Small-molecule fluorescent probes for specific RNA targets

Article abstract of DOI:10.1039/c1cc10393h

A method was developed that uses small molecules as fluorescent probes to detect specific mRNAs. In this approach, the fluorescence of fluorophore-quencher conjugates is restored by the binding of an mRNA aptamer tag to the quencher segment of the molecules. The method allows real-time detection of mRNA transcripts in vitro. The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c1cc10393h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 4712–4714
www.rsc.org/chemcomm
COMMUNICATION
Small-molecule fluorescent probes for specific RNA targetsw
a
a
Asako Murata,z Shin-ichi Sato,z Yoshinori Kawazoeb and Motonari Uesugi*ab
Received 20th January 2011, Accepted 25th February 2011
DOI: 10.1039/c1cc10393h
A method was developed that uses small molecules as fluorescent
probes to detect specific mRNAs. In this approach, the fluorescence
of fluorophore–quencher conjugates is restored by the binding
of an mRNA aptamer tag to the quencher segment of the
molecules. The method allows real-time detection of mRNA
transcripts in vitro.
Real-time visualization of mRNAs in living cells would permit
detailed analysis of mRNA dynamics during biological
processes.1 A number of powerful methods for visualizing
RNAs have been developed, which allow real-time imaging
of specific RNA targets. In one method, cells are transfected
with molecular beacons, whose fluorescence intensity changes
upon binding to endogenous mRNA targets.2 Another method
uses GFP-tagged proteins that bind to specific RNA tags.3
Although these methods have provided valuable information
about mRNA dynamics in living cells, methods based on
nucleic acids or proteins have the respective drawbacks of
poor cell permeability and a low signal-to-noise ratio. An
alternative strategy using small molecules would avoid these
problems.
Fig. 1 Functional design of the fluorophore–BHQ1 conjugates.
Several groups have reported small molecule-based
methods, in which a short RNA aptamer serves as a tag that
binds to a non-fluorescent dye, whereupon the dye becomes
fluorescent.4 The advantages of this approach are that small
molecules can be added to the sample at any time point, and
co-expression of proteins that detect specific RNA tags is not
required. However, the design of fluorogenic probes with
desirable properties is challenging, and imaging of mRNA in
live cells using small molecules has not previously been
achieved.
Fig. 2 Structures of fluorophore–BHQ1 conjugates 2–4.
fluorescein) via a PEG linker (Fig. 2; ESIw, Scheme 1). The
length of the linker (o30 A) is within the Forster distance, so
¨
that the fluorescence of the probe is quenched. BHQs have
broad absorption spectra that span the entire visible region,
and can be paired with all common reporter fluorescent dyes.
In fact, all of the synthesized fluorophore–BHQ1 conjugates
(compounds 2–4) displayed lower fluorescence than the
original fluorophore (ESIw, Fig. S1). We hypothesized that
the fluorescence intensity of the probes might be restored by
binding an aptamer to BHQ1, which would mask energy
transfer from the fluorophore to BHQ1.
In this study, we describe the development of another small-
molecule-based method that is potentially useful for the future
achievement of live cell imaging of RNA. A series of potential
probes were synthesized by coupling a black hole quencher
(BHQ1)5 and a fluorophore (Fig. 1). BHQ1 was conjugated
covalently with one of three fluorophores (Alexa594, Cy3, or
a Institute for Integrated Cell-Material Sciences,
Kyoto University Gokasho, Uji, Kyoto 611-0011, Japan.
E-mail: uesugi@scl.kyoto-u.ac.jp; Fax: +81 (0)774-38-3226;
Tel: +81 (0)774-38-3225
In vitro selection (SELEX)6 was performed to isolate
aptamers for BHQ1. We prepared an RNA library, containing
a 60-base random region flanked by 20-base primer regions on
the 50 and 30 ends. After 13 rounds of selection against
BHQ1-immobilized agarose resins, we obtained four RNA
sequences (A1–A4) that shared a conserved sequence of 17
nucleotides, and one RNA sequence (A5) that did not contain
b Institute for Chemical Research, Kyoto University, Uji,
Kyoto 611-0011, Japan
w Electronic supplementary information (ESI) available: Experimental
details, synthetic procedures, and supplemental figures. See DOI:
10.1039/c1cc10393h
z These authors contributed equally to the work.
c
4712 Chem. Commun., 2011, 47, 4712–4714
This journal is The Royal Society of Chemistry 2011

View Article Online
Fig. 4 (A) Nucleotide sequences of A1 and its mutated (A1m) and
short (A1-sh and A1m-sh) versions. The conserved sequences are
shown in blue letters. (B) Fluorescence intensity changes following
binding of A1, A1m, A1-sh or A1m-sh to conjugate 2.
Fig. 3 (A) Nucleotide sequences of the N60 randomized region in
RNA selected against BHQ1. The conserved sequences found in clones
A1–A4 are shown in blue letters. (B) Fluorescence intensity changes
following binding of clones A1–A4 to conjugate 2.
A1-sh enhanced the fluorescence intensity of conjugate 2,
although to a lesser extent than parental A1 (Fig. 4B; Kd =
142.1 ꢁ 10.8 mM). In contrast, A1m-sh failed to restore the
fluorescence of conjugate 2.
the conserved sequence (Fig. 3A). The five RNAs were tested
for their ability to restore fluorescence of conjugate 2
(Alexa594–BHQ1). A1 had the highest ability to restore
fluorescence (4.5-fold increase at 3 mM), with an estimated
Kd of 4.7 (Fig. 3B). Although A2 and A3 were less potent
than A1, they had similar Kd values. A4, which contained a
These results indicate that the stem–loop region of A1 is
important in binding and masking BHQ1. Perhaps the
conserved loop region interacts directly with BHQ1, and the
non-conserved stem region may preorganize the loop for
the interaction.
one-base mismatch in the conserved sequence (AUU vs.
ꢀ
AGU), had much lower affinity than A1–A3, suggesting that
ꢀ
To determine if the conjugates could be used for real-time
monitoring of transcript levels in a cell-free system, we
prepared two DNA templates, each containing a T7 promoter
and a downstream region encoding either A1 or A1m
(Fig. 5A). The templates were transcribed by T7 RNA
polymerase at 30 1C in the presence of conjugate 2, and
fluorescence was measured during the course of RNA synthesis.
Fluorescence of conjugate 2 showed a time-dependent increase
in response to the expression of transcripts of the A1 template
(Fig. 5B); however, expression of A1m had little effect.
the sequence of the conserved region is important in the
recognition of BHQ1. In contrast to A1–A4, A5 failed to
enhance the fluorescence intensity of conjugate 2. Although
the sequence of A5 was enriched in the final pool of
BHQ1-binding RNA, A5 bound to the agarose resin and not
to BHQ1 (ESIw, Fig. S2).
A1 was selected for further analysis. A1 restored the
fluorescence intensities of all three fluorophore–BHQ1
conjugates (2–4) in a concentration-dependent manner (ESIw,
Fig. S4). However, the three conjugates responded differently
to A1. For example, the fluorescence intensity of conjugate 3
was enhanced 7.4-fold by 3 mM A1, while that of conjugate 4
was enhanced only 1.9-fold. Conjugates 2–4 had similar
quenching efficiencies relative to the unquenched dyes (92%,
98%, and 98%, respectively, at 0.3 mM of each conjugate;
ESIw, Fig. S1). Therefore, the different responses to A1 may be
due to different dequenching efficiencies of the fluorophores:
Alexa594 and Cy3 are better suited than fluorescein for RNA
detection.
The aptamers A1–A4 were predicted to form a stem–loop,
with the conserved sequence in the loop region (ESIw, Fig. S3).
We examined the effects of disruption of the stem–loop
structure on the binding affinity of A1 to conjugate 2. A
mutated version of A1 (A1m) was prepared by replacing the
30-face of the stem GCCUGGG with CGGACCC (Fig. 4A),
and subjected to fluorescence titration. The prevention of
Fig. 5 (A) Schematic diagram of DNA templates. (B) Real-time
monitoring of transcripts corresponding to A1 or A1m using
conjugate 2. (C) Time-dependent increase of transcripts synthesized
by T7 RNA polymerase from DNA templates encoding A1 or A1m.
Gels were stained with ethidium bromide.
stem–loop formation caused
a decrease in fluorescence
restoration of conjugate 2 by A1 (Fig. 4B). Thus, we also
prepared a short version of A1 (A1-sh), containing only the
stem–loop region, and A1m-sh, its mutated version (Fig. 4A).
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 4712–4714 4713

View Article Online
detection of specific RNA transcripts in cell-free systems. To
apply the methodology to living cells, the quenching efficiency
and chemical properties of the Cy3–BHQ1 conjugate need to
be optimized by changing the linker length or the structure of
the fluorophore. Aptamers with a lower Kd for BHQ1 are also
needed for real-time detection of expression of low-abundance
mRNAs. Optimization of the stem–loop structure of A1 may
generate a short RNA aptamer with a Kd in the nM or
pM range.
This work was supported in part by grants from
MEXT(JSPS) KAKENHI (21310140 to M.U., 21870016 to
S.S., and 20611007 to Y.K.), and the Naito Foundation
(M.U.). We thank the members of the Uesugi research group
for helpful discussion and support. We also thank T. Morii for
access to his sequencing and mass spectroscopy instruments.
The Kyoto research group participates in the Global COE
program ‘‘Integrated Material Sciences’’ (#B-09). A.M.
is an Inoue Fellow supported by the Inoue Foundation of
Science.
Fig.
6
(A) Schematic diagram of DNA templates for GFP
expression. (B) mRNAs encoded by GFP-A1 (blue) or GFP-A1m
(red), monitored by 610 nm fluorescence of conjugate 2 (open circles).
Protein synthesis monitored by 495 nm fluorescence of GFP (filled
squares). Data were normalized to fluorescence in a sample lacking the
DNA template. (C) Western blot analyses of GFP expression by
anti-His antibody. Loaded amounts of overall proteins are essentially
the same (shown in Fig. S9, ESIw).
Notes and references
1 (a) D. St Johnston, Nat. Rev. Mol. Cell Biol., 2005, 6, 363–375;
(b) A. J. Rodriguez, J. Condeelis, R. H. Singer and J. B. Dictenberg,
Semin. Cell Dev. Biol., 2007, 18, 202–208; (c) S. Tyagi,
Nat. Methods, 2009, 6, 331–338; (d) K. C. Martin and
A. Ephrussi, Cell (Cambridge, Mass.), 2009, 136, 719–730.
2 (a) S. Tyagi and F. R. Kramer, Nat. Biotechnol., 1996, 14, 303–308;
(b) D. L. Sokol, X. L. Zhang, P. Z. Lu and A. M. Gewitz, Proc.
Natl. Acad. Sci. U. S. A., 1998, 95, 11538–11543; (c) T. Matsuo,
Biochim. Biophys. Acta, Gen. Subj., 1998, 1379, 178–184;
(d) J. Perlette and W. H. Tan, Anal. Chem., 2001, 73, 5544–5550;
(e) D. P. Bratu, B. J. Cha, M. M. Mhlanga, F. R. Kramer and
S. Tyagi, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 13308–13313;
(f) X. H. Peng, Z. H. Cao, J. T. Xia, G. W. Carlson, M. M. Lewis,
W. C. Wood and L. Yang, Cancer Res., 2005, 65, 1909–1917;
(g) W. J. Rhee, P. J. Santangelo, H. J. Jo and G. Bao, Nucleic
Acids Res., 2008, 36, e30.
Similar results were obtained for conjugate 3 (ESIw, Fig. S5).
Densitometric analysis of the transcript bands on a gel showed
that amounts of A1 and A1m transcripts were similar at given
time points (Fig. 5C; ESIw, Fig. S6). Thus, conjugate 2 can be
used to monitor specific RNA transcripts in real-time in a
cell-free system.
We next examined whether specific protein-coding mRNA
transcripts could be detected in cell extracts. We used an
expression vector encoding GFP with A1 or A1m in 30 UTR
as a reporter gene (Fig. 6A). The templates were transcribed
and translated in vitro by a coupled transcription/translation
reaction. The fluorescence intensities of conjugate 2 (610 nm)
and GFP (495 nm) were simultaneously monitored during
transcription and translation. The sample containing the
GFP-A1 vector showed an immediate and time-dependent
increase in fluorescence of conjugate 2, while expression
of GFP-A1m had little effect (Fig. 6B; ESIw, S7AB). The
fluorescence of GFP emerged B30 min later than that of
conjugate 2, due to the delay in protein synthesis from mRNA.
Western blot analysis confirmed that GFP proteins were
expressed equally from the A1 and A1m constructs (Fig. 6C;
S7C, ESIw). These results demonstrate that conjugate 2 is
capable of detecting a specific protein-coding mRNA in cell
extracts.
3 (a) E. Bertrand, P. Chartrand, M. Schaefer, S. M. Shenoy,
R. H. Singer and R. M. Long, Mol. Cell, 1998, 2, 437–445;
(b) D. L. Beach, E. D. Salmon and K. Bloom, Curr. Biol., 1999,
9, 569–578; (c) K. M. Forrest and E. R. Gavis, Curr. Biol., 2003, 13,
1159–1168; (d) J. Bi, X. Hu, H. H. Loh and L. N. Wei, Mol.
Pharmacol., 2003, 64, 594–599; (e) N. Daigle and J. Ellenberg,
Nat. Methods, 2007, 4, 633–636; (f) H. Liora, Z. Gadi, A.
Stella and E. G. Jeffrey, Nat. Methods, 2007, 4, 409–412;
(g) S. Mili, K. Moissoglu and I. G. Macara, Nature, 2008, 453,
115–119.
4 (a) J. R. Babendure, S. R. Adams and R. Y. Tsien, J. Am. Chem.
Soc., 2003, 125, 14716–14717; (b) B. A. Sparano and K. Koide,
J. Am. Chem. Soc., 2005, 127, 14954–14955; (c) B. A. Sparano and
K. Koide, J. Am. Chem. Soc., 2007, 129, 4785–4794; (d) S. Sando,
A. Narita, M. Hayami and Y. Aoyama, Chem. Commun., 2008,
3858–3860; (e) J. Lee, K. H. Lee, J. Jeon, A. Dragulescu-Andrasi,
F. Xiao and J. Rao, ACS Chem. Biol., 2010, 5, 1065–1074.
5 (a) G. (P.) Yang, D. D. Erdman, M. L. Tondella and B. S. Fields,
J. Virol. Methods, 2009, 162, 288–290; (b) J. G. Bruno,
M. P. Carrillo, T. Phillips, N. K. Vail and D. Hanson,
J. Fluoresc., 2008, 18, 867–876.
6 (a) A. D. Ellington and J. W. Szostak, Nature, 1990, 346, 818–822;
(b) C. Tuerk and L. Gold, Science, 1990, 249, 505–510;
(c) D. L. Robertson and G. F. Joyce, Nature, 1990, 344, 467–468.
In summary, the present study designed and demonstrated
the usefulness of fluorophore–BHQ1 conjugates for real-time
c
4714 Chem. Commun., 2011, 47, 4712–4714
This journal is The Royal Society of Chemistry 2011