Combining aminocyanine dyes with polyamide dendrons: A promising strategy for imaging in the near-infrared region

Article abstract of DOI:10.1002/chem.201002268

Cyanine dyes are known for their fluorescence in the near-IR (NIR) region, which is desirable for biological applications. We report the synthesis of a series of aminocyanine dyes containing terminal functional groups such as acid, azide, and cyclooctyne groups for further functionalization through, for example, click chemistry. These aminocyanine dyes can be attached to polyfunctional dendrons by copper-catalyzed azide alkyne cycloaddition (CuAAC), strain-promoted azide alkyne cycloaddition (SPAAC), peptide coupling, or direct S_NR1 reactions. The resulting dendron-dye conjugates were obtained in high yields and displayed high chemical stability and photostability. The optical properties of the new compounds were studied by UV/Vis and fluorescence spectroscopy. All compounds show large Stokes shifts and strong fluorescence in the NIR region with high quantum yields, which are optimal properties for in vivo optical imaging.

Full text of DOI:10.1002/chem.201002268

FULL PAPER
DOI: 10.1002/chem.201002268
Combining Aminocyanine Dyes with Polyamide Dendrons: A Promising
Strategy for Imaging in the Near-Infrared Region
Cꢀtia Ornelas,[a, b] Rachelle Lodescar,[a, b] Alexander Durandin,[b] James W. Canary,[b]
Ryan Pennell,[c] Leonard F. Liebes,[c] and Marcus Weck*[a, b]
Abstract: Cyanine dyes are known for
their fluorescence in the near-IR (NIR)
region, which is desirable for biological
applications. We report the synthesis of
a series of aminocyanine dyes contain-
ing terminal functional groups such as
acid, azide, and cyclooctyne groups for
further functionalization through, for
example, click chemistry. These amino-
cyanine dyes can be attached to poly-
functional dendrons by copper-cata-
lyzed azide alkyne cycloaddition
(CuAAC), strain-promoted azide
alkyne cycloaddition (SPAAC), peptide
coupling, or direct SNR1 reactions. The
resulting dendron–dye conjugates were
obtained in high yields and displayed
high chemical stability and photostabil-
ity. The optical properties of the new
compounds were studied by UV/Vis
and fluorescence spectroscopy. All
compounds show large Stokes shifts
and strong fluorescence in the NIR
region with high quantum yields, which
are optimal properties for in vivo opti-
cal imaging.
Keywords: CuAAC · cycloaddition ·
dendrimers · dendrons · dyes/pig-
ments · imaging agent · SPAAC
Introduction
is determined by the optical properties of the tissue of inter-
est. Hemoglobin has a strong absorption at wavelengths
lower than 600 nm and significant background fluorescence
from endogenous biomolecules can be detected up to
680 nm.[8] Near infrared (NIR) light can overcome these lim-
itations by penetrating more deeply into tissue, because
light scattering decreases with increasing wavelength. Thus,
organic fluorophores with absorption and emission proper-
ties in the NIR region are ideal for in vivo optical imaging.[8]
Among the organic near-infrared dyes, cyanine-based
compounds are of particular interest. Their basic structure
includes two aromatic or heterocyclic rings linked by a poly-
methine chain with conjugated carbon–carbon bonds.[9,10]
These compounds have found their use in day-to-day appli-
cations in the medical field. For example, indocyanine green
(ICG) (Scheme 1) is a Food and Drug Administration
(FDA)-approved cyanine dye that is currently used for
imaging in humans.[8] It is considered a safe imaging agent
and has been used in tens of thousands of patients with a re-
ported side-effect rate below 0.15%, an extremely favorable
index when compared to other imaging agents.[8,11–13] This
impressive biocompatibility has initiated a search for cya-
nine dye derivatives containing functional handles available
for further functionalization in order to attach these dyes to
peptides, targeting groups and/or drug carriers.[10] Unfortu-
nately, the synthesis of ICG derivatives is often difficult and
low yielding mainly due to low solubility and chemical sta-
bility.[9,10] One strategy to overcome these shortcomings has
been introduced by Patonay and co-workers who showed
that the insertion of a cyclohexenyl ring into the poly-
In life science, the search for new methods of early detec-
tion of diseases has stimulated researchers to target the de-
velopment of efficient imaging agents.[1–5] Different imaging
modalities present different strengths and weaknesses. For
example, magnetic resonance imaging (MRI) has a very
high spatial resolution but its sensitivity of around 10ꢀ3–
10ꢀ5 m can be limiting.[4] In some particular cases, positron
emission tomography (PET) is more appropriate due to its
high sensitivity.[1,6,7] Among the different imaging modalities,
optical imaging is the one that presents less safety concerns,
because organic fluorescent dyes often display good biocom-
patibility, a significant advantage over radioactive markers,
which are frequently toxic.[2] Thus, intensive research efforts
have been devoted to the design and synthesis of new water-
soluble organic fluorophoric systems.[3,8] The optimal wave-
length range for in vivo fluorescence excitation and emission
[a] Dr. C. Ornelas, R. Lodescar, Prof. M. Weck
Molecular Design Institute, Department of Chemistry
New York University, 31 Washington Place
New York, NY 10003-6688 (USA)
Fax : (+1)2129954895
E-mail: marcus.weck@nyu.edu
[b] Dr. C. Ornelas, R. Lodescar, Dr. A. Durandin, Prof. J. W. Canary,
Prof. M. Weck
Department of Chemistry, New York University
31 Washington Place, New York NY, 10003-6688 (USA)
[c] R. Pennell, Prof. L. F. Liebes
Cancer Institute, New York University School of Medicine
550 First Avenue, New York, NY 10016 (USA)
AHCTUNGTREGmNNNU ethine chain increases solubility and photostability and
provides an additional site for further functionalization
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201002268.
(Scheme 1).[14]
Chem. Eur. J. 2011, 17, 3619 – 3629
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3619

polyamide-based materials and follow the 1!
ACHTUNGTREN(NUNG 2+1) connec-
tivity pioneered by Newkome et al.[54] All aminocyanine
dyes and dendron–dye conjugates were characterized by
using NMR spectroscopy, mass spectrometry, and elemental
analysis, and their optical properties were investigated by
UV/Vis and fluorescence spectroscopy. The new compounds
show large Stokes shifts and strong fluorescence in the NIR
region with high quantum yields, which are optimal proper-
ties for in vivo imaging. Furthermore, the dendrons possess
terminal groups available for further functionalization with
targeting groups and/or drugs, making these materials suita-
ble for theranostics.
Results and Discussion
Scheme 1. a) Indocyanine green (ICG) and b) cyanine dye containing a
cyclohexenyl ring on the polymethine chain.
Because our main goal is the use of NIR dyes as imaging
agents for biological applications, we investigated cyanine
dyes that contain two sulfonate groups at the end of the
propyl chains on the nitrogen of the indolenine ring, which
are known to increase water solubility.[16] Furthermore, the
negative charges on the sulfonate groups should enhance
the biocompatibility of the dye–dendron conjugates because
negatively charged dendrimers are less toxic than their posi-
tively charged analogues.[35,36]
Alcohol[15–18] or thiol[19–22] groups are often employed to
replace the chlorine atom at the central ring. The resulting
ethers or thioethers unfortunately present relatively low
chemical stability limiting their use as fluorescent probes in
vivo.[23,24] The Achilefu group has reported the use of a
cross-coupling reaction as an alternative reaction to func-
tionalize cyclohexene-containing cyanine dyes resulting in
[23]
ꢀ
cyanine derivatives with C C bonds at the central ring.
These dyes are more stable than their ether or thioether an-
alogues. However, their optical properties are not optimal
due to the low stokes shift (<30 nm).[23] Gao and co-workers
have shown that the replacement of the chlorine atom with
an amine group results in the formation of aminocyanine
dyes with high quantum yields and large Stokes shifts
Synthesis of the aminocyanine dyes: The chlorocyanine dye
was synthesized in three steps from 2,3,3-trimethylindole-
nine (1), cyclohexanone (2), and 1,3-propane sultone (3) in
close analogy to the literature (Scheme 2).[15,16] The hepta-
ACHTUNGTRENNUNG
(>140 nm).[24] These materials have been explored as fluo-
rescent probes for metal ions, however, reports about their
use as imaging agents are still scarce.[25–29]
Our research group has been developing synthetic strat-
egies for the synthesis of new drug delivery systems based
on well-defined multifunctional dendrons and dendri-
ACHTUNGTRENNUNG
mers.[30–33] We were very intrigued by the above-described
aminocyanine dyes as potential imaging agents for our mul-
tifunctional dendrimer-based delivery system for applica-
tions in theranostics.[34–43] Orthogonality between functional
groups is crucial to the synthesis and functionalization of
such complex structures,[44] thus, our attention was focused
on the derivatization of the NIR dyes with functional groups
suitable for orthogonal chemistry. In this contribution we
present an extensive study on functionalization strategies for
chlorine-containing cyanine dyes with a library of amines
containing different functional groups and evaluated the
chemical stability and optical properties of the resulting ma-
terials. The dyes were then attached to either one terminal
branch or the focal point of polyfunctional dendrons by
using copper-catalyzed alkyne azide cycloaddition
(CuAAC), the most popular “click” reaction,[45–50] strain-
promoted alkyne azide cycloaddition (SPAAC),[33,51–53] pep-
tide coupling, or direct radical nucleophilic substitution
(SRN1) reactions. The dendrons employed in this work are
Scheme 2. Synthesis of the NIR chlorocyanine dye 6.
methine chain of 6 was assembled from the aldol-like con-
densation of the indolinium salt 4 and the iminium salt 5
(Scheme 2). Dye 6 was obtained as a deep green solid in
70% overall yield and characterized by mass spectrometry
showing its molecular ion peak [MꢀNa]ꢀ at m/z 697.4 (m/z
calcd for C36H42ClN2O6S2: 697.2).
3620
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3619 – 3629

A Promising Strategy for Imaging in the Near-Infrared Region
FULL PAPER
and the materials were charac-
terized by 1H and 13C NMR
spectroscopy and mass spec-
trometry (see the Supporting
Information).
To obtain a NIR dye with an
alkyne group, a SRN1 reaction
Scheme 3. General SRN1 reaction of chlorocyanine dye 6 with various amines.
between 6 and propargyl amine
was carried out. After heating
The reaction of 6 with amines was carried out in dimethyl-
formamide (DMF) at 808C.[24] In most cases, the substitution
of the chlorine atom by the amine group (Scheme 3) was ob-
served easily by a color change of the initial deep green so-
lution to deep blue. Because our target was the conjugation
of the dyes to dendritic molecules by using high yielding re-
actions, such as peptide coupling, CuAAC, and SPAAC, we
focused on the funtionalization of 6 with acid, amine, azide,
and alkyne terminal groups.
The SRN1 reactions of 6 with 3-aminopropyl azide, 6-ami-
nohexanoic acid, 11-aminoundecanoic acid, N-Boc-1,6-hexa-
nediamine (Boc=tert-butoxycarbonyl), and tert-butyl 12-
amino-4,7,10-trioxadodecanoate were carried out successful-
ly yielding dyes 7–11 (Scheme 4). After precipitation with
ethyl ether, these dyes were obtained as deep blue solids.
Final purification was performed by dialysis against water
at 808C for 2 h, the reaction mixture was analyzed by mass
spectrometry, which showed the molecular ion peak
[MꢀNa]ꢀ of 12 at m/z 716.4 (m/z calcd for C39H46N3O6S2:
716.3). However, even after several attempts, compound 12
could not be isolated probably due to low chemical stability.
We suggest that the close proximity of the alkyne group to
the cyanine p system could be the cause of the low stability.
A synthetic strategy was designed to obtain the dye–cyclo-
octyne conjugate with an amine-containing cyclooctyne ring
(13d) as the key intermediate (Scheme 5). Bicyclic com-
pound 13a (8,8-dibromobicyclo
sized according to the procedure described by Skattebol and
Solomon for the synthesis of 9,9-dibromo
[6.1.0]nonane.[55,56]
ACHTUNGTNER[NUNG 5.1.0]octane) was synthe-
AHCTUNGTRENNUNG
Silver perchlorate was then used for the electrocyclic ring-
opening to the trans-allylic cation, which was captured with
3-(Boc-amino)-1-propanol affording the bromo-trans-cyclo-
Scheme 4. Functionalization of the chlorocyanine dye 6 with a library of amines.
Chem. Eur. J. 2011, 17, 3619 – 3629
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3621

M. Weck et al.
Scheme 5. Synthesis of dye–cyclooctyne 13. DBU=1,8-diazabicycloACHTNUTRGNEUNG[5.4.0]undec-7-ene, TFA=trifluoroacetic acid.
octene 13b. Cyclooctyne 13c was obtained by elimination of
bromide in DMSO at 608C, by using DBU as the base.
However, the resulting cyclooctyne ring was not stable
under acidic conditions (formic acid or TFA) used for the
deprotection of the Boc group. Thus, we removed the Boc
protecting group and attached the resulting amine to 6 prior
to the elimination reaction affording 13. The reaction of
amine 13e with 6 and the elimination were carried out by
using a one-pot procedure. Amine 13e and 6 were heated at
808C in DMF together with two equivalents of DBU. After
24 h, a mass spectrum of the reaction showed the molecular
ion peak [MꢀNa]ꢀ of 13 f at m/z 922.4 (m/z calcd for
C47H61BrN3O7S2: 922.3) suggesting that the substitution of
the chlorine atom by the amine group was complete. How-
ever, no evidence for the elimination reaction was visible.
Ten more equivalents of DBU were added and the solution
was heated at 808C for another 48 h. After this time, the
mass spectrum of the solution showed the molecular ion
peak MꢀNa]ꢀ of 13 at m/z 842.6 (m/z calcd for
C47H60N3O7S2: 842.4) (see the Supporting Information for
the mass spectra).
spectra (Table 1). These features are in agreement with pre-
vious reports on aminocyanine dyes and result from an in-
tramolecular charge transfer (ICT).[24,28] The maximum ab-
sorption of the dyes is highly solvent dependent (Table 1).
However, contrary to the reported aminocyanine dyes with
alkyl or phenyl termini, the solvatochromism of the new
Table 1. Optical properties of the aminocyanine dyes.
Dye
Absorption[a]
DMF
l [nm]
(e)
Emission[b]
PBS
l [nm]
(e)
DMSO
l [nm]
(e)
EtOH
l [nm]
(e)
MeOH
l [nm]
(e)
[c]
l
fF
[nm]
7
622
(1500)
611
(12900)
607
(3400)
616
(30500)
633
(33600)
595
632
(2400)
634
(13100)
634
(4000)
635
(31300)
645
(31900)
636
624
(1400)
617
(11600)
614
(3400)
620
(28000)
635
(25500)
616
636
(3200)
629
(18300)
629
(5600)
635
(36200)
648
(40400)
627
637
(4000)
628
(21500)
623
(6400)
628
(39800)
648
(44000)
613
755
758
752
752
754
753
0.13
0.10
0.13
0.11
0.16
0.11
8
9
10
11
13
(2100)
(4100)
(2600)
(6300)
(7600)
Optical properties of the aminocyanine dyes: All aminocya-
nine dyes 7–13 are deep blue in solution and the solid state.
The general features of their absorption spectra are: 1) a
strong blue shift when compared to 6, from 780 nm to 595–
645 nm, 2) a large stokes shift (>105 nm), and 3) no mirror
image relationship between the absorption and fluorescence
[a] The absorption spectra were recorded at 258C and e is expressed in
mꢀ1 cmꢀ1 at the maximum of the highest peak. PBS=phosphate buffer
aqueous solution. [b] Excitation wavelength was 632 nm for all com-
pounds. [c] Relative fluorescence quantum yields (fF) were determined
in methanol by using the known quantum yield of ICG as a reference
(fF =0.078 in methanol).[11]
3622
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3619 – 3629

A Promising Strategy for Imaging in the Near-Infrared Region
FULL PAPER
dyes does not follow a specific
trend based on solvent polarity.
For dyes 8–10, which possess an
alkyl linker between the dye
and the functional group, the
value of absorption (lmax) fol-
lows the trend DMSO>
EtOH>MeOH>DMF>PBS,
whereas for dye 11, in which
the linker is a polyethylene-
glycol chain, the absorption lmax
follows the trend MeOH=
EtOH>DMSO>DMF>PBS
(Table 1, Figure 1). When com-
paring the absorption spectra of
the aminocyanine dyes in the
same solvent, their lmax also
show variations. These results
suggest that the terminal func-
tional group, as well as the
nature and length of the linker
have a strong influence on dye–
solvent interactions.
Cyanine dyes have a tenden-
cy to aggregate in solution
Figure 1. Absorption spectra (concentrations of 1ꢁ10ꢀ5 m at 258C) of a) dye
ACHTUNGTRENNUNG
ꢀ
ꢀ
ꢀ
through plane-to-plane stacking
(parallel aggregates) or through
an end-to-end arrangement
(head-to-tail). The transition to
the upper state in parallel ag-
gregates leads to a hypsochro-
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ACHUTNGERN(NUG CH2)10 COOH (9), c) dye ACTHUNGTRENNUNG
glycol) in different solvents (blue=PBS, green=DMSO, orange=DMF, red=EtOH, and black=MeOH).
“H” indicates H aggregates and “J” indicates J aggregates.
mic (red) shift and these groups of molecules are called H
aggregates, whereas the transition to a lower state in a head-
to-tail arrangement leads to a bathochromic (blue) shift and
these assemblies are called J aggregates.[9,57–62] The absorp-
tion spectra of the new aminocyanine dyes show that they
have low tendency to aggregate because their main absorp-
tion bands correspond to the monomeric species. However,
evidence for the formation of limited aggregation can be
found in the absorption spectra of some compounds. Close
analysis shows that the type and amount of aggregation de-
pends significantly on the nature and the length of the
ꢀ
ꢀ
linker, dye
ACHTUNGTRENNUNG(CH2)5 COOH (8) shows both H and J aggre-
ꢀ
ꢀ
gates, whereas dye ACHTUNGTRENNUNG(CH2)10 COOH (9) shows only H aggre-
ꢀ
ꢀ
gates (Figure 1). Dye ACHTNUGTRNEGNU(CH2)6 NHBoc (10) forms only J ag-
ꢀ
ꢀ
gregates and dye PEG COOtBu (tBu=tert-butyl) (11)
shows no evidence of aggregation (Figure 1, for all other
spectra see the Supporting Information).
Next, we investigated the fluorescence properties of the
new aminocyanine dyes. All dyes show fluorescence in the
NIR region around 754 nm. Contrary to the absorption
spectra, the linkers and the terminal groups do not have a
significant influence on the emission wavelength (Figure 2).
The dyes have high quantum yields, being higher than ICG
ꢀ
ꢀ
(0.078) for all new dyes. Dye PEG COOtBu (13) is the dye
with the highest quantum yield (0.16), twice as high as ICG,
which is probably due to the presence of some aggregates
Figure 2. Absorption (solid line) and emission (dash line) spectra of
ꢀ
ꢀ
ꢀ
ꢀ
a) dye ACTHNUGRTNEUNG(CH2)5 COOH (8) and b) dye PEG COOtBu (11).
Chem. Eur. J. 2011, 17, 3619 – 3629
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3623

M. Weck et al.
for all other dyes resulting in some self-quenching (13 does
not show any kind of aggregation). The fluorescence spectra
of all new dyes can be found in the Supporting Information.
acterized by mass spectrometry showing its molecular ion
peak [MꢀNa]2ꢀ at m/z 935.0 (m/z calcd for
(C96H145N10O23S2)2ꢀ: 935.0. The deep blue coloration of 16 in
the solid state slowly changed to deep brown after 24 h, and
the mass spectrum no longer showed the molecular ion peak
of 16. The color change and the disappearance of the molec-
ular ion peak in the mass spectra suggest that compound 16
degrades over time, which could have been promoted by the
presence of copper traces from the CuAAC reaction.[25,33]
The reaction between the acid group of 14 and 3-amino-
propyl azide by using HATU yielded dendron 17 containing
one azide group in 97% yield. SPAAC between 17 and 13
was carried out by mixing both products in H2O/MeOH
(1:1) at room temperature. The reaction was followed by
mass spectrometry and was complete after 48 h. After purifi-
cation by dialysis, dendron–dye 18 was obtained as a deep
blue waxy product and was characterized by mass spectrom-
etry through its molecular ion peak [MꢀNa]2ꢀ at m/z 998.3
(m/z calcd for (C104H159N10O24S2)2ꢀ: 998.1).
Introduction of the aminocyanine dyes into one branch of
AB6C1-type dendrons: Cyanine dyes were selectively at-
tached to one branch of AB6X1 dendrons (A=NO2; B=
COOtBu or COOH; X=CꢁCH,=N3, or NH2) through
three different synthetic approaches: CuAAC, SPAAC, and
direct SRN1. Multifunctional dendrons were synthesized
from dendron 14[31] containing a nitro group at the focal
point, six protected acids and one free acid group as de-
scribed in Scheme 6.
The coupling reaction between the acid group of 14 and
propargylamine afforded dendron 15, which contains one
alkyne group, in 97% yield. The CuAAC reaction between
dendron 15 and dye–azide 7 by using CuSO4/sodium ascor-
bate in CH2Cl2/H2O (1:1) afforded dendron–dye 16 as a
deep blue waxy product in 82% yield. Dendron 16 was char-
Scheme 6. Attachment of NIR dyes to one branch of multifunctional dendrons by using CuAAC, SPAAC, or SRN1. HATU=2-(7-aza-1H-benzotriazole-
1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate; DIPEA=N,N-diisopropylethylamine.
3624
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3619 – 3629

A Promising Strategy for Imaging in the Near-Infrared Region
FULL PAPER
Table 2. Optical properties of the dendron–dye conjugates.
Contrary to compound 16 synthesized by CuAAC, the
dye–dendron conjugate 18 synthesized by SPAAC was
stable in the solid state even after nine months of storage.
The coupling reaction between 14 and N-Boc-1,6-hexane-
diamine, followed by the deprotection of both the Boc-pro-
tected amine group and the tert-butyl-protected acid groups
by using formic acid, afforded dendron 20 containing six
acids and one amine group. The SRN1 reaction between 20
and 6 was carried out in DMF at 808C for 12 h yielding
dye–dendron 21 that was characterized by its molecular ion
peak [MꢀNa]ꢀ at m/z 1494.4 (m/z calcd for
(C72H100N7O23S2)ꢀ: 1494.6).
Dye
Absorption[a]
DMF
l [nm]
(e)
Emission[b]
PBS
l [nm]
(e)
DMSO
l [nm]
(e)
EtOH
l [nm]
(e)
MeOH
l [nm]
(e)
[c]
l
fF
[nm]
749
18
21
639
(9600)
615
636
618
628
627
0.12
0.11
(20400) (17900) (21100) (27400)
651 643 653 645
773
(27400) (29600) (18300) (36600) (41000)
772 793 790 784 780
(29200) (46800) (32300) (59600) (56800)
25[d] 764
782 778 772 765
(14000) (14600) (16700) (17600)
792
756
804
758
0.024
0.11
(8500)
–
26[e]
637
606
623
627
(7300)
797
(800)
794
(2700)
786
(6700)
783
28
774
0.064
0.027
Attachment of NIR dyes to the focal point of dendrons: We
decided to attach a NIR dye to the focal point of a dendron,
leaving the peripheral functional groups available for further
functionalization with targeting groups and/or drugs. Reduc-
tion of the nitro group at the focal point of the polyamide
dendrons affords 22 suitable for SNR1 reaction with 6. Our
first attempt to attach 6 to the focal point of 22 resulted,
after 24 h of heating at 808C in DMF, in no reaction with
the mass spectrum of the mixture showing only starting ma-
terials. The solution was heated for an additional 48 h but
no reaction was observed and the expected compound 23
was not obtained. We rationalized that the sterically bulky
tert-butyl groups on the dendrimer precluded the attachment
of 6 to the amine group at the focal point. Thus, two differ-
ent synthetic strategies to attach the NIR dye to the focal
point of the dendrons were designed: 1) removal of the tBu
groups prior to the attachment of the dye and 2) introduc-
tion of a linker between the focal point and the NIR dye.
The tBu groups of 22 were successfully removed by using
ꢀ
(30300) (40000) (40000) (54000) (56700)
30[d] 753
(8200)
773
(8200)
769
(4300)
763
760
(12400) (12000)
[a] The absorption spectra were recorded at 258C and e is expressed in
mꢀ1 cmꢀ1 at the maximum of the highest peak. [b] Excitation wavelength
was 632 nm for all compounds. [c] Relative fluorescence quantum yields
(fF) were determined in methanol by using the known quantum yield of
ICG as a reference (fF =0.078 in methanol).[11] [d] Only the band in the
NIR region is indicated. [e] Compound 26 is not soluble in aqueous
media.
before the attachment to the dendron, including high quan-
tum yields and large stokes shifts (Table 2). Dendron dye
(COOH)6 NO2 (21) shows a significant amount of mono-
ꢀ
ꢀ
N
meric species and a narrow band at higher wavelength that
corresponds to J aggregates. The amount of J aggregates in
this case is solvent dependent being lower in aqueous media
than in alcoholic solvents (Figure 3b). The high quantum
yield of this compound (0.11) is due to the presence of
monoACTHNGUTERNmUNG eric species and J aggregates that usually present
high quantum yield and the absence of H aggregates that
are known to cause self-quenching (Figures 3b and 4a, and
formic acid, affording dendrimer NH2
G
upon reaction with 6, yielded dye–dendron 25 (Scheme 7).
Following the second strategy, we coupled the amine group
of 22 with the acid group of dye 8 that contains an alkyl
linker by using HATU and DIPEA, affording dye–dendron
26 (Scheme 7). Furthermore, we introduced an ethylene
glycol linker at the focal point of 22 affording dendrons
ꢀ
ꢀ
Table 2). Dendron dye PEG ACHTNUGRTENUNG(COOtBu)9 (28) has a high
degree of organization in solution showing only J aggregates
(Figure 3e). This results in a relatively low stokes shift
(21 nm) (Figure 4c), which is typical for J aggregates. The
quantum yield for this compound is relatively high (0.064)
being close to that of ICG (0.078). Dendron–dye conjugates
25 and 30 show stochastic mixtures of H and J aggregates
with no evidence for the presence of any monomeric species
in solution (Figures 3c and f). Because the H aggregates
absorb in the visible region, 25 and 30 present different
colors in solution. Solutions of 25 in PBS, DMSO, DMF,
EtOH, and MeOH are brownish/orange. Compound 30
presents a pink coloration in PBS, EtOH, and MeOH, and
an orange coloration in DMF and DMSO. The difference in
the coloration of 30 in these two groups of solvents can also
be observed in its absorption spectra (Figure 3 f), and sug-
gests that different types of H aggregates are formed in
protic and aprotic solvents. The presence of a high amount
of H aggregates causes self-quenching of the emission spec-
tra of 25 and 30 (Figure 4b and d, and Table 2) resulting in
much lower quantum yields than all the other dyes and dye–
dendron conjugates (0.024 for 25 and 0.027 for 30).
ꢀ
ꢀ
ꢀ
ꢀ
NH2 linker ACHTUNGTRENNUNG(COOtBu)9 (27) and NH2 linker (N3)9 (29),
which were submitted to SRN1 reaction with 6 resulting in
the formation of the dye–dendrons 28 and 30, respectively
(for experimental details and characterization data of all
dendrons and dye–dendrons, see the Supporting Informa-
tion).
Optical properties of the aminocyanine dyes and the dye–
dendron conjugates: The optical properties of the new den-
dron–dye conjugates differ significantly from each other,
demonstrating that the structure and the terminal groups of
the dendron dictate the dyesꢂ behavior in solution. Den-
ꢀ
ꢀ
ꢀ
ꢀ
drons dye
ACHTUNGTRENNUNG(COOtBu)6 NO2 (18) and dye AHCTUNGTREN(NUGN CH2)5
ACHTUNGTRENNUNG(COOtBu)9 (26) show no significant aggregation, the major
absorption band corresponding to their monomeric species
(Figures 3a and d). Because these two dye–dendron conju-
gates have a low tendency to aggregate in solution, they
present optical properties similar to the aminocyanine dyes
Chem. Eur. J. 2011, 17, 3619 – 3629
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3625

M. Weck et al.
Scheme 7. Attachment of the NIR dye to the focal point of dendrons. Fmoc=fluorenylmethoxycarbonyl.
ꢀ
ꢀ
Cytotoxicity of the dye–dendron conjugate dye PEG (N3)9
(30): As our ultimate goal was the application of the NIR
aminocyanine dyes as imaging agents in theranostics, it is es-
sential to evaluate the cytotoxicity of the dye–dendron con-
3626
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3619 – 3629

A Promising Strategy for Imaging in the Near-Infrared Region
FULL PAPER
use of these aminocyanine dyes
as imaging agents in dendritic
drug delivery systems.
Conclusion
In this manuscript, we report
the synthesis of aminocyanine
dyes containing useful function-
al groups such as acids, azides,
and cyclooctyne groups. The
aminocyanine dyes can be at-
tached to polyfunctional den-
drons by using CuAAC,
SPAAC, peptide coupling, or
direct SNR1 reactions. The ami-
nocyanine dyes and dendron–
dye conjugates were obtained
in high yields and showed good
to excellent chemical and opti-
cal stabilities. The optical prop-
erties of the new compounds
were studied by UV/Vis and
fluorescence spectroscopy. All
compounds show fluorescence
in the NIR region. It was ob-
served that the formation of H
and J aggregates in solution
strongly depends on the nature
and length of the linker and the
functional group on the amino-
cyanine dyes, as well as the
dendron structure and terminal
groups in the dye–dendron con-
jugates. Finally, cytotoxicity
studies of the dye–dendron con-
jugates demonstrate that these
materials are non toxic, which
is an encouraging result for pur-
suing our studies on the use of
these aminocyanine dyes as
imaging agents in dendritic
drug delivery systems.
ꢀ
ꢀ
Figure 3. Absorption spectra of a) dendron dye ACTHNUGTRENUNG(COOtBu)6 NO2 (18) synthesized by SPAAC, b) dendron
ꢀ
ꢀ
ꢀ
E
ꢀ
G
ꢀ
dye
G
ACHTUNGTRNE(NUGN COOtBu)9 (26), e) den-
ꢀ
dron dye PEG ACHTUNGTRENNUNG
green=DMSO, orange=DMF, red=EtOH, and black=MeOH). All spectra were taken at concentrations of
1ꢁ10ꢀ5 m at 258C. “M” indicates monomeric species, “H” indicates H aggregates and “J” indicates J aggre-
gates.
jugates. Thus, we evaluated the toxicity of 30 in human
brain glioblastoma cells (T98G cell line) that were grown in
the presence of different concentrations of 30. After 72 h of
incubation, living cells were quantified by using the WST-1
proliferation assay, and Figure 5 shows the results obtained
for the proliferation assays for 0.5, 1.0, 5.0, 10, and 50 mm of
30.
Acknowledgements
We thank New York University for financial support of this research.
The proliferation assay demonstrates that even after an
incubation time of 72 h, 30 does not present significant tox-
icity for T98G cells at concentrations up to 50 mm. These re-
sults show that besides the presence of aggregates in solu-
tion, the dye–dendron conjugate 30 is not toxic to these
cells, which is encouraging for pursuing our studies on the
Chem. Eur. J. 2011, 17, 3619 – 3629
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3627

M. Weck et al.
[7] T. J. McCarthy, W. A. Banks,
C. L. Farrell, S. Adamu, C. P. Der-
deyn, A. Z. Snyder, R. Laforest,
D. C. Litzinger, D. Martin, C. P.
LeBel, M. J. Welch, J. Pharmacol.
Exp. Ther. 2002, 301, 878–883.
[8] R. Weissleder, V. Ntziachristos,
Nat. Med. 2003, 9, 123–128.
[9] A. Mishra, R. K. Behera, P. K.
Behera, B. K. Mishra, G. B.
Behera, Chem. Rev. 2000, 100,
1973–2012.
[10] R. Pandey, N. James, Y. Chen, M.
Dobhal, Top. Geterocycl. Chem.
2008, 14, 41–74.
[11] R. C. Benson, H. A. Kues, Phys.
Med. Biol. 1978, 23, 159–163.
[12] D. R. Guyer, L. A. M. Yannuzzi,
J. S. M. Slakter, J. A. M. Soren-
son, D. C. Orlock, Curr. Opin.
Ophthalmol. 1993, 4, 3–6.
[13] M. Hope-Ross, L. A. Yannuzzi,
E. S. Gragoudas, D. R. Guyer,
J. S. Slakter, J. A. Sorenson, S.
Krupsky, D. A. Orlock, C. A. Pu-
liafito, Ophthalmology 1994, 101,
529–533.
[14] L. Strekowski, M. Lipowska, G.
Patonay, J. Org. Chem. 1992, 57,
4578–4580.
[15] N. Narayanan, G. Patonay, J. Org.
Chem. 1995, 60, 2391–2395.
ꢀ
ꢀ
Figure 4. Absorption (solid line) and emission (dash line) spectra of a) dendron dye ACTHNUTGRNEG(UN COOH)6 NO2 (21),
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
b) dendron dye ACHTUNGTRENNUNG(COOH)9 (25), c) dendron dye PEG ACHTUGNTERN(NUGN COOtBu)9 (28), and d) dendron dye PEG (N3)9 (30).
[16] J. H. Flanagan, S. H. Khan, S.
Menchen, S. A. Soper, R. P.
Hammer, Bioconjugate Chem.
1997, 8, 751–756.
[17] E. Sasaki, H. Kojima, H. Nishimatsu, Y. Urano, K. Kikuchi, Y.
Hirata, T. Nagano, J. Am. Chem. Soc. 2005, 127, 3684–3685.
[18] C. Li, T. R. Greenwood, Z. M. Bhujwalla, K. Glunde, Org. Lett.
2006, 8, 3623–3626.
[19] Y. Chen, A. Gryshuk, S. Achilefu, T. Ohulchansky, W. Potter, T.
Zhong, J. Morgan, B. Chance, P. N. Prasad, B. W. Henderson, A.
Oseroff, R. K. Pandey, Bioconjugate Chem. 2005, 16, 1264–1274.
[20] S. A. Hilderbrand, K. A. Kelly, R. Weissleder, C.-H. Tung, Bioconju-
gate Chem. 2005, 16, 1275–1281.
[21] W. Wang, S. Ke, S. Kwon, S. Yallampalli, A. G. Cameron, K. E.
Adams, M. E. Mawad, E. M. Sevick-Muraca, Bioconjugate Chem.
2007, 18, 397–402.
[22] C. Bouteiller, G. Clave, A. Bernardin, B. Chipon, M. Massonneau,
P.-Y. Renard, A. Romieu, Bioconjugate Chem. 2007, 18, 1303–1317.
[23] H. Lee, J. C. Mason, S. Achilefu, J. Org. Chem. 2006, 71, 7862–7865.
[24] X. Peng, F. Song, E. Lu, Y. Wang, W. Zhou, J. Fan, Y. Gao, J. Am.
Chem. Soc. 2005, 127, 4170–4171.
[25] B. Tang, H. Huang, K. Xu, L. Tong, G. Yang, X. Liu, L. An, Chem.
Commun. 2006, 3609–3611.
[26] K. Kiyose, H. Kojima, Y. Urano, T. Nagano, J. Am. Chem. Soc.
2006, 128, 6548–6549.
Figure 5. Results obtained for the proliferation assays carried out on
human brain glioblastoma cells (T98G cell line) incubated for 72 h, in the
presence of 0.5, 1.0, 5.0, 10, and 50 mm of 30. Y-axis values correspond to
percentage of living cells (the lowest concentration of the dye–dendron
conjugate was considered as 100% of living cells). For each concentra-
tion, the proliferation assays were carried out in triplicate and the grey
columns represent the average of the three assays.
[27] B. Tang, L. J. Cui, K. H. Xu, L. L. Tong, G. W. Yang, L. G. An,
ChemBioChem 2008, 9, 1159–1164.
[28] K. Kiyose, S. Aizawa, E. Sasaki, H. Kojima, K. Hanaoka, T. Terai,
Y. Urano, T. Nagano, Chem. Eur. J. 2009, 15, 9191–9200.
[29] A. B. Descalzo, K. Rurack, Chem. Eur. J. 2009, 15, 3173–3185.
[30] K. Yoon, P. Goyal, M. Weck, Org. Lett. 2007, 9, 2051–2054.
[31] P. Goyal, K. Yoon, M. Weck, Chem. Eur. J. 2007, 13, 8801–8810.
[32] C. Ornelas, M. Weck, Chem. Commun. 2009, 5710–5712.
[33] C. Ornelas, J. Broichhagen, M. Weck, J. Am. Chem. Soc. 2010, 132,
3923–3931.
[1] S. M. Ametamey, Chem. Rev. 2008, 108, 4036–4036.
[2] M. S. T. GonÅalves, Chem. Rev. 2009, 109, 190–212.
[3] A. Louie, Chem. Rev. 2010, 110, 3146–3195.
[4] E. Terreno, D. D. Castelli, A. Viale, S. Aime, Chem. Rev. 2010, 110,
3019–3042.
[5] J. P. Celli, B. Q. Spring, I. Rizvi, C. L. Evans, K. S. Samkoe, S.
Verma, B. W. Pogue, T. Hasan, Chem. Rev. 2010, 110, 2795–2838.
[6] D. E. Reichert, M. J. Welch, Coord. Chem. Rev. 2001, 212, 111–131.
[34] C. C. Lee, J. A. MacKay, J. M. J. Frꢃchet, F. C. Szoka, Nat. Biotech-
nol. 2005, 23, 1517–1526.
3628
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3619 – 3629

A Promising Strategy for Imaging in the Near-Infrared Region
FULL PAPER
[35] J. B. Wolinsky, M. W. Grinstaff, Adv. Drug Delivery Rev. 2008, 60,
1037–1055.
[36] R. K. Tekade, P. V. Kumar, N. K. Jain, Chem. Rev. 2009, 109, 49–87.
[37] D. Astruc, E. Boisselier, C. Ornelas, Chem. Rev. 2010, 110, 1857–
1959.
[38] M. V. Backer, T. I. Gaynutdinov, V. Patel, A. K. Bandyopadhyaya,
B. T. S. Thirumamagal, W. Tjarks, R. F. Barth, K. Claffey, J. M.
Backer, Mol. Cancer Ther. 2005, 4, 1423–1429.
[48] M. Meldal, C. W. Tornoe, Chem. Rev. 2008, 108, 2952–3015.
[49] G. Franc, A. Kakkar, Chem. Commun. 2008, 5267–5276.
[50] J. D. Megiatto, D. I. Schuster, J. Am. Chem. Soc. 2008, 130, 12872–
12873.
[51] N. J. Agard, J. A. Prescher, C. R. Bertozzi, J. Am. Chem. Soc. 2004,
126, 15046–15047.
[52] N. J. Agard, J. M. Baskin, J. A. Prescher, A. Lo, C. R. Bertozzi, Acs
Chem. Biol. 2006, 1, 644–648.
[39] D. A. Tomalia, L. A. Reyna, S. Svenson, Biochem. Soc. Trans. 2007,
35, 61–67.
[53] E. M. Sletten, C. R. Bertozzi, Angew. Chem. 2009, 121, 7108–7133;
Angew. Chem. Int. Ed. 2009, 48, 6974–6998.
[40] H. Kobayashi, Y. Koyama, T. Barrett, Y. Hama, C. A. S. Regino,
I. S. Shin, B.-S. Jang, N. Le, C. H. Paik, P. L. Choyke, Y. Urano, ACS
Nano 2007, 1, 258–264.
[41] A. Almutairi, W. J. Akers, M. Y. Berezin, S. Achilefu, J. M. J. Frꢃ-
chet, Mol. Pharm. 2008, 5, 1103–1110.
[54] G. R. Newkome, H. J. Kim, C. N. Moorefield, H. Maddi, K. S. Yoo,
Macromolecules 2003, 36, 4345–4354.
[55] L. Skattebol, S. Solomon, Org. Synth. 1969, 49, 35.
[56] A. B. Neef, C. Schultz, Angew. Chem. 2009, 121, 1526–1529; Angew.
Chem. Int. Ed. 2009, 48, 1498–1500.
[42] A. Almutairi, S. J. Guillaudeu, M. Y. Berezin, S. Achilefu, J. M. J.
Frꢃchet, J. Am. Chem. Soc. 2008, 130, 444–445.
[43] E. I. Altinoglu, T. J. Russin, J. M. Kaiser, B. M. Barth, P. C. Eklund,
M. Kester, J. H. Adair, ACS Nano 2008, 2, 2075–2084.
[57] E. E. Jelley, Nature 1936, 138, 1009–1010.
[58] W. West, B. H. Carroll, J. Chem. Phys. 1951, 19, 417–427.
[59] L. G. S. Brooker, F. L. White, D. W. Heseltine, G. H. Keyes, S. G.
Dent, E. J. VanLare, J. Photogr. Sci. 1953, 1, 173.
[60] E. G. McRae, M. Kasha, J. Chem. Phys. 1958, 28, 721.
[61] M. Kasha, H. R. Rawls, M. A. El-Bayoumi, Pure Appl. Chem. 1965,
11, 371–392.
¨
[44] R. K. Iha, K. L. Wooley, A. M. Nystrom, D. J. Burke, M. J. Kade,
C. J. Hawker, Chem. Rev. 2009, 109, 5620–5686.
[45] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056–2075; Angew. Chem. Int. Ed. 2001, 40, 2004–2021.
[46] C. W. Tornoe, C. Christensen, M. Meldal, J. Org. Chem. 2002, 67,
3057–3064.
[62] M. Wittmann, F. Rotermund, R. Weigand, A. Penzkofer, Appl.
Phys. B 1998, 66, 453–459.
[47] C. Ornelas, J. R. Aranzaes, E. Cloutet, S. Alves, D. Astruc, Angew.
Chem. 2007, 119, 890–895; Angew. Chem. Int. Ed. 2007, 46, 872–
877.
Received: August 6, 2010
Published online: February 17, 2011
Chem. Eur. J. 2011, 17, 3619 – 3629
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3629