Synthesis of pH-sensitive benzothiazole cyanine dye derivatives containing a pyridine moiety at the meso position

Article abstract of DOI:10.1016/j.dyepig.2021.109268

Four benzothiazole-derived pentamethine cyanine dyes were synthesized with a pyridine moiety at the meso-position of the methine chain with a reaction yield of 89–93%. These dyes were prepared with high purity in a three-step synthesis as prospective pH-responsive probes. Each compound absorbed and fluoresced between 630 and 670 nm within the far-red range of the visible spectrum with high molar absorptivity values calculated between 134,910 L mol ^?1 cm^?1 and 163,816 L mol ^?1 cm^?1 and quantum yields between 9% and 17%. When a 20 μM solution of dye derivatives are exposed to hydrochloric acid in the range of pH 5–6 in ethanol, all the dyes experienced a blueshift in absorbance and emission as well as a decrease in spectral intensity that enables them to operate as pH probes. These properties were further validated via computational studies showing pyridine's influence on each molecules' spectral properties, and pathways for electron delocalization in the presence and absence of acid were proposed.

Full text of DOI:10.1016/j.dyepig.2021.109268

Dyes and Pigments 190 (2021) 109268
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: http://www.elsevier.com/locate/dyepig
Synthesis of pH-sensitive benzothiazole cyanine dye derivatives containing
a pyridine moiety at the meso position
,
,
b
,
,c,*
Wheeler R. Lovetta 1, Alaa Al Hamda 1, Stefanie Casaa, Maged Henarya
a Department of Chemistry, Georgia State University, USA
b Department of Chemistry, College of Science for Women University of Baghdad, Al-Jadiriyah, Iraq
c Center for Diagnostics and Therapeutics, Georgia State University, 145 Piedmont Avenue, Atlanta, GA, 30303, USA
A R T I C L E I N F O
A B S T R A C T
Keywords:
Four benzothiazole-derived pentamethine cyanine dyes were synthesized with a pyridine moiety at the meso-
position of the methine chain with a reaction yield of 89–93%. These dyes were prepared with high purity in
a three-step synthesis as prospective pH-responsive probes. Each compound absorbed and fluoresced between
630 and 670 nm within the far-red range of the visible spectrum with high molar absorptivity values calculated
between 134,910 L mol ꢀ 1 cmꢀ 1 and 163,816 L mol ꢀ 1 cmꢀ 1 and quantum yields between 9% and 17%. When a
Benzothiazole dye
Cyanine
Chemosensor
pH dye
Pyridine dye
Meso-substitution dye
20 μM solution of dye derivatives are exposed to hydrochloric acid in the range of pH 5–6 in ethanol, all the dyes
experienced a blueshift in absorbance and emission as well as a decrease in spectral intensity that enables them
to operate as pH probes. These properties were further validated via computational studies showing pyridine’s
influence on each molecules’ spectral properties, and pathways for electron delocalization in the presence and
absence of acid were proposed. © 2021 Elsevier Science. All rights reserved.
1. Introduction
are synthesized traditionally by using a fluorophore attached to a tar-
geting ligand that will aggregate to the desired biological area. This
The cyanine dye scaffold was initially discovered in the United
Kingdom in 1856 by Williams after he reacted a solution of quinoline
and 4-methyl-quinoline with pentyl iodide and ammonia, generating a
brilliant blue solid [1]. This compound was the first of a new general
scaffold. These structures can be simple and have an open terminal
amine group or a terminal heterocyclic group. Three classes of cyanine
dyes are shown in Fig. 1: streptocyanines, hemicyanines, and closed
chain cyanines; each of these classes differs by the end groups [2]. Much
of the classification of cyanine dyes is based on the length of the poly-
methine chain separating the nitrogen atoms (mono-, tri-, penta-,
hepta-) and the choice of heterocycles on the ends of the chain. These
heterocycles have a significant effect on the optical profiles of the dyes
compared to standard amines with electron-rich scaffolds such as pyr-
idinium, 4-pyrilium, 4-thiopyrilium, indolium, benzothiazolium, and
benz[c,d]indolium, each contributing to redshifts in the absorption and
emission spectra [3–6].
trend is generally seen in cancer-focused research by functionalizing
cyanine dyes with cancer-targeting peptides such as prostate-specific
antigen (PSA) or by dyes that aggregate and perfuse through tumor
tissue [10,11]. Work done by our group has included the functionali-
zation of cyanine dyes to target endocrine tissue [12]. Additionally, the
applications of these materials have also extended to the theragnostic
uses and imaging [10]. We recently reported our research on structure
inherent targeting in cyanine dyes as well with uptake seen in thyroid
and adrenal tissues [12]. Imaging with these compounds is advanta-
geous due to their operation in the near-infrared region (NIR) of the
electromagnetic spectrum. Biological tissue is mostly transparent to this
region between 650 and 800 nm (NIR window 1), and photons can go in
and out of samples being imaged without much relative interference
[15,16].
The benzothiazolium salt is particularly of interest for cyanine dye
development due to its planarity and orbital hybridization around the
sulfur atom contributing to redshifting compounds. Fluorophores using
this heterocycle were first synthesized via the condensation of 2-
The cyanine scaffold has been seen in a majority of applications such
as bioimaging and chemosensor design [7–14]. Bioimaging compounds
* Corresponding author. Department of Chemistry, Center for Diagnostics and Therapeutics, Georgia State University, 145 Piedmont Avenue, Atlanta, GA, 30303.
Tel.: +1 404 413 5566; Fax: +1 404 413 5505.
E-mail address: mhenary1@gsu.edu (M. Henary).
1
Co-first Author.
https://doi.org/10.1016/j.dyepig.2021.109268
Received 8 May 2020; Received in revised form 3 February 2021; Accepted 20 February 2021
Available online 8 March 2021
0143-7208/© 2021 Elsevier Ltd. All rights reserved.

W.R. Lovett et al.
Dyes and Pigments 190 (2021) 109268
methylbenzothiazole and N-pentyl-benzothiazolium iodide in a solution
of ammonia [17]. Benzothiazole-derived dyes have been used in various
biological imaging. They have shown effectiveness in binding to DNA
complexes [18–20] and interfere with the function of the arginine
methyltransferase enzyme involved in posttranslational protein modi-
fication. These compounds have also been used to bypass the
blood-brain barrier and image amyloid plaques in Alzheimer’s disease
models [21]. One of the applications of fluorescent dyes is pH respon-
siveness, which is indicated by some change in the spectroscopic qual-
ities of a dye in solution in response to the concentration of free protons
[22–24]. The benzothiazole heterocycle has been deployed in
pH-sensitive azo dyes. One literature example is composed of an amine
substituted benzothiazole moiety bonded to an aniline molecule via an
azo nitrogen double bond [25]. The uncharged asymmetrical dyes have
also been prepared from benzothiazole and quinoline bonded into a
monomethine bridge that does not exhibit fluorescence at pH 10, but at
high enough proton concentrations, induces a push-pull system from a
protonated nitrogen to a non-protonated nitrogen. This appears as a 3.5x
increase in absorbance as the band-gap is dramatically lowered enough
meso-substitution with pyridyl and nitrobenzene groups in benzothia-
zole with a small family of symmetrical dyes with meso substituted dyes
[34]. However, this paper’s extent did not address the capability for
these compounds to operate as pH-sensitive probes. Therefore, we report
herein a class of benzothiazole carbocyanine derived dyes with a
meso-pyridyl group, and its optical properties measured. This data in-
cludes the pH-dependent absorbance and fluorescence spectra and
computational modeling of the molecule’s electron orbital and electro-
static forces. The idea behind the scope of our work was that the pyridyl
group nitrogen would be a pH-responsive functional group that would
be directly conjugated into the fluorophore of the molecule and would
change the molecular orbital distribution and absorption band gap of the
dye once the nitrogen lone pair binds with a free proton (Fig. 2). The
fluorophore system energetics can be modeled in a simple, linear model
of an electron, the particle in a box model. From this simple mathe-
matical model, the only variable that is not constant is the path length
that the excited electron takes. Since this value is inversely related to the
system’s energy, the shorter the path length that HOMO electrons
resonate across, the higher the energy required to excite a fluorophore.
Due to the Planck relationship, higher energies have shorter wave-
lengths. This model can be applied to describe the energetics of the dyes
[35]. The synthesized compounds discussed below show a lower
absorbance value with the addition of an acid, and thus, these dyes must
be at higher energy than the non-protonated forms. As seen in Path A,
the protonated pathway is localized to the meso-substituted pyridine
and has a smaller valence molecular orbital than the non-protonated
pathway. Therefore, the protonated compound exists at a higher ener-
getic level and would be perceptible as a blueshift in absorbance or
emission. Path B demonstrates electron delocalization across the poly-
methine chain, similar to other cyanine dyes [36].
to absorb visible light ~500 nm [26]. The merocyanine,
a
poly-substituted cyanine derivative, has also been functionalized with
the benzothiazole heterocycle and a phenolic group pendant to the
polymethine chain that enables a pH response. Unlike the uncharged
monomethine dye, which engages in increased absorbance and fluo-
rescence in low pH, this benzothiazole-derived merocyanine experi-
ences a decrease in absorbance and fluorescence as well as a blueshift in
absorbance and fluorescence [27]. With symmetrical benzothiazole
dyes, pH response has been reported in tri- and pentamethine dyes via
pendant groups to each nitrogen that contains a carboxylic acid group.
In lower pH environments, this results in a decrease in absorbance and
fluorescence signal but no bathochromic shift due to the protonation
mechanism not directly involving the aromatic system of the fluo-
rophore [28]. This type of pH-responsive compound has been seen in
traditional symmetrical cyanines derived from indolium salt as well
[29].
2. Synthesis of benzothiazole pentamethine cyanine dye
derivatives
The synthetic procedure for pentamethine dyes 9–12 comprises
three separate steps (Scheme 1). In the first step, 4-methylpyridine was
treated to create the appropriately substituted pentamethine bridge.
This reaction consists of electronically priming dimethylformamide with
the oxidizer phosphorus oxychloride in trichloroethylene to promote
nucleophilic attack from the protonated 4-methylpyridine. This reaction
generates the dialdehyde intermediate 1, which is isolated by precipi-
tating the product in ice water and filtering. The second step in syn-
thesizing heterocyclic salts is the alkylation of the nitrogen in
benzothiazoles 2–4 using an alkyl halide. For compounds 5 and 6, the
alkyl halide used is methyl iodide. For compound 7, the alkylating agent
used is ethyl iodide, and for compound 8, (1-bromopropyl) benzene is
added. This reaction is performed in a solution of acetonitrile with a
slight excess of the alkylating agent to ensure the reaction is pushed to
completion. In the third step, linker 1 was then reacted with the corre-
sponding benzothiazolium salt 5–8 to yield the cyanine dyes 9–12.
The condensation of pentamethine dyes 9–12 requires a high boiling
solvent to achieve ideal temperature levels for the reaction to progress.
Therefore, the reaction mixture was carried out in butanol at 110 ◦C for
10–20 h. Triethylamine was used as a base to deprotonate the salts 5–8,
which were then allowed to react in the nucleophilic attack into the
linker 1. This reaction was performed with 2:1 equivalents of the desired
salt compared to the linker with a slight excess of salt to ensure
Dyes designed to measure pH usually have a functional group that is
pH responsive. Compounds with uncharged nitrogen in the polymethine
chain can be protonated and thus charged. One such group of dyes was
developed with sulfonate groups pendant to the heterocycles and an
additional feature of substitution of hydrogen in the meso position of the
polymethine chain with a bromine or chlorine atom [30]. This
meso-functionality is not related to the dye’s pH response, but other
research ventures have seen substitutions on the meso-position of the
polymethine chain that is responsive to pH conditions. The functional-
ization of a heptamethine cyanine with a pH-responsive pyrazole unit at
the meso position has been reported to blueshift with decreasing pH as
well as decrease fluorescence and absorbance intensity [31]. This meso
substitution can be accomplished by choosing a polymethine chain
precursor that can maintain its functionality through a Vilsmeier-Haack
formulation to form the electron-rich polymethine reagent [32]. Addi-
tional functionality to the meso carbon of the polymethine chain has
allowed for the attachment of targeting ligands, the intermolecular
binding of the dye to biological substrates, and anti-bacterial properties
[33].
The primary basis for this research is developing a chemosensor that
would be able to respond to pH and work within the far-right end of the
´
electromagnetic spectrum. The Kral group has performed research into
Fig. 1. Classifications of cCyanine dyes.
2

W.R. Lovett et al.
Dyes and Pigments 190 (2021) 109268
Fig. 2. Proposed alternate mechanism of protonated meso-pyridine pentamethine dye 9.
Scheme 1. Synthesis of benzothiazole meso-pyridine substituted carbocyanine dyes 9–12.
completion of the reaction. This reaction mixture was monitored by
thin-layer chromatography (TLC) to ensure the consumption of linker 1
as well as UV–Vis spectroscopy to confirm that the half-condensed dye is
pushed to reaction completion. Finally, the compound was worked up
and dried under reduced pressure. Each compound was isolated with
excellent yields of 89–93%.
(37% w/w, Fisher, Pittsburg, PA), sodium carbonate, sodium chloride,
DMSO (99% HPLC grade, Matrix Scientific, Elgin, SC), dimethylforma-
mide (99% HPLC), acetonitrile (99% HPLC), methanol (99% HPLC),
ethanol (97%), isopropanol (97%), butanol (99% HPLC), and pentanol
(99% HPLC) were all obtained commercially for use in spectroscopic
assays.
3. Experimental
3.2. General synthesis of benzothiazolium salts 5-8
3.1. Chemicals and synthesis
The benzothiazolium salts 5–8 were prepared based on published
procedures from our research group and others as described [18,37,38].
The benzothiazole precursors 1,5-dimethylbenzothiazole 2, 5-fluoro-1--
methylbenzothiazole 3, and 1-methylbenzothiazole 4 were individually
added to a round bottom flask with 25 mL of DMF and a slight molar
excess of the alkylating agent. For compounds 5 and 6, the alkyl halide
used is methyl iodide. For compound 7, reagent ethyl iodide is used, and
compound 8, (1-bromopropyl) benzene. The resulting solution was
allowed to react at 60 ◦C for 24 h as the resulting benzothiazolium salt
precipitated out of the solution. The formed salts were washed with cold
diethyl ether then dissolved in minimal methanol before recrystalliza-
tion from diethyl ether. Linker 1 is synthesized from a Vilsmeier-Haack
reaction. Dimethylformamide was added to a cold three-neck round
bottom flask equipped with rubber septa on the vestigial arms and a
condenser column on the center. This mixture was cooled to 0 ◦C in an
ice bath then phosphorous oxychloride was added dropwise over 15
min. The resulting Vilsmeier-Haack product was known to be formed
The four meso-pyridine pentamethine dye derivatives were synthe-
sized using commercially available reagents from Fisher Scientific,
Matrix Scientific, and Sigma Aldrich. All solvents used were HPLC grade
from Sigma Aldrich. The 1H NMR and 13C NMR spectra were recorded
on a Bruker Avance (400 MHz) spectrometer with DMSO‑d6 containing
tetramethylsilane (TMS) as an internal calibration standard (Cambridge
Isotope Laboratory, Tewksbury, MA). Processing of NMR was done using
Topsin 3.1 software. High-resolution accurate mass spectra (HRMS)
were obtained for further characterization using a Waters Q-TOF micro
(ESI-Q-TOF) mass spectrometer or Waters Micromass LCT TOF ES +
Premier Mass Spectrometer at the Mass Spectrometry Facility at Georgia
State University. The different reagents for each reaction are discussed
further below, with primarily Indocyanine green (ICG) being purchased
(Sigma-Aldrich, MO, USA) for its use as a reference standard for molar
absorptivity, quantum yield, and stability studies. Hydrochloric acid
3

W.R. Lovett et al.
Dyes and Pigments 190 (2021) 109268
when the solution turned a pale yellow. This reaction mixture was
slowly allowed to reach room temperature while 4-methylpyridine was
added dropwise over 30 min. The solution was then heated under reflux
for 4 h to ensure the full conversion of all the 4-methylpyridine before
being cooled back to room temperature. The crude slurry was then
poured over ice water to precipitate the final product 1. This compound
was filtered and washed with ice water then dried under reduced pres-
sure. This linker 1 was used without further purification for dye
condensation according to the previous procedure [38].
Hz, 2H), 7.77 (d, J = 8.4 Hz, 2H), 7.99 (m, 6H), 8.927 (d, J = 6.0 Hz,
2H). 13C NMR (100 MHz, DMSO‑d6): δ 21.53, 32.34, 46.54, 98.60,
114.43, 123.74, 126.09, 126.14, 126.55, 127.65, 128.56, 128.72,
128.85, 140.96, 141.14, 141.73, 143.90, 148.90, 166.51, 172.5. HRMS
(MS ES+) calcd for C42H38N3S2: m/z 648.2502 [M]+. Found: m/z
648.2501 [M]+.
Molar absorptivity was recorded for each synthesized dye using a
Varian Spectrophotometer (Varian Inc. Palo Alto, CA.). This instrument
was paired to Cary WinUV Scan Application v3.00 on a PC to visualize
and export data. Each sample was recorded at increasing molar con-
centrations (μM) in ethanol using standard polystyrene cuvettes with a
3.3. General synthesis of dye derivatives 9-12
path length of 1 cm. Each sample was prepared from a stock solution of
the dye. Preparation was done in dimethyl sulfoxide and measured on an
analytical balance to approximate a 1.0 mM solution of the dye. The
stock solutions were chosen to be prepared from DMSO due to its high
solubility for both polar and polyaromatic substances and for DMSO’s
ability to dissolve into ethanol effectively. Before recording their
spectra, each sample was sonicated for 30 min, then vortexed for 1 min
to ensure complete dissolution. Fluorescence emission for each synthe-
A mixture of individual benzothiazolium salt 5–8 (4 mmol), and
linker 1 (2 mmol) was added to a 50 mL solution of butanol in a two-
neck round-bottom flask equipped with a magnetic stir bar, a
condenser running vertically, and a rubber septum in the other neck. To
this reaction mixture, triethylamine (4.5 mmol) was added dropwise as
◦
the solution was brought to 110 C over an oil bath. As the reaction
progressed, a dark blue color was formed in situ. A drop of the reaction
mixture was analyzed via TLC for the consumption of linker 1, and
UV–Vis spectroscopy was used to monitor the conversion of a half dye
peak (~450–550 nm) to the pentamethine dye peak (~650 nm). Upon
reaction mixture completion, the round bottom flask was removed from
the heat and transferred to a rotary evaporation unit to remove butanol
under vacuum while it was still hot. The resulting crude residue was
then dissolved in a minimal amount of methanol (1–3 mL) then
precipitated from ethyl acetate or diethyl ether. This recrystallization
protocol was performed thrice before the resulting blue solid was puri-
fied over silica chromatography with a gradient solution of 2% methanol
and dichloromethane before 1H NMR, 13C NMR, and HRMS analysis
(Figs. S9–S20).
sized dye was recorded with
a Shimadzu RF-5301 Spectro-
fluorophotometer (Shimadzu Corporation Analytical Instruments
Division, Duisburg, F.R. Germany). This data was analyzed through a PC
operating the RF-5301 acquisition software to visualize and record data.
The experimental parameters for each fluorophore emission spectral
acquisition were as follows: slit width of excitation set to 5 nm, slit width
of emission set to 5 nm, sensitivity set to high, gain set to medium,
wavelength speed set to medium, a light source consistent with a 150 W
Xenon lamp bulb, and an excitation wavelength set to 10 nm blueshift
from the absorbance of each respective dye sample. Four-faced poly-
styrene fluorescence cuvettes with pathlengths of 1 cm were used for
each sample. Each sample was taken from getting its absorbance profile
recorded then diluted 20x to ensure an absorption value of 0.1 A U. to
minimize any inner filter effects from the concentration being too high.
The cuvettes were then sonicated for 30 min and vortexed for an addi-
tional minute to ensure full dissolution and dispersion. The data was
analyzed, visualized, and all corresponding calculations were performed
using Microsoft Excel (Microsoft Corporation, Redmond, WA.)
2-((1E,3Z,5Z)-5-(3,5-dimethylbenzo[d]thiazol-2(3H)-ylidene)-
3-(pyridin-4-yl)penta-1,3-dien-1-yl)-3,5-dimethylbenzo[d]thiazol-
3-ium iodide (9). (0.35 g, 92%). M.p. 249–251 ◦C. Vis/NIR abs, λmax
:
648 nm. 1H NMR (400 MHz, DMSO‑d6): δ 2.44 (s, 6H), 3.72 (s, 6H), 6.19
(d, J = 12.0 Hz, 2H), 7.29 (d, J = 8.0 Hz, 2H), 7.59 (s, 2H), 7.93 (m, 6H),
8.86 (d, J = 8.0 Hz, 2H). 13C NMR (100 MHz, DMSO‑d6): δ 21.57, 34.28,
98.69, 114.58, 114.72,112.74, 123.10, 127.22, 127.55, 138.94, 142.54,
142.73, 144.11, 148.25, 166.90. HRMS (MS ES+) calcd for C28H26N3S2+:
m/z 468.1563 [M]+. Found: m/z 468.1565 [M]+.
3.4. pH responsiveness data
All meso-pyridyl fluorophores were tested for their absorbance and
emission response to pH. Each sample was prepared in a polystyrene
fluorescence cuvette with a path length of 1 cm. These were then set to a
specific pH using a stock solution of 1 M HCl in ethanol or 1 M Na2CO3,
pure ethanol, and the requisite volume of stock solution of each dye to
5-fluoro-2-((1E,3Z,5Z)-5-(5-fluoro-3-methylbenzo[d]thiazol-2
(3H)-ylidene)-3-(pyridin-4-yl)penta-1,3-dien-1-yl)-3-methylbenzo
◦
[d]thiazol-3-ium iodide (10). (0.28 g, 89%). M.p. 225–229 C. Vis/
NIR abs, λmax: 652 nm. 1H NMR (400 MHz, DMSO‑d6): δ 3.71 (s, 6H),
6.14 (d, J = 14.0 Hz, 2H), 7.35 (t, J = 8.4 Hz, 2H), 7.76 (d, J = 9.2 Hz,
2H), 7.86 (d, J = 4.8 Hz, 2H), 8.09 (m, 4H), 8.91 (d, J = 5.6 Hz, 2H). 13C
NMR (100 MHz, DMSO‑d6): δ 34.40, 98.96, 101.98, 121.37, 124.99,
126.42, 143.71, 143.83, 148.74,161.98, 163.56, 163.72, 167.68. HRMS
(MS ES+) calcd for C26H20F2N3S2+: m/z 476.1061 [M]+. Found: m/z
476.1063 [M]+.
reach a total of 2.500 mL and 20 μM concentration of the dye. Each
sample was sonicated for 30 min then vortexed for 1 min before the
acquisition. The pH-resolved absorbance profiles were recorded in a
Varian Spectrophotometer linked to Cary WinUV Scan Application
v3.00 on a PC to visualize and export data. The pH-resolved emission
profiles for each dye were recorded in a Shimadzu RF-5301 Spectro-
fluorophotometer. This data was measured through a PC operating the
RF-5301 acquisition software to visualize and record the data. The
exported data was visualized and used to calculated response parame-
ters using Microsoft Excel.
3-ethyl-2-((1E,3Z,5Z)-5-(3-ethylbenzo[d]thiazol-2(3H)-yli-
dene)-3-(pyridin-4-yl)penta-1,3-dien-1-yl)benzo[d]thiazol-3-ium
iodide (11). (0.33 g, 93%). M.p. 223–226 ◦C. Vis/NIR abs, λmax: 647 nm.
1H NMR (400 MHz, DMSO‑d6): δ 1.21 (t, J = 7.2 Hz, 6H), 4.29 (d, J =
7.2 Hz, 4H), 6.09 (d, J = 12.8 Hz, 2H), 7.45 (t, J = 7.8 Hz, 2H), 7.59 (t, J
= 7.8 Hz, 2H), 7.71 (d, J = 5.6 Hz, 2H), 7.77 (d, J = 8.4 Hz, 2H), 8.00 (m,
4H), 8.82 (d, J = 5.6 Hz, 2H). 13C NMR (100 MHz, DMSO‑d6): δ 13.35,
42.26, 98.48, 114.32, 114.54, 123.61, 123.86, 126.26, 127.27, 127.43,
128.98, 141.51, 145.54, 149.29, 166.10. HRMS (MS ES+) calcd for
3.5. Computational methods
HOMO and LUMO orbital geometries were calculated using a
restricted hybrid Hartree Fock/Density Functional Theory (HF-DFT)
self-consistent fields (SCF) algorithm. These calculations were per-
formed using a Geometric Direct Minimization and a 6-31G* basis set
and B3LYP methodology via Spartan18. Additionally, QSAR parameters
such as polar surface area (PSA) and LogP were also calculated using the
same software. The data acquired from this software was then further
analyzed using Microsoft Excel.
C
28H26N3S+2 : m/z 468.1563 [M]+. Found: m/z 468.1568 [M]+.
3-(3-phenylpropyl)-2-((1E,3Z,5Z)-5-(3-(3-phenylpropyl)benzo
[d]thiazol-2(3H)-ylidene)-3-(pyridin-4-yl)penta-1,3-dien-1-yl)
benzo[d]thiazol-3-ium bromide (12). (0.31 g, 90%). M.p.
216–218 ◦C. Vis/NIR abs, λmax: 651 nm. 1H NMR (400 MHz, DMSO‑d6):
δ 1.96 (m, 4H), 2.65 (t, J = 7.6 Hz, 4H), 4.32 (t, J = 7.0 Hz, 4H), 6.10 (d,
J = 14 Hz, 2H), 7.19 (m, 10H), 7.47 (t, J = 7.6 Hz, 2H), 7.59 (t, J = 7.8
4

W.R. Lovett et al.
Dyes and Pigments 190 (2021) 109268
4. Results and discussion
intensity over a week in an ethanol solution, while dye 9 is comparable
to ICG’s performance. The trend of increasing dark-stability is consistent
with the increasing hydrophobicity, indicating that the increased
dark-stability may be due to the formation of hydrophobic aggregates in
a polar solvent. Aggregations should show less propensity to decompo-
sition as fewer molecules of the dye are exposed to the possible reactive
solvent solution or any other quenching molecules such as atomic oxy-
gen. When in ethanol and exposed to constant UV-irradiation, ICG de-
composes fully after 24 h. Comparatively, fluorophore derivatives 9–12
showed increased photostability averaging 18% percent. The derivatives
with the more hydrophobic substituents (11 and 12) showed the highest
photostability just as they showed the highest dark stability with
roughly a 15% and 10% respective reduction in the two dyes absorbance
intensity after 24 h of irradiation. These two parameters are indicative of
compounds’ practical utilization ability. Dark-stability is directly pro-
portional to any compound’s given shelf-life, and photostability is
related to the useable timescales in which the derivatives can operate as
fluorophores before becoming too decomposed to be accurate.
4.1. Optical properties
All the synthesized dyes, when dissolved in ethanol, gave an average
absorption wavelength near 650 nm with a range of 5 nm and an average
emission wavelength of 669 nm for an average Stokes shift of 19 nm
(Table 1). Absorption and emission peaks can also be visualized in the
spectrum for each dye in Fig. 3. These dyes operate in the far-red range
of the visible light spectrum, thus causing the dyes to have a brilliant
blue color. This observation is typical compared to other benzothiazole
pentamethine dyes that operate in the same wavelength [39].
As outlined in Table 1, each compound had an average Stokes shift of
19 nm. The energy loss between the absorbed and emitted photons
involved in the system is due to non-radiative intermolecular processes,
including bond vibrations. Using the Planck relationship, the wave-
length of absorption and emission can be converted into energy in
electron volts for an average in the prepared compounds 9–12 of 0.07 eV
as well as inverse centimeters at 16,120 cmꢀ 1. This data would be
interpreted on an IR spectrum as the vibrational motion of aromatic
4.2. pH responsiveness
–
–
–
–
C
C and C N bonds. Compounds 9 and 10 can provide insight into the
energetics of these non-radiative transitions by comparing the structural
differences between the two dyes is; compound 9 has two methyl groups
functionalized to C5 of the benzothiazole ring, and compound 10 has
two fluorine atoms. The fluorine atom in compound 10 slightly de-
creases the Stokes shift due to the reduced total number of C–H bonds
present in the molecule. The fewer C–H bonds mean that there are not as
many opportunities to radiate off energy. The molar absorptivity of a
compound gives an understanding of the amount of photon absorption
per molar equivalent of the compound. The synthesized compounds all
show the same trend in their molar absorptivity. For compounds 9–12,
this is an average of 152,375 L molꢀ 1 cmꢀ 1 with the highest molar ab-
sorptivity in compound 11, at 163,816 L molꢀ 1 cmꢀ 1, and the lowest in
the fluorinated dye 10 at 134,910 L molꢀ 1 cmꢀ 1. When comparing dyes
10 and 12’s molar absorptivity, the fluorine atom in dye 10 seems to
lower the absorptivity of the compound by 7%, indicating that some of
the electronic influence of the halogen decreases the fluorophore’s
ability to absorb light efficiently.
The capabilities of each dye were tested as a turn-off pH probe by
measuring the reduction in the signal that protonation induces to ab-
sorption and emission between pH 5 and 6. The pKa’s calculated for
each dye through a titration study averaged to be 5.32; therefore, this
range should give optimal feedback for incremental pH changes
(Fig. S2). The highest pKa is present in the fluorinated dye 10, and the
lowest pKa corresponds to the propylphenyl substituted dye 12, the
higher pKa could be due to the extra contribution of the fluorine’s full
electron shells donating into the group dye 10. The lower pKa could be
primarily due to poorer solvation of the more hydrophobic dye 12. The
change of absorbance and emission intensity in response to varying pH
was linearized to yield a response coefficient that is useful in comparing
the derivatives as pH sensors. All the prepared compounds showed a
similar trend in pH responses due to consistent linearization of the
absorbance regardless of the functional groups present in the dye
(Fig. S3). The exposure of acid results in a blueshift of the absorption
spectra of 8 nm and an approximately 20% decrease in the absorbance
maxima of compound 12, as shown in Fig. 4. This spectral change in
response to a chemical reaction (the binding of an acid) allows the
calculation of the pKa of that reaction. The spectral effects of dyes 9, 10,
and 11 are available in Figs. S4–S6.
The extra electron donation from the meso-pyridyl group causes an
on average 10 nm redshift in the compounds as well as providing the
dyes to operate as a chemosensor, as will be discussed. Some drawbacks
coincide with the addition of the group regarding specific optical
properties of the dye compared to previously prepared compounds in
our group. The quantum yields of these dyes were an average of 14%.
This value is lower than the typically seen quantum yields of benzo-
thiazoles without a meso-substitution. It can be attributed to increased
vibrational or rotational modes relaxation modes or unstable excited-
state molecular geometries due to the added bulky, electron-rich,
meso-substituent. Dye 10 has a marginally higher quantum yield than
its counterpart, dye 12, which can be attributed to the increased electron
density stabilizing the excited state of the fluorophore.
The pKa calculated for dye 9 was found to be 5.27, which is only
marginally higher than that of protonated pyridinium and similar to
other protonated aromatic amines. (Table 2). Each dye’s responsiveness
can be calculated by acquiring a higher pH resolved set of spectra,
normalizing the data, and linearly plot it against pH. The lead compound
9, for example, has the value of the slope of this linear plot for absor-
bance at 0.1712 (Fig. S4). This value is higher for fluorescence at 0.3710
due to a more significant relative decrease in absorbance. This difference
is most likely due to decreasing stability of the excited state due to the
redistributed valence molecular orbitals towards the pyridyl group (Path
B) as this group is more exposed to the solvent environment and more
likely to dissipate energy non-radiatively than if the energy had
remained in the polymethine core (Path A); electron delocalization of
the two pathways can be seen in Fig. 2.
Fluorophores 9–12 all showed markedly improved photostability
and dark-stability when compared to FDA approved imaging compound
ICG in ethanol [40]. As shown in Fig. S1A, ICG decomposes by 20%
percent after 7 days in the dark, while the new compounds 9–12 have an
average decomposition of 7%. The propylphenyl substituted dye 12
showed the highest degree of stability with only a 4% loss in absorbance
The combination of a blueshift in the absorbance and emission of dye
12 as well as a reduction in the intensity of the absorbance and emission
is due to the fluorophore of the protonated system being at a lower
energy level but having a less stabilized excited transition state (Fig. 4).
This blueshift behavior appears consistently in all dyes 9–12
(Figs. S4–S6). A blueshift of as much as 13 nm corresponds to the en-
tirety of the systems band gap lowered by 0.04 eV. The absorbance in-
tensity of the dyes was reduced by roughly 25%. It represents a
proportionate number of dyes in the solution that is no longer entering
an excited state compared to a neutral pH. The loss of nearly 60% of the
Table 1
Optical properties of meso-pyridyl pentamethine dyes 9–12.
Dye
λabs (nm)
λem (nm)
Stokes Shift (nm)
ε
(L mol ꢀ 1 cmꢀ 1
)
ɸf (%)
9
648
652
647
651
668
670
666
671
20
18
19
20
154,215
134,910
163,816
156,560
15
17
13
9
10
11
12
5

W.R. Lovett et al.
Dyes and Pigments 190 (2021) 109268
Fig. 3. Normalized absorption and emission spectrum of synthesized derivatives a). 9, b). 10, c). 11, and d) 12 in ethanol at 20
μ
M.
fluorescence intensity of the dye at low pH is also due to a fraction of the
dye not entering an excited state, but the 35% difference between the
absorbance and emission is due to the excited state not having the same
stability as the protonated form of the dye. The higher the linear
response coefficient, the better the dye operates as a chemosensor. Dyes
10 and 11 perform similarly with absorption and emission response
coefficients only 0.013 and 0.005 units apart, but dyes 11 and 12 are
starkly better. Dyes 11 and 12 are the more hydrophobic dyes with the
highest LogP values and give linear absorption coefficients of 0.2535
and 0.4179, respectively, a 50% and 150% increase in responsivity.
Emission is even more responsive with values of 0.3710 and 0.3661 for
dyes 9 and 10 and values of 0.4307 and 0.7105 for dyes 11 and 12.
benzothiazole dye 12 nm when compared to the standard methyl pen-
tamethine benzothiazole dye with absorption at 650 nm for dye 12 [27].
When dye 12 is protonated, there is no noticeable change in the HOMO
with the same distribution across the benzothiazole rings and poly-
methine chain. There is, however, a significant change in the molecular
orbital geometry of the LUMO, with most of the orbital pulling towards
the pyridyl group’s newly acquired proton. There is, however, a signif-
icant change in the molecular orbital geometry of the LUMO, with much
of the orbital pulling towards the pyridyl group’s newly acquired proton.
Note that while the LUMO is primarily localized on the pyridine group,
there is still traces of the molecular orbital extending down into the
polymethine chain. This observation speaks to the hypothesis that the
capture of a proton will effectively shift the electronic resonance path-
ways towards that acquired proton and shorten the size of the molecular
orbital as well as raise its energy. The surface area and volumes of
pentamethine dyes 9–12 are all within the ranges of viable cell perme-
ability. The large propyl-phenyl arms of dye 12 increase the volume of
the compounds close to 700 Å2. The average polar surface area for each
dye is 8.38 Å2 and increases 3.65 Å2 upon protonation as the newly
cationic pyridyl nitrogen accounts for more polar space on the molecule.
The dipole moment is primarily used to understand the relationship
between the dye, their solvent environments, and oscillating electro-
magnetic fields. The fluorine atoms present in dye 7 lower the dipole
moment compared to dye 9 due to the electron density pulling the
overall polarity to the opposite of the molecule as the pyridyl group. The
protonation of each dye causes an average shift in the dipole moment by
an average of 5.04 debye though this is as high as 7.55 debye in dye 10
and 0.32 debye in dye 12. Benzathiozole pentamethine dye 10 experi-
ences this high shift because the electron of the fluorine atoms is easily
overcome by the newly shifted electron density towards the newly ac-
quired proton. The computed surface geometries of dyes 9–11 are pre-
sented in Figs. S7 and S8, and dye 12 is featured in Fig. 5.
4.3. Computational data
Physical parameters for the synthesized dyes 9–12 were calculated to
illuminate the inner electronic workings of each fluorophore system
(Table 3). Qualitative Structural Activity Relationship (QSAR) parame-
ters such as molecular surface area and volume, polar surface area
(PSA), dipole moment, and LogP/LogD are all numerical descriptors
used by various algorithms to model and match different drug-like
molecules to possible pharmacodynamic and pharmacokinetic abilities
and related compounds. These values help contextualize the compounds
within the greater library of synthesized drug-like dyes and insight into
how they would act in vivo. However, a more extensive value set is
needed for an accurate pharmacokinetic evaluation.
All of the prepared compounds showed a consistent trend in their
QSAR parameters when exposed to acid. There was a consistent increase
in the polar surface area and dipole moment in all compounds 9–12 and
a consistent decrease in the LogP/LogD value of each molecule. The
color’s coded structures in Fig. 5 correspond to the geometric locations
of the HOMO and LUMO modeled in red and blue and as a red to blue
heat map corresponding to the relatively electrical density for dye 12.
The modeled HOMO and LUMO orbitals for compound 12 showed a
distribution along with the polymethine bridge and benzothiazole rings.
The molecular orbitals do not seem to spatially interact with the pyridyl
group at the meso-position. However, the electron-donating effects of
this substituent are still seen, with it providing help to redshift the
5. Conclusion
Four meso-pyridyl benzothiazole pentamethine dyes were synthe-
sized with excellent yields. All the synthesized derivatives absorbed
between 645 nm and 660 nm, within the far-red range, had high molar
6

W.R. Lovett et al.
Dyes and Pigments 190 (2021) 109268
Fig. 4. Spectral effects of pH on dye 12 in EtOH at 20
μ
M. a) pH response in absorption, b) emission, c) linear response coefficient of absorbance and d) emission.
Table 2
pH response properties of dyes 9–12 in ethanol at 20 μM.
Dye
pKa
Absorbance Bathochromic
Shift
Absorbance
Reduction
Absorbance Response
Coefficient
Emission Bathochromic
Shift
Emission
Emission Response
Coefficient
Reduction
9
5.27
5.49
5.32
5.19
8 nm
18%
18%
24%
32%
0.1712
0.1844
0.2535
0.4179
7 nm
8 nm
9 nm
5 nm
36%
34%
37%
57%
0.3710
0.3661
0.4307
0.7105
10
11
12
9 nm
12 nm
13 nm
Table 3
Spartan18 calculated QSAR properties of pentamethine dyes 9–12 and their protonated equivalents.
Dye
Area (Å2)
Volume (Å3)
PSA (Å2)
Dipole (debye)
LogP/LogD
HOMO (eV)
LUMO (eV)
ΔE (eV)
9
506.82
506.41
467.17
474.62
499.39
503.31
698.24
705.93
487.65
490.73
456.59
461.91
488.08
491.59
692.2
8.483
6.35
4.25
3.88
2.82
2.46
4.57
4.2
ꢀ 7.49
ꢀ 9.86
ꢀ 7.74
ꢀ 10.16
ꢀ 7.55
ꢀ 9.96
ꢀ 7.49
ꢀ 9.8
ꢀ 5.21
ꢀ 8.16
ꢀ 5.40
ꢀ 8.29
ꢀ 5.27
ꢀ 8.20
ꢀ 5.20
ꢀ 8.09
ꢀ 2.28
ꢀ 1.70
ꢀ 2.34
ꢀ 1.87
ꢀ 2.28
ꢀ 1.76
ꢀ 2.29
ꢀ 1.71
9Hþ
10
12.145
8.484
12.72
4.68
10Hþ
11
12.122
8.246
12.23
6.24
11Hþ
12
11.907
8.289
12.16
23.81
24.13
7.16
6.79
12Hþ
696.-6
11.909
absorptivity and quantum yields ranging from 9% to 17%. These com-
pounds were also shown to have much higher dark-stability and pho-
tostability when compared to FDA-approved fluorophore, indocyanine
green (ICG), with only 4% loss of compounds in the dark and a 10% loss
in UV-irradiation. The synthesized derivatives were also responsive to
pH changes with a resulting blueshift in the absorbance and emission
7

W.R. Lovett et al.
Dyes and Pigments 190 (2021) 109268
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgments
The authors thank the Department of Chemistry at Georgia State
University for supporting M.Sc for WL and SC. MH wishes to thank the
Brains and Behavior Seed Grant, the Health Innovation Platform Grant,
and the Georgia Research Alliance Venture Grant Phase 1, and Baghdad
University for the generous support. MH would like to thank Mr. Chia
Hung Chen for assitance in collecting NMR spectra.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.dyepig.2021.109268.
References
[1] Williams CG. XXVI.—researches on chinoline and its homologues. Trans R Soc
Edinb 1857;21(3):377–401.
Fig. 5. Computed surface geometries of valence molecular orbitals and electron
density of dye 12 using Spartan18.
[2] Ilina K, Henary M. Cyanine dyes containing quinoline moieties: history, synthesis,
optical properties, and applications. Chem Eur J 2020;26:1–20.
[3] Kovalska VB, Tokar VP, Losytskyy MY, Deligeorgiev T, Vassilev A, Gadjev N, et al.
Studies of monomeric and homodimeric oxazolo[4,5-b]pyridinium cyanine dyes as
fluorescent probes for nucleic acids visualization. J Biochem Biophys Methods
2006;68(3):155–65.
spectra and a decrease in the spectral intensities. All compounds shared
the same trend in their responses to acid, but the hydrophobic com-
pound 12 showed the highest response coefficient and the most signif-
icant change in absorbance values. Computational studies performed on
the dyes confirmed the HOMO and LUMO changes that correspond to
our proposed alternative resonance pathways. Protonation of these dyes
results in smaller valence orbitals localized around the pyridine moiety
that reside at a higher energy level for the molecule’s band gap. Cal-
culations showed that protonation of the dyes increased LogD/LogP,
polar surface area, and dipole moment. All of these are parameters that
describe the responsiveness of the synthesized compounds to solvent
environment conditions, indicating that the excited state of the pro-
tonated dyes is more responsive to intermolecular forces. It is a combi-
nation of these increased intermolecular interactions and the modified
valence molecular orbitals that result in the blueshifted and reduced
absorption and emission spectra. Continued exploration of dyes with the
meso-pyridine moiety would provide an insight into the expansion of
other cyanine pH-responsive dyes for their future applications.
[4] Yarmoluk SM, Kostenko AM, Dubey IY. Interaction of cyanine dyes with nucleic
acids. Part 19: new method for the covalent labeling of oligonucleotides with
pyrylium cyanine dyes. Bioorg Med Chem Lett 2000;10(19):2201–4.
[5] Kamimura M, Takahiro S, Yoshida M, Hashimoto Y, Fukushima R, Soga K. Over-
1000 nm near-infrared fluorescent biodegradable polymer nanoparticles for deep
tissue in vivo imaging in the second biological window. Polym J 2017;49(12):
799–803.
[6] Soriano E, Holder C, Levitz A, Henary M. Benz[c,d]indolium-containing
monomethine cyanine dyes: synthesis and photophysical properties. Molecules
2015;21(1):E23–.
[7] Frankaer CG, Sorensen TJ. A unified approach for investigating chemosensor
properties - dynamic characteristics. Analyst 2019;144(7):2208–25.
[8] Mayer M, Baeumner AJ. A megatrend challenging analytical chemistry: biosensor
and chemosensor concepts ready for the internet of things. Chem Rev 2019;119
(13):7996–8027.
[9] Ptaszek M. Rational design of fluorophores for in vivo applications. Prog Mol Biol
Transl Sci 2013;113:59–108.
[10] Shi C, Wu JB, Pan D. Review on near-infrared heptamethine cyanine dyes as
theranostic agents for tumor imaging, targeting, and photodynamic therapy.
J Biomed Optic 2016;21(5):50901.
[11] Wu J, Pan D, Chung LW. Near-infrared fluorescence and nuclear imaging and
targeting of prostate cancer. Transl Androl Urol 2013;2(3):254–64.
[12] Owens EA, Henary M, El Fakhri G, Choi HS. Tissue-specific near-infrared
fluorescence imaging. Acc Chem Res 2016;49(9):1731–40.
Consent for publication
[13] Njiojob CN, Owens EA, Narayana L, Hyun H, Choi HS, Henary M. Tailored near-
infrared contrast agents for image guided surgery. J Med Chem 2015;58(6):
2845–54.
Not applicable.
[14] Owens EA, Hyun H, Tawney JG, Choi HS, Henary M. Correlating molecular
character of NIR imaging agents with tissue-specific uptake. J Med Chem 2015;58
(10):4348–56.
Ethics approval and consent to participate
[15] Zhu S, Tian R, Antaris AL, Chen X, Dai H. Near-infrared-II molecular dyes for
cancer imaging and surgery. Adv Mater 2019;31(24):e1900321.
[16] Haque A, Faizi MSH, Rather JA, Khan MS. Next generation NIR fluorophores for
tumor imaging and fluorescence-guided surgery: a review. Bioorg Med Chem 2017;
25(7):2017–34.
MH designed the research, WL and AH performed the experiments,
MH, WL. SC wrote the manuscript, and all authors interpreted the data.
Funding
[17] Mills W. The cyanine dyes. Part IV. Cyanine dyes of the benzothiazole series.
J Chem Soc Trans 1922;121:455–66.
This research was funded by the Brains and Behavior Seed Grant, the
Health Innovation Platform Grant, and the Georgia Research Alliance
Venture Grant Phase 1 and by the support provided to Dr. Al Hamd from
Baghdad University.
[18] Owens EA, Huynh HT, Stroeva EM, Barman A, Ziabrev K, Paul A, et al. Second
generation G-quadruplex stabilizing trimethine cyanines. Bioconjugate Chem
2019;30(10):2647–63.
[19] Turaev AV, Tsvetkov VB, Tankevich MV, Smirnov IP, Aralov AV, Pozmogova GE,
et al. Benzothiazole-based cyanines as fluorescent “light-up” probes for duplex and
quadruplex DNA. Biochimie 2019;162:216–28.
[20] Lu Y-J, Wang Z-Y, Hu D-P, Deng Q, Huang B-H, Fang Y-X, et al. Benzothiazole-
substituted benzofuroquinolinium dyes as new fluorescent probes for G-quadruplex
DNA. Dyes Pigments 2015;122:94–102.
Availability of data and methods
[21] Handen BL, Cohen AD, Channamalappa U, Bulova P, Cannon SA, Cohen WI, et al.
Imaging brain amyloid in nondemented young adults with Down syndrome using
Pittsburgh compound B. Alzheimer’s Dementia 2012;8(6):496–501.
[22] Salim MM, Owens EA, Gao T, Lee JH, Hyun H, Choi HS, et al. Hydroxylated near-
infrared BODIPY fluorophores as intracellular pH sensors. Analyst 2014;139(19):
4862–73.
The data pertaining to the conclusions presented in the work are
available to the public through Elsevier Publications’ digital services.
Declaration of competing interest
The authors declare that they have no known competing financial
8

W.R. Lovett et al.
Dyes and Pigments 190 (2021) 109268
[23] Yang MX, Tang Q, Yang M, Wang Q, Tao Z, Xiao X, et al. pH-stimulus response dye-
cucurbituril sensor for amino acids in aqueous solution. Spectrochim Acta Mol
Biomol Spectrosc 2020;230:118076.
[32] Levitz A, Marmarchi F, Henary M. Introduction of various substitutions to the
methine bridge of heptamethine cyanine dyes via substituted dianil linkers.
Photochem Photobiol Sci 2018;17(10):1409–16.
[24] Nath N, Godat B, Zimprich C, Dwight SJ, Corona C, McDougall M, et al.
Homogeneous plate based antibody internalization assay using pH sensor
fluorescent dye. J Immunol Methods 2016;431:11–21.
[33] Abd El-Aal RM, Younis M. Synthesis and antimicrobial activity of meso-substituted
polymethine cyanine dyes. Bioorg Chem 2004;32(4):193–210.
´
ˇ
´ˇ
´
´
´
ˇ
[34] Rimpelova Silvie, Bríza Tomas, Kralova Jarmila, Zaruba Kamil, Zdenek Kejík,
´
ˇ
´
´
´ˇ
´
[25] Coelho FL, Braga CdA, Zanotto GM, Gil ES, Campo LF, Gonçalves PFB, et al. Low
Císarova Ivana, Martasek Pavel, Ruml Tomas, Kral Vladimír. Rational design of
chemical ligands for selective mitochondrial targeting. Bioconjugate Chem 2013;
24(9):1445–54.
pH optical sensor based on benzothiazole azo dyes. Sensor Actuator B Chem 2018;
259:514–25.
[26] Nourmohammadian F, Gholami MD. Two novel push-pull series of benzothiazole-
based dyes: synthesis and characterization. Helv Chim Acta 2012;95(9):1548–55.
[35] Albrecht T, Kornyshev A, Bjornholm T. Charge transport in nanoscale junctions.
J Phys Condens Matter 2008;20(37):370301.
´
[27] Coelho FL, da Costa Duarte R, de Avila Braga C, Toldo JM, Bruno Gonçalves PF, da
[36] Shindy HA. Fundamentals in the chemistry of cyanine dyes: a review. Dyes
Pigments 2017;145:505–13.
Silveira Santos F, et al. Benzothiazole merocyanine dyes as middle pH optical
sensors. Dyes Pigments 2020;176:108193.
[37] Hu Hao, Eric A, Owens, Hairui Su, Leilei Yan, Andrew Levitz, Xinyang Zhao,
Maged Henary, Yujun George Zheng. Exploration of cyanine compounds as
selective inhibitors of protein arginine methyltransferases: synthesis and biological
evaluation. J Med Chem 2015;58(3):1228–43.
[28] Ikeda S, Okamoto A. pH-dependent fluorescence of uncharged benzothiazole-based
dyes binding to DNA. Photochem Photobiol Sci 2007;6(11):1197–201.
`
[29] Mazieres MR, Duprat C, Wolf JG, Roshal AD. pH dependent spectral properties and
electronic structure of benzothiazol containing cyanine dyes. Dyes Pigments 2009;
80(3):355–60.
[38] Amaresh RR, Perumal PT. A novel route to the synthesis of 3-pyridine
carboxaldehydes by vilsmeier reagent. Synth Commun 2000;30(13):2269–74.
[39] Briza Tk V. Pentamethinium fluorescent probes: the impact of molecular structure
on photophysical properties and subscellular localization. Dyes Pigments 2014;
107:51–9.
[30] Hilderbrand SA, Weissleder R. Optimized pH-responsive cyanine fluorochromes for
detection of acidic environments. Chem Commun 2007;(26):2747–9.
[31] Lee H, Berezin MY, Tang R, Zhegalova N, Achilefu S. Pyrazole-substituted near-
infrared cyanine dyes exhibit pH-dependent fluorescence lifetime properties.
Photochem Photobiol 2013;89(2):326–31.
[40] Holzer W, Mauerer M, Penzkofer A, Szeimies RM, Abels C, Landthaler M, et al.
Photostability and thermal stability of indocyanine green. J Photochem Photobiol
B Biol 1998;47(2):155–64.
9