A practical approach for the preparation of monofunctional azulenyl squaraine dye

Article abstract of DOI:10.1016/S0040-4039(03)00819-0

The synthesis of monofunctional azulenyl squaraine dye NIRQ₇₀₀ is described. The essential azulene intermediate 3, 1-(methoxycarbonyl)-2-methylazulene, was achieved via [8+2] cycloaddition between lactone 2, 2H-3-methoxycarbonyl-cyclohepta[b]furan-2-one, and the in situ generated vinyl ethers under high temperature and pressure conditions. Methylation on the cycloheptatriene ring of 2-methyl azulene 6 via Meisenheimer-type intermediate following Schrott's method formed the carboxylic acid intermediate 9, 3-(2-methyl-azulen-4-yl)-propionic acid. Condensation of 9 with squaric acid provided the title compound NIRQ₇₀₀ at moderate yields. The non-fluorescent squaraine dye NIRQ₇₀₀ absorbed in a 600-700 nm range and potentially can be used to quench a number of available NIR fluorochromes in order to extend the spectrum of biological quenching assays.

Full text of DOI:10.1016/S0040-4039(03)00819-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 3975–3978
A practical approach for the preparation of monofunctional
azulenyl squaraine dye
Wellington Pham, Ralph Weissleder and Ching-Hsuan Tung*
Center for Molecular Imaging Research, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA 02129,
USA
Received 25 February 2003; revised 18 March 2003; accepted 18 March 2003
Abstract—The synthesis of monofunctional azulenyl squaraine dye NIRQ700 is described. The essential azulene intermediate 3,
1-(methoxycarbonyl)-2-methylazulene, was achieved via [8+2] cycloaddition between lactone 2, 2H-3-methoxycarbonyl-cyclohep-
ta[b]furan-2-one, and the in situ generated vinyl ethers under high temperature and pressure conditions. Methylation on the
cycloheptatriene ring of 2-methyl azulene 6 via Meisenheimer-type intermediate following Schrott’s method formed the carboxylic
acid intermediate 9, 3-(2-methyl-azulen-4-yl)-propionic acid. Condensation of 9 with squaric acid provided the title compound
NIRQ700 at moderate yields. The non-fluorescent squaraine dye NIRQ700 absorbed in a 600–700 nm range and potentially can be
used to quench a number of available NIR fluorochromes in order to extend the spectrum of biological quenching assays. © 2003
Elsevier Science Ltd. All rights reserved.
Fluorescence resonance-energy transfer has long been
used to study various biological events in vitro, such as
protease kinetics1 or nucleic acid hybridization.2 To
impart high signal changes in protease assays, a fluores-
cent donor and a non-fluorogenic chromophore are
often covalently attached to the ends of a specific
enzyme substrate. Resonance-energy transfer from the
excited state of a fluorophore to a non-fluorogenic
chromophore results in quenching of a fluorescence
signal. Upon proteolytic cleavage of the substrate by
enzymes, a fluorescent dye and a quencher are sepa-
rated from each other, resulting in fluorescence.1
Fluorochromes typically used for the above assays
fluoresce in the visible range (u=400–600 nm), so that
the fluorescence signal can be conveniently visualized
by spectrophotometers or by fluorescence microscopes.
However, light in this range is not ideal for many in
vitro and in vivo applications, because of autofluores-
cence in the visible spectrum, and because of strong
absorption of photons by tissue and blood. Recently
various near-infrared (NIR, u=700–900 nm) probes
have shown great promise for in vivo imaging of vari-
ous target molecules or biological events, such as recep-
tors,3–5 tumor associated proteases,6–8 osteolastic
activity9 and thrombin.10
We have recently reported the application of a near-
infrared fluorescence quencher, NIRQ750 (Fig. 1) for
caspase activity detection. This probe was synthesized
by dimerizing of commercially available guaiazulene
with squaric acid.11 During the course of work we
found that the electronic donating isopropyl moiety on
guaiazulene ring contributes to the bathochromic shift
of the dye in the NIR region (u=750 nm). In a contin-
ued work we would like to look for more applications
of azulenyl squaraine dyes as quenchers for commercial
NIR fluorochromes in the 650–700 nm range. It is
anticipated that substituting the guaiazulene half dye of
NIRQ750 by an azulene will hypsochromically shift the
absorbance wavelength by about 40–50 nm (Du).
The synthesis of NIRQ700 was started with a known
procedure12 by reacting the activated tropolone 1 with
Figure 1. Molecular structure of the conjugatable NIR
quenching, non-fluorescent azulenyl squaraine dyes NIRQ700
* Corresponding author. Tel.: (617) 726-5779; fax: (617) 726-5708;
e-mail: tung@helix.mgh.harvard.edu
and NIRQ750
.
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00819-0

3976
W. Pham et al. / Tetrahedron Letters 44 (2003) 3975–3978
Scheme 1. Preparation of azulene derivatives and attempted synthesis of a azulenyl squaraine dye.
dimethyl malonate in the presence of sodium methoxide
to obtain lactone 2 (Scheme 1). The crude product was
purified by recrystallization using ethanol and
dichloromethane to provide 84% yield. An azulene ring
was prepared via [8+2] cycloaddition of lactone 2 with
vinyl ether—a product of thermolysis of acetals.13,14
This reaction was temperature and solvent dependent.
In the absence of solvent, only a black tar-like decom-
position material was recovered. The expected brown-
ish-red liquid product 3 was obtained by heating 2 with
2,2-dimethoxypropane in anhydrous toluene at 200°C
under pressure.14
were able to introduce an alkyl chain into the aromatic
ring of compound 6 via Friedel–Crafts alkylation using
similar conditions (data not shown). For compound 7,
perhaps the electron density in the five-membered ring
of azulene now delocalized to the squaryl ring making
it inactive for alkylation.
An alternative approach of preparing monofunctional
squaraine azulene analog was started from methylation
on the seven-membered ring of compound 6 with
methyllithium at room temperature in diethyl ether
(Scheme 2). The mixture was refluxed overnight to form
the azulenate ions of the Meisenheimer-type
intermediate15,16 as the color changed from deep blue to
a pale yellow suspension. Addition of methanol at
−70°C provided a colorless solution which subsequently
dehydrogenated with p-chloranil gave a dark blue oil
2,4-dimethyl azulene 8 at 52% efficiency.17,20 The
absorbance maximum of 2,4-dimethyl azulene 8 was
similar to that of intermediate 4. A first trial of depro-
tonation of the methyl group with t-BuOK in THF
under conditions described by Song et al. was not
successful.18 Using Schrott’s method,19 treating 2,4-
dimethyl azulene 8 with n-BuLi at −40°C in the pres-
ence of diisopropyl amine, the nucleophilic substitution
was carried out by adding bromoacetic acid dropwise
to the reaction mixture at −40°C, at the end of the
The methyl ester of 3 was removed by saponification
using potassium hydroxide and the product was
purified by recrystallization with methanol and chloro-
form to yield the pink solid acid 4. Subsequent conden-
sation was carried out by refluxing compound 4 with
squaric acid; unfortunately, instead of obtaining the
expected azulenyl squaraine product 5 with the intact
free carboxylic acid, an unexpected product 7 was
observed. The absorption of 7 (green color) found
about 180 nm higher than compound 4 in acetonitrile
clearly demonstrating that the condensation process
was successful (Fig. 2). However, the evidence of this
decarboxylation was provided in the 1H NMR and
mass spectrometry. We thought that the 2-methyl
group on the azulene five-membered ring may serve as
an electron-donating group that foster the decarboxyla-
tion process. We then treating compound 3 in acidic
condition using anhydrous phosphoric acid to support
the hypothesis. The decarboxylation was smoothly
completed in 5 min at 100°C and the brownish red
solution of compound 3 turned to purple color of
compound 6 quantitatively.
In order to regain a functional group for future conju-
gation, an insertion of a carboxylic moiety after con-
densation was attempted. The 2-methyl azulene 6 was
first condensed with squaric acid to yield 7. But, the
subsequent Friedel–Crafts alkylation of introducing the
4-carboxylic group into the aromatic ring by succinic
anhydride was not successful. In a separate reaction, we
Figure 2. Absorption spectra of azulene derivatives in aceto-
nitrile.

W. Pham et al. / Tetrahedron Letters 44 (2003) 3975–3978
3977
Scheme 2. Synthesis of azulenyl squaraine dye, NIRQ700
.
reaction the suspension was acidified with 2 M HCl and
the carboxylic acid product 920 was extracted by diethyl
ether. Condensation between 9 and squaric acid under
similar conditions to those described earlier11 provided
8. Weissleder, R.; Tung, C.-H.; Mahmood, U.; Bogdanov,
A., Jr. Nat. Biotechnol. 1999, 17, 375–378.
9. Zaheer, A.; Lenkinski, R. E.; Mahmood, A.; Jones, A.
G.; Cantley, L. C.; Frangioni, J. V. Nat. Biotechnol. 2001,
19, 1148–11454.
20
the mono- and di-ester, NIRQ700 and 10 at 26 and
38% yield, respectively. The prepared NIRQ700 has an
absorption maximum at umax=700 nm and a broad
absorbance between u=600 and 750 nm (Fig. 2).
10. Tung, C. H.; Gerszten, R. E.; Jaffer, F. A.; Weissleder,
R. Chembiochem. 2002, 3, 207–211.
11. Pham, W.; Weissleder, R.; Tung, C.-H. Angew. Chem.,
Int. Ed. Engl. 2002, 41, 3659–3662.
12. Yokota, T.; Yanagisawa, T.; Kosakai, K.; Wakabayashi,
S.; Tomiyama, T.; Yasunamim, M. Chem. Pharm. Bull.
1994, 865–871.
13. Nozoe, T.; Wakabayashi, H.; Ishikawa, S.; Wu, C. P.;
Yang, P. W. Heterocycles 1990, 31, 17–22.
14. Pham, W.; Weissleder, R.; Tung, C.-H. Tetrahedron Lett.
2001, 43, 19–20.
15. Hafner, K.; Hartung, J.; Syren, C. Tetrahedron 1992, 48,
4879–4884.
In summary, the synthesis reported above describes a
novel approach of preparing a monofunctional azulenyl
squaraine dye NIRQ700. Similar to the previously
reported NIRQ750, the newly synthesized compound
NIRQ700 has no fluorescence, and is expected to be an
efficient
FRET
quencher
for
600–750
nm
fluorochromes.13 It is noteworthy that the absorption
could be tuned conveniently by replacing the substitu-
tion group on the azulene rings.
16. McDonald, R. N.; Petty, H. E.; Wolfe, N. L.; Paukstelis,
J. V. J. Org. Chem. 1974, 39, 1877–1887.
17. Chen, S. L.; Klein, R.; Hafner, K. Eur. J. Org. Chem.
1998, 423–433.
18. Song, J.; Hansen, H.-J. Helv. Chim. Acta 1999, 82, 309–
314.
Acknowledgements
This research was supported by NIH P50-CA86355,
R33-CA88365, and NSF BES-0119382.
19. Schrott, W.; Neumann, P.; Brosius, S.; Barzynski, H.;
Schomann, K. D.; Kuppelmaier, H. In Eur. Pat. Appl.
(BASF A.-G., Fed. Rep. Ger.): Ep 310080, 1989, p. 40.
20. Data and procedure for representative products:
2. Rf=0.48 (9.5:0.2 CH2Cl2/MeOH); 1H NMR (400
MHz, CDCl3) l 3.95 (s, 3H), 7.34 (ddd, J=3.3, 3.8, 3.3
Hz, 1H), 7.50 (t, J=4.1 Hz, 2H), 7.64 (m, 1H), 8.86 (d,
J=11.3 Hz, 1H); 13C NMR (200 MHz, CDCl3) l 51.6,
96.3, 119.2, 130.6, 134.0, 136.1, 139.6, 154.6, 158.6, 163.8,
165.1; FAB-MS: calcd (M+H)+ (C11H9O4) 205.18, found
205.11; elemental anal. calcd for C11H8O4: C, 64.71; H,
References
1. Wang, G. T.; Matayoshi, E.; Huffaker, H. J.; Krafft, G.
A. Tetrahedron Lett. 1990, 31, 6493–6496.
2. Sokol, D. L.; Zhang, X.; Lu, P.; Gewirtz, A. M. Proc.
Natl. Acad. Sci. USA 1998, 95, 11538–11543.
3. Achilefu, S.; Dorshow, R. B.; Bugaj, J. E.; Rajagopalan,
R. Invest. Radiol. 2000, 35, 479–485.
4. Becker, A.; Hessenius, C.; Licha, K.; Ebert, B.;
Sukowski, U.; Semmler, W.; Wiedenmann, B.;
Grotzinger, C. Nat. Biotechnol. 2001, 19, 327–331.
5. Licha, K.; Hessenius, C.; Becker, A.; Henklein, P.; Bauer,
M.; Wisniewski, S.; Wiedenmann, B.; Semmler, W. Bio-
conjugate Chem. 2001, 12, 44–50.
3.95. Found: C, 64.26; H, 3.89; UV–vis (MeCN) umax
=
400 nm.
3. Rf=0.63 (CH2Cl2); 1H NMR (400 MHz, CDCl3) l
2.83 (s, 3H), 3.98 (s, 3H), 7.13 (s, 1H), 7.39 (t, J=11.1
Hz, 1H), 7.50 (t, J=11.1 Hz, 1H), 7.69 (t, J=11.1 Hz,
1H), 8.28 (d, J=10.6 Hz, 1H), 9.48 (d, J=10.6 Hz, 1H);
13C NMR (200 MHz, CDCl3) l 18.1, 50.8, 86.2, 115.1,
6. Bremer, C.; Bredow, S.; Mahmood, U.; Weissleder, R.;
Tung, C. H. Radiology 2001, 221, 523–529.
7. Tung, C. H.; Mahmood, U.; Bredow, S.; Weissleder, R.
Cancer Res. 2000, 60, 4953–4958.

3978
W. Pham et al. / Tetrahedron Letters 44 (2003) 3975–3978
120.2, 126.8, 127.7, 135.8, 137.2, 142.1, 143.2, 154.1,
NMR (200 MHz, CDCl3) l 2.65 (s, 3H), 2.85 (s, 3H),
7.02–7.15 (m, 3H), 7.20 (d, J=6.4 Hz, 1H), 7.42 (t,
J=9.8 Hz, 1H), 8.17 (d, J=9.5 Hz, 1H); 13C NMR (200
MHz, CDCl3) l 16.6, 24.3, 116.5, 118.7, 122.0, 126.5,
134.5, 138.1, 140.7, 144.1, 148.4; MALDI-TOF MS calcd.
(M+H)+ (C12H13) 157.23, found 157.42; UV–vis (MeCN)
umax=548 nm.
166.6; MALDI-TOF MS: calcd (M+H)+ (C13H13O2)
201.23, found 201.22; UV–vis (MeCN) umax=524 nm. 4.
1
Rf=0.35 (CH2Cl2); H NMR (200 MHz, THF-d8) l 2.82
(s, 3H), 7.13 (s, 1H), 7.38 (t, J=9.46 Hz, 1H), 7.43 (t,
J=2.44 Hz, 1H), 7.69 (t, J=10.1 Hz, 1H), 8.30 (d,
J=9.76 Hz, 1H), 9.59 (d, J=10.1 Hz, 1H); 13C NMR (50
MHz, THF-d8) l 18.3, 116.6, 120.9, 127.4, 128.0, 136.5,
136.9, 137.9, 143.2, 144.2, 155.0, 167.4; MALDI-TOF
MS: calcd (M+H)+ (C12H11O2) 187.22, found 187.19;
UV–vis (MeCN) umax=522 nm.
9. A 2.5 M solution of n-BuLi (846 mL, 2.12 mmol) was
added dropwise to a solution of 2,4-dimethyl azulene 8
(220 mg, 1.41 mmol) and diisopropyl amine (316 mL, 2.26
mmol) in ether (10 mL) at −40°C then reaction mixture
was warmed up to 0°C for 30 min. A solution of bro-
moacetic acid (195.90 mg, 1.41 mmol) in ether (1 mL)
was added dropwise to the reaction mixture after cooling
the flask to −40°C. The reaction was stirred overnight as
the temperature slowly raised up to room temperature.
The reaction mixture was poured onto ice-cold water (30
mL) and the unreacted starting material 8 was extracted
with ether (2×10 mL). The aqueous phase was acidified
by 2 M HCl (20ml), and extracted with ether (20 ml). The
organic solution was collected, dried over MgSO4,
filtered, and concentrated to a wet purple residue. Chro-
matography with CH2Cl2 afforded 91 mg (30%) of a dark
purple oil: Rf=0.1 (CH2Cl2); MALDI-TOF MS calcd.
(M+H)+ (C14H15O2) 215.26, found 215.30.
NIRQ700. 4-Azulenepropanoic acid, 2-methyl 9 (120 mg,
0.56 mmol) and squaric acid (31.90 mg, 0.28 mmol) were
refluxed for 5 h in a solvent mixture containing toluene
(20 mL) and n-BuOH (20 mL), accompanied by water
removal using Dean–Stark apparatus. The solvents were
removed by rotavapor then the residue was redissolved in
chloroform. Chromatography with 9:1 CH2Cl2/acetone
afforded an oily green material (41 mg, 26%). Rf=0.34
(9.5:0.5 CHCl3/acetone); 1H NMR (200 MHz, CDCl3,
CD3OD) l 0.93 (t, J=7.3 Hz, 3H), 1.33 (m, J=7.9 Hz,
2H), 1.61 (q, J=6.4 Hz, 2H), 2.84 (t, J=7.0 Hz, 4H),
3.14 (s, 6H), 3.50 (t, J=7.9 Hz, 4H), 4.11 (t, J=6.7 Hz,
2H), 7.44 (s, 2H), 7.77 (m, 6H), 10.29 (d, d, J=4.27, 4.27,
2H); 13C NMR (50 MHz, CDCl3, CD3OD) l 13.7, 19.2,
29.8, 30.8, 34.1, 35.2, 65.2, 125.4, 131.0, 134.8, 139.0,
141.4, 146.8, 148.3, 149.4, 149.9, 155.5, 173.1, 175.1,
182.7, 185.8; ES-HRMS calcd (M+H)+ (C36H35O6)
563.2433, found 563.2431; UV–vis (MeCN) umax=700
nm, m (CH2Cl2, cm−1 M−1)=84,000.
6. Rf=0.8 (9.5:0.5 hexane/Et2O); 1H NMR (400 MHz,
CDCl3) l 2.67 (s, 3H), 7.14 (t, J=9.8 Hz, 2H), 7.18 (s,
2H), 7.47 (t, J=9.6 Hz, 1H), 8.16 (d, J=9.4 Hz, 2H); 13C
NMR (200 MHz, CDCl3) l 16.7, 118.3, 123.0, 134.1,
135.3, 140.7, 150.3; HRMS (CI) calcd M+ (C11H10)
142.2000, found 142.0781; UV–vis (MeCN) umax=560
nm.
7. Rf=0.7 (9.5:0.5 CH2Cl2/acetone); 1H NMR (200 MHz,
CDCl3) l 3.20 (s, 6H), 7.28 (s, 2H), 7.48 (m, 2H), 7.70
(m, 4H), 8.14 (d, J=9.5 Hz, 2H), 10.48 (m, 2H); 13C
NMR (50 MHz, 1:1 CDCl3:MeOD) l 19.0, 125.2, 128.1,
132.3, 132.8, 137.2, 140.2, 141.7, 147.4, 151.2, 156.5,
183.2, 185.3; MALDI-TOF MS: calcd (M+H)+
(C26H19O2) 363.43, found 363.36; UV–vis (MeCN) umax
=
700 nm, m (CH2Cl2, cm−1 M−1)=120,000.
8. A 1.4 M solution of MeLi (3.20 mL, 4.39 mmol) was
added to a flame-dried flask containing a solution of
2-methyl azulene 6 (520 mg, 3.66 mmol) in anhydrous
ether at room temperature. After refluxing overnight, the
resulting suspension was added with MeOH (2 mL) at
−70°C then with 2N HCl (10 mL) at room temperature.
The organic layer was separated, and the aqueous solu-
tion was extracted with ether (3×20 mL). The combined
organic solution was dried over MgSO4, and evaporated
under vacuum. The brown residue was redissolved in
benzene (10 mL), then p-chloranil (899 mg, 3.66 mmol)
was added portionally at room temperature. During this
time, the solution changed gradually from brown to
purple. After 48 h, the reaction was quenched with 1N
NaOH (20 mL). The organic layer was washed with
water, dried over MgSO4, filtered, and concentrated to a
purple oil. Chromatography with hexane afforded 300
mg (52%) of a dark purple oil: Rf=0.4 (hexane); 1H