Multi-tasking with single platform dendrimers for targeting sub-cellular microenvironments

Article abstract of DOI:10.1002/chem.201000241

Double trouble! Orthogonally functionalised dendrimers have been constructed by using a versatile approach that is widely applicable for therapeutics and diagnostics by combining imaging agents and therapeutics for synergistic effects. Dendrimers containing a covalently linked dipyrromethene boron difluoride (BODIPY) dye and α-lipoic acid perform complementary biological functions and target intracellular lipid droplets (see figure). (Figure Presented)

Full text of DOI:10.1002/chem.201000241

DOI: 10.1002/chem.201000241
Multi-tasking with Single Platform Dendrimers for Targeting Sub-Cellular
Microenvironments
Rami Hourani,[a] Manasi Jain,[b] Dusica Maysinger,*[b] and Ashok Kakkar*[a]
The fabrication of multi-functional biocompatible materi-
als with a high degree of complexity at the molecular level
constitutes the key to the future of biotechnology.[1] Much
remains to be done to develop efficient physiologically or
therapeutically active nanostructures that can not only trans-
port biologically active cargo across cell membranes, but
also ensure delivery to specific intracellular locations.[2] Mac-
romolecular architectures with well-defined size and shape
are of eminent interest for biomedical applications, such as
the delivery of therapeutics and tissue imaging.[3–7] Dendrim-
ers are monodisperse and hyperbranched macromolecules
with unique architecture and properties that can be tailored
for specific applications.[8] One of the key challenges in the
development of dendrimers for applications in medicine is
to incorporate complementary biological tasks, including
imaging and therapeutic delivery, in the same dendrimer ar-
chitecture.[9] However, introducing multiple functionalities
into a single dendrimer platform has presented significant
synthetic challenges.[10,11] Another important issue related to
the biomedical applications of dendrimers pertains to their
cytotoxicity,[12,13] and the need for versatile, clean and high
yielding chemistry with essentially no side products, contin-
ues to be the key tenet and a topical area of research.[14] We
demonstrate herein that orthogonally functionalised den-
drimers with diverse groups such as a fluorescent dye, useful
for organelle imaging, and a therapeutic agent can be easily
constructed by using a “click” chemistry[15–17] approach.
These multi-tasking dendrimers show no significant cytotox-
icity, and represent the first example of the simultaneous di-
rection of a drug and a fluorophore to lipid droplets.
Lipid droplets play a crucial role in lipid metabolism and
might have a much broader repertoire of “tasks”, such as
managing the availability of proteins and serving as generic
sites of protein sequestration.[18] a-Lipoic acid (LA) is an es-
sential cofactor for many enzymes, particularly in aerobic
metabolism.[19,20] It is readily taken up by cells and reduced
to its potent dithiol form, dihydrolipoate, much of which is
rapidly effluxed out from cells. We demonstrate that by link-
ing LA to a dendrimer, its intracellular retention could be
increased, thereby providing an enhanced effectiveness with
smaller therapeutic doses. Dipyrromethene boron difluoride
(BODIPY) PM605, abbreviated here as PM, is an efficient
lipophilic dye that has characteristic narrow absorption and
emission bands in the visible region of the spectrum with
high fluorescence quantum yields, and it readily diffuses
into cells and accumulates in lipid bodies.
The synthetically tailored linkage of PM and LA to the
same assembly was made possible by the design and synthe-
sis of an ABB’ building block that facilitates performing
CuI-catalysed click reactions in a sequential manner, and
eventually the construction of multi-valent dendrimers. It
contains two acetylenes (BB’) protected with two different
groups (trimethylsilyl (TMS) and triisopropylsilyl (TIPS)),
which are individually removed when desired. It was con-
structed by starting from 3-bromo-5-iodobenzylalcohol, and
performing Sonogashira coupling to introduce triisopropyla-
cetylene at the iodo position at room temperature
(Scheme 1), and the substitution of bromo group with trime-
thylsilylacetylene under reflux. It was subsequently followed
by bromination of the benzylalcohol, and a mild deprotec-
tion procedure to eliminate the trimethylsilyl group. The re-
sulting compound (4) is a highly versatile unit that can carry
multiple functionalities on itself, or introduce them on the
dendritic macromolecule at a later stage. We used its free
acetylene arm to click a long-chain alcohol upon reaction
with 11-azido-undecan-1-ol that was subsequently used to
covalently link LA. The benzyl bromide was then activated
by using a simple azidation (6).
[a] R. Hourani, Dr. A. Kakkar
Department of Chemistry, McGill University
801 Sherbrooke St. West, Montreal, Quebec H3A 2K6 (Canada)
Fax : (+1)514-398-3797
E-mail: ashok.kakkar@mcgill.ca
[b] M. Jain, Dr. D. Maysinger
Department of Pharmacology and Therapeutics
McGill University, 3655 Sir William Osler
Montreal, QC, H3G 1Y6 (Canada)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000241.
6164
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6164 – 6168

COMMUNICATION
was subsequently used for attaching LA through a quantita-
tive esterfication reaction.
We first assessed the cellular internalisation kinetics of
the carriers reported herein, and whether they exert any cy-
totoxicity in human cells in culture. We demonstrated that
the compounds of interest were internalised in human hepa-
tocytes, however, the quantity and rate of entry varied sig-
nificantly for each compound (Figure 1a). The free PM dye
rapidly entered the cells and reached a maximum intracellu-
lar fluorescence within an hour before gradually decreasing
to approximately 11% of its initial value. Compounds 7 and
8 were also rapidly internalised by the cells, however, to a
lesser degree and with slower kinetics. Interestingly, the
mean fluorescence intensities remained significantly higher
in cells exposed to the bifunctional dendrimer incorporating
both LA and dye (7), as well as the monofunctional dendri-
mer containing the dye alone (8), for up to 48 h of incuba-
tion, in comparison with the free dye.
The PM dye and dendritic compounds functionalised with
PM (in 0.1–10 mm concentrations) did not markedly reduce
mitochondrial metabolic activity or alter the cell morpholo-
gy of human hepatocytes (data not shown). When the com-
pounds were assessed at equal concentrations of 1 mm, for
an extended period, the corresponding metabolic activity
did not decrease post 24, 48 and even 72 h of incubation
(Figure 1b), thus indicating the neither the free dye, nor the
dendrimers, 7 and 8, are cytotoxic under the conditions eval-
uated.
The therapeutic potential of the LA-containing dendrimer
7, and free LA (at equimolar concentrations of LA) was
evaluated for cytoprotectivity against H2O2 insult in human
breast cancer (MCF-7) cells (Figure 1c). The bifunctional
dendrimer, after both 1 and 48 h of pre-incubation, was able
to completely protect the cells against H2O2 cytotoxicity
(*** p<0.0001). Free LA at equimolar concentrations was
only able to partially protect cells from the oxidative stress
(**p<0.01). These results imply that when bound to den-
drimers, LA may be therapeutically effective at reduced
concentrations, perhaps due to the greater retention time
permitted by the dendrimer structure.
Since most of the cell-labelling protocols call for short
times of exposure to the fluorophore, and our initial inter-
nalisation experiments showed a rapid increase in fluores-
cence intensities in the treated cells, we assessed the fluoro-
phore entry by fluorescence microscopy. The imaging of
single cells (see Figure S7 in the Supporting Information,)
clearly supported the spectrofluorometric measurements
and showed that the PM dye and the dendritic molecules
were internalised at different rates, and that their sub-cellu-
lar distribution patterns were different. Fluorescent micros-
copy studies indicated that the red PM and the lipid droplet
tracker, green BODIPY dyes, were at least partially co-lo-
calising in lipid droplets, as indicated by the yellow fluores-
cent signal resulting from the overlay of red and green fluo-
rescence. In the case of the bifunctional dendrimer (7), the
yellow fluorescent signal was most intense and clearly de-
tectable as early as two minutes post-treatment as a bright
Scheme 1. Synthesis of building blocks: i) TIPS-acetylene, [PdCl
2ACHTUNGRTNE(NUNG PPh3)2],
CuI, NEt2H, RT, 48 h; ii) TMS-acetylene, [PdCl2A(PPh3)2], CuI, NEt3/ben-
CTHUNGTRENNUNG
zene, reflux, overnight; iii) CBr4, triphenylphosphine (TPP), THF, 2 h;
iv) acetone/water, K2CO3, overnight; v) CuSO4·5H2O, sodium ascorbate,
H2O/THF, 508C, overnight; vi) NaN3, DMF, RT, 2 h.
To synthesise the desired carrier, building block 6 was
clicked to the 1,3,5-triethynylbenzene core, followed by re-
moval of the TIPS groups (Scheme 2). To covalently link
the PM dye to the resulting free acetylenic centres, it was
functionalised with an azide group (PM-N3), by forming an
ester linkage between the alcohol form of the dye with 6-
azido-hexanoic acid (Scheme 3). The hydroxyl terminated
arms were then treated with LA by using an esterfication re-
action to obtain the desired asymmetrically functionalised
dendrimer (7; Scheme 2). The synthetic goals of this study
included examining the role of the dendritic structure on the
internalisation process. For this purpose, we synthesised a
monofunctional dendrimer (8; Scheme 3 by clicking three
azide terminated PM-N3 dye molecules to the 1,3,5-triethy-
nylbenzene (TEB) core.
We had also anticipated that the structure of the carrier
may influence its overall behaviour. Therefore, a linear ana-
logue of the dendrimer containing the PM dye and LA (9;
Scheme 4) was constructed by using (4-ethynyl-phenylethy-
nyl)triisopropylsilane as the central unit on which two Cu-
catalysed click reactions were carried out in sequence. The
free acetylene arm of (4-ethynyl-phenylethynyl)triisopropyl-
silane was first functionalised with a long-alkane-chain alco-
hol by using 11-azidoundecan-1-ol. Upon subsequent depro-
tection, the other acetylene unit was made available for the
second click reaction with PM-N3. The free primary alcohol
Chem. Eur. J. 2010, 16, 6164 – 6168
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6165

A. Kakkar, D. Maysinger et al.
green BODIPY, the mitochon-
dria with Mitotracker Deep
Red 633, and the lysosomes
with Lysotracker DND 26.
Dendrimer 8 and PM co-lo-
calisation with lipid droplets
was partial; however, the bi-
functional dendrimer (7) dem-
onstrated complete co-localisa-
tion with the lipid droplets as
revealed by merging of the indi-
vidual photomicrographs. Simi-
larly, dual labelling of either the
mitochondria and dendrimers,
or lysosomes and dendrimers
(within 5 min), failed to show
any significant co-localisation
of these organelles and the den-
drimers or free PM.
The model therapeutic com-
pound, LA, easily diffuses in
and out of the cell,[21] but when
it is covalently bound to the
dendrimer, its diffusion was
more restrained. Thus, as con-
cluded by Tirosh et al., conjuga-
tion of LA may improve cellu-
lar retention as well as enhance
its therapeutic potential.[22] Re-
sults reported herein suggest
that the residence time of LA
Scheme 2. Synthesis of the bifunctional dendrimer 7: i) CuSO4·5H2O/sodium ascorbate, H2O/THF, 508C, 48 h;
ii) tetrabutylammonium fluoride (TBAF)/THF; iii) same as in i); iv) dicyclohexylcarbodiimide (DCC)/ 4-dime-
thylaminopyridine (DMAP), CH2Cl2, 2 h.
Scheme 4. Synthesis of linear analogue 9: i) CuSO4·5H2O/sodium ascor-
bate, H2O/THF, 508C, overnight; ii) TBAF/THF; iii) PM-N3, same as in
i); iv) LA, DCC/DMAP, CH2Cl2, 2 h.
Scheme 3. Synthesis of monofunctional dendrimer 8: i) DCC/DMAP,
CH2Cl2, 2 h and ii) CuSO4·5H2O/sodium ascorbate, H2O/THF, 508C, 48 h.
can be enhanced by binding to the dendrimer, and its export
from the cell can be slowed down. The latter could contrib-
ute to an enhanced antioxidant effect, as seen after cell
treatment with H2O2.
The synthetic methodology developed herein is versatile
and could be easily elaborated to other pharmacological
yellow signal, depicting almost perfect overlapping with the
lipid droplet stain.
To more precisely assess dendrimer sub-cellular localisa-
tion, confocal microscopy was employed (Figure 2). Organ-
elles were fluorescently labelled, the lipid droplets with
6166
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6164 – 6168

Single Platform Dendrimers
COMMUNICATION
Figure 2. Intracellular localisation of fluorescent PM dye and dendrimers.
Confocal photomicrographs of primary human hepatocytes stained with
green BODIPY, or labelled with Lysotracker and Mitotracker in the
presence of PM (10 min), 7 and 8. Marked co-localisation was observed
between lipid droplets and 7 compared with free PM and 8. Note the ab-
sence of co-localisation between the lysosomes and any of the com-
pounds in the early minutes post treatment when stained with Lysotrack-
er. Note the absence of co-localisation between any of the compounds
and mitochondria when stained with Mitotracker. For more details, see
the Experimental Section.
bodies may function as dynamic and specialised intracellular
sites, accommodating signalling molecules and storing lipids,
approaches to normalise lipid accumulation and utilisation
are of significant therapeutic value when lipid homeostasis
is impaired (e.g., obesity, diabetes). Another application of
such click chemistry materials targeting cells with an exces-
sive number of lipid droplets could be in pathologies associ-
ated with inflammation in which non-steroidal anti-inflam-
matory drugs (NSAID) are linked with dendrimers.[23]
Experimental Section
Spectrophotometric measurements of internalised free and labelled den-
drimers: Cells were seeded at a concentration of 150000 cells per well in
a 24 well plate (Sarstedt). Prior to the addition of fluorescent dyes, the
media was changed to serum-free RPMI 1640. Cells were treated with
various forms of the red fluorescent dye (free dye, linear analogues and
dendrimers), and allowed to incubate for several subsequent hours as in-
dicated in Figure 2b. Upon termination of the incubation period, media
from the cells was aspirated and the cells were gently rinsed with PBS.
DMSO was added to each well and pipetted in triplicate to a black 96-
well plate (Costar). Red fluorescence of the samples was then measured
by using a BMG spectrofluorometer.
Figure 1. Cellular internalisation, cytotoxicity and cytoprotectivity of den-
dritic compounds. a) Time course of cellular internalisation of the den-
drimers (7,8), and free dye (PM) were evaluated over 48 h. b) Mitochon-
drial metabolic activity of cells was assessed by the MTT assay (conver-
sion of a tetrazolium salt to formazan by enzymes in living cells). None
of the three compounds reduced mitochondrial metabolic activity after
24, 48 or 72 h of incubation. c) Cytoprotective effects of 7 in cells ex-
posed to H2O2 (200 mm, for 24 h) were evaluated. For more details, see
the Experimental Section.
Cellular internalisation, distribution and cytotoxicity of the linear and
dendritic compounds (Figure 1): Sub-cellular compartments and antici-
pated distribution of free lipoic acid, free PM dye and the bifunctional
dendrimers. Hepatocytes were treated with equimolar concentrations
(1 mm) of PM dye, of all three compounds; subsequent spectrofluoromet-
ric measurements of intracellular fluorescence show significant differen-
ces in the internalisation kinetics. Intracellular fluorescence intensities of
agents and organelle markers, thereby providing a wide plat-
form for bi- and multi-functional materials for biomedical
applications. For instance, dendrimers could carry inhibitors
of lipid synthesis or storage, and provide clinically useful
“nano-regulators” of cellular lipid homeostasis. Since lipid
Chem. Eur. J. 2010, 16, 6164 – 6168
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6167

A. Kakkar, D. Maysinger et al.
7 and 8 were found to be significantly different from the free dye, PM
(***p<0.0001) after 24 h of incubation. After 48 h, the two dendrimers
showed statistically significant differences from the free dye, PM (***p<
0.0001).
Keywords: biological activity
dendrimers · lipids · multiple functionalities
·
click chemistry
·
Time course of cellular internalisation of the dendrimers (7, 8), linear
compound (PM-LA) and free dye (PM) were evaluated over 48 h. Hepa-
tocytes were treated with equimolar concentrations (1 mm) of PM dye, of
all four compounds; subsequent spectrofluorometric measurements of in-
tracellular fluorescence show significant differences in the internalisation
kinetics. Intracellular fluorescence intensities of 7, 8 and 9 were found to
be significantly different from the free dye, PM (***p<0.0001), after
24 h of incubation. After 48 h, only the two dendrimers showed statisti-
cally significant differences from the free dye, PM (***p<0.0001). Of the
four compounds, only 9 reduced mitochondrial metabolic activity after
24, 48 and 72 h of incubation (***p<0.0001).
[1] C. C. Lee, J. A. MacKay, J. M. Frꢁchet, F. C. Szoka, Nat. Biotechnol.
2005, 23, 1517–1526.
[2] H. K. Bayele, C. Ramaswamy, A. F. Wilderspin, K. S. Srai, I. Toth,
A. T. Florence, J. Pharm. Sci. 2006, 95, 1227–1237.
[3] R. Savic, A. Eisenberg, D. Maysinger, J. Drug Targeting 2006, 14,
343–355.
[4] M. Poupot, L. Griffe, P. Marchand, A. Maraval, O. Rolland, L. Mar-
tinet, F. E. LꢂFaqihi-Olive, C. O. Turrin, A. M. Caminade, J. J. Four-
nie, J. P. Majoral, R. Poupot, FASEB J. 2006, 20, 2339–2351.
[5] E. R. Gillies, E. Dy, J. M. Frꢁchet, F. C. Szoka, Mol. Pharmaceutics
2005, 2, 129–138.
[6] O. Zava, J. Mattsson, B. Therrien, P. J. Dyson, Chem. Eur. J. 2010,
16, 1428–1431.
[7] Y. Hong, H. Xiong, J. W. Y. Lam, M. Hꢃußler, J. Liu, Y. Yu, Y.
Zhong, H. H. Y. Sung, I. D. Williams, K. S. Wong, B. Z. Tang, Chem.
Eur. J. 2010, 16, 1232–1245.
[8] a) D. A. Tomalia, J. M. Frꢁchet, J. Polym. Sci. Part A: Polym. Chem.
2002, 40, 2719–2728; b) R. Hourani, A. Kakkar, Macromol. Rapid
Commun. 2010, DOI: 10.1002/amrc.200900712.
[9] T. P. Thomas, I. J. Majoros, A. Kotlyar, J. F. Kukowska-Latallo, A.
Bielinska, A. Myc, J. R. Baker, Jr., J. Med. Chem. 2005, 48, 3729–
3735.
[10] A. P. Goodwin, S. S. Lam, J. M. Frꢁchet, J. Am. Chem. Soc. 2007,
129, 6994–6995.
[11] P. Antoni, Y. Hed, A. Nordberg, D. Nystrom, H. von Holst, A. Hult,
M. Malkoch, Angew. Chem. 2009, 121, 2160–2164; Angew. Chem.
Int. Ed. 2009, 48, 2126–2130.
[12] R. Duncan, L. Izzo, Adv. Drug Delivery Rev. 2005, 57, 2215–2237.
[13] S. Shaunak, S. Thomas, E. Gianasi, A. Godwin, E. Jones, I. Teo, K.
Mireskandari, P. Luthert, R. Duncan, S. Patterson, P. Khaw, S. Broc-
chini, Nat. Biotechnol. 2004, 22, 977–984.
[14] S. Svenson, D. A. Tomalia, Adv. Drug Delivery Rev. 2005, 57, 2106–
2129.
[15] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056–2075; Angew. Chem. Int. Ed. 2001, 40, 2004–2021.
[16] C. A. DeForest, B. D. Polizzotti, K. S. Anseth, Nat. Mater. 2009, 8,
659–664.
[17] P. Wu, A. K. Feldman, A. K. Nugent, C. J. Hawker, A. Scheel, B.
Voit, J. Pyun, J. M. Frꢁchet, K. B. Sharpless, V. V. Fokin, Angew.
Chem. 2004, 116, 4018–4022; Angew. Chem. Int. Ed. 2004, 43, 3928–
3932.
[18] S. Murphy, S. Martin, R. G. Parton, Biochim. Biophys. Acta 2009,
1791, 441–447.
[19] M. L. Bolognesi, A. Minarini, V. Tumiatti, C. Melchiorre, Mini-Rev.
Med. Chem. 2006, 6, 1269–1274.
Mitochondrial metabolic activity of cells was assessed by the MTT assay.
Cytoprotective effects of 7 in cells exposed to H2O2 (200 mm, for 24 h).
Each compound was preincubated with the cells for either 1 h or the op-
timal time required to ensure maximal cellular uptake (24 h for 9, 48 h
for 7). LA was preincubated with cells for 1 h prior to H2O2 addition.
LA, 9 and the dendrimer 7 treatments were utilised in equimolar concen-
trations of LA (5.6 mm). Compound 9 regardless of incubation time was
not protective, but significantly cytotoxic (p<0.0001) and further en-
hanced the cytotoxicity derived from H2O2 (**p<0.01). Compound 7 was
cytoprotective (p<0.0001) at both preincubation times, whereas LA, at
these sub-therapeutic conditions was only mildly cytoprotective (**p<
0.01).
Confocal laser scanning microscopy: Confocal laser scanning microscopy
was carried out by using a Zeiss LSM 510 microscope equipped with the
following lasers: 1) HeNe LASOS LGK 7786 P/Power supply 7460 A:
543 nm, 1 mW, 2) Argon LASOS LGK 7812 ML-1/LGN: 458, 488,
514 nm; 25 mW; Laser class 3D and 3) Titanium:Sapphire The Coherent
Mira Model 900-F Laser tuneable from 710 to 1000 nm for two photon
microscopy set to pulse at 800 nm. Cells for imaging were grown in eight-
well chambers (Lab-Tek, Nalge Nunc International, Rochester, NY,
USA). Prior to staining the cells, the cell media was changed to serum
free and fluorescent organelle dyes were added. Human primary hepato-
cytes were stained with the blue fluorescent nuclear dye, Hoechst 33342,
and lipid droplets were labelled with 493/503 (20 mm, 10 min; lex
=
493 nm, lem =503 nm). Lipid bodies were stained with green BODIPY
(Invitrogen) (20 mm, 10 min; lex =644 nm, lem =665 nm), mitochondria
were stained with Mitotracker Deep Red 633 (1 mm, 1 min; lex =644 nm,
lem =665 nm), but then pseudo-coloured green, and nuclei were stained
with Hoechst 33342 (Invitrogen) (10 mm, 1 h; lex =350 nm, lem =461 nm).
Red, linearised Red and dendrimers with red were added to designated
wells and the cells were incubated for 10 min. Before imaging, cells were
washed with PBS or with serum-free medium. No background fluores-
cence of cells was detected under the settings used. Images were also re-
corded of cells labelled with Lysotracker Red DND99 (500 nm; 3 min;
lex =577, lem =590) and Mitotracker Deep Red 633.
[20] L. Packer, E. H. Witt, H. J. Tritschler, Free Radical Biol. Med. 1995,
19, 227–250.
[21] V. N. Totskii, Biokhimiya 1976, 41, 1094–1105.
[22] O. Tirosh, C. K. Sen, S. Roy, M. S. Kobayashi, L. Packer, Free Radi-
cal Biol. Med. 1999, 26, 1418–1426.
Acknowledgements
[23] M. Zentella de Pina, H. Vazquez-Meza, C. Agundis, M. A. Pereyra,
J. P. Pardo, R. Villalobos-Molina, E. Pina, Auton. Autacoid Pharma-
col. 2007, 27, 85–92.
A.K. and D.M. would like to thank NSERC of Canada for financial assis-
tance. R.H. thanks FQRNT (Quebec, Canada) and M.J. thanks FRSQ
for a post-graduate scholarship.
Received: January 27, 2010
Published online: May 3, 2010
6168
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 6164 – 6168