A ratiometric near-infrared fluorescent probe based on a novel reactive cyanine platform for mitochondrial pH detection

Article abstract of DOI:10.1039/d1tb00643f

A near-infrared reactive cyanine platform (probeA) was prepared by condensation of 9-chloro-1,2,3,4-tetrahydro-10-methyl-acridinium iodide with Fisher's aldehyde. A near-infrared fluorescent probe (probeB) was prepared by modifying a reactive chlorine atom of probeAwithtert-butyl(2-aminoethyl)carbamate through a substitution reaction. The deprotection of the Boc group of probeBwas achieved under an acidic condition, affording an amine-functionalized cyanine dye (probeC). A near-infrared ratiometric fluorescent probe (probeD) for mitochondrial pH detection was synthesized by conjugating a FRET coumarin donor to a FRET cyanine acceptor (probeC) through an amide bond connection. ProbeAshows low fluorescence of 2% due to an electron-withdrawing chlorine atom, while probesB-Ddisplay high fluorescence quantum yields of 60%, 32%, and 35% in aqueous solutions containing 10% ethanol, respectively. ProbesB-Dshow strong fluorescence with push-pull molecular structures in neutral and basic pH conditions. However, protonation of the probe's second amine at the 9-position under acidic condition disrupts the push-pull feature of the probes, resulting in fluorescence quenching of the new cyanine fluorophores. The probes can selectively stain mitochondria, while probeDwas employed to detect pH changes in HeLa cells andDrosophila melanogasterfirst-instar larvae.

Full text of DOI:10.1039/d1tb00643f

Journal of
Materials Chemistry B
View Article Online
View Journal | View Issue
PAPER
A ratiometric near-infrared fluorescent probe
based on a novel reactive cyanine platform for
mitochondrial pH detection†
Cite this: J. Mater. Chem. B, 2021,
, 5150
9
a
a
a
a
b
Shulin Wan, Shuai Xia, Jerry Medford, Emma Durocher, Tessa E. Steenwinkel,
a
a
a
b
Lexi Rule, Yibin Zhang,
Rudy L. Luck, * Thomas Werner* and
a
Haiying Liu
*
A near-infrared reactive cyanine platform (probe A) was prepared by condensation of 9-chloro-1,2,3,4-
tetrahydro-10-methyl-acridinium iodide with Fisher’s aldehyde. A near-infrared fluorescent probe (probe B)
was prepared by modifying a reactive chlorine atom of probe A with tert-butyl(2-aminoethyl)carbamate
through a substitution reaction. The deprotection of the Boc group of probe B was achieved under an
acidic condition, affording an amine-functionalized cyanine dye (probe C). A near-infrared ratiometric
fluorescent probe (probe D) for mitochondrial pH detection was synthesized by conjugating a FRET
coumarin donor to a FRET cyanine acceptor (probe C) through an amide bond connection. Probe A shows
low fluorescence of 2% due to an electron-withdrawing chlorine atom, while probes B–D display high
fluorescence quantum yields of 60%, 32%, and 35% in aqueous solutions containing 10% ethanol, respec-
tively. Probes B–D show strong fluorescence with push–pull molecular structures in neutral and basic pH
conditions. However, protonation of the probe’s second amine at the 9-position under acidic condition
disrupts the push–pull feature of the probes, resulting in fluorescence quenching of the new cyanine fluoro-
phores. The probes can selectively stain mitochondria, while probe D was employed to detect pH changes
in HeLa cells and Drosophila melanogaster first-instar larvae.
Received 25th March 2021,
Accepted 6th June 2021
DOI: 10.1039/d1tb00643f
rsc.li/materials-b
dye, it is difficult to synthesize cyanine dyes with on–off switches.
Recently, near-infrared hemicyanine dyes bearing spironolactone,
Introduction
Cyanine dyes are a unique group of near-infrared fluorophores, spirolactam, and oxazolidine switches have also been developed
consisting of pentamethine or heptamethine backbones with two for pH sensing and the detection of reactive nitrogen, oxygen, and
1–7
4,11–18
nitrogen-containing heterocycles.
The U.S. Food and Drug sulfur species.
Most neutral fluorescent probes based on
Administration has approved indocyanine green for clinical appli- spirolactam switches have been used to detect pH changes in
4,8,9
12,14,15,19–25
cations due to its appealing near-infrared optical property.
lysosomes,
while positively charged fluorescent probes
Many fluorescent probes have been developed by modifying a with oxazolidine and spiropyran switches were employed to deter-
18,26
central active chlorine atom of traditional cyanine dyes with mine pH changes in mitochondria.
Despite significant progress
amine-, thiol- and boron ester-functionalized compounds through in cyanine dye research, there are few reports on new cyanine
substitution reactions for a variety of sensing and imaging platforms with reactive sites for on–off switchable probes. Therefore,
1–5,7,10
applications.
Positively charged cyanine dyes have been used it is very important to develop new near-infrared cyanine platforms
to specifically target mitochondria. However, due to the high-lying with unique chemical structures and easy post-functionalization for
occupied molecular orbital (HOMO) energy levels of the cyanine sensing and imaging applications.
In this paper, we developed a novel near-infrared fluorescent
a
cyanine platform (probe A) with a reactive chlorine atom by
condensing 9-chloro-1,2,3,4-tetrahydro-10-methyl-acridinium
iodide with Fisher’s aldehyde (Chart 1). We further modified
the reactive chlorine atom on probe A with tert-butyl(2-amino-
ethyl)carbamate by a substitution reaction, affording a secondary
amine-functionalized pH sensitively ‘‘on–off’ switchable cyanine
dye (probe B). The Boc group in probe B was then removed,
resulting in an amine-terminating cyanine dye (probe C). In order
Department of Chemistry, Michigan Technological University, Houghton,
MI 49931, USA. E-mail: hyliu@mtu.edu, rluck@mtu.edu
b
Department of Biological Sciences, Michigan Technological University,
Houghton, MI 49931, USA. E-mail: twerner@mtu.edu
†
Electronic supplementary information (ESI) available: It includes NMR spectra
of probes A–D, mass spectra of the probes in neutral and acidic conditions,
calculations of the probe fluorescence quantum yields, pK values, the probes’
a
reversible fluorescence responses to pH changes, and computational results. See
DOI: 10.1039/d1tb00643f
5150
|
J. Mater. Chem. B, 2021, 9, 5150–5161
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry B
Chart 1 Reactive cyanine dye platform (probe A) and its derivatives (probes B, C, and D) for pH detection in mitochondria.
to achieve ratiometric imaging of pH changes, we introduced a transfer from the donor to the acceptor since the acceptor
coumarin moiety as a F ¨o rster Resonance Energy Transfer (FRET) absorption has significant overlap with the donor emission.
donor to probe C through an amide bond connection to prepare a Ratiometric fluorescent probes with self-calibration capabilities
2
7–30
near-infrared cyanine dye, probe D.
FRET from a fluorophore can effectively overcome systematic errors of intensity-based only
donor to a fluorophore acceptor is often used to construct fluorescent probes caused by excitation light fluctuations, probe
2
7,29–31
ratiometric fluorescent probes.
significantly on the connection distance between the donor and locations.
acceptor and the overlap degree between the donor emission and C–D show pH-sensitive responses (Chart 1), which are completely
FRET efficiency depends concentration variations, and different cellular compartmental
2
7,29
Probe A is insensitive to pH changes, while probes
27,29–31
acceptor absorption.
In probe D, the coumarin donor was different from the acid-activated enhanced fluorescence of
connected to the cyanine acceptor through a very short two- fluorescent probes based on spirolactam, spiropyran, and
1
2,14,15,18–26
methylene tethered spacer to achieve high efficiency of energy oxazolidine switches.
Gradual pH decreases lead to
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. B, 2021, 9, 5150–5161 | 5151

View Article Online
Journal of Materials Chemistry B
Paper
Scheme 1 Synthetic approach to prepare a near-infrared reactive cyanine dye backbone (probe A) and its derivatives (probes B–D).
fluorescence decreases of near-infrared fluorescence of probes 6.98 (d, J = 7.5 Hz, 1H), 5.77 (d, J = 13.0 Hz, 1H), 4.48 (s, 3H),
B–C. Probe D shows ratiometric responses to pH decreases with 3.53 (s, 3H), 2.95 (t, J = 5.8 Hz, 2H), 2.81 (t, J = 6.2 Hz, 2H), 1.97
1
3
fluorescence decreases of the cyanine acceptor and fluorescence (t, J = 6.0 Hz, 2H), 1.58 (s, 6H). C NMR (101 MHz, CDCl ) d
3
increases of the coumarin donor. Ratiometric responses of 174.39, 170.82, 158.19, 145.56, 141.27, 140.90, 133.95, 133.76,
probe D to pH arise from a high-efficiency energy transfer rate 128.55, 127.59, 125.65, 124.16, 123.91, 122.30, 120.10, 116.73,
of 97.7% from the coumarin donor to the cyanine acceptor. The 110.99, 109.72, 104.94, 96.36, 49.19, 48.74, 47.93, 29.89, 28.98,
+
fluorescence decreases of the cyanine fluorophores arise from 28.40, 28.14, 20.77. MS (ESI): calcd for [C34
the disruption of push–pull molecular structures upon acidic-pH found 415.1927.
protonation of the secondary amine in the 9-position of the probes
H
43
N
4
O
2
] 415.1936,
(Chart 1). The positively charged probes A–D can specifically target
Synthesis of probe B
mitochondria through electrostatic interactions with the negatively
charged mitochondrial matrix. Probe D shows sensitive fluores-
cence responses to pH changes in live cells and Drosophila
melanogaster (D. melanogaster) first-instar larvae.
Probe A (200 mg, 0.48 mmol) was added to 6 mL dimethylforma-
mide (DMF) containing boc-ethylenediamine (153 mg, 0.96 mmol).
This mixture was stirred at 50 1C for 2 h, whereupon it was poured
into 100 mL of dichloromethane in a separatory funnel and
washed with 50 mL saturated saline solution three times. The
2 4
organic layer was collected, dried with anhydrous Na SO , and
condensed under reduced pressure. The residue was subjected to
column chromatography (packed with silica gel, particle size,
Materials and instruments
Details of regents and instruments are documented in the ESI.†
0
.063–0.2 mm Merck) and eluted with a mixture of dichloro-
Synthesis of probe A
methane and MeOH at ratios from 70 : 1 to 50 : 1, yielding the
Fisher’s aldehyde (5) (112 mg, 0.56 mmol) was added to 6 mL of pure probe B in the form of red crystals (207 mg with a reaction
12,23
1
acetic anhydride containing compound 4 (200 mg, 0.56 mmol).
3
yield of 80%). H NMR (400 MHz, CDCl ) d 8.56 (d, J = 10.1 Hz, 1H),
The mixture was then stirred at 50 1C for 2 h, Scheme 1. The 7.85 (d, J = 4.2 Hz, 1H), 7.63–7.49 (m, 2H), 7.19 (dd, J = 17.5, 7.7 Hz,
reaction mixture was evaporated under vacuum. The crude pro- 2H), 6.99–6.85 (m, 2H), 6.77 (d, J = 8.3 Hz, 1H), 5.95 (s, 1H), 5.46 (d,
duct was purified through column chromatography (packed with J = 12.6 Hz, 1H), 4.16 (s, 3H), 4.10–4.02 (m, 2H), 3.72–3.61 (m, 2H),
silica gel, particle size, 0.063–0.2 mm Merck) and eluted with a 3.27 (s, 3H), 2.94–2.83 (m, 2H), 2.76–2.66 (m, 2H), 1.99–1.86 (m,
13
mixture of dichloromethane and MeOH at ratios from 100 : 1 to 2H), 1.54 (s, 6H), 1.37 (s, 9H). C NMR (101 MHz, CDCl ) d 163.92,
3
3
0 : 1, yielding pure probe A in the form of blue crystals (233 mg 158.02, 156.80, 153.62, 144.22, 141.09, 139.12, 136.26, 133.04,
1
with a reaction yield of 70%). H NMR (400 MHz, CDCl ) d 8.31 128.27, 125.76, 121.90, 121.54, 118.86, 118.32, 118.05, 115.94,
3
(
7
d, J = 8.6 Hz, 1H), 8.15 (d, J = 9.7 Hz, 1H), 7.97–7.86 (m, 1H), 107.49, 92.12, 79.97, 50.92, 46.70, 43.43, 40.89, 29.99, 29.71,
+
.60 (t, J = 7.7 Hz, 1H), 7.43 (d, J = 13.3 Hz, 1H), 7.32 (d, J = 28.70, 26.83, 22.74. MS (ESI): calcd for [C34
43 4 2
H N O ] 539.3481,
7.2 Hz, 1H), 7.24 (d, J = 7.5 Hz, 1H), 7.11–7.03 (m, 1H), found 539.3364.
5152
|
J. Mater. Chem. B, 2021, 9, 5150–5161
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry B
Synthesis of probe C
5 mM probe A, B, or C with commercially available mitochondria-
targeting MitoView Blue (5 mM) for 15 minutes. In order to avoid
fluorescence overlap between probe D and MitoView Blue, we
incubated HeLa cells with probe D with a mitochondria-targeting
near-infrared cyanine dye (IR-780) (5 mM) for the colocalization
experiment. A 0.2% DMSO concentration was used during the cell
incubation to facilitate the dye solubility in the cell culture
medium. The fluorescence imaging was conducted after the cells
were washed twice with the cell culture medium, and added with
the cell culture medium. In order to detect intracellular pH
changes in live cells, we incubated HeLa cells with 5 mM probe
D in the cell culture medium for 15 minutes, washed with the
medium twice, and then further incubated in different pH buffer
After a solution of probe B (200 mg, 0.37 mmol) in 6 mL
dichloromethane was cooled by being surrounded by a mixture
of ice and water, 6 mL of trifluoracetic acid (TFA) were added
dropwise. The reaction mixture was stirred at room temperature
for 6 h, and the solvent evaporated under reduced pressure.
The crude product was subjected to column chromatography
(packed with silica gel, particle size, 0.063–0.2 mm Merck) and
eluted with a mixture of dichloromethane and MeOH at a ratio
of 10 : 1, yielding probe C in the form of red crystals (146 mg
with a reaction yield of 90%). H NMR (400 MHz, CDCl
1
3
) d 8.49
(d, J = 9.0 Hz, 1H), 7.96 (s, 2H), 7.80 (s, 2H), 7.59–7.51 (m, 1H),
7.22–7.17 (t, 1H), 7.14 (d, J = 7.5 Hz, 1H), 6.95–6.84 (m, 2H), 6.74
ꢀ1
+
+
solutions in the presence of 5 mg mL
nigericin, a K /H
(
3
d, J = 7.2 Hz, 1H), 5.44 (d, J = 12.8 Hz, 1H), 5.26 (s, 1H), 4.13 (s,
H), 3.43–3.48 (m, 2H), 3.24 (s, 3H), 2.86–2.78 (m, 2H), 2.68–2.66
m, 2H), 1.93–1.84 (m, 2H), 1.51 (s, 6H), 1.39 (d, J = 7.0 Hz, 2H).
ionophore for 20 minutes, to reach equilibrium between intra-
3
2–34
cellular pH and extracelluar pH.
We conducted fluorescence
(
1
3
imaging of HeLa with different intracellular pH values after the
cells were rinsed twice with the medium. Fluorescence of Mito-
Vew Blue was collected from 435 nm to 475 nm under excitation
at 405 nm, while the fluorescence of probes A, B, and C were
collected from 625 nm to 675 nm under 559 nm excitation. The
fluorescence of the cyanine dye (IR-780) was collected from
C NMR (101 MHz, CDCl ) d 163.79, 161.77, 161.43, 157.37,
3
1
1
1
53.60, 144.26, 140.85, 139.22, 136.40, 132.83, 128.24, 125.91,
25.37, 121.91, 121.32, 119.16, 118.67, 118.47, 117.97, 117.17,
15.55, 107.38, 92.27, 46.63, 42.99, 39.97, 29.85, 29.55, 22.61. MS
+
(
ESI): calcd for [C29
H
N
35 4
] 439.2856, found 439.2843.
750 nm to 800 nm at 635 nm excitation, while the fluorescence
Synthesis of probe D
of probe D was collected from 475 nm to 525 nm (for the
coumarin donor emission) and from 625 nm to 675 nm (for
cyanine acceptor fluorescence) under 405 nm excitation. The
fluorescence of probe D was also collected for acceptor fluores-
cence from 625 nm to 675 nm under 559 nm excitation.
To activate the carboxylic acid group of 7-(diethylamino)-
coumarin-3-carboxylic acid (7), BOP (301 mg, 0.68 mmol) and
N,N-diisopropylethylamine (DIPEA) (176 mg, 1.36 mmol) were
added into a solution of compound 7 (118 mg, 0.45 mmol) in
6
mL dichloromethane and then stirred at room temperature
for 20 min. After probe C (197 mg, 0.45 mmol) was added to the
reaction mixture, the mixture was stirred at room temperature
for 6 h under a nitrogen atmosphere. The solvent was then
removed by evaporation under vacuum, and the product was
subjected to column chromatography (packed with silica gel,
particle size, 0.063–0.2 mm Merck) and eluted with mixtures of
dichloromethane and MeOH at ratios from 100 : 1 to 50 : 1,
Fluorescence imaging of Drosophila melanogaster larvae
In order to demonstrate the feasibility of probe D for an in vivo
system, we submerged freshly hatched fruit fly larvae in three
different pH buffer solutions containing 10 mM probe D and
0.5% DMSO for 2 h, respectively. Fluorescence imaging was
conducted after the fruit fly larvae were washed three times with
deionized water.
resulting in pure probe D in the form of a red solid (180 mg,
1
6
1
1
6
1
4
3
1
0% reaction yield). H NMR (400 MHz, CDCl
3
) d 9.16–8.97 (m,
H), 8.70 (d, J = 9.3 Hz, 1H), 8.59 (s, 1H), d 7.81 (d, J = 5.6 Hz,
H), 7.60 (s, 1H), 7.38 (d, J = 8.9 Hz, 1H), 7.23–7.09 (m, 2H),
.95–6.83 (m, 2H), 6.74 (d, J = 8.0 Hz, 1H), 6.61 (dd, J = 8.9, Design and synthesis of the probes
.9 Hz, 1H), 6.41 (d, J = 1.9 Hz, 1H), 5.43 (d, J = 11.9 Hz, 1H),
.13 (s, 3H), 3.68–3.58 (m, 2H), 3.46–3.34 (m, 4H), 3.24 (s, 3H),
.13–3.03 (m, 2H), 2.99–2.92 (m, 2H), 2.74–2.64 (m, 2H),
Results and discussion
We have developed a new cyanine-based fluorescence platform
consisting of conjugated indole and acridine moieties to pro-
duce near-infrared pH-responsive probes. This platform has a
long p-conjugation system and a reactive chlorine atom at the
13
.96–1.87 (m, 2H), 1.52 (s, 6H), 1.27–1.09 (t, 6H). C NMR
101 MHz, CDCl ) d 166.43, 164.31, 162.74, 157.99, 157.20,
(
3
9-position of acridine, which facilitates functionalization with
1
1
1
4
9
53.29, 148.66, 144.21, 141.27, 139.36, 136.77, 133.13, 131.69,
28.14, 125.77, 124.68, 121.93, 121.52, 118.76, 118.67, 117.98,
15.92, 110.65, 108.57, 108.43, 107.45, 96.64, 92.20, 52.26, 47.32,
6.75, 45.46, 43.27, 40.72, 29.87, 29.53, 26.76, 26.33, 22.70, 12.72,
amine- and thiol derivatives, resulting in probes that can be
utilized for various sensing and imaging applications. The
functionalization of this new platform with amine derivatives
allows us to alter optical properties of the probes by modulating
the electron-donating or electron-withdrawing status of the
nitrogen atom at the 9-position varies upon pH changes.
We synthesized 9-chloro-1,2,3,4-tetrahydro-acridine (3) through
+
.15. MS (ESI): calcd for [C H N O ] 682.3752, found 682.3731.
4
3 48 5 3
Fluorescence imaging of HeLa cells
The HeLa cell line was acquired from ATCC and cultured accord- condensation of anthranilic acid (1) and cyclohexanone (2) in fresh
10,12,14–16,19,20,23,32–34
35–37
ing to our previously reported procedures.
For POCl
3
at high temperature
and then converted compound 3
colocalization experiments, the HeLa cells were incubated with to 9-chloro-1,2,3,4-tetrahydro-10-methyl-acridinium iodide (4)
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. B, 2021, 9, 5150–5161 | 5153

View Article Online
Journal of Materials Chemistry B
Paper
by reacting compound 3 with iodomethane in sulfolane at 50 1C yield of 2% and 0.53% in an aqueous solution containing 10%
Scheme 1). The near-infrared cyanine dye with a reactive chlorine ethanol and 0.1% DMSO, respectively. The low fluorescence
atom (probe A) was prepared by a condensation reaction of Fish- quantum yield was observed in the near-infrared cyanine dye
(
34
er’s aldehyde (5) with compound 4. A subsequent substitution (IR 780) with a fluorescence quantum yield of 7% in a pure
reaction of the reactive chlorine atom of probe A with an amine ethanol solution. We used a substitution reaction to replace the
residue from tert-butyl(2-aminoethyl)carbamate (6) afforded a chlorine atom in probe A with a nitrogen atom, affording
1
6
Boc-functionalized cyanine dye (probe B). The deprotection of probes B, C, and D, respectively. Probes B, C, and D possess
the Boc group of probe B with trifluoroacetic acid gave a primary high fluorescence quantum yields of 60%, 32%, and 35% in an
amine-functionalized cyanine dye (probe C) (Scheme 1). In order to aqueous solution containing 10% ethanol, respectively. Probes
detect pH changes ratiometrically, we prepared a near-infrared B, C, and D display reduced fluorescence quantum yields of
fluorescent probe (D) by introducing a 7-(diethylamino)coumarin- 18.2%, 10%, and 5.2% in pH 7.0 buffer solution containing
3-carboxylic acid (7) as a FRET donor to a near-infrared FRET 0.1% DMSO (Table S1, ESI†). Fluorescence quenching of the
cyanine acceptor (probe C) through an amide connection probes in almost pure buffer solvents is due to the aggregation-
1
13
(
Scheme 1). The probes were characterized by H NMR,
NMR, and high-resolution mass spectrometry (Fig. S1–S14, ESI†). hydrophobic feature (Table S1, ESI†).
Based on the fluorescent quantum yields of probes A, B, and
C, we conclude that an electron-withdrawing group at 9-position
C
induced quenching effect, especially for probe D with a more
Optical properties of probes A–D
We found that the optical properties of this novel fluorescent will quench the fluorophore, while a donating group at the
platform are highly dependent on the electron donating ability 9-position will enhance the fluorescence intensity of the fluoro-
of the substituents attached at the 9-position of the acridine phores (probes B, C, and D). The conclusion inspired us to
moiety. Probe A shows an absorption peak at 592 nm and an introduce an amine as the pH-responsive group to the reaction
emission peak at 660 nm with a low fluorescence quantum site on the 9-position of the acridine moiety. The amine group at
Fig. 1 (A) pH-dependent colors of buffer solutions containing probe D under visible light and UV light. Absorption (B) and fluorescence spectra (C) of
probe D in different pH buffer solutions.
5154
|
J. Mater. Chem. B, 2021, 9, 5150–5161
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry B
the 9-position is a strong electron-donating group under neutral from 9.01 to 3.2 for probe D, respectively (Fig. S22, ESI†). The
conditions and can be changed into a strong electron-withdrawing stability of probes B, C, and D was also confirmed by high-
group upon protonation with acid. This is expected to cause resolution ESI mass spectroscopy under strongly acidic conditions
fluorescence quenching due to disruption of the push–pull mole- (Fig. S23–S25, ESI†).
cular structure feature if the pH of the solution is decreased.
Probe A shows insensitivity to pH changes as we expected (Fig. S15, pH conditions, we investigated the H NMR spectrum of probe C
ESI†). Probe B shows absorption and fluorescence peaks at 510 nm in the absence and presence of TFA in a CDCl solution. As shown
To explore the mechanism of the probe responses to acidic
1
3
and 650 nm, respectively (Fig. S16, ESI†). Gradual decreases of the in Fig. S26 (ESI†), at acidic pH, peak a, assigned to the ethyl
0
pH from 10.40 to 2.0 lead to decreases in both absorbance and protons (adjacent to the amino group), shifts to peak a . Peak b,
fluorescence, which arise from protonation of the second amine of assigned to the proton on the NH group, disappears after the
the cyanine dye at the 9-position (Fig. S16, ESI† and Chart 1). Probe addition of TFA, indicating that the NH group adjacent to the
B displays a pK value of 2.84 (Fig. S17, ESI†). Probe C displays benzyl ring was protonated at acidic pH. Additionally, the benzyl
a
similar responses to pH changes with a slightly lower pK value of protons c merged after treatment with TFA, indicating that the
a
2.18 than probe B (Fig. S18 and S19, ESI†). Probe D shows two electronic conjugation within probe C changed under acidic
absorption peaks at 436 nm and 540 nm, responding to the conditions. We also observed a color change of probe C in the
absorption of the coumarin donor and cyanine acceptor (Fig. 1). NMR tube from dark red to yellow after the acid treatment.
Gradual pH decreases from 10.40 to 2.20 cause absorbance
Under neutral and basic pH conditions, there are three amine
decreases of both the coumarin donor and the cyanine acceptor groups with two electron-donating and electron-withdrawing
at 436 nm and 540 nm, respectively (Fig. 1B). It undergoes a drastic features in probes B–D, which possesses push–pull features and
color change from pink to yellow when buffer pH decreases from results in strong fluorescence of the probes (Chart 2). However,
9
.02 to 3.20 under daylight (Fig. 1A). However, probe D displays protonation of the secondary amine at 9-position in acidic
ratiometric fluorescence responses to pH changes from 7.0 to condition leads to fluorescence quenching because the systems
.0 with gradual fluorescence decreases of the cyanine acceptor possess two electron-withdrawing amine groups and one
and gradual fluorescence increases of the coumarin donor electron-donating amine group, disrupting push–pull features
Fig. 1C). The ratiometric responses of probe D arise from a of the fluorophore. This unique structure of the probes allows
3
(
high-efficiency energy transfer from the coumarin donor to the for direct modulation of the probe absorption and fluorescence
cyanine acceptor with an energy transfer rate of 97.7%. This is properties by altering the electron-donating/withdrawing nature
because the emission spectrum of the coumarin donor overlaps of the central amine at 9-position through pH changes (Chart 2).
significantly with the absorption spectrum of the cyanine
Probe selectivity and photostability
acceptor (Fig. S21, ESI†), which satisfies the requirement for
fluorescence resonance energy transfer (FRET) from the donor to Probes B, C and D show good selectivity to pH over anions and
2
+
2+
3+
2+
2+
2+
2+
the acceptor. Solutions of probe D under UV light are red and green cations such as Zn , Ca , Fe , Hg , Cd , Ni , and Mg ,
ꢀ
ꢀ
ꢀ
3ꢀ
2ꢀ
at pH values of 7.0 and 3.0, respectively (Fig. 1A). Probe D also NO
displays a higher pK value of 6.01 (Fig. S21, ESI†) due to the 10 mM cations or anions does not cause any significant fluores-
bonding of the electron-withdrawing coumarin donor to the cyanine cence changes (Fig. S27, ESI†).
3
, Cl , HCO
3
, PO
4
, and CO
3
since the presence of
a
acceptor. Also, probes B, C, and D display reversible fluorescence
Probes A, B, C, and D were excited continuously at 520 nm for
responses to pH changes from 7.0 to 2.2 for probe B and C, 10 min intervals and the fluorescence intensity was measured
Chart 2 pH response mechanism of the probes based on the new cyanine platform.
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. B, 2021, 9, 5150–5161 | 5155

View Article Online
Journal of Materials Chemistry B
Paper
every 10 min at pH 7.4. The result indicates that probes A, B, C,
and D all displayed excellent photostability with less than
5% decrease only in fluorescence intensity under two-hour
excitation at pH 7.4 (Fig. S28, ESI†).
Cytotoxicity of the probes
Standard MTT assays were carried out through 24 h incubation
of HeLa cells with 1, 2, 5, 10, and 20 mM concentrations of
probes A, B, C, or D (Fig. 2). The viability of the cells incubated
with 10 mM probe A, B, C, and D are more than 85%, demonstrating
that the probes own decent biocompatibility for biological studies.
Probe D shows slightly lower cell viability than probes B–D
because of its more hydrophobic feature with an additional
coumarin donor (Fig. 2).
+
Fig. 3 Overlay of probe D (blue) and probe DH (red), showing the
different positions of the coumarin moieties.
and 0.9728 for the observed transitions at 345 and 600 nm
(Table S3, ESI† and Fig. 1). Probes B with attached carbamate
and C with attached ethylenediamine and their protonated
compounds did not differ substantially in their calculated transi-
tions (Fig. 4), which may be due to a lack of orbital overlap and
thus conjugation with these added substituents. Probe D was
calculated to have a visible absorption at 494 nm (compared to
537 nm expt.), which was composed of a transition from the
indole to the acridine moiety, as described earlier for Probes A–C.
The higher energy transition calculated at 369 nm appears to be
mainly based on an absorption located on the coumarin section
of the probe (Table S13 and Fig. S49, ESI†). The profiles of the
calculated absorptions for probe D matched those obtained
experimentally (compare Fig. 1 and Fig. S49, ESI†) due to the
contribution of transitions with high oscillating strength from
excited states 5 and 6 (Table S13, ESI†). Upon protonation and
formation of DH , the visible transition appears to be from the
coumarin section of the molecule to a delocalized state consisting
of orbitals localized on the acridine and indole moieties
Table S15, ESI†). The final calculated structure for probe DH
appeared to have a sterically hindered geometry, but changing
torsion angles and subsequent refining of the model resulted in
a similar conformation. There are many ways to calculate pK
values for molecules containing functional groups including
amines. For probes B–D, we utilized DFT together with the
Solvation Model based on the Density (SMD) solvent model.
This method resulted in substantial differences between the
Theoretical results
Computational models of the probes and their expected protonated
versions were calculated as outlined in the ESI.† The conformations
38
obtained in a Polarizable Continuum Model (PCM) of water
contained dihedral angles between the indole and tetrahydro-
acridine planes that varied slightly between the neutral and proto-
+
nated versions of the probes with A, 41.311; B, 44.321; BH , 38.071; C,
+
+
3
6.441; CH , 39.061; D, 39.301 and DH , 36.271. However, while there
were slight differences in this dihedral angle in probe D, there was a
big difference in the location of the coumarin moieties resulting
after protonation, as shown in Fig. 3.
Electronic transition calculations established that in probe A,
a movement of electron density from the indole end of the
molecule to the tetrahydro-acridine moiety occurs at 538 nm, as
compared to the experimentally observed 600 nm (Fig. 4).
The higher energy transition was calculated to occur at
+
363 nm (345 nm expt. in water, see Fig. 1), which was in closer
agreement and corresponded to an excited state with a higher
electron density on the acridine moiety (Fig. S34, ESI†). For
probe A, the calculated oscillating strengths of the two transitions
matched the absorption profile for the spectrum obtained experi-
mentally, as the UV transitions had calculated values of 0.3379
+
(
a
3
9
calculated and experimental data with an average DpK
a
of 4.60
(Table 1). In a study on alcohols, phenols, and hydroperoxides,
the addition of one water molecule directly to the site being
protonated/deprotonated improved the accuracy of the calcula-
4
0
tion by as much as 3 pK
of one water molecule (Fig. S54, ESI†) reduced the average DpK
from 4.60 to 0.64 (Table 1), which is comparable to the error
a
units. With probes B–D, the addition
a
4
0
range reported with the inclusion of a cluster of three water
molecules of 0.4 to 1.4 pK units. With the inclusion of one
a
water molecule, we found the largest deviation for probe C,
which contained two N atoms that could be protonated. The
calculated values for the pK
a
for the outer N atom on probe C at
Fig. 2 MTT assays were used to evaluate the cytotoxicity of the probes through
7.58 deviated substantially from the experimentally obtained
24 h incubation of HeLa cells with different probe concentrations. The cell
value of 2.18. The calculated value for the more centrally located
N atom at 0.27 was closer to the experimental value, suggesting that
viability was determined by the absorbance obtained at 490 nm with three
repeated measurements while cells without the probe were used as a control.
5156
|
J. Mater. Chem. B, 2021, 9, 5150–5161
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry B
Fig. 4 Illustrations of the current density difference as isosurfaces of the probes and their protonated forms, as indicated for the excited states (ES) and
the calculated (and experimental) wavelengths. The composition of specific ES, together with percentage contributions, are indicated. The numerical
range values for the color scale illustrated at the top of the figure are also listed. Drawings of the numbered MOs are available in the ESI.†
Table 1 Experimental pK
a
’s and calculated pK
a
’s using the APFD/6- adenosine triphosphate (ATP). They play very important roles in
a
3
1+G(d) level of theory for the probes
calcium homeostasis, regulation of the innate immunity, cellular
4
1
b
b
signaling, and programmed cell death. The mitochondrial
functions also significantly depend on the intracellular pH.
Therefore, it is important to accurately determine mitochondrial
Probe
Experiment
SMD
DpK
a
SMD + 1 water
2.63
DpK
a
B
C
D
2.84
2.18
6.01
ꢀ1.19
ꢀ2.20
0.62
ꢀ4.03
ꢀ4.38
ꢀ5.40
ꢀ0.21
ꢀ1.61
c
0.27 (7.58)
6.13
0.12 pH changes for an insightful understanding of mitochondrial
physiology and pathology.
a
b
Atomic positions and diagrams are in the ESI. DpK
a
= pKa,calculated
ꢀ
c
pKa,experiment
.
Added water located near the outer N atom in probe C.
Since probes A–D are positively charged, we expected that
they would specifically target mitochondria through electrostatic
interactions as negative potentials are located within mitochondrial
inner membranes. In order to test this hypothesis, we conducted
colocalization experiments by incubating HeLa cells with 5 mM
probes A–C and 5 mM mitochondria-specific Mitoview Blue for
this was the site of protonation/deprotonation. The results of
these experiments suggest that a model consisting of SMD with
one strategically placed water molecule would be a reasonable
starting point for pK calculations of large molecules involving
a
N atoms.
32
15 minutes, washing the cells with cell culture medium twice,
and then adding cell culture medium to the cells for cellular
imaging (Fig. S29–S31, ESI†). In order to avoid fluorescence
interference of Mitoview Blue with coumarin donor fluorescence
Cellular imaging of the probes
Mitochondria, double membrane-bound organelles that are the of probe D, we chose a mitochondria-specific cyanine dye IR-780
power producers within cells, convert nutrients and oxygen into for the colocalization experiment by incubating HeLa cells with
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. B, 2021, 9, 5150–5161 | 5157

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 5 Fluorescence images of HeLa cells incubated with 5 mM probe D and 5 mM cyanine dye (IR-780) for 15 minutes under excitation at 405 nm and
5
5
59 nm for probe D and under excitation at 635 nm for cyanine dye (IR-780). Scale bar: 20 mm. Fluorescence of probe D was collected from 475 to
25 nm for the donor fluorescence in channel I, and 625 nm to 675 nm for the acceptor fluorescence in channel II under the donor excitation at 405 nm.
Fluorescence of probe D was also collected from 625 nm to 675 nm for the acceptor fluorescence under the acceptor excitation at 559 nm.
5
mM probe D and 5 mM cyanine dye IR-780 according to the same Pearson correlation coefficients of 0.888, 0.857, and 0.901 were
colocalization procedure used for probes A–C (Fig. 5). High observed among probes A, B, and C incubated with Mitoview Blue
Fig. 6 Fluorescence images of HeLa cells incubated with 5 mM probe D under different cellular pH conditions with scale bars of 50 mm. HeLa cells were
incubated with 5 mM probe D for 15 minutes, washed with cell culture medium twice, and then incubated in different pH buffers in the presence of
ꢀ1
+
+
5
mg mL nigericin, a K /H ionophore for 20 minutes. Channel 1: the coumarin donor fluorescence collected from 475 to 525 nm at 405 nm excitation.
Channel 2: the cyanine acceptor fluorescence collected from 600 to 700 nm at 405 nm excitation. Channel 3: bright-field. Channel 4: merged images of
channels 1, 2, and 3. Channel 5: ratiometric images of channel 1 divided by channel 2, which were acquired with Image-Pro software.
5158
|
J. Mater. Chem. B, 2021, 9, 5150–5161
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry B
Probe D displays ratiometric fluorescent responses to intra-
cellular pH changes with fluorescence decreases of the cyanine
acceptor in channel 2 and fluorescence increases of the coumarin
donor in channel 1 upon intracellular pH decreases from 9.0 to
pH 4.5 (Fig. 6 and 7). The merged images (channel 4) of channels
1, 2, and 3 show drastic color changes from red to yellow and
then to yellowish-green upon intracellular pH changes from 9.0
to 4.5. Furthermore, probe D also exhibits significant ratiometric
color changes in channel 5 from bluish yellow to bluish green
upon intracellular pH changes from 9.0 to 4.5 (Fig. 6).
Fluorescence visualization in D. melanogaster first-instar larvae
We chose probe D to test the feasibility of visualizing pH
changes in vivo. We incubated first-instar D. melanogaster larvae
with probe D in three different pH buffers for 2 h before
Fig. 7 Fluorescence intensity of the probe donor (channel 1) and accep-
tor (channel 2) in HeLa cells obtained from Fig. 6.
3
0,42
conducting confocal fluorescence microscopy.
showed strong fluorescence of the cyanine acceptor and relatively
Fig. 5–8), while the colocalization of probe D with the cyanine weak fluorescence of the coumarin donor after incubation in a
dye IR-780 resulted in a high Pearson correlation coefficient of pH 7.0 buffer (Fig. 7). However, the larvae displayed weak
.946. These results confirm that probes A–D do selectively stain fluorescence of the cyanine acceptor and relatively strong fluores-
The larvae
(
0
mitochondria (Fig. S29–S31, ESI† and Fig. 5).
cence of the coumarin donor after the larvae were incubated with
probe D in a pH 5.0 buffer, which agrees with the ratiometric
fluorescence responses of probe D to pH changes in buffer
solutions (Fig. 1).
Intracellular pH imaging of probe D
We further used probe D to verify whether it can respond to pH
changes in mitochondria as it possesses a higher pK value. We
a
incubated probe D with HeLa cells in different pH buffer
solutions in the presence of 5 mM nigericin, a K /H ionophore,
Conclusion
+
+
which is used to adjust intracellular pH to the buffer pH We have developed a new near-infrared cyanine dye with an
1
0,12,14–16,19,20,23,32–34
through pH equilibration.
easily substituted chlorine atom as well as three pH-sensitive
Fig. 8 Fluorescence images of D. melanogaster first-instar larvae incubated with fluorescent probe D in three different buffer solutions. Channel 1: the
donor fluorescence collected from 475 to 525 nm at 405 nm excitation. Channel 2: the acceptor fluorescence collected from 600 to 700 nm at 405 nm
excitation. Channel 3: bright-field. Channel 4: the acceptor fluorescence collected from 600 to 700 nm under 559 nm excitation. Scale bar: 200 mm.
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. B, 2021, 9, 5150–5161 | 5159

View Article Online
Journal of Materials Chemistry B
Paper
near-infrared fluorescent probes with high fluorescence quantum 15 S. Xia, J. B. Wang, J. H. Bi, X. Wang, M. X. Fang, T. Phillips,
yields. These three probes B–D were obtained by substituting the
chlorine atom on probe A for a nitrogen atom bonded to three
A. May, N. Conner, M. Tanasova, F. T. Luo and H. Y. Liu,
Sens. Actuators, B, 2018, 265, 699–708.
different substituents. Probes B–C show fluorescence decreases 16 Y. B. Zhang, J. H. Bi, S. Xia, W. Mazi, S. L. Wan, L. Mikesell,
upon pH decreases, while probe D exhibits ratiometric fluores-
R. L. Luck and H. Y. Liu, Molecules, 2018, 23, 2679.
cence responses to pH decreases with fluorescence decreases of 17 H. A. Shindy, Dyes Pigm., 2017, 145, 505–513.
the cyanine acceptor and fluorescence increases of the coumarin 18 W. Mazi, Y. N. Yan, Y. B. Zhang, S. Xia, S. L. Wan, M. Tajiri,
donor. We have further demonstrated that probe D can be
R. L. Luck and H. Y. Liu, J. Mater. Chem. B, 2021, 9, 857–863.
used to ratiometrically visualize pH changes in HeLa cells and 19 J. B. Wang, S. Xia, J. H. Bi, Y. B. Zhang, M. X. Fang,
D. melanogaster first-instar larvae.
R. L. Luck, Y. B. Zeng, T. H. Chen, H. M. Lee and
H. Y. Liu, J. Mater. Chem. B, 2019, 7, 198–209.
2
2
0 M. X. Fang, R. Adhikari, J. H. Bi, W. Mazi, N. Dorh, J. B.
Wang, N. Conner, J. Ainsley, T. G. Karabencheva-Christova,
F. T. Luo, A. Tiwari and H. Y. Liu, J. Mater. Chem. B, 2017, 5,
Conflicts of interest
There are no conflicts to declare.
9579–9590.
1 T. H. Chen, S. W. Zhang, M. Jaishi, R. Adhikari, J. H. Bi,
M. X. Fang, S. Xia, Y. B. Zhang, R. L. Luck, R. Pati, H. M. Lee,
F. T. Luo, A. Tiwari and H. Y. Liu, ACS Appl. Bio Mater., 2018,
Acknowledgements
1
, 549–560.
We greatly appreciate financial support for this research work by
the National Institute of General Medical Sciences of the National
Institutes of Health under Award Numbers R15GM114751 and
2
2
2
2 Y. B. Zhang, S. Xia, M. X. Fang, W. Mazi, Y. B. Zeng,
T. Johnston, A. Pap, R. L. Luck and H. Y. Liu, Chem.
Commun., 2018, 54, 7625–7628.
3 J. B. Wang, S. Xia, J. H. Bi, M. X. Fang, W. F. Mazi,
Y. B. Zhang, N. Conner, F. T. Luo, H. P. Lu and H. Y. Liu,
Bioconjugate Chem., 2018, 29, 1406–1418.
4 S. W. Zhang, T. H. Chen, H. M. Lee, J. H. Bi, A. Ghosh,
M. X. Fang, Z. C. Qian, F. Xie, J. Ainsley, C. Christov,
F. T. Luo, F. Zhao and H. Y. Liu, ACS Sens., 2017, 2, 924–931.
2R15GM114751-02 (to H. Y. Liu). The calculations of the fluo-
rescent probe were conducted by using a high-performance
computing infrastructure at Michigan Technological University.
References
1
J. L. Bricks, A. D. Kachkovskii, Y. L. Slominskii, A. O. 25 G. K. Vegesna, J. Janjanam, J. H. Bi, F. T. Luo, J. T. Zhang,
Gerasov and S. V. Popov, Dyes Pigm., 2015, 121, 238–255.
A. P. Gorka, R. R. Nani and M. J. Schnermann, Org. Biomol.
Chem., 2015, 13, 7584–7598.
H. A. Shindy, Mini-Rev. Org. Chem., 2012, 9, 352–360.
W. Sun, S. G. Guo, C. Hu, J. L. Fan and X. J. Peng, Chem.
Rev., 2016, 116, 7768–7817.
C. Olds, A. Tiwari and H. Y. Liu, J. Mater. Chem. B, 2014, 2,
4500–4508.
2
26 Y. B. Zhang, S. Xia, L. Mikesell, N. Whisman, M. X. Fang,
T. E. Steenwinkel, K. Chen, R. L. Luck, T. Werner and
H. Y. Liu, ACS Appl. Bio Mater., 2019, 2, 4986–4997.
27 L. L. Wu, C. S. Huang, B. Emery, A. C. Sedgwick, S. D. Bull,
X. P. He, H. Tian, J. Yoon, J. L. Sessler and T. D. James,
Chem. Soc. Rev., 2020, 49, 5110–5139.
3
4
5
6
7
L. Yuan, W. Y. Lin, K. B. Zheng, L. W. He and W. M. Huang,
Chem. Soc. Rev., 2013, 42, 622–661.
Y. Li, Y. M. Zhou, X. L. Yue and Z. F. Dai, Adv. Healthcare 28 R. J. Gui, H. Jin, X. N. Bu, Y. X. Fu, Z. H. Wang and Q. Y. Liu,
Mater., 2020, 9, 2001327.
Coord. Chem. Rev., 2019, 383, 82–103.
Z. Q. Guo, S. Park, J. Yoon and I. Shin, Chem. Soc. Rev., 2014, 29 L. Yuan, W. Y. Lin, K. B. Zheng and S. S. Zhu, Acc. Chem.
43, 16–29.
Res., 2013, 46, 1462–1473.
8
9
R. Cadillo-Chavez, Arch. Esp. Urol., 2019, 72, 759–764.
A. Kumari and S. Gupta, J. Biophotonics, 2019, 12, 1–7.
30 S. Xia, Y. B. Zhang, M. X. Fang, L. Mikesell, T. E. Steenwinkel,
S. L. Wan, T. Phillips, R. L. Luck, T. Werner and H. Y. Liu,
ChemBioChem, 2019, 20, 1986–1994.
10 M. X. Fang, S. Xia, J. H. Bi, T. P. Wigstrom, L. Valenzano,
J. B. Wang, W. Mazi, M. Tanasova, F. T. Luo and H. Y. Liu, 31 X. J. Zhang, Y. Hu, X. T. Yang, Y. Y. Tang, S. Y. Han, A. Kang,
Chem. Commun., 2018, 54, 1133–1136.
1 H. X. Li, H. Dong, M. M. Yu, C. X. Liu, Z. X. Li, L. H. Wei,
H. S. Deng, Y. M. Chi, D. Zhu and Y. Lu, Biosens. Bioelectron.,
2019, 138, 111314.
L. D. Sun and H. Y. Zhang, Anal. Chem., 2017, 89, 8863–8869. 32 S. A. Xia, J. B. Wang, Y. B. Zhang, N. Whisman, J. H. Bi,
1
1
2 W. Mazi, R. Adhikari, Y. B. Zhang, S. Xia, M. X. Fang,
R. L. Luck, M. Tajiri, A. Tiwari, M. Tanasova and H. Y. Liu,
Methods, 2019, 168, 40–50.
3 Y. X. Shi, X. C. Meng, H. R. Yang, L. N. Song, S. J. Liu,
A. Q. Xu, Z. J. Chen, W. Huang and Q. Zhao, J. Mater. Chem.
B, 2019, 7, 3569–3575.
T. E. Steenwinkel, S. L. Wan, J. Medford, M. Tajiri, R. L. Luck,
T. Werner and H. Y. Liu, J. Mater. Chem. B, 2020, 8, 1603–1615.
33 J. T. Zhang, M. Yang, C. Li, N. Dorh, F. Xie, F. T. Luo,
A. Tiwari and H. Y. Liu, J. Mater. Chem. B, 2015, 3,
2173–2184.
34 J. T. Zhang, M. Yang, W. F. Mazi, K. Adhikari, M. X. Fang,
F. Xie, L. Valenzano, A. Tiwari, F. T. Luo and H. Y. Liu, ACS
Sens., 2016, 1, 158–165.
1
1
4 S. Xia, M. X. Fang, J. B. Wang, J. H. Bi, W. Mazi, Y. B. Zhang,
R. L. Luck and H. Y. Liu, Sens. Actuators, B, 2019, 294, 1–13.
5160
|
J. Mater. Chem. B, 2021, 9, 5150–5161
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry B
3
3
5 M. Scipioni, G. Kay, I. L. Megson and P. K. T. Lin, Med- 38 G. Scalmani and M. J. Frisch, J. Chem. Phys., 2010, 132, 114110.
ChemComm, 2019, 10, 764–777.
39 P. Lian, R. C. Johnston, J. M. Parks and J. C. Smith, J. Phys.
Chem. A, 2018, 122, 4366–4374.
6 X. J. Cheng, J. X. Gu, Y. P. Pang, J. Liu, T. Xu, X. R. Li,
Y. Z. Hua, K. A. Newell, X. F. Huang, Y. H. Yu and Y. Liu, ACS 40 B. Thapa and H. B. Schlegel, J. Phys. Chem. A, 2017, 121,
Chem. Neurosci., 2019, 10, 3500–3509. 4698–4706.
37 M. Scheiner, D. Dolles, S. Gunesch, M. Hoffmann, M. Nabissi, 41 S. W. G. Tait and D. R. Green, J. Cell Sci., 2012, 125, 807–815.
O. Marinelli, M. Naldi, M. Bartolin, S. Petralla, E. Poeta, B. Monti, 42 Y. B. Zhang, Y. N. Yan, S. Xia, S. L. Wan, T. E. Steenwinkel,
C. Falkeis, M. Vieth, H. Hubner, P. Gmeiner, R. Maitra,
J. Medford, E. Durocher, R. L. Luck, T. Werner and H. Y. Liu,
T. Maurice and M. Decker, J. Med. Chem., 2019, 62, 9078–9102.
ACS Appl. Mater. Interfaces, 2020, 12, 20172–20179.
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. B, 2021, 9, 5150–5161 | 5161