Turn-on fluorescent probes that can light up endogenous RNA in nucleoli and cytoplasm of living cells under a two-photon microscope

Article abstract of DOI:10.1039/c6ra28284a

We have synthesized two-photon organic molecule probes (IMT-E and IMT-M), which selectively stain endogenous RNA in the nucleolus and cytoplasm of living cells in a short incubation time. Using these probes, bright two-photon excitation fluorescence (TPEF) images of living cells have been developed. Titrations and digest tests of ribonuclease indicate the markedly higher affinity of these probes for RNA, especially when using IMT-E. Upon binding to RNA, the fluorescence intensity of IMT-E increases by about 15-fold, showing that IMT-E is a turn-on probe for the detection of RNAs. MTT assays demonstrate that the mitochondria of cells maintain their electron mediating ability after being stained with the probes. These results demonstrate that IMT-E is an attractive two-photon turn-on fluorescent probe for visualizing RNA in living cells.

Full text of DOI:10.1039/c6ra28284a

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
RSC Advances
View Article Online
View Journal | View Issue
PAPER
Turn-on fluorescent probes that can light up
endogenous RNA in nucleoli and cytoplasm of
Cite this: RSC Adv., 2017, 7, 16730
living cells under a two-photon microscope†
Ruiqing Feng,‡ab Longlong Li,‡a Bing Li,a Jinhui Li,a Dan Peng,a Yitao Yu,a
a
a
b
c
*
*
*
Qiuhong Mu, Ning Zhao, Xiaoqiang Yu and Zhenhua Wang
We have synthesized two-photon organic molecule probes (IMT-E and IMT-M), which selectively stain
endogenous RNA in the nucleolus and cytoplasm of living cells in a short incubation time. Using these
probes, bright two-photon excitation fluorescence (TPEF) images of living cells have been developed.
Titrations and digest tests of ribonuclease indicate the markedly higher affinity of these probes for RNA,
especially when using IMT-E. Upon binding to RNA, the fluorescence intensity of IMT-E increases by
about 15-fold, showing that IMT-E is a turn-on probe for the detection of RNAs. MTT assays
demonstrate that the mitochondria of cells maintain their electron mediating ability after being stained
with the probes. These results demonstrate that IMT-E is an attractive two-photon turn-on fluorescent
probe for visualizing RNA in living cells.
Received 16th December 2016
Accepted 28th February 2017
DOI: 10.1039/c6ra28284a
rsc.li/rsc-advances
and essential to design and develop RNA uorescent probes for
imaging the nucleolus of living cells.
Introduction
In the eld of uorescent probes for bio-imaging, two-
photon turn-on uorescence probes have gained tremendous
attention.13 This is because two-photon uorescence micros-
copy (TPM) has unique merits in imaging living specimen, such
as high detection sensitivity, deep penetration, no image
distortion, low photodamage and reduced photobleaching.14–17
Moreover, turn-on probes possess the advantages of low back-
ground noise and high sensitivity when binding to biomole-
cules.18 For imaging nucleolus RNA in living cells, the probe not
only needs to penetrate the cell membrane, but also enter the
nucleus. Thus, RNA probes have a high requirement for
membrane permeability. Generally, large conjugated structures
could help probes to achieve better two-photon properties.19,20
However, it is a contradiction that increasing the conjugate
structure could reduce the membrane permeability.21 Recently,
small-molecule uorescent probes have become the most
practical tool for imaging living cells, due to their good
membrane permeability and chemical tractability.22 However, it
is rather difficult to design small-molecule uorescent probes
with excellent two-photon properties and membrane perme-
ability at the same time. In the eld of two-photon probes for
RNA detection, an appropriate probe is rarely available. Ohul-
chanskyy et al. found that the conventional nucleic acid dye
cyan 40 can be used to image RNA in nucleoli using TPM.
However, the two-photon action absorption cross section (d ꢀ
F) of the probe, a crucial parameter for TPEF probes, was not
reported.23 Saini et al. synthesized a new trivalent aluminium
metal complex (1) which could achieve nucleolus imaging in
a variety of cell lines, but its two-photon properties were not
Ribonucleic acid (RNA) plays an active role within living cells by
catalyzing biological reactions, controlling gene expression, and
sensing and communicating responses to cellular signals.1–4
Recently, researchers have found that the variations in nucle-
olus morphology are closely associated with disease pheno-
types, and the relationship between the morphology of the
nucleolus and cancerous status in tumor cells has attracted
increasing attention.5,6 As is known, RNA is mainly distributed
in the nucleolus and cytoplasm of living cells, and the nucleolus
is the key site in the nucleus that synthesizes and assembles
ribosomal RNAs (rRNA).7 Therefore, gaining a better under-
standing of the morphology of the nucleolus using RNA
imaging in living cells could offer powerful help for the
advancement of molecular biology, pathophysiology, and
medical diagnostics.8–11 As one of the most powerful techniques
for monitoring biomolecules in living systems, uorescence
imaging techniques are widely used to visualize morphological
details of RNA in living cells.12 Consequently, it is signicant
aShandong Key Laboratory for Adhesive Materials, Advanced Materials Institute,
Shandong Academy of Sciences, Jinan 250014, Shandong, P. R. China. E-mail:
zhaon@sdas.org
bCenter of Bio & Micro/Nano Functional Materials, State Key Laboratory of Crystal
Materials, Shandong University, Jinan 250100, P. R. China. E-mail: yuxq@sdu.edu.
cn; Tel: +86 53188366418
cShandong Analysis and Testing Center, Shandong Academy of Science, Jinan 250014,
Shandong, P. R. China. E-mail: wzh312@mail.sdu.edu.cn
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c6ra28284a
‡ Feng and Li made equal contributions to this work.
16730 | RSC Adv., 2017, 7, 16730–16736
This journal is © The Royal Society of Chemistry 2017

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
Paper
RSC Advances
reported.24 Our research group has also reported several
carbazole- and indole-based two-photon RNA probes for cell
imaging.25,26 However, because of the poor membrane perme-
ability, living cell imaging using these probes is limited.
Subsequently, in order to improve the permeability, we intro-
duced a long alkyl chain into indole-based di-cationic salts, and
synthesized two RNA probes that can stain living cells and have
two-photon properties. Unfortunately, the incubation time of
1.5 hours demonstrated that their permeability was still less
than satisfactory.27 Only Guo et al. synthesized an indole-based
cyanine complex as a two-photon RNA-selective probe that
realized successful imaging in living cells.28 Therefore, it is
a signicant challenge to develop two-photon turn-on probes
with good membrane permeability for imaging RNA in living
cells.
According to a variety of theoretical and experimental data,
the introduction of terminal donor (D) and acceptor (A) moie-
ties into small p-conjugates can result in a considerable two-
photon absorption cross section (d) value and a simultaneous
red shi in emission.20,29,30 Based on previous work, two-photon
RNA probes with good membrane permeability may be ob-
tained by using various cationic salts with a particular chemical
structure by appropriately increasing the donor–acceptor
strength and intramolecular conjugation length. In this paper,
we introduced indole as a donor group and benzothiazole
iodized salt as an acceptor group into p-conjugates, and
prepared two mono-cationic compounds (IMT-E: (E)-2-(2-(1H-
indol-3-yl)vinyl)-3-(2-hydroxyethyl)benzo[d]thiazol-3-ium, MW:
448; IMT-M: (E)-2-(2-(1H-indol-3-yl)vinyl)-3-methylbenzo[d]
thiazol-3-ium, MW: 418, see Fig. 1). Because of their smaller
molecular structures and the appropriate electron donating
ability of indole and electron withdrawing ability of benzimid-
azole iodized salt, IMT-E and IMT-M may be two-photon probes
with good membrane permeability for imaging nucleolus RNA
in living cells.
Fig. 1 The UV-vis absorption (a) and single-photon excited fluores-
cence (b) spectra of SYTO RNA-Select (I), IMT-E (II) and IMT-M (III) in
buffer, RNA solutions and DNA solutions. Buffer: 10 mM Tris–HCl,
100 mM KCl, pH ¼ 7.2. Compound concentration: 2 ꢀ 10ꢁ6 mol Lꢁ1
.
Excitation at each largest single-photon absorption wavelength. Ratio
(phosphate of NA/dye): 300 : 1 for SYTO RNA-Select (I), 500 : 1 for
IMT-E and IMT-M.
The subsequent testing of IMT-E and IMT-M gave us
a pleasant surprise. The probes were found to possess a two-
photon property, and can light up the RNA in the nucleolus
and cytoplasm of living cells with excellent biological compat-
ibility. Aer binding to RNA, the uorescence intensity from
IMT-E/IMT-M increased by about 15/10-fold, showing that the
dyes are turn-on probes for the detection of RNAs.
Scheme 1 The synthesis of IMT-E and IMT-M.
127.81, 127.27, 125.17, 124.37, 124.31, 122.89, 121.31, 116.72,
114.68, 113.62, 107.11. HRMS (m/z): [M ꢁ I]+. Calcd for
C
19H17N2OS, 321.1; found, 321.42.
IMT-M. 1H NMR (400 MHz): 12.497 (s, 1H), 8.466 (t, J ¼ 10.6,
Experimental
Synthesis
2H), 8.335 (d, J ¼ 8, 1H), 8.275 (t, J ¼ 4.42, 1H), 8.149 (d, J ¼ 8.36,
1H), 7.815 (t, J ¼ 7.72, 1H), 7.712 (t, J ¼ 7.64, 1H), 7.595 (t, J ¼
4.5, 1H), 7.525 (d, J ¼ 15.76, 1H), 7.525 (d, J ¼ 15.76, 1H), 7.370–
7.324 (m, 2H), 4.277 (s, 3H). 13C NMR (400 MHz, DMSO-d6),
d (ppm): 172.45, 144.73, 142.41, 138.37, 138.04, 129.31, 127.09,
125.20, 124.40, 124.30, 122.90, 121.48, 116.40, 114.70, 113.62,
106.22. HRMS (m/z): [M ꢁ I]+. Calcd for C18H15N2S, 291.1;
found, 291.39.
The synthetic details for IMT-E and IMT-M are shown in
Scheme 1 and the ESI.† The dyes were synthesized via Knoeve-
nagel
condensation
reactions
between
N-methyl-2-
methylbenzothiazole iodine and 4-dimethylaminobenzalde-
hyde, using piperidine as the catalyst.
IMT-E. 1H NMR (400 MHz): 12.478 (s, 1H), 8.448 (t, J ¼ 7.66,
2H), 8.342 (d, J ¼ 7.96, 1H), 8.205 (d, J ¼ 4.96, 1H), 8.16 (t, J ¼
9.72, 1H), 7.798 (t, J ¼ 7.76, 1H), 7.713 (q, J ¼ 9.29, 1H), 7.604 (t, J
¼ 5.4, 2H), 7.357 (d, J ¼ 4.36, 2H), 5.261 (t, J ¼ 5.68, 1H), 4.959 (t,
Reagents and chemicals
J ¼ 4.34, 2H), 3.965 (q, J ¼ 4.88, 2H). 13C NMR (400 MHz, DMSO- Indole was purchased from J&K Chemical (Beijing, China).
d6), d (ppm): 173.40, 144.44, 142.06, 138.38, 137.87, 129.23, Iodomethane, iodoethanol and 2-aminobenzimidazole were
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 16730–16736 | 16731

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
RSC Advances
Paper
purchased from Aladdin Co. All other chemicals were from is the concentration of the solution, n is the refractive index of
commercial sources and of analytical reagent grade, unless the solution, F is the two-photon excited uorescence integral
indicated otherwise. Ultrapure water was used throughout. Calf intensity and F is the uorescence quantum yield.
thymus DNA and E. coli RNA, used as the models for DNA and
RNA, and ribonuclease A (RNase) were obtained from Sigma.
Tris and PBS were purchased from Seikagaku Corporation
(Japan).
Cell culture and staining
Human cervical cancer cell lines (HeLa) were cultured in Dul-
becco's modied Eagle's medium supplemented with
penicillin/streptomycin and 10% bovine calf serum in a 5% CO2
incubator at 37 ꢂC. IMT-E/IMT-M were dissolved in DMSO at
a concentration of 1 mM.
Apparatus and general methods
Nuclear magnetic resonance spectra (1H and 13C) were obtained
on a Bruker Avance 400 spectrometer. The HRMS spectra were
recorded on an Agilent Technologies 6510 Q-TOF LC/MS. The
UV-visible-near-IR absorption spectra of dilute solutions were
recorded on a HITACHI U-2910 spectrophotometer using
a quartz cuvette with a path length of 1 cm. One-photon uo-
rescence spectra were obtained on a HITACHI F-2700 spectro-
uorimeter equipped with a 450 W Xe lamp. TPEF was
measured on a SpectroPro300i and the pump laser beam came
from a mode-locked Ti:sapphire laser system with a pulse
duration of 220 fs, and a repetition rate of 76 MHz (Coherent
Mira900-D). PBS buffer solution (137 mM NaCl, 2.7 mM KCl, 10
mM Na2HPO4$12H2O, 2 mM NaH2PO4$2H2O, pH ¼ 7.40) and
Tris-HCl buffer solution (10 mM Tris, 100 mM KCl, pH ¼ 7.2)
were used. All pH measurements were performed with a MET-
TLER TOLEDO FE20-FiveEasy™ pH-meter. Cell viability was
assayed using a cell proliferation Kit I, with the 492 nm absor-
bance being detected using a PerkinElmer Victor plate reader.
Two-photon uorescence images were obtained using a IX71
(Olympus) multiphoton laser scanning microscope with
a Coherent Mira900-D. The dyeing rate experiment was carried
out using ow cytometry (ImageStreamX MarkII). The data were
obtained using INSPIRE soware and analyzed using IDEAS
Application v6.0 soware.
For living cell staining experiments, cultured cells grown on
glass coverslips were stained with 5 mM IMT-E/IMT-M in culture
medium for 30 min at 37 ꢂC and then imaged using uores-
cence microscopy.
For xed cell staining experiments, cultured cells grown on
glass coverslips were pretreated according to the following
procedure: cells were rst xed by 4% paraformaldehyde for
ꢂ
30 min or pre-chilled methanol (ꢁ20 C) for 15 min and were
then permeabilized by 0.5% Triton X-100 for 2 min at ambient
temperature.
For the RNase digest test, two sets of prexed HeLa cells were
stained with 5 mM IMT-E/IMT-M in PBS (pH ¼ 7.4) for 30 min.
Aer rinsing with PBS twice, a total of 1 mL PBS (as a control
experiment) was added to one group of cells and 25 mg mLꢁ1
DNase-free RNase (GE) was added to the other group of cells.
Then the two sets of cells were incubated at 37 ꢂC in 5% CO2 for
2 h. Aer rinsing with PBS twice, both two sets of cells were
imaged using wide-eld uorescence microscopy.
Cell-viability assay. The study of the effect of IMT-E/IMT-M
on cell viability was carried out using the methylthiazolyldiphenyl-
tetrazolium bromide (MTT) assay. HeLa cells growing in log phase
were seeded into 96-well plates (ca. 1 ꢀ 104 cells per well) and were
allowed to adhere for 24 h. For examining short-term cytotoxic
effects, IMT-E/IMT-M (100 mL per well) at a concentration of 5 mM
was added into the wells of the treatment group, and 100 mL per
well DMSO diluted in DMEM at a nal concentration of 0.2% was
added to the negative control group. The cells were then incubated
for 2, 4, 8, 12 and 24 h at 37 ꢂC under 5% CO2, and then the MTT
assay was used as described. Each individual cytotoxic experiment
was repeated three times.
Spectroscopic measurements
The uorescence quantum yields (F) can be calculated using
eqn (1):
2
Ar lr ns Fs
2
Fs ¼ Fr
(1)
As ls nr Fr
where the subscripts s and r refer to the sample and the refer-
ence materials, respectively. F is the quantum yield, F is the
integrated emission intensity, A stands for the absorbance, and
n is the refractive index. In this paper, the quantum yield was
calculated by using uorescein in aqueous NaOH (pH ¼ 13, F ¼
0.95) as a standard.31
Two-photon absorption (TPA) cross sections were measured
using the two-photon induced uorescence method. Rhoda-
mine B in MeOH (d ¼ 210 GM) was used as the standard,32
whose two-photon properties have been reported in the litera-
ture, and thus, cross sections can be calculated using eqn (2):
Fluorescence imaging
Wide-eld uorescence microscopy images were acquired with
an Olympus IX71 inverted microscope coupled with a CCD and
display controller soware. A confocal microscopic image and
differential interference contrast (DIC) image were taken with
a 488 nm Argon laser on an Olympus IX71 confocal microscope.
TPEF imaging was obtained using an Olympus FV 300 Laser
Confocal Microscope. In the two-photon experiments, the
excitation wavelength was 800 nm from a Ti:sapphire femto-
second laser source (Coherent Chamelon Ultra). The incident
power on the samples was modied by means of an attenuator
and the samples were examined using a Power Monitor
(Coherent). A multiphoton emission lter (FF01-750; Semrock)
Fr Cr ns Fs
ds ¼ dr
(2)
Fs Cs nr Fr
where the subscripts s and r refer to the sample and the refer- was used to block the IR laser. The differential interference
ence materials, respectively. d is the TPA cross-section value, C contrast (DIC) images were taken with a 488 nm Ar ion laser.
16732 | RSC Adv., 2017, 7, 16730–16736
This journal is © The Royal Society of Chemistry 2017

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
Paper
RSC Advances
Results and discussion
Photophysical properties of IMT-E/IMT-M
The photophysical properties of IMT-E/IMT-M in different
solvents at room temperature were measured, as shown in
Fig. S1.† Their one-photon absorption wavelengths in various
solvents are similar, mainly in the range of 350–525 nm.
Furthermore, these probes have appropriate molar extinction
coefficients (3: 6.62 ꢀ 104 cmꢁ1 Mꢁ1 for IMT-E and 3.39 ꢀ 104
cmꢁ1 Mꢁ1 for IMT-M in PBS), making them suitable for imaging
applications. In addition, the emission maximum of the two
probes in these solvents was around 520 nm. Such long-
wavelength emission is also approved for cellular imaging, to
efficiently avoid the background uorescence of endogenous
uorophores.
Subsequently, the photophysical properties of IMT-E/IMT-
M in the presence of DNA and RNA were carefully studied
(Fig. 1 and Table 1). In order to make an intuitive comparison,
the commercial RNA stain “SYTO RNA-Select” was also
studied. In the absorption and emission spectra, IMT-E and
IMT-M display a tiny shi (Fig. 1a) when binding to DNA or
RNA, while their emission uorescence intensities show
a large change. IMT-E and IMT-M are weakly uorescent in
Tris–HCl buffer, whereas their uorescence intensities show
an approximately 15-fold (IMT-E) and 10-fold (IMT-M)
enhancement in the presence of RNA, and a 10-fold (IMT-E)
and 10-fold (IMT-M) enhancement in the presence of DNA
(Fig. 1b). As shown in Table 1, in Tris–HCl buffer, the uo-
rescence quantum yields (F) of IMT-E and IMT-M are 0.062%
and 0.031%, but the values increase to 6.157% and 2.476% in
the presence of RNA, respectively. The “turn-on” behaviors are
similar to commercial SYTO RNA-Select (RNA: 35-fold, DNA:
12-fold). The obvious “turn-on” behavior can be attributed to
the restraining of intramolecular motions when binding to
DNA and RNA. Furthermore, as shown in Fig. 2, the TPEF
properties of these dyes give rather similar results with exci-
tation at 800 nm, an optimal excitation wavelength provided
by a commercial Ti:sapphire laser source. The probes exhibit
much brighter two-photon excited uorescence in the pres-
ence of RNA, than that in the presence of DNA. The d ꢀ F
values of IMT-E and IMT-M in buffer are 0.07 GM and 0.04 GM,
Fig. 2 Two-photon excited fluorescence spectra (excitation wave-
length: 800 nm) of IMT-E (Ia) and IMT-M (IIa), in the absence and
presence of RNA or DNA (error limit: 8%). The d ꢀ F values at different
wavelengths of IMT-E (Ib) and IMT-M (IIb) in the presence of RNA and
DNA (error limit: 20%). Buffer: 10 mM Tris–HCl, 100 mM KCl, pH ¼ 7.2.
Compound concentration: 2 ꢀ 10ꢁ6 mol Lꢁ1. Ratio (phosphate of NA/
dye): 500 : 1 for IMT-E and IMT-M.
while in the presence of RNA, they increase to 5.95 GM and 4.9
GM (Table 1), respectively. The d ꢀ F values at different
wavelengths of the IMT-E and IMT-M in saturated RNA or DNA
and Tris–HCl buffer are shown in Fig. 2b. From Table 1, the
maximum d values (in the presence of RNA) for IMT-E and
IMT-M are 63.7 GM and 73.7 GM, respectively. These results
indicate that these dyes show higher uorescence responses to
RNA than to DNA. It is expected that the compounds prefer to
bind to RNA over DNA when entering cells. In addition, in the
intracellular environment, acid compartments such as lyso-
somes change the intracellular pH dramatically. Therefore,
the photophysical properties of an ideal probe in various
intracellular environments should be stable. As shown in
Fig. S2,† the peak positions of the uorescence spectra of IMT-
E and IMT-M at different pH values are unchanged, and their
uorescence intensities also show very little change. This
nding demonstrates an important advantage.
Table 1 The photophysical properties of IMT-E and IMT-Ma
Sample
Solvent
l1/nm
l2/nm
l3/nm
3/Mꢁ1 cmꢁ1
F/%
d/GM
d ꢀ F/GM
IMT-E
Tris
446
460
465
460
447
453
517
519
522
520
518
517
543
533
534
541
535
534
6.62 ꢀ 104
4.83 ꢀ 104
5.34 ꢀ 104
3.39 ꢀ 104
4.39 ꢀ 104
4.54 ꢀ 104
0.177
5.691
9.334
0.092
6.151
5.481
39.5
49.1
63.7
47.6
74.4
73.7
0.07
2.79
5.95
0.04
4.58
4.90
DNA
RNA
Tris
DNA
RNA
IMT-M
a
l1, l2 and l3 are the linear absorption, and one- and two-photon emission peaks, respectively; 3 is molar absorptivity; F is the one-photon
uorescence quantum yield determined using uorescein (F ¼ 0.95) as the standard; d is two-photon absorption cross section determined
using of Rhodamine B in MeOH (d ¼ 210 GM) as the standard; d ꢀ F is two-photon action absorption cross section. 1 GM ¼ 10ꢁ50 cm4 s per
photon; concentration of samples: 2 ꢀ 10ꢁ6 mol Lꢁ1; buffer: 10 mM Tris–HCl, 100 mM KCl, pH ¼ 7.2. DNA/RNA: in Tris–HCl buffer in the
presence of DNA/RNA. Ratio: (phosphate of NA/dye): 500 : 1 for IMT-M and IMT-E. Error limits of F, d and d ꢀ F are 10%, 20% and 32%, respectively.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 16730–16736 | 16733