Simple, high yielding synthesis of trifunctional fluorescent lanthanide chelates

Article abstract of DOI:10.1016/S0040-4039(01)00620-7

In an effort to produce strong fluorophores which are conjugated to biologically active substrates, we report a facile, high yielding synthesis of conjugable lanthanide chelates based on the cyclen (1,4,7,10-tetraazacyclododecane) macrocycle.

Full text of DOI:10.1016/S0040-4039(01)00620-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 3823–3825
Simple, high yielding synthesis of trifunctional fluorescent
lanthanide chelates
John M. M. Griffin, Anna M. Skwierawska, H. Charles Manning, John N. Marx and
Darryl J. Bornhop*
Department of Chemistry and Biochemistry, Texas Tech University, Lubbock, TX 79409-1061, USA
Received 2 February 2001; revised 4 April 2001; accepted 6 April 2001
Abstract—In an effort to produce strong fluorophores which are conjugated to biologically active substrates, we report a facile,
high yielding synthesis of conjugable lanthanide chelates based on the cyclen (1,4,7,10-tetraazacyclododecane) macrocycle. © 2001
Elsevier Science Ltd. All rights reserved.
It is the goal of this study to produce bioconjugable,
highly fluorescent, strong binding lanthanide chelates in
a simple and efficient manner. Existing synthetic
methodology for producing conjugable lanthanide
chelates has had the drawback of either requiring a
long and labored synthesis, production of multiple
active sites for conjugation1 or ligands lacking a suit-
able light harvesting moiety.2–4
veys the absorbed energy to a chelated lanthanide ion,
ionic fluorescence from the chelated lanthanide can be
obtained. This pathway not only results in emission
that is far removed from the absorption (Du]280 nm)
but is also responsible for their extremely long fluores-
cent lifetimes (0.5–3.5 ms), thus allowing detection with
inexpensive instrumentation and at very low levels of
background signal.11–14
The ability to link fluorophores to biologically active
species is a critical step toward making contrast agents
that can help elucidate cellular mechanisms and aid in
the identification of disease.5–7 While having shown
great utility, most of the currently available conjugable
fluorophores still have limitations regarding their spec-
troscopic properties.8,9 Photobleaching, overlapping
excitation and emission, and short lived fluorescence
limit these fluorophores adding to increased back-
ground noise and making gated detection rather expen-
sive and complicated. In addition, fluorophores with
more than one active conjugation site often give poor
yields of the desired product through cross-linking or
require multiple steps with protecting groups, increas-
ing the total time and cost of product production.10
Lanthanide chelates have been recently exploited as
optical imaging agents, due in part to the fact that they
have excellent spectroscopic properties as a result of a
unique luminescent pathway.11 By employing a molecu-
lar antenna to serve as a light harvesting moiety,
through which intramolecular energy transfer then con-
We report here the efficient, inexpensive synthesis, in
high yields, of a lanthanide chelating ligand based on
the cyclen (1,4,7,10-tetraazacyclododecnae) nucleus
which possesses a single carboxyl group for conjuga-
tion, a 6-substituted quinaldine light harvesting moiety,
and two phosphonic acid pendant arms for strong,
thermodynamically stable lanthanide complexation
(Fig. 1).
Selective substitution is paramount in this synthesis in
order to yield the type of molecule in Fig. 1 with just
one nitrogen linked to a light-harvesting moiety and a
O
P
N
F
HO
HO
N
N
N
N
OH
HOC
P
OH
O
O
1
Figure 1. Lanthanide chelating ligand possessing a single
conjugable acetic acid side-chain, two opposing phosphonic
acids for strong lanthanide complexation, and a 6-fluoro-
quinaldine light harvesting moiety for efficient lanthanide
sensitization.
Keywords: 1,4,7,10-tetraazacyclododecane; lanthanide; fluorescence;
conjugates; chelate.
* Corresponding author. E-mail: djbornhop@ttu.edu
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00620-7

3824
J. M. M. Griffin et al. / Tetrahedron Letters 42 (2001) 3823–3825
single carboxyl group on the opposite nitrogen. The
first attempt at monoalkylation with the quinaldine
derivative15a–c utilized 1,7-bis (methylene phosphonic
acid diethyl ester)-1,4,7,10-tetraazacyclododecane, 2a,
first described by Sherry,16 as a starting point (Scheme
1). However, the main product isolated was the unde-
sired disubstituted compound 3a with yields of the
monosubstituted compound 3b of 30%.17 Even when
employing two equivalents of 2a with one equivalent of
quinaldine antenna, the main product obtained was still
the disubstituted compound 3a.
conditions and resulting percentage yield of monosub-
stituted versus disubstituted products are summarized
in Table 1.
Alkylation of the remaining macrocyclic nitrogen in 3c
with a-chloro ethyl acetate placed the acetic acid conju-
gation unit on the ring to produce 4a. Deprotection of
the BOC groups was accomplished by catalytic transfer
hydrogenation utilizing 5% Pd/C in ethanol with cyclo-
hexene as the hydrogen donor to give 4b in up to 92%
yield. These conditions were much superior to
Sherry’s16 original conditions for similar deprotections.
Potential reduction of the quinoline ring19,20 was also
avoided.
Surprisingly, when utilizing 1,7-bis(benzyloxycarbonyl)-
1,4,7,10-tetraazacyclododecane, 2b,16 as the starting
point for the monosubtitution with the quinaldine
derivative, the desired monosubstituted product was
isolated as the main product. Selective monoalkylation
with the quinaldine light-harvesting moiety was per-
formed under fairly high dilution (0.04 M) conditions
with a 1:1 molar ratio of reactants. Simple chromato-
graphy gave the monosubstituted product 3c in 70%
yield and the disubstituted biproduct 3d in 10%
yield.18a,b This apparent high selectivity when using
benzyloxycarbonyl (BOC) protecting groups in the 1,7-
positions is likely due to a conformation that renders
one of the nitrogens inaccessible either before or after
the first substitution. Starting materials, reagents and
Next the bis methylene phosphonic acid pendant arms
were added via reaction of 4b with paraformaldehyde
and triethyl phosphite to produce 4c. Finally, acid
hydrolysis of 4c with 6 M HCl gave the desired product
1. Purification consisted of azeotropic distillation with
water to remove excess HCl. The resulting solid was
then lypholized to produce a floculant white solid with
a yield of 95%.21
The commercial availability and low cost of cyclen
coupled with the relative simplicity of this synthesis
(effective monoalkylation, soft/efficient deprotection
R
R´
R
H
H
R
N
N
N
N
N
N
N
N
a
N
R
F
3a: R=CH2PO(OC2H5)2, R´=CH2C9H5NF
3b: R=CH2PO(OC2H5)2, R´=H
3c: R=COOCH2Ph, R´=H
2a: R=CH2PO(OC2H5)2
2b: R=COOCH2Ph
3d: R=COOCH2Ph, R´=CH2C9H5NF
O
R
COC2H5
b
e
N
N
N
N
1
N
R
F
4a: R=COOCH2Ph
4b: R= H
4c: R=CH2PO(OC2H5)2
c
d
Scheme 1. (a) ClCH2C9H5NF, CH3CN, K2CO3; (b) BrCH2COOC2H5, CH3CN, K2CO3; (c) cyclohexene, Pd/C 5%, ethanol reflux;
(d) paraformaldehyde, triethyl phosphite; (e) 6 M HCl; reflux, 4 days.
Table 1. Monosubstituted and disubstituted quinaldine derivatives 3a–d via Scheme 1
Entry
Reagents and conditions
Ratio a:ba
% Yield monosubstituted
% Yield disubstituted
2a
2a
2a
2c
25°C, CH3CN, 24 h, 0.16 M
25°C, CH3CN, 24 h, 0.16 M, K2CO3
25°C, CH3CN, 48 h, 0.04 M
1:1
1:1
2:1
1:1
25
17
30
70
60
64
55
10
40°C, CH3CN, 48 h, 0.04 M, K2CO3
a a: substrate listed under entry; b: quinaldine antenna.

J. M. M. Griffin et al. / Tetrahedron Letters 42 (2001) 3823–3825
3825
and high yield) make this an attractive route to bio-
conjugable lanthanide chelates that require a lanthanide
sensitizer, opposing pendant arms for strong lanthanide
complexation and a single conjugation moiety.
1884, 17, 1699. (b): Leir, C. M. J. Org. Chem. 1977, 42,
911–913. (c) Butera, J. A.; Spinelli, W.; Anantharaman,
V.; Marcopulos, N.; Parsons, R. W.; Moubarak, I. F.;
Cullinan, C.; Bagli, J. F. J. Med. Chem. 1991, 34, 3212–
3228.
16. Kovacs, Z.; Sherry, A. D. Synthesis 1997, 759–763.
17. Compound 3a: 1H NMR (500 MHz, CDCl3) l: 1.16 (t
3H, CH3), 2.74 (br d 12H, NCH2CH2N), 2.89 (br 8H,
NCH2CH2N and PCH2N), 3.86 (d 4H, CH2Ar), 3.93 (q
8H, OCH2), 7.24–7.42 (m 4H, Ar), 7.84 (d 2H, Ar),
8.05–8.09 (m 4H, Ar); 31P NMR (300 MHz, CDCl3) l:
26.26. Compound 3b: 1H NMR (500 MHz, CDCl3) l:
1.16 (t 12H, CH3), 2.49–2.82 (br m 19H, NCH2CH2N,
and HNCH2CH2N), 3.39 (s 4H, NCH2P), 3.83–3.98 (m
10H, OCH2 and CH2Ar), 7.24–7.42 (m 2H, Ar), 7.85 (d
1H, Ar), 7.96 (t 1H, Ar), 8.03 (d 1H, Ar); 31P NMR (300
MHz, CDCl3) l: 26.19.
Acknowledgements
This work was supported by grants from the Whitaker
Foundation and The Southwest Cancer Center.
References
1. Parker, D. Macrocycle Synthesis: A Practical Approach;
Oxford: Oxford University Press, 1996.
2. Corson, D. T.; Meares, C. F. Bioconjugate Chem. 2000,
11, 292–299.
18. (a) Experimental details for 3c: To a stirring solution of
2c (1 g, 2.27 mmol) and K2CO3 (325 mesh, 0.941 g, 6.51
mmol) in dry acetonitrile (57 mL, 0.04 M) was added
2-(chloromethyl)-6-fluoroquinoline (0.444 g, 2.27 mmol).
The solution was then heated to 40°C and allowed to stir
for 48 hours. After reaction completion the K2CO3 was
filtered through Celite then the reaction volume was
reduced producing a slightly yellow viscous oil. The crude
product was then purified on silica using a 30:4:1 ethyl
acetate:methanol:NH4OH(aq) eluent system. Fractions
containing monosubstituted product were collected, the
volume was reduced then redissolved in CH2Cl2 and
washed with NH4OH(aq). The organic layer was dried
over MgSO4 and the volume was reduced producing a
3. Moran, J. K.; Greiner, D. P.; Meares, C. F. Bioconjugate
Chem. 1995, 6, 269–301.
4. Li, Min.; Meares, C. F. Bioconjugate Chem. 1993, 4,
275–283.
5. Leopold, P. L.; Kreitzer, G.; Miyazawa, N.; Rempel, S.;
Pfister, K. K.; Rodreguez-Boulan, E.; Crystal, G. R.
Hum. Gene Ther. 2000, 11, 151–165.
6. Rochon, Y. P.; Kavanagh, T. J.; Harlan, J. M. J.
Microsc. 2000, 197, 15–24.
7. Birchler, M.; Neri, G.; Tarli, L.; Halin, C.; Viti, F.; Neri,
D. J. Immunol. M. 1999, 231, 239–248.
8. Deka, C.; Lehnert, B. E.; Lehnert, N. M.; Jones, G. M.;
Sklar, L. A.; Steinkamp, J. A. Cytometry 1996, 25, 271–
279.
1
viscous colorless oil; (b) H NMR (500 MHz, CDCl3) l:
2.65–2.88 (br d 8H, NCH2CH2N), 3.23–3.61 (br 8H,
NCH2CH2N), 3.85–4.05 (br d 2H, NCH2Ar), 4.53 (d 2H,
CH2Ar), 4.81–4.97 (br d 2H, CH2Ar), 6.85–7.35 (m 10H,
Ar), 7.30–7.53 (m 3H, Ar), 7.72–7.94 (m 2H, Ar).
19. Rylander, P. N. Catalytic Hydrogenation over Platinum
Metals; Academic Press: New York, 1967; p. 385.
20. Honel, M.; Vierhapper, W. J. Chem. Soc., Perkin Trans.
1 1980, 9, 1933–1939.
9. Petit, J. M.; Denis-Gay, M.; Ratinaud, M. H. Biol. Cell.
1993, 78, 1–13.
10. Taylor, J. R.; Fang, M. M.; Nie, S. Anal. Chem. 2000, 72,
1979–1986.
11. Soini, E.; Lo¨vegren, T. CRC Crit. Rev. Anal. Chem. 1987,
18, 105–154.
12. Hubbard, D. S.; Houlne, M. P.; Kiefer, G.; Janseen, H.
F.; Hacker, C.; Bornhop, D. J. Lasers Med. Sci. 1998, 13,
14–21.
21. Compound 1 was isolated in its fully protonated form
and showed a complex proton NMR spectrum in D2O.
The phosphorous NMR spectrum showed a single peak
and high resolution MALDI produced a mass accuracy
within −3.7 ppm (theoretical [M+1]: 578.19394; experi-
13. Parker, D.; Williams, J. A. J. Chem. Soc., Perkin Trans.
2 1996, 1581–1586.
14. Li, M.; Selvin, P. R. J. Am. Chem. Soc. 1995, 117,
8132–8138.
15. Experimental details for synthesis of 6-substituted
quinaldines can be found in Refs. 15a,b. Experimental
details for synthesis of 2-chloromethylene 6-substituted
quinoline can be found in Ref. 15c. (a) Doebner, A.;
vonMiller, W. Berichte 1881, 14, 2816: 1883, 16, 2646:
1
mental [M+1]: 578.1918). H NMR (500 MHz, D2O) l:
2.84–3.89 (br m 24H, NCH2CH2N, NCH2P, NCH2CO,
and NCH2Ar), 7.82–7.88 (m 2H, Ar), 8.05 (d 1H, Ar),
8.36 (m 1H, Ar), 8.97 (d 1H, Ar); 31P NMR (500 MHz,
D2O) l: 6.85 singlet.
.