Synthesis and fluorescence properties of 4-diarylmethylene analogues of the green fluorescent protein chromophore

Article abstract of DOI:10.1016/j.tet.2015.05.073

New green fluorescent protein (GFP) chromophore analogues, namely 4-(diarylmethylene)imidazolinones (DAINs), were readily synthesized under weakly acidic conditions using a novel condensation reaction between methyl imidate (or thioimidate) and ethyl N-(diarylmethylene)glycinate. DAINs showed notable fluorescence properties. Although they were nearly non-fluorescent in the solution, visible emissions were detected in most of their frozen solution states and crystalline powder states. Therefore, control of the molecular motions significantly affected emissions by DAINs. Comparison of the fluorescence properties of DAIN 5a with those of the corresponding GFP-chromophore analogues 8 revealed that 5a possessed superior solid-state fluorescence properties.

Full text of DOI:10.1016/j.tet.2015.05.073

Tetrahedron xxx (2015) 1e12
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis and fluorescence properties of 4-diarylmethylene
analogues of the green fluorescent protein chromophore
Masahiro Ikejiri a, Kousuke Matsumoto a, Hiraku Hasegawa a, Daisuke Yamaguchi a,
Moe Tsuchino a, Yoshiko Chihara a, Takao Yamaguchi b, Kazuto Mori b,
,
Takeshi Imanishi b, Satoshi Obika b, Kazuyuki Miyashita a
*
a Faculty of Pharmacy, Osaka Ohtani University, Nishikiori-Kita 3-11-1, Tondabayashi, Osaka 584-8540, Japan
b Graduate School of Pharmaceutical Sciences, Osaka University, Yamadaoka 1-6, Suita, Osaka 565-0871, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
New green fluorescent protein (GFP) chromophore analogues, namely 4-(diarylmethylene)imidazoli-
nones (DAINs), were readily synthesized under weakly acidic conditions using a novel condensation
reaction between methyl imidate (or thioimidate) and ethyl N-(diarylmethylene)glycinate. DAINs
showed notable fluorescence properties. Although they were nearly non-fluorescent in the solution,
visible emissions were detected in most of their frozen solution states and crystalline powder states.
Therefore, control of the molecular motions significantly affected emissions by DAINs. Comparison of the
fluorescence properties of DAIN 5a with those of the corresponding GFP-chromophore analogues 8 re-
vealed that 5a possessed superior solid-state fluorescence properties.
Received 15 April 2015
Received in revised form 20 May 2015
Accepted 21 May 2015
Available online xxx
Keywords:
Fluorescence
Chromophore
Imidazolinones
Green fluorescent protein
Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction
Green fluorescent protein (GFP) is a well-known fluorescent
protein, and is widely employed as a useful imaging tool because it
can help visualise various biological events through highly sensitive
fluorescence detection.1 The chromophore within GFP is formed
from a Ser-Tyr-Gly tripeptide in the primary protein structure by
post-translational cyclodehydration and subsequent oxidation to
yield the 4-(p-hydroxybenzylidene)imidazolinone (HO-BDI) moi-
ety. The conformation of the HO-BDI chromophore is restricted to
the Z-form within the GFP b-barrel tertiary structure, which en-
ables emission by GFP.2 However, outside the protein, fluorescence
quenching of the chromophore is induced mainly by molecular
motions such as the free rotation of the arylealkene single bond
and the double bond isomerization of the Z-4-hydroxybenzylidene
moiety (‘bright’ form) to the corresponding E-isomer,3,4 which is
generally known to exhibit extremely weak fluorescence (‘dark’
form),5 although exceptions are known6 (Fig. 1A).
In an effort to prepare newsmall-molecule fluorescent materials,
we designed a series of diarylmethylene-based compounds, namely
4-(diarylmethylene)imidazolinones (DAINs), as novel analogues of
the GFP chromophore (Fig. 1B). Our design comprised three
* Corresponding author. Tel.: þ81 721 24 9652; fax: þ81 721 24 9652; e-mail
Fig. 1. A. Structure of the GFP chromophore (HO-BDI). B. Design of the model com-
pound, 4-(diarylmethylene)imidazolinone (DAIN).
address: miyasik@osaka-ohtani.ac.jp (K. Miyashita).
http://dx.doi.org/10.1016/j.tet.2015.05.073
0040-4020/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

2
M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
structural generations. The first generation contained a diphenyl-
methylene unit on the imidazolinone ring. Because this moiety is
symmetrical, the structure is the same after the double-bond
isomerization. Therefore, the fluorescence (attributed to the Z-
form) could be maintained regardless of the isomerization. The
second- and third-generation structures were designed by focusing
on the diarylmethylene moiety. As a common imidazolinone moiety
of these two generations, we selected a 1,2-pentanoimidazolinone
structure based on the fluorescence properties of the first-
generation DAINs 1e7 (Fig. 2), whose details are described in Sec-
tion 2.2.1.1. The second-generation structures were designed to
elucidate substituent effects (electron donating and electron with-
drawing) on the diphenyl moiety of the first-generation compounds
(DAINs 5b and 5c). In the third generation, the diphenyl rings were
designed to be in nearly the same plane as the imidazolinone ring by
installing a bridge between the diphenyl rings (DAINs 5d and 5e).
Such installation would prevent CePh free rotation. We expected
that this third generation would exhibit the strongest emission
because of the overall structural advantages described above. Al-
though several types of 4-benzylideneimidazolinone (BDI) ana-
logues with substantial fluorescence have been reported so far,7e10
to the best of our knowledge, there has been no report on such
diarylmethylene analogues of the GFP chromophore. In this study,
we report the synthesis of the DAIN series (Section 2.1) and their
fluorescence properties, including a comparison of DAIN 5a with
analogous BDI compounds, Z- and E-8 (Section 2.2).11
a suitable concise condensation reaction between methyl thio-
imidate 9a and ethyl N-(diphenylmethylene)glycinate 10a, as
shown in Table 1. Although the product, 1, was not yielded by
mixing 9a and 10a at room temperature, it was obtained in 52%e
60% yield in hot solvents like toluene and 1,2-dichloroethane, al-
though a long reaction time was required (Table 1, runs 1e4).
However, contrary to our expectations, the product was not
Table 1
Reaction of thioimidate 9a with N-(diphenylmethylene)glycinates 10aed
Run
10
Temp (ꢀC)
Additive [equiv]
Time (days)
Yield (%)
1
2
3
4
5
6
7
8
a
a
a
a
a
a
a
a
a
a
a
a
b
c
rt
70
80
80
Reflux
80
80
None
None
None
None
7.0
7.0
4.0
3.0
7.0
2.5
3.0
1.0
1.0
1.0
0.1
1.0
1.0
1.0
1.0
0
55
60
52a
Traceb
43
None
EtOH [5.0]
BuSH [1.0]
AcOH [1.0]
AcOH [2.0]
MsOH [2.0]
ZnCl2 [2.0]
LiCl [2.0]
AcOH [2.0]
AcOH [2.0]
AcOH [2.0]
58
64
80
2. Results and discussion
9
Reflux
rt to reflux
Reflux
Reflux
Reflux
Reflux
Reflux
78
10
11
12
13
14
15
0c
0c
2.1. Synthesis of the DAIN series
0b
7d
2.1.1. Optimization of reaction conditions. Although several syn-
thetic approaches towards BDI-based structures have been re-
ported,10,12,13 no effective approaches towards DAIN structures are
known. Notably, the syntheses used for BDI10,12 were ineffective in
the synthesis of DAIN 1. Thus, we endeavoured to develop a novel
synthetic method to obtain DAIN 1, and eventually identified
12d
13d
d
a
ClCH2CH2Cl was used as the solvent.
Almost no reaction.
Complex mixture.
b
c
d
The reaction was not complete within 24 h.
Fig. 2. Structures of DAIN series and BDI analogues.
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
3
Table 2
obtained under reflux in toluene (run 5). As this result appeared to
Reaction of (thio)imidates 9bei with N-(diphenylmethylene)glycinate 10a
suggest an important role of the volatile methanethiol and/or
ethanol produced during the course of the reaction, the reaction
was conducted in the presence of a mild protic additive (runs 6e9).
The addition of protic additives shortened the reaction time, and
acetic acid in refluxing toluene produced the best result (run 9). The
use of a stronger protic acid (methanesulfonic acid) or Lewis acids
such as ZnCl2 and LiCl were ineffective (runs 10e12). Surprisingly,
the identity of the alkyl group on the ester moiety of 10 significantly
influenced the reaction. The use of an ethyl group afforded
a promising result, whereas the use of other moieties such as
methyl, propyl and tert-butyl groups resulted in a slow reaction
(run 9 vs runs 13e15). Thus, this reaction seemed to be affected by
the steric hindrance and electron donation ability of the alkyl group.
Run
9
Temp
Time (hours)
DAIN
Yield (%)
1
2
3
4
5
6
7
8
b-S
c-S
d-S
d-S
e-S
f-S
g-S
b-O
c-O
f-O
f-O
f-O
f-O
h-O
b-O
f-O
i-S
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
rt
24
36
36
24
36
48
36
5
24
4
6
24
4
2
43
77
42
75a
68
27
89
71
85
59
29b
25c
5d
3a
3b
3b
4
5a
6
2.1.2. Synthesis of 1st-generation DAINs. To obtain the first-
generation DAINs, various thioimidates and imidates, shown in
Fig. 3, were used as the starting materials; the results are sum-
marised in Table 2. The products of the reactions with most thio-
imidates were obtained in good to moderate yields (Table 2, runs
1e7). An increased amount of acetic acid was effective in the case of
slow reaction (run 3 vs 4). Interestingly, the use of imidates was
effective (runs 8e10, and 14), and afforded better results than those
of the corresponding thioimidates in regard to the yields and re-
action times (run 1 vs 8; 2 vs 9; 6 vs 10), even though the imidates
were less stable than the thioimidates. The reaction could even be
conducted at room temperature using aliphatic imidates such as
9b-O and 9f-O, although a longer reaction time was required (runs
15 and 16). A decreased amount of 10a, the use of 10b, and AcOH-
free conditions resulted in poor yields even when imidate 9f-O was
used (run 10 vs runs 11e13). Unfortunately, aromatic imidate and
thioimidate 9i were not viable substrates in this reaction (runs 17
and 18).
2
9
3a
5a
5a
5a
5a
7
10
11
12
13
14
15
16
17
18
20
96
24
24
24
77
32
60
0
2
rt
Reflux
Reflux
5a
d
d
i-O
0
a
4.0 equiv of AcOH was used.
3.0 equiv of 10a was used.
5.0 equiv of 10b was used.
b
c
d
Without AcOH (reaction did not go to completion).
Fig. 4. Structures of N-(diarylmethylene)glycinates 10eeh.
Table 3
Reaction of imidate 9f-O with N-(diarylmethylene)glycinates 10eeh
Fig. 3. Structures of thoimidates and imidates 9bei.
Run
10 [equiv]
Conditions
5
Yield (%)
1
2
3
4
5
6
7
8
e [5.0]
f [5.0]
g [3.0]
h [3.0]
e [5.0]
f [5.0]
e [5.0]
f [5.0]
rt, 24h
rt, 34h
rt, 24h
rt, 24h
Reflux, 4h
Reflux, 8h
Reflux, 18h
Reflux, 18h
b
c
d
e
b
c
70
52
85
74
42
15
28
9
2.1.3. Synthesis of 2nd- and 3rd-generation DAINs. To obtain the
second- and third-generation DAINs, coupling reactions with imi-
date 9f-O and several glycinates with a particular diarylmethylene
moiety, as shown in Fig. 4,14 were respectively conducted, and the
results are summarized in Table 3.
All reactions proceeded smoothly even at room temperature
(runs 1e4); notably, bridged type N-(diarylmethylene)glycinates
10g and 10h reacted despite the use of fewer equivalents (3.0 equiv,
runs 3 and 4). The substituent on the benzene ring of the diphenyl
moiety affected the reaction yield; an electron-donating group
(eOMe, 10e) gave a higher yield (70%), whereas an electron-
withdrawing group (eCl, 10f) resulted in a lower yield (52%) as
compare to no substituent (eH, 10a, 60%) (runs 1 and 2 vs Table 2,
run 16). Therefore, the stability of the benzylic carbocation at the
b
c
diarylmethylene moiety seemed to affect the reactivity. Notably, at
reflux, the substituents did not seem to have a noticeable effect
(runs 5 and 6 vs Table 2, run 10) probably due to the instability of
the product at high temperatures; for instance, prolonged treat-
ment of 10e and 10f under reflux for 18 h resulted in a marked
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

4
M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
decrease in the product yields (runs 7 and 8), as opposed to that of
10a (reflux, 18h, 51% yield of 5a).
Section 2.2.2.2. IR and UVevis absorptions were also helpful in
elucidating the DAIN structures, due to characteristic absorptions
(Table 4). In the case of diphenylmethylene-containing species, two
strong absorptions were detected in the IR spectra around
1615e1640 cmꢁ1 and 1700e1715 cmꢁ1, probably owing to the two
carbon-heteroatom double bonds in the imidazolinone ring; in
addition, an absorption was detected around 360e400 nm in the
UVevis spectra, which was attributed to the DAIN ring system (runs
1e8). Modification of the diphenyl moiety shifted the absorptions
slightly; the wavenumber of the IR absorption tended to decrease
except for one absorption of 5c (1705 cmꢁ1), while the UVevis
absorption was red-shifted (runs 6 vs 9e12).
2.1.4. Reaction mechanism. A possible reaction mechanism for the
DAIN synthesis is depicted in Fig. 5. A notable feature of this re-
action is that the diarylmethylene moiety migrates from the ni-
trogen to the
and not the
a
-carbon. Specifically, an attack of the imine-nitrogen,
a-carbanion of 10, on the (thio)imidate 9 appears to be
the first step of this reaction to form iminium intermediate A. In
this step, the addition of a proton source would aid in the activation
of the (thio)imidate-nitrogen. Subsequently, following the enola-
tion and transformation into carbocation intermediate B, an
intramolecular aziridine formation would occur to afford C and
then D. Concomitant aziridine-ring cleavage of D promoted by a-
deprotonation would produce the DAIN structure, accompanied by
the migration of the diarylmethylene moiety from the nitrogen to
Table 4
Characteristic IR and UVevis absorption of DAIN series
the
a-carbon. Although we could not isolate the aziridine in-
a
Run
DAIN
IR (cmꢁ1
)
UVevis (nm)b
termediate, Ayyangar and co-workers reported an aziridine syn-
thesis from glycinate 10a.15
1
2
3
4
5
6
7
8
1
2
1620, 1715
1619, 1703
1634, 1710
1637, 1708
1620, 1706
1623, 1700
1625, 1712
1616, 1707
1601, 1696
1619, 1705
1601, 1696d
1605, 1688
357
365
370
390c
397
362
367
390
384
370
406
412
3a
3b
4
5a
6
7
9
5b
5c
5d
5e
10
11
12
a
Representative wavenumbers around 1600e1720 cmꢁ1 (diffuse reflection in
KBr).
b
c
lmax of the longest wavelength absorption (1.0ꢂ10ꢁ5 M in CH2Cl2).
Detected as a shoulder peak around 360e420 nm.
Another absorption was detected at 1626 cmꢁ1 (medium).
d
2.1.6. Synthesis of BDI analogues 8. BDI analogues 8, which are
analogous to 5a, were synthesised for comparison, as shown in
Scheme 1. Initially, we attempted to apply the synthetic method of
DAINs to the synthesis of BDIs by using imidate 9f-O and glycinate
11 instead of 10a. However, the desired product Z-8 was not ob-
tained. Because a convenient BDI synthesis based on the aza-Wittig
reaction has been reported,10 we adopted the reported method as
an alternative route. Lactam 12 was transformed into azidoacetyl
derivative 14 through chloroacetyl intermediate 13 in 33% yield
over two steps. The aza-Wittig reaction of 14 was performed by
treatment with Ph3P to afford imidazolinone 15 in 90% yield. Be-
cause the total yield of these three steps was moderate (30%), we
also adopted a modified version of Campagna’s procedure16 using
glycine methyl ester hydrochloride, by which 15 was easily ob-
tained from imidate 9f-O in good yield (96%). Finally, imidazolinone
15 was coupled with benzaldehyde in the presence of piperidine to
afford BDI Z-8 exclusively in 70% yield. The Z-benzylidene moiety in
Z-8 was isomerised under UV light to form the E-isomer E-8 in 43%
yield, as with other reported BDI analogues.3,5 The E-isomer was
reisomerised to Z-8 under UV or visible light. The Z-, E-geometry of
8 was determined by comparison of the 1H NMR chemical shifts of
the olefin hydrogen to those of structurally similar known
compounds.5
Fig. 5. A possible reaction mechanism for the DAIN synthesis.
The different reactivities of N-(diarylmethylene)glycinates
10aeh could be explained if the reaction occurred via the
aforementioned pathway. The reactivity derived from the dia-
rylmethylene moiety would be related to the carbocation for-
mation step (A to B), since a stable carbocation seemed to
accelerate the reaction as shown in Table 3. The choice of the
ester alkyl group (Table 1 runs 9, 13e15), would affect the azir-
idine formation (B to C) and the final cyclization (D to DAIN). The
aziridine formation would be facilitated by the electron donation
of the ester alkoxy group, but this step and the final cyclization
step would be interrupted by the steric hindrance of the ester
alkyl group due to the crowded reaction space. The ethyl group
could be of a suitable size to meet these factors (high electron
donation and small size). Additionally, the electron donation of
the diaryl moiety with methoxy substitutions and that of the
ester alkyl group would also accelerate the first addition step of
10 to 9.
2.2. Fluorescence evaluation
2.2.1. Results
2.1.5. Analysis of the DAIN structure. The structures of the DAIN
analogues were confirmed with 1H NMR, 13C NMR, mass spec-
trometry, and elemental analysis. In addition, for representative
samples (i.e., 1, 3a, 5d and 6), the structure was confirmed with X-
ray crystal structural analysis, whose details are described in
2.2.1.1. Fluorescence properties of 1st-generation DAINs. Initially,
we examined the fluorescence properties of the first-generation
DAINs 1e4, 5a, 6 and 7 (1st-DAINs) in CH2Cl2, benzene and di-
methyl sulfoxide (DMSO); however, all the compounds were nearly
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
5
Table 5
Fluorescence of DAINs 1e7 in the crystalline powder state
Run
DAIN
Gen.
lex
lem
R.I.a
1
2
1
2
1st
1st
1st
1st
1st
1st
1st
1st
2nd
2nd
3rd
3rd
370
369
369
370
367
370
371
369
370
371
403
491
506
504
531
540
505
513
493
535
517
486
566
563
0.03
0.17
3g
4g
5
3a
3b
4
5a
6
0.72b
0.81
1.13c
1.00d
2.20e
0.31
6
7
8
9
10
11
12
7
5b
5c
5d
5e
0.50f
0.06
<0.01
0.01
a
Emission intensity relative to that of 5a.
b
c
d
e
f
F3a: 9.8%.
F4: 19.4%.
F5a: 16.3%.
F6: 33.6%.
F5b: 8.3%.
g
Another prominent emission was detected (Fig. 8).
Scheme 1. Synthesis of BDIs 8. Reagents and conditions: (a) 11, AcOH, toluene, reflux;
(b) ClCH2COCl, Et3N, CH2Cl2, rt; (c) NaN3, DMSO, rt, 33% from 12; (d) Ph3P, toluene, rt,
90%; (e) HCl$H2NCH2CO2Me, Et3N, MeOH, reflux, 96% (f) PhCHO, piperidine, MeOH, rt,
70% (g) 365 nm UV, benzene, rt, 43% (recovery of Z-8, 42%).
non-fluorescent in the solution state, contrary to our expectations.
In contrast, a number of analogues exhibited emission visible to the
naked eye in the crystalline powder state,17 with the exception of 1.
Real-colour photographs of the fluorescence of crystalline powders
of 1st-DAINs are shown in Fig. 6 (black numbers correspond to 1st-
DAINs). Emissions ranging in colour from cyan to yellow were vi-
sually observed. The fluorescence profiles of crystalline powder
1st-DAINs are listed in Table 5 (runs 1e8), and excitation and
emission spectra of representative compounds are given in Fig. 7. As
an approximate estimation of the fluorescence intensity, the rela-
tive intensity (R.I.) of each analogue was calculated from the in-
tegral of the emission spectrum to that of standard compound 5a,
and regarding representative analogues, their quantum yields (F)
were also measured. The excitation wavelengths (lex) were detec-
ted at approximately 370 nm for most compounds, although it was
difficult to assign accurate values because of the broadness of the
spectra. In contrast, the emission wavelengths (lem) were detected
Fig. 7. Spectra of crystalline powders 1, 3a, 4, 5a, 5b and 6. A. Excitation spectra. B.
Emission spectra. lex: 370 nm for 1, 5a, 5b; 367 nm for 4, 369 nm for 3a, 371 nm for 6.
between 493 and 540 nm. Interestingly, unlike other 1st-DAINs,
imidazoquinoline-type DAINs 3a and 3b exhibited another prom-
inent emission in the crystalline powder state at a longer wave-
length (lex: 477 nm, lem: 558 nm for 3a; lex: 482 nm, lem: 559 nm
for 3b) with a good quantum yield (F3a at lex¼477 nm: 47.3%).
Contour plots of 3a and 3b, and their emission spectra corre-
sponding to the longer excitation wavelength are shown in Fig. 8.18
Because there was no UVevis absorption at the longer wavelength
in the solution state of 3a and 3b, (Table 4, runs 3 and 4) the
emissions at longer wavelengths were derived from their solid
states, probably because of aggregated states, such as a J-aggregate
system.19,20 It also should be noted that with 1st-DAINs, visible
fluorescence emission was observed when the solution was frozen
in most cases. Real-colour photographs of the fluorescence from the
frozen solutions are shown in Fig. 9A (black numbers correspond to
1st-DAINs). Visually, emissions ranging from blue to yellow were
observed. Interestingly, the emission seemed to depend on the
solvent; for example, the emission of 1 was not visible in frozen
DMSO, whereas 1 exhibited a blue emission in frozen benzene.
Notably, BDI-type compounds were reported to exhibit fluores-
cence in frozen solutions at low temperatures (in ethanol at
ꢁ196 ꢀC),21 whereas DAIN analogues showed fluorescence at
higher temperatures ranging from ꢁ20 ꢀC (e.g., in DMSO and
Fig. 6. Real-colour photographs of the fluorescence of crystalline powder DAINs 1e7
(black numbers: 1st-DAINs, red numbers: 2nd-DAINs, blue numbers: 3rd-DAINs). The
pictures were taken under visible light (top panel) and 365 nm UV lamp (6 W) (bottom
panel). The picture with asterisks (*) was obtained with a prolonged shutter speed (ꢂ
8).
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

6
M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
DAIN 5a as a model compound of the first-generation DAINs be-
cause of its promising solid-state fluorescence (crystalline powder
and frozen solution), acceptable quantum yield (16.3%), longer
wavelength emission (513 nm) and ease of synthesis.
2.2.1.2. Fluorescence properties of 2nd- and 3rd-generation
DAINs. In the solution state (in CH2Cl2, benzene and DMSO), the
second-generation DAINs 5b and 5c (2nd-DAINs) and third-
generation DAINs 5d and 5e (3rd-DAINs) were nearly non-
fluorescent, similar to the 1st-DAINs. Moderate emission from the
crystalline powder form was observed for one 2nd-DAIN (5b), as
shown in Fig. 6 (red numbers for 2nd-DAINs), Table 5 (run 9) and
Fig. 7, although the intensity was lower than that of 5a (Table 5, run
6). However, the emission of the other 2nd-DAIN (5c) was much
weaker (Table 5, run 10). The emission wavelengths of the 2nd-
DAINs were shifted to some extent from that of 5a (run 6 vs runs 9
and 10). Surprisingly, 3rd-DAINs were almost non-fluorescent even
in the crystalline powder state, although xanthene-containing 5e
emitted a slight orange fluorescence (Fig. 6, blue numbers corre-
spond to 3rd-DAINs, and Table 5, runs 11 and 12). Similar results
were also observed in frozen solutions. In particular, 5b exhibited
visible fluorescence, whereas 5c and 3rd-DAINs showed weak or
almost no fluorescence (Fig. 9, red numbers for 2nd-DAINs and
blue numbers for 3rd-DAINs).
Fig. 8. Dual emission of 3a and 3b. A. Excitationeemission contour plots of crystalline
powders 3a and 3b. (lex, lem): a1 (369 nm, 531 nm), a2 (477 nm, 558 nm), b1 (370 nm,
540 nm), b2 (482 nm, 559 nm). B. Emission spectra of crystalline powders 3a and 3b at
the longer excitation wavelength, a2 and b2. 3a: lex¼477 nm, 3b: lex¼482 nm
F3a¼47.3%, intensity of 3b relative to 3a¼1.13.
2.2.1.3. Comparison of the fluorescence properties of DAIN 5a with
those of BDIs 8. UVevis absorption and emission spectra in CH2Cl2
and emission spectra of the crystalline powder forms of DAIN 5a,
BDIs Z-8 and E-8 are shown in Fig. 10. In solution, the absorption
maximum of the longest-wavelength absorption band of DAIN 5a
appeared at a slightly longer wavelength than those of BDIs Z-8 and
E-8 (Fig. 10A, lmax: 362, 358 and 359 nm, respectively), even though
the molar absorption of 5a at the absorption maximum was weaker
than that of Z-8 (log ε: 4.09 and 4.32, respectively). The fluores-
cence in solution was quite weak for DAIN 5a and BDIs 8, as shown
in Fig. 10D; however, emission of BDI Z-8 was somewhat greater
(FZ-8: approximately 0.05%) (Fig. 10B). In contrast, the crystalline
powder of DAIN 5a and BDI Z-8 displayed naked-eye visible
emission, as shown in Fig. 10E; in this case, DAIN 5a exhibited
a higher quantum yield and a longer emission wavelength than BDI
Z-8 (F5a: 16.3%, FZ-8: 1.8%; lem 5a: 513 nm, lem Z-8: 449 nm)
(Fig. 10C). Notably, BDI E-8 was almost non-fluorescent even in the
crystalline powder state, which may support the ‘dark’ form of the
GFP-chromophore mentioned in the introduction. Although the
reason for the non-fluorescence of E-8 is unclear, we suppose that
the E-phenyl group would be twisted toward the imidazolinone
moiety due to steric hindrance, similar to nonfluorescent E-BDI
type chromoproteins.22 Overall, it appears that the DAIN structure
possessed superior fluorescence properties in the solid state to the
corresponding BDI structures in terms of the higher quantum yield
and longer emission wavelength. Therefore, the DAIN displayed
better ‘off’ to ‘on’ fluorescence upon changing from the solution to
the solid state.
Fig. 9. Real-colour photographs of the fluorescence of DAINs 1e7 in solution (black
numbers: 1st-DAINs, red numbers: 2nd-DAINs, blue numbers: 3rd-DAINs). A. Pictures
of DMSO solutions of 1st-DAINs and 2nd-DAINs (1.0ꢂ10ꢁ3 M) under 365 nm UV light
(top panel), those of frozen DMSO (FDM) solutions (middle panel) and those of frozen
benzene (FBN) solutions (1.0ꢂ10ꢁ3 M) (bottom panel). B. A picture of solidified bi-
phenyl solutions of representative DAINs (1.0ꢂ10ꢁ3 mol/kg) under 365 nm UV light. C.
Pictures of DMSO solutions of 3rd-DAINs (1.0ꢂ10ꢁ3 M) under 365 nm UV light (left
panel) and those of FDM and FBN solutions (right panel). The picture with asterisks (*)
was obtained with a prolonged shutter speed (ꢂ 8). [AeC] The DMSO and benzene
solutions were prepared at rt and then frozen at ꢁ20 ꢀC. The biphenyl solutions were
prepared at 90 ꢀC and then solidified at rt. All photographs were taken under a 365 nm
UV lamp (6 W).
2.2.2. Discussion
2.2.2.1. Fluorescence properties of the DAIN series. 2.2.2.1.1. In
the solution state. To our surprise, all compounds, regardless of
generation, were nearly non-fluorescent in the solution state. This
result would suggest that the excited energy was consumed by
molecular motions, even in the case of DAIN, similar to the case
with BDI. Although the photoinduced transformation from the Z-to
E-isomer was excluded in the DAINs, the double bond isomeriza-
tion at the diarylmethylene moiety probably occurred continu-
ously, similar to a molecular motor. In fact, some compounds with
a diarylmethylene group are known to work as light-driven
benzene) to room temperature (rt) (e.g., in biphenyl, Fig. 9B).
Overall, most 1st-DAINs (e.g., 3, 4, 5a and 6) exhibited bright
fluorescence in the solid state, such as in crystalline powder and in
frozen solution, whereas they did not exhibit fluorescence in the
solution state. For studies on the second- and third-generation
DAINs and as a comparison with the BDI analogues, we selected
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
7
a negative effect on the emission intensity, regardless of electron
donating (eOMe, 5b) or withdrawing (eCl, 5c) ability, whereas the
emission wavelength was slightly red-shifted for electron donating
5b and blue-shifted for electron withdrawing 5c (run 6 vs runs 9
and 10). To our surprise, the bridging modification at the diaryl
moiety (i.e., 3rd-DAINs) was ineffective even in the solid state,
despite the restraint of the carbonecarbon double bond rotation in
the solid in addition to the rigid and flat structure (runs 11 and 12).
As such, the fluorescence of the DAIN analogues in the solid state
probably depended on several factors; for example, not only the
conformation and electron density of the molecule, but also the
degree of vibrations and its surrounding conditions in the solid
state can be contributing factors. However, the restraint of molec-
ular motions seemed to be the most crucial factor. Details regarding
the crystalline conditions are discussed in Sections 2.2.2.2 and
2.2.2.3.
Overall, the observed fluorescence properties of the DAIN series
differed significantly from our initial expectations; however, it is
noteworthy that DAINs, particularly non-bridged DAINs, have
a useful function in that their fluorescence is turned ‘on’ when their
molecular motions are restrained. Although BDI analogues exhibit
a similar off/on property, DAINs possessed a superior off/on effect in
that the DAIN showed lower fluorescence (i.e., ‘off’) in solution
(Fig. 10B) and a higher fluorescence (i.e., ‘on’) in the solid state
(Fig. 10C). Accordingly, DAINs could be utilised in fluorescence off/
on switching systems by controlling their molecular motions.
2.2.2.2. Conformational analysis of DAINs with X-ray crystal
structures. The X-ray crystal structures of representative DAINs 1,
3a, 5d and 6 are shown in Fig. 11.26 It has been reported that the
phenyl and imidazolinone (IMN) rings in BDI, including in GFP,2 are
almost coplanar in the crystalline state (Fig. 12).27 In contrast, the
diphenyl moiety in DAIN, except for fluorine-containing 5d,
exhibited a twisted conformation, as shown in Table 6. Although
the torsion between Z-Ph and IMN was somewhat less (approxi-
mately 30e40ꢀ) than that between E-Ph and IMN (approximately
60e70ꢀ), it was not as flat as that of BDI. Nonetheless, some con-
jugation seemed to exist between the Z-Ph and IMN moieties in
DAIN, when considering the fluorescence properties of DAIN and
BDI, as mentioned above. In contrast, 5d exhibited a planar con-
formation, as expected. Therefore, factors other than its confor-
mation must be considered to explain the very weak fluorescence
of the crystalline powder form of 5d.
Fig. 10. A. UVevis absorption spectra for 5a, Z-8 and E-8 in CH2Cl2 solution
(1.0ꢂ10ꢁ5 M). lmax: 5a¼362 nm (log ε: 4.09), Z-8¼358 nm (log ε: 4.32), E-8¼359 nm
(log ε: 4.13). B. Emission spectra for 5a, Z-8 and E-8 in CH2Cl2 solution (1.0ꢂ10ꢁ5 M).
Scattered light (approximately 400e405 nm) was eliminated. The lmax of UVevis
absorption was used as the excitation wavelength for all compounds. lem: 5a¼432 nm,
Z-8¼416 nm, E-8¼429 nm. FZ-8: approximately 0.05%. C. Emission spectra for 5a, Z-8
and E-8 in the crystalline powder state. lem: 5a¼513 nm (lex 370 nm), Z-8¼449 nm
(
lex 371 nm), E-8¼465 nm (lex 370 nm). F5a: 16.3%, FZ-8: 1.8%. D. A photograph of the
CH2Cl2 solution of 5a, Z-8 and E-8 under a UV lamp (365 nm, 6 W). [E] A photograph of
the crystalline powder of 5a, Z-8 and E-8 under visible light (top panel) and 365 nm UV
lamp (6 W) (bottom panel).
molecular motors.23 Fluorescence quenching caused by twisted
intramolecular charge transfer (TICT) at the olefin moiety in the
DAIN structure may also be conceivable.24 In the case of the first-
and second-generation DAINs, energy consumption by the single-
bond rotations of the arylealkene bonds in the diaryl moiety was
also possible in addition to the double bond rotation, which would
be supposable from the structural similarity between BDI3 and
DAIN. Overall, the emission of all DAIN analogues was nearly non-
existent in the solution state, including the third-generation DAINs,
which had rigid and flat conformations.
2.2.2.1.2. In the solid state. In contrast, several DAIN analogues
exhibited naked-eye visible bright emission in the crystalline
powder state (so-called aggregation-induced emission19,25) and in
frozen solution. The emission from such solid states may have been
produced by the restraint of the molecular motions, which would
lead to a decrease in the non-radiative energy consumption. With
regard to 1st-DAINs, we initially postulated that a fused benzene
ring conjugated to the imidazolinone ring would be important for
substantial emission (Table 5, runs 1e5).11 However, further trials
revealed that such a structure was not essential. Compounds 5a and
7 showed suitable emission without the benzene ring or fused ring
structure, respectively (Table 5, runs 6 and 8). Interestingly, modi-
fication of the diaryl moiety (i.e., 2nd- and 3rd-DAINs) led to re-
duced emission in the solid state (Table 5, run 6 vs runs 9e12). The
substituent effect on the diphenyl moiety seemed to have
2.2.2.3. Molecular packing analysis of X-ray crystal structures of
DAINs. To obtain a better understanding of the fluorescence of the
crystalline powder form of DAINs, molecular packing analysis of the
X-ray crystal structures was performed with fluorescent DAINs 3a
and 6, and nearly non-fluorescent DAINs 1 and 5d. A remarkable
difference between the fluorescent and non-fluorescent DAINs was
found in their space groups, as shown in Fig. 13. Fluorescent com-
pounds 3a and 6 belonged to the P21/c group (top two panels),
while non-fluorescent 1 and 5d belonged to other groups, such as
P-1 and C2/c (bottom two panels). Regarding solid-state fluores-
cence, it is known that intermolecular
pep interactions typically
lead to fluorescence quenching (non-radiative process) because of
electron and/or energy transfers.19,28 The molecular packing of
non-fluorescent DAINs
quenching. In both cases, neighbouring molecules were packed in
an anti-parallel fashion and the -bonds overlapped each other
1 and 5d appeared to lead to such
p
ꢀ
within 3.6e5.0 A, which would be close enough for such transfers
via the Dexter mechanism, as shown in Fig. 14. These compounds
might be forming excimers with an intermolecular distance of
approximately 3.6e3.7 A, resulting in the quenched emission. In
contrast, fluorescent DAINs 3a and 6 were packed in a stepwise
form with a head-to-tail configuration, as shown in Fig. 15.
29
ꢀ
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

8
M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
ꢀ
may explain the different emission colours (i.e., 3a: 3.95 A, yellow
30
ꢀ
fluorescence; 6: 4.51 A, cyan fluorescence).
3. Conclusions
SynthesisdThe newly designed GFP chromophore analogues,
DAINs, were successfully synthesized using a novel condensation
reaction between (thio)imidate 9 and N-(diarylmethylene)glyci-
nate 10. This reaction proceeded smoothly under weakly acidic
conditions in refluxing toluene or, in some cases, at room temper-
ature. Although both thioimidates and imidates were viable com-
ponents in this reaction, imidates appeared to afford a better yield
in a shorter reaction time. This reaction was affected by properties
of the diarylmethylene moiety; the stability of the benzylic carbo-
cation at the diarylmethylene moiety seemed to influence the re-
activity. The reaction mechanism included an unprecedented
migration step of the diarylmethylene group from the nitrogen to
a-carbon probably via an aziridine intermediate.
Fluorescence propertiesdThe DAINs possessed interesting fluo-
rescence properties. They were nearly non-fluorescent in the so-
lution state regardless of their bridged or non-bridged structure at
the diarylmethylene moiety. This result would suggest that the
excited energy was consumed by molecular motions such as the
continuous photo-induced carbonecarbon double bond rotation at
the diarylmethylene moiety, similar to a light-driven molecular
motor, and the single-bond rotations of the arylealkene moieties.
In contrast, naked-eye visible fluorescence emission was detected
in the crystalline powder state and the frozen solution state of
many DAINs. This was probably because of a decrease in the energy
consumption owing to the restrained molecular motions. Thus, the
restraint of molecular motions is crucial to the emission of DAIN
analogues. In addition, regarding the solid-state fluorescence, the
crystal packing of the molecules was also important. The non-
fluorescent analogues 1 and 5d seemed to exhibit intermolecular
Fig. 11. X-ray-based ORTEP drawings of 1, 3a, 5d and 6. Ellipsoids are set at 50%
probability.26
p
e
p
interactions, which would cause fluorescence quenching by
electron and/or energy transfers, whereas the fluorescent ana-
logues 3a and 6 exhibited a stepwise arrangement where such
pep
interactions did not seem to exist. Comparison of the solid-state
fluorescence properties of DAIN 5a with those of the correspond-
ing BDIs 8 showed that the DAIN structure possessed superior
properties in terms of the higher quantum yield and the longer
emission wavelength. Although some properties of the DAIN ana-
logues differed from our initial expectations, they would be of great
use in a fluorescence off/on switching system where the restraint of
the molecular motion is applicable to the switch mechanism.
Studies on such applications are now in progress in our laboratory.
Fig. 12. Torsion angles in BDI structures. A: HO-BDI in GFP structure (PDBID: 1GFL); B:
Reported by Ramanathan et al.27
ꢀ
Although the molecules overlapped within 4.0e4.5 A in the lattices,
effective
pep interactions probably did not arise because of the
non-overlapping or non-parallel arrangement of the aromatic
rings. However, the difference in the spacing between the layers
4. Experimental section
4.1. General information
Table 6
Torsion angle (ꢀ) between the CePh bond (4) and the C]C bond (
s
) measured from
the ORTEP drawings
1H and 13C NMR spectra were recorded on a JEOL JNM-AL400
spectrometer (400 and 100 MHz, respectively); the chemical
shifts are reported in parts per million (ppm) relative to tetrame-
thylsilane or solvent signals (CDCl3: 7.26 ppm for 1H- and 77.0 ppm
for 13C NMR; DMSO-d6: 2.50 ppm for 1H- and 39.5 ppm for 13C
NMR; C6D6: 7.16 ppm for 1H NMR; acetone-d6: 2.05 ppm for 1H
NMR; CD3CN: 1.94 ppm for 1H NMR; CD2Cl2: 53.8 ppm for 13C
NMR). Mass spectra were recorded using a JEOL JMS-BU25 mass
spectrometer in positive FAB mode or using a JEOL JMS-T100GCV
mass spectrometer in positive FD mode. UVevis spectra were ob-
tained on a SHIMADZU MultiSpec-1500 spectrophotometer at
concentrations of 1.0ꢂ10ꢁ5 M in CH2Cl2. IR spectra were obtained
on a JASCO FT/IR-6300 spectrometer by diffuse reflection in KBr.
Melting points (Mp) were measured using a Yanaco melting-point
apparatus and are uncorrected. Microanalyses were performed by
DAINa
4
4 (E-Ph)
s(Z)
s(E)
b
b
c
d
(Z-Ph)
1
42.4
31.4
3.5
59.1
71.4
ꢁ6.1
63.9
5.7
3.4
9.6
1.5
3a
5d
6
ꢁ3.2
ꢁ2.5
40.8
7.7
8.2
a
Angles were measured for one of the enantiomeric structures in the unit cell.
Torsion of C(4)¼CeC(10)eC(20).
b
c
Torsion of N(3)eC(4)¼CeC(10).
d
Torsion of C(5)eC(4)¼CeC(10).
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
9
ꢀ
ꢀ
Fig. 13. X-ray crystal structures and space groups of fluorescent DAINs 3a and 6, and nearly non-fluorescent DAINs 1 and 5d. Lengths in A, angles in
.
A Rabbit Science Japan Co., Ltd. (Sagamihara, Japan). Fluorescence
spectra were recorded using a JASCO FP-6200 spectrofluorometer
equipped with an FPA-450 solid sample attachment to measure the
fluorescence of crystalline powder samples. Quantum yield mea-
surements of solid samples were carried out by JASCO Corporation
(Tokyo, Japan) or performed on a JASCO FP-8300 spectrometer. X-
ray diffraction measurements were performed on a Rigaku RAXIS-
RAPID diffractometer using CuKa radiation. Analytical thin-layer
chromatography was carried out on Merck silica gel 60 F254. Col-
umn chromatography was performed using a disposable silica gel
column RediSep (Teledyne Isco, Inc.) or Fuji Silysia PSQ 100B or Fuji
Silysia CHROMATOREX (NH-silica gel).
4.2. General procedure for DAIN synthesis
A
mixture of imidate or thioimidate 9 (0.10e1.0 mmol,
1.0 equiv), N-(diarylmethylene)glycinate 10 (3.0e5.0 equiv), and
acetic acid (2.0e4.0 equiv) in toluene (ca. 0.2 M concentration of
the imidate) was stirred under gentle reflux (or at room tempera-
ture in some cases) for several hours to days, as shown in Tables 2
and 3. After being cooled to room temperature, the mixture was
diluted with ethyl acetate and was washed with a saturated solu-
tion of NaHCO3, water, and 10% aqueous HCl. In the case of polar
products (e.g., 2, 4, 5a, 5b, 7), the acidic aqueous layer was re-
extracted with ethyl acetate after neutralization with a saturated
solution of NaHCO3. The (combined) organic layers were washed
with a saturated solution of NaHCO3 and brine, dried over Na2SO4,
and concentrated under reduced pressure. The residue was purified
by silica gel column chromatography to give DAIN as a coloured
solid. The product was recrystallised from dichloromethane/hex-
ane prior to evaluation of its properties.
Fig. 14. Intermolecular
Dotted lines represent the distances of the
p
e
p
interactions in nearly non-fluorescent DAINs 1 and 5d.
interactions.
pep
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

10
M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
ethyl acetate/CH2Cl2¼1:50. Yellow needles. Mp: 202e202.5 ꢀC. 1H
NMR (C6D6)
d
¼8.51 (d, J¼9 Hz, 1H), 7.86 (d, J¼8 Hz, 2H), 7.44 (d,
J¼8 Hz, 2H), 7.28e7.21 (m, 4H), 7.11e7.06 (m, 2H), 6.56 (dd, J¼9,
3 Hz, 1H), 6.51 (d, J¼3 Hz, 1H), 3.24 (s, 3H), 2.33 (t, J¼7 Hz, 2H), 2.14
(t, J¼7 Hz, 2H) ppm 13C NMR (CD2Cl2)
¼165.9, 160.4, 157.1, 146.6,
d
140.0, 138.4, 136.8, 132.4, 130.9, 129.5, 129.1, 128.3, 128.2, 128.0,
127.6, 118.7, 114.4, 112.5, 55.8, 26.5, 25.8 ppm. IR (KBr) nmax: 1637,
1708 cmꢁ1. UV (CH2Cl2) lmax nm (log ε): 254 (4.45), 320 (4.36), 390
(3.72). MS (FAB) m/z 381 (MþHþ). Anal. Calcd for C 78.93, H 5.30, N
7.36; found C 78.64, H 4.94, N 7.61.
4.2.5. 2-(Diphenylmethylene)-5,6-dihydroimidazo[2,1-a]isoquinolin-
3(2H)-one (4). Eluent for silica gel chromatography: ethyl acetate/
hexane¼1:10 to 1:4. Yellow needles. Mp: 163.5e164.5 ꢀC. 1H NMR
(CDCl3)
d
¼8.23 (d, J¼8 Hz, 1H), 7.73e7.70 (m, 2H), 7.51e7.34 (m,
11H), 7.30 (d, J¼8 Hz, 1H), 3.81 (t, J¼7 Hz, 2H), 3.10 (t, J¼7 Hz, 2H)
ppm 13C NMR (CD2Cl2)
d¼168.0, 156.2, 145.9, 140.2, 138.7, 138.5,
137.9,132.9,132.6,131.1,129.5,129.1,128.8,128.2,128.0,127.8,126.5,
125.4, 37.2, 28.2 ppm. IR (KBr) nmax: 1620, 1706 cmꢁ1. UV (CH2Cl2)
lmax nm (log ε): 268 (4.45), 310 (4.13), 397 (4.34). MS (FAB) m/z 351
(MþHþ). Anal. Calcd for C 82.26, H 5.18, N 7.99; found C 82.25, H
5.22, N 7.96.
4.2.6. 2-(Diphenylmethylene)-6,7,8,9-tetrahydro-2H-imidazo[1,2-a]
azepin-3(5H)-one (5a). Eluent for silica gel chromatography: ethyl
acetate/hexane¼1:3 to 1:2. Yellow needles. Mp: 194e195 ꢀC. 1H
Fig. 15. Head-to-tail configuration of fluorescent DAINs 3a and 6. Distances between
layers are shown with red arrows.
NMR (CDCl3)
d¼7.58e7.55 (m, 2H), 7.41e7.26 (m, 8H), 3.63e3.61 (br
m, 2H), 2.79e2.73 (br m, 2H), 1.84e1.74 (br m, 4H), 1.69e1.60 (br m,
4.2.1. 4-(Diphenylmethylene)-2-methyl-1-phenyl-1H-imidazol-
5(4H)-one (1). Eluent for silica gel chromatography: ethyl acetate/
hexane¼1:4 to 1:3. Yellow plates. Mp: 209e210 ꢀC. 1H NMR (CDCl3)
2H) ppm 13C NMR (CD2Cl2)
d
¼168.2, 166.5, 145.2, 140.2, 138.7, 137.5,
132.4,130.8,129.2,128.9,128.1,127.9, 40.8, 31.8, 31.1, 29.4, 26.1 ppm.
IR (KBr) nmax: 1623, 1700 cmꢁ1. UV (CH2Cl2) lmax nm (log ε): 265
(4.02), 287 (4.00), 362 (4.09). MS (FAB) m/z 317 (MþHþ). Anal. Calcd
for C 79.72, H 6.37, N 8.85; found C 79.72, H 6.44, N 8.64.
d
¼7.63e7.61 (m, 2H), 7.46e7.33 (m, 11H), 7.23e7.20 (m, 2H), 2.25 (s,
3H) ppm 13C NMR (CD2Cl2)
d
¼168.1, 160.3, 146.2, 140.1, 138.3, 136.4,
134.2, 132.5, 130.9, 129.7, 129.5, 129.1, 128.8, 128.2, 128.0, 127.8,
16.5 ppm. IR (KBr) nmax: 1620, 1715 cmꢁ1. UV (CH2Cl2) lmax nm (log
ε): 241 (4.31), 299 (4.15), 357 (4.21). MS (FAB) m/z 339 (MþHþ).
Anal. Calcd for C 81.63, H 5.36, N 8.28; found C 81.45, H 5.28, N 8.27.
4.2.7. 2-(Diphenylmethylene)-2,4,5,6-tetrahydro-1H-benzo[f]imi-
dazo[1,2-a]azepin-1-one (6). Eluent for silica gel chromatography:
ethyl acetate/hexane¼1:9 to 1:4. Yellow needles. Mp:
213.5e214 ꢀC. 1H NMR (CDCl3)
d
¼7.63e7.60 (m, 2H), 7.45 (d, J¼8 Hz,
4.2.2. 2-(Diphenylmethylene)-5,6,7,8-tetrahydroimidazo[1,2-a]pyr-
idin-3(2H)-one (2). Eluent for silica gel chromatography: ethyl ac-
etate/hexane¼1:4 to 1:2. Yellow needles. Mp: 226e227 ꢀC. 1H NMR
1H), 7.42e7.34 (m, 8H), 7.31e7.20 (m, 3H), 2.74 (t, J¼7 Hz, 2H), 2.69
(t, J¼7 Hz, 2H), 2.24e2.17 (m, 2H) ppm 13C NMR (CD2Cl2)
¼166.7,
d
162.9, 146.7, 140.1, 138.4, 136.7, 135.0, 134.0, 132.5, 130.9, 130.1,
129.5, 129.2, 128.2, 128.0, 127.8, 127.6, 124.8, 30.5, 27.3, 27.2 ppm. IR
(KBr) nmax: 1625,1712 cmꢁ1. UV (CH2Cl2) lmax nm (log ε): 244 (4.36),
306 (4.24), 367 (4.10). MS (FAB) m/z 365 (MþHþ). Anal. Calcd for C
82.39, H 5.53, N 7.69; found C 82.43, H 5.54, N 7.64.
(CDCl3)
d
¼7.56e7.54 (m, 2H), 7.41e7.29 (m, 8H), 3.49 (t, J¼6 Hz,
2H), 2.80 (t, J¼6 Hz, 2H), 1.91e1.83 (m, 4H) ppm 13C NMR (CD2Cl2)
d¼169.1, 161.5, 144.5, 140.2, 138.6, 136.9, 132.4, 130.9, 129.2, 128.9,
128.1, 127.9, 40.2, 26.8, 22.3, 20.4 ppm. IR (KBr) nmax: 1619,
1703 cmꢁ1. UV (CH2Cl2) lmax nm (log ε): 264 (3.96), 288 (3.90), 365
(4.00). MS (FAB) m/z 303 (MþHþ). Anal. Calcd for C 79.44, H 6.00, N
9.26; found C 79.45, H 6.17, N 9.34.
4.2.8. 4-(Diphenylmethylene)-1-methyl-2-phenyl-1H-imidazol-
5(4H)-one (7). Eluent for silica gel chromatography: ethyl acetate/
hexane¼1:9 to 1:4. Yellow needles. Mp: 208.5e209.5 ꢀC. 1H NMR
4.2.3. 2-(Diphenylmethylene)-4,5-dihydroimidazo[1,2-a]quinolin-
1(2H)-one (3a). Eluent for silica gel chromatography: ethyl acetate/
hexane¼1:4 to 1:3. Yellow needles. Mp: 210.5e211.5 ꢀC. 1H NMR
(CDCl3)
d
¼7.83e7.81 (m, 2H), 7.70e7.67 (m, 2H), 7.55e7.51 (m, 3H),
7.46e7.43 (m, 3H), 7.38e7.35 (m, 5H), 3.28 (s, 3H) ppm 13C NMR
(CD2Cl2)
¼169.7, 161.5, 147.0, 140.0, 138.6, 137.2, 132.9, 131.5, 131.0,
d
(CDCl3)
d
¼8.27 (d, J¼8 Hz, 1H), 7.58e7.55 (m, 2H), 7.47e7.34 (m,
130.0,129.6,129.2,129.1,128.9,128.2,128.0, 29.1 ppm. IR (KBr) nmax:
8H), 7.24e7.21 (m, 2H), 7.11 (td, J¼7, 1 Hz, 1H), 3.01 (s, 4H). 1H NMR
1616,1707 cmꢁ1. UV (CH2Cl2) lmax nm (log ε): 258 (4.34), 296 (4.04),
390 (4.17). MS (FAB) m/z 339 (MþHþ). Anal. Calcd for C 81.63, H
5.36, N 8.28; found C 81.85, H 5.43, N 8.13.
(C6D6)
d
¼8.55 (dd, J¼8, 1 Hz, 1H), 7.85e7.82 (m, 2H), 7.42e7.39 (m,
2H), 7.28e7.06 (m, 6H), 7.01 (td, J¼8, 2 Hz, 1H), 6.81 (td, J¼7, 1 Hz,
1H), 6.71 (dd, J¼8, 1 Hz, 1H), 2.31 (dd, J¼8, 6 Hz, 2H), 2.15 (dd, J¼8,
6 Hz, 2H) ppm 13C NMR (CD2Cl2)
d¼166.0, 160.3, 146.8, 140.0, 138.4,
4.2.9. 2-(Bis(4-methoxyphenyl)methylene)-6,7,8,9-tetrahydro-2H-
imidazo[1,2-a]azepin-3(5H)-one (5b). Eluent for silica gel chroma-
tography: ethyl acetate/hexane¼2:3 to 1:1. Yellow needles. Mp:
136.6, 134.1, 132.4, 130.8, 129.6, 129.2, 128.7, 128.3, 128.0, 127.9,
126.8, 125.4, 117.7, 26.3, 25.8 ppm. IR (KBr) nmax: 1634, 1710 cmꢁ1
.
UV (CH2Cl2) lmax nm (log ε): 249 (4.41), 313 (4.32), 370 (3.95). MS
(FAB) m/z 351 (MþHþ). Anal. Calcd for C 82.26, H 5.18, N 7.99; found
C 82.26, H 5.26, N 7.98.
209.5e210.5 ꢀC. 1H NMR (CDCl3)
d
¼7.53 (d, J¼9 Hz, 2H), 7.26 (d,
J¼9 Hz, 2H), 6.92 (d, J¼9 Hz, 2H), 6.87 (d, J¼9 Hz, 2H), 3.85 (s, 3H),
3.83 (s, 3H), 3.68e3.62 (br m, 2H), 2.80e2.75 (br m, 2H), 1.85e1.74
(br m, 4H), 1.71e1.62 (br m, 2H) ppm 13C NMR (CD2Cl2)
d
¼168.2,
164.8, 160.8, 160.7, 145.6, 135.9, 134.5, 132.9, 130.9, 113.4, 113.3, 55.7,
55.6, 40.7, 31.8, 31.1, 29.5, 26.2 ppm. IR (KBr) nmax: 1601, 1696 cmꢁ1
4.2.4. 2-(Diphenylmethylene)-7-methoxy-4,5-dihydroimidazo[1,2-a]
quinolin-1(2H)-one (3b). Eluent for silica gel chromatography:
.
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
11
UV (CH2Cl2) lmax nm (log ε): 318 (4.18), 384 (4.27). MS (FAB) m/z
377 (MþHþ). Anal. Calcd for C 73.38, H 6.43, N 7.44; found C 73.20,
H 6.46, N 7.41.
(KBr) nmax: 1699, 2107 cmꢁ1. HRMS (FAB) Calcd for C8H13N4O2
(MþHþ): 197.1038. Found: 197.1030.
4.3.2. 6,7,8,9-Tetrahydro-2H-imidazo[1,2-a]azepin-3(5H)-one (15).16
4.2.10. 2-(Bis(4-chlorophenyl)methylene)-6,7,8,9-tetrahydro-2H-
imidazo[1,2-a]azepin-3(5H)-one (5c). During the work-up, the
mixture was not washed with aqueous HCl. Eluent for silica gel
chromatography: ethyl acetate/hexane¼1:4 to 1:3. Pale yellow
4.3.2.1. Synthesis from 14. To a stirred solution of 14 (560 mg,
2.86 mmol) in toluene (14 mL), Ph3P (825 mg, 3.15 mmol) was
added at room temperature, and stirred at room temperature for
2.5 h. After evaporation of the solvent, the residue was purified by
silica gel column chromatography (ethyl acetate, then CHCl3/
MeOH¼10/1) to give 15 (390 mg, 90%) as a yellow oil.
needles. Mp: 187.5e188.5 ꢀC. 1H NMR (CDCl3)
d¼7.51 (d, J¼9 Hz,
2H), 7.38 (d, J¼9 Hz, 2H), 7.32 (d, J¼9 Hz, 2H), 7.23 (d, J¼9 Hz, 2H),
3.67e3.60 (br m, 2H), 2.80e2.75 (br m, 2H), 1.85e1.76 (br m, 4H),
1.69e1.66 (br m, 2H) ppm 13C NMR (CD2Cl2)
d
¼168.0, 167.4, 141.9,
4.3.2.2. Synthesis from 9. To a stirred solution of 9 (500 mg,
3.94 mmol) in MeOH (30 mL), triethylamine (1.10 mL,
7.88 mmol) and glycine methyl ester hydrochloride (993 mg,
7.91 mmol) were added at room temperature; the mixture was
stirred under reflux for 1.5 h. Subsequent to evaporation of the
solvent, the residue was dissolved in ether, and the precipitate
was removed by filtration. After concentration of the filtrate, the
residue was purified by silica gel column chromatography
(CHCl3/MeOH¼30/1) to give 15 (576 mg, 96%) as a yellow oil. 1H
138.1, 137.9, 136.6, 135.4, 135.0, 133.9, 132.4, 128.5, 128.2, 40.8, 31.8,
31.0, 29.3, 26.0 ppm. IR (KBr) nmax: 1619, 1705 cmꢁ1. UV (CH2Cl2)
lmax nm (log ε): 274 (4.25), 370 (4.21). MS (FAB) m/z 385 (MþHþ).
Anal. Calcd for C 65.46, H 4.71, N 7.27; found C 65.18, H 4.55, N 7.23.
4.2.11. 2-(9H-Fluoren-9-ylidene)-6,7,8,9-tetrahydro-2H-imidazo[1,2-
a]azepin-3(5H)-one (5d). Eluent for silica gel chromatography:
ethyl acetate/hexane¼1:20 to 1:8. Ruby red prisms. Mp:
192.5e193.5 ꢀC. 1H NMR (CDCl3)
d
¼9.45 (d, J¼8 Hz, 1H), 9.01 (d,
NMR (CDCl3)
d
¼4.10 (t, J¼1 Hz, 2H), 3.66e3.57 (m, 2H),
2.71e2.62 (m, 2H), 1.84e1.58 (m, 6H) ppm 13C NMR (CDCl3)
J¼8 Hz, 1H), 7.63e7.59 (m, 2H), 7.40e7.27 (m, 4H), 3.85e3.77 (br m,
2H), 2.92e2.86 (br m, 2H), 1.93e1.83 (br m, 4H), 1.80e1.71 (br m,
d
¼180.5, 168.4, 58.7, 40.4, 31.7, 30.5, 29.1, 25.4 ppm. IR (KBr)
2H). 1H NMR (acetone-d6)
d
¼9.52 (dd, J¼8, 1 Hz, 1H), 9.09 (dd, J¼8,
nmax: 1634, 1724 cmꢁ1. HRMS (FAB) Calcd for C8H13N2O (MþHþ):
1 Hz, 1H), 7.78 (d, J¼8 Hz, 1H), 7.75 (d, J¼8 Hz, 1H), 7.43 (td, J¼8,
1 Hz, 1H), 7.39 (td, J¼8, 1 Hz, 1H), 7.33 (td, J¼8, 1 Hz, 1H), 7.30 (td,
J¼8, 1 Hz, 1H), 3.84 (br t, J¼5 Hz, 2H), 2.96e2.92 (m, 2H), 1.94e1.85
153.1028. Found: 153.1040.
4.3.3. (Z)-2-Benzylidene-6,7,8,9-tetrahydro-2H-imidazo[1,2-a]aze-
(m, 4H), 1.80e1.74 (m, 2H) ppm 13C NMR (CD2Cl2)
d
¼169.0, 167.6,
pin-3(5H)-one (Z-8). To
6.58 mmol) in MeOH (50 mL), benzaldehyde (734
and piperidine (378 L, 3.83 mmol) were added at room tem-
a
stirred solution of 15 (1.00 g,
142.3, 142.2, 139.8, 139.2, 138.4, 136.6, 131.3, 130.9, 130.4, 129.3,
128.3, 127.9, 119.8, 119.7, 41.1, 32.2, 31.1, 29.4, 26.1 ppm. IR (KBr)
nmax: 1601,1626,1696 cmꢁ1. UV (CH2Cl2) lmax nm (log ε): 242 (4.56),
267 (4.59), 275 (4.64), 333 (4.01)., 406 (4.39). MS (FAB) m/z 315
(MþHþ). Anal. Calcd for C 80.23, H 5.77, N 8.91; found C 80.10, H
5.70, N 8.85.
mL, 7.24 mmol)
m
perature, and the mixture was stirred at room temperature
overnight. Subsequent to evaporation of the solvent, the residue
was dissolved in ethyl acetate, and the organic layer was washed
with water and brine, then dried over Na2SO4, and concentrated
under reduced pressure. The residue was purified by silica gel
column chromatography (ethyl acetate/hexane¼1/4) to give Z-8
(1.11 g, 70%) as a pale yellow solid. Pale yellow needles. Mp:
4.2.12. 2-(9H-Xanthen-9-ylidene)-6,7,8,9-tetrahydro-2H-imidazo
[1,2-a]azepin-3(5H)-one (5e). Eluent for silica gel chromatography:
ethyl acetate/hexane¼1:4. Orange powder. Mp: 193.5e194 ꢀC. 1H
119e120 ꢀC (CH2Cl2/hexane). 1H NMR (CD3CN)
d
¼8.20e8.17 (m,
NMR (CDCl3)
d
¼8.98 (dd, J¼8, 1 Hz, 1H), 8.43 (dd, J¼8, 1 Hz, 1H),
2H), 7.46e7.37 (m, 3H), 6.96 (s, 1H), 3.66 (t, J¼5 Hz, 2H),
2.80e2.76 (m, 2H), 1.85e1.74 (m, 4H), 1.69e1.63 (m, 2H) ppm 13C
7.48e7.39 (m, 2H), 7.29e7.20 (m, 4H), 3.71e3.68 (br m, 2H),
2.81e2.78 (br m, 2H), 1.85e1.77 (br m, 4H), 1.71e1.64 (br m, 2H)
NMR (CD2Cl2)
d
¼170.2, 168.6, 140.0, 134.9, 132.4, 130.2, 128.9,
ppm 13C NMR (CD2Cl2)
d¼167.8, 164.5, 153.7, 153.4, 133.5, 131.9,
126.6, 41.0, 32.0, 31.0, 29.4, 26.0 ppm. IR (KBr) nmax: 1643,
1711 cmꢁ1. UV (CH2Cl2) lmax nm (log ε): 293 (4.06), 358(4.32). MS
(FAB) m/z 241 (MþHþ). Anal. Calcd for C 74.97, H 6.71, N 11.66;
found C 74.90, H 6.75, N 11.62.
131.8, 131.4, 131.2, 130.7, 123.3, 123.2, 122.1, 119.7, 116.5, 116.4, 41.0,
31.9, 31.1, 29.4, 26.2 ppm. IR (KBr) nmax: 1605, 1688 cmꢁ1. UV
(CH2Cl2) lmax nm (log ε): 295 (4.15), 412 (4.29). MS (FAB) m/z 331
(MþHþ). Anal. Calcd for C 76.34, H 5.49, N 8.48; found C 76.04, H
5.45, N 8.47.
4.3.4. (E)-2-Benzylidene-6,7,8,9-tetrahydro-2H-imidazo[1,2-a]aze-
pin-3(5H)-one (E-8). A solution of Z-8 (400 mg, 1.67 mmol) in
benzene (16 mL) was left under a UV light (365 nm, 16 W) at room
temperature for 20 h. After evaporation of the solvent, the residue
was purified by silica gel column chromatography (ethyl acetate/
hexane¼1/2 to 1/1) in a darkroom to give E-8 (171 mg, 43%) and
recovered Z-8 (168 mg, 42%). Yellow needles. Mp: 125e126 ꢀC.
4.3. Synthesis of BDIs 8
4.3.1. 1-(2-Azidoacetyl)azepan-2-one (14). To a stirred solution of
12 (2.00 g, 17.7 mmol) in CH2Cl2 (44 mL), triethylamine (2.71 mL,
19.5 mmol) and 2-chloroacetyl chloride (1.55 mL, 19.5 mmol) were
added at room temperature, and stirred at room temperature for
2 h. After dilution of the mixture with ethyl acetate, the organic
layer was washed with water, and brine, then dried over Na2SO4,
and concentrated under reduced pressure. The residue was sub-
jected to the next reaction without purification. To a stirred solution
of crude 13 (ca. 2.2 g) in DMSO (44 mL), NaN3 (2.30 g, 35.4 mmol)
was added at room temperature, and stirred overnight at room
temperature. After dilution of the mixture with ether, the organic
layer was washed with water and brine, dried over MgSO4, and
concentrated under reduced pressure to afford 14 (1.13 g, 33% from
(CH2Cl2/hexane). 1H NMR (CD3CN)
d
¼8.27e8.23 (m, 2H), 7.46e7.40
(m, 3H), 7.21 (s, 1H), 3.67 (t, J¼4 Hz, 2H), 2.71e2.66 (m, 2H),
1.84e1.70 (m, 4H), 1.69e1.62 (m, 2H) ppm 13C NMR (CD2Cl2)
d
¼167.1, 165.3, 141.0, 136.4, 134.1, 131.8, 130.6, 128.6, 40.9, 31.7, 31.0,
29.4, 26.1 ppm. IR (KBr) nmax: 1635, 1695 cmꢁ1. UV (CH2Cl2) lmax nm
(log ε): 289 (4.09), 359 (4.13). HRMS (FAB) Calcd for C15H17N2O
(MþHþ): 241.1341. Found: 241.1339.
Acknowledgements
12) as an almost pure compound. 1H NMR (CDCl3)
3.98e3.95 (m, 2H), 2.74e2.71 (m, 2H), 1.82e1.70 (m, 6H) ppm 13C
NMR (CDCl3)
¼177.8, 171.1, 55.9, 43.6, 39.2, 29.0, 28.3, 23.5 ppm. IR
d¼4.43 (s, 2H),
This work was supported by JSPS KAKENHI Grant Number
24790029.
d
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073

12
M. Ikejiri et al. / Tetrahedron xxx (2015) 1e12
8. Tolbert’s group has reported a series of GFP-chromophore analogues: (a) Tol-
Supplementary data
bert, L. M.; Baldridge, A.; Kowalik, J.; Solntsev, K. M. Acc. Chem. Res. 2012, 45,
171e181 and references therein; (b) Baranov, M. S.; Lukyanov, K. A.; Borissova,
A. O.; Shamir, J.; Kosenkov, D.; Slipchenko, L. V.; Tolbert, L. M.; Yampolsky, I. V.;
Solntsev, K. M. J. Am. Chem. Soc. 2012, 134, 6025e6032; (c) Fery-Forgues, S.;
Veesler, S.; Fellows, W. B.; Tolbert, L. M.; Solntsev, K. M. Langmuir 2013, 29,
14718e14727; (d) Samanta, S. R.; Silva, J. P. D.; Baldridge, A.; Tolbert, L. M. Org.
Lett. 2014, 16, 3304e3307.
Supplementary data (Experimental procedures for the synthe-
ses of 9 and 10, copies of the 1H NMR and 13C NMR spectra for all
new compounds, excitationeemission contour plots of crystalline
powders 4, 5a, 5b and 6 and supplementary figures for the mo-
lecular packing of the X-ray crystal structures. Supplementary data
associated with this article can be found in the online version.)
related to this article can be found at http://dx.doi.org/10.1016/
j.tet.2015.05.073.
9. Dong, J.; Solntsev, K. M.; Tolbert, L. M. J. Am. Chem. Soc. 2009, 131, 662e670.
10. Wu, L.; Burgess, K. J. Am. Chem. Soc. 2008, 130, 4089e4096.
11. A part of this work appeared as a preliminary letter Ikejiri, M.; Tsuchino, M.;
Chihara, Y.; Yamaguchi, T.; Imanishi, T.; Obika, S.; Miyashita, K. Org. Lett. 2012,
14, 4406e4409.
12. Baldridge, A.; Kowalik, J.; Tolbert, L. M. Synthesis 2010, 2424e2436.
13. (a) Gong, X.; Yang, H.; Liu, H.; Jiang, Y.; Zhao, Y.; Fu, H. Org. Lett. 2010, 12,
3128e3131; (b) Lee, C.-Y.; Chen, Y.-C.; Lin, H.-C.; Jhong, Y.; Chang, C.-W.; Tsai, C.-
H.; Kao, C.-L.; Chien, T.-C. Tetrahedron 2012, 68, 5898e5907; (c) Baranov, M. S.;
Solntsev, K. M.; Lukyanov, K. A.; Yampolsky, I. V. Chem. Commun. 2013,
5778e5780; (d) Review Ivashkin, P. E.; Yampolsky, I. V.; Lukyanov, K. A. Russ. J.
Bioorg. Chem. 2009, 35, 652e669.
14. The syntheses of glycinates 10eeh were carried out by a partly modified
O’Donnell method with corresponding imines and glycinate hydrochloride
(Table S1 in supplementary data) O’Donnell, M. J.; Polt, R. L. J. Org. Chem. 1982,
47, 2663e2666.
15. Narayana Rao, M.; Holkar, A. G.; Ayyangar, N. R. Tetrahedron Lett. 1989, 30,
4717e4720.
16. Campagna, F.; Carotti, A.; Casini, G. J. Heterocycl. Chem. 1990, 27, 1973e1977.
17. Several BDI analogues are known to exhibit fluorescence in the solid state: see
Ref. 7l,o,r,t and 9.
18. Second emissions of other DAINs were not so prominent. Contour plots for
other typical DAINs are given as Supplementary data.
19. Review Shimizu, M.; Hiyama, T. Chem.dAsian J. 2010, 5, 1516e1531.
20. A similar dual emission was reported for a certain type of BDI analogue: see
Ref. 9.
21. Niwa, H.; Inouye, S.; Hirano, T.; Matsuno, T.; Kojima, S.; Kubota, M.; Ohashi, M.;
Tsuji, F. I. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 13617e13622 Ref. 7a.
22. Prescott, M.; Ling, M.; Beddoe, T.; Oakley, A. J.; Dove, S.; Hoegh-Guldberg, O.;
Devenish, R. J.; Rossjohn, J. Structure 2003, 11, 275e284.
23. Review Feringa, B. L. J. Org. Chem. 2007, 72, 6635e6652. Since fluorenylidene
and thioxanthenylidene moieties are typical components of the light-driven
molecular motors, we consider that the double bond isomerization is a cru-
cial factor for the fluorescence quenching of DAINs, especially for the 3rd-
generation.
References and notes
1. Review Zimmer, M. Chem. Rev. 2002, 102, 759e781.
€
2. (a) Ormo, M.; Cubitt, A. B.; Kallio, K.; Gross, L. A.; Tsien, R. Y.; Remington, S. J.
Science 1996, 273, 1392e1395; (b) Yang, F.; Moss, L. G.; Phillips, G. N., Jr. Nat.
Biotechnol. 1996, 14, 1246e1251.
3. (a) Voliani, V.; Bizzarri, R.; Nifosì, R.; Abbruzzetti, S.; Grandi, E.; Viappiani, C.;
Beltram, F. J. Phys. Chem. B 2008, 112, 10714e10722; (b) Yang, J.-S.; Huang, G.-J.;
Liu, Y.-H.; Peng, S.-M. Chem. Commun. 2008, 1344e1346.
4. A similar Z-E isomerization has been reported for benzylidene-oxazolone an-
alogues: (a) Blanco-Lomas, M.; Campos, P. J.; Sampedro, D. Org. Lett. 2012, 14,
4334e4337; (b) Blanco-Lomas, M.; Funes-Ardoiz, I.; Campos, P. J.; Sampedro, D.
Eur. J. Org. Chem. 2013, 6611e6618; (c) Funes-Ardoiz, I.; Blanco-Lomas, M.;
Campos, P. J.; Sampedro, D. Tetrahedron 2013, 69, 9766e9771.
5. Dong, J.; Abulwerdi, F.; Baldridge, A.; Kowalik, J.; Solntsev, K. M.; Tolbert, L. M. J.
Am. Chem. Soc. 2008, 130, 14096e14098.
6. Some exceptions to such a ‘dark’ E-form are known: for example Petersen, J.;
Wilmann, P. G.; Beddoe, T.; Oakley, A. J.; Devenish, R. J.; Prescott, M.; Rossjohn, J.
J. Biol. Chem. 2003, 278, 44626e44631.
7. Examples of fluorescent GFP-chromophore analogues: (a) Kojima, S.; Ohkawa,
H.; Hirano, T.; Maki, S.; Niwa, H.; Ohashi, M.; Inouye, S.; Tsuji, F. I. Tetrahedron
Lett. 1998, 39, 5239e5242; (b) He, X.; Bell, A. F.; Tonge, P. J. Org. Lett. 2002, 4,
1523e1526; (c) Bourotte, M.; Schmitt, M.; Follenius-Wund, A.; Pigault, C.;
Haiech, J.; Bourguignon, J.-J. Tetrahedron Lett. 2004, 45, 6343e6348; (d) Chen,
K.-Y.; Cheng, Y.-M.; Lai, C.-H.; Hsu, C.-C.; Ho, M.-L.; Lee, G.-H.; Chou, P.-T. J. Am.
Chem. Soc. 2007, 129, 4534e4535; (e) Stafforst, T.; Diederichsen, U. Eur. J. Org.
Chem. 2007, 899e911; (f) Huang, G.-J.; Yang, J.-S. Chem.dAsian J. 2010, 5,
2075e2085; (g) Kang, J.; Zhao, G.; Xu, J.; Yang, W. Chem. Commun. 2010,
2868e2870; (h) Li, Y.; Shi, L.; Qin, L.-X.; Qu, L.-L.; Jing, C.; Lan, M.; James, T. D.;
Long, Y.-T. Chem. Commun. 2011, 4361e4363; (i) Katritzky, A. R.; Yoshioka-
Tarver, M.; El-Gendy, B. E.-D. M.; Hall, C. D. Tetrahedron Lett. 2011, 52,
2224e2227; (j) Paige, J. S.; Wu, K. Y.; Jaffrey, S. R. Science 2011, 333, 642e646;
(k) Wenge, U.; Wagenknecht, H.-A. Synthesis 2011, 502e508; (l) Chuang, W.-T.;
Hsieh, C.-C.; Lai, C.-H.; Lai, C.-H.; Shih, C.-W.; Chen, K.-Y.; Hung, W.-Y.; Hsu, Y.-
H.; Chou, P.-T. J. Org. Chem. 2011, 76, 8189e8202; (m) Kang, J.; Fang, X.; Chen, X.;
Zhao, G.; Ren, A.; Xu, J.; Yang, W. Eur. J. Inorg. Chem. 2011, 5322e5327; (n)
Cacciarini, M.; Nielsen, M. B.; de Castro, E. M.; Marinescu, L.; Bols, M. Tetrehe-
dron Lett. 2012, 53, 973e976; (o) Huang, G.; Zhang, G.; Wu, Y.; Liao, Q.; Fu, H.;
Zhang, D. Asian J. Org. Chem. 2012, 1, 352e358; (p) Shi, L.; Li, Y.; Liu, Z.-P.; James,
T. D.; Long, Y.-T. Talanta 2012, 100, 401e404; (q) Huang, G.-J.; Ho, J.-H.; Prab-
hakar, C.; Liu, Y.-H.; Peng, S.-M.; Yang, J.-S. Org. Lett. 2012, 14, 5034e5037; (r)
Tou, S.-L.; Huang, G.-J.; Chen, P.-C.; Chang, H.-T.; Tsai, J.-Y.; Yang, J.-S. Chem.
Commun. 2014, 620e622; (s) Baranov, M. S.; Solntsev, K. M.; Baleeva, N. S.;
Mishin, A. S.; Lukyanov, S. A.; Lukyanov, K. A.; Yampolsky, I. V. Chem.dEur. J.
2014, 20, 13234e13241; (t) very recent report Huang, G.-J.; Lin, C.-J.; Liu, Y.-H.;
Peng, S.-M.; Yang, J.-S. Photochem. Photobiol. 2015, 91, 714e722.
24. Grabowski, Z. R.; Rotkiewicz, K.; Rettig, W. Chem. Rev. 2003, 103, 3899e4031.
25. Review Hong, Y.; Lam, J. W. Y.; Tang, B. Z. Chem. Soc. Rev. 2011, 40, 5361e5388.
26. CCDC-1034530e1034533 contain the supplementary crystallographic data for
this paper. 1034530(1), 1034531 (3a), 1034532 (5d), 1034533 (6). These data
can be obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
27. Bhattacharjya, G.; Savitha, G.; Ramanathan, G. J. Mol. Struct. 2005, 752, 98e103.
28. A certain type of BDI analogue has been reported to accompany a photoreac-
tion such as double bond isomerization and [2þ2] photo-dimerisation in the
solid state Naumov, P.; Kowalik, J.; Solntsev, K. M.; Baldridge, A.; Moon, J.-S.;
Kranz, C.; Tolbert, L. M. J. Am. Chem. Soc. 2010, 132, 5845e5857; for a oxazolone
ꢁ
analogue: Runcevski, T.; Blanco-Lomas, M.; Marazzi, M.; Cejuela, M.; Sampedro,
D.; Dinnebier, R. E. Angew. Chem. Int. Ed. 2014, 53, 6738e6742.
29. A similar quenching caused by intermolecular
p-overlap within dimers was
reported for some BDI analogues containing a large aromatic ring: see Ref. 7o.
30. A similar relationship between spacing and fluorescence wavelength was re-
ported for a bis(phenylethynyl)benzene derivative Thomas, R.; Varghese, S.;
Kulkarni, G. U. J. Mater. Chem. 2009, 19, 4401e4406.
Please cite this article in press as: Ikejiri, M.; et al., Tetrahedron (2015), http://dx.doi.org/10.1016/j.tet.2015.05.073