Genetically Encoded Dual-Color Light-Up RNA Sensor Enabled Ratiometric Imaging of MicroRNA

Article abstract of DOI:10.1021/acs.analchem.0c04588

MicroRNAs (miRNAs) play essential roles in regulating gene expression and cell fate. However, it remains a great challenge to image miRNAs with high accuracy in living cells. Here, we report a novel genetically encoded dual-color light-up RNA sensor for ratiometric imaging of miRNAs using Mango as an internal reference and SRB2 as the sensor module. This genetically encoded sensor is designed by expressing a splittable fusion of the internal reference and sensor module under a single promoter. This design strategy allows synchronous expression of the two modules with negligible interference. Live cell imaging studies reveal that the genetically encoded ratiometric RNA sensor responds specifically to mir-224. Moreover, the sensor-to-Mango fluorescence ratios are linearly correlated with the concentrations of mir-224, confirming their capability of determining mir-224 concentrations in living cells. Our genetically encoded light-up RNA sensor also enables ratiometric imaging of mir-224 in different cell lines. This strategy could provide a versatile approach for ratiometric imaging of intracellular RNAs, affording powerful tools for interrogating RNA functions and abundance in living cells.

Full text of DOI:10.1021/acs.analchem.0c04588

pubs.acs.org/ac
Article
Genetically Encoded Dual-Color Light-Up RNA Sensor Enabled
Ratiometric Imaging of MicroRNA
Cai-Xia Dou, Chaoyang Liu, Zhan-Ming Ying,* Wanrong Dong,* Fenglin Wang,* and Jian-Hui Jiang
Cite This: Anal. Chem. 2021, 93, 2534−2540
Read Online
ACCESS
Metrics & More
Article Recommendations
*
sı Supporting Information
ABSTRACT: MicroRNAs (miRNAs) play essential roles in regulating gene
expression and cell fate. However, it remains a great challenge to image
miRNAs with high accuracy in living cells. Here, we report a novel genetically
encoded dual-color light-up RNA sensor for ratiometric imaging of miRNAs
using Mango as an internal reference and SRB2 as the sensor module. This
genetically encoded sensor is designed by expressing a splittable fusion of the
internal reference and sensor module under a single promoter. This design
strategy allows synchronous expression of the two modules with negligible
interference. Live cell imaging studies reveal that the genetically encoded
ratiometric RNA sensor responds specifically to mir-224. Moreover, the
sensor-to-Mango fluorescence ratios are linearly correlated with the
concentrations of mir-224, confirming their capability of determining mir-
2
24 concentrations in living cells. Our genetically encoded light-up RNA
sensor also enables ratiometric imaging of mir-224 in different cell lines. This
strategy could provide a versatile approach for ratiometric imaging of intracellular RNAs, affording powerful tools for interrogating
RNA functions and abundance in living cells.
icroRNAs (miRNAs) are short non-coding RNAs that
play crucial roles in regulating gene expression and cell
,2
independent factors such as instrumental parameters and
sensor concentrations. We previously reported a genetically
encoded ratiometric RNA sensor for quantitative imaging
mRNA, coexpressing green fluorescent protein (GFP) mRNA
M
1
fate. Methods that can provide information about their
abundance, distribution, and dynamics are essential to
understand their roles and functions. Nevertheless, it remains
a great challenge to detect miRNAs in living cells due to their
small size, intrinsic non-fluorescent nature, and low abun-
18
under a single promoter as a reference. However, the
requirements of translation and maturation render GFP a less-
desirable reference. More recently, a genetically encoded
ratiometric RNA sensor utilizing two RNA aptamers,
dinitroaniline (DN) and Broccoli, was developed to quantify
3
,4
dance.
Hybridization-based probes including molecular
beacons and spherical nucleic acids have been developed to
5
−7
19,20
image miRNAs in living cells with high sensitivity.
small molecules in bacteria.
However, the development of
However, the difficulty in delivery and susceptibility to
a genetically encoded RNA sensor based on two RNA
aptamers for ratiometric imaging of RNA in mammalian cells
remains unexplored.
Here, we develop a genetically encoded dual-color RNA
sensor for ratiometric imaging of miRNAs in mammalian cells.
We envision that the utility of two light-up RNA aptamers, one
functions as an internal reference and the other functions as a
sensor module, could allow the development of a genetically
encoded ratiometric RNA sensor. Based on this rationale, we
choose two light-up RNA aptamers, Mango and SRB2, which
degradation may limit their applications in determining
8
miRNAs in living cells. Genetically encoded RNA sensors
that can be transcribed inside living cells could potentially
overcome the issues of inefficient intracellular delivery and
probe degradation.
Light-up RNA aptamers are in vitro selected RNA sequences
that can specifically bind and activate the fluorescence of
9
,10
cognate non-fluorescent small molecule fluorophores.
A
1
1
palette of light-up RNA aptamers including Broccoli,
1
2
13
14
Mango,
SRB2,
and Pepper
have been reported.
Incorporation of the analyte-responsive motif to the light-up
RNA aptamers has enabled the development of genetically
encoded fluorogenic RNA sensors for imaging various analytes
including small molecules, proteins, and RNAs in living
cells. However, current designs mainly rely on a single-
wavelength readout that prevents them from quantitatively
imaging analytes due to variations from other analyte-
Received: October 30, 2020
Accepted: January 8, 2021
Published: January 19, 2021
15
16
17
©
2021 American Chemical Society
https://dx.doi.org/10.1021/acs.analchem.0c04588
2
534
Anal. Chem. 2021, 93, 2534−2540

Analytical Chemistry
pubs.acs.org/ac
Article
can specifically bind and activate the fluorescence of thiazole
China). RNA oligonucleotides were synthesized by TaKaRa
Biotechnology (Dalian, China). A mir-224 mimic and inhibitor
were obtained from RiboBio (Guangzhou, China). Oligonu-
cleotide sequences are shown in the Supporting Information in
Table S1. Restriction endonucleases and a Monarch RNA
cleanup kit (50 μg) were purchased from New England Biolabs
(Massachusetts, USA). Recombinant human β-actin protein
was purchased from R&D Systems (Minneapolis, MN). Cell
lysis buffer was purchased from Beyotime Biotechnology
12
orange biotin (TO1-Biotin) and sulforhodamine B (SR)-
1
3
DN, respectively. Mango and SRB2 exhibit distinct spectral
properties after binding with their corresponding fluorophores,
facilitating dual-color imaging. A genetically encoded RNA
sensor is designed by inserting a dsRNA Dicer substrate
21
(
dsRNA) that can be cleaved by endogenous Dicer enzyme
between the reference and the sensor module under a single
promoter (Figure 1). This design allows a synchronous
(
Shanghai, China). HepG2, HeLa, and L-02 cells were
purchased from the cell bank of Central Laboratory of Xiangya
Hospital (Changsha, China). All other chemicals were of
analytical grade and obtained from Sinopharm Chemical
Reagent (Shanghai, China). All solutions and ultrapure water
were autoclaved and treated with diethyl pyrocarbonate to
avoid RNA degradation.
The fluorescence assay was performed using an F-7000
fluorescence spectrophotometer (Hitachi, Japan). A Tanon
4200SF gel imaging system (Tanon Science & Technology,
China) was used to visualize the gel images. Mass spectrometry
was collected on an Agilent Technologies accurate mass Q-
TOF 6530 microspectrometer. Quantitative reverse tran-
scription PCR (qRT-PCR) was carried out using an Applied
Biosystems StepOnePlus PCR instrument (ABI, Foster, CA,
USA). Confocal laser scanning microscopy (CLSM) images
were acquired on a Nikon TI*E + A1 SI confocal laser
scanning microscope (Japan) using an oil objective lens (60×).
Synthesis of TO1-Biotin Conjugate. TO1-Biotin con-
jugate was synthesized according to previous procedures with
Figure 1. Scheme for the working principle of the genetically encoded
dual-color light-up RNA sensor. The expressed Mango-dsRNA-sensor
transcript is cleaved by Dicer to separate the reference and sensor
module. The reference binds and activates the fluorescence of TO1-
Biotin, whereas the sensor module hybridizes with the target miRNA
and restores the correct confirmation of SRB2 with activated SR-DN
fluorescence. The fluorescence ratio of SR-DN over TO1-Biotin
enables ratiometric imaging of the target miRNA.
12
slight modifications. Specifically, 2-(2-methylbenzo[d]-
thiazol-3-ium-3-yl)acetate was first prepared by fluxing 2-
methylbenzothiazole (1 equiv) and bromoacetic acid (1.5
equiv) in toluene for 12 h. The resulting mixture was
condensed, filtrated, and washed with toluene to afford 2-(2-
methylbenzo[d]thiazol-3-ium-3-yl)acetate with a yield of 95%.
Then, 1-methylquinolinium iodide was obtained by refluxing
quinoline (1 equiv) and iodomethane (2.1 equiv) in 1,4-
dioxane for 1 h (50−90% yield). TO1-acetate was synthesized
by stirring 1-methylquinolinium iodide (1 equiv), 2-(2-
methylbenzo[d]thiazol-3-ium-3-yl)acetate (1.4 equiv), and
triethylamine (3.1 equiv) in dichloromethane at room
temperature overnight. The resulting mixture was precipitated
with a mixture of ethanol and diethyl ether (4:1, v/v). The
precipitates were filtrated and washed with water and acetone
to afford TO1-acetate with a yield of 30%. Finally, TO1-Biotin
was prepared by stirring a mixture composed of benzotriazole-
N,N,N′,N′-tetramethyl-uronium-hexafluorophosphate
(HBTU; 14 equiv), N,N-diisopropylethylamine (DIPEA; 28
equiv), TO1-acetate (1.0 equiv), and EZ-link amine-PEG-
biotin (4.2 equiv) overnight at room temperature. The
ultimate product, TO1-Biotin, was purified and characterized
by high-resolution−electrospray ionization−mass spectrometry
(HR-ESI-MS).
expression of the reference and the sensor module. Moreover,
the dsRNA enables separation of these modules by
endogenous Dicer to prevent their interaction. After expression
and Dicer-mediated cleavage, the reference module can bind
and activate the fluorescence of TO1-Biotin, whereas the
sensor module hybridizes with the target miRNA and restores
the correct confirmation of SRB2 with activated SR-DN
fluorescence. The fluorescence ratios of the sensor over Mango
allow ratiometric imaging of the target RNAs.
To demonstrate the concept, we choose mir-224 that is
overexpressed in various cancers, especially hepatocellular
2
2,23
carcinoma.
The sensor module is constructed by
incorporating the mir-224-responsive moiety to the SR-binding
24
loop according to our previous design strategy. Fluorescence
imaging reveals that the Mango-dsRNA-sensor responds
specifically to mir-224, and the senor-to-Mango ratios are
linearly correlated with the concentrations of mir-224
determined by quantitative reverse transcription polymerase
chain reaction (qRT-PCR). Moreover, this genetically encoded
dual-color RNA sensor also allows ratiometric imaging of mir-
Fluorescence Assay. The in vitro transcribed RNA
aptamers including 10 μM Mango, 5 μM SRB2, or 10 μM
the fusion of Mango and SRB2 (M-S) were incubated with the
corresponding fluorophore (10 μM TO1-Biotin or 1 μM SR-
DN) in a reaction buffer (40 mM HEPES, 100 mM KCl, and
2
24 in different cell lines. This strategy affords a general
platform for ratiometrically imaging intracellular RNAs,
providing versatile tools for probing RNA functions and
dynamics in living cells.
10 mM MgCl ) for 30 min at 37 °C. Fluorescence spectra for
2
TO1-Biotin were collected in the range of 515−615 nm with
an excitation wavelength of 495 nm, whereas those for SR-DN
were collected in the range of 565−650 nm with an excitation
wavelength of 550 nm.
EXPERIMENTAL SECTION
Materials and Instruments. DNA oligonucleotides
HPLC) were synthesized from Sangon Biotech (Shanghai,
■
(
2
535
https://dx.doi.org/10.1021/acs.analchem.0c04588
Anal. Chem. 2021, 93, 2534−2540

Analytical Chemistry
pubs.acs.org/ac
Article
To study the response of the sensor module to mir-224, a
given concentration of mir-224 was incubated with the sensor
module (5 μM) and 1 μM SR-DN in the reaction buffer for 30
min at 37 °C.
The selectivity study was carried out by incubating different
substances including mir-224, mir-21, Let-7f, mir-222, β-actin,
and L-02 cell lysates with the RNA sensor module and 1 μM
SR-DN in the reaction buffer for 30 min at 37 °C. L-02 cell
lysates were obtained according to the following procedures:
after washing, L-02 was treated with 200 μL of cell lysis buffer
for 5 min on ice. The mixture was centrifuged at 10,000g at 4
being washed with PBS three times. The fluorescence for TO1-
Biotin was excited with a 488 nm laser with emission collected
in the range of 495−550 nm. The fluorescence for SR-DN was
excited with a 561 nm laser with the emission collected in the
range of 570−640 nm. For data analysis, the average
fluorescence intensity of TO1-Biotin and SR-DN was
calculated by choosing 10 region of interests (ROIs) in the
cytoplasm of the cells in the same field of view.
RESULTS AND DISCUSSION
■
Synthesis and Characterization of TO1-Biotin Con-
jugate. To develop a genetically encoded dual-color light-up
RNA sensor, we chose Mango and SRB2 aptamers attributed
to their compact sizes and distinct spectral properties of their
°
C for 5 min, and the supernatant was collected.
Gel Imaging Assay. To test whether the RNA transcripts
could bind with the corresponding fluorophores, 10 μL of 1
μM Mango, SRB2, or M-S was loaded into a well of 12%
polyacrylamide gel electrophoresis (PAGE) gel. To determine
whether the RNA sensor module could bind with SR-DN after
reacting with mir-224, the RNA sensor module was incubated
with 1 μM mir-224 at 37 °C for 30 min. The reaction mixture
was loaded into a well of 12% PAGE gel. For comparison, the
RNA sensor was also loaded in another well. All electro-
phoresis experiments were performed at a constant potential of
12,13
cognate fluorophores.
We first synthesized TO1-Biotin, the
ligand for Mango, and SR-DN for SRB2 was previously
24
developed in our lab. TO1-Biotin was synthesized by
conjugating benzothiazole with methylquinoline, which was
further derived with biotin via a linker. The compound was
+
characterized with HR-ESI-MS (calculated for [M + H] m/z
7
49.3155, found 749.3136) (Figure S1). PAGE analysis
showed that Mango especially stained by the TO1-Biotin
conjugate was also colored by 4S GelRed, a nucleic acid dye,
confirming that TO1-Biotin could form a complex with
Mango. Its cell permeability was investigated by incubating
HeLa cells with varying concentrations of TO1-Biotin.
Confocal laser scanning microscopy (CLSM) imaging showed
negligible fluorescence for cells treated with 10 μM TO1-
Biotin, which gradually increased as its concentration increased
from 10 to 50 μM (Figure S2). Further, z-stacked projection
analysis showed typical cytosol distribution of TO1-Biotin for
cells treated with 50 μM TO1-Biotin. These results
demonstrated good cell permeability of TO1-Biotin. We
chose 10 μM TO1-Biotin for subsequent intracellular studies
due to its low intracellular background.
9
0 V in TBE buffer (pH 8.3) for 1 h. The PAGE gels were
stained with 1 μM SR-DN or 10 μM TO1-Biotin followed by
staining with a 4S GelRed nucleic acid stain. The gel was
imaged with a Tanon 4200SF gel imaging system.
Vector Construction. The plasmid was constructed using
pcDNA3.1(+) as a mammalian expression vector. For
improving the stability of RNA, the RNA aptamers and RNA
sensor were inserted into a tRNA scaffold. Overlapping PCR
was utilized to obtain the DNA sequences for tRNA-scaffolded
Mango-SRB2, Mango-dsRNA-SRB2, Mango-dsRNA-sensor,
and Mango-dsRNA-control flanked by Hind III and PmeI
cleavage sites. DNA sequences are listed in Table S1 in the
Supporting Information. The PCR products were purified
using a PureLink quick gel extraction kit (Thermo Scientific),
according to the manufacturer’s protocol. The purified
products were cloned into the pcDNA3.1(+) vector under a
CMV promoter via the Hind III and PmeI restriction sites to
construct plasmids of Mango-SRB2, Mango-dsRNA-SRB2,
Mango-dsRNA-sensor, and Mango-dsRNA-control, respec-
tively. The resulting plasmids were confirmed by sequencing.
Cell Culture and Transfection. HepG2, HeLa, and L-02
cell lines were cultured in a 37 °C incubator with 5% CO2
using Dulbecco’s modified Eagle medium (DMEM; Thermo
Scientific Hyclone) added with 10% fetal bovine serum. The
cells were plated on 35 mm sterilized dishes with 14 mm wells.
Upon reaching a confluence of 40−60%, the cells were
transfected with different plasmids using Lipofectamine 3000
Characterization of Dual-Color RNA Aptamers. The
potential to construct a dual-color light-up RNA sensor using
Mango and SRB2 was first explored. The transcripts of Mango,
SRB2, and a fusion of Mango and SRB2 (M-S) were
transcribed in vitro. TO1-Biotin was essentially non-emissive,
but its fluorescence at ∼540 nm was dramatically increased in
the presence of Mango (Figure 2A). Similarly, SR-DN
exhibited negligible fluorescence but its fluorescence at ∼600
nm was remarkably enhanced in the presence of SRB2 (Figure
2B). These results revealed that Mango and SRB2 possessed
distinct spectral properties, holding great potentials for dual-
color fluorescence imaging. Surprisingly, there was no
fluorescence enhancement for TO1-Biotin or SR-DN when
incubated with the M-S transcripts, which might be due to the
proximal interaction between Mango and SRB2 that hinders
their correct folding. Gel electrophoresis showed consistent
results that majority of M-S transcripts could not form
complexes with TO1-Biotin, while M-S transcripts formed
negligible complexes with SR-DN (Figure 2C,D and Figure
S3). To reduce the interaction between Mango and SRB2 in
vivo, a dsRNA Dicer substrate, which can be cleaved by
(
Thermo Scientific), according to the manufacturer’s protocol.
Typically, the cells were transfected with 500 ng of the plasmid
and 2 μL of Lipofectamine 3000 in OptiMEM. The cells were
then cultured in DMEM supplied with 10% FBS for 6 h before
changing to a complete medium. CLSM images were obtained
2
4 h after transfection. To modulate the expression levels of
mir-224 in HeLa cells, mir-224 mimics or inhibitors at a given
dose were cotransfected with the plasmid Mango-dsRNA-
sensor.
21
abundant endogenous Dicer enzyme, can be introduced to
separate these two aptamers. We constructed plasmid Mango-
dsRNA-SRB2 to express tRNA-scaffolded Mango-dsRNA-
SRB2 under a CMV promoter (Figure S4). Gel electrophoresis
analysis and positive clone sequencing confirmed its successful
construction. To confirm that the transcript could be cleaved
by Dicer enzyme in vivo, reverse transcription PCR (RT-PCR)
CLSM Imaging and Data Analysis. The cells in different
conditions were washed with PBS three times and incubated
with a DMEM medium containing 10 μM TO1-Biotin, 1 μM
SR-DN conjugate, 40 mM HEPES, 100 mM KCl, and 10 mM
MgCl for 30 min at 37 °C. CLSM images were obtained after
2
2
536
https://dx.doi.org/10.1021/acs.analchem.0c04588
Anal. Chem. 2021, 93, 2534−2540

Analytical Chemistry
pubs.acs.org/ac
Article
Figure 2. (A) Fluorescence spectra of TO1-Biotin in the presence of
Mango and M-S transcripts (a fusion of Mango and SRB2). (B)
Fluorescence spectra of SR-DN in the presence of SRB2 and M-S
transcripts. (C) PAGE analysis for Mango and M-S stained by TO1-
Biotin and 4S GelRed. (D) PAGE analysis for SRB2 and M-S stained
by SR-DN and 4S GelRed, respectively. The lengths for Mango,
SRB2, and M-S are 31, 50, and 91 nucleotides, respectively. M: DNA
marker (25−500 bp).
Figure 3. (A) CLSM images for HeLa cells transfected with plasmid
Mango-dsRNA-SRB2 at different times. (B) Average fluorescence
intensity of Mango and SRB2 in 10 ROIs.
In Vitro Characterization of the Sensing Module. The
feasibility of constructing an RNA sensor using SRB2 as the
sensing module was first studied in vitro. We chose mir-224 as
the target miRNA as it was overexpressed in various types of
22,23
of the total RNA extracts from HeLa cells transfected with
plasmid Mango-dsRNA-SRB2 was performed using three
primers designed for Mango, SRB2, and Mango-dsRNA-
SRB2. Agarose gel electrophoresis analysis of the RT-PCR
products revealed that the bands of Mango and SRB2 were
significantly more intense than that for Mango-dsRNA-SRB2,
confirming that the transcript of Mango-dsRNA-SRB2 was
efficiently cleaved by intracellular Dicer enzyme (Figure S5).
Further, CLSM imaging showed that HeLa cells transfected
with plasmid Mango-dsRNA-SRB2 displayed strong fluores-
cence for TO1-Biotin and SR-DN (Figure S6). In contrast,
cells transfected with a control plasmid expressing Mango-
SRB2 without a Dicer substrate exhibited very dim
fluorescence for TO1-Biotin and SR-DN, indicating that the
separation of Mango and SRB2 was essential for fluorescence
activation. In addition, z-stack images revealed that the
fluorescence of TO1-Biotin and SR-DN was distributed
throughout the cytoplasm (Figure S7). Together, these results
demonstrated that Mango and SRB2 hold the potential for
developing genetically encoded dual-color RNA sensors.
The capability of developing a reliable ratiometric sensor
using Mango and SRB2 was then investigated. The SRB2-to-
Mango ratios were analyzed for cells transfected with plasmid
Mango-dsRNA-SRB2 at different time intervals. CLMS
imaging showed that there was intense fluorescence from
SRB2 and Mango 24 h post-transfection, which remained
stable up to 48 h followed by a slight decrease of 60 h after
transfection (Figure 3A). Time-dependent changes in the
expression levels of SRB2 and Mango were confirmed by qRT-
PCR analysis of the total RNA extracts from cells transfected
with the plasmid Mango-dsRNA-SRB2 at different times
tumors including hepatocellular carcinoma.
The sensing
module was constructed by incorporating the mir-224-
responsive stem-loop motif to the SR binding loop of SRB2.
There was negligible SRB2 fluorescence in the absence of mir-
224, indicating that the integration of the mir-224-responsive
motif disrupted the correct formation of SRB2 (Figure 4A).
The fluorescence at 600 nm was enhanced by ∼5-fold upon
incubation with 5 μM mir-224. These results demonstrated
that the sensing module was responsive to mir-224. PAGE
analysis gave direct evidence that SR-DN was only capable of
staining the sensing module in the presence of mir-224 (Figure
S9). Further investigation showed that the fluorescence for SR-
Figure 4. (A) Fluorescence responses for the sensor module: SR-DN:
(
Figure S8). Nonetheless, further analysis showed that the
average ratios of SRB2 to Mango in 10 region of interests
ROIs) in the CLSM images remained constant from 24 to 60
1
μM SR-DN; blank: 1 μM SR-DN plus 5 μM sensor module; target
RNA: 1 μM SR-DN and 5 μM sensor module plus 5 μM mir-224. (B)
Selectivity study for the RNA sensor module. A fluorescence intensity
at 600 nm was plotted against different substances. (C) CLSM images
for cells transfected with the plasmid Mango-dsRNA-sensor for 24 h.
(D) Mean fluorescence intensity of Mango and SRB2 in 10 ROIs in
panel (C).
(
h after transfection (Figure 3B). Together, these results
demonstrated that the SRB2-to-Mango ratios were independ-
ent of their expression levels, affording robust ratiometric
imaging.
2
537
https://dx.doi.org/10.1021/acs.analchem.0c04588
Anal. Chem. 2021, 93, 2534−2540

Analytical Chemistry
pubs.acs.org/ac
Article
DN increased dynamically as the concentrations of mir-224
changed from 1 nM to 5 μM (Figure S10). Additional
selectivity studies revealed that the sensing module exhibited
negligible fluorescence enhancement upon incubation with
other intracellular components, including RNAs (mir-21,
Let7f, and mir-222), proteins (β-actin), and L-02 cell lysate
(
Figure 4B). Together, these results demonstrated that the
constructed sensing module was capable of detecting mir-224
with high sensitivity and specificity in vitro.
Ratiometric Imaging of MicroRNA. To develop a
genetically encoded ratiometric sensor for mir-224, we
constructed a plasmid expressing the reference, a Dicer
substrate, and the sensing module (Mango-dsRNA-sensor)
under a single CMV promoter. The capability of the Mango-
dsRNA-sensor to image mir-224 was demonstrated in two cell
lines, L-02 (with a negative expression of mir-224) and HeLa
22,25
cells (with an overexpression of mir-224).
L-02 and HeLa
cells transfected with the plasmid Mango-dsRNA-sensor
exhibited bright fluorescence of TO1-Biotin in the cytosol
(
Figure 4C,D). By contrast, there was negligible SRB2
fluorescence for transfected L-02 cells, whereas transfected
HeLa cells displayed intense SRB2 fluorescence in the cytosol.
These results indicated that the sensing module was responsive
to the expression levels of mir-224 in living cells, and Mango
could function as a reference. Additionally, HeLa cells
transfected with a plasmid expressing Mango, a Dicer substrate,
and a control loop that was not complementary to mir-224
exhibited negligible fluorescence from SR-DN (Figure S11).
These results demonstrated that the Mango-dsRNA-sensor
responded to mir-224 in living cells with high specificity.
The capability of the Mango-dsRNA-sensor for ratiometric
imaging of mir-224 was then investigated using HeLa cells
treated with different concentrations of the mir-224 mimic and
mir-224 inhibitor. It was known that RNA mimics can increase
the levels of mature miRNAs in living cells, whereas the
miRNA inhibitors can reduce the miRNA concentrations via
Figure 5. (A) CLSM images of HeLa cells transfected with the
plasmid Mango-dsRNA-sensor under different conditions. (a) 250
nM mir-224 mimic. (b) 50 nM mir-224 mimic. (c) Plasmid only. (d)
50 nM mir-224 inhibitor. (e) 250 nM mir-224 inhibitor. (B) Relative
concentrations of mir-224 obtained from qRT-PCR under different
conditions. (C) Liner correlation between mir-224 expression levels
and average sensor-to-Mango ratios in 10 ROIs.
26
specific hybridization. CLMS images showed that there was
evident and constant Mango fluorescence for cells in different
conditions, suggesting that it is a reliable reference (Figure
5
A). By contrast, there was varied SRB2 fluorescence for cells
in different conditions. Cells treated with mir-224 mimics
displayed stronger SRB2 fluorescence than the control cells
without treatment, whereas cells treated with the mir-224
inhibitor showed weaker SRB2 fluorescence. Moreover, cells
treated with higher concentrations of mir-224 mimics exhibited
stronger SRB2 fluorescence, while those treated with higher
concentrations of the mir-224 inhibitor delivered lower SRB2
fluorescence. Further ratiometric images using the sensor-to-
Mango fluorescence ratios showed a distinct color shift for cells
under different treatments. Moreover, the average of sensor-to-
Mango fluorescence ratios in 10 ROIs was linearly correlated
with the relative concentrations of mir-224, as determined by
qRT-PCR using U6 as a reference RNA (Figure 5B,C).
Collectively, these results demonstrated that the Mango-
dsRNA-sensor was capable of ratiometrically imaging mir-224
in living cells.
Figure 6. (A) CLSM images of different cell lines transfected with the
plasmid Mango-dsRNA-sensor. (B) qRT-PCR analysis of mir-224 for
different cell lines. The red, blue, and black lines denote the cell line
of HepG2, HeLa, and L-02, respectively. (C) Correlation of relative
concentrations calculated based on sensor-to-Mango ratios in panel
(
A) using the calibration curve in Figure 5C (x axis) and those
determined by qRT-PCR analysis in panel (B) using U6 as a reference
y axis).
The Mango-dsRNA-sensor was further utilized to image
mir-224 in different cells. Three cell lines, HepG2, HeLa, and
L-02, with different expression levels of mir-224 were chosen.
There was bright fluorescence from Mango for all the cells,
whereas the fluorescence for SRB2 varied in different cells
(
fluorescence. By contrast, the SRB2 fluorescence in L-02 was
very dim. These results demonstrated that HepG2, HeLa, and
L-02 cells had decreased mir-224 levels, which were consistent
(
Figure 6A). HepG2 cells displayed the most intense SRB2
22,25
fluorescence, while HeLa cells showed intermediate SRB2
with the previous studies
and qRT-PCR results (Figure
2
538
https://dx.doi.org/10.1021/acs.analchem.0c04588
Anal. Chem. 2021, 93, 2534−2540

Analytical Chemistry
B). The relative expression levels of mir-224 in different cell
pubs.acs.org/ac
Article
6
Engineering, Hunan University, Changsha, Hunan 410082,
China
Chaoyang Liu − State Key Laboratory of Chemo/Bio-Sensing
and Chemometrics, College of Chemistry and Chemical
Engineering, Hunan University, Changsha, Hunan 410082,
China
Jian-Hui Jiang − State Key Laboratory of Chemo/Bio-Sensing
and Chemometrics, College of Chemistry and Chemical
Engineering, Hunan University, Changsha, Hunan 410082,
China; orcid.org/0000-0003-1594-4023
lines were calculated based on the calibration curve in Figure
C using the average sensor-to-Mango ratios and determined
5
based on the qRT-PCR results using U6 as the internal
reference (Figure S12). It was revealed that the relative
concentrations of mir-224 determined by utilizing the
calibration curve in Figure 5C are highly correlated with
those obtained from qRT-PCR (Figure 6C). These results
further demonstrated that our genetically encoded ratiometric
sensor could accurately image the levels of mir-244 in different
cells.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.analchem.0c04588
CONCLUSIONS
■
We developed a novel genetically encoded dual-color light-up
RNA sensor using Mango as the reference and SRB2 as the
sensing module for ratiometric imaging of miRNA in living
mammalian cells. We demonstrated that a direct fusion of
SRB2 and Mango failed to show fluorescence enhancement for
both RNA aptamers in vitro and in living cells. Splitting of the
RNA aptamers by integration of a Dicer substrate restored the
fluorescence for both SRB2 and Mango. We demonstrated that
the fluorescence of SRB2-to-Mango ratios remained constant
at different times upon transfection, affording a reliable
ratiometric imaging platform. This genetically encoded sensor
was successfully utilized to ratiometrically imaging mir-224 in
tumor cells under different treatments. It was revealed that the
sensor-to-Mango fluorescence ratios were linearly correlated
with the concentrations of mir-224 obtained from qRT-PCR.
The developed sensor also allowed ratiometric imaging of mir-
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was financially supported by the National Natural
Science Foundation of China (21904034, 21806033,
21974041, and 91753107) and the Natural Science Founda-
tion of Hunan Province (2017JJ3027).
REFERENCES
■
(
(
(
(
1) Bartel, D. P. Cell 2004, 116, 281−297.
2) Vidigal, J. A.; Ventura, A. Trends Cell Biol. 2015, 25, 137−147.
3) Baker, M. Nat. Methods 2012, 9, 787−790.
4) Qing, Z.; Xu, J.; Hu, J.; Zheng, J.; He, L.; Zou, Z.; Yang, S.; Tan,
W.; Yang, R. Angew. Chem., Int. Ed. 2019, 58, 11574−11585.
2
24 in different cell lines. Our design may afford a robust
(5) Xue, C.; Zhang, S.; Li, C.; Yu, X.; Ouyang, C.; Lu, Y.; Wu, Z. S.
platform for developing genetically encoded light-up RNA
sensors for quantitative imaging of RNAs in living cells.
Anal. Chem. 2019, 91, 15678−15685.
(6) Li, L.; Feng, J.; Liu, H.; Li, Q.; Tong, L.; Tang, B. Chem. Sci.
2
016, 7, 1940−1945.
7) Rouge, J. L.; Sita, T. L.; Hao, L.; Kouri, F. M.; Briley, W. E.;
Stegh, A. H.; Mirkin, C. A. J. Am. Chem. Soc. 2015, 137, 10528−
0531.
8) Bao, G.; Rhee, W. J.; Tsourkas, A. Annu. Rev. Biomed. Eng. 2009,
(
ASSOCIATED CONTENT
sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.analchem.0c04588.
■
*
1
(
1
(
(
1, 25−47.
9) Paige, J. S.; Wu, K. Y.; Jaffrey, S. R. Science 2011, 333, 642−646.
Experimental method, DNA and RNA sequence, scheme
of plasmid expression, characterization of TO1-Biotin,
plasmid map, gel electrophoresis analysis, fluorescence
spectra, CLSM images, and additional data (PDF)
10) Neubacher, S.; Hennig, S. Angew. Chem., Int. Ed. 2019, 58,
1
(
266−1279.
11) Filonov, G. S.; Moon, J. D.; Svensen, N.; Jaffrey, S. R. J. Am.
Chem. Soc. 2014, 136, 16299−16308.
12) Dolgosheina, E. V.; Jeng, S. C. Y.; Panchapakesan, S. S. S.;
(
Cojocaru, R.; Chen, P. S. K.; Wilson, P. D.; Hawkins, N.; Wiggins, P.
AUTHOR INFORMATION
Corresponding Authors
Zhan-Ming Ying − State Key Laboratory of Chemo/Bio-
Sensing and Chemometrics, College of Chemistry and
Chemical Engineering, Hunan University, Changsha, Hunan
■
A.; Unrau, P. J. ACS Chem. Biol. 2014, 9, 2412−2420.
(
13) Sunbul, M.; Jaschke, A. Angew. Chem., Int. Ed. 2013, 52,
̈
13401−13404.
(
14) Chen, X.; Zhang, D.; Su, N.; Bao, B.; Xie, X.; Zuo, F.; Yang, L.;
Wang, H.; Jiang, L.; Lin, Q.; Fang, M.; Li, N.; Hua, X.; Chen, Z.; Bao,
C.; Xu, J.; Du, W.; Zhang, L.; Zhao, Y.; Zhu, L.; Loscalzo, J.; Yang, Y.
Nat. Biotechnol. 2019, 37, 1287−1293.
4
10082, China; Email: zmying@hnu.edu.cn
Wanrong Dong − State Key Laboratory of Chemo/Bio-
Sensing and Chemometrics, College of Chemistry and
Chemical Engineering, Hunan University, Changsha, Hunan
(15) Su, Y.; Hickey, S. F.; Keyser, S. G. L.; Hammond, M. C. J. Am.
Chem. Soc. 2016, 138, 7040−7047.
16) Song, W.; Strack, R. L.; Jaffrey, S. R. Nat. Methods 2013, 10,
73−875.
17) Cawte, A. D.; Unrau, P. J.; Rueda, D. S. Nat. Commun. 2020,
1, 1283.
18) Ying, Z. M.; Yuan, Y. Y.; Tu, B.; Tang, L. J.; Yu, R. Q.; Jiang, J.
H. Chem. Sci. 2019, 10, 4828−4833.
19) Wu, R.; Karunanayake Mudiyanselage, A. P. K. K.; Shafiei, F.;
(
8
(
1
4
10082, China; orcid.org/0000-0002-6541-2853;
Email: wanrongdong@hnu.edu.cn
Fenglin Wang − State Key Laboratory of Chemo/Bio-Sensing
and Chemometrics, College of Chemistry and Chemical
Engineering, Hunan University, Changsha, Hunan 410082,
China; orcid.org/0000-0002-5872-2693;
(
(
Email: fengliw@hnu.edu.cn; Fax: +86-731-88821916
Zhao, B.; Bagheri, Y.; Yu, Q.; McAuliffe, K.; Ren, K.; You, M. Angew.
Chem., Int. Ed. 2019, 58, 18271−18275.
Authors
(20) Wu, R.; Karunanayake Mudiyanselage, A. P. K. K.; Ren, K.;
Cai-Xia Dou − State Key Laboratory of Chemo/Bio-Sensing
and Chemometrics, College of Chemistry and Chemical
Sun, Z.; Tian, Q.; Zhao, B.; Bagheri, Y.; Lutati, D.; Keshri, P.; You, M.
ACS Appl. Bio Mater. 2020, 3, 2633−2642.
2
539
https://dx.doi.org/10.1021/acs.analchem.0c04588
Anal. Chem. 2021, 93, 2534−2540

Analytical Chemistry
pubs.acs.org/ac
Article
(
3
(
21) Sinha, N. K.; Iwasa, J.; Shen, P. S.; Bass, B. L. Science 2018, 359,
29−334.
22) Wang, Y.; Lee, A. T. C.; Ma, J. Z. I.; Wang, J.; Ren, J.; Yang, Y.;
Tantoso, E.; Li, K.-B.; Ooi, L. L. P. J.; Tan, P.; Lee, C. G. L. J. Biol.
Chem. 2008, 283, 13205−13215.
(23) Murakami, Y.; Yasuda, T.; Saigo, K.; Urashima, T.; Toyoda, H.;
Okanoue, T.; Shimotohno, K. Oncogene 2006, 25, 2537−2545.
24) Ying, Z.-M.; Wu, Z.; Tu, B.; Tan, W.; Jiang, J.-H. J. Am. Chem.
Soc. 2017, 139, 9779−9782.
25) Liao, X.; Wang, Q.; Ju, H. Chem. Commun. 2014, 50, 13604−
3607.
26) Li, J.; Tan, S.; Kooger, R.; Zhang, C.; Zhang, Y. Chem. Soc. Rev.
014, 43, 506−517.
(
(
1
(
2
2
540
https://dx.doi.org/10.1021/acs.analchem.0c04588
Anal. Chem. 2021, 93, 2534−2540