Synthesis and spectroscopic characterization of fluorescent boron dipyrromethene-derived hydrazones

Article abstract of DOI:10.1007/s10895-010-0723-0

Derivatives of 4,4-difluoro-4-bora-3a,4a,diaza-s-indacene (BODIPY or BDP) that possess a hydrazine substituent on position 5 are potential "turn-on" fluorophores for labeling aldehydes The unnatural amino acid L-3-formyltyrosine can be incorporated into a

Full text of DOI:10.1007/s10895-010-0723-0

J Fluoresc (2011) 21:347–354
DOI 10.1007/s10895-010-0723-0
ORIGINAL PAPER
Synthesis and Spectroscopic Characterization of Fluorescent
Boron Dipyrromethene-Derived Hydrazones
Ozlem Dilek & Susan L. Bane
Received: 13 June 2010 /Accepted: 8 September 2010 /Published online: 1 October 2010
#
Springer Science+Business Media, LLC 2010
Abstract Derivatives of 4,4-difluoro-4-bora-3a,4a,diaza-s-
indacene (BODIPY® or BDP) that possess a hydrazine
substituent on position 5 are potential “turn-on” fluoro-
phores for labeling aldehydes The unnatural amino acid L-
3-formyltyrosine can be incorporated into a protein or
peptide; thus, these hydrazines are potentially site specific
labels for such polymers. In this work, model compounds
were synthesized to assess whether the photochemical
properties of the BDP-hydrazone would be suitable for
protein labeling. Hydrazones were synthesized from the
fluorophore 3-chloro-5-hydrazino-BDP and different alde-
hydes, and the absorption and emission spectra of the
products were compared. The hydrazone of an unsubsti-
tuted aromatic aldehyde displays absorption and emission
maxima (531 nm and 559 nm, respectively in dioxane) that
are red shifted relative to those of a hydrazone from an
aliphatic aldehyde (513 nm and 543 nm, respectively, in
dioxane) and an increased quantum yield (0.21 vs. 0.11,
respectively, in dioxane). The presence of a hydroxyl group
ortho- to the aldehyde produces a hydrazone in which the
absorption and emission maxima are slightly red shifted
(528 nm and 564 nm, respectively in dioxane) from the
unsubstituted aromatic hydrazone, but the quantum yields
of the two hydrazones are equivalent. Thus, an ortho-
hydroxy substituted aromatic aldehyde is a suitable electro-
phile for “turn on” protein labeling using the hydrazino-BDP.
The specificity of this labeling reaction for the unnatural
amino acid was demonstrated through fluorescent labeling of
just the 3-formyltyrosine-containing α-subunit of α,β-
tubulin.
.
.
Keywords BODIPY Hydrazones Bioorthogonal
.
.
labeling Fluorescent probes Unnatural amino acids
Introduction
Boron dipyrromethene-based fluorescent dyes (BODIPY®
or BDP) are versatile and valuable probes for many
biological systems. Structural modifications of the core
fluorophore can produce molecules with a large range of
the absorption and emission energies—from visible light to
the infrared and even microwave region of the electromag-
netic spectrum [1, 2]. Absorption and emission energies of
the dye can be tuned by the nature of the substituents on the
dipyrromethene template and the boron atom. Rohand et al.
showed that molecule 1 is amenable to substitution at the 3-
and 5- positions by a variety of nucleophiles [3] and their
photochemical properties of the BDP are highly dependent
on the nature of these substitutents [4]. We are interested in
exploiting this property of BDPs to create bioorthogonal
protein labeling systems in which the BDP fluorescence is
“turned on” by formation of the fluorophore-biomolecule
covalent bond. Reactions between a modified biomolecule
and its chemical partner can be exploited to selectively
incorporate a reporter molecule into a chosen site within the
biological target [5, 6]. For example, reactions between
azides and phosphine ligands or alkynes have been
successfully exploited to tag proteins, nucleic acid poly-
mers and carbohydrates [7, 8].
:
O. Dilek S. L. Bane (*)
Department of Chemistry,
State University of New York at Binghamton,
Vestal Parkway,
Binghamton, NY 13902, USA
e-mail: sbane@binghamton.edu
We have chosen to focus on the reaction of aldehydes
with hydrazine-containing fluorophores, since these reac-
tions are known to occur in aqueous solution to form a

348
J Fluoresc (2011) 21:347–354
Fig. 1 Structures of BDP
derivatives
5 R =
NH N C
H
1 R = Cl
2 R = NHNH2
HO
3 R = NH-N=CH-CH2CH3
N
N
F
6 R =
NH N C
H
OCH3
4 R =
NH N C
B
H
R
Cl
F
H3CO
stable product [9]. In addition, the optical properties of the
fluorophore may change upon hydrazone formation by
fluorophores in which the hydrazine functional group is
directly bonded to the conjugated system. We have shown
previously that while 5-hydrazino-BDP’s are weakly
fluorescent in solution, hydrazone formation causes a 30-
45 fold increase in the BDP quantum yield [10]. Moreover,
we found that the nature of the aldehyde affects the
fluorescence of the hydrazone. Specifically, reaction of the
BDP-hydrazine 2 with propanal yields the aliphatic
hydrazone 3, which has an emission maximum at higher
energy than that of the aromatic hydrazone 4 (Fig. 1). This
result indicates that a protein containing an aromatic
aldehyde could be distinguished from endogenous aliphatic
aldehydes that may be present in the biological system.
We are engaged in developing methodologies for
incorporating aromatic aldehydes into proteins. In one
manifestation of this work, we have shown that the
aromatic amino acid L-3-formyltyrosine can be selectively
bonded to the C-terminus of just one of the subunits of the
tubulin heterodimer [11]. The aromatic aldehyde is there-
fore available for reaction with hydrazine-containing
fluorophores. Since the electrophile on the protein is a
substituted benzaldehyde, it is of interest to evaluate the
effect of an ortho-substituent on the aromatic aldehyde on
the reaction of 5-hydrazinyl-BDP’s and the photochemical
properties of the resulting product (Fig. 2).
the previously characterized aliphatic and aromatic BDP-
hydrazones.
It is also important to confirm that the hydrazone
reaction occurs under conditions suitable for protein
labeling. Kinetics of hydrazone formation at neutral pH
was therefore examined using the model compound
salicylaldehyde. Finally, the potential applicability of this
fluorophore-unnatural amino acid reactive pair was
assessed using 3-formyltyrosinated tubulin.
Experimental
Materials
Dry methanol and dichloromethane were purchased from
Acros Chimica and Aldrich. Other solvents and aldehydes
were also obtained from Acros or Aldrich and were used as
received. Deuterated solvents were obtained from Cam-
bridge Isotope Laboratories. All other chemicals are
commercially available and were used without further
purification. BDP compounds 1–4 were synthesized as
described previously [10].
General Synthetic Procedures
1H, 13C, 11B NMR spectra were recorded on instruments
1
Hydrazones formed between salicylaldehyde and aryl
hydrazines are capable of undergoing tautomerization into
quinoid structures, which are normally not fluorescent [12].
It was therefore important to assess the effect of hydroxyl
substituent on the fluorescence spectra of the BDP-
hydrazone. In this work, we synthesized two model
compounds by allowing BDP-hydrazine 2 to react with
salicylaldehyde or 2,4-dimethoxybenzaldehyde and com-
pared their absorption and emission spectra with those of
operating at a frequency of 360 MHz. H and 13C NMR
spectra were referenced to CDCl3 (7.26 ppm or 77.00 ppm).
11B NMR spectra were referenced to BF3.OEt2 (0 ppm).
Chemical shift multiplicities are reported as s=singlet, d=
doublet, t=triplet, q=quartet and m=multiplet. Flash
column chromatography was performed using Baker silica
gel 60–200 mesh or 200–400 mesh. All flash columns were
performed using the same solvent conditions used for TLC.
All moisture and air sensitive reactions were carried out
Fig. 2 Structures of aromatic
aldehydes and hydrazone
tautomers
OH
O
H
NH
H
NH
H
OH
N
O
HN
R
R = H, salicylaldehyde
R = CH2CH(NH2)COOH, 3-formyltyrosine
quinoid tautomer
hydrazone

J Fluoresc (2011) 21:347–354
349
under nitrogen or argon atmosphere using oven dried
glassware.
3-Chloro-3a, 4a-diaza-4,4-difluoro-5-(2,4-dimethoxy)
benzylidenehydrazino-8-phenyl borondipyrromethene (6)
Synthesis of BDP-hydrazones
One equivalent of 2,4-dimethoxybenzaldehyde (7.5 mg)
was added to 15 mg (0.045 mmol) of compound 2 in 90 μl
CH2Cl2. Neutral Alumina (Brockman Activity I, 0.14 g)
was added to the mixture. The color of the mixture changed
immediately after addition of 2,4-dimethoxybenzaldehyde.
The reaction was followed by TLC and UV-vis. The crude
residue was purified by silica gel flash column chromatogra-
phy to afford the desired product 6 as pink solid (7 mg, 35 %
yield); Rf=0.52 (100% CH2Cl2); IR (cm-1, CDCl3): υ 3324,
2926, 1727, 1573, 1469, 1412, 1276, 1084. 1H NMR
(CDCl3): δ 3.86 (s, CH3), 3.88 (s, CH3), 6.18 (s, aromatic),
6.21 (d, 1H, J=4.40 Hz), 6.26 (d, 1H, J=5.50 Hz), 6.39 (d,
2H, J=4.40 Hz), 6.46 (s, aromatic), 6.55 (m, J=8.43 Hz ),
6.82 (t, aromatic, J=5.86 Hz), 6.92 (d, 1H, J=5.50 Hz), 7.46
(m, 5H, aromatic), 7.87 (d, aromatic, J=8.43 Hz), 8.35 (s,
CH), 9.29 (s, NH). 13C NMR (CDCl3): δ 55.4, 55.6, 97.9,
106.0, 112.5, 112.9, 114.5, 115.1, 117.7, 120.9, 127.5, 128.0,
128.3, 128.7, 129.3, 130.3, 130.5, 133.3, 134.9, 135.3,
143.8, 163.3. 11B NMR (CDCl3): δ 0.95 (t, J=36.6 Hz).
3-Chloro-3a, 4a-diaza-4,4-difluoro-5-salicylidenehydrazino-
8-phenyl boron dipyrromethene (5)
One equivalent of salicylaldehyde (6.6 mg) was added to
18 mg (0.054 mmol) of compound 2. The mixture was
stirred without solvent for 1.5 h. The color of the mixture
changed immediately after addition of salicylaldehyde. The
reaction was followed by TLC and UV-vis. The crude
residue was purified by silica gel flash column chromatog-
raphy to afford the desired product 5 as pink solid (16 mg,
76% yield); Rf=0.64 (100% CH2Cl2); IR (cm-1, CDCl3): υ
3256, 2950, 1609, 1481, 1412, 1197, 1096. 1H NMR
(CDCl3): δ 6.25 (d, 1H, J=4.28 Hz), 6.49 (d, 1H, J=
4.28 Hz), 6.59 (d, 1H, J=4.70 Hz), 6.97 (d, 1H, J=
4.70 Hz), 7.00-7.24 (m, 2H, aromatic), 7.35 (m, 2H,
aromatic), 7.50 (m, 5H, aromatic ), 8.12 (s, 1H, CH), 9.29
(s, 1H, NH), 9.91 (s, 1H, OH) 13C NMR (CDCl3): δ 110.8,
113.9, 117.0, 120.1, 123.0, 128.0, 128.4, 129.8, 129.9,
130.4, 131.3, 132.1, 132.5, 133.0, 133.4, 133.6, 135.6,
150.0, 157.7, 164.7. 11B NMR (CDCl3): δ 0.50 (t, J=
33.5 Hz).
Absorption and Fluorescence Spectroscopy
Absorption spectra were obtained using an HP 8453
absorption spectrophotometer equipped with a circulating
water bath for temperature control. Kinetic data were
obtained using absorption difference spectroscopy [13].
Equal volumes of each reactant were placed separately
placed in a dual chambered quartz cuvette. The instrument
was blanked and the solutions were mixed by inversion of
the cuvette. Absorption spectra were collected every 30 s
until a steady state was achieved.
0.10
a
0.08
0.06
0.04
0.02
Kinetic data were collected in 0.1 M phosphate buffer
(pH 4.0 or 7.0) at 23°C. Experiments were performed by
0.00
400 425 450 475 500 525 550 575 600
80
60
40
20
0
Wavelength (nm)
0.06
b
0.05
0.04
0.03
0.02
0.01
500
525
550
575
600
625
650
675
700
Wavelength (nm)
0.00
400 425 450 475 500 525 550 575 600
Fig. 4 Normalized fluorescence emission spectra of 2–6 in dioxane.
From left to right, compound 2 (black), compound 3 (red), compound
4 (blue), compound 5 (green), compound 6 (light blue). Molecules
were excited at their absorption maxima at room temperature (23°C)
Wavelength (nm)
Fig. 3 Normalized absorption spectra of BDP hydrazones 5 (panel A)
and 6 (panel B) in dioxane (black) and methanol (red)

350
J Fluoresc (2011) 21:347–354
mixing 65 μM hydrazine 2 with an excess of aldehyde
(salicylaldehyde: 650 μΜ, lmax=326 nm, ε326=3400 cm-1
M-1; propanal: 2 mM)
Fluorescence spectra were collected using a Spex
FluoroMax-3 spectrofluorometer. The fluorescence quan-
tum yields (fF) were determined in dilute solutions with an
absorbance below 0.1 at the excitation wavelength. Quinine
sulfate in 0.1 M H2SO4 (λex=347 nm, ϕF=0.57) was used
as a standard (Lakowicz, 2006). All spectra were recorded
at 23°C. Quantum yields were calculated using the
following equation:
À
ÁꢀÀ
Á
À
ꢀ
Á
À
ꢀ
Á
fsample ¼ fFstandard ꢀ Fsample ꢁ Fsolvent
Fstandard ꢁ Fsolvent ꢀ hsample hstandard ꢀ Astandard Asample
F
where F denotes the area under the fluorescence band, A
denotes the absorbance at the excitation wavelength, and
η denotes the refractive index of the solvent. Integration
of the emission bands was performed using SigmaPlot
10.0.
Results and Discussion
In a previous study, we showed that BDP’s with a
hydrazino substituent at the 5-position of BDP are weakly
fluorescent, and that hydrazone formation induces an
increase in fluorescence quantum yield and a red shift in
the emission spectrum [10].
Covalent Labeling of α-tubulin with BDP-hydrazine 2
The emission energies and quantum yields are different
for an aliphatic and aromatic hydrazine. Since the unnatural
amino acid we hope to covalently label is a substituted
benzaldehyde, it was necessary to ascertain whether a
corresponding substituent on the aromatic ring affects the
emission properties of the BDP-hydrazone. We therefore
synthesized two model hydrazones (5 and 6).
Absorption and emission spectra of the two hydrazones
were obtained in an apolar, aprotic solvent (dioxane) and a
polar, protic solvent (methanol). The absorption spectra are
broad (Fig. 3) and feature two narrower bands centered at
about 500 nm and around 530 nm. The overall broadness
may be attributed to the asymmetrical substitution of the
dipyrromethene at the 3,5-positions [16] and the amine-like
substituent on the 3-position. The absorption spectrum of a
similarly substitute molecule is also broad in protic solvent
and sharper in aprotic solvent [17] The relative intensity of
these two bands is affected by solvent and by the
substituent on the 5-position, but the shape of the emission
is the same when the molecule is excited in either peak.
Because both hydrazones possess these broad bands and
since no quionoid tautomer was observed in the NMR
Tubulin was purified from bovine brain as described by
Williams and Lee [14]. The carboxy terminal of α-tubulin
was cleaved by carboxypeptidase A and was replaced with
either L-tyrosine or L-3-formyltyrosine using tubulin
tyrosine ligase as described in Banerjee et al. [11].
Retyrosinated tubulin was prepared in PME buffer (0.1 M
PIPES, 1 mM MgSO4 and 2 mM EGTA, pH 6.9 at 25°C).
BDP-hydrazine 2 was dissolved in DMSO to a final
concentration of 1 mM. While vortexing the protein
solution, the dye 2 in DMSO was slowly added to each
reaction solution to a final concentration of 100 μM (1:10
molar ratio of tubulin: dye). The reactions were incubated
for 2.5 h at 25°C on a rocker. Unreacted dye was removed
by rapid gel filtration using Sepharose G-25 columns [15].
After addition of sample buffer, the samples were boiled
for 5 min and 6.5 μg of protein was loaded on an SDS-
PAGE gel. After electrophoresis, the gel was placed in
the destaining solution (1:5:4 acetic acid: methanol: dd-
water) overnight and photographed. The next day, the gel
was stained with Coumassie Blue, destained and photo-
graphed.
Table 1 Absorption and Emission Maxima and Quantum Yields of BDPs 3–6
Solvent
Dioxane
Methanol
Compound
labmax (nm)
lemmax(nm)
ΦF(at 23°C)
labmax (nm)
lemmax(nm)
ΦF(at 23°C)
3
4
5
6
513
543
559
564
572
0.11
0.21
0.20
0.061
502
539
555
559
545
0.064
0.098
0.062
0.006
531
521
500, 528a
502, 534a
506, 536 (sh)b
500, 533 (sh)b
a Absorption spectrum shows two peaks of similar intensity. b Absorption spectrum is broad and less structured than in dioxane

J Fluoresc (2011) 21:347–354
351
spectra of 5, the two main components in the absorption
bands are most likely the 0-0 and 0-1 vibrational bands of
the lowest energy electronic transition [18].
The rate of hydrazone formation at pH 7 was also measured
and it was slower than pH 4 (not shown). The 2nd rate order
constant calculated from the data was 27 M-1min-1.
The difference in the spectrum in dioxane and methanol is
more pronounced for compound 5 than for compound 6. The
ability of the ortho substituent to be both a hydrogen bond
donor as well as a hydrogen bond acceptor may explain the
different behavior of the two molecules in solvent.
Figure 4 shows the normalized emission spectra of
compounds 2–6 in dioxane. The emission maxima of all
hydrazones are red shifted relative to the hydrazine 2, and
all of the aromatic hydrazines emit at longer wavelength
than the aliphatic hydrazone 3.
The quantum yields of the model hydrazones are
compared in Table 1. Addition of the ortho-substituent
to the benzaldehyde produces a small red shift in the
emission maximum of the hydrazone in dioxane. Interest-
ingly, an ortho-hydroxyl group does not affect the
quantum yield of the hydrazone in this solvent, while the
hydrazone possessing an ortho-methoxy group has a
quantum yield that is about 1/3rd that of the other
aromatic hydrazones. A similar trend in emission maxima
and relative quantum yields of the hydrazones in methanol
is observed, although the magnitudes of the quantum
yields are lower methanol than in dioxane. This result
indicates that the BDP hydrazones will be more effective
“turn on” fluorophores in a hydrophobic than in a
hydrophilic environment.
BDP-hydrazine 2 with Propanal
An isolated protein can be prepared to contain just an
aromatic aldehyde. Cells, however, may also contain
aliphatic aldehydes. It was therefore of interest to evaluate
0.08
a
0.06
0.04
0.02
0
-0.02
-0.04
-0.06
300
400
500
600
Wavelength (nm)
OH
We therefore conclude that an ortho-hydroxyl group,
such as in 3-formyltyrosine, will not adversely affect the
photochemical properties of the BDP-hydrazone. Thus, the
hydrazone formed between 2 and 3-formyltyrosine will be
suitably fluorescent for BDP-hydrazine labeling of proteins.
O
N
N
F
H
N
N
F
B
B
NHNH
Cl
2
F
HN
F
N
HO
Cl
salicylaldehyde
2
0.05
0.04
0.03
0.02
0.01
0.00
Kinetics Studies
b
BDP-hydrazine 2 with Salicylaldehyde
The kinetics of hydrazone formation was examined in order
to determine experimental conditions that could be used for
protein labeling. Reaction of 3-chloro-5-hydrazino-BDP 2
with salicylaldehyde, the model compound for the unnat-
ural amino acid 3-formyltyrosine, was followed by absorp-
tion difference spectroscopy. The reaction at non-
physiological pH was measured first, since the maximum
reaction rate of the hydrazone formation is near the pKa of
the hydrazine [19]. The reaction was completed in 2 h and a
red shift of the absorption was seen, as expected. Figure 5a
displays the absorption difference spectrum produced as a
function of time. The change in absorption at 550 nm as a
function of time was fit to a pseudo 1st order reaction
(Fig. 5b). The 2nd rate order constant calculated from the
data is 37 M-1min-1.
0
2000
4000
Time (s)
6000
Fig. 5 a Absorption difference spectra for the reaction of 650 μM
salicylaldehyde with 65 μM BDP 2 in 0.1 M phosphate buffer at pH 4
at room temperature. Spectra shown were collected immediately after
mixing and at 30 s intervals thereafter. b Plot of the change in
absorbance at 550 nm as a function of time. Solid line: fit of the data
as a pseudo-first order reaction

352
J Fluoresc (2011) 21:347–354
a
b
0.03
0.02
0.01
0.00
0
1000
2000
Time (s)
3000
4000
0.000
-0.002
-0.004
-0.006
-0.008
-0.010
-0.012
-0.014
-0.016
-0.018
-0.020
0.05
0.04
0.03
0.02
0.01
0.00
c
d
0
500 1000 1500 2000 2500 3000 3500 4000 4500
0
1000
2000
Time (s)
3000
4000
Time (s)
Fig. 6 a Absorption difference spectra for the reaction of 2 mM
propanal with BDP 2 in 0.1 M phosphate buffer at pH 4 at room
temperature. Spectra shown were collected immediately after mixing
and at 30 s intervals thereafter. b Change in absorption at 550 nm as a
function of time. Solid line: Data fit as a pseudo-first order reaction. c
Change in absorption at 482 nm as a function of time. d Data from
panel C subtracted from data in panel B. Solid line: Data fit as a
pseudo-first order reaction
the kinetics of hydrazone formation with an aliphatic
aldehyde.
Hydrazones are known to be susceptible to oxidation [9]
and aliphatic hydrazones oxidize 10-100-fold more rapidly
than aromatic hydrazones [20]. We hypothesized that the
formation of an oxidation product might account for the
lack of isosbestic point and decrease in absorption at
550 nm. Plotting the decrease of absorption at 482 nm, the
isosbestic point early in the reaction time course, produced
a curve that fit a first order decay (Fig. 6c). Subtracting the
absorption at 482 nm from the change in absorption at
550 nm yielded a curve that reasonably fit a pseudo-first
order reaction (Fig. 6d). The 2nd rate order constant for
hydrazone formation is therefore estimated to be about
110 M-1min-1. The rate of hydrazone formation at pH 7 was
The kinetics reaction between propanal and 3-chloro-5-
hydrazino-BDP 2 was monitored by absorption difference
spectroscopy. Figure 6a shows absorption difference spec-
tra. The change in absorption at 550 nm was plotted as a
function of time (Fig. 6b).
Note that the reaction with the aliphatic aldehyde does
not show a sharp isosbestic point (Figs. 5a vs 6a). Also, the
change in absorption as a function of time displays normal
saturation kinetics for the aromatic aldehyde, but change in
absorption intensity for the aliphatic aldehyde reaction
reaches a maximum and then slowly decreases with time.
Table 2 Rate constants for hydrazone formation of BDP 2 with the aliphatic and aromatic aldehyde
Reactants
pH
2nd order rate constant, M-1min-1
1st order decomposition constant, min-1
BDP 2+propanal
4
7
4
7
~110
~30
37
0.03
0.01
–
BDP 2+propanal
BDP 2+salicylaldehyde
BDP 2+salicylaldehyde
27
–

J Fluoresc (2011) 21:347–354
353
a
b
occurs at neutral pH at a modest rate, and the product does
not appear to oxidize under the reaction conditions.
Reaction of 2 with 3-formyltyrosinated tubulin fluores-
cently labels just the subunit that possesses the hydrox-
ybenzaldehyde functional group. Therefore, the reactive
pair BDP 2 and 3-formyltyrosine should be generally useful
for site specifically fluorescent labeling of proteins.
Acknowledgments Thanks to Dr. Jürgen Schulte for helping
collecting 11B NMR spectra and to Dr. Abhijit Banerjee for providing
the formyltyrosinated tubulin and for assistance with the protein. We
thank Professor Rebecca Kissling for helpful discussions and valuable
scientific assistance. We also thank to David Tuttle for photography.
Financial support from NIH (R01 CA69571 and R15 GM 093941) is
gratefully acknowledged.
Lane
1
2
1
2
Fig. 7 SDS-PAGE of tyrosinated tubulin (Lane 1) and 3-
formyltyrosinated tubulin (Lane 2) after treatment with BDP 2. The
gel was photographed in Panel A under irradiation with long
wavelength UV light and in Panel B after staining with Coumassie
blue
also measured and it was significantly slower than at pH 4.
The data were analyzed in the same manner as for the pH 4
reaction, and the results are shown in Table 2.
References
1. Loudet A, Burgess K (2007) BODIPY dyes and their derivatives:
Syntheses and spectroscopic properties. Chem Rev 107:4891–4932
2. Ziessel R, Ulrich G, Harriman A (2007) The chemistry of Bodipy:
A new El Dorado for fluorescence tools. New J Chem 31:496–501
3. Rohand T, Baruah M, Qin WW, Boens N, Dehaen W (2006)
Functionalisation of fluorescent BODIPY dyes by nucleophilic
substitution. Chem Commun 266–268.
4. Qin WW, Rohand T, Baruah M, Stefan A, Van der Auweraer M,
Dehaen W, Boens N (2006) Solvent-dependent photophysical
properties of borondipyrromethene dyes in solution. Chem Phys
Lett 420:562–568
5. van Swieten PF, Leeuwenburgh MA, Kessler BM, Overkleeft HS
(2005) Bioorthogonal organic chemistry in living cells: novel
strategies for labeling biomolecules. Org Biomol Chem 3:20–27
6. Prescher JA, Bertozzi CR (2005) Chemistry in living systems.
Nature Chem Biol 1:13–21
7. Agard NJ, Baskin JM, Prescher JA, Lo A, Bertozzi CR (2006) A
comparative study of bioorthogonal reactions with azides. ACS
Chem Biol 1:644–648
8. Wang CCY, Seo TS, Li ZM, Ruparel H, Ju JY (2003) Site-specific
fluorescent labeling of DNA using Staudinger ligation. Bioconjug
Chem 14:697–701
9. Smith PAS (1983) Derivatives of Hydrazine and Other Hydro-
nitrogens Having N-N Bonds. The Benjamin/Cummings Publish-
ing Company Inc, Canada
10. Dilek Ö, Bane SL (2008) Synthesis of boron dipyrromethene
fluorescent probes for bioorthogonal labeling. Tetrahedron Lett
49:1413–1416
11. Banerjee A, Panosian TD, Mukherjee K, Ravindra R, Gal S,
Sackett DL, Bane, SL (2010) Site-specific orthogonal labeling of
carboxy terminus of alpha –tubulin. ACS Chem Biol 5 (in press).
12. Ball P, Nicholls CH (1982) Azo-hydrazone tautomerism of
hydroxyazo compounds—a review. Dyes Pigm 3:5–26
13. Shanker N, Bane SL (2008) Basic aspects of absorption and
fluorescence spectroscopy and resonance energy transfer methods.
Methods Cell Biol 84:213–242
14. Williams RC, Lee JC (1982) Preparation of tubulin from brain.
Methods in Enzymol 85 Pt B:376–85.
15. Penefsky HS (1977) Reversible binding of inorganic phosphate by
beef heart mitochondrial adenosine triphosphatase. J Biol Chem
252:2891–2899
Fluorescent Labeling of α-tubulin with BDP 2
The data with the model compounds have shown that the
quantum yield of aromatic hydrazone of BDP 2 is essentially
unaffected by an ortho- hydroxyl substituent, indicating that
a covalent bond between 3-formyltyrosinated tubulin and
BDP 2 will produce a fluorescent product. Furthermore, the
reaction takes place at room temperature and neutral pH in a
reasonable time, conditions that will not normally denature a
protein such as tubulin. Finally, although an aliphatic
aldehyde reacts more quickly with BDP 2 than the aromatic
aldehyde, there is no evidence that the aromatic hydrazone
oxidizes under the reaction conditions.
The unnatural amino acid 3-formyltyrosine was covalently
attached to the carboxy terminus of α-tubulin using the method
recently described [11]. The formyltyrosinated protein and a
tyrosinated control were incubated with BDP 2 and the
samples were subjected to SDS-PAGE. Figure 7a shows the
gel irradiated with long wavelength UV light from a hand-
held lamp. The only fluorescent band in the gel is α-tubulin
in from the formyltyrosinated protein, indicating that covalent
bond formation is formed between 2 and 3-formyltyrosine.
Conclusion
We have shown previously that the hydrazino derivative of
BDP, compound 2, reacts with an unsubstituted benzalde-
hyde, and that the fluorescence of the hydrazone is stronger
and considerably red shifted from that of the hydrazine. We
show here that the presence of an ortho-hydroxy substituent
on benzaldehyde does not adversely alter the spectroscopic
properties of the hydrazone. Moreover, hydrazone forma-
tion between 2 and the model compound salicylaldehyde
16. Qin WW, Leen V, Dehaen W, Cui J, Xu C, Tang XL, Liu WS,
Rohand T, Beljonne D, Van Averbeke B, Clifford JN, Driesen K,
Binnemans K, Van der Auweraer M, Boens N (2009) 3, 5-Dianilino

354
J Fluoresc (2011) 21:347–354
substituted difluoroboron dipyrromethene: Synthesis, spectroscopy,
photophysics, crystal structure, electrochemistry, and quantum-
chemical calculations. J Phys Chem C 113:11731–11740
18. Qin W, Baruah M, Van der Auweraer M, De Schryver FC, Boens
N (2005) Photophysical properties of borondipyrromethene
analogues in solution. J Phys Chem A 109:7371–7384
19. Jencks WP (1964) Mechanism and catalysis of simple carbonyl
group reactions. Prog Phys Org Chem 2:63–128
17. Liras M, Prieto JB, Pintado-Sierra M, Arbeloa FL, Garcia-Moreno
I, Costela AS, Infantes L, Sastre R, Amat-Guerri F (2007)
Synthesis, photophysical properties, and laser behavior of 3-
amino and 3-acetamido BODIPY dyes. Org Lett 9:4183–4186
20. Harej M, Dolenc D (2007) Autoxidation of hydrazones. Some
new insights. J Org Chem 72:7214–7221