Synthesis and properties of BF2-3,3′- dimethyldiarylazadipyrromethene near-infrared fluorophores

Article abstract of DOI:10.1021/ol401434c

The first synthesis of both organic and aqueous soluble BF₂ chelated 3,3′-dimethyl-5,5′-diarylazadipyrromethenes has been achieved. The fluorophores are emissive in organic and aqueous solvents with high quantum yields in the key biological near-infrared (NIR) spectral region of 675-700 nm. Following efficient cellular uptake from aqueous media the fluorophore can be readily visualized with confocal microscopy.

Full text of DOI:10.1021/ol401434c

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis and Properties of
BF2‑3,30-Dimethyldiarylazadipyrromethene
Near-Infrared Fluorophores
Dan Wu and Donal F. O’Shea*
School of Chemistry and Chemical Biology, University College Dublin, Belfield,
Dublin 4, Ireland
donal.f.oshea@ucd.ie
Received May 21, 2013
ABSTRACT
The first synthesis of both organic and aqueous soluble BF2 chelated 3,30-dimethyl-5,50-diarylazadipyrromethenes has been achieved. The
fluorophores are emissive in organic and aqueous solvents with high quantum yields in the key biological near-infrared (NIR) spectral region of
675ꢀ700 nm. Following efficient cellular uptake from aqueous media the fluorophore can be readily visualized with confocal microscopy.
The ability to visualize in vitro and in vivo using molecular
near-infrared (NIR) fluorescence is a rapidly evolving rese-
arch field. Opportunities exist for the development of new
fluorescent platforms with strong absorption and emissions
in the low energy spectral regions (650ꢀ800 nm) that most
readily pass through biological tissues. Potential biomedical
uses for such techniques range from clinical diagnosis of
disease states to real time imaging for intraoperative guided
surgery.1 Our recent interest in this field stems from our
development of boron chelates of tetraarylazadipyrro-
methenes 3 (Scheme 1).2 These fluorophores exhibit NIR
absorption and emissions and have high fluorescence
quantum yields (0.3ꢀ0.4) and excellent photostability.3
Their synthesis derives from tetraarylazadipyrromethenes
2 which were first reported in 1943 as unexpected, deep blue,
colored products obtained from treatment of diaryl-γ-nitro
ketones 1 with ammonium formate (Scheme 1).4 In more
recent times we have developed new and optimized routes to
these compounds and specifically explored their boron
chelated derivatives as NIR fluorophores 3 (Scheme 1).5
Their advantageous photophysical properties have encour-
aged investigations for potential applications such as fluoro-
chromes,6 sensors/energy transfer cassettes,7 and donor/
acceptor conjugates for solar cells.8 Despite their growing
importance, the 3,30,5,50-tetraaryl substituted derivatives
(4) (a) Rogers, M. A. T. Nature 1943, 151, 504. (b) Rogers, M. A. T.
J. Chem. Soc. 1943, 590–596.
(1) (a) Escobedo, J. O.; Rusin, O.; Lim, S.; Strongin, R. M. Curr.
Opin. Chem. Biol. 2010, 14, 64–70. (b) Nguyen, Q. T.; Olson, E.;
Aguilera, T. A.; Jiamg, T.; Scadeng, M.; Ellies, L. G.; Tsien, R. Y. Proc.
Natl. Acad. Sci. U.S.A. 2010, 107, 4317–4322. (c) Razansky, D.; Distel,
M.; Vinegoni, C.; Ma, R.; Perrimon, N.; Koster, R. W.; Ntziachristos,
V. Nat. Photonics 2009, 3, 412–417. (d) Weissleder, R.; Pittet, M. J.
Nature 2008, 452, 580–589.
(5) (a) Gorman, A.; Killoran, J.; O’Shea, C.; Kenna, T.; Gallagher,
W. M.; O’Shea, D. F. J. Am. Chem. Soc. 2004, 126, 10619–10631. (b)
Grossi, M.; Palma, A.; McDonnell, S. O.; Hall, M. J.; Rai, D. K.;
Muldoon, J.; O’Shea, D. F. J. Org. Chem. 2012, 77, 9304–9312.
(6) Tasior, M.; O’Shea, D. F. Bioconj. Chem. 2010, 7, 1130–1133.
€
(7) (a) Jokic, T.; Borisov, S. M.; Saf, R.; Nielsen, D. A.; Kuhl, M.;
(2) (a) Palma, A.; Alvarez, L. A.; Scholz, D.; Frimannsson, D. O.;
Grossi, M.; Quinn, S. J.; O’Shea, D. F. J. Am. Chem. Soc. 2011, 133,
19618–19621. (b) Murtagh, J.; Frimannsson, D. O.; O’Shea, D. F. Org.
Lett. 2009, 11, 5386–5389. (c) Palma, A.; Tasior, M.; Frimannsson,
D. O.; Vu, T. T.; Meallet-Renault, R.; O’Shea, D. F. Org. Lett. 2009, 11,
3638–3641.
(3) (a) Batat, P.; Cantuel, M.; Jonusauskas, G.; Scarpantonio, L.;
Palma, A.; O’Shea, D. F.; McClenaghan, N. D. J. Phys. Chem. A 2011,
115, 14034–14039.
Klimant, I. Anal. Chem. 2012, 84, 6723–6730. (b) Loudet, A.; Bandichhor,
R.Wu, L.; Burgess, K. Tetrahedron 2008, 3642–3654. (c) Gao, L.;
Deligonul, N.; Gray, T. G. Inorg. Chem. 2012, 51, 7682–7688.
(8) (a) Flavin, K.; Lawrence, K.; Bartelmess, J.; Tasior, M.; Navio,
C.; Bittencourt, C.; O’Shea, D. F.; Guldi, D. M.; Giordani, S. ACS Nano
2011, 5, 1198–1206. (b) Leblebici, S. Y.; Catane, L.; Barclay, D. E.;
Olson, T.; Chen, T. L.; Ma, B. ACS Appl. Mater. Interfaces 2011, 3,
4469–4474. (c) Amin, A. N.; El-Khouly, E.; Subbaiyan, N. K.; Zandler,
M. E.; Fukuzumi, S.; D’Souza, F. Chem. Commun. 2012, 48, 206–208.
ꢀ
r
10.1021/ol401434c
XXXX American Chemical Society

Scheme 1. Synthesis of 3,30,5,50-Tetraarylazadipyrromethenes
Scheme 2. Previous Attempted Syntheses of 5,50-Diarylazadi-
pyrromethene 4
have been the only substitution pattern explored to date.
For some specific uses it would be beneficial to reduce the
lipophilicity of this fluorophore class, though this would
need to be achieved without loss of their valuable photo-
physical properties. One approach to achieving this would
be removal of the aryl rings from the 3 and 30 positions of 3
and would, in effect, reduce the molecular weight of the
fluorophore by approximately one-third (Scheme 1). An
examination of the literature reveals that the synthesis of
5,50-diaryl substrates 4 has been independently reported
three times via two different routes (Scheme 2).9aꢀc It has
also been reported that attempts to produce 4 by a similar
reaction used to produce 2 from the reaction of 4-nitro-
1-arylbutanone 5 with ammonium acetate failed to generate
4 (Scheme 2).4b,9a
The first claimed synthesis was by Knott in 1947 in
which the reaction of 3-benzoylpropionitrile 7 (Ar = Ph)
with hydroxylamine hydrochloride gave a dark blue co-
lored chromophore which was assigned the structure 4
(Scheme 2).9a Due to the era of this report, the structural
assignment was only based upon compound color and in
analogy with the earlier publication4 describing the synthe-
sis of tetraaryl derivatives 2. It was also shown that other
ammonium sources such as ammonium acetate failed to
give the product assigned as 4. This synthesis was later
reproduced by Boyer et al. in 1993 though no characteriza-
tion data was reported for the product other than an
absorbance λmax of 595 nm and, tellingly, attempts to
In our hands we were unable to obtain sufficient quan-
tities of the darkblue chromophore from the route utilizing
Grignard addition to succinonitrile,9c so we turned
our attention to the reaction of 3-benzoylpropionitrile
7 (Ar = Ph) with hydroxylamine hydrochloride.9a Follow-
ing the heating of 7 with NH2OH HCl in EtOH for 2 h, a
3
darkblue colored precipitate was obtained. Theprecipitate
was isolated and washed to remove excess hydroxylamine.
The resultant solid was bench stable but decomposed in a
matter of hours as a solution in THF, MeCN, acetone, or
MeOH. The absorption spectrum taken in CHCl3 corre-
sponded with the previously reported λmax value of 595 nm
(Figure S1 (Supporting Information, SI)).9b,c But the
1H NMR did not contain the expected signals for four
β-pyrrole protons and the 13C NMR showed 16 signals
where 8 would be predicted for structure 4 (SI). Addition-
ally, ES-MS analysis gave an [MþH]þ value of 314.1304
corresponding to a molecular formula of C20H15N3O in-
dicating that the synthesis of 4remains unsubstantiated (SI).
Due to the issues concerning 4 the 3,30-dimethyl-5,50-
diaryl substituted derivative 8 was chosen as a more
suitable alternative target structure that would still achieve
our goal of providing access to derivatives of this fluoro-
phore class with lower lipophilicity (Figure 1).
convert this material into a BF2 chelate using BF3 OEt2
3
and TEA failed, giving only “intractable mixtures”.9b In a
1992 report Bird et al. described access to 4 (Ar = Ph) by
the reaction of succinonitrile with phenylmagnesium bro-
mide followed by heating of the resulting crude extract
with ammonium chloride.9c Unfortunately again, the only
analytical data provided to confirm the assigned structure
as 4 was an absorption maximum of 595 nm. Our insight
gained from a mechanistic investigation into the synthesis
of the tetraaryl derivatives 25b and the lack of character-
ization data raised doubts about these results, so repeat
syntheses of 4 from 6 and 7 were attempted.
Figure 1. 3,30-Dimethyl-5,50-diarylazadipyrromethene.
The route selected was via the 3-methyl-4-nitro-1-aryl-
butan-1-ones 12, which could be accessed in three steps
from crotonaldehyde 9 (Scheme 3). 1,2-Addition to croto-
naldehyde with phenyllithium and p-methoxyphenylmag-
nesium bromide gave the allylic alcohols 10a,b, and
subsequent oxidation with MnO2 generated the R,β-un-
saturated ketones 11a,b.
(9) (a) Knott, E. B. J. Chem. Soc. 1947, 1196–1201. (b) Sathyamoorthi,
G.; Soong, M.-L.; Ross, T. W.; Boyer, J. H. Heteroatom. Chem. 1993, 4,
603–608. (c) Bird, C. W.; Jiang, L. Tetrahedron Lett. 1992, 33, 7253–
7254.
Conjugate addition of nitromethane to 11a,b in EtOH
with diethylamine (DEA) gave the nitro-ketones 12a,b in
B
Org. Lett., Vol. XX, No. XX, XXXX

Conversion of 8a,b to their corresponding BF2 chelates
13a,b was achieved in excellent yields by their treatment
with BF3 etherate and N,N-diiopropylethylamine (DIPEA)
in CH2Cl2 (Scheme 4). This boron chelation restricts rota-
tion about the central bridging CꢀNꢀC bonds which limits
excited state quenching and provided the first diaryl sub-
stituted azadipyrromethene fluorophores. To expand the
aryl substituent pattern the bismethoxy substituted 8b was
monodemethylated with BBr3 giving 8c which upon chela-
tion provided the monophenol-monomethoxyphenyl sub-
stituted derivative 13c (Scheme 4).
Scheme 3. Synthesis of 3-Methyl-4-nitro-1-arylbutan-1-ones
The structure of 13b was confirmed with X-ray analysis
of a single crystal grown by slow evaporation of a dichloro-
methane solution (Scheme 4, inset). Analysis of the crystal
structure of 13b confirmed the overall conjugated nature of
the fluorophore with a planar 5ꢀ6ꢀ5 fused ring system.
The methoxy-substituted aryl rings had torsion angles of
23.9(3)° and 34.2(3)° with the pyrrole rings, indicating a
strong electronic interaction with the central fluorophore
unit which would impact upon absorption and emission
wavelengths.
good yields (Scheme 3). Substrate 12b was selected to opti-
mize reaction conditions for its transformation into azadi-
pyrromethene 8b (Table 1). Screening of ammonia sources
showed that heating 8b in EtOH with NH2OH HCl, NH4Cl,
3
(NH4)2SO4, and (NH4)2CO3 all failed to give any product
formation (entries 1ꢀ4). In contrast, ammonium formate
and ammonium acetate both gave the desired product 8b in
15 and 17% yield respectively after 10 h of reflux (entries 5, 6).
A further examination of reaction times and solvents
(EtOH, MeOH, iPrOH, BuOH) identified reflux in MeOH
for 10 h as optimal conditions providing a 28% yield of 8b
(entries 8ꢀ11). Applying these conditions to 12a gave the
corresponding product 8a in a 19% yield (entries 12, 13).
NMR and HRMS analyses for 8a,b were consistent with
these structures, and revealingly, the absorption spectrum
of 8a (λmax 561 nm) differed significantly from the com-
pound previously claimed to be structure 4 (Figure S1 (SI)).
Scheme 4. Synthesis of NIR Fluorophores 13aꢀd (Inset: X-ray
crystal structure of 13b with thermal ellipsoids drawn at 50%
probability level)
Table 1. Investigation of 3,30-Dimethyldiarylazadipyrromethene
Formation
sub-
NH3
time
(h)
yield
(%)
entry
strate
source
solvent
8
1
12b
12b
12b
12b
12b
12b
12b
12b
12b
12b
12b
12a
12a
NH2OH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
MeOH
iPrOH
BuOH
EtOH
MeOH
20
20
20
20
10
10
20
6
8b
8b
8b
8b
8b
8b
8b
8b
8b
8b
8b
8a
8a
0
2
NH4Cl
0
3
(NH4)2SO4
(NH4)2CO3
HCOONH4
NH4OAc
NH4OAc
NH4OAc
NH4OAc
NH4OAc
NH4OAc
NH4OAc
NH4OAc
0
4
0
5
15
17
10
26
28
21
22
19
19
6
7
8
Using CHCl3 as a representative solvent the absorption
spectra of 13aꢀc showed sharp bands (fwhm between 35
and 43 nm) between 618 and 651 nm (Table 2). The aryl
p-OMe substituents of 13b had a pronounced bathochro-
mic shift of over 30 nm when compared to the unsubsti-
tuted phenyl derivative13a (entries 1, 2). Emission maxima
9
10
4
10
11
12
13
1
6
10
Org. Lett., Vol. XX, No. XX, XXXX
C

Table 2. UVꢀvis and Fluorescence Properties of 13aꢀca
Figure 3. Left: Confocal imaging of 13d in HeLa cells after 1 h of
incubation. Right: Overlay with cell nuclei stained blue fluo-
rescent with DAPI. Scale bar = 10 μm.
pattern 13c was converted into the water solubilized 13d by
its reaction with 1,3-propanesultone (Scheme 4).
λ
max abs
fwhm
(nm)b
ext. coeff.
λ
max flu
entry
13
(nm)
(Mꢀ1 cmꢀ1
)
(nm)
Φflu
c
Absorption and emission spectra of 13d taken in phos-
phate buffered saline (PBS) were encouraging with minor
broadening observed in the UVꢀvisible spectrum (λmax
651 nm) and an emission maximum at 691 nm (Figure 2).
Incubation of 13d with HeLa cells in Dulbecco’s modified
eagle’s medium (DMEM) for 60 min and confocal imaging
showed efficient internalization of the fluorophore (Figure 3).
Confocal Z-stack imaging of individual cellular focal
planes showed localization of 13d restricted to the cytosol
(Figure S2 (SI)).
1
2
3
a
b
c
618
648
651
35
39
39
64 000
81 000
81 000
642
678
679
0.41
0.44
0.45
a 13a (black), 13b (blue), 13c (red). In CHCl3. b Full width at
half-maximum height. c Compound 3a (Ar = Ph, Ar0 = p-MeOC6H4;
Φ = 0.36) used as a standard.3
In conclusion, we have described the first synthesis of
BF2 chelated 3,30-dimethyl-5,50-diarylazadipyrromethenes
which exhibit excellent NIR photophysical properties in
organic and aqueous solutions. These results are encoura-
ging indications that they could be adapted for in vitro and
in vivo NIR-fluorescence imaging. Development of routes
to further substitution patterns is ongoing. Literature
confusion with respect to claimed syntheses of 5,50-diaryl-
azadipyrromethenes has been clarified.
Acknowledgment. The authors are grateful to Science
Foundation Ireland for funding. W.D. thanks the China
€
Scholarship Council for a Ph.D. fellowship. H. Muller-
Bunz (University College Dublin) for X-ray crystallogra-
phy and S. Chueng (University College Dublin) for con-
focal imaging are also thanked.
Figure 2. Absorbance (blue) and fluorescence (red) spectra of
13d in PBS at pH 7.2 (solid lines) (λmax abs 651 nm, em 691 nm)
and DMEM (dashed lines) (λmax abs 654 nm, em 676 nm).
Supporting Information Available. All experimental
procedures, NMR spectra, X-ray data, and confocal
Z-stack images. This material is available free of charge
via the Internet at http://pubs.acs.org.
for 13b and c were at 678 and 679 nm with high quantum
yields of 0.44 and 0.45 respectively (Table 2, entries 2, 3).
Examination of abs and flu properties in various organic
solvents showed little solvent effects (Table S1, (SI)). To
complete the first phase of our work on this new substitution
The authors declare the following competing financial interest(s): A
patent application has been filed by DOS on azadipyrromethene based
NIR fluorophores (PCT/EP2010/065991).
D
Org. Lett., Vol. XX, No. XX, XXXX