Excited state tautomerization of azaindole

Article abstract of DOI:10.1039/b506652b

Fluorescent tryptophan analogs, like azatryptophan, offer an advantage for exploring protein and peptide structure and dynamics. The chromophoric moieties, azaindole, of the azatryptophan analogs are investigated for their potential as fluorescent probes. The photophysical properties of 4-azaindole (4AI) and 5-azaindole (5AI) and their tautomers are characterized through computational and experimental methods. Both 4AI and 5AI undergo excited state tautomerization in the presence of 1 M NaOH. The protonated forms of 4AI and 5AI have a fluorescence emission of 415 and 410 nm, respectively, while the tautomers of 4AI and 5AI have a fluorescent emission of 480 and 450 nm, respectively. Gas phase computations (B3LYP/6-31+G**) show that the N¹H azaindole tautomer is lower in energy in the ground state by as much as 12.5 kcal mol^-1, while the N″H azaindole tautomer is lower in energy in the excited state by as much as 18.1 kcal mol^-1. Solvent effects on the tautomer energy differences were computed using the isodensity polarized continuum model (IPCM). The polarity of the solvent helps to reduce the energy difference between the tautomers in the ground state by as much as 5.8 kcal mol^-1, but not enough to reverse the ground state tautomer preference. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b506652b

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Excited state tautomerization of azaindole
Michael T. Cash, Peter R. Schreiner and Robert S. Phillipsa
a
a,b
a
Department of Chemistry, University of Georgia, Athens, Georgia, 30602, USA.
E-mail: prs@chem.uga.edu
Justus-Liebig-Universit a¨ t Giessen, Institut f u¨ r Organische Chemie, Heinrich-Buff-Ring 58,
b
35392, Giessen, Germany. E-mail: prs@org.chemie.uni-giessen.de
Received 12th May 2005, Accepted 10th August 2005
First published as an Advance Article on the web 14th September 2005
Fluorescent tryptophan analogs, like azatryptophan, offer an advantage for exploring protein and peptide structure
and dynamics. The chromophoric moieties, azaindole, of the azatryptophan analogs are investigated for their
potential as fluorescent probes. The photophysical properties of 4-azaindole (4AI) and 5-azaindole (5AI) and their
tautomers are characterized through computational and experimental methods. Both 4AI and 5AI undergo excited
state tautomerization in the presence of 1 M NaOH. The protonated forms of 4AI and 5AI have a fluorescence
emission of 415 and 410 nm, respectively, while the tautomers of 4AI and 5AI have a fluorescent emission of 480 and
1
4
50 nm, respectively. Gas phase computations (B3LYP/6–31+G**) show that the N H azaindole tautomer is lower
−
1
n
in energy in the ground state by as much as 12.5 kcal mol , while the N H azaindole tautomer is lower in energy in
the excited state by as much as 18.1 kcal mol . Solvent effects on the tautomer energy differences were computed
using the isodensity polarized continuum model (IPCM). The polarity of the solvent helps to reduce the energy
difference between the tautomers in the ground state by as much as 5.8 kcal mol , but not enough to reverse the
ground state tautomer preference.
−
1
−
1
3
Introduction
the zwitterionic form of tryptophan in experimental studies.
Therefore, the azaindole isomers are studied in this work to avoid
the complications associated with the azatryptophan isomers.
The four azaindole isomers of interest, 4-azaindole (4AI), 5-
azaindole (5AI), 6-azaindole (6AI) and 7-azaindole (7AI), are
shown in Fig. 1. The picture in Fig. 2 shows the fluorescence
of the four different azaindole isomers (10 mM aqueous
solution) when irradiated with UV light (254 nm). Visually, the
Tryptophan has been used extensively as an intrinsic fluorescent
probe to investigate protein dynamics and structure; tryptophan
fluorescence is particularly useful due to its sensitivity to the local
environment. A red shift in the fluorescence emission is observed
when a tryptophan residue is exposed to solvent relative to a
tryptophan buried in a hydrophobic pocket of a protein. In
addition, charged amino acid side chains that are in close prox-
imity quench the fluorescence decay of tryptophan, providing
additional information about the local environment of the tryp-
tophan residue. Several tryptophan analogs with significantly
improved fluorescence have been investigated for incorporation
into peptides and proteins. The most commonly used analogs
include 5-hydroxytryptophan and 7-azatryptophan. Each of
these analogs offer advantages over natural tryptophan for
exploring protein dynamics. The advantages include greater
fluorescence intensity, red-shifted fluorescence, and greater
1–5
6,8
7
sensitivity to the electric field of the local environment.
Ideally, the tryptophan analog incorporated should be struc-
turally similar to natural tryptophan to minimize the potential
for perturbation of protein structure and function. The azatryp-
tophan analogs are closer in structure to natural tryptophan
than the other analogs bearing hydroxyl and fluoro substituents,
which makes azatryptophan ideally suited as an optical spectral
probe. In addition, the incorporation of azatryptophan analogs
into proteins has proven to be relatively simple compared
Fig. 1 Azaindole isomers and tautomers explored in this work.
6,7
to other synthetic amino acid probes. The azatryptophan
analogs can be incorporated into proteins by expressing the
protein in an E. coli tryptophan auxotroph while replacing the
azatryptophan analogs for natural tryptophan in the growth
media. An integrated approach has been established that allows
domain specific incorporation of azatryptophan analogs using
25
the auxotroph followed by expressed protein ligation. Both
-azatryptophan and 6-azatryptophan have been successfully
incorporated into proteins and peptides with very promising
7
7,25–26
results.
Two other azatryptophan analogs, 4-azatryptophan
and 5-azatryptophan, may also be useful as probes for protein
dynamics and structure.
Indole is the chromophoric moiety of tryptophan and the
photophysics of indole are explored to avoid complications of
Fig. 2 A 10 mM solution of each azaindole isomer under a UV lamp
(k = 254 nm). From left to right: 4AI, 5AI, 6AI, and 7AI.
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 0 1 – 3 7 0 6
3 7 0 1

View Article Online
fluorescence intensity of 4AI and 6AI are significantly greater
than the other isomers, which makes 4AI and 6AI of additional
interest.
Theoretical and experimental
A. Computations
Another interesting property of the azaindole isomers is their
ability to undergo excited state tautomerization, which has been
demonstrated with 7AI and 6AI. Excited state tautomerization
of 7AI has been thoroughly explored. There are two established
mechanisms for the tautomerization of 7AI, an excited state
double proton exchange (Fig. 3a) and cyclic intermediate
The ground state studies of the azaindole isomers in both gas
phase and solvent were done using density functional theory.
All stationary structures were characterized as minima by
computation of second derivatives of the energy (no imaginary
frequencies). We employed Becke’s three parameter exchange
11
functional, and the Lee–Yang–Parr nonlocal correlation func-
tional in conjuction with a B3LYP/6–31+G** basis set imple-
9,10
involving one water molecule (Fig. 3b). Recently, it was shown
that 6AI also undergoes excited state tautomerization evident by
the presence of a fluorescent band at 440 nm in strong base
and methanol–water mixtures. The tautomerism of 6AI was
12
mented in Gaussian 94. Nucleus independent chemical shifts
23
(
NICS), a measure of aromaticity, were computed using density
functional theory as well.
6
proposed to occur by protonation at N in the excited state,
Bulk solvent effects were included with self consistent reaction
field (SCRF) computations using the isodensity polarized con-
1
7
followed by deprotonation at N also in the excited state.
13
tinuum model (IPCM). The IPCM model was chosen because
the effects of solvation are incorporated into the iterative SCF
computation, which results in a proper variational condition.
The dielectrics of the three solvents mimic chloroform (4.8),
methanol (32.7) and water (78.4) at 298 K.
27
Both time dependent DFT (TDDFT) and ab initio methods
were used for the excited state transition energies. The ab
initio method chosen is the configuration interaction with
14
single excitations (CIS) procedure. CIS was used to obtain
the optimized excited state energies and geometries. The same
basis set was used (6–31+G**) as for the DFT computations.
16
TDDFT computations utilized Gaussian 98 while all other
15
computations were carried out using Gaussian 94.
B. Synthesis of 4AI (Scheme 1)
Two routes were explored for obtaining 2-methyl-3-nitro-
pyridine from 2-chloro-3-nitropyridine.
Method
Chloro-3-nitropyridine (2.0 g, 12.5 mmol) was added to 0.3 g
12.5 mmol) NaH in 10 mL of dry DMF with stirring under
1. Diethyl-2-(3-nitropyridyl)malonate
(2).2-
Fig. 3 (a) Excited state double proton transfer of 7-azaindole; (b)
excited state proton transfer with one water molecule.
(
nitrogen. Diethyl malonate (2.0 g, 12.5 mmol) was added
dropwise, resulting in a brownish-red mixture. After the
reaction was allowed to stir for 1 h, the solvent was evaporated
giving a brown oil. The oil was diluted with 50 mL of water,
followed by neutralization with a few drops of acetic acid. The
precipitate was extracted with dichloromethane three times,
Indole is the base structure of a number of very important
biological molecules. The azaindole isomers could also be useful
for other applications besides peptide and protein dynamics.
7
AI is currently being explored as a molecular probe for use
in auxin physiology; 7AI is the chromophoric moiety of 7-
and the combined organic layers were dried with MgSO . The
4
24
azaindole-3-acetic acid. Considering the great potential for its
use as a fluorescent probe, we are interested in characterizing
the fluorescence properties of 4AI and 5AI. The ability of the
azaindole isomers to tautomerize in solution is an important
aspect regarding the fluorescence, because each tautomer has
a unique fluorescent spectrum. Through experimental and
computational methods, we determined the ground state and
excited state characteristics of each azaindole isomer and the
potential for excited state tautomerization of 4AI and 5AI. In
addition, we computed the effects of solvation on the energy
differences of each azaindole isomer.
solvent was removed and the remaining oil was then passed
over a silica column and eluted with a 30% ethyl acetate in
hexanes. The product diethyl-2-(3-nitropyridyl)malonate (1.8 g)
was isolated as yellow crystals (50% yield). All analytical data
17
agree with previously reported synthesis.
2-Methyl-3-nitropyridine (3). Diethyl-2-(3-nitropyridyl)-
malonate (1.0 g, 3.9 mmol) was dissolved in 50 mL of 6 N
HCl and refluxed for 8 h. The solvent was evaporated leaving a
brownish oil that solidified upon cooling. 50 mL of saturated
sodium carbonate was added, and the solution was extracted
three times with 25 mL of dichloromethane. The organic layers
Scheme 1
3
7 0 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 0 1 – 3 7 0 6

View Article Online
n
were combined and dried with MgSO
4
followed by removal of
Table 1 Energy differences between tautomers. (energy of N –H-
1
energy of N –H) at B3LYP/6–31+G**
solvent yielding 0.42 g of 2-methyl-3-nitropyridine as a brown
solid (86% yield). All analytical data agree with previously
−
1
−1
17
Isomer
4AI
Ground state/kcal mol
Excited state/kcal mol
reported synthesis.
Method 2. 2-Methyl-3-nitropyridine (3).
A
mixture of
11.6
11.4
10.1
12.5
−15.5
−17.6
−14.0
−18.1
5
6
7
AI
AI
AI
(
0.16 g, 1 mmol) 2-chloro-3-nitropyridine, tetrakis(triphenyl-
phosphine)Pd (0.12 g, 0.1 mmol), methylboronic acid (0.07 g,
.1 mmol) and potassium carbonate (0.4 g, 3 mmol) was added
1
to dioxane and refluxed for 2 d. The solvent was evaporated
and the mixture was separated on a silica column using an ethyl
acetate–hexanes solvent. 2-methyl-3-nitropyridine (0.06 g) was
obtained as a light brown solid (43% yield). All analytical data
2
93 nm and 270 nm for 5AI. The excitation wavelengths were
also computed as described above for comparison.
18
agree with previously reported synthesis.
Results and discussion
Preparation of the enamine (4). 2-Methyl-3-nitropyridine
0.5 g, 3.6 mmol) was dissolved in 10 mL of dry DMF and
Many theoretical and experimental studies have been reported
on 7AI including its tautomerization. It has been well established
that the N –H tautomer of both 7AI and 6AI is the most stable
tautomer in the ground state.
van reported that all four azaindole isomers are more stable
in the N –H tautomeric form based on ab initio computations
using the minimal STO-3G basis set at the HF level of theory, not
taking into account effects of electron correlation. In Table 1
we summarize the results of computing the ground states of
the tautomers of the azaindole isomers B3LYP/6–31+G**. The
N –H tautomer is more stable by more than 10 kcal mol for
each of the azaindole isomers; the N –H tautomer has often
been referred to as the “normal” tautomer for this reason.
Among the normal tautomers, 6AI is the lowest in energy
while 7AI is the highest, with 2.4 kcal mol separating the
two. On the other hand, in the excited state the other tautomer
(
stirred under nitrogen. DMF–DMA (0.86 g, 7.2 mmol) was
1
◦
added dropwise. The reaction was heated to 90 C. After 15 min
6,7,10,20,21
Rasmussen and Mahade-
of heating, a deep reddish color began to appear, and it was
allowed to react for 4 h. After 4 h, the solvent was removed
by evaporation. The red oil obtained (0.65 g) was used without
further purification. All analytical data agree with previously
1
19
17
reported synthesis.
4-azaindole (5). The crude enamine (0.65 g, 3.4 mmol) was
dissolved in 10 mL of 8.8% formic acid in methanol. The
mixture was added dropwise to a flask containing 10 mL of 8.8%
formic acid in methanol with 0.2 g of 5% Pd/C, stirred under
nitrogen. The reaction proceeded for 4 h, until the red color
had completely disappeared. The Pd catalyst was removed by
filtration through celite, and the filtrate was then concentrated.
After sitting overnight, the product, 4AI, crystallized (0.12 g).
The crystallized product was filtered and washed with hexanes.
The filtrate was then evaporated to dryness, and the remaining
solid was recrystallized with ethyl acetate and hexanes to give a
combined yield of 0.31 g (78%). All analytical data agree with
1
−1
1
6
−
1
n
(
referred to as N –H, where n = 4, 5, 6 or 7 depending on
−
1
the azaindole isomer) is more stable, by at least 15 kcal mol .
Previous experimental work agrees with the computations, as
both 7AI and 6AI do tautomerize in the excited state, provided
that the solvent conditions are suitable (i.e., alkaline solution or
17
previously reported synthesis.
6,7
combination of water with a less polar co-solvent).
The following investigation was to see if 5AI and 4AI also
tautomerize in the excited state. Computations indicate that the
C. Synthesis of 5AI
17
n
−1
5
AI was synthesized as previously described.
N –H tautomers are more stable by at least 15 kcal mol in the
excited state and therefore, the possibility of tautomerization
is likely. However, due to the geometric arrangement of the
pyridine and pyrrole nitrogens of 4AI and 5AI, the mechanism
of tautomerization will differ from that of 7AI (Fig. 3). The
mechanism of tautomerization would be similar to 6AI, a
proposed mechanism being excitation of the normal tautomer,
D. General procedure for the synthesis of the
N-methyl-azaindole isomers (Scheme 2)
19
6
followed by solvent proton abstraction at N in the excited state,
1
7
and the deprotonation of N also in the excited state.
1
The relative energies of the N –H azaindole isomers are listed
in Table 2. The values are reported relative to 4AI. 7AI is
lower in energy by approximately 4 kcal mol in the ground
state compared to other N –H azaindole isomers. 4AI, 5AI and
AI all have nearly the same ground state energy as the N –H
Scheme 2
−
1
1
The azaindole isomer (100 mg, 0.85 mmol) was dissolved in
mL dry DMF. Iodomethane (120 mg, 0.85 mmol) was added
1
6
2
tautomer. Likewise, in the excited state, 7AI is again the lower
to the mixture, and the reaction was allowed to stir for 3 h.
The solvent was evaporated, leaving a brown oil. The remaining
brown oil was dissolved in 25 mL dichloromethane and extracted
three times with 15 mL of saturated sodium carbonate. The
−
1
1
in energy by as much as 9.4 kcal mol in the N –H tautomer.
n
The relative energies of the N –H tautomers are summarized
in Table 2. N –H 5AI is more stable by about 5 kcal mol in the
ground state compared to other N –H azaindole isomers. N –H
5
−1
n
4
organic layer was dried with MgSO followed by evaporation
4
and the mixture was then loaded onto an alumina column.
The parent azaindole was eluted first with chloroform. The N-
methyl azaindole was eluted with 20% ethanol–chloroform to
Table 2 Energies of azaindole isomers relative to 4AI at B3LYP/6–
3
1+G**
19
−
1
−1
give approximately 20 mg of each azaindole isomer (18% yield).
Isomer
Ground state/kcal mol
Excited state/kcal mol
Absorbance and fluorescence spectroscopy
4AI
AI
0.0
−0.3
0.4
0.0
1.0
2.8
5
The absorbance spectra were obtained on a Cary 1E UV/Vis
spectrophotometer. The fluorescence experiments were obtained
on a Fluorolog Spectrofluorimeter by Instruments S.A., Inc.
The path length was 1 cm and all experiments were performed
6AI
7AI
−4.4
−6.6
4
N H 4AI
0.0
0.0
5
N H 5AI −1.4
−1.1
4.4
−9.2
6
◦
N H 6AI −1.0
at 20 C. Each of the azaindole isomers was analyzed in water
7
N H 7AI
3.6
and 1 M NaOH solution. The excitation wavelength for 4AI was
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 0 1 – 3 7 0 6
3 7 0 3

View Article Online
6
4
AI and N –H 6AI have nearly the same ground state energy as
Table 3 Fluorescence maxima of the tautomers and protonated azain-
dole isomers. Both 6AI and 7AI fluorescence maxima taken from the
specified reference
5
7
N –H 5AI, while N –H 7AI is highest in energy However, in the
excited state, N –H 7AI is the lowest in energy by as much as
3.6 kcal mol .
7
−
1
1
Isomer
4AI
Wavelength/nm
Spectroscopic studies
N/A
415
480
N/A
410
+
4
AI–H
The fluorescence spectra of 5AI and 4AI were measured to
determine if these isomers tautomerize in the excited state and
to characterize the fluorescence wavelength of the tautomer and
the protonated species. The N -methyl and N -methyl derivatives
were synthesized to help characterize the fluorescence spectra,
since the presence of a methyl group locks the azaindole into the
4
N H 4AI
AI
5AI–H
5
+
4
5
5
N H 5AI
450
350
380
440
385
440
500
a
6
6
AI
+
a
AI–H
6
a
N H 6AI
n
N –H tautomeric form. As seen from the spectra in Fig. 4 and
b
7
7
AI
5
, at high pH (i.e., pH = 14), 4AI and 5AI both tautomerize.
+
b
AI–H
7
b
In the case of 4AI, we see the presence of two bands, one at
N H 7AI
4
3
90 nm and another at 480 nm. The N -methyl-4AI fluorescence
spectrum shows a single band at 480 nm in the presence of
a
b
Values from ref. 7. Values from ref. 6.
4
1
M NaOH. The band at 480 nm is assigned to the N –H 4AI
4
tautomer based on the fluorescence spectrum of N -methyl-
AI. The fluorescence spectrum of 5AI in 1 M NaOH shows
Table 4 Computed and experimental absorbance wavelengths
4
5
only one broad band with a maximum at 455 nm. N -methyl-5AI
has a virtually identical fluorescence spectrum with a maximum
at 460 nm. The fluorescent band at 455 nm is assigned to
a
Isomer
4AI
Experimental
CIS
TDDFT
293
370
270
310
300
355
296
350
242
336
248
358
240
318
251
373
270
383
260
355
264
345
271
387
5
4
the N –H 5AI tautomer. For comparison, the spectra of 5AI
N H 4AI
5
AI
and 4AI in water are shown. These fluorescent bands are the
protonated form of the azaindole isomer. The protonated 4AI
has a maximum at 415 nm, and the protonated 5AI has a
5
N H 5AI
AI
6
4
N H 6AI
AI
7
7
N H 7AI
a
Values taken from ref. 19.
maximum at 410 nm. Table 3 summarizes the fluorescence data
for each of the azaindole isomers.
We explored the excitation energies using two different
methods. First, we used time-dependent DFT, which computes
22
a vertical excitation from the optimized ground state. In order
to estimate the effect of adiabatic transistions, the optimized
excited state geometries and energies were computed using the
CI Singles method; TD-DFT optimizations can currently not
yet be performed. The computed and experimental excitation
wavelength and excited state energies are listed in Table 4. The
experimental wavelengths were taken from ref. 19 Across the
board, TD-DFT vertical (without geometry optimization) ab-
sorptions were closer to experiment than those from optimized
excited state geometries at the CIS level of theory. This is easily
understandable as electronic absorption phenomena are vertical
Fig. 4 Normalized fluorescence spectra of 4AI in H
2
O (dot-dash line),
M NaOH (solid line) and N -Methyl-4AI in 1 M NaOH (dashed line).
4
1
5
in nature. With the exception of N –H 5AI and 6AI, all TDDFT
excitation wavelengths were within 25 nm of the experimental
values. The computed and experimental values listed in Table 4
both indicate a significant excitation energy difference between
1
n
the N –H and N –H tautomers. The difference arises from the
1
n
lower energy of the N –H species relative to the N –H tautomer
n
in the ground state. However, in the excited state, the N –H
tautomer is lower in energy relative to the N –H tautomer
1
(Table 1). As expected, the greater difference in the HOMO–
LUMO energy gap is reflected in the higher excitation energy
1
observed in the N –H azaindole tautomer.
Solvent effects
The energy of each azaindole tautomer in the presence of a
continuous solvent dielectric was explored to determine the
effects of a solvent dielectric on the tautomer energy differences.
All the azaindole isomers were stabilized by the dielectric
constant of the solvent, but the largest decrease in the energy
difference between tautomers occurred for 5AI and 6AI as seen
in Fig. 6 and Table 5. The energy difference of both isomers was
2
Fig. 5 Normalized fluorescence spectra of 5AI in H O (dot-dash line),
5
1
M NaOH (solid line) and N -Methyl-*5AI in 1 M NaOH (dashed
−
1
line).
decreased by 5.8 and 4.8 kcal mol , respectively. The energy
3
7 0 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 0 1 – 3 7 0 6

View Article Online
Table 5 Energy differences between tautomers in the presence of
different dielectrics
Table 7 NICS(1) values at B3LYP/6–31+G**
Molecule
AI
Pyridine ring
Pyrrole ring
Isomer
Gas phase Chloroform
Methanol
Water
4
−10.6
−9.1
−10.1
−12.9
−10.3
−13.2
−10.5
−13.4
−10.2
−12.5
—
4
4
5
6
7
AI
AI
AI
AI
11.6
11.4
10.1
12.5
9.1
7.4
6.7
8.3
5.8
5.5
9.1
8.2
5.6
5.3
9.0
N H 4AI
5AI
−10.8
−9.1
5
N H 5AI
10.0
6AI
−10.9
−8.6
4
N H 6AI
7
AI
−10.6
−9.4
7
N H 7AI
Benzene
−10.2
7
Fig. 7 Structure comparisons of 7AI, N -7AI and fulvene.
Fig. 6 Energy differences between tautomers in the presence of a
continuous solvent dielectric; × = 7AI, ꢀ = 6AI, ꢁ = 5AI, ᭜ =
shown that computing the chemical shielding at the center
of an aromatic ring can provide a qualititative measurement
4
-azaindole.
23
of the aromaticity of the ring. These types of computations
have been given the term NICS, nucleus independent chemical
difference of 4AI and 7AI tautomers decreased by 3.4 and 3.5
−
1
kcal mol , respectively.
˚
shieldings. NICS were computed 1 A above the center (NICS(1))
n
The effect of solvation was greater on the N –H tautomer than
of both the pyrrole and pyridine rings for each of the azaindole
1
on the N –H tautomer as shown in Table 6. The introduction
23
isomers (Table 7). Interestingly, a NICS(1) computation of
of a solvent dielectric constant induces an increase in the dipole
1
n
each ring shows that both the N –H tautomer and the N –
n
of the molecule, which occurs to a greater extent on the N –H
H tautomer are almost equally aromatic. In fact, the pyrrole
tautomer. As the solvent dielectric increases, a linear increase in
n
ring of the N –H tautomer is even more aromatic than the
n
the dipole of each isomer is observed. Since each of the N –H
1
N –H tautomer although the resonance structure suggests an
1
isomers has a higher dipole moment than the N –H tautomers,
exocyclic p bond. The NICS(1) value of −10.2 for benzene was
included as a reference point. This can be rationalized by dipolar
resonance structures that avoid the fulvene-type double bond
configuration and allow for complete cyclic, and hence aromatic,
delocalization (Fig. 7).
the solvation energy of that tautomer is greater. The higher
n
solvation energy of the N –H tautomer results in the lower
energy difference between the tautomers. Table 6 shows the sol-
vation stabilization energies of each isomer and their tautomer
n
by the water dielectric. Clearly, each of the N –H isomers has
a greater solvation stabilization energy than the corresponding
1
N –H tautomer. The greatest solvation stabilization energy was
Conclusions
5
6
observed for the N –H 5AI and the N –H 6AI isomers, while
1
the solvation energy of the corresponding N –H tautomers are
Both 4AI and 5AI display excited state tautomerization in
the presence of strong base demonstrated by the fluorescence
spectra in comparison with the N –CH
nearly the same, which explains the decreased energy difference
shown in Fig. 5.
n
3
analogs. The protonated
Gordon and co-workers suggested that part of the reason
form of 4AI has an emission maximum at 415 nm and the
1
7
4
for the greater stability of the N –H tautomer over the N –H
N –H tautomer has a fluorescence maximum at 480 nm. The
1
tautomer of 7AI is due to aromatic stabilization of the N –
protonated form of 5AI has an emission maximum at 410 nm
21
7
5
H tautomer. In the N –H tautomer, the pyrrole portion has
an exocyclic p bond, similar to fulvene, which is considered
to be non-aromatic (Fig. 7). Schleyer and co-workers have
and the N –H tautomer has a fluorescence maximum at 450 nm.
1
All the azaindole isomers are more stable in the N –H tautomer
−
1
by at least 10 kcal mol in the ground state. In the excited state,
n
all the N –H tautomers are lower in energy. The polarity of
the solvent helps to reduce the energy difference between the
Table 6 Energy differences from gas phase to solvent dielectric equal
−
1
tautomer in the ground state by as much as 5.8 kcal mol , but
not enough to reverse the ground state tautomer preference.
to water (78.4) at B3LYP/6–31+G**
1
−
1
The preference for the N –H tautomer appears to not be due to
Isomer
AI
Solvation stabilization energy/kcal mol
aromatic stabilization considering both tautomers are equally
aromatic.
4
8.1
11.5
8.3
14.2
8.0
12.8
6.5
4
N H 4AI
AI
5
5
N H 5AI
AI
Acknowledgements
6
4
N H 6AI
AI
This work was supported by the National Science Foundation
(PRS, CHE-0209857) and the National Institute of Health (RSP,
GM-42588).
7
7
N H 7AI
9.9
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 0 1 – 3 7 0 6
3 7 0 5

View Article Online
References
16 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr.,
R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D.
Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone,
M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford,
J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J.
Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
C. Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen,
M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle and J. A.
Pople, GAUSSIAN 98 (Revision A.9), Gaussian, Inc., Pittsburgh,
PA, 1998.
1
2
3
4
5
6
A. Clayton and W. Sawyer, Euro. Biophys. J., 2002, 31, 9.
C. Royer, Biophys J., 1993, 65, 9.
J. Lakowicz, Photochem. Photobiol., 2000, 72, 421.
Y. Engelborghs, Spectrochim. Acta, Part A, 2001, 57, 2255.
D. Creed, Photochem. Photobiol., 1984, 39, 537–562.
A. Smirnov, D. English, R. Rich, J. Lane, L. Teyton, A. Schwabacher,
S. Luo, R. Thornburg and J. Petrich, J. Phys. Chem. B, 1997, 101,
2
758.
7
S. Twine, L. Murphy, R. Phillips, P. Callis, M. Cash and A. Szabo,
J. Phys. Chem. B, 2003, 107, 637.
8
9
Q. Li, H. Du and H. Hu, Biopolymers, 2003, 72, 116.
C. Chang and T. Lin, Huaxue, 1997, 55, 83.
1
1
1
1
0 J. Catalan and M. Kasha, J. Phys. Chem. A, 2000, 104, 10812.
1
1
1
2
2
2
7 M. Sloan and R. Phillips, Bioorg. Med. Chem. Lett., 1992, 2, 1053.
8 C. Niu, J. Li, T. Doyle and S. Chen, Tetrahedron, 1998, 54, 6311.
9 I. Mahadevan and M. Rasmussen, Tetrahedron, 1993, 49, 7337.
0 M. Gordon, J. Phys. Chem., 1996, 100, 3974.
1 A. Becke, J. Chem. Phys., 1993, 98, 5648.
2 C. Lee, W. Yang and E. Parr, Phys. Rev. B, 1988, 37, 785.
3 C. Curutchet, C. J. Cramer, D. G. Truhlar, M. F. Ruiz-Lopez, D.
Rinaldi, M. Orozco and F. J. Luque, J. Comput. Chem., 2003, 24,
1 G. Chaban and M. Gordon, J. Phys. Chem. A, 1999, 103, 185.
2 C. Jamorski Jodicke and H. P. L u¨ thi, J. Am. Chem. Soc., 2003, 125,
2
84.
1
1
4 J. Foresman, M. Head-Gordon, J. Pople and M. Frisch, J. Phys.
2
52.
Chem., 1992, 96, 135.
2
2
2
2
2
3 P. v. R. Schleyer, M. Manoharan, Z. Wang, B. Kiran, H. Jiao, R.
Puchta and N. van Eikema Hommes, J. Org. Lett., 2001, 3, 2465.
4 S. Antolic, B. Kojic-Prodic and V. Magnus, Acta Crystallogr., Sect.
C, 2000, 56, 1026.
5 M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G.
Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Petersson,
J. A. Montgomery, K. Raghavachari, M. A. Al-Laham, V. G.
Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B. Stefanov,
A. Nanayakkara, M. Challacombe, C. Y. Peng, P. Y. Ayala, W. Chen,
M. W. Wong, J. L. Andres, E. S. Replogle, R. Gomperts, R. L. Martin,
D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker, J. P. Stewart, M. Head-
Gordon, C. Gonzalez, and J. A. Pople, GAUSSIAN 94, Gaussian,
Inc., Pittsburgh, PA, 1995.
5 V. Muralidharan, J. Cho, M. Trester-Zedlitz, L. Kowalik, B. Chait,
D. Raleigh and T. Muir, J. Am. Chem. Soc., 2004, 126, 14004.
6 V. De Filippis, S. De Boni, E. De Dea, D. Dalzoppo, C. Grandi and
A. Fontana, Protein Sci., 2004, 13, 1489.
7 M. Petersilka, U. J. Gossmann and E. K. U. Gross, Phys. Rev. Lett.,
1
996, 76, 1212.
3
7 0 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 3 7 0 1 – 3 7 0 6