Diamine catalyzed hemicyanine dye formation from nonfluorescent precursors through DNA programmed chemistry

Article abstract of DOI:10.1021/ja0753602

A hemicyanine fluorescent dye was generated by a diamine-catalyzed, DNA-templated, aldol-type condensation from nonfluorescent precursors. Studies of reaction rate and yield as a function of catalyst structure indicated the diamine catalyst operates in a

Full text of DOI:10.1021/ja0753602

Published on Web 02/27/2008
Diamine Catalyzed Hemicyanine Dye Formation from Nonfluorescent
Precursors through DNA Programmed Chemistry
Yumei Huang* and James M. Coull
Ensemble DiscoVery Corporation, 99 Erie Street, Cambridge, Massachusetts 02139
Received July 18, 2007; E-mail: yhuang@ensemblediscovery.com
DNA templated synthesis has evolved dramatically from its use
to study interactions of oligonucleotides,1 to a method for producing
DNA-like structures,2 to a general approach for compound synthesis
and reaction discovery.3,4 Essentially, DNA programmed chemistry
(DPC) is the coupling of organic chemical reactions to a biological
recognition event (DNA hybridization). As such, DPC has tremen-
dous potential for biodetection.5 DPC-based biodetection employs
recognition elements such as antibodies, aptamers, or polypeptides
to drive DNA hybridization, and a resultant analyte-dependent
synthesis of a fluorescent reporter compound. In a previous
publication,6 we demonstrated that DPC can be used to generate a
coumarin label from a less fluorescent azidocoumarin by phosphine
reduction following similar work by Taylor’s lab.7 However,
phosphines are susceptible to oxidation by air and through reaction
with disulfides, somewhat limiting their utility for biodetection.
We then turned our attention to hemicyanine dyes which have
been widely used as fluorescent probes.8 Hemicyanine dyes are
generally synthesized by aldol-type condensations of heterocyclic
quaternary salts bearing an active methyl group with aldehydes
under anhydrous conditions. However, recent reports have shown
asymmetric aldol condensation can be performed in aqueous media
in the presence of a Lewis acid9 or enamine catalyst10 with an im-
proved reaction rate. Preliminary screening of catalysts in a model
reaction indicated that organocatalyst (S)-pyrrolidine methylpyrroli-
dine,10c ((S)-PMP)) performed the best under the aqueous conditions
chosen (Table S2).
Both end-of-helix (E)3band middle-of-helix (M) DNA architec-
tures were then tested for (S)-PMP-mediated DNA-templated
hemicyanine formation (Figure 1). For the E-architecture, the
reactants were coupled to a pair of complementary DNA strands,
such that, upon DNA hybridization, the reactants would be
juxtaposed at the end of the helix. Figure 1A shows fluorescence
emission spectra of the reaction mixture of the indolinium- and
aldehyde-labeled complementary 15-mers, I_DNA4 and A_DNA2,
and various control reactions. First, there was no background
fluorescence emission for hemicyanine precursors I_DNA4 and
A_DNA2 alone (2 and 3 in Figure 1A) as compared to the buffer
blank. Second, without the addition of (S)-PMP, there was no
fluorescent signal generated. Third, increasing the pH alone from
8.4 to 10.0 had no effect on the yield (5 and 6 in Figure 1A). Last,
the DPC reaction was sequence-specific (DNA dependent), as no
fluorescence signal was generated using DNA strands having three
mismatched bases (DNA3) after 140 min, and only a trace amount
of fluorescence signal was detected after 16 h (data not shown).
For the matched sequences, LC-MS showed a major product peak
and a small residual peak of A_DNA2 after 16 h (Figure S3).
In the M architecture, the reactants were coupled to DNAs
designed to bind contiguously to a DNA template placing the
reactive groups in the middle of the helix upon DNA hybridization.
The experimental results showed that product and signal were
formed only in the presence of the complementary template DNA1
Figure 1. Relative fluorescent intensity of hemicyanine DPC reaction
mixtures. DPC reactions were performed using 0.2 µM reagent strands (and
equimolar template for M architecture) at RT for 140 min in 10 mM (S)-
PMP, 50 mM sodium phosphate (NaPi), 1 M NaCl, pH 8.4 unless otherwise
specified. Results for E and M architecture are shown in (A) and (B),
respectively.
(Figure 1B). The quantum yields and extinction coefficients mea-
sured for purified template-free DPC products (DNA2_H_DNA5,
H: hemicyanine; Φ: 0.1; ꢀ550 (H2O) : 75 000 cm-1 M-1) produced
in the M architecture and DNA2_H_DNA4 produced in the E
architecture (Φ: 0.1; ꢀ550 (H2O) : 87 000 cm-1 M-1) were similar,
as were the DPC reaction kinetics for the two architectures (Figure
S4). However, we noticed that the relative fluorescence intensity
of the M reaction mixture (Figure 1B, reaction 1) was approximately
twice that of the E reaction mixture (Figure 1A, reaction 1). This
2-fold difference in intensity between E and M architectures might
have been due to intercalation and/or restricted rotation of the two
heteroaromatic fragments of the hemicyanine within the M archi-
tecture. The observed increase in fluorescence of cyanines upon
DNA binding through intercalation is a well-known phenomena.11
When DNA1 template was added back to purified template-free
DNA2_H_DNA5 in the presence of salt, the fluorescence intensity
was doubled consistent with this hypothesis (Figure S5).
To understand the catalytic role of PMP and its analogues in
hemicyanine formation by DPC, we performed a structure-activity
study whereby we removed or varied the substitution of the two
nitrogen atoms (N1 and N1′) and the chirality at the C2′ position
of the ring (Table 1, pyrrolidine analogue). We found that replacing
N1 with a hydroxyl group resulted in a low yield (PM). When N1
was a primary amine (AMMP), the catalyst was much more active
as when N1 was a secondary or tertiary amine substituted with
aryl or alkyl moieties (PMP and ANMP). The poor water solubility
of ANMP also contributed to the low product yield. Moreover,
whether N1′ of the pyrrolidine ring was substituted or not had little
9
3238
J. AM. CHEM. SOC. 2008, 130, 3238-3239
10.1021/ja0753602 CCC: $40.75 © 2008 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Relative Fluorescence of DPC Reaction Mixtures in the
for (S)-AMEP, DMEDA, DMAPA, and 60% for DMABA (Figure
S9). That no DPC product is formed in the presence of DMABA
despite the Schiff’s base formation demonstrates the importance
of the close proximity of the two nitrogen centers in support of the
proposed mechanism.
Presence of Different Catalysts after 2 ha
To date, almost all DPC approaches have relied on the proximity
and intrinsic chemical reactivity of components. Exceptions are
palladium(0) and copper(I) catalyzed, cross-coupling, and Huisgen
reactions, which, although rapid, damage and degrade DNA.4c The
finding that simple diamines can act in a concerted fashion upon
two otherwise unreactive moieties to rapidly catalyze a high yield
of product without side reactions illustrates that other methods of
control may be used to facilitate DPC reactions.
The DPC-mediated synthesis of hemicyanine dyes from non-
fluorogenic precursors was undertaken to allow development of
homogeneous fluorescent bioassays. We have found that matrices
such as serum that are rich in primary amines and other nucleophiles
do not decrease the rate and yield of hemicyanine formation and
resulting fluorescent signal (Figure S11). Preliminary results indicate
that hemicyanine creation can be used broadly for detection of
biomolecules when the chemistry is linked to a variety of recogni-
tion elements such as oligonucleotides, aptamers, and antibodies.
Moreover, it is likely the functionality of the simple diamine
catalysts described here can be engineered into the template and/
or reactant strands. Such “all-in-one” designs might allow for
hemicyanine-based biodetection in living systems where the
introduction of an excess of free diamine would be experimentally
difficult and potentially toxic to the cell or organism being studied.
a
DPC reactions were performed using 0.2 µM reagent strands (and
equimolar template for M architecture) at RT in 10 mM catalyst, 50 mM
b
NaPi, 150 mM NaCl, pH 8.4. RFUs were recorded using a microplate
spectrofluorometer (Molecular Device, GeminiXPS). c Product yields after
2 h were calculated from RFUs based on standard curves (Figure S6). d In
a separate experiment, the T1/2 of the DMEDA reaction was determined to
be 52 min (Figure S8).
Scheme 1
Acknowledgment. We thank Scott Harbeson, Nick Terrett, Ben
Seigal, and Larry Haff for useful discussions.
Supporting Information Available: Experimental details and
figures. This material is available free of charge via the Internet at
http://pubs.acs.org.
effect on the product yield (AMEP vs AMMP). In a separate
experiment, we determined that no DPC products were formed when
simple primary amines such as isobutylamine, cyclopentylamine,
ethylamine, and isopropylamine were used. Inversion of the chiral
center at C2′ had little effect on product yield. Finally, we examined
the effect of concentration of (S)-PMP and found that increasing
the concentration of catalyst increased the reaction rate to a plateau
at ∼15 to 20 mM catalyst (Figure S7). Taken together these data
suggested that two basic centers were required, that one must be a
primary amine, that catalyst concentration had a significant impact,
and that catalyst chirality while having some influence on reaction
rate was of lesser importance.
This led us to propose a mechanism by which pyrrolidine
diamines were facilitating hemicyanine DPC through Schiff’s base
formation (Scheme 1). The observed requirement for a high catalyst
concentration was consistent with the need to establish and maintain
a Schiff’s base under aqueous conditions throughout the initial
reaction sequence in order to localize the catalyst at the reaction
site. Further evidence to support this mechanism came from studies
of acyclic diamines with different carbon chain lengths (Table 1).
Increasing the distance between the primary and tertiary amine
centers by a single carbon atom decreased the product yield by a
factor of 2 (DMAPA), and extending the distance by two carbon
atoms effectively abolished all catalytic effect (DMABA). Addition
of NaCNBH3 to DPC reaction mixtures provided direct evidence
for the Schiff’s base intermediate, as LC-MS showed quantitative
formation of the reduced Schiff’s base (secondary amine formation)
References
(1) Naylor, R.; Gilham, P. T. Biochemistry 1966, 5, 2722-2728.
(2) (a) Bo¨hler, C.; Nielsen, P. E.; Orgel, L. E. Nature 1995, 376, 578-581.
(b) Bolli, M.; Micura, R.; Eschenmoser, A. Chem. Biol. 1997, 4, 309-
320. (c) Luther, A., Branksch, R.; Kiedrowski, G. v. Nature 1998, 396,
245-249. (d) Gat, Y.; Lynn, D. G. Biopolymers 1998, 48, 19-28. (e)
Xu, Y.; Kool, E. T. J. Am. Chem. Soc. 2000, 122, 9040-9041.
(3) (a) Czlapinski, J. L.; Sheppard, T. L. J. Am. Chem. Soc. 2001, 123, 8618-
8619. (b) Garner, Z. J.; Liu, D. R. J. Am. Chem. Soc. 2001, 123, 6961-
6963.
(4) For reviews in DNA-templated synthesis, see: (a) Orgel, L. E. Acc. Chem.
Res. 1995, 28, 109-118. (b) Summerer, D.; Marx, A. Angew. Chem.,
Int. Ed. 2002, 41, 89-90. (c) Li, X.; Liu, D.R. Angew. Chem., Int. Ed.
2004, 43, 4848-4870.
(5) Silverman, A., P.; Kool, E. T. Chem. ReV. 2006, 106, 3775-3789.
(6) Haff, L. A.; Wilson, C. G. M.; Huang, Y.; Benton, B. K.; Bond, J. F.;
Stern, A. M.; Begley, R. F.; Coull, J. M. Clin. Chem. 2006, 52, 2147-
2148.
(7) Cai, J.; Li, X.; Yue, X.; Taylor, J. S. J. Am. Chem. Soc. 2004, 126, 16324-
16325.
(8) Jedrzejewska, B.; Kabatc, J.; Pietrzak, M.; Paczkowski, J. Dyes Pigm.
2003, 58, 47-58.
(9) (a) Kobayashi, S.; Nagayama, S.; Busujima, T. J. Am. Chem. Soc. 1998,
120, 8287-8288. (b) Kobayashi, S.; Busujima, T.; Nagayama, S.
Tetrahedron Lett. 1998, 39, 1579-1582. (c) Kawano, Y.; Fujisawa, H.;
Mukaiyama, T. Chem. Lett. 2005, 34, 614-615.
(10) (a) Dziedzic, P.; Zou, W.; Ha´fren, J.; Co´rdova, A. Org. Biomol. Chem.
2006, 4, 38-40. (b) Kofoed, J.; Darbre, T.; Reymond, J. Chem. Commun.
2006, 1482-1484. (c) Mase, N.; Nakai, Y.; Ohara, N.; Yoda, H.; Takabe,
K.; Tanaka, F.; Barbas, C. F., III. J. Am. Chem. Soc. 2006, 128, 734-
735.
(11) Isacsson, J.; Westman, G. Tetrahedron Lett. 2001, 42, 3207-3210.
JA0753602
9
J. AM. CHEM. SOC. VOL. 130, NO. 11, 2008 3239