Near IR excitation of heavy atom free Bodipy photosensitizers through the intermediacy of upconverting nanoparticles

Article abstract of DOI:10.1039/c4cc03387f

Orthogonal dimers of Bodipy were recently shown to be efficient generators of singlet oxygen. However, these dyes require green light for excitation, which would be very quickly attenuated inside the mammalian tissues. We now demonstrate that when these d

Full text of DOI:10.1039/c4cc03387f

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Near IR excitation of heavy atom free Bodipy
photosensitizers through the intermediacy of
upconverting nanoparticles†
Seda Demirel Topel,a Gunseli Turgut Cina and Engin U. Akkaya*bc
Cite this: Chem. Commun., 2014,
50, 8896
Received 6th May 2014,
Accepted 14th June 2014
DOI: 10.1039/c4cc03387f
www.rsc.org/chemcomm
Orthogonal dimers of Bodipy were recently shown to be efficient Over the years, a number of classes of compounds were proposed
generators of singlet oxygen. However, these dyes require green as PDT photosensitizers, and within the last few years Bodipy dyes
light for excitation, which would be very quickly attenuated inside have been considered as viable alternatives to established examples.7
the mammalian tissues. We now demonstrate that when these dyes Although the parent dye absorbs in the middle of the visible
are covalently attached to UCNPs, near IR irradiation results in very region, it is possible to move the absorption band up to the
efficient generation of singlet oxygen.
near IR region with straightforward transformations.8
Parent (unmodified) Bodipy dyes typically are not effective
Recent years witnessed exponentially growing interest in a class of photosensitizers. As a general rule, heavy atom substitution on
compounds known as Bodipy dyes (a.k.a., boradiazaindacenes or the p-system results in enhanced intersystem crossing.9 However,
borondipyrrin dyes).1 It is remarkable that every single member of the incorporation of heavy atoms for photosensitizers geared for
the tricyclic boradiazaindacene ring system, including the compo- photodynamic therapy would not be the optimal choice, because
nents of the difluoroboron bridge, can be derivatized, substituted while the sensitization of molecular oxygen is very likely with
or otherwise functionalized in a straightforward manner, yielding this approach, increased dark toxicity is typically an undesired
new spectral features, including red shifted absorption and outcome. Dark toxicity limits the inherent selectivity of photo-
emission,2 or large intersystem crossing rates.3 The latter is of dynamic therapy agents, which rests on the fact that the PDT
particular importance if these dyes are to be proposed as potential agent is expected to be non-toxic unless/until it is at the right
singlet oxygen generators in a photodynamic therapeutic regime. location for absorbing the light of excitation.
Photodynamic therapy (PDT) is a non-invasive treatment modality
In our earlier work,10 followed by computational studies,11
for certain cancers and non-cancerous disease states.4 The treatment and later corroborated by Zhang,12 it was demonstrated that
is based on the cytotoxicity of singlet oxygen at the region where it is orthogonal-Bodipy dyes, whether connected in a 8,2- or a
generated, following sensitization of the dissolved oxygen by a dye 8,8-fashion, resulted in a very peculiar first excited state with
with high intersystem crossing efficiency. There is an important tetraradicalic character and have very high propensity to undergo
caveat however; the dye has to have a strong absorption in the red intersystem crossing. This finding was noteworthy, because in
or the near IR region of the spectrum, otherwise the excitation of principle, the design is applicable to essentially all other chromo-
the dye will be highly ineffective beyond the first few millimeters phores, offering an alternative to heavy atom substitution as a
of the tissues.5 This is due to the fact that mammalian tissues safe bet for enhanced intersystem crossing (another approach
are highly scattering and absorbing for light with wavelengths for heavy-atom free photosensitization was developed by Zhao
shorter than 630 nm.6
and co-workers13). While the detailed photophysical experimenta-
Historically, porphyrins and derivatives were considered as tion eliminating other possibilities, such as singlet fission, is
the photosensitizers of choice for this purpose, but the extinc- awaited, one limitation expressed in the previous work10 was that
tion coefficients in the therapeutic window are not very high. the dimeric Bodipy dyes needed excitation at around 500 nm,
however for true photodynamic therapy potential, the excitation
a Department of Chemistry, Faculty of Science, Akdeniz University, 07058, Antalya,
should be done in the therapeutic window, i.e., between 650 and
1000 nm. With Bodipy dyes, conjugation extension by styryl
modification has been shown to result in bathochromic shifts
in the absorption band, but unfortunately, computational and
experimental work showed that the excited state loses its tetra-
radicalic character in the longer wavelength absorbing derivatives
Turkey
b UNAM-National Nanotechnology Research Center, Bilkent University, 06800,
Ankara, Turkey. E-mail: eua@fen.bilkent.edu.tr
c Department of Chemistry, Bilkent University, 06800, Ankara, Turkey
† Electronic supplementary information (ESI) available: Synthesis procedures
and additional spectroscopic data. See DOI: 10.1039/c4cc03387f
8896 | Chem. Commun., 2014, 50, 8896--8899
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
of the dimeric chromophores, becoming much less efficient visible region. Compound 1 shows a major peak at 638 nm, red shifted
singlet oxygen generators. due to dimethylaminostyryl conjugation. Dimers 2 and 3 show a major
In consideration of this, we investigated the possibility of using band peaking at 508 and 509 nm, respectively, corresponding to S0–S1
upconverting nanoparticles (UCNPs) as a solution to this problem. transitions and the bands in the visible region are somewhat
UCNPs can be excited with near IR irradiation and depending on broadened. The peak shapes and positions are not affected by
the exact composition, emission at shorter wavelength regions of concentration changes, eliminating the possibility of any aggrega-
the spectrum is obtained. UCNPs have these remarkable properties tion induced changes for spectral differences.
due to the presence of trivalent lanthanide ions as dopants (as
There are many successful techniques for the preparation of
absorbers and emitters) in an inorganic matrix.14 This combination lanthanide-doped upconversion luminescence NPs, such as
provides a multitude of metastable states with relatively long coprecipitation18 and thermal decomposition.19 However, these
lifetimes. The main difference between two-photon absorbance methods involve organic solvent or high temperatures and usually
(TPA) and upconversion is that, in upconversion, the virtual require heat after treatment or produce toxic byproducts.20 There-
states of TPA are replaced by actual states, which are long-lived fore, in this work, we preferred to prepare upconverting nano-
enough to be excited once more, under intense irradiation. The particles by the solvothermal method21 which employs milder
typical emitters are erbium, thulium and holmium ions due to conditions and water based reaction medium leading to the
the availability of equally spaced energy levels which would be formation of polyethyleneimine (PEI) covered nanoparticles for
optimal for absorption and energy transfer processes.15
further functionalization with BODIPY derivatives. PEI is bio-
The first goal of the present work was to synthesize appro- compatible22 and PEI-nanoparticles are soluble in aqueous
priately derivatized orthogonal Bodipy dimers (Scheme 1). First of all, systems. It has been reported that free PEI can be toxic to cells,
a positive control compound 1 was synthesized (ESI†). This com- but when it is in particulate form, the hazardous effects are greatly
pound has two iodo substituents on the Bodipy core, ensuring good decreased.23 When used for DNA delivery PEI was reportedly non-
singlet oxygen generation upon excitation in the presence of dissolved toxic at concentrations used in that study.24 In principle, the
molecular oxygen. Nagano16 and others17 demonstrated that such functional groups or water solubility conferring groups can be
halogenated Bodipy compounds are indeed efficient photosensitizers. incorporated during synthesis or can be placed by post-synthetic
As for the dimeric compounds, we have previously reported the modifications. We opted for in situ preparation which includes
synthesis of the dimeric parent compounds,10 but for covalent attach- the addition of polyethyleneimine (PEI) groups during the nano-
ment, additional functionalization is needed. In the orthogonal particle synthesis. The synthesis procedure of the upconverting
dimer 2, the carboxylate anchor is placed on the meso-phenyl sub- nanoparticles is presented in detail in ESI.† The size and
stituent. Formylation and the construction of the second Bodipy core morphology of the UCNPs were examined by TEM which shows
in the usual manner yield the desired compound. The other dimer 3 an average size of 35 nm and this size distribution was also
carries the carboxylate in the 2-position of a Bodipy core, which is easily determined by DLS measurements. The TEM-EDAX spectrum
placed by a reaction with acrylic acid under palladium(II) catalysis shows that the rare earth cations were successfully doped into
(ESI†). Absorption spectra of the Bodipy compounds show peaks in the the UCNPs (Fig. 1). When excited using a near IR laser at 980 nm
Fig. 1 TEM images of PEI-coated NaYF4:(Yb+3, Er+3)/Ce+3 nanoparticles
(A–D), (E) SAED (selected area electron diffraction) pattern, (F) histogram of
the particle size distribution from DLS measurements, (G) emission from
Scheme 1 Structures of the carboxy-functionalized sensitizers: styryl-
Bodipy 1, orthogonal Bodipy dimers 2, and 3.
UCNP solution in methanol (1.0 mg/1.0 mL) upon excitation with a 980 nm
laser, (H) TEM-EDAX spectrum of the PEI-coated UCNP.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 8896--8899 | 8897

View Article Online
ChemComm
Communication
(fluence rate 50 mW cmÀ2), green emission from a solution the dyes is short, efficient energy transfer is to be expected upon
containing these nanoparticles is clearly observed which is excitation of the UCNPs using a 980 nm laser (ESI,† Fig. S38).
composed of two main peaks at 540 and 654 nm (Fig. S28, ESI,†
Finally, we wanted to demonstrate the efficiency of singlet oxygen
and Fig. 1G). Further characterization at this stage was done by generation with specifically designed Bodipy dyes attached covalently
acquiring XPS and XRD data (Fig. S26 and S25, ESI,† respectively). to the UCNPs. The generation of singlet oxygen was monitored
XPS shows a particular peak at 402 eV, clearly belonging to N (1s) by using 1,3-diphenylisobenzofuran (DPBF) as a trap.25 When
which corroborates that UCNP was covered by PEI in situ. The XRD DPBF reacts with singlet oxygen, it is initially converted to its
spectrum also shows that UCNPs are in highly crystalline form endoperoxide form, and then further transformed through ring
including both hexagonal and cubic phases (Fig. S25, ESI†).
opening into 1,2-dibenzoylbenzene, resulting in a decrease of
For UCNP functionalization, the photosensitizers have to be the absorbance peak of DPBF. Therefore, the singlet oxygen
prepared in an appropriately derivatized form. Orthogonal Bodipy production rate can be assessed following the change in the
dimer and styryl-Bodipy functionalization is shown in Scheme S1 absorption of DPBF at the wavelength of absorption maximum
(ESI†). Detailed synthesis procedures are provided, here it will suffice (B410 nm) (Fig. S30–S32, ESI†).
to say that orthogonal Bodipy dimers with such remarkable excited
The procedure we followed for studying the relative rates of
state properties proved to be amenable to modifications leading singlet oxygen generation with functionalized UCNPs (Fig. S39,
to derivatives with conjugatable side groups. The next step was to ESI†) is as follows: all BODIPY conjugated UCNPs were exposed
conjugate these dyes onto the PEI coated UCNPs. This step pro- to a NIR laser (50 mW cmÀ2) for 36 min (following 9 min dark
ceeded smoothly, leading to sensitizer dye functionalized UCNPs incubation to eliminate any source for absorbance change due
(ESI,† Fig. S37) in the presence of DCC/DMAP coupling reagents.
to a dark reaction). As expected, the adsorption of DPBF in the
BODIPY conjugated nanoparticles obtained in this way were first iodinated BOD1-UCNP (B26%) showed a decrease, but dimeric
characterized physically. Following attachment of the BODIPY dyes BOD2-UCNP showed a rapid decrease (B36%) during irradia-
to the UCNPs, no appreciable changes in the size were observed tion. The smallest change in the absorbance was observed in
based on TEM analysis and also the TEM image (Fig. 2B) shows that BOD3-UCNP (B23%). These results show that Bodipy dyes were
there are fringes on the UCNPs, which prove that crystallinity is effectively sensitized by the energy transfer from UCNPs. It is
preserved, this also corroborated with the SAED image (Fig. 2C). The surprising that even a minimal overlap between the UCNP
EF-TEM (Energy Filtered-TEM) map shows (Fig. 2, bottom three emission and the Bodipy emission in BOD2-UCNP is sufficient
pictures) a halo of boron nuclei around the nanoparticles following for effective sensitization. This is probably due to large extinc-
the conjugation procedure.
tion coefficients of the orthogonal Bodipy dyes and short
On the other hand, the FTIR spectrum also supports covalent distance between the dye and the particles.
modification: there are new peaks at 1692–1686 cmÀ1 (1st amide
In order to eliminate other possibilities for singlet oxygen genera-
bond, n(CQO)), 1665–1656 cmÀ1 (2nd amide bond, n(N–H)) and tion, control experiments were carried out as follows: unmodified
1428–1422 cmÀ1 (3rd amide bond, n(C–N)) attesting to the formation UCNPs and Bodipy dyes alone were subjected to near IR irradiation
of the amide functionality between BODIPY dyes and UCNP. Before (980 nm) separately, no singlet oxygen generation (Fig. S33 and S34
the conjugation reaction (i.e., PEI-covered UCNP) the nanoparticles ESI†) was observed under either of these conditions.
show bands corresponding to the internal vibration of NH2
bands at 1630 and 1420 cmÀ1 and CH2 stretching at 2930 and conjugated orthogonal Bodipy dimers which require excitation at
2850 cmÀ1 (Fig. S27, ESI†).
around 500 nm were successfully excited by a much more tissue
In summary, we were able to show conclusively that UCNP-
The emission wavelength of UCNPs matched with the penetrating 980 nm laser through the intermediacy of the upcon-
absorption wavelength of BODIPY dyes resulting in reasonable version process. Thus, Bodipy dimers with no heavy atoms can be
spectral overlap, and since the distance between the core and considered as potential photosensitizers with practical utility
when used in conjunction with UCNPs. Work towards making
biocompatible nanoparticles with sensitizer functionality, aiming
this particular goal, is in progress in our laboratory.
The authors gratefully acknowledge support from TUBITAK.
The authors acknowledge the Akdeniz University Coordination
Unit of Scientific Research Projects (project No. 2011.03.0121.017)
for their financial support.
Notes and references
1 (a) A. Kamkaew, S. H. Lim, H. B. Lee, L. V. Kiew, L. Y. Chung and
K. Burgess, Chem. Soc. Rev., 2013, 42, 77; (b) S. G. Awuah and Y. You,
RSC Adv., 2012, 2, 11169; (c) G. Ulrich, R. Ziessel and A. Harriman,
Angew. Chem., Int. Ed., 2008, 47, 1184.
2 (a) T. Mizuno, K. Umezawa, Y. Shondo, D. Citterio, K. Oka and
Fig. 2 TEM images (A, B), SAED (Selected Area Electron Diffraction) image (C)
K. Suzuki, J. Fluoresc., 2013, 23, 1007; (b) A. Poirel, A. De Nicola and
and EF-TEM boron mapping of BOD1-UCNP, BOD2-UCNP and BOD3-UCNP,
R. Ziessel, J. Org. Chem., 2011, 76, 4489; (c) S. Kolemen, Y. Cakmak,
left to right. Orange halo indicates boron nuclei (bottom three pictures).
S. Erten Ela, Y. Altay, J. Brendel, M. Thelakkat and E. U. Akkaya,
8898 | Chem. Commun., 2014, 50, 8896--8899
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Org. Lett., 2010, 12, 3812; (d) Z. Dost, S. Atilgan and E. U. Akkaya, 14 (a) F. Wang, D. Banerjee, Y. Liu, X. Chen and X. Liu, Analyst, 2010,
¨
Tetrahedron, 2006, 62, 8484.
135, 1839; (b) H. Schafer and M. Haase, Angew. Chem., Int. Ed., 2011,
3 M. J. Ortiz, I. Garcia Moreno, A. R. Agarrabeita, G. Duran Sampedro,
50, 5808.
A. Costela, R. Sastre, F. Lopez Arbeloa, J. Banuelos Prietoand and 15 H. X. Mai, Y. W. Zhang, R. Si, Z. G. Yan, L. D. Sun, L. P. You and
I. Lopez Arbeloa, Phys. Chem. Chem. Phys., 2010, 12, 7804. C. H. Yan, J. Am. Chem. Soc., 2006, 128, 6426.
4 (a) D. E. J. G. J. Dolmans, D. Fukumura and R. K. Jain, Nat. 16 T. Yogo, Y. Urano, Y. Ishitsuka, F. Maniwa and T. Nagano, J. Am.
Rev. Cancer, 2003, 3, 380; (b) Z. Huang, Technol. Cancer Res. Treat.,
2005, 4, 283.
5 K. Plaetzer, B. Krammer, J. Berlanda, F. Berr and T. Kiesslich, Lasers
Med. Sci., 2009, 24, 259.
6 T. Vo-Dinh, Editor-in-Chief, Biomedical Photonics Handbook, CRC
Press, Boca Raton, Florida, 2003.
Chem. Soc., 2005, 127, 12162.
17 (a) C. S. Zhang, J. Z. Zhao, S. Wu, Z. L. Wang, W. H. Wu, J. Ma, S. Guo
and L. Huang, J. Am. Chem. Soc., 2013, 135, 10566–10578;
(b) S. Atilgan, Z. Ekmekci, A. L. Dogan, D. Guc and E. U. Akkaya,
Chem. Commun., 2006, 4398.
¨
18 (a) S. Heer, K. Kompe and H. U. Gudel, Adv. Mater., 2004, 16, 2102;
7 (a) Y. Cakmak, T. Nalbantoglu, T. Durgut and E. U. Akkaya, Tetrahedron
(b) J. H. Zeng, J. Su, Z. H. Li, X. Ja and Y. D. Li, Adv. Mater., 2005, 2119.
Lett., 2014, 5, 538; (b) S. Erbas Cakmak and E. U. Akkaya, Angew. Chem., 19 (a) J. C. Boyer, F. Vetrone, L. A. Cuccia and J. A. Capobianeo, J. Am.
Int. Ed., 2013, 52, 11364; (c) S. Erbas, M. Kocakusakogullari and
E. U. Akkaya, Chem. Commun., 2009, 4956.
8 S. Ozlem and E. U. Akkaya, J. Am. Chem. Soc., 2009, 131, 48.
Chem. Soc., 2006, 128, 7444; (b) H. X. Mail, Y. W. Zhang and R. Si,
J. Am. Chem. Soc., 2006, 128, 6426.
20 F. Wang and X. Liu, Chem. Soc. Rev., 2009, 38, 976.
9 N. Adarsh, R. R. Avirah and D. Ramaiah, Org. Lett., 2010, 21 (a) X. F. Yu, Z. Sun, M. Li, Y. Xiang, Q. Q. Wang, F. Tang, Y. Wu,
12, 5720.
Z. Cao and W. Li, Biomaterials, 2010, 31, 8724; (b) X. Yu, M. Li,
M. Xie, L. Chen and Q. Wang, Nano Res., 2010, 3, 51.
10 Y. Cakmak, S. Kolemen, S. Duman, Y. Dede, Y. Dolen, B. Kilic,
Z. Kostereli, L. Tatar Yildirim, A. L. Dogan, D. Guc and E. U. Akkaya, 22 F. Wang, D. K. Chatterjee, Z. Li, Y. Zhang, X. Fan and M. Wang,
Angew. Chem., Int. Ed., 2011, 50, 11937. Nanotechnology, 2006, 17, 5786.
11 S. Duman, Y. Cakmak, S. Kolemen, E. U. Akkaya and Y. Dede, J. Org. 23 W. T. Godbey, K. K. Wu and A. Mikos, J. Controlled Release, 1999, 60, 149.
Chem., 2012, 77, 4516.
12 X. F. Zhang and X. Yang, J. Phys. Chem. B, 2013, 117, 9050.
24 R. C. Lambert, Y. Maulet, J. L. Dupont, S. Mykita, P. Craig, S. Volsen
and A. Feltz, Mol. Cell. Neurosci., 1996, 7, 239.
13 L. Huang, X. R. Yui, W. H. Wu and J. Z. Zhao, Org. Lett., 2012, 25 S. Erbas Cakmak, O. A. Bozdemir, Y. Cakmak and E. U. Akkaya,
14, 2594.
Chem. Sci., 2013, 4, 858.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 8896--8899 | 8899