 C. C. Winterbourn, M. B. Hampton, J. H. Livesey and
A. J. Kettle, J. Biol. Chem., 2006, 281, 39860.
5 D. E. Dolmans, D. Fukumura and R. K. Jain, Nat. Rev.
Cancer, 2003, 3, 380.
6 E. Hideg, Cent. Eur. J. Biol., 2008, 3, 273.
7 N. Hananya, O. Green, R. Blau, R. Satchi-Fainaro and
D. Shabat, Angew. Chem., Int. Ed., 2017, 56, 11793.
8 S. Kim, M. Fujitsuka and T. Majima, J. Phys. Chem. B, 2013,
117, 13985.
9 D. Liang, Y. Zhang, Z. Wu, Y. J. Chen, X. Yang, M. Sun,
R. Ni, J. Bian and D. Huang, Sens. Actuators, B, 2018, 266,
645.
10 S. K. Pedersen, J. Holmehave, F. H. Blaikie, A. Gollmer,
T. Breitenbach, H. H. Jensen and P. R. Ogilby, J. Org.
Chem., 2014, 79, 3079.
11 S. Kim, T. Tachikawa, M. Fujitsuka and T. Majima, J. Am.
Chem. Soc., 2014, 136, 11707.
Determination of cytotoxicity of 1O2 probes towards HL-60
cells. The cells, grown as described above, were centrifuged,
and the medium was replaced with RPMI 1640 medium (5%
FCS, 1% L-glutamine, 1% penicillin/streptomycin) to obtain
suspensions containing 106 cells per mL. Stock solutions
(5 µL) of compound 8d (0.1, 1, 10 and 100 µM in DMF) were
added to the cellular suspensions (500 µL) and incubated for
48 h. At the end of the incubation period propidium iodide
(PI, 1 µL, 1.5 mM in water) was added to label dead cells and
in 5 min the suspension of the cells was analysed by using
flow cytometry: λex = 488 nm, λem = 570–606 nm (FL2 channel).
Viable cells remain PI-negative: no signal detected in the FL2
channel. Four replicates were tested for each concentration of
the prodrug. Three independent experiments were conducted
to confirm the effect of the probes on the viability of HL-60
cells.
12 D. Arian, L. Kovbasyuk and A. Mokhir, J. Am. Chem. Soc.,
2011, 133, 3972.
13 M. Sun, S. Krishnakumar, Y. Zhang, D. Liang, X. Yang,
M. W. Wong, S. Wang and D. Huang, Sens. Actuators, B,
2018, 271, 346.
Conclusions
We developed the synthesis of a series of fluorescein deriva- 14 D. Arian, E. Clo, K. V. Gothelf and A. Mokhir, Chem. – Eur.
tives coupled to 9-anthracenyl fragments and have shown that J., 2010, 16, 288.
some of these compounds (4, 8d, 8e) exhibit excellent fluoro- 15 D. Arian, L. Kovbasyuk and A. Mokhir, J. Am. Chem. Soc.,
genic response in the presence of 1O2 in cell-free settings
2011, 133, 3972.
exceeding in the best case 20-fold. We found that these new 16 A. Meyer and A. Mokhir, Angew. Chem., Int. Ed., 2014, 53,
chemodosimeters are 1O2 specific (as compared to red light
irradiation, H2O2, O2
12840.
and HO•). Finally, the selected 17 A. Meyer, M. Schikora and A. Mokhir, Chem. Commun.,
•−
compound 8d was found to be non-toxic to human promyelo-
2015, 51, 13324.
9890 | Org. Biomol. Chem., 2019, 17, 9883–9891
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
18 A. Meyer, M. Schikora, V. Starkuvienne and A. Mokhir, 21 O. Zozulia, T. Bachmann, N. S. Deussner-Helfmann,
Photochem. Photobiol. Sci., 2016, 15, 1120.
19 H. Pieper, S. Chercheja, S. Eigler, C. Halbig,
F. Beierlein, M. Heilemann and A. Mokhir, Chem.
Commun., 2019, 55, 10713.
M. R. Filipovic and A. Mokhir, Angew. Chem., Int. Ed., 2016, 22 J. Rigaudy, A. M. Seuleiman and N. K. Cuong, Tetrahedron,
55, 405. 1982, 38, 3157.
20 O. Zozulia, T. Bachmann and A. Mokhir, Bioconjugate 23 U. Müller, V. Enkelmann, M. Adam and K. Müllen, Chem.
Chem., 2019, 30, 2023.
Ber., 1993, 126, 1217.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9883–9891 | 9891