Organometallic activation of a fluorogen for templated nucleic acid detection

Article abstract of DOI:10.1021/ol800878b

(Chemical Equation Presented) A nucleic acid detection scheme that employs DNA-mediated delivery of an organomercury activator to unmask a fluorophore is described. The approach relies on adjacent hybridization of two oligonucleotide conjugates containing organomercury and caged rhodamine functionalities. Postsynthetic conjugation of amino-modified DNAs enabled efficient preparation of these probes. Complementary DNA templates yielded fluorescence signals arising from metal-assisted rhodamine uncaging.

Full text of DOI:10.1021/ol800878b

ORGANIC
LETTERS
2008
Vol. 10, No. 14
2935-2938
Organometallic Activation of a
Fluorogen for Templated Nucleic Acid
Detection
Raphael M. Franzini and Eric T. Kool*
Department of Chemistry, Stanford UniVersity, Stanford, California 94305-5080
kool@stanford.edu
Received April 16, 2008
ABSTRACT
A nucleic acid detection scheme that employs DNA-mediated delivery of an organomercury activator to unmask a fluorophore is described.
The approach relies on adjacent hybridization of two oligonucleotide conjugates containing organomercury and caged rhodamine functionalities.
Postsynthetic conjugation of amino-modified DNAs enabled efficient preparation of these probes. Complementary DNA templates yielded
fluorescence signals arising from metal-assisted rhodamine uncaging.
The biomedical importance of point mutations has generated
a substantial demand for methods that detect nucleic acids
with single nucleotide specificity. Detecting RNAs in vivo
makes it possible to bypass amplification and purification
steps and significantly reduce the costs and time associated
with RNA detection. Possible applications of cellular RNA
detection include pathogen identification and detection of
cancer-associated mutations. Nucleic acid templated reactions
that activate fluorescence are among the most sensitive and
sequence-selective nucleic acid detection approaches1–8 and
can be applied to detecting RNAs in cells.3,4 This strategy
exploits the target strand as a template for a chemical reaction
between two functionalized DNAs.2 The probes hybridize
to adjacent sites on the target sequence, increasing the
effective concentration of the attached functionalities, and
accelerating the reaction. Templated reactions have enabled
the sensing of RNAs in both prokaryotes3 and eukaryotic
cells4 using an SN2 quencher displacement strategy.5 How-
ever, substantial background emission from nonspecific
interactions with cellular components currently limits in vivo
detection to highly expressed RNAs.4 Several other chemical
(5) (a) Miller, G. P.; Silverman, A. P.; Kool, E. T. Bioorg. Med. Chem.
2008, 16, 56. (b) Abe, H.; Kool, E. T. J. Am. Chem. Soc. 2004, 126, 13980.
(c) Sando, S.; Abe, H.; Kool, E. T. J. Am. Chem. Soc. 2004, 126, 1081. (d)
Sando, S.; Kool, E. T J. Am. Chem. Soc. 2002, 124, 2096. (e) Xu, Y.;
Karalkar, N. B.; Kool, E. T. Nat. Biotechnol. 2001, 19, 148.
(6) (a) Dose, C.; Seitz, O. Bioorg. Med. Chem. 2008, 16, 65. (b)
Grossmann, T. N.; Seitz, O. J. Am. Chem. Soc. 2006, 128, 15596.
(7) (a) Pianowski, Z. L.; Winssinger, N. Chem. Commun. 2007, 3820.
(b) Cai, J.; Li, X.; Yue, X.; Taylor, J. S. J. Am. Chem. Soc. 2004, 126,
16324.
(1) Silverman, A. P.; Kool, E. T. Chem. ReV. 2006, 106, 3775
.
(2) (a) Li, X.; Liu, D. R. Angew. Chem., Int. Ed. 2004, 43, 4848. (b)
Ma, Z.; Taylor, J. S. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 11159. (c) Xu,
Y.; Kool, E. T. Tetrahedron Lett. 1997, 38, 5595. (d) Gryaznov, S. M.;
Letsinger, R. L. J. Am. Chem. Soc. 1993, 115, 3808
(3) (a) Silverman, A. P.; Baron, E. J.; Kool, E. T. ChemBioChem 2006,
7, 1890. (b) Silverman, A. P.; Kool, E. T. Nucleic Acids Res. 2005, 33,
.
4978. (c) Sando, S.; Kool, E. T J. Am. Chem. Soc. 2002, 124, 2096
.
(4) Abe, H.; Kool, E. T. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 263
.
(8) Huang, Y.; Coull, J. M. J. Am. Chem. Soc. 2008, 130, 3238.
10.1021/ol800878b CCC: $40.75
Published on Web 06/13/2008
2008 American Chemical Society

transformations such as native chemical ligation,6 the
Staudinger reaction,7 and diamine-catalyzed aldol condensa-
tion8 have been applied to DNA-templated activation of
fluorescence. However, the design of fluorogenic probes that
are active under physiological conditions but inert to non-
target cellular components continues to be a challenge.
Metal complexes may provide complementary function-
alities for the templated activation of fluoresence. Kra¨mer
et al. have combined DNA templated chemistry with metal
catalysis in vitro.9 Also, the templated complementation of
luminescent coordination complexes has been described.10
However, the use of labile coordination complexes likely
precludes the application of such probes in a cellular context.
Many organometallic reactants not only have unique
chemical reactivities but also metal-ligand interactions that
are inert to physiological conditions.11 Organometallic
molecules may provide stable, bioorthogonal reactant pairs
that can be applied to templated fluorescence activation in
living cells. This letter describes a DNA-templated reaction
between an organomercury-DNA probe and a caged fluo-
rophore conjugate. In the presence of a complementary target,
the reaction generates a strong fluorescence turn-on signal.
unlocks the rhodamine spirolactam ring and induces a
substantial fluorescence increase. We hypothesized that
Rhops would be sensitive to HgBA although with a reduced
reactivity relative to inorganic Hg2+. Rhops has excellent
photophysical properties such as long wavelength emission
and a high turn-on ratio; this fluorogenic reagent is biostable
and is under investigation for cellular imaging of inorganic
Hg2+.13 The sequences of the modified probes are comple-
mentary to adjacent positions at a single nucleotide poly-
morphism locus of the H-ras oncogene (Table 1).14
Table 1. Sequences of Probea and Templateb Strands Used for
Fluorescence Activation Studies
strand
sequence
Rhops-
DNA1 5′-Rhops-(CH2)3-OPO3--TGT GGG CAA GAG T-3′
Rhops-
DNA2 5′-CCG TCG G-OPO3--(CH2)3-Rhops-3′
Hg-
DNA1 5′-CCG TCG G-OPO3--(CH2)3-HgBA-3′
Hg-
DNA2 5′-(HgBACH2)2CH-OPO3--TGT GGG CAA GAG T-3′
mut
mut A
5′-GCA CTC TTG CCC ACA CCG ACG GCG-3′
5′-GCA CTC TTG CCC ACA ACC GAC GGC G-3′
Scheme 1. Chemistry of DNA-Templated Activation of Rhops
mut AA 5′-GCA CTC TTG CCC ACA AAC CGA CGG CG-3′
wt A 5′-GCA CTC TTG CCC ACA ACC GCC GGC G-3′
Fluorescence by Phenylmercury-DNA Conjugates
a Rhops)rhodamineBphenylthiosemicarbazide.HgBA)p-mercuriobenzamide.
b Italic letters indicate probe hybridization sites. The underlined base in wt
A specifies a single nucleotide mismatch.
The postsynthetic condensation of amino-modified DNAs
with activated p-chloromercurio benzoate15 allowed straight-
forward preparation of HgBA-DNA conjugates (Scheme 2).
HgBA-DNA was stable to HPLC purification in CH3CN/
aqueous eluent systems as confirmed by MALDI-TOF mass
spectrometry (Supporting Information).
Preparation of Rhops-DNA1 (Table 1) involved the
postsynthetic conjugation of Rhops-isothiocyanate to amino-
modified DNA (Scheme 2). The reaction of rhodamine B
hydrazide with phenylene 1,4-diisothiocyanate afforded
Rhops-isothiocyanate in 89% yield (Supporting Informa-
tion). This caged dye-DNA conjugate was purified by HPLC
and its identity confirmed by MALDI-TOF mass spectrom-
etry.
WeinvestigatedthefluorescencepropertiesofRhops-DNA1
and its changes upon reaction with Hg-DNA1. Incubation
of 1 µM Rhops-DNA1 with 5 µM Hg-DNA1 in the
prescence 1 µM of the complementary template mut AA (15
h at 37 °C in 70 mM tris-borate and 10 mM MgCl2, pH 7)
generated a strong fluorescence signal with an emission
maximum at λem ) 600 nm when excited at λex ) 530 nm
The outlined reaction strategy involves two functionalized
DNA probes. When hybridized to the nucleic acid target,
one strand delivers a p-mercuriobenzoate (HgBA) group to
a fluorogenic, mercury-sensitive reagent attached to the
second strand (Scheme 1). The thiophilicity of mercury and
the inertness of the Hg-C bond to aqueous conditions makes
HgBA a promising organometallic reactant for bioanalytical
applications. Rhodamine B phenylthiosemicarbazide12 (Rhops)
serves as the masked fluorophore. In the prescence of Hg2+,
the thiosemicarbazide functionality undergoes a cyclization
reaction generating an oxadiazole. This transformation
(9) (a) Boll, I.; Kra¨mer, R.; Brunner, J.; Mokhir, A. J. Am. Chem. Soc.
2005, 127, 7849. (b) Zelder, F. H.; Brunner, J.; Kra¨mer, R. Chem. Commun.
2004, 902. (c) Brunner, J.; Mokhir, A.; Kra¨mer, R. J. Am. Chem. Soc. 2003,
125, 12410.
(10) (a) Kitamura, Y.; Ihara, T.; Tsujimura, Y.; Tazaki, M.; Jyo, A.
Chem. Lett. 2005, 34, 1606. (b) Oser, A.; Valet, G. Angew. Chem., Int. Ed.
1990, 29, 1167.
(13) (a) Yang, Y.-K.; Ko, S.-K.; Shin, I.; Tae, J. Nat. Protoc. 2007, 2,
1740. (b) Ko, S.-K.; Yang, Y.-K.; Tae, J.; Shin, I. J. Am. Chem. Soc. 2006,
128, 14150.
(11) Metzler-Nolte, N. Chimia 2007, 61, 736.
(14) Reddy, E. P.; Reynolds, R. K.; Santos, E.; Barbacid, M. Nature
1982, 300, 149.
(12) Yang, Y.-K.; Yook, K.-J.; Tae, J. J. Am. Chem. Soc. 2005, 127,
16760.
(15) Warning: p-chloromercuriobenzoic acid is acutely toxic!
2936
Org. Lett., Vol. 10, No. 14, 2008

Scheme 2
.
Post-Synthetic Preparation of DNA-Reactant
Conjugatesa
aDNA sequences are summarized in Table 1.
Figure 2. Template dependence of Rhops-DNA1 fluorescence
activation by Hg-DNA1. (a) Representative time courses of
fluorescence emission at λem ) 600 nm (λex ) 575 nm). (b) Relative
Rhops-DNA1 fluorescence signal in the presence of various
templates after 110 min reaction with Hg-DNA1. Error bars are
standard deviations of triplicate experiments.
(Figure 1). The emission intensity of Rhops-DNA1 in-
creased ca. 150-fold upon uncaging, which is comparable
to the best current DNA-templated fluorescence activation
designs.
Next, we assessed the template dependence of the reaction
between Rhops-DNA1 and Hg-DNA1. The fluorescence
signal of the DNA probes incubated at 37 °C in the presence
of various DNA targets was monitored as a function of time
(Figure 2a). Each sample consisted of 0.2 µM Rhops-DNA1
and template oligonucleotide, respectively, and 0.6 µM
Hg-DNA1. The fluorescence emission after 110 min
incubation served for quantitative comparison of the reaction
rates (Figure 2b).
To optimize probe alignment, we determined the rate of
reaction between Rhops-DNA1 and Hg-DNA1 in the
presence of complementary target strands with probe hy-
bridization sites separated by 0 to 2 nucleotides (mut, mut
A, and mut AA). All three templates accelerated the reaction
significantly. The fluorescence emission of Rhops-DNA1
in the presence of mut A and mut AA after 110 min was 5.9
and 6.4 times higher, respectively, than the emission in the
absence of Hg-DNA1. In the presence of mut, Hg-DNA1
was less efficient in unmasking Rhops-DNA1 (4.2-fold
relative to no Hg-DNA) than in the presence of mut A. Short
gaps between the sites of hybridization thus enable the
optimal alignment of the probes.
Altering a single nucleobase at the center of mut A’s
Hg-DNA1 binding site (wt A) significantly decreased the
reactivity of Hg-DNA1 toward Rhops-DNA1. The emis-
sion of Rhops-DNA1 after 110 min in the presence of wt
A (Table 1) was 2.6 times the emission measured without
Hg-DNA1. This result translated into a ∼2-fold reduction
of the reactivity between Hg-DNA and Rhops-DNA1
compared to mut A. A control sample containing Hg-DNA1
but no template exhibited a similar emission after 110 min
as a sample containing the mismatched strand wt A.
We hypothesized that DNA probes with two HgBA
functionalities might exhibit enhanced reactivity to Rhops-
Figure 1. Fluorescence spectra (λex ) 530 nm) of 1 µM
Rhops-DNA before (red trace) and after reaction with 5 µM Hg-
DNA1 in the presence of 1 µM of the template mut AA (blue trace).
Org. Lett., Vol. 10, No. 14, 2008
2937

template mut A was 11.3 times the fluorescence of back-
ground uncaging (no Hg-DNA2) compared to a 5.9-fold
increase for the experiments with Hg-DNA1. In addition,
the relative rate of reaction with Hg-DNA2 in the presence
of the mismatched template wt A increased 4.9-fold relative
to spontaneous Rhops-DNA unmasking (as compared to
2.6-fold for Hg-DNA1). On the other hand, the effect of a
second organomercury group in absence of template was
insignificant. This outcome demonstrates that two appended
HgBA residues yield improved efficiency and selectivity for
this templated reaction.
In summary, we have described a new DNA-templated
reaction that uses an organometallic reactant. In the presence
of a DNA target strand, an organomercury-functionalized
DNA probe induces the fluorogenic unmasking of a
Rhops-DNA conjugate. This system could be useful as a
fluorogenic reporter of nucleic acid sequence in solution, and
it gives readable signals with single nucleotide discrimination
in minutes. The fluorescence turn-on ratio of the Rhops probe
is superior to most current templated detection schemes.
However, the mismatch sensitivity, while significant, is only
moderate compared to previous methodologies. It is possible
that a different mercury-sensitive caged fluorophore will
afford a more predictable template dependence than Rhops.
We are currently investigating alternative organometallic
reactants for the development of template-dependent fluo-
rescence activation schemes that may be applied to the
sensing of nucleic acids with single mismatch discrimination
in vitro and in vivo.
It merits mention that in addition to this novel templated
reaction scheme, the approach we have developed for
phenylmercury conjugation may find applications in deriva-
tization of DNA. For example, organomercury compounds
react with thiols under very mild conditions, forming inert
but reversible Hg-S bonds.16 Organomercury functionalities
may further enable postsynthetic radiolabeling17 or func-
tionalization by metal-catalyzed cross-coupling reactions.18
Figure 3. Template dependence of Rhops-DNA2 fluorescence
activation by the doubly modified Hg-DNA2. (a) Representative
time courses of fluorescence emission at λem ) 600 nm (λex ) 575
nm). (b) Relative Rhops-DNA2 fluorescence signal in the presence
of various templates after 110 min reaction with Hg-DNA2. Error
bars are standard deviations of triplicate experiments.
DNA relative to Hg-DNA1. Rate acceleration could result
either from an increased effective concentration of HgBA
or from the concerted reaction of two HgBA groups with
Rhops.
The preparation of the bis-HgBA probe Hg-DNA2 (Table
1) involved 5′-diamino modification followed by postsyn-
thetic DNA conjugation of activated HgBA in analogy to
Hg-DNA1 (Scheme 2). The synthesis of the 5′-diamino
modifier required two steps. The reaction of 1,3-diamino-
2-propanol with trifluoroacetic anhydride afforded 1,3-
bis(trifluoroacetamido)-2-propanol in 83% yield. Treatment
of the amino-protected alcohol with 2-cyanoethyl N,N-
diisopropylchlorophosphoramidite provided the 5′-diamino
modifier (Supporting Information). The introduction of bis-
HgBA at the 5′-terminus facilitated the preparation of Hg-
DNA2. The associated DNA-fluorogen conjugate Rhops-
DNA2 (Table 1) was prepared in analogy to Rhops-DNA1.
The reaction of Hg-DNA2 with Rhops-DNA2 was
investigated as outlined for the reaction between Hg-DNA1
and Rhops-DNA1 (Figure 3). Hg-DNA2 accelerated the
template-dependent Rhops-DNA2 activation approximately
2-fold relative to the monosubstituted Hg-DNA1. After 110
min, the fluorescence in the presence of the complementary
Acknowledgment. We thank the U.S. National Institutes
of Health (Grant No. GM068122) for support. R.M.F.
acknowledges a Roche Research Fellowship and an Abbott
Summer Fellowship.
Supporting Information Available: Synthetic procedures
and analytical characterizations; protocols for postsynthetic
probe modification including MALDI-TOF characterization;
supporting kinetics experiments. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL800878B
(16) (a) Sigrist, H.; Kisling, E.; Rottenberg, M. Biochem. Biophys. Res.
Commun. 1989, 161, 640. (b) Dale, R. M. K.; Martin, E.; Livingston, D. C.;
Ward, D. C. Biochemistry 1975, 14, 2447.
(17) Dale, R. M. K.; Ward, D. C.; Livingston, D. C.; Martin, E. Nucleic
Acids Res. 1975, 2, 915.
(18) (a) Matsui, M.; Nishiyama, Y.; Ueji, S.-I.; Ebara, Y. Bioorg. Med.
Chem. Lett. 2007, 17, 456. (b) Bergstrom, D. E.; Ogawa, M. K. J. Am.
Chem. Soc. 1978, 100, 8106.
2938
Org. Lett., Vol. 10, No. 14, 2008