Trimethine Cyanine Dyes as NA-Sensitive Probes for Visualization of
Cell Compartments in Fluorescence Microscopy
Daria Aristova, Roman Selin, Hannah Sophie Heil, Viktoriia Kosach, Yuriy Slominsky, Sergiy Yarmoluk,
Vasyl Pekhnyo, Vladyslava Kovalska, Ricardo Henriques, Andriy Mokhir, and Svitlana Chernii*
Cite This: ACS Omega 2022, 7, 47734−47746
Read Online
ACCESS
Metrics & More
Article Recommendations
*
sı
Supporting Information
ABSTRACT: We propose symmetrical cationic trimethine cyanine
dyes with β-substituents in the polymethine chain based on modified
benzothiazole and benzoxazole heterocycles as probes for the
detection and visualization of live and fixed cells by fluorescence
microscopy. The spectral-luminescent properties of trimethine
cyanines have been characterized for free dyes and in the presence
of nucleic acids (NA) and globular proteins. The studied cyanines
are low to moderate fluorescent when free, but in the presence of
NA, they show an increase in emission intensity up to 111 times; the
most pronounced emission increase was observed for the dyes T-2 in
the presence of dsDNA and T-1 with RNA. Spectral methods
showed the binding of all dyes to nucleic acids, and different
interaction mechanisms have been proposed. The ability to visualize
cell components of the studied dyes has been evaluated using
different human cell lines (MCF-7, A2780, HeLa, and Hs27). We have shown that all dyes are cell-permeant staining nucleus
components, probably RNA-rich nucleoli with background fluorescence in the cytoplasm, except for the dye T-5. The dye T-5
selectively stains some structures in the cytoplasm of MCF-7 and A2780 cells associated with mitochondria or lysosomes. This effect
has also been confirmed for the normal type of cell line−human foreskin fibroblasts (Hs27). The costaining of dye T-5 with
MitoTracker CMXRos Red demonstrates specificity to mitochondria at a concentration of 0.1 μM. Colocalization analysis has
shown signals overlapping of dye T-5 and MitoTracker CMXRos Red (Pearson’s Coefficient value = 0.92 ± 0.04). The
photostability study shows benzoxazole dyes to be up to ∼7 times more photostable than benzothiazole ones. Moreover, studied
benzoxazoles are less cytotoxic at working concentrations than benzothiazoles (67% of cell viability for T-4, T-5 compared to 12%
for T-1, and ∼30% for T-2, T-3 after 24 h). Therefore, the benzoxazole T-4 dye is proposed for nucleic acid detection in vitro and
intracellular fluorescence imaging of live and fixed cells. In contrast, the benzoxazole dye T-5 is proposed as a good alternative to
commercial dyes for mitochondria staining in the green-yellow region of the spectrum.
■ INTRODUCTION
Fluorescence microscopy has become a prevalent research
method in modern cell biology since it gives incomparable
opportunities to understand cellular processes. Fluorescent
protein tags, organic dyes, carbon dotes, and other ways to
fluorescently label biomolecules, compartments, and organelles
provide a range of tools to investigate virtually any cellular
process under the microscope.1,2 For instance, it was reported
that fluorescent carbon dots could stain nuclei in live and fixed
cells. Besides, fluorescent carbon dots are widely applicable in
imaging, sensing, drug delivery, and nucleolus-related biological
behaviors.3,4 Although carbon dots possess favorable spectral-
luminescent characteristics, high photostability, and low
cytotoxicity, they still suffer from low biomolecule selectivity
and difficulties tuning biomolecule specificity. Most of them do
not provide higher sensitivity and selectivity compared to other
probes, mainly because of their low affinity and low enhancing/
quenching efficiency induced by the analyte of interest.5
Meanwhile, organic dyes are appealing stringing agents for
live-cell visualization because of their simultaneously favorable
optical properties, such as brightness and photostability, and
high specificity toward biomolecules. In addition, they can be
specifically designed to be membrane-permeable or membrane-
impermeable, depending on what we want to observe�
processes inside cells or on the surface.6
One of the most widespread classes of fluorescent organic
dyes is cyanines. The history of cyanine dyes began more than
Received:
August 15, 2022
Accepted:
November 24, 2022
Published: December 13, 2022
Article
http://pubs.acs.org/journal/acsodf
© 2022 The Authors. Published by
American Chemical Society
47734
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
Downloaded via 89.115.116.29 on January 5, 2024 at 11:35:45 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

----!@#$NewPage!@#$----
150 years ago when Williams reported the first synthesis of a
blue solid in 1856.7 Since then, cyanine dyes have been widely
applied in many areas, especially fluorescent imaging, due to
their good photophysical properties, high quantum yield values,
excellent biocompatibility, and low toxicity to biological
systems.8,9 Monomethine and polymethine cyanine dyes are
widely used for nucleic acid imaging. Their spectral properties
are more advanced than those of other noncyanine dyes, such as
Ethidium Bromide, DAPI, and Hoechst.10 Moreover, they can
be both membrane-permeant and membrane-impermeant. For
example, TOTO or TO-PRO dyes for NA-staining in the gel or
SYTOX dyes for dead-cell staining cannot penetrate inside a live
cell. For instance, SYTOX Green nucleic acid stain quickly
penetrates cells with compromised plasma membranes. This dye
is a simple and quantitative dead-cell indicator successfully used
to detect mammalian cells and bacteria viability.11−13
On the other hand, cell-permeant cyanine dyes, such as SYTO
dyes, can be used for NA-staining in live cells. Nevertheless,
these dyes do not have selectivity and can stain DNA and RNA;
furthermore, they act as nuclear stains in live cells and can also
stain other compartments, such as mitochondria.14 For
fluorescent probes, the low fluorescence quantum yield in the
free state and the high quantum yield in the bound state, that is, a
sharp increase in fluorescence intensity upon binding to a
biomolecule, are essential.15 These fluorescent dyes differ from
fluorescent labels such as cyanines of the Cy3-Cy5 family dyes,
which work as fluorescent labels attached chemically to the
target molecule.16 Dyes from the Cy family have high intrinsic
fluorescence, quantum yield, and extinction coefficient and
cannot be used as fluorescently sensitive probes for noncovalent
binding and visualization of cell components. Thus, fluorescent
probes should have a wide range of opposite properties, which is
difficult to achieve in one molecule.15
Another example is SYBR family dyes, e.g., SYBR Green I,
which intensely stained DNA in live cells but has the
disadvantage that the fluorescence fades rapidly, and observa-
tion must be done as quickly as possible.17 Moreover, it has been
reported that monomethine cyanine dye Cyan 40 can be used
for the two-photon fluorescent visualization of KB cells (oral
epidermoid carcinoma).18 We have also shown that this dye in
low concentration could stain RNA-rich nucleoli without a
background in the cytoplasm and have reported highly specific
green-emitted monomethine cyanine dyes to stain RNA-
containing structures in cells.19 Besides this, our research
group proposed near-infrared fluorescent probes among
pentamethine cyanines and merocyanines as powerful far-red
fluorescent probes applicable for the highly sensitive detection
of albumins.20,21
Here we present our research of symmetrical cationic
trimethine cyanine dyes with β-substituents in the polymethine
chain based on substituted benzothiazole and benzoxazole
heterocycles (ethyl group at the 3 positions and ethoxy group at
the 6 positions for dye T-1, methyl group at the 3 positions for
dyes T-2−T-5 with additional chloro-group at the 5 positions
for dye T-5) as probes for nucleic acids detection and
visualization by fluorescence microscopy (Figure 1).
For the symmetrical β-methyl-substituted benzoxazole dye T-
4, also known as Cyan 2-O, fluorescent sensitivity in complex
with dsDNA was shown to be higher than the corresponding
values for its benzothiazole analog Cyan 2.22 It was also
previously shown the fluorescent sensitivity to nucleic acids for
trimethine cyanine dyes T-2 (Cyan βPh) and T-3 (Cyan βPr),23
but their organelle specificity has not yet been investigated. For
effective primary binding to DNA, cationic cyanine dyes are
usually used.24 It was previously shown that introducing
substituents into β-position in the polymethine chain leads to
a decrease in the fluorescence quantum yields of free dyes, with
fluorescence response growing stronger with DNA presence as
compared to carbocyanine dyes.25
Thus, the spectral-luminescent properties of trimethine
cyanines have been studied in a free state in an aqueous buffer
and in the presence of double-stranded DNA (dsDNA), RNA,
human serum albumin (HSA), and beta-lactoglobulin (BLG).
Furthermore, the ability of the studied dyes to penetrate the cell
and stain its components has been investigated using human
cancer MCF-7, A2780, HeLa, and normal Hs27 cell lines as
models. First, different dye concentrations in live and fixed
MCF-7 cells were used to determine the working conditions of
each dye. Next, the ability to stain cells was confirmed for other
cancer cell lines and normal Hs27 cells. According to
Figure 1. Chemical structure of studied symmetrical trimethine cyanine dyes T-1−T-5.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47735

----!@#$NewPage!@#$----
photostability evaluation, benzoxazole cyanine dyes (T-4, T-5)
are more photostable than benzothiazoles (T-1, T-2, and T-3).
We have tested dyes for cytotoxicity toward the A2780 cell line
using standard MTT assay. Benzoxazoles are also shown lower
cytotoxicity. Moreover, the dye T-5 has a lower concentration
selectivity toward cell components in the cytoplasm. Thus, the
T-5 dye was chosen for further studies. For this dye, a human
foreskin fibroblast cell line (Hs27) was used to check whether
the dye T-5 at working concentration has the specificity staining
pattern in cells. Finally, the mitochondria staining by
benzoxazole T-5 dye was confirmed by colocalization analysis
with MitoTracker CMXRos Red.
■ MATERIAL AND METHODS
Materials. Dimethyl sulfoxide (DMSO), methanol
(MeOH), distilled water, and 0.05 M Tris-HCl buffer (pH
7.9) were used as solvents. In addition, high molecular weight
dsDNA (double-stranded DNA) from Salmon testis, RNA,
human serum albumin (HSA), and beta-lactoglobulin (BLG)
were purchased from Sigma-Aldrich Co. MitoTracker CMXRos
Red, LysoTracker Deep Red, Hoechst 33342, and Thiazolyl
Blue tetrazolium bromide (MTT) were purchased from
ThermoFisher Scientific (Invitrogen).
The Synthesis of Symmetrical Trimethine Cyanine
Dyes. 6-Ethoxy-2-((1Z,3Z)-3-(6-ethoxy-3-ethylbenzo[d]-
thiazol-2(3H)-ylidene)-2-methoxyprop-1-en-1-yl)-3-
ethylbenzo[d]thiazol-3-ium iodide (T-1). Dye T-1 was
synthesized by condensation of 6-ethoxy-3-ethyl-2-methyl-1,3-
benzothiazolium iodide with triethyl orthoformate.26
1H NMR (400 MHz, dmso) δ 7.73 (d, J = 9.1 Hz, 2H), 7.69 (s,
2H), 7.20 (dd, J = 8.9, 2.6 Hz, 2H), 6.45 (s, 2H), 4.46 (q, J = 6.9,
6.9, 6.8 Hz, 4H), 4.10 (q, J = 7.0, 6.9, 6.9 Hz, 4H), 2.57 (s, 3H),
1.37 (m, 12H). ES-API MS (positive range) m/z: Calcd for
[C25H28N2O3S2+H]+: 469.2. Found: 469.0
3-Methyl-2-[3-(3-methyl-2,3-dihydro-1,3-benzothiazol-2-
ylidene)-2-phenyl-1-propenyl]-trimethinium iodide (T-2).
Dye T-2 also known as Cyan βPh23 was synthesized as
described27,28 using iodomethylates of the corresponding salts
of alkylthio derivatives of 3-methyl-benzothiazole.
3-Methyl-2-[2-(3-methyl-2,3-dihydro-1,3-benzothiazol-2-
ylidenemethyl)-1-pentenyl]-trimethinium iodide (T-3). Dye
T-3 also known as Cyan βPr23 was synthesized as described27,28
using iodomethylates of the corresponding salts of alkylthio
derivatives of 3-methyl-benzothiazole.
3-Methyl-2-[2-methyl-3-(3-methyl-2,3-dihydro-1,3-ben-
zooxazol-2-ylidene)-1-propenyl)-1,3-benzooxazol]-trime-
thine iodide (T-4). B-methyl-substituted benzoxazole dye T-4
also known as Cyan 2-O(J6)22,29 was synthesized as described in
refs 22 and 23 using the orthoester method.
5-Chloro-2-((E)-2-((Z)-(5-chloro-3-methylbenzo[d]thiazol-
2(3H)-ylidene)methyl)but-1-en-1-yl)-3-methylbenzo[d]-
thiazol-3-ium perchlorate (T-5). Dye T-5 was synthesized by
condensation of 5-cloro-2,3-dimethylbezoxazole ethyl sulfate
with triethyl orthoformate by the procedure described in ref 30.
1H NMR (400 MHz, dmso) δ 7.91 (m, 2H), 7.83 (m, 2H),
7.43 (d, J = 8.9 Hz, 2H), 5.83 (m, 2H), 3.74 (s, 6H), 2.50 (m, 2H
+DMSO), 1.31 (t, J = 7.4, 7.4 Hz, 3H). ES-API MS (positive
range) m/z: Calcd for [M + H]+: 434.0. Found: 434.0.
The structure of the compounds was confirmed by NMR H1
spectroscopy and element analysis.
The Preparation of Stock/Working Solutions of Dyes,
Nucleic Acids, and Proteins. Dye stock solutions were
prepared by dissolving the dyes at 2 mM concentration in
DMSO. Working solutions of free dyes were prepared by
diluting the dye stock solution in 0.05 M Tris-HCl buffer or
MeOH. The dye working concentrations amounted to 5 μM.
Working solutions of the dyes in the presence of nucleic acids
were prepared by adding the aliquot of the dye stock solution to
the nucleic acid working solution. Nucleic acid stock solutions
were equal to 6 × 10−3 M for dsDNA from Salmon testis and 1.2
× 10−2 M for RNA. Working solutions were prepared by diluting
the stock solution 100 times to final dsDNA and RNA
concentrations of 6 × 10−5 M bp and 1.2 × 10−4 M bases,
respectively. Protein stock solutions (HSA, BLG) were prepared
by dissolving in 0.05 M Tris-HCl buffer pH 7.9 at a
concentration equal to 0.2 mg/mL. Working solutions were
prepared by adding the dye aliquots to the protein stock
solutions. All working solutions were prepared immediately
before the experiments.
Spectroscopic Measurements. Spectroscopic measure-
ments were performed in a standard quartz cuvette (10 × 10
mm). Fluorescence excitation and emission spectra were
recorded using the Cary Eclipse fluorescence spectrophotom-
eter (Varian, Australia). Absorption spectra were registered
using the spectrophotometer Shimadzu UV-3600. All the
spectral-luminescent characteristics of dyes were studied at
room temperature.
Cell Culture Cultivation. The human breast adenocarci-
noma cell line (MCF-7) was obtained from the Bank of Cell
Lines of the R. E. Kavetsky Institute of Experimental Pathology,
Oncology and Radiobiology, NASU (Ukraine). Human foreskin
fibroblasts HFF Hs27 (ATCC CRL-1634) were gifted by
Jonathan Howard from the Instituto Gulbenkian de Ciência
(Portugal). The human ovarian cell line (A2780) was purchased
from Sigma-Aldrich. HeLa(CCL-2) cell line was obtained from
ATCC.
The MCF-7 and HHF Hs27 cells were cultivated in DMEM
culture medium (Gibco, USA) supplemented with 10% fetal
bovine serum (FBS), 4 mM L-glutamine, 50 units/mL penicillin,
and 50 μg/mL streptomycin at 37 °C in 5% CO2. In addition,
A2780 cells were cultivated in RPMI-1640 medium supple-
mented with 10% FBS, 1% L-glutamine, and 1% penicillin/
streptomycin at 37 °C in 5% CO2 to 80−90% confluence. Cells
were detached from the flask by using trypsin/EDTA solution
(0.025%/0.01%, w/v, Biochrom GmbH, Germany) in DPBS.
MCF-7 cells were seeded on sterile glass coverslips in 24-well
plates 48 h before further experiments at the concentration of 2
× 104 cells per well in 1 mL of DMEM medium containing 10%
FBS, 4 mM glutamine, 50 units/mL penicillin, and 50 μg/mL
streptomycin. HeLa and Hs27 cells were seeded in an 8-well-
chambered cover glass with #1.5 high-performance cover glass
(Cellvis) 24 h before the experiment at the concentration of 2 ×
104 cells per well in 0.4 mL of DMEM medium containing 10%
FBS, 4 mM glutamine, 50 units/mL penicillin, and 50 μg/mL
streptomycin. The samples were stored at 37 °C in 5% CO2
overnight. The A2780 cells were seeded at 80 cells/μL per glass-
bottom fluorescent dish (μ-Dish 35 mm, high, ibidi GmbH,
Germany) in RPMI-1640 medium with 5% FBS. Cell number
counting was performed using a hemocytometer or via Guava
easyCyteTM 6−2L Flow cytometer (Merck Millipore, Darm-
stadt, Germany).
MCF-7 Live-Cell and Fixed Cell Fluorescence Micros-
copy. For live-cell imaging of MCF-7, cells were washed twice
with preheated PBS. After that, the cells were incubated with
cyanine dyes in FluoroBrite DMEM (Gibco, USA) without FBS
for 30 min at 37 °C in 5% CO2. Cells were then rewashed with
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47736

----!@#$NewPage!@#$----
PBS and placed in FluoroBrite DMEM. For fixed-cell imaging,
MCF-7 cells were fixed in 10% neutral buffered formalin (Sigma-
Aldrich, USA) for 15 min at room temperature (RT). Next, cell
membranes were permeabilized with 0.2% Triton X-100 in PBS
for 10 min at RT. After that, cells were incubated with 10 mM
cupric sulfate and 50 mM ammonium acetate (pH 5.0) for 30
min at RT to reduce autofluorescence. Next, cyanine dyes in
PBS were added to the samples for 30 min at RT in the dark.
During incubations, cells were washed three times with PBS. In
the end, samples were embedded into Mowiol 4−88 mounting
medium (Sigma-Aldrich, USA) containing 2.5% DABCO
(Sigma-Aldrich, USA). Microscopy image acquisition of MCF-
7 cells was performed using Leica DM 1000 epifluorescent
microscope with the following excitation (ex) filters: ex 355−
425 nm for Hoechst 33342, ex 450−490 nm for benzoxazole
dyes, and ex 515−560 nm for benzothiazoles.
A2780, HeLa, and Hs27 Live-Cell Fluorescence
Microscopy and Colocalization Study. HeLa and HHF
Hs27 cells were washed twice with preheated PBS for live-cell
imaging. After that, a solution of Hoechst 33342 in DMEM
culture medium (Gibco, USA) was added, and cells were left for
10 min at 37 °C in 5% CO2. The cells were then rewashed twice
with PBS, and the solution of the cyanine dyes was added to the
DMEM culture medium. Cells were incubated for 20 min at 37
°C in 5% CO2 and rewashed twice after with PBS. FluoroBrite
DMEM medium was then added. Images were acquired with the
commercial Nikon High Content Screening microscope
equipped with an Andor Zyla 4.2 sCMOS camera and heating
and atmospheric control chamber, using a 40× 1.30NA Oil
objective and excitation lasers: 380 nm for Hoechst 33342, 480
nm for benzoxazole dyes T-4−T-5, 550 nm for benzothiazole
dyes T-1−T-3. Images were proceeded using free software Fiji/
ImageJ v1.53f51.31
For the T-5 colocalization with LysoTracker DeepRed, the
A2780 cell line was used. Every fluorescent dish with 2 mL of
total volume was left at 37 °C in 5% CO2 overnight. On the next
day, for the colocalization study, solutions of Hoechst 33342,
LysoTracker DeepRed, and the dye T-5 in DMSO were added
into separate and the same dishes for 30 min of incubation for T-
5, 15 min for Hoechst 33342 and LysoTracker DeepRed (1 and
0.1 μM for T-5, and 50 nM for LysoTracker DeepRed). For
other dyes, the following concentrations were used: 10 μM for
T-1, 1 μM for T-2 and T-3, and 0.1 μM for T-4. Then the cells
were washed twice with DPBS, and 5% Medium (2 mL) was
added. The fluorescence images were taken with a Zeiss Axio
Vert.A1 and filter set: ex 625−655/em 665−715 nm (deep red)
for the detection of LysoTracker DeepRed, ex 335−383/em
420−470 nm (blue) for the detection of Hoechst 33342, and ex
450−490/em 500−550 nm (green) for the detection of T-5, T-
4, and ex 538−562/em 570−640 nm for dyes T-1, T-2, and T-3.
Objective: 40×1.30. Oil (DIC). Images were proceeded using
free software Fiji/ImageJ v1.53f51.31
For the colocalization study with MitoTracker CMXRos Red,
the HHF Hs27 cell line was used. Before the experiment, the
growth medium was removed, and the Hs27 cells were washed
twice with preheated PBS. After that, the cells were incubated
with MitoTracker CMXRos Red in DMEM culture medium
(Gibco, USA) for 15 min at 37 °C in 5% CO2. The cells were
then rewashed with PBS, and the solution of the cyanine dye T-5
was added to FluoroBrite DMEM medium. Cells were incubated
with this solution for 20 min at 37 °C in 5% CO2. Images of
Hs27 cells were acquired on the Zeiss LSM980 system,
equipped with a dark heating and atmospheric control chamber,
using Airyscan SR mode, a 63× 1.4NA Oil immersion objective,
with 488 nm laser for the excitation of the dye T-5 and 561 nm
laser for the excitation of MitoTracker CMXRos Red with the
same laser power and scanning speed for all acquisitions. Images
were acquired at 37 °C in 5% CO2. Obtained Airyscan images
were processed with ZEN Blue 3.3 and analyzed using free
software Fiji/ImageJ v1.53f51.31 For colocalization analysis,
seven images of cells stained with T-5 dye and MitoTracker
CMXRos Red were acquired, and Pearson’s coefficient of
correlation was obtained using Fiji plugin JACoP32 with a
followed average value and standard deviation calculation.
The Time-Dependent Photostability Study. The irre-
versible bleaching of trimethine dyes fluorescence was
monitored under constant irradiation as a function of time.
HeLa and HHF Hs27 cells were seeded at concentration 2 × 104
cells per well in 8 Well Chambered Cover Glass with #1.5 high-
performance cover glass (Cellvis) 24 h prior to the experiment.
Before the experiment, the growth medium was removed, and
the cells were washed twice with preheated PBS. After that, the
cells were incubated with Hoechst 33342 in DMEM culture
medium (Gibco, USA) for 10 min at 37 °C in 5% CO2. Then
cells were rewashed with PBS, and a solution of the cyanine dyes
in DMEM culture medium was added at their working
concentrations (given in Figure 6) for 20 min. After cells were
rewashed twice with PBS and FluoroBrite DMEM was added.
Fluorescence was excited with 380 nm laser for Hoechst33342,
550 nm for T-1−T-3, and 480 nm for T-4−T-5 using the
commercial Nikon High Content Screening microscope
equipped with an Andor Zyla 4.2 sCMOS camera and heating
and atmospheric control chamber, using a 40× 1.30NA Oil
objective. The fluorescence intensity was measured using a
special mode in NIS-Elements AR 4.60 software for 8 min. The
software recorded values every 0.05 or 0.1 s. Fluorescence
intensity was normalized as I/I0, where I0 is the fluorescence
intensity at zero time point (0 min 00 s) and I is the fluorescence
intensity at a given time point (up to 8 min).
Cytotoxicity Assay. Cytocompatibility of trimethine dyes
toward A2780 cells was assessed by the standard MTT
Table 1. Characteristics of Vis Absorption Spectra of Dyesa
Tris-HCl buffer, pH 7.9
Tris-HCl buffer, with RNA
Tris-HCl buffer, with dsDNA
MeOH
dye
λmax (nm)
λmax (nm)
λmax (nm)
λmax (nm)
ε (105 M−1cm−1)
T-1
500/552*
502*/562
461*/530*/565
559
1.1
T-2
507*/543
514*/548
510*/544
545
0.7
T-3
504*/541
514*/545
512*/544
543
1.2
T-4
460*/486
463*/487
470*/495
460*/488
0.9
T-5
465*/492
475*/497
473*/497
467*/494
1.7
aλmax, maximum wavelength of the absorption spectrum; *, characteristic shoulder at the edge of the main band; ε, molar extinction coefficient at
λmax, calculated from the main absorption maximum optical density.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47737

----!@#$NewPage!@#$----
cytotoxicity assay based on the reduction of MTT tetrazolium to
MTT formazan pigment by the metabolic activity of living cells.
For this, A2780 cells were seeded at the concentration of 250
cells/μL per well in a 96-well plate. Three separate plates with
three replicate wells were used for each DMSO-containing
control and tested concentrations. T-1−T-5 dyes in three
different concentrations of 10, 1, and 0.1 μM (1 μL of DMSO
stock solution) were added to the cells (100 μL per well into 96-
well plates) and incubated for 6 and 24 h. After incubation, the
MTT dye (20 μL in PBS, 5 mg/mL stock solution) was added to
the cells and incubated for 3 h. Then, 10% SDS solution (90 μL
per well) was added. After overnight incubation, the absorption
of suspensions at λ 590 and 690 nm was measured. The value of
A(590 nm)−A(690 nm) counted as a value proportional to the
percentage of living cells (compared to DMSO-containing
samples as control ones).
■ RESULTS AND DISCUSSION
Photophysical Properties of the Symmetrical Trime-
thine Cyanine Dyes. We have obtained Vis spectra of the
trimethine cyanine dyes in methanol and aqueous buffer 0.05 M
Tris-HCl, pH 7.9 (Table 1, Figure S1). The absorption maxima
of the trimethine cyanines in methanol are located in the range
of 488−559 nm, and the molar extinction values are in the range
(0.7−1.7) × 105 M−1 cm−1. The spectra of the cyanine dyes in
methanol are suggested to belong to their monomeric form. We
have observed that the positions of the absorption maximum are
different and depend on heterocycles in dyes structures. Thus,
the maximums of benzothiazole cyanines (T-1, T-2, T-3) are
located in the longer wavelength region of the spectrum (543−
559 nm) compared to the benzoxazole dyes T-4 and T-5 (488−
494 nm). The absorption of cyanines is determined by the
existence of the delocalized π-electron system. Therefore, it
depends on the length of the polymethine chain. At the same
time, heavy atoms in the heterocycle change the effective length
of the π-electron system, affecting the absorption and emission
maxima.33 As most symmetric cyanines, T-1, T-4, and T-5 have
an utter short-wavelength vibronic maximum as a characteristic
shoulder at the short-wavelength edge of the main band in their
absorption spectra (Figure S1a).34
The aggregation morphology of dyes can significantly affect
their optical properties.35 Benzoxazole dyes (T-4, T-5) are
presented in monomeric form in methanol and aqueous buffer
solution (Figure S1a, b). Benzothiazole cyanines T-2 and T-3
undergo self-organization in an aqueous solution. The reduction
in the absorption of the monomer band and the appearance of a
new maximum at shorter wavelengths 504−507 nm has been
observed (Figure S1a, b). In the case of the T-3, the ε of the
aggregates maximum is almost equal to the ε of the monomers
maximum. The addition of these hypsochromic peaks is a
hallmark of forming H-aggregates (H-type dimers).33,36
Dimerization of cyanines can lead to longer excited singlet
lifetimes but weak or no fluorescence.35 The dye T-1 in the
buffer solution exists as H-dimers with a blue-shifted absorption
peak at 502 nm compared to the monomers (552 nm) with
higher ε (Figure S1b). We can suggest the strong aggregation
behavior of benzothiazole dyes (T-1, T-2, T-3) in an aqueous
solution.
We have also estimated absorption spectra in the presence of
dsDNA and RNA (Table 1, Figure S1c, d). A characteristic
feature of the cyanine spectra in the presence of dsDNA and
RNA is the bathochromic shift of the monomer absorption
maxima relative to free dyes up to 10 nm (Table 1). According to
the literature, such behavior is suggested to be caused by a
different immediate environment of chromophores due to
complexation with nucleic acids.37 The long-wavelength shift in
the absorption maxima is known to be observed due to a
decrease in the nucleophilicity of the environment.37
In addition, there is an increase in optical density for
monomer bands and a decrease for H-type dimer bands in the
case of T-2 and T-3 dyes (Figure S1c, d). This behavior indicates
a dye-nucleic acid interaction, preferably in monomeric form.
For T-4 and T-5 dyes, the shoulder at the edge of the main band
disappears almost entirely in the presence of dsDNA (Figure
S1c) with an immediate effect in the presence of RNA (Figure
S1d). This behavior also indicates the dye-nucleic acid
interaction in monomeric form.
For the dye T-1, different aggregation behavior is observed
depending on the type of nucleic acid (Figure 2a, b). Thus, in the
presence of RNA, broad maxima of the monomer (longwave)
and dimer (shortwave) are observed (Figure 2b). The optical
density of the dye monomer is twice less than the optical density
of the dimer. However, the shape of the spectra differs slightly
from that of the free dye in the buffer solution. Increasing RNA
concentration leads to a slightly decreasing absorbance of the H-
band, indicating the slow breakdown of the H-aggregates with
the minimal recovery of the monomer bands, which reach
equilibrium at the concentration of 0.32 mg/mL. Further
addition of RNA does not affect the H-dimers of the dye T-1.
Figure 2. Absorption spectra of the trimethine cyanine dye T-1 in dsDNA (a) and RNA (b) presence in 0.05 M Tris-HCl (pH 7.9). Dye concentration
5 μM.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47738

----!@#$NewPage!@#$----
In the presence of dsDNA, besides the absorption maxima of
the monomer and dimer of the T-1, there is also the third blue-
shifted absorption maximum band that appears at lower
wavelengths with a large Stokes shift relative to the monomer
(104 nm, Figure 2a). In most cases, the H-type aggregates of
cyanine dyes are limited to H-dimers and extended H-type
aggregates when dye molecules form head-to-head H-type
aggregates, and ideally appear commonly as a broad absorption
band with quenching properties.38 As an exception, the
formation of the chiral arrangement of cyanine dyes with a
narrow absorption band, a huge Stokes shift of cyanine dye
bound to gemini surfactants, has been previously shown.35 We
can assume a similar aggregation behavior of the T-1 dye on the
dsDNA template (Figure 2a). This effect is not observed for
RNA, most likely because it exists mainly in a single-stranded
form. Thus, increasing the concentration of dsDNA leads to a
gradually decreasing absorbance of the narrow absorption H-
band, indicating the progressive disruption of the H-aggregates.
As shown in Figure 2a, the second H-band almost completely
disappears upon adding 0.2 mg/mL of dsDNA. Based on the
absorption spectra, all the dyes studied interact with nucleic
acids but have different preferable types of binding.
Spectral-luminescent properties of the symmetrical trime-
thine cyanine dyes free in Tris-HCl buffer pH 7.9 solution and
the presence of nucleic acids are presented in Table 2. These
dyes possess low to moderate fluorescence intensity in the
aqueous buffer. Introducing bulky substituents into the cyanine
polymethine chain could decrease its free-state fluorescence
intensity compared to the unsubstituted analogs. This effect is
observed for the dye T-2 compared to T-3, which are
structurally identical except for the substituent in the
polymethine chain. Similar to the main absorption bands,
fluorescence emission maxima depend on the presence of heavy
atoms in heterocycles. In the case of the benzoxazole ring,
trimethine cyanine fluorescence is observed at 501−506 nm,
while those of benzothiazole have a maximum of 564−580 nm.
The Stokes shift value for free dyes is in the range of 12 to 18 nm.
The addition of nucleic acids leads to an increase in
fluorescence intensity (15−111 times). The maximums of the
fluorescence emission spectra of dyes in the presence of RNA
and dsDNA are shifted in the long-wavelength region for 4−13
nm. The increase in Stokes shift is observed for T-2 and T-3 dyes
in the presence of nucleic acids (Table 2). The most pronounced
fluorescent response is observed for T-1 in the presence of
dsDNA (111 times) and up to 83 times for dye T-2 in the
presence of RNA. The fluorescent sensitivity of the studied dyes
to dsDNA and RNA is not similar, as seen in Table 2. The T-2
and T-3 dyes were 1.5−2.2 times more sensitive to RNA than to
dsDNA. T-1 dye was 1.5 times more sensitive to dsDNA than to
RNA. For other dyes, the IRNA/IDNA ratio was 0.7−0.89 (Table
2), probably meaning no selective specificity to NA.
The binding of dyes to nucleic acids, depending on their
structure, can occur by (i) electrostatic interaction between the
cationic dyes and the anionic phosphodiester groups of the
dsDNA/RNA backbone, (ii) intercalation between adjacent
base pairs, (iii) minor groove binding,10,39 and (iv) half-
intercalation interaction mode.40,41 It is also known that groove
binding to DNA becomes more prevalent for the cyanine dyes
with more than one methine bridge. For example, trimethine-
bridged cyanine dyes have been shown to bind to the minor
DNA groove as monomers and dimers.42
For dyes T-2 and T-3 with nucleic acids (both dsDNA and
RNA), there is an increase in the value of the Stokes shift (ΔS)
by 11−16 nm. The increase in the ΔS for cyanine dyes occurs
with increasing electron asymmetry of the molecule.43 The
electrical asymmetry of the dye molecule increases with the half-
intercalating mode of binding to DNA when the molecule is
simultaneously oriented in the dipole medium between base
pairs and negative charges of phosphate groups in the groove.
Thus, given the increase in ΔS values and the partial destruction
of dimers in the absorption spectra in the presence of dsDNA, it
can be assumed that the dyes T-2 and T-3 also bind to dsDNA
due to a half-intercalating mechanism in monomeric form.
As mentioned above, we have observed that dyes T-2 and T-3
have slightly higher fluorescence sensitivity to RNA than
dsDNA. The different sensitivity of benzothiazole cyanine
dyes for different NA was explained in.44 These dyes could
selectively form fluorescent J-aggregates when bound to the
DNA molecule. However, based on our results in Table 1, the
studied benzothiazole cyanine dyes do not form J-aggregate
peaks in the absorbance spectra. Thus, we have considered
another explanation for this phenomenon. For example, forms of
secondary structures of RNA make the binding sites for these
dyes more preferable. In ref 45, perturbations of the A-form helix
of the RNA molecule induced by un- or mispaired bases widen
the major groove to provide a surface-exposed binding pocket,
and pockets found in RNA present varying degrees of
electronegative potentials. The various folds adopted by the
different secondary structures offer specific binding sites further
characterized by their electronegative landscape. Binding to
such pockets can lead to changes in the electrical asymmetry of
the dye molecules, which is confirmed by an increase in the ΔS
for T-2 and T-3 in the presence of RNA. Therefore, we can
suggest that the addition of the longer β-substituents in the
polymethine chain led to stronger binding to the secondary
structure of the RNA molecule than dsDNA.
The most significant changes occur in the electronic spectra of
the dsDNA complex with the T-1 dye (Figure 2a), which shows
the highest increase in fluorescence in its presence. Redis-
Table 2. Spectral-Luminescent Properties of Trimethine Cyanine Dyes in a Free State and the Presence of Nucleic Acids in 0.05
M Tris-HCl Buffer (pH 7.9)a
free state
with RNA
with dsDNA
λex. (nm)
λem. (nm)
I (a.u.)
ΔS (nm)
λex. (nm)
λem. (nm)
IRNA/I (t)
ΔSRNA (nm)
λex. (nm)
λem. (nm)
IDNA/I (t)
ΔSDNA (nm)
T-1
562
580
15
18
567
584
75
17
566
578
111
12
T-2
554
567
46
13
550
575
83
25
545
569
56
24
T-3
552
564
118
12
548
574
33
26
545
567
15
22
T-4
488
501
124
12
499
511
65
12
500
511
73
11
T-5
493
506
312
13
505
519
26
14
505
517
37
12
aλex. (λem.), maximum wavelength of fluorescence excitation (emission) spectrum; I, emission intensity of dye in a free state (I) and in the presence
of nucleic acids (IRNA, IDNA); a.u., arbitrary units; t, the number of times; ΔS, Stokes shift.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47739

----!@#$NewPage!@#$----
tribution between three electronic bands upon adding dsDNA
indicates three distinct binding modes. We suggest that the
addition of dsDNA leads to the formation of the arrangement of
dye molecules at 461 nm using dsDNA as a template. These
aggregates can interact with dsDNA through electrostatic
binding or small groove binding. Subsequent addition of
dsDNA leads to the destruction of the aggregates and increases
the absorption maximum of the monomers, accompanied by an
Figure 3. Fluorescence imaging of live (a) and fixed (b) MCF-7 cells stained with the studied dyes T-1−T-5 (dyes concentrations for live cell imaging:
T-1−10 μM; T-2, T-3−1 μM; T-4, T-5−0.1 μM; for fixed cell imaging: T-1, T-4−1 μM; T-2, T-3, T-5−0.1 μM). Scale bars: 10 μm.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47740

----!@#$NewPage!@#$----
increase in fluorescence. The other absorption peak at 565 nm
proportionally increases with the destruction of aggregates and
is most probably associated with an intercalated complex. These
observations indicate that the dye T-1 is bound to dsDNA in
monomeric form with an increased DNA/dye ratio and that the
preferred binding mode can be intercalative. In the presence of
proteins (HSA and BLG), no significant increase in the
fluorescence intensity of the dyes was observed (Table S1).
Thus, the fluorescence specificity and interaction with nucleic
acids by spectral methods are shown for all studied dyes.
The Fluorescence Microscopy Study of Staining
Properties of Trimethine Cyanine Dyes in Cancer Cells.
To understand if the studied trimethine cyanine dyes are
suitable for cell visualization by fluorescence microscopy, we
have stained human breast cancer cells MCF-7 with the
following visualization. We suggested that the dyes may be
specific for nucleic acids in cells (RNA-containing cell organelles
or nuclei). Therefore, we have investigated the ability of dyes to
penetrate the cell and stain its components (Figure 3).
The MCF-7 cell line was chosen for this purpose because it is a
well-characterized, widely used breast cancer model with
adherent and nicely spread cells suitable for imaging subcellular
architecture.46 MCF-7 cells were incubated with different dye
concentrations (10 μM, 1 μM, 0.1 μM) to determine the
working conditions of the dyes. We have used 450−490 nm
irradiation for benzoxazoles and 515−560 nm irradiation for
benzothiazoles to excite the fluorescence of the cyanine dyes
studied.
We have found that all dyes proved to be membrane-
permeant and stain the subcellular components both in live and
fixed cells with different sensitivity (Figure 3). Optimal results
for live cell staining were obtained at the following
concentrations: 10 μM for dye T-1, 1 μM for dyes T-2 and T-
3, and 0.1 μM for dyes T-4 and T-5. For fixed cell staining,
optimal conditions were 1 μM for dyes T-1, and T-4, 0.1 μM for
dyes T-2, T-3, and T-5. All studied dyes penetrated the nuclear
membrane and stained large structures within the nucleus, most
probably nucleoli, except for the dye T-5, which at the
concentration of 0.1 μM, specifically stains some structures in
the cytoplasm of MCF-7 cells.
In fixed cells, the dyes studied possessed a similar behavior
with slightly stronger fluorescence in the cytoplasm. The
exception is the T-5 dye, which shows a different staining
pattern from live to fixed cells at the same concentration of 0.1
μM: (i) staining compartments (mitochondria or lysosomes) in
the cytoplasm without nucleus penetration in live cells and (ii)
penetrates the nucleus and staining nucleoli with background
fluorescence in the cytoplasm in fixed cells. Usually, a charge of
the mitochondria membrane is involved in binding dyes to this
organelle, but in fixed cells, the charge disappears, and binding
does not occur. Thus, we don’t see the dye accumulation in the
cytoplasm and, consequently, weak fluorescent green back-
ground with quite a bright staining of nucleoli.
Staining two more cancer cell lines (human ovarian
adenocarcinoma cell line A2780 and cervical cancer cell line
HeLa) was performed to check if the staining pattern remains
the same regardless of cell type. It was shown that dyes penetrate
cytoplasm membranes and stain A2780 and HeLa cell
compartments at the same working concentration as MCF-7
(Figures S2 and S3).
The cytocompatibility of the dyes has been investigated using
cell line A2780. A standard cytotoxicity MTT test was
performed for 6 and 24 h (Figure 4). After 24 h, studied
benzoxazoles are less cytotoxic to cancer cells at working
concentrations than benzothiazoles (67% of cell viability
compared to 12% for T-1 and ∼30% for T-2, T-3). In contrast,
after 6 h, benzoxazoles T-4/T-5 and benzothiazoles T-2/T-3 at
working concentrations show almost the same cytotoxicity
(∼85% of cell viability). However, the toxicity of benzothiazoles,
but not benzoxazoles, increases with increasing concentration.
Thus, the dyes T-4 and T-5 have the highest cytocompati-
bility, ensuring the applicability of this dye as a potentially useful
probe for long-term live-cell imaging.
Staining of Human Foreskin Fibroblasts by Trime-
thine Cyanine Dyes. The membrane potential (Vm) in cancer
cells changes. Namely, electrophysiological analyses in many
cancer cell types have shown a depolarized Vm.47 As a result, the
cancer cell surface is negatively charged, and the cytoplasm has a
positive charge. Furthermore, cancer cells have a higher
intracellular pH (≥7.4) and a lower extracellular pH (∼6.7−
7.1).48 These pH conditions in cancer cells are entirely opposite
to the pH gradient of normal cells.49,50 This hallmark of cancer
cells lies in the overactivation of plasma membrane ion pumps
and transporters that extrude protons and intrude other ions.
Figure 4. Comparisons of the cytotoxicity of trimethine cyanines T-1−T-5 at different concentrations (0.1, 1, and 10 μM) toward A2780 cells for 6 (a)
and 24 (b) h. Working concentrations are marked with a circle. The standard deviation is presented as error bars.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47741

----!@#$NewPage!@#$----
Moreover, the mitochondria membrane of cancer cells is
hyperpolarized compared to normal cells.51 Another character-
istic of cancer cell metabolism is the higher ROS levels observed
in the tumor cells than in healthy ones. Because the studied dyes
carry a positive (cationic) charge, it was decided to test whether
differences in these parameters in cancer and normal cells affect
the ability of the dyes to penetrate the cell and their organelle
specificity. For this purpose, the human foreskin fibroblast cell
line (Hs27) was selected as a model object for live-cell imaging.
The fluorescent microscopy study has shown that all studied
dyes keep the same staining pattern in normal Hs27 cells as in
cancer cells MCF-7, except for dye T-1. Thus, this dye does not
penetrate the nucleus of Hs27 and stains some roundish
structures in the cytoplasm that could be swollen mitochondria
or lysosomes (Figure 5).
The Time-Dependent Photostability of Trimethine
Cyanine Dyes in HeLa and Hs27 Cells. Photodegradation,
commonly referred to as photobleaching, depletes the
fluorophore concentration under prolonged irradiation and
represents one of the most severe limitations in fluorescence
microscopy.33 We have investigated the photostability of dyes
under continuous irradiation in cancer HeLa and normal Hs27
cells. Benzoxazole cyanine dyes are more photostable under
direct irradiation than benzothiazole ones (Figure 6a).
Increasing the length of the polymethine chain for cyanine
was shown to reduce photostability.52 After 1 min 30 s of
Figure 5. Fluorescence imaging of live Hs27 cells stained with the studied dyes T-1−T-5. (T-1 at the concentration of 10 μM, T-2, and T-3 dyes at the
concentration of 1 μM, T-4, and T-5 at the concentration of 0.1 μM). Scale bars: 10 μm.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47742

----!@#$NewPage!@#$----
constant irradiation in Hs27 cells, benzoxazole T-4 and T-5
remain at 22% and 15% of their fluorescence intensity,
respectively. In the case of HeLa cells, they remain at 10% and
15% of fluorescence intensity.
In contrast, dyes T-2/T-3 in Hs27 cells were up to 2 times less
photostable (and the dye T-1 is ∼7 times less stable compared to
T-4). All benzothiazoles in HeLa cells are 7 times less stable than
benzoxazoles. However, a lower concentration of T-3 (0.1 μM)
had to be used in cancer cells due to the signal’s overbrightness.
Similar results were previously shown on monomethine cyanine
dyes, where benzoxazole derivative was ∼5 times more
photostable than benzothiazoles.19 We suggest that a heavy
atom in the heterocycle of benzoxazole reduces the effective
length of the π-electron system (thus, the length of the
polymethine chain), positively affecting the stability of T-4
and T-5 dyes. The obtained photostability results are consistent
with the location of the absorption maximums and the relation
between the polymethine chain length and photostability. We
assume symmetrical cationic benzoxazole cyanines as more
suitable probes for cell visualization, and the dye T-5 was chosen
for further studies.
Cells Staining by the Benzoxazole Dye T-5 and Its
Colocalization with MitoTracker CMXRos Red. T-5
staining ability was examined in another cancer cell line�
A2780. At a working concentration equal to 0.1 μM, the same
staining pattern as above was shown: the dye could penetrate
inside the cell, without penetration through the nuclear
membrane and stain some structures inside the cytoplasm
(Figure S4b). Also, as already shown for MCF-7, this dye lost its
specificity to these structures in the higher concentration. At 1
μM, T-5 brightly stains cytoplasm in A2780 cells entirely
without any specificity and penetrates the nucleus staining
nucleoli (Figure S4a). Colocalization of T-5 and lysosome-
targeting standard dye LysoTracker DeepRed (50 nM) was
carried out to understand the specificity of the dye in cell
imaging (Figure S5). This study showed no colocalization of T-5
with LysoTracker DeepRed. Thus, the dye in a low
concentration does not penetrate the nucleus and does not
colocalize with lysosomes in the cytoplasm.
We have shown that the dye T-5 is membrane-permeant and
stains some compartments in cell cytoplasm at a concentration
of 0.1 μM in all studied cell lines. It was assumed as a pattern of
mitochondria staining. Thus, a colocalization analysis was
carried out for the benzoxazole cyanine dye T-5 with the
MitoTracker CMXRos Red (a dye specific for mitochondria).
As seen in Figure 7c, T-5 shows signals overlapping with the
mitochondria-targeting dye MitoTracker CMXRos Red with
Person’s value r = 0.92 ± 0.04, proving its mitochondria
localization (7 separate calculations are given in Figure S6).
■ CONCLUSIONS
We have presented the synthesis of cationic symmetrical
trimethine cyanine dyes and discussed their spectral-lumines-
cence properties and applicability in cell imaging. All dyes
exhibit a strong fluorescent ‘light-up’ response in the presence of
nucleic acids, including RNA and double-stranded DNA.
Figure 6. Time-dependent confocal fluorescence microscopy imaging of HeLa (a) and Hs27 (b) cells stained by trimethine cyanine dyes T-1−T-5 (at
the concentration of 10 μM for T-1, 1 μM for T-2, T-3, 0.1 μM for T-4 and T-5 for Hs27, and the same for HeLa, but the concentration for T-3 is 0.1
μM) under constant irradiation for 8 min.
Figure 7. Fluorescence live-cell imaging of Hs27 cells. Colocalization analysis of the dye T-5 (green, a) with MitoTracker CMXRos Red (red, b) and
merged image (c). Scale bars: 10 μm.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47743

----!@#$NewPage!@#$----
Wavelengths of fluorescence maxima of these dyes depend on
heterocycles in dyes structures: benzoxazole trimethine cyanine
fluorescence in a free state is observed on 501−506 nm, while
dyes with benzothiazole heterocycle have a maximum of 564−
580 nm with excitation wavelength nearly on 490 and 550−560
nm, respectively. These dyes possess low to moderate
fluorescence intensity in the aqueous buffer. The addition of
nucleic acids leads to an increase in fluorescence intensity (15−
111 times). Spectral methods showed the binding of all dyes to
nucleic acids, and different interaction mechanisms have been
proposed.
The ability to visualize cell components of the studied dyes
was tested on different cell lines (MCF-7, A2780, HeLa, Hs27).
We have shown that all dyes are cell-permeant staining nucleus
components, probably RNA-rich nucleoli with background
fluorescence in the cytoplasm, except for the dye T-5, which at a
concentration of 0.1 μM, specifically stains some structures in
the cytoplasm. Colocalization analysis of the dye T-5 with
MitoTracker CMXRos Red shows that the T-5 dye stains
mitochondria (Pearson’s Coefficient value = 0.92 ± 0.04).
Fluorescence imaging of Hs27 cells by dye T-5 shows no
difference in mitochondrial selectivity between normal and
cancer cells. The photostability study of dyes in cells under
constant irradiation showed that benzoxazole dyes are up to ∼7
times more photostable than benzothiazole ones. Moreover,
studied benzoxazoles are less cytotoxic to cancer cells at working
concentrations than benzothiazoles (67% of cell viability
compared to 12% for T-1 and ∼30% for T-2, T-3 after 24 h).
Because of its sensitive fluorescence response to nucleic acids
with a significant fluorescence increase, moderate photostability,
and low cytotoxicity, benzoxazole T-4 dye can be used for
nucleic acid detection in vitro and intracellular fluorescence
imaging in live and fixed cells. On the other hand, the
benzoxazole dye T-5 is proposed as a good alternative to
commercial dyes for mitochondria staining in the green-yellow
region of the spectrum.
■ ASSOCIATED CONTENT
*
sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acsomega.2c05231.
The absorption spectra of trimethine cyanines in MeOH,
in 0.05 M Tris-HCl pH 7.9, in dsDNA presence and RNA
presence; luminescent properties of trimethine cyanines
in the presence of globular proteins; fluorescence live-cell
imaging of A2780, HeLa cells, and colocalization
experiment with LysoTracker DeepRed using A2780;
separate merged CLSM images and Pearson’s coefficient
for colocalization of T-5 with MitoTracker CMXRos Red
using Hs27 (PDF)
■ AUTHOR INFORMATION
Corresponding Author
Svitlana Chernii − Institute of Molecular Biology and Genetics
NASU, 03143 Kyiv, Ukraine; V.I. Vernadsky Institute of
General and Inorganic Chemistry NASU, 03142 Kyiv,
Ukraine;
orcid.org/0000-0003-2034-8429;
Email: chernii.sv@gmail.com
Authors
Daria Aristova − Institute of Molecular Biology and Genetics
NASU, 03143 Kyiv, Ukraine; Instituto Gulbenkian de Ciência,
2780-156 Oeiras, Portugal
Roman Selin − V.I. Vernadsky Institute of General and
Inorganic Chemistry NASU, 03142 Kyiv, Ukraine; Organic
Chemistry II, Friedrich-Alexander-University of Erlangen-
Nuremberg, 91058 Erlangen, Germany
Hannah Sophie Heil − Instituto Gulbenkian de Ciência, 2780-
156 Oeiras, Portugal
Viktoriia Kosach − Institute of Molecular Biology and Genetics
NASU, 03143 Kyiv, Ukraine
Yuriy Slominsky − Institute of Organic Chemistry NASU,
02094 Kyiv, Ukraine
Sergiy Yarmoluk − Institute of Molecular Biology and Genetics
NASU, 03143 Kyiv, Ukraine
Vasyl Pekhnyo − V.I. Vernadsky Institute of General and
Inorganic Chemistry NASU, 03142 Kyiv, Ukraine
#Vladyslava Kovalska − Institute of Molecular Biology and
Genetics NASU, 03143 Kyiv, Ukraine
Ricardo Henriques − Instituto Gulbenkian de Ciência, 2780-
156 Oeiras, Portugal
Andriy Mokhir − Organic Chemistry II, Friedrich-Alexander-
University of Erlangen-Nuremberg, 91058 Erlangen,
Germany;
orcid.org/0000-0002-9079-5569
Complete contact information is available at:
https://pubs.acs.org/10.1021/acsomega.2c05231
Author Contributions
Conceptualization: Vladyslava Kovalska. Formal analysis:
Svitlana Chernii, Selin Roman, Daria Aristova, Hannah Sophie
Heil. Investigation: Daria Aristova, Svitlana Chernii, Hannah
Sophie Heil, Viktoriia Kosach, Roman Selin. Methodology:
Yuriy Slominsky. Project administration: Andriy Mokhir,
Ricardo Henriques. Resources: Andriy Mokhir, Ricardo
Henriques, Vasyl Pekhnyo. Supervision: Andriy Mokhir,
Ricardo Henriques. Validation: Vasyl Pekhnyo, Sergiy Yarmo-
luk. Visualization: Svitlana Chernii, Daria Aristova. Writing−
original draft: Svitlana Chernii, Daria Aristova, Hannah Sophie
Heil. Writing−review and editing: Andriy Mokhir, Ricardo
Henriques, Svitlana Chernii, Daria Aristova. The authors have
read and agreed to the published version of the manuscript.
Notes
The funders had no role in the design of the study; in the
collection, analyses, or interpretation of data; in the writing of
the manuscript; or in the decision to publish the results.
The authors declare no competing financial interest.
#Dr. Vladyslava Kovalska, who initiated and inspired this work,
suddenly passed away on 03 December 2020.
■ ACKNOWLEDGMENTS
D.A., R.S., A.M., and S.Ch. are supported by the European
Union’s Horizon 2020 research and innovation programme
under the Marie Skłodowska-Curie grant agreement No.
872331: “NoBiasFluors”. In addition, S.Ch. and R.S. thank the
grant of a group of young scientists of NAS of Ukraine No.
0122U002204 for 2022−2023. D.A., H.S.H, and R.H. also
received support from the European Research Council (ERC)
under the European Union’s Horizon 2020 research and
innovation programme (grant agreement No. 101001332),
Horizon 2021 INFRA programme (grant agreement No.
101057970), the European Molecular Biology Organization
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47744

----!@#$NewPage!@#$----
(EMBO) Installation Grant (EMBO-2020-IG4734), the Chan
Zuckerberg Initiative Visual Proteomics Grant (vpi-
0000000044). H.S.H. is supported by European Molecular
Biology Organization (ALTF 499-2021) and Fundação para a
Ciência e Tecnologia, Portugal (FCT) fellowship CEECIND/
01480/2021.
■ REFERENCES
(1) Hickey, S. M.; Ung, B.; Bader, C.; Brooks, R.; Lazniewska, J.;
Johnson, I. R. D.; Sorvina, A.; Logan, J.; Martini, C.; Moore, C. R.;
Karageorgos, L.; Sweetman, M. J.; Brooks, D. A. Fluorescence
Microscopy − An Outline of Hardware, Biological Handling, and
Fluorophore Considerations. Cells 2022, 11 (1), 35.
(2) Ettinger, A.; Wittmann, T. Fluorescence live cell imaging. Methods
Cell Biol. 2014, 123, 77−94.
(3) Hua, X.-W.; Bao, Y.-W.; Zeng, J.; Wu, F.-G. Nucleolus-Targeted
Red Emissive Carbon Dots with Polarity-Sensitive and Excitation-
Independent Fluorescence Emission: High-Resolution Cell Imaging
and in Vivo Tracking. ACS Appl. Mater. Interfaces. 2019, 11 (36),
32647−32658.
(4) Hua, X.-W.; Bao, Y.-W.; Wu, F.-G. Fluorescent Carbon Quantum
Dots with Intrinsic Nucleolus-Targeting Capability for Nucleolus
Imaging and Enhanced Cytosolic and Nuclear Drug Delivery. ACS
Appl. Mater. Interfaces. 2018, 10 (13), 10664−10677.
(5) Liu, J.; Li, R.; Yang, B. Carbon Dots: A New Type of Carbon-
Based Nanomaterial with Wide Applications. ACS Cent. Sci. 2020, 6
(12), 2179−2195.
(6) Specht, E. A.; Braselmann, E.; Palmer, A. E. A Critical and
Comparative Review of Fluorescent Tools for Live-Cell Imaging. Annu.
Rev. Physiol. 2017, 79, 93−117.
(7) Williams, C. XXVI.�Researches on Chinoline and its
Homologues. Trans. - R. Soc. Edinburgh 1857, 21, 377−401.
(8) Ma, X.; Shi, L.; Zhang, B.; Liu, L.; Fu, Y.; Zhang, X. Recent
advances in bioprobes and biolabels based on cyanine dyes. Anal.
Bioanal. Chem. 2022, 414, 4551−4573.
(9) Sun, W.; Guo, S.; Hu, Ch.; Fan, J.; Peng, X. Recent Development
of Chemosensors Based on Cyanine Platforms. Chem. Rev. 2016, 116,
7768−7817.
(10) Yarmoluk, S.; Kovalska, V.; Volkova, K. Optimized dyes for
protein and nucleic acid detection. Springer Ser. Fluoresc 2011, 113,
161−199.
(11) Jones, L. J.; Singer, V. L. Fluorescence Microplate-Based Assay
for Tumor Necrosis Factor Activity Using SYTOX Green Stain. Anal.
Biochem. 2001, 293, 8−15.
(12) Roth, B. L.; Poot, M.; Yue, S. T.; Millard, P. J. Bacterial viability
and antibiotic susceptibility testing with SYTOX green nucleic acid
stain. Appl. Environ. Microbiol. 1997, 63, 2421−31.
(13) Lebaron, P.; Catala, P.; Parthuisot, N. Effectiveness of SYTOX
Green stain for bacterial viability assessment. Appl. Environ. Microbiol.
1998, 64, 2697−700.
(14) Johnson, I. D.; Spence, M. T. Z. The Molecular Probes Handbook:
A Guide To Fluorescent Probes And Labeling Technologies, 11th ed.;
Academic Press, Life Technologies: Carlsbad, CA, USA, 2010, 307−
314.
(15) Pronkin, P. G.; Tatikolov, A. S. Photonics of Trimethine Cyanine
Dyes as Probes for Biomolecules. Molecules 2022, 27, 6367.
(16) Moreira, B. G.; You, Y.; Owczarzy, R. Cy3 and Cy5 dyes attached
to oligonucleotide terminus stabilize DNA duplexes: predictive
thermodynamic model. Biophys Chem. 2015, 198, 36−44.
(17) Suzuki, T.; Fujikura, K.; Higashiyama, T.; Takata, K. DNA
Staining for Fluorescence and Laser Confocal Microscopy. J. Histochem.
Cytochem. 1997, 45, 49−53.
(18) Ohulchanskyy, T. Yu; Pudavar, H. E.; Yarmoluk, S. M.;
Yashchuk, V. M.; Bergey, E. G.; Prasad, P. N. A monomethine cyanine
dye Cyan 40 for two-photon-excited fluorescence detection of nucleic
acids and their visualization in live cells. Photochem. Photobiol. 2003, 77,
138−145.
(19) Aristova, D.; Kosach, V.; Chernii, S.; Slominsky, Yu.; Balanda, A.;
Filonenko, V.; Yarmoluk, S.; Rotaru, A.; Ozkan, H. G.; Mokhir, A.;
Kovalska, V. Monomethine cyanine probes for visualization of cellular
RNA by fluorescence microscopy. Methods Appl. Fluoresc. 2021, 9,
045002.
(20) Aristova, D.; Volynets, G.; Chernii, S.; Losytskyy, M.; Balanda,
A.; Slominskii, Y.; Mokhir, A.; Yarmoluk, S.; Kovalska, V. Far-red
pentamethine cyanine dyes as fluorescent probes for the detection of
serum albumins. R. Soc. Open Sci. 2020, 7, 200453.
(21) Syniugina, A.; Chernii, S.; Losytskyy, M.; Syniugin, A.;
Slominskii, Y.; Balanda, A.; Yarmoluk, S.; et al. The synthesis and
study of novel merocyanine probes for protein detection and cells
visualization. Photochem. Photobiol. 2021, 7, 100046.
(22) Yarmoluk, S. M.; Kovalska, V. B.; Volkova, K. D. Optimized Dyes
for Protein and Nucleic Acid Detection. Springer Ser. Fluoresc. 2011,
113, 161−199.
(23) Lukashov, S. S.; Losytskyy, M. Yu.; Slominskii, Yu. L.; Yarmoluk,
S. M. Interaction of cyanine dyes with nucleic acids. 7. Carbocyanine
dyes, substituted in polymethine chain, as possible probes for
fluorescent nucleic acid detection [in Ukrainian]. Biopolym. Cell.
2001, 17 (2), 169−177.
(24) Armitage, B. A. Armitage Cyanine Dye−Nucleic Acid
Interactions. Top. Heterocycl. Chem. 2008, 14, 11−29.
(25) Anikovsky, M. Y.; Tatikolov, A. S.; Kuzmin, V. A. Fluorescent
Properties of Some Thia- and Oxacarbocyanine Dyes in the Presence of
DNA. High Energy Chem. 2002, 36, 179−183.
(26) Kiprianov, A. I.; Sych, K. D. Zapysky Instytutu Khimii Akademii
Nauk URSR [In Russian] 1946, 8, 105−125.
(27) Hamer, F. M. The cyanine dyes and related compounds. The
chemistry of heterocyclic compounds. XVIII; Academic Press: New York,
1964; p 790.
(28) Brooker, L. G. S.; White, F. L. Studies in the Cyanine Dye Series.
II. Carbocyanines with Substituents in the Three-Carbon Chain. J. Am.
Chem. Soc. 1935, 57 (12), 2480−2488.
(29) Lukashov, S. S.; Makovenko, I. E.; Losytskyy, M.Yu.; Slominskii,
Yu.L.; Yarmoluk, S. M. Interaction of cyanine dyes with nucleic acids.
Meso-methylsubstituted trimethincyanines, as possible probes for
fluorescent nucleic acid detection. Biopolym. Cell. 2001, 17 (5), 448−
454.
(30) Kabatc, J.; Pa̧czkowski, J. The photophysical and photochemical
properties of the oxacarbocyanine and thiacarbocyanine dyes. Dyes
Pigm. 2004, 61 (1), 1−16.
(31) Schindelin, J.; Arganda-Carreras, I.; Frise, E.; Kaynig, V.; Longair,
M.; Pietzsch, T.; Preibisch, S.; Rueden, C.; Saalfeld, S.; Schmid, B.;
Tinevez, J.; White, D.; Hartenstein, V.; Eliceiri, K.; Tomancak, P.;
Cardona, A. Fiji: an open-source platform for biological-image analysis.
Nat. Methods. 2012, 9, 676−682.
(32) Bolte, S.; Cordelières, F. P. A guided tour into subcellular co-
localization analysis in light microscopy. J. Microsc. 2006, 224, 213−
232.
(33) Levitus, M.; Ranjit, S. Cyanine dyes in biophysical research: The
photophysics of polymethine fluorescent dyes in biomolecular
environments. Q. Rev. Biophys. 2011, 44 (1), 123−151.
(34) Pronkin, P.; Tatikolov, A. Isomerization and Properties of
Isomers of Carbocyanine Dyes. Sci. 2019, 1 (1), 19.
(35) Ryu, N.; Okazaki, Y.; Pouget, E.; Takafuji, M.; Nagaoka, S.; Ihara,
H.; Oda, R. Fluorescence emission originated from the H-aggregated
cyanine dye with chiral gemini surfactant assemblies having a narrow
absorption band and a remarkably large Stokes shift. Chem. Commun.
2017, 53 (63), 8870−8873.
(36) Bricks, J. L.; Slominskii, Y. L.; Panas, I. D.; Demchenko, A. P.
Fluorescent J-aggregates of cyanine dyes: basic research and
applications review. Methods Appl. Fluoresc. 2018, 6 (1), 012001.
(37) Lukashov, S. S.; Kachkovskyy, G. O.; Losytskyy, M. Y.; Yarmoluk,
S. M. The interaction of cyanine dyes with nucleic acids. 22. Spectral-
luminescent properties of monomethine pyrylium and pyridinium
cyanines and their DNA complexes [In Ukrainian]. Biopolym. Cell.
2001, 17 (3), 242−248.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47745

----!@#$NewPage!@#$----
(38) Rösch, U.; Yao, S.; Wortmann, R.; Würthner, F. Fluorescent H-
Aggregates of Merocyanine Dyes. Angew. Chem. 2006, 118 (42), 7184−
7188.
(39) Zhytniakivska, O.; Zabrudska, A.; Tarabara, U.; Vus, K.; Trusova,
V.; Gorbenko, G.; Kurutos, A.; Deligeorgiev, T. Competitive Binding of
Novel Cyanine Dye AK3−5 and Europium Coordination Complexes to
DNA. East Eur. J. Phys. 2019, 3, 63−70.
(40) Yarmoluk, S. M.; Kovalska, V. B.; Kovtun, Yu.P. Interaction of
cyanine dyes with nucleic acids. 5. Towards model of “half
intercalation” of monomethyne cyanine dyes into double-stranded
nucleic acids. Biopolym. Cell. 1999, 15, 75−82.
(41) Yarmoluk, S. M.; Lukashov, S. S.; Ogul’chansky, T.Yu.;
Losytskyy, M.Yu.; Kornyushyna, O. S. Interaction of cyanine dyes
with nucleic acids. XXI. Arguments for half- intercalation model of
interaction. Biopolymers. 2001, 62, 219−227.
(42) Sovenyhazy, K. M.; Bordelon, J. A.; Petty, J. T. Spectroscopic
studies of the multiple binding modes of a trimethine-bridged cyanine
dye with DNA. Nucleic Acids Res. 2003, 31 (10), 2561−2569.
(43) Ishchenko, A. A. Structure and spectral-luminescent properties of
polymethine dyes. Russ. Chem. Rev. 1991, 60 (8), 865−884.
(44) Yarmoluk, S.; Kovalska, V.; Losytskyy, M. Symmetric cyanine
dyes for detecting nucleic acids. Biotechnol. Histochem. 2008, 83 (3−4),
131−145.
(45) Thomas, J. R.; Hergenrother, P. J. Targeting RNA with Small
Molecules. Chem. Rev. 2008, 108 (4), 1171−1224.
(46) Comşa, Ş.; Cîmpean, A. M.; Raica, M. The Story of MCF-7
Breast Cancer Cell Line: 40 years of Experience in Research. Anticancer
Res. 2015, 35 (6), 3147−3154.
(47) Yang, M.; Brackenbury, W. J. Membrane potential and cancer
progression. Front Physiol. 2013, 4, 185.
(48) Gillies, R. J.; Raghunand, N.; Karczmar, G. S.; Bhujwalla, Z. M.
MRI of the tumor microenvironment. J. Magn. Reson. Imaging. 2002, 16,
430−450.
(49) Webb, B. A.; Chimenti, M.; Jacobson, M. P.; Barber, D. L.
Dysregulated pH: A perfect storm for cancer progression. Nat. Rev.
Cancer. 2011, 11, 671−677.
(50) White, K. A.; Grillo-Hill, B. K.; Barber, D. L. Cancer cell
behaviors mediated by dysregulated pH dynamics at a glance. J. Cell Sci.
2017, 130, 663−669.
(51) Bhat, T. A.; Kumar, S.; Chaudhary, A. K.; Yadav, N.; Chandra, D.
Restoration of mitochondria function as a target for cancer therapy.
Drug Discovery Today. 2015, 20, 635−643.
(52) Pham, W.; Lai, W. F.; Weissleder, R.; Tung, C. H. High efficiency
synthesis of a bioconjugatable near-infrared fluorochrome. Bioconjugate
Chem. 2003, 14, 1048−51.
ACS Omega
http://pubs.acs.org/journal/acsodf
Article
https://doi.org/10.1021/acsomega.2c05231
ACS Omega 2022, 7, 47734−47746
47746
 Recommended by ACS
Synthesis of Aza-BODIPYs, Their Differential Binding for
Cu(II), and Results of Bioimaging as Fluorescent Dyes of
Langerhans β-Cells
Yaneth C. Pino, Ignacio A. Rivero, et al.
NOVEMBER 15, 2022
ACS OMEGA
READ 
Benchmarking Thiolate-Driven Photoswitching of Cyanine
Dyes
Lucas Herdly, Sebastian van de Linde, et al.
JANUARY 13, 2023
THE JOURNAL OF PHYSICAL CHEMISTRY B
READ 
Light-Deactivated Fluorescent Probes (FLASH-Off) for
Multiplexed Imaging
Elias A. Halabi and Ralph Weissleder
APRIL 05, 2023
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
READ 
General Strategy To Improve the Photon Budget of Thiol-
Conjugated Cyanine Dyes
Yuan Zhang, Chunlai Chen, et al.
FEBRUARY 09, 2023
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
READ 
Get More Suggestions >