Ylidenemalononitrile enamines as fluorescent "turn-on" indicators for primary amines

Article abstract of DOI:10.1021/ja509058u

Ylidenemalononitrile enamines undergo rapid amine exchange followed by a cyclization with primary amines to yield fluorescent products with emission intensities as high as 900 times greater than the starting materials. After identifying the fluorescent species by X-ray crystallography, we demonstrate that the rate of amine exchange is substrate dependent and that by simple structural variation the fluorescence can be tuned over the entire visible spectrum. We further demonstrate their potential application in biomolecule labeling.

Full text of DOI:10.1021/ja509058u

Communication
pubs.acs.org/JACS
Ylidenemalononitrile Enamines as Fluorescent “Turn-On” Indicators
for Primary Amines
Ashley R. Longstreet, Minyoung Jo, Rebecca R. Chandler, Kenneth Hanson, Naiqian Zhan,
Jeremy J. Hrudka, Hedi Mattoussi, Michael Shatruk, and D. Tyler McQuade*
Department of Chemistry and Biochemistry, Florida State University, Tallahassee, Florida 32306, United States
*
S Supporting Information
Scheme 1. Reaction Between 1 and a Primary Amine to
Produce 2
ABSTRACT: Ylidenemalononitrile enamines undergo
rapid amine exchange followed by a cyclization with
primary amines to yield fluorescent products with emission
intensities as high as 900 times greater than the starting
materials. After identifying the fluorescent species by X-ray
crystallography, we demonstrate that the rate of amine
exchange is substrate dependent and that by simple
structural variation the fluorescence can be tuned over the
entire visible spectrum. We further demonstrate their
potential application in biomolecule labeling.
hromophores and fluorophores reactive toward specific
C
functional groups, such as amines, are valuable for
1
detection and labeling. While many amine reactive labels or
dyes exist, they often lack the fluorescence “turn-on” and/or
shift in λmax upon reacting that are ideal, and sometimes
necessary, for the differentiation of the labeled species and
2
unbound dye. The invariant photophysical properties make
these labels less ideal for in situ studies where purification of the
3
labeled species and unbound dye is undesired or difficult.
Recently, new probes/labels containing functional groups
Previously, we demonstrated that enamines such as 1a could
be synthesized in moderate to high yields from ylidenemalo-
nonitriles and N,N-dimethylformamide dimethyl acetal in the
presence of acetic anhydride and could lead to polysubstituted
4
that react with primary amines to form push−pull anilines,
5
6
7
imines, hemiaminals, or enamines have emerged. These next
generation compounds yield products with shifted λmax and/or
an increased quantum yield relative to the starting material.
One example is the commercially available natural product
epicocconone that, upon the addition of primary amines, is
transformed from a weak green emitter to a bright orange
8
nicotinitriles after a subsequent Pinner cyclization. We then
discovered that species 1a−c react with primary amines to yield
fluorescent species (2a−c) (Scheme 1) when we attempted to
produce 2-aminopyridines by treating 1a with propylamine and
7
9
1
emitter. Despite these successes, most “turn-on” probes
heat. Analysis by H NMR spectroscopy demonstrated that 1a
reacts with propylamine (∼300 min; room temperature) to
yield a new species with changes to the vinyl protons (Figure
possess large molecular weights and are difficult to modify in
order to change the emission properties. A broader range of
applications could be realized with a smaller, more easily
modifiable class of “turn-on” probes.
1
a). These resonances suggested that the new product was
10 11
either the isomerized cis-enamine or the acyclic amidines 2a
Scheme 1).
The ambiguity of the H NMR spectra was resolved by
(
Here we introduce a new class of weakly fluorescent
ylidenemalononitrile enamines (1) that undergo a “turn-on”
mechanism when reacted with primary amines. The basic
elements are shown in Scheme 1 where an amine exchange and
cyclization occurs converting virtually nonfluorescent enamines
into fluorescent cyclic amidines (2). The products exhibit both
a spectral shift and as high as 900-fold greater emission
intensity enhancement relative to the starting materials. The
mechanism for the “turn-on” enables the system to be selective
toward alkyl primary amines. Additionally, the fluorescent
species can be tuned to yield colors ranging from cyan to
orange by altering the enamine substituents.
1
obtaining the crystal structures of 2a and 2b (Figure 1b,c).
Analysis of the structures revealed that the reaction between the
enamine and primary amine yields rigid, cyclic amidines. The
amidine moieties of 2a and 2b are plannar suggesting that each
is fully conjugated; however, 2a exhibits uneven bond-length
distribution when compared to 2b indicating that 2a is less
aromatic than 2b.
Received: September 10, 2014
Published: October 14, 2014
©
2014 American Chemical Society
15493
dx.doi.org/10.1021/ja509058u | J. Am. Chem. Soc. 2014, 136, 15493−15496

Journal of the American Chemical Society
Communication
(
(
0.1%) to 2a (8.0%), 1b (0.035%) to 2b (7.0%), and 1c
0.016%) to 2c (14.2%), there is an approximate 80-, 200-, and
9
00-fold increase in Φ , respectively. Presumably, the lack of
PL
emission in compounds 1a−c at room temperature is due to
torsional relaxation of the dimethylamine or photoinduced
electron transfer (PET). At 77 K, in glassy solvent (2-MeTHF),
the emission intensity and lifetime of 1a−c significantly
increase suggesting torsional and vibrational relaxations are
the primary nonradiative pathways, which is inhibited by the
rigid matrix.
The real-time fluorescent “turn-on” time course of 1a is
shown in Figure 3a. A solution of 1a in acetonitrile was treated
with propylamine, and the emission was monitored over time.
In 80 min, there is a ∼75-fold increase in emission intensity.
The visible emission change associated with 1a to 2a is
illustrated in Figure 3b.
The reaction rate between propylamine and 1a (Figure 3a)
and 1b (data not shown) is slow and requires concentrations
too high to function within real-time sensors or for in vivo
studies. On the other hand, the reaction of 1c to generate 2c
under the same conditions is visibly faster than that observed
with 1a and 1b. The formation of 2c is also visible in ambient
light, where the color of the solution changes from magenta to
orange as 1c transforms into 2c. Therefore, the reaction
kinetics of 1c in the presence of propylamine was easily
monitored by UV−vis spectroscopy. The absorption spectrum
of 1c (0.045 mM in acetonitrile) was monitored once a second
for 300 s after the addition of 355 equiv (16 mM) of
propylamine. The spectral changes indicate a transition from 1c
1
Figure 1. (a) H NMR spectra of the reaction between 1a (100 mM)
and propylamine (12 equiv) in acetonitrile-d3 over time. 1a is
highlighted in red, 2a is highlighted in blue, and a possible
intermediate(s) is highlighted in yellow. Crystal structures of (b) 2a
and (c) 2b. For clarity, larger images are located in the Supporting
Information.
(
red) into 2c (pink) (Figure 4).
Spectral changes could be satisfactorily fit with the
DFT calculations of compounds 2a−c were performed at the
B3LYP/6-31G* level with a static electric field representing
SPECFIT/32 global analysis software package based on a
model with two consecutive first-order reactions, A → B →
C. The rate constant for the first step (kA→B = 0.034 s ) is
three times slower than the second (kB→C = 0.109 s ). Under
the assumption that the amine exchange occurs first (yielding a
propylamine enamine intermediate), the reaction is rate limited
by the amine exchange, which is then followed by a faster
cyclization step. The enhanced reaction rate of 1c, relative to 1a
and 1b, may be due to the increased conjugation stabilizing the
formation of the intermediate. An intermediate species, B, was
CH Cl . Based on the trend in HOMO−LUMO gap, we
2
2
anticipated an increasingly lower energy absorption and
emission in the order of 2a > 2b > 2c. In accordance with
this prediction, a broad range of emission colors can be
achieved (Figure 2). When 1a−c were treated with propyl-
12
−1
−1
1
also observed when the reaction with 1a was monitored by H
NMR spectroscopy (Figure 1a).
The reactivity and selectivity of emission “turn-on” with a
variety of amines were then investigated (Figure 5). Steric
effects play a significant role in the rate of “turn-on.” Where
propylamine (i in Figure 5) is nearly complete at 5 min,
isopropylamine (ii), cyclohexylamine (v), and benzylamine (vi)
take at most 2 h to induce a full “turn-on”. With very bulky
amines, such as t-butylamine (iii), there was a nominal increase
in emission intensity even after 24 h. We also investigated the
effect a tethered alcohol or amine would have on the “turn-on”.
The ethanolamine (iv) reaction is nearly complete in 5 min
indicating that the alcohol group does not greatly affect the
reaction kinetics or emission yield when compared to
propylamine (i). The tethered amine present in 3-(dimethyl-
amine)-1-propylamine (DMAPA) (vii), however, causes a large
decrease in emission intensity. Since the amine exchange was
Figure 2. Normalized emission spectra of 2a−c in CH Cl at room
2
2
temperature. Photographs shown are of 1 mM solutions of 2a−c in
CH Cl under 365 nm irradiation.
2
2
amine in a solution of CH Cl or acetonitrile, 2a, 2b, and 2c
2
2
provided cyan (CIE coordinates: 0.24, 0.38), green (0.33, 0.55),
and orange (0.48, 0.51), respectively (Figure 2). The
photophysical properties of complexes 1a−c and 2a−c are
summarized in Table 1.
1
confirmed to be complete within minutes by H NMR, this
The emission intensity “turn-on” of the complexes is best
decrease may be due to PET between the dimethylamine and
the fluorophore. Nonalkyl primary amines show no significant
increase in emission (viii, ix, x, and xi). Having a mixture of a
exemplified by comparing the fluorescent quantum yields (Φ )
PL
for 1 and 2 as summarized in Table 1. When comparing 1a
1
5494
dx.doi.org/10.1021/ja509058u | J. Am. Chem. Soc. 2014, 136, 15493−15496

Journal of the American Chemical Society
Communication
Table 1. Photophysical Properties of 1a−c and 2a−c at Room Temperature and 77 K
a
d
emission at room temperature
emission at 77 K
4
−1
a
λmax (nm) τ (ns) ΦPL (%) k (10 s−1)b knr (108s−1
7
c
compound
absorbance λ (nm) (ε, × 10 M cm−1
388 (4.9), 377 (4.6)
395 (2.9), 296 (1.0), 261 (1.0)
)
)
λmax (nm) τ (ns)
r
e
1a
410
560
605
500
530
555
−
−
−
0.1
−
−
−
−
−
−
410
540
609
465
490
541
2.0
6.4
e
e
1b
0.035
0.016
1
c
a
541 (1.3), 520 (1.2), 356 (2.8), 300 (1.5), 267 (1.6)
390 (4.8)
12.3
18.0
16.1
17.7
f
2
8.0
8.0
1.00
1.15
f
2b
425 (0.5), 276 (2.4)
3.3
8.5
7.0
2.12
1.67
2.81
1.01
f
2c
511 (0.4), 481 (0.5), 390 (1.0), 368 (0.9), 341 (2.9), 325 (2.0)
14.2
a
b
c
d
e
In CH Cl . k = Φ/τ. k = (1 − Φ)/τ. In 2-MeTHF. Relative quantum yields calculated with the corresponding amidine as a reference.
2
2
r
nr
f
Obtained by using an integrating sphere.
Figure 3. (a) Emission spectra of 1a (100 mM) in acetonitrile at 0, 5,
10, 20, 40, and 80 min after the addition of propylamine (12 equiv).
Figure 5. Relative emission intensities at 625 nm depicting the “turn-
on” selectivity of 1c toward alkyl primary amines when excited at 482
nm at 5 min and 2 h. Samples were prepared by treating a solution of
1c (0.9 mM, acetonitrile) with the appropriate amine(s) (12 equiv, 11
mM). Samples were diluted (0.025 mM) before obtaining emission
spectra. Amines are as followed: (i) propylamine, (ii) isopropylamine,
An aliquot of the reaction was diluted to 1 mM before obtaining the
emission spectra. (b) Solutions of 1a and 2a (1 mM) in CH Cl under
2
2
365 nm irradiation.
(
iii) t-butylamine, (iv) ethanolamine, (v) cyclohexylamine, (vi)
benzylamine, (vii) DMAPA, (viii) aniline, (ix) pyrrolidine, (x) TEA,
xi) pyridine, (xii) 1:1 mixture of TEA and propylamine.
(
consumed. Denaturing of the protein and long reaction times
were not necessary for sufficient labeling, which is ideal for
maintaining the initial function of most proteins. The labeled
transferrin was easily purified by size-exclusion chromatography
before the emission spectrum was obtained to ensure the
protein was labeled (Figure 6).
Figure 4. Changes in the absorption spectra of 1c (0.045 mM in
acetonitrile) after the addition of propylamine (16 mM, 355 equiv).
Spectra were acquired once a second for 300 s (0 s (red) to 300 s
In conclusion, we have shown that ylidenemalononitrile
enamines can act as fluorescent “turn-on” indicators toward
alkyl primary amines. This acyclic to cyclic “turn-on”
(
pink)). Only every 10th spectra is included in the figure for clarity.
Product generation is 90% complete in approximately 40 s.
nonreactive amine, triethylamine (TEA), with propylamine
(
xii) has no effect on the “turn-on” process.
While emission “turn-on” upon reaction with common
amines is applicable toward detecting chemical leakage or food
4
,5b,13
spoilage,
smaller, longer-wavelength emitting fluorescent
molecules could have great impact as protein labeling
3
b
reagents. To demonstrate the enamines’ ability to react and
label relevant biomolecules, human transferrin glycoprotein
(
310 μM, 50:50 vol % DMSO/PBS buffer pH = 8.5) was
labeled with 1c (3.1 mM). Labeled transferrin conjugates are
very useful as endosome markers for cellular studies.
However, purification of the labeled conjugates from the
unlabeled dye is known to be difficult, and the “turn-on” dye
could eliminate the need for purification. Within 4 h of the
reaction, the magenta solution turned yellow signifying 1c was
1
4
Figure 6. Emission spectra of transferrin, 1c (30 μM, 50:50 vol %
DMSO/PBS buffer, pH = 7), and purified 1c-transferrin conjugates
(∼30 μM after purification, PBS buffer, pH = 7) at room temperature.
Excitation at 420 nm.
3c
1
5495
dx.doi.org/10.1021/ja509058u | J. Am. Chem. Soc. 2014, 136, 15493−15496

Journal of the American Chemical Society
Communication
mechanism is unique, and we suggest that is the origin of the
excellent primary amine selectivity observed. We also
demonstrate that the emission maximum and the reaction
kinetics are easily tuned by making changes to the position
flanking the ylidenemalononitrile. Having a series of primary
amine reactive probes with different exchange rates might have
application in both real-time biosensing and array-based sensor
strategies, as we have already demonstrated their ability to label
the transferrin glycoprotein.
(10) For an example of a cis-enamine formation when an
̌
ylidenemalononitrile is treated with a primary amine: Skofic, P.;
Dambrot, C.; Kozelj, M.; Golobic, A.; Barrio, J. R.; Petric, A. Acta
Chim. Slov. 2005, 52, 391.
11) For an example of an amidine formation when an
̌
̌
̌
(
ylidenemalononitrile is treated with a primary amine: Lacroix, A.;
Fleury, J.-P. Tetrahedron Lett. 1978, 19, 3469.
(12) Binstead, R. A.; Jung, B.; Zuberbuhler, A. D. SPECFIT/32 global
analysis software, version 3.0.40 for 32-bit Windows systems; Spectrum
Software Associates: Marlborough, MA, 2007.
(13) (a) Maynor, M. S.; Nelson, T. L.; O’Sulliva, C.; Lavigne, J. J.
ASSOCIATED CONTENT
Supporting Information
Experimental procedures, spectroscopy data, and CIF files. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Org. Lett. 2007, 9, 3217. (b) Gao, T.; Tillman, E. S.; Lewis, N. S. Chem.
Mater. 2005, 17, 2904.
■
*
S
(14) Medintz, I. L.; Pons, T.; Delehanty, J. B.; Susumu, K.; Brunel, F.
M.; Dawson, P. E.; Mattoussi, H. Bioconjugate Chem. 2008, 19, 1785.
AUTHOR INFORMATION
Corresponding Author
mcquade@chem.fsu.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the Clinton Health Access Initiative,
Corning Glass, Gates Foundation, and NSF (CHE-1152020)
for support.
REFERENCES
■
(
4
1) Wu, J.; Liu, W.; Ge, J.; Zhang, H.; Wang, P. Chem. Soc. Rev. 2011,
0, 3483.
(
2) (a) Morales, A. R.; Schafer-Hales, K. J.; Marcus, A. I.; Belfield, K.
D. Bioconjugate Chem. 2008, 19, 2559. (b) Meng, Q.; Yu, M.; Zhang,
H.; Ren, J.; Huang, D. Dyes Pigm. 2007, 73, 254. (c) Hoefelschweiger,
B. K.; Duerkop, A.; Wolfbeis, O. S. Anal. Biochem. 2005, 344, 122.
(
(
d) Banks, P. R.; Paquette, D. M. Bioconjugate Chem. 1995, 6, 447.
e) Mornet, D.; Ue, K. Biochemistry 1985, 24, 840.
(
3) (a) Yamaguchi, T.; Asanuma, M.; Nakanishi, S.; Saito, Y.;
Okazaki, M.; Dodo, K.; Sodeoka, M. Chem. Sci. 2014, 5, 1021.
(
(
b) Chan, J.; Dodani, S. C.; Chang, C. J. Nat. Chem. 2012, 4, 973.
c) Tynan, C. J.; Clarke, D. T.; Coles, B. C.; Rolfe, D. J.; Martin-
Fernandez, M. L.; Webb, S. E. D. PLoS One 2012, 7, e36265. (d) Jun,
M. E.; Roy, B.; Ahn, K. H. Chem. Commun. 2011, 47, 7583.
(
8
(
2
4) Soga, T.; Jimbo, Y.; Suzuki, K.; Citterio, D. Anal. Chem. 2013, 85,
973.
5) (a) Hettie, K. S.; Klockow, J. L.; Glass, T. E. J. Am. Chem. Soc.
014, 136, 4877. (b) Shi, L.; Fu, Y.; He, C.; Zhu, D.; Gao, Y.; Wang,
Y.; He, Q.; Cao, H.; Cheng, J. Chem. Commun. 2014, 50, 872. (c) Lee,
B.; Chen, S.; Heinis, C.; Scopelliti, R.; Severin, K. Org. Lett. 2013, 15,
3
456. (d) Wu, J.-S.; Liu, W.-M.; Zhuang, X.-Q.; Wang, F.; Wang, P.-F.;
Tao, S.-L.; Zhang, X.-H.; Wu, S.-K.; Lee, S.-T. Org. Lett. 2006, 9, 33.
e) Secor, K. E.; Glass, T. E. Org. Lett. 2004, 6, 3727. (f) Feuster, E. K.;
Glass, T. E. J. Am. Chem. Soc. 2003, 125, 16174.
6) (a) Ryu, D.; Park, E.; Kim, D.-S.; Yan, S.; Lee, J. Y.; Chang, B.-Y.;
(
(
Ahn, K. H. J. Am. Chem. Soc. 2008, 130, 2394. (b) Mohr, G. J. Anal.
Chim. Acta 2004, 508, 233. (c) Mertz, E.; Zimmerman, S. C. J. Am.
Chem. Soc. 2003, 125, 3424. (d) Mertz, E.; Beil, J. B.; Zimmerman, S.
C. Org. Lett. 2003, 5, 3127.
(
́
7) (a) Chatterjee, S.; Karuso, P.; Boulange, A.; Peixoto, P. a.; Franck,
X.; Datta, A. J. Phys. Chem. B 2013, 117, 14951. (b) Coghlan, D. R.;
Mackintosh, J. a.; Karuso, P. Org. Lett. 2005, 7, 2401. (c) Bell, P. J. L.;
Karuso, P. J. Am. Chem. Soc. 2003, 125, 9304.
(
8) Longstreet, A. R.; Campbell, B. S.; Gupton, B. F.; McQuade, D.
T. Org. Lett. 2013, 15, 5298.
9) Villemin, D.; Belhadj, Z.; Cheikh, N.; Choukchou-Braham, N.;
Bar, N.; Lohier, J.-F. Tetrahedron Lett. 2013, 54, 1664.
(
1
5496
dx.doi.org/10.1021/ja509058u | J. Am. Chem. Soc. 2014, 136, 15493−15496