Conformationally restricted dipyrromethene boron difluoride (BODIPY) dyes: Highly fluorescent, multicolored probes for cellular imaging

Article abstract of DOI:10.1002/chem.200800309

Novel fluorescent, conformationally restricted dipyrromethene boron difluoride (BODIPY) dyes have been prepared by introducing a naphthalenyl group at the meso position of the BODIPY core. These BODIPY dyes exhibit increased fluorescence quantum yields co

Full text of DOI:10.1002/chem.200800309

DOI: 10.1002/chem.200800309
Conformationally Restricted Dipyrromethene Boron Difluoride (BODIPY)
Dyes: Highly Fluorescent, Multicolored Probes for Cellular Imaging
Qingdong Zheng,* Gaixia Xu, and Paras N. Prasad*[a]
Abstract: Novel fluorescent, conforma-
tionally restricted dipyrromethene
boron difluoride (BODIPY) dyes have
been prepared by introducing a naph-
thalenyl group at the meso position of
the BODIPY core. These BODIPY
dyes exhibit increased fluorescence
quantum yields compared with dyes
that have a meso-position phenyl group
with internal rotation. The absorption
and emission wavelengths of such con-
formationally restricted BODIPY dyes
can be easily tuned to the near-IR
range by derivatization through a con-
densation reaction with benzaldehyde
derivatives. The two-photon absorption
properties of these BODIPY dyes were
also investigated and the results show
that they exhibit increased two-photon
excited fluorescence compared to ana-
logue dyes that contain a phenyl group.
The one- and two-photon fluorescence
imaging of living cells by using selected
BODIPY dyes has been successfully
demonstrated.
Keywords: dyes/pigments · fluores-
cent probes · imaging agents · lumi-
nescence · two-photon absorption
Introduction
easily introduced.[8,9] Although arylation at the BODIPY
meso position does not have much effect on the linear ab-
sorption and emission wavelengths, different sterical config-
urations between the aryl group at the meso position and
the BODIPY core may greatly affect other photophysical
properties of these dyes. For example, the fluorescence
quantum yield of meso-phenyl compound PY1 is much less
than that of its more substituted analogue PY2
(Scheme 1).[10,11] At the same time, compound PY2 shows
hypsochromically shifted absorption and emission compared
with PY1. Such differences are widely attributed to b sub-
stituents that prevent free rotation of the phenyl group and,
therefore, reduce the loss of energy from excited states via
nonirradiative molecular relaxation. The two methyl groups
at the b positions also lead to an increase in the dihedral
Fluorescence microscopy is one of the most versatile imag-
ing techniques in biomedical research, and allows the nonin-
vasive imaging of cells and tissues with molecule specificity.
Consequently, fluorescent probes that have high fluores-
cence quantum yields and that emit light of various colors
are in great demand because of the key role they play in
fluorescence microscopy.[1–4] So far, cyanine dyes have at-
tracted much attention in fluorescence imaging, owing to
the easy tunability of their absorption and emission wave-
lengths up to several hundred nm.[5] However, these cyanine
chromophores often have low fluorescence quantum yields
(less than 10–15%), which limits their applications. In con-
trast, dipyrromethene boron difluoride (BODIPY) dyes gen-
erally have high fluorescence quantum yields as well as in-
tense absorption.[6,7] To label specific biological targets, func-
tional groups can be introduced to the BODIPY core, while
retaining their inherent properties. The meso position of
BODIPY dyes has often been used as a connection into
which aryl substitutes with various functionalities can be
[a] Dr. Q. Zheng, Dr. G. Xu, Prof. Dr. P. N. Prasad
Department of Chemistry
Institute for Lasers, Photonics, and Biophotonics
State University of New York at Buffalo
Buffalo, NY 14260 USA
Fax : (+1)716-645-6945
E-mail: qzheng7@jhu.edu
Scheme 1. Chemical structures and photophysical properties of PY1 and
PY2. f is the fluorescence quantum yield, lmax is the linear absorption
spectrum maximum, and Emax is the linear emission spectrum maximum.
pnprasad@buffalo.edu
5812
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 5812 – 5819

FULL PAPER
angle between the BODIPY core and the phenyl group,
which results in blue-shifted absorption and emission spec-
tra.[12] These results suggest that the introduction of a more
bulky aryl group, such as a naphthalenyl group, may further
increase the fluorescence quantum yields for these kinds of
dyes because free rotation of the naphthalenyl group would
be more difficult.
With this consideration in mind, we report herein the
design and synthesis of two conformationally restricted
BODIPY dyes (2a,b) through the introduction of a naph-
thalenyl group at the meso position. For comparison purpos-
es, we also prepared two phenyl-substituted analogues
(1a,b). In addition, an attempt was made to tune the emis-
sion wavelengths of these conformationally restricted
BODIPY dyes to the near-IR range by derivatization
through condensation between the methyl groups at the a
positions and a substituted benzaldehyde. The linear and
nonlinear photophysical properties of these BODIPY dyes
were also investigated. Finally, the one- and two-photon cel-
lular imaging of cancer cells was demonstrated with the aid
of selected BODIPY dyes.
Scheme 2. Synthesis of BODIPY dyes 1a,b and 2a,b. i) TFA, CH2Cl2,
12 h, RT; ii) DDQ; iii) NEt3, BF3·OEt2.
1
compounds were confirmed by H and 13C NMR spectrosco-
py, HRMS, and elemental analysis.
Crystal structures of 1a and 2a: Crystals suitable for X-ray
analysis were grown by slow evaporation of n-hexane into a
solution of 1a or 2a in dichloromethane. Cell parameters
and refinement details are summarized in Table 1. Note that
there is only one crystallographically independent molecule
in 1a, whereas there are two crystallographically independ-
ent molecules in 2a. Figure 1 shows that the angles of N1-
B1-N2 and F1-B1-F2 in both compounds indicate a tetrahe-
dral BF2N2 configuration, and are within the range of the
published data.[13,14] The average bond length of N1 C1 indi-
À
cates double-bond character, compared with the single-bond
À
character of N1 C4. For both compounds, there is strong p-
electron delocalization between the central six-membered
ring and the two adjacent pyrrole rings. However, almost no
p-electron delocalization is observed between the aryl group
at the meso position and the BODIPY core (indacene
À
plane), as indicated by the average length of the C5 C14
À
bond, which is almost the same length as a single C C
bond. Figure 2 shows how the structures of 1a and 2a are
À
stabilized by p–p and C H···p interactions.
It can be seen from the crystal structures that the inda-
cene planes of 1a and 2a are essentially planar. For both
compounds, the deviation of the pyrrole planes with respect
to each other and the central ring is typically only a few de-
grees. However, the indacene plane is almost perpendicular
to the aryl group at the meso position. On examining the di-
hedral angle between the phenyl/naphthalenyl group and in-
Results and Discussion
Synthesis: As depicted in Scheme 2, compounds 1 and 2
were synthesized in yields of 35 to 45% by reacting alde-
hydes with 3-ethyl-2,4-dimethylpyrrole (or 2,4-dimethylpyr-
role) in the presence of trifluoroacetic acid (TFA) and 2,3-
dichloro-5,6-dicyano-1,4-benzo-
quinone (DDQ), followed by
the addition of BF3·OEt2. As
shown in Scheme 3, compounds
3 and 4 were prepared by a
Knoevenagel-type condensation
reaction between 2b and 4-di-
methylaminobenzaldehyde
in
the presence of acetic acid, pi-
peridine, and molecular sieves
in toluene. The products result-
ing from the single and double
condensation reactions could be
separated by using column
chromatography with dichloro-
Scheme 3. Synthesis of monostyryl- and distyryl-BODIPY dyes 3 and 4. i) Piperidine, AcOH, toluene, and mo-
methane as the eluent. All new lecular sieves, reflux, 24h.
Chem. Eur. J. 2008, 14, 5812 – 5819
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5813

Q. Zheng, P. N. Prasad, and G. Xu
Table 1. Crystallographic data and structure refinement for 1a and 2a.[a]
1a
2a
formula
Mr
cell system, space group
C22H23BF2N2O3
412.23
triclinic, P1
C52H50B2F4N4O6
924.58
monoclinic, P21/c
90(2)
14.8040(6)
21.718(1)
16.782(1)
90.00
123.061(3)
90.00
4522.0(3)
4
1.358
0.10
0.056
¯
T [K]
a [Š]
b [Š]
c [Š]
a [8]
b [8]
g [8]
V [Š3]
Z
90(2)
10.5126(9)
10.628(1)
10.762(1)
102.117(2)
101.129(1)
116.535(2)
993.7(2)
2
1calcd [MgmÀ3
]
1.378
0.10
0.042
m [mmÀ1
Rint
]
Figure 2. Molecular packing in the crystal structures of 1a (left) and 2a
(right).
qmax [8]
29.6
0.037
0.1040.192
1.02
26.0
0.067
R [F2 >2s
G
wR
S
A
1.02
no. of reflns
no. of params
N
5597
305
0.002
0.41, À0.22
8896
613
<0.0001
0.90, À0.90
the dihedral angle between the aryl group and indacene
À
plane, and a slight decrease in the C5 C14bond length be-
À3
tween the aryl group and indacene plane, from 1.496(1) Š in
]
1a to 1.489(4) Š in 2a.
[a] Criterion for observed reflections: I>2s(I). Refinement on F2.
Linear absorption and emission spectra: Linear absorption
spectra for compounds 1–4 in toluene were recorded on a
Shimadzu UV-3101 PC spectrophotometer. One-photon ex-
cited fluorescence spectra were measured in dilute solutions
by using a Jobin–Yvon FluoroLog-3 (model FL-3-11) spec-
trofluorometer and 10 mm path-length cuvettes. The fluores-
cence lifetimes for these compounds were measured by
using a high-speed streak camera with a resolution of 20 ps.
Table 2 summarizes the linear and two-photon absorption
(TPA) properties for 1a,b and 2a,b in toluene and shows
the linear absorption maxima, molar extinction coefficients,
emission maxima, Stokes shifts, fluorescence quantum
yields, fluorescence lifetimes, TPA peak, and TPA cross-sec-
tion peak values. Figure 3 depicts the linear absorption and
emission spectra for 1a,b and 2a,b in toluene. As shown in
the figures, the linear absorption and emission spectra for
dilute 1a,b and 2a,b in toluene are quite similar, except that
2a and 2b exhibit 3 nm red-shifted linear absorption and
Figure 1. Perspective views of 1a (left) and 2a (right), showing 50%
probability displacement ellipsoids. Selected bond lengths [Š] and angles
Table 2. Photophysical properties of compounds 1–4 in toluene.
À
À
À
À
[8] for 1a: F B1 1.402(2), F2 B1 1.393(1), N1 C1 1.346(1), N1 C4
[d]
[e]
emax ”105
[mÀ1 cmÀ1
lmax
Emax
Dn[a]
f[b]
t[c]
l2
dTPA
max
À
À
À
À
1.405(1), N1 B1 1.545(1), C1 C2 1.406(1), C2 C3 1.383(1), C3 C4
A
]
[nm] [nm]
A
]
[%] [ns] [nm]
[GM]
À
À
1.431(1), C4 C5 1.396(1), C5 C141.496(1), F2-B1-F1 108.67(8), F2-B1-
N2 110.49(9), F1-B1-N2 110.39(9), F2-B1-N1 110.65(9), F1-B1-N1
110.10(9), N2-B1-N1 106.52(8); selected bond lengths [Š] and angles [8]
1a 0.97
2a 1.00
1b 0.94527 52554.1
504517 98.94
507
60
493.1
5.1
519.3
828.1
707.7 0 4 2.8
3.7
98
950
5.8
109
7.0
82
950
520
78
545
112
128
990
À
À
À
À
for 2a: F1 B1 1.398(3), F2 B1 1.386(3), N1 C1 1.349(3), N1 C4
2b 0.95
530
624658
714752
99
643.8
990
–
–
À
À
À
À
1.403(4), N1 B1 1.549(4), C1 C2 1.405(4), C2 C3 1.379(4), C3 C4
3
4
1.03
1.05
–
–
À
À
1.432(4), C4 C5 1.398(4), C5 C14 1.489(4), F2-B1-F1 109.0(2), F2-B1-N2
110.9(2), F1-B1-N2 110.0(2), F2-B1-N1 111.1(2), F1-B1-N1 109.3(2), N2-
B1-N1 106.6(2).
[a] Stokes shift. [b] Fluorescence quantum yield. Rhodamine 6G (94% in
methanol) was used as a reference for 1a,b and 2a,b,[15] Cresyl Violet
(54% in methanol) was used as a reference for 3,[16] and Nile Blue (26%
in ethylene glycol) was used as a reference for 4.[17] [c] Fluorescence life-
time, recorded at a resolution of 20 ps. [d] TPA peak wavelength. [e] TPA
cross-section peak value, 1 GM=10À50 cm4 s per photon, experimental un-
certainty Æ15%.
dacene plane, we find a dihedral angle of 82.4(1)8 in 1a,
whereas in 2a this angle is 78.0(3) or 79.2 (3)8. The introduc-
tion of a naphthalenyl group leads to a small reduction in
5814
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 5812 – 5819

Dipyrromethene Boron Difluoride Dyes
FULL PAPER
low radiative probability,[12] and is energetically less accessi-
ble in a BODIPY dye that has a more conformationally re-
stricted group in the meso position. Obviously, the naphtha-
lenyl group is more sterically hindered than the phenyl
group, which is the main reason that 2a has a fluorescence
quantum yield that is 63% greater than 1a. This explanation
is also valid in the case of 1b and 2b. The quantum yield of
2b in toluene is close to 100%, which is 27% greater than
that of 2a. To explore the fluorescence dynamics of these
chromophores, their fluorescence decay profiles in toluene
were collected. The fluorescence decay traces of these chro-
mophores could all be fitted by a single-exponential curve,
and the fitting results are listed in Table 2. The naphthalen-
yl-containing 2a and 2b have longer fluorescence lifetimes
compared with their phenyl analogues 1a and 1b. For exam-
ple, compound 2a has a fluorescence lifetime of 5.8 ns,
whereas compound 1a has a fluorescence lifetime of 3.7 ns.
Compounds 1 and 2 have a small Stokes shift in the range
of 493 to 525 cmÀ1. The monostyryl compound 3 and the dis-
tyryl compound 4 are both derived from 2b. As shown in
Figure 4, compound 3 has a linear absorption maximum at
Figure 3. a) Linear absorption (10À5 m) and b) emission (10À6 m) spectra of
1a,b and 2a,b in toluene.
emission spectra. By examining the geometric arrangements
of 1a and 2a, one can see that the phenyl/naphthalenyl
group and the indacene plane are almost perpendicular to
each other, and that the dihedral angle between the phenyl
group and indacene plane (1a) is around 3 or 48 greater
than the angle between the naphthalenyl group and inda-
cene plane (2a), which may contribute to the bathochromi-
cally shifted absorption and emission spectra observed for
chromophores that contain a naphthalenyl group. Further-
more, the introduction of a 5-membered heteroaromatic
ring, such as a furan ring, at the meso position of the
BODIPY core leads to a decrease in the dihedral angle and
results in red-shifted linear absorption and emission spectra;
these results will be published elsewhere. However, com-
pounds 2a and 2b have higher fluorescence quantum yields
(98 and 99% for 2a and 2b, respectively) compared with 1a
and 1b (60 and 78% for 1a and 1b, respectively). The only
difference between 1a and 2a is the aryl group at the meso
position. From the crystal structures we know that these aryl
groups are actually perpendicular to the indacene plane. It
has been shown that the excited state of an aryl-substituted
BODIPY dye has a second, lower-energy minimum in which
the nonhindered aryl group rotates closer to the mean plane
of the BODIPY core, which itself undergoes some distor-
tion. This relaxed, distorted, excited-state conformation has
Figure 4. Linear absorption (c; 10À5 m) and emission (b; 10À6 m)
spectra of 3 and 4 in toluene.
l=624nm and an emission maximum at l=658 nm. The
linear absorption band of 4 is centered at l=714nm, and
the emission peak is located at l=752 nm. An extra styryl
group leads to more red-shifted linear absorption and emis-
sion spectra. The fluorescence quantum yields for 3 and 4
are 63 and 40%, respectively. Although these values are
smaller than the value for 2b, they are much higher than the
values for general cyanine dyes (<10–15%).[5] At the same
time, compounds 3 and 4 show decreased fluorescence life-
times compared with 2b, which is probably due to their
elongated p-conjugated system.
Two-photon absorption and emission spectroscopy: There
has been a lot of research focused on the development of
materials with large TPA and high fluorescence quantum
yields owing to their applications in two-photon bioimaging
and two-photon biosensing.[18–22] In Figure 3a, one can see
Chem. Eur. J. 2008, 14, 5812 – 5819
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5815

Q. Zheng, P. N. Prasad, and G. Xu
that 1a,b and 2a,b have no linear absorption above l=
600 nm. However, when these chromophores were excited
by a focused IR laser beam (l=730–1050 nm), we observed
strong frequency up-converted fluorescence. As shown in
Figure 5, the fluorescence spectrum for each compound is
Figure 5. Main graph: TPE fluorescence spectra for 1a,b and 2a,b (10À4
in toluene). Inset: Fluorescence peak intensities versus laser intensity at
m
~
:
&
*
^
l=950 nm. : 1a, slope=1.98; : 2a, slope=1.93; : 1b, slope=2.01;
2b, slope=2.02. These results demonstrate two-photon absorption.
exactly the same as that induced by one-photon excitation,
which indicates that this fluorescence is from the same fluo-
rescing excited state. However, measurements from laser-in-
tensity dependence studies of the fluorescence of 1a,b and
2a,b showed that the fluorescence results from TPA, be-
cause the emission intensity depends quadratically on the
excitation power (Figure 5, inset). Owing to the high fluores-
cence quantum yields of these chromophores, the TPA spec-
tra for 1a,b and 2a,b can be readily obtained by using the
two-photon excited (TPE) fluorescence method.[23] The TPA
action cross-section (fd) spectra for 1a,b and 2a,b are
shown in Figure 6. The TPA maximium is at l=950 nm for
both 1a and 2a (Figure 6a), which is a little less than twice
the wavelength of linear lmax. However, compound 2a has
an fd peak value of 110 GM, which is 124% larger than the
value for 1a (49 GM). One can see from Table 2 that the
fluorescence quantum yields for 1a and 2a are 60 and 98%,
respectively. Accordingly, compounds 1a and 2a have TPA
cross-sections of 82 and 112 GM, respectively. Similarly,
compound 2b has an fd peak value of 127 GM at l=
990 nm, which is 49% larger than the value for 1b (85 GM
at l=990 nm). The fluorescence quantum yields for com-
pounds 1b and 2b are 78 and 99%, respectively. Therefore,
compounds 1b and 2b have TPA cross-sections of 109 and
128 GM, respectively. A comparison between compounds 1a
and 2a and compounds 1b and 2b shows that ethyl substitu-
tions at the two b positions of the BODIPY core lead to a
bathochromically shifted TPA band and an increased TPA
peak value.
!
*
Figure 6. Two-photon excitation spectra for a) 1a ( ) and 2a ( ) and
!
*
b) 1b ( ) and 2b ( ) in toluene.
Selected compounds for cellular imaging: To test the practi-
cal applicability of these highly fluorescent probes, HeLa
(human cervix epitheloid carcinoma) cells were treated with
2b, 3, and 4 in water (containing 0.1% dimethylsulfoxide,
DMSO). Figure 7a, b, and c shows confocal microscopy
images of HeLa cells stained with 2b, 3, and 4, respectively,
and were obtained by excitation with l=514, 543, and
643 nm lasers, respectively. Figure 7, middle shows the corre-
sponding transmission images, and Figure 7, bottom shows
the overlays of the fluorescence and transmission images. As
can be seen in the images, these molecules can be easily
taken up by the cells. We did not observe any signs of mor-
phological damage to the cells upon treatment with these
chromophores for 2 to 4h, which demonstrates their non-
toxicity. As shown herein, with chemical modification of
these conformationally restricted BODIPY dyes, tuning of
the fluorescence activity across the visible and into the near-
IR spectrum was achieved. These probes can be easily con-
jugated with biomacromolecules, such as antibodies and
DNAs, because they all contain acetic acid methyl ester
groups that can be converted into acetic acid groups by a
simple hydrolysis reaction.
Significant TPE fluorescence was observed from 2b as a
result of its high fluorescence quantum yield, although its
TPA cross-section is only moderate. Therefore, two-photon
scanning microscopy imaging was carried out on HeLa cells
treated with 2b. Figure 8a shows the two-photon scanning
5816
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 5812 – 5819

Dipyrromethene Boron Difluoride Dyes
FULL PAPER
Conclusions
Novel fluorescent BODIPY dyes have been synthesized by
introducing a naphthalenyl group at the meso position of
the BODIPY core. A condensation reaction between 4-di-
methylaminobenzaldehyde and a methyl group at the a po-
sition of the BODIPY core afforded chromophores that
emitted in the red-color and near-IR regions. All new com-
1
pounds were characterized by using H and 13C NMR spec-
troscopy, HRMS, and elemental analysis. Compounds 1a
and 2a were further characterized by X-ray diffraction to
study the structure–property relationship of this class of
dyes. BODIPY dyes that had a naphthalenyl group at the
meso position exhibited increased fluorescence quantum
yields and longer excited-state lifetimes compared with
those that had a phenyl group at the meso position. This
effect is attributed to a reduction in the loss of energy from
nonirradiative molecular relaxation of the excited state for
chromophores with a naphthalenyl group because free rota-
tion is more difficult for a naphthalenyl group than for a
phenyl group. The red and near-IR spectrum BODIPY dyes
exhibit higher fluorescence quantum yields compared with
common cyanine dyes. The two-photon excitation spectra of
1a,b and 2a,b show that conformationally restricted
BODIPY dyes show increased TPE emission spectra com-
pared with dyes that have a phenyl group at the meso posi-
tion. The one- and two-photon fluorescence imaging of
living cells was successfully
Figure 7. Top: Confocal microscopic images of HeLa cells treated with
a) 2b, b) 3, and c) 4. Confocal microscopy images were obtained by using
laser excitation at l=514( 2b), 543 (3), and 643 nm (4). Middle: The cor-
responding transmission images. Bottom: Overlays of the fluorescence
and transmission images. All scale bars are 47.62 mm long.
demonstrated by using selected
BODIPY dyes. This work pres-
ents a novel type of optical
probe for biological applica-
tions, such as multiplex imaging
and multiphoton microscopy.
Experimental Section
Materials and instruments: All chemi-
cals were purchased from Aldrich and
were used without further purification.
1H and 13C NMR spectra were record-
ed on an Inova-500 spectrometer
(500 MHz) and Gemini-300
Figure 8. a) Two-photon confocal microscopic image and b) localized spectrum of the TPE fluorescence of
HeLa cells treated with 2b. Confocal microscopy image and two-photon fluorescence were obtained by using
laser excitation at l=1028 nm. The scale bar is 47.62 mm long.
a
(75 MHz), respectively. Elemental
analysis was carried by Atlantic Anal-
ysis Inc., Norcross, GA (USA). The
linear absorption spectra of dilute sol-
microscopy image, and Figure 8b depicts the localized spec-
utions (10À5 m in toluene) were recorded by using a Shimadzu UV-3101
trum of TPE fluorescence in cells treated with 2b and
DMSO (0.1%) in water and excited by a laser at l=
1028 nm. Compared with UV/Vis-excited imaging, this two-
photon imaging technique leads to the possibility of long-
term imaging of cellular processes and deep tissue penetra-
tion, and causes less photodamage because of the long-
wavelength excitation. These features enhance the potential
of BODIPY dyes for biological applications in multiplex
imaging and multiphoton microscopy, especially for complex
multi-imaging tasks.[22]
PC spectrophotometer.
(4-Formylphenoxy)acetic acid methyl ester:[24] Anhydrous K2CO3 (17 g,
123 mmol) was added to a solution of 4-hydroxybenzaldehyde (10 g,
82 mmol) in dry acetone (120 mL), then ethyl bromoacetate (10.9 mL,
120 mmol) was added dropwise at room temperature. The mixture was
heated at reflux for 4h, then filtered and the solvent was removed under
reduced pressure. The residue was dissolved in AcOEt (50 mL), then
washed with NaOH (2n) and water. The organic layer was dried over an-
hydrous Na2SO4, filtered, and the solvent was removed under reduced
pressure to give the title compound as a white solid that was sufficiently
pure to be used in the following step (9.5 g, 45.7 mmol, 92%). 1H NMR
Chem. Eur. J. 2008, 14, 5812 – 5819
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5817

Q. Zheng, P. N. Prasad, and G. Xu
(500 MHz, CDCl3, 258C, TMS): d=9.90 (s, 1H), 7.85 (d, J=9.0 Hz, 2H),
7.01 (d, J=9.0 Hz, 2H), 4.73 (s, 2H), 3.82 ppm (s, 3H); 13C NMR
(125 MHz, CDCl3, 258C, TMS): d=190.6, 168.5, 162.6, 131.9, 130.9,
114.9, 65.1, 52.4 ppm.
4-Formylnaphthalene-1-yloxyacetic acid methyl ester:[25] The synthesis of
this precursor is analogous to that of (4-formylphenoxy)acetic acid
methyl ester (92%). 1H NMR (500 MHz, CDCl3, 258C, TMS): d=10.22
(s, 1H), 9.29 (d, J=8.0 Hz, 1H), 8.43 (d, J=8.0 Hz, 1H), 7.89 (d, J=
8.0 Hz, 1H), 7.72 (t, J=8.0 Hz, 1H), 7.61 (t, J=8.0 Hz, 1H), 6.79 (d, J=
8.0 Hz, 1H), 4.92 (s, 2H), 3.85 ppm (s, 3H); 13C NMR (75 MHz, CDCl3,
258C, TMS): d=192.2, 168.3, 138.8, 131.9, 129.8, 126.7, 125.8, 125.4,
124.8, 122.5, 103.6, 65.4, 52.5 ppm.
Monostyryl- and distyryl-BODIPY dyes (3 and 4): 4-Dimethylaminobenz-
aldehyde (0.29 g, 2 mmol), 2b (0.52 mg, 1 mmol), glacial acetic acid
(0.7 mL), and piperidine (0.8 mL) were heated at reflux for 24h in dry
toluene (50 mL) in the presence of a small amount of activated 4Š mo-
lecular sieves. The mixture was cooled to room temperature, the solvents
were removed under reduced pressure, and the crude product was placed
on a silica column and eluted with methylene chloride. The blue-colored
fraction was collected and recrystallized from methylene chloride/metha-
nol to give 3 as a glossy red powder (0.123 g, 19%), and the green-col-
ored fraction was collected and the solvent was removed under reduced
pressure to give 4 as a black crystalline solid (0.062 g, 8%).
Monostyryl-BODIPY 3: 1H NMR (500 MHz, CDCl3, 258C, TMS): d=
8.42 (d, J=8.5 Hz, 1H), 7.78 (d, J=8.5 Hz, 1H), 7.61 (d, J=16.5 Hz,
1H), 7.55–7.50 (m, 3H), 7.46 (t, J=7.5 Hz, 1H), 7.28 (s, 1H), 7.20 (d, J=
16.5 Hz, 1H), 6.81 (d, J=7.5 Hz, 1H), 6.71 (d, J=9.0 Hz, 2H), 4.91 (s,
2H), 3.87 (s, 3H), 3.02 (s, 6H), 2.60 (s, 3H), 2.56–2.53 (m, 2H), 2.28–2.22
(m, 2H), 1.10 (t, J=7.5 Hz, 3H), 1.03 (s, 3H), 1.01 (s, 3H), 0.95 ppm (t,
J=7.5 Hz, 3H); 13C NMR (125 MHz, CDCl3, 258C, TMS): d=169.0,
154.2, 150.8, 136.2, 133.3, 128.7, 127.8, 127.0, 126.2, 126.1, 125.8, 125.6,
125.2, 122.2, 112.2, 104.8, 65.6, 52.4, 40.3, 18.4, 17.1, 14.6, 14.0, 11.3,
11.0 ppm; HRMS: m/z: calcd for C39H42BF2N3O3: 649.3396; found:
649.3390; elemental analysis calcd (%) for C39H42BF2N3O3: C 72.11, H
6.52, N 6.47; found: C 71.96, H 6.49, N 6.42.
Distyryl-BODIPY 4: 1H NMR (500 MHz, CDCl3, 258C, TMS): d=8.43
(d, J=8.0 Hz, 1H), 7.81 (d, J=8.0 Hz, 1H), 7.67 (d, J=16.0 Hz, 2H),
7.56–7.52 (m, 4H), 7.46 (t, J=8.0 Hz, 1H), 7.29–7.27 (m, 4H), 7.21 (d,
J=16.0 Hz, 2H), 6.81 (d, J=8.0 Hz, 1H), 6.75 (d, J=8.0 Hz, 2H), 4.91 (s,
2H), 3.87 (s, 3H), 3.03 (s, 12H), 2.55–2.52 (m, 4H), 1.11 (t, J=7.5 Hz,
6H), 1.03 ppm (s, 6H); 13C NMR (75 MHz, CDCl3, 258C, TMS): d=
169.2, 154.6, 150.7, 136.1, 130.6, 128.8, 126.1, 125.4, 122.2, 116.2, 112.3,
104.8, 65.7, 52.3, 40.4, 17.9, 14.0, 11.0 ppm; HRMS: m/z: calcd for
C48H51BF2N4O3: 780.4017; found: 780.4026; elemental analysis calcd (%)
for C48H51BF2N4O3: C 73.84, H 6.58, N 7.18; found: C 73.39, H 6.72, N
7.04.
General procedure for the preparation of 1a,b and 2a,b: Benzaldehyde
(5 mmol) and 2,4-dimethylpyrrole (10 mmol) were dissolved in dry
CH2Cl2 (250 mL) under an argon atmosphere. One drop of TFA was
added and the solution was stirred overnight at room temperature. After
complete consumption of the aldehyde (monitored by TLC; silica,
CH2Cl2 eluent), a solution of DDQ (5 mmol) in CH2Cl2 was added with
stirring. Triethylamine (14.3 mmol) was added to the resulting solution,
followed by BF3·OEt2 (19.8 mmol). The mixture was stirred for 40 min,
then washed with water. The aqueous solution was extracted with
CH2Cl2, and the combined organic fractions were dried over Na2SO4, fil-
tered, and the solvent evaporated under reduced pressure. The crude
compound was purified by flash chromatography (silica gel, CH2Cl2
eluent).
1,3,5,7-Tetramethyl-8-(4-methoxycarbonylmethyloxyphenyl)-4,4’-difluoro-
boradiazaindacene (1a): Red crystalline solid (0.72 g, 35%). 1H NMR
(500 MHz, CDCl3, 258C, TMS): d=7.20 (d, J=8.5 Hz, 2H), 7.03 (d, J=
8.5 Hz, 2H), 5.98 (s, 2H), 4.70 (s, 2H), 3.83 (s, 3H), 2.55 (s, 6H),
1.41 ppm (s, 6H); 13C NMR (75 MHz, CDCl3, 258C, TMS): d=168.9,
158.3, 155.4, 143.1, 141.3, 131.7, 129.4, 128.2, 121.2, 115.3, 65.3, 52.3, 14.6,
14.5 ppm; HRMS: m/z: calcd for C22H23BF2N2O3: 412.1764; found:
412.1769; elemental analysis calcd (%) for C22H23BF2N2O3: C 64.10, H
5.62, N 6.80; found: C 64.13, H 5.63, N 6.80.
X-Ray crystallography: A crystal of 1a or 2a was mounted on the dif-
fractometer, and data were collected at 90 K. The reflection intensities
were collected by using a Bruker APEX II CCD diffractometer installed
at a rotating anode source (MoKa radiation, l=0.71073 Š) and equipped
with an Oxford Cryosystems nitrogen flow apparatus. The collection
method involved 0.38 scans in w at 208 in 2q. Data integration up to a
resolution of 0.70 Š was carried out by using SAINT V7.34[26a] with re-
flection spot size optimization. Absorption corrections were made with
the program SADABS.[26a] The structure was solved by using the direct
methods procedure and refined by using least-squares methods against F2
with SHELXS-97 and SHELXL-97.[26b] Nonhydrogen atoms were refined
anisotropically, and hydrogen atoms were allowed to ride on the respec-
tive atoms. Crystal data and details of the data collection and refinement
for 1a and 2a are summarized in Table 1. CCDC-678553 (1a) and
-678554( 2a) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
2,6-Diethyl-1,3,5,7-tetramethyl-8-(4-methoxycarbonylmethyloxyphenyl)-
4,4’-difluoroboradiazaindacene (1b): Red crystalline solid (0.87 g, 37%).
1H NMR (500 MHz, CDCl3, 258C, TMS): d=7.19 (d, J=8.5 Hz, 2H),
7.02 (d, J=8.5 Hz, 2H), 4.71 (s, 2H), 3.83 (s, 3H), 2.53 (s, 6H), 2.30 (q,
J=7.5 Hz, 4H), 1.31 (s, 6H), 0.98 ppm (t, J=7.5 Hz, 6H); 13C NMR
(75 MHz, CDCl3, 258C, TMS): d=169.0, 158.2, 153.6, 139.7, 138.3, 132.7,
131.0, 129.6, 129.1, 115.2, 65.3, 52.3, 17.0, 14.6, 12.5, 11.8 ppm; HRMS: m/
z: calcd for C26H31BF2N2O3: 468.2390; found: 468.2391; elemental analy-
sis calcd (%) for C26H31BF2N2O3: C 66.68, H 6.67, N 5.98; found: C
66.50, H 6.62, N 6.00.
1,3,5,7-Tetramethyl-8-(4-methoxycarbonylmethyloxynaphthalen-1-yl)-4,4’-
difluoroboradiazaindacene (2a): Dark red crystalline solid (0.88 g, 38%).
1H NMR (500 MHz, CDCl3, 258C, TMS): d=8.42 (d, J=8.5 Hz, 1H),
7.73 (d, J=8.5 Hz, 1H), 7.54(t, J=7.5 Hz, 1H), 7.47 (t, J=7.5 Hz, 1H),
7.26 (s, 1H), 6.81 (d, J=8.0 Hz, 1H), 5.93 (s, 2H), 4.89 (s, 2H), 3.86 (s,
3H), 2.58 (s, 6H), 1.10 ppm (s, 6H). 13C NMR (75 MHz, CDCl3, 258C,
TMS): d=168.8, 155.5, 154.5, 143.0, 140.1, 132.8, 132.3, 128.0, 126.3,
125.7, 124.7, 122.4, 121.1, 104.8, 65.6, 52.3, 14.6, 14.0 ppm; HRMS: m/z:
calcd for C26H25BF2N2O3: 462.1921; found: 462.1926; elemental analysis
calcd (%) for C26H25BF2N2O3: C 67.55, H 5.45, N 6.06; found: C 67.55, H
5.46, N 6.01.
Fluorescence properties and fluorescence quantum yields: Steady-state
fluorescence spectroscopic studies were performed by using a Jobin–
Yvon Fluorolog-3 (model FL-3-11) spectrofluorometer. The relative
quantum yields of fluorescence of 1a,b and 2a,b were obtained by com-
paring the area under the corrected emission spectrum of the test sample
with that of a solution of Rhodamine 6G in methanol, which has a quan-
tum yield of 94% according to the literature.[15] The relative quantum
yields of 3 and 4 were obtained by using Cresyl Violet (54% in metha-
nol)[16] and Nile Blue (26% in ethylene glycol),[17] respectively, as exter-
nal references. For the fluorescence experiments, dilute solutions with an
optical density of below 0.2 at the absorption maximum were used.
2,6-Diethyl-1,3,5,7-tetramethyl-8-(4-methoxycarbonylmethyloxynaphtha-
len-1-yl)-4,4’-difluoroboradiazaindacene (2b): Dark red crystalline solid
(1.17 g, 45%). 1H NMR (500 MHz, CDCl3, 258C, TMS): d=8.42 (d, J=
8.5 Hz, 1H), 7.75 (d, J=8.5 Hz, 1H), 7.54(t, J=7.5 Hz, 1H), 7.46 (t, J=
7.5 Hz, 1H), 7.26 (s, 1H), 6.81 (d, J=8.0 Hz, 1H), 4.90 (s, 2H), 3.87 (s,
3H), 2.56 (s, 6H), 2.25 (q, J=7.5 Hz, 4H), 1.01 (s, 6H), 0.94ppm (t, J=
7.5 Hz, 6H); 13C NMR (75 MHz, CDCl3, 258C, TMS): d=168.9, 154.3,
153.7, 138.4, 138.2, 133.0, 132.6, 131.6, 127.9, 126.6, 126.2, 125.8, 125.6,
125.0, 122.3, 104.8, 65.6, 52.3, 17.0, 14.6, 12.5, 11.2 ppm; HRMS: m/z:
calcd for C30H33BF2N2O3: 518.2547; found: 518.2549; elemental analysis
calcd (%) for C30H33BF2N2O3: C 69.51, H 6.42, N 5.40; found: C 69.40, H
6.36, N 5.35.
TPA measurements: The TPA spectra were determined by using laser
pulses with a repetition rate of 80 MHz, a pulse duration of about 140 fs,
and a wavelength range of l=680–1080 nm that was generated by a
mode-locked Ti/sapphire laser (Chameleon Ultra II, Coherent, Inc.). All
data were obtained in toluene (100 mm) by using the TPE fluorescence
method with Rhodamine B (100 mm in methanol) as a reference.[23]
5818
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 5812 – 5819

Dipyrromethene Boron Difluoride Dyes
FULL PAPER
In vitro cellular imaging with tumor cells: HeLa cells (American Type
Culture Collection, Manassas, VA) were cultured in minimum essential
medium (MEM) supplemented with 10% fetal bovine serum (FBS), ac-
cording to the instructions supplied by the vendor. The cells were trypsi-
nized and resuspended in MEM alpha medium with 10% FBS at a con-
centration of 7.5”105 cellsmLÀ1, then plated in 35 mm culture plates by
using 2.5 mL of medium that contained 0.10 mL of the cell suspension.
The plates were incubated overnight at 378C under a 5% CO2 atmos-
phere. Then the cells (50% confluency) were carefully rinsed with phos-
phate-buffered saline (PBS), and the medium (2.0 mL), which contained
the aqueous dispersion (80 mL) also containing solutions of selected com-
pounds (10À4 m) in 2.5% DMSO, was added to plate and mixed gently.
The treated cells were reincubated for 1 h. Then the plates were rinsed
three times with sterile PBS, and fresh media were added. The cells were
then directly imaged by using a Leica laser scanning confocal inverted
microscope (TCS SP2, Leica Microsystems Semiconductor GmbH). This
microscope was equipped with multiple visible laser lines (l=514, 543,
and 633 nm) for one-photon excitation and confocal imaging, and a Ti/
sapphire laser (Chameleon Ultra II, Coherent, Inc.) for two-photon exci-
tation (l=1028 nm). The lightsource under the microscope was con-
trolled by using neutral-density filters and was maintained at 10 mW for
the two-photon imaging process. A 63.0” oil immersion objective (HCX
PL APO CS, 63”, NA 1.40) was used for all cell-imaging processes.
Three different imaging channels at l=520–580, 600–690, and 710–
780 nm were used for 2b, 3, and 4, respectively, to remove the scattering
influence from the excitation lights.
[5] J. Fabian, H. Nakazumi, M. Matsuoka, Chem. Rev. 1992, 92, 1197–
1226.
[6] M. Shah, K. Thangaraj, M. L. Soong, L. Wolford, J. H. Boyer, I. R.
Politzer, T. G. Pavlopoulos, Heteroat. Chem. 1990, 1, 389–399.
[7] A. Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891–4932.
[8] Z. Dost, S. Atilgan, E. U. Akkaya, Tetrahedron 2006, 62, 8484–8488.
[9] J. S. Li, H. Wang, L. W. Cao, H. S. Zhang, Talanta 2006, 69, 1190–
1199.
[10] M. Kollmannsberger, K. Rurack, U. Resch-Genger, W. Rettig, J.
Daub, Chem. Phys. Lett. 2000, 329, 363–369.
[11] W. Qin, M. Baruah, M. Van der Auweraer, F. C. De Schryver, N.
Boens, J. Phys. Chem. A 2005, 109, 7371–7384.
[12] H. L. Kee, C. Kirmaier, L. H. Yu, P. Thamyongkit, W. J. Young-
blood, M. E. Calder, L. Ramos, B. C. Noll, D. F. Bocian, W. R.
Scheidt, R. R. Birge, J. S. Lindsey, D. Holten, J. Phys. Chem. B 2005,
109, 20433–20443.
[13] Z. Shen, H. Rohr, K. Rurack, H. Uno, M. Spieles, B. Schulz, G.
Reck, N. Ono, Chem. Eur. J. 2004, 10, 4853–4871.
[14] N. Kuhn, A. Kuhn, M. Speis, D. Blaser, R. Boese, Chem. Ber. 1990,
123, 1301–1306.
[15] W. Baumler, A. Penzkofer, Chem. Phys. 1990, 140, 75–97.
[16] D. Magde, J. H. Brannon, T. L. Cremers, J. Olmsted, J. Phys. Chem.
1979, 83, 696–699.
[17] R. Sens, K. H. Drexhage, J. Lumin. 1981, 24/25, 709–712.
[18] T. R. Krishna, M. Parent, M. H. V. Werts, L. Moreaux, S. Gmouh, S.
Charpak, A. M. Caminade, J. P. Majoral, M. Blanchard-Desce,
Angew. Chem. 2006, 118, 4761–4764; Angew. Chem. Int. Ed. 2006,
45, 4645–4648.
[19] C. Stosiek, O. Garaschuk, K. Holthoff, A. Konnerth, Proc. Natl.
Acad. Sci. USA 2003, 100, 7319–7324.
[20] H. J. Koester, D. Baur, R. Uhl, S. W. Hell, Biophys. J. 1999, 77,
2226–2236.
Acknowledgements
This work was supported in part by a grant from the Chemistry and Life
Sciences Directorate of the Air Force Office of Scientific Research and
in part by the John R. Oishei Foundation. Partial support from the
Center of Excellence in Bioinformatics and Life Sciences at University at
Buffalo is also acknowledged. We thank Dr. Shao-liang Zheng for obtain-
ing the crystal structures and Dr. Hong Ding for providing the HeLa
cells.
[21] Q. Zheng, T. Y. Ohulchanskyy, Y. Sahoo, P. N. Prasad, J. Phys.
Chem. C 2007, 111, 16846–16851.
[22] P. N. Prasad, Introduction to Biophotonics, Wiley, New York, 2003.
[23] C. Xu, W. W. Webb, J. Opt. Soc. Am. B 1996, 13, 481–491.
[24] M. D. Bowman, R. C. Jeske, H. E. Blackwell, Org. Lett. 2004, 6,
2019–2022.
[25] P. Madsen, A. Ling, M. Plewe, C. K. Sams, L. B. Knudsen, U. G.
Sidelmann, L. Ynddal, C. L. Brand, B. Andersen, D. Murphy, M.
Teng, L. Truesdale, D. Kiel, J. May, A. Kuki, S. Shi, M. D. Johnson,
K. A. Teston, J. Feng, J. Lakis, K. Anderes, V. Gregor, J. Lau, J.
Med. Chem. 2002, 45, 5755–5775.
[1] A. S. Waggoner, L. Stryer, Proc. Natl. Acad. Sci. USA 1970, 67, 579–
589.
[2] R. Y. Tsien, Annu. Rev. Neurosci. 1989, 12, 227–253.
[3] X. H. Fang, J. W. J. Li, J. Perlette, W. H. Tan, K. M. Wang, Anal.
Chem. 2000, 72, 747A–753A.
[4] A. L. Tatarets, I. A. Fedyunyayeva, T. S. Dyubko, Y. A. Povrozin,
A. O. Doroshenko, E. A. Temetschnig, L. D. Patsenker, Anal. Chim.
Acta 2006, 570, 214–223.
[26] a) Bruker AXS, SMART and SAINTPLUS, Bruker AXS, Madison,
Wisconsin (USA), 2006; b) G. M. Sheldrick, SHELXS-97 and
SHELXL-97, Bruker Analytical Instrumentation, Madison, Wiscon-
sin (USA), 2000.
Received: February 19, 2008
Published online: May 21, 2008
Chem. Eur. J. 2008, 14, 5812 – 5819
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5819