Azido-Substituted BODIPY Dyes for the Production of Fluorescent Carbon Nanotubes

Article abstract of DOI:10.1002/chem.201501817

A series of azido-dyes were synthesized through Knoevenagel reactions of an azido-BODIPY with aromatic aldehydes. The nature of the substituents allowed the fine tuning of their spectroscopic properties. The dyes were used to decorate oxidized multiwalled

Full text of DOI:10.1002/chem.201501817

DOI: 10.1002/chem.201501817
Full Paper
&
Fluorescent Probes
Azido-Substituted BODIPY Dyes for the Production of Fluorescent
Carbon Nanotubes
Stefano Fedeli,[a] Paolo Paoli,[b] Alberto Brandi,[a] Lorenzo Venturini,[a]
Giuliano Giambastiani,[c, d] Giulia Tuci,[c] and Stefano Cicchi*[a]
Abstract: A series of azido-dyes were synthesized through
Knoevenagel reactions of an azido-BODIPY with aromatic al-
dehydes. The nature of the substituents allowed the fine
tuning of their spectroscopic properties. The dyes were used
to decorate oxidized multiwalled carbon nanotubes (ox-
MWCNTs), bearing terminal triple bond groups, by CuAAC
reactions, affording fluorescent materials. This decoration al-
lowed the efficient determination of the internalization of
the ox-MWCNT derivatives by different model cancer cells,
such as MCF7.
Introduction
BODIPY dyes is described in which the reactivity of the methyl
groups present on the pyrrole rings is exploited in a series of
Knoevenagel reactions to extend the conjugation of the core
and to obtain derivatives with spectroscopic properties that
were modulated by the nature of the substituents. This syn-
thetic approach is a straightforward way of extending the mo-
lecular conjugation of the BODIPY dye,[7] and is applied on
azido-substituted BODIPY dyes for the first time. The availabili-
ty of a library of BODIPY dyes bearing an azido group adds
a useful alternative for their application in medicinal[8] and ma-
terial chemistry[9] and, in our case, revealed to be a pratical
method for the easy decoration of ox-MWCNTs. The coupling
of properly modified ox-MWCNTs with BODIPY dyes afforded
fluorescent material, and internalization of this material inside
cells was monitored through cytofluorimetry and confocal mi-
croscopy analyses.
BODIPY-type molecules have found a widespread application
as fluorescent probes.[1] The chemical stability of the BODIPY
core goes along with its synthetic versatility.[2] The various re-
views on the subject propose a large number of different
structures.[3] Such structural variety allows the coverage of
a great part of the UV/Vis spectrum, both in absorption and
emission. However, despite their wide applications, BODIPY
probes have been used sparingly with nanostructured carbo-
naceous materials.[4] In our ongoing study for the production
of drug delivery systems based on carbon nanotubes (CNTs),[5]
it was decided to decorate oxidized multiwalled carbon nano-
tubes (ox-MWCNTs), bearing terminal alkyne groups, with a flu-
orescent probe through a simple CuAAC reaction.[6] The choice
for the use of MWCNTs is related, among other more practical
reasons (cost, easy control of oxidation) to their being less ad-
dressed for use in biological systems. For this goal, it was nec-
essary to synthesize simple dyes, bearing an azido group, the
absorption and emission properties of which could be finely
modulated to fulfill the requirements for the detection during
biological tests. Herein, the synthesis of azido-substituted
Results and Discussion
The starting material is represented by the BODIPY dye 5, the
synthesis of which has already been described.[10] Compound 5
is easily obtained from 2,4-dimethylpyrrole (2) and 4-nitroben-
zaldehyde (3). Among the several different procedures that
have been reported, we found more convenient to use the
mechanochemical approach[11] up to the intermediate 3, while
the treatment with BF3·Et2O was performed in dry CH2Cl2.[10]
The reduction of the nitro group with Fe powder/HCl afforded
the corresponding amino group,[12] and the final conversion
into the azido derivative 5 was performed with TMS-N3 and
isoamyl nitrite (Scheme 1).[13] The reactivity of the two methyl
groups in positions 3 and 5 of the BODIPY core allowed to per-
form the Knoevenagel reaction with four different aromatic al-
dehydes, 6a–d, to afford the corresponding mono- and disub-
stituted derivatives (Scheme 1). Previous examples of the Knoe-
venagel reaction on BODIPY derivatives were performed treat-
ing the dye with an excess of aromatic aldehyde in the pres-
[a] Dr. S. Fedeli, Prof. A. Brandi, Dr. L. Venturini, Prof. S. Cicchi
Dipartimento di Chimica “Ugo Schiff”, Università di Firenze
Via della Lastruccia 3–13, 50019, Sesto Fiorentino (Italy)
E-mail: stefano.cicchi@unifi.it
[b] Dr. P. Paoli
Dipartimento di Scienze Biomediche, Sperimentali e Cliniche “Mario Serio”,
Università di Firenze
Viale Morgagni, 50-50134 Firenze (Italy)
[c] Dr. G. Giambastiani, Dr. G. Tuci
Istituto di Chimica dei Composti OrganoMetallici (ICCOM)
Consiglio nazionale delle Ricerche (CNR)
Via Madonna del Piano 10, 50019 Sesto Fiorentino (Italy)
[d] Dr. G. Giambastiani
Kazan Federal University, 420008 Kazan (Russian Federation)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201501817.
Chem. Eur. J. 2015, 21, 15349 – 15353
15349
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 1. Absorption spectra (normalized) of compounds, 5, 7b, 8b, 8c
(DMF), 7d (H2O/DMF), and 7a, 7c (CH2Cl2).
from 500 to 650 nm. For the sake of clarity, the UV spectra of
the dyes were normalized. The molar attenuation coefficients
of the compounds are reported.
Table 1 shows also a remarkable solvatochromic effect. Com-
pound 7b, sufficiently soluble in water, showed a 32 nm red-
shift moving from CH2Cl2 to H2O.
Also the maxima (see Figure 2) in the emission spectra were
scattered, covering a range of 150 nm. The Stokes shift values
are very low, as usually happens with this kind of fluorescent
dyes, ranging from 8 nm (compound 8c) to 18 nm (compound
7c).[3]
Scheme 1. Synthesis of compound 5 and its transformation into azido-sub-
stituted dyes 7a–d and 8b,c.
The presence of the azido group on all the substituents
allows their use in CuAAC reactions with compounds bearing
a terminal triple bond. This reactivity was used for the decora-
tion of ox-MWCNT 9 bearing terminal triple bonds (for the syn-
thesis and characterization, see the Supporting Information,
Figures S1–S7).[16,17] The CuAAC reaction with dyes 5, 7b, and
8c was performed in DMF and afforded the fluorescent ox-
MWCNT derivatives 10–12 (Scheme 2). The decoration of the
MWCNTs afforded fluorescent materials for which absorption
and emission properties can be easily tuned using differently
substituted bodipy dyes. This material was found to be stable,
ence of acetic acid and pyrrolidine in refluxing toluene[14] or in
absolute ethanol.[15]
However, the presence of the azido group on compound 5
made it sensitive to prolonged heating and required a compro-
mise between conversion of the starting material and the final
yields of the adducts. Even after a short reflux in toluene, the
crude reaction material showed the presence of compounds in
which the azido group had been lost. The best reaction condi-
tions revealed to be reflux in MeCN for a very short time,
checking the reaction mixture by TLC. The results are reported
in Scheme 1. The reaction conditions were not optimized for
the synthesis of the double or mono substituted derivatives
but for the obtainment, when possible, of both compounds.
This choice was due to the need to explore the potentiality of
these compounds to cover a large part of the visible spectrum,
in absorption as well as in emission. The reaction with benzal-
dehyde (6a) and 4-HO2C-benzoic acid (6d) afforded exclusively
the doubly substituted compound 7a[13] and 7d, respectively,
while 4-hydroxybenzaldehyde (6b) and pyridine-2-carbalde-
hyde (6c) afforded a mixture of mono- and disubstituted com-
pounds (Scheme 1). The structure of the final compounds were
Table 1. UV max and molar attenuation coefficient e of compounds 5, 7,
and 8.
Compound
UVmax [nm]
solvent
e [LmolÀ1 cmÀ1
]
5
7a
503
626
641
652
673
570
576
624
560
560
633
DMF
7.5”103
1.4”105
6.8”104
6.5”104
8.7”103
9.9”103
8.7”103
3.2”105
2.0”105
1.1”105
3.7”103
CH2Cl2
CH2Cl2
DMF
H2O
CH2Cl2
DMF
CH2Cl2
CH2Cl2
DMF
7b
1
8b
7c
8c
7d
easily determined by H NMR spectroscopy, showing the sig-
nals of the AX system related to the newly formed E double
bonds.
In line with our expectation, the absorption spectra
(Figure 1) covered a wide part of the visible spectrum ranging
H2O/DMF
Chem. Eur. J. 2015, 21, 15349 – 15353
www.chemeurj.org
15350
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Emission spectra (normalized) of compounds 5, 7b, 8b, 8c (DMF),
7d (H2O/DMF), and 7a, 7c (CH2Cl2).
Figure 3. Absorption spectra of compounds 10–12. All spectra were ac-
quired in DMF (0.5 mgmLÀ1).
Figure 4. Emission spectra (normalized) of compounds 10–12. All spectra
were acquired in DMF (0.5 mgmLÀ1).
Scheme 2. The CuAAc reaction between CNT 9 and BODIPY dyes 5, 7b, and
8c.
A main issue, working with CNTs, is the determination of
their functionalization degree obtained after a reaction. Very
often this information is obtained through elemental or ther-
mogravimetric analysis.
as after several months, no apparent degradation of the fluo-
rescence was observed.
Figures 3 and 4 show the absorption and emission spectra
of compounds 10–12. The decoration of the CNT walls did not
alter significantly the absorption and the emission maxima of
the BODIPY derivatives. This is evident when comparing the
UV and fluorescence maxima of compounds 5 and 7b with
those of 10 and 12, respectively: the wavelengths of absorp-
tion and emission (measured in DMF) are the same for each
sample. This might be rationalized by considering that the flu-
orescent probe is efficiently solvated by DMF reducing to
a minimum the interaction with the CNTs. On the contrary,
compound 8c shifted its UV and fluorescence maxima by 11
and 22 nm, respectively, when anchored to ox-MWCNTs to give
compound 11. This might be due to the presence of a pyridine
ring on the dye and its interaction with the carboxylic groups
present on the ox-MWCNTs.
The presence of the BODIPY dyes, bearing a boron atom, al-
lowed the use of inductively coupled plasma atomic emission
spectroscopy (ICP-AES) for a quantitative determination of the
functionalization degree of decorated ox-MWCNTs. In the case
of compound 10, the analysis revealed a presence of 12.4 mg
of Boron per gram of material (1.14 mmolgÀ1 of material)
while for compound 11 the analysis revealed the presence of
6.2 mg per gram of material (0.57 mmolgÀ1 of material). The
confirmation of the capacity of the fluorescent probes to show
the internalization of CNTs inside cancer cells was achieved by
a series of cytofluorimetric analyses performed on three differ-
ent cancer cell lines : MCF7 (human breast adenocarcinoma),
PC3 (prostatic small cell carcinoma), and HT29 (human colon
cancer) after incubation with compound 10, 11, and 12
(Figure 4 and Table 2).
Chem. Eur. J. 2015, 21, 15349 – 15353
www.chemeurj.org
15351
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
showed a threefold value of the final fluorescence. Since the
functionalization degree of the compound used is the same, it
is possible to correlate, at least qualitatively, the degree of in-
ternalization with the enhancement of fluorescence.
Table 2. Fluorescence data from cytofluorimetry analysis after incubation
of three different cell lines with compounds 10–12.[a]
10
11
12
Cell line
Control
Treated
Control
Treated
Control
Treated
Finally, compound 11 was used in a confocal microscopy
analysis performed on MCF7 cells after incubation (Figure 6).
HT29
PC3
MCF7
136
93
180
597
531
2650
344
247
275
4190
4879
1380
69
158
275
1350
1260
2900
[a] Each cell culture was incubated for 90 min with a water dispersion of
the decorated nanotubes 10–11 (10 mgmLÀ1). Data reported in the table
represent the mean values of the fluorescence of the cells.
Figure 5 shows the results of three cytofluorimetric analyses
performed with compounds 10–12 on MCF7 cancer cell lines.
The Figure shows the remarkable increase of fluorescence of
the cells, which is directly related to the internalization of the
CNTs. Table 2 summarizes all of the cytofluorimetric data col-
lected for the three different cancer cell lines. In all cases there
is a net increase of the cellular fluorescence, demonstrating
that the decorated carbon nanotubes are efficiently internal-
ized by different cell lines.[18] While it is not possible to com-
pare the internalization degree of the different compounds
inside one kind of cell cultures, as any enhancement or
quenching effect might be different, it is possible to compare
the difference between the internalization degrees of the same
compound inside different kind of cells. For example, MCF7
cells showed a much higher fluorescence when incubated with
10 respect to HT29 and PC3 (five times the final fluorescence
value). The opposite holds for 12 for which HT29 and PC3 cells
Figure 6. Confocal microscopy analysis of a MCF7 cell culture after incuba-
tion with MWCNT 11.
Compound 11 was chosen since it best fitted the excitation
frequency (571 nm) of the laser used in the confocal microsco-
py analysis. Despite the fact that MCF7 cells were those that
evidenced the smaller increase in fluorescence upon incuba-
tion with compound 11, the confocal analysis afforded a clear
picture (Figure 6): the fluorescence of compound 11 is well dis-
tributed in the cytoplasm of the cells while the nuclei remain
unaffected. Concerning the amount of material internalized,
during the biological tests it was not possible to determine
any difference between the cell cultures incubated with simple
Ox-MWCNTs and those incubated with compound 11.
To verify any possible toxic effect of the nanostructured ma-
terials, MTT assays were performed and the results are given in
the Supporting Information, Figure S8: neither simple ox-
MWCNTs nor compound 11 show any sign of cytotoxicity to-
wards MCF7 cell lines using a concentration of 10 mgmLÀ1 of
the material.
Conclusion
A series of Knoevenagel reactions of an azido-bodipy deriva-
tive allowed the production of a small library of azido-substi-
tuted fluorescent dyes, the spectroscopic properties of which
are tuned by the nature of their substituents. The presence of
an azido group on the fluorescent dyes allowed their easy an-
choring onto properly substituted ox-MWCNTs. This affords flu-
orescently labeled nanostructured carbon materials that can
be internalized inside cells after a brief incubation time. Fur-
Figure 5. Cytofluorimetric analysis of MCF7 cell lines after incubation with
fluorescent CNTs: a) control experiment, b) incubation with 10, c) incubation
with 11, d) incubation with 12.
Chem. Eur. J. 2015, 21, 15349 – 15353
www.chemeurj.org
15352
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Y. Cakmak, G. Barin, S. Erten-Ela, M. Marszalek, J.-H. Yum, S. M. Zakeerud-
din, M. K. Nazeeruddin, M. Graetzel, E. U. Akkaya, Chem. Sci. 2011, 2,
949–954; c) N. Boens, W, Qin, M. Baruah, W. M. De Borggraeve, A. Fi-
larowski, N. Smisdom, M. Ameloot, L. Crovetto, E. M. Talavera, J. M. Al-
varez-Pez, Chem. Eur. J. 2011, 17, 10924–10934; d) Z. Kostereli, T. Ozde-
mir, O. Buyukcakir, E. U. Akkaya, Org. Lett. 2012, 14, 3636–3639; e) S.
Zhu, J. Zhang, G. Vegesna, A. Tiwari, F.-T. Luo, M. Zeller, R. Luck, H. Li, S.
Green, H. Liu, RSC Adv. 2012, 2, 404–407; f) E. Palao, A. R. Agarrabeitia,
J. BaÇuelos-Prieto, T. Arbeloa Lopez, I. Lopez-Arbeloa, D. Armesto, M. J.
Ortiz, Org. Lett. 2013, 15, 4454–4457.
ther studies for use of these materials as platform for the selec-
tive delivery of drugs into cancer cells are ongoing in our labo-
ratories.
Acknowledgments
A.B., S.C., and S.F. gratefully acknowledge support from the
Italian Ministry of Education, University and Research (MIUR—
Rome): grant FIRB RBAP11ETKA_002 Approcci nanotecnologici
per la teragnostica dei tumori.
[8] a) E. Lallana, A. Sousa-Herves, F. Fernandez-Trillo, R. Riguera, E. Fernan-
dez-Megia, Pharm. Res. 2012, 29, 1–34; b) J. Hou, X. Liu, J. Shen, G.
Zhao, P. G. Wang, Expert Opin. Drug Discovery 2012, 7, 489–501.
[9] S. Campidelli, Curr. Org. Chem. 2011, 15, 1151–1159.
[10] M. Yu, J. K.-H. Wong, C. Tang, P. Turner, M. H. Todd, P. J. Rutledge, Beil-
stein J. Org. Chem. 2015, 11, 37–41.
Keywords: BODIPY · carbon nanotubes · click reaction ·
[11] L. P. Jameson, S. V. Dzyuba, Beilstein J. Org. Chem. 2013, 9, 786–790.
[12] D. Yim, H. Yoon, C.-H. Lee, W.-D. Jang, Chem. Commun. 2014, 50,
12352–12355.
[13] M. Di Donato, A. Iagatti, A. Lapini, P. Foggi, S. Cicchi, L. Lascialfari, S.
Fedeli, S. Caprasecca, B. Mennucci, J. Phys. Chem. C 2014, 118, 23476–
23486.
fluorescent probes
[1] N. Boens, V. Leen, W. Dehaen, Chem. Soc. Rev. 2012, 41, 1130–1172.
[2] G. Ulrich, R. Ziessel, A. Harriman, Angew. Chem. Int. Ed. 2008, 47, 1184–
1201; Angew. Chem. 2008, 120, 1202–1219.
[3] a) A. Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891–4932; b) M. Ven-
drell, D. Zhai, J. C. Er, Y. T. Chang, Chem. Rev. 2012, 112, 4391–4420;
c) A. Bessette, G. Hanan, Chem. Soc. Rev. 2014, 43, 3342–3405.
[4] a) S. Erbas, A. Gorgulu, M. Kocakusakogullari, E. U. Akkaya, Chem.
Commun. 2009, 4956–4958; b) S. H. Yoshimura, S. Khan, H. Maruyama,
Y. Nakayama, K. Takeyasu, Biomacromolecules 2011, 12, 1200–1204; c) K.
Flavin, K. Lawrence, J. Bartelmess, M. Tasior, C. Navio, C. Bittencourt,
D. F. O’Shea, D. M. Guldi, S. Giordani, ACS Nano 2011, 5, 1198–1206;
d) J. Bartelmess, E. D. Luca, A. Signorelli, M. Baldrighi, M. Becce, R. Bres-
cia, V. Nardone, E. Parisini, L. Echegoyen, P. P. Pompa, S. Giordani, Nano-
scale 2014, 6, 13761–13769.
[5] a) G. Tuci, C. Vinattieri, L. Luconi, M. Ceppatelli, S. Cicchi, A. Brandi, J. Fil-
ippi, M. Melucci, G. Giambastiani, Chem. Eur. J. 2012, 18, 8454–8463;
b) G. Tuci, L. Luconi, A. Rossin, F. Baldini, S. Cicchi, S. Tombelli, C. Trono,
A. Giannetti, I. Manet, S. Fedeli, A. Brandi, G. Giambastiani, ChemPlu-
sChem 2015, 80, 704–714.
[14] T. Uppal, N. V. S. Dinesh, K. Bhupathiraju, M. GraÅa, H. Vicente, Tetrahe-
dron 2013, 69, 4687–4693.
[15] J.-S. Lee, N.-y. Kang, Y. K. Kim, A. Samanta, S. Feng, H. K. Kim, M. Ven-
drell, J. H. Park, Y.-T. Chang, J. Am. Chem. Soc. 2009, 131, 10077–10082.
[16] a) S. Campidelli, B. Ballesteros, A. Filoramo, D. D. Díaz, G. de La Torre, T.
Torres, G. M. A. Rahman, C. Ehli, D. Kiessling, F. Werner, et al., J. Am.
Chem. Soc. 2008, 130, 11503–11509; b) K. Flavin, M. N. Chaur, L. Eche-
goyen, S. Giordani, Org. Lett. 2010, 12, 840–843.
[17] M. S. P. Shaffer, X. Fan, A. H. Windle, Carbon 1998, 36, 1603–1612.
[18] K. Kostarelos, L. Lacerda, G. Pastorin, W. Wu, S. Wieckowski, J. Luangsivi-
lay, S. Godefroy, D. Pantarotto, J.-P. Briand, S. Muller, M. Prato, A. Bianco,
Nat. Nanotechnol. 2007, 2, 108–113.
Received: May 8, 2015
[6] J. E. Hein, V. V. Fokin, Chem. Soc. Rev. 2010, 39, 1302–1315.
[7] a) R. Ziessel, G. Ulrich, A. Harriman, M. A. H. Alamiry, B. Stewart, P. Retail-
leau, Chem. Eur. J. 2009, 15, 1359–1369; b) S. Kolemen, O. A. Bozdemir,
Revised: July 15, 2015
Published online on August 24, 2015
Chem. Eur. J. 2015, 21, 15349 – 15353
www.chemeurj.org
15353
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim